Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

X-ray crystallographic studies of five group III compounds Rettig, Steven J. 1974-12-31

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata

Download

Media
UBC_1974_A1 R48.pdf [ 10.49MB ]
Metadata
JSON: 1.0061057.json
JSON-LD: 1.0061057+ld.json
RDF/XML (Pretty): 1.0061057.xml
RDF/JSON: 1.0061057+rdf.json
Turtle: 1.0061057+rdf-turtle.txt
N-Triples: 1.0061057+rdf-ntriples.txt
Original Record: 1.0061057 +original-record.json
Full Text
1.0061057.txt
Citation
1.0061057.ris

Full Text

X-RAY CRYSTALLOGBAPHIC STUDIES 01 FIVE GROUP III COMPOUNDS 49. by STEVEN J. EETTIG B.S., University of Illinois at Chicago Circle, 1970 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in the Department of CHEMISTRY We accept this thesis as conforming tc the required standard THE UNIVERSITY OF BRITISH COLUMBIA APRIL 1974 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the Head of my Department or by his representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of Chemistry  The University of British Columbia Vancouver 8, Canada Date June 1Q7A ii ABSTRACT Supervisor: Professor James Trotter The structures of five compounds of group III elements have been determined by single crystal X-ray diffraction, three boron compounds, one aluminum compound, and one gallium compound: 1» B,B-diphenylboroxazolidine (2-atninoethyl diphenyl-borinate) , C^H^BNO. 2. B ,B-bis (£-f luorophenyl) boroxazolidine , C-iL^H^BF2 NO. 3. 4 , U-dimethyl-2,2-diphenyl-1,3-dioxa-4-azonia-2-bor-anatacyclopentane, Cj^H^gBNG^. 4. N-methyldiethanolaminogallane dimer, C^QH^G^®2°k' 5. (£entaha£tocyclopentadienyl) hydridomolybdenum-><-di-me th ylaluminum-/<-[ methylaluminum-di- iju.-£entaha_p_to-(monoha£to)cyclopentadienyl)dimethylaluminum] (£entaha]3-tocyclopentadienyl)hydridomolybdenum, C25 H35A ^3Mo2* Crystals of B,B-diphenylboroxazolidine are monoclinic, a = 13.840(1), b = 8.9169(5), c = 10.170(1) A, = 98.85 (1)° , Z = 4, space group P2^/n. The structure was determined by direct methods, and refined by electron-density and full-matrix least-squares procedures to R 0.041 for 1458 reflexions. The five-membered boroxazolidine ring is in the half-chair conformation. Bond angles in the ring range from 99.7 for OBN to 110.1° for BOC. Bond lengths are as follows: mean B-C, 1.616(2), B-H, 1 .653 (3), B-0, 1.484(3), C-N, 1 .485 (3), C-O, 1.413(3), mean C-C (aromatic) , 1 .392(11 ), and iii C-C, 1.505(4) A. The structure consists of discrete molecules linked by 0...H-N hydrogen bonds (2,874(2) A) to form continuous spirals about the 2^ axes. Crystals of B,B-bis (£-fluorophenyl)boroxazolidine are orthorhomfcic, a = 13.442 (4), b = 10.214(3), c = 9. 823 (2) ft, Z = 4, space group P2^ 2\ 2\. The structure was solved by direct methods, and refined by electron-density and full-matrix least-squares procedures to R 0.047 for 1234 reflexions. The five-membered boroxazolidine ring is in a distorted half-chair conformation , Bond angles in the ring range from 99.9(2) for OBH to 108.2(2)° for BOC. Bond lengths are: mean B-C, 1.621:, (3), B-N, 1.652 (4), B-O, 1.471(4), C-N, 1.491(4), C-0, 1.4 18(4), mean C-F, 1.371(1), mean C-C (aromatic) , 1 .390 (13), and C (sp_3) -C (s£3) , 1.494 (6) A. The structure consists of discrete molecules each linked to six others by an extensive network of O...H-N (0...N = 2.941(3) A), F...H-N (F...N = 3.171(4) A), ana F...H-C (F...C = 3.318(5) A) hydrogen bonds. Crystals of 4,4-dimethyl-2,2-diphenyl-1,3-dioxa-4-azonia-2-boranatacyclopentane are orthorhombic, a = 17.043 (3), b= 6.289 (1), c= 13.024 (2) I, Z= 4, space group Pna2i. The structure was determined by direct methods, and was refined by full-matrix least-squares procedures to R 0,071 for 1100 reflexions. Bond angles in the five-membered ring , which has a distorted half-chair conformation, range from 101.5(4) for OBO to 107.1(4)° for NOB. Bond lengths are: mean B-C, 1.632 (8), B-0, 1.506(7) and 1.556 (8), N-0, iv 1.409(5), C-O, 1. 378 (9), C-N, 1.467-1.509(7-10), mean C-C(aromatic), 1.395(25)A. The structure consists of discrete molecules separated by normal van der Waals distances. Crystals of the N-methyldiethanolaminogallane dimer are orthorhomfcic, a = 19.112(4), b = 9.947(2), c = 7.709(2) A, Z = 4, space group ^2\2\2i . The structure was determined by Patterson and Fourier synthesis and was refined by full-matrix least-squares procedures to a final R of 0.056 for 1477 reflexions. The structure provides the first known crystallographic example of pentacoordinate gallium, the dimerization of HeH (CH2CH2°)2GaH occurring via the formation of a four-membered Gn2°2 ^in9« Tne coordination about the gallium is distorted trigonal bipyramidal with an angle of 151.2(4)° between the axial substituents. The mean bond distances are: Ga-N, 2.192(5), and Ga-O, 2.018(2) for axial ligands; Ga-O, 1.847(2), 1.960 (8), and Ga-H, 1.41(4) for equatorial ligands; O-C, 1.419(14), C-N, 1.470(7), C-C, 1.520 (12), and C-H, 1.00 (13) A. The molecule has C2 symmetry to within experimental error. There are possible C-H...0 hydrogen bonds (C...O, 3.13 (1)-3.44 (1) A) in the structure. Crystals of the hydridomolybdenum complex, C2^H^^A l-^MOg, are orthorhombic, a = 19.398 (4) , b = 14.438 (9) , c = 9.0 35 (2) I, Z =4, space group V.2\2\2\, The structure was determined by Patterson and Fourier syntheses, and refined by full-matrix least-sguares procedures to R 0.066 and Rw 0.063 for 1213 reflexions. The molecular structure exhibits several unusual features: C^E^ groups which are £entaha£to to the molybdenum atoms and are involved via the unique carbon atom in multicentre bonding to two aluminum atoms, one of which occurs as an Al (Me)2 unit and the other an AlMe unit which also bridges the two molybdenum atoms. The third aluminum atom is probably involved in a Mo-H-Al (Me)2-H-Mo linkage. Mean bond distances are: Mo-Al, 2.659 and 2.974, Al-C(terminal) , 2.00, ftl-C (bridge), 2.05 and 2.33, Mo-C (cyclopentadienyl) , 2. 285, and C-C (cyclopentadienyl) , 1.389 TABLE OF CONTENTS Page TITLE PAGE i ABSTRACT ii TABLE OF CONTENTS vLIST OF TABLES x LIST OF FIGURES xii ACKNOWLEDGEMENTS v GENERAL INTRODUCTION 1 PART 1. CRYSTAL AND MOLECULAR STRUCTURE OF B,B-DIPHENYLBOROX AZOLIDINE (2-A MINOETHYLDIPHENYLBORINATE) 4 Introduction 5 Experimental 6 Structure analysis .............................. 7 Analysis of thermal motion 9 Results and discussion .......................... 12 PART 2. CRYSTAL AND MOLECULAR STRUCTURE OF B, B-BIS (J3-FLUORO-PHENYLBOROXAZOLIDINE 27 Introduction 28 ExperimentalStructure analysis .............................. 30 Analysis of thermal motion 34 Results and discussion .......................... 38 vii PART 3. CRYSTAL AND MOLECULAR STRUCTURE OF 4,4-DIMETHYL-2,2-DIPHENYL-1 ,3-DIOXA-4 - A ZONIA-2-BOR ANAT ACYCLOPENTAN E . 52 Introduction 53 Experimental 5 Structure analysis 57 Analysis of thermal motion 61 Results and discussion 5 PART 4. CRYSTAL AND MOLECULAR STRUCTURE OF THE N-METHYLDIETHANOLAMINOGALLANE DIMER 77 Introduction 78 Experimental 9 Structure analysis 81 Analysis of thermal motion ...................... 82 Results and discussion . . 88 PART 5. CRYSTAL AND MOLECULAR STRUCTURE CF (PENTAHAPTOCYCLO-PENTADIENYL)HYDRIDOHOLYBDENUM-^-DIMETHYLALUMINUM^-[ METH YL ALUM IN UM-DI- yu. - £ ENT AH APT 0 (RONOHAPTO) CYCLOP FN-TADIENYL)DIMETHYLALUMINUM] (PENTAHAPTOCYCLOPENT ADIEN-YL) HYDRIDOMOLYBDENUM 103 Introduction 104 ExperimentalStructure analysis 106 Results and discussion 108 viii PART 6. THE COMPUTER PROGRAM "SIGCOR" 129 Introduction 130 General description 13Program instructions 9 Source listing 143 Example output 9 Discussion 15SUMMARY . 157 REFERENCES 160 ix LIST OF TABLES Table Page B,B-Diphenylboroxazolidine 1 Starting set of reflexions 8 2 Final atomic coordinates 10 3 Final thermal parameters 1 4 Rigid-body thermal parameters ................... 13 5 Bond lengths 17 6 Bond angles 8 7 Torsion angles 20 8 Mean planes 1 9 Non-bonded contacts 25 B,B-bis (g-Fluorophenyl) boroxazolidine 10 Starting set of reflexions 31 11 Results of the phase determination procedure .... 32 12 Final atomic coordinates 35 13 Final thermal parameters ........................ 36 14 Rigid-body thermal parameters 37 15 Bond lengths 40 16 Bond angles 1 17 Torsion angles 43 18 Mean planes 7 19 Non-bonded contacts 50 X 4,4-Dimethyl-2,2-diphenyl-1,3-dioxa-4-azonia-2-boranatacyclopentane 20 Starting set of reflexions ...................... 59 21 Results of the phase determination procedure .... 60 22 Final atomic coordinates 62 23 Final thermal parameters ........................ 63 24 Rigid-body thermal parameters 65 25 Bond lengths 67 26 Bond angles 8 27 Torsion angles 69 28 Mean planes 72 29 Non-bonded contacts 73 N-Methyldiethanolaminogallane Dimer 30 Results of Hamilton's test 83 31 Final atomic coordinates 4 32 Final thermal parameters 85 33 Rigid-body thermal parameters 87 34 Bond lengths ,. 89 35 Bond angles 90 36 Structural data for some gallium complexes 94 37 Mean planes 6 38 Torsion angles 97 39 Non-bonded contacts 101 xi Hydridomolybdenum complex 40 Final atomic coordinates 109 41 Final thermal parameters 110 42 Calculated hydrogen positions 111 43 Results of Hamilton»s test 113 44 Bond lengths 116 45 Bond angles ... 7 46 Mean planes 121 47 Non-bonded contacts 123 48 Structural data for some molybdenum - cyclopentadienyl complexes ......... 124 Computer program SIGCOR 49 Results of sample calculations 154 xii LIST OF FIGURES Figure Page B,B-Diphenylboroxazolidine 1 The molecule viewed down b 14 2 View of the molecule showing bond lengths ........ 15 3 The structure viewed down c 24 BfB-bis(jD-Fluorophenyl)boroxazolidine 4 A stereo view of the molecule 39 5 The structure viewed along b .................... 44 6 The structure viewed down c 44,4-Dimethyl-2,2-diphenyl-1,3-dioxa-4-azonia-2-boranatacyclopentane 7 A stereo view of the molecule 64 8 The structure viewed along b 7N-Methyldiethanolaminogallane Dimer 9 A stereo view of the molecule ................... 86 10 Coordination about the gallium atoms ............ 93 11 The structure viewed along c .................... 99 12 The structure viewed along b 100 xiii Hydridomolybdenum complex 13 A stereo view of the molecule 114 14 The structure viewed down c ..................... 115 15 The structure viewed along b 1116 The Al-C-C-Al bridging system 120 17 Idealized structure of bis (cyclopentadienyl) -transition metal complexes with canted Cp rings . 127 The computer program SIGCOR 18 Bond contraction vs. s character ................ 136 19 Bond order vs. bond contraction 137 xiv ACKNOWLEDGEMENTS I wish to thank Professor James Trotter for giving me the opportunity to join his research group and for the help he has given me during the past four years. I am also indebted to my fellow graduate students and postdoctoral fellows, in particular Drs. Ian Nowell and Bill Harrison, for the assistance they have given me. I would also like to thank Prof. W. Kliegel, Technischen Universitat Braunschwieg, for providing the sample and introductory material for the study of 4,4-Dimethyl-2, 2-diphenyl-1,3-dioxa-4-azonia-2-boranatacyclopentane (Part 3) and Prof. H. Noth, University of Munich, for running the 11 B NMR spectrum. I thank Dr. Alan Storr for providing background material, details of the preparations, and crystals of the N-Methyldiethanolaminogallane dimer (Part 4) and the hydridomolybdenum complex (Part 5) . I am grateful to the National Research Council of Canada for a postgraduate scholarship (1972-73 and 1973-74). ENEBAL INTRODUCTION 2 The historical background and established principles of X-ray crystallography are dealt with in a number of standard texts (1-5) . The crystallographic symbols and nomenclature appearing throughout this thesis have their conventional meanings described in the "International Tables for X-ray Crystallography" (6). The main body of the thesis, parts 1-5, consists of the crystallographic studies of the five compounds containing group III elements. Each part includes introductory material relevant to that particular compound as well as details of the structure determination and a discussion of the results. The final part of the thesis describes a computer program which calculates approximate valence bond orders from observed molecular geometry. It is based on a general relationship which associates bond order with the fractional difference between the observed interatomic distance and the calculated single bond distance. Hybridization and electronegativity effects are considered in the calculation of the single bond distances. For each of the five crystal structures the least-squares refinement was based on the minimization of2w(Fo-Fc)2 where Fo and Fc are the observed and calculated structure factors and w is the assigned weighting factor. The anisotropic thermal factors employed in the refinement are y_i ^ in the expression: f = f °exp[-2ff2 (y_nh2a*2 + U22JS2b*2 + ^j£2£*z 3 + 2U12hka*b* + 2D1^h/a*c* + 2U2j^£b*c*) ] where f° is the tabulated scattering factor and f is that corrected for thermal motion. The isotropic thermal parameters have the form: f = fOexp[-B(sin 9/*) * ] where B is related to the mean-square displacement, U2, of the atom from its mean position by the expression: B = 8*r202 PART 1 CRYSTAL AND MOLECULAR STRUCTURE OF B,B-DIPHEN YLBOROXAZOLIDINE (2-AMINOETHYL DIPHENYLBORINATE) 5 INTRODUCTION In recent years the boroxazclidines have teen extensively studied, the primary concern being the unusual stability of the aminoalcohcl esters with respect to boron esters of ordinary alcohols. B,B-diphenylboroxazclidine Q) was originally prepared by Letsinger and Skoog (7), who correctly proposed the cyclic structure of the ester. The possibility of the N—>B dative bond was first proposed by Brown and Fletcher (8) for triethanolaraineborate {triptych boroxazolidine) in 1951. The first substantial evidence for the existence of the N->B dative bond in boroxazolidines was the result of detailed kinetic studies of the acid hydrolysis of these compounds by Zimmerman and co-workers, the details of which are the subject of a review article by Zimmerman (9). The X-ray crystallographic study cf £,B-diphenylboroxazolidine was undertaken to provide conclusive proof of the existence of the boroxazolidine ring. 1 6 EXPERIMENTAL B,B-Diphenylboroxazolidine was prepared as previously described (7, 10). Recrystallization from 1:1 ethanol-carbon tetrachloride gave colorless needles, elongated along b, with (100) , (001) , and (101) variously developed, The crystal chosen for study was mounted with b parallel to the goniostat axis and was ca. 0.5 mm in length with a cross section of 0.3 x 0.3 mm. Unit-cell and space group data were obtained from film and diffTactometer measurements. The unit-cell parameters were refined by a least-squares treatment of sin2© values for 22 reflexions measured on a diffTactometer with Cu radiation. Crystal data are: cl^Hl6BN0 f,w* = 225*1 Monoclinic, a = 13.840(1), b = 8.9169(5), c = 10.170(1) A, j8 = 98.85 (1)° , V = 1240.1(2) A3. Dm = 1.201(5), Z = 4, D x = 1.2055 (3), F(000) = 480 (20° C, Cu K*, a = 1.5418 A, ^ = 5.9 cm-*). Absent spectra: OkO, k * 2n and h0^, h + £ # 2n define uniquely the space group P2^/n(C"|^, No. 14). Intensities were measured on a Datex-automated General Electric XRD 6 diffTactometer, with a scintillation counter, Cu Kc< radiation (nickel filter and pulse height analyser) , and a G-2€ scan at 2° min-1 over a range of (1.80 + 0.86 tan G) degrees in 29, with 20 s background counts being measured at each end of the scan. Data were measured to 29 = 145° o (minimum interplanar spacing 0.81 A). A check reflexion was monitored every 40 reflexions throughout the data collection. 