UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Amidophosphine complexes of electron-poor metals Giesbrecht, Garth Ronald 1998

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1998-271471.pdf [ 19.97MB ]
Metadata
JSON: 831-1.0061549.json
JSON-LD: 831-1.0061549-ld.json
RDF/XML (Pretty): 831-1.0061549-rdf.xml
RDF/JSON: 831-1.0061549-rdf.json
Turtle: 831-1.0061549-turtle.txt
N-Triples: 831-1.0061549-rdf-ntriples.txt
Original Record: 831-1.0061549-source.json
Full Text
831-1.0061549-fulltext.txt
Citation
831-1.0061549.ris

Full Text

AMIDOPHOSPHINE COMPLEXES OF ELECTRON-POOR M E T A L S by Garth Ronald Giesbrecht  B.Sc. (Hons.), University of Winnipeg, 1990  A thesis submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in T H E F A C U L T Y OF G R A D U A T E STUDIES (Department of Chemistry)  We accept this thesis as conforming to the required standard  T H E U N I V E R S I T Y OF BRITISH C O L U M B I A January 1998 © Garth Ronald Giesbrecht, 1998  In  presenting  this  degree at the  thesis  in  partial fulfilment  of  University of  British Columbia,  I agree  freely available for reference copying  of  department  this or  and study.  his  or  her  requirements that the  I further agree  thesis for scholarly purposes by  the  representatives.  may be It  is  publication of this thesis for financial gain shall not be permission.  Department of  Qk^MfeW/  The University of British Columbia Vancouver, Canada  Date  DE-6 (2/88)  that  for  an  advanced  Library shall make it  permission for extensive  granted  by the  understood  that  allowed without  head  of  my  copying  or  my written  ABSTRACT  The preparation of new scandium phosphine complexes that contain two hydrocarbyl groups is reported. Reaction of the amidodiphosphine ligand precursor LiN(SiMe2CH2PPr 2)2 i  with ScCl3(THF)3 leads to the formation of ScCl2(THF)[N(SiMe2CH2PPr 2)2]. Addition of i  alkyl lithium reagents RLi (R = Me, Et, and CH2SiMe3) results in bis(hydrocarbyl) derivatives, ScR2[N(SiMe2CH2PPr 2)2]. A number of reactions with small molecules such as CO, C O 2 and i  H2 were attempted but generally led to decomposition, or reaction at the scandium-amide bond. A chloride ligand was replaced with a cyclopentadienyl ligand to yield Sc(T| 5  C5H )Cl[N(SiMe2CH PPr 2) ]. Metathesis of the chloride ligand with RLi (R = Me, Ph, BH4) i  5  2  2  yields Sc(ri -C5H5)R[N(SiMe2CH2PPr 2)2]. i  5  The methyl complex was inert to H2, but found to  react with amines to produce amide complexes. The borohydride complex reacts with PMe3 to form a putative scandium-hydride complex. To study the effect the amide linkage has on reactivity, ScCl2[T| -C5H3-l,3-(SiMe2CH2PPr 2)2] was synthesized. 5  i  Substitution of the  chlorides with bulky alkyl groups (neopentyl or trimethylsilylmethyl) proceeds; however these dialkyls decompose when exposed to small molecules as a result of oc-hydrogen abstraction. The preparation of new four- and five-coordinate aluminum amidodiphosphine complexes is reported. The reaction of LiN(SiMe2C F^PPr^te with AICI3 leads to the formation of AlCl2[N(SiMe2CH2PPr 2)2]- Addition of the alkyl lithium reagents RLi (R = Me, 1  Et) or dialkyl magnesium reagents R2Mg (R = Me, CFl2Ph) yields AlR [N(SiMe2CH2PPr 2)2]. i  2  Variable temperature NMR studies are consistent with rapidly fluxional species at ambient temperature. These compounds are inert to transformation. The reaction of the lithium salt LiN(SiMe2CH2PPr 2)2 with RAICI2 (R = Me, Et) produces mixtures containing predominantly i  Al(R)X[N(SiMe2CH PPr 2) ] (R = Me, Et). Treatment of the monoethyl derivative with i  2  2  excess AICI3 in an effort to generate a cationic species instead affords the adduct, AlCl2[N(SiMe CH PPr 2)2] • AICI3. i  2  2  ii  The preparation of L i C l , LiAlMe4, LiAlEty, L i B E U and N a B E u adducts of the lithium salt of the tridentate ligand precursor LiN(SiMe2CH2PPr 2)2 is reported. i  The reaction of  HN(SiMe2CH Cl)2 with L i P P r ^ leads to the isolation of { L i N C S i f t ^ C J ^ P P r ^ h h L i C l , under 2  certain conditions. The X-ray crystal structure shows it to exist as a 2:1 adduct with pseudo Ci symmetry in which a L i C l molecule is sandwiched between two LiN(SiMe2CH2PPr 2)2 J  monomers.  Variable temperature  3 1  P and L i N M R spectroscopy indicate that the basic 7  structural features of this compound are maintained in solution. The addition of LiAlMe4 to LiN(SiMe2CH2PPri2)2 results in the formation of {LiN(SiMe2CH2PPr 2)2*LiAlMe4}2- The X i  ray crystal structure of this product indicates that a 2:2 dimer of C2 symmetry is present. Variable temperature N M R studies are consistent with a highly fluxional molecule under ambient  conditions.  The addition of NaBEt4  to LiN(SiMe2CH2PPr 2)2 i  affords  {LiN(SiMe2CH2PPr 2)2*NaBEt4}x- The X-ray crystal structure of this compound shows it to i  be an infinite one-dimensional polymer. Comparison of the three crystal structures illustrates that even with varying adducts (i.e., L i C l , LiAlMe4, and NaBEt4) the basic geometry of the LiN(SiMe2CH2PPr 2)2 unit remains similar. i  Investigations on the preparation of four- and five-coordinate aluminum, gallium and indium bis(amidophosphine) derivatives are reported. The reaction of the macrocyclic ligand precursor anri-Li (THF)2[P2N2] ([P2N2] = [PhP(CH SiMe2NSiMe CH2)2PPh]) with AICI3 or 2  2  2  GaCl3 leads to the formation of the four-coordinate species anri-MCl[P2N2] (M = A l , Ga). The  X-ray crystal structure of a/zft*-GaCl[P2N2] shows the [P2N2] ligand bound through a single phosphorus donor, resulting in the retention of the ami- conformation. The addition of AICI3, GaCl3  and InCl3 to s;y«-Li2(dioxane)[P2N2] yields the five-coordinate complexes syn-  MCIIP2N2] ( M = A l , Ga, In).  The X-ray crystal structure of 5yrt-GaCl[P2N2] reveals the  coordination of both phosphorus atoms, with retention of the syn- conformation. Heating the anti- complexes results in the clean conversion to the syn- complexes, with pyramidal inversion observed at phosphorus. Barriers to pyramidal inversion (AG*) were calculated to be 29.1 and 30.1 kcal/mol for aluminum and gallium, respectively; these are ~2 to 3 kcal/mol lower than iii  that determined for the metal-free compounds anti- and syn-H2[P2N2]. The complexes synMC1[P2N2] ( M = A l , Ga, In) were suitable starting materials for the generation of monomelic aluminum and gallium hydrides yy«-MH[P2N2] ( M = A l , Ga). Reduced forms of gallium and indium were also accessible; yyrc-MCl[P2N2] ( M = Ga, In) could be reduced with KCs to yield the dimers {^y«-M[P2N2] }2 ( M = Ga, In).  iv  T A B L E OF CONTENTS ABSTRACT  ii  T A B L E OF CONTENTS  v  LIST OF FIGURES  xiv  LIST OF TABLES  xx  GLOSSARY OF TERMS  xxiii  ACKNOWLEDGMENTS  xxviii  Chapter 1:  INTRODUCTION T O AMIDOPHOSPHINE C O M P L E X E S O F ELECTRON-POOR METALS  1  1.1 Introduction  1  1.2 Organometallic Complexes of Scandium  3  1.3 Organometallic Complexes of Aluminum  6  1.4 Mono- and Non-Cyclopentadienyl Derivatives of the Group 3 Metals and the Lanthanides  8  1.5 Phosphine Complexes of the Group 3 Metals  9  1.6 Phosphine Complexes of the Group 13 Metals  11  1.7 Summary of Previous Work from our Laboratory  12  1.7.1 First Generation Ligand Design  12  1.7.2 Second Generation Ligand Design  14  1.7.3 Third Generation Ligand Design  15  1.8 Group 3 Complexes of the PNP and P2N2 Ligand Systems  17  1.9 The Scope of this Thesis  19  1.10 References  21  v  Chapter 2: SYNTHESIS, C H A R A C T E R I Z A T I O N AND R E A C T I V I T Y O F PHOSPHINE COMPLEXES OF SCANDIUM  30  2.1 Introduction  30  2.2.1 Synthesis of ScCl (THF)[N(SiMe2CH PPr 2)2] (1)  31  2.2.2 Structure of ScCl2(THF)[N(SiMe2CH PPr )2] (1)  32  i  2  2  i  2  2  2.3.1 Synthesis of Scandium Dialkyls ScR [N(SiMe2CH PPr 2)2] (R = Me (2), i  2  2  Et (3), CH SiMe (4)) 2  34  3  2.3.2 Structure of ScEt2[N(SiMe CH PPr 2)2] (3)  35  i  2  2  2.3.3 Structure of Sc(CH SiMe3)2[N(SiMe2CH PPr 2)2] (4)  38  i  2  2  2.4 Solvent Dependency of the Reaction Between ScCl2(THF)[N(SiMe2CH PPr )2] (1) and MeLi (2 equiv)  39  i  2  2  2.5 Reactivity of Scandium Dialkyls ScR2[N(SiMe2CH PPr 2)2] (R = Me (2), i  2  Et(3), C H S i M e 3 (4))  47  2  2.6 Electronic Structure of the Model Compound ScMe2[N(SiMe2CH2PH2)2]  50  2.7 Monoalkyl Scandium Complexes Utilizing the N(SiMe2CH2PPr 2)2 i  Ligand System  53  2.8.1 Synthesis of Sc(ri5-C5H )Cl[N(SiMe CH2PPr 2)2] (6)  54  2.8.2 Structure of Sc(Ti5-C H5)Cl[N(SiMe CH2PPr 2)2] (6)  55  i  5  2  i  5  2  2.9.1 Synthesis of Scandium Complexes Sc(rj -C5H )R[N(SiMe2CH2PPr 2)2] 5  i  5  (R = Me (7), Ph (8), BH4 (9))  57  2.9.2 Structure of Sc(ri5-C5H )Ph[N(SiMe2CH2PPr 2)2] (8)  58  2.9.3 Structure of Sc(Ti -C5H )(BH4)[N(SiMe2CH2PPr ) ] (9)  60  2.10.1 Reactivity of Sc(ri5-C H5)Me[N(SiMe2CH2PPr 2)2] (7)  61  2.10.2 Reactivity of Sc(Ti5-C5H5)(BH4)[N(SiMe2CH PPr 2)2] (9)  66  i  5  5  i  5  2  2  i  5  i  2  2.11 Scandium Dialkyl Complexes Utilizing the Tl -C5H3-1,35  (SiMe CH2PPr 2) Ligand System  71  i  2  2  2.12.1 Synthesis of ScCl2[Tl -C H3-l,3-(SiMe2CH2PPr 2)2] (13) 5  i  5  vi  72  2.12.2 Structure of ScCl [Tl -C5H3-l,3-(SiMe2CH2PPri2)2] (13)  73  5  2  2.13 Synthesis of Scandium Dialkyls ScR2[ri -C5H3-l,3-(SiMe2CH PPr 2) ] 5  i  2  2  (R = CH CMe3 (14), CH SiMe (15)) 2  2  75  3  2.14 Reactivity of ScR2[rt -C H3-l,3-(SiMe2CH PPr 2)2] (R = C H C M e (14), 5  i  5  CH CMe 2  3  2  2  3  (15))  76  2.15 Conclusions  77  2.16 Experimental Section  78  2.16.1 Procedures  78  2.16.2 Materials  79  2.16.3 Syntheses  80  2.16.3.1 ScCl2(THF)[N(SiMe CH PPr 2)2] (1)  80  2.16.3.2 ScMe2[N(SiMe CH PPr 2)2] (2)  80  2.16.3.3 ScEt2[N(SiMe CH PPr 2)2]  81  i  2  2  i  2  2  i  2  2  (3)  2.16.3.4 Sc(CH2SiMe3)2[N(SiMe CH PPr 2)2] (4)  82  i  2  2  2.16.3.5 (Et 0)Li(n-Cl)(|i-Me)ScMe2  [N(SiMe CH PPr )2]LiCl(Et20) (5) i  2  2  82  2  2.16.3.6 Reaction of Sc(CH SiMe3)2[N(SiMe2CH2PPr 2) ] (4) i  2  2  with CO  83  2.16.3.7 Reaction of ScEt2[N(SiMe CH PPr 2)2] (3) with C 0 i  2  2  2.16.3.8 Sc(Ti5-C H5)Cl[N(SiMe CH2PPr 2)2] 5  (6)  i  2  2  84  2.16.3.9 Sc(ri5-C H5)Me[N(SiMe CH2PPr )2] (7) i  2  5  84  85  2  2.16.3.10 Reaction of ScR2[N(SiMe2CH PPri2)2] (R = M e (2), 2  Et (3)) and Sc(Tl5-C H5)Me[N(SiMe2CH PPr ) ] (7) i  5  2  with C2H4  2  2  86  2.16.3.11 Sc(T| -C H5)Ph[N(SiMe2CH PPr 2)2] (8)  87  2.16.3.12 Sc(Ti5-C5H5)(BH4)[N(SiMe2CH2PPri )2] (9)  87  2.16.3.13 Sc(Ti5-C H5)NHPh[N(SiMe2CH2PPr 2)2] (10)  88  5  i  5  2  2  i  5  vii  2.16.3.14 Sc(Ti5-C H )NHBu [N(SiMe2CH2PPT 2)2] (11)  89  2.16.3.15 Sc(ri5-C H5)H[N(SiMe2CH PPr 2)2] (12)  90  2.16.3.16 ScCl [(ri5-C5H3-l,3-(SiMe2CH2PPr 2)2] (13)  91  t  5  i  5  i  5  2  i  2  2.16.3.17 Reaction of ScCl2(THF)[N(SiMe2CH PPr 2)2] (1), i  2  Sc(rj5-C H5)Cl[N(SiMe2CH2PPr2)2] (6) and i  5  ScCl [(ri -C5H3-l,3-(SiMe2CH2PPr 2)2] (13) with 5  i  2  C H 2  91  4  2.16.3.18 Sc(CH CMe3) [(ri5-C5H3-l,3-(SiMe2CH2PPr 2)2] (14)  92  2.16.3.19 Sc(CH SiMe )2[(Tl -C5H3-l,3-(SiMe2CH2PPri2)2] (15)  92  2.16.3.20 Molecular Orbital Calculations  93  i  2  2  5  2  3  2.17 References  94  Chapter 3: SYNTHESIS, C H A R A C T E R I Z A T I O N AND R E A C T I V I T Y O F AMIDOPHOSPHINE COMPLEXES OF ALUMINUM  99  3.1 Introduction  99  3.2.1 Synthesis of AlCl2[N(SiMe CH PPr 2)2] (1)  100  3.2.2 Structure of AlCl2[N(SiMe2CH PPrJ2)2] (1)  101  i  2  2  2  3.3.1 Synthesis of Aluminum Dialkyls AlR2[N(SiMe CH PPr 2)2J (R = Me (2), i  2  2  Et (3), CH Ph(4))  104  2  3.3.2 Structure of Al(CH Ph)2[N(SiMe CH2PPr 2)2] (4)  105  i  2  3.3.3  3 1  2  P and A1 NMR Studies and Coordination Number of 27  AlR2[N(SiMe CH PPr 2)2] (R = Me (2), Et (3), CH Ph (4))  107  i  2  2  2  3.4 Reactivity of Aluminum Dialkyls AlR2[N(SiMe2CH2PPr )2] (R = Me (2), i  2  Et (3), CH Ph (4))  117  2  3.5 Synthesis of the Monoalkyl Derivatives Al(R)Cl[N(SiMe2CH PPr 2)2] i  2  (R = Me(5), Et (6))  117  3.6.1 Synthesis of AlCl [N(SiMe2CH PPr 2)2]*AlCl3 (7) i  2  2  viii  120  3.6.2 Structure of AlCl [N(SiMe CH2PPr 2)2]»AlCl3 (7)  121  i  2  2  3.6.3 Mechanism of Formation of AlCl2[N(SiMe CH2PPr 2) ]»AlCl3 (7)  124  3.7 Conclusions  126  3.8 Experimental  127  i  2  2  3.8.1 Procedures  127  3.8.2 Materials  127  3.8.3 Syntheses  128  3.8.3.1  A l C l [ N ( S i M e C H P P r _) ] (1)  128  3.8.3.2  AlMe [N(SiMe CH2PPr )2]  128  3.8.3.3  AlEt [N(SiMe2CH PPr )2]  3.8.3.4  Al(CH2Ph) [N(SiMe CH PPri2) ] (4)  130  3.8.3.5  Al(Me)Cl[N(SiMe CH PPr 2)2]  131  3.8.3.6  Al(Et)Cl[N(SiMe2CH PPr_)2]  3.8.3.7  AlCl [N(SiMe CH PPr_) ].AlCl3  2  2  2  2  i  2  2  2  i  2  2  2  2  2  (2)  129  (3)  2  2  i  2  2  2  2  (5)..'  2  2  131  (6)  2  132  (7)  3.9 References  132  Chapter 4: SYNTHESIS, CHARACTERIZATION AND SOLUTION DYNAMICS OF A L K A L I - M E T A L CHLORIDE, A L U M I N A T E AND B O R A T E ADDUCTS OF T H E TRIDENT A T E AMIDOPHOSPHINE LIGAND PRECURSOR LiN(SiMe CH PPr 2)2  138  i  2  2  4.1 Introduction  138  4.2.1 Synthesis of { L i N ( S i M e C H P P r ) } L i C l (1)  139  4.2.2 Structure of { L i N C S i M e ^ H ^ P r ^ h h L i C l (1)  140  i  2  2  2  2  2  4.2.3 Variable Temperature N M R Studies of { L i N ( S i M e C H P P r ) } 2 L i C l (1)  143  4.2.4 Mechanism of Formation of {LiN(SiMe2CH2PPr 2)2} LiCl (1)  150  4.3.1 Synthesis of {LiN(SiMe2CH PPr 2) «LiAlMe }2 (2)  152  4.3.2 Structure of {LiN(SiMe CH2PPr )2*LiAlMe4}2 (2)  154  i  2  2  i  2  i  2  2  i  2  2  ix  4  2  2  4.3.3 Variable Temperature NMR Studies of {LiN(SiMe2CH2PPr )2'LiAlMe } 2 (2)  156  i  2  4  4.3.4 Mechanism of Formation of {LiN(SiMe2CH2PPr 2)2*LiAlMe4}2 (2)  162  4.4.1  164  i  LiAlEt4, LiBEty and NaBEt4 Adduct Syntheses  4.4.2 Structure of {LiN(SiMe2CH PPr )2*NaBEu}x (6)  165  4.5 Similarity of Ligand Geometry and Trends in Adduct Formation  169  4.6 Conclusions  170  4.7 Experimental  170  i  2  2  4.7.1 Procedures  -170  4.7.2 Materials  171  4.7.3 Syntheses  171  4.7.3.1 {LiN(SiMe2CH PPr )2}2LiCl (1)  171  i  2  2  4.7.3.2 {LiN(SiMe CH2PPr ) »LiAlMe } (2)  172  4.7.3.3 LiN(SiMe2CH2PPri2)2«AlMe3 (3)  173  4.7.3.4 LiN(SiMe CH PPr 2)2'LiAlEt4 (4)  174  4.7.3.5 LiN(SiMe2CH2PPri2)2*LiBEt (5)  174  4.7.3.6 {LiN(SiMe CH PPr 2)2'NaBEt } (6)  175  i  2  2  2  4  2  i  2  2  4  i  2  2  4  x  4.8 References Chapter 5:  175  SYNTHESIS, C H A R A C T E R I Z A T I O N AND R E A C T I V I T Y O F M A C R O C Y C L I C BIS(AMIDOPHOSPHINE) C O M P L E X E S OF ALUMINUM, G A L L I U M AND INDIUM  5.1 Introduction  179 179  5.2 Pyramidal Inversion at Phosphorus Facilitated by the Presence of Aluminum, Gallium and Indium  180  5.3.1 Synthesis of anti-[P2^2\ Complexes of Aluminum and Gallium  183  5.3.2 Structure of anr/-GaCl[P2N ] (2)  185  5.4.1 Synthesis of ^y«-[P2N2] Complexes of Aluminum, Gallium and Indium  187  2  5.4.2 Structure of syn-AlCl[P N ] (3)  189  5.4.3 Structure of sy«-GaCl[P N ] (4)  192  2  2  2  2  5.5.1 Conversion of „nfi-MCl[P N ] Complexes to yyn-MCl[P N2] Complexes 2  2  2  (M = Al, Ga)  193  5.5.2 Pyramidal Inversion at Phosphorus in „nfi-H [P N ] and syn-H [P N ] 2  2  2  2  2  2  200  5.5.3 Mechanism of Pyramidal Inversion at Phosphorus in MC1[P N ] 2  2  (M = A L Ga)  207  5.6 Hydrides of Aluminum, Gallium and Indium  212  5.6.1 Synthesis of yyn-[P N ] Hydride Complexes of Aluminum and 2  2  Gallium  213  5.6.2 Structure of 5y«-AlH[P N ] (6) 2  217  2  5.6.3 Solution structure of ^yrt-AlH[P N ] (6)  220  5.6.4 Reactivity of syn-A1H[P N ] (6)  222  2  2  2  2  5.7 Metal-Metal Bonded Dimers of Aluminum, Gallium and Indium  222  5.7.1 Synthesis of .syrc-[P N ] Metal-Metal Bonded Dimers of 2  2  Gallium and Indium  ..223  5.7.2 Structure of {^yn-Ga[P N ]} (8)  225  5.7.3 Structure of {sy«-In[P N ] } (9)  227  2  2  2  2  2  2  5.7.4 Solution structure of {5y«-Ga[P N ]} (8) and {iT«-In[P N ]} (9) ....228 2  2  2  2  2  2  5.8 Conclusions  229  5.9 Experimental  230  5.9.1 Procedures  230  5.9.2 Materials  '.  5.9.3 Syntheses  230 231  5.9.3.1 a«ri-AlCl[P N ] (1)  231  5.9.3.2 „/m-GaCl[P N ] (2)  231  5.9.3.3 sy«-AlCl[P N ] (3)  232  2  2  2  2  2  2  xi  5.9.3.4 sy«-GaCl[P N ] (4)  232  5.9.3.5 syn-InCl[P N ] (5)  233  5.9.3.6 syn-AlH[P N ] (6)  234  5.9.3.7 syn-GaH[P N ] (7)  235  5.9.3.8 {.ryrc-Ga[P N ] } (8)  236  5.9.3.9 { ^ « - I n [ P N ] } ( 9 )  236  2  2  2  2  2  2  2  2  2  2  2  2  2  2  5.9.4 Kinetics of the Inversion Reactions  237  5.10 References  237  Chapter 6: FUTURE W O R K  246  6.1 Introduction  246  6.2 Scandium Compounds Involving New Ligands  246  6.2.1 Synthesis and Structure of ScCl [N(SiMe CH NMe ) ] (1) 2  2  2  2  2  6.3 Bis(amidophosphine) Complexes of Titanium  248 251  6.3.1 Synthesis and Structure of a«ri-TiCl [P N ] (2) 2  2  251  2  6.3.2 Phosphorus Inversion Behavior of a«ri-TiCl [P N ] (2)  253  6.3.3 Reactivity of anri-TiCl [P N ] (2)  254  2  2  2  2  2  2  6.4 Bis(arnidophosphine) Complexes of Vanadium  254  6.4.1 Synthesis of yy«-VCl[P N ] (3) 2  255  2  6.4.2 Synthesis and Structure of {syn-[P N ]V} (^i-N ) (4) 2  2  2  2  256  6.5 Synthesis and Structure of a«ri-(CdCl) [P N ] (5)  261  6.6 Conclusions  266  6.7 Experimental Section  266  2  2  2  6.7.1 Procedures  266  6.7.2 Materials  267  6.7.3 Syntheses  267  6.7.3.1 ScCl [N(SiMe CH NMe ) ] (1)  267  6.7.3.2 <wri-TiCl [P N ] (2)  268  2  2  2  2  2  2  xii  2  2  6.7.3.3 yy»-VCl[P N ] (3) 2  268  2  6.7.3.4 {sy«-[P N ]V} (n-N ) (4)  268  6.7.3.5 a«n--(CdCl) [P N ] (5)  269  2  2  2  2  2  2  2  6.8 References  270  APPENDIX 1: X-ray Crystal Structure Data  273  APPENDIX 2: Cartesian Coordinates for the Model Compound ScMe [N(SiMe CH PH ) ] 2  2  2  2  322  2  APPENDIX 3: Kinetic Data for the Conversion of syn-H2[^2^2\ to a«ri-H [P N ] in 2  4:1 C6D Br/C6D 5  2  2  325  6  xiii  List of Figures Figure Figure 2.1  Caption  Page  Molecular structure of ScCl2(THF)[N(SiMe2CH PPr _)2] (1); i  2  32  33% probability thermal ellipsoids are shown. Hydrogen atoms are omitted for clarity. Figure 2.2  Molecular structure of ScEt2[N(SiMe2_H2PPr )2] (3); 33% 1  2  36  probability thermal ellipsoids are shown. Hydrogen atoms are omitted for clarity. Figure 2.3  Molecular structure of Sc(CH SiMe3)2[N(SiMe CH2PPr _) ] (4); 2  2  2  38  33% probability thermal ellipsoids are shown. Hydrogen atoms are omitted for clarity. Figure 2.4  202.4 MHz P{ lH} NMR spectrum of 5 in _6D .  40  Figure 2.5  Splitting diagram for an ABX spin system (A,B = P , X = Li).  40  Figure 2.6  NMR-II simulations of the P{ *H} NMR spectra of a) 5a alone,  41  31  6  31  7  31  b) 5b alone and c) 5a and 5b together. Figure 2.7  202.4 MHz 31p.3lp COSY NMR spectrum of 5a and 5b in C D .  42  Figure 2.8  Tetrahedral lithium centre proposed to be present in 5a and 5b.  43  Figure 2.9  194.4 MHz Li{ !H} NMR spectrum of 5a and 5b in _6D .  44  Figure 2.10  Proposed  46  6  6  7  6  solution  structure  of  (Et20)Li(|i-Cl)(p.-  Me)ScMe[N(SiMe2CH PPr 2) ]LiCl(Et20) (5). i  2  Figure 2.11  2  Frontier orbitals of bis(pentamethylcyclopentadienyl)scandium  50  alkyls. Figure 2.12  Molecular  orbital  diagram  ScMe2[N(SiMe2CH PH ) ]. 2  2  2  xiv  for the model  compound  52  Figure 2.13  Molecular structure of Sc(Ti5-C H5)Cl[N(SiMe2CH2PPr 2)2] (6); i  5  56  33% probability thermal ellipsoids are shown. Hydrogen atoms are omitted for clarity. Figure 2.14  Molecular structure of Sc(r|5-C5H5)Ph[N(SiMe2CH2PPr 2)2] (8); i  59  33% probability thermal ellipsoids are shown. Hydrogen atoms are omitted for clarity. Figure2.15  96.3 MHz HBf !!} NMR spectrum of Sc(Ti5-C H )(BH4)1  5  5  67  [N(SiMe2CH2PPri2)2] (9) and 9' in C6D6Figure 2.16  Variable temperature 96.3 MHz B{ H} NMR spectra for the 11  1  reaction between Sc(Ti -C H )(BH4)[N(SiMe2CH2PPr 2)2] 5  (9)  i  5  68  5  and PMe3 (-100 equiv) in C6D6. Figure 2.17  van't Hoff plot for the equilibrium between Sc(rj S-CsHs)-  70  (BH4)[N(SiMe2CH PPri2)2] (9), PMe , PMe »BH and Sc(rj52  3  3  3  C H5)H[N(SiMe CH PPr 2)2] (12) (r = 0.997). i  5  Figure 2.18  2  2  Molecular structure of ScCl2[Tl -C5H3-l,3-(SiMe2CH2PPr )2] 5  i  2  74  (13); 33% probability thermal ellipsoids are shown. Hydrogen atoms are omitted for clarity. Figure 3.1  Molecular structure of AlCl [N(SiMe2CH PPr 2)2] (1); i  2  2  33%  102  probability thermal ellipsoids are shown. Hydrogen atoms and disorder not shown for clarity. Figure 3.2  Molecular structure of Al(CH2Ph) [N(SiMe CH2PPr 2)2] (4); i  2  2  106  33% probability thermal ellipsoids are shown. Hydrogen atoms are omitted for clarity. Figure 3.3  Aluminum-27 chemical shift range as a function of coordination number.  xv  108  Figure3.4  161.9  MHz  3 1  P{ H}  M A S NMR spectrum  1  of  109  AlCl2[N(SiMe2CH2PPr 2) ] (1) with a spinning rate of 4.551 i  2  kHz. The small multiplets are spinning side bands. Figure 3.5  Variable temperature 202.4 MHz P{!H} NMR spectra of 31  112  Al(CH Ph)2[N(SiMe CH2PPr 2)2] (4) in CD C1 . i  2  Figure3.6  2  2  2  Variable temperature 121.4 MHz P{ H} NMR spectra of 31  1  114  Al(CH Ph) [N(SiMe2CH PPr 2)2] (4) in C D 0 . i  2  Figure 3.7  van't  2  2  Hoff  plot  4  for  the  8  equilibrium  between  119  AlCl [N(SiMe2CH PPr 2)2] (1), AlMe2[N(SiMe CH PPri2)2] (2) i  2  2  2  2  and Al(Me)Cl[N(SiMe2CH PPr 2)2] (5). i  2  Figure 3.8  Molecular structure of AlCl2[N(SiMe2CH PPr 2)2]»AlCl3 (7); i  2  122  33% probability thermal ellipsoids are shown. Hydrogen atoms are omitted for clarity. Figure 4.1  Molecular structure of {LiN(SiMe2CH PPr . h h L i C l (1); 33% 2  141  probability thermal ellipsoids are shown. Hydrogen atoms are omitted for clarity. Figure4.2  Variable temperature 121.4 MHz P{ H} NMR spectra of 31  1  {LiN(SiMe CH2PPr 2)2}2LiCl (1) in C D . i  2  7  8  144  The peak marked  with an asterisk (*) is due to LiN(SiMe2CH2PPri2)2. Figure 4.3  Experimental (left) and simulated (right) variable temperature 116.7  MHz  7  L i { H} J  NMR  spectra  of  spectrum  of  145  {LiN(SiMe2CH PPri2)2}2LiCl (1) in C D . 2  Figure4.4  194.5  7  MHz  7  Li  NOESY  8  NMR  {Li N(SiMe2CH PPr 2) }2LiCl (lb) in C D 15  i  2  2  7  8  at 223 K. The  small coupling present is due to L i - N coupling (UU-ISN = 5.8 7  Hz).  xvi  1 5  147  Figure 4.5  Arrhenius plot for the apparent exchange of lithium centres in  148  {LiN(SiMe CH PPr )2}2LiCl (1). i  2  Figure 4.6  2  2  Molecular structure of {LiN(SiMe2CH2PPr^)2»LiAlMe4h (2);  154  33% probability thermal ellipsoids are shown. Hydrogen atoms are omitted for clarity. Figure 4.7  Variable temperature 44.1 MHz L i { H } NMR spectra of 6  1  158  {6Lil5N(SiMe2CH2PPr 2)2' LiAlMe4}2 (2b) in C7D8/pentane. i  Figure 4.8  6  Arrhenius plot for the exchange of lithium centres in  159  {6Li 5N(SiMe2CH PPr 2)2» LiAlMe } 2 (2b). 1  i  6  2  Figure 4.9  4  a) Molecular structure of {LiN(SiMe2CH2PPri2)2»NaBEt4}x (6);  167  33% probability thermal ellipsoids are shown. Hydrogen atoms are omitted for clarity, b) Chem 3D view of the monomer unit of {LiN(SiMe2CH2PPrJ2)2'NaBEt4}xFigure 5.1  a) Molecular structure of a«ft'-GaCl[P2N2] (2); 33% probability  186  thermal ellipsoids are shown. Hydrogen atoms are omitted for clarity, b) Chem 3D view of a«ri-GaCl[P N2] (2). 2  Figure 5.2  a) Molecular structure of jyn-AlCl[P N2] (3); 33% probability 2  190  thermal ellipsoids are shown. Hydrogen atoms are omitted for clarity, b) Chem 3D view of sy/i-AlCl[P2N2] (3). Figure 5.3  Molecular structure of syrc-GaCl[P2N2] (4); 33% probability  192  thermal ellipsoids are shown. Hydrogen atoms are omitted for clarity. Figure 5.4  First-order rate plots for the conversion of awri-AlCl[P2N2] (1) to  195  syn-A1C1[P2N ] (3) in C D . 2  Figure 5.5  7  8  First-order rate plots for the conversion of anri-GaCl[P2N ] (2) to 2  5y«-GaCl[P2N2] (4) in C D . 7  8  xvii  196  Figure 5.6  Arrhenius plots for the conversion of anri-MCl[P2N2] to  197  yyn-MCl[P N ] (M = Al, Ga) in C D . 2  Figure 5.7  2  7  8  Concentration study for the conversion of anri-AlCl[P2N2] (1) to  198  yyn-AlCl[P2N ] (3) in C D . 2  Figure 5.8  7  8  a) Rate plots for the conversion of „«ri-H2[P N2] to  202  2  yyn-H [P N ] in C D . b) Pre-equilibrium first order rate plots 2  2  2  7  8  for the conversion of „«r/-H2[P2N2] to syn-H2\^2^2] in C D . 7  Figure 5.9  8  a) Rate plots for the conversion of syn-H [P2N2] to  203  2  a«ft'-H [P2N2] in C D . b) Pre-equilibrium first order rate plots 2  7  8  for the conversion of ^yn-H2[P N2] to „«ri-H [P2N2] in C D . 2  Figure 5.10  2  7  Arrhenius plots for the conversion of anti/syn-H2[P2^2]  8  t o  205  syn/anti-H2[^2^2\ in C D . 7  8  Figure 5.11  Transition state o~-overlap diagrams for MC1[P2N2] (M = Al, Ga).  210  Figure 5.12  Transition state a-overlap diagrams for MCHP2N2]  211  (M = Al, Ga, Sc). Figure 5.13  Molecular structure of ^yn-AlH[P2N2] (6); 33% probability  218  thermal ellipsoids are shown. Hydrogen atoms are omitted for clarity. Figure 5.14  Molecular structure of {,sy/t-Ga[P2N2]}2 (8); 50% probability  225  thermal ellipsoids are shown. Hydrogen atoms, silyl methyls and phenyl rings are omitted for clarity. Figure 5.15  Molecular structure of {^yn-In[P2N2]}2 (9); 33% probability  227  thermal ellipsoids are shown. Hydrogen atoms, silyl methyls and phenyl rings are omitted for clarity. Figure 6.1  Molecular structure of ScCl2[N(SiMe CH NMe2) ] (1); 33% 2  2  2  probability thermal ellipsoids are shown. Hydrogen atoms are omitted for clarity. xviii  249  Figure 6.2  Molecular structure of a«ri-TiCl2[P2N2] (2); 33% probability  252  thermal ellipsoids are shown. Hydrogen atoms are omitted for clarity. Figure 6.3  a) Experimental and b) simulated room temperature solution  258  ESR spectrum of {5yn-[P2N ]V} 0i-N2) (4). 2  Figure 6.4  Molecular structure  of  2  {^«-[P2N2]V)2(|i-N2) ( 4);  33%  259  probability thermal ellipsoids are shown. Hydrogen atoms are omitted for clarity. Figure 6.5  a) 202.4 MHz 31p{lH} and b) 66.7 MHz HSCdOH} NMR  263  spectra of ann'-(CdCl) [P2N2] (5) in C6D6. Satellites in the 2  31  Figure 6.6  P{ !H} NMR spectrum are due to Vp-Cd couplings.  Molecular structure of a«ft'-(CdCl) [P2N2] (5); 33% probability 2  264  thermal ellipsoids are shown. Hydrogen atoms are omitted for clarity. Figure A2.1  Chem 3D view of the model compound  322  ScMe2[N(SiMe CH2PH ) ]. 2  Figure A3.1  2  2  a) Rate plots for the conversion of syn-K2i^2^2\ to  325  anft'-H2[P2N2] in4:l C6D Br/C6D6. b) Pre-equilibrium first 5  order rate plots for the conversion of syn-H2[P2^2\ to aitft-H2[P2N2] in 4:1 C 6 D B r / C D . 5  Figure A3.2  6  6  Arrhenius plot for the conversion of syn-R2[P2^2] to anri-H [P2N ] in 4:1 C D B r / C D . 2  2  6  5  6  xix  6  326  List of Tables Table  Table 1.1  Title  Page  Structurally Characterized Phosphine Complexes of the Group 3  10  Metals. Table 2.1  Selected bond lengths (A) and angles (deg) for  34  ScCl2(THF)[N(SiMe CH PPr 2)2] (1). i  2  Table 2.2  2  Selected bond lengths (A) and angles (deg) for ScEt [N(SiMe CH PPr ) ] (3) i  2  2  2  (left) and Sc(CH SiMe ) -  a  2  37  2  2  3  2  [N(SiMe CH PPri ) ] (4) (right). 2  2  2  2  Table 2.3  Microanalytical data for 5.  45  Table 2.4  Selected bond lengths (A) and angles (deg) for  57  Sc(ri5.C H5)Cl[N(SiMe CH PPri ) ](6). 5  Table 2.5  2  2  2  2  Selected bond lengths (A) and angles (deg) for  59  Sc(Ti5-C H )Ph[N(SiMe CH PPri ) ](8). 5  Table 2.6  5  2  2  2  2  van't Hoff data for the equilibrium between Sc(r| S-CsHs)-  70  (BH )[N(SiMe CH PPr ) ] (9), PMe , PMe «BH and Sc(ri5i  4  2  2  2  2  3  3  3  C H )H[N(SiMe CH PPri ) ] (12). 5  Table 2.7  5  2  2  2  2  Selected bond lengths (A) and angles (deg) for  74  ScCl [ri5-C H -l,3-(SiMe CH PPri ) ](13). 2  Table 3.1  5  3  2  2  2  2  Selected bond lengths (A) and angles (deg) for  103  AlCl [N(SiMe CH PPri ) ] (1). 2  Table 3.2  2  2  2  2  Selected bond lengths (A) and angles (deg) for  107  Al(CH Ph) [N(SiMe CH PPri2) ] (4). 2  Table 3.3  van't  2  2  Hoff  2  data  2  for  the  equilibrium  between  AlCl [N(SiMe CH PPri2) ] (1), AlMe [N(SiMe CH PPri ) ] (2) 2  2  2  2  2  and Al(Me)Cl[N(SiMe CH PPr ) ] (5). i  2  2  2  xx  2  2  2  2  2  119  Table 3.4  Selected bond lengths (A) and angles (deg) for  123  AlCl2[N(SiMe CH2PPr 2)2]»AlCl3 (7). i  2  Table 4.1  Selected bond lengths (A) and angles (deg) for  142  {LiN(SiMe2CH PPr ) }2LiCl (1). i  2  Table 4.2  2  2  Arrhenius data for the apparent exchange of lithium centres in  149  {LiN(SiMe CH2PPr )2} LiCl (1). i  2  Table 4.3  2  2  Selected bond lengths (A) and angles (deg) for  156  {LiN(SiMe CH2PPr 2)2«LiAlMe4} 2 (2). i  a  2  Table 4.4  Arrhenius data for the exchange of lithium centres in  159  {6Li 5N(SiMe CH PPr 2)2« LiAlMe4} (2b). 1  i  2  Table 4.5  6  2  2  Selected bond lengths (A) and angles (deg) for  168  {LiN(SiMe CH PPr 2)2'NaBEt4} (6). i  2  Table 4.6  2  x  Selected Bond Lengths in {LiN(SiMe CH PPr )2}2LiCl (1), i  2  2  2  169  {LiN(SiMe2CH PPr )2*LiAlMe } 2 (2), i  2  2  4  and {LiN(SiMe2CH PPr 2)2«NaBEt4} (6). i  2  Table 4.7  x  Selected Bond Angles in {LiN(SiMe CH PPri2) } LiCl (1), 2  2  2  2  170  {LiN(SiMe CH PPr_)2*Li AlMe } (2), 2  2  4  2  and {LiN(SiMe2CH PPr 2)2«NaBEt4} (6). i  2  Table 5.1  x  Selected bond lengths (A) and angles (deg) for  187  „«ri-GaCl[P2N ] (2). 2  Table 5.2  Selected bond lengths (A) and angles (deg) for  191  A7«-A1C1[P2N ] (3) and5y«-GaCl[P N ] (4). 2  Table 5.3  2  2  Rate data for the conversion of „«r/-MCl[P N2] to 2  196  ^y/i-MCl[P N2] (M = Al, Ga) in C D . 2  Table 5.4  7  8  Transition state parameters for the conversion of „«ri-MCl[P2N2] to S3W-MC1[P2N2] (M = Al, Ga) in C D . 7  xxi  8  197  Table 5.5  Rate data for the conversion of attri/^«-H [P N ] to 2  204  Transition state parameters for the conversion of  205  2  2  sy«/artn-H [P N ] in C7D8. 2  Table 5.6  2  2  ann7syrt-H [P N ] to syn/a«n'-H [P N ] in C7D8. 2  Table 5.7  2  2  2  2  2  Selected bond lengths (A) and angles (deg) for  220  sy«-AlH[P N ] (6). 2  Table 5.8  2  Selected bond lengths (A) and angles (deg) for  226  {sy«-Ga[P N ]} (8) and {syn-In[P N ]} (9). 2  Table 6.1  2  2  2  2  Selected bond lengths (A) and angles (deg) for ScCl [N(SiMe CH NMe ) ] (l). 2  Table 6.2  2  2  2  2  249  a  2  Selected bond lengths (A) and angles (deg) for  253  a/m -TiCl [P N ] (2). -  2  Table 6.3  2  2  Selected bond lengths (A) and angles (deg) for  260  {syn-[P N ]V} (|i-N )(4). 2  Table 6.4  2  2  2  Selected bond lengths (A) and angles (deg) for  265  anft--(CdCl) [P N ] (5)» 2  2  2  Table A l . l  Crystallographic Data, Fractional Atomic Coordinates and  to A1.27  Beq (A ) for reported crystal structures (twenty two sets).  Table A2.1  Cartesian coordinates for the model compound  273 to 321  2  322  ScMe [N(SiMe CH PH ) ]. 2  Table A3.1  2  2  2  2  Rate data for the conversion of jy/i-H [P N ] to arcft'-H [P N ] in 2  2  2  2  2  2  326  4:1 C D Br/C6D . 6  Table A3.2  5  6  Transition state parameters for the conversion of 5y«-H [P N ] to 2  a«ri-H [P N ] in 4:1 C6D Br/C D6. 2  2  2  5  6  xxii  2  2  327  GLOSSARY OF TERMS The following abbreviations, most of which are commonly found in the literature, used in this thesis.  A  angstrom (10" m)  Anal.  analysis  atm  atmosphere  HBfiH}  observe boron-11 while decoupling proton  Beq  equivalent isotropic thermal parameter  B.M.  Bohr Magneton  br  broad  Bu  10  l  tertiary butyl group, -C(CH3)3  Bz  benzyl group, -CH2PI1  "CflH}  observe carbon-13 while decoupling proton  3Cd{lH}  observe cadmium-113 while decoupling proton  n  °C  degrees Celcius  cal  calories  calc  calculated  Chem 3D  molecular modelling program for the Macintosh computer  cm  wave number  -1  COSY  Correlated SpectroscopY  Cp  cyclopentadienyl group, {C5H5}-  Cp*  pentamethylcyclopentadienyl group, {Cs(CH3)5}-  d  doublet  D  deuterium  dd  doublet of doublets  d  n deuterated  n  xxiii  d of sept  doublet of septets  deg  degrees  DME  1,2-dimethoxyethane  E  activation energy  a  EI  electron ionization  equiv  equivalent(s)  ESR  Electron Spin Resonance  Et  ethyl group, -CH2CH3  eV  electron volt  g  gram  AG°  Gibb's standard free energy of reaction  AG*  Gibb's energy of activation  GC-MS  Gas Chromatography/Mass Spectrometry  h  hour  *H  proton  AH°  enthalpy of reaction  AH*  enthalpy of activation  HOMO  Highest Occupied Molecular Orbital  Hz  Hertz, second"  I  spin  INDO  Intermediate Neglect of Differential Overlap  IR  infrared  n  /A-B  1  rc-bond  scalar coupling constant between nuclei A and B  K  degrees Kelvin, or equilibrium constant  ki  forward rate constant  k_i  reverse rate constant  kobs  observed rate constant xxiv  kcal  kilocalories  kHz  kiloHertz  6  L i { H} X  observe lithium-6 while decoupling proton  7  Li{ *H}  observe lithium-7 while decoupling proton  In  natural log  Ln  lanthanide  LUMO  Lowest Unoccupied Molecular Orbital  m  multiplet, or medium for infrared data  m  meta  M  molar, or central metal atom  M  parent ion  +  MAS  Magic Angle Spinning  Me  methyl group, -CH3  mg  milligram  MHz  megaHertz  mL  millilitre  mmol  millimole  MO  Molecular Orbital  mol  mole  ms  millisecond  MS  Mass Spectrometry  15  N{ H} 1  observe nitrogen-15 while decoupling proton  NOESY  Nuclear Overhauser Enhancement SpectroscopY  NMR  Nuclear Magnetic Resonance  o  ortho  *P{*H}  observe phosphorus-31 while decoupling proton  3  p  para xxv  Ph  phenyl group, -C6H5  ppm  parts per million  n-Pr  propyl group, - C H 2 C H 2 C H 3  Pr  isopropyl group, -CH(CH3)2  q  quartet  R  hydrocarbyl substituent  r  correlation  1  29  Si{ !H}  observe silicon-29 while decoupling proton  45  Sc{ ^H}  observe scandium-45 while decoupling proton  s  singlet, or seconds, or strong for infrared data  AS°  entropy of reaction  AS* entropy of activation sext  sextet  SN2  substitution, nucleophilic, 2nd order  t  triplet  T  temperature  THF  tetrahydrofuran  VT  Variable Temperature  X  chloride or alkyl group  ZENDO  Zerner Intermediate Neglect of Differential Overlap  ot  on a carbon that is directly bonded to a metal  P  on a carbon that is two bonds away from a metal  8 A A1/2 Tl  n  M-calc  chemical shift heat ' peak width at half height n-hapto calculated magnetic moment xxvi  effective magnetic moment X-bridging ligand vibrational band for bond x-  xxvii  ACKNOWLEDGMENTS I would like to express my thanks to Mike Fryzuk for his guidance throughout the course of this work. Mike's enthusiasm for chemistry is matched only by his patience and humour. I would also like to thank all the past and present members of the Fryzuk group, for their friendship and advice, and for making the lab an enjoyable place to work. I only wish I had time to play practical jokes on all of you. I am indebted to Steve Rettig for solving the crystal structures in this thesis. Without his help, much of this work would have been merely speculative. My thanks also to Glenn Yap from the University of Windsor who solved three of the structures. I greatly appreciate the help of the NMR staff, particularly Liane Darge, Marietta Austria, Nick Burlinson and Orson Chan, who cheerfully assisted me whenever I had questions or concerns. I would also like to mention the support staff here at UBC (Peter Borda for the microanalyses, Marshall Lapawa for the mass spectrometry) for their prompt and accurate service. Special mention must also be made of Elizabeth Varty in illustration for all her help with the illustrations in this thesis.  xxviii  Chapter 1  Introduction to Amidophosphine Complexes of Electron-Poor Metals  1.1 Introduction The modern study of coordination compounds began at the turn of the century with Alfred Werner, whose investigations of octahedral Co(III) compounds of the type [Co (NH3)x] " showed that a neutral molecule such as ammonia could bind to a metal and act as m  3  a ligand.  Such coordination compounds signalled the emergence of an important field of  1  inorganic chemistry, and prompted the study of other simple a-donors, such as H2O, as ligands. Since Werner's original discovery, the field of coordination compounds has experienced continual growth, encompassing both neutral and anionic ligands, as well as a- and rc-donors.  2  Currently, ligand design is subject to an ever-growing complexity, to suit the specific needs of the metal studied.  3  Organometallic chemistry is concerned with those coordination compounds which contain a metal-carbon bond. The stability and reactivity of organometallic complexes are a function of the metal's position in the transition series and the nature of the ligands present. While the former provides chemical diversity, it is the latter feature that allows the reactivity of a particular metal complex to be controlled. Such complexes can be described by the general 2  formula L M X , where Ln represent ancillary ligands and X n  m  m  are defined as reactive ligands.  Ancillary ligands provide the metal centre with an appropriate steric and electronic environment to facilitate reactivity, but in general remain inert to a particular transformation of interest. In  1  References  begin  on page  21  Chapter  1: Introduction  to Amidophosphine  Complexes...  contrast, reactive ligands are those groups which undergo changes. Some typical examples of reactive ligands would be hydrocarbyl groups, which could undergo insertion with CO, or hydrides that would undergo olefin insertion. Thus, it is often found that changes in the ancillary ligand can result in different reactivity patterns of the reactive ligands.  4  For example, the  influence that the electronic and steric properties of an ancillary ligand can have on the reactivity of a metal-carbon bond is illustrated by the following series (Scheme 1.1). Scheme l . l  5  R = CH(SiMe )  3 2  2  References  begin  on page  21  Chapter 1: Introduction to Amidophosphine Complexes... In the above examples, the rate of hydrogenolysis of the same yttrium alkyl group is very dependent  on  the  choice  of  the  ancillary  ligand.  While  the  bis(pentamethylcyclopentadienyl)yttrium complex rapidly reacts with hydrogen to form the corresponding yttrium hydride, more forcing conditions are necessary when the metal is coordinated by two amidinate ligands. In contrast, the porphyrin yttrium system does not undergo hydrogenolysis. This series emphasizes the importance of the ancillary ligands in controlling the reactivity of the metal-carbon bond. The complexation of metals by ancillary ligands, and their effect upon the chemistry of the reactive ligands is the subject of much of modern organometallic and coordination chemistry.  1.2 Organometallic Complexes of Scandium The organometallic chemistry of the elements of the first transition series has been extensively investigated during the past four decades. ' ' " 1 2 6  virtually ignored.  10  Scandium, however, has been  9  In 1956, Birmingham and Wilkinson reported that "It seems to be fairly  certain that alkyl and aryl derivatives of scandium, yttrium, lanthanum and the rare earth elements either do not exist or have an existence so transitory that they cannot be isolated".  11  Wilkinson showed that tris(cyclopentadienyl)scandium could be obtained as a stable species;  11  later work proved it to have an unusual polymeric structure. The first homoleptic alkyls and 12  aryls were also polymeric materials, and as a result were intractable and thus inert to chemical transformation.  13  The high price of scandium did little to encourage the development of organoscandium chemistry. Although scandium's presence in the earth's crust is fifty times that of silver, it is only thinly distributed in ores, predominantly throughout Norway, in the form of the mineral thortvetite (Sc2Si207).  14  Today, the major source of scandium is the processing of uranium ores,  3  References begin on page 21  Chapter  1: Introduction  to Amidophosphine  Complexes...  which generates -0.02% S C 2 O 3 . It was not until 1960 that the first pound of 99% pure metal was produced. By the mid 1970's, only 12 compounds with scandium-carbon bonds were known. 14  15  Although scandium is a member of Group 3, its small size and Lewis acidity more closely resembles that of its main group congener, aluminum.  16  Early work by Lappert  17  determined that the scandium alkyl (r| -C5H5)2ScMe was more Lewis acidic than the yttrium or 5  lanthanum analogues, or even trimethylaluminum. Organoscandium and organoaluminum compounds exhibit similar behaviors, particularly the tendency to form dimers with 3-centre, 2electron (M-(|i-R)-M) bridges.  17,18  Scandium complexes have many properties in common with  those of aluminum as well as the lanthanides. This hybrid character of scandium sets it apart from the other members of the transition series. Aside from the research group of Lappert, " Bercaw et al. have concentrated on 17  19  devising preparative routes to organometallic compounds of scandium and exploring their structures. ' ' ' 10  15  20  27  The utilization of two bulky pentamethylcyclopentadienyl groups results in  monomelic, salt- and solvent-free scandium alkyls A . with C-H bonds (a-bond metathesis)  10,15,27  1 0 , 2 7  These compounds partake in reactions  as well as olefins, nitriles  20,26  and C O  1 5  (insertion  reactions). In addition, these complexes are active ethylene polymerization catalysts.  21  A  B  4  References  begin  on page  21  Chapter 1: Introduction to Amidophosphine Complexes.. Although ethylene polymerization is well defined for these systems, the inability to insert oc-olefins such as propylene led to the investigation of less sterically encumbered organoscandium derivatives such as B, in which the pentamethylcyclopentadienyl rings are tethered back by a SiMe2 unit;  22,23  this metallocene slowly oligomerizes propylene, while the  cyclopentadienyl-amido system C is an active polymerization catalyst for 1-butene and 1pentene.  24,25  Titanium, zirconium and hafnium complexes incorporating the ligand system in C  are used industrially for the copolymerization of ethylene and cc-olefins. " 28  30  The study of these scandium systems has lent much to our understanding of the cj-bond metathesis reaction. The activation of carbon-hydrogen bonds has been a challenge that has been actively pursued in recent years. " 31  34  C-H activation by electron-rich, late metal complexes is  accepted to proceed via an oxidative addition/reductive elimination sequence.  However, the  2  resistance of Sc(III) to reduction prohibits this pathway from being operative. Instead, C-H activation involving d° (or d ? ) metals, referred to as a-bond metathesis, is proposed to occur 0  1  via a four-centered transition state without any formal change in oxidation state (Scheme 1 2  ).35.36  Scheme 1.2  it  'R' C p * S c R + R'H 2  / Cp* Sc5 \ 2  \ 8 H 8" /  5 +  +%  C p * S c R ' + RH 2  The activation of stable, saturated molecules such as SiMe4 or CH4 is a reflection of the favorable orientation of the frontier orbitals of the Cp* Ln fragment, which are ideally suited to 2  support the transition state (see Section 2.6).  27  5  References begin on page 21  Chapter  1: Introduction  to Amidophosphine  Complexes...  1.3 Organometallic Complexes of Aluminum In contrast to the relative infancy of organoscandium chemistry, organometallic compounds of aluminum have been known for more than a century. ' 6  16,37  However, they  attracted little attention until the work of Ziegler in the 1950's. Ziegler's discovery of the direct synthesis of trialkylaluminum compounds from aluminum metal, olefins and hydrogen revolutionized the petrochemical industry.  38  Fifteen years later, organoaluminum compounds  were available in 100,000 lb tank-car loads at prices of about $l/kg.  37  Aluminum is the most common metallic element in the earth's crust (8.8%) and occurs mainly in silicates and the ore bauxite, a hydrated aluminum oxide. Industrially it is prepared via the electrolysis of its molten salts, known as the Hall process.  16  Due to aluminum's strong Lewis  acidity, the organometallic chemistry of aluminum in many ways resembles the chemistry of the high valent, electrophilic early transition metals.  39  It is most often compared to scandium,  mainly due to the stability of the M(III) state. Perhaps the most influential of all reactions with organoaluminum compounds are carbaluminations. Ziegler discovered that the addition of olefins to alkylaluminum compounds resulted in the formation of carbon-carbon bonds (equation l . l ) .  R AI— R 2  +  H C=CH 2  •  2  40  R AI—CH — C H — R 2  2  2  [i.i]  Repeated addition of ethylene to the Al-C bond of a trialkylaluminum followed by olefin displacement affords a valuable route to oligomers of moderate length (Cio - C2o)- This process is also used to produce unbranched primary alcohols by oxidation and subsequent hydrolysis of  6  References  begin  on page  21  Chapter  the oligomer,  1: Introduction  to Amidophosphine  Complexes...  whose sulfates are produced in bulk quantities to serve as biodegradable  41  surfactants ROSO3H in the detergent industry.  6,9  The polymerization of olefins by trialkylaluminum species has two drawbacks: 1) the number of insertions before chain displacement becomes dominant is small and 2) larger aolefins undergo only single insertions. Although applications for these insertions have been found (n-Pr2AlH catalyzes the dimerization of propene, which is an important technical process in the production of synthetic rubber), ' Ziegler discovered that combining alkylaluminum 6 9  compounds with complexes of the early transition metals resulted in longer chain lengths. The most effective catalyst resulted from the combination of TiCU and AlEt3; under ambient pressure and temperature, polyethylene consisting of 10 to 10 monomer units could be produced. 3  4  42  The application of Ziegler-type catalysts to the polymerization of propylene was the contribution of Natta, who discovered that closely related systems could polymerize propylene to a molecular weight of 10 - 10 . 5  6  43  Natta also established a link between the stereochemical  structure and bulk properties of the polymer.  44,45  For their contributions to polymer chemistry,  Ziegler and Natta were awarded the Nobel prize in 1963. In the last two decades, olefin polymerization has been dominated by metallocene-based systems. The discovery that trace amounts of water actually led to higher activities resulted in the use of methylalumoxane  46  Kaminsky  as a cocatalyst for the polymerization of oc-olefins by  and Brintzinger-type catalysts  47,48  4 9  Methylalumoxane is valuable in its role as a  water scavenger, an alkylation reagent and a halide abstractor to generate active catalysts of the type Cp2MMe . +  50  Currently, the US chemical industry produces over 15 billion kg of polyolefin  annually.  9  7  References  begin  on page  21  Chapter  1: Introduction  to Amidophosphine  Complexes..  1.4 Mono- and Non-Cyclopentadienyl Derivatives of the Group 3 Metals and the Lanthanides The recent growth in organolanthanide chemistry has produced a wealth of new reactions including ethylene ' 21  27,51  (3-methyl elimination.  36  "  and propylene polymerization  54  5 2  methane transmetallation  27,36  and  Furthermore, lanthanide hydrides are reported to be excellent catalysts  for olefin hydrogenation ' and ethylene polymerization. 53 55  56  This chemistry has been limited  mainly to complexes that contain the bis(pentamethylcyclopentadienyl) set as ancillary ligands, and although these systems partake in a number of facile reactions, the presence of two bulky Cp* ligands, as well as a single reactive lanthanide-carbon bond may limit their usefulness. Unlike compounds of the type Cp* LnR, the number of mono-ligand lanthanide complexes of 2  the formula Cp*LnR is limited. " The presence of a single Cp* ligand results in greater steric 57  66  2  and electronic unsaturation. As a result, mono(ligand) compounds are often undesirably complexed by salt and solvent molecules, which significantly diminish their reactivity. In addition, the syntheses of such species are often accompanied by the bis(ligand) and unsubstituted products. Traditionally, Group 3 and lanthanide organometallic chemistry has focused on cyclopentadienyl ligands and their derivatives. Approximately 90% of all previously described organometallic complexes of the Group 3 and lanthanide elements are compounds containing cyclopentadienyl ligands;  67,68  these anionic Tt-ligands are ideally suited for stabilizing highly  reactive lanthanide-alkyl and -hydride species. However, if one can impart the same electronic environment while decreasing the steric bulk around the metal centre, a much more reactive complex can be obtained.  15  Thus, anionic o-ligands such as amides, bulky alkyl groups, and  anionic heterocycles have recently received more attention.  3  The use of nitrogen-based donors  as supporting ligands, such as porphyrins or amidinates has been the subject of much work. " 69  72  Conversely, the use of phosphorus-based ligands has been more limited, even though the high Lewis basicity of these donors could be valuable in stabilizing complexes of electrophilic metals.  8  References  begin  on page  21  Chapter 1: Introduction to Amidophosphine Complexes... If the reactivity of a metal-carbon bond is dependent upon the ancillary ligands present, new ligand designs could result in different patterns of reactivity.  1.5 Phosphine Complexes of the Group 3 Metals Although the phosphine ligand is prevalent in transition metal coordination and inorganic chemistry, there are still areas of the periodic table for which metal-phosphine complexes are rare. Phosphine complexes of the lanthanides and the early transition metals, in which the metal 2  is usually in a high oxidation state and resistant to reduction, are such examples. recently have these species become isolable.  73,74  Only  Prior to 1986 there were no structurally  characterized examples of a Group 3 phosphine compound and today, only a handful of these complexes are known (Table 1.1). Most scandium phosphine derivatives are stabilized by the small monodentate PMe3 ligand.  In general, these ligands undergo rapid dissociation in solution, generating a  coordinatively and electronically unsaturated metal centre. Cp*SiMe2[CpCH2CH2PBu 2], with a t  tertiary phosphine linked to a cyclopentadienyl group, was prepared to examine the reactivity of a scandium alkyl with a high local phosphine concentration, due to the chelate effect. 84  Although {(Ti5-C Me )SiMe2[Ti -C5H3CH2CH2P(Bu )2]}ScCH(SiMe3)2 does not contain a t  5  5  4  scandium-phosphine bond, the scandium hydride, produced upon exposing the scandium alkyl to hydrogen, shows spectroscopic evidence for such coordination (equation 1.2).  85  9  References begin on page 21  Chapter  1: Introduction  to Amidophosphine  Complexes...  Table 1.1 Structurally Characterized Phosphine Complexes of the Group 3 Metals.  Complex [ {Cp*SiMe2NBu } Sc(PMe )] (|i-H) t  3  2  24 2  {Cp* SiMe } ScH(PMe3)  73  2  2  [ {Cp*SiMe NBut} Sc(PMe )] (|i,ri ,Ti 2  2  C H ) 2  3  2  2  Year  2.996(1)  1990  2.752(1)  1990  2.825(3)  1994  2.752(2)  1994  2.617(1)  1994  3.045(2)  1988  2.817(3) - 3.005(3)  1991  2.892(1), 2.931(1)  1992  2.940(1) - 3.035(1)  1986  3.165(6)  1992  3.127(2)-3.216(2)  1993  3.141(4)  1993  3.038(1) - 3.233(1)  1994  2.857(1), 2.861(1)  1996  2 5  4  [ {mt?w-SiMe [Bu Cp] } Sc(PMe )] Te t  2  2  Sc[(Me P) CSiMe ] 2  m  M-P Bond Length (A)  2  3  3  75  2  76 3  [Y(OCBu«2CH PMe ) ] 2  2  77  3  ' Y[N(SiMe CHPMe )(SiMe2CH PMe )]2  2  2  [N(SiMe CH PMe ) ] 2  2  2  2  74  2  {Y(ri3-C H )[N(SiMe CH PMe ) ]} (n3  Cl)  5  2  2  2  2  2  78 2  [La{(Ph P) CH} ] 2  2  79  3  [La{N(SiMe ) } (PPh )(Ph PO) ] 3  2  2  3  La[TeSi(SiMe ) ] (dmpe) 3  3  3  80  2  81 2  [(Pri N) La{^-P(C6H OMe-o) ] Li(thf)] 2  2  4  La[(Me P) CSiMe ] 2  2  3  2  82  2  76 3  (But P) La[(^i-PBu'2) Li(thf)]  83  2  2  2  10  References  begin  on page  21  Chapter  1: Introduction  to Amidophosphine  Complexes...  1.6 Phosphine Complexes of the Group 13 Metals The chemistry of organoaluminum compounds is dominated by their extreme Lewis acidity, and the tendency for aluminum to build up an octet of electrons.  37  This results in 1:1  adducts of trialkylaluminums with neutral bases such as ethers or amines. Phosphine complexes of aluminum are also numerous. For example, only N M e 3 binds more strongly to A l M e 3 than PMe3.  16  Organometallic complexes of aluminum are stronger Lewis acids than organo-  derivatives of boron, gallium or indium, and less sterically hindered than those of boron. The reactions of AIR3 with tertiary amines usually terminate with adduct formation; in the case of secondary and primary amines, elimination of alkane can be induced (Scheme 1.3).  6  Scheme 1.3  NMe ^ — • Me AI-NMe 3  3  Me AI 3  N M e  2  3  > Me AI-NMe H — - ^ - ^ - • 1 / 2 (Me2AINMe )2  H  3  NMeHp  2  M^AI-NMeH  2  2  -  2  C  H  % y  7  ( AINMe) Me  7  This reaction sequence is general to a number of Lewis bases, with alcohols giving organoaluminum alkoxides, thiols generating thiolates and phosphines giving phosphides. Both neutral and anionic phosphorus-based ligands are common for aluminum, although less so for the larger members of Group 13. Gallium and indium are less Lewis acidic, and not as likely to form adducts with Lewis bases. Unlike AIR3 which is dimeric in solution through electron deficient bridges, GaR3 and InR3 are monomeric in solution, although alkyl scrambling in MRR'R" (M = Ga, In) is observed upon heating.  6  11  References  begin  on page  21  Chapter  1: Introduction  to Amidophosphine  Complexes...  1.7 Summary of Previous Work from our Laboratory 1.7.1 First Generation Ligand Design Research in our laboratories has centered on the development of hybrid ligands of the general type I, which incorporate both an amide, NR2", and phosphines, PR3, in a chelating array. '  4 86  These donor types were chosen for a number of reasons: although amide ligands were  common amongst early metals, their utility was less noted for the later members of the transition series.  87  In contrast, phosphine complexes of the late metals were prevalent, although those of  the early transition metals were more elusive.  88  More specifically, phosphine complexes of the  group 3 metals and those of the lanthanides were virtually unknown. This can be ascribed to the hard Lewis acidic nature of the early metals and the soft Lewis basic nature of the phosphine donor, resulting in a mismatch of donor and acceptor.  1,89  In an effort to overcome mismatches in coordination, a tridentate hybrid ligand system was developed. It was anticipated that the hard amide donor NR2" would serve to anchor the ligand to early metals, inducing chelation of the pendant phosphine donors by virtue of their proximity; alternately amides of the later transition elements would be accessible since the phosphine interaction would help stabilize the amide ligation.  12  References  begin  on page  21  Chapter  1: Introduction  to Amidophosphine  Complexes..  This ligand system is readily synthesized from commercially available starting materials. Reaction of l,3-bis(chloromethyl)-l,l,3,3-tetramethyldisilazane with 3 equiv LiPPr^ displaces the chlorides and deprotonates the amine to yield the lithium salt of the ligand. This lithium salt is a convenient starting material for metathesis reactions with metal halide salts (Scheme 1.4).  90  As a formal six electron donor, the ligand system (referred to as PNP) is comparable to the Cp ligand. The tridentate PNP ligand most often exhibits meridional coordination, although -  facial coordination is occasionally observed.  91,92  The steric and electronic properties of the  ligand can be further tuned by the appropriate choice of the substituents on the phosphine, from the small, electron-donating methyl group to the larger, electron-withdrawing phenyl substituent. In this way, the basicity of the phosphine, and the resultant electron density at the metal centre, can be controlled.  86,93  13  References  begin  on page  21  Chapter  1: Introduction  to Amidophosphine  Complexes..  The popularity of phosphine ligands in organometallic chemistry is due in part to the NMR activity of the phosphorus nucleus. Both the abundance (100%) and the sensitivity (6.6% that of H) of the phosphorus-31 nucleus (I = 1/2) allow for a sensitive and convenient handle for 1  characterization. The chemical shift window (>600 ppm wide) and 94  1  _  4  /HP  and  ^ P M  coupling  constants are also useful probes. " 95  97  1.7.2 Second Generation Ligand Design In the course of our investigations of metal complexes utilizing the PNP ligand system, certain limitations became evident. Two of the main drawbacks to this type of ligand system are the hydrolytic instability of the NSi2 unit in the ligand backbone and the basicity of the amide linkage. In an effort to overcome these two limitations, our laboratories developed a second generation ligand design in which the formally negatively-charged, basic amide donor has been replaced by a cyclopentadienyl unit as in II. 15)  99  98  The lower basicity of cyclopentadiene (pA^ =  as compared to a disilylamine (pAT = 25.8) a  a  100  solves the basicity issue. As this second  generation ligand incorporates carbon-silicon linkages in the backbone, the problem of hydrolytic instability of the ligand backbone is obviated.  14  References  begin  on page  21  Chapter  1: Introduction  to Amidophosphine  Complexes..  A number of interesting transformations have been noted with the [(T| -C5H.3-1,35  (SiMe2CH2PPrJ2)2] [P2Cp] ligand system, including the stabilization of the first zirconium alkylidene via a-hydrogen abstraction of a Zr(IV) dibenzyl species.  101  What emerges from study  of these and related zirconium species is that coordination of the pendant phosphine arms becomes less favorable as the number and size of alkyl substituents on the metal increases. Consequently, fluxional processes such as phosphine dissociation become prevalent. One solution to this problem is to tether the two phosphines together, reducing the tendency for dissociation. Such a macrocyclic ligand is described below.  1.7.3 Third Generation Ligand Design While nitrogen-based macrocycles such as porphyrins and phthalocyanines are well established in the literature, those containing phosphorus are much less common. syntheses of macrocycles have employed high dilution techniques,  103  102  The  and more recently, metal  template reactions have been utilized in an effort to control stereochemistry.  104  Recent work in  our laboratory has resulted in the preparation of a new macrocyclic bis (amidophosphine) ligand system based on the original [N(SiMe2CH2PR2)2l framework. In this system, the "open end" has been complexed with another disilylamido unit to generate a macrocyclic ligand with mutually trans disposed amido and phosphine groups. The synthesis of this ligand yields two isomers, one in which the phenyl rings on the phosphines are oriented syn (or cis) across the macrocycle (III), while these substituents are disposed in an anti (or trans) relationship in the other (IV).  105  15  References  begin  on page  21  Chapter  1: Introduction  to Amidophosphine  Complexes..  Scheme 1.5  Me? Si  Me ,Si.  2  Cl  ci  H  Me Me ,SL .. ^Si. 2  2 LiPHPh THF O'C (> 90%)  2  I  H  Ph-P  I  P-Ph I H  H  T  v  Si Me  Si Me  9  Me. .Si  T |  2  4 BuLi THF 0°C  Me? ,Si  SL  Ph<  s = THF Me2  S>V Me  Me2  syn-Li (THF)[P N2] m 2  Me? .Si  Me2  2  Me  2  2  anf/-Li (THF) [P N ] IV 2  2  2  2  In a similar manner to the synthesis of HN(SiMe2CH2PPh2)2, the reaction between 2 equiv L i P H P h and H N ( S i M e 2 C H 2 C I ) 2 yields the secondary  diphosphinodisilazane  HN(SiMe2CH2PHPh)2. The slow addition of B u L i (4 equiv) to an equimolar mixture of this diphosphine and the starting dichlorodisilazane forms the dilithio salt of the macrocycle  16  References  begin  on page  21  Chapter  1: Introduction  to Amidophosphine  Complexes...  