7 The intensity of the check reflexion remained within ± 2.5% of its initial value during the data collection, the final value being equal to the initial value. Lorentz and polarization corrections were applied, and the structure amplitudes were derived. No absorption correction was made in view of the low value of //.. Of the 1837 independent reflexions measured, 369 had intensities less than 3<r(I) above background where <r2 (I) = S + B + (0.05S)2 with S = scan count and B = time averaged background count. These reflexions were not included in the refinement. Structure Analysis The structure was solved by direct methods. Sixteen sets of signs for 254 reflexions with normalized structure factor |E| > 1.50 were determined by a computer program which uses Sayre relationships in an iterative procedure (11). The starting set of reflexions is given in Table 1. One set of signs was outstanding in that it converged in 5 cycles to a set having the highest consistency index (11) (0.85) with 130 positive signs and 124 negative signs. An E-raap was computed using the 254 signed values of E from this set. The 17 non-hydrogen atoms accounted for the 17 highest peaks on the map. A structure factor calculation based on the positions from the E-map gave R 0.211. Two cycles of full-matrix least-squares refinement of the positions and isotropic temperature factors of the boron, nitrogen, oxygen, and carbon atoms reduced B to 0.151. All 16 hydrogen atoms were then located from a difference Fourier. One cycle with the non-hydrogen 8 Table 1 Basic starting set of reflexions for C^H^BNO h IS A HI 6 1 -8 4. 00-, I 3 1 -1 1 2.25|- origin determining 1 2.20J 0 2 1 6 3 0 3.14 1 2 -4 2. 47 9 3 -1 3.19 10 3 4 3. 21 9 atoms having anisotropic temperature factors and the hydrogen atoms isotropic resulted in R = 0.058. Convergence was reached after two more cycles at R = 0.041 for 1458 reflexions with I > 3<r(I) (10 reflexions were given zero weight in the final stages of refinement due to suspected extinction errors: 0 0 2, 10 -1, 2 0 0, 111, 3 1-1, 3 2 0, 0 2 0, 0 2 1 , 2 2 -1, and 1 3 0). The scattering factors of ref. 12 were used for the boron, nitrogen, oxygen, and carbon atoms and those of ref. 13 for the hydrogen atoms. The weighting scheme: w = 1 if |Fo| < 10; w = (10/JFo|)2 if |p0| > 10, and w = 0,49 for the weak reflexions gave constant average values of w (Fo - Fc)2 over ranges of |Fo|, and was employed in the final stages of refinement. On the final cycle of refinement, no parameter shift was greater than 0.33 standard deviations. The final positional and thermal parameters are given in Tables 2 and 3 respectively. Measured and calculated structure amplitudes are available on request. THERMAL MOTION AND CORRECTION OF MOLECULAR GEOMETRY The thermal motion has been analysed in terms of the rigid-body modes of translation (T) , libration (L) , and screw (S) motion using the computer program MGTLS (14), Four analyses were carried out: the 17 non-hydrogen atoms were considered first, then each of the three rings in the molecule was analysed for rigid-body motion. The analysis of the five-membered ring and attached atoms C (3) and C (9) Table 2 Final positional parameters (fractional x 10* ) with estimated standard deviations in parentheses Atom X 2 z B 7785 (2) 2708 (2) 3710 (2) 0 7244 (D 4803 (D 3574 (1) N 7069 O) 1504 (2) 2904 (2) C(1) 6243 (2) 3698 (3) 3325 (3) C (2) 6182 (2) 2385 (3) 2386 (3) C(3) 8799 (D 2978 (2) 2988 (2) C (4 ) 8782 (2) 2916 (3) 1616 (2) C(5) 9594 (2) 3304 (3) 1032 (3) C (6) 10440 (2) 3758 (3) 1796 (3) C(7) 10487 (2) 3833 (3) 3157 (3) C (8) 9678 (2) 3449 (2) 3730 (2) C (9) 8111 (D 2110 (2) 5216 (2) C (10) 8600 (2) 759 (3) 5526 (3) • C (11) 8829 (2) 247 (3) 6827 (3) C (12) 859 1 (2) 1086 (3) 7858 (3) C (13) 8111 (2) 2418 (4) 7595 (2) C (14) 7876 (2) 2912 (3) 6289 (2) H(D 6010 (18) 3454 (30) 4199 (27) H (1 ') 5853 (20) 4484 (32) 2942 (25) H (2) 6285 (22) 2691 (34) 1518 (31) H (2 ') 5598 (19) 1826 (28) 2317 (23) H (N ') 7276 (16) 955 (27) 2321 (24 ) H (N) 6900 (20) 776 (34) 3565 (28) H (4) 8142 (19) 2619 (28) 1026 125) H (5) 9514 (19) 3250 (30) 80 (29) H (6) 11013 (19) 3993 (28) 1398 (25) H (7) 11087 (21) 4 149 (30) 3742 (25) H (8) 9720 (15) 3520 (26) 4694 (24) H (10) 8819 (18) 226 (30) 4837 (25) H (1 1) 9176 (23) -700 (38) 6942 (29) H (12) 8777 (21) 744 (32) 8783 (30) H (1 3) 7910 (21) 3063 (35) 8278 (31) H (14) 7529 (17) 3834 (27) 6127 (21) 11 Table 3 Final thermal parameters and their estimated standard deviations (a) Anisotropic thermal parameters (U^j x 100 A2) Atom 5ll «22 u33 "12 Ul3 "23 B 5.2 (1) 4.0 (1) 5.7 (1) -0. 1(1) 1. 1 ( 1) -0.6 ( 1) 0 3.9(1) 3.3 (1) 5.6 (1) 0.1 (1) 0.9 ( 1 ) 0. 1 ( 1) N 4.9 (1) 3.9 (1) 5. 6(1) -0. 4(1) 1. 1 ( 1) -'0. 8 ( 1) C(1) 4.3(1) 4.8 (1) 8.3 (2) 0.2 (1 ) 0.8 ( 1) 0. 1 ( 1) C(2) 4.8 (2) 5. 9 (2) 7.8 (2) -0.8(1) -0.5 ( 1) -0.4 ( 1) C(3) 4.3 (1) 3.1(1) 4.5 (1) 0.3 (1) 0.8 ( 1 ) 0. 0 ( 1) C(4) 5.3 (1) 5. 4 (1) 4.8 (1) -0. 1(1) 1.2( 1) -0. 2 ( 1) C (5) 7.3(2) 6.5 (2) 5.2 (2) 0.1 (1 ) 2.5 ( 1 ) 0. 2 ( 1) C{6) 5.7 (2) 5. 6 (2) 8.0 (2) 0.3(1) 3.2 ( 1) 1.2 ( 1) C{7) 4.4 (1) 5.7 (2) 7.7 (2) -0.4 (1) 1. 1 < 1 ) 0.4 ( 1) C(8) 4.7 (1) 5.0 (1) 5.0 (1) -0.2 (1) 0.9 ( 1) -0. 1 ( 1) C (9) 4.0(1) 3.6 (1) 4.9 (1) -0.5 (1 ) 1.0 \ 1 ) 0. 1 ( 1) C (10) 7.3 (2) 4.0 (1) 6.3 (2) 0.4(1) 0.8 ( 1) 0.3 ( D C(11) 7.9(2) 4.7 (2) 8.1 (2) -0.4 (1) -0.8 i 2) 2. 1 ( 1) C(12) 6.7 (2) 8. 1 (2) 5.5 (2) -2.5 (2) -0. 1 ( 1) 1.8 ( 2) C (13) 6.0(2) 8.7 (2) 4.7 (1) -0.9 (1 ) 1.3 | 1) -0. 1 ( 1) C (14) 4.8 (1) 5.5 (2) 5. 1 (1) 0. 1 (1) 1.2 ( 1) -0. 1 ( 1) (b) Isotropic thermal parameters Atom B (Az) Atom B (A2) H(1) 5.9(6) H (6) 5.9 (6) H (1«) 5.7 (6) H(7) 6.0 (6) H (2) 7.1 (8) H (8) 4.6 (5) H (2 •) 5.0 (5) H(10) 5.4 (6) H (W) 6.8 (7) H (11) 7.5 (7) H (N* ) 4. 1 (5) H(12) 6.9(7) H(4) 5.8 (6) H (13) 7.7 (8) H (5) 6. 2 (6) H (14) 4.2 (5) 12 failed to give a positive-definite L tensor. The r.m.s. 4 U^ ^ of 0.0055 A2 for the molecule as a whole is significantly larger than the r.m.s. standard deviation in the thermal parameters UJJ^ (0.0013 A2), indicating that the molecule as a whole is not a good rigid-body. The analyses of the two phenyl groups were successful and the results appear in Table 4. The r.m.s. 4 U^ values cf 0.0019 and 0.0023 A2 for the phenyl groups indicate that the thermal motion of the groups is adequately described by the rigid-body parameters in Table 4, Both groups show nearly isotropic translational motion and anisotropic librational motion. The orientation of the principal axes of L is as expected: the largest oscillations, L^, correspond to rotations about the B-C bonds, the angles between the axes and the bonds being 7.7 (C(3)) and 14.5° (C(9)). The appropriate bond distances and angles in the phenyl groups have been corrected for libration (15,16) using shape parameters (j2 of 0.08 for all atoms. Corrected bond distances appear in Table 5 and both corrected and uncorrected bond angles in Table 6, RESULTS AND DISCUSSION The X-ray analysis confirms the cyclic structure first proposed for this compound by Letsinger and Skoog (7). Figure 1 shows the molecule viewed down the b axis. Individual bond lengths (not corrected for libration) with their standard 13 Table 4 algid-body thermal parameters1 C(3)-C (8) , B C (9) -C (14) , B r 38(6) 6(3) -12 (3) r 18 (2) 2 (2) 14 (4) n L (deg2) I 20 (2) -3(2) | | 17(3) - 12(4) | L 15(3) J L 47(8) J Principal axes of L r. m. s. Amplitude Direction cosines (X103) 6.7° -882 -273 385 7.5° -321 274 -907 4.2 -269 961 65 4.4 -685 -728 23 3.2 -388 -46 -921 2.7 -654 628 421 Principal axes of reduced T r. m. s. Amplitude Direction cosines (X103) 0.21 6 A 522 12 853 0. 22 A -230 126 -965 0.21 832 215 -511 0.19 -851 445 262 0.18 -189 977 102 0. 18 463 887 5 Displacement of axes from intersecting 0 (A) Parallel to L(1) 0.09 0. 69 Parallel to L(2) -0.01 0.2Parallel to L(3) 0.11 -0.03 e Effective screw translations (A) Parallel to L(1) -0.003 0.014 Parallel to L(2) -0.017 -0.025 Parallel to L(3) 0.028 0.003 Fractional coordinates of unique origin (x10*) x 8830 8156 % 3087 2595 2 3030 ^ 5136 Fractional coordinates of centre of gravity (x104) x 9398 8288 2 3286 1734 z 2559 6181 r.m.s. AjJj^ 0.0019 0.0023 A2 lAxes of reference are orthogonal angstrom axes. E.s.d.'s of components of L are given in parentheses in units of the last places shown. Figure 1 The molecule viewed down b, showing crystallographic numbering scheme. H(K) H(ll) Figure 2 A general view of the molecule with bond distances (A) and their standard deviations in parentheses. The C(2)-H(2') distance is 0.94(3) A. 16 deviations are shown in Figure 2 and mean bond lengths are given in Table 5. The five-membered boroxazolidine ring is approximately e in the half-chair conformation, with C (2) displaced 0,08 A e from the BON plane and C(1) lying 0.50 A on the opposite side of the BON plane. The dihedral angles in the ring (see Table 7) are in good agreement with those obtained from minimum energy calculations for UJ^ = 25° (17) , also shown in Table 7. The observed magnitudes of the dihedral angles are slightly smaller than the calculated values since the mean valence angle in the ring, 104.8°, is slightly greater than the calculated value of 104.2°. Angles in the five-membered ring range from 99.7(1) at B to 110.1(2)° at 0. The angular strain in the ring is partially relieved by a shortening of the C(1)-C(2) bond to 1.505 A from the expected value of 1 .537 A (18) for a C (SJ>3)-C (s£3) bond and a lengthening of the B-N O e bond to 1.653 A from the mean value of 1.55 A (18) for B (S£3)-N (S£3) bonds. The isoelectronic N-C (1.485 A) and 0-B (1.484 A) bonds as well as the C-0 bond (1.413 A) are, normal single bonds (18). The two phenyl groups are planar within experimental error (see Table 8). All of the phenyl hydrogen atoms lie in the respective phenyl planes with the exception of H(10) e which lies 0.08 A (3 standard deviations) below the C (9)-C(14) plane. The boron atom deviates significantly from both phenyl planes, being displaced 0.15 from the C(3)-C(8) plane 0 and -0.02 A from the C(9)-C(14) plane representing a slight 17 Table 5 (a) Mean bond lengths (A) , with r.m. s. deviations in parentheses* Atoms number of values uncorrected corrected B-C 2 1.611(2) 1.616(2) B-N 1 1 .653 (3) B-0 1 1.484(3) C-C 1 1.505(4) C-C(ar) 12 1. 383 (1 1) 1.392 (1 1) C-N .1 1 .4 85 (3) C-0 1 1.413(3) C-H 4 0.96 (3) C-H (ar) 10 0.974 (25) N-H 2 0.92 (7) (b) Bond lengths corrected for libration Atoms distance Atoms distance B-C (3) 1. 617 B-C(9) 1.614 C(3)-C(4) 1.404 C (9)-C (10) 1.405 C (4)-C (5) 1.398 C (10) -C ( 11) 1.395 C(5)-C(6) 1.373 C (11)-C (12) 1.381 C(6)-C(7) i 1. 388 C (12) -C (13) 1.377 C(7)-C(8) 1 .386 C (13)-C (14) 1.395 C (8)-C (3) 1. 404 C (14) -C (9) 1.397 •For single value parameters, the least-squares standard deviation is given in parentheses. Table 6 Bond angles (deg) with estimated standard deviations in parentheses (a) Non-hydrogen atoms Atoms nncort. corr. 0-B-C (3) 108.9 (2) 0-B-C (9) 113.7 (2) 0-B-N 99.7 (1) C (3)-B-C (9) 1 14.0 (2) C (3) -B-H 1 12.9 (2) C (9)-B-N 106. 8 (2) B-0-C(1) 110.1 (2) C (2)-N-B 106. 1 (2) 0-C (1) -C (2) 105.2 (2) H-C (2) -C (1) 102.9 (2) B-C (3) -C (4) 124.1 (2) 124.0 B-C (3) -C (8) 120.0(2) 119.9 C(8)-C(3)-C(4) 1 15.6 (2) 1 15.8 C (3)-C (4)-C(4) 121.8 (2) 121.7 C(4)-C(5)-C(6) 120.6 (2) 120.5 C (5)-C (6)-C (7) 119. 3(2) 119.6 C(6)-C(7)-C(8) 119.7 (2) 1 19.6 C (7)-C (8)-C(3) 122.9 (2) 122.8 B-C (9) -C (10) 122.0 (2) 121.8 B-C (9) -C (14) 122.4(2) 122.3 C (10) -C (9) -C (14) 115.6 (2) 115.9 C (9)-C (10) -C (11) 122. 1 (3) 121.9 C (10) -C (11) -C (12) 120.2 (3) 120. 1 C (1 1)-C (12) -C (13) 1 19. 5(3) 1 19.7 C (12) -C (13) -C (14) 119.8 (3) 119.7 C (13)-C (14)-C (9) 122.8 (2) 122.7 continued... 19 (b) Angles involving hydrogen atoms Atoms value Atoms value H (N) -N-H (N •) 104 (2) H (5)-C (5)-C (4) 1 16 (2) H (N) -N-B 107 (2) H (5)-C(5) -C(6) 123 (2) H(N) -N-C(2) 109 (2) H (6)- C(6) -C(5) 121 (2) H (N •)-N-C(2) 1 14 (2) H (6)-C (6)-C (7) 120 (2) H (N 1 ) -N-B 1 17 (2) H (7)-C(7)-C(6) 122 (2) H (1) -C (1) -H (1 •) 107 (2) H (7)-C (7)-C (8) 1 19 (2) H(D -C (1)-0 109 (D H (8)-C(8) -C(3) 1 18 (D H (1) -C(1) -C(2) 113 (2) H (8)-C (8)-C (7) 1 19 (D H (1 •)-C (1 )-0 112 (2) H (10) -C (10)-C (9) 1 18 (2) H (1 »)-C(1)-C(2) 110 (2) H (10) -C (10) -C (1 1) 120 (2) H (2) -C(2) -H (2») 109 (2) H (11) -C (1 1 )-C (10) 1 16 (2) H (2] -C (2) -N 104 (2) H (11) -C (1 1)-C (12) 124 (2) H(2) -C(2)-C(1) 111 (2) H (12) -C (12)-C (11 ) 120 (2) H (2' )-C (2)-N 1 14 (2) H (12) -C (1 2) -C (13) 120 (2) H(2< »)-C(2)-C(1) 1 15 (2) H (13) -C (1 3) -C (12) 124 (2) H (4) -C(4) -C(3) 118 (D H (13) -C (13)-C (14) 1 16 (2) H (4] -C (4)-C (5) 120 (D H (14) -C (14)-C (13) 1 18 (D H <1«*) -C (14)-C (9) 1 19 (D Table 7 Intra-annular torsion angles (deg) Boroxazolidine ring Bond observed calc. B-C -22.0 (2) -25.0 0-C(1) 39.6 (2) 41.6 C(1)-C (2) -39.3 (2) -42. 3 C(2) -H 24.8 (2) 25.9 B-B -3.1 (2) -1.3 21 Table 8 Weighted least-sguares mean planes (a) Distances (A) of relevant atoms from the mean planes Atom d A torn d Plane 1: C(3)-C(8) Plane 2: C (9) -C (14) C(3) 0,000 0.0 C (9) -0,001 0.4 C (4) -0,001 0. 4 C(10) -0.003 1.2 C(5) 0.001 0.5 C (11) 0,007 2.4 C (6) 0.000 0. 1 C(12) -0.004 1.4 C(7) -C.001 0.3 C (13) -0.001 0.5 C{8) 0.00 1 0. 3 C(14) 0.003 1.3 B 0. 150 67.4 B -0.025 10.7 H (4) 0.040 1.6 H (10) -0.080 3.1 H(5) 0.020 0.7 H (11) -0.007 0.2 H (6) -0.032 1.3 H (12) -0.035 1.2 H(7) -0.007 0.3 H (13) 0.006 0.2 H (8) 0.012 0.5 H(14) 0.023 1.0 (b) Equations of planes: £X + wY • nZ = g, where X, Y, and Z are orthogonal angstrom coordinates derived as follows: i-X-, r a 0 ccos T rxn III = I 0 b 0 | in LZJ «- o 0 csin J «-ZJ Plane £ m n £ (1) -0.3074 0.9482 -0.0800 -1.3230 (2) -0.8579 -0.4573 -0.0823 -10.4443 The dihedral angle between plane normals is 100° 22 folding of the two planes away from each other. The angle between phenyl plane normals is 100°. The two rings are not equivalent as the C(3)-C(8) ring is twisted 21° with respect to the BNC(3) plane while the C(9)-C(14) ring lies nearly in the B0C(9) plane, dihedral angle 7°. The C-C (ar) distances range from 1.373 to 1. 405 A with a mean value of 1.392(1 1 ) A, o in good agreement with the accepted mean of 1.394 A (18). There are, however, significant differences between the individual C-C distances in the phenyl rings. There is a noticeable trend toward shortening of the C-C distances as they are removed from the boron substituent. This is due to a combination of steric and electronic effects which are discussed in more detail in Part 2. The borcn-carbcn distances, mean 1.616(2), are significantly shorter than the B-C distances of 1. 63 1 (9) - 1. 646 (8) A found in the tetraphenyl borate anion (19), in accord with electron delocalization. The mean bond angles at tetrahedrally and trigonally coordinated atoms are 109.4 and 120.0° respectively. There are a number of significant deviations from these values, resulting from steric and charge delocalization effects. Intramolecular contacts between atom pairs N and C(4), N and C(10), and 0 and C(14) are responsible for angular distortions at the boron atom, and at carbon atoms C(3) and C(9). Expansion of the angles NBC (3), 0BC(9), BC(3)C(4), BC(9)C{14), and BC(9)C(10) [ 112.9, 113.7, 124.0, 122. 3, and 121.8° respectively] allows the distances C(4)...N (3.143 ), O...C(14) (2.955 ), and N...C(10) (3.210 A) to be equal tc or slightly greater than the sum of van der Waals radii. The 23 expansion of 0BC{9) and NBC (3) causes a contraction of NBC (9) to 106.8° which is balanced by an expansion of BC(9)C(10) (as above) to allow the N...C(10) contact to be normal. The phenyl C-C-C angles at C(3) and C(9) are both contracted to a mean value of 115.9° as a result of expansion of the B-C-C angles. This, in turn, makes angular adjustments at the remaining phenyl carbon atoms necessary to retain the planarity of the phenyl rings. The magnitude of these distortions is also dependent on the electron delocalization in as much as the C-C distances are not all equal. The angle opposite the small OBN angle (99.7°) is opened to 114.0° and is normal for the angle between two bulky substituents. The interior angles in the boroxazolidine ring, as previously mentioned, are all contracted as are the H-N-H and H-C-H angles opposite them, all of which are less than, but not significantly different from the tetrahedral angle. The remaining angles involving the ring hydrogen atoms are generally greater than the tetrahedral angle. The angles H(N»)-N-C(2) (113.9°) and H(N»)-N-B (116.9°) represent bending 'of H(N') toward the oxygen atom to which it is hydrogen bonded. Bond angles involving phenyl hydrogen atoms show a trend that when adjacent C-C distances are different, so are the corresponding H-C-C angles. The H-C-C angle which involves the carbon atom nearer the vertex atom is larger than the other H-C-C angle. As the difference between adjacent C-C distances increases, so does that between the H-C-C angles. An example is C(11), where C(10)-C(1 1) (1. 395) is 0 five standard deviations longer than C(11)-C(12) (1.381 A) OO NO Co B • Ho 5A Figure 3 The structure viewed down c; hydrogen bonds are represented by broken lines. 25 Table 9 (a) Selected intra- and intermolecular contacts Intramolecular Intermolecular* Atoms distance Atoms distance 0. . .C (14) 2.955 (3) N...C (3) 1 3.437 (3) 0...C (8) 3. 395 (3) N...C (4) 1 3.470 (3) 0. . .C (4) 3.299 (2) C(4)...H(N)2 2.72 (3) M. . .C (4) 3.143 (3) C (7)...H (11)3 2.84 (3) N...C (10) 3.201 (3) C(11) . . . H (1 *) * 2.87 (3) C (12) . . .H (14) s 2. 83 (2) C (14) . . . H (7) 6 2.99 (3) (b) Hydrogen-bond 0 data (distances in A and angles in deg) D-H • • • A H. .. A D...A /DHA /XAH N-H (N«) ., .07 2.06 (3) 2. 874 (2) 160 (2) 1 19. 5(6) ,119.6 (6) •Superscripts refer to atoms at positions: 1 3/2-x 1-1/2 1/2-2 5 3/2-x 1-1/2 3/2-z 2 3/2-x 1/2+1 1/2-2 6 2-x 1-1 1-z 3 2-x ~I 1-z 7 3/2-x 1-1/2 1/2 + z * 1/2+x 1/2-1 1/2+z 26 and the angles H-C (11) -C (10) (115.9) ana H-C (11)-C (1 2) (123.8°) differ by more than four standard deviations. The mean C-H, C-H (ar), and N-H bond lengths of 0.96(3), o 0.97(3), and 0.92(7) A are as expected. The distances are shorter than those obtained spectroscopically indicating that the hydrogen electron has been pulled toward the atom to which it is bonded. Figure 3 shows the structure viewed down c. The crystal structure consists of discrete molecules cf B,B-diphenylboroxazolidine which are linked by O...H-N hydrogen o bonds (O...N = 2.874 A) to form continuous spirals about the 2± axes along b. Details of the hydrogen bonding scheme are given in Table 9 as well as inter- and intramolecular contacts less than 3.5 A. There are only two heavy atom intermolecular contacts less than 3.5 A: N...C(3), 3.437, and N-C (4), 3.470 A (apart from the hydrogen bond). These, and all other intermolecular contacts correspond to van der Waals interactions. PART 2 CRYSTAL AND MOLECULAR STRUCTURE OF B-BIS( D -FLUOROPHENYL)BOROXAZOLIDINE 28 I NT RO DU CTION The cyclic structure of B,B-diphenylboroxazolidine Q, Part 1) has been established as has that of triethanclairine borate (TFAB) (20), confirming the existence of the N—>B dative bond in these esters. The X-ray analysis of B,B-bis(£-fluoropheny1)boroxazolidine (2) was undertaken to study the structural effects of the fluorine substituent both in the phenyl rings and in the five-membered ring. The density cf 2 and crystallization in a different space group than J suggested the possibility of an F...H-N hydrogen bond for which there are only limited structural data, particularly for organic structures. Recrystallization of B,B-bis (n-fluorophenyl) boroxazol idine from ethanol gave colorless, regular crystals elongated along b. The specimen used for data collection was bounded by the (011) and (101) planes, at distances of 0.27 and 0.13 mm 1 2 EXPERIMENTAL 29 respectively from an internal origin and was mounted with b parallel to the goniostat axis. unit-cell and space group data were obtained from film and diffTactometer measurements. The unit-cell parameters were refined by a least-squares treatment of sin29 values for 30 reflexions measured on a diffTactometer with Cu radiation. Crystal data are: C1i4.H1/4.BF2 NO f.w. = 261.1 Orthorhombic, a = 13.442(4), b = 10.214(3), c = 9.283 (2) A, V = 1274.5 (6) A3, Dm = 1.37 (flotation in aqueous KI) , Z = 4, Dx = 1.361 g cm-3, F(000) = 544 (20°C, Cu 1^, A= 1.5418 A,/< = 9.0 cm-*). Absent reflexions: hOO, h # 2n, OkO, k * 2n, and 00j£, J * 2n define uniquely the space group B2^2^2^ (C|, No. 19) . Intensities were measured on a Datex-automated General Electric XRD 6 diffTactometer, with a scintillation counter, Cu radiation (nickel filter and pulse height analyser), and a 6-20 scan at 2° min-1 over a range of (1,80 + 0.86 tan 0) degrees in 29, with 20 s background counts being measured at each end of the scan. Data were measured to 29 = 145° (minimum interpianar spacing 0.81 A). The r.m.s. deviation of the intensity of the check reflexion, measured every 40 reflexions throughout the data collection, from its initial value was 1.4%. The final intensity was 99% of the initial value. Lorentz, polarization, and absorption corrections were applied, and the structure amplitudes were derived. Of 1481 independent reflexions measured, 231 had intensities less than 3<r(I) above background where <r2 (I) = S + E + (0.03S)2 30 with S = scan count and B = background count, corrected to time of scan. These reflexions were not included in the refinement. Structure Analysis The structure was solved by direct methods, 200 reflexions with normalized structure factor |E| > 1.45 being used in the symbolic addition procedure for non-centrosymmetrie crystals (21). The phases of the 11 1 0, 2 0 5, and 6 7 0 reflexions were fixed to define the origin and that of 10 3 was fixed at +250mc to specify the enantiomorph. During a manual expansion, carried out among the 70 reflexions with largest |E| values, symbol phases were assigned to the 0 5 1, 10 10 1, and 13 3 6 reflexions. The phase of 0 5 1 must be ±250 mc and manual indications gave two possible values for each of the other symbols, near ±250 mc for both 10 10 1 and 13 3 6. These seven reflexions comprise the basic starting group given in Table 10. Eight starting sets were generated by allowing each of the three symbol phases to have initial values of ±250 mc. These sets were used as input to a computer program which determines phases using the tangent formula (22,23). The values of overall t overall*, Q, and Rk on the final cycle for each of the sets are given in Table 11. Set 1, which had the lowest value of Rk, was expanded to 228 reflexions with |EJ > 1.40 by using as starting values for the symbols a, b, and c those calculated in set 1, +250, +277, and +135 mc \ Table 10 Basic starting set of reflexions for Cj^H^BF^ NO h k 111 phase (mc) 11 1 0 4.58 250T 2 0 5 3.71 j 250F • origin determining 6 7 0 3. 51 I OJ 1 0 3 2.43 250 enantiomorph 0 5 1 2.21 a 10 10 1 2.27 b 13 3 6 2.26 c 32 Table 11 Results for the eight starting sets in the phase determination procedure Set a (mc) b (mc) c (mc) t 2 JiS 1 1 250 250 250 0. 59 180 0. 39 0.2 0 193 2 250 250 -250 0.54 129 0.44 0.35 178 3 250 -250 250 0. 58 156 0. 40 0. 35 178 4 250 -250 -250 0.55 155 0.43 0.33 177 5 -250 250 250 0. 53 122 0. 46 0.38 172 6 -250 250 -250 0.54 141 0.45 0. 33 173 7 -250 -250 250 0. 55 138 0. 44 0.35 176 8 -250 -250 -250 0.57 155 0.41 0.33 173 33 respectively, A new symbol, 10 6 0, was allowed to take either of its two possible values, 0 or 500 mc. The two resulting values of 8k were 0.17 with 10 6 0 having a phase of 500 mc and 0.36 with 10 6 0 at 0 mc. An E -map based on the set of 214 determined phases with Bk = 0.17 clearly gave the structure, the 19 highest peaks corresponding to the 19 non-hydrogen atoms. Two cycles of full-matrix least-squares refinement of the positional and isotropic thermal parameters of the non-hydrogen atoms gave B 0.121. This was followed by two cycles of anisotropic refinement which reduced B to 0.093. A difference map at this point revealed the positions of all 14 hydrogen atoms which were included in all subsequent refinement with isotropic thermal parameters. Convergence was reached after two more cycles with R = 0.047 for 1234 reflexions with I > 3<r(I) (16 reflexions were given zero weight in the final stages of refinement due to suspected extinction errors). The mirror image (j to -yj was also refined to convergence giving an R value of 0.047. Application of Hamilton's test (24) did not show a significant difference between the R factors for the two enantiomorphs, The scattering factors for the F, 0, N, C, and E atoms were taken from ref. 12 and those for hydrogen from ref. 13. The values used for anomalous dispersion corrections 4f1 and 4f" were as follows: 0.068 and 0.056 for F, 0.050 and 0.032 for 0, 0.034 and 0.019 for W, and 0.020 and 0.010 for C. The 34 values for C and 0 are those of Hope, de la Camp, and Thiessen (25). The weighting scheme: w = 1 if |Fo| < 7 ; w = (7/|Fo|)2 if | Fo | > 7 ; and w = 0.2025 for the weak reflexions gave constant average values of w (Fo - Fc)2 over ranges of |Fo|, and was employed in the final stages of refinement. On the final cycle of refinement, no parameter shift was greater than 0.19 standard deviations. Final positional and thermal parameters appear in tables 12 and 13 respectively. Observed and calculated structure amplitudes are available on request. THERMAL MOTION AND CORRECTION OF MOLECULAR GEOMETRY The ellipsoids of thermal motion for the non-hydrogen atoms are shown in figure 4. The thermal motion has been analysed in terms of the rigid-body modes as previously described. Four analyses were carried out: the 19 non-hydrogen atoms were considered first and indications of significant independent motion in the phenyl rings prompted separate analyses of the fluorophenyl groups along with the boron atom; finally an analysis of the five-membered ring and atoms C (3) and C(9) failed to give a positive-definite L tensor (as for the parent molecule B,B-diphenylboroxazolidine in Part 1). The results of the analyses for the two fluorophenyl groups are compiled in table 14. The r.m.s. standard deviation in the temperature factors o U^^ is 0.0016 A2 which indicates that the entire molecule (r.m.s. AU** = 0.0063 A2) is not a good rigid-body (this was Table 12 Final positional parameters (fractional x 10* ) with estimated standard deviations in parentheses Atom X I z F(1) 11462 (2) 6938 (2) 3863 (3) F(2) 8699 (2) -1240 (2) 7146 <3) 0 7244 (D 4697 (2) 5901 1 (2) N 7190 (2) 3839 (3) 3503 (3) C(1) 6247 (3) 4345 (4) 5584 (4) C(2) 6182 (3) 4267 (5) 3979 (5) C (3) 8903 (2) 4762 (3) 4603 (3) C(4) 9786 (2) 4182 (3) 4154 <4) C (5) 10644 (2) 4892 (4) 3886 (4) C(6) 10613 (2) 6221 (3) 4063 (3) C(7) 9770 (3) 6852 (3) 4498 (5) C(8) 8921 (2) 6120 (3) 4749 (4) C (9) 8102 (2) 2467 (3) 5523 (3) C(10) 8091 (3) 2218 (3) 7007 (4) C(11) 8297 (3) 984 (4) 7550 (4) C(12) 8501 (2) -17 (3) 6622 (4) C (13) 8502 (3) 157 (4) 5167 (4) C(14) 83 01 (3) 1411 (3) 4648 <4) B 7903 (2) 3942 (3) 4949 (3) H(N 1) 7142 (27) 3036 (38) 3166 <4V) H (M2) 7413 (30) 4315 (38) 2642 (50) H(1A) 5361 (34) 5032 (47) 5882 (50) H (1B) 6082 (29) 3462 (44) 5906 (43) H (2A) 5673 (43) 3722 (56) 3625 (65) H (2B) 6132 (30) 5143 (44) 3539 (47) H(4) 9773 (28) 3228 (38) 3943 (41 ) H (5) 11246 (38) 4516 (51) 3733 (55) H(7) 9769 (38) 7868 (51) 4574 (52) H (8) 8361 (27) 6565 (36) 5040 (42) H(10) 7962 (33) 2925 (40) 7618 448) H (11) 8219 (46) 736 (58) 8513 (71) H(13) 8656 (3 8) -615 (52) 4683 (53) H (14) 8267 (30) 1454 (40) 3569 (47) 36 Table 13 Final thermal parameters and their estimated standard deviations (a) Anisotropic thermal parameters (U.* * x 100 A2) Atom ^22 o33 2l2 2i3 «23 F(1) 4.8 (1) 7.7 (2) 7.0 (1) -2.0 (1) 0.8(1) -0. 1 ( 1) F (2) 8.7 (2) 4.9 (1) 9.6 (2) 0.3 (1) -1.1 (2) 3.0 ( D 0 4.1 (1) 4.3 (1) 5.0 (1) -0. 1(1) 0.6(1) - 1.0 { 1) N 4.6(1) 3.4 (1) 4.2 (1) -0.0 (1 ) -0.4 (1) 0.4 ( 1) C(1) 4.0 (2) 5.9 (2) 6.8 (2) -0. 2 (2) 0.8(2) - 1. 1 ( 2) C(2) 3,7(2) 7.8 (3) 6.4 (2) 0.1 (2) -0.4 (2) 0. 3 ( 2) C (3) 3,7 (1) 3.8 (1) 3.7 (1) 0.2(1) 0.1 (1) -0.0 ( 1) C(4) 4.2(2) 4.3 (2) 5.9 (2) 0.5 (1 ) 0.1 (1) -0. 6 i 2) C{5) 3.6 (2) 6. 0 (2) 5.8 (2) 0.6 (2) 0.5(2) -0.4 ( 2) C(6) 4.1 (2) 5.<* (2) 4.2 (2) -0.8 (1 ) 0.3 (1 ) 0. 2 ( 2) C (7) 5.6 (2) 3. 6 (2) 7.7 (2) -0.3(1) 0.9 (2) 0.4 ( 2) C{8) 4.3(2) 4.2 (2) 6.1 (2) 0.3 (1 ) 0.7 (2) -0,0 ( 2) C (9) 3.6 (1) 3.8 (1) 4.0 (2) -0.4 (1) 0. 1(1) 0.4 ( D, C(10) 7.8 (2) 4.7 (2) 3.3 (2) -1.3 (2) -0.8 (2 ) 0. 2 ( 1) C (1 1) 8,2 (3) 6. 3 (2) 4.3 (2) -1.4(2) -1.5(2) 1.8 ( 2) C(12) 4.3(2) 4.1 (2) 6.8 (2) -0.3 (1) -1.0 (2) 1.9 ( 2) C (13) 5.9 (2) 4.5 (2) 6.3(2) 1.4(2) 0.4(2) 0.3 ( 2) C(14) 6.6(2) 4.8 (2) 4.4 (2) 1.2 (2) 0.6 (2) 0.4 | D B 3.8 (2) 3. 8 (2) 3.6(1) -0,1 (1) 0.2(1) -0.3 ( 1) (b) Isotropic thermal parameters Atom B (A2) Atom B (Az) H (N1) 3.5 (7) H (5) 6, 1 (11 ) H (M2) 4.8(9) H(7) 6.6(12) H(1A) 5.1 (10) H (8) 3,4 (7) H(1B) 4.1(8) H(10) 5.2(9) H(2A) 7.2(13) H(11) 8.6(15) H(2B) 4.7(9) H(13) 5.9(11) H(4) 3.9(7) H(14) 4.9(9) 37 Table 14 Rigid-body thermal parameters1 F(1), C(3)-C(8), B F(2), C(9)-C(14), B r 53 (6) 22 (3) -11 (3) -, r 14(3) -14(6) 4(3) L(deg2) | 17(4) -3(2) | | 78(10)-20 (5) | «- ' 13(2) J >• 18(5) J Principal axes of L r.m.s. Amplitude Direction cosines (x103) 8.1° 887 413 -206 9. 3° 198 -939 283 3.3 37 377 925 3.4 543 -136 -829 2.3 460 -828 321 3.2 816 317 482 Principal axes of reduced T r.m.s. Amplitude Direction cosines (x103) 0.20 A -738 -661 132 0.20 A -267 888 -373 0.18 -166 372 914 0.18 -232 316 920 0.18 -656 650 -383 0.16 935 332 122 Displacement of axes from intersecting (A) Parallel to 0.37 0. 89 Parallel to L2 -0.08 0.68 Parallel to 0.26 0.09 o Effective screw translations (A) Parallel to L1 0.015 -0.032 Parallel to L2 -0.008 0.066 Parallel to -0.041 0.023 Fractional coordinates of unique origin (x104) x 8888 7840 J 4804 2548 z 4668 5470 Fractional coordinates of centre of gravity (x104) x 9889 8332 2 5606 1038 z 4304 6163 r.m.s. 0.0023 0.0036 A2 lkxes of reference are orthogonal angstrom axes. E.s.d.'s of components of L are given in parentheses in units of the last places shown. 38 also noted in Part 1). Examination of the individual A ^ shows significant independent motion of the phenyl groups and also of atoms in the five-membered ring. The r.m.s. A U^^ o values of 0.0023 and 0.0036 A2 for the analyses of the phenyl groups indicate that these groups do behave as rigid bodies. Both groups show nearly isotropic translationa1 motion and anisotropic librational motion. The principal axes of L are oriented as expected: the largest axes, , correspond to rotations about the B-C bonds, the angles between the axes and the bonds being 6.9° (C (3) ) and 2.6° (C (9) ) . The unique origins are in expected locations, for the C(3)-C(8) ring approximately at C (3) , and for the C(9)-C(14) ring near C (9) . The appropriate bond distances and angles in the phenyl groups have been corrected for libration (15,16) using shape parameters cj2 of 0.08 for all atoms. Biding motion corrections based on the <A^ (26,27) have been applied to the C-F bonds. Both corrected and uncorrected bond lengths and angles appear in Tables 15 and 16. RESULTS AND DISCUSSION Figure 4 shows a general view of the molecule and the crystallographic numbering scheme. Figures 5 and 6 show the packing arrangement viewed along b and c respectively. Intra-annular torsion angles defining the conformation of the boroxazolidine ring are given in Table 17 and some weighted least-squares mean planes in Table 18. Non-bonded intra- and intermolecular distances and details of the hydrogen-bonding 19 n • S) Figure 4 A stereo view of the molecule showing 50'J? probability thermal motion ellipsoids for the non-hydrogen atoms and the crystallographic numbering of the atoms. scheme appear in table 19. The crystal structure consists of discrete molecules of B,B-bis(£-fluorophenyl)boroxazolidine, each linked to six others by an extensive and interesting network of hydrogen bonds. 