PhP(CH2SiMe2NSiMe2CH2)2PPh (abbreviated as [P2N2]) as two isomers (Scheme 1.5). These two isomers are easily separated after fractional crystallization as colorless, crystalline solids.  1.8 Group 3 Complexes of the PNP and P2N2 Ligand Systems Previous work has studied the complexation of the tridentate ligand system [N(SiMe2CH2PR2)2]" with the heavier members of Group 3, Y and L a . 7 4  complexes, YCl[N(SiMe2CH2PMe2)2]2, are analogous to Cp*2YR, transformations have been reported.  56,108 1 1 6  1 0 6 , 1 0 7  The bis(ligand)  for which many  Using the PNP ligand system, thermally unstable  hydrocarbyl complexes YR[N(SiMe2CH2PMe2) ]2 were produced which underwent first order 2  intramolecular elimination of RH to yield a cyclometallated product (Scheme 1.6). '  4 74  Scheme 1.6  Previously, we also investigated the reaction chemistry of the mono(ligand) complex, YCl2[N(SiMe2CH2PPr 2)2]. i  While monomelic, salt- and base-free metal complexes were  accessible, the conversion to dialkyl or diaryl derivatives was problematic. The addition of 2 equiv PhLi to the starting dichloride YCl [N(SiMe2CH2PR2)2] proceeded with the metathesis of 2  a single chloride to generate the monoaryl complex; this compound underwent intermolecular  17  References  begin  on page  21  Chapter 1: Introduction to Amidophosphine Complexes... elimination of benzene to produce a dimeric yttrium species (Scheme 1.7). The addition of typical alkylating reagents to the yttrium dichloride resulted in intractable materials.  117  Scheme 1.7  In contrast to the aforementioned yttrium complexes, {.ryn-[p2n2]Y}2(u.-Cl)2 is dimeric in the solid state, but alkylation with large alkyl groups affords well-defined monomelic species (Scheme 1.8). Yttrium alkyls of the general formula ,syrc-YR[p2N2] (R = Ph, CH2SiMe3) do not undergo intramolecular elimination of alkane, but have been shown to be involved in C-C bond formation processes.  118  Scheme 1.8  (silyl methyl groups omitted for clarity)  18  References begin on page 21  Chapter  1: Introduction  to Amidophosphine  Complexes...  This series of yttrium complexes (mono-[PNP], bis-[PNP], [P2N2]) illustrates that the electronic and steric properties imparted to the metal by the ligand system are important aspects of ligand design and the resultant reactivity of the metal-carbon bonds.  1.9 The Scope of this Thesis In light of the transformations noted for yttrium mono-[PNP] complexes, and the paucity of scandium phosphines in the literature, we embarked upon a study of the'coordination and reactivity properties of the complexes ScR2[N(SiMe2CH2PPr )2] (X = halide, alkyl). The i  2  majority of previously reported scandium phosphine complexes involve coordination to PMe3; the study of compounds with pendant donor groups is rare. Our focus was to examine the effect of this ligation upon the reactivity of the scandium-carbon bonds.  M = Al, Sc  Chapter 2 presents the synthesis of 12-electron scandium dialkyl complexes and an examination of their reactivity patterns with small molecules. We have also replaced an alkyl group with a cyclopentadienyl ring to stabilize these complexes by increasing the steric bulk around the metal and the number of available electrons.  19  Chapter 2 concludes with some  References  begin  on page  21  Chapter  1: Introduction  to Amidophosphine  Complexes...  preliminary studies of the scandium [P Cp] system and the effect of the substitution of Cp~ for 2  NR ". 2  We also studied the analogous aluminum species since the synthesis and reactivity of these congeners provides a direct comparison to the scandium system. Chapter 3 addresses phosphine dissociation in these complexes and their nature in solution; the potential for developing cationic complexes of aluminum was also explored. The chemistry in Chapter 4 grew out of serendipitous results obtained regarding the lithium salt of the PNP ligand system; the complexation of the lithiated ligand with various salts allowed for an indirect study of the solid state structure of the PNP ligand system. As for the aluminum complexes studied in Chapter 2, the structure of these species in solution was of interest. Chapter 5 is a continuation of Chapter 3. By utilizing the macrocyclic third generation ligand system [P N ], the effect of incorporating phosphine donors into a macrocyclic array was 2  studied.  2  The coordination of both anti- and syn- isomers of the macrocycle, and their  interconversion, is presented. In addition, the organometallic chemistry of aluminum, gallium and indium was investigated to see if the bulky macrocyclic ligand [P N ] could induce new 2  2  coordination compounds of these Lewis acidic metals.  Me Si  2  Me  Si  2  P—Ph  Ph  M = Al.-Ga, In  20  References  begin  on page  21  Chapter 1: Introduction to Amidophosphine Complexes... Finally, Chapter 6 advances some second thoughts, and includes some results which point toward future avenues of study. Included are some preliminary results regarding paramagnetic complexes of Ti(UI) and V(IU), and the isolation of a dinuclear vanadium dinitrogen complex.  1.10 References (1)  Huheey, J. E. Inorganic Chemistry; Principles of Structure and Reactivity; 3rd ed.;  Harper & Row: New York, 1983. (2)  Collman, J. P.; Hegedus, L. S.; Norton, J. R.; Finke, R. G. Principles and Applications of Organotransition Metal Chemistry; University Science Books: Mill Valley, 1987.  (3)  For a review of non-cyclopentadienyl ligands in lanthanide chemistry, see: Edelmann, F. T. Angew. Chem., Int. Ed. Engl. 1995,34, 2466., and references therein.  (4)  Fryzuk, M. D.; Haddad, T. S.; Berg, D. J.; Rettig, S. J. Pur. App. Chem. 1991, 63, 845.  (5)  Duchateau, R.; Wee, C. T. v.; Meetsma, A.; Duijnen, P. T. v.; Teuben, J. H. Organometallics 1996,15,2279.  (6)  Elschenbroich, C ; Salzer, A. Organometallics; 2nd ed.; VCH Publications: New York, 1992.  (7)  Comprehensive Organometallic Chemistry; Wilkinson, G.; Stone, F. G. A.; Abel, E. W., Ed.; Pergamon Press: Oxford, 1982.  (8)  Comprehensive Organometallic Chemistry II; Wilkinson, G.; Stone, F. G. A.; Abel, E. W., Ed.; Pergamon Press: Oxford, 1995.  (9)  Crabtree, R. H. The Organometallic Chemistry of the Transition Metals; 2nd ed.; John  Wiley & Sons: USA, 1994.  21  References begin on page 21  Chapter  1: Introduction  to Amidophosphine  Complexes...  (10)  Thompson, M. E.; Bercaw, J. E. Pur. App. Chem. 1984,56, 1.  (11)  Wilkinson, G.; Birmingham, J. M. J. Am. Chem. Soc. 1956, 78, 42.  (12)  Atwood, J. L.; Smithe, K. D. / . Am. Chem. Soc. 1973, 95, 1488.  (13)  Hart, F. A.; Massey, A. G.; Saran, M. S. / . Organomet. Chem. 1970,21, 147.  (14)  Greenwood, N. N.; Earnshaw, A. Chemistry of the Elements; 1st ed.; Pergamon Press: Oxford, 1984.  (15)  Piers, W. E.; Shapiro, P. J.; Bunel, E. E.; Bercaw, J. E. Synlett. 1990, 74.  (16)  Cotton, F. A.; Wilkinson, G. Advanced Inorganic Chemistry; 4th ed.; John Wiley & Sons: New York, 1980.  (17)  Lappert, M. F.; Holton, J.; Ballard, D. G. H.; Pearce, R.; Atwood, J. L.; Hunter, W. E. /. Chem. Soc, Dalton Trans. 1979, 45.  (18)  Lappert, M . F.; Singh, A.; Atwood, J. L.; Hunter, W. E. J. Chem. Soc, Chem. Commun. 1983, 206.  (19)  Lappert, M . F.; Singh, A.; Atwood, J. L. / . Chem. Soc, Chem. Commun. 1981, 1190.  (20)  Bercaw, J. E.; Davies, D. L.; Wolczanski, P. T. Organometallics 1986, J, 443.  (21)  Burger, B. J.; Thompson, M . E.; Cotter, W. D.; Bercaw, J. E. / . Am. Chem. Soc. 1990, 112, 1566.  (22)  Hajela, S.; Bercaw, J. E. Organometallics 1994,13, 1147.  (23)  Piers, W. E.; Bercaw, J. E. / . Am. Chem. Soc. 1990,112, 9406.  (24)  Shapiro, P. J.; Bunel, E.; Schaefer, W. P.; Bercaw, J. E. Organometallics 1990,9, 867.  22  References  begin  on page  21  Chapter 1: Introduction to Amidophosphine Complexes... (25)  Shapiro, P. J.; Cotter, W. D.; Shaefer, W. P.; Labinger, J. A.; Bercaw, J. E. / . Am. Chem. Soc. 1994,116, 4623.  (26)  St. Clair, M. A.; Santarsiero, B. D.; Bercaw, J. E. Organometallics 1989, 8, 17.  (27)  Thompson, M. E.; Baxter, S. M.; Bulls, A. R.; Burger, B. J.; Nolan, M. C ; Santarsiero, B. D.; Schaefer, W. P.; Bercaw, J. E. / . Am. Chem. Soc. 1987,109, 203.  (28)  Canich, J. A. U. S. Patent 5,026,798,1991 (Exxon).  (29)  Canich, J. A. PCT Application WO 91/04257, April 4, 1991 (Exxon).  (30)  Stevens, J. C ; Timmers, F. J.; Wilson, D. R.; Schmidt, G. F.; Nickias, P. N.; Rosen, R. K.; Knight, G. W.; Lai, S. Y. European Patent Application EP-416-815-A2, March 13, 1991 (Dow).  (31)  Green, M. L. H.; O'Hare, D. Pur. App. Chem. 1985,57, 1897.  (32)  Crabtree, R. H. Chem. Rev. 1985, 85, 245.  (33)  Rothwell, I. P. Polyhedron 1985,4, 111.  (34)  Halpern, J. Inorg. Chim. Acta. 1985,100,41.  (35)  Simpson, S. J.; Turner, H. W.; Andersen, R. A. / . Am. Chem. Soc. 1979,101, 7729.  (36)  Watson, P. L. / . Am. Chem. Soc. 1983,105, 6491.  (37)  Mole, T.; Jeffrey, E. A. Organoaluminum Compounds; Elsevier: Amsterdam, 1972.  (38)  Ziegler, K.; Gellert, H. G.; Zosel, K.; Lehmkuhl, W.; Pfohl, W. Angew. Chem. 1955, 67, 424.  (39)  Shapiro, P. J.; Vij, A.; Yap, G. P. A.; Rheingold, A. L. Polyhedron 1995,14, 203.  23  References begin on page 21  Chapter 1: Introduction to Amidophosphine Complexes... (40)  Ziegler, K.; Gellert, H. G. Angew. Chem. 1952, 64, 323.  (41)  Ziegler, K.; Krupp, F.; Zosel, K. Angew. Chem. 1955, 67, 425.  (42)  Ziegler, K.; Breil, H.; Holzkamp, E.; Martin, H. German Patent 973,626, 1953.  (43)  Natta, G.; Pino, P.; Mazzanti, G. Italian Patent 535,712, 1954.  (44)  Natta, G.; Pino, P.; Corradini, P.; Danusso, F.; Mantica, E.; Mazzanti, G.; Moraglio, G. J. Am. Chem. Soc. 1955, 77, 1708.  (45)  Natta, G. / . Am. Chem. Soc. 1955, 77, 1708.  (46)  Sinn, H.; Kaminsky, W.; Vollmer, H. J.; Woldt, R. Angew. Chem., Int. Ed. Engl. 1980, 19, 390.  (47)  Sinn, H.; Kaminsky, W.; Vollmer, H. J.; Woldt, R. Angew. Chem. 1980, 92, 396.  (48)  Sinn, H.; Kaminsky, W. Adv. Organomet. Chem. 1980,18, 99.  (49)  Kaminsky, W.; Kulper, K.; Brintzinger, H. H.; Wild, F. R. W. P. Angew. Chem., Int. Ed. Engl. 1985,24, 507.  (50)  Jordan, R. F. / . Chem. Ed. 1988, 65, 285.  (51)  Watson, P. L.; Parshall, G. W. Acc. Chem. Res. 1985, 25, 51.  (52)  Watson, P. L. / . Am. Chem. Soc. 1982,104, 337.  (53)  Jeske, G.; Lauke, H.; Mauermann, H.; Schumann, H.; Marks, T. J. / . Am. Chem. Soc. 1985, 707,8111.  (54)  Bunel, E.; Burger, B. J.; Bercaw, J. E. / . Am. Chem. Soc. 1988,110,976.  24  References begin on page 21  Chapter  (55)  1: Introduction  to Amidophosphine  Complexes...  Conticello, V. P.; Beard, L.; Giardello, M. A.; Tsuji, Y.; Sabat, M.; Stern, C. L.; Marks, T. J. / . Am. Chem. Soc. 1992,114, 2761.  (56)  Ballard, D. G. H.; Courtis, A.; Holton, J.; McMeeking, J.; Pearce, R. / . Chem. Soc, Chem. Commun. 1978, 994.  (57)  Schaverien, C. J. Organometallics 1994,13, 69.  (58)  Hogerheide, M. P.; Grove, D. M.; Boersma, J.; Jastrzebski, J. T. B. H.; Kooijman, H.; Spek, A. L.; van Koten, G. Chem. Eur. J. 1995,1, 343.  (59)  Albrecht, I.; Schumann, H. J. Organomet. Chem. 1986,310, C29.  (60)  Schumann, H.; Albrecht, I.; Pickardt, J.; Hahn, E. / . Organomet. Chem. 1984,276, C5.  (61)  Heeres, H. J.; Meetsma, A.; Teuben, J. H. / . Chem. Soc, Chem. Commun. 1988, 962.  (62)  Heijden, H. v. d.; Pasman, P.; Boer, E. J. M. d.; Schaverien, C. J.; Orpen, A. G. Organometallics 1989, 8, 1459.  (63)  Heijden, H. v. d.; Schaverien, C. J.; Orpen, A. G. Organometallics 1989,8,255.  (64)  Hazin, P. N.; Huffman, J. C ; Bruno, J. W. Organometallics 1987, 6, 23.  (65)  Tilley, T. D.; Andersen, R. A. Inorg. Chem. 1981,20, 3267.  (66)  Tsutui, M.; Chen, L. B.; Bergbreiter, D. E.; Miyamoto, T. K. / . Am. Chem. Soc. 1982, 104, 855.  (67)  Schumann, H.; Meese-Marktscheffel, J. A.; Esser, L. Chem. Rev. 1995,95, 865.  (68)  Schaverien, C. J. Adv. Organomet. Chem. 1994,36,283.  (69)  Hagadorn, J. R.; Arnold, J. Organometallics 1996,15,984.  25  References  begin on page  21  Chapter  1: Introduction  to Amidophosphine  Complexes...  (70)  Schaverien, C. J. Inorg. Chem. 1991,30, 4968.  (71)  Arnold, J.; Hoffman, C. G. / . Am. Chem. Soc. 1990,112, 8620.  (72)  Arnold, J.; Hoffman, C. G.; Dawson, D. Y.; Hollander, F. J. Organometallics 1993,12, 3645.  (73)  Fryzuk, M. D.; Berg, D. J.; Haddad, T. S. Coord. Chem. Rev. 1990, 99, 137.  (74)  Fryzuk, M . D.; Haddad, T. S.; Rettig, S. J. Organometallics 1991,10, 2026.  (75)  Piers, W. E.; Ferguson, G.; Gallagher, J. F. Inorg. Chem. 1994,33, 3784.  (76)  Karsch, H. H.; Ferazin, G.; Koojman, H.; Steigelmann, O.; Schier, A.; Bissinger, P.; Hiller, W. / . Organomet. Chem. 1994,482, 151.  (77)  Hitchcock, P. B.; Lappert, M. F.; MacKinnon, I. A. / . Chem. Soc, Chem. Commun. 1988, 1557.  (78)  Fryzuk, M. D.; Haddad, T. S.; Rettig, S. J. Organometallics 1992,11, 2967.  (79)  Karsch, H. H.; Appelt, A.; Muller, G. Angew. Chem., Int. Ed. Engl. 1986, 9, 823.  (80)  Aspinall, H. C.; Moore, S. R.; Smith, A. K. / . Chem. Soc, Dalton Trans. 1992, 153.  (81)  Cary, D. R.; Arnold, J. / . Am. Chem. Soc. 1993,115, 2520.  (82)  Aspinall, H. C.; Moore, S. R.; Smith, A. K. / . Chem. Soc, Dalton Trans. 1993, 993.  (83)  Rabe, G. W.; Riede, J.; Schier, A. Inorg. Chem. 1996,35, 40.  (84)  Schaefer, W. P.; Kohn, R. D.; Bercaw, J. E. Acta. Cryst. 1992, c48, 251.  (85)  Bercaw, J. E., personal communication.  26  References  begin  on page  21  Chapter  1: Introduction  to Amidophosphine  Complexes...  (86)  Fryzuk, M. D. Can. J. Chem. 1992, 70, 2849.  (87)  Lappert, M. F.; Power, P. P.; Sanger, A. R.; Srivastava, R. C. Metal and Metalloid Amides; Wiley: New York, 1979.  (88)  McAuliffe, C. A. In Comprehensive Coordination Chemistry; G. Wilkinson, R. D. Gilliard and J. A. McCleverty, Ed.; Pergamon: London, 1987; Vol. 2.  (89)  Pearson, R. G. J. Chem. Ed. 1968,45, 581.  (90)  Fryzuk, M. D.; MacNeil, P. A.; Rettig, S. J.; Secco, A. S.; Trotter, J. Organometallics  1982,1, 918. (91)  Fryzuk, M. D.; Gao, X.; Joshi, K.; MacNeil, P. A.; Massey, R. L. / . Am. Chem. Soc.  1993,115, 10581. (92)  Fryzuk, M. D.; Carter, A.; Westerhaus, A. Inorg. Chem. 1985, 24, 642.  (93)  Tolman, C. A. Chem. Rev. 1977, 77, 313.  (94)  Harris, R. K. Nuclear Magnetic Resonance Spectroscopy; Pitman Books Limited: London, 1983.  (95)  Garrou, P. E. Chem. Rev. 1981,81,229.  (96)  Meek, D. W.; Mazanec, T. J. Acc. Chem. Res. 1981,14,266.  (97)  Nixon, J. F.; Pidcock, A. Ann. Rev. NMR Spec. 1969,2, 345.  (98)  Fryzuk, M. D.; Mao, S. S. H.; Duval, P. B.; Rettig, S. J. Polyhedron 1995,14, 11.  (99)  Dessy, R. E.; Okuzumi, Y.; Chen, A. / . Am. Chem. Soc. 1962, 84, 2899.  (100) Fraser, R. R.; Mansour, T. S. / . Org. Chem. 1984,49, 5284.  27  References  begin  on page  21  Chapter 1: Introduction to Amidophosphine Complexes... (101) Fryzuk, M. D.; Mao, S. S. H.; Zaworotko, M. J.; MacGillivray, L. R. / . Am. Chem. Soc. 1993,115, 5336. (102) Caminade, A. M.; Majoral, J. P. Chem. Rev. 1994, 94, 1183. (103) Kyba, E. P.; Davis, R. E.; Hudson, C. W.; John, A. M.; Brown, S. B.; McPhaul, M. J.; Liu, L. K.; Glover, A. C. / . Am. Chem. Soc. 1981,103, 3868. (104) Coles, S. J.; Edwards, P. G.; Fleming, J. S.; Hurthouse, M. B.; Liyanange, S. S. / . Chem. Soc, Chem. Commun. 1996, 293. (105) Fryzuk, M. D.; Love, J. B.; Rettig, S. J. / . Chem. Soc, Chem. Commun. 1996, 2783. (106) Fryzuk, M. D.; Haddad, T. S. / . Am. Chem. Soc. 1988,110, 8263. (107) Fryzuk, M. D.; Haddad, T. S. / . Chem. Soc, Chem. Commun. 1990, 1088. (108) Booij, M.; Deelman, B. J.; Duchateau, R.; Postma, D. S.; Meetsma, A.; Teuben, J. H. Organometallics 1993,12, 3531. (109) Casey, C. P.; Hallenbeck, S. L.; Pollock, D. W.; Landis, C. R. / . Am. Chem. Soc. 1995, 117, 9770. (110) Molander, G. A.; Hoberg, J. O. / . Am. Chem. Soc 1992,114, 3123. (111) Haan, K. H. D.; Wielstra, Y.; Teuben, J. H. Organometallics 1987, 6, 2053. (112) Heeres, H. J.; Teuben, J. H. Organometallics 1991,10,1980. (113) Deelman, B. J.; Booij, M.; Meetsma, A.; Teuben, J. H.; Kooijman, H.; Spek, A. L. Organometallics 1995,14, 2306. (114) Coughlin, E. B.; Bercaw, J. E. / . Am. Chem. Soc. 1992,114, 7606.  28  References begin on page 21  Chapter 1: Introduction to Amidophosphine Complexes...  (115) Schaverien, C. J. Organometallics 1994,13, 69. (116) Gagne, M. R.; Stern, C. L.; Marks, T. J. J. Am. Chem. Soc. 1992,114, 275. (117) Haddad, T. S. Ph. D. Thesis, University of British Columbia, 1990. (118) Fryzuk, M. D.; Love, J. B.; Rettig, S. J. / . Am. Chem. Soc. 1997,119, 9071.  29  References begin on page 21  Chapter 2  Synthesis, Characterization and Reactivity of Phosphine Complexes of Scandium  2.1 Introduction Compared to most metals in the transition series and the other group 3 elements, the organometallic chemistry of scandium is still largely unexplored. However, recendy there has been a surge of activity in this area " mostly due to the recognition that the enhanced Lewis 1  4  acidity of Sc(III) can result in new reactivity patterns with small molecules. Although a number 5  of homoleptic alkyls and aryls of scandium have been reported, ' by far the vast majority of 6 7  organometallic complexes of this element utilize modified-cyclopentadienyl ancillary ligands to stabilize particular coordination geometries. " 4,8  14  For example, the o-bond metathesis reaction  owes its existence to early work with bis(pentamethylcyclopentadienyl) systems such as 15  Cp*2ScR; in fact, the combination of two substituted cyclopentadienyl supporting ligands is a common theme in the recent chemistry of scandium. " 5,16  20  Our approach to examining the chemistry of the early transition metals and the elements of group 3 has been to use the tridentate, mixed ligand system [N(SiMe2CH2PR2)2]"; " the 21  24  hard amido donor serves to anchor the ligand to those electropositive metal centers and this forces the phosphine donors to bind by virtue of their proximity. In this way, we have been able to prepare phosphine complexes of many metals for which the phosphine ligand may not be especially well suited, for example, phosphine complexes of the heavier elements of group 3,  30  References begin on page 94  Chapter 2: ...Phosphine Complexes of Scandium yttrium and lanthanum.  25,26  Here, we describe our efforts in the coordination chemistry with  Sc(UI) particularly in the formation of dialkyl derivatives.  2.2.1 Synthesis of ScCl2(THF)[N(SiMe CH PPr 2)2] (1) i  2  2  The reaction of the lithium salt of the ligand, LiN(SiMe2CH2PPr 2)2 i  2 7  with ScCl3(THF)3  in toluene at 100 °C for 24 h generates good yields of the mono(ligand) starting material ScCl2(THF)[N(SiMe2CH2PPri2)2] (1) (equation 2.1). Even though ScCl3(THF)3 is insoluble in toluene, this solvent is necessary for the metathesis to proceed. The coordinated THF can be removed by pumping the solid under vacuum as evidenced by the elemental analysis; the coordinated THF however can be detected by NMR spectroscopy and was also evident in the particular crystal used for the X-ray analysis (see later). The solution spectroscopic data for the dichloro derivative 1 is consistent with a meridional coordination of the tridentate ancillary ligand and a trans orientation of the two chloride ligands with the coordinated THF ligand thus trans to the amide donor; this is evident from the fact that the silyl methyl proton resonances in the ! H NMR spectrum appear as a singlet indicating C2v symmetry. The P{ H} NMR 31  1  spectrum of 1 consists of a broad singlet (A1/2 = 480 Hz) slightly shifted from that of the free ligand; the broadness of the peak persists at all temperatures and is a result of the quadrupolar 45  S c nucleus ( Sc, I = 7/2,100% natural abundance). 45  toluene  [2.1]  2  31  References begin on page 94  Chapter 2: ..Phosphine Complexes of Scandium  The T H F adduct 1 is the preferred starting reagent over the THF-free material ScCl [N(SiMe CH PPri ) ] simply because the latter is less soluble in hydrocarbon solvents 2  2  2  2  2  than 1. The coordinated T H F does not hinder further metathesis chemistry; this is in contrast to Cp*2ScCl(THF), which is quite inert.  15  The coordinated T H F must be removed by sublimation  to yield Cp* ScCl, which is a useful starting material for the synthesis of permethylscandocene 2  derivatives.  2.2.2 Structure of ScCl2(THF)[N(SiMe CH PPr 2)2] (1) i  2  2  Slow evaporation of a toluene solution of ScCl2(THF)[N(SiMe2CH PPr 2)2] (1) yielded i  2  colorless plates that were suitable for single crystal X-ray diffraction. The molecular structure and numbering scheme are illustrated in Figure 2.1.  Figure 2.1 Molecular structure of ScCl (THF)[N(SiMe2CH2PPr )2] (1); 33% probability i  2  2  thermal ellipsoids are shown. Hydrogen atoms are omitted for clarity.  32  References begin on page 94  Chapter 2: .. .Phosphine Complexes of Scandium  The coordination environment around the Sc centre can best be described as distorted octahedral, with trans-disposed chlorides and the THF molecule trans to the amide nitrogen of the ancillary tridentate ligand. The PNP ligand is bound in a meridional fashion with the P(l)-Sc(l)-P(2) angle bent back to 164.6(1)°. The distorted octahedral geometry is evidenced by both the Cl(l)Sc(l)-Cl(2) bond angle of 168.1(1)° and the 0(1)-Sc(l)-N(l) bond angle of 174.4(3)°. In addition, the backbone of the ancillary amidodiphosphine ligand is twisted with respect to the plane defined by the NPScP atoms; the dihedral angle between the NSi2 and NPScP planes is 34.43° for 1. The scandium-chlorine distances of 2.448(3) A and 2.418(3) A are slightly longer than those reported for the only other characterized compound containing terminal scandiumchlorine bonds, ScCl3(THF)3, for which an average Sc-Cl distance of 2.413 A was cited.  28  By  comparison, Cp2ScCl, which is known to have a dimeric structure in the solid state, contains bridging Sc-Cl bond lengths of 2.575 A (ave), the lengthening of a bond to a bridging halide being quite common  2 9 , 3 0  The scandium-phosphorus bond lengths of 2.752(3) A and 2.756(3) A  are comparable to.those found in [Me2Si(r) -C5Me4)2]ScH(PMe3) (2.752(1) A ) . 5  31  As is the case  with [Me2Si(T| -C5Me4)2]ScH(PMe3), it appears that the Sc-P bond distance, approximately 5  0.21 A larger than the sum of the covalent radii,  31  is indicative of a weak Sc-P bond. It is  noteworthy that the two other previously characterized phosphine complexes of scandium, [(Cp*SiNR)(PMe )Sc]2(f^2-'n -Tl -C2H4) and [(Cp*SiNR)(PMe )Sc] (^-H)2 each possess 2  2  3  3  2  uncharacteristically long Sc-P bonds (2.825(3) and 2.996(1) A respectively  10,32  ); however, in  each of these cases, PMe3 dissociation was determined to be rapid. Complex 1 also contains an unremarkable scandium-amide bond distance of 2.101(7) A which is intermediate between those reported for Cp*-based systems and the somewhat longer Sc-N bonds found in (OEP)ScR 10  (OEP = octaethylporphyrinato; R = alkyl, aryl) derivatives.  1,10,33  Table 2.1 lists bond lengths  and angles for 1.  33  References begin on page 94  Chapter 2: ...Phosphine Complexes of Scandium  Table 2.1 Selected bond lengths (A) and angles (deg) for ScCl2(THF)[N(SiMe2CH2PPr2)2] (1). i  P(l)-Sc(l)  2.752(3)  PQ)-Sc(l)  2.756(3)  Cl(l)-Sc(l)  2.448(3)  Cl(2)-Sc(l)  2.418(3)  N(l)-Sc(l)  2.101(7)  0(1)-Sc(l)  2.271(7)  P(l)-Sc(l)-P(2)  164.6(1)  Cl(l)-Sc(l)-Cl(2)  168.1(1)  N(l)-Sc(l)-0(1)  174.4(3)  Cl(l)-Sc(l)-P(l)  85.2(1)  Cl(l)-Sc(l)-P(2)  95.1(1)  Cl(2)-Sc(l)-P(l)  99.0(1)  Cl(2)-Sc(l)-P(2)  88.8(1)  P(l)-Sc(l)-N(l)  83.2(2)  P(2)-Sc(l)-N(l)  81.5(2)  2.3.1 Synthesis of Scandium Dialkyls ScR2[N(SiMe CH PPr 2)2] (R = Me (2), Et (3), i  2  2  C H S i M e (4)) 2  3  The formation of bis(hydrocarbyl) derivatives of scandium can be effected by a variety of alkylating reagents; however, the selection of solvent is as important as it is in the case of the starting dichloro compound 1 (equation 2.2). By using alkyl lithium (RLi) reagents, dialkyl derivatives could be isolated in pure form, although these species showed a propensity for incorporation of LiCl. The use of Grignard or organozinc reagents did not result in the formation of characterizable materials.  ScCI^THFJINCSiMegCHgPPr'^g]  + 2 RLi  >  ( g  1 ScR [N(SiMe CH PPr ) ]«(LiCI) l  2  2  2  2  2  x  [2.2]  2 : R = Me 3 : R = Et 4 : R = CH SiMe 2  34  3  References begin on page 94  Chapter 2: ...Phosphine Complexes of Scandium These metatheses were best performed using toluene as a reaction solvent for both the diethyl, 3, and the bis(trimethylsilylmethyl), 4; however, for the formation of the dimethyl complex, 2, the use of hexanes/Et20 was necessary. (The effect of solvent on this reaction will be discussed later.) These colorless, thermally-stable dialkyl derivatives are extremely soluble in hydrocarbon solvents and can only be crystallized with difficulty from (Me3Si)20. The elemental analyses are indicative of the incorporation of variable amounts of LiCl which can be reduced by repeated recrystallizations. The *H NMR spectroscopic parameters for these species are quite simple: the ligand backbone resonances are consistent with a highly symmetric environment (one silyl methyl resonance) and the equivalent hydrocarbyl groups show coupling of two equivalent phosphorus-31 nuclei to the a-H's. The P{ *H} NMR spectra of 2 - 4 show very broad singlets 31  (Al/2 = 90 - 230 Hz) that are not particularly sensitive to temperature. There is no feature that corresponds directly to the presence of LiCl nor do any of the single crystals show evidence of regular LiCl incorporation in the lattice.  2.3.2 Structure of ScEt [N(SiMe CH PPr 2)2] (3) i  2  2  2  Both of the bis(hydrocarbyl) derivatives 3 and 4 are isolated as colorless prisms after slow evaporation of hexamethyldisiloxane solutions. The results of the single-crystal X-ray diffraction analyses are presented in Figures 2.2 and 2.3, respectively. In both cases, the dialkyls are isolated free from the interaction of a THF donor molecule and can best be described as possessing a distorted trigonal bipyramidal geometry in the solid state. The diethyl derivative 3 maintains C2v symmetry, and thus has two N(l)-Sc(l)-C(10) bond angles measuring 117.59(8)°. Again, as in 1, the backbone of the ancillary ligand is twisted out of the axial plane (defined by the NPScP atoms) with a dihedral angle of 36.62°; in the equatorial plane, the C(10)-Sc(l)C(10)* bond angle is 124.8(2)°. The scandium-phosphorus bonds are in the axial positions with a P(l)-Sc(l)-P(l)* bond angle of 166.95(4)°; here again, the phosphines are cantilevered slightly  35  References begin on page 94  Chapter 2: ...Phosphine Complexes of Scandium  back from a purely trans disposition although less than that observed for the octahedral geometry found in 1.  Figure 2.2 Molecular structure of ScEt2[N(SiMe2CH PPr 2)2] (3); 33% probability thermal i  2  ellipsoids are shown. Hydrogen atoms are omitted for clarity.  Compound 3 also exhibits scandium-carbon bond lengths of 2.242(3) A , almost identical to those reported for C p * S c M e , 4  2  15  (OEP)ScMe and (OEP)ScCH(SiMe ) . 3  2  1  What is remarkable  is the lack of any evidence for an agostic interaction between the coordinatively unsaturated scandium centre and the {3-hydrogens of the two ethyl groups.  In our systems, the dialkyl  derivatives S c R [ N ( S i M e C H P P r ) ] are formally 12-electron compounds and, hence, are i  2  2  2  2  2  electron deficient. As 3 is isolated free from Lewis-base interactions, and given the Lewis-acidic nature of scandium, it would be reasonable to expect 3 to harbor an agostic Sc-C-C-H  36  References begin on page 94  Chapter 2: ...Phosphine Complexes of Scandium  interaction. Previously, the monoethyl derivative Cp*2ScCH-2CH3 was inferred to possess an agostic (5-C-H bond interaction, although disorder in the crystal prevented an accurate determination of the relevant bond lengths and angles.  15  However, the Sc-C(10)-C(l 1) angle in  3 is 115.5(2)° and not acute enough to suggest a |3-C-H interaction. Bond lengths and angles for 3 and 4 are available in Table 2.2.  Table  2.2 Selected bond lengths (A) and angles (deg) for ScEt2[N(SiMe2CH2PPri2)2] (3) (left) a  and Sc(CH SiMe3)2[N(SiMe CH2PPr 2)2] (4) (right). i  2  Pd)-Sc(l)  C(10)-Sc(l)  2.779(2)  2.242(3)  P(D-Sc(l)  2.794(2)  P(2)-Sc(l)  2.788(2)  C(19)-Sc(l)  2.224(5)  C(23)-Sc(l)  2.204(5) . 2.126(4)  N(l)-Sc(l)  2.103(3)  N(l)-Sc(l)  P(l)-Sc(l)-P(2)  166.95(4)  P(l)-Sc(l)-P(2)  154.92(6)  C(10)-Sc(l)-C(10*)  124.8(2)  C(19)-Sc(l)-C(23)  113.5(2)  C(10)-Sc(l)-P(l)  97.33(7)  C(19)-Sc(l)-P(l)  92.2(2)  C(19)-Sc(l)-P(2)  102.3(1)  C(23)-Sc(l)-P(l)  99.7(2)  C(23)-Sc(l)-P(2)  93.1(1)  C(19)-Sc(l)-N(l)  113.2(2)  C(23)-Sc(l)-N(l)  133.3(2)  C(10*)-Sc(l)-P(l)  C(10)-Sc(l)-N(l)  a  2  88.72(8)  117.59(8)  Symbols refer to symmetry operations: (*) 1-A:, y, 1/2-z.  37  References begin on page 94  Chapter 2: ...Phosphine Complexes of Scandium  2.3.3 Structure of Sc(CH SiMe3)2[N(SiMe2CH PPr 2)2] (4) i  2  2  The solid state structure of the bis(trimethylsilylmethyl) complex 4 is similar to that of 3 in that it too exhibits a coordination sphere that can best be described as trigonal bipyramidal, although deviations from this ideal geometry are more pronounced. The twist angle of the N S i  2  portion of the backbone is found to be 40.82°. Complex 4 possesses N(l)-Sc(l)-C(19, 23) bond angles of 133.3(2)° and 113.2(2)° which are the greatest indication of dissymmetry in the molecule.  Also of note is the reduced P(l)-Sc(l)-P(2) bond angle of 154.92(6)° as compared  with that of 166.95(4)° for 3.  This is undoubtedly due to the greater steric demands of the  CH SiMe3 moiety as compared to that of C H C H 3 . It should also be noted that the Sc(l)-C(19)2  2  Si(3) bond angle of 131.4(3)° is probably a result of steric crowding around Sc and not due to any a-agostic interaction. Compounds 3 and 4 both contain comparable scandium-amide and scandium-phosphorus bond lengths; as was found for 1, the latter are indicative of weak Sc-P bonds.  C(15)  Figure 2.3 Molecular structure of Sc(CH SiMe ) [N(SiMe CH PPri ) ] (4); 33% probability 2  3  2  2  2  2  2  thermal ellipsoids are shown. Hydrogen atoms are omitted for clarity.  38  References begin on page 94  Chapter 2: ...Phosphine Complexes of Scandium The dialkyl complexes ScR2[N(SiMe2CH2PPr 2)2] can be considered 12-electron species i  assuming that the amido ligand is a four-electron donor. Such coordinative unsaturation might be expected to lead to agostic interactions  20,34  with either the cc-C-H's of the dimethyl complex 2  and the bis(trimethylsilylmethyl) derivative 4, or (3-C-H's of the diethyl complex 3. There is no evidence to suggest the presence of such an interaction; *H NMR data of the diethyl complex 3 are not diagnostic of a static or rapidly fluxional agostic structure and solid or solution IR spectroscopy reveal no low-energy C-H bands. In addition, the carbon-hydrogen coupling constants (Vc-H = 119, 125 Hz) for the a- and P-hydrogens of the diethyl derivative 3 are typical for sp C-H bonds, further support that no additional interaction with the unsaturated Sc centre is 3  occurring.  2.4 Solvent Dependency of the Reaction Between ScCl2(THF)rN(SiMe2CH PPr 2)2] (1) and i  2  MeLi (2 equiv) As previously shown, the metathesis reaction between ScCl2(THF)[N(SiMe2CH2PPr )2] i  (1)  and MeLi  (2  equiv)  proceeds  smoothly  in hexanes to yield the  2  dialkyl  ScMe2[N(SiMe2CH2PPr 2) ] (2). However, if toluene is used as the reaction solvent, a mixture i  2  of products is obtained, the structures of which are unknown. The P{ H} NMR spectrum of 31  1  the products, labeled as 5 (Figure 2.4), is indicative of this change. Compound 5 is isolated as an extremely soluble colorless solid. Instead of a single broad peak, as for 1 - 4, a much more complex pattern is found. The lack of quadrupolar broadening of any of the peaks is due to the absence of phosphine coordination to scandium. The features of the P{ ^H) NMR spectrum are 31  best described as two A B X multiplets (A,B =  3 1  P, X = Li) 7  3 5  arising from two similar but  uncoupled products. In other words, the spectrum seen is the result of two overlapping AB doublets which are then further split by coupling to lithium-7 (I = 3/2,92.6% abundance) (Figure 2.5).  39  References begin on page 94  Chapter 2: ...Phosphine Complexes of Scandium  if -3  -4  * * i *  -5  ppm  -8  Figure 2.4 202.4 MHz 31p{ 1H} NMR spectrum of 5 in C6D . 6  PA  B  •'P-P + # * * * •  #•  0.  I.  %\ %  «• •» v«  'P-Li  Figure 2.5 Splitting diagram for an ABX spin system (A,B = 3lp, x = Li). 7  40  References begin on page 94  Chapter 2: ...Phosphine Complexes of Scandium The spectral features may be simulated by NMR-U using the following parameters: 5a; 8 = -3.3, -8.6 ppm; / _ = 12 Hz, l / . 2  P  P  P  L i  = 53 Hz; 5b; 8 = -3.9, -6.7 ppm; 2 / . = 6 Hz, l / . p  p  P  L i  = 54 Hz  (Figure 2.6).  41  References begin on page 94  Chapter 2: ...Phosphine Complexes of Scandium  When the spectra from the two products are overlaid, a reasonable facsimile of the spectrum presented in Figure 2.4 can be generated.  This implies that the reaction of  ScCl2(THF)[N(SiMe2CH2PPr2)2] (1) and MeLi (2 equiv) in toluene yields two products, in an i  approximately equal ratio. Heating the mixture to 100 °C does not induce any obvious change to the 31p{lH} NMR spectrum aside from a loss of resolution. Thus, the two products are not involved in any kind of exchange process, either intermolecular or intramolecular. The 3lp _ 31p COSY NMR spectrum of the reaction products confirms the presence of two independent species (Figure 2.7); the cross-peaks illustrate that the sets of resonances centered at -3.3 ppm and -8.7 ppm are coupled (5a), as are those at -3.9 and -6.7 ppm (5b). The COSY spectrum also shows that 5a is not coupled to 5b.  ppm Figure 2.7 202.4 MHz 3lp.3lp COSY NMR spectrum of 5a and 5b in C D . 6  42  6  References begin on page 94  Chapter 2: ...Phosphine Complexes of Scandium  When MeLi is added to ScCl2(THF)[N(SiMe2CH2PPr2)2] (1) the solution immediately i  becomes  cloudy.  This  implies  that  the  metathesis  reaction  to  produce  ScMe2[N(SiMe2CH2PPr )2] (2) and 2 equiv LiCl occurs. If the P { H ) NMR spectrum is i  31  1  2  obtained immediately after the reaction (~5 min) a broad peak at 0.1 ppm is observed; over time this single peak gradually evolves into the more complex pattern discussed above. Thus the desired dialkyl 2 is synthesized, but then rearranges to coordinate lithium, presumably in the form of LiCl, instead of scandium. As we will see in Chapter 3, aluminum can also be displaced in favor of lithium by the addition of excess methyl lithium to an aluminum dialkyl. The incorporation of LiCl and other lithium containing species by Group III complexes and those of the lanthanides is quite common.  30,36  "  41  We suggest that a tetrahedral lithium centre is present,  coordinated by two inequivalent phosphines, a chloride and a solvent molecule (Figure 2.8). A tetrahedral geometry is the most common for lithium  4 2  Cl  , Li  B  Figure 2.8 Tetrahedral lithium centre proposed to be present in 5a and 5b.  A secondary interaction with the rest of the complex, or some asymmetry present in the remaining portion of the complex must then render the phosphines inequivalent. The lithium-7 NMR spectrum is consistent with the geometry proposed in Figure 2.8. A triplet at 0.1 ppm (^Li-P =  53 Hz) is present, the result of coupling to two phosphorus-31 nuclei (Figure 2.9).  43  References begin on page 94  Chapter 2: ...Phosphine Complexes of Scandium  -2 ppm Figure 2.9 194.4 MHz 7Li{ lH} NMR spectrum of 5a and 5b in C D . 6  6  If two inequivalent phosphorus nuclei were present, one would expect a doublet of doublets, and not a triplet. No change to the spectrum is evident even upon cooling to -90 °C in dg-toluene.  This discrepancy may be due to the different time scales of lithium-7 and  phosphorus-31 N M R ; 35  43,44  or it may be that the two coupling constants  JU-PA  1  and JU-PB are 1  coincident. The presence of only one triplet, and not two, which would be expected from two distinct products, may be the result of the relatively narrow chemical shift window available to lithium (generally 0-2 ppm for neutral species)  42  A broad peak centered at -1.0 ppm can also be  seen, which corresponds to associated LiCl, perhaps involving an interaction with the scandium nucleus. Compounds 5a and 5b were also studied by *H and Sc{!H} NMR spectroscopy. The 45  H NMR spectrum consisted of a forest of peaks from which no useful information could be  l  44  References begin on page 94  Chapter 2: ...Phosphine Complexes of Scandium obtained. As well, S c NMR spectroscopy revealed a broad peak at -220 ppm (A1/2 = 5.1 KHz), 45  with no coupling to phosphorus or lithium. The microanalysis of the product may be interpreted in two ways, as shown in Table 2.3; the results are consistent with ScMe2[N(SiMe2CH2PPr 2)2] (2) further complexed by either i  (Et20)(LiCl)2.3 or (Et20)2(LiCl)3.i. The second formulation allows us to propose a structure to account for the features observed in the P{ H} and Li{ H} NMR spectra. 31  l  7  l  Table 2.3 Microanalytical data for 5.  Complex  C  H  N  Calculated for ScMe2[N(SiMe2CH2PPri2)2]  51.36  10.78  2.99  Calculated for  45.09  9.46  2.19  45.00  9.44  1.87  45.03  9.36  2.12  ScMe [N(SiMe2CH PPri2)2] (Et 0)(LiCl) 3 2  2  2  2  Calculated for ScMe [N(SiMe2CH PPri2)2](Et20) (LiCl)3i 2  2  2  Found  The  proposed  structure  of  the  species  (Et2 O ) L i (|i -Cl)(|i-  Me)ScMe[N(SiMe2CH2PPri2)2]LiCl(Et20) (5), although speculative, is in accord with the NMR spectra and microanalytical results.  A lithium chloride molecule is coordinated by two  phosphines and an ether molecule; the stereogenic scandium centre renders the phosphines magnetically distinct, rationalizing the ABX multiplet present in the P { H ) NMR spectrum. 31  1  The presence of two products (or two ABX multiplets) may be accounted for by two differing salt interactions at scandium; the (Et20)»Li unit may alternately bridge the two methyl groups  45  References begin on page 94  Chapter 2: ...Phosphine Complexes of Scandium  instead of a methyl group and a chloride. The type of coordination experienced by the scandium centre, specifically that to LiCl»(Et20), is well known in lanthanide chemistry.  45  CI  Pr : 1  :"7  Et 02  CI  Si Me  --*y .Sc  \  ^ 'Me"  >r  2  2  Me  Figure  Et 0 2  PB  Me2S  <«* Li  \  UCI 1.1  2.10 Proposed solution structure of (Et20)Li(u,-Cl) (u,-Me)ScMe[N(SiMe CH PPr )2]LiCl(Et20) (5). i  2  2  2  We have previously encountered the undefined association of salt to Zr(III)-alkyl complexes of [N(SiMe2CH PPr 2)2]- as well as in some of the dialkyls synthesized here. Finally, we must i  46  2  qualify this argument by stating that all the presented data is circumstantial, and in the absence of an X-ray crystal structure determination, the true identity of the products of this reaction can only be speculated upon.  46  References begin on page 94  Chapter 2: ...Phosphine Complexes of Scandium  2.5 Reactivity of Scandium Dialkyls ScRitNtSiMezCHiPPr^h] (R = Me (2), Et (3), C H S i M e (4)) 2  3  In an effort to probe the reactivity of these species, we performed a number of standard reactions with the dialkyl complexes ScR2[N(SiMe2CH2PPr2)2] to test their ability to activate i  small molecules. Unfortunately, in every case that we examined, the complex decomposed to give a mixture of intractable materials including some protonated ligand, HN(SiMe2CH2PPr2)2. i  For example, reaction with H gave elimination of alkane (observed by *H NMR spectroscopy) 2  but no hydrides could be detected and the P{ H} NMR spectrum of the crude material 31  1  indicated a complex mixture of products. Similar results were obtained when the dialkyl complexes were allowed to react with CH3I, CO, nitriles, isocyanides and phosphines. An analogous decomposition has been observed with lanthanide dialkyls of the tridentate monoanionic ligand system 2,6-bis[(dimethylamino)methyl]benzenide (Scheme 2.1).  41  van  Koten et al. proposed that bis(trimethylsilylmethyl) lutetium derivatives decomposed via the elimination of tetramethylsilane and a lutetium alkylidyne to yield the protonated ligand. While the scandium dialkyls are thermally stable over long periods of time, it may be that upon exposure to small molecules, phosphine dissociation becomes favored, and the above decomposition pathway becomes operative. Attempts were made to characterize some of the products of these reactions. Thus, in the reaction of the bis(trimethylsilylmethyl) complex 4 with CO, volatiles and involatiles were subjected to NMR analyses, GC-MS and IR spectroscopy; the only products detected were SiMe4 (by H NMR spectroscopy and GC-MS) and l  HN(SiMe2CH2PPr2)2 (by P{ H} and H NMR spectroscopy and IR spectroscopy). The fate i  31  1  l  and function of the carbon monoxide in this decomposition process remain unknown, although CO coordination to the scandium centre may serve to trigger this reaction via phosphine dissociation. Although a number of lanthanide alkylidynes have been isolated (Ln = Y, Nd, Lu),  41,47  we were unable to locate any scandium-containing species.  47  References begin on page 94  Chapter 2: ...Phosphine Complexes of Scandium  A common product in the decomposition of dialkyl complexes of the general formula Cp*LnR (where R = CH SiMe or CH(SiMe3) ) is the alkane SiMe or CH (SiMe3) ; - - oc37  2  2  3  2  4  2  48  49  2  hydrogen abstraction is a facile decomposition route for lanthanide complexes stabilized by a single ancillary ligand. The inability to detect other products in these reaction mixtures is evidence of the instability of lanthanide carbenes. The reaction of the diethyl complex 3 with CO2 is typical of the reactions of dialkyls 2-4 and did show evidence for the formation of Sc(0 CEt) type residues by *H NMR spectroscopy. 2  Most noticeably, the resonances due to the methylene protons of the ethyl groups (ScC// CH3) 2  shifted from 0.55 ppm in the starting diethyl compound (t of q) to 3.26 ppm after exposure to C 0 (q). 2  This shift, and the loss of coupling to 2 equivalent phosphorus-31 nuclei are  characteristic of C 0 insertion into the scandium-alkyl bonds (A). A similar insertion has been 2  noted for Cp* Sc(p-tolyl) and (OEP)ScR. 50  2  48  1  However, substantial decomposition was also  References begin on page 94  Chapter 2: ...Phosphine Complexes of Scandium  observed and we were unable to determine if CO2 insertion into the Sc-N bond was also occurring (B). Previously, CO2 has been found to preferentially insert into metal amide bonds to form carbamate species instead of acetates, i.e., upon exposure to CO2, the ethyl resonances of Et2Hf[N(SiMe3)2]2  remain  Et2Hf[C«2CN(SiMe3)2]2-  51  31  unperturbed,  consistent  with  the  formation  of  P{ *H} NMR spectroscopy revealed peaks consistent with both A  andB.  The dialkyls 2 - 4  were also exposed to olefins to examine their potential as  polymerization catalysts. Both the dimethyl derivative 2 and diethyl 3 were active for the polymerization of ethylene (2 = 120 g mol" h" ; 3 = 55 g mol" h ), in contrast to 1  1  1  _1  bis(trimethylsilylmethyl) 4, which did not react with ethylene. To further study these catalyses, we followed the course of the reaction between 2 and ethylene in a sealed NMR tube. Even though copious quantities of polyethylene were observed to form in the NMR tube during the course of the experiment (~8 h), we did not uncover any significant change in the C{ H] NMR 13  l  spectra; thus it is hard to definitively determine the nature of the active catalyst (i.e., small impurities present may be the actual active species).  Alternately, a much faster rate of  propagation as compared to initiation may also account for these results. Exposing a mixture of the dichloride 1 and methylalumoxane (1:500) to ethylene does not produce any polyethylene.  49  References begin on page 94  Chapter 2: ...Phosphine Complexes of Scandium This implies that the species responsible for these transformations is most likely a neutral Group 3 derivative and not a cationic scandium complex. Finally, complexes 1-4 showed no activity with propylene.  2.6 Electronic Structure of the Model Compound ScMe2tN(SiMe CH2PH )2] 2  2  The singular lack of controllable reactivity of these particular organoscandium complexes was disappointing. The compounds would appear to be kinetically stable enough to be isolated and handled under standard Schlenk or glovebox conditions, but in the presence of a variety of small molecules, these five-coordinate, dialkyl scandium complexes decompose. It is well established that the bis(pentamethylcyclopentadienyl) ligand ancillary set is remarkable in terms of generating reactive early metal fragments; the frontier orbitals of the bent Cp*2M unit have 52  a symmetry that allows the interaction of certain small molecules with the fragment in a way that promotes insertion and metathesis reactions (Figure 2.11).  Cp*<  R  Cp*  Figure 2.11 Frontier orbitals of bis(pentamethylcyclopentadienyl)scandium alkyls.  We therefore examined our complex to determine if the frontier orbitals of these mononuclear bis(hydrocarbyl)s showed any peculiarities. We performed INDO/1 level calculations using the ZINDO program on the CAChe system; the model we used was symmetrized to C  2 v  and the  isopropyl substituents at phosphorus were replaced with hydrogens. The parameters are those  50  References begin on page 94  Chapter 2: ...Phosphine Complexes of Scandium developed in the literature and are generally applicable for first and second row transition metals.  53  The orbital diagram is shown in Figure 2.12. The orbitals of interest are the sets of highest occupied and lowest unoccupied MO's. The HOMO corresponds to the lone pair of electrons on the amide nitrogen p orbital; the next two occupied orbitals, HOMO - 1 and HOMO - 2, are essentially the Sc-C a bonds and are more stabilized by 1.4 eV. The two lowest unoccupied orbitals are two d orbitals at Sc. By comparison to bent metallocene complexes, the frontier orbitals of these bis(hydrocarbyl) systems are noticeably different.  Instead of the  characteristic ai and b unoccupied orbitals of the metallocene systems (Figure 2.11), one can 2  see that the comparable LUMO's of the Sc[PNP] system are d orbitals ofrc-symmetry(a and bi) 2  and cannot accept electron density from filled orbitals of molecules like CO. Thus from a symmetry point of view, one can rationalize that the two important empty orbitals at Sc are not appropriate for overlap with typical small molecules. In addition, the steric bulk of the ancillary amidodiphosphine ligand is such that approach to the LUMO at Sc perpendicular to the plane defined by ScNP is impeded by the backbone of the tridentate ligand, while approach to LUMO 2  + 1 is opposite to the Sc-N bond and here the large isopropyl groups at phosphorus are possibly in the way. Nevertheless, it is probably the symmetry aspect that is the overriding feature of these systems. As a caveat, we point out that rehybridization of these MO's upon interaction with a small molecule is possible and thus these simple ideas do suffer from lack of sophistication. In addition, such an analysis completely ignores kinetic issues and these may be the important reasons for the reactivity differences for the bis(hydrocarbyl) complexes of Sc versus the bent metallocene derivatives. For example, if there is phosphine dissociation, new orbitals would become available that have correct symmetry requirements for interaction with typical small molecules; moreover, phosphine dissociation might also be the trigger to decomposition under certain conditions.  51  References begin on page 94  Chapter 2: ...Phosphine Complexes of Scandium  E(eV)  H  Me»,.. C  2  AT \.  HOMO - 1 A  ..•••CH  3  Me»«...  c  \  C  _-PH  H  2  2  Figure 2.12 Molecular orbital diagram for the model compound ScMe2[N(SiMe2CH2PH2)2J.  52  References begin on page 94  Chapter 2: .. .Phosphine Complexes of Scandium One final point should be mentioned regarding the frontier orbitals of this fragment. We note that there are no appropriate empty orbitals that can overlap with the existing a-C-H or (3-CH bonds in the ground state structures; both the LUMO and the LUMO + 1 are too far away to interact with these alkyl groups without considerable distortion of Sc-C-C angles. In other words, the lack of agostic interactions from both solution and solid state structures is supported by this MO analysis.  2.7 Monoalkyl Scandium Complexes Utilizing the N(SiMe2CH2PPr'2)2 Ligand System As discussed in Sections 2.5 and 2.6, the scandium dialkyls C suffer a similar fate to many other Group 3 and lanthanide dialkyls stabilized by a single ancillary ligand. Although the dialkyls are stable enough to be isolated, they decompose to multiple, unidentifiable products when reacted with small molecules. One approach to this problem is to replace a chloride with a larger, unreactive ligand, which will result in a more coordinatively saturated complex. Additionally, if the ligand can be chosen to donate more than two electrons to the metal, the compound will have greater electronic stability. This leaves a site at the metal at which to pursue reaction chemistry. In an effort to generate more stable derivatives, and better defined reaction pathways, we metathesized a chloride with NaCp to generate compounds of the type Sc(rj 5  C5H5)R[N(SiMe2CH2PPri2)2] (D), which have added steric bulk and an additional four electrons, as compared to the ScR2[N(SiMe2CH2PPri2)2] series.  53  References begin on page 94  Chapter 2: ...Phosphine Complexes of Scandium  2.8.1 Synthesis of Sc(Ti5.C5H )Cl[N(SiMe2CH2PPri )2] (6) 5  The  addition  of  2  a T H F solution  of  NaCp(DME)  to  a  solution  ScCl2(THF)[N(SiMe CH PPri2)2] (1) in THF maintained at -78 ° C results in 2  2  C5H5)Cl[N(SiMe CH PPr ) ] (6) (equation 2.3). i  2  2  2  2  of  Sc(r\ 5  Additionally, the same metathesis can be  effected by the addition of solid NaCp(DME) to a toluene solution of the starting dichloride, although this route requires a longer reaction time (1 week).  ScCl2(THF)[N(SiMe CH PPr 2)2] + NaCp(DME) i  2  2  THF, -78 °C (- NaCI)  [2.3]  54  References begin on page 94  Chapter 2: ...Phosphine Complexes of Scandium  Compound 6 is isolated as a slightly yellow, waxy solid and forms colorless crystals upon slow evaporation of a saturated toluene solution. The *H NMR spectrum of 6 shows a doubling of the ligand resonances, due to the presence of a chloride and a cyclopentadienyl ligand. Heating a ds-toluene solution of 6 to 100 °C does not result in the averaging of the diastereotopic ligand resonances, indicating that the positions of the Cp and CI moieties are fixed, and not subject to exchange via a process such as phosphine dissociation, followed by rotation about the scandium-amide bond and re-coordination. The P{ H} NMR spectrum of 6 exhibits a broad 31  1  singlet (8-1.2 ppm, A1/2 = 425 Hz) at all temperatures. As well, the five equivalent Cp protons couple to two equivalent phosphorous-31 nuclei, exhibiting a triplet in the *H NMR spectrum at 6.51 ppm ( / H - P =1.3 Hz). The identity of the compound was confirmed by mass spectral, 3  microanalytical and X-ray crystallographic studies.  2.8.2 Structure of Sc(Ti5.C H )Cl[N(SiMe2CH PPr 2)2] (6) i  s  5  2  The evaporation of a saturated toluene solution of 6 yielded large colorless crystals for Xray analysis. The molecular structure and numbering scheme are presented in Figure 2.13. The complex is monomeric with one of the chlorides of 1 having been replaced by a Cp ligand. The scandium centre has a geometry intermediate between trigonal bipyramidal and square pyramidal; if the Cp ligand occupies the apical position, the square pyramidal geometry is completed by the amide, the two phosphines and the chloride. The angles defined by the Cpscandium bond, and the four members of the base of the pyramid range from 105.4° to 127.9° (Cp = cyclopentadienyl centroid). The P(l)-Sc(l)-P(2) angle of 147.08(6)° is similar to that in bis(trimethylsilylmethyl) 4 (154.92(6)°). In terms of sterics, the Cp ligand is comparable to the trimethylsilylmethyl ligand. The phosphorus-scandium bond lengths (2.804(2) and 2.822(2) A) are slightly longer than those found for dichloro 1, as is the scandium-amide bond length of 2.116(4) A (compared to 2.101(7) A in 1). The donation of six electrons by the Cp ligand to the  55  References begin on page 94  Chapter 2: ...Phosphine  Complexes  of  Scandium  Molecular structure of Sc(ri -C5H5)Cl[N(SiMe2CH2PPr 2)2] (6); 33% probability 5  Figure 2.13  i  thermal ellipsoids are shown. Hydrogen atoms are omitted for clarity.  metal, as opposed to two offered by a chloride ligand, makes the Sc centre in 6 less electrophilic and this is reflected in the longer bond lengths in 6. The scandium centre is equidistant from all five Cp carbons (2.480(9) - 2.511(7) A ) , consistent with an r^-bound cyclopentadienyl ligand. The scandium-Cp bond length (2.21 A) is similar to that found for (Cp2ScCl)2 (OEP)Sc(ri -C5H5) 5  33  (2.19 and 2.20 A , respectively).  and  Bond lengths and angles for 6 are  tabulated below (Table 2.4).  56  29  References begin on page 94  Chapter 2: ...Phosphine Complexes of Scandium Table 2.4 Selected bond lengths (A) and angles (deg) for Sc(Tl5.C5H5)Cl[N(SiMe2CH PPri2)2](6). 2  P(l)-Sc(l)  2.804(2)  P(2)-Sc(l)  2.822(2)  Cp-Sc(l)  2.21  N(l)-Sc(l)  2.116(4)  Cl(l)-Sc(l)  2.383(2)  P(l)-Sc(l)-P(2)  147.08(6)  P(l)-Sc(l)-Cp  106.9  P(2)-Sc(l)-Cp  105.4  N(l)-Sc(l)-Cp  127.9  Cl(l)-Sc(l)-Cp  116.6  2.9.1 Synthesis of Scandium Complexes Sc^-CsHsMNtSilVUsCI^PPr^h] (R = Me (7), Ph (8),BH (9)) 4  Metathesis of the chloride ligand of Sc(ri -C5H5)Cl[N(SiMe2CH2PPr 2)2] (6) is 5  i  undertaken by a number of alkyl, aryl or borohydride lithium reagents (equation 2.4). These reactions show a much lower dependency upon solvent than the dialkyls 2 - 4 .  As well, the  presence of the large Cp ring reduces the tendency for these complexes to incorporate LiCl as evidenced by the microanalyses. The synthesis of methyl 7 proceeds in toluene, in contrast to that found for the dimethyl derivative 2 (see Section 2.4). Attempts to synthesize Sc(T) 5  C H )Et[N(SiMe2CH PPr 2)2] or Sc(ri5-C5H5)(CH2SiMe3)[N(SiMe2CH PPri2)2], analogous to i  5  5  2  2  compounds 3 and 4, resulted in decomposition.  Sc(ri -C5H5)CI[N(SiMe CH2PPr 2)2] i  5  2  6  + RLi  *  (- LiCl) "  v1e CH PPr ) ] Sc(ri -C5H5)R[N(SiMe CH PPr ) ] b  i  2  2  ,2  2  2  2  2  2  7 : R = Me  8 : R = Ph 9 : R = BH  57  [2.4]  4  References begin on page 94  Chapter 2: .. .Phosphine Complexes of Scandium  The phenyl complex 8 and the borohydride complex 9 could be isolated as thermallystable, off-white solids, which were subsequently obtained as air- and moisture-sensitive colorless crystals from hexamethyldisiloxane  solutions.  However, methyl 7 resisted  crystallization and was used as a yellow, sticky solid. The *H NMR spectrum of each complex shows a doubling of the ligand backbone resonances, due to the presence of an alkyl, aryl or borohydride ligand, and a Cp ligand. In the case of methyl 7, the methyl protons (ScGF/3) resonate as a triplet at -0.19 ppm ( / H - P = 4.3 Hz) due to coupling to two equivalent phosphorus3  31 nuclei. For all three derivatives, the five Cp protons are equivalent, appearing as a triplet around 6.5 ppm. The P{ K} NMR spectra of 7 - 9 show broad, temperature invariant singlets l  31  at about 0 ppm (A 1/2 = 100 - 200 Hz). The structure of the phenyl derivative 8 was investigated by X-ray crystallography.  2.9.2 Structure of Sc(Ti .C H5)Ph[N(SiMe2CH2PPr 2)2] (8) 5  i  5  Colorless crystals suitable for X-ray crystallographic analysis were grown from a saturated hexamethyldisiloxane solution of phenyl 8. Figure 2.14 shows the numbering scheme and solid state structure. As for the chloride derivative 6, the scandium centre has a distorted square pyramidal geometry. In this case, the four angles involving the Cp ring, the scandium centre and the bases of the pyramid range from 104.0° to 130.2°. The P(l)-Sc(l)-P(2) bond angle (149.77(6)°) is in the same range as that determined for 6 and the bis(trimethylsilylmethyl) complex 4. All of the bond lengths involving scandium are normal and show little variation from the parent chloride. The scandium-carbon bond length lacks literature precedent; compound 8 appears to be the first compound containing a scandium-phenyl bond to be structurally characterized. The 2.229(6) A spacing between the scandium and ,s/? -carbon atom may be 2  compared with the distances of 2.242(3) A for diethyl 3 and 2.204(5) and 2.126(4) A for bis(trimethylsilylmethyl) 4. The hybridization of the carbon atom has little effect upon the scandium-carbon bond length. Table 2.5 presents selected bond lengths and angles for 8.  58  References begin on page 94  Chapter 2: ...Phosphine Complexes of Scandium  Figure 2.14 Molecular structure of Sc(Tj -C H )Ph[N(SiMe2CH2PPr 2)2] (8); 33% probability 5  i  5  5  thermal ellipsoids are shown. Hydrogen atoms are omitted for clarity.  Table 2.5 Selected bond lengths (A) and angles (deg) for Sc(Ti5-C H5)Ph[N(SiMe CH2PPr 2)2] (8). i  5  2  Pd)-Sc(l)  2.822(2)  P(2)-Sc(l)  2.836(2)  Cp-Sc(l)  2.24  N(l)-Sc(l)  2.174(4)  C(24)-Sc(l)  2.229(6)  P(l)-Sc(l)-P(2)  149.77(6)  P(l)-Sc(l)-Cp  105.5  P(2)-Sc(l)-Cp  104.0  N(l)-Sc(l)-Cp  130.2  C(24)-Sc(l)-Cp  115.3  59  References begin on page 94  Chapter 2: ...Phosphine Complexes of Scandium  2.9.3 Structure of Sc(Ti5.C H5)(BH4)[N(SiMe2CH PPri2)2] (9) 5  2  Colorless crystals of the borohydride complex 9 remained after the evaporation of a saturated hexamethyldisiloxane solution. Although the basic structural features of the complex were confirmed, a refined structure was not obtained due to disorder. The labile nature of the [BH4]- unit, and its proximity to the quadrupolar scandium nucleus, prevented these resonances  from being observed in the *H NMR spectrum. The absence of peaks assignable to BH4 protons is a characteristic of the U NMR spectrum of [T) -C5H3(SiMe3)2]2Sc(BH4) as well. l  5  9  The  fluxionality of the [BH4]- moiety is readily demonstrated by B NMR spectroscopy: at ambient n  temperature, the spectrum consists of a 1:4:6:4:1 quintet at -27.6 ppm, consistent with coupling to 4 equivalent protons ( ^ B - H = 85 Hz, A1/2 = 32 Hz). When a C7D8 toluene solution of 9 is cooled to -80 °C, the fluxional exchange process between bridge (Hb) and terminal (H ) t  hydrogens is retarded; thus, a single broad peak at -27.0 ppm is observed (A1/2 = 160 Hz).  This  is faster than the exchange in [T| -C5H3(SiMe3)2]2Sc(BH4) which must be heated to 100 °C 5  before a quintet is observed ( ^ B - H = 75 Hz).  9  The solid state infrared spectrum of 9 contains a sharp, strong band at 2478 cm and a -1  broader, weaker signal at 2211 cm . These data are consistent with a B H 4 ligand bound in a -1  tridentate fashion.  54  These numbers compare favorably with a series of lanthanide  tetrahydroborates [Ti -C5H3(SiMe3)2]2M(BH4) (M = La, Pr, Nd, Sm) which feature IR peaks 9  5  diagnostic of tridentate coordination at 2420 and 2240 cm" . 1  to the B-H £).  5 4  The band at 2478 cm' is assigned 1  stretch (A\) and the broad band at 2211 cm" to the B-Hb stretching modes (Ai and 1  t  In contrast to 9, the previously characterized scandium tetrahydroborates [rj 5  C H (SiMe3)2]2Sc(BH4), [TolC(NSiMe )2]2Sc(BH )(THF) and Sc(BH4)3«THF bind B H 9  5  3  30  3  55  4  in a bidentate manner, the former having been structurally characterized. The solution IR spectrum of 9 (hexanes) is identical to that of the solid state.  