0...H-N (0... N = 2.941(3) A) and weak F...H-N (F...N = 0 3.171(4) A) hydrogen bonds form continuous spirals about alternate twofold screw axes along c, thereby forming 'nets' of molecules normal to the a axis. The hydrogen bonding • scheme is completed by a weak F...H-C (F...C = 3.318(5) A) interaction which forms spirals about alternate twofcld screw axes along b, linking adjacent 'nets' to form a three-dimensional network which employs all available acceptors in the molecule. 40 Table 15 o Bond lengths (A) with estimated standard deviations in parentheses (a) Non-hydrogen atoms Atoms uncorr. corr. Atoms uncorr. corr. F(1)-C(6) 1. 369 (4) 1. 372 C(3)- C (8) 1. 394 (5) 1. 407 F (2)-C (12) 1. 367 (4) 1. 370 C(4)- C(5) 1. 384(5) 1. 3 87 0-C(1) 1. 418 (4) - — C(5)- C(6) 1. 369 (5) 1. 38 1 O-B 1. 471 (4) -— C (6)-C (7) 1. 365 (5) 1. 374 B-B 1. 652 (4) - — C(7)- C(8) 1. 384 (5) 1. 3 87 C(3)-B 1. 616(4) 1. 619 C (9)-C (10) 1. 401 (4) 1. 416 C (9) -B 1. 620(5) 1. 623 C(9)- C (14) 1. 377 (5) 1. 388 N-C (2) 1. 491 (4) - — C (10) -C (11) 1. 385 (6) 1, 390 C (1)-C (2) 1. 494 (6) - — C(11) -C(12) 1. 366 (6) 1 . 378 C(3)-C(4) 1. 391 (5) 1. 399 C(12) ~C(13) 1. 362 (6) 1. 377 C(13) -C (14) 1. 395 (5) 1. 400 (b) Bonds involving hydrogen atoms Atoms distance Atoms distance N-H (N1) 0. 88 (4) C(5)- H(5) 0.91 (5) N-H (N2) 0. 98 (4) C (7)-H (7) 1.04 (5) C (1) -H (1A) 0. 92 (5) C(8)- H(8) 0.92 (4) C(1)-fl(1B) 0. 98 (4) C (10) -H (10) 0.93 (U ) C (2) -H (2A) 0. 94 (6) C (1 1) -H(1 1) - 0.94(6) C(2)-fl(2B) 0. 99 (4) C (13) TH (13) 0.93 (5) C (4)-H (4) 0. 99 (4) C(14) -H(14) 1.00 (4) Table 16 Bond angles (deg) with estimated standard deviations in parentheses (a) Non-hydrogen atoms Atoms uncorr. corr. C (1) -0-B 108. 2(2) C (2) -N-B 105.5 (2) 0- B- N 99. 9(2) 0- B-C (3) 110.4 (2) 0-B-C (9) 112.9 (2) N-B-C (3) 110,7 (2) N-B-C (9) 107.7 (2) C(3) -B-C (9) 114.2 (2) 0-C (1) -C (2) 106.0 (3) N-C(2)-C (1) 105.0 (3) B-C (3)-C (4) 123. 3(3) 123. 1 B-C { 3) -C (8) 120.7 (3) 120.6 C (4)-C (3)-C(8) 1 16.0 (3) 116.3 C(3)-C(4)-C(5) 122.8 (3) 122.6 C(4)-C (5)-C{6) 118. 2(3) 118.1 F(1)-C(6)-C(5) 119.3 (3) 119.2 F (1)-C (6)-C(7) 118.7 (3) 118.5 C(5)-C(6)-C(7) 121.9 (3) 122.2 C(6)-C (7)-C(8) 118.7(3) 118.5 C(3)-C(8)-C(7) 122.3 (3) 122.2 B-C (9)-C (10) 119. 4(3) 119.2 B-C (9) -C (14) 124.5 (3) 124.3 C (10)-C (9) -C (14) 116. 1 (3) 116.5 C(9) -C (10)-C (11) 121.4 (3) 121.2 C (10)-C (11)-C(12) 119.5 (3) 1 19.3 F(2)-C(12)-C(11) 120.0 (3) 119.7 F (2) -C (12) -C (13) 118.2(3) 118. 1 C(11)-C(12)-C(13) 121.8 (3) 122. 2 C (12)-C (13) -C (14) 1 17.6 (4) 1 17.4 C(9) -C (14) -C (13) 123.6 (3) 123,4 continued... 42 (b) Angles involving hydrogen atoms Atoms value Atoms value B-N-H(N1) 113 (2) C (3)-C (4)-H (4) 1 17 (2) B-N-H (N2) 1 17 (2) c (5)- C(4)-H(4) 1 19 (2) C(2)-N-H(N1) 108 (2) c (4)- C(5) -H(5) 123 (3) C (2)-N-H (N2) 112 (2) c (6)-C (5)-H (5) 1 18 (3) H (N1)-N-H (N2) 101 (3) c (6)-C(7)-H(7) 119 (3) 0-C (1) -H (1A) 106 (3) c (8)- C (7)-H (7) 122 (3) 0-C (1)-H (IB) 113 (2) c (3)- C(8) -H(8) 120 (2) C(2)-C(1)-H(1A) 108 (3) c (7)-C (8)-H (8) 1 17 (2) C (2) -C (1)-H (1B) 104 (2) c (9)-C (10)-H (10) 1 17 (3) H (1A)-C (1)-H (1B) 119 <«») c (11) -C (10) -H (10) * 121 (3) N-C (2) -H (2A) 1 13 (3) c (10) -C (1 1 )-H (1 1 ) 125 (4) N-C (2)-H (2B) 102 (2) c (12) -C (11)-H (1 1) 115 (4) C(1)-C(2) -H (2A) 115 <«») c (12) -C (13)-H (13) 1 12 (3) C (1)-C (2)-H (2B) 112 (3) c (14) -C (1 3) -H (1 3) 131 (3) H(2A)-C(2)-H(2B) 110 (4) c (9)- C(14) -H (14) 123 (2) c (13) -C (14) -H (14) 113 (2) Table 17 Intra-annular torsion angles (deg) in the boroxazolidine ring Bond observed a b calc, B-0 -22.2(2) 32.9 (3) -37.7 0-C (1) 39.6 (2) 42. 1 (3) 43, 8 C(1)-C{2) -39.3(2) 31.2 (3) -33.3 C (2) -N 24.8 (2) 10. 2(3) 10.0 H-B -3.1 (2) 12.7 (3) 17.2 B,B-diphenylboroxazclidine, Part 1 This work Figure 5 The structure viewed along b, O...H-N and F...H-N hydrogen bonds are represented by broken lines. Figure 6 The structure viewed into c, broken lines represent hydrogen bonds. 45 The geometrical data for all three hydrogen bonds are quite reasonable, the angles at the hydrogen and at the acceptor atoms are within the expected limits (28). The 0...N distance is near the accepted mean while the F...N distance of 3.171 is longer than the mean value of 2.92(11) A (29) and probably represents a relatively weak interaction. It should" be noted, however, that the mean N...F distance is based on only 10 examples, most (if not all) of which occur in inorganic structures in which the interaction is highly ionic in nature. In the present case the N and F atoms carry only small partial charges and evidence indicates that the nitrogen atom in this structure probably carries a net negative charge. These factors are probably responsible for the long N...F distance. The F.,.H distances in the F...H-N and F...H-C a interactions are 2.35(4) and 2.34(4) A, both of which are significantly less than the sum of van der Waals radii. The geometry of the F...H-C system is more nearly ideal than that of the F...H-N hydrogen bond and with the F...H distances equal there is little doubt that the F...H-C interaction is a weak hydrogen bond. Aside from the hydrogen bonds there is only one other intermolecular contact which is significantly less than the sum of van der Waals radii, H (7) ..,H (13) , 2.16(7) A. All other intermolecular contacts, the shortest of which are listed in Table 19, correspond to normal van der Waals interactions. 46 The conformation of the five-membered boroxazolidine ring is different from that in _1 as can be seen by comparison of the corresponding dihedral angles in Table 17. The two carbon atoms were on opposite sides of the NBO plane in J while in the present structure both C(1) and C(2) lie on the 0 same side of the NBO plane, displaced -0.73 and -0.32 A from the plane. The observed dihedral angles in the ring are in good agreement with those obtained from energy minimization calculations for 10\ - 10° (17), also shown in Table 17. The observed magnitudes of the dihedral angles are slightly smaller than the calculated values since the mean valence angle in the ring, 104.9°, is slightly greater than the calculated value of 104.2°, but in good agreement with the mean of 104.8° in J_« The individual values range from 99.7(2) at B to 108.2(2)° at 0. There are small but significant differences between the angles at 0, N, an'd C(1) in this structure and in 1 which are a result of conformational and electronic differences. The angular strain inherent in the five-membered ring is, as in \, partially relieved by a significant shortening of the C(1)-C(2) bond (1.494(6) A) from the value of 1.537 A expected for a C (sp.3)-C (S£3) single e bond. The B-N distance of 1.652 (4) A agrees well with chemically similar bonds: 1 .653 in 1 , 1.647 in triethanolamine borate (20), and 1.638 A in (Et2NBF2)2 (30). The two phenyl rings are planar within experimental error (see Table 18). The eight phenyl hydrogen atoms lie in the respective mean planes while the boron and fluorine atoms 47 Table 18 Weighted least-squares mean planes (a) Distances (A) of relevant atoms from the mean planes Atom d <V<r A torn d <V<r Plane 1: C(3)-C(8) Plane 2: C (9)-C (14) C(3) -0.005 1.6 C (9) -0.007 2.4 C (4) 0.003 0.8 C(10) 0.010 2.3 C(5) 0.000 0.1 C (11) 0.000 0.0 C (6) 0.000 0.0 C(12) -0.007 2.1 C(7) -0.005 1.1 C (13) 0.007 1.6 C (8) 0.008 2.0 C(14) 0.004 1.0 B -0.031 9.5 B -0.059 18. 4 F(1) -0.059 24. 1 F(2) -0.008 3.1 H(4) 0.091 2.4 H (10) -0.004 0. 1 H (5) -0.128 2.5 H(11) 0. 123 2.0 H(7) .0.039 0.8 H (13) -0.003 0. 1 H{8) 0.005 0. 1 H(14) 0.079 2.0 (b) Equations of planes: Vk + mY + nZ = jo, where X, Y, and Z are orthogonal angstrom coordinates derived as follows: I-XT r a 0 0 T rXi III = I 0 b 0 I IYl t-ZJ L 0 0 c J «-ZJ Plane m n 1 -0.2755 0.1068 -0.9554 -6.8546 2 -0.9743 -0.2217 -0.0395 - 1 1.3654 The dihedral angle between plane normals is 74° 48 i are significantly displaced from the planes, F(1) and B by -0.06 and -0.03 A from the C(3)-C(8) plane and F(2) and E by -0.01 and -0.06 A from the C(9)-C(14) plane. This is probably a result of intra- and intermolecular steric forces. The dihedral angle between the plane normals is 74° compared to 100° in J. The two fluorophenyl groups are not equivalent, the rings being unequally rotated about the E-C tends. The dihedral angles C (8) [ C (3)-B ]0 and C (10) [ C (9)-B ]0 are 21.7(3) and 30.2(3)° compared to values of 78.0(2) and 7.0(2)° in J.. The difference in the orientation of the phenyl groups in the two structures is a result of packing considerations, among which the C(14)-H (14)...F (1) hydrogen bond may be an important factor. The corrected C-C bond lengths in the phenyl groups range from 1.374 to 1.416 A with a mean value of 1.390 A. There is a significant variation in the individual bond distances, the bond lengths decreasing as they are removed from the boron substituent. The mean values for the three 0 groups are 1.403, 1.391, and 1.378 A, similar to the corresponding values for 1 (corrected for libration) of 1.403, 1.394, and 1.380 A. The B-C distances, mean 1.621, are c slightly longer than the value of 1.616 A in 1 and shorter than in the tetraphenyl borate anion (1 .631-1.648(8) A) (19). The angles in the phenyl rings have a mean value of 120° but the individual values, ranging from 116.3 to 123.4°, show significant deviations'from 120°. The mean angle at the carbon atom carrying the boron substituent is 116.4° and the 49 other mean values are 122.4, 118.3, and 122.2° for atoms 2£tho , meta , and £ara to the boron group respectively. These variations have been explained in terms of the electronegativities of the substituent groups (31). The angles at C(6) and C(12) carrying the fluorine atoms, mean 122.2°, are as expected for an electron withdrawing group. The angles at C(3) and C (9), mean 116.4°, carrying the boron substituent are indicative that this group is releasing electron density into the aromatic system. The distribution of bond lengths is in agreement with this observation, indicating small residual positive charge at atoms ortho to F and negative charge ortho to B, the overall donating and withdrawing effects of the para substituents cancelling each other to result in electronic neutrality of the aromatic TT systems. There is both theoretical and physical evidence which indicates that in spite of the N-»B dative bond, the boron atom remains more positively charged than the nitrogen atom as a result of charge redistributions occurring in the remainder of the molecule (30,32). This offers an explanation for the small differences between the B-C, B-0, C-N, C-0, and C(1)-C(2) bond lengths in this structure and those in 1_, where the negative charges which occur at the fluorine atoms in this structure result in delocalization effects in the o molecule. The mean C-F distance of 1.371 A is close to that of 1.368 A in o-fluorobenzoic acid (33) but significantly longer than the mean C(ar)-F distance of 1. 328 A in ref. 18. The mean bond angles at tetrahedrally and trigonally coordinated atoms are 109.4 and 119.9°. There are a number of 50 Table 19 (a) Selected intra- and intermolecular contacts Intramolecular Intermolecular* atoms distance atoms distance 0.. ,C(8) 2. 888 (4) F (1) . ..F (2) i 3.229 (4) 0. . .C (10) 2.961 (4) F(1) . . .C (1 1) z 3.484 (5) N • • . C (14) 3.084 (4) F (1) . ..C(12) * 3. 400 (4) N...C (8) 3. 489 (4) F (1) . 3. 446 (4) F (2) . ..C(7) * 3.451 (5) C (6) . ..C (11) 2 3.477 (5) C (1) . . ,H(N2) 5 2.96 (4) C (5) . . . H (1 1 ) 2 2.98 (7) C (6) . ..H(1 1) 2 2.79 (7) C (6) . . . H (14) 3 2.88 (4) C (7) . . . H ( 1 3) 6 2.99 (5) C (13) ...H (11) * 2.92(6) C(13) ...H(7) * 2.94 (5) H (7). . . H (13) 6 2. 16 (7) (b) Hydrogen-bond data (distances in 0 A and angles in deg) D-H.. .A H. .. A D. . . A ^DHA ^X AH N-H (N2) .. .08 1.96 (5) 2.941 (3) 176 (4) 122 (1 ) ,129 (1) N-H (N1) .. . F (2) 7 2.35 (4) 3. 171 (4) 155(3) 140(1 ) C(14)-H(14) ...F (1) 9 2.34 (4) 3.318 (5) 165 (3) 99 ( 1) •Superscripts refer to atoms at positions: 1 1/2-x 1/2-2 1-2 6 X 1+y. 2 2 2-x 1/2+1 3/2-2 7 3/2-x -2 z- 1/2 3 2-x V2+I 1/2-2 e 3/2-x 1-2 2-1/2 * X 1-1 2 9 2-2 2-1/2 1/2-z S 3/2-x 1-1 1/2 + 2 51 significant deviations from the mean values resulting from steric and electronic effects. Interior angles in the rings have already been discussed. Asymmetry of the packing appears to be responsible for significant differences between corresponding angle pairs 0-B-C, N-B-C, and C-C-F. The C (3)-B-C(9) angle is equal to that in J to within experimental error. The mean C(sj>3)-H, C(ar)-H, and N-H distances of 0.95, 0.96, and 0.93 A are as expected for X-ray data. The bond angles involving the hydrogen atoms are generally as expected. There are significant differences between the C-C-H angles at C(13) and C(14) which are probably a result of van der Waals contacts F(1)...H(14) and H(7)...H(13) (see Table 19) . PART 3 CRYSTAL AND MOLECULAR STRUCTURE OF 4,4-DIMETHYL-2,2-DIPHENYL-1,3-DIOXA-4-AZONIA-2-BORANATACYCLOPENTANE 53 INTRODUCTION From the reaction of N-hydroxydialkylamine Q) and formaldehyde an addition can be expected either at the nitrogen or at the oxygen atom to give 2 or 4. The addition products, originally regarded as N-hydrcxymethyl-oxydialkylamines (4) by Zinner and Ritter (34,35), react with diphenylboron-supplying reagents (Pl^B-X) to yield crystalline compounds which were initially assigned the structure 5 containing intramolecular N—»B coordination. This assignment was based on the earlier studies of the •boroxazolidines• by Weidmann and Zimmerman (36-38) and was subsequently employed for compounds of this type (39-43). R ^OH 11 0© 0=CH, \ 0 0—'H R RX '0 I Ph^B-X - HX + Ph^B-X -HX -Ph Ph R- CH, R = CH3 Ph-- C4HS X = a) 0-BPh2 b) Ph c) 0-CH4-CHt-NH2 d) 0-CH2-CH2-N(CH3)t R. R' P,h 7Ph NcHrCHz R\@/CH* C^ /N 0 R V ©/ 0—B—Ph / 7 Ph There is, however, some evidence which indicates that the alternate structures 2 and 3 are probably favored over 54 the originally proposed structures 4 and 5: 1. The alkylation of N-hydroxydialkylamines normally leads to tertiary amine oxides. 2. N-Oxides show stronger basicity and possess better nucleophilic or donor gualities than the isomeric N-alkyloxyamines (44,45). & hydrogen bridge chelate of the type 2 should therefore be more stable than 4. 3. If both forms 2 and 4 existed, possibly in a state of equilibrium, the reaction with an electrophilic reagent such as Ph2B-X should shift the (hypothetical) equilibrium to the side of the better donor molecule, i.e. the N-oxide (2). 4. The N-oxide form not only facilitates the approach of the Lewis acid Ph2B-X to the donor (oxygen) atom but also results in a sterically favored chelate structure (3) . 