60  References begin on page 94  4  Chapter 2: ...Phosphine Complexes of Scandium  A fluxional process that would exchange the terminal and bridging hydrogens could involve a monodentate or bidentate intermediate.  Given that monodentate borohydride  compounds are rare, and the preference of BH4 units to coordinate to scandium via 2 or 3 bridging hydrogens, a bidentate intermediate seems most plausible (Scheme 2.2).  54  Scheme 2.2  S c ^  H  l  ^ B -  •H  sc:  2  Hi  B  ..«m«Hi1 n  w S  C  ^  H  1 V ^  B  -  •Hi  Hi  2.10.1 Reactivity of Sc(Ti5.C5H5)Me[N(SiMe CH2PPr'2)2] (7) 2  In contrast to the facile decomposition of dialkyls 2 - 4 with H2, the scandium methyl complex 7 is inert to H2 (4 atm, 5 days). Since H2 is small, the unreactive nature of Sc(ri 5  C5H5)Me[N(SiMe CH PPr ) ] (7) towards H2 is likely a reflection of the low acidity of i  2  2  2  2  dihydrogen. Thus, we examined the reactivity of 7 with more acidic proton sources such as amines. Exposure of methyl 7 to E^NPh or H^NBu (toluene solution, 1 equiv) yields the amides 1  Sc(ti5-C H5)NHPh[N(SiMe2CH2PPr 2)2] (10) i  5  and Sc(ri -C5H )NHBu [N(SiMe2CH2PPr )2] 5  t  5  i  2  (11) with the elimination of methane (equation 2.5).  61  References begin on page 94  Chapter 2: ...Phosphine Complexes of Scandium  Me Si 2  + CH  4  Me2Si  [2.5]  7  10: R = Ph 11: R = Bu'  These reactions proceed analogously to those between Cp*2ScMe and H2NR (R = H, Me, Ph, CH2CMe3).  14  It is likely that these reactions occur by interaction of the nitrogen lone pair  with the scandium centre followed by a four-centered elimination of methane. The syntheses of 10 and 11 are accompanied by a competing decomposition to produce HN(SiMe2CH2PPr2)2 and i  an insoluble scandium-containing precipitate; thus the yields are quite low.  (The amides  Cp*2ScNHR are also produced in low yields. ) The similar solubility of HN(SiMe2CH2PPr )2 14  i  2  to amides 10 and 11 hindered our attempts to isolate them as pure solids. If a large excess of amine is used (4 equiv), only the protonated ligand is observed. We assume the amine protiolytically attacks the scandium-amide bond, resulting in decomposition accompanied by loss of HN(SiMe2CH2PPri2)2.  Cp*2ScMe reacts similarly with hydrazine; with 1 equiv,  Cp*2ScN(H)NH2 is formed although excess NH2NH2 results in decomposition to Cp*H.  56  In  order to ensure the identity of the products, 10 and 11 were also synthesized by a salt-elimination route between the chloro species Sc(r| -C5H5)Cl[N(SiMe2CH2PPr 2)2] (6) and LiNHR (R = Ph, 5  Bu ) (equation 2.6). l  i  The low yields of 10 and 11 by this route as well suggest that the  decomposition to HN(SiMe CH2PPr )2 may be due to reaction of the scandium-amide bond of i  2  2  10 (or 11) with the NH bond of another molecule of 10 (or 11).  62  References begin on page 94  Chapter 2: ...Phosphine Complexes of Scandium  Me Si 2  Me Si  [2.6]  2  6  10: R = Ph 11: R = Bu  The methyl resonance of 7 is no longer present in the *H NMR spectrum of phenylamide 10. The ligand backbone and Cp resonances are shifted from the starting methyl complex, and peaks due to the phenyl ring are visible.  The N / / proton resides at 5.55 ppm (br s); by  comparison, the NH resonance in [PhC(NSiMe3)2]2Sc(NHC6H3-2,6-Pr2) is located at 5.86 i  ppm.  30  Even when synthesized from the lithium amides, 10 and 11 are contaminated with  HN(SiMe2CH PPr 2)2, suggesting that intermolecular cleavage of Sc-N bonds by Sc(r| i  5  2  C5H5)NHR[N(SiMe2CH2PPr 2)2] may also occur.  The related hydrazido complex  i  Cp* ScN(H)NH2 also undergoes a concentration dependent decomposition to eliminate Cp*H.  56  2  The *H NMR spectrum of rm-butylamide 11 contains a NCMe^ resonance at 1.37 ppm (s) and a broad peak which integrates for a single proton at 4.70 ppm. The NH peak for the related compound Cp* Sc(NHCH CMe3) resides at 4.75 ppm. 2  2  14  As for 10, the absence of a methyl  resonance due to the starting methyl complex proposes that 11 is the result of C-H activation of 7.  Phenylamide 10 and rm-butylamide 11 exhibit broad peaks (A 1/2 = 80 and 90 Hz,  respectively) around 0 ppm in the P{ *H} NMR spectra. 31  63  References begin on page 94  Chapter 2: ...Phosphine Complexes of Scandium We also investigated the reaction of these amides with hydrogen (Scheme 2.3). Tertbutylamide 11 undergoes hydrogenation of the scandium-silylamide bond to yield HN(SiMe2CH2PPr 2)2 and unidentified products (Pathway I). The absence of rm-butylamine i  from the *H NMR spectrum of the reaction products proves that hydrogenolysis of the scandiumrm-butylamide bond does not occur. The other reaction product, presumably CpScH(NHBu ), l  decomposes to a number of species as no new NH or Cp resonances were observed.  Scheme 2.3  decomposition  64  References begin on page 94  Chapter 2: ...Phosphine Complexes of Scandium Phenylamide 10 appears to offer both scandium-amide bonds for hydrogenation. As for 11, the *H NMR spectrum of the reaction products includes peaks assignable to HN(SiMe2CH2PPr 2)2, consistent with pathway I. Although new Cp (5 6.50) and NH (8 5.73) i  resonances are present, the absence of peaks attributable to a phenyl group make the existence of CpScH(NHPh) speculative. However, peaks due to aniline are clearly present in the aromatic region of the *H NMR spectrum. This supports addition of H2 across the scandium-phenylamide bond as well (Pathway II).  Alternately, HN(SiMe2CH2PPrJ2)2 may be the result of  decomposition of the scandium-hydride produced by the hydrogenation of the scandiumphenylamide  bond of  The reaction of H2NPh  11.  with  the  hydride  SC(TJ 5  C5H5)H[N(SiMe2CH2PPri2)2] may also be the source of HN(SiMe2CH PPr 2)2. The instability i  2  of scandium hydrides may explain the decomposition of dialkyls 2-4 and 14 and 15 (see later) when exposed to hydrogen.  Reactions between methyl 7 and primary and secondary phosphines were much less well behaved. The reaction of Sc(Ti5-C H5)Me[N(SiMe2CH2PPr 2)2] with H PPh or HPPh (1 equiv i  5  2  2  in THF) results in total decomposition to HN(SiMe2CH2PPr )2 and an insoluble white i  2  precipitate. This is a reflection of the instability of scandium phosphides which have so far eluded characterization. Attempts to generate scandium phosphides of the general formula Sc(rj5-C5H5)(PR2)[N(SiMe2CH PPr 2)2] or Sc(PR )2[N(SiMe2CH PPr )2] (R = Ph, Pr ) via salt i  2  2  i  2  1  2  metathesis routes from the chlorides 1 or 6 and LiPPh2 or LiPPr^ resulted in unidentifiable mixtures. We also examined the potential of these systems as ethylene polymerization catalysts. Although phenyl 8 and borohydride 9 show no activity towards ethylene, methyl 7 polymerizes C2H4 at a rate of 170 g mol" h" . This difference can be accounted for by the smaller size of 1  1  -CH3 as opposed to -C6D5 or -BH4, allowing for coordination and insertion of ethylene. If the chloride 6 is subjected to ethylene using methylalumoxane as a cocatalyst (~500 equiv), the activity increases 10-fold (1600 g mol" h* ). 1  65  1  In its role as a cocatalyst, methylalumoxane  References begin on page 94  Chapter 2: ...Phosphine Complexes of Scandium  methylates the chloride to produce methyl 7. We suspect that the large excess of MAO also coordinates to phosphorus, resulting in a four-coordinate metal centre (7'). This less saturated metal centre would be more electrophilic and less sterically encumbered, resulting in a higher rate of polymerization.  2.10.2 Reactivity of Sc(Ti5.C5H5)(BH4)[N(SiMe2CH2PPri )2] (9) 2  Compounds that contain direct metal-hydrogen bonds are important due to their implication in a large number of stoichiometric and catalytic processes.  57  In certain cases,  transition-metal tetrahydroborate complexes serve as precursors to metal hydride derivatives if B H 3 can be cleaved from the [BH4]- unit. " 58  61  Typically, strong a-donors such as tertiary  phosphines or amines are effective in this regard. Borohydride 9 shows evidence for cleavage of B H 3 by both intra- and intermolecular Lewis bases. In addition to a quintet attributable to  C H5)(BH4)[N(SiMe2CH PPr 2) ] (9) in the room temperature 5  i  2  2  n  SC(TJ 5  B NMR spectrum (5 -27.6,  ^ B - H = 85 Hz), a minor peak at -51.8 ppm is visible. This peak, ~10% the area of the major peak, is a broad multiplet; with proton decoupling it appears as a broad doublet ( V B - P = 60 Hz)  66  References begin on page 94  Chapter 2: ...Phosphine Complexes of Scandium  (Figure 2.15). In addition, a new peak in the *H NMR spectrum (5 5.82 ppm) corresponds to the Cp protons of 9'.  Figure 2.15 96.3 MHz B{ H} NMR spectrum of Sc(ri5-C H )(BH4)[N(SiMe2CH2PPr 2)2] (9) 11  1  i  5  5  and 9' in C6D6.  The *B NMR results can be rationalized if one assumes that the borohydride 9 engages 1  in intramolecular cleavage of BH3 by one of the phosphine donors (Scheme 2.4). There is precedent for this type of behavior with the [N(SiMe2CH2PPr2)2]" ligand set: the first step in i  the formation of the zirconium (IV) species Zr(r|5-C5H )(BH4)(HCOPPr 2CH2SiMe2)i  5  N(SiMe2CH2PPr 2BH3) is suggested to be phosphine dissociation and removal of BH3 to form a i  zirconium hydride.  The resultant hydride 9' would exhibit the broad doublet at -51.8 ppm.  62  However, the relative peak areas of 9 and 9' show little temperature dependence, suggesting that this equilibrium is governed primarily by entropy (AH° = 0).  Another possibility is that  borohydride 9 and hydride 9' are two separate products of the reaction between  Sc(T| 5  C H5)Cl[N(SiMe2CH PPr 2)2] (6) and L1BH4. i  5  2  67  References begin on page 94  Chapter 2: ...Phosphine Complexes of Scandium Scheme 2.4  9'  9  We also performed an NMR tube reaction between 9 (or rather 9 and 9', since the two species are inseparable) and a large excess of PMe3.  The reaction of Sc(r| 5  C H5)(BH4)[N(SiMe2CH PPr 2)2] (9) with PMe3 (-100 equiv) was also followed by "B^H} i  5  NMR  2  spectroscopy (Figure 2.16).  i ••i i i i  -25 Figure 2.16  -30  i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—j—  -35  -40  ppm  Variable temperature 96.3 MHz H B ^ H }  -45  -50  NMR spectra for the reaction between  Sc(Ti5-C5H )(BH4)[N(SiMe2CH PPr 2)2] (9) and PMe (-100 equiv) in C6D . i  5  2  68  3  6  References begin on page 94  Chapter 2: ...Phosphine Complexes of Scandium The spectra consist of a large singlet due to the borohydride complex 9 at -27.6 ppm and a doublet at -47.8 ppm ( ^ B - P = 48 Hz). If the spectrum is collected with proton coupling, this doublet appears as a quartet of doublets  ( /B-H 3  = 86 Hz). This pattern is due to PMe3»BH3. As  the temperature is increased, the relative area of the peak due to PMe3»BH3 increases. This indicates that an equilibrium is present between Sc(T] -C5H5)(BH4)[N(SiMe2CH2PPr2)2] (9), 5  i  PMe , PMe «BH and the putative hydride Sc(t|5-C5H )H[N(SiMe2CH PPr 2)2] (12) (Scheme i  3  3  3  5  2  2.5).  Scheme 2.5  The equilibrium data and the van't Hoff plot for the above equilibrium are presented in Table 2.6 and Figure 2.17.  From the van't Hoff plot, the following thermodynamic parameters were  obtained: AH° = 9 ± 2 kcal mol" and AS" = 10 ± 8 cal K 1  69  _ 1  moH.  References begin on page 94  Chapter 2: ...Phosphine Complexes of Scandium  1/T Figure 2.17 van't Hoff plot for the equilibrium between Sc(rj -C5H5)(BH4)[N(SiMe2CH2PPr 2)2] 5  i  (9), PMe , PMe »BH and Sc(ri5-C H )H[N(SiMe2CH2PPr )2] (12) (r = 0.997). i  3  3  3  5  5  2  Table 2.6 van't Hoff data for the equilibrium between Sc(Ti -C5H5)(BH4)[N(SiMe2CH2PPr 2)2] 5  i  (9), PMe , PMe »BH and Sc(ri -C5H )H[N(SiMe CH2PPr 2)2] (12). 5  3  3  i  3  5  2  temp, K  1/T  K  293  0.00341  3.96 x IO"  -10.14  313  0.00319  1.27 x IO"  -8.97  328  0.00305  1.79 x 10-  -8.63  341  0.00293  3.62 x IO"  -7.92  InK 5  4  4  4  From the equilibrium data, it appears unlikely that a scandium hydride can be isolated via this route. Even with a 100-fold excess of PMe at 68 °C, the equilibrium still favors the starting 3  borohydride complex. We should also note that by studying the equilibrium by B{ H} NMR 11  70  1  References begin on page 94  Chapter 2: ...Phosphine Complexes of Scandium  spectroscopy, we only have indirect evidence for the formation of the scandium hydride 12. Although the observation of PMe3»BH3 implies the existence of this species, we were not able to observe the scandium-hydride signal in the *H spectrum due to quadrupolar broadening; however new ligand and Cp resonances were evident (see Experimental Section 2.16.3.15). The scandium hydride containing product(s) may be monomelic, dimeric, stabilized by additional PMe3 interactions or some combination of all three.  2.11 Scandium Dialkyl Complexes Utilizing the ri5.C H3-l,3-(SiMe CH2PPr 2)2 Ligand i  5  2  System Sections 2.7 to 2.10 present the synthesis, characterization and reactivity of scandium alkyls ligated by a tridentate [NtSift/h^CF^PPr^h]" ligand and a cyclopentadienyl group. Although the reaction chemistry with molecules such as amines is more controlled, reactivity at the scandium amide bond is still favored. In a further effort to understand the reaction pathways of scandium dialkyl complexes, the [N(SiMe2CH2PPr2)2]" ligand system may be replaced by the i  second generation ligand system [T| -C5H3-l,3-(SiMe2CH2PPr 2)2] . 5  i  _ 63  R  E In this ligand, the amide in [N(SiMe CH PPr 2)2]- is replaced with a Cp ring to generate a i  2  2  coordination environment around the metal centre devoid of an amide linkage. In this way, the  71  References begin on page 94  Chapter 2: ...Phosphine Complexes of Scandium  effect of substituting a cyclopentadienyl group (pK = 15) for an amide group (pK = 25.8) on the a  a  reactivity and decomposition of scandium dialkyls can be systematically investigated. The remaining portion of this chapter deals with the study of scandium dialkyls E stabilized by the tridentate Cp-based [Ti5-C5H -l,3-(SiMe2CH2PPr 2)2]- ligand system. i  3  2.12.1 Synthesis of ScCl [ri5.C5H3-l^-(SiMe2CH2PPri2)2] (13) 2  If a reactor bomb, charged with the lithium salt Li[ri -C5H3-l,3-(SiMe2CH2PPr 2)2], 5  i  ScCl3(THF)3 and toluene, is heated at 100 °C for 48 h, good yields of the monomelic dichloride ScCl [T|5-C5H3-l,3-(SiMe2CH2PPr 2)2] (13) are obtained (equation 2.7). i  2  Lih -C H -1,3-(SiMe CH PPr ) ] 5  5  3  2  2  2  2  + ScCI (THF) 3  toluene 3  (- LiCl)  [2.7]  •ScPr'g  4V» ci C l  •P P r J 2  Cl  13  Compound 13 can be isolated as a slightly yellow, waxy solid after filtration and solvent removal. Slow evaporation of a toluene solution yields large colorless crystals of the dichloride. The ^II NMR spectrum consists of two sets of ligand resonances due to the 1,3- substitution of the Cp ring. The H NMR spectrum displays two cyclopentadienyl proton resonances in a 1:2 l  72  References begin on page 94  Chapter 2: ...Phosphine Complexes of Scandium  ratio, while the remaining ligand peaks are consistent with a compound having C symmetry. s  The P{ H} NMR spectrum exhibits a broad singlet at 2.0 ppm (A1/2 = 550 Hz) due to coupling 31  l  to the quadrupolar scandium-45 nucleus. In contrast to dichloro 1, the coordination of THF to the metal is not observed in the *H NMR spectrum, the microanalysis or the X-ray crystallographic analysis. ScCl2[r| -C5H3-l,3-(SiMe2CH2PPr 2)2] (13) tenaciously holds on to 5  i  LiCl (see Section 2.16.3.16). As for the dialkyl complexes 3 and 4, repeated crystallizations decrease the salt content but also result in lower yields. The isolation of crystals of 13 allowed us to investigate its solid state structure.  2.12.2 Structure of ScCl [ri -C5H3-l,3-(SiMe2CH2PPri2)2] (13) 5  2  Evaporation of a toluene solution of 13 yielded large colorless blocks that were used for X-ray crystallography. The molecular structure and numbering scheme are presented below (Figure 2.18). The X-ray crystal structure shows that dichloride 13 is monomeric in the solid state.  As seen for Sc(r| 5- C H5)Cl[N(SiMe2CH PPr 2)2 ] (6) and i  5  2  Sc(Tl 5-C H )Ph5  5  [N(SiMe2CH2PPr2)2] (8), the coordination of a large cyclopentadienyl ring to the scandium i  centre forces the metal to approach a square pyramidal geometry. The angles defined by the apical Cp moiety and the other members of the pyramid range from 106.1° to 116.6°. The larger Ti -C5H3-l,3-(SiMe2CH2PPr 2)2 ligand causes the scandium nucleus to sit farther into the 5  i  coordination cavity; the phosphine donors must "reach back" to bind to the metal (P(l)-Sc(l)P(2) = 146.18(3)) in contrast to ScCl2(THF)[N(SiMe CH PPr 2)2] (1), in which the metal i  2  2  perches out of the cavity. The scandium-phosphorus bond lengths are similar to those in chloride 6 and phenyl 8. The dianionic ligand system [Ti -C3H5-SiMe2N-r<?rr-butyl-3-CH2CH NMe2] 5  2  2  offers [(Cp  NMe  scandium a similar environment to SiNR)ScH] . 2  64  13  in the  complex lR-trans-VS-  The scandium-Cp distance in this bridging hydride (2.1635(18) A) is  comparable to that in 13 (2.17 A). Bond angles and lengths of interest are listed below (Table 2.7).  73  References begin on page 94  Chapter  2: ..Phosphine  Complexes  of  Scandium  Figure 2.18 Molecular structure of ScCl2h -C H3-13-(SiMe CH2PPr 2)2] (13); 33% 5  i  5  2  probability thermal ellipsoids are shown. Hydrogen atoms are omitted for clarity.  Table 2.7 Selected bond lengths (A) and angles (deg) for ScCl [ri -C5H3-1,3-(SiMe CH PPri2)2] (13). 5  2  2  2  P(l)-Sc(l)  2.813(1)  P(2)-Sc(l)  2.836(1)  Cl(l)-Sc(l)  2.384(1)  Cl(2)-Sc(l)  2.411(1)  Cp-Sc(l)  2.17  P(l)-Sc(l)-P(2)  146.18(3)  Cl(l)-Sc(l)-Cl(2)  130.99(4)  P(l)-Sc(l)-Cp  106.1  P(2)-Sc(l)-Cp  106.8  Cl(l)-Sc(l)-Cp  116.6  Cl(2)-Sc(l)-Cp  112.4  74  References begin on page 94  Chapter  ...Phosphine  2:  Complexes  of  Scandium  2.13 Synthesis of Scandium Dialkyls ScR [ri -C5H3-l,3-(SiMe2CH2PPr 2)2] (R = 5  i  2  C H C M e (14), CH SiMe (15)) 2  3  2  3  The synthesis of scandium dialkyl derivatives proceeds only with bulky alkyl groups. In contrast to systems involving the [N(SiMe2CH2PPr )2]" ligand set, the reaction of dichloro 13 i  2  with small alkyl lithium (MeLi, EtLi, Bu Li) or Grignard reagents (MeMgCl, MeMgBr, EtMgBr) l  leads to decomposition or mixtures of mono- and bis-alkylated species. Reaction with the larger trimethylsilylmethyl- and neopentyllithium reagents (2 equiv) results in stable dialkyl species (equation 2.8).  + 2 RLi  ScC^^^CsHa-I.S-CSiMesCHsPPr^g] 13  (-2 LiCl)  ScR [r| -C5H3-1,3-(SiMe CH2PPr 2)2] 5  l  2  2  14 : R = C H C M e 15 : R = CH SiMe 2  2  [2  .8]  3  3  Bis(neopentyl) 14 and bis(trimethylsilymethyl) 15 are isolated as air- and moisturesensitive yellow oils which resist crystallization. If a slight excess of the alkylating reagent is used, the products are contaminated with small amounts of Li[rj -C5H3-l,3-(SiMe2CH2PPr 2)2]5  1  The presence of excess LiR (R = CH2CMe3, CH2SiMe3) promotes the formation of the lithium salt and  SCR3.  This is reminiscent of the decomposition of ZrMe3[r| -C5H3-l,35  (SiMe2CH2PPr )2] when exposed to a small excess of MeLi. i 2  65  The *H NMR spectra of 14 and  15 reflect C symmetric structures. For bis(neopentyl) 14, the CH2CMe3 protons appear as a s  singlet at 1.36 ppm, whereas in bis(trimethylsilymethyl) 15, the CFf2SiMe3 resonance is partially obscured by the ligand SiMe protons at 0.30 ppm. The P{ H} NMR spectrum of both 31  1  2  dialkyls consists of a singlet at —1.0 ppm.  75  These peaks are much sharper (A1/2 = 40 Hz) than  References  begin  on page  94  Chapter  2:  .. .Phosphine  Complexes  of  Scandium  that observed for any other scandium-phosphorus derivatives synthesized in this work. The sharpness of this peak points towards a fluxional process such as phosphine dissociation. This likely involves the exchange of phosphines between 2 species with pseudo-tetrahedral scandium centres. A similar process has been observed for ZrBzCl2[Tl -C5H3-l,3-(SiMe2CH2PPr 2)2], 5  i  65  where the bulky benzyl group does not allow for the simultaneous coordination of both phosphine donors (Scheme 2.6). Scheme 2.6  14: R = CH CMe 2  3  15: R = CH SiMe 2  3  2.14 Reactivity of ScR2[Ti -C H3-l,3-(SiMe2CH2PPri ) ] (R = C H C M e (14), C H C M e 5  5  2  2  2  3  2  3  (15)) Bis(neopentyl) 14 and bis(trimethylsilymethyl) 15 are hydrogenated with H2 (4 arm, 5 days) to give the free alkanes; however, no new organoscandium complex could be isolated or identified.  This is analogous to the dialkyl species Cp*Lu(CH2SiMe3)(CH(SiMe3)2) and  Cp*Lu(CH2CMe3)2 which undergo similar decompositions with H2.  48  Of the handful of  mono(ligand) lanthanide dialkyl complexes in the literature, these appear to be the only, ones whose reaction chemistry with small molecules has been investigated. Most of those reported,  76  References  begin  on page  94  Chapter  2:  ...Phosphine  Complexes  of  Scandium  such as Cp*La(CH(SiMe3)2)2 and Cp*Ce(CH(SiMe3)2)2 slowly decompose, eliminating 37  CH2(SiMe3)2.  49  Even the lutetium derivatives slowly deteriorate in the solid state, and their  reaction chemistry with small molecules was studied as sealed NMR tube reactions. The decomposition of 14 and 15, stabilized by a Cp and two pendant phosphines, mirrors that of lanthanide dialkyls ligated by a single, bulkier pentamethylcyclopentadienyl group. Finally, the catalytic potential of the dichloride ScCl2[ri -C5H3-l,3-(SiMe2CH2PPr 2)2] 5  i  (13) was investigated. By using MAO as a cocatalyst (500 equiv), an activity of 90 g mol" h" for  the  polymerization of  ethylene  was  observed.  This  is  1  1  in contrast  to  ScCl2(THF)[N(SiMe2CH2PPr 2)2] (1) which shows no activity under identical conditions. We i  assume that the absence of a strongly bound THF molecule in 13 results in the observed activity of this system.  2.15 Conclusions The use of the bulky amidodiphosphine ancillary ligand [N(SiMe2CH2PPr2)2]" has i  allowed for the isolation of a series of mononuclear, dialkyl complexes of Sc(III). In particular, both the diethyl and the bis(trimethylsilylmethyl) complexes 3 and 4 were characterized in solution and in the solid state. No evidence for agostic interactions was observed either in solution or in the solid state. Reactivity studies were attempted, but in all cases studied, decomposition of the dialkyl complexes was observed. It would appear that the ancillary ligand allows for the kinetic stabilization of these 12-electron species but they are poised to decompose in the presence of small molecules. In an attempt to generate more stable scandium-alkyls, a chloride was replaced with a cyclopentadienyl ligand. The phenyl complex 8 was characterized in solution and the solid state. The methyl complex 7 was inert to H2, but found to react with amines via C-H activation to yield phenylamide 10 and tert- butylamide 11 although a competing decomposition to HN(SiMe2CH2PPr 2)2 was observed. The borohydride complex 9 reacts with i  77  References  begin  on page  94  Chapter 2: ...Phosphine Complexes of Scandium PMe to form a putative scandium-hydride complex 12, as studied by  11  3  spectroscopy.  The amide linkage in [N(SiMe2C K^PPr^]"  m  B{ H} NMR 1  a y be replaced with a  cyclopentadienyl group to form the second generation ligand set  [rj -C5H3-l,35  (SiMe2CH2PPr2)2]~; this ligand system supports dialkyls of Sc(III). As for 2 - 4, bis(neopentyl) i  14 and bis(trimethylsilylmethyl) 15 decompose when exposed to small molecules, cc-hydrogen abstraction appears to be a facile decomposition route in many of these species, as is generally the case for mono(ligand) lanthanide dialkyls.  2.16 Experimental Section 2.16.1 Procedures Unless otherwise stated all manipulations were performed under an atmosphere of dry, oxygen-free dinitrogen or argon by means of standard Schlenk or glovebox techniques.  66  The  glovebox used was a Vacuum Atmospheres HE-553-2 model equipped with a MO-40-2H purification system and a -40 °C freezer. H, P{ H}, l  31  1  1 3  C , ^ ^ H } and Sc{ H} NMR 45  1  spectroscopy were performed on a Varian XL-300 instrument operating at 299.9, 121.4, 75.5, 96.3 and 72.9 MHz, respectively.  7  Li{ H} NMR spectroscopy was performed on a Bruker 1  AMX-500 instrument operating at 194.5 MHz. *H NMR spectra were referenced to internal C6D5H  (7.15 ppm) or C 6 D C D / / (2.09 ppm). 5  2  31  P{!H} NMR spectra were referenced to  external P(OMe)3 (141.0 ppm with respect to 85% H3PO4 at 0.0 ppm).  1 3  C spectra were  referenced to internal C^D^ (128.0 ppm). B { U} NMR spectra were referenced to neat 2?Et3 n  (0.0 ppm). 7  45  l  Sc{ H} NMR spectra were referenced to external 5c(OH)3 in D2O (0.0 ppm). 1  Li{ !H} NMR spectra were referenced to external LiCl in D2O (0.0 ppm). The P COSY NMR 3 1  spectrum was obtained on a Bruker AMX-500 instrument operating at 202.4 MHz. Mass spectral studies were carried out on a Kratos MS 50 using an EI source. Infrared spectra were recorded on a BOMEM MB-100 spectrometer. Solution samples were recorded on a 0.1 mm  78  References begin on page 94  Chapter  2:  ...Phosphine  Complexes  of  Scandium  KBr cell and solid samples were recorded as KBr pellets. Microanalyses (C, H, N, Cl) were performed by Mr. P. Borda of this department. Low carbon analyses may in part be due to the metal-catalyzed formation of silicon carbide, which does not burn completely during combustion analysis.  2.16.2 Materials LiN(SiMe CH2PPr 2)2 and Li[r| -C5H3-l,3-(SiMe2CH2PPr )2] i  27  5  i  2  2  65  were prepared by  published procedures. MeLi (1.4 M solution in ether) was purchased from Aldrich and used as received. Methylalumoxane was purchased from Aldrich as a toluene solution; the solvent was removed under vacuum and the MAO used in the solid form. LiBH4 and NaBFLj. were purchased from Strem and used as received.  H.2NPh and H^NBu were purchased from Aldrich and 1  distilled over sodium benzophenone. PMe3 was purchasedfromAldrich and vacuum transferred over sodium benzophenone. LiNHPh and LiNHBu were prepared by the addition of BuLi to an 1  ether solution of the amine. ScCl3(THF) , EtLi, LiCH SiMe , 5  3  67  2  3  68  NaCp(DME), PhLi and 69  70  LiCH2CMe3 were prepared according to published procedures. 71  Hexanes, toluene, Et20, THF and hexamethyldisiloxane were refluxed over CaH2 prior to a final distillation from either sodium metal or sodium benzophenone ketyl under an Ar atmosphere. Deuterated solvents were dried by distillation from sodium benzophenone ketyl; oxygen was removed by 3 freeze-pump-thaw cycles. CO was purchasedfromPraxair and used as received. C O 2 was purchased from Praxair and subjected to 3 freeze-pump-thaw cycles to remove oxygen followed by transferfroma -50 °C dry ice/ethanol bath to remove residual water. Ethylene was purchased from Praxair and purified either by 3 freeze-pump-thaw cycles or by stirring for 1 h over a toluene solution of methylalumoxane. Hydrogen was purchased from Matheson and purified by passage through a column of activated 4A molecular sieves and MnO supported on vermiculite.  79  References  begin  on page  94  Chapter 2: ...Phosphine Complexes of Scandium  2.16.3 Syntheses 2.16.3.1 ScCl2(THF)[N(SiMe CH PPr 2)2] (1) i  2  2  To a slurry of ScCl3(THF)3 (992 mg; 2.70 mmol) in toluene (10 mL) was added a toluene solution (10 mL) of LiN(SiMe2CH2PPr 2)2 (985 mg; 2.48 mmol). The head space was i  evacuated and the reaction vessel was then heated at 100 °C for 24 h. The reaction mixture was cooled and passed through a frit lined with Celite to remove LiCl. The solvent was removed in vacuo to yield a yellow waxy solid. The residue was taken up in a minimum amount of toluene (approx. 4 mL).  Slow evaporation of the solvent afforded large colorless plates (1.193 g; 83%  yield). H NMR (C D ): 8 4.08 (t, 4H, 0 ( C / / C H ) l  6  6  2  C//Me , / - H = 7.3 Hz, 3  2  2  H  /  1.10 (dd, 24H, CUMeMe', 0.45 (s, 12H, S\Me ). 2  mle 507 (M+ - THF).  31  H  3  - P  2  2>  V H - H = 6.5 Hz), 1.91 (d of sept, 4H,  = 4.8 Hz), 1.36 (t, 4H, 0 ( C H C / / ) , 2  / -H = H  7.3 Hz,  3  7 -P H  2  2  JR-H  3  6.5 Hz), 1.20 and  =  = 7.0 Hz), 1.05 (d, 4H, C// P,  9.7 Hz),  !  6  6  The THF can be removed by pumping in vacuo. *H NMR (C6D ) for 6  i  2  /H-P=  P{ H} NMR (C D ): 8 1.8 (480 Hz peak width at half height). MS:  ScCl2[N(SiMe CH PPr )2]: 8 1.59 (d of sept, 4H, C//Me , / - H = 6.9 Hz, 2  2  2  3  2  2  1.04 and 1.02 (dd, 24H, CUMeMe', 7 -H = 6.9 Hz, 3  H  3  7 -P H  H  /  - P =  H  1.5 Hz),  = 7.0 Hz), 0.47 (d, 4H, C// P, 2  3.7 Hz), 0.29 (s, 12H, SiMe ). l p { l H } NMR ( C D ) : 3  2  2  6  6  8 -4.9.  2  /  H  - P =  Anal. Calcd. for  Ci8H4 Cl NP ScSi (THF-free material): C, 42.51; H, 8.72; N, 2.75. Found: C, 42.33; H, 8.65; 4  2  2  2  N, 2.58.  2.16.3.2 ScMe [N(SiMe CH PPri ) ] (2) 2  2  2  2  2  To a hexanes solution (5 mL) of ScCl (THF)[N(SiMe CH PPr ) ] (252 mg; i  2  2  2  2  2  0.434 mmol) was added MeLi (0.65 mL of a 1.4 M Et 0 solution, diluted with 5 ml ether; 0.910 2  mmol) dropwise. The reaction mixture was then stirred overnight, after which time it was filtered through a frit lined with Celite to remove LiCl. The solvent was removed in vacuo to  80  References begin on page 94  Chapter 2: ...Phosphine Complexes of Scandium yield an orange-yellow oil (197 mg; 97% yield). Purification by recrystallization was not possible either due to the compound's extreme solubility in hydrocarbon solvents or because it was low melting; as a result, characterization by MS or elemental analysis was not possible. *H NMR (C D ): 8 1.84 (d of sept, 4H, C//Me , / H - H = 7.3 Hz, / - P = 5.6 Hz), 1.12 and 1.08 3  6  6  2  2  (dd, 24H, CHMeMe',  3  / - H = 7.3  Hz,  H  3  H  8.0 Hz), 0.80 (d, 4H, C// P,  /H-P=  2  2  ^H-P=  6.8 Hz), 0.31  (s, 12H, SiMe ), 0.27 (t, 6H, ScMe , / . p = 2.1 Hz). P{!H} NMR (C5D6): 8 -2.0 (160 Hz 3  2  2  31  H  peak width at half height).  2.16.3.3 ScEt [N(SiMe CH PPri ) ] (3) 2  2  2  2  2  To a toluene solution (10 mL) of ScCl (THF)[N(SiMe CH PPri2) ] (569 mg; 2  2  2  2  0.980 mmol) was added EtLi (74 mg; 2.06 mmol) in toluene (10 mL). The reaction mixture was left to stir overnight, and filtered through a frit lined with Celite to remove LiCl, followed by solvent removal in vacuo. The residue was taken up in a minimum amount of hexamethyldisiloxane (approx. 5 mL). Subsequent slow evaporation resulted in the formation of colorless prisms (335 mg; 69% yield). *H NMR (C D ): 8 1.86 (d of sept, 4H, CHMoz, / H - H = 3  6  6  7.5 Hz, / - P = 4.7 Hz), 1.82 (t, 6H, ScCH Me, / - H = 8.4 Hz), 1.13 and 1.08 (dd, 24H, 2  3  H  2  H  CHMeMe', 7 - H = 7.5 Hz, / - P = 6.6 Hz), 0.81 (d, 4H, C// P, / H - P = 6.6 Hz), 0.55 (t of q, 4H, 3  3  2  H  H  2  ScC// Me, / - H = 8.4 Hz, / - P = 4.0 Hz), 0.31 (s, 12H, SiMe ). 3  2  3  H  H  2  1 3  C NMR (C5D6): 8 23.25  (mult, AC, CHMe ), 19.78 and 19.20 (mult, 8C, CHMeMe'), 12.56 (mult, 2C, CH P), 8.66 (mult, 2  2C, ScCH Me, 2  Hz).  31  1  /  C  2  - H =  125 Hz), 6.25 (mult, 4C, SiMe ), 6.00 (mult, 2C, ScCH Me, 2  2  1  /c-H=  119  P{lH} NMR (QD6): 8 -0.8 (230 Hz peak width at half height). MS: mle 496 (M+).  Anal. Calcd. for C H54NP ScSi (+ 0.5 equiv LiCl): C, 51.12; H, 10.53; N, 2.71. Found: C, 22  2  2  51.17; H, 10.58; N, 3.11.  81  References begin on page 94  Chapter 2: ...Phosphine Complexes of Scandium  2.16.3.4 SctCH^iMeshLNtSiMe^H^Pr ^] (4) 1  A solution of Me3SiCH2Li (164 mg; 1.74 mmol) in toluene (10 mL) was added to a toluene solution (15 mL) of ScCl2(THF)[N(SiMe CH PPr 2)2] (500 mg; 0.862 mmol). The i  2  2  reaction mixture was stirred overnight, after which it was filtered through a frit lined with Celite to remove LiCl. The solvent was then removed under vacuum and the residue taken up in a minimum amount of hexamethyldisiloxane (approx. 5 mL).  Slow evaporation of the solvent  resulted in the formation of colorless prisms (380 mg; 72% yield). *H NMR (QD6): 8 1.96 (d of sept, 4H, CHM&z, / - H = 7.2 Hz, / - P = 4.8 Hz), 1.18 and 1.14 (dd, 12H, CUMeMe', / - H 3  2  3  H  H  H  = 7.2 Hz, / - p = 5.8), 0.84 (d, 4H, CH P, / H - P = 12.0 Hz), 0.41 (s, 18H, CH SiMe?3), 0.35 (s, 3  2  H  2  2  12H, SiMe ), 0.21 (t, 4H, C// SiMe , / - P = 5.0 Hz). P{!H} NMR (C D ): 8 -0.7 (90 Hz 3  2  2  3  31  H  6  6  peak width at half height). MS: mle 524 (M+ - CH SiMe3). Anal. Calcd. for C 6H66NP ScSi4 2  2  2  (+ 0.80 equiv LiCl): C, 48.34; H, 10.30; N, 2.17. Found: C, 48.57; H, 10.48; N, 2.28. After an additional recrystallization: Anal. Calcd. for C 6H66NP ScSU (+ 0.58 equiv LiCl): C, 49.05; H, 2  2  10.45; N, 2.20. Found: C, 49.06; H, 10.28; N, 2.36. After 3rd recrystallization: Anal. Calcd. for C 6H66NP ScSi4 (+ 0.43 equiv LiCl): C, 49.55; H, 10.55; N, 2.22; CI, 2.42. Found: C, 2  2  49.96; H, 10.27; N, 2.38; CI, 2.45.  2.16.3.5 (Et 0)Li(u,-CI)(u.-Me)ScMe[N(SiMe CH PPri ) ]LiCl(Et 0)(5) 2  2  2  2  2  2  MeLi (0.49 mL of a 1.4 M Et 0 solution, diluted with 5 mL ether; 0.692 mmol) 2  was added dropwise to a toluene solution (5 mL) of ScCl (THF)[N(SiMe CH PPr ) ] (200 mg; i  2  2  2  2  2  0.345 mmol). The reaction mixture immediately became cloudy and slowly darkened. The mixture was stirred overnight, after which time it was filtered through a frit lined with Celite to remove LiCl. The solvent was then removed in vacuo to yield an orange-brown oil. Repeated crystallizations resulted in colorless crystals (72 mg; 30% yield). P{ B) NMR (QDg): 8 -3.3, 31  82  l  References begin on page 94  Chapter 2: ...Phosphine Complexes of Scandium -8.6 (ABX mult (dd of 1:1:1:1 q), / . = 12 Hz, / . = 53 Hz), -3.9, -6.7 (ABX mult (dd of 2  1  P  P  P  L i  1:1:1:1 q), / _ = 6 Hz, / . i = 54 Hz). ^Li^U) NMR (C6D ): 8 0.1 (t, P Li, I f a - P = 53 Hz), 2  1  P  -1.0 (br s).  45  P  P  L  6  2  Sc{ H} NMR (C6D ): 8 -220 (br s, 5.1 KHz peak width at half height). Anal. 1  6  Calcd. for C28H7oCl2Li N02P2ScSi (+ 1.1 equiv LiCl): C, 45.00; H, 9.44; N, 1.87. Found: C, 2  2  45.03; H, 9.36; N, 2.12.  2.16.3.6 Reaction of Sc(CH SiMe ) [N(SiMe CH PPr ) ] (4) with CO i  2  3  2  2  2  2  2  Method 1: In the glovebox, approx. 20 mg of 4 was dissolved in 0.5 mL C6D6 and placed in a sealable NMR tube equipped with a Kontes-valve needle adapter. The NMR tube assembly was attached to a second sealable NMR tube by means of a glass bridge. The apparatus was then taken onto the line and the solution of 4 was subjected to 3 freeze-pump-thaw cycles. The solution was exposed to 1 atm CO under which it was left to stand for 12 h. The volatiles were condensed at liquid N temperature into the second NMR tube under static 2  vacuum. The second NMR tube containing the volatiles was then flame-sealed. C6D6 was then added to the first NMR tube via syringe and the tube flame-seale. The contents of both tubes were analyzed by H and P{ *H} NMR spectroscopy. l  31  Method 2: In the glovebox, approx. 100 mg of 4 was dissolved in 10 mL toluene and placed in a small reactor bomb, which was then attached to a second reactor bomb by means of a glass bridge. The apparatus was taken onto the line and the solution of 4 was subjected to 3 freeze-pump-thaw cycles. The solution was exposed to 1 atm CO under which it was left to stand for 12 h. The volatiles were then condensed at liquid N2 temperature into the second bomb under static vacuum. The contents of the second bomb were analyzed by GC-MS and solution IR spectroscopy. Toluene was then added to the first bomb via syringe and the contents examined by GC-MS and solution IR spectroscopy.  83  References begin on page 94  Chapter 2: ...Phosphine Complexes of Scandium  2.16.3.7 Reaction of ScEt2[N(SiMe CH PPr 2)2] (3) with C 0 i  2  2  2  In the glovebox, 184 mg of 3 was dissolved in 10 mL toluene and placed in a small reactor bomb. The bomb was attached to the line by means of a constant volume glass bulb (9.1 mL) equipped at either end with a Kontes needle valve. The yellow solution of 3 was subjected to 3 freeze-pump-thaw cycles. With the solution frozen, the apparatus was evacuated and C 0 introduced to the glass bulb. The bulb was then closed to the C 0 source and opened to 2  2  the evacuated bomb, and the C 0 condensed into the bomb over a period of 30 min. The bomb 2  was then sealed and left to stir overnight, during which time the solution turned colorless. The toluene was removed in vacuo to yield a slightly yellow oil. This reaction was repeated using varying amounts of 3 and/or glass bulbs of differing volume. *H NMR (C6D6): 5 3.26 (q, 4H, ScC0 C// Me, 2  2  3  /  H  -H  = 7.0 Hz), 1.47 (d of sept, 4H, C//Me , / H - H = 7.1 Hz, 3  2  2  /  H  -P  = 1.5 Hz),  1.11 (t, 6H, ScC0 CH Me, / H - H = 7.0 Hz), 0.91 and 0.90 (dd, 24H, CHMeMe', / - H = 7.1 Hz, 3  2  3  2  H  3/ _p = 4.5 Hz), 0.30 (d, 4H, C// P, / - p = 3.0 Hz), 0.08 (s, 12H, SiMe ). 2  H  2  H  2  2.16.3.8 Sc(ti5-C H )Cl[N(SiMe CH PPr 2)2] (6) i  5  5  2  2  Method 1: To a solution of ScCl (THF)[N(SiMe CH PPr ) ] (572 mg; 0.985 i  2  2  2  2  2  mmol) in toluene (10 mL) was added solid NaCp(DME) (176 mg; 0.985 mmol). The reaction mixture was stirred at room temperature for 1 week. The reaction mixture was then passed through a frit lined with Celite to remove NaCl. The solvent was removed in vacuo to yield a yellow waxy solid. The residue was taken up in a minimum amount of toluene (approx. 5 mL). Slow evaporation of the solvent afforded large colorless plates (358 mg; 68% yield). Method 2: 10 mL of a THF solution of NaCp(DME) (400 mg, 2.25 mmol) was added dropwise to a THF solution (10 mL) of ScCl2(THF)[N(SiMe2CH PPr 2) ] (1.270 g; 2.19 i  2  2  mmol) maintained at -78 °C. This was stirred overnight and left to warm to room temperature.  84  References begin on page 94  Chapter 2: ...Phosphine Complexes of Scandium  The THF was then removed in vacuo and the residue extracted with toluene (3 x 10 mL) and passed through a frit lined with Celite to remove NaCl. The solvent was removed in vacuo to yield a white waxy solid, which resulted in colorless plates upon crystallization from toluene (655 mg; 56% yield). H NMR (C D ): 5 6.51 (t, 5H, C5//5, / - P = 1.3 Hz), 2.15 and 1.78 (d l  3  6  of sept, 4H, CHMe ,  3  2  CHMeMe', P  3  /  H  -H  /H-H  = 7.7 Hz,  3  /  6  = 7.7 Hz, H  - P =  H  /  2  - P =  H  3.4 Hz), 1.34, 1.14, 1.11 and 1.06 (dd, 24H,  6.5 Hz), 0.88 and 0.65 (dd, 4H, C// P, 2  = 7.8 Hz), 0.28 and 0.24 (s, 12H, SiMe ).  31  2  2  = 14.5 Hz, / 2  /H-H  H  P{ H} NMR (C D ): 8 -1.2 (425 Hz peak width 1  6  6  at half height). MS: mle 538 (M+). Anal. Calcd. for C23H4 ClNP ScSi2: C, 51.33; H, 9.18; N, 9  2  2.60. Found: C, 51.40; H, 8.98; N, 2.83.  2.16.3.9 Sc(r|5.C5H )Me[N(SiMe CH2PPr 2)2] (7) i  5  2  To a toluene solution (5 mL) of Sc(Ti -C5H5)Cl[N(SiMe2CH2PPri2)2] (302 mg; 5  0.561 mmol) was added MeLi (0.40 mL of a 1.4 M Et20 solution, diluted with 5 mL ether; 0.560 mmol), dropwise. The reaction mixture became cloudy and was stirred overnight, after which time it was filtered through a frit lined with Celite to remove LiCl. The solvent was removed in vacuo to yield an orange-yellow waxy solid (187 mg; 65% yield). Because the product was low melting, elemental analysis was not possible. H NMR (C^D^): 8 6.49 (t, 5H, C5//5, / H - P =1-1 l  Hz), 2.04 and 1.79 (d of sept, 4H, CHMe ,  3  2  3  / - H = 7.4 Hz, / - P = 4.1 Hz), 1.22, 1.15, 1.13 and 2  H  H  1.06 (dd, 24H, CHMeMe', / - H = 7.4 Hz, / - P = 7.0 Hz), 0.83 and 0.69 (dd, 4H, CH P, 3  3  H  = 14.0 Hz, 31  2  /H-P  H  2  = 4.0 Hz), 0.31 and 0.28 (s, 12H, SiMe ), -0.19 (t, 3H, ScC// , 2  3  3  /  H  - P =  2  /H-H  4.3 Hz).  P{ H) NMR (C6D ): 8 -2.0 (135 Hz peak width at half height). MS: mle 502 (M+ - Me). l  6  85  References begin on page 94  Chapter 2: ...Phosphine Complexes of Scandium  2.16.3.10 Reaction of ScR2[N(SiMe CH PPr 2)2] (R = Me (2), Et (3)) and i  2  2  Sc(Ti -C H5)Me[N(SiMe CH2PPr 2)2] (7) with C H 5  i  2  5  2  4  Method 1: Solutions of 2, 3 and 7 were each placed in a bomb (10 mg catalyst, 10 mL solvent) and the bomb attached to a glass bridge. The solutions were subjected to 3 freeze-pump-thaw cycles. Ethylene was introduced to a bomb approx. twice the volume of the bomb containing the catalyst solution, and freeze-pump-thawed 3 times. The catalyst solution was then frozen and the ethylene condensed onto it. The bomb was sealed and the contents left to warm up with stirring (final pressure ~2 atm). After 24 h, white insoluble material was present. The polymerization was quenched with methanol, and the polyethylene removed, dried and weighed. The activity was calculated from the amount of catalyst used (mmol), the time of the polymerization (h) and the weight of polymer isolated (g).  For 2:  10 mg of  ScMe2[N(SiMe2CH2PPr )2] (0.0214 mmol) resulted in 62 mg polyethylene in 24 h, equal to an i  2  activity of 120 g mol" h" . For 3: 10 mg of ScEt [N(SiMe CH PPr ) ] (0.0202 mmol); 27 mg 1  1  i  2  2  2  2  2  polyethylene; 55 g mol" h" . For 7: 10 mg of Sc(r| -C5H5)Me[N(SiMe CH PPr ) ] (0.0193 1  1  5  i  2  2  2  2  mmol); 79 mg polyethylene; 170 g mol" h" . 1  1  Method 2: Approx. 1 mL of a 0.1 M solution of 2 was placed in a sealable NMR tube equipped with a Kontes valve. The NMR tube was attached to the line and the contents freeze-pump-thawed 3 times. The solution was frozen, the head space evacuated, and then filled with ethylene once the solution had been thawed. The ethylene was condensed and the tube flame-sealed. This resulted in an ethylene pressure of ~2 atm. The ^ C ^ H } NMR spectrum of the contents were taken at 15 minute intervals. Over the course of the experiment (8 h), large amounts of polyethylene were observed to form in the tube. No change in the NMR spectrum was evident.  86  References begin on page 94  Chapter  ...Phosphine  2:  Complexes  of  Scandium  2.16.3.11 Sc(ri5-C H )Ph[N(SiMe CH PPri ) ] (8) 5  5  2  2  2  2  To a toluene solution (10 mL) of Sc(rj5-C H5)Cl[N(SiMe CH2PPr 2)2] (300 mg; i  5  2  0.558 mmol) maintained at -60 °C was added an ethereal solution of PhLi (1.2 mL of a 0.57 M solution; 0.684 mmol). The reaction mixture was left to warm up and stir overnight, and then filtered through a frit lined with Celite to remove LiCl, followed by solvent removal in vacuo. The residue was taken up in a minimum amount of hexamethyldisiloxane (approx. 5 mL). Subsequent slow evaporation resulted in the formation of colorless prisms (104 mg; 32% yield). H NMR ( Q D 6 ) : 8 7.93 (d, 2H, o-Ph, 3 / . = 8.2 Hz), 7.23 (t, lH,/?-Ph, 3 / . = 8.0 Hz), 7.15  l  H  H  H  H  (m, 2H, m-Ph), 6.62 (t, 5H, C5//5, 3 / - P = 1.3 Hz), 1.71 (d of sept, 4H, C//Me , / - H = 7.3 Hz, 3  H  2  /  H  2  = 2.3 Hz), 1.05, 1.03, 0.85 and 0.55 (dd, 24H, CHMeMe',  - P  V  H  - H=  1  3  H  Z  >  H  3  ^H-P=  7.0 Hz),  1.00 and 0.67 (dd, 4H, C / / P , / - H = 13.5 Hz, / - p = 5.0 Hz), 0.42 and 0.32 (s, 12H, SiMe ). 2  2  2  H  H  2  3lp{lH} NMR (C D ): 5 -2.8 (100 Hz peak width at half height). MS: mle 580 (M+). Anal. 6  6  Calcd. for C29H 4NP2ScSi : C, 60.07; H, 9.39; N , 2.42. Found: C, 59.62; H, 9.17; N, 2.08. 5  2  2.16.3.12 Sc(ri5-C5H5)(BH4)[N(SiMe2CH2PPri )2] (9) 2  Method 1: Solid L i B H 4 (37 mg; 1.70 mmol) was added to a solution of S C ( T | 5  C5H )Cl[N(SiMe CH PPr ) ] (200 mg; 0.372 mmol) in toluene (10 mL). The reaction mixture i  5  2  2  2  2  was stirred at room temperature for 1 week, and then passed through a frit lined with Celite to remove LiCl and excess L1BH4. The solvent was removed in vacuo to yield a yellow waxy solid. The residue was taken up in a minimum amount of hexamethyldisiloxane (approx. 5 mL). Slow evaporation of the solvent yielded large colorless plates (45 mg; 23% yield). Method 2:  Sc(rj5-C H )Cl[N(SiMe2CH2PPr 2)2] (130 mg; 0.242 mmol) and i  5  5  NaBFL; (90 mg; 2.38 mmol) were placed in a bomb. 10 mL THF was distilled onto the mixture and left to stir overnight. The solvent was removed under vacuum and the residue extracted into  87  References  begin  on page  94  Chapter 2: ...Phosphine Complexes of Scandium toluene (3x5 mL). The reaction mixture was then passed through a frit lined with Celite to remove NaCl and excess NaBPfy. The solvent was then removed in vacuo to yield a white solid. The residue was taken up in a minimum amount of hexamethyldisiloxane (approx. 2 mL).  Slow  evaporation of the solvent gave large colorless plates (83 mg; 66% yield). *H NMR (C6D6): 8 6.41 (t, 5H, C5//5,  3  /  H  - P =  1.3 Hz), 1.99 and 1.80 (d of sept, 4H, O/Mes,  = 5.1 Hz), 1.23, 1.11, 1.08 and 1.03 (dd, 24H, CUMeMe', and 0.65 (dd, 4H, CH P, 2  31  2  /H-H  = 14.5 Hz,  2  /  H  - P =  /H-H  = 7.0 Hz,  2  /  H  - P  / - H = 7.0 Hz, / - P = 6.0 Hz), 0.90  3  3  H  H  4.8 Hz), 0.31 and 0.28 (s, 12H, SiMe ). 2  P{ !H} NMR (C D ): 8 0.5 (200 Hz peak width at half height). 6  3  H  6  B NMR (Cgl^, 20 °C): 8  -27.6 (major product (-90%), quint, BH4, JB-H = 85 Hz, 32 Hz peak width at half height), -51.8 1  (minor product (-10%), mult, P-BH3), (C7D8, -80 °C): 8 -27.0 (br s, BH4, 160 Hz peak width at half height). B { *H} NMR (C6D ): 8 -27.6 (major product, s, BU4, 32 Hz peak width at half n  6  height), -51.8 (minor product, d, P-.BH3, ^ B - P = 60 Hz, 50 Hz peak width at half height). IR (KBr disc): V - H = 2478 (Ht), 2211 cm" (H ), (hexanes solution): V - H = 2476 (Ht), 2212 cm1  B  b  B  1 (H ). MS: mle 516 (M+ - H). Anal. Calcd. for C23H 3BNP ScSi2: C, 53.38; H, 10.32; N, b  2  5  2.71. Found: C, 53.79; H, 10.40; N, 2.86.  2.16.3.13 Sc(Ti5-C H )NHPh[N(SiMe2CH PPr 2)2] (10) i  5  5  2  Method 1: To Sc(Ti5-C H5)Me[N(SiMe2CH2PPr 2) ] (92 mg; 0.178 mmol) in i  5  2  toluene (10 mL) was added a toluene solution (10 mL) of aniline (23 mg, 0.247 mmol). The yellow color of the solution gradually dissipated and the solution became cloudy. The reaction mixture was stirred overnight, and the contents of the flask filtered through Celite to remove a white precipitate, followed by solvent removal. 10 was isolated as an oil; the presence of HN(SiMe2CH PPri2)2 (by NMR spectroscopy) did not allow for a pure sample to be obtained 2  (30 mg; 28% yield (estimated)).  88  References begin on page 94  Chapter  ...Phosphine  2:  Complexes  of  Scandium  Method 2: An ethereal solution (5 mL) of LiNHPh (14 mg, 0.141 mmol) was slowly added by cannula to a toluene solution (10 mL) of Sc(T| -C5H5)Cl[N(SiMe2CH2PPr 2)2] 5  i  (75 mg, 0.139 mmol) maintained at -78 °C. The solution was allowed to warm to room temperature, during which time it became cloudy. The solution was filtered to remove LiCl and the solvent removed in vacuo to yield a yellow oil (25 mg; 30% yield (estimated)). *H NMR (C D ): 6 7.20 (t, 2H, m-Ph, 6  6  3  /H-H=  6.0 Hz), 6.78 (d, 2H, o-Ph,  3  /  H  - H  = 6.0 Hz), 6.66 (t, IH, o-  Ph, / - H = 6.0 Hz), 5.55 (br s, IH, NH), 6.48 (t, 5H, C5//5, / - P = 1.5 Hz), 1.87 and 1.73 (d of 3  3  H  H  sept, 4H, C#Me , / H - H = 6.8 Hz, / - p = 1.5 Hz), 1.13, 1.09, 0.99 and 0.92 (dd, 24H, 3  2  2  CHMeMe',  3  /  H  - H  H  = 6.8 Hz, / H - P = 6.5 Hz), 0.75 and 0.68 (dd, 4H, CH P, 3  /H-H=  13.4 Hz, / 2  H  = 4.3 Hz), 0.30 and 0.28 (s, 12H, SiMe ). P{ H] NMR (C D6): 8 -1.0 (80 Hz peak width at 31  P  2  2  l  2  6  half height).  2.16.3.14 Sc(Ti5-C H5)NHBu [N(SiMe2CH2PPr 2)2] (11) i  t  5  Method 1: To Sc(T|5-C H5)Me[N(SiMe2CH PPr 2)2] (90 mg; 0.174 mmol) in i  5  2  toluene (10 mL) was added a toluene solution (10 mL) of rerr-butylamine (17 mg, 0.232 mmol). The yellow color of the solution dissipated and the solution became cloudy. The reaction mixture was stirred overnight, the contents of the flask filtered through Celite to remove a white precipitate, and the solvent removed.  1 1 was isolated as an oil; the presence of  HN(SiMe2CH2PPri2)2 (by NMR spectroscopy) did not allow for a pure sample to be obtained (41 mg; 41% yield (estimated)). Method 2: An ethereal solution (5 mL) of LiNHBu (11 mg, 0.149 mmol) was 1  slowly added by cannula to a toluene solution (10 mL) of Sc(r| -C5H5)Cl[N(SiMe2CH2PPr 2)2] 5  i  (75 mg, 0.139 mmol) maintained at -78 °C. The solution was allowed to warm to room temperature, during which time it became cloudy. The solution was filtered to remove LiCl and the solvent removed in vacuo to yield a yellow oil (20 mg; 25% yield (estimated)). *H NMR  89  References  begin  on page  94  Chapter 2: ...Phosphine Complexes of Scandium ( C D ) : 6 6.48 (s, 5H, C5//5), 4.70 (br s, 1H, NH), 1.80 and 1.71 (d of sept, 4H, C#Me , / H - H 3  6  6  2  = 7.4 Hz, 2/H-P = 3.4 Hz), 1.37 (s, 9H, NMej), 1.15, 1.06, 1.04 and 1,00 (dd, 24H, CUMeMe', 3  2  ^H-H  = 14.8 Hz,  3  /  -P  H  = 7.4 Hz), 0.67 (other peak obscurred, dd, 4H, Ctf P, JR-H  = 10.2 Hz,  2  2  /H-p = 4.8 Hz), 0.42 and 0.35 (s, 12H, SiMe ).  3 1  2  P { H } NMR (C6D ): 8 1.0 (90 Hz peak 1  6  width at half height).  2.16.3.15 Sc(Ti5-C Hs)H[N(SiMe2CH2PPri )2] (12) 2  5  A sealable NMR tube equipped with a Kontes valve was loaded with a 1 mL C7D8 solution of Sc(r|5-C5H5)(BH4)[N(SiMe CH PPr 2) ] (approx. 0.1 M). The NMR tube was i  2  2  2  attached via a glass bridge to a bomb containing PMe3. The contents of the NMR tube were frozen and the head space evacuated. The head space was then opened to the bomb of PMe3 and allowed to fill with vapour. The bomb was closed, and the vapour condensed into the NMR tube. The NMR tube was flame-sealed and carefully allowed to warm to room temperature. This method resulted in the exposure of Sc(ri -C5H5)(BH4)[N(SiMe CH PPr ) ] to ~1 mL 5  i  2  2  2  2  PMe3 (approx. 100 equiv). Due to the large excess of PMe3, the reaction was followed by llB{ lH} NMR spectroscopy. *H NMR (C D ): 8 6.30 (t, 5H, C5//5, / H - P =1.1 Hz), 1.95 and 3  6  6  1.85 (d of sept, 4H, C#Me , / - H = 7.1 Hz, / - P = 4.5 Hz), 1.15, 1.13, 1.12 and 1.03 (dd, 24H, 3  2  2  H  H  CUMeMe', 7 -H= 7.2 Hz, / - P = 10.5 Hz), 0.22 and 0.16 (s, 12H, SiMe ), methylene protons 3  3  H  H  2  obscurred by PMe . P{ U) NMR (C6D ): 8 0.3 (430 Hz peak width at half height). 31  B{ U}  l  3  n  6  l  NMR (C D ): 8 -27.6 (s, Sc(r|5-C H5)(flH4)[N(SiMe CH PPr^) ]), -47.8 (d, P M e « 5 H , 1 / - P 6  5  6  2  2  2  3  3  B  = 48 Hz). HB NMR ( C D ) : 8 -27.6 (quint, Sc(Ti5-C H5)(fiH4)[N(SiMe2CH PPri ) ]), -47.8 (q 6  5  6  2  2  2  of d, PMe3«5H3, V B - P = 48 Hz, / B - H = 86 Hz). The peak at -47.8 ppm increased in area as the 3  temperature was elevated.  90  References begin on page 94  Chapter  2:  ...Phosphine  Complexes  of  Scandium  2.16.3.16 ScCl2[(Tl -C5H3-l,3-(SiMe2CH2PPri )2] (13) 5  2  A large reactor bomb was charged with Li[(rj -C5H3-l,3-(SiMe2CH2PPr 2)2] 5  i  (1.02 g; 2.28 mmol), ScCl3(THF)3 (1.07 g; 2.93 mmol) and 30 mL toluene. The head space was evacuated and the contents heated to 100 °C for 48 h. The reaction mixture was cooled and passed through a frit lined with Celite to remove LiCl. The solvent was removed in vacuo to yield a yellow waxy solid. The residue was taken up in a minimum amount of toluene (approx. 5 mL). Slow evaporation of the solvent afforded large colorless blocks (1.00 g; 79% yield). *H NMR (C D ): 8 7.10 (t, IH, C5//3, 7 -H = 1.5 Hz), 6.72 (d, 2H, C5//3, / - H = 1.5 Hz), 2.11 4  6  6  4  H  H  and 2.05 (d of sept, 4H, C//Me , / H - H = 7.7 Hz, / . p = 7.5 Hz), 1.36, 1.33, 1.28 and 1.12 (dd, 3  2  2  24H, CHMeMe', 2  3  H  / - H = 7.7 Hz, / . p = 8.0 Hz), 1.10 and 0.88 (dd, 4H, CH P, 3  H  H  2  / - P = 10.3 Hz), 0.44 and 0.27 (s, 12H, SiMe ). H  2  31  2  / H - H = 15.2 Hz,  P{!H} NMR (C D ): 8 2.0 (550 Hz peak 6  6  width at half height). MS: mle 557 (M+). Anal. Calcd. for C23H4 Cl2P2ScSi2 (+ 1.00 equiv. 7  LiCl): C, 46.04; H, 7.90. Found: C, 45.74; H, 8.48. After an additional recrystallization: Anal. Calcd. for C23H 7Cl P ScSi2 (+ 0.77 equiv. LiCl): C, 46.80; H, 8.03. Found: C, 46.80; H, 4  2  2  8.47.  2.16.3.17 Reaction of ScCl (THF)[N(SiMe CH PPr' )2] (1), 2  2  2  2  Sc(ri5.C H5)Cl[N(SiMe2CH2PPri2) ] (6) and 5  2  ScCl [(ri -C5H3-l,3-(SiMe2CH2PPri2) ] (13) with C H 5  2  2  2  4  A large Schlenk tube was charged with a toluene solution (10 mL) of methylalumoxane (-500 equiv). The head space was evacuated and ethylene introduced. This caused the solution to smoke and heat up slightly. This was then stirred for 1 h until the ethylene was scrubbed, and no further smoke evolved. A 3 mL toluene solution of the catalyst 1, 6 or 13 (-10 mg) was then added by syringe to the rapidly stirred solution. This was left to stir for 24 h. The next day, white insoluble polyethylene was observed in the flask. THF (30 mL) was added  91  References  begin  on page  94  Chapter  ...Phosphine  2:  Complexes  of  Scandium  to the flask, followed by water, which caused an exothermic reaction as the MAO was quenched. The contents of the flask were then filtered, and the polyethylene collected, dried and weighed. For 1: 10 mg of ScCl2(THF)[N(SiMe CH2PPri2)2] (0.0202 mmol); 0 mg polyethylene; 0 g mol" 2  1  h" .  10 mg of Sc(T)5-C5H5)Cl[N(SiMe2CH PPr 2)2] (0.0185 mmol); 710 mg  For 6:  1  i  2  polyethylene; 1600 g mol" h-1. For 13: 1  10 mg of ScCl [(ri -C5H3-l,3-(SiMe2CH2PPr 2)2] 5  i  2  (0.0179 mmol); 39 mg polyethylene; 90 g mol" h" . 1  1  2.16.3.18 Sc(CH CMe3) [(Tl5.C H3-l,3-(SiMe2CH2PPri2)2] (14) 2  2  5  A solution of M e 3 C C H 2 L i (123 mg; 1.58 mmol) in toluene (10 mL) was added to a toluene solution (10 mL) of ScCl [(ri -C5H3-l,3-(SiMe2CH PPr 2)2] (410 mg; 0.736 mmol). 5  i  2  2  The reaction mixture was stirred overnight, after which it was filtered through a frit lined with Celite to remove LiCl. The solvent was then removed under vacuum to yield a yellow oil. Purification by recrystallization was not possible either due to the compound's extreme solubility in hydrocarbon solvents or because it was low melting; as a result, characterization by MS or elemental analysis was not possible (254 mg; 55% yield). *H NMR (C6D6): 8 7.03 (t, 1H,  C5H3,  4  /  H  - H  = 1-0 Hz), 6.90 (d, 2H, C 5 / / 3 , / - H = 10 Hz), 1.65 (d of sept, 4H, CHMei, 4  H  3  /  H  - H  = 7.2 Hz, 2 / . = 2.0 Hz), 1.36 (s, 18H, CH CMe ), 1.04, 1.03, 0.97 and 0.93 (dd, 24H, H  CUMeMe',  3  /  H  p  - H  2  = 7.2 Hz,  3  / -P= H  7.0 Hz), 0.75 and 0.65 (d, 4H, C// P,  2  /H-H  2  2  and 0.43 (s, 12H, SiMe ), C// CMe not observed. 2  3  3  31  = 17.8 Hz), 0.49  P{lH} NMR (C6D ): 8 -0.9 (40 Hz peak 6  width at half height).  2.16.3.19 Sc(CH SiMe )2[(r|5-C5H3-l,3-(SiMe2CH2PPri2)2] (15) 3  2  A solution of Me3SiCH Li (125 mg; 1.33 mmol) in toluene (10 mL) was added to 2  a toluene solution (10 mL) of ScCl [(r| -C5H3-l,3-(SiMe2CH2PPr 2)2] (340 mg; 0.610 mmol). 5  i  2  92  References  begin  on page  94  Chapter  2:  ...Phosphine  Complexes  of  Scandium  The reaction mixture was stirred overnight, after which it was filtered through a frit lined with Celite to remove LiCl. The solvent was then removed under vacuum to yield a yellow oil. Purification by recrystallization was not possible either due to the compound's extreme solubility in hydrocarbon solvents or because it was low melting; as a result, characterization by MS or elemental analysis was not possible (177 mg; 44% yield). *H NMR (C6D6): 8 7.00 (d, 2H, C5//3, /H-H = 4  1.0 Hz), 6.72 (t, IH,  C5//3, / H - H = 4  1.0 Hz), 1.64 and 1.57 (d of sept, 4H,  C//Me , / H - H = 7.6 Hz, / - P = 3.1 Hz), 0.99, 0.97, 0.93 and 0.91 (dd, 24H, CHMeMe', / - H = 3  2  2  3  H  H  7.6 Hz, / - p = 3.4 Hz), 0.89 and 0.65 (dd, 4H, C// P, 3  H  2  2  / -H H  = 6.2 Hz,  2  /H-P=  6.2 Hz), 0.45 and  0.29 (s, 12H, SiMe ), 0.30 (s, 18H, CH SiMe ), C// SiMe3 not observed. 2  2  2  2  31  P{ H} NMR 1  (C D ): 8-1.0 (40 Hz peak width at half height). 6  6  2.16.3.20 Molecular Orbital Calculations All molecular orbital calculations were performed on the CAChe Worksystem, a product developed by Tektronix. The parameters used in the INDO/1 semiempirical molecular orbital calculations on the model compound ScMe [N(SiMe CH PH ) ] were taken from the 2  literature.  53  2  2  2  2  The bond lengths were taken from the X-ray crystal structure analysis of 3 and the  model was restricted to C  2 v  symmetry. The -PPri groups were replaced with -PH groups and 2  2  the -CH CH3 groups replaced with their methyl analogues. The -PH , - C H and SiMe moieties 2  2  3  2  were all eclipsed to effect maximum symmetry. The Cartesian coordinates of the model can be found in Appendix 2. For the model, the following standard bond lengths were used: P-H, 1.380  A; C-H, 1.090 A.  93  References  begin  on page  94  Chapter 2: ...Phosphine Complexes of Scandium  2.17 References (1)  Arnold, J.; Hoffman, C. G.; Dawson, D. Y.; Hollander, F. J. Organometallics 1993,12, 3645.  (2)  Cloke, F. G. N. Chem. Soc. Rev. 1993,22,17.  (3)  Housecraft, C. E. Coord. Chem. Rev. 1993,127, 131.  (4)  Thompson, M. E.; Bercaw, J. E. Pure Appl. Chem. 1984,56, 1.  (5)  Piers, W. E.; Shapiro, P. J.; Bunel, E. E.; Bercaw, J. E. Synlett 1990, 74.  (6)  Lappert, M. F.; Pearce, R. J. Chem. Soc, Chem. Commun. 1973, 126.  (7)  Hart, F. A.; Massey, A. G.; Saran, M. S. / . Organomet. Chem. 1970,21,147.  (8)  Lappert, M. F.; Singh, A.; Atwood, J. L. / . Chem. Soc, Chem. Commun. 1981, 1190.  (9)  Lappert, M . F.; Singh, A.; Atwood, J. L.; Hunter, W. E. / . Chem. Soc, Chem. Commun. 1983, 206.  (10)  Shapiro, P. J.; Cotter, W. D.; Shaefer, W. P.; Labinger, J. A.; Bercaw, J. E. / . Am. Chem. Soc. 1994,116, 4623.  (11)  Bulls, A. R.; Bercaw, J. E.; Manriquez, J. M.; Thompson, M. E. Polyhedron 1988, 7, 1409.  (12)  Parkin, G.; Bunel, E.; Burger, B. J.; Trimmer, M. S.; van Asselt, A.; Bercaw, J. E. / . Mol. Catal. 1987,41, 21.  (13)  Bunel, E.; Burger, B. J.; Bercaw, J. E. / . Am. Chem. Soc. 1988,110, 976.  (14)  Bercaw, J. E.; Davies, D. L.; Wolczanski, P. T. Organometallics 1986,5,443.  94  References begin on page 94  Chapter 2: ...Phosphine Complexes of Scandium (15)  Thompson, M. E.; Baxter, S. M.; Bulls, A. R.; Burger, B. J.; Nolan, M. C ; Santarsiero, B. D.; Schaefer, W. P.; Bercaw, J. E. / . Am. Chem. Soc. 1987,109, 203.  (16)  Piers, W. E.; Bercaw, J. E. / . Am. Chem. Soc. 1990,112,9406.  (17)  Hajela, S.; Bercaw, J. E. Organometallics 1994,13, 1147.  (18)  Campion, B. K.; Heyn, R. H.; Tilley, T. D. Inorg. Chem. 1990, 29, 4355.  (19)  Campion, B. K.; Heyn, R. H.; Tilley, T. D. / . Am. Chem. Soc. 1990,112, 2011.  (20)  Burger, B. J.; Thompson, M. E.; Cotter, W. D.; Bercaw, J. E. / . Am. Chem. Soc. 1990, 112, 1566.  (21)  Fryzuk, M. D.; Haddad, T. S.; Berg, D. J.; Rettig, S. J. Pure App. Chem. 1991, 63, 845.  (22)  Fryzuk, M. D. Can. J. Chem. 1992, 70, 2849.  (23)  Fryzuk, M. D.; Mylvaganam, M.; Zaworotko, M. J.; MacGillivray, L. R. / . Am. Chem. Soc. 1993,115,10360.  (24)  Fryzuk, M. D.; Haddad, T. S.; Mylvaganam, M.; McConville, D. H.; Rettig, S. J. / . Am. Chem. Soc. 1993,115,2782.  (25)  Fryzuk, M. D.; Haddad, T. S.; Rettig, S. J. Organometallics 1991,10, 2026.  (26)  Fryzuk, M. D.; Haddad, T. S.; Rettig, S. J. Organometallics 1992,11, 2967.  (27)  Fryzuk, M. D.; Carter, A.; Westerhaus, A. Inorg. Chem. 1985,24, 642.  (28)  Atwood, J. L.; Smith, K. D. / . Chem. Soc, Dalton Trans. 1974, 921.  (29)  Atwood, J. L.; Smith, K. D. / . Chem. Soc, Dalton Trans. 1973,2487.  (30)  Hagadorn, J. R.; Arnold, J. Organometallics 1996,15,984.  95  References begin on page 94  Chapter 2: ...Phosphine Complexes of Scandium (31)  Fryzuk, M. D.; Haddad, T. S.; Berg, D. J. Coord. Chem. Rev. 1990, 99, 137.  (32)  Shapiro, P. J.; Bunel, E.; Schaefer, W. P.; Bercaw, J. E. Organometallics 1990, 9, 867.  (33)  Arnold, J.; Hoffman, C. G. / . Am. Chem. Soc. 1990,112, 8620.  (34)  Brookhart, M.; Green, M. L. H.; Wong, L.-L. Prog. Inorg. Chem. 1988,36, 1.  (35)  Harris, R. K. Nuclear Magnetic Resonance Spectroscopy; Pitman Books Limited: London, 1983.  (36)  Edelmann, F. T. Angew. Chem., Int. Ed. Engl. 1995,34, 2466.  (37)  Heijden, H. v. d.; Schaverien, C. J.; Orpen, A. G. Organometallics 1989, 8, 255.  (38)  Albrecht, I.; Schumann, H. / . Organomet. Chem. 1986,310, C29.  (39)  Schumann, H.; Albrecht, I.; Pickardt, J.; Hahn, E. / . Organomet. Chem. 1984,276, C5.  (40)  Booij, M.; Kiers, N. H.; Heeres, H. J.; Teuben, J. H. / . Organomet. Chem. 1989,364, 79.  (41)  Hogerheide, M. P.; Grove, D. M.; Boersma, J.; Jastrzebeski, J. T. B. H.; Kooijman, H.; Spek, A. L.; van Koten, G. Chem. Eur. J. 1995,1, 343.  (42)  Sapse, A. M.; Schleyer, P. v. R. Lithium Chemistry; John Wiley and Sons, Inc.: New York, 1995.  (43)  Sanders, J. K. M.; Hunter, B. K. Modern NMR Spectroscopy; 2nd ed.; Oxford University Press: New York, 1993.  (44)  Hore, P. J. Nuclear Magnetic Resonance; Oxford University Press: New York, 1995.  (45)  Watson, P. L.; Whitney, J. F.; Harlow, R. L. Inorg. Chem. 1981,20, 3271.  96  References begin on page 94  Chapter 2: ...Phosphine Complexes of Scandium (46)  Fryzuk, M. D.; Mylvaganam, M.; Zaworotko, M. J.; MacGillivray, L. R. Polyhedron 1996,15, 689.  (47)  Dolgoplosk, B. A.; Tinyakova, E. I.; Guzman, I. S.; Vollerstein, E. L.; Chigir, N. N.; Bondarenko, G. N.; Sharaev, O. K.; Yakovlev, V. A. J. Organomet. Chem. 1980,201, 249.  (48)  Heijden, H. v. d.; Pasman, P.; Boer, E. J. M. d.; Schaverien, C. J.; Orpen, A. G. Organometallics 1989, 8, 1459.  (49)  Heeres, H. J.; Meetsma, A.; Teuben, J. H. / . Chem. Soc, Chem. Commun. 1988, 962.  (50)  St. Clair, M. A.; Santarsiero, B. D. Acta Cryst. 1989, C45, 850.  (51)  Anderson, R. A. Inorg. Chem. 1979, 25, 2928.  (52)  Albright, T. A.; Burdett, J. K.; Whangbo, M. H. Orbital Interactions in Chemistry; Wiley: New York, 1985.  (53)  Anderson, W. P.; Cundari, T. P.; Drago, R. S.; Zerner, M. C. Inorg. Chem. 1990,29, 1.  (54)  Marks, T. J.; Kolb, J. R. Chem. Rev. 1977, 77, 263.  (55)  Morris, J. FL; Smith, W. E. / . Chem. Soc, Chem. Commun. 1970, 245.  (56)  Shapiro, P. J.; Henling, L. M.; Marsh, R. E.; Bercaw, J. E. Inorg. Chem. 1990,29, 4560.  (57)  Collman, J. P.; Hegedus, L. S.; Norton, J. R.; Finke, R. G. Principles and Applications of Organotransition Metal Chemistry; University Science Books: Mill Valley, 1987.  (58)  Fryzuk, M. D.; Williams, H. D. Organometallics 1983,2, 162.  (59)  Fryzuk, M. D.; Rettig, S. J.; Westerhaus, A.; Williams, H. D. Inorg. Chem. 1985,24, 4316.  97  References begin on page 94  Chapter 2: ...Phosphine Complexes of Scandium (60)  Wolczanski, P. T.; Bercaw, J. E. Organometallics 1982,1,793.  (61)  James, J. D.; Nauda, R. K.; Wallbridge, M. G. H. Inorg. Chem. 1967, 6, 1979.  (62)  Fryzuk, M. D.; Mylvaganam, M.; Zawarotko, M. J.; MacGillivray, L. R. Organometallics 1996,15, 1134.  (63)  Fryzuk, M. D.; Mao, S. S. H.; Zaworotko, M. J.; MacGillivray, L. R. / . Am. Chem. Soc. 1993,115, 5336.  (64)  Mu, Y.; Piers, W. E.; MacQuarrie, D. C ; Zawarotko, M. J.; Young, Jr., V. G. Organometallics 1996,15, 2720.  (65)  Fryzuk, M. D.; Mao, S. S. H.; Duval, P. B.; Rettig, S. J. Polyhedron 1995,14, 11.  (66)  Brown, T. L.; Dickerhoof, D. W.; Bafus, D. A.; Morgan, G. L. Rev. Sci. Inst. 1962,33, 491.  (67)  Brookhart, M.; Green, M. L. H. / . Organomet. Chem. 1983,250, 395.  (68)  Brookhart, M.; Green, M. L. H.; Pardy, R. B. A. / . Chem. Soc, Chem. Commun. 1983, 691.  (69)  Smart, J. C.; Curtis, C. J. Inorg. Chem. 1977,16,1788.  (70)  Schlosser, M.; Ladenburger, V. / . Organomet. Chem. 1967,8,193.  (71)  Schrock, R. R.; Fellmann, J. D. / . Am. Chem. Soc. 1978,100, 3359.  98  References begin on page 94  Chapter 3  Synthesis, Characterization and Reactivity of Amidodiphosphine Complexes of Aluminum  3.1 Introduction Our previous studies on the coordination properties of the potentially tridentate, mixed donor ligand [N(SiMe2CH.2PR2)2]" have centered on the transition metals and the lanthanoids. " 1  3  In an effort to expand the territory to which this ligand can be applied, we have initiated a study of the coordination chemistry of certain main group elements. As a starting point we examined the synthesis of complexes of aluminum using this ligand since both amide and phosphine type donors have been used previously with this metal although never before at the same time. Our primary goal was to try to use this ancillary ligand system to stabilize cationic Al(III) alkyl systems " of the formula {AiR[N(SiMe2CH.2PR2)2]} - While this has not yet been achieved, 4  14  +  along the way we did uncover some intriguing results that are reported here. Typically, Al(III) tends to form complexes that are four- or five-coordinate. Simple adducts of AICI3 or AIR3 with phosphines  15 2 4  or amines  10,14,25  "  35  tend to generate four  coordinate species; chelating ligands can enforce higher coordination numbers by virtue of the chelate effect. With amido or imido type ligands, dimers, trimers or cage structures can be obtained.  26,36  Because our mixed donor system is potentially tridentate we anticipated that five-  99  References begin on page 132  Chapter 3: ...Amidodiphosphine Complexes of Aluminum coordinate complexes of the general formula AlX2[N(SiMe2CH PR2)2] would result. While this 2  was generally found to be true, four-coordinate derivatives were also found to form under certain conditions.  3.2.1 Synthesis of AlCl2[N(SiMe CH PPr 2)2] (1) i  2  2  The reaction of the lithium salt of the ancillary ligand, L i N ( S i M e 2 C H P P r ) , i  2  A I C I 3 in toluene at 25  2  2  37  with  °C for 24 h generates the monomeric starting dichloride  AlCl2[N(SiMe2CH PPr )2] (1) in good yield (equation 3.1). The *H NMR spectrum of the i  2  2  product is indicative of a C  2 v  symmetric complex: one observes resonances due to equivalent  silyl methyl protons (S1CH3), methylene (SiC// P) and isopropyl methyls (PCH(C//3)2) and 2  methines (PC//(CH3) ) at typical chemical shifts. The P{ H} NMR spectrum of 1 consists of 31  1  2  a single broad resonance at -10.5 ppm, which is upfield as compared to the starting lithium derivative. The broadness of the peak (A 1/2 = 400 Hz) is undoubtedly the result of coordination to the quadrupolar A l nucleus ( A 1 , 1 = 5/2, 100% natural abundance), 2 7  27  38,39  while the upfield  shift upon coordination is in contrast to that found for other complexes of this ligand system for which a downfield shift is typical.  Small upfield or negative shifts upon coordination have  40  been previously reported for the phosphorus-31 resonances of phenyl- and benzylphosphines coordinated to trimethylaluminum.  17  Elemental analyses of 1 are indicative of some LiCl  incorporation that can be removed by successive recrystallizations. This type of behavior has been observed previously in aluminum complexes ' 29  32,36,41  and also during our investigations of  this ligand system with early transition metals (see Chapter 2 )  100  4 0 , 4 2  References begin on page 132  Chapter  3:  LiN(SiMe2CH2PPr'2)2 + AICI Pr' P  Me Si 2  Complexes  of  Aluminum  toluene 3  (- LiCl)  2  [3.1]  \ ,* ^  \  Me Si  ...Amidodiphosphine  CI  :N—AI  ^  2  3.2.2 Structure of AICl [N(SiMe2CH PPri )2] (D 2  2  2  Slow evaporation of a saturated toluene solution of AlCl [N(SiMe CH PPr ) ] (1) i  2  2  2  2  2  resulted in colorless needles that were suitable for single crystal X-ray diffraction. The molecular structure and numbering scheme are illustrated in Figure 3.1. The structural analysis confirms the monomeric nature of 1 and illustrates the coordination geometry of the complex that is best described as trigonal bipyramidal. It is evident that monomeric, salt- and base-free complexes of aluminum are possible utilizing the ancillary tridentate [N(SiMe CH PPr 2) ] i  2  2  _  2  ligand. The value of bulky ligands or alkyl groups in producing species of this type has been a common theme in the chemistry of organoaluminum compounds. " 43  49  The ancillary  amidodiphosphine ligand is bound in a meridional fashion with the phosphine donors occupying axial positions. The greatest deviation from ideal trigonal bipyramidal geometry is evident in the P(1)-A1(1)-P(2) bond angle of 171.8(3)°. The C1(1)-A1(1)-C1(2) bond angle of 118.2(3)° is close to that expected (120°) and is complemented by C1(1)-A1(1)-N(1) and C1(2)-A1(1)-N(1) bond angles of 121.3(5)° and 120.5(4)° respectively. Presumably, the twist in the backbone of the tridentate ligand is due to the relatively small Al(III) radius; the dihedral angle between the AlNSi and A1NP planes is 37.8°. In addition, this twist may be necessary in order to maintain 2  2  101  References  begin  on page 132  Chapter 3: ...Amidodiphosphine  Complexes of Aluminum  a trigonal bipyramidal environment for the A l centre, the preferred geometry for five-coordinate aluminum. 50  51  This has been observed previously in early-metal complexes that contain this  same ancillary ligand system  40,52  and compares to values ranging from 34.4° - 40.8° for similar  compounds of scandium (see Chapter 2).  40  Selected bond lengths and angles for 1 can be found  in Table 3.1.  F i g u r e 3.1  Molecular structure of AlCl2[N(SiMe2CH2PPri  2  )2]  (1); 33% probability thermal  ellipsoids are shown. Hydrogen atoms and disorder not shown for clarity.  The aluminum-chloride distances of 2.183(6) and 2.164(7) A compare well with those previously reported for terminal A l - C l bonds, 1U6,20,21,33,53 falling just outside the cited range of the longer bridging chlorides. 45  54  The aluminum-phosphorus bond lengths of 2.509(7) and  2.542(7) A are also common in terms of A l - P bond distances for phosphine adducts of aluminum, which usually span the range from ~2.4 - 2.8 A .  102  1 7 , 2 2 , 5 5  The intermediate solvation of  References begin on page 132  Chapter 3: ...Amidodiphosphine Complexes of Aluminum the aluminum centre by the phosphine donors is further illustrated by comparison with the bond lengths of 2.53(1) A and 2.535(1) A in Me3AlPMe3 and Me3AlPPh respectively.  19  3)  The  aluminum-nitrogen bond distance of 1.89(1) A falls well within the usual range for aluminumamide bonds.  25,26  ' " 28  30,36,56  As is expected, the amide bond present is slightly shorter than the N-  donor amine adducts of aluminum (-2.0 - 2.2 A )  30,33,35  but is similar to the bond lengths cited in  structures retaining an AI2N2 or bridging nitrogen c o r e ,  12,26,32,36  in which a lower degree of  dative bonding is usually invoked. The Al-N bond distance in 1 compares well with the similar amidodiamine compound Al(CH3)2[(NCH2CH2NEt2)2] (Al-N(amido) = 1.84(1) A) which also exhibits a trigonal bipyramidal Al coordination geometry.  29  It is unclear to what extent this bond  length, approximately 0.1 A shorter than the sum of the covalent radii, is indicative of some degree of p-p 7t bonding.  25  Table 3.1 Selected bond lengths (A) and angles (deg) for AlCl2[N(SiMe2CH PPr 2)2] (1). i  2  Pd)-Al(l)  2.542(7)  P(2)-A1(1)  2.509(7)  C1(1)-A1(1)  2.183(6)  C1(2)-A1(1)  2.164(7)  N(1)-A1(1)  1.89(1)  P(1)-A1(1)-P(1)  171.8(3)  C1(1)-A1(1)-C1(2)  118.2(3)  C1(1)-A1(1)-P(1)  93.5(2)  C1(1)-A1(1)-P(2)  89.3(2)  C1(2)-A1(1)-P(1)  89.8(2)  C1(2)-A1(1)-P(2)  95.7(3)  C1(1)-A1(1)-N(1)  121.3(5)  C1(2)-A1(1)-N(1)  120.5(4)  P(1)-A1(1)-N(1)  84.6(4)  P(2)-A1(1)-N(1)  87.4(4)  103  References begin on page 132  Chapter 3: ...Amidodiphosphine Complexes of Aluminum  3.3.1 Synthesis of Aluminum Dialkyls AlR [N(SiMe2CH PPr 2)2] (R = Me (2), Et (3), i  2  2  CH Ph (4)) 2  In typical metathesis fashion, the starting dichloride 1 reacts with lithium or magnesium alkyls to yield monomeric aluminum hydrocarbyl complexes of the general formula AlR [N(SiMe CH PPr ) ] (R = Me (2), Et (3), CH Ph (4)) with concomitant formation of the i  2  2  2  2  2  2  appropriate metal-halide salt as a byproduct in each case (Scheme 3.1).  Scheme 3.1  /  2MeLi  • or Me Mg»S (S = dioxane)  AIMestNCSiMesCHsPPr';2)2]  2  2  1  (S = THF) 4  104  References begin on page 132  Chapter 3: ...Amidodiphosphine Complexes of Aluminum The reactions to form 2 and 3 were carried out at 25 °C in toluene whereas the metathesis reaction to form the dibenzyl 4 proved to be more successful at -60 °C using THF as the solvent. The dimethyl complex 2 may also be generated via the reaction of the starting lithium salt LiN(SiMe2CH2PPr 2)2 with M e 2 A l C l at room temperature in hexanes. i  All three dialkyl-  aluminum compounds are isolated in good yields as either colorless oils, as in the case of 2 and 3, or as pale yellow crystals, as found for 4. These thermally stable derivatives are soluble in aromatic solvents as well as THF; however, their solubility is more limited in hydrocarbon solvents such as pentane or hexanes. As is the case with 1, the elemental analysis of 4 is valid only if one allows for the inclusion of salt, in this case M g C l 2 . All three of these air- and moisture-sensitive hydrocarbyl derivatives exhibit room temperature *H NMR spectra indicative of highly symmetric environments. Interestingly, and in contrast to that found for 1, the solution 31  P{ *H} NMR spectra of the dialkyl complexes 2 - 4 all consist of sharp singlets in the range of  -3.6 to -4.9 ppm, consistent with very weak phosphine coordination and/or rapid exchange. In an effort to probe these aspects of phosphine ligation we undertook some solid-state structural analyses where possible.  3.3.2 Structure of Al(CH2Ph)2[N(SiMe2CH PPr 2)2] (4) i  2  Evaporation of a saturated toluene solution of Al(CH2Ph) [N(SiMe2CH2PPr 2)2] i  2  (4)  resulted in the deposition of pale yellow needles that were suitable for X-ray diffraction. The molecular structure and numbering scheme of this compound can be seen in Figure 3.2. The structural study reveals 4 to be four-coordinate via coordination to only one phosphine arm of the ancillary [N(SiMe2CH2PPr 2)2]" ligand. The steric repulsion between the aryl rings of the benzyl i  groups and the isopropyl substituents on the phosphorus donors is great enough to force one of the weakly bound phosphines to dissociate, allowing the Al centre to adopt a distorted tetrahedral geometry. This is the most prevalent and preferred mode of coordination for Al(III).  57  The six  tetrahedral angles, varying from 92.1(2)° to 117.5(3)°, indicate the extent of the distortion from  105  References begin on page 132  Chapter 3: ...Amidodiphosphine  Complexes of Aluminum  pure tetrahedral with the most acute angle, defined by the P(l), A l ( l ) and N ( l ) atoms, being the result of the bidentate ligand "pinching back". The single aluminum-phosphorus bond measures 2.453(3) A and is slightly shorter than the A l - P bond lengths of 1. The aluminum-nitrogen bond is also similar to that ascertained for 1, although the slightly longer silicon-amide bond (1.745(5) A vs. 1.72(1) A) coupled with the marginally shorter A l - N bond (1.853(5) A vs. 1.89(1) A) may point to a slightly higher degree of jt-donation from the amide lone pair to the metal. Compound 4 also possesses unremarkable aluminum-carbon bond lengths of 1.991(7) A and 1.987(6) A which are akin to those previously reported for terminal A l - C bonds. 10,25,45-47,58 j  n  es  t g e s t similarity is with the A l - C distances ron  found in Al(CH2Ph)3 for which average values of 1.989(6) A were found.  48  am  Figure 3.2 Molecular structure of Al(CH2Ph)2[N(SiMe2CH2PPri2)2] (4); 33% probability thermal ellipsoids are shown. Hydrogen atoms are omitted for clarity. 106  References  begin on page 132  Chapter  3: ...Amidodiphosphine  Complexes  of  Aluminum  Table 3.2 Selected bond lengths (A) and angles (deg) for Al(CH Ph)2[N(SiMe2CH2PPr 2)2] (4). i  2  P(D-A1(1)  2.453(3)  N(1)-A1(1)  1.853(5)  C(19)-A1(1)  1.991(7)  C(26)-A1(1)  1.987(6)  C(19)-A1(1)-C(26)  114.1(3)  P(1)-A1(1)-N(1)  92.1(2)  P(1)-A1(1)-C(19)  106.3(2)  P(1)-A1(1)-C(26)  110.4(2)  N(1)-A1(1)-C(19)  113.6(2)  N(1)-A1(1)-C(26)  117.5(3)  The recent X-ray structure of Cr(CH2Ph)[N(SiMe2CH2PPh2)2], which possesses an re52  bound benzyl fragment, prompted us to further examine the structure of 4. The Al-Qp  SO  lengths  of 2.883 and 2.974 A, coupled with Al-Cbenzylic-Qpso bond angles of 110.4(4)° and 116.2(4)°, were outside of the range normally found for an r | interaction. Furthermore, the Al-C rtho 2  0  distances of 3.589 - 3.924 A were convincing evidence against an rj formulation of any type. 3  The CbenzylicrQpso bond lengths of 1.505(8) A also fall within the desired range for a-bound benzyl groups devoid of any additional interactions, and are longer than that found in the related chromium derivative. A list of pertinent bond lengths and angles is presented in Table 3.2.  3.3.3 3lp and A l NMR Studies and Coordination Number of AIR2[N(SiMe CH PPr 2)2] 2 7  i  2  2  (R = Me (2), Et (3), CH Ph (4)) 2  While the X-ray structure of dichloride 1 is compatible with the solution H and P{ H} l  31  1  NMR studies, the simple H NMR spectrum and single sharp resonance in the P{ H} NMR l  31  1  spectrum of the dibenzyl 4 is at odds with the bidentate mode of coordination of the ancillary ligand found in the solid state. Even though we were unable to isolate either dimethyl 2 or diethyl 3 as solids, we were curious about their solution structures. Previously, A l NMR 2 7  107  References  begin  on page 132  Chapter 3: ...Amidodiphosphine Complexes of Aluminum spectroscopy has proved helpful in correlating chemical shift with coordination number,  38,39  even at temperatures where highly dynamic molecules show only averaged *H and P{ H} 31  1  signals.  17  Aluminum-27 NMR chemical shifts span a range of about 300 ppm which may be broadly categorized according to the degree of substitution at the aluminum atom (Figure 3.3). The presence of Al-C bonds and the stronger tendency towards lower coordination in aluminum alky Is results in resonances from -100 to 280 ppm low field of reference.  Compounds  containing aluminum bonded to elements of Group 15, 16 and 17 are found to high field of the aluminum alkyls, falling into two distinct regions of either tetrahedral (140 to 40 ppm) or octahedral (40 to -40 ppm) geometries. The few five-coordinate species measured come between 70 and 30 ppm. Although there is some overlap between regions, and exceptions to these guidelines exist (eg. [AII4]- appears at -28 ppm), in general the shift is very indicative of structure. For example,  2 7  A l NMR has been useful in assigning the tetrahedral and octahedral  sites in a number of geologically important materials such as aluminum oxides and alkali and alkali earth aluminates.  38,39  alkyl Al 3 coord  4 coord  5 coord tetrahedral Al 5 coord Al octahedral Al  j 250  j 200  j 150  j 100  j 50  j 0  j -50  ppm Figure 3.3 Aluminum-27 chemical shift range as a function of coordination number.  108  References begin on page 132  Chapter  3:  ..Amidodiphosphine  Complexes  of  Aluminum  As already mentioned, the solution spectroscopic data for dichloride 1 reinforce the results of the solid state X-ray analysis. The H NMR spectrum is indicative of a molecule with 1  overall C  2 v  symmetry and the single broad resonance (A1/2 = 400 Hz) in the -^PpH} NMR  spectrum at -10.5 ppm suggests some aluminum-phosphorus interaction. The P{!H} NMR 31  spectrum is temperature invariant down to the solvent limit (-90 °C, ds-toluene). The P MAS 3 1  NMR spectrum of a solid sample of dichloride 1 is consistent with the same sort of Al-P coupling that is present in solution, exhibiting a 1:1:1:1:1:1 sextet due to coupling to the I = 5/2 aluminum nucleus (Vp-Al = 265 Hz) (Figure 3.4).  20  -20  -40 ppm  Figure 3.4 161.9 MHz P{ lH} MAS NMR spectrum of AlCl2[N(SiMe2CH2PPr )2] (1) with a 31  i  2  spinning rate of 4.551 kHz. The small multiplets are spinning side bands.  The solution A l NMR spectrum contains a very broad peak (A1/2 = 900 Hz) at 65 ppm, 2 7  which is well within the typical range for five-coordinate aluminum derivatives.  38,39  Large peak  widths appear to be common when observing quadrupolar nuclei in heteroleptic environments.  23,59  Dichloride 1 does not form adducts with THF, pyridine or acetonitrile in  solution. Presumably, the steric constraints of the ancillary ligand system prevent the aluminum centre from attaining a higher coordination number even in the presence of excess donor molecules. Interestingly, the addition of a strongly Lewis acidic moiety such as AICI3 to 1  109  References  begin  on page 132  Chapter 3: ...Amidodiphosphine Complexes of Aluminum results in the reduction of the aluminum coordination number to four with a further AICI3 interaction being observed with a free phosphine (see later). The P{ H} NMR spectrum of dimethyl 2 features a much narrower peak (A1/2 = 5 Hz) 31  l  than would be expected for a compound with an Al-P bond. The position of this peak, -4.9 ppm, is also close to that of the starting lithium salt LiN(SiMe2CH2PPr )2, which resonates at -4.7 i  2  ppm.  As the temperature is lowered, the peak broadens slightly, but no evidence of  decoalescence behaviour is observed even down to -90 °C. It is unclear whether this indicates a three-coordinate structure, or a rapidly fluxional process that could not be frozen out. The solution  2 7  A l NMR spectrum of 2, however, consists of a broad peak at 70 ppm (A1/2 = 4.3  KHz), indicating a similar geometry to that found in 1. As well, the equivalent methyl groups (AlMe2) appear as a triplet in the *H NMR spectrum, the result of coupling to two equivalent phosphorus-31 nuclei. All the spectroscopic data taken together point to a highly fluxional structure for 2, presumably a result of rapid phosphine-arm dissociation and reassociation on the NMR time scale, via a five-coordinate intermediate (Scheme 3.2). Scheme 3.2 r  <  X R  P— * - A L * - P -"R R' 2 R V  R  R'-P, R  P—*^AI  110  References begin on page 132  Chapter 3: ..Amidodiphosphine Complexes of Aluminum  Examination of the *H and P{ H} NMR spectra of diethyl 3 reveals similar properties 31  1  to those found for 2. The *H NMR spectrum is characteristic of a highly symmetric molecule in which the methylene protons of the ethyl groups (AIC//2CH3) appear as a quartet of triplets due to coupling to two equivalent phosphines as well as the terminal -CH3 protons of the ethyl group. As in 2, the P{ H} NMR spectrum of 3 exhibits only a single sharp resonance at -4.2 ppm, 31  1  which broadens but does not decoalesce as the temperature is lowered. However, the A l NMR 2 7  spectrum is shifted downfield from that of 1 or 2 to 140 ppm, well within the range normally associated with four-coordinate species.  38,39  Again, the data point toward a rapidly fluxional  structure; however, to account for the A l NMR spectral result, one can suggest that diethyl 3 2 7  spends a comparatively greater amount of time in the dissociated state as compared to dimethyl 2. That this is the case can be rationalized on the similarity in size between the -CI and -CH3 groups in 1 and 2, and the larger -CH2CH3 groups in 3 which undoubtedly destabilize coordination by both of the bulky isopropyl-substituted phosphine donors. The crystal structure of Al(CH2Ph)2[N(SiMe2CH2PPr2)2] (4) illustrates that with large i  enough alkyl groups, aluminum will reside in its preferred tetrahedral geometry. Once again, the *H NMR spectrum displays resonances indicative of a C \ symmetric complex although instead 2  of a triplet for the benzylic protons, a broad peak at 3.50 ppm is observed. No useful information could be garnered from variable temperature *H NMR spectroscopy due to a systematic broadening of all resonances with no obvious low temperature limit. However, the sharp singlet at -3.6 ppm in the P{ *H} NMR spectrum does display typical coalescence behavior; the singlet 31  broadened and shifted slightly upfield as the temperature was lowered and at -73 °C, two singlets of equal intensity at -5.4 and -6.8 ppm were evident (Figure 3.5). Unfortunately, monitoring a fluxional process over such a limited temperature range (-62 to -73 °C, at 202.4 MHz) results in large inherent errors in the kinetic parameters, AG*, AH* and AS*. In cases such as this, errors in T, and subsequently 1/T, become large enough to render the numerical value of AS* meaningless. One solution to this problem is to examine the variable temperature behavior at a variety of 111  References begin on page 132  Chapter 3: ..Amidodiphosphine Complexes of Aluminum  Figure 3.5 Variable temperature 202.4 MHz P{ !H} NMR spectra of Al(CH Ph)231  2  [N(SiMe CH PPri2)2] (4) in CD C1 . 2  2  2  112  2  References begin on page 132  Chapter 3: ...Amidodiphosphine Complexes of Aluminum magnetic field strengths since fluxional processes are magnetic field strength dependent.  60  Repeating the study at 121.4 MHz reveals a similar coalescence between -73 and -84 °C. Although a large error is still associated with the numerical value of AS^ the large negative value ascertained in both cases is evidence for an associative mechanism of exchange being operative (Scheme 3.2). The solution  2 7  A l NMR spectrum consists of a broad peak at 54 ppm, both at room  temperature and at -74 °C. This suggests that at room temperature in solution, 4 is rapidly fluxional through a five-coordinate intermediate (associative pathway). That dibenzyl 4 would be five-coordinate in solution, while diethyl 3 would remain four-coordinate, is counterintuitive, based on the relative sizes of the ethyl and benzyl groups. In other words, one would expect the solution A l NMR shift for dibenzyl 4 to be similar to diethyl 3 on the basis of the solid state 2 7  structure of 4. It may be that the more electron-withdrawing nature of the benzyl groups results in a more Lewis-acidic aluminum centre which subsequently has a greater attraction for the phosphine donors. In contrast to the coalescence behavior of 4 observed in non-coordinating solvents (dstoluene, CD2CI2) the presence of a coordinating solvent such as THF establishes an equilibrium. At -75 °C, a broad singlet at -5.5 ppm in the P{ H) NMR spectrum is observed (Figure 3.6). 31  l  However, as the temperature is lowered, two additional peaks of equal intensity at -4.7 and -7.1 ppm slowly grow in, until finally at -86 °C, only these peaks remain. These data suggest that in this temperature range, a species with equivalent phosphines (represented by the singlet at -5.5 ppm) is in equilibrium with a second species in which the phosphines are inequivalent. A number of possible equilibria can be envisioned to explain the results of the variable temperature NMR study (Scheme 3.3). Since the singlet at -5.5 ppm does not reside at the midpoint of the two peaks at -4.7 and -7.1 ppm, one can assume that a difference in solvation must occur, i.e., one species has a bound THF molecule while the other does not. Of the possibilities shown in Scheme 3.3, we propose that equilibrium a) is operative.  113  References begin on page 132  Chapter 3: ...Amidodiphosphine Complexes of Aluminum  Figure 3.6 Variable temperature 121.4 MHz P{ !H} NMR spectra of Al(CH2Ph)231  [N(SiMe CH PPr 2) ] (4) in C D 0 . i  2  2  2  4  114  8  References begin on page 132  Chapter  3:  ...Amidodiphosphine  Complexes  of  Aluminum  The reasons for this are two-fold: first, the starting dibenzyl compound exhibits a singlet close to the resonance assigned to the uncoordinated phosphine (-4.7 ppm) of the final asymmetric product. This implies that no coordination of the phosphines to the aluminum centre is occurring in the starting compound. Thus, b), in which both phosphines are coordinated, seems less likely. Although we have previously stated that associative intermediates are probable when accounting for phosphine exchange in the aluminum dialkyls 2 - 4, it should be noted that these were all studied in non-coordinating solvents. Upon dissolution in ds-THF, the phosphine(s) are displaced from the coordination sphere in favor of solvent. Second, although a similar structure with uncoordinated phosphines can be invoked in c), aluminum has an overwhelming tendency to favor tetrahedral geometries when possible. This is also confirmed by the solid state structure of 4, which coincides with the low temperature limiting structure proposed in a). Thus, Scheme 3.3.a), in which both species involved in the equilibrium are tetrahedral, is presented. Due to the proximity of these resonances and the broadness of these peaks, equilibrium data such as AS", AG° and AH° could not be determined to support or negate our postulate.  115  References  begin  on page 132  Chapter  3:  ...Amidodiphosphine  Complexes  of  Aluminum  Scheme 3.3  b) Me2 Me SL _>»Si,  r  Me  2  <  R - P--*-Ah*--R--R R' R  A  Me  2  2  I  R  + THF  c) R  ^  S  i  Me2 ^ /  S  N  THF  Al  i  Me2 N /  P  C  Me Me / k../Si S  R  R  THF  \  2  2  P -R V  I  R  /\'H/> THF  P-*-AI.  + THF  116  References  begin  on page  132  Chapter 3: ...Amidodiphosphine Complexes of Aluminum  3.4 Reactivity of Aluminum Dialkyls AlR [N(SiMe CH PPr )2] (R = Me (2), Et (3), i  2  2  2  2  CH Ph (4)) 2  In contrast to the instability noted for the analogous scandium systems, the dialkyl compounds produced are remarkably inert when exposed to a variety of small molecules. Reactions with ethylene, CO, H , CH3I, C H 3 C N or Bu NC failed to result in any appreciable l  2  transformation. In retrospect, this lack of reactivity is not surprising. There is a paucity of reaction chemistry for aluminum alkyls, the most documented being the elimination of alkane via the reaction of primary and secondary amines and phosphines with aluminum aikyig 15,26,32,36,61 Singular examples of B u ^ C insertion into Cp-based and tetradentate nitrogen donor systems 62  28  have been noted.  3.5 Synthesis of the Monoalkyl Derivatives Al(R)Cl[N(SiMe CH PPr ) ] (R = Me (5), i  2  2  2  2  Et (6)) The difference in structure between dichloride 1 and the dialkyls 2-4 prompted us to examine the preparation of the monoalkyl derivatives, Al(R)Cl[N(SiMe CH PPr ) ]. It would i  2  2  2  2  stand to reason that intermediate behaviour could be expected for these latter species; in addition, these derivatives might be useful starting materials for cationic derivatives by halide abstraction. ' " 8,9 11  14  Crown ether ligands have been the most successful in this regard, although  nitrogen-based ancillary ligands have also been of some use.  7,10  Unfortunately, our attempts to  selectively remove one alkyl group from the dialkyls 2 or 4 with the strong Lewis acid B(C F )3 ' 6  5  63  64  or with Brookhart's acid, [H(OEt ) ][B{3,5-(F 0 C H3}4], 2  2  3  2  6  65  resulted in a  mixture of intractable products.  117  References begin on page 132  Chapter 3: ...Amidodiphosphine Complexes of Aluminum When dichloride 1 was subjected to the slow addition of one equivalent of methyl lithium or equivalent Grignard reagent (i.e., MeMgBr, MeMgCl or 1/2 equiv Me2Mg(dioxane)), equimolar mixtures of starting dichloride 1 and dimethyl 2 were obtained. Similarly, addition of EtLi (one equiv) to 1 also produced a mixture of starting dichloride 1 and diethyl 3 (equation 3.2).  AICI [N(SiMe CH PPr )2] i  2  2  2  2  + 1 equiv RLi  toluene, RT (- LiCl)  (R = Me, Et)  1  1/2 AICI [N(SiMe CH PPr ) ] i  2  2  2  2  +  2  1  1/2 AIR [N(SiMe CH PPr ) ! i  2  2  2  2  2  2,3  To account for this behaviour, we suggest that the monoalkyl intermediate reacts faster than 41  dichloro 1 because in the former species, phosphine dissociation becomes more likely thus opening the coordination sphere to attack by nucleophiles. However, we did find that the mixture of dichloride 1 and dimethyl 2 could be induced to equilibrate to a mixture that contained the monomethyl derivative Al(Me)Cl[N(SiMe2CH2PPr 2)2] (5); it was found that i  heating at 100 °C resulted in approximately 85% conversion, as monitored by *H NMR spectroscopy (equation 3.3).  AICI [N(SiMe CH2PPr ) ] i  2  2  2  2  D  ^  1 AIMe [N(SiMe CH PPr ) ] 2 i  2  c  2  2  2  2  1 0 0  °  118  6  C  ^  2 AI(Me)CI[N(SiMe CH PPr ) ] i  2  5  2  2  2  [3.3]  References begin on page 132  Chapter 3: ..Amidodiphosphine Complexes of Aluminum The following thermodynamic data for the equilibrium were calculated using the van't Hoff equation: AH° = 25 ± 2 kcal mol" and AS° = 74 ± 6 cal K- mol" . The van't Hoff plot and 1  1  1  data are presented in Table 3.3 and Figure 3.7.  1/T  Figure 3.7 van't Hoff plot for the equilibrium between AlCl2[N(SiMe2CH2PPr 2)2] (1), L  AlMe [N(SiMe2CH PPr 2)2] (2) and Al(Me)Cl[N(SiMe CH PPr 2)2] (5) (r = 0.990). i  2  i  2  2  2  Table 3.3 van't Hoff data for the equilibrium between AlCl2[N(SiMe2CH2PPr 2)2] (1), L  AlMe2[N(SiMe2CH PPr )2] (2) and Al(Me)Cl[N(SiMe CH PPr 2)2] (5). i  2  i  2  2  2  temp, K  1/T  K  InK  293  0.00341  0.00327  -5.723  318  0.00314  0.0400  -3.219  333  0.00300  0.279  -1.277  343  0.00292  0.742  -0.298  373  0.00268  39.16  3.668  119  References begin on page 132  Chapter 3: ...Amidodiphosphine Complexes of Aluminum  In an effort to obtain pure Al(Me)Cl[N(SiMe2CH2PPr 2)2] (5), we studied the reaction of i  the starting lithium salt LiN(SiMe2CH2PPr 2)2 with M e A l C l 2 - Although in this reaction we were i  able to prepare more pure Al(Me)Cl[N(SiMe2CH2PPr 2)2] (5), small quantities of 1 and 2 were i  present, and these impurities persisted even if the reaction was done under dilute conditions at -78 °C. Given the equilibrium between 5 and dichloro 1 and dimethyl 2, this last result is not surprising. As mentioned above, the reaction of 1 with 1 equiv of EtLi produced an equimolar mixture of 1 and diethyl 3; however, in this case prolonged heating (2 weeks, 100 °C) did not result in any observable exchange to produce the monoethyl derivative as monitored by *H NMR spectroscopy. As a result, we examined the reaction of the lithium salt LiN(SiMe2CH2PPr 2)2 i  with EtAlCl2, in hopes of producing Al(Et)Cl[N(SiMe2CH2PPr 2)2] (6). The H NMR spectrum i  !  of this reaction mixture indeed did show predominantly 6, but small amounts (< 5%) of 1 were invariably present, due to trace amounts of AICI3 present in EtAlCl2- The complete absence of diethyl 3 is consistent with the lack of an exchange equilibrium being present. In other words, once the monoethyl complex 6 forms, it does not disproportionate into 1 and 3, in contrast to the monomethyl derivative 5.  3.6.1 Synthesis of AlCl [N(SiMe CH PPr 2)2]»AlCl3 (7) i  2  Although  we  2  were  Al(R)Cl[N(SiMe2CH2PPr 2)2], i  2  unable  to  isolate  pure  monoalkyl  derivatives  we decided to attempt the formation of the cation  {AlR[N(SiMe2CH2PPr 2)2]} by reaction of the slightly impure monoethyl complex 6 with i  +  AICI3. Upon addition of the solution of 6 to a slurry of AICI3 in toluene (approx. 2.5 equiv), the  solution turned slightly amber and most of the AICI3 was taken up into solution. Upon workup, colorless crystals were isolated that displayed the following spectroscopic characteristics: the 31  P{ H} NMR spectrum consisted of two broad peaks at 0.0 and -11.8 ppm of roughly equal 1  120  References begin on page 132  Chapter 3: ...Amidodiphosphine Complexes of Aluminum intensity and the *H NMR spectrum indicated a single clean but asymmetric product, with no resonances due to the ethyl group present. The elemental analysis, coupled with the P{ H} 31  1  and ! H NMR data pointed towards an AICI3 adduct of dichloro 1, in which the [N(SiMe2CH2PPr 2)2]~ i  ligand was bound to an AICI2 unit in a bidentate fashion via the amide  function and one phosphine donor, with the remaining phosphine arm of the ancillary ligand coordinated to an AICI3 molecule (Scheme 3.4).  Scheme 3.4  7  3.6.2 Structure of AlCl [N(SiMe CH PPri ) ].AlCI (7) 2  2  2  2  2  3  X-ray quality crystals of 7 were grown by slow evaporation of a saturated toluene solution. The molecular structure and numbering scheme are depicted in Figure 3.8. The most important feature of this molecule is that two four-coordinate aluminum centres are present. One Al is bound in a bidentate fashion to the tridentate ligand with the remaining phosphine arm coordinated to the AICI3 unit. We presume that the presence of this coordinated AICI3 unit in 7 serves the same function as the large benzyl groups in 4, in that they both allow for the central aluminum to retain a tetrahedral geometry.  121  References begin on page 132  Chapter 3: ...Amidodiphosphine  Complexes of  Aluminum  Cl(3)  Figure 3.8 Molecular structure of AlCl2[N(SiMe CH2PPr 2)2] AlCl3 (7); 33% probability i  ,  2  thermal ellipsoids are shown. Hydrogen atoms are omitted for clarity.  The bond angles involving this aluminum range from 115.9(1)° to 95.5(1)°. The smallest angle, as in 4, involves chelation of the ligand to the metal (P(1)-A1(1)-N(1)). The second aluminum also exhibits a four-coordinate, tetrahedral geometry. The angles involving the phosphine in this case are slightly smaller (105.79(8)° - 108.42(9)°) than those between the chlorides (110.30(9)° 112.1(1)°) and are the result of the donation of the lone pair from the phosphine to the metal centre.  The aluminum-chlorine bond lengths are all similar (2.106(2) A - 2.132(2) A) and  unexceptional. The aluminum-phosphorus bond lengths of 2.403(2) A (P(l)-Al(l)) and 2.412(2) A (P(2)-A1(2)) are both smaller than those found in 1 and 4.  122  References begin on page 132  Chapter 3: ...Amidodiphosphine Complexes of Aluminum  The H NMR spectrum of 7 is consistent with its structure in that a doubling of all ligand l  resonances is seen as a result of the lack of symmetry. The P{ H} NMR spectrum consists of 31  l  two broad singlets, due to the coordination of each phosphorus nucleus to a quadrupolar Al centre. Also, the methylene protons associated with the "free" arm (C(2)), appear as a broad doublet. Examination of the crystal structure reveals a short intramolecular C(2)-H(3) •Cl(2) distance of 3.082 A. It appears this interaction slows the rotation of the free arm on the NMR time scale, resulting in a broadening of these protons. Selected bond lengths and angles are presented in Table 3.4.  Table 3.4 Selected bond lengths (A) and angles (deg) for AlCl2[N(SiMe2CH PPr 2) ]«AlCl3(7). i  2  2  P(D-A1(1)  2.403(2)  P(2)-A1(2)  2.412(2)  C1(1)-A1(1)  2.132(2)  C1(2)-A1(1)  2.124(2)  C1(3)-A1(2)  2.115(2)  C1(4)-A1(2)  2.106(2)  C1(5)-A1(2)  2.112(2)  N(1)-A1(1)  1.828(3)  C1(1)-A1(1)-P(1)  110.02(7)  C1(2)-A1(1)-P(1)  110.29(8)  C1(1)-A1(1)-N(1)  115.4(1)  C1(2)-A1(1)-N(1)  115.9(1)  C1(1)-A1(1)-C1(2)  108.90(8)  P(1)-A1(1)-N(1)  95.5(1)  C1(3)-A1(2)-P(2)  105.79(8)  C1(4)-A1(2)-P(2)  108.32(8)  C1(5)-A1(2)-P(2)  108.42(9)  Polynuclear aluminum compounds are numerous in the literature although those incorporating amido ligands are more rare. Complex 7 is one of only three known polynuclear aluminum complexes with a tridentate amido ligand.  29,66  Its asymmetric nature is similar to that  of Al3(CH3)8(Et NCH2CH2NCH2CH2NEt2) although ligation to the amide lone pair is not 2  123  References begin on page 132  Chapter  observed in our case.  3:  ...Amidodiphosphine  Complexes  of  Aluminum  Exposure of 7 to additional AICI3 did not result in any further  rearrangement.  3.6.3 Mechanism of Formation of AlCl [N(SiMe2CH PPr 2)2]«AlCl3 (7) i  2  2  The most striking feature of the structure of 7 is the absence of an ethyl group bound to the central aluminum. It appears that monoethyl 6 undergoes two separate reactions with AICI3 to form 7: first, a ligand redistribution reaction at the central aluminum to eliminate an ethyl group, and second, ligation of a second AICI3 molecule by a dissociated phosphine arm. The ligand redistribution behavior of mixed alkyl-halo aluminum compounds is well documented and is most often called upon to explain a deficiency of alkyl groups in organoaluminum compounds.  For example, redistribution of Et2AlCl to generate EtAlCl2 and Et3Al has been  67  suggested to occur through the formation of the unsymmetrical dimer Et2Al(u.-Cl)(u,-Et)AlEtCl. Such a process accounts for the shortage of ethyl groups (relative to that expected) for a number of organoaluminum compounds found when Et AlCl is added to ligand systems as varied as 2  crown ethers, macrocyclic and open-chain multidentate amines 12  dimers.  16,20  Me2AlCl  21  33,34  and aluminum-phosphorus  A similar explanation has been invoked to account for analogous results using Asymmetric bridge cleavage is thought to be involved in the formation of cationic  species of crown ethers.  12  It is possible that a similar type of process operates in our system. As shown in Scheme 3.5, exposure of monoethyl 6 to AICI3 could result in the equilibrium formation of a mixed ethylchloro-bridged species. For this intermediate to form, dissociation of one phosphine arm alleviates steric crowding that would otherwise occur in a six-coordinate species. The dangling Lewis basic phosphine then coordinates to a second equivalent of AICI3, effectively negating any further chance for aluminum coordination.  The transient alkyl-halo-bridged dimer then  eliminates EtAlCl to yield 7. 2  124  References  begin  on page 132  Chapter  3:  ...Amidodiphosphine  Complexes  of  Aluminum  Scheme 3.5  Some support for this mechanism comes from the following observations: (i) the number of equivalents of AICI3 required for this reaction is two. The addition of 6 to a toluene slurry of one equiv AICI3 does not result in the formation of 7; (ii) the trace amount of AlCl2[N(SiMe2CH PPr 2)2] (1) present in starting samples of Al(Et)Cl[N(SiMe CH2PPr )2] (6) i  2  2  reacts in a similar manner to that of 6 as no AlCl2[N(SiMe CH PPr 2)2] i  2  the reaction with AICI3.  2  References  2  (1) can be found after  Thus, the addition of two equivalents  125  i  begin  of AICI3  on page 132  to  Chapter  3:  ...Amidodiphosphine  Complexes  of  Aluminum  AlCl2[N(SiMe2CH2PPr 2)2] (1) also forms the AICI3 adduct 7, although one equivalent does not. i  This suggests that if the phosphine arms of AlCl2[N(SiMe2CH2PPr 2)2] (1) were rapidly i  fluxional, and an AICI3 molecule simply trapped a transiently dangling phosphine arm, the addition of one equiv of AICI3 should be sufficient to form 7. The need for two equiv AICI3 supports our contention that the formation of a bridged species is necessary to dissociate one phosphine arm, which is then trapped by the excess AICI3. In an effort to track down the E t A l C l 2 postulated to be generated, we collected the volatiles from the reaction and examined them by ^H. NMR spectroscopy. Although peaks at 0.93(t) and 0.22(q) ppm, corresponding to the methyl and methylene protons of the ethyl group, respectively, could be seen, the spectrum was further complicated by additional resonances in the range 0-2 ppm. In the light of the K NMR results, it is difficult to conclusively determine the l  presence or absence of the desired aluminum alkyl halide. That E t A l C l 2 is produced and then forms other products as the result of exchange cannot be ruled out. The analogous reaction of the methyl derivative Al(Me)Cl[N(SiMe2CH2PPr )2] (5) with i  2  AICI3 provides a mixture of products, amongst them a small amount of 7 (as observed by K l  NMR spectroscopy). The smaller -CH3 group likely experiences a more facile exchange, and quickly forms other products as well. No attempt to further characterize the products of this addition was made.  3.7 Conclusions Bis(hydrocarbyl) aluminum derivatives of the ancillary, potentially tridentate ligand [N(SiMe2CH2PPr 2)2]"  were synthesized via metathesis of the starting dichloride  i  AlCl2[N(SiMe2CH2PPr 2)2] (1) with various alkylating reagents. The solution structures of i  these compounds are postulated to be highly fluxional on the basis of their variable temperature NMR spectroscopic properties. In all cases, these compounds resisted attempts at chemical 126  References  begin  on page 132  Chapter 3: ...Amidodiphosphine Complexes of Aluminum transformation. The aluminum alkyl halides could be synthesized, upon addition of RAICI2 (R = Me, Et) to the lithium salt LiN(SiMe2CH2PPr 2)2 but attempts to abstract a chloride resulted in i  the formation of AlCl2[N(SiMe2CH2PPr 2)2]*AlCl3 (7), a compound retaining two fouri  coordinate tetrahedral aluminum centres rather than a cationic structure. It is postulated that an exchange process is responsible for the lack of an alkyl group in the final product. It appears that even in the presence of a tridentate anionic ligand, the aluminum centre will still tend to reside in a four-coordinate environment if at all possible.  3.8 Experimental Section 3.8.1 Procedures Unless otherwise stated, general procedures were performed according to Section 2.16.1. 2 7  A l NMR spectroscopy was performed on a Varian XL-300 instrument operating at 75.48 MHz.  3 1  P MAS NMR spectroscopy was performed on a Bruker MSL 400 instrument operating at  161.9 MHz.  2 7  A l spectra were referenced to external A/CI3 in D2O (0.0 ppm).  3.8.2 Materials AICI3 was purchased from Aldrich and sublimed prior to use. Me 2 AlCl (1.0 M solution  in hexanes) was purchased from Aldrich and used as received. M e A l C l 2 (1.0 M solution in hexanes) and EtAlCh (1.0 M solution in hexanes) were purchased from Aldrich, crystallized and used in the solid form. Mg(CH2Ph)2(THF) and MgMe2(dioxane) were prepared according to 2  published procedures.  68,69  127  References begin on page 132  Chapter 3: ..Amidodiphosphine Complexes of Aluminum  3.8.3 Syntheses 3.8.3.1 AlCl2[N(SiMe CH PPr 2)2] (1) i  2  2  To a slurry of AICI3 (650 mg; 4.88 mmol) in toluene (10 mL) was added a toluene solution (10 mL) of LiN(SiMe CH PPr ) (1.78 g; 4.47 mmol). The reaction mixture was i  2  2  2  2  stirred for 12 h, and then passed through a frit lined with Celite to remove LiCl. The solvent was then removed in vacuo to yield a white waxy solid. The residue was taken up in a minimum amount of toluene (approx. 4 mL).  Slow evaporation of the solvent afforded large colorless  needles (1.84 g; 84% yield). !H NMR (C D ): 8 1.91 (d of sept, 4H, C#Me2, / - H = 7.0 Hz, 3  6  2  6  H  /H-p = 1.0 Hz), 1.21 and 1.11 (dd, 24H, CHMeMe',  3  / - H = 7.0 Hz, / - P = 7.0 Hz), 0.65 (d, 3  H  H  4H, CH P, / - P = 9.0 Hz), 0.27 (s, 12H, SiMe ). P{ ^H} NMR (C6D ): 5 -10.5 (400 Hz peak 2  2  31  H  2  width at half height).  3 1  P MAS NMR:  6  8-10.5 (1:1:1:1:1:1 sext, J .M = 265 Hz). 1  2 7  P  (C6D ): 8 65 (900 Hz peak width at half height).  A l NMR  MS: mle 490 (M+). Anal. Calcd. for  6  C i 8 H 4 C l N P S i (+0.34 equiv LiCl): C, 42.81; H, 8.78; N, 2.77. Found: C, 42.84; H, 8.34; 4  2  2  2  N, 2.48. After an additional recrystallization: Anal. Calcd. for Ci8H44Cl NP Si : C, 44.07; H, 2  2  2  9.04; N, 2.86. Found: C, 44.08; H, 9.02; N, 2.66.  3.8.3.2 AlMe2[N(SiMe CH PPr 2)2] (2) i  2  2  Method 1: To a toluene solution (5 mL) of AlCl [N(SiMe CH PPr ) ] (500 mg; i  2  2  2  2  2  1.02 mmol) was added MeLi (1.50 mL of a 1.4 M E t 0 solution, diluted with 5 mL ether, 2.08 2  mmol) dropwise. The reaction mixture was then stirred for 3 h after which time it was filtered through a frit lined with Celite to remove LiCl. The solvent was then removed in vacuo to yield 2 as a clear colorless oil (398 mg; 87% yield). Purification by recrystallization was not possible either due to the compound's extreme solubility in hydrocarbon solvents or because it was low  128  References begin on page 132  Chapter 3: ...Amidodiphosphine Complexes of Aluminum melting; as a result, characterization by either mass spectrometry or elemental analysis was not possible. Method 2: To Me2AlCl (0.60 mL of a 1.0 M hexanes solution, diluted with 10 mL hexanes; 0.60 mmol) was added LiNtSiN^CF^PPr^te (200 mg; 0.502 mmol) in 10 mL toluene. The reaction mixture was stirred overnight, after which time it was filtered through a frit lined with Celite to remove LiCl. The solvent was then removed in vacuo to yield 2 as a clear colorless oil (174 mg; 72% yield). Method 3: AlCl2[N(SiMe2CH2PPr2)2] (250 mg; 0.51 mmol) was dissolved in i  10 mL THF. The solution was then cooled to -60 °C and Me2Mg(dioxane) (3.7 mL of a 0.14 M THF/Et20 solution; 0.52 mmol) added. The reaction mixture was then warmed to RT and stirred overnight. The THF/Et20 was removed under vacuum and toluene added to the solution, after which time it was filtered through a frit lined with Celite to remove LiCl. The solvent was then removed in vacuo to yield 2 as a clear colorless oil (140 mg; 58% yield). *H NMR (C6D6): 8 1.63 (d of sept, 4H, C / / M e , / - H = 7.5 Hz, / - P = 5.0 Hz), 0.99 and 0.97 (dd, 24H, 3  CUMeMe',  3  /  H  -H  = 7.5 Hz,  3  /  2  2  H  H  - P =  H  10.0 Hz), 0.60 (d, 4H, C// P, 2  2  = 10.0 Hz), 0.44 (s, 12H,  /H-P  SiMe ), -0.24 (t, 6H, AlMe , / H - P = 2.4 Hz). lP{lH} NMR (C6D ): 8 -4.9. 3  2  3  2  6  2 7  A l NMR  (C6D ): 8 70 (4.3 KHz peak width at half height). 6  3.8.3.3 AlEt [N(SiMe CH2PPri )2] (3) 2  2  2  To a toluene solution (5 mL) of AlCl2[N(SiMe2CH PPr 2)2] (200 mg; 0.408 i  2  mmol) was added EtLi (30 mg; 0.83 mmol) in toluene (5 mL). The reaction mixture was left to stir overnight and then filtered through a frit lined with Celite to remove LiCl, followed by solvent removal in vacuo to yield a clear, colorless oil (144 mg; 74% yield). Purification by recrystallization was not possible either due to the compound's extreme solubility in hydrocarbon  129  References begin on page 132  Chapter 3: ...Amidodiphosphine Complexes of Aluminum solvents or because it was low melting; as a result, characterization by mass spectrometry or elemental analysis was not possible. *H NMR (C^Ds): 8 1.64 (d of sept, 4H, C//Me , ^ H - H = 3  2  6.9 Hz, / H - P = 3.5 Hz), 1.42 (t, 6H, AlCH Me, / - H = 7.7 Hz), 0.99 and 0.98 (dd, 24H, 2  3  2  CHMeMe',  3  /  H  = 6.9 Hz,  - H  3  /  - P  H  = 6.3 Hz), 0.60 (d, 4H, C// P,  -4.2.  2 7  2  2  SiMe ), 0.34 (t of q, 4H, AlC// Me, 2  H  2  3  /  H  - H  = 8.4 Hz,  /  3  - P =  H  7 -P H  2.5 Hz).  31  = 7.7 Hz), 0.43 (s, 12H, P{ H) NMR (C D ): l  6  6  8  A l NMR (C D ): 8 140 (5.8 KHz peak width at half height). 6  6  3.8.3.4 Al(CH Ph) [N(SiMe CH PPri ) ] (4) 2  2  2  2  2  2  A solution of Mg(CH Ph) (THF) (150 mg; 0.431 mmol) in THF (10 mL) was 2  2  2  added to a THF solution (15 mL) of AlCl [N(SiMe CH PPri2) ] (198 mg; 0.404 mmol) at -60 2  2  2  2  °C. The reaction mixture was left to warm to room temperature and stirred overnight, after which the solvent was removed in vacuo. Toluene was then added to the reaction vessel and the mixture filtered through a frit lined with Celite to remove LiCl. The solvent was then reduced in volume to approx. 5 mL. Slow evaporation of the solvent resulted in the formation of pale yellow needles (197 mg; 81% yield). *H NMR (C D ): 8 7.28 (d, 4H, o-Ph, / - H = 7.0Hz), 3  6  6  H  7.20 (t, 2H,p-Vh, / - H = 3.5 Hz), 6.96 (t, 4H, m-Ph, / . = 7.0 Hz), 3.50 (br s, 4H, AlC// Ph), 3  3  H  H  H  2  1.52 (d of sept, 4H, C / / M e , / - H = 7.0 Hz, / - P = 3.8 Hz), 0.89 and 0.85 (dd, 24H, 3  2  CHMeMe',  3  2  H  H  / - H = 7.0 Hz, / - p = 5.6), 0.39 (d, 4H, C// P, / - p = 7.8 Hz), 0.33 (s, 12H, 3  2  H  H  2  H  SiMe ). 31P{1H} NMR ( C D , 20 °C): 8 -3.6, (CD C1 -73 °C): 8 -5.4 and -6.8 (s), (d -THF, 2  6  -86 °C): 8 -4.7 and -7.1 (s).  6  27  2  2>  8  A1 NMR (C D ): 8 54 (5.6 KHz peak width at half height). MS: 6  6  mle 510 (M+ - Bz). Anal. Calcd. for C H AlNP Si -(+ 0.39 equiv MgCl ): C, 60.14; H, 9.15; 32  58  2  2  2  N, 2.19. Found: C, 60.11; H, 9.56; N, 1.94.  130  References begin on page 132  Chapter 3: ...Amidodiphosphine Complexes of Aluminum  3.8.3.5 Al(Me)CI[N(SiMe2CH PPr 2) ] (5) i  2  2  To a hexanes solution (5 mL) of L i N t S i M e ^ H ^ P r ^ (204 mg; 0.511 mmol) 1  was added MeAlCl (69 mg; 0.61 mmol) in hexanes (5 mL) dropwise. The reaction mixture was 2  then stirred overnight, after which time it was filtered through a frit lined with Celite to remove LiCl. The solvent was then removed in vacuo to yield a clear colorless oil (144 mg; 59% yield). Purification by crystallization was hampered by contamination of small amounts of 1 and 2. *H NMR (C D ): 8 1.71 and 1.62 (d of sept, 4H, C#Me , / - H = 6.6 Hz, / - P = 4.8 Hz), 1.05, 3  6  6  2  2  H  H  1.02, 0.95 and 0.92 (dd, 24H, CUMeMe', / - H = 6.6 Hz, / - P = 10.0 Hz), 0.68 and 0.61 (d, 4H, 3  3  H  Ctf P, 2  31  2  /H-P  H  = 12.0 Hz), 0.48 and 0.40 (s, 12H, SiMe ), -0.05 (t, 3H, AlMe , 2  2  3  2.8 Hz).  /H-P=  P{ H} NMR(C6D ): 5-6.6. 1  6  3.8.3.6 Al(Et)Cl[N(SiMe CH PPr ) ] (6) i  2  2  2  2  To a hexanes solution (5 mL) of LiN(SiMe CH PPr ) (198 mg; 0.495 mmol) i  2  2  2  2  was added EtAlCl (70 mg; 0.55 mmol) in hexanes (5 mL) dropwise. The reaction mixture was 2  then stirred overnight, after which time it was filtered through a frit lined with Celite to remove LiCl. The solvent was then removed in vacuo to yield a clear colorless oil (208 mg; 87% yield). Purification by recrystallization was not possible due to the persistence of small amounts of 1 ( < 5%) in the reaction mixture. *H NMR (C6D6): 8 1.70 and 1.64 (d of sept, 4H, C//Me , / H - H = 3  2  7.5 Hz, / - P = 3.4 Hz), 1.41 (t, 3H, AlCH Me, / - H = 8.7 Hz), 1.05, 1.02, 0.96 and 0.94 (dd, 2  3  H  24H, CUMeMe',  2  3  / -H= H  H  7.5 Hz, / - p = 7.3 Hz), 0.67 and 0.65 (d, 4H, CH P, 3  H  2  2  /H-P  = 8.6 Hz),  0.48 and 0.40 (s, 12H, SiMe ), 0.46 (t of q, 4H, AlC// Me, J - H = 8.0 Hz, / - P = 1.6 Hz). 3  2  31  2  3  H  H  P{ H} NMR (C6D ): 8-2.0. 1  6  131  References begin on page 132  Chapter 3: ...Amidodiphosphine Complexes of Aluminum 3.8.3.7 AlCl2[N(SiMe2CH PPr 2)2]'AlCl3(7) i  2  Method 1: To a slurry of AICI3 (88 mg; 0.66 mmol) in toluene (5 mL) was added a 5 mL toluene solution of Al(Et)Cl[N(SiMe2CH2PPr2)2] (150 mg; 0.310 mmol) dropwise. The i  reaction mixture turned slightly amber and was then stirred overnight, after which time it was filtered through a frit lined with Celite to remove LiCl. The solvent was then removed in vacuo and the residue taken up in a minimum amount of toluene (approx. 5 mL). Slow evaporation of the solvent resulted in the formation of colorless plates (157 mg; 81% yield). Method 2: To a slurry of AICI3 (500 mg; 3.75 mmol) in toluene (10 mL) was added a 10 mL toluene solution of LiN(SiMe2CH2PPr 2)2 (498 mg; 1.25 mmol) dropwise. The i  reaction mixture was then stirred overnight, after which time it was filtered through a frit lined with Celite to remove LiCl. The solvent was then removed in vacuo and the residue taken up in a minimum amount of toluene (approx. 5 mL).  Slow evaporation of the solvent resulted in the  formation of colorless plates (615 mg; 78% yield). *H NMR (C6D6): 5 2.11 and 1.48 (d of sept, 4H, C//Me ,  3  2  /H-H  = 7.3 Hz,  2  /  H  - P =  2.0 Hz), 1.56 (br d, 2H, CH P, 2  1.08, 0.86 and 0.81 (dd, 24H, CHMeMe',  3  / -H H  = 7.3 Hz,  3  /H-P=  2  /H-P=  14.1 Hz), 1.09,  7.5 Hz), 0.69 and 0.24 (s, 12H,  SiMe ), 0.26 (d, 2H, C// P, / - P = 13.5 Hz). lP{ lH) NMR (Qd^): 8 0.0 (800 Hz peak width 2  2  2  3  H  at half height) and -11.8 (500 Hz peak width at half height). MS: mle 490 (M+ - AICI3). Anal. Calcd. for Ci H3 Al2Cl NP2Si2: C, 34.65; H, 7.11; N, 2.25. Found: C, 34.32; H, 7.26; N, 2.07. 4  6  5  3.9 References (1)  Fryzuk, M. D.; Haddad, T. S.; Berg, D. J. Coord. Chem. Rev.  (2)  Fryzuk, M. D.; Haddad, T. S.; Berg, D. J.; Rettig, S. J. Pure App. Chem. 1991, 63, 845.  (3)  Fryzuk, M. D. Can. J. Chem. 1992, 70, 2849.  132  1990,99,137.  References begin on page 132  Chapter 3: ..Amidodiphosphine Complexes of Aluminum (4)  Robinson, G. H.; Zhang, H.; Atwood, J. L. Organometallics 1987, 6, 887.  (5)  Robinson, G. H.; Sangokoya, S. A. / . Am. Chem. Soc. 1988,110, 1494.  (6)  Taghoif, M.; Heeg, M. J.; Bailey, M.; Dick, D. G.; Kumar, R.; Hendershot, D. G.; Rahbarnoohi, H.; Oliver, J. P. Organometallics 1995,14, 2903.  (7)  Atwood, J. L.; Robinson, K. D.; Jones, C ; Raston, C. L. / . Chem. Soc, Chem. Commun. 1991, 1697.  (8)  Bott, S. G.; Alvanipour, A.; Morley, S. D.; Atwood, D. A.; Means, C. M.; Coleman, A. W.; Atwood, J. L. Angew. Chem., Int. Ed. Engl. 1987,26, 485.  (9)  Atwood, J. L.; Bott, S. G.; Harvey, S.; Junk, P. C. Organometallics 1994,13, 4151.  (10)  Engelhardt, L. M.; Kynast, U.; Raston, C. L.; White, A. H. Angew. Chem., Int. Ed. Engl. 1987,26, 681.  (11)  Bott, S. G.; Elgamal, H.; Atwood, J. L. J. Am. Chem. Soc. 1985,107, 1796.  (12)  Self, M. F.; Pennington, W. T.; Laske, J. A.; Robinson, G. H. Organometallics 1991,10, 36.  (13)  Knajazhansky, S. Y.; Nomerotsky, I. Y.; Bulychev, B. M.; Belsky, V. K.; Soloveichik, G. L. Organometallics 1994,13, 2075.  (14)  Davidson, M. G.; Lambert, C ; Lopez-Solera, I.; Raithby, P. R.; Snaith, R. Inorg. Chem. 1995,34, 3765.  (15)  Krannich, L. K.; Watkins, C. L.; Schauer, S. J. Organometallics 1995,14, 3094.  (16)  Sangokoya, S. A.; Pennington, W. T.; Robinson, G. H.; Hrncir, D. C. / . Organomet. Chem. 1990,385,23.  133  References begin on page 132  Chapter 3: ...Amidodiphosphine Complexes of Aluminum (17)  Muller, G.; Lachmann, J.; Rufinska, A. Organometallics 1992,11,2970.  (18)  Schmidbaur, H.; Lauteschlager, S.; Muller, G. / . Organomet. Chem. 1985,281, 33.  (19)  Wierda, D. A.; Barron, A. R. Polyhedron 1989, 8, 831.  (20)  Wells, R. L.; McPhaul, A. T.; Self, M. F.; Laske, J. A. Organometallics 1993,12, 3333.  (21)  Sangokoya, S. A.; Lee, B.; Self, M. F.; Pennington, W. T.; Robinson, G. H. Polyhedron 1989, 5, 1497.  (22)  Bennet, F. R.; Elms, F. M.; Gardiner, M. G.; Koutsantonis, G. A.; Raston, C. L.; Roberts, N. K. Organometallics 1992,11, 1457.  (23)  Barron, A. R. / . Chem. Soc, Dalton Trans. 1988, 3047.  (24)  Janik, J. F.; Duesler, E. N.; McNamara, W. F.; Westerhausen, M.; Paine, R. T. Organometallics 1989, 8, 506.  (25)  Petrie, M. A.; Ruhlandt-Senge, K.; Power, P. P. Inorg. Chem. 1993, 32, 1135.  (26)  Waggoner, K. M.; Power, P. P. / . Am. Chem. Soc 1991,113, 3385.  (27)  Atwood, D. A.; Rutherford, D. Organometallics 1995,14, 3988.  (28)  Moise, F.; Pennington, W. T.; Robinson, G. H. J. Coord. Chem. 1991,24, 93.  (29)  Trepanier, S. J.; Wang, S. Organometallics 1994,13,2213.  (30)  Steiner, A.; Stalke, D. Organometallics 1995,14, 2422.  (31)  Vliet, M. R. P. v.; Buysingh, P.; van Koten, G.; Vrieze, K.; Kojic-Prodic, B.; Spek, A. L. Organometallics 1985,4,1701.  134  References begin on page 132  Chapter 3: ..Amidodiphosphine Complexes of Aluminum (32)  Janik, J. F.; Duesler, E. N.; Paine, R. T. Inorg. Chem. 1987,26, 4341.  (33)  Sangokoya, S. A.; Moise, F.; Pennington, W. T.; Self, M. F.; Robinson, G. H. Organometallics 1989, 8, 2584.  (34)  Robinson, G. H.; Sangokoya, S. A. Organometallics 1988, 7, 1453.  (35)  Healy, M. D.; Barron, A. R. / . Am. Chem. Soc. 1989, 111, 398.  (36)  Choquette, D. M.; Timm, M. J.; Hobbs, J. L.; Rahim, M. M.; Ahmed, K. J.; Planalp, R. P. Organometallics 1992,11, 529.  (37)  Fryzuk, M. D.; Carter, A.; Westerhaus, A. Inorg. Chem. 1985,24, 642.  (38)  Delpuech, J. J. NMR of Newly Accessible Nuclei; Academic Press: New York, 1983; Vol. 2.  (39)  Akitt, J. W. Prog. NMR Spec. 1989, 27, 1.  (40)  Fryzuk, M. D.; Giesbrecht, G. R.; Rettig, S. J. Organometallics, 1996,15, 3329.  (41)  Sierra, M. L.; Mel, V. S. J. d.; Oliver, J. P. Organometallics 1989, 8, 2313.  (42)  Fryzuk, M. D.; Mylvaganam, M.; Zaworotko, M. J.; MacGillivray, L. R. Polyhedron 1995, 75,689.  (43)  Wehmschulte, R. J.; Ellison, J. L.; Ruhlandt-Senge, K.; Power, P. P. Inorg. Chem. 1994, 33, 6300.  (44)  Cowley, A. H.; Isom, H. S.; Decken, A. Organometallics 1995,14,2589.  (45)  Lalama, M. S.; Kampf, J.; Dick, D. G.; Oliver, J. P. Organometallics 1995,14, 495.  (46)  Mel, V. S. J. D.; Oliver, J. P. Organometallics 1989,8, 827.  135  References begin on page 132  Chapter 3: ..Amidodiphosphine Complexes of Aluminum (47)  Jerius, J. J.; Hahn, J. M.; Rahman, A. F. M. M.; Mols, O.;flsley,W. FL; Oliver, J. P. Organometallics 1986,5, 1812.  (48)  Rahman, A. F. M. M.; Siddiqui, K. F.; Oliver, J. P. Organometallics 1982,1, 881.  (49)  Beachley, O. T.; Victoriano, L. Organometallics 1988, 7, 63.  (50)  Robinson, G. H.; Sangokoya, S. / . Am. Chem. Soc. 1987,109, 6852.  (51)  Robinson, G. H.; Sangokoya, S. A.; Moise, F.; Pennington, W. T. Organometallics 1988, 7,1787.  (52)  Fryzuk, M. D.; Leznoff, D. B.; Rettig, S. J. Organometallics 1995,14, 5193.  (53)  Lewinski, J.; Pasynkiewicz, S.; Lipowski, J. Inorg. Chim. Acta 1990,178, 113.  (54)  Schonberg, P. R.; Paine, R. T.; Campana, C. F. / . Am. Chem. Soc. 1979,101, 7726.  (55)  Atwood, J. L.; Butz, K. W.; Gardiner, M. G.; Jones, C.; Koutsantonis, G. A.; Raston, C. L.; Robinson, K. D. Inorg. Chem. 1993,32, 3482.  (56)  Heine, A.; Stalke, D. Angew. Chem., Int. Ed. Engl. 1992,31, 854.  (57)  Mole, T.; Jeffrey, E. A. Organoaluminum Compounds; Elsevier: Amsterdam, 1972.  (58)  Fisher, J. D.; Wei, M. Y.; Willett, R.; Shapiro, P. J. Organometallics 1994,13, 3324.  (59)  Wehmshulte, R. J.; Power, P. P. Inorg. Chem. 1994,33, 5611.  (60)  Harris, R. K. Nuclear Magnetic Resonance Spectroscopy; Pitman Books Ltd: London, 1983.  (61)  Beachley, O. T.; Victoriano, L. Inorg. Chem. 1986,25,1948.  136  References begin on page 132  Chapter 3: ...Amidodiphosphine Complexes of Aluminum  (62)  Shapiro, P. J.; Vij, A.; Yap, G. P. A.; Rheingold, A. L. Polyhedron 1995,14, 203.  (63)  Yang, X.; Stern, C. L.; Marks, T. J. / . Am. Chem. Soc. 1991,113, 3623.  (64)  Yang, X.; Stern, C. L.; Marks, T. J. / . Am. Chem. Soc. 1994,116, 10015.  (65)  Brookhart, M.; Grant, M.; Volpe, J. A. F. Organometallics 1992,11, 3920.  (66)  Wannagat, U.; Blumenthal, T.; Brauer, D. J.; Burger, H. / . Organomet. Chem. 1983,249, 33.  (67)  Ziegler, K. Organometallic Chemistry; Reinhold: New York, 1960.  (68)  Brookhart, M.; Green, M. L. H. / . Organomet. Chem. 1983,250, 395.  (69)  Dryden, N. Ff.; Legzdins, P.; Trotter, J.; Yee, V. C. Organometallics 1991,10, 2857.  137  References begin on page 132  Chapter 4  Synthesis, Characterization and Solution Dynamics of Alkali-Metal Chloride, Aluminate and Borate Adducts of the Tridentate Amidodiphosphine Ligand Precursor LiN(SiMe2CH2PPr'2)2  4.1 Introduction Lithium amide compounds, LiNR (R = alkyl, aryl or silyl), have proven to be ubiquitous 2  in organometallic chemistry as amide transfer reagents and also as mild deprotonation reagents for organic synthesis. Recently, in an attempt to better explain the reactivity of organolithium 1  reagents, structural studies have been undertaken to probe the extent and nature of their solvation as well as the degree of aggregation present in solution. " 2  15  These studies have revealed a  remarkable diversity of bonding types in solution and the solid state, with solvent-separated ion pairs, monomers, dimers, trimers, tetramers, larger oligomers, polymers, ladders and cages all being known. " 1,16  18  What becomes evident from these reports is that the coordination geometry  of organolithium reagents is not straightforward, and can often vary between the solution and the solid states. Research in our group has been concerned with the use of the lithium amidodiphosphine reagent, LiN(SiMe2CH2PR2)2, as a starting material for the introduction of a potentially tridentate ligand to both early and late transition metals, and to some main group elements. " While this 19  28  ligand system imbues crystallinity on many of its complexes, the actual lithium precursors have 138  References  begin  on page  175  Chapter  4:  ..Adducts  ofLiN(SiMe2CH2PP1*2)2  so far not yielded single crystals to allow crystallographic analyses. However, we have been able to isolate and crystallographically characterize LiCl, LiAlMe4 and NaBEu adducts of LiN(SiMe2CH2PPr 2)2, in which three different coordination modes were observed. Here we i  provide details of these solid state structures and include variable temperature NMR studies to illustrate that the integrity of these compounds is not always maintained in solution.  4.2.1 Synthesis of {LiN(SiMe CH PPr )2}2LiCl (1) i  2  2  2  The original preparation of the lithium amidodiphosphine reagent LiN(SiMe2CH2PPri2)2 requires the addition of three equivalents of LiPPr^ to a THF solution of l,3-bis(chloromethyl)tetramethyldisilazane (equation 4.1); two equivalents of the phosphide are necessary to displace 19  the chlorides and form the phosphorus-carbon bonds while the third equivalent deprotonates the amine.  Me  Me2  2  3 LiPPr's H  CI  n  A. CI  •  LiN(SiMe CH PPr )2 x± u i  2  2  2  -78°C (- 2 LiCl)  L 4 , A J  (- HPPr 2) J  The *H NMR spectrum consists of peaks attributable to a C2v symmetric complex with no evidence of any phosphorus-lithium interaction in both the variable temperature P{ H} and 31  7  1  Li{ H} NMR spectra (singlets are observed at all temperatures). The microanalytical data 1  support the above formulation. All of our attempts to grow crystals suitable for X-ray diffraction were foiled due to the ease with which this compound melts near or slightly above room  139  References  begin  on page 175  Chapter 4: ...Adducts ofLiN(SiMe2CH2PPr 2)2 i  temperature. However, in one particular preparation, several large colorless crystals could be isolated from a saturated hexanes solution maintained at -40 °C. On the basis of NMR spectroscopy, elemental analysis and, most convincingly X-ray crystallography, it became apparent that these crystals, which we label as compound 1 for convenience, were different from the material originally reported as LiN(SiMe2CH2PPr 2)2. i  The *H NMR spectrum of 1 is nearly identical to that of authentic LiN(SiMe2CH2PPr 2)2, i  with the exception that the resonances due to the methylene protons (SiC//2P) arid the methine protons of the isopropyl groups (PC//(CH3)2) are noticeably broadened. The P{ H} NMR 31  1  spectrum, however, consists of a broad singlet at -3.8 ppm which is indicative of an interaction of phosphorus-31 with a quadrupolar nucleus, in this case lithium-7. This is in contrast to the sharp singlet at -4.7 ppm due to LiN(SiMe2CH2PPri2)2. The combustion analysis of the crystals of 1 is consistent with a complex incorporating a molecule of LiCl.  Ultimately, an X-ray  crystallographic study was necessary to determine the nature of the LiCl interaction in this compound.  4.2.2 Structure of {LiN(SiMe CH2PPr' )2}2LiCl 2  2  (1)  Cooling a saturated hexanes solution of 1 to -40 °C for 2 weeks resulted in large colorless blocks that were suitable for single crystal X-ray diffraction. The molecular structure and numbering scheme are illustrated in Figure 4.1. The structure of 1 can best be described as a three-rung ladder, in which a LiCl unit is sandwiched between two LiN(SiMe2CH2PPri2)2 monomers. The formulation of 1 as {LiN(SiMe2CH2PPr 2)2)2LiCl is further supported by the i  microanalytical data. Compound 1 exhibits pseudo C2 symmetry with the axis passing through the Li(3)-Cl(l) axis. The structural features are similar to those of the four other three-rung ladders of LiCl and two lithium amides previously reported. ' ' 3  140  29  30  References begin on page 175  Chapter 4: ...Adducts  Figure 4.1  ofLiN(SiMe CH PPr 2)2 i  2  2  Molecular structure of { L i N ( S i M e C H P P r ) } L i C l (1); 33% probability thermal i  2  2  2  2  2  ellipsoids are shown. Hydrogen atoms are omitted for clarity.  Compound 1 contains three lithium atoms in two distinct environments: Li(l) and Li(2) maintain approximate tetrahedral geometries in which bonding to the amide nitrogen and two neutral phosphines of a L i N ( S i M e C H P P r ) molecule is observed, in addition to the planar u i  2  2  3  2  2  chloride associated with L i C l , while Li(3) resides in a distorted trigonal planar environment. The bridging lithium Li(3) binds to the chloride and two amido anions. The atoms in the three-rung ladder, L i ( l ) , Cl(l), Li(2), N(2), Li(3) and N(l), define a slightly twisted plane in which the mean deviation from the plane is 0.1022 A. Examination of the structure shows that 1 involves a L i C l molecule linking two L i N ( S i M e C H P P r ) 2 units together. However, closer inspection reveals i  2  2  2  four similar lithium-nitrogen bond lengths (2.06(1) - 2.10(1) A ) , and a lengthened Li(3)-Cl(l)  141  References begin on page 175  Chapter 4: ..Adducts ofLiN(SiMe2CH2PPr 2)2 i  interatomic distance of 2.403(10) A, which is longer than the distance from the chloride to each of the lithium atoms associated with the LiN(SiMe2CH2PPri2)2 units (2.336(10) and 2.333(10) A). These data reflect the inherent stability of the ladder form in alkali metal chemistry. ' 6  12,16,31  " , 33  and have implications with regard to the mechanism of formation of 1 (see later). No unusually short intramolecular contacts were noted. A list of bond lengths and angles for 1 can be found in Table 4.1.  Table 4.1 Selected bond lengths (A) and angles (deg) for {LiN(SiMe2CH2PPr 2)2}2LiCl (1). i  PG)-Li(l)  2.59(1)  PQ)-Li(l)  2.63(1)  P(3)-Li(2)  2.61(1)  P(4)-Li(2)  2.58(1)  N(l)-Li(l)  2.06(1)  N(l)-Li(3)  2.08(1)  N(2)-Li(2)  2.10(1)  N(2)-Li(3)  2.10(1)  Cl(l)-Li(l)  2.34(1)  Cl(l)-Li(2)  2.33(1)  Cl(l)-Li(3)  2.40(1)  P(l)-Li(l)-P(2)  117.9(4)  P(3)-Li(2)-P(4)  118.9(4)  Li(l)-Cl(l)-Li(2)  145.2(4)  Li(l)-Cl(l)-Li(3)  72.0(3)  Li(2)-Cl(l)-Li(3)  73.2(4)  Li(l)-N(l)-Li(3)  84.6(4)  N(l)-Li(3)-N(2)  161.0(5)  Li(2)-N(2)-Li(3)  84.6(4)  Cl(l)-Li(l)-N(l)  102.5(4)  142  References begin on page 175  Chapter  4:  ...Adducts  ofLiN(SiMe2CH2PPr 2)2 i  4.2.3 Variable Temperature NMR Studies of {LiNtSiMeiCHiPPr^hhLiCl (1) The solution *H NMR spectrum of 1, recorded in Q D 6 , is in accord with the structure in the solid state. Each LiN(SiMe2CH2PPr 2)2 unit possesses local C i  2v  symmetry, and the pseudo  mirror plane that is defined by the L i - C l bond renders each LiN(SiMe CH2PPr 2)2 ligand i  2  equivalent. The broadened resonances associated with the methylene (SiCH P) and methine 2  (PC//(CH3>2) protons reflect a kinetic process in which the phosphine arms rapidly dissociate and recoordinate to the lithium atom bound to the central amide. This postulate is further supported by the variable temperature P{ H} NMR data (Figure 4.2); at room temperature, a broad singlet 31  l  at -3.8 ppm is observed for 1, which upon cooling (in C7D8) to -27 °C, decoalesces into a 1:1:1:1 quartet at -6.2 ppm. The quartet, a result of coupling to a single quadrupolar lithium-7 nucleus (I = 3/2, natural abundance = 92.6%), displays a typical Vp-Li coupling of 48.5 H z .  34  At room temperature the ^ i ^ H } NMR spectrum also consists of a single broad resonance at -0.7 ppm, even though two distinct lithium environments are present in the solid state. Upon lowering the temperature gradually to -38 °C, one observes broadening of the singlet and decoalescence into two resonances of 2:1 intensity: a triplet at -0.4 ppm (Vp-Li = 48.5 Hz) and a broad singlet at -1.4 ppm (Figure 4.3). The triplet, which can be attributed to the two lithium ions from the LiN(SiMe2CH2PPr 2)2 units, coincides with the decoalescence behavior i  observed in the P{ H} spectra, while the broad singlet can be assigned to the lithium bridged 31  1  between the two amide donors. At first glance, one might conclude that the Li{ H} NMR 7  1  spectra reflect the averaging of the two lithium environments on the NMR time scale, perhaps by lithium site exchange or via the rapid association and dissociation of the phosphine arms to Li(l) and Li(2), as well as Li(3). However, it should be noted that phosphine dissociation cannot by itself average the two environments if the three rung ladder is maintained in solution. This is illustrated in Scheme 4.1; if one assumes that the ladder structure is maintained in solution, then phosphine migration from one type of lithium to the other never exchanges the two different lithium environments present.  143  References  begin  on page 175  Chapter 4: ...Adducts ofLiN(SiMe2CH2PPr 2)2 i  22 °C  -7 °C  13 °C  -12 °C  3 °C  •17 °C  -2*C  -27  Figure  °C  4.2 Variable temperature 121.4 M H z 3lp{ lH} N M R spectra of PPr 2)2}2LiCl i  (1) in C7D8.  {LiN(SiMe2CH 2  The peak marked with an asterisk (*) is due to  LiN(SiMe CH2PPri2)2. 2  144  References begin on page 175  Chapter 4: ..Adducts ofLiN(SiMe2CH2PPr 2)2 i  Figure  4.3 Experimental (left) and simulated (right) variable temperature 116.7 MHz Li{!H} 7  NMR spectra of {LiN(SiMe2CH PPr )2}2LiCl (1) in C D . i  2  145  2  7  8  References begin on page 175  Chapter 4: ..Adducts ofLiN(SiMe2CH2PPr 2)2 i  Scheme 4.1  C  B  (isopropyl groups at phosphorus omitted for clarity)  This was confirmed by the  7  L i NOESY spectrum of the  15  N-labeled  derivative,  {Li N(SiMe2CH PPr 2) }2LiCl, l b , run in C D at -50 °C. As shown in Figure 4.4, no off15  i  2  2  7  8  diagonal peaks were observed upon varying t [ from 5 to 50 ms, which previous work found to m  x  be sufficient for observing dynamic processes in the slow-exchange limit for analogous systems. " 35  37  146  References begin on page 175  Chapter 4: ...Adducts ofLiN(SiMe2CH2PPr 2)2 i  l l l l l l l ' l l l l l l l l  -2.0  -1.0  0.0  1.0  Figure 4.4 194.5 MHz L i NOESY NMR spectrum of {Lil5N(SiMe CH PPri2)2}2LiCl (lb) 7  2  2  in C7D8 at 223 K. The small coupling present is due to L i - N coupling (i/u-isN 7  1 5  = 5.8 Hz).  The low temperature L i NOESY NMR spectrum illustrates that lithium site-exchange does not 7  occur to any appreciable extent and that the structural integrity of 1 is retained in solution. Other LiCl adducts such as {(|i -Cl)Li H4[(Me3SiNCH CH2)3N]2} also retain their ladder structures 3  3  even up to 65 ° C .  2  29  147  References begin on page 175  Chapter 4: ...Adducts ofLiN(SiMe2CH2PPr 2)2 l  From the variable temperature Li{ H} NMR data, we were able to derive kinetic 7  1  parameters regarding the apparent exchange of three lithium atoms over two sites. The Arrhenius plot and rate data are available in Table 4.2 and Figure 4.5. From the Arrhenius plot, AG^c*, AH* and AS* were determined to be 13 ± 1 kcal mol , 11 ± 1 kcal mol and -7 ± 8 cal K" mol" , -1  -1  1  1  respectively, at T = 1 °C. The value of AS*, although small and subject to considerable error, is c  further evidence of the retention of the ladder conformation in solution and complements the L i 7  NOESY results. If the positions of the lithium atoms in 1 were exchanged by a process whereby the ladder structure was lost, one would expect a much different value for AS*. That the lithium environments are averaged by a non-intrusive mechanism such as phosphine dissociation and recoordination is supported by the minimal value determined for AS*.  7  o o ©  1/T  Figure 4.5 Arrhenius plot for the apparent exchange of lithium centres in {LiN(SiMe CH PPri2)2}2LiCl (1) (r = 0.979). 2  2  148  References begin on page 175  Chapter 4: ..Adducts ofLiN(SiMe2CH2PPr 2)2 i  Table 4.2 Arrhenius data for the apparent exchange of lithium centres in {LiN(SiMe2CH PPr 2)2}2LiCl (1). i  2  temp, K  1/T  k, s-  Ink  294  0.00340  600  6.40  275  0.00364  200  5.30  265  0.00377  90  4.50  260  0.00385  40  3.69  255  0.00392  20  2.99  245  0.00408  10  2.30  235  0.00426  5  1.61  1  Yet the fact remains that at the high temperature limit of the variable temperature Li{ H} 7  J  NMR spectra (Figure 4.3), a broadened singlet is observed. This is unusual for two reasons: (i) there should be two environments for the two different types of lithiums present in the ladder structure even at high temperature, and (ii) the coupling information from the phosphorus-31 nuclei has been lost. To rationalize the apparent observation of only one environment, we suggest that this is a result of the phosphine exchange process and the relatively small chemical shift range for L i . As shown in Scheme 4.1, in the fast exchange limit, each lithium can be found 7  with two, one or no coordinated phosphine ligands. Given the ionic nature of lithium compounds, and the fact that the small chemical shift range for lithium-7 is not that sensitive to environmental changes, the apparent observation of a single environment in the high temperature limit is reasonable. The other unusual aspect is that this resonance appears as a singlet, in other words, coupling information from the phosphorus-31 nuclei has been lost. For a purely intramolecular process that involves fast phosphine dissociation and reassociation, coupling information should still be present albeit statistically weighted to reflect residence time. However, in this case,  149  References begin on page 175  Chapter 4: ...Adducts ofLiN(SiMe2CH2PPr 2)2 i  because L i is quadrupolar, as the temperature is raised quadrupolar relaxation becomes fast with 7  respect to /u-P resulting in the formation of a singlet, arid not the expected triplet.  38  We undertook other heteronuclear NMR spectroscopic studies on 1 in hopes of further confirming that the structure determined in the solid state was also indicative of its solution geometry.  The  2 9  Si{ H} 1  NMR  spectrum  of  the  labeled  derivative  {Li N(SiMe2CH2PPr 2)2}2LiCl (lb) exhibited only a broad singlet (8 -17.2 ppm) which can be 15  i  viewed as indicative of either a compound with C \ symmetry, or a fluxional process on the NMR 2  time scale. Similar results were garnered from the N{!H} NMR spectra, which maintained 15  sharp singlets at both room temperature and -50 °C (8 5.1 ppm). The lack of coupling from N 1 5  to L i in the ^ N p H } NMR spectra is expected based on the quadrupolar moment of L i . 7  7  3 9  However, the Li{ !H} NMR spectrum obtained at -50 °C (194.4 MHz) exhibited a small V T J - I S N 7  coupling of 5.8 Hz. Although all organolithium compounds are believed to be predominantly ionic with small covalent contributions^ it is evident that this small coupling points to some 40  degree of covalent contribution to the Li-N bond. Even though we were unable to obtain a 9  suitable solution molecular weight measurement of 1 (Signer apparatus, isopiestic method), we conclude that the NMR spectroscopic data indicate that the basic structural features of 1 are maintained in solution, and the crystal structure determined is not merely a chance feature of aggregation.  4.2.4 Mechanism of Formation of {LiN(SiMe CH PPr )2}2LiCl (1) i  2  2  2  The serendipitous discovery of 1 may be the result of incomplete removal of THF from the original reaction mixture. We suggest that the presence of small quantities of THF serves to both solubilize the LiCl coproduct and solvate the lithium amide unit, presumably in a dimeric form. Although an early report suggested that structures such as 1 arose from the trapping of a LiCl molecule by two lithium amide monomers, recent work argues that three-rung ladders are a 30  result of the incorporation of a LiCl molecule into an open (R2NLiS )2 dimer x  15Q  3,15,41  (Scheme 4.2).  References begin on page 175  Chapter 4: ...Adducts ofLiN(SiMe2CH2PPr 2)2 i  Scheme 4.2  Additionally, we suggest that the solvent molecules, in this case THF, are displaced by the more Lewis-basic phosphines of the tridentate LiN(SiMe2CH2PPr 2)2} 2 ligand to render a solvent-free i  structure. Although the structure of 1 is free from any THF interaction, if one considers the solvation of lithium by PRPr^, these results are in line with the MNDO calculations carried out by Collum and Gade, who found a strong preference for cyclic oligomers over three-rung ladders in the unsolvated form, but the reverse upon solvation. " 41  43  One can also generate 1 from the  addition of LiCl to L i N C S i ^ ^ C ^ P P r ^ (2 equiv) in either THF or toluene; the resultant crystalline solids exhibited identical *H and P{ H} NMR spectra as 1, although no attempt was 31  1  made to further analyze them. The three-rung ladder structure of 1 does not hinder further metathesis chemistry, however, as reaction with metal halides, M X i results in the expected n +  MX [N(SiMe2CH PPr 2)2] products (equation 4.2). 26  i  n  27  2  151  References begin on page 175  Chapter 4: ...Adducts ofLiN(SiMe2CH2PPr 2)2 i  {LiN(SiMe CH PPr 2) }2LiCI + 2 MCI^THF), i  2  2  2  toluene  M = Al, x = 0 M = Sc, x = 3 [4.2]  2 MCI (THF) [N(SiMe CH PPr ) ] + 3 LiCl i  2  x  2  2  2  2  M = Al, x = 0 M = Sc, x = 1  As a final note, we should clarify that the original preparation of LiN(SiMe2CH2PPr 2)2 19  i  is accurate as long as all the THF from the preparation is removed before extraction of the product with hexanes; in this way, all of the LiCl precipitates and formation of the LiCl adduct 1 is prevented.  4.3.1 Synthesis of {LiN(SiMe CH PPr 2)2»LiAlMe4}2 (2) i  2  2  While examining the reaction chemistry of aluminum derivatives of the general formula AlR2[N(SiMe2CH2PPri2) ] (R = Cl, Me, Et, Bz) (see Chapter 3), we serendipitously discovered 26  2  the formation of yet another adduct of LiN(SiMe2CH2PPr 2)2. i  Upon the addition of four  equivalents of MeLi to the starting aluminum dichloride, AlCl2[N(SiMe2CH2PPr )2], we i  2  obtained a clear colorless oil that could be crystallized from toluene. The *H NMR spectrum of this compound exhibits characteristic ligand backbone resonances in addition to a large singlet at -0.21 ppm that corresponds to 12 protons. The upfield position of this peak, and the absence of a triplet in the same vicinity (due to six methyl protons coupling to two equivalent phosphorus-31 nuclei, as is found for AlMe2[N(SiMe2CH2PPr 2)2]) points towards an aluminate type structure. i  Also, the P{ H} NMR spectrum consists of only a single sharp peak at -3.5 ppm. Previously, 31  l  152  References begin on page 175  Chapter 4: ..Adducts ofLiN(SiMe2CH2PPr 2)2 l  we have found that phosphines bound to quadrupolar nuclei (i.e., S c , I = 7/2; A l , I = 5/2) 45  2 7  display broadened signals in the absence of fast exchange (see Chapter 2).  26,27  The microanalysis  of this product is also consistent with the empirical formulation LiN(SiMe2CH2PPr 2)2»LiAlMe4. i  Additionally, if L i N ^ i f t ^ C F ^ P P r ^ is added to a slurry of LiAlMe4 (1 equiv) in toluene, a product exhibiting identical H , P{ H} and Li{ H} NMR spectra is obtained (Scheme 4.3). 1  31  1  7  1  All of these factors point towards 2 as being a L i A l M e 4 adduct of LiN(SiMe2CH2PPr )2. The i  2  isolation of crystals allowed us to undertake a structural study of 2 and to show that it has a dimeric molecular formulation.  Scheme 4.3  AICI [N(SiMe CH PPr ) ] + 4Mel_i ,  2  2  2  2  LiN(SiMe CH PPr ) + LiAIMe  toluene  i  2  2  toluene  2  2  2  r  \  PH ">! 2  Me SiS,  \  2  <  Li  Me AILi--4  Me Sif 2  4  PH  2  2  153  References begin on page 175  2  Chapter 4: ...Adducts of  LiN(SiMe CH PPr 2)2 i  2  2  4.3.2 Structure of {LiN(SiMe2CH PPr 2)2»LiAlMe }2 (2) i  2  4  Evaporation of a saturated toluene solution of 2 resulted in large colorless plates that were suitable for single crystal X-ray diffraction. The molecular structure and numbering scheme are illustrated in Figure 4.6.  Figure 4.6 Molecular structure of {LiN(SiMe2CH2PPr 2)2 LJAlMe }2 (2); 33% probability i  ,  4  thermal ellipsoids are shown. Hydrogen atoms are omitted for clarity.  The structure is made up of two LiN(SiMe2CH2PPr 2)2*LiAlMe units which form a 1  4  dimer with C2 symmetry.  The dimer may also be viewed as two [Li2N(SiMe CH2PPr 2)2] i  2  cations bridged by two [AlMe ] anions. Each [N(SiMe2CH2PPri2)2]~ ligand chelates a lithium 4  154  References begin on page 175  Chapter 4: ...Adducts ofLiN(SiMe2CH2PPr 2)2 i  cation in a geometry intermediate between facial and meridional. The coordination of each [N(SiMe2CH2PPr 2)2]" ligand to a lithium rather than an aluminum is noteworthy, particularly i  since aluminum was the central metal in the starting material. Each lithium in the dimer maintains a roughly tetrahedral geometry:  Li(l) is coordinated by two phosphines of a  [N(SiMe2CH2PPr2)2]" ligand and an amide, in addition to a carbon associated with a bridging i  [AlMe4] unit. The deviation from ideal is evident in the P(l)-Li(l)-P(2) angle of 124.1(3)°, and _  the two P-Li(l)-N(l) angles of 96.3(3)° and 96.9(3)°. The phosphorus-lithium bond lengths of 2.538(7) and 2.675(7) A in 2 are both similar to those found in 1. The distances determined between lithium and both nitrogen and carbon are well within normal values associated with these types of bonds. The second lithium atom resides in a much less conventional geometry: Li(2) shares short contacts with the same amide as Li(l), two carbons, three hydrogens and an aluminum atom. The distorted environment around the Li(2) atom is also apparent in an almost linear N(l)-Li(2)-Al(l') angle of 164.6(4)°. If one discounts the contacts present with three hydrogens (two on C(20) and one on C(21)), Li(2) can be said to reside in a greatly distorted tetrahedral environment, in which the two carbon atoms and the aluminum atom are greatly bent back from the Li(2)-N(l) bond. It may be that the steric bulk of the silyl methyl groups of the [N(SiMe2CH2PPr2)2]" ligand prevents a more ideal geometry from being attained. It is well i  known that the environment of the small lithium cation is governed primarily by steric factors.  1,44  The bridging tetrahedral [AlMe4] groups present no unusual bond angles or bond lengths. A -  partial list of bond lengths and bond angles for 2 are presented in Table 4.3.  155  References begin on page 175  Chapter 4: ...Adducts ofLiN(SiMe2CH2PPr 2)2 i  Table 4.3 Selected bond lengths (A) and angles (deg) for {LiN(SiMe2CH PPr 2)2»LiAlMe4} 2 (2). i  a  2  a  PQ)-Li(l)  2.538(7)  P(2)-Li(l)  2.675(7)  P(3)-Li(3)  2.656(7)  P(4)-Li(3)  2.556(7)  N(l)-Li(l)  2.075(8)  N(l)-Li(2)  1.988(9)  N(2)-Li(3)  2.076(8)  N(2)-Li(4)  1.989(8)  Al(l)-Li(2)'  2.736(8)  Al(2)-Li(4)"  2.714(8)  P(l)-Li(l)-P(2)  124.1(3)  P(3)-Li(3)-P(4)  123.9(3)  P(l)-Li(l)-N(l)  96.9(3)  P(2)-Li(l)-N(l)  96.3(3)  P(3)-Li(3)-N(2)  96.3(3)  P(4)-Li(3)-N(2)  97.7(3)  Li(l)-N(l)-Li(2)  107.1(3)  Li(3)-N(2)-Li(4)  105.2(3)  Symbols refer to symmetry operations: (') 1-x, l-y, -z; (") 2-x, -y, 1-z.  4.3.3 Variable Temperature NMR Studies of {LiN(SiMe CH PPr )2»LiAlMe4}2 (2) i  2  2  2  The fluxional behavior of {LiN(SiMe2CH2PPr )2 LiAlMe4}2 (2) was studied by variable i  ,  2  temperature NMR spectroscopy. All of the methyl groups of the tetramethyl aluminate moiety were found to be equivalent since the resonance corresponding to these protons (*H NMR spectra) remained a singlet down to the solvent limit (-90 °C, C7D8), with no coupling to phosphorus or lithium being observed. The shift of this resonance was also found to be solvent dependent (-0.21 ppm in C6D6, -0.38 ppm in C7D8) which is perhaps indicative that the [AlMe4]" group is mobile in solution. Previous work with lithium amides has shown that the quadrupolar interactions present ( Li, N) often prevent well defined spectra from being attained. " 7  14  4,5,7  14  Thus  we performed P { H } , L i { ! H } and ^Nf !!} NMR experiments on the labeled derivative 3 1  1  6  1  156  References begin on page 175  Chapter 4: ...Adducts ofLiN(SiMe2CH2PPr 2>2 i  { Li N(SiMe2CH2PPr 2)2» LiAlMe4)2 (2b), which was synthesized in an analogous fashion to 2 6  15  6  i  from AlCl2[ N(SiMe CH2PPr 2)2] and CH36LL 15  i  2  The P { !H} NMR spectra of 2b experienced no change as the temperature was lowered 31  from 20 °C to -88 °C in C7D8; the sharp singlet present at room temperature persisted at all temperatures and shifted upfield to -8.6 ppm (from -3.5 ppm) at the low temperature limit. This behavior is similar to that observed for a series of aluminum dialkyls of this ligand system which we have attributed to the highly fluxional character of electropositive main group compounds of N(SiMe2CH2PPri2)2 (see Chapter 3). The room temperature Li{ H} NMR spectrum of 2b in 6  l  CyDs/pentane is also a sharp singlet (8 1.4 ppm) (Figure 4.7). Upon cooling to -48 °C, the singlet gradually broadens and evolves into a triplet of doublets at 1.2 ppm ^/qj-p = 10.8 Hz, ^ I J - I S N = 3.2 Hz) due to the coupling of both lithium atoms to two phosphines and a single N nucleus (I = 1 5  1/2). In this intermediate temperature regime, the presence of only one multiplet is consistent with fast lithium exchange between the two equivalent sites. Further cooling (-78 °C) results in the loss of all coupling information to regenerate a broad singlet. Finally, at -108 °C, two resonances of equal intensity are evident: a triplet of doublets at 1.9 ppm (^Li-p = 19.9 Hz, J6Ul  i5N  = 2.0 Hz) due to a lithium bound to two phosphines and an amide of a  [ N(SiMe2CH2PPr 2)2] ligand, and a doublet at 0.7 ppm ( V S L I - I S N = 4.0 Hz) arising from a 15  i  _  lithium atom bound only to the amide nitrogen.  Thus, the low temperature Li{ H} NMR spectrum of 2b is similar to that of 1, and is in 6  l  accord with the solid state structure, in which two distinct lithium environments were determined. However,  the  solution  data  LiN(SiMe2CH2PPr 2)2 LiAlMe4. i  ,  are also consistent with  the  monomeric  adduct,  From the decoalescence behavior of 2b between -78 and -108  °C, the following activation parameters were determined: AGr * = 9 ± 1 kcal m o l , AH* = 8 ± 1 -1  c  kcal mol" and AS* = -4 ± 8 cal K" mol" at T = -79 °C. The Arrhenius plot and rate data are 1  1  1  c  presented below (Figure 4.8, Table 4.4).  157  References begin on page 175  Chapter 4: ..Adducts ofLiN(SMe2CH2PPr 2)2 i  Figure 4.7 Variable temperature 44.1 MHz Li{ R) NMR spectra of 6  l  {6Lil5N(SiMe CH PPri2)2* LiAlMe4}2 (2b) in C D8/pentane. 6  2  2  7  158  References begin on page 175  Chapter 4: ...Adducts ofLiN(SiMe2CH2PPr 2)2 i  1/T  Figure 4.8 Arrhenius plot for the exchange of lithium centres in { Li 5N(SiMe2CH2PPr 2)2» LiAlMe }2 (2b) (r = 0.995). 6  1  i  6  4  Table 4.4 Arrhenius data for the exchange of lithium centres in { Li 5N(SiMe2CH2PPr 2) » LiAlMe4} 2 (2b). 6  1  i  6  2  temp, K  1/T  k, s-  Ink  195  0.00514  150  5.01  190  0.00527  75  4.32  185  0.00541  30  3.40  181  0.00522  20  3.00  174  0.00573  10  2.30  169  0.00590  5  1.61  165  0.00607  1  0.69  1  159  References begin on page 175  Chapter 4: ...Adducts ofLiN(SiMe2CH2PPr 2)2 i  The similarity of these numbers to those determined for the chloride adduct 1 suggest that the mechanism of lithium exchange in this low temperature range is also a result of phosphine dissociation and re-association. Such a process is outlined in Scheme 4.4.  Scheme 4.4 Me  —  2A  S  i  k \  n  "Me  Me* —Al-  Pr' p'\/ 5 N 2  Y V.  Pr^P*  t  Me  Me M  Me S  I  1/2 Me'  / \ ...^  PPra  I I I I  U  2  V  .Me  Me  \  | Me, " Si-  s  I  -Al.... | "'Me Me  -Me  Me  M  Li  Al M&2 Si  /  X/Si Me  M  A. Me * S i V  v  2  (solid state)  Me  2  YV  .PPr'  Me  2  MeMe.  -Al- - M e  D'  Me  As mentioned above, the variable temperature ^if^H} NMR spectra point towards a monomeric form being present at most temperatures in solution. Additional support for this comes from the room temperature solution molecular weight determination (Signer apparatus, isopiestic method, 430 ± 40 g mol ) of 2b which corresponds to a simple 1:1 L i A l M e 4 adduct of -1  LiN(SiMe2CH2PPr 2)2. Since the dimer is held together in the solid state by two relatively weak i  interactions that involve only one methyl group of the aluminate moiety and the lithium bound to the phosphine donors, the formation of the monomeric form D is reasonable (Scheme 4.4). At  160  References begin on page 175  Chapter 4: ...Adducts ofLiN(SiMe2CH2PPr 2)2 i  room temperature, the molecule is highly fluxional, with rapidly dissociating phosphines and exchanging lithium centers.  Upon cooling to -48 °C, rapid lithium exchange is still occurring,  although coupling to phosphorus-31 and nitrogen-15 is evident. To exchange the lithiums, a symmetrical intermediate such as E can be envisioned since it would involve dissociation of one arm of the ligand from one Li and re-coordination to the other concomitant with the [AlMe4]moiety bridging the two lithiums. Cycling through structures D, E and D' provides the necessary pathway to exchange the lithium centers as well as the methyls on Al. The ^Lif^H) NMR spectrum at -108 °C indicates that two lithium types are present, one of them not exhibiting lithium-phosphorus coupling and therefore consistent with the monomeric form D or D' in the slow exchange limit. It should be pointed out that the fluxional process detailed in Scheme 4.4 is only applicable to the temperature regime below -48 °C; above this temperature, loss of coupling from 6  L i to both  1 5  N and P is observed which suggests that an intermolecular process becomes 3 1  operative that exchanges lithium centers. One possibility is dissociation of the monomeric units D or D' into L i A l M e 4 and LiN(SiMe2CH2PPr 2)2. In contrast to that previously discussed for the i  ladder structure of 1, the loss of coupling for 2b at higher temperatures is not due to quadrupolar relaxation. Because 2b is the lithium-6 labeled material, the lower quadrupolar moment of L i (I 6  = 1, -0.0008 Barns) as compared to L i (I = 3/2, -0.04 Barns), used in 1, effectively removes 7  45  this as a mechanism for relaxation. The results of the Si{ H} and N{ R) NMR studies were similar to those found for the 29  1  15  1  labeled lithium chloride adduct lb. The Si{ H} NMR spectrum of 2b consists of a singlet at 29  l  -18.5 ppm, which again is consistent with a highly symmetric molecule (as in the crystal structure of 2), or a highly fluxional molecule. The ^Nf !!} NMR spectra at ambient temperature (5 27.3 1  ppm) or at -108 "C both showed no N - L i coupling. 1 5  6  The behavior of 2 was found to vary greatly upon dissolution in coordinating solvents. In the H NMR spectrum in ds-THF, the [AlMe4] resonance, a singlet at -0.21 ppm in C 6 D 6 , l  _  161  References begin on page 175  Chapter 4: ...Adducts ofLiN(SMe2CH2PPr 2)2 l  becomes a 1:1:1:1:1:1 sextet at -1.36 ppm which does not change upon 31p decoupling. This pattern, due to aluminum-proton coupling ( 7H-A1 = 6.1 Hz) is indicative of a very symmetric 2  environment around the A l nucleus and is identical in shift and coupling to that of a dg-THF solution of LiAlMe4. The resonances due to LiNCSiM^C^PPr^^ are shifted slightly upfield and sharpened considerably. As with L i A l M e 4 , these resonances are analogous to those of a simple dg-THF solution of L i N C S i N ^ C ^ P P r ^ .  From these data, one can conclude that  coordinating solvents such as T H F break up the adduct and results in what is likely three discrete species in solution (equation 4.3), as has been observed for other lithium amides.  46  The Li{ H] 6  l  NMR spectrum of 2b in dg-THF was comprised of a singlet at 0.1 ppm, which most likely reflects an averaged signal from both LiN(SiMe2CH PPri2)2 and [Li(THF) ]+[AlMe4]-. We did not 2  x  probe the nature of this exchange any further.  LiN(SiMe CH PPr )2 j  2  2  2  d -THF  [4.3]  8  [Li(THF) ][AIMe ] x  4  4.3.4 Mechanism of Formation of { L i N ( S i M e C H P P r ) » L i A l M e } (2) i  2  2  2  2  4  2  The formation of a lithium aluminate adduct of LiN(SiMe CH PPri2) upon reaction of 2  2  2  AlCl2[N(SiMe2CH2PPr )2] with four equivalents of MeLi may be rationalized if one allows for i  2  the insertion/addition of MeLi into/across the aluminum-amide bond. As previously reported, the first two equivalents of MeLi form the dimethyl complex AlMe2[N(SiMe2CH2PPrJ2)2]; the third equivalent of MeLi then either inserts into the aluminum-amide bond to form 3 or adds across the aluminum-amide bond to form 3' (Scheme 4.5). The fourth equivalent of MeLi then results in the  162  References begin on page 175  Chapter 4: ...Adducts ofLiN(SiMe2CH2PPr 2)2 l  formation of the product 2. In either case, the final product is the result of an aluminum atom being displaced from the [N(SiMe2CH2PPr 2)2]" coordination sphere in favor of lithium. To test i  our  hypothesis,  we added three equivalents  of MeLi to the starting dichloride  AlCl2[N(SiMe2CH2PPr 2)2]. The U NMR spectrum of the product obtained showed a singlet l  i  corresponding to nine protons at -0.30 ppm and a sharp singlet in the P{ ^H) NMR spectrum at 31  -4.0 ppm; these data are best rationalized using structure 3 as the intermediate rather than 3' since the latter would be expected to exhibit a methyl resonance with phosphorus-31 coupling in the H l  NMR spectrum and a broadened P{ H} NMR spectrum due to the presence of aluminum-27. 31  1  We have previously suggested a structure similar to 3 in earlier iridium amidodiphosphine chemistry.  47  Although no resonances indicative of the alternative structure 3' were evident, the  initial formation of this complex and its rapid conversion to 3 cannot be ruled out. Scheme 4.5  • MeLi  Me AI—-N—Li 3  Me Si** 2  f Pr'  \  ,»*'Vlfc  1  N — A l \  2  3  •  Me AILi—;.N—Li 4  v  Me£\f  I Pr'  MeLi  f PHg  2  2  MeLi  163  References begin on page 175  Chapter 4: ..Adducts ofLiN(SiMe2CH2PPr 2)2 i  4.4.1 LiAlEt4, LiBEt4 and NaBEt Adduct Synthesis 4  The ease with which we were able to generate a lithium aluminate adduct of LiN(SiMe2CH2PPr 2)2 by the addition of LiAlMe4 prompted us to examine the potential of other i  alkali metal -ates as complexing agents. Thus, toluene solutions of LiNCSiM^CF^PPr^te were added to slurries of LiAlEt4, LiBEt4 and NaBEty in toluene to examine the effect of the systematic variation of the cation (Li to Na), the tripositive metal of the anion (Al to B) and the alkyl group present (Me to Et). In all three cases, addition of the ligand system resulted in the slow dissolution of the alkali metal -ate complex, which we view as being indicative of the formation of a {LiN(SiMe2CH2PPr 2)2«AER } (A = Li, Na, E = Al, B, R = Me, Et) type of i  4  x  adduct. In the case of LiAlEty (4) and LiBEt4 (5), the product could only be isolated as a yellow oil; all attempts to generate the adducts as solids failed. Interestingly, by analogy with 2, the addition of four equivalents of EtLi to AlCl2[N(SiMe CH.2PPr 2)2] i  2  LiN(SiMe2CH.2PPr2)2  and  i  did not  form the  LiAlEt4;  we  same product as  only  observed  the  the  reaction of  formation  of  AlEt2[N(SiMe2CH2PPr 2)2]. It is probable that the larger ethyl group is unable to effectively i  insert into the aluminum amide bond as we propose is likely for the smaller methyl derivative. The Yi NMR spectra of 4 and 5 are as expected, with no coupling to lithium present in l  any of the resonances, as is the case for 2. The ethyl resonances in 4 remain a triplet (-CH2C//3) and quartet (-C//2CH3) even at -90 °C (C7D8) and point to the same rapid tumbling behavior on the NMR time scale as was found for the [AlMe4]- group in 2. The broadened ethyl resonances in 5 are the result of the close proximity of the ethyl groups of the [BE14]- unit to both the quadrupolar lithium and boron atoms. The singlets in the P{ H) and Li{ H) NMR spectra in 31  l  7  l  spite of these broadened H NMR signals are further evidence of the dynamic nature of these l  adducts. Even though we could not obtain these compounds as solids, the uptake of LiAlEt4 and LiBEt4 by toluene solutions of LiN(SiMe2CH2PPr 2)2 as well as the donation of the amide lone i  pair to an additional lithium atom in the structures of both 1 and 2 point towards the existence of  164  References begin on page 175  Chapter 4: ..Adducts ofLiN(SiMe2CH2PPr 2)2 i  the adducts 4 and 5. However, in the absence of structural confirmation, we did not attempt variable temperature P{ *H} or Li{ H) NMR spectroscopic studies of these species. 31  7  l  The addition of LiN(SiMe2CH2PPr 2)2 to a toluene slurry of NaBEty resulted in a yellow i  oil, the *H NMR of which was similar to that of 5 in that many of the alkyl resonances were greatly broadened, even with B decoupling (equation 4.4). As for 2 - 5, the P{ H} NMR was 1 1  31  1  of no use in establishing the formation of an adduct, although the solvation of the NaBEt4 upon addition of ligand solution was again positive evidence in this regard. In this case, crystals were isolated which allowed us to further probe the extent and nature of this interaction.  LiN(SiMe CH PPr ) i  2  2  2  2  +  NaBEt  Me Si£  toluene  2  ^  EUBNa-fiN—Li Me Si^ f P. Pr' ^  4  [4.4]  2  2  4.4.2  Structure of {LiN(SiMe2CH2PPr>2)2'NaBEt4} (6) x  Evaporation of a saturated toluene solution of 6 resulted in large colorless plates that were suitable for single crystal X-ray diffraction. The molecular structure and numbering scheme are illustrated in Figure 4.9. The structure is best described as being a loosely associated onedimensional polymer, the repeating unit of which contains two LiN(SiMe2CH2PPr 2)2 NaBEt4 i  ,  fragments. The first [N(SiMe2CH2PPr2)2]" ligand coordinates to a lithium in a tridentate fashion, i  such that the metal center maintains a roughly tetrahedral geometry (92.5(4)° - 128.0(4)°), its coordination sphere completed by a weak interaction with three hydrogens on a neighboring  165  References begin on page 175  Chapter 4: ...Adducts ofLiN(SiMe2CH2PPr 2)2 i  [BEuY anion (Li(2) - H(124*, 125*, 126*) = 2.27 - 2.34 A). The stereochemical importance of achieving coordinative saturation at an electron deficient metal center via metal-alkane hydrogen interactions has long been known.  48  The bond lengths from the lithium atom to the  [N(SiMe2CH2PPri2)2]" ligand are similar to those determined for 1 and 2. As in the structures previously discussed, the amide nitrogen is more tetrahedral than planar, owing to the donation of electrons to an alkali metal, in this case sodium. The N(2)-Na(2) distance of 2.447(5) A is ~ 0.43 A longer than the distance from this nitrogen to Li(2). The sodium atom serves to bridge the LiN(SiMe2CH2PPr )2 and [BEty] units via ion-dipole interactions with the four methylene i  -  2  protons of two ethyl groups of the  [BEt4]-  group. These bond distances range from 2.44 to 2.59  A, indicative of the larger atomic radius of Na as compared to Li. Another close contact (2.48 A) is present between Na(2) and H(120*), which originates from the same [ B E t 4 ] group as the initial _  Li-H contacts. An additional interaction with H(124*) (2.69 A), which coordinates to Li(2), results in an unusual seven-coordinate sodium ion. The central [BEty] group exhibits local C -  2v  symmetry and is normal in terms of usual borate geometries.  This general scheme is then  49  repeated in reverse to yield a LiN(SiMe2CH PPr 2)2-Na-BEt4-Na-LiN(SiMe2CFl2PPr 2)2 chain. i  i  2  However, the second sodium atom differs from the first in that it is now nine-coordinate. In addition to interactions with the methylene protons as seen for Na(2), Na(l) now not only shares the electrons of N(l) from the second [N(SiMe2CH2PPr )2]~ ligand, but also those of two i  2  hydrogens from a third [BEt4]- unit in addition to two silyl methyl protons arising from the ligand system, although the increased bond lengths point to a weaker interaction than the [BEt4]  _  methylene protons. The geometry of the second LiN(SiMe2CH PPr 2)2 ligand is similar to that of i  2  the first, the most notable exception being the increased P-Li-P angle of 136.1(4)°. As before, the second lithium fulfills its coordination sphere with the three donors of the tridentate ligand and three ion-dipole interactions with neighboring [ B E t 4 ] - hydrogens.  166  References begin on page 175  Chapter 4: ...Adducts of  LiN(SiMe CH PPr 2)2 i  2  2  a)  b)  Figure 4.9.a) Molecular structure of {LiN(SiMe2CH PPr )2»NaBEt4}x (6); 33% i  2  2  probability  thermal ellipsoids are shown. Hydrogen atoms are omitted for clarity, b)  Chem 3D view of the monomer unit of { L i N ( S i M e 2 C H P P T ) 2 » N a B E t 4 } i  2  167  2  x  References begin on page 175  (6).  Chapter 4: ..Adducts ofLiN(SiMe2CH2PPr 2)2 i  The overall structure of 6 may best be viewed as [LiNCSiN^CF^PPr^te-Na-BEu-NaLiN(SiMe2CH2PPri2)2] chains bridged perpendicularly by [BEt4]- groups. +  50,51  Notably, the  "bridging" [BEty] units differ from the "chain" [BEty]- groups in that the former coordinate to -  two lithium atoms as well as two sodium atoms, whereas the latter only interact with two sodium atoms. It is clear from examination of the crystal determination that this structure is held together by predominantly ion-dipole interactions and the donation of electrons from the [N(SiMe2CH PPri2)2]" ligand amide lone pairs to adjacent sodium ions. A partial list of bond 2  lengths and bond angles for 6 is presented in Table 4.5.  Table 4.5 Selected bond lengths (A) and angles (deg) for {LiN(SiMe2CH PPr )2»NaBEu} (6). i  2  2  x  P(l)-Li(l)  2.57(1)  P(2)-Li(2)  2.60(1)  P(3)-Li(2)  2.59(1)  P(4)-Li(2)  2.60(1)  N(l)-Li(l)  2.05(1)  N(2)-Li(2)  2.03(1)  N(l)-Na(l)  2.431(5)  N(2)-Na(2)  2.447(5)  P(l)-Li(l)-P(2)  136.1(4)  P(3)-Li(2)-P(4)  128.0(4)  P(l)-Li(l)-N(l)  94.9(4)  P(2)-Li(l)-N(l)  92.7(4)  P(3)-Li(2)-N(2)  96.2(5)  P(4)-Li(2)-N(2)  92.5(4)  The polymeric nature of 6 is most likely the result of aggregation upon solvent evaporation; in other words, this structure is not maintained in solution. The solution molecular weight of this compound (Signer apparatus, isopiestic method, 520 ± 50 g mol ), which -1  corresponds to a simple 1:1 adduct of LiN(SiMe2CH2PPr 2)2 and NaBEt4, coupled with the i  decreased ease with which crystals of this compound can be redissolved after isolation are further proof of the polymeric nature of 6 in the solid form, but the simple monomeric form that 6  168  References begin on page 175  Chapter 4: ...Adducts ofLiN(SiMe2CH2PPr 2)2 i  maintains in solution. We did not probe the nature of this interaction by variable temperature NMR spectroscopy.  4.5 Similarity of Ligand Geometry and Trends in Adduct Formation Although the crystal structure of LiN(SiMe2CH2PPri2)2 still remains elusive, we have been able to isolate three adducts of our ligand system in which three different modes of coordination are observed: a three-rung ladder with LiCl; a 2:2 dimer with L i A l M e 4 , and an infinite one-dimensional polymer with NaBEt4. It may be that the degree of aggregation depends upon the size of the anion present: CI" (three-rung ladder) < [AlMe4]" (2:2 dimer) < [BEt4]~ (polymer). The presence of the counter-anion in these complexes determines the amount of steric crowding around the corresponding alkali metal and undoubtedly the final stoichiometry. What is striking is the remarkable similarity in the geometry of the LiN(SiMe2CH2PPr' )2 unit in the face 2  of varying coordination modes (see Tables 4.6 and 4.7). With the exception of the small variations in Li-P bond lengths and the P-Li-P bond angles, the geometric parameters can be said to be identical within experimental error.  Table 4.6 Selected Bond Lengths in {LiN(SiMe2CH PPr )2}2LiCl (1), i  2  2  {LiN(SiMe CH PPr )2«LiAlMe4}2 (2), and {LiN(SiMe CH PPr ) »NaBEt4}x (6). i  2  2  i  2  2  2  2  2  Ligand Bond Lengths  {LiN(SiMe2CH2PPr )2)2  {LiN(SiMe2CH2PPri2)2'  {LiN(SiMe2CH2PPrJ2)2*  (A)  LiCl (1)  LiAIMe4}2 (2)  NaBEt4} (6)  Li-N  2.06(1), 2.10(1)  2.075(8)  2.05(1), 2.03(1)  Li-P  2.59(1), 2.63(1)  2.538(7), 2.675(7)  2.57(1), 2.60(1)  i  2  2.61(1), 2.58(1) N-Si  1.702(4), 1.687(5) 1.702(5), 1.694(4)  169  x  2.59(1), 2.60(1) 1.698(3), 1.707(3)  1.701(5), 1.700(5) 1.693(5), 1.690(5)  References begin on page 175  Chapter 4: ...Adducts ofLiN(SiMe2CH2PPr 2)2 i  Table 4.7 Selected Bond Angles in {LiN(SiMe2CH PPr 2)2hLiCl (1), i  2  {LiN(SiMe CH PPr 2)2»LiAlMe4}2 (2), and {LiN(SiMe CH PPr ) «NaBEt } (6). i  2  Ligand Bond Angles  i  2  2  {LiN(SiMe CH PPr ) }2 {LiN(SiMe2CH2PPr )2' i  2  2  i  2  2  2  2  2  2  4  x  {LiN(SiMe2CH2PPr 2)2i  O  LiCl (1)  LiAlMe4)2 (2)  NaBEt4) (6)  P-Li-P  117.9(4), 118.9(4)  124.1(3)  136.1(4), 128.0(4)  P-Li-N  97.1(4), 96.5(4)  96.9(3), 96.3(3)  94.9(4), 92.7(4)  x  97.0(4), 95.5(4) Li-N-Si  107.2(4), 110.7(4)  96.2(5), 92.5(4) 109.9(2), 104.8(2)  107.3(3), 110.1(3)  110.0(4), 111.1(4) 110.8(5), 110.5(4)  4.6 Conclusions Adducts of the lithium salt of the ancillary, potentially tridentate ligand precursor LiN(SiMe2CH2PPr 2)2 were synthesized by the addition of the suitable lithium- or sodium- 'ate i  species to LiNXSiJv^CKkPPr^te. The solution structures of these compounds are postulated to be highly fluxional on the basis of their variable temperature heteronuclear NMR spectroscopic properties. The solid state structures of these compounds were found to vary to that found in solution; the exception is the three-rung ladder lithium chloride adduct 1 for which the solution data indicate a similar structure to that in the solid state.  4.7 Experimental Section 4.7.1 Procedures Unless otherwise stated, general procedures were performed according to Section 2.16.1. ^SipH}, ^ N ^ H } and SLipH} NMR spectroscopy were performed on a Varian XL-300 spectrometer operating at 59.9, 30.3 and 44.1 MHz, respectively. ^SipH} NMR spectra were 170  References begin on page 175  Chapter 4: ...Adducts ofLiN(SiMe2CH2PPr 2)2 i  referenced to external hexamethyldisiloxane in C7D8 (0.0 ppm). ^NpH} NMR spectra were referenced to external MJ.4C1 in D 2 O (0.0 ppm). The reference for the Li{ H} NMR spectra 15  6  1  was external L/C1 (0.0 ppm) in D2O. The L i NOESY NMR spectrum was obtained on a Bruker 6  7  AMX-500 instrument operating at 194.4 MHz in C7D8 at 223 K; t\ was incremented in 512 steps and the data zero-filled to 1024 words before Fourier transformation. 16 scans were recorded for each increment with r jx = 5 or 50 ms with a relaxation time of 800 ms. Simulations of the m  variable temperature L i and L i spectra were performed by DNMR-SIM. Solution molecular 7  6  weight determinations were carried out in toluene by means of the isopiestic method in a Signer molecular weight apparatus; LiN(SiMe2CH2PPr 2)2 was used as a reference. 52  i  4.7.2 Materials HN(SiMe2CH2Cl)2  and LiPPr^  were prepared by published procedures.  19  Li N(SiMe2CH2PPr )2 and AlCl2[ N(SiMe2CH2PPri2)2] were prepared in a fashion analogous 15  i  to their  1 4  15  2  N congeners from the starting labeled disilazane H N ( S i M e 2 C H 2 C l ) 2 l5  prepared by the addition of CH3CI to an ethereal L i solution. 6  54  53  Me Li was 6  L i A l M e 4 was prepared by the  addition of MeLi to a toluene solution of AlMe3. L i A l E t 4 and LiBEty were prepared in a similar manner with EtLi and either AlEt3 or B E t 3 .  NaBEt4 was obtained from Strem and used  5 1 , 5 5  without further purification.  4.7.3 Syntheses 4.7.3.1 {LiN(SiMe CH PPri )2}2LiCl (1) 2  2  2  To a solution of HN(SiMe CH2Cl) (1.73 g, 7.51 mmol) in 20 mL THF was added 2  2  a 20 mL THF solution of LiPPr^ (2.92 g, 23.5 mmol) at -78 °C, dropwise. The solution was left to stir and warm up to room temperature overnight. The THF was then removed in vacuo and the  171  References begin on page 175  Chapter 4: ..Adducts ofLiN(SiMe2CH2PPr 2)2 i  residue extracted with hexanes (3 x 20 mL).  After filtration through Celite to remove LiCl, the  hexanes were removed to yield an orange oil (2.21 g; 70% yield). This was then taken up in a minimum volume of hexanes and cooled to -40 "C. Large blocks formed upon standing for 2 weeks. *H NMR (C D ): 8 1.76 (d of sept, 4H, C//Me , / H - H = 6.5 Hz, / - P = 3.1 Hz), 1.11 3  6  6  2  2  H  and 1.06 (dd, 24H, CHMeMe', / - H = 6.5 Hz, / H - P = 5.0 Hz), 0.73 (br d, 4H, CH P, 3  3  H  Hz), 0.45 (s, 12H, SiMe ). 2  31  2  2  / - P = 5.5 H  P{ H) NMR (C D , 20 °C): 8 -3.8 (s, 490 Hz peak width at half 1  6  6  height), ( C D , -27 °C): 8 -6.2 (1:1:1:1 q, l/p.u = 48.5 Hz). Li{ H} NMR (QD6, 20 °C): 8 7  7  1  8  -0.7 (s, 50 Hz peak width at half height), (C D , -38 °C): 8 -0.4 (t, 2Li, ' / u - P = 48.5 Hz), -1.4 (br 7  s, ILi, 110 Hz peak width at half height).  8  15  N{!H} NMR (C D ): 8 5.1 (s). ^Si^H} NMR 7  8  (C D ): 8-17.2 (s, 40 Hz peak width at half height). Anal. Calcd. for C36H ClLi3N P Si4: C, 7  8  88  2  4  51.38; H, 10.54; N, 3.33. Found: C, 51.13; H, 10.41; N, 3.50.  4.7.3.2 {LiN(SiMe CH PPr ) »LiAlMe4} (2) i  2  2  2  2  2  Method 1: To a 10 mL toluene solution of AlCl [N(SiMe CH PPr 2) ] (204 mg, i  2  2  2  2  0.416 mmol) was added MeLi (1.20 mL of a 1.4 M solution, diluted with 10 mL ether, 1.68 mmol) dropwise at room temperature. The solution immediately turned cloudy white. The reaction mixture was then stirred for 3 h after which time it was filtered through a frit lined with Celite to remove LiCl. The solvent was then removed in vacuo to yield 2 as a clear colorless oil. This was then taken up in a minimum amount of toluene.  Slow evaporation afforded 2 as  colorless plates (94 mg; 46% yield). Method 2: LiAlMe4 (23 mg, 0.245 mmol) was slurried in approx. 10 mL toluene. To this was added a toluene solution (10 mL) of LiN(SiMe CH PPr 2) (101 mg, 0.253 mmol). i  2  2  2  Upon addition, the LiAlMe4 slowly dissolved until only a clear solution remained. The reaction mixture was then stirred 2 h, after which time the solvent was removed in vacuo. The resulting clear, colorless oil was taken up in a minimum amount of toluene. Slow evaporation of the solvent resulted in the deposition of colorless plates (96 mg; 71% yield). *H NMR (CeD^): 8 172  References begin on page 175  Chapter 4: ...Adducts ofLiN(SiMe2CH2PPr 2)2 i  1.54 (d of sept, 4H, CHM&i, / - H = 7.3 Hz, 7 - P = 3.9 Hz), 0.93 and 0.88 (dd, 24H, CUMeMe', 3  2  H  3/H-H =  7  3  H Z  H  > H - P = 6.5 Hz), 0.49 (d, 4H, C// P, / - P = 7.8 Hz), 0.21 (s, 12H, SiMe ), -0.21 3 /  2  H  2  2  (s, 12H, AlMe ). l H NMR (d -THF): 81.71 (d of sept, 4H, CtfMe^ / - H = 7.6 Hz, / H - P = 2.5 3  4  2  H  8  Hz), 1.05 (dd, 24H, CUMeMe', / - H = 7.6 Hz, / . p = 4.6 Hz), 0.47 (d, 4H, C// P, / - P = 5.0 3  2  3  H  H  H  2  Hz), -0.05 (s, 12H, SiMe ), -1.36 (1:1:1:1:1:1 sext, 12H, A l M e , / - A l = 6.1 Hz). P{lH} NMR 2  2  ( C D , 20 °C): 8 -3.5 (s), (-88 °C): 8 -8.6 (s). 7  31  4  31  8  H  P{!H} NMR (d -THF, 20 °C): 8 -0.3 (s). 8  6Li{!H} NMR (C D , 20 °C): 8 1.4 (s), (C Ds/pentane, -48 °C): 8 1.2 (t of d, USLUP = 10.8 Hz, 7  8  7  l / 6 i - i 5 = 3.2 Hz), (-108 °C): 5 1.9 (t of d, ILi, l/en.p = 19.9 Hz, ^ U - I S N = 2.0 Hz), 0.7 (d, ILi, N  L  i/eu.isN = 4.0 Hz). 6Li{lH) NMR(ds-THF, 20°C): 8 0.1 (s).  15  N{!H} NMR (C D ): 8 27.3 7  8  (s). ^SipH} NMR (C D ): 8-18.5 (s). Mol wt. (Signer apparatus, isopiestic method, toluene): 7  8  430 ± 40; calcd. 493.67. Anal. Calcd. for C H A l L i N P S i : C, 53.53; H, 11.43; N, 2.84. 2 2  5 6  2  2  2  Found: C, 52.67; H, 11.10; N, 2.90.  4.7.3.3 LiN(SiMe CH PPr 2)2*AlMe (3) i  2  2  3  To a 10 mL toluene solution of AlCl [N(SiMe CH PPr ) ] (200 mg, 0.408 mmol) i  2  2  2  2  2  was added MeLi (0.87 mL of a 1.4 M solution, diluted with 10 mL ether, 1.22 mmol) dropwise at room temperature. The solution immediately turned cloudy white. The reaction mixture was then stirred for 3 h after which time it was filtered through a frit lined with Celite to remove LiCl. The solvent was then removed in vacuo to yield 2 as a clear colorless oil (123 mg; 64% yield). This species was only analyzed in solution. H NMR (C6D6): 8 1.56 (d of sept, 4H, C#Me , / H - H = !  3  2  7.8 Hz, / - P = 3.4 Hz), 0.94 and 0.89 (dd, 24H, CUMeMe', / - H = 7.8 Hz, / . p = 6.4 Hz), 0.51 3  2  H  3  H  H  (d, 4H, C / / P , / - P = 7.3 Hz), 0.23 (s, 12H, SiMe ), -0.30 (s, 9H, AlMe ). 2  2  H  (C6D ): 8 -4.0 (s). 7Li{lH} 6  2  N  M  R  3  31  P{lH} NMR  (C6D6): 8 -0.9 (s). Mol wt. (Signer apparatus, isopiestic  method, toluene): 450 ± 40; calcd. 471.70.  173  References begin on page 175  Chapter 4: ..Adducts ofLiN(SiMe2CH2PPr 2)2 i  4.7.3.4 LiN(SiMe CH PPr» )rLiAlEt (4) 2  2  2  4  L i A l E t 4 (113 mg, 0.752 mmol) was slurried in approx. 10 mL toluene. To this was  added a toluene solution (10 mL) of LiN(SiMe CH PPr ) (300 mg, 0.751 mmol). Upon i  2  2  2  2  addition, the L i A l E t 4 slowly dissolved until only a clear solution remained. The reaction mixture was stirred 2 h, after which time the solvent was removed in vacuo, resulting in a yellow oil (384 mg; 84% yield). Purification by recrystallization was not possible either due to the compound's extreme solubility in hydrocarbon solvents or because it was low melting; as a result, characterization by elemental analysis was not possible. H NMR l  C//Me , 3 / . 2  H  = 8.3 Hz, / - p = 3.6 Hz), 1.49 (t, 12H, A l C H C / / , / - H = 7.9 Hz), 0.96 and 2  H  8 1.58 (d of sept, 4H,  (C(DQ): 3  H  0.90 (dd, 24H, CHMeMe',  3  2  3  H  / - H = 8.3 Hz, 7 -P = 4.0 Hz), 0.51 (d, 4H, Ctf P, / - P = 7.1 Hz), 3  H  2  H  2  H  0.19 (s, 12H, SiMe ), 0.03 (q, 8H, A1C// CH , / - H = 7.9 Hz). P{!H} NMR (C D ): 8 -3.6 3  2  2  3  31  H  6  6  (s). Li{lH} NMR(C D ): 8-1.3(s). 7  6  6  4.7.3.5 LiN(SiMe CH PPr>2)2*LiBEt4 (5) 2  2  LiBEt4 (100 g » 0.746 mmol) was slurried in approx. 10 mL toluene. To this was m  added a toluene solution (10 mL) of LiN(SiMe CH PPri2) (300 mg, 0.751 mmol). The reaction 2  2  2  mixture was then stirred 2 h, after which time the solvent was removed in vacuo, resulting in a yellow oil (364 mg; 81% yield). Purification by recrystallization was not possible either due to the compound's extreme solubility in hydrocarbon solvents or because it was low melting; as a result, characterization by elemental analysis was not possible. *H NMR (C^Dg): 8 1.60 (d of sept, 4H, C//Me , / - H = 7.3 Hz, / - P = 2.9 Hz), 1.39 (br t, 12H, B C H C / / , / - H = 7.7 Hz), 3  2  2  H  0.98 and 0.91 (mult, 24H, CHMeMe',  3  2  H  3  3  H  / - H = 7.3 Hz), 0.64 (d, 4H, C# P, / - P = 7.2 Hz), 0.58 2  H  2  H  (mult, 8H, B C / / C H ) , 0.29 (s, 12H, SiMe ). lP{lH} NMR (C6D ): 8 -4.0 (s). Li{lH} NMR 3  2  3  2  7  6  (C D ): 8-1.4(s). 6  6  174  References begin on page 175  Chapter 4: ..Adducts ofLiN(SiMe2CH2PPr 2)2 i  4.7.3.6 {LiN(SiMe2CH PPr )2*NaBEt4}x (6) i  2  NaBEt4  2  (40 mg, 0.274 mmol) was slurried in approx. 10 mL toluene. To this was  added a toluene solution (10 mL) of LiN(SiMe CH PPr ) (105 mg, 0.263 mmol). Upon i  2  2  2  2  addition, the NaBEt4 slowly dissolved until only a clear solution remained. The reaction mixture was then stirred a further 12 h, after which time the solvent was removed in vacuo. The resulting yellow oil was taken up in a minimum amount of toluene. Slow evaporation of the solvent resulted in the deposition of colorless plates (91 mg; 60% yield). *H NMR (CgDg): 8 1.60 (d of sept, 4H, CHMe , ^ H - H = 7.0 Hz, 2 / . = 2.1 Hz), 1.38 (br t, 12H, B C H C / / ) , 1.00 and 0.91 3  2  H  p  2  3  (dd, 24H, CUMeMe', / - H = 7.0 Hz, / - P = 5.2 Hz), 0.56 (d, 4H, C// P, / . p = 5.0 Hz), 0.38 3  3  H  2  H  2  (br q, 8H, B C / / C H ) , 0.19 (s, 12H, SiMe ). 2  3  2  apparatus, isopiestic method, toluene):  31  H  P{!H} NMR (C D ): 8 -4.2 (s). Mol wt. (Signer 6  6  520 ± 50; calcd. 549.67.  Anal. Calcd. for  C H B L i N N a P S i : C, 56.81; H, 11.74; N, 2.55. Found: C, 57.21; H, 11.74; N, 2.41. 22  56  2  2  4.8 References (1)  Sapse, A. M.; Schleyer, P. v. R. Lithium Chemistry; John Wiley and Sons, Inc.: New York, 1995.  (2)  Reich, H. J.; Gudmundsson, B. O. / . Am. Chem. Soc. 1996,118, 6074.  (3)  Henderson, K. W.; Dorigo, A. E.; Liu, Q.; Williard, P. G.; Schleyer, P. v. R.; Bernstein, P. R. / . Am. Chem. Soc. 1996,118, 1339.  (4)  Lucht, B. L.; Collum, D. B. / . Am. Chem. Soc. 1996,118, 2217.  (5)  Lucht, B. L.; Collum, D. B. / . Am. Chem. Soc. 1995,117, 9863.  (6)  Henderson, K. W.; Walther, D. S.; Williard, P. G. / . Am. Chem. Soc. 1995,117, 8680.  (7)  Romesberg, F. E.; Collum, D. B. / . Am. Chem. Soc. 1994,116, 9198. 175  References begin on page 175  Chapter 4: ..Adducts ofLiN(SMe2CH2PPr 2)2 i  (8)  Nichols, M. A.; Waldmuller, D.; Williard, P. G. / . Am. Chem. Soc. 1994,116, 1153.  (9)  Collum, D. B. Acc. Chem. Res. 1993,26, 227.  (10)  Bernstein, M. P.; Romesberg, F. E.; Fuller, D. J.; Harrison, A. T.; Collum, D. B.; Liu, Q.Y.; Williard, P. G. J. Am. Chem. Soc. 1992,114, 5100.  (11)  Romesberg, F. E.; Gilchrist, J. H.; Harrison, A. T.; Fuller, D. J.; Collum, D. B. / . Am. Chem. Soc. 1991,113, 5751.  (12)  Hall, P. L.; Gilchrist, J. H.; Harrison, A. T.; Fuller, D. J.; Collum, D. B. / . Am. Chem. Soc. 1991,113, 9575.  (13)  Galiano-Roth, A. S.; Collum, D. B. / . Am. Chem. Soc. 1989, 111, 6772.  (14)  DePue, J. S.; Collum, D. B. J. Am. Chem. Soc. 1988,110, 5518.  (15)  Williard, T. G.; Salvino, J. M. Tetrahedron Lett. 1985,26, 3931.  (16)  Mulvey, R. E. Chem. Soc. Rev. 1991,209,167.  (17)  Neumann, F.; Hampel, F.; Schleyer, P. v. R. Inorg. Chem. 1995,34, 6553.  (18)  Gregory, K.; Schleyer, P. v. R.; Snaith, R. Adv. Inorg. Chem. 1991,37,47.  (19)  Fryzuk, M. D.; Carter, A.; Westerhaus, A. Inorg. Chem. 1985,24, 642.  (20)  Fryzuk, M. D.; Berg, D. J.; Haddad, T. S. Coord. Chem. Rev. 1990, 99, 137.  (21)  Fryzuk, M. D.; Haddad, T. S.; Berg, D. J.; Rettig, S. J. Pure App. Chem. 1991, 63, 845.  (22)  Fryzuk, M. D. Can. J. Chem. 1992, 70, 2849.  (23)  Fryzuk, M. D.; Haddad, T. S.; Rettig, S. J. Organometallics 1992,11,2967.  176  References begin on page 175  Chapter 4: ..Adducts ofLiN(SiMe2CH2PPr 2)2 i  (24)  Fryzuk, M. D.; Mylvaganam, M.; Zaworotko, M. J.; MacGillivray, L. R. Polyhedron 1995, 75, 689.  (25)  Fryzuk, M. D.; Leznoff, D. B.; Rettig, S. J. Organometallics 1995,14, 5193.  (26)  Fryzuk, M. D.; Giesbrecht, G. R.; Olovsson, G.; Rettig, S. J. Organometallics 1996,15, 4832.  (27)  Fryzuk, M. D.; Giesbrecht, G. R.; Rettig, S. J. Organometallics 1996,15, 3329.  (28)  Fryzuk, M. D.; Mylvaganam, M.; Zaworotko, M. J.; MacGillivray, L. R. Polyhedron 1996, 75, 689.  (29)  Duan, Z.; Young, V. G.; Verkade, J. G. Inorg. Chem. 1995,34, 2179.  (30)  Mair, F. S.; Clegg, W.; O'Neil, P. A. / . Am. Chem. Soc. 1993, 775, 3388.  (31)  Gardiner, M. G.; Raston, C. L. Inorg. Chem. 1995,34, 4206.  (32)  Armstrong, D. R.; Barr, D.; Clegg, W.; Mulvey, R. E.; Reed, D.; Snaith, R.; Wade, K. / . Chem. Soc, Chem. Commun. 1986, 869.  (33)  Armstrong, D. R.; Barr, D.; Clegg, W.; Hodgson, S. M.; Mulvey, R. E.; Reed, D.; Snaith, R.; Wright, D. S. / . Am. Chem. Soc. 1989, 777, 4719.  (34)  Fryzuk, M. D.; Love, J. B.; Rettig, S. J. / . Chem. Soc, Chem. Commun. 1996, 2783.  (35)  Arnold, J. / . Chem. Soc, Chem. Commun. 1990, 976.  (36)  Arnold, J.; Dawson, D. Y.; Hoffman, C. G. / . Am. Chem. Soc. 1993, 775, 2707.  (37)  Briere, K. M.; Dettman, H. D.; Detellier, C. / Mag. Res. 1991, 94, 600.  (38)  Harris, R. K. Nuclear Magnetic Resonance Spectroscopy; Pitman Books Limited: London, 1983. 177  References begin on page 175  Chapter 4: ..Adducts ofLiN(SiMe2CH2PPr 2)2 i  (39)  Brand, H.; Capriotti, J. A.; Arnold, J. Inorg. Chem. 1994,33, 4334.  (40)  Novoa, J. J.; Whangbo, M . H.; Stucky, G. D. / . Org. Chem. 1991,56, 3181.  (41)  Romesberg, F. E.; Collum, D. B. / . Am. Chem. Soc. 1994,116,9187.  (42)  Gade, L. H.; Becker, C ; Lauher, J. W. Inorg. Chem. 1993,32, 2308.  (43)  Hellmann, K. W.; Gade, L. H.; Li, W. S.; McPartlin, M. Inorg. Chem. 1994,33, 5974.  (44)  Schade, C ; Schleyer, P. v. P. Adv. Organomet. Chem. 1988,27,169.  (45)  Fuller, G. H. / . Phys. Chem. Ref. Data 1976,5, 835.  (46)  Gade, L. H.; Mahr, N. / . Chem. Soc, Dalton. Trans. 1993,489.  (47)  Fryzuk, M. D.; Huang, L.; McManus, N. T.; Paglia, P.; Rettig, S. J.; White, G. S. Organometallics 1992,11, 2979.  (48)  Rhine, W. E.; Stucky, G.; Peterson, S. W. / . Am. Chem. Soc 1975, 97, 6401.  (49)  Fryzuk, M. D.; Lloyd, B. R.; Clentsmith, G. K. B.; Rettig, S. J. / . Am. Chem. Soc. 1994, 116, 3804.  (50)  Barr, D.; Snaith, R.; Mulvey, R. E.; Perkins, P. G. Polyhedron 1988, 7, 2119.  (51)  Gerteis, R. L.; Dickerson, R. E.; Brown, T. L. Inorg. Chem. 1964,3, 872.  (52)  Signer, R. Liebigs Ann. Chem. 1930,478, 246.  (53)  Fryzuk, M. D.; MacNeil, P. A. / . Am. Chem. Soc. 1984,106, 6993.  (54)  Schlosser, M. Organometallics in Synthesis; John Wiley and Sons Ltd.: West Sussex, 1994.  (55)  Honeycutt, J. B.; Riddle, J. M. / . Am. Chem. Soc 1961,55, 369. 17 g  References begin on page 175  Chapter 5  Synthesis, Characterization and Reactivity of Macrocyclic Bis(amidophosphine) Complexes of Aluminum, Gallium and Indium  5.1 Introduction The work presented in Chapters 3 and 4 illustrates the challenges of studying early metals complexed with the [PNP] ligand system. Even when present as pendant arms, coordination of the phosphine donors cannot always be assumed. In these cases fluxional processes that involve phosphine dissociation dominate. Thus, in these systems we speculate that the amide unit becomes the dominant ligand with the phosphine donors serving only a minor role.  1  Recent work in our group has extended the scope of hybrid ligands to macrocycles. In 2  these ligands, phosphine dissociation should be less prevalent due to the macrocycle effect.  1  Although metal complexes in which the coordination of only a single phosphorus can result, these compounds are static, and the fluxionality experienced by the acyclic analogues is absent (see later). The [P2N2] ([P2N2J = [PhP(CH SiMe2NSiMe CFi2)2PPh]) ligand system is at least 2  2  an 8 electron donor, imparting additional electronic, as well as steric, stability to the metal centre, as compared to the acyclic analogue, [PNP]. Our study of macrocyclic bis(amidophosphine) complexes MC1[P2N2] of Group 13 metals is an extension of the work on A1[PNP] systems described in Chapter 3. Our rationale in studying aluminum, gallium and indium complexes coordinated by the [P2N2] ligand system was that the bulky macrocycle might allow for the isolation of unusual and/or reactive Group 13 179  References begin on page 237  ChapterS: ...Bis(amidophosphine) Complexes of Al, Ga and In  species. In the course of our studies, we discovered that these metal systems could participate in the inversion of configuration of a phosphorus centre in the macrocycle as well. Thus, this chapter is divided into two parts; first, a kinetic study of the effects of metal coordination upon the facility of phosphorus inversion is presented.  Second, the metathesis of the remaining  chloride to generate non-alkyl metal complexes is oudined.  5.2 Pyramidal Inversion at Phosphorus Facilitated by the Presence of Aluminum, Gallium and Indium Although the pyramidal structure of species such as NH3 is widely accepted, in reality 3  amines are not rigid molecules, but undergo rapid inversion of configuration at the nitrogen centre. This isomerization proceeds via a planar transition state in which the nitrogen is sp -2  hybridized (Scheme 5.1). For amines, this inversion is highly favored; at room temperature, 4  NH3 undergoes this process over 10 times per second. 10  5  Scheme 5.1  For amines, the barrier to this inversion is quite low (~5 kcal mol" ), " and the 1  6  9  conversion of A to B under ambient conditions is rapid. Thus, amines can be incorporated into a macrocyclic array with little regard to stereochemistry, due to the facile interconversion of stereoisomers (Scheme 5.2).  10  180  References begin on page 237  Chapter 5: ...Bis(amidophosphine) Complexes ofAl, Ga and In  Scheme 5.2  On the other hand, the barrier to pyramidal inversion is much higher for the heavier members of Group 15. For trialkylphosphines (PR3), this barrier is usually between 35 and 38 kcal mol- , 1  4,11  " and even higher for the arsenic analogues (ASR3). While this implies that the 14  synthesis of chiral phosphines is possible, ' 11  12,15,16  it also suggests that macrocyclic  polyphosphines can exist as a mixture of diastereomers under ambient conditions (Scheme 5.3).  10,17  "  24  Despite the fact that phosphines are ubiquitous in coordination chemistry, the  number of macrocyclic polyphosphines reported to date pales in comparison to its Group 15 analogue, nitrogen. " 25  27  Scheme 5.3  181  References begin on page 237  Chapter 5: ...Bis(amidophosphine) Complexes of Al, Ga and In  Apart from the synthetic problems typically associated with the preparations of macrocycles,  1,22,23,28  "  31  the use of macrocyclic polyphosphines may be limited due to the  separation that is required of the mixtures of isomers produced. For example, the macrocycle [18]aneP402 (C) is synthesized as a mixture of five isomers (a - e) in a combined yield of 12%. ' 17  19  20  Separation of these isomers was undertaken by ion-exchange chromatography on  aqueous solutions of the nickel(II) derivatives.  The pure isomers were then obtained by  cyanolysis of the nickel(II) complexes.  182  References begin on page 237  Chapter 5: ...Bis(amidophosphine) Complexes of Al, Ga and In  In addition to their arduous synthesis, the presence of stereoisomers clearly complicates the study of the coordination chemistry of these macrocycles, especially since their separation is not trivial and often results in poor yields of the individual isomers.  Although template  procedures have recently been exploited to not only improve yields, stereochemistry,  10,19,34  32,33  but also control  these problems are still significant hurdles for the synthetic chemist.  Previously, our group reported the stereoselective preparation of the new macrocyclic ligand precursors £y/z-Li (THF)[P N ] and flm/-Li (THF)2[P N ].  2  2  2  2  2  2  2  The reaction of these lithium  salts with a variety of first row metal dichlorides leads to the stereospecific formation of the corresponding syn- and anti- macrocyclic complexes, M[P2N ] for M = Cr, Co and N i .  35  2  However, the synthesis of the early metal complex {.S37i-[P2N2]Sc}2(u.-Cl)2 was found to be accompanied by inversion of configuration at phosphorus.  35  The ease of this conversion did not  allow for a kinetic study to be undertaken. Herein, we report the synthesis of the Group 13 complexes a«ft'-MCl[P2N ] (M = Al, Ga) and their conversion to the syn- invertomers. 2  5.3.1 Synthesis of a/ift'-[P N ] Complexes of Aluminum and Gallium 2  2  The reaction of the dilithio salt a/jrf-Li2(THF)2[P2N2] with AICI3 or GaCl3 in toluene at 25 °C for 12 h generates good yields of the monomeric chlorides anri-AlCl[P2N2] (1) and antiGaCl[P2N2] (2) (equation 5.1). The *H NMR spectra of both products are similar and show four singlets arising from the silyl methyl protons (SiC/fj) and two distinct ortho- phenyl resonances. The  31  P { H } NMR spectrum of 1 consists of two singlets; a sharp peak at -38.8 ppm !  characteristic of an uncoordinated phosphorus-31 nucleus within the [P2N2] macrocycle, and a broad resonance upfield at -44.7 ppm, due to the coordinated phosphine (A1/2 = 120 Hz). The broadness of this latter peak is the result of coordination to the quadrupolar A l nucleus ( A 1 , 1 2 7  = 5/2, 100% natural abundance).  36,37  27  The upfield shift upon coordination is similar to that found  for AlCl2[N(SiMe2CH2PPr )2] (Chapter 3). The gallium analogue 2 also exhibits two peaks in i  2  183  References begin on page 237  Chapter 5: ...Bis(amidophosphine) Complexes of Al, Ga and In  the P{ H} NMR spectrum; however, the chemical shifts of the peaks arising from coordinated 31  ]  and uncoordinated phosphine donors are as expected, with the resonance due to the coordinated phosphine downfield of the free phosphine (-36.6 and -38.0 ppm, respectively). As in 1, the peak arising from the coordinated phosphine in 2 is broadened by interaction with a quadrupolar nucleus ( Ga, I = 3/2, 60.4% abundance; G a , I = 3/2, 39.4% abundance) - although to a 69  71  36  37  lesser extent (A 1/2 = 20 Hz) than in the aluminum analogue 1.  Me2  Me2  .Si  COT 7V  SL  N  toluene •Ph  +  MCI3  (-2 LiCl)  .N Si  Si  Me2  Me2  Me .Si.  2  Me Si  2  [5.1]  aA7f/-Li (THF) [P2N2] 2  2  M-*—P—Ph  Ph'  'Si Me  2  a/?f/-MCI[P N ] 2  2  Si Me  2  M = Al  1  Ga 2  While the gallium species 2 was isolated as a crystalline solid, the aluminum derivative 1 resisted crystallization and could only be obtained as an air- and moisture-sensitive oil which  184  References begin on page 237  ChapterS: ...Bis(amidophosphine) Complexes of Al, Ga andln  prevented elemental analyses. Although we were unable to isolate 1 in the solid state, the similarity of the *H and P{ H} NMR spectra of 1 and 2 indicate that the general structural 31  1  features present in the structural analysis of 2 also hold true for 1. Attempts to generate the corresponding awtt'-indium complex, artri-InCl[P N ] were 2  2  unsuccessful. The elevated temperatures needed to metathesize the indium chloride bonds were also sufficient to invert the uncoordinated phosphorus, yielding ^y«-InCl[P2N2] (5).  5.3.2 Structure of antf-GaCl[P N ] (2) 2  2  Evaporation of a saturated toluene solution of 2 resulted in large colorless plates that were suitable for single crystal X-ray diffraction. The molecular structure and numbering scheme are illustrated in Figure 5.1. In 2 the [P2N ] ligand retains an anti- conformation, with coordination 2  of a single phosphorus to the central metal observed. Although both phenyl rings appear to be oriented towards the chloride ligand, the anti- geometry can be seen if one envisions the coordination of the second phosphorus to the gallium atom. This allows the gallium atom to adopt a distorted tetrahedral geometry, which is the typical geometry for lighter members of Group 13.  38  The macrocyclic ligand does impose some distortion as can be seen from the N(l)-  Ga(l)-N(2) angle of 116.7(1)°. The other angles anchored by Ga range from 116.0(1)° to 96.5(1)°. P(l) must distort slightly in order to coordinate to gallium, as is illustrated by the Ga(l)-P(l)-C(13) angle of 125.9(2)°. Otherwise, P(l) and P(2) exhibit typical bond angles, falling within normal ranges associated with tetrahedral geometries. The Ga-Cl bond length of 2.189(1) A is only slightly shorter than those reported for a number of similar macrocyclic systems (i.e., Ga(tetraphenylporphyrinato)Cl = 2.196(2) A , Ga(phthalocyaninato)Cl = 2.217(1) 3 9  A,  4 0  Ga(dibenzotetramethyltetraaza[14]annulene)Cl  = 2.222(2) A ) , but still longer than for 41  Cp*Ga systems.  42  185  References begin on page 237  Chapter 5: ...Bis(amidophosphine)  Complexes ofAl, Ga and In  a)  b)  Figure 5.1.a)  Molecular structure of a«n-GaCl[P2N2] (2); 33% probability thermal ellipsoids are shown. Hydrogen atoms are omitted for clarity,  b) Chem 3 D view of anr/-GaCl[P N ] (2). 2  186  2  References begin on page 237  ChapterS: ...Bis(amidophosphine) Complexes of Al, Gaandln  The Ga(l)-P(l) distance of 2.360(1) A is comparable to the shortest reported galliumphosphorus distance of 2.353(2) A found in GaCl3»PMe3  43  This distance, indicative of the  Lewis acidic nature of the metal centre, is approximately the sum of the covalent radii (~2.30 A).  44  Most neutral phosphine adducts of gallium exhibit Ga-P bond distances in the range of 2.4  - 2.7 A . " 45  48  The Ga-N distances of 1.899(3) and 1.887(3) A are normal in terms of gallium-  amide bond lengths ' 49  50  and are somewhat shorter than those reported for the above mentioned  macrocyclic systems, where some degree of derealization is present. comparison is with the compounds Ga[N(SiMe3)2]3 (1.870(6) A ) ' 4 9  5 1  The most valid  39,40  and GaCl[N(SiMe3)2]2  (1.8344(4) and 1.844(4) A ) in which the Ga-N distances were determined to fall well within the 5 1  established range of 1.85 - 1.92 A for gallium-amide bonds  4 9  A list of selected bond lengths and  angles for 2 can be found in Table 5.1.  Table 5.1 Selected bond lengths (A) and angles (deg) for anri-GaCl[P2N ] (2). 2  Cl(l)-Ga(l)  2.189(1)  P(l)-Ga(l)  2.360(1)  N(l)-Ga(l)  1.899(3)  N(2)-Ga(l)  1.887(3)  Cl(l)-Ga(l)-P(l)  112.49(5)  Cl(l)-Ga(l)-N(l)  114.3(1)  Cl(l)-Ga(l)-N(2)  116.0(1)  P(l)-Ga(l)-N(l)  97.5(1)  P(l)-Ga(l)-N(2)  96.5(1)  N(l)-Ga(l)-N(2)  116.7(1)  Ga(l)-P(l)-C(13)  125.9(2)  5.4.1 Synthesis of syn-[P2^2\ Complexes of Aluminum, Gallium and Indium Good yields of the monomeric chlorides syrt-AlCl[P N2] (3), 5y«-GaCl[P N ] (4) and 2  2  2  .yy«-InCl[P N2] (5), were effected via the reaction of the dilithio-dioxane salt of the ancillary 2  187  References begin on page 237  Chapter 5: ...Bis(amidophosphine) Complexes of Al, Ga andln  ligand .yy«-Li2(dioxane)[P2N2], with slurries of AICI3, GaCl3 and L1CI3 in toluene at 25 °C for 12 h (equation 5.2).  The H NMR spectra of the products are indicative of C v symmetric A  2  complexes: the four silyl methyl proton resonances (SiCHi) found for 1 and 2 are replaced by two peaks of equal intensity in 3, 4 and 5, while only a single ortho- proton signal is now present. The P{ ^H} NMR spectrum of 3, 4 and 5 each consist of a single peak at —42 ppm, 31  whose broadness is mirrored by its quadrupole moment, i.e., In (A1/2 = 400 Hz) > Al (A1/2 = 340 Hz) > Ga (A1/2 = 65 Hz).  Me2  Me2  Me  Me  2  2  syA>Li (dioxane)[P2N2]  [5.2]  2  sy/7-MCI[P N ] 2  2  M = AI  3  Ga In  4  5  Compounds 3 - 5 are colorless, thermally-stable, air- and moisture-sensitive crystalline solids. The *H and P{ R} NMR spectra indicate that 3,4 and 5 have symmetric solution structures. 31  l  Structural studies of 3 and 4 were undertaken to further study the nature of these compounds.  188  References begin on page 237  ChapterS: ...Bis(amidophosphine) Complexes of Al, Ga and In  5.4.2 Structure of syn-A1CI[P N ] (3) 2  2  Evaporation of a saturated toluene solution of 3 resulted in large colorless plates that were suitable for single crystal X-ray diffraction. The molecular structure and numbering scheme are illustrated in Figure 5.2. The structural analysis confirms that 3 is monomeric and illustrates the distorted trigonal bipyramidal geometry of the metal centre, similar to that found for AlCl [N(SiMe CH PPri2) ] (see Chapter 3). 2  2  2  2  Although the preferred geometry for five-  coordinate aluminum is trigonal bipyramidal, macrocyclic nitrogen ligand systems such as phthalocyanin (Pc)  40  and octaethylporphyrin (OEP) induce square pyramidal coordinations 52  from aluminum. The difference between these systems and the [P N ] system can be attributed 2  2  to the more rigid nature of the delocalized N4 ring system in Pc and OEP. As in AlCl [N(SiMe CH PPr ) ], the phosphine donors of 3 occupy the axial positions, i  2  2  2  2  2  defining a P(1)-A1(1)-P(2) angle of 165.72(5)°. The two amides, the Al centre and the chloride define a plane (mean deviation = 0.0033 A) whose members form angles of 121.5(1)°, 118.6(1)° and 119.9(1)° with each other. The metal perches slightly above the plane defined by the two amides and the two phosphines. Both phosphines experience normal tetrahedral geometries with two notable exceptions: P(l) maintains an Al(l)-P(l)-C(13) angle of 131.0(1)°, -20° more than that expected for four-coordinate phosphorus. A similar distortion is evident for P(2) (Al(l)P(2)-C(19) = 128.7(1)°). This bending back of the phenyl rings on the phosphorus donors is an indication of the slight distortion required for the [P N ] ligand system to accommodate the 2  2  small aluminum nucleus.  189  References begin on page 237  Chapter 5: ...Bis(amidophosphine)  Figure 5.2.a)  Complexes ofAl, Ga and In  Molecular structure of jyn-AlCl[P2N2] (3); 33% probability thermal ellipsoids are shown. Hydrogen atoms are omitted for clarity,  b) Chem 3 D view of 5yn-AlCl[P N ] (3). 2  2  190  References begin on page 237  Chapter 5: ...Bis(amidophosphine) Complexes ofAl, Ga and In  The aluminum chloride distance of 2.197(1) A also compares well with those found in AlCl [N(SiMe2CH2PPr 2)2] (2.183(6) A and 2.164(7) A) as well as Al(Pc)Cl (2.179(6) A ) . i  40  2  The aluminum-phosphorus bond lengths of 2.481(1) A and 2.465(1) A are well within the normal range associated with such bonds (-2.4 - 2.8  A ) " and are slightly shorter than those 53  55  reported for AlCl2[N(SiMe2CH PPri2)2] (2.509(7) A and 2.542(7) A). The aluminum-nitrogen 2  distances of 1.886(3) A and 1.903(3) A are normal in terms of aluminum-amide bonds.  51  Table  5.2 lists the pertinent bond lengths and angles for 3 and 4.  Table 5.2 Selected bond lengths (A) and angles (deg) for jyrt-AlCl[P2N2] (3) and syrt-GaCl[P N ] (4). 2  2  Cl(l)-Al(l)  2.197(1)  Cl(l)-Ga(l)  2.2525(9)  P(D-A1(1)  2.481(1)  P(D-Ga(l)  2.4930(8)  P(2)-A1(1)  2.465(1)  P(2)-Ga(l)  2.4695(8)  N(1)-A1(1)  1.886(3)  N(l)-Ga(l)  1.968(2)  N(2)-A1(1)  1.903(3)  N(2)-Ga(l)  1.966(2)  P(1)-A1(1)-P(2)  165.72(5)  P(l)-Ga(l)-P(2)  164.63(3)  N(1)-A1(1)-N(2)  119.9(1)  N(l)-Ga(l)-N(2)  119.4(1)  C1(1)-A1(1)-P(1)  97.44(5)  Cl(l)-Ga(l)-P(l)  98.04(3)  C1(1)-A1(1)-P(2)  96.84(5)  Cl(l)-Ga(l)-P(2)  97.32(3)  C1(1)-A1(1)-N(1)  121.5(1)  Cl(l)-Ga(l)-N(l)  122.06(8)  C1(1)-A1(1)-N(2)  118.6(1)  Cl(l)-Ga(l)-N(2)  118.54(8)  Al(l)-P(l)-C(13)  131.0(1)  Ga(l)-P(l)-C(13)  130.3(1)  Al(l)-P(2)-C(19)  128.7(1)  Ga(l)-P(2)-C(19)  127.9(1)  191  References begin on page 237  Chapter 5: ...Bis(amidophosphine)  Complexes of Al, Ga and In  5.4.3 Structure of syn-GaCI[P2N2] (4)  Colorless plates that were amenable to single crystal X-ray diffraction remained after the evaporation of a saturated toluene solution of 4. The molecular structure and numbering scheme are illustrated in Figure 5.3. 5>'«-GaCl[P2N2J (4) is isomorphous with the aluminum derivative described above. The two compounds crystallize in the same space group and there is only slight variation in the bond lengths and angles of interest (see Table 5.2).  Figure 5.3 Molecular structure of sy/i-GaCl[P2N2] (4); 33% probability thermal ellipsoids are shown. Hydrogen atoms are omitted for clarity.  The five bond lengths involving gallium are 0.082 to 0.0045 A longer than in the aluminum analogue 3. Even though gallium is a heavier element than aluminum, its covalent  192  References begin on page 237  Chapter 5: ...Bis(amidophosphine) Complexes of Al, Ga and In  radius (1.25 A) is actually smaller than that of aluminum (1.3 A ) .  56  Since gallium immediately  follows a row of transition elements, it has 10 d electrons, which are less efficient at shielding the nucleus than s or p electrons. This results in roughly equal covalent radii for Al and Ga. For example, the gallium-carbon bond lengths in (Mes*)2GaH (Mes* = 2,4,6-Bu 3C6H2) (1.995(3) t  and 2.009(3) A )  5 7  are essentially the same as those in (Mes*)2AlH (1.976(6) and 2.007(6) A ) .  58  However, in aluminum or gallium amide complexes, the ionic contribution to the bond must be considered. Since the difference in electronegativity between aluminum and nitrogen is greater than the difference between gallium and nitrogen, the A l - N bond has greater ionic character than the Ga-N bond. This results in a shorter aluminum amide bond.  50  The homoleptic amides  M[N(SiMe ) ]3 (M = A l , Ga) exhibit longer Ga-N bonds (1.868(1) A) than Al-N bonds (1.78(2) 3  A).  51  2  This general trend is also borne out in the series investigated here.  5.5.1 Conversion of anft'-MCI[P2N2l Complexes to sy/i-MCl[P2N2] Complexes (M = Al, Ga) Previous work from our group established that the addition of awft'-Li2(THF)2[P2N2] to ScCl3(THF)3 at room temperature resulted in the formation of {5y«-[P2N2]Sc)2(|J.-Cl)2.  35  That  the syn- compound was produced instead of the anti- analogue was surprising, indicating that pyramidal inversion and subsequent coordination of the free phosphine was occurring. However, the high rate of inversion did not allow us to study the kinetics of this system. Similarly, InCl3 reacts with a«ri-Li2(THF)2[P2N2] at elevated temperatures to yield ,yyn-InCl[P N2] (5) (equation 2  5.3). The reaction between InCl3 and antf-Li2(THF)2[P2N2] at room temperature is incomplete, implying that elevated temperatures are needed to form anri-InCl[P2N2J; these temperatures are also sufficient to invert the phosphorus and form ,yy«-InCl[P2N2].  193  References begin on page 237  ChapterS: ...Bis(amidophosphine) Complexes ofAl, Ga and In  Me  Ph'  Me  2  2  P—•Li L i ^ — P . — P h + Si Me?  Si Me  lnCI  toluene, 70 °C 3  (-2 LiCl)  Me  9  Me  2  [5.3]  2  amALi (THF) [P N ] 2  2  2  2  Ph—P  *Hnr«—P—Ph Si Me?  Si Me  5  syn-lnCI[P N2] 5 2  The isolation of the four-coordinate and- compounds 1 and 2, and the observation that these species could be converted to the five-coordinate syn- compounds 3 and 4 resulted in a more in depth study of the activation parameters of metal-assisted pyramidal inversion (equation 5.4).  Me  -SL .%./ Ph— \ P  Me  ^Si  2  /  V C I ^  Si  M  Me  2  ^ ~ r "  d -toluene 8  P  h A  ^Si  Me2  Me  M = AI  *  [ 5  Me  2  M = Al  Ga 2  4 ]  ^ S i  Me  1  2  \ \ [ Ph—P—^M-—P—Ph . ^Si  2  Me  2  2  3  Ga 4  194  References begin on page 237  Chapter 5: ...Bis(amidophosphine)  Complexes of Al, Ga and In  The first order conversion of arc//-MCl[P N ] (M = A l (1), Ga (2)) to ,yyn-MCl[P N ] ( M 2  2  2  2  = A l (3), Ga (4)) was followed by ' H N M R spectroscopy. Sealed tubes containing ~0.1 M solutions of 1 and 2 in C7D8 were maintained at constant temperatures between 60 and 100 °C (± 2 °C) for predetermined lengths of time. The irreversibility of the reaction (i.e., conversion of five-coordinate syn-MCl[P N ] complexes to four-coordinate anft'-MCl[P N ] complexes was not observed) 2  2  2  2  precluded the need for quenching the samples by freezing prior to H N M R analysis. As well, no l  conversion was observed under ambient conditions. Kinetic data for the two systems can be found in Figures 5.4 and 5.5. Selected data regarding the rates of conversion are presented in Table 5.3. A l l kinetics runs were performed to approximately five half-lives, after which time meaningful integration of the peaks of interest became difficult.  From the rate data we produced Arrhenius  plots (Figure 5.6) in order to compare the activation parameters of the two systems.  Selected  transition state parameters are listed in Table 5.4.  time (hours)  Figure 5.4 First-order rate plots for the conversion of anri-AlCl[P N ] (1) to sy/2-AlCl[P N ] (3) 2  2  2  in C D . 7  8  195  References  begin on page 237  2  Chapter 5: ...Bis(amidophosphine)  Complexes of Al, Ga and In  •  60 °C  O  70 °C  o  80 °C 90 °C 100 °C  time (hours)  Figure 5.5 First-order rate plots for the conversion of anri-GaCl[P2N2] (2) to sytt-GaCl[P2N2] (4) in C D . 7  8  Table 5.3 Rate data for the conversion of anri-MCl[P N ] to .yy/z-MCl[P2N2] (M = A l , Ga) in 2  2  C D . 7  8  M = Al  M = Ga  temp, "C  kobs* S~l  correlation, R  temp, °C  kobs> ^  correlation, R  60  1.77 ± 0.12 x  0.998  60  5.50 ± 0.44 x  0.999  io70  10-7  6  2.78 ± 0.14 x IO"  0.994  1.10 ± 0.13 x  0.999  70  6  80  io90  io100  io-  80  0.999  3.17 ± 0.17 x  0.997  io0.996  90  0.999  100  196  0.997  6  1.79 ± 0.09 x  io-  5  6  4.83 ± 0.39 x  io-  5  6.83 ± 0.69 x  1.62 ± 0.09 x IO" 6  5  1.92 ±0.11 x  s  0.997  5  References begin on page 237  Chapter 5: ...Bis(amidophosphine)  Complexes  ofAl, Ga and In  •  anrJAlCl[P N ]  6  a n r J G a C l [ P N ] tosy>z-GaCl[P N ]  2  2  2  to.ryn-AlCl[P N ] 2  2  2  2  2  1/T  Figure 5.6 Arrhenius plots for the conversion of anri-MCl[P2N2] to 5yn-MCl[P N2] ( M = A l , Ga) 2  in C D . 7  8  Table 5.4 Transition state parameters for the conversion of anft'-MCl[P N ] to 2  2  SVAI-MCIIP2N2]  ( M = A l , Ga) in C D . 7  function Eg, kcal mol"  M = Al 1  A H * , kcal mol" AS*,  1  M = Ga  20.8  ± 1.4  18.2 ± 1.2  20.0  ± 1.4  17.5 ± 1.2  cal K " m o l  1  -24.4 ± 3.0  -33.8  kcal m o l  1  29.1 ± 2 . 7  30.1 ± 2 . 7  0.985  0.985  AG3 3*, 7  8  1  correlation, R  197  ± 2.5  References begin on page 237  Chapter 5: ...Bis(amidophosphine) Complexes of Al, Ga and In  Plots of l n { l / % a>m'-MCl[P N2]} vs. time yielded straight lines at all temperatures 2  observed for both the aluminum and the gallium systems. The first-order nature of the inversion was further confirmed by a concentration study of the aluminum system (Figure 5.7) in which various concentrations of 1 were heated at 80 and 100 °C for 8 h. The invariance of the rate of inversion with concentration points to a simple intramolecular process being operative.  13  3.5  cl  u <  80 °C  •  100 °C  concentration [M]  Figure  5.7 Concentration study for the conversion of anfz'-AlCl[P2N ] 2  (1) to 5yn-AlCl[P2N ] (3) 2  in C D . 7  8  From the kinetics, it is also evident that anri-AlClfP N J (1) transforms into the invertomer syn2  A1C1[P2N2] (2)  2  approximately four times faster than the gallium system (see later). To further  illustrate that the process being studied is a simple pyramidal inversion, a variable temperature 3 1  P { H } N M R study was performed on arc//-AlCl[P N ] 1  2  2  (1) and anft'-GaCl[P N ] (2). 2  2  Aluminum-amide compounds of the type AlR [N(SiMe CH2PPr 2)2] (R = Me, Et and Bz) have i  2  198  2  References begin on page 237  Chapter 5: ...Bis(amidophosphine) Complexes of Al, Ga and In  been found to be highly fluxional (see Chapter 3). The P{ H} NMR spectra of 1 and 2 at 90 31  !  °C were identical to the room temperature spectra of these species, and thus the solid state structure of 2 was concluded to be representative of the room temperature solution structures of both 1 and 2. The transitions state parameters for the two systems are quite similar: the activation energy for the pyramidal inversion process was determined to be 20.8 and 18.2 kcal mol for -1  aluminum and gallium respectively. The Gibb's free energy of activation (at 100 °C) was found to be 29.1 kcal mol' for the aluminum system, and 30.1 kcal mol' for the gallium system. This 1  1  represents a lowering of the barrier to pyramidal inversion of about 5 kcal mol" ; activation 1  energies for trialkylphosphines are typically in the range 35 - 38 kcal mol" . 1  4,11  "  14  Previous  studies have established that such inversions occur via a planar transition state where the phosphine is 5p -hybridized. Consequently, the p-orbital can engage in overlap with orbitals on 2  adjacent atoms to stabilize the transition state and lower the barrier to inversion. Most often this interaction has been viewed as a 7t-overlap as in phenylphosphines diphosphines " or silylphosphines (Pn-dn)- ' 60  62  12 63  (Pn~Pn)} ' '  2 13 59  The derealization of the arene 7C-electrons  into the ,sp -hybridized phosphorus atom in arylphosphines results in a lowering of the barrier of 2  -2-3  kcal mol" per aryl substituent. 1  4  Similar overlap with the J-orbitals of silicon or  phosphorus in silyl- and diphosphines has also been invoked. Incorporation of the phosphorus centre into a phosphole ring results in 67t-electron systems in which electron derealization is at a maximum in the planar transition state relative to the pyramidal ground state.  24,64,65  This results  in barriers to inversion most often in the range of 15 - 18 kcal mol" . However, to our 1  knowledge, the effect of an electropositive metal centre in proximity to a phosphorus inversion process has never been cited. To more accurately assess the role of the metal, it was necessary to obtain the transition state parameters for the inversion of phosphorus in the metal-free, protonated analogues syn- and anti-H2[^2^2\-  199  References begin on page 237  ChapterS: ...Bis(amidophosphine) Complexes of Al, Ga and In  5.5.2 Pyramidal Inversion at Phosphorus in anri-H2[P2N2] and SV/1-H2IP2N2] The metal-free protonated derivatives of [P2N2] were prepared by the addition of Me3N»HCl (2 equiv) to the lithium salts, an«'-Li (THF)2[P2N ] and sy«-Li (dioxane)[P2N2]. 2  2  35  2  This reaction generates the diamines as white solids in high yields. The same procedure used to study the kinetics of phosphorus inversion in the metal systems was used for the protonated derivatives, although in this case, higher temperatures were employed (100 °C to 140 °C). a«ri-H.2[P2N2] and .sy«-H2[P2N2] undergo pyramidal inversion at phosphorus in a different manner than the metal complexes a«tf-MCl[P2N2] (M = A l , Ga). Whereas the aluminum and gallium complexes undergo irreversible first-order reactions, the protonated derivatives are in equilibrium (equation 5.5).  The barrier to pyramidal inversion at phosphorus is high enough such that the kinetics may be studied as the system approaches equilibrium. This high barrier also means that samples do not need to be quenched or frozen prior to P{ H} NMR spectroscopic analysis. The 31  1  observed rate constant in this case is twice the rate constant of the inversion. If two species, A and B, are in equilibrium, the observed rate constant is the sum of the forward and reverse rate constants (equation 5.6).  200  References begin on page 237  Chapter 5: ...Bis(amidophosphine) Complexes of Al, Ga and In  [5.6]  K)bs - ki + k_i  In the case of a macrocyclic bis(phosphine), the inversion of a second phosphorus centre results in the same product as the "double" inversion of a single phosphorus. Thus, ki is equal to k_i, and the observed rate constant is equal to 2 ki. With the metal complexes studied previously, the reaction is irreversible (k.j = 0) and the observed rate constant is equal to ki. In order to ensure that the protonated derivatives were in equilibrium, the kinetics were studied in both directions, starting with anri-H2[P2N2] and also with SVH-H2[P2N2]. Plots of ln{l/% starting protonated ligand} vs. time are presented in Figures 5.8.a) and 5.9.a) as a comparison to the inversion behavior of the metal systems. All kinetic runs were performed until the contents of the tubes had equilibrated to a -50/50 mixture of the syn- and and- isomers. Plots of lnfao-ae/at-ae) vs. time (where arj = [starting macrocycle] when t = 0, ae = [starting macrocycle] at equilibrium, at = [starting macrocycle] at time, t) can be found in Figures 5.8.b) and 5.9.b). The rates of inversion were determined from these plots. The kinetics of the two systems are tabulated in Table 5.5.  201  References begin on page 237  Chapter 5: ...Bis(amidophosphine)  Complexes of Al, Ga and In  0.8  time (hours)  ff rt b)  sr i  o rt  •  100 °c  0  110 °c  o  120 °C  A  130 °C  S  140 °C  time (hours) Figure 5.8.a) Rate plots for the conversion offl/j?z-H2[P2N2]to 5VA1-H2P2N2] in C7Dg. b) Pre-equilibrium first order rate plots for the conversion of anti-H2[P2^2\ to syrt-H [P2N ] in C D . 2  2  7  8  202  References  begin on page 237  Chapter 5: ...Bis(amidophosphine) Complexes of Al, Ga and In  0.8  o  in  o  <r>  o  >n  Cl  V)  tS  O  Cl  © »/">  time (hours)  Figure 5.9.a) Rate plots for the conversion of syn-H2[P2N ] to anti-H2[P2N2] in C7D8. 2  b) Pre-equilibrium first order rate plots for the conversion of yyn-H2[P2N2] to arcft'-H2[P2N2] in C7D8.  203  References begin on page 237  Chapter 5: ...Bis(amidophosphine) Complexes of Al, Ga and In  Table 5.5 Rate data for the conversion of anti/syn-H2[^2^2\ to syn/anti-H2[P2^2\ in C7D8.  anti- to syntemp, °C  syn- to anti-  kobs, s"  correlation, R  temp, °C  kbs> s'  correlation, R  100  9.14 ± 0.61 x 10-7  0.991  100  9.83 ± 0.37 x 10-7  0.998  110  2.51 ± 0.10 x 10-  0.996  110  3.79 ± 0.25 x IO"  0.990  4.40 ± 0.26 x IO"  0.993  4.23 ± 0.45 x 10-  0.982  7.18 +0.54 x 10-  0.989  8.39 ± 0.59 x 10-  0.992  1.31 + 0.19 x 10-5  0.971  2.37 ± 0.30 x IO"  0.974  1  6  120  6  120  6  130  6  130  6  140  1  0  6  140  5  As Figures 5.8.a) and 5.9.b) illustrate, both systems experience an initial period of inversion, which follows first order kinetics, before a roughly equal mixture of the two isomers is obtained, after which time no further study is possible.  Prolonged heating establishes an  equilibrium mixture of the two isomers, the ratio of which has very little dependence on temperature. This suggests that AH° = 0 for the reaction. At all temperatures, the anti- isomer is slightly favored over the syn- isomer. Since AH" = 0, the slight excess of the anti- isomer must be a reflection of a small AS° term. For a 50/50 equilibrium, one would expect the forward reaction (anti- to syn-) to occur at an equal rate to the reverse reaction (syn- to anti-). The small difference in rates in the forward and reverse reactions is likely a reflection of the quality of the data. It is worth noting that the protonated derivatives invert ~20 to 70x slower than the gallium and aluminum complexes, respectively. The Arrhenius plots and transition state parameters of the two systems are presented below (Figure 5.10 and Table 5.6).  204  References begin on page 237  Chapter 5: ...Bis(amidophosphine)  Complexes of Al, Ga and In  •  anft'-H [P N ] to s y w - F y P ^ ]  o  syn-H2[P N ] to anft-H [P N ]  7  2  2  2  2  2  2  2  1/T  Figure 5.10 Arrhenius plots for the conversion of antil syn-H2\?2^2]to syn/anti-H2[P2^ ] 2  C D . 7  8  Table 5.6 Transition state parameters for the conversion of antilsyn-H [P2^ ] 2  to synlanti-  2  H2P2N2] in C D . 7  function Ea, kcal m o l  anti- to syn1  AH*, kcal mol" AS*,  7  1  cal K - m o l  AG3 3*,  8  1  kcal moh  correlation, R  1  1  syn- to anti-  19.6 ± 2.3  US  18.9 ± 2.3  21.0 ± 2 . 2  -36.0 ± 4.2  -30.0 ± 3 . 1  32.3 ± 2.5  32.2 ± 2.5  0.993  0.983  205  ±2.2  References begin on page 237  m  Chapter 5: ...Bis(amidophosphine) Complexes of Al, Ga and In  As expected for an equilibrium, the transition state parameters are the same (within experimental error) regardless of the direction studied. The AG373* values for the two systems are 32.3 (anti- to syn-) and 32.2 (syn- to anti-) kcal mol . These numbers are in agreement with -1  those found in the literature; monoaryldialkylphosphines usually have barriers of about 32 kcal mol" . This is ~3 kcal mol" lower than that of a trialkylphosphine. What is unusual is the large 1 4  1  negative value of AS* (-36.0, -30.0 cal K" mol" ). Since the planar transition state is more 1  1  ordered than the pyramidal ground state, a small negative entropy of activation for acyclic phosphines would be expected (—3 cal K" mol" ). 1  1  11  Two factors may contribute to the large  negative entropy of activation observed here: 1) ligand rearrangement and 2) solvation effects. First, by incorporating the phosphines into a macrocycle, the planar transition state may be more strained than that in an acyclic phosphine. This disfavorable arrangement would be reflected in a large negative value of AS*. However, phosphetanes, in which the phosphorus atom is part of a four-membered ring, exhibit normal barriers to pyramidal inversion and AS* values of -8 cal K" mol" . 1  66  1  If AS* is only -8 cal K" mol" for a phosphorus centre which is part of a 4-membered 1  1  ring, in which one would expect ring strain in the transition state to be high, it would stand to reason that if the ring size was increased, the transition state would be less strained. This would argue against ring strain being the cause of the large AS* values in our system. Second, if the ground state and the transition state interact with solvent molecules to different degrees, this could result in anomalously large entropy terms. This has been found previously for nitrogenbased systems. For example, the value of AS* for 1,2,2-trimethylaziridine in a halogenated solvent such as chloroform (38 cal K" mol" ) is much larger than in benzene or acetone (15 cal 1  K" mol" ). 1  1  67  1  This is due to greater hydrogen bonding in the ground state than in the transition  state. To examine the possible role that solvent may play in these systems, we repeated the study of the conversion of syn-H2[^2^2\ to anti-H2[V2^2\  in 4:1 C6D5BT/C6D6.  The rate and  Arrhenius plots for this system can be found in Appendix 3. The transition state data for the pyramidal inversion of phosphorus in 4:1 C6D5Br/CgD6 are similar to those determined in C7D8: E = 12.4 ± 1.8 kcal mol" , AG373* = 31.4 ± 3.7 kcal mol" , AH* = 11.7 ± 1.8 kcal mol" and AS* 1  1  1  a  206  References begin on page 237  Chapter 5: ...Bis(amidophosphine) Complexes of Al, Ga and In  = -53.0 ± 4.8 cal K" mol . The large negative value of AS* implies that C^DsBr interacts with 1  -1  the transition state to a greater degree than the ground state. As well, AS* has a larger negative value than for the inversion in C7Dg. This argues that the difference in the solvation of the ground and transition states is greater in a halogenated solvent than in C7D8. The entropy of activation in C7D8 (-36.0, -30.0 cal K" mol" ) points toward the transition state being stabilized 1  1  by 7t-stacking or some other sort of interaction with the solvent. However, in the absence of further study, this still remains speculative.  5.5.3 M e c h a n i s m of P y r a m i d a l Inversion at Phosphorus i n MCI[P2N2] ( M = A l , G a )  The results obtained for the metal complexes may now be compared with the metal-free protonated analogues. The aluminum and gallium compounds have AG373* values which are 3.2 and 2.2 kcal mol" lower than that of the protonated derivatives, respectively. However, given 1  the uncertainty present in the data (± 2.5 kcal mol" in the diamines, ± 2.7 kcal mol" in the metal 1  1  complexes), it cannot be determined whether or not this represents a real effect due to the presence of the metals. Nevertheless, two possible mechanisms of phosphorus inversion are outlined in Scheme 5.4. In the pathway I, the metal centre directly assists in the inversion of the phosphorus. As seen in the crystal structure of 2, the minor lobe of the lone pair on the .yp -hybridized 3  phosphorus atom is oriented towards the metal centre, as the ligand geometry disposes the more hindered face of the tetrahedral phosphine in the direction of the Lewis acid. The electrophilic metal atom attracts the minor lobe towards it, resulting in a planar i p -hybridized phosphorus ,  2  atom. Note that this interaction is not a 7t-overlap, but a a-overlap between a phosphorus-based p orbital and a p-type orbital on the metal. The attraction of the metal for the lone pair results in the simultaneous formation of a dative bond and inversion of the phosphorus. The mechanism  207  References begin on page 237  Chapter 5: ...Bis(amidophosphine) Complexes of Al, Ga and In  Scheme 5.4  syn-MCI[P N ] 2  208  2  References begin on page 237  Chapter 5: ...Bis(amidophosphine) Complexes ofAl, Ga and In  proposed is much like that of a typical SN2 reaction in which the carbon atom is attacked from the more hindered side, passing through a planar ip -hybridized carbocation transition state on 2  its way to the final product whose chirality is opposite that of the original reactant.  68  Pathway II illustrates a second possibility. In this case, the uncoordinated phosphine inverts without the aid of the metal. Once the phosphine has inverted, the lone pair is directed towards the metal and little rearrangement of the ligand is required to form a dative P-M bond. Although the first steps involving the inversion of the phosphorus reflect an equilibrium, the overall mechanism resulting in the ,syn-MCl[P2N2] complexes is irreversible. The two pathways cannot be distinguished solely on the values of AG*. However, the entropy of activation, AS*, does lend some support to pathway I. If the inversion occurred via pathway II, the substitution of gallium for aluminum would have no effect on AS*, since the transition state does not involve bond formation. The difference in AS* for Al (-24.4 cal K"  1  mol" ) vs. Ga (-33.8 cal K* mol" ) is consistent with a transition state involving the formation of 1  1  1  a phosphorus-metal bond. The conversion of a«tt'-H2[P2N2] to ^yn-H2[P2N2], and the reverse conversion, should theoretically yield similar values of AS*, although here they differ by ~6 cal K" mol" . However, this may be a reflection of the quality of the data. We should add that the 1  1  data obtained for the metal complexes has less uncertainty and is likely more accurate, reflecting a real difference in entropy. Although certain evidence supports the presence of pathway I, we cannot conclusively determine from this data whether the inversion of the phosphines in MC1[P2N2] type complexes is assisted by the metal or not. We also noted previously that the protonated macrocyeles anti-H^l^i]  and syn-  H2IP2N2] invert ~70x slower than the aluminum compound. Pathway II presents no clear reason why this should be so. The slower rate implies that the transition state for the inversion of the diamines anti-E^l^l]  and ,syrc-H2[P2N2] must involve a greater conformational rearrangement  than that in the inversion of the metal complexes MC1[P2N ] (M = Al, Ga). This is also more 2  consistent with pathway I than pathway II. 209  References begin on page 237  Chapter  5:  ...Bis(amidophosphine)  Complexes  of Al, Ga and In  Another curious aspect of these systems is the faster rate of inversion for the aluminum compound even though the activation energy is ~3 kcal mol" higher than the gallium analogue. 1  This we attribute to a difference in the entropy of activation. The AS* values for the aluminum and gallium systems indicate that both systems are entropically disfavored (-24.4 and -33.8 cal K" mol" , respectively). However, AS* for the formation of the gallium compound is ~10 cal K" 1  1  1  mol" more negative than that of the corresponding aluminum complex. This means that the 1  gallium system must attain a more ordered transition state presumably by undergoing a greater conformational rearrangement, i.e., a planar phosphorus engaging in a-overlap with the gallium atom. If we can assume the inversion occurs via pathway I, the longer gallium-amide bonds as compared to aluminum-amide bonds (see Table 5.2) necessitate the tetrahedral gallium perching higher above the [P2N2] cavity than aluminum, resulting in less effective overlap (Figure 5.11).  M = Al  CI  M = Ga  Figure 5.11 Transition state a-overlap diagrams for MC1[P N ] (M = Al, Ga). 2  210  2  References  begin  on page  237  Chapter 5: ...Bis(amidophosphine) Complexes of Al, Ga and In  To attain a similar degree of orbital interaction as the aluminum compound, the [P2N2] ligand must contort to a greater degree, as evidenced by the larger value of AS*. The 10 cal K '  1  mol' difference in AS* accounts for a ~120x rate enhancement of the aluminum system over the 1  gallium system; conversely the enthalpy term, AH* favors the gallium system by a factor of 30. The negative AS* value for both systems implies a more ordered transition state as compared to the starting compound, which is consistent with a bond-making reaction of this type. Thus, the difference in rates is primarily a size effect. A slight variation of this proposal can also be applied to the previously mentioned case of Sc, which promotes pyramidal inversion under even milder conditions (but not amenable to kinetic analysis) than those studied here, implying a lowering of the barrier. This d° transition metal offers a larger 3d-orbital to overlap with the phosphorus lone pair, as opposed to the nporbital (where n = 3 for Al, n = 4 for Ga) for the main group metals, resulting in more effective overlap even though the metal itself is larger and perches higher above the cavity (Figure 5.12). This overlap facilitates a further lowering of the barrier as the anti- conformation of the Sc complex is never observed.  M = Al, G a  M = Sc  Figure 5.12. Transition state a-overlap diagrams for MC1[P2N2] (M = Al, Ga, Sc). 211  References begin on page 237  ChapterS: ...Bis(amidophosphine) Complexes of Al, Ga and In  Finally, we must qualify our arguments by stating that the evidence for Pathway I is not definitive. Although most of our observations can be explained by this hypothesis, the large negative value of AS* for the protonated derivatives a«ri-H2[P2N2] and .syn-H2[P2N2] is worrisome. Further studies of the kinetics in solvents such as di2-cyclohexane may shed more light on the extent of solvent interaction in the transition state, but is beyond the scope of this thesis. The transition state parameters determined for the metal complexes MC1[P2N2] (M = Al, Ga) are reasonable and consistent with the mechanism postulated in Scheme 5.4.  5.6 Hydrides of Aluminum, Gallium and Indium The coordination chemistry of the Group 13 metals aluminum, gallium and indium has been extensively investigated, with many recent reports concentrating on the isolation of unusual Group 13 metal complexes such as hydrides, ' 57  cationic structures. " 86  96  58,69  "  77  metal-metal bonded dimers " and 78  85  The complexation of these metals by bulky ligands has been a common  theme in the literature and, in certain cases, disfavors aggregation (see Chapter 3). Recent work has shown that it is possible to use bulky aryl groups to synthesize unassociated halide derivatives of aluminum, gallium and indium, and hydride species of aluminum and gallium.  58,69,77  Monomeric hydride compounds are important starting materials for a number of  transformations. Specifically, the synthesis of dihydride species of aluminum and gallium has been investigated as possible precursors to thin films of reduced forms of these metals.  97,98  The  generation of monomeric, salt- and base-free Group 13 hydrides has been the subject of intense work, as well as being intimately linked to the study of the lower oxidation states of aluminum, gallium and i n d i u m .  58,69,73,77  212  References begin on page 237  Chapter  5:  ...Bis(amidophosphine)  Complexes  of Al, Ga and In  5.6.1 Synthesis of syn-lP^Nil Hydride Complexes of Aluminum and Gallium The reaction of the monomeric chloride sy«-AlCl[P2N2] (3) with L1AIH4 in ether at -78 °C generates syn-A\H[P2^2] (6) in moderate yield (equation 5.7). The * H NMR spectrum of the product indicates a C2v symmetric complex: two silyl methyl peaks of equal intensity and a single ortho- proton signal are present. The P { H } NMR spectrum of 6 consists of a sharp 3 1  1  singlet at -46.5 ppm; this is in contrast to the broad peak found for ,syw-AlCl[P2N2] (3) and that expected for a phosphorus bound to a quadrupolar nucleus.  The gallium 4 and indium 5  analogues of the starting aluminum chloride undergo facile transmetallation reactions with LiAlFLt; treatment of these Group 13 chlorides with the reducing agent also results in synA1H[P2N2] (6), although in lower yields. This type of behavior has been seen before for Group 13 metals complexed by the [2,6-bis(dimethylaminomethyl)phenyl] ligand system.  58,69,77  Due to  the quadrupolar aluminum centre, we were not able to observe the hydride resonance in the * H NMR spectrum.  syn-MCI[P N ] 2  +  2  UAIH4  M = AI 3 Ga 4 In 5  ether (-LiCl) (-MH3)  Me Si.  2  Me SL  2  [5.7]  Ph—P-  213  References  begin  on page 237  Chapter 5: ...Bis(amidophosphine) Complexes ofAl, Ga and In  The transmetallation reaction of the indium chloride 5 with LiAlfLt is accompanied by the formation of LiCl, and the deposition of indium metal, due to the instability of (InH3) (L1H3 x  -> ln(0) + 3/2 H ). 2  The aluminum hydride 6 is isolated as a colorless, air- and moisture-sensitive solid. Mass spectral and microanalytical data suggest that 6 is monomeric and free from interaction with LiCl or external base. An X-ray diffraction study, presented in Section 5.6.2 allowed us to study the solid state structure of this compound. The transmetallation reactions observed between 5yn-MCl[P2N2] (M = Al, Ga, In) and L i A l H 4 suggest that a different hydride source is needed to make the analogous gallium  compound sv«-GaH[P2N2] (7). LiGaFLj is synthesized from GaCl3 and LiH at low temperature (equation 5.8)."  4 LiH + GaCI — _ ^ f ^ h  3  » LiGaH + 3 LiCl 4  .8]  [5  LiGalL; is a milder reducing agent than LiAlH4 and is also less stable. At room temperature it decomposes slowly to give lithium hydride, hydrogen and gallium metal, whereas solution or solid samples of LiAlIL; are stable indefinitely. Synthetic procedures utilizing LiGaFLj. usually involve in situ preparation. The slow addition of an ethereal solution of .syrt-GaCl[P2N2] (4) to a solution of LiGaH4 (prepared in situ) maintained at -78 °C yields a mixture of products (equation 5.9).  Upon  warming, the solution becomes cloudy and darkens. The K NMR spectrum of the product has l  the same features as the aluminum hydride 6: two silyl methyl peaks of equal intensity and a single ortho- proton signal are present. The lower quadrupolar moment of gallium allows for the  214  References begin on page 237  Chapter  GSLH  5:  ...Bis(amidophosphine)  Complexes  of Al, Ga and In  signal to be observed (8 5.25), although the peak is still broad enough to mask any  coupling. This is in the region reported for other molecules with a terminal GaH moiety.  69,77  The P{ *H} NMR spectrum of syn-GaH[P2^2\ (?) consists of a singlet at -34.7 ppm. 31  syn-MCI[P N ] 2  2  +  LiGaH  e  t  h  e  r  4  (-LiCl) M = Ga 4 In  (-MH ) 3  5  The presence of syn-H.2\?2N2\  (detected by *H NMR spectroscopy) in the reaction  mixture and the darkening of the solution upon warming, consistent with metal deposition, illustrate the instability of the hydride. If syn-GaH[l?2^2\ is left to stand in solution, gallium metal gradually appears. This decomposition may be assisted by GaH3(Et20) produced as a byproduct in the reaction, serving as a source of protons. The transmetallation reaction between yy«-InCl[P2N2] (5) and LiGafU involves a more facile decomposition and produces lower yields  215  References  begin  on page 237  Chapter  5:  ...Bis(amidophosphine)  Complexes  of Al, Ga and In  of the gallium hydride 7. The accelerated rate of metal formation reflects the greater instability of (InH3) compared to GaH3(Et20) (equations 5 . 1 0 and x  GaH (Et 0) 3  •  2  o  lnH  2  2  Ga(0) + 3/2 H + Et 0 2  2  [5.10]  r  •  3  2 syA7-GaH[P N ] + H  5.11).  ln(0) + 3/2H  2  • 2 syn-H [P N ] + 2 Ga(0)  2  2  2  2  [5  .ii]  The decomposition of syn-GaHIT^^] did not allow for characterization by mass spectroscopy, elemental analysis or X-ray diffraction. Our assignment of the product of this reaction is based on the *H and P{ H} NMR spectral analysis, and literature precedent. The 31  1  greater thermal stability of ,syrc-AlH[P2N2] (6) vs. yy«-GaH[P2N2] (7) may also be due to aluminum's preference for higher coordination numbers. For example, progressively increasing the number of tertiary donor sites results in higher thermal stability of alane, whereas polydentate tertiary donors destabilize gallane relative to monodentate donors. This is due to the preference of aluminum in alane to attain higher coordination numbers, while gallium is content to be fourcoordinate. Even though tetramethylethylenediamine (tmen) initially forms a 1:1 adduct with gallane, it slowly decomposes to give the 4-coordinate species [(GaH3)2(tmen)].  46  The analogous synthesis of an indium hydride is much more problematic. Although some cationic compounds containing In-H bonds have been structurally characterized, neutral indium hydrides exist in the literature.  70,71  ' no 74  (For example, the unstable ether adduct  [(InH3) (Et20) ] decomposes to a polymeric In(I) hydride of unknown structure. ) This is due 100  x  n  to the lack of a suitable hydride transfer reagent with which to metathesize the In-Cl bond (Scheme 5.5); LilnFLj. has been reported to be stable only for short periods of time below - 6 0 216  References  begin  on page  237  Chapter  °C.  101  5:  ...Bis(amidophosphine)  Complexes  ofAl,  Ga and In  Attempts at in situ reactions with LiInH4, as well as those with reagents such as LiH, KH,  or KEt3BH either result in no reaction or the generation of syrt-H [P N ], indium metal, and 2  2  2  other unknown products. Scheme 5.5  syA7-AICI[P N ] 2  • syA7-AIH[P N ]  2  2  2  5.6.2 Structure of syn-AlH[P N ] (6) 2  2  Large colorless plates resulted from the evaporation of a saturated toluene solution of 6. The molecular structure and numbering scheme are illustrated in Figure 5.13. Compound 6 is monomeric in the solid state and free from interaction with salt (LiCl) or external base (such as Et 0). As the hydride ligand was located, but not refined, information regarding the aluminum 2  hydride bond length or angles involving this proton are not available. As is the case for the aluminum chloride yy«-AlCl[P N ] (3), the metal centre exhibits trigonal bipyramidal geometry. 2  2  217  References  begin  on page  237  Chapter 5: ...Bis(amidophosphine)  Complexes of Al, Ga and In  The bond angles involving Al(l), P(l) and P(2) are nearly identical in the hydride 6 and the chloride 3.  The effect of substituting a chloride for a hydride is minimal in terms of ligand  rearrangement.  Figure 5.13 Molecular structure of syn-AlH[P2N2] (6); 33% probability thermal ellipsoids are shown. Hydrogen atoms are omitted for clarity.  The aluminum-phosphorus bond lengths (2.517(3) and 2.522(3) A) are longer than those in syra-AlCl[P2N2] (2.481(1) and 2.465(1) A). The electron-withdrawing chloride ligand results in a higher effective Lewis acidity in 3 than in 6 and thus shorter aluminum-phosphorus bonds. No variation is seen in the aluminum-amide bond lengths of 1.903(5) and 1.910(5) A , which are similar to those in the starting aluminum chloride 3 as well as that determined for AlCl [N(SiMe CH PPr 2)2] (1-89(1) A) (see Chapter 3). i  2  2  2  218  References begin on page 237  Chapter  5:  ...Bis(amidophosphine)  Complexes  of Al, Ga and In  i  The hydride complex 6 represents only the second structurally characterized aluminum mono-hydride species which does not interact with either salt or external base. Previous work by Cowley et al. demonstrated that the alkyl ligand 2,4,6-Bu 3C6H was sufficiently bulky to t  2  stabilize a monomeric aluminum hydride of the formula (2,4,6-Bu 3C6H2)2AlH ( D ) . t  58  The tendency toward oligomerization is impeded by the use of this bulky ligand, due to the shielding of the Al-H bond by the o-Bu groups. In our system, the phenyl groups attached to the l  phosphorus, and the syn- geometry of the ligand likely play a similar role, preventing two molecules from approaching close enough to form hydride bridges. The same group has also reported an aluminum dihydride complex, employing intramolecular base stabilization.  [2,6-  Bis(dimethylaminomethyl)phenyl]aluminumdihydride (E) is the first base-stabilized monomeric aluminum hydride. Intramolecular base stabilization and bulky alkyl ligands are both effective 69  routes to monomeric hydrides. A list of selected bond lengths and angles for 6 are presented in Table 5.7.  219  References  begin  on page  237  Chapter  5: ...Bis(amidophosphine)  Complexes  of Al, Ga and In  Table 5.7 Selected bond lengths (A) and angles (deg) for 5y«-AlH[P2N2] (6).  N(1)-A1(1)  1.903(5)  N(2)-A1(1)  1.910(5)  P(1)-A1(1)  2.522(3)  P(2)-A1(1)  2.517(3)  N(1)-A1(1)-N(2)  114.9(2)  N(1)-A1(1)-P(1)  89.2(2)  N(1)-A1(1)-P(2)  83.3(2)  N(2)-A1(1)-P(1)  82.6(2)  N(2)-A1(1)-P(2)  89.3(2)  P(1)-A1(1)-P(2)  165.53(11)  Al(l)-P(l)-C(6)  130.0(3)  Al(l)-P(2)-C(12)  131.2(3)  5.6.3 Solution structure of syn-A1H[P2N2] (6)  The structural characterization of 6 does not correlate with the P{ *H} NMR data. The 31  bound phosphines in the X-ray diffraction study are not observed in the P{!H} NMR spectrum 31  of 6, which consists of a sharp singlet. One possibility is that 6 is highly fluxional and the single sharp peak at room temperature represents a molecule undergoing rapid phosphine exchange (Scheme 5.6). Scheme 5.6  220  References  begin  on page 237  Chapter  5: ...Bis(amidophosphine)  Complexes  of Al,  Ga and In  Cooling a ds-toluene solution of 6 to -90 °C does not result in any appreciable difference in the  3 1  P { H} N M R spectrum. Even at low temperature, the molecule may be highly fluxional l  as is the case for AlR2[N(SiMe2CH2PPr 2)2] (R = M e , Et) (Chapter 3). In this case, the P { H} i  NMR  3 1  l  spectrum may reflect the average environment o f two species which are rapidly  exchanging.  The tendency for aluminum to be four-coordinate, and our studies o f similar  systems (Chapter 3) favor a tetrahedral species in solution.  Another possibility is that 6  undergoes a monomer-dimer equilibrium in solution i.e., upon dissolution, 6 changes from being monomeric with a five-coordinate metal centre and a terminal hydride F, to dimeric with two bridging hydrides per metal centre G (Scheme 5.7).  