5. Ethanolaroine esters of diphenylborinic acid are intramolecular N—»B coordinated cyclic complexes (6), as recently proved conclusively for Ph2B-0-CH2CH2NH2 (Part 1) and (£-FC6H^) 2B-0-CH2CH2NH2 (Part 2). Despite the stability of these *boroxazolidines' (9,36-38) the rechelation of the applied examples (6, E = H, CH^) was possible with the formaldehyde adduct. This also 55 supports structures 2 and 3 since it is not very plausible that a weakly basic N-hydroxymethyl-ox ydialkylamine HO-CH2ONR2 {4) in an equimolar quantity can displace the isosteric but more basic aminoalcohol H0-CH2CH2Ni<2-6, Finally there exists an analogy between 3 and the diphenyl boron chelates (7) of N-(2-hydroxyalkyl)-dialkylamine-N-oxides and other similar cyclic boron-nitrogen-betaines (46-50) which are closely related to 3 both in means of preparation and in their chemical and physical behavior. These considerations and also the chemical and physical data obtained to date are consistent with the betaine-type chelate 3, but do net provide unambiguous proof. To this end the full X-ray crystallographic study of the homologue with R = CH^ has been carried out. EXPERIMENTAL HJ.1Z ge th Y 112L 2- d iphgny_ 1- 1 F_ 3 - dio xa-A solution of N-hydroxydimethylamine (5 mmole) in 5 ml of ethanol was mixed with an aqueous solution of formaldehyde (40%, 5 mmole). After addition of: a) 2.5 mmole oxybisdiphenylborane cr b) 5.0 mmole triphenylborane or 56 c) 5.0 mmole B-(2-aminoethyloxy)diphenylborane or d) 5.0 mmole B-(2-dimethylaminoethyloxy)diphenylborane the mixture was heated until initial boiling and then allowed to cool. During the cooling or after the dissolution of the boron component the precipitation began. Yields: a) 99%, b) 90%, c) 98%, d) 85% m.p. 191-192° C (acetonitrile) ; Lit. (6): m.p. 191-192° (ethanol) C15H18BN02 (255.1) Calc. C 70.62 H 7.11 B 4.24 N 5.49 Found 70.96 7.21 4.18 , 5.45 1H-NMR (100 MHz, d6"DMSO/TMS) (ppm) : 6.84 s (6, CH^), 5.25 s (2, CH2) , 2.6-3.1 m (10, Ph) 11B-NMR (32.1 MHz, DMSO) : SjBF^OEt^) = -11.1 ppm Crystals suitable for X-ray analysis were obtained by recrystallization from 3:1 acetone-carbon tetrachloride. The crystal used for data collection was bounded by the (001), (010), and (100) planes at distances of 0.14, 0.35, and 0.14 mm from an internal origin and was mounted with b parallel to the goniostat axis. Dnit-cell and space group data were obtained from film and diffractometer measurements. The unit-cell parameters were refined by a least-squares treatment of sin2© values for 27 reflexions measured on a dif fractometer with Cu K,*. radiation. Crystal data are: 57 C15H18BM02 f,w* = 255.1 Orthorhomfcic, a = 17.043 (3), b = 6.289(1), c = 13.024(2) A, V = 1395.9 (5) A3, Dm = 1.225 (flotation in aqueous KI), Z = 4, Dx = 1.214(1) g cm-3, F (000) = 544 (20° C, Cu K*, fl = 1.5418 A, /x. - 6.4 cm-1). Absent reflexions: 0k/, k + * 2n g and h0^, h # 2n, space group Pna2^ (P-^v' No* 33)* Intensities were measured on a Datex-automated General Electric XKD 6 diffractometer, with a scintillation counter, Cu radiation (nickel filter and pulse height analyser), and a +3-29 scan at 2° min-1 over a range of (1.80 + 0. 86 tan 9) degrees in 29, with 20 s background counts being measured at each end of the scan. Data were measured to 29 = 145° (minimum interplanar spacing 0.81 A). The r.m.s. deviation of the intensity of the check reflexion, measured every 40 reflexions throughout the data collection, from its initial value was 2.4%. The final intensity was 1.045 times the initial value. Lorentz, polarization, and absorption corrections were applied, and structure amplitudes were derived. Of 1450 independent reflexions measured, 324 had intensities less than 3<r(l) above background where <rz (I) = S + B + (0.06S)2 with S = scan count and B = background count, corrected to time of scan. These reflexions were not included in the refinement. Structure Analysis The space group was assumed to be Pna2^ from systematic absences and the number of molecules in the unit-cell (Z = 58 4). The structure was solved by direct methods, 158 reflexions with normalized structure factor |E| > 1.55 being used in the symbolic addition procedure for non-centrosymmetric crystals (21). The phases of the 5 5 0, 8 3 0, and 14 2 1 reflexions were fixed to define the origin and the enantiomorph was fixed by allowing one of the symbol phases to take only values between 0 and 500 mc. During a manual expansion, carried out among the 75 reflexions with largest |E| values, it became apparent that there were eight reflexions from which the three symbol phases could be chosen. After several unsuccessful runs, a combination of symbol phases which gave a promising set of trial phases was found. The three symbol phases; 1 1 11, 3 1 13, and 14 8; along with the origin determining phases comprise the basic starting group given in Table 20. Eight starting sets were generated by allowing symbols a and b to have initial values of ±250 mc and c to have initial values of 125 and 375 mc (thereby fixing the enantiomorph). These sets were used as input to a computer program which determines phases using the tangent formula (22,23). The values of overall t, overall*, Q, and Rk on the final cycle for each of the sets are given in Table 21. Set 4, which had the lowest value of Rk, was expanded to 185 reflexions with |E| > 1.50 by starting with the same symbol values as in set 4. The final value of Rk was 0.23 with 180 phases assigned. An E-map based on these 180 phases gave positions for the 19 non-hydrogen atoms among the 40 highest peaks. Table 20 Basic starting set of reflexions for Ci^H^gBNC^ h k III phase (mc) 5 5 0 3.28 8 3 0 2. 07 1 Of origin determining i 14 2 1 1.99 1 0-> 1 1 11 2.66 a 3 1 13 2.62 b 1 4 8 2.44 c 60 Table 21 Results for the eight starting sets in the phase determination procedure set a {toe) b (mc) c (mc) t Rk 1 1 250 250 125 0.69 158 0. 30 0.30 146 2 250 250 3 75 0.72 158 0.27 0.31 144 3 250 -250 125 0. 69 165 0. 30 0. 35 14 1 4 250 -250 375 0.68 160 0.31 0.24 154 5 -250 250 125 0.69 160 0.30 0.30 145 6 -250 250 375 0.71 163 0.28 0.34 143 7 -250 -250 125 0.71 163 0. 28 0.28 149 8 -250 -250 375 0.58 135 0.41 0.36 141 61 Two cycles of full-matrix least-sguares refinement of the positional and isotropic thermal parameters of the non-hydrogen atoms gave R 0.156. This was followed by two cycles of anisotropic refinement which reduced R to 0. 103. A difference map at this point revealed the positions of seven of the ten phenyl hydrogen atoms. The remaining hydrogen atom positions were calculated and all 18 hydrogen atoms were included in subsequent cycles of refinement with isotropic thermal parameters. The refinement was concluded at R 0.071 for 1100 reflexions with I > 3<r(I) (26 reflexions were given zero weight in the final stages of refinement due to suspected extinction or counter errors). The scattering factors for the non-hydrogen atoms were taken from ref. 12 and those for the hydrogen atoms from ref. 13. The weighting scheme: w = 1/<r2(F) where <rz (F) is derived from the previously defined <rz (I) , gave constant average values of w(Fo-Fc)2 over ranges of |Fo| and was employed in the final stages of refinement. On the final cycle of refinement no parameter shift was greater than 0.33(rfor non-hydrogen atoms except for the y_ coordinates of methyl carbon atoms C(2) and C(3) which shifted by 0.80 <r". The shifts were less than 1.5 <r for the methyl hydrogens and less than 1.0<r for the remaining hydrogen atoms. The final positional and thermal parameters appear in Tables 22 and 23 respectively. Observed and calculated structure amplitudes are available on request. THERMAL MOTION AND CORRECTION OF MOLECULAR GEOMETRY 62 Table 22 Final positional parameters (fractional x 10*t x 103 for H atoms) with estimated standard deviations in parentheses Atom x j z 0 (1) 3418 (2) 4469 (6) 1466 (4) 0 t2) 3126 (2) 962 (6) 915 N 3315 (2) 977 (7) 1968 (4) C{ [D 3207 (4) 3249 (14) 2300 (6) c (2) 4138 (4) 108(17) 2051 (7) c [3) 2771 (5) -431 (22) 2513 (6) c (4) 2139 (3) 3904 (9) 385 (4) c (5) 1578 (4) 2389 (11) 255 (7) c (6) 801 (3) 2872 (13) 39(7) c (7) 571 (3) 4941 (1 1) -20 (5) c (8) 1100 (3) 6539 (13) 91 (6) c (9) 1907 (3) 6006 (10) 308 (5) c (10) 3584 (3) 3654 (9) -463 (5) c (11) 3659 (3) 1995 (10) -1166 (5) c (12) 4062 (4) 2244 ( 13) -2085 (6) c (13) 4409 (4) 4265 (14) -2305 (6) c (14) 4327 (4) 5879 (14) - 1610 (7) c (15) 3922 (3) 5542 (12) -688 (6) B 3054 (3) 3323 (1 1) 570 (5) H d&) 354 (4) 364 (11) 307 (6) H (1B) 255 (4) 340 (12) 255 (5) H (2A) 434 (5) 129 (11 ) 159 (8) H (2B) 422 (9) 2(21) 298 (14) H (2C) 425 (5) -177 (14) 170 (7) H <3A)- 242 (11) -13 (29) 228 (16) H (3B) 311 (4) -248 (10) 226 (5) H (3C) 285 (7) -43 (16) 336 (10) H (5) 170 (3) 99 (9) 23 (4) H (6) 47 (3) 149 (7) -25(4) H (7) 5(4) 525 (11) -25 (5) H (8) 9 1 (3) 796 (9) 12(4) H (9) 220 (6) 706 (17) 53 (8) H (11) 342 (5) 50 (12) -116(7) H (12) 404 (3) 129 (9) -259 (5) H (13) 479 (5) 459 (11) -292 (7) H (14) 443 (6) 776 (13) -186 (7) H (15) 384 (4) 685 (10) -33(5) 63 Table 23 Final thermal parameters and their estimated standard deviations 0 (a) Anisotropic thermal parameters (U. . x 100 A2) Atom %1 222 u33 %2 . %3 u23 0(1) 4.7 [2) 7. 5 < 3) 4.6 (2) -1.2(2) -1.3(2) 0.3 ( 2) 0(2) 4.6 ( 2) 6. 8 | 3) 3.8 (2) -0.4 (2) -1.0 (2) 0.3 | 2) N 3.5 (2) 6.0 | 3) 3. 4 (2) 0.2(2) -0.5(2) 0.5 ( 2) C(1) 6.7 ( 4) 10. 1 6) 5.7 (4) -0. 1 (4) -0.9 (3) 0. 3 | 4) C (2) 4.8 [3) 14. 2 | 7) 6.0 (4) 3.9(4) -1.8(3) -0.4 ( 5) C(3) 6.5 ( 4) 18.7 | 11) 4.6 (4) -3.3 (6) 1.4 (3) 2.6 | 5) C (4) 3.7 [2) 5.9 | 3) 3. 3 (2) -1. 1 (2) 0.1 (2) -0. 1 ( 2) C(5) 5.1 ( 3) 4.4 | 4) 10.4 (5) 0.3 (2) -1.9 (3) 1. 4 4) C (6) 3.8 [3) 8.7 | 5) 10. 1 (5) -1. 1 (3) -1.8(3) 1. 1 ( 4) C(7) 3.4 ( 2) 8.0 4) 4.7 (3) 0.2 (3) -0.6 (2) 0.0 3) C (8) 4.5 [3) 8. 4 [5) 6.6 (4) 2.3(3) -0.5(3) - 1.3 ( 3) C(9) 3.9( 3) 5.7 5.8 (3) 0.4 (2) -0,4 (2) -0.5 3) C{10) 2.9 [2) 6. 7 < 3) 4. 2 (3) 0. 3 (2) -1.1 (2) 1. 1 ( 3) C(11) 4.1 | 3) 5.6 4) 4.9 (3) 0.9 (2) -0.3 (2) 0.5 3) C (12) 5.3 (3) 8. 5 < 5) 5.8(4) 2. 9 (3) 0.2(3) 0.6 ( 4) C(13) 3.9( 3) 12.3 | 6) 6.0 (4) 1. 2 (4 ) 1.3 (3) 1.4 4) C (14) 4.5 [3) 9. 1 5) 8. 6 (5) 0. 0 (3) 1.7(3) 2.9 ( 4) C(15) 4.1 | 3) 7.5 | 4) 5.6 (3) -1.0 (3) -0.8 (2) 1. 1 3) B 3.7 (3) 5. 2 | 3) 4.2(3) -0.7(2) -0.5(2) 0,3 ( 3) (b) Isotropic thermal parameters Atom B(A2) Atom B (A2) H(1A) 3.8 (16) H (6) 2.0 (9) H <1B) 5. 3(15) H(7) 5. 1 (13) H(2A) 6.0(18) H (8) 3.1(11) H (2B) 15.0 (44) H(9) 11.9 (27) H (2C) 8.8 (21) H (11) 5.6 (16) H (3 A) 22.2 (67) H(12) 3.0 (12) H(3B) 4.3 (14) H (13) 7. 1 (17) H (3C) 11.2 (30) H(14) 9. 2 (25) H(5) 2.5(9) H (15) 4.4 |13) 64 Figure 7 A stereoscopic view of the molecule showing crystallographic numbering scheme. 50% probability ellipsoids are shown for the non-hydrogen atoms. The ellipsoids of thermal motion for the non-hydrogen atoms are shown in Figure 7. The thermal motion has been analysed in terms of the rigid-body modes of translation (T), libration (L) , and screw (S) motion using the computer program MGTLS (14). Four analyses were carried out: the 19 non-hydrogen atoms were considered first; then each of the phenyl groups along with the boron atom; and finally the five-membered ring and attached carbon atoms which failed to give a positive-definite L tensor. The results of the analyses of the two phenyl groups appear in Table 24. The r.m.s. standard deviation in the temperature factors e U|i is 0.C035 A2 which indicates that the mclecule as a whole (r.m.s. AU±j = 0.0124 A2) is not a good rigid-body whereas the thermal motion of the phenyl groups is adegu.itly 65 Table 24 Rigid-body thermal parameters1 C(4)-C (9) , B C (10)-C (15) , B r104 (17)-18 (7) 26(10)-, r 23(9) -2(5) -12(9) n L(aeg2) | 9(7) -5(5) | | 13(6) 0(6) | «- 17(5) J «• 33(14)J Principal axes of 1 r, m. s. Amplitude Direction cosines (x103) 10.7° 949 -171 264 6.4° -562 26 827 3.0 268 2 -963 4.0 -702 512 -495 2.5 165 985 48 3.5 -436 -859 -269 Principal axes of reduced T r.m.s. Amplitude Direction cosines (x103) 0.22 A 45 -998 36 0.22 A 152 -857 -492 0.19 998 46 32 0.20 407 508 -759 0.17 -33 35 999 0.14 901 -84 426 Displacement of axes from intersecting (A) Parallel to 0.74 1.60 Parallel to I2 0.13 0.55 Parallel to 0.40 -0.03 0 Effective screw translations (A) Parallel to 0.014 0.002 Parallel to L2 -0.005 0.000 Parallel to "ij -0.053 -0.004 Fractional coordinates of unique origin (x10*) x 2412 3051 I 4353 3786 z 329 -388 Fractional coordinates of centre of gravity (x10*) x 1572 3871 • y. 4296 385z 227 -1132 r.m.s. 4UJJ^ 0.0052 0.0046 A2 iAxes of reference are orthogonal angstrom axes. E.s.d.'s of components of L are given in parentheses in units of the last places shown, 66 described by the rigid-body parameters (r.m.s. A U^j = 0.0052 0 and 0.0046 A2). Both groups show somewhat anisotropic translational motion and anisotropic librational motion, particularly the C(4)-C(9), B group (see Table 24). The principal axes of L are oriented as expected: the largest oscillations, , correspond to rotations about the B-C bonds, the angles between the axes and the B-C bonds being 7.3 (C(4)) and 8.9° (C(10)). The unigue origins (14) are in the expected locations for both groups, lying between the B and attached phenyl C atoms. The appropriate bond distances and angles in the phenyl groups have been corrected for libration (15,16) using shape parameters cj2 of 0.08 for all atoms. Both corrected and uncorrected bond lengths and angles appear in Tables 25 and 26 respectively. RESULTS AND DISCUSSION The X-ray analysis has shown that the betaine-type structure (3) is correct. Figure 7 shows a general view of the molecule and the crystallographic numbering scheme. Figure 8 shows the packing arrangement viewed along b. Intra-annular torsion angles defining the conformation of the five-membered ring are given in Table 27 and some weighted least-squares mean planes through the molecule in Table 28. Non-bonded intra- and intermolecular contacts are listed in Table 29. Henceforth, the molecules (C^H^)gBCCHgCHgNHg and (p-FC6H4)2B0CH2CH2SH2 (Parts 1 and 2 ) will be referred to as 6a 67 Table 25 0 Bond lengths (A) with estimated standard deviations in parentheses (a) Non-hydrogen atoms Atoms uncorr. corr. Atoms uncorr. corr. 0(1)- C (1) 1. 378 (9) C(5)-C(6) 1. 388 (9) 1. 391 0(2)- N 1. 409 (5) C(6)-C (7) 1. 361 (10) 1. 377 0(1)- B 1. 506 (7) C(7)-C(8) 1. 359 (10) 1. 372 0(2)- B 1. 556 (8) C (8)-C (9) 1. 443 (8) 1. 446 C (4)-B 1. 620 (7) 1 .624 C(10) -C (11) 1. 394 (8) 1. 402 C(10) -B 1. 634 (8) 1 .639 C(10) -C (15) 1. 353 (9) 1. 363 C (1 )-N 1. 505 (10) C(11) -C(12) 1. 390 (9) 1. 396 C(2)- N 1. 509(7) 1 .551* C (12) -C (13) 1. 431 (10) 1. 441 C (3)-N 1. 467 (9) 1 .520* C(13) -C (14) 1. 367 ( 1 1) 1. 374 C(4)- C(5) 1. 360 (8) 1 .373 C (14) -C (15) 1. 401 (10) 1. 406 C{4)~ C{9) 1. 383 (8) 1 . 399 (b) Bonds involving hydrogen atoms Atoms distance Atoms distance CO) -H (1A) 1. 18(7) C(6)- H(6) 1. 10 (5) C(1) -H(1B) 1. 17(7) C (7)-H (7) 0.95 (7) C (2) -H (2A) 1. 02 (9) C(8)- H(8) 0.95 (6) C(2) -H(2B) 1. 22 (18) C (9)-H (9) 0.88 (11) C(2) -H (2C) 1. 28 (9) C (11) -H(11) 1.02 (8) C(3) -H (3A) 0. 70 (20) C (12) -H (12) 0.89 (6) C (3) -H (3B) 1. 45 (6) C(13) -H(13) 1.05 (9) C(3) -H(3C) 1. 11 (13) C (14) -H (14) 1.24 (8) C(5) -H (5) 0. 90 (5) C(15) -H(15) 0. 96 (6) •riding motion correction only. 68 Table 26 Bond angles (deg) with estimated standard deviations in parentheses (a) Non-hydrogen atoms Atoms uncorr. Atoms uncorr. corr, C{1) -0(1)-B 103.8 (4) B-C (4) -C (5) 122. 5 (5) 122. 1 N-0(2)-B 107. 1 (4) B-C (4) -C (9) 120. 1 (4) 119. 9 0(2) -N-C(1) 105.0 (4) C(9)-C(4)-C(5) 117. 4 (5) 118. 0 0 (2) -N-C (2) 106. 4 (4) C(4)-C(5)-C(6) 122. 8 (6) 122. 5 0 (2) -N-C(3) 108.8 (5) C (5)-C (6)-C (7) 119. 7 (6) 1 19. 5 C(1) -N-C (2) 115.9 (6) C(6)-C(7)-C(8) 120. 7 (5) 121. 2 C(1) -N-C(3) 110.9 (7) C(7)-C (8)-C(9) 118. 8 (6) 118. 5 C (2) -N-C (3) 109.6 (7) C (8) -C (9) -C (4) 120. 5 (6) 120. 3 0(1) -B-0 (2) 101.5 (4) B-C (10)-C (11) 119. 7 (5) 119, 6 0(1) -B-C (4) 113.8 (5) B-C (10) -C (15) 121. 8 (5) 121. 7 0(1) -B-C ( 10) 110.5 (4) C (15) -C (10)-C (11 ) 118. 4 (6) 118. 5 0 (2) -B-C (4) 109.5 (4) C (10) -C (1 1) -C ( 12) 121. 8 (6) 121. 7 0(2) -B-C( 10) 108.4 (4) C (11) -C (12)-C (13) 118. 5 (7) 118. 5 C (4) -B-C (10) 112.4 (4) C(12) -C(13)-C(14) . 119. 0 (6) 119. 1 0(1) -C(1)-N 105.7 (6) C (13) -C (14)-C (15) 120. 4 (7) 120. 3 C (14) -C (15) -C (10) 121. 9 (7) 121. 9 (b) Angles involving hydrogen atoms Atoms value Atoms value 0 (1 )-C (1) -H (1 A) 1 16 (4) C (5)- C(6)-H(6) 113 (2) 0(1)-C(1) -H (1B) 115 (3) c (7)- C (6)-H (6) 126 (2) N-C (1 )-H (1A) 112 (4) c (6)- C(7)-H(7) 118 (4) N-C (1) -H ( IB) 106 (4) c (8)- C (7)-H(7) 120 (4) H (1A)-C (1)-H (1B) 102 (5) c (7)- C(8)-H(8) 118 (3) N-C(2)-H(2A) 90 (4) c (9)- C (8)-H (8) 123 (3) N-C (2)-H (2B) 101 (7) c (8)- C (9)-H (9) 122 (7) N-C{2)-H (2C) 1 16 (4) c (4)- C(9) -H(9) 116 (7) H (2A) -C (2) -H (2B) 126 (9) c (10) -C (1 1 )-H (11 ) 130 (5) H (2A)-C (2)-H (2C) 115 (6) c (12) -C(11)-H(11) 108 (5) H (2B) -C (2)-H (2C) 107 (8) c (11) -C (12)-H (12) 123 (4) N-C (3)-H (3A) 100 (16) c (13) -C(12)-H(12) 118 (4) N-C (3)-H (3B) 100 (2) c (12) -C (13)-H (13) 126 (4) N-C (3)-H (3C) 1 14 (6) c (14) -C (13) -H (13) 1 15 (4) H(3A)-C(3)-H(3B) 119 (17) c (13) -C (14 )-H (14) 121 (4) H (3A)-C (3)-H (3C) 122 (17) c (15) -C (14) -H ( 14) 116 (4) H(3B)-C(3)-H(3C) 100 (6) c (14) -C(15)-H(15) 126 (4) C (4)-C(5) -H (5) 121 (3) c (10) -C (15)-H (15) 1 11 (4) C(6)-C(5)-H (5) 1 15 (3) Table 27 Intra-annular torsion angles (cleg) Five-membered ring Bond obs. calc. B-0(1) 34.3 (5) 36. 4 0(1)-C(1) -42.3 (5) -43.9 C (1)-N 33.7 (5) 34. 8 N-0 (2) -10.5 (5) -12.3 0 (2)-B -13.4 (4) - 15.0 70 and 6b respectively. The conformation of the five-membered ring is nearly the same as that of the isosteric •boroxazolidine* ring in 6b, four of the five torsion angles being equal within experimental error while the last differs by 2.5° (4 standard deviations) . Atoms C(1) and N both lie on the same side of o the OBO plane, displaced -0.75 and -0.31 A from the plane. The observed torsion angles in the ring are in good agreement with those obtained from energy minimization calculations for ui i = 10° (17), also given in Table 27. The observed magnitudes of the torsion angles are slightly smaller than the calculated values since the mean angle in the ring, 104.6°, is slightly greater than the calculated value of 104.2° but in good agreement with the values of 104.8 and 104.9° in the structures 6a and 6b. The individual values range from 101.5(4) at B to 107.1(4)° at 0(2). The angle at B is slightly, but significantly, greater than the mean value of 99.8(1)° in the boroxazolidines. The bond distances in the five-membered ring differ from their expected values as a result of steric strain and electron distribution in the molecule, analogous to that occurring in systems with N—>B interactions (see eg. 12 and 32). The 0(1)-C(1) bond, 1.378 (9) A, is significantly shorter than the usual value of 1.426 A as well as the values of 1.413 in 6a and 1.418 A in 6b. The C(1)-N bond, 1. 505(10) A, is somewhat longer than those in 6a and 6b (1.485 and 1.491 0 A) but is not significantly longer than a normal C (sp3)-71 N (sp3) bond. The N-0 (2) distance of 1.409(5) A is 0 significantly longer than the sum of covalent radii (1.36 A) but lies in the range of 1.34-1.44 A usually observed for N-0 single bonds (18,51). The two B-0 distances, 1.506(7) and 1.556(8) A, are significantly different. The pattern of one o bond close to the normal value and one on the order of 0.1 A longer than normal also occurs in the boroxazolidines 6a and 6b where B-0 distances are 1.484 and 1.471 A and the B-N o o bonds are 1.653 and 1.652 A, each about 0.1 A longer than normal. The exocyclic C-N distances have been corrected for riding motion and are egual within experimental error. Bearing in mind that the riding model approach usually overcorrects, it still appears that these bonds are somewhat longer than normal (see Table 25). The two phenyl rings are planar within experimental error (see Table 28). Two hydrogen atoms, H (6) and H(14), are significantly displaced from their respective mean planes, probably as a' result of inaccuracy in the hydrogen atom positions due to thermal effects. The boron atom is significantly displaced from both phenyl mean planes, by 0.07 from the C(4)-C(9) plane and by 0.11 A from the C(10)-C(15) plane, representing a slight folding of the phenyl groups away from each other. The dihedral angle between the mean planes is 74°. The two phenyl groups are not equivalent, the rings being rotated unequally about the E-C bonds. The dihedral angles C (9) [ C (4)-B ]0 (1) , C (15) [C (10 )-B ]0 (1) , C (5)[C (4)-B ]0 (2) , and C (11) [ C (10)-B ]0 (2) are -52.1(6), 39.9(6), 17.9(6), and -33.4(6)° respectively. The orientation 72 Table 28 Weighted least-squares mean planes (a) 0 Distances (A) of relevant atoms from the mean planes Atom d <V<r Atom d <V«r Plane 1: C(U)-C(9) Plane 2: C (10) -C (15) C(4) -0.001 0.1 C (10) 0.005 0.5 C (5) -0.004 0.5 C (11) 0.006 0.0 C(6) 0.012 1.3 C (12) -0.007 0. 1 C (7) -0.009 1. 3 C{13) -0.007 0.4 C(8) 0.007 0.8 C (14) 0.007 1.2 C (9) 0.000 0.0 C(15) -0.006 1.1 B 0.070 10.5 B 0.105 18.0 H (5) 0.088 1.6 H(11) 0.062 0.8 H(6) 0.285 6.1 H (12) 0. 143 2.5 H(7) 0. 116 1.7 H(13) -0.137 1.7 H(8) -0. 104 1.9 H (14) 0.357 3.7 H (9) -0.195 1.8 H(15) 0. 150 2.5 (b) Equations of planes: JLX + mY + nZ = £, where X, Y, and Z are orthogonal angstrom coordinates derived as follows: IH = r a 0 | 0 b >- 0 0 0 0 c 1 III J LzJ Plane m n 2 1 0.1930 -0.0163 0.9811 0. 1760 2 -0.8458 0.29 17 -0.4468 -4.2290 The dihedral angle between the planes is 74°. 73 Table 29 Selected intra- and intermolecular contacts Intramolecular Intermolecular* Atoms distance Atoms distance 0(1) • • .H(2A) 2.55 (7) C (3) .. .C (11) i 3. 393 (9) 0(1] • • ,C(2) 3. 100 (8) C (3).. .C (12) * 3.488 (10) 0 (1] * • • C(9) 3. 137 (7) 0(1) .. , H ( 3B) z 2.24 (6) 0(1) • • • C(15) 3.013 (8) C (1) . . . H (3B) 2 2.69 (6) 0 (2) • • . H(2A) 2. 26 (8) C(4) .. . H ( 3C) 3 2.67 (12) 0 (2) * • » H(3A) 2.26 (19) C (7) .. . H (2B) 3 2.63 (19) 0 (2) * • • H (5) 2. 58 (5) C (9) .. . H(3C) 3 2.73 (13) 0(2) * * • C(5) 2.916 (8) H (2C) . . .H(14) * 2.46 (1 1) 0 (2) • • .C(11) 2. 932 (7) H (5) . . .H (8)5 2.34 (7) C(1) • • • C(4) 3. 1 17 (9) H (6) . . . H(8)s 2.39 (7) *Superscripts refer to atoms at positions: 1 V2-X 1-1/2 1/2*z * 1-x -j 1/2+z 2 x 1+Y z 5 x _y-1 z 3 1/2-x 1/2+1 1-1/2 74 Figure 8 The packing arrangement viewed along b, hydrogen atoms have been omitted for clarity. of the phenyl groups represents a minimization of intra- and intermolecular steric interactions. The corrected C-C bond lengths in the phenyl groups range from 1.363 to 1.446 with a mean value of 1.395 A. There is a significant variation in the individual bond distances, the C(10)-C(17) bond, 1.363 A, being significantly shorter and the C(8)-C(9), 1 .446, and C(12)-C(13), 1.441 A, bonds significantly longer than the normal value of 1.394 A (18,51). The means over chemically equivalent groups of bonds (as they are removed from the boron substituent) are 1.384, 1.410, and 1.391 A. This pattern is different from that observed in the two boroxazolidine structures (Parts 1 and 2) where the bond lengths decrease as they are removed from the boron substituent. The B-C distances are equal within 75 experimental error and their mean value, 1.632 A, is longer than in the structures 6a (1.616) and 6b (1.621) but shorter than in the tetraphenyl borate anion (1.631-1.648(8) A) (19). The angles in the phenyl rings have a mean value of 120.0°, but the individual values, ranging from 118.0 to 122.5°, show some significant deviations from 120°. The mean angle at the carbon atom carrying the bcron group is 118.3° and the other mean values are 121.6, 119.2, and 120.2° for atoms orthox metaA and £ara to the boron group. These angular deviations have the same pattern as those in 6a and 6b but the magnitudes of the distortions are one-half as great. These variations have been explained in terms of the electronegativities of the substituent groups (31). The angles at C(4) and C(10), mean 118.3°, carrying the boron substituent indicate that this group is weakly electron releasing. The overall geometry of the molecule suggests that the formal charges on B and H in 3 are delocalized in a way such that, formally, the B and 0(2) carry partial negative charges while N and 0(1) carry partial positive charges. This is in accord with the observed pattern of bond distances, particularly the difference between the two B-0 distances. The mean bond angles in the molecule are as expected. There are a number of significant deviations from the mean values resulting from steric and electronic effects. Interior angles in the rings have already been discussed. The C (4)-B-C(10) angle, 112.4(4)°, is significantly smaller than in 6a 76 and 6b, but is generally as expected. Asymmetry in the packing arrangement appears to be responsible for significant differences between corresponding angle pairs 0-B-C and C-N-C. The geometry involving hydrogen atoms is as follows: O 0 mean C(ar)-H, 0.99 A, mean C(sj33)-H, 1.14 A, mean C(ar) -C(ar) -H, 119°, mean H-C (S£3)-H, 107°, and mean H-C(sj33)-H, 113°. The distances are long for X-ray data, probably as a result of relatively large thermal motion in the sample. The crystal structure consists of discrete molecules separated by normal van der Waals distances, the shortest of which are listed in Table 29. 77 PART U CRYSTAL AND MOLECULAR STRUCTURE OF THE N-METHYLDIETHANOLAMINOGALLAN E DIMER 78 INTRODUCTION Trimethylamine-gallane is known to react with compounds containing active hydrogen to eliminate molecular hydrogen and trimethylamine and form coordinatively unsaturated intermediates which then undergo cyclization to give oligomers whose size depends upon a balance between steric, mechanistic, and valency angle effects (52-54). The present work is part of an extension of this type of reaction involving aminoalcohols where the active hydrogen is attached to oxygen and/or nitrogen atoms. The title compound is derived from N-methyldiethanolamine and trimethylamine-gallane reacted in 1:1 molar ratio. In the title compound four-coordination about the gallium atom can be achieved in monomer units, analogous to similar boron compounds (9,20,Parts 1 and 2), by coordination of two oxygen atoms, one nitrogen atom, and the remaining hydrogen atom, after elimination of two moles of hydrogen and one mole of trimethylamine from the reaction sphere. The metal is indeed coordinated to these atoms but instead of discrete monomer units a novel dimerization through bridging oxygen atoms is realized, to give a distorted trigonal bipyramidal arrangement about each five-coordinate* gallium atom (_1). 1 A preliminary report of the structure of the five-coordinate complex chlorobis-(8-hydroxy-2-methylquinolin-ato)gallium (III) by K. Dymock and G. J. Palenik, Chem. Comm., 884 (1973) appeared during the preparation of this thesis. The amount of structural information therein does not warrant inclusion of this data in the discussion. 79 1 EXPERIMENTAL The N-methyldiethanclaminogallane dimer was prepared by reacting N-methyldiethanolamine (0.226 g; 1.9 mmoles) with trimethylamine - gallane (0.250 g; 1.9 mmoles) in benzene. Hydrogen (84.5 ml; 3.77 mmoles) was evolved at room temperature to leave the product in benzene as a clear solution: MeN (CH2CH2OH) 2 + Me^NGaH^ > MeN (CH2CH20 )2GaH + 2H2 + Me^N Removal of all volatiles gave a white air-sensitive sclid. [ Analysis: reguired for MeN (CH2CH20)2GaH: Ga, 37.1??; hydrol. H, 0.535? found: Ga, 36.9??; hydrol. H, 0.54%.] The compound was redissolved in benzene and the solution cooled to 5° C. Large colorless crystals were deposited from solution after a prolonged period of time. Crystals suitable fcr X-ray analysis were positioned in capillaries under a nitrogen 80 atmosphere to avoid the rapid hydrolysis which occurred in contact with moist air. The capillaries were then flame sealed. The crystal chosen for study was mounted with the [2 1 1] vector parallel to the goniostat axis and had dimensions of ca. 0.3 x 0.3 x 0.5 mm. Unit-cell .and space group data were obtained from film and diffractometer measurements. The unit-cell parameters were refined by a least-squares treatment of sin2 9 values for 30 reflexions measured on a diffractometer with Cu radiation. Crystal data are: C10H24Ga2N20^ f.w. = 375.8 Orthorhombic, a = 19.112(4), b = 9.947 (2), c = 7.709 (2) A, V = 1465.5 (5) A3, Z = 4, Dx = 1.703 (1) g cm~3, F(000) = 768 (20° C, Cu K^, 9\ = 1.5418 A, ^tU. = 49.7 cm-*). Absent reflexions: hOO, h * 2n, OkO, k * 2n, and 00^, £ * 2n define ii, uniguely the space group P212^21 (Dg, So. 19). Intensities were measured on a Datex-automated General Electric XRD 6 diffractometer, with a scintillation counter, Cu radiation (nickel filter and pulse height analyser), and a 9-20 scan at 2° min-1 over a range of (1.80 + 0.86 tan 9) degrees in 29, with 20 s background counts being measured at each end of the scan. Data were measured to 29 = 145° 0 (minimum interplanar spacing 0.81 A). A check reflexion was monitored every 40 reflexions throughout the data collection. The r.m.s. deviation of the intensity of the check reflexion 81 from its initial value was 2.2% and the final intensity was 1.014 times the initial value. Lorentz and polarization corrections were applied., and the structure amplitudes were derived. No absorption correction was attempted due to the irregularity of the crystal surface (in particular re-entrant angles). Of the 1697 independent reflexions measured, 180 had intensities less than 3<r(I) above background where <rz (I) = S + B + (0.05S)2 with S = scan count and B = background count, corrected to time of scan. These reflexions were not included in the refinement. Strueture Analysis The positions of the two gallium atoms were determined from the three-dimensional Patterson function, Three cycles of full-matrix least-squares refinement of the positional and isotropic thermal parameters of the gallium atoms gave R 0.27. A difference map revealed the positions of all the C, N, and 0 atoms. All the non-hydrogen atoms were refined isotropically for three cycles giving R 0.096 and then anisotropically for two cycles giving R 0.076. A difference map revealed the positions of the gallium H atoms and 12 of the 16 methylene protons. The remaining hydrogen atoms were assigned calculated positions. The hydrogen atoms were included in all subsequent cycles of refinement with isotropic temperature factors. The refinement was concluded after four more cycles with R = 0.056 for 1477 reflexions with I > 3<r(I) . 82 The absolute configuration of the complex (for the particular crystal used) has been determined through the anomalous scattering of the non-hydrogen atoms. Enantiomorph (A) is represented by the coordinates in Table 31 referred to a right-handed axial system and enantiomorph (B), the mirror image of (A), was generated by changing the x coordinates of (A) to 1-x. Both enantiomorphs were refined and Hamilton's test (24) applied to the resulting R factor ratios. Enantiomorph (A) was clearly indicated as correct. The results of Hamilton's test are compiled in Table 30. The scattering factors of ref. 55 were used for the non-hydrogen atoms and those of ref. 13 for the hydrogen atoms. Anomalous scattering factors from ref. 56 were used for Ga, 0, N, and C atoms. The weighting scheme: w = 1 if |Fo| < 11; w = (11/1Fo |)2 if |Fo | > 11, and w = 0.49 for the weak reflexions gave constant average values of w(Fo-Fc)2 over ranges of |Fo| and was employed in the final stages of refinement. On the final cycle of refinement the mean parameter shift was 0.29<r, the largest shifts were 0.85<r- for non-hydrogen and 1.70<*~for hydrogen atoms, both of which were associated with the C{10) methyl group. The final positional and thermal parameters are given in Tables 31 and 32 respectively. Measured and calculated structure amplitudes are available on request. THERMAL MOTION AND CORRECTION OF MOLECULAR GEOMETRY The ellipsoids of thermal motion for the uon-hydrogen 83 Table 30 Results of Hamilton's Test Parameter compared Value for enantiomorph Sig. (ft) (B) (B/A) level1 Conventional R (3<r data) 6. 126 6. 210 1. .0137 >99, .5 Conventional R (all F) 6. 491 6. 573 1. .0125 >99. .5 Weighted R (3<r data) 8. 738 8. 878 1. .0161 >99. .5 Weighted R (all F) 9. 162 9. 232 1. .0145 >99. ,5 *This is the 5? probability that enantiomorph (A) is the correct absolute configuration. Table 31 Final positional parameters (fractional x 10*f Ga x10s, H x103) with estimated standard deviations in parentheses Atom X 2 z Ga(1) 41623 (5) 20988 (9) 38421 (13 ) Ga (2) 40777 (5) 27950 (8) 76738 (12) 0(1) 4292 (3) 1238 (5) 6085 (7) 0 (2) 329 1 (4) 2739(6) 3234 (10) 0 (3) 4335 (3) 3645 (5) 5470 (8) 0 (4 ) 3170 (4) 2235 (6) 8000 (9) N (1) 3704 (4) 126 (7) 3277 (10) N (2) 3662 (4) 4807 (7) 8153 (11) C (1) 3999 (6) -794 (8) 4579 (14) C (2) 4036 (6) -97(8) 6316 (13) C (3) 2961 (5) 424 (9) 3574 (18) C (4 ) 2800 (5) 1779 (9) 2715 (16) C (5) 3837 (9) -343 (12) 1516 (18) C (6) 4032 (5) 5690 (8) 6919(15) C (7) 4118 (6) 4996 (8) 5206 (13) C (8) 2928 (5) 4611 (9) 7696 (15) C (9) 2686 (5) 3265 (9) 8447 (17) C (10) 3763 (7) 5237 (1 1) 9954 (16) H (Gal) 475 (5) 217 (9) 282 (1 1) H (Ga2) 461 (4) 268 (7) 899 (9) H (1A) 445 (5) -90 (10) 430 (14) H (1E) 365 (5) -151 (9) 463 ( 12) H (2A) 426 (8) -81 (14) 713 (18) H (2B) 353 (5) 5(11) 653 (15) H (3A) 290 (6) 34 (12) 476 (18) H (3B) 258 (6) -27 (10) 312(13) H (4A) 273 (4) 165 (8) 141 (11) H (4B) 241 (10) 194 (17) 304 (26) H(5A) 342(10) -102 (20) 142 (29) H (5B) 380 (8) 32 (18) 76 (22) H (5C) 426 (5) -41 (10) 131 (13) H (6 A) 460 (6) 566 (13) 735 ( 15) H (6B) 375 (7) 641 (13) 679 (17) H (7A) 443 (7) 548 (14) 467 (18) H (7B) 377 (7) 490 (12) 460 (15) H (8A) 289 (5) 434 (11) 629 (17) H (8B) 276 (11) 516 (19) 359 (28) H (9A) 263 (6) 318(11) 997 (15) H (9B) 213 (6) 290 (11) 795 (14) H (10A) 334 (8) 598 (19) 990 (22) H(10B) 340 (9) 453 (17) 1086 (24) H (10C) 431 (6) 548 (12) 1021 (15) 85 Table 32 Final thermal parameters and their estimated standard deviations (a) Anisotropic thermal parameters (U« * x 100 A2) Atom hi P-22 533 "12 P-13 H23 Ga(1) 5.47 (6) 3. 