Scheme 5.7  2  F  G  (silyl methyl groups omitted for clarity)  The steric congestion resulting from the bridging hydrides would be enough to cause the phosphines to dissociate and the aluminum centre to be four-coordinate.  However, solution  molecular weight measurements (isopiestic method, Signer apparatus, 520 ± 50 g m o l ) indicate - 1  a monomeric solution structure. A s well, phosphines which do not coordinate to a metal centre  221  References  begin  on page 237  Chapter  5:  ...Bis(amidophosphine)  Complexes  of Al, Ga and In  in the [P2N2] ligand system most often appear at —38 ppm in the ^ P ^ H } NMR spectrum. The quadrupolar A l nucleus does not allow for the observation of the A l - H signal in the H N M R 2 7  l  spectrum, thus a definitive assignment of a terminal or bridging hydride from the / H - P coupling 2  was not possible. Compound 6 exhibits a v(Al-H) stretch at 1594 cm" in the IR spectrum; this 1  medium intensity peak shifts to 1132 cm" upon deuteration. This is at the lower end of the 1  determined range for a terminal aluminum hydride. The IR spectrum of (2,4,6-Bu 3C6H2)2AlH t  features a sharp absorption at 1869 c m . - 1  5 8  The low solubility of 6 in ether or toluene did not  allow for a definitive solution IR peak assignment to be made. Although IR measurements alone are not enough to conclusively determine the monomeric nature of 6, there is no evidence for the alternate structures proposed and we conclude that 6 is monomeric both in the solid and solution states.  5.6.4 Reactivity of sy/i-AIH[P N ] (6) 2  2  The aluminum hydride 6 is surprisingly inert to a variety of substrates.  Insertion  chemistry with C O , Bu*CN and ethylene was not observed and unreacted starting material was recovered in each case. We also attempted to abstract the hydride to generate a more reactive cationic aluminum derivative. There was no sign of reaction upon exposure of yy«-AlH[P2N2] (6) to either A l H » N M e or [H(OEt )2][B{3,5-(F C)2C6H }4]. 3  3  2  3  3  5.7 Metal-Metal Bonded Dimers of Aluminum, Gallium and Indium The synthesis and characterization of derivatives of the main-group elements in low oxidation states provide the focus for an ongoing area of chemistry. Dimeric complexes of gallium dihalides were first reported 40 years a g o .  102  With monodentate ligands, these  complexes have the stoichiometry Ga2X4»4L, and Ga2X4»2L with bidentate ligands. Original  222  References  begin  on page  237  Chapter  5:  ...Bis(amidophosphine)  Complexes  of Al, Ga and In  reports suggested that these complexes were ionic with structures such as [GaL4] [GaX4]". +  102,103  More recently, crystallographic studies have shown that species such as Ga2X4»(dioxane)2 (X = CI, Br) have ethane-like structures and Ga-Ga bonds.  80,104  Current work in the area has  concentrated on the use of bulky alkyl groups instead of halides to stabilize Ga(II), substituted aryl groups, amides 83  105,106  82,84  although  and carboranes have also been reported. The synthetic 107  route to these organo-gallium derivatives has most often been metathesis reactions of the dimeric gallium(II) dihalides.  85  The reduction of a gallium(III) chloride with K or Na to yield a dimer  with a direct metal-metal bond is a relatively new development in the syntheses of these complexes and appears to be accessible only to those systems supported by bulky groups.  84,106  5.7.1 Synthesis of syn-[P N2] Metal-Metal Bonded Dimers of Gallium and Indium 2  The addition of an ethereal solution of either syn-GaCl[P2N2] (4) or sy«-InCl[P2N2] (5) to a slurry of K C (1.1 equiv) in ether is accompanied by the gradual dissipation of the brown 8  color of the reducing agent to yield a black slurry of graphite (equation 5.12). Extraction into toluene yields a white solid which crystallizes as colorless needles. The *H NMR spectra of the products indicate a single symmetric product: two silyl methyl peaks as well as a single orthoresonance are visible. The P{ H} NMR spectra exhibit a single slightly broadened peak 31  1  downfield of the starting chloride (8 -25.5 ppm, A1/2 = 10 Hz for {.yyn-Ga[P2N2]}2 (8); 8 -37.7 ppm, Ai/2 = 24 Hz for {syrc-In[P N ]} (9)). 2  2  2  223  References  begin  on page 237  Chapter  5: ...Bis(amidophosphine)  Complexes  ofAl,  Ga and In  [5.12]  In 9 (silyl methyl groups omitted for clarity)  The indium (II) species 9 can be prepared in moderate yields by the above method. The reduction to produce the gallium analogue 8 does not go to completion; K C must be continually 8  added to the reaction mixture to produce enough {5y«-Ga[P2N2]}2 (8) to characterize, and the dimer 8 must be isolated pure by fractional crystallization. The attempted reduction of synA1C1[P2N2] (3) with K C did not yield a characterizable product, although metal-metal dimers of 8  Al(II) have been r e p o r t e d .  85,108,109  The unfavorable steric interaction of the phenyl rings of two  syn-A\[P2^2] units may prevent the close approach of two monomers and a metal-metal bond from being formed. The reduced species {syn-Ga[P2^2\) 2 (8) and {5y«-In[P2N2]}2 (9) were studied by X-ray crystallography.  224  References  begin  on page  237  Chapter 5: ...Bis(amidophosphine) Complexes of Al, Ga and In  5.7.2 Structure of {sy/i-Ga[P N ]}2 (8) 2  2  Small colorless needles were collected following the slow evaporation of a saturated toluene solution of 8. Figure 5.14 presents the molecular structure and numbering scheme. The compound contains an unsupported metal-metal bond between two Ga(II) centres.  Figure 5.14 Molecular structure of {5 y«-Ga[P2N ]} (8); 50% probability thermal ellipsoids are ,  2  2  shown. Hydrogen atoms, silyl methyls and phenyl rings are omitted for clarity.  The gallium-gallium bond length of 2.5206(8) A is similar to that reported for the tetra(amido) species (tmp)2Ga-Ga(tmp) (tmp = 2,2,6,6-tetramethylpiperidino) (Ga-Ga = 2  2.525(1) A ) .  1 0 6  The metal-metal bond lengths in gallium(II) dimers fall into two approximate  categories: those supported by bulky alkyls such as substituted aryls or organosilyls (~2.5 - 2.6  225  References begin on page 237  Chapter  A)  82-84  *  106  5: ...Bis(amidophosphine)  and those ligated by halides (-2.4 A ) ,  are also known.  80,81  Complexes  of Al, Ga and In  although some anomalously short contacts  105,107  Table 5.8 Selected bond lengths (A) and angles (deg) for {syn-Ga[P2^2\ h (8) and Uyn-In[P N ]} (9). 2  2  2  Ga(l)*-Ga(l)  2.5206(8)  In(l)*-In(l)  2.7618(12)  P(D-Ga(l)  3.1459(11)  Pd)-In(l)  2.870(4)  P(2)-Ga(l)  2.6591(11)  P(2)-In(l)  2.879(4)  N(l)-Ga(l)  1.958(3)  N(l)-In(l)  2.170(8)  N(2)-Ga(l)  1.968(3)  N(2)-In(l)  2.160(8)  P(l)-Ga(l)-P(2)  155.22(3)  P(l)-In(l)-P(2)  149.87(10)  N(l)-Ga(l)-N(2)  105.46(12)  N(l)-In(l)-N(2)  107.32(9)  Ga(l)*-Ga(l)-P(l)  95.00(3)  In(l)*-In(l)-P(l)  102.79(10)  Ga(l)*-Ga(l)-P(2)  109.75(3)  In(l)*-In(l)-P(2)  125.6(2)  Ga(l)*-Ga(l)-N(l)  125.47(9)  In(l)*-In(l)-N(l)  131.1(2)  Ga(l)*-Ga(l)-N(2)  126.63(9)  In(l)*-In(l)-N(2)  103.3(3)  Ga(l)-P(l)-C(13)  156.51(13)  In(l)-P(l)-C(13)  143.9(5)  Ga(l)-P(2)-C(19)  144.86(13)  In(l)-P(2)-C(19)  142.8(4)  Compound 8 exhibits a slightly staggered geometry; the dihedral angle between the two GaN planes is 27.92°. By comparison, the GaC planes in (Trip) Ga-Ga(Trip) (Trip = 2,4,62  2  triisopropylphenyl) are skewed by 4 3 . 8 ° , Ga(Si(SiMe3) )2 by 8 0 ° . 3  82  83  and the GaSi  2  2  2  planes in ((Me Si)3Si) Ga3  2  The gallium-nitrogen bond lengths (1.958(3) and 1.968(3) A) are  slightly longer than in other amido-gallium systems (-1.84 - 1.90 A ) 226  1 0 5 , 1 0 6  References  and reflect the steric  begin  on page 237  Chapter 5: ...Bis(amidophosphine)  constraints of the macrocyclic [P2N2] ligand system.  Complexes of Al, Ga and In  Compound 8 also contains two short  (2.6591(11) A) and two long (3.1459(11) A) gallium-phosphorus bonds.  Whether or not the  longer Ga-P bond length represents a bonding interaction is debatable. The gallium centre is best described as a distorted tetrahedron, with angles involving the metal ranging from 126.63(9)° to 81.59(9)°. Bond lengths and angles for 8 and 9 are presented in Table 5.8.  5.7.3 Structure of {sy/i-In[P N ]} (9) 2  2  2  Evaporation of a saturated toluene solution of 9 resulted in large colorless needles that were suitable for single crystal X-ray diffraction.  The molecular structure and numbering  scheme are illustrated in Figure 5.15. As for 8, {.syrt-In[P N ]} (9) is dimeric with a metal2  metal bond between two In(II) nuclei.  2  2  {5>'n-In[P N ]} (9) and {syrt-Ga[P N ]} (8) are 2  2  2  2  2  2  isomorphous, crystallizing in the same space group with similar unit cell parameters.  C(19)  Figure 5.15 Molecular structure of {5>'n-In[P N ]} (9); 33% probability thermal ellipsoids are 2  2  2  shown. Hydrogen atoms, silyl methyls and phenyl rings are omitted for clarity.  227  References begin on page 237  Chapter  5:  ...Bis(amidophosphine)  Complexes  of Al, Ga and In  The indium centres in 9 exhibit distorted trigonal bipyramidal geometries (Table 5.8). The P(l)-In(l)-P(2) and N(l)-In(l)-N(2) bond angles (149.87(10)° and 103.3(3)° respectively) are smaller than expected (180° and 120°) and indicate larger deviations from ideal than in synA1C1[P2N2] (3) or ^y«-GaCl[P N2] (4) . In this manner, 9 approaches a square pyramidal 2  geometry. The indium centres are each equally coordinated by two phosphine donors (2.870(4) and 2.879(4) A) in contrast to the situation found for the gallium analogue. Only four indium(II) dimers have been structurally characterized; these are supported by bulky alkyl groups,  110  cyclic bis(amides) or halogens. 79  111  78  or aryl  The metal-metal bond in 9 (2.7618(12) A) is  similar to that in the cyclic bis(amido) species (MeSi)2(NBu )4ln-In(NBu )4(MeSi)2 (2.768(1) t  A).  79  t  It is worth noting that a number of compounds of In (I) with weak metal-metal interactions  have been cited. These compounds, of the general type [Cp In(I)] (Cp = substituted Cp) #  #  x  contain weakly interacting metal centres separated by ~3.6 - 4.0 A, and are higher polymers or aggregates.  112-115  The dihedral angle between the two InN2 planes in 9 is 24.65°, which is  slightly less than that observed in the gallium analogue 8.  5.7.4 Solution structure of {syn-Ga[P N ]} (8) and {syn-In[P N ]} (9) 2  Since {i y«-Ga[P N ]} ,  2  monomers, as the G a / I n 2+  2  2+  2  2  2  2  2  2  (8) and {.yyn-In[P N ]} (9) are diamagnetic, they cannot be 2  2  2  ion must be paramagnetic. The observation of H and P{ H) l  31  l  NMR spectra for 8 and 9 indicate that these monomers couple via metal-metal bonds and are dimeric in solution. A challenging problem is the stabilization of a monomeric M X 2 species by the appropriate choice of ligand. Recent studies have shown that the formation of dimeric M(II) species is a reflection of the reactivity of the unpaired electron, and not the instability of complexes of M(II).  116  The reduction of the metals in 8 and 9 is reflected in the broadening of  the resonances in the H NMR spectra; however the presence of sharp peaks in the P{ H} l  31  NMR spectra of 8 and 9 are also consistent with a fluxional process (Scheme 5.8).  1  Such a  process, as proposed for the aluminum hydride 6, would average the distinct phosphorus centres 228  References  begin  on page 237  Chapter 5: ..JSis(amidophosphine) Complexes ofAl, Ga and In  present in the solid state structure of 8 and would minimize the broadening effects of coordination to a quadrupolar nucleus. We were unable to reach a limiting structure for either 8 or 9 by variable temperature P{ H} NMR spectroscopy (C7D8, -90 °C); as a result, this 31  1  mechanism is not definitive. Scheme 5.8  M = G a 8, In 9 (1/2 of dimer shown for clarity)  5.8 Conclusions By utilizing the macrocyclic ligand system [P2N2] ([P2N2] = [PhP(CH2SiMe2NSiMe2CH2)2PPh]),  we were able to isolate and characterize both anti- and syn- isomers of the  chlorides of aluminum and gallium, and the syn- isomer of indium. It can be argued that the proximity of the metals to an uncoordinated phosphine in complexes of the type anri-MCl[P2N2] (M = Al (1), Ga (2)) assists in the inversion and subsequent coordination of this phosphine to yield the syn- isomers 5yn-MCl[P2N2].  The barrier to pyramidal inversion (AG373*) at  phosphorus in these systems was found to be ~2 to 3 kcal mol" lower than that determined for 1  the metal-free compounds anti- and sy«-H2[P2N2]. The complexes syn-MCl[P2N2] (M = Al (3), Ga (4), In (5)) were suitable starting materials for the generation of monomeric aluminum and gallium hydrides syn-MR[P2^2\  (M = Al (6), Ga (7)). The crystal structure of the aluminum  derivative represents only the second characterized monomeric aluminum hydride. Reduced  229  References begin on page 237  Chapters: ...Bis(amidophosphine) Complexes of Al, Ga and In  forms of gallium and indium were also accessible; SVH-MC1[P2N2] ( M = Ga (4), In (5)) could be reduced with K C to yield gallium and indium (II) species, which contained unsupported metal8  metal bonds. Group 13 metal complexes utilizing the [P2N2] ligand system experience a more versatile chemistry as compared to those complexed with the acyclic ligand system [PNP] ([PNP] = N(SiMe2CH2PPri2)2) (Chapter 3).  5.9 Experimental Section 5.9.1 Procedures Unless otherwise stated, general procedures were performed according to Section 2.16.1.  5.9.2 Materials. anft'-Li2(THF)2[P2N2 ] and .^n-Li2(dioxane)[P2N2] were prepared by published procedures.  2  anti-fylPiNi] and 5y«-H2[P2N2] were synthesized by the addition of Me3N»HCl  (2 equiv) to toluene solutions of the lithium salts.  35  GaCl3 was purchased from Strem and used  as received. InCl3 was purchased from Alfa Chemicals and sublimed prior to use. LiAlFLi was purchased from Aldrich and used as received. LiGaFL; was prepared by a literature method." K C was prepared by the addition of potassium metal to solid graphite at 160 °C under a purge of 8  117 argon/ ' 1  230  References begin on page 237  Chapter  5:  ...Bis(amidophosphine)  Complexes  of Al, Ga and In  5.9.3 Syntheses 5.9.3.1 a«ri-AICl[P N ] (1) 2  2  To a slurry of AICI3 (60 mg; 0.450 mmol) in toluene (5 mL) was added a toluene solution (10 mL) of a«ri-Li (THF) [P N ], (296 mg; 0.428 mmol). The reaction mixture was 2  2  2  2  stirred for 12 h and then passed through a frit lined with Celite to remove LiCl. The solvent was then removed in vacuo to yield a clear, colorless oil (224 mg; 88% yield). *H NMR (C7D8): 7.76 and 7.19 (t, 4H, o-Ph, / - H = 7.5 Hz, 3/ -P = 7.5 Hz), 7.58 and 7.54 (dd, 4H, m-Ph, 3  H  3  8  / -H H  H  = 2.5 Hz, 7H-H = 7.5 Hz), 6.99 (m, 2H, pPh), 2.33 and 1.01 (dd, 4H, ring CH , / H - H = 14.5 Hz, 2  3  2  2  /H-p = 3.5 Hz), 0.76 (ABX m, 4H, ring CH , / H - H = 13.5 Hz, / - P = 9.0 Hz), 0.58, 0.33, 0.30 2  2  2  H  and 0.12 (s, 12H, SiMe ). 31p{lH} NMR (C6D ): 8 -38.8 (s, IP, uncoordinated phosphine), 2  6  -44.7 (s, IP, coordinated phosphine, 120 Hz peak width at half height).  5.9.3.2 anri-GaCl[P N ] (2) 2  2  To a slurry of GaCl3 (77 mg; 0.437 mmol) in toluene (5 mL) was added a toluene solution (10 mL) of anri-Li (THF)2[P N ] (303 mg; 0.438 mmol). The reaction mixture was 2  2  2  then stirred for 12 h. The reaction mixture was passed through a frit lined with Celite to remove LiCl. The solvent was then removed in vacuo to yield a white solid. The residue was taken up in a minimum amount of toluene (approx. 3 mL). Slow evaporation of the solvent afforded large colorless plates (206 mg; 74% yield). *H NMR (C D ): 8 7.66 and 7.19 (t, 4H, o-Ph, / - H = 3  7  H  8  8.5 Hz, V H - P = 8.5 Hz), 7.58 and 7.56 (dd, 4H, m-Ph, / - H = 7.0 Hz, / - H = 8.5 Hz), 7.00 (m, 3  3  H  H  2H, p-Ph), 2.37 and 1.02 (dd, 4H, ring CH , / - H = 14.5 Hz, / - P = 3.5 Hz), 0.68 (ABX m, 2  2  2  H  H  4H, ring CH , / - H = 16.5 Hz, / - p = 9.5 Hz), 0.58, 0.36, 0.28 and 0.12 (s, 12H, SiMe ). 2  2  2  H  H  2  31P{ lH) NMR (C D ): 8 -36.6 (s, IP, coordinated phosphine, 20 Hz peak width at half height), 6  6  231  References  begin  on page 237  Chapter  5: ...Bis(amidophosphine)  -38.0 (s, IP, uncoordinated phosphine).  Complexes  MS: mle 638 (M+).  of Al, Ga and In  Anal. Calcd. for  C24H42ClGaN P2Si4: C, 45.18; H, 6.63; N, 4.39. Found: C, 45.42; H, 6.59; N, 4.30. 2  5.9.3.3 sy«-AlCl[P N ] (3) 2  2  Method 1: To a slurry of AICI3 (110 mg; 0.824 mmol) in toluene (10 mL) was added a toluene solution (10 mL) of syn-Li (dioxane)[P N ] (500 mg; 0.788 mmol). The 2  2  2  reaction mixture was stirred for 12 h and then passed through a frit lined with Celite to remove LiCl. The solvent was removed in vacuo to yield a white solid. The residue was taken up in a minimum amount of toluene (approx. 4 mL).  Slow evaporation of the solvent yielded large  colorless plates (470 mg; 100% yield). Method 2: a«ri-AlCl[P N ] (198 mg; 0.336 mmol) was dissolved in 10 mL 2  2  toluene and placed in a reactor bomb. The head space was evacuated and the contents of the bomb heated to 100 °C for 3 days with stirring. The solvent was then removed in vacuo to yield a white solid, which was taken up in a minimum amount of toluene (approx. 2 mL). Slow evaporation of the solvent quantitatively afforded colorless plates. *H NMR (C6D6): 8 7.95 (t, 4H, o-Ph, / - H = 8.5 Hz, / . p = 8.5 Hz), 7.10 (m, 6H, m,/>-Ph), 0.97 (ABX m, 8H,ringCH , 3  3  H  2  H  2  / H - H = 15.0 Hz, / H - P = 9.5 Hz), 0.29 and 0.22 (s, 12H, SiMe ). 2  2  -43.6 (s, 340 Hz peak width at half height).  31  P{ H} NMR (C6D ): 5 1  6  MS: mle 595 (M+). Anal. Calcd. for  C H4 AlClN P2Si4: C, 48.42; H, 7.11; N, 4.71. Found: C, 48.68; H, 7.12; N, 4.55. 24  2  2  5.9.3.4 syn-GaCl[P N ] (4) 2  2  Method 1: To a slurry of GaCl3 (226 mg; 1.28 mmol) in toluene (10 mL) was added a toluene solution (10 mL) of sy«-Li (dioxane)[P N ], (748 mg; 1.18 mmol). The 2  2  2  reaction mixture was stirred for 12 h. The reaction mixture was then passed through a frit lined 232  References  begin  on page 237  Chapter 5: ..Ms(amidophosphine) Complexes ofAl, Ga and In  with Celite to remove LiCl. The solvent was removed in vacuo to yield a white solid. The residue was taken up in a minimum amount of toluene (approx. 5 mL). Slow evaporation of the solvent yielded large colorless plates (662 mg; 88% yield). Method 2: arcft'-GaCl[P N ] (200 mg; 0.313 mmol) was dissolved in 10 mL 2  2  toluene and placed in a reactor bomb. The head space was evacuated and the contents of the bomb heated to 100 °C for 3 days with stirring. The solvent was removed in vacuo and the resultant colorless oil taken up in a minimum of toluene.  Slow evaporation of the solvent  resulted in colorless plates (185 mg, 92% yield). *H NMR (C6D6): 8 7.90 (t, 4H, o-Ph, 3/ _ = H  7.9 Hz, / - p = 7.9 Hz), 7.09 (m, 6H, m,p-Ph), 0.96 (ABX m, 8H, ring CH , /H-H  = 13.0 Hz,  2  3  H  H  2  2/H-p = 14.0 Hz), 0.28 and 0.19 (s, 12H, SiMe ). P{ H) NMR (C6D6): 5 -41.5 (s, 65 Hz peak 31  l  2  width at half height).  5.9.3.5 sjn-InCl[P N ] (5) 2  2  Method 1: To a slurry of IX1CI3 (175 mg; 0.791 mmol) in toluene (10 mL) was added a toluene solution (10 mL) of sy«-Li (dioxane)[P N ], (502 mg; 0.790 mmol). The 2  2  2  reaction mixture was stirred for 12 h. The reaction mixture was then passed through a frit lined with Celite to remove LiCl. The solvent was removed in vacuo to yield a waxy white solid (430 mg; 80% yield). Method 2. InCl3 (174 mg, 0.787 mmol) and a«r/-Li (THF) [P N ] (500 mg, 2  2  2  2  0.787 mmol) were placed in a reactor bomb, and THF (20 mL) distilled onto the mixture. The head space was evacuated and the contents of the bomb heated to 70 °C for 3 days with stirring. The solvent was then removed in vacuo and the solid extracted with toluene (3 x 10 mL). The solution was then filtered through Celite to remove LiCl and the solvent removed under vacuum to yield a white solid (402 mg, 75% yield). *H NMR (C D ): 8 7.95 (t, 4H, o-Ph, 3/ _ = 8.5 6  6  H  H  Hz, / . p = 8.5 Hz), 7.07 (m, 6H, m,/?-Ph), 0.90 (ABX m, 8H, ring CH , / - H = 15.5 Hz, 7 - P 3  2  H  2  233  2  H  H  References begin on page 237  Chapter  5:  ...Bis(amidophosphine)  Complexes  of Al, Ga and In  = 14.0 Hz), 0.28 and 0.20 (s, 12H, SiMe ). 31p{lH} NMR (C D ): 8 -40.5 (s, 400 Hz peak 2  6  6  width at half height). MS: mle 684 (M+). Anal. Calcd. for C24H42lnClN P2Si4: C, 42.19; H, 2  6.20; N, 4.10. Found: C, 42.55; H, 6.20; N, 3.99.  5.9.3.6 svn-AlH[P N ] (6) 2  2  Method 1: To a -78 °C solution of LiAlHj (12 mg; 0.32 mmol) in ether (10 mL) was added an ethereal solution (10 mL) of SVH-AICIIP2N2] (156 mg; 0.262 mmol). The reaction vessel was left to warm with stirring. The reaction mixture was then passed through a frit lined with Celite to remove LiCl. The solvent was removed in vacuo and the residue extracted into a minimum amount of toluene (approx. 3 mL). Slow evaporation of the solvent gave large colorless plates (54 mg; 37% yield). Method 2: To a -78 "C solution of LiAlFLi (12 mg; 0.32 mmol) in ether (10 mL) was added an ethereal solution (10 mL) of ,sy«-GaCl[P2N2] (170 mg; 0.266 mmol). The reaction mixture was left to warm with stirring. The reaction mixture was then passed through a frit lined with Celite to remove LiCl. The solvent was removed in vacuo and the residue extracted into toluene. Evaporation of the solvent afforded a white powder (45 mg; 30% yield). Method 3: To a -78 °C solution of L1AIH4 (12 mg; 0.32 mmol) in ether (10 mL) was added an ethereal solution (10 mL) of syn-InCl[P2N2] (202 mg; 0.296 mmol). Upon warming, the mixture became black and deposited indium metal. The reaction mixture was then passed through a frit lined with Celite to remove LiCl and ln(0). The solvent was removed in vacuo and the residue extracted into toluene. Evaporation of the solvent yielded a white powder (26 mg; 16% yield). *H NMR (C D ): 8 7.77 (t, 4H, o-Ph, 3 / . 6  6  H  H  = 6.9 Hz, 3 / . = 6.9 Hz), H  p  7.08 (m, 6H, m,p-?h), 1.01 (ABX m, 8H, ring CH , / - H = 13.8 Hz, / - P = 6.9 Hz), 0.34 and 2  2  2  H  H  0.21 (s, 12H, SiMe ). 31p{lH} NMR (C D ): 8 -46.5 (s). IR (KBr disc): VAI-H = 2  6  1  5  9  4  6  c™"  1  (s), VAI-D = H32 cm* (s). MS: mle 559 ( M - H). Mol wt. (Signer, isopiestic method, toluene): 1  +  234  References  begin  on page 237  Chapter  5:  ...Bis(amidophosphine)  Complexes  of Al, Ga and In  520 + 50; calcd. 570. Anal. Calcd. for C24H43AlN P2Si4: C, 51.39; H , 7.73; N, 4.99. Found: 2  C, 51.26; H , 7.65; N, 4.86.  5.9.3.7 sy/i-GaH[P N ] (7) 2  2  Method 1: To a freshly prepared solution of LiGaFL; maintained at -78 °C (21 mg; 0.257 mmol) in ether (10 mL) was added an ethereal solution (10 mL) of yyn-GaCl[P N ] 2  2  (150 mg; 0.235 mmol). The reaction mixture was left to warm with stirring, during which time some metal deposition was evident. The reaction mixture was then passed through a frit lined with Celite to remove LiCl and Ga(0). The solvent was removed in vacuo and the residue extracted into toluene. Evaporation of the solvent afforded an impure white powder (34 mg; 24% yield). Method 2: To a freshly prepared solution of LiGaFL; maintained at -78 °C (21 mg; 0.257 mmol) in ether (10 mL) was added an ethereal solution (10 mL) of svH-InCl[P N ] 2  2  (156 mg; 0.228 mmol). Upon warming, the mixture became black and deposited gallium and indium metal. The reaction mixture was then passed through a frit lined with Celite to remove LiCl, Ga(0) and ln(0). The solvent was removed in vacuo and the residue extracted into toluene. Evaporation of the solvent gave an impure white powder (15 mg; 11% yield). The unstable nature of 7 did not allow for a suitable elemental analysis or mass spectrum to be obtained. Prolonged standing (-days) of solutions of 7 resulted in the slow decomposition to iyn-H [P N ] 2  2  2  and gallium metal. R NMR (C6D ): 5 7.47 (t, 4H, o-Ph, / - H = 6.6 Hz, / - P = 6.6 Hz), 7.11 3  l  3  H  6  H  (m, 6H, m,/?-Ph), 5.25 (br s, IH, GaH), 1.75 and 0.93 (ABX m, 8H, ring CH , V H - H = 13.2 Hz, 2  2  /H-P = 7.6 Hz), 0.48 and 0.25 (s, 12H, SiMe ). P{ lH} NMR (QDe): 5 -34.7 (s). 31  2  235  References  begin  on page  237  Chapter  5:  ..JBis(amidophosphine)  Complexes  of Al, Ga and In  5.9.3.8 {sj7i-Ga[P2N ]h (8) 2  To a slurry of KCs (25 mg; 0.185 mmol) in ether (10 mL) was added an ethereal solution (10 mL) of .yyn-GaCl[P N ] (105 mg; 0.165 mmol). The reaction mixture was stirred 2  2  for 24 h. The reaction mixture was then passed through a frit lined with Celite to remove KC1 and C(gr). The solvent was removed in vacuo to yield a white solid. The residue was taken up in a minimum amount of toluene (approx. 2 mL). Slow evaporation of the solvent afforded small colorless needles (33 mg; 33% yield). H NMR (C D ): 8 7.48 (br t, 4H, o-Ph), 7.20 and 7.08 l  6  6  (m, 6H, m,p-Ph), 0.93 (ABX m, 8H, ring CH , / H - H = 12.9 Hz, / - P = 5.9 Hz), 0.52 and 0.33 2  2  2  (s, 12H, SiMe ).  31  2  H  P{ H} NMR (C6D ): 8 -25.5 (10 Hz peak width at half height). MS: mle !  6  603 (M+(monomer)). Anal. Calcd. for C48H84Ga N4P Si8: C, 47.84; H, 7.02; N, 4.65. Found: 2  4  C, 47.54; H, 6.94; N, 4.49.  5.9.3.9 {syn-In[P N ]} (9) 2  2  2  To a slurry of KCs (22 mg; 0.163 mmol) in ether (10 mL) was added an ethereal solution (10 mL) of ^yn-InCl[P N ] (98 mg; 0.143 mmol). The reaction mixture was stirred for 2  2  24 h. The reaction mixture was then passed through a frit lined with Celite to remove KC1 and C(gr). The solvent was removed in vacuo to yield a white solid. The residue was taken up in a minimum amount of toluene (approx. 2 mL).  Slow evaporation of the solvent afforded large  colorless needles (54 mg; 58% yield). H NMR (C D ): 8 7.48 (br t, 4H, o-Ph), 7.02 (br m, 6H, l  6  6  m,/?-Ph), 1.13 (br ABX m, 8H, ring CH ), 0.42 and 0.39 (s, 12H, SiMe ). 2  2  31  P{!H} NMR  (C6D ): 8 -37.4 (24 Hz peak width at half height). MS: mle 647 (M+(monomer)). Anal. Calcd. 6  for C48H84ln N P Si8: C, 44.50; H, 6.54; N, 4.32. Found: C, 44.14; H, 6.23; N, 4.19. 2  4  4  236  References  begin  on page  237  Chapter  5:  ...Bis(amidophosphine)  Complexes  of Al, Ga and In  5.9.4 Kinetics of the Inversion Reactions The first-order conversion of the a«ri-MCl[P2N2] complexes to the ,yy«-MCl[P2N2] complexes (M = A l , Ga) was monitored by *H NMR spectroscopy by following the disappearance of the anti- complex and the appearance of the syn- complex over time. ~0.1 M C7D8 solutions of the pure anti- complexes were placed in sealed NMR tubes and immersed in oil baths (maintained at a specific temperature) for predetermined lengths of time. The tubes were then removed from the oil baths and the spectra taken at room temperature. The irreversibility of the conversion did not necessitate quenching the reaction by freezing the samples.  A similar protocol was followed in examining the behavior of the protonated  derivatives a«ri-H2[P2N2] and ^yn-H2[P2N2]; in this case the conversion was followed by 31  P{ H} NMR spectroscopy in both C7D8 and a 4:1 mixture of C6DsBr/C6D6. !  5.10 References (1)  Lindoy, L. F. The Chemistry of Macrocyclic Ligand Complexes; Cambridge University Press: New York, 1989.  (2)  Fryzuk, M. D.; Love, J. B.; Rettig, S. J. / . Chem. Soc, Chem. Commun. 1996,24,2783.  (3)  Cotton, F. A.; Wilkinson, G. Advanced Inorganic Chemistry; 4th ed.; John Wiley & Sons: New York, 1980.  (4)  Rauk, A.; Allen, L. C ; Mislow, K. Angew. Chem., Int. Ed. Engl. 1970, 9, 400.  (5)  Moore, W. J. Basic Physical Chemistry, Prentice-Hall, Inc: New Jersey, 1983.  (6)  Swalen, J. D.; Ibers, J. A. / . Chem. Phys. 1962,36,1914.  (7)  Anet, F. A. L.; Trepka, R. D.; Cram, D. J. / . Am. Chem. Soc. 1967,89, 357. 237  References  begin  on page  237  Chapter 5: ..JBis(amidophosphine) Complexes of Al, Ga and In  (8)  Gribble, G. W.; Easton, Jr, N. R.; Eaton, J. T. Tetrahedron Lett. 1970,13, 1075.  (9)  Lambert, J. B.; Oliver, J. / . Am. Chem. Soc. 1969, 91, 111 A.  (10)  Edwards, P. G.; Fleming, J. S.; Liyanage, S. S. Inorg. Chem. 1996,35,4563.  (11)  Baechler, R. D.; Farnham, W. B.; Mislow, K. / . Am. Chem. Soc. 1969, 91, 5686.  (12)  Baechler, R. D.; Mislow, K. J. Am. Chem. Soc. 1970, 92, 3090.  (13)  Baechler, R. D.; Mislow, K. / . Am. Chem. Soc. 1971, 93,113.  (14)  Lehn, J. M.; Munsch, B. J. Chem. Soc, Chem. Commun. 1969, 1327.  (15)  Pabel, M.; Willis, A. C.; Wild, S. B. Inorg. Chem. 1996,35, 1244.  (16)  Naumann, K.; Zon, G.; Mislow, K. / . Am. Chem. Soc. 1969, 91, 2788.  (17)  Ciampolini, M.; Dapporto, P.; Nardi, N.; Zanobini, F. Inorg. Chem. 1983,22, 13.  (18)  Ciampolini, M.; Nardi, N.; Zanobini, F.; Cini, R.; Orioli, P. L. Inorg. Chim. Acta. 1983, 76, L17.  (19)  Ciampolini, M.; Dapporto, P.; Dei, A.; Nardi, N.; Zanobini, F. Inorg. Chem. 1982,21, 489.  (20)  Ciampolini, M.; Dapporto, P.; Nardi, N.; Zanobini, F. / . Chem. Soc, Chem. Commun. 1980, 177.  (21)  Ciampolini, M.; Dapporto, P.; Nardi, N.; Zanobini, F. Inorg. Chim. Acta. 1980,45, L239.  (22)  Kyba, E. P.; John, A. M.; Brown, S. B.; Hudson, C. W.; McPhaul, M. J.; Harding, A.; Larsen, K.; Niedzwiecki, S.; Davis, R. E. / . Am. Chem. Soc. 1980,102,139.  238  References begin on page 237  Chapter 5: ...Bis(amidophosphine) Complexes of Al, Ga and In  (23)  Kyba, E. P.; Davis, R. E.; Hudson, C. W.; John, A. M.; Brown, S. B.; McPhaul, M. J.; Liu, L. K.; Glover, A. C. / . Am. Chem. Soc. 1981,103, 3868.  (24)  Laporte, F.; Mercier, F.; Ricard, L.; Mathey, F. J. Am. Chem. Soc. 1994,116, 3306.  (25)  McAuliffe, C. A. Comprehensive Coordination Chemistry; Pergamon: Oxford, 1987; Vol. 2.  (26)  Caminade, A. M.; Majoral, J. P. Chem. Rev. 1994, 94, 1183.  (27)  Stelzer, O.; Langhans, K. P. The Chemistry of Organophosphorus Compounds; Wiley: Chichester, 1990; Vol. 1.  (28)  Kyba, E. P.; Chou, S. S. P. / . Org. Chem. 1981,41, 860.  (29)  Kyba, E. P.; Clubb, C. N.; Larson, S. B.; Schueler, V. J.; Davis, R. E. / . Am. Chem. Soc. 1985,107, 2141.  (30)  Vogtle, F.; Wittig, G. / . Chem. Educ. 1973, 650.  (31)  Vogtle, F. Chem. Ind. (London) 1972, 346.  (32)  Buter, J.; Kellogg, R. M. / . Org. Chem. 1981,46, 4481.  (33)  For an example of the effect of metal templating on the yield of a macrocycle, see: de Groot, B.; Giesbrecht, G. R.; Loeb, S. J.; Shimizu, G. K. H. Inorg. Chem. 1991,30, 111.  (34)  Coles, S. J.; Edwards, P. G.; Fleming, J. S.; Hursthouse, M. B.; Liyanange, S. S. / . Chem. Soc, Chem. Commun. 1996, 293.  (35)  Fryzuk, M. D.; Love, J. B. unpublished results.  (36)  Akitt, J. W. Prog. NMR Spec. 1989,21,1. 239  References begin on page 237  ChapterS: ...Bis(amidophosphine) Complexes of Al, Ga andln  (37)  Delpuech, J. J. NMR of Newly Accessible Nuclei; Academic Press: New York, 1983; Vol. 2.  (38)  Mole, T.; Jeffrey, E. A. Organoaluminum Compounds; Elsevier: Amsterdam, 1972.  (39)  Coutsolelos, A.; Guilard, R.; Bayeul, D.; Lecomte, C. Polyhedron 1986,5, 1157.  (40)  Wynne, K. J. Inorg. Chem. 1984,23, 4658.  (41)  Atwood, D. A.; Atwood, V. O.; Cowley, A. H.; Atwood, J. L.; Roman, E. Inorg. Chem. 1992,3i,3871.  (42)  Beachley, O. T.; Hallock, R. B.; Zhang, H. M.; Atwood, J. L. Organometallics 1985,4, 1675.  (43)  Carter, J. C ; Jugie, G.; Enjalbert, R.; Galy, J. Inorg. Chem. 1978,17, 1248.  (44)  Boardman, A.; Small, R. W. H.; Worrall, I. J. Inorg. Chim. Acta. 1986,119, L13.  (45)  Wells, R. L.; Aubuchon, S. R.; Self, M. F.; Jasinski, J. P.; Woudenberg, R. C ; Butcher, R. J. Organometallics 1992,11, 3370.  (46)  O'Hare, D.; Foord, J. S.; Page, T. C. M.; Whitaker, T. J. / . Chem. Soc, Chem. Commun. 1991, 1445.  (47)  Banks, M. A.; Beachley, O. T.; Maloney, J. D.; Rogers, R. D. Polyhedron 1990, 9, 335.  (48)  Bradley, D. C ; Chudzynska, H.; Faktor, M. M.; Frigo, D. M.; Hursthouse, M. B.; Hussain, B.; Smith, L. M. Polyhedron 1988,14, 1289.  (49)  Atwood, D. A.; Atwood, V. O.; Cowley, A. H.; Jones, R. A.; Atwood, J. L.; Bott, S. G. Inorg. Chem. 1994,33, 3251.  240  References begin on page 237  Chapter 5: ...Bis(amidophosphine) Complexes ofAl, Ga and In  (50)  Waggoner, K. M.; Ruhlandt-Senge, K.; Wehmschulte, R. J.; He, X.; Olmstead, M. M.; Power, P. P. Inorg. Chem. 1993,32, 2557.  (51)  Brothers, P. J.; Wehmschulte, R. J.; Olmstead, M. M.; Ruhlandt-Senge, K.; Parkin, S. R.; Power, P. P. Organometallics 1994,13,2792.  (52)  Dolphin, D. The Porphyrins; Academic Press: New York, 1978; Vol. 1.  (53)  Muller, G.; Lachmann, J.; Rufinska, A. Organometallics 1992,11, 2970.  (54)  Bennet, F. R.; Elms, F. M.; Gardiner, M. G.; Koutsantonis, G. A.; Raston, C. L.; Roberts, N. K. Organometallics 1992,11, 1457.  (55)  Atwood, J. L.; Butz, K. W.; Gardiner, M. G.; Jones, C.; Koutsantonis, G. A.; Raston, C. L.; Robinson, K. D. Inorg. Chem. 1993,32, 3482.  (56)  Huheey, J. E. Inorganic Chemistry: Principles of Structure and Reactivity; 3rd ed.; Harper & Row: New York, 1983.  (57)  Wehmschulte, R. J.; Ellison, J. J.; Ruhlandt-Senge, K.; Power, P. P. Inorg. Chem. 1994, 33, 6300.  (58)  Cowley, A. H.; Isom, H. S.; Decken, A. Organometallics 1995,14,2589.  (59)  Rakshys, J. W.; Taft, R. W.; Sheppard, W. A. J. Am. Chem. Soc. 1968, 90, 5236.  (60)  Lambert, J. B.; Mueller, D. C. / . Am. Chem. Soc. 1966, 88, 3669.  (61)  Lambert, J. B.; Jackson, G. F.; Mueller, D. C. / . Am. Chem. Soc. 1968, 90, 6401.  (62)  Lambert, J. B.; Jackson, G. F.; Mueller, D. C. / . Am. Chem. Soc. 1970, 92, 3093.  (63)  Baechler, R. D.; Mislow, K. / . Am. Chem. Soc. 1970,92, 4758.  241  References begin on page 237  Chapter 5: ...Bis(amidophosphine) Complexes ofAl, Ga and In  (64)  Egan, W.; Tang, R.; Zon, G.; Mislow, K. / . Am. Chem. Soc. 1970, 92, 1442.  (65)  Egan, W.; Tang, R.; Zon, G.; Mislow, K. / . Am. Chem. Soc. 1971, 93, 6205.  (66)  Cremer, S. E.; Chorvat, R. J.; Davis, D. W. Tetrahedron Lett. 1968,55, 5799.  (67)  Jautelat, M.; Roberts, J. D. / . Am. Chem. Soc. 1969, 91, 642.  (68)  Fessenden, R. J.; Fessenden, J. S. Organic Chemistry; 4th ed.; Brooks/Cole Publishing Co.: Pacific Grove, 1990.  (69)  Cowley, A. H.; Gabbai, F. P.; Isom, H. S.; Decken, A. / . Organomet. Chem. 1995,500, 81.  (70)  Avent, A.; Eaborn, C ; Hitchcock, P. J.; Smith, J. D.; Sullivan, A. C. / . Chem. Soc, Chem. Commun. 1986, 988.  (71)  Beachley, O. T.; Chao, S. H. L.; Churchill, M. R.; See, R. F. Organometallics 1992,11, 1486.  (72)  Cowley, A. H.; Gabbai, F. P.; Isom, H. S.; Carrano, C. J.; Bond, M. R. Angew. Chem., Int. Ed. Engl. 1994,33, 1253.  (73)  Cowley, A. H.; Gabbai, F. P.; Atwood, D. A.; Carrano, C. J.; Mokry, L. M.; Bond, M. R. /. Am. Chem. Soc. 1994,116,1559.  (74)  Churchill, M. R.; Lake, C. H.; Chao, S. H. L.; Beachley, O. T. / . Chem. Soc, Chem. Commun. 1993, 1577.  (75)  Wehmschulte, R. J.; Power, P. P. Inorg. Chem. 1994,33, 5611.  (76)  Klein, C ; Noth, H.; Tacke, M.; Thomann, M. Angew. Chem., Int. Ed. Engl. 1993,32, 886.  242  References begin on page 237  Chapter 5: ..Bis(amidophosphine) Complexes of Al, Ga and In  (77)  Isom, H. S.; Cowley, A. H.; Decken, A.; Sissingh, F.; Corbelin, S.; Lagow, R. J. Organometallics 1995,14,2400.  (78)  Uhl, W.; Layh, M.; Hiller, W. J. Organomet. Chem. 1989,368, 139.  (79)  Veith, M.; Goffing, F.; Becker, S.; Huch, V. / . Organomet. Chem. 1991,406,105.  (80)  Beamish, J. C ; Small, R. W. H.; Worrall, I. J. Inorg. Chem. 1979,18, 220.  (81)  Beamish, J. C ; Boardman, A.; Small, R. W. H.; Worrall, I. J. Polyhedron 1985,4, 983.  (82)  Linti, G.; Kostler, W. Angew. Chem., Int. Ed. Engl. 1996,35, 550.  (83)  He, X.; Bartlett, R. A.; Olmstead, M. A.; Ruhlandt-Senge, K.; Sturgeon, B. E.; Power, P. P. Angew. Chem., Int. Ed. Engl. 1993, 32, 111.  (84)  Uhl, W.; Layh, M.; Hildebrand, T. / . Organomet. Chem. 1989,364, 289.  (85)  Uhl, W. Angew. Chem., Int. Ed. Engl. 1993,32, 1386.  (86)  Bott, S. G.; Elgamal, H.; Atwood, J. L. / . Am. Chem. Soc. 1985,107,1796.  (87)  Bott, S. G.; Alvanipour, A.; Morley, S. D.; Atwood, D. A.; Means, C. M.; Coleman, A. W.; Atwood, J. L. Angew. Chem., Int. Ed. Engl. 1987,26,485.  (88)  Engelhardt, L. M.; Kynast, U.; Raston, C. L.; White, A. H. Angew. Chem., Int. Ed. Engl. 1987,26, 681.  (89)  Davidson, M . G.; Lambert, C ; Lopez-Solera, I.; Raithby, P. R.; Snaith, R. Inorg. Chem. 1995,34, 3765.  (90)  Atwood, J. L.; Robinson, K. D.; Jones, C ; Raston, C. L. / . Chem. Soc, Chem. Commun. 1991,1697.  243  References begin on page 237  Chapter 5: ...Bis(amidophosphine) Complexes of Al, Ga and In  (91)  Robinson, G. H.; Zhang, H.; Atwood, J. L. Organometallics 1987, 6, 887.  (92)  Robinson, G. H.; Sangakoya, S. A. J. Am. Chem. Soc. 1988,110,1494.  (93)  Self, M. F.; Pennington, W. T.; Laske, J. A.; Robinson, G. H. Organometallics 1991,10, 36.  (94)  Taghoif, M.; Heeg, M. J.; Bailey, M.; Dick, D. G.; Kumar, R.; Hendershot, D. G.; Rahbarnoohi, H.; Oliver, J. P. Organometallics 1995,14, 2903.  (95)  Knajazhansky, S. Y.; Nomerotsky, I. Y.; Bulychev, B. M.; Belsky, V. K.; Soloveichik, G. Organometallics 1994,13, 2075.  (96)  Atwood, J. L.; Bott, S. G.; Harvey, S.; Junk, P. C. Organometallics 1994,13, 4151.  (97)  Butz, K. W.; Elms, F. M.; Raston, C. L.; Lamb, R. N.; Pigram, P. J. Inorg. Chem. 1993, 32, 3985.  (98)  Bent, B. E.; Nuzzo, R. G.; DuBois, L. H. / . Am. Chem. Soc. 1989, 111, 1634.  (99)  Greenwood, N. N.; Storr, A.; Wallbridge, M. G. H. Inorg. Chem. 1963,2, 1036.  (100) Wiberg, E.; Amberger, E. Hydrides of the Elements of Main Groups I-IV; Elsevier: Amsterdam, 1971. (101) Storr, A. Ph. D; Thesis, University of Newcastle-Upon-Tyne, 1963. (102) Ali, S. M.; Brewer, F. M.; Chadwick, J.; Garton, G. / . Inorg. Nucl. Chem. 1959, 9, 124. (103) Brewer, F. M.; Chadwick, J. R.; Garton, G. / . Inorg. Nucl. Chem. 1961,23,45. (104) Small, R. W. H.; Worrall, I. J. Acta. Cryst. 1982, B38, 250. (105) Brown, D. S.; Decken, A.; Cowley, A. H. / . Am. Chem. Soc. 1995,117, 5421.  244  References begin on page 237  Chapter 5: ...Bis(amidophosphine) Complexes of Al, Ga and In  (106) Linti, G.; Frey, R.; Schmidt, M. Z. Naturforsch 1994,49b, 958. (107) Saxena, A. K.; Zhang, H.; Maguire, J. A.; Hosmane, N. S.; Cowley, A. H. Angew. Chem., Int. Ed. Engl. 1995,34, 332. (108) Uhl, W. Z. Naturforsch 1988,43b, 1113. (109) Wehmschulte, R. J.; Ruhlandt-Senge, K.; Olmstead, M. M.; Hope, H.; Sturgeon, B. E.; Power, P. P. Inorg. Chem. 1993,32, 2983. (110) Schluter, R. D.; Cowley, A. H.; Atwood, D. A.; Jones, R. A.; Bond, M. R.; Carrano, C. J. /. Am. Chem. Soc. 1993,115, 2070. (111) Gabbai, F. P.; Schier, A.; Riede, J.; Schmidbaur, H. Inorg. Chem. 1995,34, 3855. (112) Beachley, O. T.; Blom, R.; Churchill, M. R.; Faegri, K.; Fettinger, J. C ; Pazik, J. C ; Victoriano, L. Organometallics 1989, 8, 346. (113) Hosmane, N. S.; Lu, K.-J.; Saxena, A. K.; Zhang, H.; Maguire, J. A.; Cowley, A. H.; Schluter, R. D. Organometallics 1994,13, 979. (114) Beachley, O. T.; Pazik, J. C ; Glassman, T. E.; Churchill, M. R.; Fettinger, J. C ; Blom, R. Organometallics 1988, 7,1051. (115) Schumann, H.; Janiak, C ; Gorlitz, F.; Loebel, J.; Dietrich, A. / . Organomet. Chem. 1989, 363, 243. (116) Tuck, D. G. Chem. Soc. Rev. 1993,113,269. (117) Love, J. B., personal communication.  245  References begin on page 237  Chapter 6  Future Work  6.1 Introduction This thesis has been concerned with the stabilization of high valent, early and main group metals with ligands incorporating both amides and phosphines.  This work began with an  examination of the chemistry of scandium, aluminum and lithium complexes using the [N(SiMe2CH2PPri2)2l" ligand system, and continued with a study of aluminum, gallium and indium compounds utilizing the third generation ligand system [P2N2] ". As with any research 2  project, not all avenues of research were explored, with some topics only briefly touched upon or future directions alluded to. In this final chapter, we report some ideas for future research, as well as some initial findings which bear mention.  6.2 Scandium Compounds Involving New Ligands Much of the work in this thesis has centered on the synthesis and reactivity of scandium dialkyl complexes.  Chapter 2 presented the effect of different ligand environments on the  stability of scandium alkyls and dialkyls (A - C). Although characterizable species were attainable in all cases, only B yielded alkyl complexes which underwent controlled reactions with proton sources, even though these too were accompanied by insertion into the scandium-amide bond, and decomposition to the protonated ligand, HN(SiMe CH PPr )2. i  2  246  2  2  References  begin  on page 270  Chapter  6: Future  Work  Pr'2  Me Si^  ^^R  Me Si  Ue S\(  f  Me?Si  S  2  2  P Pr  2  R  P PH  ! 2  2  A  C  B  The scarcity of scandium phosphine complexes in the literature attests to the poor match of donor and acceptor in these complexes. " Even though scandium-phosphine complexes of the general 1  5  type A - C are stable enough to be crystallographically characterized, phosphine dissociation is the likely source of their instability when reacted with small molecules. Recent work has shown that scandium alkyls can be stabilized by a ligand system in which all the donors are nitrogens. Complexes of the type CnMX (Cn = l,4,7-trimethyl-l,4,7-triazacyclononane; M = Sc, Y; X = 3  Cl, Me) can be reacted with B ( C 6 F 5 ) or [H(CH3)2NPh]+[B(C6F5)4]" to give cationic species 3  which polymerize olefins.  6  CH  X  3  I  X  X  M = Sc, Y X = Cl, C H  247  3  References  begin  on page 270  Chapter  6: Future  Work  6.2.1 Synthesis and Structure of ScCl2[N(SiMe CH NMe2)2] (1) 2  2  The ease with which the [N(SiMe2CH2PR2)2]~ ligand system attaches to a wide range of metals provided the initiative to examine the related all nitrogen-donor ligand [N(SiMe2CH2NMe2)2]"-  7  Since amines are known to be better donors for early, high valent  metals than phosphines, complexes of the type ScX2[N(SiMe2CH2NMe2)2] (X = chloride, 8  alkyl) should be more stable than ScX2[N(SiMe2CH2PPr 2)2] (Chapter 2). The synthesis of this i  starting material is detailed below (equation 6.1).  1/2 {LiN(SiMe2CH NMe )2}2 +  ScCI (THF)  2  2  Me N  Me Si 2  Me Si  \  3  toluene 3  (- LiCl)  2  ic  [6.1]  cl  2  N Me  1 1 2  The reaction of 0.5 equiv of the lithium amide dimer with ScCl3(THF)3 in toluene at room temperature yields scandium dichloride 1 as a colorless powder. The *H and ^Cf^H} NMR spectra both exhibit three narrow resonances for the SiMe2, CH2 and NMe2 groups, suggesting a symmetric solution structure. Crystals of 1 suitable for X-ray analysis were obtained from slow evaporation of a saturated toluene solution; Figure 6.1 shows the molecular structure and numbering scheme. The complex is monomeric with exact C2 symmetry and adopts a distorted trigonal bipyramidal geometry, with the ligand bound in a meridional fashion.  248  References  begin  on page  270  Chapter 6: Future Work  Figure 6.1 Molecular structure of ScCl [N(SiMe2CH NMe2)2] (1); 33% probability 2  2  thermal ellipsoids are shown. Hydrogen atoms are omitted for clarity.  Table 6.1 Selected bond lengths (A) and angles (deg) for S c C l [ N ( S i M e C H N M e ) ] (l). 2  a  Cl(l)-Sc(l)  2.3758(6)  N(2)-Sc(l)  2.319(2)  Cl(l)-Sc(l)-Cl(l)*  2  2  2  a  2  N(l)-Sc(l)  2.047(2)  116.51(4)  N(2)-Sc(l)-N(2)*  165.77(9)  N(l)-Sc(l)-N(2)  82.89(4)  Cl(l)-Sc(l)-N(l)  121.75(2)  Cl(l)-Sc(l)-N(2)  93.26(4)  Cl(l)-Sc(l)-N(2)*  94.21(5)  Symbols refer to symmetry operations: (*) 1-x, y, l/2-z.  Many of the parameters determined for dichloride 1 are similar to those of ScEt [N(SiMe2CH2PPr 2) ] (see Section 2.3.2); the angle defined by the two amines and the 1  2  2  249  References begin on page 270  Chapter  6: Future  Work  scandium nucleus (165.77(9)°) is similar to the P(l)-Sc(l)-P(2) angle of 166.95(4)°. Also, the Cl(l)-Sc(l)-Cl(l)* angle in 1 of 116.51(4)° can be compared to the C(10)-Sc(l)-C(10)* angle of 124.8(2)° in the diethyl species. The scandium-amide bond (2.047(2) A) is shorter than that in either of the ScR2[N(SiMe2CFJ.2PPr 2)2] derivatives and is similar to the metal-amide bond i  distance in the zirconium and hafnium analogues, MCl3[N(SiMe2CH2NMe2)2] (M = Zr, Hf) (2.056(3) and 2.048(4) A, respectively). The scandium-amine bonds (2.319(2) A) compare well 7  with the pendant amine-scandium bond in l/?-fra^-l'5-[(Cp SiNR)ScH]2 (2.452(5) A). A NMe  9  partial list of bond angles and lengths can be found in Table 6.1. The metathesis of the chloride ligands of 1 with typical alkyl lithium reagents should be straightforward; as for ScCl2(THF)[N(SiMe2CH2PPr 2)2], i  monomeric scandium dialkyls are  expected (equation 6.2).  The chemistry of these scandium dialkyls with small molecules would provide a comparison to those studied in Chapter 2, and may help to illuminate the role that the pendant donor group plays is the stability of these scandium alkyl complexes.  250  References  begin  on page 270  Chapter 6: Future Work  6.3 Bis(amidophosphine) Complexes of Titanium Chapter 5 presented a study of the effect of a proximate Group 13 Lewis acid on the pyramidal inversion of phosphorus in the macrocyclic ligand [P2N2]. Earlier studies showed that this type of behavior was not limited to Group 13 metals, but was also promoted by early metals such as scandium. However, the ease of this conversion did not allow for the isolation of anti10  ScCl[P2N2]. In an effort to study the effect of a proximate transition metal on the rate of pyramidal inversion of phosphorus, we recognized it was necessary to synthesize antiMC1 [P2N2] (M = transition metal), in which only one phosphorus of the macrocycle was bound X  to the metal. Our investigations led us to examine the coordination chemistry of titanium. Although this titanium system did not ultimately undergo pyramidal inversion, some preliminary studies of the chemistry of this species were undertaken.  6.3.1 Synthesis and Structure of anri-TiCl [P N ] (2) 2  2  2  The reaction of a/iri-Li (THF)2[P2N2] with TiCL* in toluene at 25 "C for 12 h generates 2  the monomeric dichloride a«ft'-TiCl2[P2N2] (2) in excellent yield (equation 6.3). The H NMR l  spectrum of the product shows four singlets arising from the silyl methyl protons and two distinct ortho- phenyl resonances. The P{ H} NMR spectrum of 2 consists of two singlets at -5.5 and 31  l  -40.8 ppm. As for the complexes discussed in Chapter 5, these spectral features are indicative of a metal centre which is ligated by two amides, but only a single phosphine. Evaporation of a saturated toluene solution of 2 yielded small red crystals suitable for X-ray crystallography. The molecular structure and numbering scheme are presented in Figure 6.2.  As in the crystal  structure of anft-GaCl[P2N2] (Chapter 5), the complex is monomeric, with one phosphine remaining uncoordinated.  251  References begin on page 270  Chapter 6: Future  Me  2  Me?  SL  Ph  Si Me  2  SI Me  +  TiCI  toluene 4  (-2 LiCl)  Me  2  .Si 2  2  2  Ph-  Me  2  anfALi (THF) [P N ] 2  Work  Si.  / 0  CI  [6.3]  Ti-«—Fi  Si  Me  Si  Me  2  ( ) N(2) Sl  2  2  2  C(8)  C(10)  C(22)  Figure 6.2.  Molecular structure of anft-TiCl2[P2N2] (2); 33% probability thermal ellipsoids are shown. Hydrogen atoms are omitted for clarity.  252  References begin on page 270  Chapter 6: Future Work  The titanium nucleus exhibits trigonal bipyramidal geometry, the axial positions being occupied by P(l) and Cl(2) (P(l)-Ti(l)-Cl(2) = 163.77(6)°). The angles between the equatorial atoms (N(l), N(2) and Cl(l)) and titanium range from 107.9(2) to 124.54(14)°, the smallest angle (N(l)-Ti(l)-N(2)) reflecting the constraints of the ligand system. The two titanium-chloride bonds define an angle of 86.72(6)°, with each chloride directed towards a different phosphine. The titanium-amide bond lengths of 1.893(5) and 1.905(4) A are normal and compare to a value of 1.896(3) A in the bis(silyl)amide TiCl3[N(SiMe2CH2NMe2)2]. Bond lengths and angles of 7  interest are given in Table 6.2.  Table 6.2 Selected bond lengths (A) and angles (deg) for artri-TiCl2[P2N2] (2).  Cl(l)-Ti(l)  2.339(2)  Cl(2)-Ti(l)  2.3233(15)  P(l)-Ti(l)  2.542(2)  N(l)-Ti(l)  1.905(4)  N(2)-Ti(l)  1.893(5)  Cl(l)-Ti(l)-Cl(2)  86.72(6)  N(l)-Ti(l)-N(2)  107.9(2)  Cl(l)-Ti(l)-P(l)  80.82(6)  Cl(2)-Ti(l)-P(l)  163.77(6)  Cl(l)-Ti(l)-N(l)  122.12(15)  Cl(l)-Ti(l)-N(2)  124.54(14)  Cl(2)-Ti(l)-N(l)  108.63(13)  Cl(2)-Ti(l)-N(2)  99.56(13)  N(l)-Ti(l)-P(l)  86.91(13)  N(2)-Ti(l)-P(l)  79.35(12)  6.3.2 Phosphorus Inversion Behavior of anft'-TiCl2[P2N2l (2) To study the potential for phosphorus inversion, solutions of 2 were heated at 50, 70 and 100 °C. Unfortunately, the results of this study were not straightforward. For example, the 31  P{ H} NMR spectrum of the sample heated to 100 °C exhibited a small peak (<10%) at 0.9 1  253  References begin on page 270  Chapter 6: Future Work  ppm due to ^y«-TiCl2[P2N2].  10  However, the spectrum was complicated by numerous other  resonances which were not assignable. We were unable to isolate 5yw-TiCl2[P2N2] cleanly from this reaction, despite numerous attempts.  Heating at 70 °C yielded a different product,  characterized by an A B doublet in the P { H } N M R spectrum at -7.7 and -12.3 ppm ( /p.p = 31  1  2  120 Hz), while no change was observed at 50 °C. We cannot state with certainty the nature of the product which gives rise to the A B doublet.  6.3.3 Reactivity of an/i-TiCI [P N ] (2) 2  2  2  The reaction of anri-TiCl [P N ] (2) with alkylating reagents such as methyl-, neosilyl2  2  2  or bis(trimethylsilyl)methyllithium (two equiv) either resulted in reduction to Ti(III) species, or mixtures  of  unidentifiable  products.  This pattern of reactivity  mirrors that  of  TiCl3 N(SiMe CH PPr ) ], which leads to mixtures of products upon reaction with numerous r  i  2  alkyl sources.  11  2  2  2  We also attempted the reduction of anri-TiCl [P N ] (2) with both KCs and M g 2  2  2  in an effort to synthesize a titanium dinitrogen complex. Both TiCl3[N(SiMe CH PPr 2)2] and i  2  5y«-TiCl2[P2N2]  10  11  2  can be reduced to the corresponding dimeric dinitrogen compounds, the  former having been crystallographically characterized. In contrast to these systems, the reaction of arttt'-TiCl2[P2N2] (2) with either KCg or M g yielded Ti(III) products which did not coordinate dinitrogen.  6.4 Bis(amidophosphine) Complexes of Vanadium As part of our continued interest in the coordination of early metals by the macrocyclic ligand system [P N2], we examined the chemistry of some [P2N2] complexes of vanadium. Part 2  of the impetus to study vanadium compounds was that they might serve as precursors to dinitrogen complexes of vanadium. The fixation of dinitrogen by vanadium has attracted  254  References begin on page 270  Chapter  6: Future  Work  considerable attention since the discovery by Shilov that a vanadium/magnesium system could catalytically transform dinitrogen into N2H4 and  NH3.  1 2  More recently, the characterization of a  vanadium-containing nitrogenase in nitrogen-fixing bacteria such as Azotobacter and 13  Anabaena has prompted investigations into the chemistry and structure of vanadium complexes 14  containing dinitrogen fragments. " 15  18  Earlier work from our laboratory described that the vanadium-dinitrogen complex {[N(SiMe2CH2PPr 2)2]VCl}2(P--N2) could be synthesized by exposing the vanadium (II) starting i  material {[N(SiMe2CH PPr^)2]V(u,-Cl}2Li(TMEDA) to N 2 . 2  11  Unfortunately, we were unable  to generate the analogous macrocyclic V(U) species using either the anti- or the syn- isomers of the [P2N2] ligand. Another potential route to vanadium-dinitrogen species is via the reduction of V(III) compounds in the presence of N2. A similar methodology has been used to prepare the macrocyclic Zr(IV) complex {£yn-[P2N2]Zr} (|J.-N2).  1!>  2  Since our attempts to synthesize anti-  VC1[P2N2] also met with failure, we examined the use of the syn- isomer of the [P2N2] ligand system to yield the vanadium (III) precursor, ,syw-VCl[P2N2].  6.4.1 Synthesis of syn-VCl[P N ] (3) 2  2  The addition of a toluene solution of £ytt-Li (dioxane)[P2N2] to a slurry of VCl3(THF)3 2  results in 5yn-VCl[P2N2] (3) (equation 6.4). The mass spectrum (EI) shows a parent ion peak at 619 mle and the microanalysis is consistent with VC1[P2N2]. Although the MS indicates a monomeric structure, the solution magnetic moment (toluene) of 3 (|i ff = 2.44 B.M., |J. alc = e  C  2.89 B.M., V(III), d ) may result from weak association in solution via chloride bridges, or spin2  orbit coupling.  20  We attempted the synthesis of a dinitrogen complex of vanadium by the  reduction of this compound in the presence of N2.  255  References  begin  on page 270  Chapter  Me  Ph-  2  Me Si  + VCI (THF) 3  3  Ph  Si Me  P  toluene (-2 LiCl)  Me Si.  on Me2  .Si  sy/7-Li (dioxane)[P N ] 2  Work  2  OO 'S Me  6: Future  2  2  Ph—P  r  /*Si  »~V  v 1 a  2  1-  \  [6.4]  •Ph  Si  Me  M62  2  6.4.2 Synthesis and Structure of {syn-[P2N ]V}2(|i-N2) (4) 2  The addition of a toluene solution of ^yn-VCl[P2N2] (3) to a slurry of KCs (1.1 equiv) in toluene under an atmosphere of N2 yields the dinitrogen species {.yyrt-[P2N2]V}2(M.-N2) (4) (equation 6.5). Compound 4 is isolated as a dark purple solid which crystallizes from toluene as purple needles. The H and P{ H} NMR spectra of 4 consist of broadened and shifted l  31  1  resonances which yield little useful information. The solution magnetic moment (toluene) of 1.39 B.M. per metal centre indicates an oxidation state of +4 for each vanadium nucleus, and is consistent with 1 unpaired electron per metal. This value is lower than the spin only value of 1.73 B.M. and may be due to spin-orbit coupling.  20  The ESR spectrum (toluene solution)  exhibits an octet of binomial triplets at g = 1.980 due to coupling to the I = 7/2 vanadium nucleus, and two equivalent phosphorus-31 nuclei (Figure 6.3).  256  References  begin  on page 270  Chapter  Me  2  6: Future  Work  Me2  [6.5]  (silyl methyl groups omitted for clarity)  The hyperfine coupling constant a( P) of 26.1 G is comparable to the vanadium 31  dinitrogen complex {[N(SiMe2CH2PPr ) ]VCl}2( t-N2) (a( P) = 28.0 G), which exhibits a i  31  2  2  r  similar ESR spectrum. The hyperfine coupling (a( V) = 84.5 G) to a single vanadium nucleus 11  51  ( V, I = 7/2, 99.8% natural abundance) implies that the electron is relatively confined, and 51  21  does not couple across the N2 bridge. This is also the case for {[N(SiMe2CH2PPr )2]VCl}2(|J.i 2  N ) (a( V) = 88.4 G) which is known to be dimeric in the solid state. Due to limitations of the 51  11  2  ESR simulation software, the tumbling of the molecule (i.e., outer peaks are less intense than inner peaks) could not be modelled.  257  References  begin  on page 270  Chapter  6: Future  Work  a)  b)  Figure 6.3 a) Experimental and b) simulated room temperature solution ESR spectrum of {5yn-[P N2]V}2(il-N2) (4). 2  The mass spectrum (EI) of 4 shows a parent ion peak at 1194 mle, corresponding to a dimeric species. This is supported by the IR spectrum (KBr disc), which shows no peaks diagnostic of an N2 unit. Following the evaporation of a toluene solution, deep purple crystals of 4 remained. The molecular structure and numbering scheme are shown in Figure 6.4.  258  References  begin  on page 270  Chapter 6: Future Work  CI39I  Figure 6.4 Molecular structure of {syn-[P N2]V}2(fx-N2) (4); 33% probability thermal 2  ellipsoids are shown. Hydrogen atoms are omitted for clarity.  The X-ray analysis reveals that 4 is dimeric, with two VP2N2] units bridged by an endon N molecule, the overall molecule having C2 symmetry. 2  Each vanadium centre can be  described as a distorted trigonal bipyramid, the largest deviation displayed by the P(l)-V(l)-P(2) angle of 154.24(11)°. P(3)-V(2)-P(4) exhibits a similar distortion (159.66(11)°). The V - N - V 2  linkage is essentially linear, with V - N - N angles of 178.2(7)° and 179.3(6)°. The two V [ P N ] 2  2  fragments are rotated with respect to each other (44.4°) to minimize steric repulsion between the phenyl groups. The N2 unit has a bond length of 1.232(9) A , which is similar to those previouslyreported for neutral vanadium dinitrogen species.  15-18  The vanadium-phosphorus bond lengths  range from 2.472(3) to 2.491(3) A , and are slightly shorter than the V - P bonds in the acyclic analogue, {[N(SiMe CH2PPr ) lVCl}2(pi-N2) (2.519(2) and 2.530(2) A ) . i  2  2  2  11  Bond lengths and  angles for 4 are tabulated below.  259  References begin on page 270  Chapter 6: Future Work  Table 6.3 Selected bond lengths (A) and angles (deg) for {.yyrt-[P N ]V} (u.-N ) (4). 2  2  2  2  N(l)-V(l)  1.763(8)  N(4)-V(2)  1.772(8)  N(2)-V(l)  2.060(7)  N(5)-V(2)  2.054(7)  N(3)-V(l)  2.032(7)  N(6)-V(2)  2.065(7)  P(l)-V(l)  2.473(3)  P(3)-V(2)  2.472(3)  P(2)-V(l)  2.491(3)  P(4)-V(2)  2.472(3)  N(l)-N(4)  1.232(9)  P(l)-V(l)-P(2)  154.24(11)  P(3)-V(l)-P(4)  159.66(11)  N(l)-V(l)-N(2)  122.6(3)  N(4)-V(l)-N(5)  122.7(3)  N(l)-V(l)-N(3)  119.9(3)  N(4)-V(l)-N(6)  124.3(3)  N(2)-V(l)-N(3)  117.5(3)  N(5)-V(l)-N(6)  112.9(3)  N(4)-N(l)-V(l)  178.2(7)  N(l)-N(4)-V(2)  179.3(6)  The N-N distance of 1.232(9) A, considerably longer than that of free dinitrogen (1.098 A) suggests that the dinitrogen fragment has been reduced. This distance falls in the general range of nitrogen atoms connected by a double bond, or a N has a bond distance of 1.255 A .  2 2  2 _ 2  unit. By comparison, PhN=NPh  However, the correlation between the N-N bond length and  the extent of reduction of the N unit is not as strong for vanadium-dinitrogen complexes as for 2  other early metal-dinitrogen systems. The traditional views of end-on dinitrogen bonding are shown below. In D , the N ligand has a bond order of 3 and a formal oxidation state of 0; 2  therefore, the predicted N-N bond length would be short (close to that of free dinitrogen) and the M-N distances typical of single bonds. The alternate view, E , involves a reduced form of dinitrogen, in which the N ligand has a formal charge of -4. This structure is characterized by 2  short M-N and long N-N bond lengths.  260  References begin on page 270  Chapter  M-  •M  •N=N-  M—N  6: Future  Work  N—M  Although the V-Ndinitrogen bond lengths (1.763(8) and 1.772(8) A) in 4 are relatively long compared to most known V=N bonds in V(IV) or V(V) systems (usually 1.60 - 1.68 A ) ,  23  they  compare favorably with those determined for {[N(SiMe2CH2PPr^)2]VCl}2(|i-N2) (1.760(4) A),  11  which is thought to contain two V(IV) centres connected by a hydrazido unit. A medium  intensity absorption at 833 c m  - 1  in the IR spectrum of 4 can be assigned to the V=N stretch, in  good agreement with that observed for [(Me3CCH2)3V] (p.-N ) (858 c n r ) 1  2  dinitrogen complexes of Mo and W . 2 4  2 5  1 7  2  and a number of  Thus, the vanadium-dinitrogen complex 4 is best  described as containing two V(IV) centres, implying that the dinitrogen fragment is likely a hydrazido unit, N  4 _ 2  , as in E .  6.5 Synthesis and Structure of a/itf-(CdCl) [P N2] (5) 2  2  Although all previously synthesized transition metal [P2N2] complexes have been mononuclear, some recent work shows that dinuclear derivatives are also possible. A compound with a dinuclear structure was isolated from the reaction of anri-Li2(THF)2[P2N2] with C d C l 2 in THF at -78 °C (equation 6.6).  This reaction produces at least two major products; anti-  (CdCl)2[P2N2] (5) can be obtained pure from a toluene solution by fractional crystallization.  261  References  begin  on page  270  Chapter 6: Future Work  Me  2  Me  2  [6.6]  + other products (silyl methyl groups omitted for clarity)  The two silyl methyl peaks and single ortho proton signal in the ! H NMR spectrum of 5 are indicative of a symmetric product. The presence of two spin-active isotopes of cadmium ( Cd, 113  I = 1/2, 12.3% natural abundance; C d , I = 1/2, 12.8% natural abundance) is evident in the n i  31  21  P{ *H} NMR spectrum (Figure 6.5a)), which consists of a singlet (8 -27.0 ppm) flanked by two  sets of satellite peaks due to the phosphorus-cadmium interactions (Vp.incd = 1739 Hz, Vp.nicd = 1659 Hz). The singlet also implies that the two phosphorus nuclei are in similar magnetic environments.  