23 (5) 4.90 (6) -0.30(4) 0.36 (4) 0. 20 (4) Ga (2) 5.16(6) 2.86 (5) 4.97 (6) 0.08 (4 ) -0. 27 (4) 0. 14 (4) 0(1) 5.8 (3) 3. 1 (2) 4.7 (3) 0.5 (2) 0.0(3) 0.2 (2) 0(2) , 7.5(4) 2.9(3) 8.1 (4) -0.1 (3) -1.9 (3) 1. 4 (3) 0 (3) 6.3 (3) 2. 8 (2) 5. 1 (3) -0.9 (2) 0. 1(3) 0. 2 (2) 0(4) 6.5(3) 3.1 (3) 7.5 (4) -0.1 (3) 1.0 (3) 0. 4 (3) N (1) 5.7 (4) 3. 1 (3) 5.5 (4) -0. 1 (3) -0.5(3) 0.3 (3) N (2) 5.6(4) 2.9 (3) 6.3 (4) 0. 1 (3) 0.5 (3) 0.5 (3) C(1) ,6.5(6) 3. 2 (4) 6.5(5) 0.7 (4) -0.9 (5) -0. 1 (4) C(2) 6.8(6) 3.6 (4) 5.3 (5) -0.1 (4) -0.6 (4) 0.6 (4) C(3) 5.8 (5) 3.7 (4) 9.0 (8) -0. 1 (4) -0.5(5) 0.7 (5) C (4) 5.9(5) 4.2 (4) 8.3 (7) -0.1 (4) -1.8 (5) 1. 1 (4) C(5) 9.3 (10) 4.9 (5) 7. 1 (7) -0.4 (6) 0.6(7) -0.8 (5) C(6) 6.3(6) 2.7 (4) 7.6 (6) -0.5 (4) 0.7 (5) 0. 0 (4) C(7) 6 .3 (5) 3.7 (4) 6.0 (5) -0.7 (4) 1.0(5) 1.0 (4) C (8) 5.8(5) 3.5 (4) 7.1 (6) -0.5 (3) 0.5 (4) 0. 9 (4) C (9) 6.0 (5) 3. 8 (4) 9.0 (7) 0. 6 (4) 1.8(5) 1.4(4) C(10) 10.1 (9) 4.2 (5) 6.6 (6) 0.5 (5) -0.4 (6) -1. 4 (5) (b) Isotropic thermal parameters o Atom B (A2) Atom B (A2) H (Ga1) 3.9 (19) H (Ga2) 2. 3(13) H(1A) 4.3 (21) H (6A) 7.0 (29) H (1 B) 3.0 (17) H (6B) 5. 4(28) H (2A) 8.8 (30) H (7A) 6.5 (29) H (2B) 3.6 (21) H(7B) 4.3 (28) H(3A) 4.8 (27) H (8A) 2.6 (23) H (3B) 4.0 (20) H (8B) 11.9 (52) H (4A) 1.7(12) H (9A) 3.0 (24) H (4B) 12. 3 (51) H (9B) 10.2(23) H(5A) 17.4 (50) H (10A) 13.7 (44 ) H (5B) 8.0 (40) H (10B) 4. 7(47) H(5C) 2.1 (18) H (10C) 5.2 (24) 86 H<2A> Figure 9 A stereo view of the molecule along the Ca axis showing the atom numbering and 50% probability thermal ellipsoids for the non-hydrogen atoms. Broken lines show possible C-H...0 hydrogen bonds. atoms are shown in Figure 9. The thermal motion has been analysed in terms of the rigid-body modes of translation (T), libration (L), and screw (S) motion using the computer program MGTLS (14). The r.m.s. standard deviation in the temperature factors ^ is 0.0042 A2 which indicates that the thermal motion of the molecule as a whole (r.m.s. A. = 0.0054 A2) is adequately described by the rigid-hody parameters in Table 33. The indicated modes of motion are physically reasonable; the translational and lihraticnal motions are both somewhat anisotropic. The orientation of the principal axes of L is nearly coincident with that of the principal axes of inertia, the largest librational motion occurring about the least axis of inertia. Table 33 Rigid-body thermal parameters1 87 all non-hydrogen atoms r 49(9) -14(9) -4(8) 1 L (x 10 deg2) | 157 (21) 46 (13) | «- !, 69 (15) J Principal axes of L r.m.s. Amplitude Direction cosines (x103) 4.2° -108 914 392 2.2 -506 288 -813 2.2 -856 -286 430 Principal axes of reduced T r.m.s. Amplitude Direction cosines (x103) 0.23A -1 -268 -963 0.22 997 -73 20 0.16 -76 -960 267 Displacement of axes from intersecting (A) Parallel to 0.65 Parallel to Ig -0.53 Parallel to L3 -0.32 o Effective screw translations (A) Parallel to 0.036 Parallel to L2 -0.039 Parallel to -0.031 Fractional coordinates of unique origin (x10*) x 3862 jr 2358 z 5740 Fractional coordinates of centre of gravity (x10A) x 3818 2 2462 z 5721 r.m.s. A0jj_ (A2) 0.0054 JAxes of reference are orthogonal angstrom axes. E.s.d.*s of components of L are given in parentheses in units of the last places shown. 83 The appropriate bond distances and angles have been corrected for libration (15,16) , using shape parameters q2 of 0.08 for all the atoms involved, and appear in Tables 34 and 35 respectively. RESULTS AND DISCUSSION The X-ray analysis has provided the first known crystallographic example of pentacoordinate gallium as well as the first reported Ga-H distances. The numbering scheme is shown in Figure 9, in which the molecule is viewed along its approximate C2 axis. Figure 10 shows the coordination about the gallium atoms and Figures 11 and 12 show the crystal structure viewed along c and b respectively. Ga-N and Ga-0 bond distances in related four and six-coordinate structures are compiled in Table 36. Some weighted least-squares mean planes through the molecule are given in Table 37 and the dihedral angles in the five fused rings of the molecule in Table 38, Selected inter- and intramolecular contacts are listed in Table 39. The molecule has C2 symmetry within the limits of experimental error. The bond distances, valence angles, and dihedral angles averaged assuming C2 symmetry also appear in the appropriate tables and will be employed in the discussion of the molecular geometry. The molecule exhibits, in part, the structure expected for the monomeric boron analogue (on the basis of the structures of triethanolamine borate * (20), B,B-diphenylboroxazolidine (Part 1), and B,B-bis(£-fluorophenyl) boroxazolidine (Part 2), and supporting 89 Table 34 o Bond lengths (A) with estimated standard deviations in parentheses (a) Non-hydrogen atoms Atoms uncorr. corr. Atoms uncorr. corr. mean* Ga (1) -0 (1) 1. 945 (6) 1. 952 Ga(2) -0(3) 1. 960 (6) 1. 967 1. 960 (8) Ga(1) -0(2) 1. 843(7) 1. 848 Ga (2) -0 (4) 1. 839 (7) 1. 845 1. 847 (2) Ga (1) -0 (3) 2. 012 (6) 2. 016 Ga (2) -0 (1) 2. 016 (6) 2. 019 2. 018 (2) Ga(1) "N (1) 2. 193 (7) 2. 196 Ga (2) -N(2) 2. 184 (7) 2. 187 2. 192 (5) 0(1)- C(2) 1. 427 (10) 1. 429 0 (3)-C (7) 1. 422 (10) 1. 424 1. 427 (3) 0 (2)-C(4) 1. 398 (11) 1. 399 0(4)- C(9) 1. 422 (1 1) 1. 424 1. 412 (13) N(1)- C{1) 1. 471 (12) 1. 475 N (2)-C(6) 1. 475 (12) 1. 479 1. 477 (2) H (D-C (3) 1. 468 (13) 1. 471 N(2)- C(8) 1. 460 (12) 1. 464 1. 468 (4) N(1)- C(5) 1. 458(15) 1. 460 N(2)- C(10) 1 .466 (14) 1. 468 1. 464 (4) C (1)-C{2) 1. 510 (14) 1. 512 C(6) -C(7) 1. 499 (14) 1. 501 1. 507 (6) C(3)- C(4) 1. 533(13) 1. 535 C (8)-C (9) 1. 531 (12) 1. 533 1. 534 (1) (b) Bonds involving hydrogen atoms Atoms distance Atoms distance Ga(1) -H (Ga1) 1.37 (8) Ga (2) -H (Ga2) 1. 45 (7) C (1)-H (1&) 0.90 (10) C(6)- H (6A) 1. 14 (12) C(1)- H(1B) 0.98(9) C(6)-H (6B) 0. 90 (13) C(2)- H (2A) 1.04 (14) C (7)-H (7A) 0. 87 (14 ) C (2)-H (2B) 0.99 (10) C(7)-H(7B) 0. 82 (12) C(3)- H ( 3A) 0.93 (14) C (8)-H (8A) 1. 12 (13) C(3)- H (3B) 1.06 (11) C(8)- H (8B) 0. 94 (20) C(4)- H (4A) 1.02(8) C (9)-H (9A) 1. 18 (11) C (4)-H (4B) 0.81 (19) C(9)-H(9B) 1. 20 (11) C(5)- H (5A) 1.05 (20) C (10) -H (10A) 1. 10 (13) C(5)-H (5B) 0.89 (18) C (10) - H(10B) 1. 21 (16) C(5)- H (5C) 0.83 (9) C (10) -H (10C) 1. 09 (12) •Average of bonds related by the £2 axis, number in parentheses is r.m.s. deviation from the mean. Table 35 Bond angles (deg) with estimated standard deviations in parentheses (a) Non-hydrogen atoms Atoms uncorr. corr. A toms uncorr. corr. mean* 0(1 0(1 0(1 0 (2 0(2 0(3 Ga (1) Ga (1) Ga (2) Ga ( Ga ( Ga ( Ga ( C(1 C(1 C(3 N(1 N(1 C(1 C(3 •Ga( 1 Ga (1 •Ga{ 1 Ga (1 Ga (1 Ga (1 -0(1 -0(1 -0(1 -0(2 - N (1 - N ( 1 -N(1 N (1) N (1) N (1) C(1) C (3) C(2) C (4) -0(2) -0 (3) -N (1) -0(3) -N (1) -N(1) -Ga (2) -C (2) -C(2) -C(4) -C (1) -C(3) -C{5) C(3) C(5) C(5) C(2) C (4) 0(1) 0(2) 1 19. 5 (3 76. 2 (2 80.4 (3 92.5 (3 84. 1 (3 150.9 (3 100.1 (2 1 18.5(5 124.8 (5 116.4(5 105.6 (5 100.1 (5 1 13.7 (7 113.0 (8 111.6(8 1 12.2 (9 109.7 (7 107.7 (8 109.5(7 1 10.0 (8 119. 5 0(3)- Ga (2) -0 (4) 119. 1 (3) 119. 1 119.3 (2) 76. 2 0(3)- Ga (2) -0(1) 75. 8 (2) 75. 7 76. 0 (3) 80. 4 0(3)- Ga (2) -N(2) 80. 9 (3) 80. 9 80. 7 (3) 92. 5 0(4)- Ga (2) -0 (1) 92. 4 (3) 92. 4 92. 5 (1) 84. 0 0(4)- Ga (2) -N (2) 84. 9(3) 84. 9 84. 5(5) 150. 8 0(1)- Ga (2) - « (2) 151 . 6 (3) 15 1. 6 151.2 (4) 100. 2 Ga (2) -0 (3) -Ga (1) 99. 7(2) 99. 8 100. 0 (2) 118. 5 Ga (2) -0(3) -C (7) 117. 3 (6) 117. 3 117. 9 (6) 124. 7 Ga (1) -0 (3) -C(7) 125. 8(6) 125. 7 125. 2(5) 116. 4 Ga (2) -0 (4 ) -C (9) 115. 4 (5) 115. 3 1 15. 9 (6) 105. 5 Ga (2) -N (2) -C(6) 105. 2 (5) 105. 2 105. 4 (2) 100. 1 Ga (2) -N (2) -C(8) 100. 7 (5) 10 0. 8 100. 5 (4) 113. 6 Ga (2) -N (2) -C (10) 112. 3 (6) 112. 3 1 13.0 (7) 112. 9 C(6)- N (2)" C(8) 112. 6(8) 112. 6 112.8 (2) 111. 7 C (6) -N (2)- C (10) 112. 0 (8) 112. 0 111.9 (2) 112. 3 C(8)- M (2)- C(10) 113. 2(9) 113. 2 112.8 (5) 109. 8 N (2)- C (6)- C (7) 110. 2 (7) 110. 3 1 10. 1 (3) 107. 7 N (2) -C (8)- C(9) 108. 4 (8) 108. 4 108. 1 (4) 109. 4 C (6)- C (7)- 0(3) 109. 9(8) 109. 9 109. 7 (3) 110. 1 C(8)- C (9)- 0 (4) 110. 0 (8) 110. 0 110.1 (1 ) continued. (b) Angles involving hydrogen atoms Atoms value Atoms va lue 0 (1) -Ga (1 )-H (Ga1) 115 (4) 0 (3) -Ga (2) -H (Ga2) 118 (3) 0 (2) -Ga (1)-H(Ga1) 125 (4) 0 (4) -Ga (2) -H (Ga2) 123 (3) 0 (3)-Ga (1 )-H (Ga1) 10 1 (4) 0 (1)-Ga (2)-H (Ga2) 103 (3) N (1) -Ga (1)-H (Ga1) 105 (4) N (2) -Ga (2) -H (Ga2) 102 (3) N (1)^C(1) -H(1A) 106 (7) N (2)-C (6)-H (6A) 105 (6) N (1 )-C (1 )-H (1B) 102 (5) N (2) -C (6)-H (6B) 104 (8) C (2)-C(1)-H (1A) 103 (7) C (7) -C (6)-H (6A) 98 <6) c (2)-C (1) -H (1 B) 109 (5) C (7) -C (6) -H (6B) 109 (8) H (1A) -C(1)-H(1B) 126 (8) H (6 A) -C (6) -H (6E) 129 (10) 0 (1 )-C (2) -H (2A) 125 (8) 0 (3)-C(7)-H(7A) 113 (9) 0 (1) -C (2) -H (2B) 103 (6) 0 (3) -C(7)-H (7B) 102 (9) c (1)-C(2) -H(2A) 104 (7) C (6)-C (7)-H (7A) 104 (9) c (1)-C (2)-H (2B) 100 (7) C (6) -C(7) -H (7B) 118 (8) H (2A) -C (2)-H (2B) 113 (11) H (7A) -C (7) -H (7B) 1 10 (12) N (1 )-C (3) -H (3A) 105 (7) N (2) -C (8) -H (8A) 109 (5) N (1)-C(3) -H(3B) 118 (6) N (2) -C (8) -H (8B) 94 (12) C (4)-C (3) -H (3A) 119 (8) C (9)-C(8)-H (8A) 98 (6) C (4) -C(3) -H (3B) 107 (5) C (9)-C (8)-H (8B) 98 (12) H (3A)-C (3)-H (3B) 101 (9) H (8A) -C (8) -H (8B) 146 (15) 0 (2) -C (4) -H (4A) 117 (4) 0 (4)-C (9)-H (9A) 104 (6) 0 (2)-C (4) -H (4B) 114 (13) 0 (4)-C (9)-H (9B) 107 (5) C (3) -C (4)-H (4A) 110 (4) C (8) -C(9)-H (9A) 118 (6) C (3) -C (4) -H (4B) 103 (13) C (8)-C (9)-H (9B) 1 15 (5) H (4A)-C (4)-H (4B) 103 (15) H (9A) -C (9) -H (9B) 103 (8) N (1) -C (5) -H (5A) 98 (12) « (2) -C (10) -H (10 A) 94 (9) H (1 )-C (5) -H (5B) 111 (10) N (2) -C(10) -H (10B) 107 (9) N (1)-C(5) -H(5C) 112 (7) N (2)-C (10) -H (10C) 1 1 1 (6) H <5A)-C (5)-H (5B) 111 (14) H (10A)-C(10)-H(10B) 89 (11) H (5A) -C (5)-H (5C) 133 (13) H (10A)-C (10) -H (10C) 124 (11) H (5B)-C (5)-H (5C) 91 (12) H (10B)-C (10)-H (10C) 125 (10) •mean of corrected angles related by the axis, number in parentheses is the r.m.s. deviation from the mean. 92 chemical evidence (9)) in that the MeN (CH^CH^C^ acts as a tridentate ligand, the nitrogen and two oxygen atoms all being coordinated to the same gallium atom. Here the similarity to the boron compounds ends as dimerization occurs through bridging oxygen atoms, creating a four-membered Ga^C^ ring which results in the formation of a molecule possessing a system of five fused rings. A polymeric structure might be expected of the compound produced by the reaction of trimethylamine-gallane and N-methyldiethanolamine even though gallium has a high tendency to form four-membered rings with oxygen (57,58). Thus polymerization through nitrogen or oxygen atoms (ie. the MeN (CH2CH2O)2 acting as a bridging ligand) to form large heterocycles with tetrahedral coordination about the gallium atoms was originally suspected. The resulting structure contains pentacoordinate gallium atoms with distorted trigonal bipyramidal geometry. The nitrogen and two oxygen atoms of each MeN (CH2CH20)2 ligand occupy respectively an axial and two eguatorial positions about the associated gallium atom. One of the oxygen atoms bridges the two gallium atoms, occupying an axial position of the second gallium atom. The remaining eguatorial site is occupied by the hydrogen atom (see Fig. 10). o The axial Ga-N distance of 2.192(5) A is longer than the e observed distances for tetrahedral (mean 1.97 A) and for octahedral gallium (mean 2.12 A) shown in Table 36. The distance compares well with a sterically similar bond in Figure 10 A view of the coordination about the gallium atoms. octahedral GaH (EOTA) . H20 (59) of 2.182(5) A and the axial Al-N distance of 2.18 A in AlH^(NMe^)2 (60). The bond nevertheless appears to be be weaker than a normal single bond as a result cf steric strain as in the related aminoalcohol boron compounds (20,Parts 1 and 2). The three types of Ga-0 bonds are all significantly different. The equatorial non-bridging distance is 1.847(2) A and for the bridging oxygen the equatorial distance is 1.960(8) and the axial is 2.018(2) A. The equatorial bonds involve ££2 hybrids at the gallium atoms which reduces the covalent radius of gallium to 1.21 A for these bonds. The expected equatorial Ga-0 distance is then about 1.87 A. The mean Ga-0 distances in related four-coordinate and octahedral complexes are 1.96 and 1.959 A respectively. In this structure the 1.847 A •terminal1 Ga-0 distance corresponds to a strong bond while 94 Table 36 Comparison of Ga-O and Ga-N bond distances Compound Ga Coord, no. Ga-O Ga-N ref. GaN 4 1.94 62 (H2GaNCH2CH2) 3 4 1.97 53 [ D2Ga (N2C3H3) ]2 4 1.980 54 [ (CH3) 2GaOH]4  1 .94 ,1 .98 63 (CH3)3NGaH3 4 1.97 64 [CH3N(CH2CH20)2GaH]2 5 1.843-2.019 2.187,2.196 * [Ga2 (0H)2C12 (C124.H17N3)2 ]C12. H20 6 1. 908, 2.0 17 2. 083-2. 132 58 GaH(EDTA) .H20 6 1 .924-1 .996 2.097,2. 182 59 [GaCl2 (bipy)2 ]+[GaCl^ ]- 6 2.097, 2. 105 65 GaCl3(terpy) 6 2.034-2.115 66 •this work 95 the bridge bonds, 1.960 and 2.018 A, seem to be of nearly egual strength considering that one is axial and the other equatorial and both these distances are within the range of previously reported Ga-0 bond lengths (see Table 36). Using 0 an effective radius of 0.23 A for hydrogen (61), the expected Ga-H bond length is 1.44 A in good agreement with the mean Ga-H distance of 1.41(4) A. The distortion of the trigonal bipyramid occurs as a deformation of the angle between the axial groups from the ideal 180° to 151.2(4)°. The equatorial GaOOH groups are both planar within experimental error (see Table 37) and the mean O-Ga-O, 119.3(2)°, and O-Ga-H angles, 120.3°, are close to the expected 120°. The equatorial-Ga-axial angles range from 76° (in the four-membered ring) to 104° (mean N-Ga-H), each 14° from the ideal 90°. The angular distortions are a result of the steric constraints inherent in the fused ring system. The four-membered Ga202 ring is non-planar with all intra-annular dihedral angles equal within experimental error, the mean value being 21.6°. Angles in the ring are 76.0(3) at Ga and 100.0(2)° at 0. The ring is different from the planar centrosymmetric Ga202 ring in the octahedral complex [ Ga2 (OH) 2Cl2 (C^^H^yN^) 2 ]Cl2. H20 (16) in which there 0 is one strong and one weak Ga-0 bond (1.908 and 2.017 A). The difference between the •terminal' and 'bridging1 oxygen atoms is carried into the five-membered GaOCCN rings which have distinct geometries which may be ascribed to steric and electronic differences between the two classes of 96 Table 37 Intra-annular torsion angles (deg) (a) Five membered rings Bond obs. Bond obs. mean calc. Ga{1)-0 (1) 2. 1 (5) Ga (2) -0 (3) 3.4(5) 2.8(7) 5.0 0 (1)-C(2) -25.0 (8) 0(3)-C (7) -26.4 (8) -25.7 (7) -29. 8 C (2)-C(1) 43.3 (8) C (7)-C (6) 44. 1 (8) 43.7 (4) 43. 0 C (1)-N (1) -39.8 (7) C(6)-N(2) -39. 2(7) -39.5(3) -40.0 N(1)-Ga (1) 21.4 (5) N (2) -Ga (2) 20. 1 (5) 20,8 (7) 21. 8 Bond obs. Bond obs. mean calc. Ga (1)-0 (2) 1.2 (5) Ga (2) -0(4) 2. 7(5) 2. 0(8) 1.3 0(2)-C(4) -27.5 (8) 0(4)- C (9) -28. 3 (8) -27. 9(4) -27. 1 C (4)-C(3) 50. 