262  References begin on page 270  Chapter  6: Future  Work  a)  b)  ' i i  i i  i | i  i i  470  Figure 6.5 a) 202.4 MHz P{ lH} and b) 66.7 MHz 31  113  i  i i i ppm  i  Cd{ B] NMR spectra of antil  (CdCl) [P N ] (5) in C6D . Satellites in the P{ H) NMR spectrum are due to 31  2  2  2  l  6  ^P-Cd couplings.  The  113  Cd{ H} NMR spectrum (Figure 6.5b)) displays a doublet at 477.8 ppm. Each 1  cadmium nucleus couples to a single phosphorus donor, with the coupling matching that found in the P{ H} NMR spectrum (Vcd-P = 1739 Hz). These data are consistent with the 31  1  complexation of two cadmium centres per [P N ] ligand. Crystals of 5 were grown from 2  263  2  References  begin  on page 270  Chapter 6: Future  Work  toluene; the molecular structure and numbering scheme are shown in Figure 6.6.  The anti-  configuration of the ligand is retained, while binding to two tetrahedral cadmium centres.  C(12)  Figure 6.6 Molecular structure of anri-(CdCl)2[P2N2] (5); 33% probability thermal ellipsoids are shown. Hydrogen atoms are omitted for clarity.  The molecule exhibits C h symmetry with the rotation axis passing through the N ( l ) 2  N(1A) vector. Each cadmium nucleus is best described as a distorted tetrahedron with the bond angles ranging from 86.72(5)° to 134.01(3)°. Although Cd[N(SiMe )2]2 was first reported over 3  30 years ago,  26  the number of structurally characterized complexes containing cadmium-amide  bonds is small. The four bonds defining the Cd2N2 plane in 3 are slightly longer than those in (CdCp*[N(SiMe3)2])2,  27  but  [C H402(Me SiNBu )2]2Cd2. t  6  2  28  similar  to  the  cadmium-amide  bond  lengths  in  The Cd-Cd(A) distance of 3.2094(3) A is slightly longer than  the sum of the covalent radii of two tetrahedral cadmium atoms (2.96 A ) , placing it outside the 29  264  References begin on page 270  Chapter  6: Future  Work  range normally associated with a metal-metal bond. The cadmium-phosphorus bonds (2.5198(6) A) are shorter than the bonds found in simple donor species such as CdCl2(PPh3)2 (2.635(5) and 2.632(6) A ) or Cdl2(PPh >2 (2.630(3) and 2.654(3) A ) . 2 9  30  3  Selected bond lengths and angles for  5 can be found in Table 6.4.  Table 6.4 Selected bond lengths (A) and angles (deg) for a«f/-(CdCl)2[P2N2] (5).  a  N(l)-Cd  2.333(2)  N(1A)-Cd  2.342(2)  Cl-Cd  2.3737(7)  P(l)-Cd  2.5198(6)  Cd-Cd(A)  3.2094(3)  N(l)-Cd-N(1A)  93.28(6)  N(l)-Cd-Cl  118.05(5)  N(1A)-Cd-Cl  122.74(5)  N(l)-Cd-P(l)  86.72(5)  N(l)-Cd-P(1A)  90.77(5)  Cl-Cd-P(l)  134.01(3)  Symbols refer to symmetry operations: (A) -x, -y+1, -z+2  Although the Cd-Cd distance in 5 is probably too long for a metal-metal interaction to be present, the choice of a suitable metal may result in dinuclear species F, with the metals in close proximity. Alternately, the simultaneous complexation of two different metals by the [P2N2] ligand system may result in heterobimetallic systems G .  3 1  To date, CpTi(|i.-NMe2)3M(CO)3 (M  = Cr, Mo, W) is the only reported early metal-late metal heterobimetallic system involving bridging amides.  32  265  References  begin  on page 270  Chapter 6: Future Work  R  R  Ph  Ph  Ph R  F:  Mi  =M  G l M i if M  2  2  (silyl methyl groups omitted for clarity) 6.6 Conclusions These are some suggestions for future work. It is our opinion that the high reactivity of scandium was never fully realized by the ligand systems used here; the modification of the coordination environment along the lines suggested may result in a fuller appreciation of the useful properties of scandium as well as other efi lanthanide metals. The advent of the macrocyclic ligand [P2N2] allows for a new level of investigation into early transition metal complexes when coordinating in a "traditional" mode, while the new coordination geometries briefly mentioned here may suggest new examinations into the more established chemistry of the later members of the transition series.  6.7 Experimental Section 6.7.1 Procedures Unless otherwise stated, general procedures were performed according to Section 2.16.1. 13  C{ H) NMR spectroscopy was performed on a Bruker AC-200 spectrometer operating at 50.3 l  266  References begin on page 270  Chapter 6: Future Work MHz.  113  Cd{ H) NMR spectroscopy was performed on a Varian XL-300 instrument operating l  at 66.7 MHz.  113  Cd{ H} NMR spectra were referenced to external CdC\ in D 0 (0.0 ppm). 1  2  2  ESR spectroscopy was performed on a Bruker ECS 106 spectrometer. ESR simulations were done using WIN-Simphonia. The values of line broadening and the coupling constants were obtained from the simulated spectra. Magnetic moments were measured by a modification of the Evans' method using C6D5H or Cp Fe as a reference peak.  33,34  2  6.7.2 Materials {LiN(SiMe CH NMe ) } was prepared by a published procedure. TiCU was used as a 7  2  2  2  2  2  I. 0 M solution in toluene (Aldrich). VCl3(THF)3 was prepared by a published procedure.  35  CdCl was purchased from Mallinkrodt and sublimed prior to use. 2  6.7.3 Syntheses 6.7.3.1 ScCl [N(SiMe CH NMe ) ] (1) 2  2  2  2  2  To a slurry of ScCl3(THF)3 (798 mg; 2.18 mmol) in toluene (10 mL) was added a toluene solution (10 mL) of {LiN(SiMe CH NMe ) } (501 mg; 0.990 mmol). The reaction 2  2  2  2  2  vessel was then stirred for 12 h. The reaction mixture was passed through a frit lined with Celite to remove LiCl. The solvent was then removed in vacuo to yield a white solid. The residue was taken up in a minimum amount of toluene (approx. 5 mL). Slow evaporation of the solvent afforded large colorless blocks (210 mg; 29% yield). H NMR (C6D6): 8 2.38 (s, 12H, NMe ), !  2  2.03 (s, 4H, CH ), 0.06 (s, 12H, SiMe ). ^Cf^H} NMR (C D ): 8 52.5 (CH ), 50.8 (NMe ), 2  2  6  6  2  2  3.0 (SiMe )- MS: mle 363 (M+). Anal. Calcd. for Ci H 8Cl N3ScSi : C, 33.14; H, 7.79; N, 2  0  2  2  2  II. 60. Found: C, 33.34; H, 7.93; N, 11.47.  267  References begin on page 270  Chapter 6: Future Work  6.7.3.2 a/irt-TiCl [P N ] (2) 2  2  2  A toluene solution of TiCLt (1.5 mL of a 1.0M solution, 1.50 mmol) was added to a toluene solution (10 mL) of a/jri-Li (THF) [P N ], (1000 mg; 1.47 mmol). The solution 2  2  2  2  became a deep cloudy red. The reaction mixture was stirred for 12 h, and then passed through a frit lined with Celite to remove LiCl. The solvent was then removed in vacuo and extracted with hexanes to yield a red solid. Crystallization from toluene yielded small, red prisms (912 mg; 96% yield). H NMR (C D ): 5 7.84 and 7.63 (t, 4H, o-Ph, 3/ _ = 7.8 Hz, 3/ .p = 7.8 Hz), l  7  8  H  H  H  7.20 - 6.95 (m, 6H, m,p-Ph), 2.77 (br d, 2H, ring CH , / - H = 13.0 Hz), 1.48 (dd, 2H, ring CH , 2  H  2  2  2  / H - H = 13.0 Hz, 2/H-P = 13.0 Hz), 0.98 (ABX m, 4H, ring CH , 2 / . = 9.0 Hz, / . = 2.0 2  2  H  H  H  P  Hz), 0.57,0.45,0.32 and 0.12 (s, 12H, SiMe ). P{ H} NMR (C D ): 5 -5.5 (s, IP, coordinated 31  X  2  6  6  phosphine), -40.8 (s, IP, uncoordinated phosphine). MS: mle 616 (M+ - Cl). Anal. Calcd. for C H 4 C l N P S i 4 T i : C, 44.23; H, 6.50; N, 4.30. Found: C, 44.18; H, 6.54; N, 4.10. 24  2  2  2  2  6.7.3.3 syn-VCl[P N ] (3) 2  2  To a slurry of VC1 (THF) (147 mg; 0.393 mmol) in toluene (10 mL) was added a 3  3  toluene solution (10 mL) of ^yn-Li (dioxane)[P N ], (250 mg; 0.394 mmol). The reaction 2  2  2  mixture was stirred for 12 h, and then passed through a frit lined with Celite to remove LiCl. The solvent was then removed in vacuo and the oil extracted with hexanes to yield an orange solid (181 mg; 74% yield).  MS: mle 619 (M+). |i ff = 2.44 B.M. e  Anal. Calcd. for  C 4H42ClN P2Si4V: C, 46.55; H, 6.84; N, 4.52. Found: C, 46.88; H, 6.71; N, 4.41. 2  2  6.7.3.4 {sy/i-[P N ]V} (p.-N ) (4) 2  2  2  2  To a slurry of K C (40 mg; 0.296 mmol) in toluene (10 mL) was added a toluene 8  solution (10 mL) of 5y/7-VCl[P N ] (160 mg; 0.258 mmol). The reaction mixture was stirred for 2  2  268  References begin on page 2 70  Chapter  6: Future  Work  12 h, during whichtimethe solution gradually changed from orange to dark purple. The reaction mixture was then passed through a frit lined with Celite to remove KC1 and C(gr). The solvent was removed in vacuo to yield a purple solid. The residue was taken up in a minimum amount of toluene (approx. 2 mL) and placed in the fridge. Overnight, large deep purple needles formed (105 mg; 35% yield). ESR (C H ): g = 1.980; a ( V ) = 84.5 G, I V , a( P) = 26.1 G, 2P; 51  7  31  8  linewidth used for simulation = 10.0 G. MS: mle 1196 (M+). |i ff = 1.39 B.M. IR (KBr disc): e  V = N = 833 cm-1 (m). Anal. Calcd. for C4 H84N6P4Si8V (+ 0.5 C H ) : C, 49.81; H, 7.14; N, V  8  2  7  8  6.77. Found: C, 49.85; H, 7.19; N, 6.25.  6.7.3.5  anri-(CdCl) [P N ] 2  2  (5)  2  To a cooled (-78 °C) solid mixture of CdCl (53 mg; 0.289 mmol) and anti2  Li (THF) [P N ] (200 mg; 0.289 mmol) was added 10 mL TFfF. The resultant solution became 2  2  2  2  a cloudy pale yellow. The mixture was left to slowly warm to room temperature with stirring, during which time the cloudiness abated. The volatiles were removed under vacuum and the residue extracted with toluene. The solution was filtered through Celite and the volume of the solvent reduced. Slow evaporation afforded ann'-(CdCl) [P N ] as large blocks (35 mg; 15% 2  2  2  yield). l H NMR (C D ): 5 7.75 (t, 4H, o-Ph, / - H = 12.0 Hz, 3/ -P = 12.0 Hz), 7.50 (m, 6H, 3  6  6  H  H  m,p'Ph), 1.60 (ABX m, 8H, ring CH , / H - H = 12.0 Hz, / - P = 12.0 Hz), 0.45 and 0.15 (s, 12H, 2  2  2  H  SiMe ). 31p{lH} NMR (C D ): 5 -27.0 (s, 2  6  1 1 3  6  C d satellites, J .m d  = 1739 Hz; l " C d  l  P  C  satellites, Vp-mcd = 1659 Hz). USCdpH} NMR ( C ^ ) : 5 477.8 (d, J jnca l  P  = 1739 Hz).  MS: mle 828 (M+). Anal. Calcd. for C 4H4 N P Si4Cl2Cd : C, 34.79; H, 5.11; N, 3.58. Found: 2  2  2  2  2  C, 36.04; H, 5.75; N, 2.50.  269  References  begin  on page 270  Chapter 6: Future Work  6.8 References (1)  Fryzuk, M. D.; Berg, D. J.; Haddad, T. S. Coord. Chem. Rev. 1990, 99, 137.  (2)  Piers, W. E.; Shapiro, P. J.; Bunel, E. E.; Bercaw, J. E. Synlett. 1990, 74.  (3)  Schaefer, W. P.; Kohn, R. D.; Bercaw, J. E. Acta. Cryst. 1992, c48, 251.  (4)  Shapiro, P. J.; Bunel, E.; Schaefer, W. P.; Bercaw, J. E. Organometallics 1990,9, 867.  (5)  Shapiro, P. J.; Cotter, W. D.; Shaefer, W. P.; Labinger, J. A.; Bercaw, J. E. / . Am. Chem. Soc. 1994,116, 4623.  (6)  Hajela, S.; Schaefer, W. P.; Bercaw, J. E. / . Organomet. Chem. 1997,532, 45.  (7)  Fryzuk, M. D.; Hoffman, V.; Kickham, J. E.; Rettig, S. J.; Gambarotta, S. Inorg. Chem. 1997,36, 3480.  (8)  Togni, A.; Venanzi, L. M. Angew. Chem., Int. Ed. Engl. 1994,33,497.  (9)  Mu, Y.; Piers, W. E.; MacQuarrie, D. C ; Zawarotko, M. J.; Young, V. G. Organometallics 1996,15, 2720.  (10)  Fryzuk, M. D.; Love, J. B., unpublished results.  (11)  Chow, P. C , M. Sc. Thesis, University of British Columbia, 1993.  (12)  Shilov, A. E.; Denisov, N. T.; Efimov, O. N.; Shubalov, N. F.; Shuvalova, N. I.; Shilova, E. Nature (London) 1971,231, 460.  (13)  Hales, B. F.; Case, E. E.; Morningstar, J. E.; Dzeda, M. F.; Mauterer, L. A. Biochemistry 1986,25,7251.  270  References begin on page 270  Chapter 6: Future Work  (14)  Kentemich, T.; Dannenberg, G.; Hundeshagen, B.; Bothe, H. FEMS Microbiol. Lett.  1988,57, 19. (15)  Edema, J. J. H.; Meetsma, A.; Gambarotta, S. / . Am. Chem. Soc. 1989, 111, 6878.  (16)  Bemo, P.; Hao, S.; Minhas, R.; Gambarotta, S. / . Am. Chem. Soc. 1994,116, 7417.  (17)  Buijink, J. K. F.; Meetsma, A.; Teuben, J. H. Organometallics 1993,12, 2004.  (18)  Song, J. I.; Berno, P.; Gambarotta, S. / . Am. Chem. Soc. 1994,116, 6927.  (19)  Fryzuk, M. D.; Love, J. B.; Rettig, S. J.; Young, V. G. Science 1997,275, 1445.  (20)  Cotton, F. A.; Wilkinson, G. Advanced Inorganic Chemistry; 4th ed.; John Wiley & Sons: New York, 1980.  (21)  Ebsworth, E. A. V.; Rankin, D. W. H.; Cradock, S. Structural Methods in Inorganic Chemistry; Blackwell Scientific Publications: Oxford, 1987.  (22)  Allen, F. Ff.; Kennard, O.; Watson, D. G.; Brammer, L.; Orpen, A. G.; Taylor, R. / . Chem. Soc, Perkin Trans. 1987, //, SI.  (23)  Nugent, W. A.; Mayer, J. M. Metal-Ligand Multiple Bonds; John Wiley & Sons: New York, 1988.  (24)  O'Regan, M. B.; Liu, A. FL; Finch, W. C ; Schrock, R. R.; Davis, W. M. / . Am. Chem. Soc. 1990,772,4331.  (25)  Schrock, R. R.; Kolodziej, R. M.; Liu, A. W.; Davis, W. M.; Vale, M. G. / . Am. Chem. Soc. 1990, 772,4338.  (26)  Burger, Ff.; Sawadny, W.; Wannagat, U. / . Organomet. Chem. 1965,3,113.  (27)  Cummins, C. C ; Shrock, R. R.; Davis, W. M. Organometallics 1991,10, 3781. 271  References begin on page 2 70  Chapter 6: Future Work  (28)  Veith, M.; Woll, K. L. Chem. Ber. 1993,126, 2383.  (29)  Cameron, A. F.; Forrest, K. P.; Ferguson, G. / . Chem. Soc., A 1971, 1286.  (30)  Kessler, J. M.; Reeder, J. H.; Vac, R.; Yeung, C ; Nelson, J. H.; Frye, J. S.; Alcock, N. W. Magn. Reson. Chem. 1991,29, S94.  (31)  Stephan, D. W. Coord. Chem. Rev. 1989, 95, 41.  (32)  Bradley, D. C ; Kasenally, A. S. / . Chem. Soc, Chem. Comm. 1968, 1430.  (33)  Evans, D. F. / . Chem. Soc. 1959, 2003.  (34)  Sur, S. K. / . Magn. Res. 1989, 82,169.  (35)  Manzer, L. E. Inorg. Synth. 1982,21, 155.  272  References begin on page 270  Appendix 1: X-Ray Crystal Structure Data  Table A l . l Crystallographic Data for Chapter T w o ' . a  Compound  1 ScCl (THF)[PNP]  Empirical Formula  3 ScEt [PNP]  b  4 Sc(Ns) [PNP]  6 ScCpCiPNPl  C22H52Cl2NOP2ScSi2C22H54NPScSi2  C26H NP2ScSi4  C 3H49ClNP ScSi  fw  580.64  495.75  612.06  538.17  Color, habit  colorless, plate  colorless, prism  colorless, prism  colorless, plate  Crystal system  orthorhombic  monoclinic  monoclinic  monoclinic  Space group  Pca2i  C2/c  P2 /c  P2x/c  a, A  21.482(4)  19.316(1)  12.862(2)  17.409(2)  b, A  10.597(5)  10.3386(9)  12.515(3)  10.0803(7)  c, A  14.668(3)  15.832(1)  24.582(3)  17.688(2)  a, deg  90  90  90  90  P.deg  90  104.853(2)  91.15(1)  94.606(8)  Y.deg  90  90  90  90  V,A?  3339(1)  3056.1(4)  3956(1)  3094.0(4)  Z  4  4  4  4  1.155  1.077  1.028  1.155  F(000)  1248  1088  1344  1160  Radiation  Mo  Cu  Mo  Cu  |i, cm"  5.61  39.24  4.01  46.10  Crystal size, mm  0.05 x 0.35 x 0.40  0.10x0.20x0.35  0.25 x 0.30 x 0.35  0.05x 0.25 x 0.50  Transmission factors  0.90-1.00  0.598-1.00  0.968-1.000  0.440-1.000  Scan type  co-26  co-20  co-29  co-29  Pcalc. g /  c m 3  1  2  2  2  2  66  l  273  2  2  2  Appendix 1: X-Ray Crystal Structure Data  Scan range, co°  0.89+0.35 tan 0  0.73+0.20 tan 0  0.82+0.35 tan 0  0.94+0.20 tar  16.0 (8 rescan)  16.0 (up to 9 scans)  16 (up to 9 scans)  16 (up to 9 sc  2e ,deg  45  155  50  155  Crystal decay, %  12.85  -  1.75  -  Total reflections  2514  3387  7679  6949  Unique reflections  -  3287  7320  6758  -  0.023  0.043  0.038  Number w/7 > 3a(7) 1255  2314  2643  2414  Variables  279  129  317  272  R  0.035  0.038  0.043  0.047  R  0.034  0.046  0.036  0.042  gof  1.39  2.72  1.85  1.98  Max A/a (final cycle) 0.003  0.0005  0.02  0.002  Residual density e/A -0.19 to 0.18  -0.20 to 0.20  -0.21 to 0.19  -0.21 to 0.25  Compound  8 ScCpPh[PNP]  13 ScCl2[P2Cp]  Empirical Formula  C 9H54NP ScSi2  C 3H47Cl2P ScSi2  fw  579.83  557.60  Color, habit  colorless, irregular  colorless, plate  Crystal system  triclinic  monoclinic  Space group  PI  P2ifc  a, A  10.5129(15)  14.478(1)  b,k  19.124(2)  13.226(1)  c, A  9.114(2)  16.920(1)  a, deg  95.582(13)  90  Scan speed, °min'  1  max  merge  w  3  2  2  2  2  274  Appendix 1: X-Ray Crystal Structure Data  P.deg  107.047(12)  109.509(4)  Y.deg  90.042(11)  90  V,A?  1742.8(5)  3054.0(4)  Z  2  4  1.105  1.213  F(OOO)  628  1192  Radiation  Mo  Cu  u,, cm  3.85  55.99  Crystal size, mm  0.25 x 0.35 x 0.40  0.06 x 0.25 x 0.38  Transmission factors  0.910-1.000  0.47-1.00  Scan type  (0-28  co-20  Scan range, co°  1.25+0.35 tan 0  0.89+0.20 tan 0  Scan speed, "min"  16 (up to 9 scans)  8 (up to 9 scans)  2e  50  155.3  Crystal decay, %  13.42  8.69  Total reflections  6485  6779  Unique reflections  6107  6523  0.045  0.026  Number W//>3CT(/)  2844  3817  Variables  316  272  R  0.054  0.036  R  0.049  0.036  gof  2.77  2.03  Pcalc g /  c m 3  -1  1  max  ,deg  n merge  w  Max A/a (final cycle) 0.0005  0.0004  Residual density e/A -0.27 to 0.33  -0.22 to 0.23  3  275  Appendix  a  1: X-Ray  Crystal  Structure  Data  Complex 1,4,8: Temperature 294 K, Rigaku AFC62 diffractometer, MoK^X = 0.71069 A) radiation; by Dr. S. J.  Rettig at UBC. Complex 3, 6,13: Temperature 294 K, Rigaku AFCS6S diffractometer, CaKotfr = 1.54178 A) radiation; by Dr. S. J. Rettig at UBC. b  For Rigaku structures, function minimized Ew(lF l-IF l) where w = 4F /a (F ), 2  0  (Lw(\F \AF \) /Lw\F \ ) , 2  0  c  2 l/2  0  2  c  0  2  2  0  R = 2IF I-IF !I/£IF I, R = 0  C  and gof = [Iw(IF l-lF l) /(m-n)] (where m = number of observations, 2  0  1/2  c  0  n=  number of  variables). Values given for R, R , and gof are based on those reflections with / > 3a(7). w  Table A1.2 Fractional Atomic Coordinates and B  c e q  (A ) for 2  ScCl2(THF)[N(SiMe CH2PPr 2)2] (1). 2  i  atom  X  y  z  Bea  Sc(l)  0.05665  0.25970  0.35990  4.35  Cl(l)  0.0310  0.3971  0.4887  6.42  Cl(2)  0.0613  0.1385  0.2204  6.05  P(l)  -0.0092  0.0782  0.4531  5.05  P(2)  0.1489  0.4139  0.2918  5.57  Si(l)  0.1248  0.1031  0.5269  5.01  Si(2)  0.2016  0.1596  0.3616  5.58  0(1)  -0.0199  0.3677  0.2869  6.7  N(l)  0.1337  0.1707  0.4202  4.3  C(l)  0.0398  0.0833  0.5543  5.4  C(2)  0.2049  0.2890  0.2728  6.1  C(3)  0.1621  -0.0549  0.5396  7.3  C(4)  0.1590  0.2070  0.6191  7.9  C(5)  0.2723  0.1830  0.4376  8.1  C(6)  0.2120  0.0069  0.3003  7.5  C(7)  -0.0914  0.0880  0.4941  7.3  276  |  w  Appendix 1: X-Ray Crystal Structure Data  c  B  C(8)  0.0016  -0.0835  0.4073  6.1  C(9)  0.1450  0.5080  0.1860  8.8  C(10)  0.1844  0.5140  0.3890  12.9  C(ll)  -0.0999  0.1550  0.5815  10.6  C(12)  -0.1315  0.1500  0.4190  12.1  C(13)  -0.0388  -0.1040  0.3230  9.0  C(14)  -0.0086  -0.1899  0.4750  8.2  C(15)  0.1140  0.4330  0.1110  12.1  C(16)  0.2067  0.5570  0.1470  13.2  C(17)  0.1456  0.6330  0.3950  16.4  C(18)  0.2510  0.5330  0.3800  19.3  C(19)  -0.0579  0.3220  0.2145  8.3  C(20)  -0.1100  0.4100  0.2030  11.7  C(21)  -0.0902  0.5220  0.2560  15.9  C(22)  -0.0362  0.4950  0.3020  12.5  = 8/3Jc (C/n(aa*) + £/22(bb*) + C/33(cc*) + 2£/i2aa*bb*cosy + 2tfi3aa*cc*cosP + 2E/23bb*cc*cosa) 2  e q  T a b l e A1.3  2  2  2  Fractional Atomic Coordinates and B c (A ) for ScEt2[N(SiMe CH2PPr )2] (3). 2  i  e q  2  2  atom  X  y  z  Sc(l)  0.50000  0.53358  0.2500  3.93  P(l)  0.62707  0.56412  0.37891  4.38  P(2)  0.37293  0.56412  0.12109  4.38  Si(l)  0.53116  0.81549  0.34792  5.10  Si(2)  0.46884  0.81549  0.15208  5.10  277  Appendix 1: X-Ray Crystal Structure Data  0.5000  0.7370  0.2500  0.5974  0.7080  0.4262  0.4586  0.8473  0.4038  0.5768  0.9743  0.3419  0.7123  0.6139  0.3562  0.6496  0.4506  0.4726  0.6997  0.7153  0.2856  0.7518  0.5004  0.3288  0.7140  0.4905  0.5459  0.6534  0.3118  0.4436  0.4535  0.4331  0.3484  0.4312  0.2970  0.3275  0.5465  0.4331  0.1516  0.5414  0.8473  0.0962  0.4232  0.9743  0.1581  0.2877  0.6139  0.1438  0.3504  0.4506  0.0274  0.5688  0.2970  0.1725  0.4026  0.7080  0.0738  0.3003  0.7153  0.2144  0.2482  0.5004  0.1712  0.2860  0.4905  -0.0459  0.3466  0.3118  0.0564  278  Appendix 1: X-Ray Crystal Structure Data  Table A1.4 Fractional Atomic Coordinates and B  c e q  (A ) for Sc(CH2SiMe3)22  [N(SiMe2CH PPr 2) ] (4). i  2  atom  2  X  y  z  Sc(l)  0.24931  0.14096  0.36055  3.72  P(l)  0.0410  0.1813  0.3832  5.24  P(2)  0.4601  0.1911  0.3553  5.09  Si(l)  0.1868  0.2866  0.4666  5.637  Si(2)  0.3319  0.3883  0.3874  5.26  Si(3)  0.2547  -0.1486  0.3919  5.10  Si(4)  0.2374  0.1582  0.2020  5.23  N(l)  0.2559  0.2835  0.4076  3.9  C(l)  0.0665  0.2026  0.4558  6.2  C(2)  0.4353  0.3312  0.3419  5.4  C(3)  0.2612  0.2280  0.5252  8.2  C(4)  0.1411  0.4226  0.4884  9.2  C(5)  0.3983  0.4643  0.4435  8.7  C(6)  0.2593  0.4887  0.3453  7.8  C(7)  -0.0730  0.0914  0.3761  7.8  C(8)  -0.0035  0.3149  0.3609  7.2  C(9)  -0.0731  0.0033  0.4137  14.8  C(10)  -0.0869  0.0530  0.3189  13.5  C(ll)  -0.1040  0.3547  0.3868  13.6  C(12)  -0.0063  0.3268  0.3013  12.3  C(13)  0.5501  0.1479  0.3027  12.2  C(14)  0.5308  0.1861  0.4219  9.7  279  Appendix 1: X-Ray Crystal Structure Data  C(15)  0.6330  0.2310  0.2852  11.3  C(16)  0.5437  0.0385  0.2882  15.1  C(17)  0.6159  0.2659  0.4276  14.9  C(18)  0.5600  0.0795  0.4398  16.3  C(19)  0.2511  -0.0066  0.4110  4.8  C(20)  0.3904  -0.1957  0.3860  9.0  C(21)  0.1856  -0.1768  0.3261  8.2  C(22)  0.1886  -0.2348  0.4422  9.2  C(23)  0.2355  0.1094  0.2725  6.0  C(24)  0.1040  0.1649  0.1716  7.8  C(25)  0.3176  0.0706  0.1581  8.4  C(26)  0.2938  0.2951  0.1958  8.2  Table A1.5 Fractional Atomic Coordinates and B c (A ) for Sc(r|5-C H5)C12  5  eq  [N(SMe2CH2PPr 2) ] (6). i  2  atom  X  y  z  Sc(l)  0.24886  0.18110  0.21131  4.26  Cl(l)  0.15861  0.11290  0.29803  7.45  P(l)  0.11514  0.22380  0.11506  5.03  P(2)  0.35370  -0.00500  0.28002  4.47  Si(l)  0.1941  -0.0342  0.0744  6.56  Si(2)  0.3628  -0.0004  0.1104  5.96  N(l)  0.2695  0.0423  0.1256  4.3  C(l)  0.1013  0.0480  0.0969  6.0  280  Appendix 1:  0.4198  -0.0019  0.2055  0.2033  -0.0246  -0.0318  0.1791  -0.2116  0.0991  0.4088  0.1192  0.0471  0.3766  -0.1685  0.0673  0.0242  0.2920  0.1460  0.1198  0.2941  0.0191  0.3201  -0.1790  0.2759  0.4105  0.0062  0.3718  0.0344  0.4060  0.1972  -0.0295  0.2010  0.1709  0.0558  0.2537  -0.040  0.1321  0.4383  0.0160  0.2569  -0.2065  0.3267  0.3832  -0.2841  0.2882  0.3594  0.0391  0.4353  0.4815  0.0690  0.3767  0.2511  0.4291  0.2011  0.3137  0.3856  0.1676  0.3643  0.3277  0.2246  0.3291  0.3401  0.2900  0.2600  0.4018  0.2770  281  X-Ray Crystal Structure  Data  Appendix 1: X-Ray Crystal Structure  Table A1.6 Fractional Atomic Coordinates and B  c e q  (A ) for Sc(r|5-C5H5)Ph2  [N(SiMe CH PPri2)2] (8). 2  2  atom  X  y  z  Sc(l)  0.33795  0.26039  0.42077  4.39  P(l)  0.35910  0.37187  0.65430  5.22  P(2)  0.23580  0.12485  0.28140  5.05  Si(l)  0.1031  0.2823  0.6107  5.05  Si(2)  0.0107  0.2252  0.2816  5.66  N(l)  0.1370  0.2579  0.4407  4.30  C(l)  0.2487  0.3353  0.7485  5.5  C(2)  0.0801  0.1578  0.1620  5.4  C(3)  0.0788  0.2066  0.7170  6.6  C(4)  -0.0519  0.3357  0.5895  7.7  C(5)  -0.0654  0.2933  0.1539  8.6  C(6)  -0.1306  0.1790  0.3253  8.3  C(7)  0.2928  0.4606  0.6000  7.9  C(8)  0.5136  0.3999  0.8087  8.0  C(9)  0.1746  0.4531  0.4584  10.2  C(10)  0.2628  0.5069  0.7264  14.8  C(ll)  0.6218  0.4239  0.7432  11.1  C(12)  0.5613  0.3572  0.9280  10.6  C(13)  0.1795  0.0668  0.4033  6.4  C(14)  0.3022  0.0626  0.1576  8.6  C(15)  0.2959  0.0294  0.5053  8.4  C(16)  0.0678  0.0129  0.3134  8.7  282  Data  Appendix 1: X-Ray Crystal Structure  C(17)  0.4509  0.0585  0.2129  8.0  C(18)  0.2446  0.0555  -0.0080  11.6  C(19)  0.5196  0.3177  0.3346  8.0  C(20)  0.4838  0.2567  0.2423  7.3  C(21)  0.3556  0.2610  0.1517  6.9  C(22)  0.3105  0.3272  0.1893  7.7  C(23)  0.4879  0.2099  0.6037  8.8  C(24)  0.4587  0.1834  0.7233  4.6  C(25)  0.5404  0.1527  0.8435  6.7  C(26)  0.6734  0.1446  0.8553  6.6  C(27)  0.7145  0.1680  0.7375  8.0  C(28)  0.6273  0.1991  0.6203  8.1  C(29)  0.4158  0.3602  0.3043  6.8  Table A1.7 Fractional Atomic Coordinates and B  c e q  (A ) for ScCl2[(ri -C5H3-l,32  5  (SiMe2CH PPri2)2] (13). 2  atom  X  y  z  Sc(l)  0.22856  0.45753  0.48156  3.02  Cl(l)  0.08243  0.55517  0.42824  4.89  Cl(2)  0.37368  0.45236  0.44189  4.17  P(l)  0.15644  0.34198  0.33406  3.58  P(2)  0.32001  0.64054  0.55452  3.52  Si(l)  0.20124  0.17114  0.46588  4.02  Si(2)  0.40388  0.48466  0.70301  3.96  283  Data  Appendix 1: X-Ray Crystal Structure Data  0.2216  0.2832  0.5363  0.3313  0.3254  0.5829  0.3019  0.4083  0.6309  0.1985  0.4167  0.6140  0.1510  0.3422  0.5577  0.2111  0.2177  0.3645  0.3004  0.0757  0.5093  0.0813  0.1123  0.4562  0.0225  0.3168  0.2928  0.1859  0.3782  0.2390  -0.0350  0.4072  0.2459  -0.0058  0.2224  0.2385  0.1837  0.4929  0.2282  0.2841  0.3365  0.2381  0.4203  0.6052  0.6500  0.5195  0.4117  0.7267  0.3790  0.5080  0.8026  0.2572  0.7428  0.5914  0.3869  0.7044  0.4920  0.1819  0.8005  0.5201  0.2078  0.7018  0.6505  0.3221  0.7256  0.4020  0.4466  0.7968  0.5342  284  Appendix 1: X-Ray Crystal Structure  Table A1.8 Crystallographic Data for Chapter Three ' . b  d  Compound  1A1C1 [RNP]  Empirical Formula  C 8H44AlCl2NP Si2 C 2H5 AlNP Si  fw  490.56  601.92  623.90  Color, habit  colorless, prism  colorless, needle  colorless, prism  Crystal system  monoclinic  triclinic  monoclinic  Space group  Flila  PI  P21/C  a, A  10.015(3)  11.742(3)  12.227(2)  b, A  16.132(5)  18.649(4)  17.388(2)  c, A  17.951(7)  9.047(3)  15.935(2)  a, deg  90  98.38(2)  90  P.deg  101.41(3)  98.06(2)  102.987(8)  Y.deg  90  73.92(2)  90  V,A?  2842(1)  1870.0(9)  3301.0(6)  Z  4  2  4  1.146  1.069  1.255  F(000)  1056  656  1312  Radiation  Mo  Mo  Cu  u, cm"  4.61  2.18  62.01  Crystal size, mm  0.45 x 0.45 x 0.40  0.12x0.20x0.40  0.20 x 0.40 x 0.45  Transmission factors  -  0.939-1.000  0.362-1.000  Scan type  co-29  co-29  co-29  Scan range, co°  1.52+0.35 tan 9  1.15+0.35 tan 9  0.89+0.20 tan 9  Scan speed, "min"  8 (up to 9 scans)  16 (up to 9 scans)  16 (up to 8 scans)  2e  55  50  155  Pcalc.  g/  c m 3  1  1  max  ,deg  4 Al(CH Ph) [PNP] 7 A1C1 [PNP](A1C13)  2  1  2  2  3  8  285  2  2  2  2  Ci8H44Al2Cl NP Si 5  2  2  Appendix 1: X-Ray Crystal Structure  Crystal decay, %  -  6.3  7.36  Total reflections  6048  6933  7300  Unique reflections  5781  6574  6992  1.667  0.045  0.073  Number w/ 1 > 3a(7)  1405  2569  3611  Variables  245  343  272  R  0.081  0.056  0.046  R  0.076  0.050  0.046  gof  3.88  2.43  2.00  Max A/a (final cycle) 0.07  0.002  0.0003  Residual density e/A -0.47 to 0.47  -0.28 to 0.28  -0.28 to 0.30  merge  w  3  a  Complex 1: Temperature 294 K, Rigaku AFC62 diffractometer, MoK (k= a  Olovsson at UBC.  Data  0.71069 A) radiation; by Dr. G.  Complex 4: Temperature 294 K, Rigaku AFC62 diffractometer, MoK^X = 0.71069 A)  radiation; by Dr. S. J. Rettig at UBC.  Complex 7: Temperature 294 K, Rigaku APCS6S diffractometer, CuKa(k =  1.54178 A) radiation; by Dr. S. J. Rettig at UBC.  Table A1.9 Fractional Atomic Coordinates and B  c e q  (A ) for AlCl2[N(SiMe2CH2PPr 2)2] (1). 2  L  atom  X  y  z  Bea  Al(l)  0.07390  0.2107  0.3654  4.8  Cl(l)  -0.09170  0.1204  0.3465  8.2  Cl(2)  0.09880  0.2895  0.2715  7.8  P(l)  -0.0669  0.3223  0.4113  5.7  P(2)  0.2338  0.0995  0.3394  5.4  Si(D  0.2216  0.3213  0.4991  6.5  286  Appendix 1: X-Ray Crystal Structure Data  0.2682  0.1373  0.5064  0.1890  0.2234  0.4620  -0.301  0.332  0.299  -0.200  0.387  0.347  -0.147  0.455  0.312  -0.160  0.293  0.485  -0.129  0.220  0.526  -0.203  0.364  0.531  0.072  0.389  0.458  0.374  0.367  0.471  0.246  0.327  0.601  0.177  0.100  0.583  0.451  0.157  0.553  0.262  0.055  0.434  0.407  0.141  0.329  0.405  0.170  0.248  0.522  0.078  0.356  0.193  0.016  0.267  0.101  0.051  0.198  0.149  -0.063  0.293  287  Appendix  1: X-Ray Crystal Structure  Table ALIO Fractional Atomic Coordinates and B q (A ) for c  2  e  Al(CH Ph)2[N(SiMe2CH PPr 2)2] (4). i  2  2  atom  X  y  z  Al(l)  0.2948  0.1772  0.7999  3.32  P(l)  0.4190  0.2320  0.6767  3.65  P(2)  -0.1957  0.3768  0.7749  4.75  Si(l)  0.1904  0.2098  0.4810  4.03  Si(2)  0.0189  0.2318  0.7144  3.71  N(l)  0.1636  0.2084  0.6622  3.1  C(l)  0.3456  0.2203  0.4851  4.6  C(2)  -0.0344  0.3366  0.7599  4.4  C(3)  0.1883  0.1206  0.3586  6.4  C(4)  0.0883  0.2879  0.3808  6.3  C(5)  0.0129  0.1861  0.8819  4.9  C(6)  -0.0883  0.2006  0.5614  5.8  C(7)  0.5803  0.1909  0.6667  4.6  C(8)  0.4015  0.3335  0.7304  4.0  C(9)  0.6101  0.1078  0.6150  7.1  C(10)  0.6454  0.2054  0.8220  6.8  C(ll)  0.2705  0.3745  0.7351  5.7  C(12)  0.4594  0.3691  0.6259  6.6  C(13)  -0.2067  0.4182  0.9722  6.7  C(14)  -0.2285  0.4624  0.6800  9.7  C(15)  -0.1565  0.3624  1.0810  12.7  C(16)  -0.3368  0.4556  0.9950  9.0  288  Data  Appendix 1: X-Ray Crystal Structure Data  C(17)  -0.1480  0.5127  0.7220  16.2  C(18)  -0.2400  0.4357  0.5060  15.6  C(19)  0.3665  0.0666  0.7768  4.2  C(20)  0.2803  0.0263  0.8112  4.0  C(21)  0.2770  0.0093  0.9562  5.1  C(22)  0.1935  -0.0245  0.9876  6.2  C(23)  0.1112  -0.0423  0.8740  6.5  C(24)  0.1124  -0.0257  0.7310  6.1  C(25)  0.1962  0.0076  0.6992  4.8  C(26)  0.2842  0.2227  1.0118  4.2  C(27)  0.3976  0.2361  1.0969  4.0  C(28)  0.4118  0.3083  1.1308  5.5  C(29)  0.5164  0.3212  1.2060  6.3  C(30)  0.6090  0.2609  1.2496  6.6  C(31)  0.5955  0.1895  1.2181  5.7  C(32)  0.4898  0.1779  1.1431  4.7  Table A l . l l Fractional Atomic Coordinates and B  c e q  (A ) for 2  AlCl [N(SiMe CH2PPr 2)2]»AlCl3 (7). i  2  atom  2  X  y  z  Al(l)  0.36530  0.32215  0.47474  3.29  Al(2)  -0.11320  0.17532  0.35900  4.57  Cl(l)  0.40310  0.35809  0.60599  5.60  Cl(2)  0.21610  0.37799  0.40880  5.99  289  Appendix 1:  -0.26280  0.15590  0.26500  -0.07880  0.29352  0.37770  -0.12040  0.11896  0.47510  0.51830  0.35355  0.40900  0.03487  0.11781  0.30270  0.50060  0.18087  0.45236  0.25370  0.16335  0.45733  0.3715  0.2184  0.4575  0.5872  0.2610  0.4190  0.1718  0.1607  0.3404  0.4915  0.1027  0.3714  0.5800  0.1464  0.5589  0.1670  0.2076  0.5260  0.2905  0.0650  0.5000  0.4734  0.3739  0.2947  0.3949  0.3125  0.2474  0.4210  0.4531  0.2774  0.6206  0.4277  0.4554  0.6746  0.4099  0.5480  0.7097  0.4417  0.4036  0.0448  0.0153  0.3303  -0.0675  -0.0256  0.2982  0.1402  -0.0292  0.3048  -0.0050  0.1305  0.1849  -0.0239  0.2154  0.1625  0.0724  0.0942  290  0.1345  X-Ray Crystal Structure  Data  Appendix 1: X-Ray Crystal Structure Data  Table A1.12 Crystallographic Data for Chapter F o u r - . b  e  Compound  1 {Li[PNP]} LiCl  Empirical Formula  C H ClLi3N2P4Si4 C44Hi Al2Li4N2P4 C26H64BLiNNaP Si2  2  36  2 {LJ2AlMe[PNP]}2  88  6 {LiNaBEt4[PNP]}  x  12  Si  2  4  fw  841.62  987.35  549.66  Color, habit  colorless, irregular  colorless, irregular  colorless, irregular  Crystal system  monoclinic  triclinic  monoclinic  Space group  P2i/n  PI  «i/c  a, A  11.878(2)  16.215(1)  12.519(2)  b,k  34.137(3)  19.672(1)  35.067(6)  c, A  13.683(2)  11.5163(9)  17.871(2)  a, deg  90  91.106(7)  90  P.deg  98.98(1)  108.477(6)  109.37(1)  Y.deg  90  100.085(6)  90  V , A 3  5480(1)  3419.3(5)  7400(1)  Z  4  2  8  1.020  0.959  0.987  F(000)  1840  1088  2432  Radiation  Cu  Cu  Cu  H, cm"  27.23  21.21  18.85  Crystal size, mm  0.30 x 0.30 x 0.40  0.25 x 0.30 x 0.30  0.20 x 0.20 x 0.50  Transmission factors  0.730-1.000  0.908-1.000  0.875-1.000  Scan type  co-20  (0-26  co-20  Scan range, co°  1.00+0.20 tan 0  0.89+0.20 tan 9  0.89+0.20 tan 0  Scan speed, min"!  16 (up to 8 scans)  8 (up to 9 scans)  16 (up to 9 scans)  ^max'deg  120  155  120  Pcalc g /  c m 3  1  0  291  Appendix 1: X-Ray Crystal Structure  Crystal decay, %  6.46  9.26  Data  ! 5.2 i  Total reflections  8794  14474  Unique reflections  8346  13942  11245  0.049  0.030  0.052  Number w/7 > 3a(7)  3127  6549  3568  Variables  464  551  614  R  0.045  0.056  0.050  R  0.043  0.050  0.041  gof  1.93  3.64  1.93  Max A/a (final cycle) 0.0003  0.001  0.001  Residual density e/A -0.16 to 0.18  -0.32to 0.40  -0.16 to 0.33  n  merge  w  3  e  ;ii82i  Complex 1, 2, 6: Temperature 294 K, Rigaku AFCS6S diffractometer, CuK(^X= 1.54178 A) radiation; by Dr. S.  J. Rettig at UBC.  Table A1.13 Fractional Atomic Coordinates and B  c e q  (A ) for 2  {LiN(SiMe CH PPri2)2}2LiCl (1). 2  2  atom  X  y  z  Bea  Li(l)  0.2945  0.1426  0.3846  6.3  Li(2)  0.4274  0.0746  0.6560  6.1  Li(3)  0.2725  0.1344  0.5833  6.4  Cl(l)  0.3962  0.09788  0.4938  6.67  P(l)  0.4171  0.19788  0.3254  7.14  P(2)  0.1105  0.1203  0.2647  7.12  P(3)  0.3541  0.00408  0.6844  6.98  292  Appendix 1: X-Ray Crystal Structure Data  0.5973  0.10195  0.7778  0.2947  0.21718  0.5049  0.0724  0.17834  0.4360  0.1876  0.07199  0.7134  0.3562  0.12683  0.8229  0.2138  0.1764  0.4770  0.2992  0.1036  0.7175  0.3522  0.2361  0.3930  0.0274  0.1326  0.3612  0.4192  0.2058  0.6020  0.2234  0.2599  0.5548  0.0207  0.2219  0.3578  -0.0148  0.1788  0.5392  0.5693  0.1979  0.3869  0.6347  0.2356  0.3761  0.6330  0.1617  0.3616  0.4201  0.2198  0.2024  0.3055  0.2341  0.1540  0.4638  0.1892  0.1357  0.0480  0.1521  0.1613  -0.0780  0.1493  0.1308  0.1100  0.1483  0.0722  0.0656  0.0702  0.2277  0.1059  0.0421  0.3098  0.1040  0.0556  0.1352  0.2372  0.0205  0.7475  0.5106  0.1403  0.8161  293  Appendix 1: X-Ray Crystal Structure  C(21  0.1049  0.0706  0.5859  9.0  C(22;  0.0778  0.0834  0.7945  10.9  C(23  0.3585  0.0971  0.9396  9.0  C(24;  0.2830  0.1740  0.8433  9.4  C(25  0.4195  -0.0357  0.7664  9.8  C(26;  0.4598  -0.0200  0.8722  12.1  C(27  0.5194  -0.0513  0.7321  14.7  C(28  0.2797  -0.0200  0.5724  11.4  C(29  0.3381  -0.0198  0.4883  11.6  C(30;  0.2010  -0.0564  0.5861  14.0  C(31  0.6517  0.0744  0.8915  10.5  C(32  0.7397  0.0443  0.8710  17.9  C(33  0.6940  0.0985  0.9801  18.6  C(34;  0.7272  0.1268  0.7442  9.9  C(35  0.7011  0.1394  0.6391  14.1  C(36  0.7807  0.1586  0.8051  14.1  Table A1.14 Fractional Atomic Coordinates and B q (A ) for c  2  e  {LiN(SiMe2CH PPr 2)2»LiAlMe4} (2). i  2  2  atom  X  y  z  Li(l)  0.27432  0.36737  0.02528  6.3  Li(2)  0.39352  0.52127  0.12918  7.8  P(l)  0.14474  0.40200  -0.13722  6.66  P(2)  0.25794  0.26824  0.17678  6.78  294  Data  Appendix 1: X-Ray Crystal Structure Data  0.20787  0.48882  0.11928  0.34223  0.41950  0.29178  0.50764  0.35723  -0.08322  0.29772  0.45047  0.15268  0.15332  0.48407  -0.05252  0.27822  0.33197  0.30478  0.23852  0.58417  0.16558  0.12232  0.45317  0.19238  0.45922  0.41137  0.31298  0.34592  0.47527  0.42668  0.13792  0.42317  -0.29452  0.03312  0.35417  -0.15992  0.35652  0.22657  0.22368  0.16632  0.20007  0.18938  0.21952  0.43757  -0.31592  0.07472  0.47177  -0.35062  0.01812  0.28467  -0.23012  -0.01208  0.35587  -0.09232  0.36262  0.18077  0.12158  0.37032  0.19017  0.34348  0.13582  0.14677  0.08348  0.08842  0.22847  0.19818  0.39432  0.35407  -0.04912  0.49292  0.39077  -0.25132  0.59892  0.42657  0.04648  0.54442  0.26697  -0.07232  295  Appendix 1: X-Ray Crystal Structure  Table A1.15 Fractional Atomic Coordinates and B q (A ) for c  2  e  {LiN(SiMe CH2PPri2) 'NaBEt4} (6). 2  2  x  atom  X  y  z  Li(l)  0.2064  0.1437  0.4545  4.3  Li(2)  -0.4309  0.1284  0.8850  5.1  B(l)  -0.1216  0.0964  0.6631  5.4  B(2)  0.3831  0.0874  0.6607  5.1  P(l)  0.1425  0.2106  0.4798  6.93  P(2)  0.2584  0.1102  0.3412  5.85  P(3)  -0.3424  0.1955  0.8889  7.33  P(4)  -0.4917  0.0953  0.9945  6.66  Si(D  -0.0516  0.1559  0.3843  5.48  Si(2)  0.0431  0.0783  0.3715  5.17  Si(3)  -0.1716  0.1284  0.9576  5.78  Si(4)  -0.2991  0.0550  0.9485  6.12  Na(l)  0.0551  0.0971  0.5479  6.76  Na(2)  -0.3022  0.0858  0.7770  6.96  N(l)  0.04630  0.1208  0.4181  4.3  N(2)  -0.2828  0.0991  0.9154  4.6  C(l)  -0.0058  0.1976  0.4540  6.4  C(2)  0.1863  0.0681  0.3609  7.5  C(3)  -0.0726  0.1737  0.2812  7.6  C(4)  -0.1975  0.1441  0.3823  7.5  C(5)  0.0260  0.0368  0.4333  8.5  C(6)  -0.0732  0.0708  0.2749  8.1  C(7)  0.1413  0.2428  0.3975  13.7  296  Data  Appendix 1:  0.1710  0.2419  0.5678  0.1831  0.1215  0.2380  0.4014  0.0923  0.3466  0.0750  0.2782  0.3900  0.2500  0.2480  0.3870  0.1350  0.2262  0.6288  0.2930  0.2534  0.5987  0.2206  0.1574  0.2088  0.1736  0.0889  0.1781  0.4830  0.1225  0.3642  0.4180  0.0561  0.3253  -0.2024  0.1756  0.9038  -0.4420  0.0514  0.9640  -0.1374  0.1379  1.0668  -0.0324  0.1123  0.9502  -0.1869  0.0386  1.0408  -0.3042  0.0172  0.8739  -0.3487  0.2373  0.8224  -0.3256  0.2173  0.9881  -0.4065  0.0990  1.1011  -0.6356  0.0850  0.9963  -0.3380  0.2243  0.7474  -0.4580  0.2585  0.8072  -0.2352  0.2456  1.0191  -0.4385  0.2298  0.9942  -0.4193  0.0654  1.1529  -0.4192  0.1365  1.1388  297  X-Ray Crystal Structure  Data  Appendix 1:  -0.7095  0.0666  0.9215  -0.6940  0.1219  1.0049  -0.1439  0.0737  0.5786  -0.2504  0.0497  0.5463  -0.0052  0.1213  0.6804  0.0300  0.1454  0.7546  -0.1137  0.0655  0.7332  -0.0165  0.0376  0.7552  -0.2299  0.1249  0.6567  -0.2544  0.1571  0.5959  0.2736  0.0612  0.6607  0.2997  0.0266  0.7167  0.3377  0.1183  0.5870  0.4254  0.1460  0.5769  0.4831  0.0609  0.6501  0.4536  0.0354  0.5775  0.4345  0.1109  0.7469  0.3555  0.1395  0.7653  298  X-Ray Crystal Structure  Data  Appendix 1: X-Ray Crystal Structure Data  Table A1.16 Crystallographic Data for Chapter Fivers.  Compound  2 antt-GaCl[P2N ]  Empirical Formula  C 4H4 ClGaN2P Si4 C 4H4 AlClN2P2Si4 C24H42ClGaN P2Si4 C 4H43AlN P Si4  fw  638.07  595.33  638.07  560.88  Color, habit  colorless, plate  colorless, irregular  colorless, irregular  colorless, plate  Crystal system  orthorhombic  orthorhombic  orthorhombic  monoclinic  Space group  Pbca  /'2 2 2  Fl\1\2\  C2/c  a, A  17.237(2)  17.553(2)  8.9331(8)  37.865(2)  b,k  24.650(2)  21.155(1)  17.5651(6)  9.1760(4)  15.887(2)  18.924(1)  21.2421(7)  22.687(1)  a, deg  90  90  90  90  P, deg  90  90  90  123.954(1)  y, deg  90  90  90  90  v, A3  6750(1)  3313.7(5)  3333.1(3)  6538.6(5)  z  8  4  4  8  Pcaic S/cm  1.256  1.193  1.271  1.140  F(000)  2672  1264  1336  2400  Radiation  Cu  Cu  Mo  Mo  H, cm"  42.20  36.99  11.61  3.22  Crystal size, mm  0.08 x 0.20 x 0.45  0.25 x 0.40 x 0.40  0.15x0.30x0.35  0.20x0.10x0.10  Transmission factors  0.559-1.000  0.690-1.000  0.7581-1.0059  -  Scan type  co-29  co-29  co-29  co-29  Scan range, co°  0.94+0.20 tan 9  0.84+0.20 tan 9  -  -  16 (up to 9 scans)  8 (up to 9 scans)  -  -  155  155.5  60.1  21  c,  2  A  3  1  Scan speed, °min"  2e ,deg max  1  2  2  3 jyw-AlCl[P2N2]  2  4 ,ryw-GaCl[P2N2]  2  2  1  1  6 yy/i-A!H[P2N2]  2  1  299  2  2  2  Appendix 1: X-Ray Crystal Structure  Crystal decay, %  4.62  Total reflections  7791  4000  Unique reflections merge  30632  10782  8895  3502  0.075  0.0786  Number w/ / > 3a(l) 3197  3072  Variables  308  308  307  301  R  0.036  0.029  0.075  0.0815  R...  0.032  0.029  0.075  0.01307  gof  1.55  1.73  1.83  1.187  Max A/a (final cycle) 0.0004  0.0005  0.002  Residual density e/A -0.19 to 0.22  -0.18to0.16  -1.12 to 1.03  Compound  8 {,ryn-Ga[P2N2])2  9 {.sy/i-Infl^^lh  Empirical Formula  C4gH84Ga2N4P4Si8 C48H84ln N4P4Si8  fw  1205.24  1295.44  Color, habit  colorless, needle  colorless, prism  Crystal system  tetragonal  tetragonal  Space group  P42ic  P42ic  a, A  16.9622(4)  17.149(2)  b,k  -  -  c,A  21.2288(2)  21.649(2)  oc,deg  90  90  P.deg  90  90  Y, deg  90  90  V,A?  6107.9(2)  6366.5(13)  3  2  300  -0.211 to 0.210  Data  Appendix 1: X-Ray Crystal Structure  z  4  4  1.311  1.351  F(OOO)  2536  2680  Radiation  Mo  Mo  (j., cm"  11.79  9.93  Crystal size, mm  0.10x0.15x0.50  0.22 x 0.25 x 0.35  Transmission factors  0.797-1.004  0.970-1.000  Scan type  co-20  co-20  Scan range, co°  -  -  -  -  2e ,deg  60.1  60  Crystal decay, %  -  -  Total reflections  56346  5252  Unique reflections  4353  5247  0.054  0.107  Pcalc &/  Cm3  1  Scan speed, ° m i n  _1  max  merge  Number W//>3CT(/)  -  Variables  298  298  R  0.052  0.038  0.040  0.032  R  w  gof  1.95  1.61  Max A/a (final cycle) 0.004  0.005  Residual density e/A -1.19 to 1.10  -0.44 to 0.56  3  f  Data  Complex 2, 3: Temperature 294 K, Rigaku AFC6S diffractometer, CuK (k = 1.54178 A) radiation; by Dr. S. J. a  Rettig at UBC. Complex 4,8,9: Temperature 180 K, Rigaku/ADSC CCD diffractometer, MoK (k = 0.71069 A) a  radiation; by Dr. S. J. Rettig at UBC.  Complex 6:  Temperature 298 K, Siemens SMART Platform CCD  diffractometer, MoK (k = 0.71073 A) radiation; by Dr. G. P. Yap at Univ. of Windsor. a  301  Appendix 1: X-Ray Crystal Structure Data  8 For Siemens structures, function minimized Iw(IF l-IF l) where w' = a ( F ) + 0.0010F ,/? = EF -IF I/£IF I, 2  0  R  2  2  7  0  1  2  2  0  0  2  G  = ZI(w / (F -/ c)l/2(w) / F l, and gof = [lLw(\F \-\F \) /(m-n)] V . 1  w  1  c  2  c  o  0  c e q  w  (A ) for anri-GaCl[P2N2] (2). 2  atom  X  y  z  Bea  Ga(l)  0.55794  0.20679  0.30112  3.84  Cl(l)  0.49302  0.21278  0.18305  5.33  P(D  0.51914  0.27254  0.40058  4.74  P(2)  0.65406  0.07231  0.17780  5.08  Si(l)  0.52298  0.15919  0.47192  6.29  Si(2)  0.68678  0.28794  0.34382  5.58  Si(3)  0.72980  0.18592  0.24721  4.74  Si(4)  0.52694  0.07993  0.32661  5.05  N(l)  0.5342  0.1444  0.3663  4.24  N(2)  0.6632  0.2276  0.2963  4.40  C(l)  0.4831  0.2286  0.4817  5.6  C(2)  0.6124  0.2997  0.4305  5.7  C(3)  0.6826  0.1441  0.1629  4.7  C(4)  0.5521  0.0748  0.2123  4.9  C(5)  0.6279  0.1612  0.5220  9.6  C(6)  0.4670  0.1133  0.5349  9.9  C(7)  0.7829  0.2876  0.3979  8.4  C(8)  0.6812  0.3456  0.2697  8.1  C(9)  0.7795  0.1409  0.3246  6.7  302  0  Values given for R, R and gof are based  on those reflections with I > 4a(7).  T a b l e A1.17 Fractional Atomic Coordinates and B  C  Appendix 1: X-Ray Crystal Structure Data  C(10)  0.8044  0.2261  0.1885  6.6  C(ll)  0.5916  0.0323  0.3848  7.7  C(12)  0.4254  0.0533  0.3351  8.3  C(13)  0.4529  0.3280  0.3816  4.8  C(14)  0.4445  0.3482  0.3019  6.1  C(15)  0.3957  0.3921  0.2874  7.4  C(16)  0.3554  0.4147  0.3517  7.8  C(17)  0.3637  0.3950  0.4318  8.0  C(18)  0.4118  0.3508  0.4481  6.5  C(19)  0.6442  0.0542  0.0660  5.0  C(20)  0.5767  0.0548  0.0210  5.6  C(21)  0.5750  0.0405  -0.0636  6.8  C(22)  0.6410  0.0258  -0.1042  7.8  C(23)  0.7087  0.0255  -0.0613  8.9  C(24)  0.7105  0.0393  0.0227  7.3  Table A1.18 Fractional Atomic Coordinates and B  c e q  (A ) for yyn-AlCl[P2N2] (3). 2  atom  X  y  z  Bea  Al(l)  0.46862  0.36490  0.3709  2.93  C1Q)  0.52367  0.32532  0.1708  5.295  P(l)  0.58473  0.37487  0.5277  3.36  P(2)  0.33768  0.36564  0.2712  3.24  Si(l)  0.49859  0.25967  0.6076  4.06  Si(2)  0.49510  0.49553  0.5259  3.46  303  Appendix 1: X-Ray Crystal Structure Data  0.40952  0.49115  0.2341  0.34274  0.32039  0.5915  0.4353  0.3128  0.5294  0.4539  0.4539  0.3817  0.5918  0.3031  0.6380  0.5459  0.4360  0.6482  0.3255  0.4396  0.1733  0.2921  0.3735  0.4525  0.5181  0.1933  0.4785  0.4694  0.2256  0.7913  0.4249  0.5382  0.6468  0.5664  0.5556  0.4625  0.4736  0.4987  0.0709  0.3708  0.5707  0.2793  0.3372  0.3590  0.7785  0.2888  0.2453  0.5978  0.6803  0.4016  0.4813  0.6978  0.4177  0.3358  0.7706  0.4379  0.2963  0.8253  0.4420  0.4059  0.8091  0.4261  0.5501  0.7363  0.4059  0.5883  0.2878  0.3038  0.1710  0.3237  0.2469  0.1440  0.2881  0.2001  0.0620  0.2174  0.2102  0.0048  0.1803  0.2663  0.0285  304  Appendix 1: X-Ray Crystal Structure  C(24)  0.2155  0.3126  Table A1.19 Fractional Atomic Coordinates and B  0.1121  c e q  4.44  (A ) for ,yyrc-GaCl[P2N2] (4). 2  atom  X  y  z  Ga(l)  0.63044  0.53091  0.36484  2.478  Cl(l)  0.83650  0.47455  0.32533  4.97  P(l)  0.47110  0.41503  0.37490  2.90  P(2)  0.72749  0.66254  0.36578  2.75  Si(l)  0.39180  0.50066  0.25893  3.49  Si(2)  0.47330  0.50411  0.49665  2.90  Si(3)  0.76620  0.59126  0.49184  3.21  Si(4)  0.40570  0.65788  0.31995  2.97  N(l)  0.4656  0.5659  0.3106  2.67  N(2)  0.6175  0.5465  0.4563  2.76  C(l)  0.3612  0.4085  0.3041  4.37  C(2)  0.3528  0.4544  0.4363  3.34  C(3)  0.8257  0.6744  0.4401  3.86  C(4)  0.5454  0.7078  0.3733  3.50  C(5)  0.5249  0.4783  0.1937  7.2  C(6)  0.2108  0.5297  0.2230  6.3  C(7)  0.3499  0.5733  0.5393  4.92  C(8)  0.5377  0.4325  0.5554  5.1  C(9)  0.7240  0.6305  0.5710  5.2  C(10)  0.9320  0.5268  0.4991  6.1  305  Data  Appendix 1: X-Ray Crystal Structure  C(ll)  0.2189  0.6623  0.3594  4.65  C(12)  0.3985  0.7127  0.2450  5.6  C(13)  0.5182  0.3197  0.4013  3.33  C(14)  0.4094  0.2633  0.4058  4.62  C(15)  0.4528  0.1902  0.4259  6.3  C(16)  0.5950  0.1748  0.4414  6.1  C(17)  0.7040  0.2297  0.4379  5.9  C(18)  0.6655  0.3028  0.4174  4.5  C(19)  0.8280  0.7117  0.3038  2.84  C(20)  0.8860  0.7847  0.3124  3.73  C(21)  0.9703  0.8192  0.2656  4.51  C(22)  0.9941  0.7827  0.2104  4.9  C(23)  0.9347  0.7112  0.2000  5.1  C(24)  0.8541  0.6756  0.2476  4.21  Table A1.20 Fractional Atomic Coordinates and B  c e q  (A ) for ,yyrt-AlH[P N ] (6). 2  2  2  atom  X  y  z  Bea  Al(l)  0.1439  1.0604  0.2012  5.5  P(l)  0.0879  1.1643  0.0814  5.7  P(2)  0.1859  0.9076  0.3119  6.6  Si(l)  0.1015  0.8441  0.0717  6.0  Si(2)  0.1908  0.7899  0.1937  6.6  Si(3)  0.0951  0.9422  0.2661  7.4  Si(4)  0.0688  1.2286  0.1880  7.9  306  Data  Appendix 1: X-Ray Crystal Structure  0.1441  0.8867  0.1556  4.9  0.0991  1.0734  0.2160  5.7  0.0594  1.3086  -0.0474  9.8  0.0669  1.3822  -0.0933  13.3  0.1064  1.4015  -0.0771  12.9  0.1407  1.3525  -0.0148  11.7  0.1337  1.2782  0.0321  8.2  0.0936  1.2572  0.0162  6.1  0.2499  1.0573  0.4257  9.0  0.2772  1.1006  0.4961  11.4  0.2708  1.0422  0.5457  12.0  0.2398  0.9482  0.5265  15.3  0.2134  0.9028  0.4562  12.5  0.2188  0.9583  0.4052  6.8  0.0610  0.9939  0.0418  6.4  0.2164  0.7859  0.2937  8.1  0.1404  0.8093  0.2970  7.4  0.0532  1.2717  0.0956  7.4  0.1160  0.8316  0.0052  9.8  0.0743  0.6720  0.0679  9.6  0.2282  0.8808  0.1765  9.5  0.1839  0.5969  0.1638  9.5  0.0992  1.0174  0.3461  11.9  0.0447  0.8314  0.2144  11.7  0.0997  1.3865  0.2443  15.6  0.0179  1.2159  0.1827  14.4  307  Data  Appendix 1: X-Ray Crystal Structure  Table A1.21 Fractional Atomic Coordinates and B  c e q  (A ) for {syn-Ga[P2N2] h (8). 2  atom  X  y  z  Ga(l)  0.01173  0.42663  0.14674  1.774  Ga(2)  -0.01173  0.57337  0.14674  1.774  P(l)  -0.16570  0.38187  0.12263  2.03  P(2)  0.16431  0.39739  0.16355  2.03  P(3)  0.16570  0.61813  0.12263  2.01  P(4)  -0.16431  0.60261  0.16355  2.01  Sid)  -0.10239  0.36720  0.25784  1.92  Si(2)  -0.03572  0.28586  0.04934  1.92  Si(3)  0.11180  0.37813  0.02769  2.19  Si(4)  0.05419  0.28799  0.24029  2.19  Si(5)  0.10239  0.63280  0.25784  2.28  Si(6)  0.03572  0.71414  0.04934  2.28  Si(7)  -0.11180  0.62187  0.02769  1.98  Si(8)  -0.05419  0.71201  0.24029  1.98  N(l)  -0.0159  0.3544  0.2151  1.69  N(2)  0.0291  0.3593  0.0725  1.69  N(3)  0.0159  0.6456  0.2151  2.05  N(4)  -0.0291  0.6407  0.0725  2.05  Cd)  -0.1918  0.3662  0.2047  2.28  C(2)  -0.1250  0.2850  0.1016  2.28  C(3)  0.1976  0.3946  0.0825  2.09  C(4)  0.1428  0.2957  0.1875  2.09  C(5)  -0.1244  0.2866  0.3163  2.58  C(6)  -0.0996  0.4608  0.3039  2.58  308  Data  Appendix 1: X-Ray Crystal Structure Data  0.0063  0.1827  0.0535  -0.0693  0.2987  -0.0344  0.1015  0.4684  -0.0218  0.1447  0.2958  -0.0251  0.0246  0.1821  0.2373  0.0842  0.3069  0.3234  -0.2594  0.3789  0.0797  -0.3331  0.3735  0.1085  -0.4021  0.3730  0.0735  -0.4007  0.3809  0.0089  -0.3278  0.3875  -0.0211  -0.2591  0.3881  0.0143  0.2514  0.4220  0.2100  0.3187  0.4559  0.1827  0.3823  0.4795  0.2178  0.3824  0.4676  0.2814  0.3191  0.4319  0.3103  0.2544  0.4112  0.2745  0.1918  0.6338  0.2047  0.1250  0.7150  0.1016  -0.1976  0.6054  0.0825  -0.1428  0.7043  0.1875  0.1244  0.7134  0.3163  0.0996  0.5392  0.3039  -0.0063  0.8173  0.0535  0.0693  0.7013  -0.0344  -0.1015  0.5316  -0.0218  309  Appendix 1: X-Ray Crystal Structure  C(34)  -0.1447  0.7042  -0.0251  2.37  C(35)  -0.0246  0.8179  0.2373  2.36  C(36)  -0.0842  0.6931  0.3234  2.36  C(37)  0.2594  0.6211  0.0797  2.33  C(38)  0.3331  0.6265  0.1085  2.33  C(39)  0.4021  0.6270  0.0735  2.83  C(40)  0.4007  0.6191  0.0089  2.83  C(41)  0.3278  0.6125  -0.0211  3.38  C(42)  0.2591  0.6119  0.0143  3.38  C(43)  -0.2514  0.5782  0.2100  4.15  C(44)  -0.3187  0.5441  0.1827  4.15  C(45)  -0.3823  0.5205  0.2178  4.17  C(46)  -0.3824  0.5324  0.2814  4.17  C(47)  -0.3191  0.5681  0.3103  3.42  C(48)  -0.2544  0.5888  0.2745  3.42  Table A1.22 Fractional Atomic Coordinates and B  c e q  (A ) for {.ryn-In[P2N2]}2 (9). 2  atom  X  y  z  Bea  In(l)  0.00136  0.41949  0.14057  3.291  In(2)  -0.00136  0.58051  0.14057  3.291  P(l)  -0.1567  0.3670  0.12742  3.94  P(2)  0.1644  0.3851  0.1565  3.94  P(3)  0.1567  0.6330  0.1274  4.68  P(4)  -0.1644  0.6149  0.1565  4.68  310  Data  Appendix 1: X-Ray Crystal Structure Data  Si(l)  -0.1007  0.3613  0.2614  4.21  Si(2)  -0.0408  0.2652  0.0501  4.21  Si(3)  0.1068  0.3566  0.0247  4.73  Si(4)  0.0567  0.2861  0.2452  4.73  Si(5)  0.1007  0.6387  0.2614  4.84  Si(6)  0.0408  0.7348  0.0501  4.84  Si(7)  -0.1068  0.6434  0.0247  4.32  Si(8)  -0.0567  0.7139  0.2452  4.32  N(l)  -0.0180  0.3455  0.2204  3.4  N(2)  0.0245  0.3371  0.0674  3.4  N(3)  0.0180  0.6545  0.2204  3.7  N(4)  -0.0245  0.6629  0.0674  3.7  C(l)  -0.1888  0.3578  0.2071  4.3  C(2)  -0.1235  0.2683  0.1067  4.3  C(3)  0.1952  0.3657  0.0777  4.3  C(4)  0.1396  0.2906  0.1886  4.3  C(5)  -0.1208  0.2866  0.3218  6.7  C(6)  -0.1042  0.4571  0.3004  6.7  C(7)  0.0021  0.1632  0.0563  5.0  C(8)  -0.0786  0.2729  -0.0302  5.0  C(9)  0.0995  0.4499  -0.0174  6.5  C(10)  0.1339  0.2801  -0.0329  6.5  C(ll)  0.0292  0.1804  0.2476  6.8  C(12)  0.0896  0.3122  0.3253  6.8  C(13)  -0.2456  0.3781  0.0814  6.8  C(14)  -0.3191  0.3654  0.1059  6.8  C(15)  -0.3854  0.3730  0.0700  7.7  311  Appendix 1:  -0.3807  0.3934  0.0078  -0.3090  0.4044  -0.0155  -0.2421  0.3960  0.0209  0.2522  0.4151  0.1995  0.3119  0.4520  0.1688  0.3769  0.4799  0.2021  0.3818  0.4688  0.2633  0.3238  0.4346  0.2940  0.2601  0.4073  0.2620  0.1888  0.6422  0.2071  0.1235  0.7317  0.1067  -0.1952  0.6343  0.07770  -0.1396  0.7094  0.1886  0.1208  0.7134  0.3218  0.1042  0.5429  0.3004  -0.0021  0.8368  0.0563  0.0786  0.7271  -0.0302  -0.0995  0.5501  -0.0174  -0.1339  0.7199  -0.0329  -0.0292  0.8196  0.2476  -0.0896  0.6878  0.3253  0.2456  0.6219  0.0814  0.3191  0.6346  0.1059  0.3854  0.6270  0.0700  0.3807  0.6066  0.0078  0.3090  0.5956  -0.0155  0.2421  0.6040  0.0209  312  X-Ray Crystal Structure  Data  Appendix 1: X-Ray Crystal Structure Data  C(43)  -0.2522  0.5849  0.1995  8.4  C(44)  -0.3119  0.5480  0.1688  8.4  C(45)  -0.3769  0.5201  0.2021  8.6  C(46)  -0.3818  0.5312  0.2633  8.6  C(47)  -0.3238  0.5654  0.2940  7.1  C(48)  -0.2601  0.5927  0.2620  7.1  313  Appendix 1: X-Ray Crystal Structure  Data  Table A1.23 Crystallographic Data for Chapter Sixg. . h  Compound  1SCC12[N3]  2 anf/-TiCl2[P2N2]  4 [syn-\P2N ]V}22  5 a«ri-(CdCl)2[P2N ] 2  CN2) Empirical Formula  C H28Cl2N ScSi2 10  3  C 4H42Cl2N P2Si4Ti C H 4 N P S i V 2 2  2  4 8  8  6  4  8  C24 2Cd2Cl2 H  4  N2P2SU fw  362.38  651.71  1260.19  828.60  Color, habit  colorless, irregular  yellow, plate  purple, prism  colorless, prism  Crystal system  monoclinic  triclinic  monoclinic  triclinic  Space group  C2/c  PI  a, A  14.612(1)  8.5671(7)  11.768(2)  8.9383(1)  b, A  8.844(2)  13.945(2)  28.392(7)  9.2294(2)  c A  15.3397(8)  18.613(2)  21.933(4)  11.3609(2)  a, deg  90  104.170(5)  90  110.431(1)  P.deg  104.184(3)  100.116(2)  101.64(1)  91.93(1)  Y.deg  90  106.033(2)  90  96.643(1)  V,A?  1921.9(3)  2000.4(4)  7177(2)  869.60(3)  Z  4  2  4  1  1.252  1.235  1.166  1.582  F(000)  768  784  2668  416  Radiation  Mo  Mo  Mo  Mo  p., cm"  7.74  5.70  5.18  16.23  Crystal size, mm  0.15x0.35x0.55  0.03 x 0.25 x 0.45  0.20x0.10x0.10  0.20x0.20x0.20  Transmission factors  0.8103-0.9984  0.422-1.001  -  -  Scan type  co-29  co-29  co-29  co-29  Pcalc g /  c m 3  1  PI  Scan range, co°  314  Appendix 1: X-Ray Crystal Structure  Data  Scan speed, °min' 2e  ,deg  60.1  63.6  21  25  Crystal decay, %  -  -  -  -  Total reflections  8962  ,18659  24450  4618  Unique reflections  2383  8065  7659  2995  0.048  0.064  0.1065  0.0139  Number w/1 > 3,0(1)-  -  -  -  Variables  139  400  669  163  R  0.038  0.081  0.0953  0.0219  R  0.061  0.079  0.1697  0.0603  gof  1.00  2.14  1.184  1.053  Max No (final cycle) 0.004  0.06  -  -  Residual density e/A -0.47 to 0.50  -1.17 to 1.75  -0.305 to 0.537  -0.317 to 0.447  max  P merge  w  3  Complex 1, 2: Temperature 180 K, Rigaku/ADSC CCD diffractometer, MoK (X = 0.71069 A) radiation; by Dr. S. a  J. Rettig at UBC.  Complex 3,4: Temperature 298 K, Siemens SMART Platform CCD diffractometer, MoK (X = a  0.71073 A) radiation; by Dr. G. P. Yap at Univ. of Windsor.  Table A1.24 Fractional Atomic Coordinates and B  c e q  (A ) for ScCl2[N(SiMe2CFl2NMe2)2] (1). 2  atom  X  y  z  Sc(l)  0.5000  0.46973  0.2500  1.14  Cl(l)  0.42085  0.32839  0.34109  3.03  Si(D  0.45699  0.78281  0.14694  1.42  N(l)  0.5000  0.7012  0.2500  1.26  N(2)  0.3640  0.5022  0.1362  1.56  C(l)  0.3937  0.6187  0.0777  1.76  315  Bea  Appendix 1: X-Ray Crystal Structure  C(2)  0.5512  0.8477  0.0933  3.13  C(3)  0.3759  0.9460  0.1464  3.04  C(4)  0.2863  0.5637  0.1719  2.45  C(5)  0.3288  0.3643  0.0839  2.88  T a b l e A1.25  Fractional Atomic Coordinates and B  c e q  (A ) for <wri-TiCl2[P2N2] 2  (2).  atom  X  y  z  Ti(l)  0.55660  0.13781  0.27183  1.740  Cl(l)  0.52521  -0.03861  0.25638  4.28  Cl(2)  0.84466  0.17143  0.31016  2.65  P(l)  0.24017  0.06563  0.25217  1.95  P(2)  0.79826  0.43322  0.24931  2.20  Si(l)  0.31349  0.20161  0.15439  1.88  Si(2)  0.40130  0.18416  0.41667  2.44  Si(3)  0.65002  0.36576  0.38897  2.16  Si(4)  0.66242  0.20864  0.13264  2.02  N(l)  0.5101  0.1877  0.1866  1.89  N(2)  0.5304  0.2304  0.3584  1.96  C(l)  0.1899  0.1732  0.2262  1.80  C(2)  0.2181  0.0711  0.3478  2.45  C(3)  0.8242  0.4137  0.3452  2.10  C(4)  0.8499  0.3249  0.1916  2.33  C(5)  0.3214  0.3343  0.1463  2.24  C(6)  0.1934  0.0992  0.0600  2.77  316  Data  Appendix 1: X-Ray Crystal Structure  C(7)  0.3080  0.2817  0.4648  3.75  C(8)  0.5144  0.1363  0.4867  4.14  C(9)  0.5113  0.4431  0.3722  3.54  C(10)  0.7586  0.4038  0.4928  3.70  C(ll)  0.7144  0.0867  0.1003  2.77  C(12)  0.5874  0.2401  0.0436  3.23  C(13)  0.0852  -0.0563  0.1871  2.25  C(14)  0.1276  -0.1257  0.1317  2.55  C(15)  0.0069  -0.2201  0.0845  2.97  C(16)  -0.1519  -0.2434  0.0907  3.22  C(17)  -0.1991  -0.1765  0.1451  3.41  C(18)  -0.0761  -0.0814  0.1917  2.69  C(19)  0.9939  0.5450  0.2699  2.43  C(20)  0.9777  0.6364  0.2592  3.37  C(2.1)  1.1210  0.7219  0.2724  4.40  C(22)  1.2777  0.7159  0.2934  4.40  C(23)  1.2950  0.6237  0.3037  3.98  C(24)  1.1536  0.5396  0.2907  3.14  Table A1.26 Fractional Atomic Coordinates and B  c e q  (A ) for {syn-[P2^2\^}2(^2) 2  (4).  atom  X  y  z  Bea  V(l)  0.8979  0.1058  0.6631  3.1  V(2)  0.9144  0.2666  0.72265  3.4  Si(l)  0.7471  0.1166  0.5264  4.5  317  Data  Appendix 1:  0.6800  0.0380  0.6067  1.0402  0.0521  0.7819  1.0870  0.0238  0.6554  0.7725  0.3252  0.8141  0.6546  0.3086  0.6813  1.0710  0.3560  0.7004  1.1815  0.2833  0.7949  0.9975  0.1067  0.5740  0.7901  0.0678  0.7367  0.9699  0.2563  0.8406  0.8618  0.3064  0.6244  0.9064  0.1652  0.6871  1.0170  0.0559  0.7030  0.7668  0.0844  0.5936  0.9093  0.2069  0.7029  1.0648  0.3054  0.7426  0.7727  0.3044  0.7416  1.1914  0.1634  0.6147  1.2972  0.1836  0.6105  1.3468  0.1742  0.5594  1.2906  0.1445  0.5125  1.1849  0.1243  0.5166  1.1353  0.1337  0.5677  0.6120  0.1300  0.7454  0.5110  0.1437  0.7642  0.4586  0.1136  0.8003  0.5073  0.0698  0.8176  318  X-Ray Crystal Structure  Data  •I  Appendix 1: X-Ray Crystal Structure Data  0.6084  0.0561  0.7988  4.6  0.6607  0.0862  0.7627  5.6  0.8928  0.1206  0.5034  4.5  0.7396  0.0168  0.6895  5.5  0.8945  0.0476  0.8044  5.3  1.0268  0.0434  0.5721  7.9  0.6415  0.0918  0.4588  7.3  0.7002  0.1788  0.5408  7.4  0.6892  -0.0133  0.5534  7.6  0.5243  0.0536  0.6013  12.2  1.1317  0.0003  0.8158  9.6  1.1066  0.1080  0.8186  7.7  1.0574  -0.0414  0.6582  7.9  1.2495  0.0324  0.6732  9.4  0.9470  0.2191  0.9553  11.1  0.8945  0.1910  0.9936  9.1  0.8084  0.1593  0.9676  8.2  0.7747  0.1556  0.9032  6.3  0.8272  0.1837  0.8650  4.6  0.9133  0.2155  0.8910  7.6  0.8577  0.3149  0.4971  9.1  0.8985  0.3063  0.4429  9.3  0.9909  0.2756  0.4440  10.0  1.0427  0.2536  0.4993  7.6  1.0020  0.2623  0.5535  4.8  0.9095  0.2929  0.5524  5.0  0.9206  0.3128  0.8631  5.2  319  Appendix 1: X-Ray Crystal Structure  C(38)  0.7058  0.3134  0.6056  5.2  C(39)  0.9264  0.3637  0.6465  5.7  C(40)  1.1254  0.2555  0.8617  10.4  C(41)  0.7559  0.3907  0.8164  8.3  C(42)  0.6585  0.2983  0.8507  10.6  C(43)  0.5597  0.3602  0.6854  10.2  C(44)  0.5617  0.2544  0.6754  11.1  C(45)  1.0927  0.4112  0.7499  11.0  C(46)  1.1864  0.3541  0.6529  8.9  C(47)  1.2571  0.2356  0.7592  8.3  C(48)  1.2922  0.3284  0.8290  16.0  Table A1.27 Fractional Atomic Coordinates and B  c e q  (A ) for <wri-(CdCl)2[P2N2]) (5). 2  atom  X  y  z  Bea  Cd(l)  -0.0613  0.6504  1.09595  3.0  P(l)  0.2090  0.7776  1.1545  3.2  Si(l)  0.1475  0.7029  0.8639  3.4  Si(2)  0.1692  0.4708  1.2103  3.7  N(l)  -0.0029  0.5804  0.8864  2.9  Cl(l)  -0.2986  0.7374  1.1472  6.1  C(l)  0.1871  1.0123  1.3823  5.1  C(2)  0.2434  1.1403  1.4887  6.9  C(3)  0.3925  1.1991  1.5046  6.5  C(4)  0.4876  1.1337  1.4163  5.7  320  Data  Appendix 1: X-Ray Crystal Structure Data  C(5)  0.4344  1.0056  1.3101  4.3  C(6)  0.2832  0.9441  1.2926  3.5  C(7)  0.2700  0.8158  1.0165  4.7  C(8)  0.2736  0.5941  0.7479  6.5  C(9)  0.0875  0.8544  0.8072  8.1  C(10)  0.2975  0.6129  1.1602  3.8  C(ll)  0.2779  0.3063  1.1966  6.9  C(12)  0.1363  0.5711  1.3796  7.0  321  Appendix 2: Cartesian Coordinates for the Model Compound ScMe2[N(SiMe CH PH2)2] 2  2  Figure A2.1 Chem 3D view of the model compound ScMe2[N(SiMe2CH2PH2)2]-  Table A2.1 Cartesian coordinates for the model compound ScMe2[N(SiMe2CH2PH2)2l-  atom  X  y  z  Sc(l)  9.444  4.354  5.194  P(2)  11.992  5.014  6.027  H(3)  9.942  2.665  7.373  N(4)  9.048  6.419  5.263  H(5)  12.939  4.602  5.112  322  Appendix 2: ..Model Compound  ScMe2[N(SiMe2CH2PH2)2l  H(6)  8.275  2.389  6.813  H(7)  8.652  3.805  7.824  C(8)  11.962  6.833  6.136  H(9)  12.273  4.462  7.260  H(10)  11.189  3.006  3.466  H(ll)  9.526  2.792  2.870  H(12)  10.393  4.315  2.559  C(13)  6.262  5.926  4.265  C(14)  9.024  3.148  7.038  Si(15)  10.266  7.511  5.786  C(16)  10.240  3.508  3.276  P(17)  6.815  4.191  4.327  Si(18)  7.505  7.008  4.796  H(19)  12.272  7.122  7.140  C(20)  6.813  7.930  6.281  H(21)  6.063  3.454  5.219  H(22)  6.759  3.600  3.081  C(23)  9.650  8.316  7.370  H(24)  12.668  7.233  5.409  C(25)  10.453  8.819  4.450  C(26)  7.798  8.218  3.388  H(27)  5.993  6.191  3.242  H(28)  5.397  6.060  4.914  H(29)  7.488  7.812  7.129  H(30)  5.834  7.524  6.537  H(31)  6.715  8.988  6.039  H(32)  8.854  8.212  3.117  323  Appendix 2: ...Model Compound  ScMe2[N(SiMe2CH2PH2)2l  H(33)  7.200  7.924  2.525  H(34)  7.510  9.221  3.705  H(35)  10.376  8.155  8.166  H(36)  8.695  7.874  7.654  H(37)  9.520  9.386  7.206  H(38)  11.491  8.853  4.117  H(39)  9.809  8.572  3.606  H(40)  10.169  9.792  4.851  324  Appendix 3: Kinetic Data for the Conversion of syn-HxlPi^l] anti-HzlPlNz]  to  in 4:1 C6D5Br/C6D6  0.8  time (hours)  Figure A3.1a) Rate plots for the conversion of syrt-H2[P2N2] to anr/-H2[P2N ] in 4:1 2  C D5Br/C6D6. 6  b) Pre-equilibrium first order rate plots for the conversion of  anri-H [P2N ] in 4:1 C D 5 B r / C D . 2  2  6  6  325  6  syn-H2[P2^2\ to  Appendix 3: Kinetic  Data...  Table A3.1 Rate data for the conversion of syn-U2[P2^2\ to anti-E.2[P2^2\ in 4:1 C D Br/C D . 6  5  6  6  temp, °C  kobs, s"  correlation, R  100  2.92 x IO" ± 0 . 2 4  0.997  120  5.42 x 10" ± 0.47  0.996  140  1.77 x IO" ± 0 . 2 7  0.989  160  2.55 x 10-6 + 0.32  0.990  1  6  6  5  326  Appendix 3: Kinetic  Data..  Table A3.2 Transition state parameters for the conversion of syn-H.2[P2^2] to a«n'-H2[P2N2] in  4:lC D Br/C6D6. 6  5  function E , kcal mol  syn- to anti12.4 ± 1.8  -1  a  AH*, kcal mol'  11.7 ± 1.8  1  AS*, cal K" mol"  1  -53.0 ± 4.8  AG373*, kcal mol  -1  31.4 ± 3 . 7  1  correlation, R  0.982  327  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share