3 (9) C(9)- C (8) 49. 6 (9) 50. 0 (4) 42. 3 C (3)-N (1) -44.1 (7) C (8) - N(2) -42. 6(7) -43. 4(8) -41.6 N(1)-Ga (1) 25.3 (6) N (2)- Ga (2) 23. 5 (5) 24. 4 (11) 25.0 (b) Four-membered ring Bond obs. Ga (1) -0(1) -21 .3 (2) 0(1)- Ga (2) 22 .0(2) Ga (2) -0(3) -21 .2 (2) 0(3)- Ga(1) 21 .9(2) 97 Table 38 Weighted least-squares mean planes (a) Distances (A) of relevant atoms from the mean planes Atom d d/o- Atom d d/^ 1: Ga(1), 0(1), 0(2), H (Ga 1) 2: Ga(2), 0(3), 0(4), H (Ga2) Ga(1) 0.000 0.0 Ga(2), 0. 000 0.0 0 (1) 0.000 0.0 0 (3) 0.000 0.0 0 (2) 0.000 0.0 0 (4) 0.000 0.0 H(Ga1) 0.005 0. 1 H(Ga2) 0.042 0.6 Equations of planes: Jrl * El * nZ = £, where X, Y, and are orthogonal angstrom coordinates derived as follows: I-XT r a 0 0 III = I 0 b 0 1 III 0 0 c Plane m n 1 0.2240 0.8857 0.4066 4.8355 2 -0.2331 0.8959 0.3783 2.9117 The angle between the plane normals is 26.5° 98 rings. The mean dihedral angles in each type of ring are compared with those obtained from energy minimization calculations (17) in Table 38. The conformational differences between the two types of five-membered rings are small yet the rings with the 'bridging' oxygen atoms (A rings) have a conformation which is closest to that calculated for w = 5.0° while those containing the 'terminal' oxygens (E rings) have a conformation nearest to that calculated for u)^ = 25.0°. Both ring types show some strain relative to the minimum energy conformations but the B rings show higher strain (4.0° r.m.s. deviation between ^ obs an^ "^calc compared to 2.4° for the A rings), this occurring primarily in the twist about the C-C bonds. Bond angles in the A rings range from 80.7(3)° at Ga to 117,9(6)° at 0 and in the B rings from 84.5(5)° at Ga to 115.9(6)° at 0 with mean values of 104.8 and 103.8° in A and B rings respectively compared to the calculated value of 104.2° (17) and observed values in BOCCN rings of 104.8° (Part 1) and 104.9° (Part 2). The C-0, C-C, and C-H distances are 1. 427(3), 1.507 (6), and1.477(2) A in the A rings and 1.412 (13), 1.534 (1), and 1.468 (4) A in the B rings. The differences in the bond lengths and angles in the two types of rings are a result of steric and electronic differences between the corresponding atoms in the ring, and to some extent are indicative of the charge distribution in the molecule. The mean C-0, C-C, and C-N distances in the two structures with BOCCN rings (Parts 1 and 2) are 1.416(3), 1.500 (6), and 1.488 (3) A. The mean N-C(methyl) distance of 99 1.164(4) A in the present structure is as expected. The mean angle at nitrogen is 109.4° but the individual angles all differ significantly from the mean and range from 100.5(4)° for Ga-N-C(B ring) to 113.0(7)° for Ga-N-C (methyl) . The angle at N between the A and B rings is 112.8(2)°. The distortion of the nitrogen tetrahedron results from steric constraints imposed by the fused-ring system. 0 The mean C-H distance of 1.00(13) A is as expected for X-ray data (61). All angles involving hydrogen atoms (R-C-H, R = N#0,C,H) are within three standard deviations of the mean value of 109°. Figure 11 The packing arrangement viewed along c. The crystal structure consists of discrete [ CH3N (CH2 CH20)2GaH ]2 molecules which are separated by normal i 100 Figure 12 The packing arrangement viewed along b. van der Waals distances, the shortest of which are listed in Table 39, except for one C-H...0 interaction (C...C = 3.14(1) • A) which may correspond to a weak hydrogen bond. There are also two possible intramolecular C-H...0 hydrogen bonds present (related by the two-fold rotation axis) which are indicated by broken lines in Figure 9. The geometrical data for these C-H...0 interactions are given in Table 39. The asymmetry introduced by the intermolecular 0 (4) . . .H (3B)-C (3) interaction is a reasonable explanation for the difference between the C(4)-0 (2), 1. 399, and C(9)-0(4), 1 . 424 A, bond distances (which represents the largest deviation from symmetry in the molecule). The non-bonded contacts in the gallium coordination spheres and other intramolecular non-bonded contacts which correspond to steric interactions within the molecule are also listed in Table 39. 101 Table 39 (a) Selected intra- and intermolecular contacts Intramolecular Intermolecular* Atoms distance Atoms distance C(2) . • • C(3) 2. 99 (2) 0 (4) . ..C (3) i 3.44 (D C (7). • • C(8) 3. 00 (D Ga(1) ...H (10C)2 3.4 1 (12) C(4) . • * C(5) 3. 04 (2) Ga(2) ...H(6A) 2 3. 30 (12) C (9) . • • C(10) 3. 07 (2) Ga (2) . . H (5B) 3 3. 46 (18) C(2). H (3A) 2. 52 (12) Ga (2) ...H(2A) * 3.48 (15) C (7) . * • H (8A) 2. 58 (11) 0 (1) . ..H (6 A) 2 2, 50 (12) C (3). • • H (2B) 2. 55 (11) 0(2) . ..H(10B) s 2.57 (17) C(9) . • * H(10B) 2. 68 (18) C (4) . . . H (9 A) 5 2.56 (11) H (2B) • • .H (3A) 1. 84 (17) C(9) . ..H(4A) 3 2.79 (8) H(7B) • • .H (8A) 2. 20 (16) H (1B) .. .H(9B) i 2. 40 (14) H (9A) • • . H (10B) 2. 10 (19) H (4A) . . . H (9A) 5 1.90 (15) (b) Gallium coordination sphere Atoms distance Atoms dista nee Ga (1) 0(1) 0 (1) 0(2) 0(2) 0 (1) 0(2) 0(3) H (1) . .Ga (2) .0(2) • N(1) .0(3) .H (Ga1) .H (Ga1) .H (Ga1) . H (Ga1) 3.038 (2) 3.273 (8) 2.678 (10) 2.787 (9) 2.716 (9) 2.82 (8) 2.86 (9) 2.63 (8) 2.87 (9) 0(1) 0 (3) 0 (3) 0 (4) 0 (4) 0 (3) 0 (4 ) 0 (1) N (2) .0(3) .0(4) • N (2) • 0(1) .N (2) . H (Ga2) .H (Ga2) . H (Ga2) .H (Ga2) 2.442 (8) 3. 275 (9) 2.695 (10) 2. 785 (9) 2.728 (9) 2.92 (7) 2.90 (8) 2.73 (7) 2.86 (7) (c) Possible C-H...0 hydrogen bonds D- H . . . A H. .. A D...A /DHA /X AH C (7) - H (7B) . .. 0 (2) 2.56 (12) 3.14(1) 129 (10) C (2)-H (2B)...0(4) 2.55(11) 3.13(1) 118(8) C (3)-H(3B) . . .0 (4) i 2.43 (1 1) 3.44(1 ) 160 (8) 82 (3) , 158 (3) 88 (3) , 154 (2) 144 (3) ,10 1 (3) The H...0...H angle at 0(4) is 59(3)° •Superscripts refer to atoms at positions: 1 1/2-x ' 2 1-x 3 X -1 1-1/2 1 z-1/2 3/2-z 1 + z 4 5 1-x 1/2+y, 3/2-z x y z-1 102 The infrared spectrum of the title compound in benzene solution showed a very strong Ga-H stretching absorption at 1900 cm-1 with a weak shoulder at 1810 cm-1. A medium intensity band at 770 cm-1 is assigned to the Ga-H wagging mode. The low frequency spectrum displayed a number of absorptions attributable to * Ga-O1, 'Ga-N * and ring modes (615 sh, 595 vs, 540 vs, 510 vs, 420 s, 390 s, 380 sh) but no assignment of this part of the spectrum is attempted at this time. Coates and Hayter (57), by chemical tests, postulated that dimerization in [MegNCR^C^OGaMeg ]g probably occurs via a four-membered Ga202 ring leaving the gallium atoms four-coordinate and the normally stronger nitrogen donor atoms not utilized in coordinate-type bonding. It is tempting, as a result of the present study, to postulate that in the above dimer five-coordinate gallium atoms might again be featured, the bonding about the metal atoms again involving both nitrogen and oxygen atoms to give a fused-ring system. This possibility is under study for the analogous gallane dimer, [ Me2NCH2CH2OGaH2 ]2 . 103 PART 5 CRYSTAL AND MOLECULAR STRUCTURE CF (PENTAHAPTOCYCLOPENTADIENYL) HYDRIDOMOLYBDENUM-/--DIMETHYLALUMINUM->5t-[METHYLALUMINUfi-DI- (/<-PENTAHAPTO (MONOHAPTO) CYCLOPENTADIENY.L) DIMETHYLALUMINUM ] (PENTAHAPTO CYCLOPENTADIENYL)HYDRIDOMOLYBDEN UM 104 INTRODUCTION An earlier report (67) indicated that slow decomposition of the adduct CP2M0H2.AlMe^ occurs in benzene solution at room temperature. Methane is liberated and eventually a solid is deposited from solution. From one such solution a small amount of crystalline material was produced suitable for X-ray analysis and an investigation was carried out to determine the extent of the expected Mo-Al network in the crystals. The novel* structure which resulted (shown in Fig. 13) contained two molybdenum and three aluminum atoms per molecular unit. EXPERIMENTAL The small amount of crystalline material deposited as a result of the slow methane elimination from benzene solutions of the parent compound, Cp2M0H2.AlMe^, was sufficient only for the crystal structure investigation and consequently no chemical analyses are reported. The molecular formula given in the title was derived from the experimental X-ray data collected on the sample. The air-sensitive crystals were mounted in glass 1 During the preparation of this thesis a preliminary report of this structure by Dr. C. K. Prout and co-workers appeared in Chem. Comm., 426 (1973). Correspondence with Dr. Prout, who will in the future publish an account on both this structure and that of the symmetric complex [ (C^H/j.) ^MoH]2Al£j,Me6, is acknowledged.' 105 capillary tubes under a nitrogen atmosphere and subsequently sealed off. An irregularly shaped crystal with dimensions of ca. 0.15 x 0.15 x 0.15 mm was mounted with the [0 11] vector parallel to the goniostat axis. Unit-cell and space group data were obtained from film and diffTactometer measurements. The unit-cell parameters were refined by a least-sguares treatment of sin2 6 values for 30 reflexions measured on a diffractometer with Mo radiation. Crystal data are: c25H35Al3Mo2 f*w* = 608.a Orthorhombic, a = 19.398 (4), b = 14.438(9), c = 9.035 (2) k, V = 253 1 (2) A3, Z = 4, Dx = 1. 597 (1) g cm"3, F(000) = 1232 (20° C, Mo K*, 7\ = 0.71069 A, /<: = 10.9 cm-»). Absent reflexions: hOO, h # 2n, OkO, k * 2n, and OOi, JL * 2n define k uniquely the space group V2^2^2^ (P2 , No. 19). Intensities were measured on a Datex-automated General Electric XRD 6 diffractometer, with a scintillation counter, Mo K<* radiation (zirconium filter and pulse height analyser), and a f)-26 scan at 2° min-1 over a range of (1. 80 + 0.86 tan 9) degrees in 20, with 20 s background counts being measured at each end of the scan. Data were measured to 29 = 45° (minimum interplanar spacing 0.93 A). Later data for /. = 0 to 7 were collected between 29 = 45 and 50° (minimum interplanar spacing 0.84 A). Data collection in the 20 = 45-50° shell was discontinued at £ = 7 due to a very low percentage of reflexions with I > 3cr(I). A check reflexion was monitored every 40 reflexion throughout the data collection. The intensity of the check reflexion remained within 1095 of its 106 original value during the data collection, the final measurement giving 95% of the original count, Lorentz and polarization corrections and check reflexion scaling were applied in deriving the structure amplitudes. No absorption correction was made in view of the relatively small value of JU+ Of the 2352 independent reflexions measured, 1113 had intensities less than 3<r[l) above background where <r2 (I) = S + B + (0.03S)2 with S = scan count and B = time averaged background count. These reflexions were not included in the refinement. Structure Analysis The positions of the two molybdenum atoms were determined from the three-dimensional Patterson function. One cycle of isotropic full-matrix least-squares refinement gave R 0.25. A subsequent difference map revealed three large peaks, two of which were clearly the bridging aluminum atoms. The third peak was thought to be anomalous at the time and was left out of the calculations. The molybdenum and two aluminum atoms were refined isotropically for one cycle and a second difference Fourier was calculated. The R factor at this point was 0.200. The difference map showed the same large peak as the previous one, which was deduced to be a third aluminum atom, as well as probable positions for sixteen carbon atoms. The molybdenum atoms were then refined anisotropically and the three aluminum and sixteen carbon atoms isotropically for one cycle, giving R 0.130. After one additional cycle of refinement and difference Fourier 107 synthesis all 25 carbon atoms had been located. Refinement with anisotropic carbon atoms gave an R value of 0.051 but three carbon atoms had non-positive definite temperature factors. Since the number of strong reflexions was relatively low it was decided to refine the structure with isotropic thermal parameters for the carbon atoms. Hydrogen atom positions were calculated with C-H = 0.97 0 A for the methyl and cyclopentadienyl groups. The hydrogen atoms were assigned isotropic temperature factors o approximately 1.5 A2 larger than the mean B for the carbon atom type to which they are bonded. Difference maps did not unambiguously reveal the position of the molybdenum hydrogen atoms. With the 33 methyl and cyclopentadienyl hydrogen atoms fixed, the remainder of the structure was refined to convergence with the carbon atoms isotropic, Mo and Al atoms anisotropic. The final agreement factors were R 0.066 and Rw 0.063 for 1213 reflexions with I > 3<r(l) . The absolute configuration of the complex (for the particular crystal used) has been determined through the anomalous scattering of the molybdenum and aluminum atoms. Enantiomorph (A) is represented by the coordinates in Tables 40 and 42. Enantiomorph (B) was generated by changing the x coordinates of (A) to 1-x. (B) was then refined to convergence and Hamilton's test (24) was applied to the resulting R factor ratios. The results, summarized in Table 43, indicate that enantiomorph (A) is most probably the correct absolute configuration, assuming the data to be free 108 of systematic error. The scattering factors of ref. 12 were used for the non-hydrogen atoms and those of ref. 13 for the hydrogen atoms. Corrections for anomalous scattering have been made for the molybdenum and aluminum atoms(13). A standard errors weighting scheme was used (see Part 3) giving constant average values of w (Fo -Fc)2 over ranges of |Fo| in the final stages of refinement. On the final cycle of refinement the largest parameter shift was 0.310"; Final positional parameters appear in Table 40 and thermal parameters in Table 41. The calculated positions of the hydrogen atoms and their assigned temperature factors appear in Table 42. In the final stages of refinement 26 reflexions believed to be suffering from counter errors or which had ratios of greater than 10:1 between the two background counts were given zero weight. Observed and calculated structure amplitudes are available on request. RESULTS AND DISCUSSION Bond distances and angles appear in Tables 44 and 45 respectively. Weighted least-squares mean planes are given in Table 46 and some important non-bonded contacts in Table 47. Table 48 gives structural data for related molybdenum cyclopentadienyl complexes. Stereoscopic views of the structure viewed along the c and b axes are shown in Figures 14 and 15. The crystal structure consists of discrete molecular 109 Table 40 Final positional parameters (fractional x 10* ) with estimated standard deviations in parentheses Atom X 2 z Mo (1) 4031 (D 9007 (D 5698 (2) Mo (2) 2971 (D 6488 (D 3182 (2) Al (1) 3720 (2) 8025 (4) 3296 (7) Al(2) 3440 (3) 7152 (4) 6166 (6) Al (3) 5033 (3) 7308 (4) 2544 (8) C (1) 3610 (8) 8777 (11) 1476 (20) C (2) 2652 (9) 7157 (12) 7586 (21) C(3) 4111 (9) 6284 (11) 7097 (19) C (4) 53 11 (12) 7847 (17) 585 (30) C(5) 5728 (10) 6463 (15) 3457 (24) C (11) 4825 (8) 8301 (12) 4091 (20) C (12) 4980 (10) 8119 (14) 5585 (26) C (13) 5159 (10) 8891 (15) 6371 (22) C (14) 5093 (11) 9578 (15) 5439 (27) C (15) 4895 (9) 9309 (12) 4061 (22) C (21) 4121 (8) 6588 (12) 2430 (18) C (22) 3675 (9) 6443 (14) 1167 (20) C (23) 3358 (11) 5571 (15) 1344 (24) C (24) 3548 (10) 5133 (13) 2729 (24) C(25) 4013 (9) 5804 (11) 3290 (20) C(3 1) 2859 (10) 9264 (12) 6137 (20) C (32) 3082 (12) 9808 (15) 494 1 (25) C (33) 3592 (12) 10424 (17) 5446 (30) C (34) 3646 (12) 10331 (17) 6871 (3 1) C (35) 3220 (11) 9617 (15) 74 15 (25) C (4 1) 1845 (11) 6082 (16) 3250 (30) C (42) 1906 (10) 6793 (14) 4219 (25) C (43) 2123 (12) 7557 (15) 3564 (26) C (44) 2152 (10) 7337 (13) 2022 (25) C(45) 1967 (11) 6402 (15) 1954 (26) 110 Table 41 Final thermal parameters and their estimated standard deviations 0 (a) Anisotropic thermal parameters (U^^ x 100 A2) Atom U 11 "22 u33 2l2 2l 3 y 23 Mo(1) 2. 7(1) 2.8(1) 3.5 (1) 0.3 (1) 0 .3 (1) -0. 5 (1) Mo (2) 2. 4 (1) 3. 2(1) 3.9(1) -0.4(1) -0 .1(1) -o. 2(1) aid) 3. 1 (3) 2.6 (3) 3.4 (4) -0.1 (2) 0 .0 (3) 0. 3 (3) Al<2) 3. 9 (4) 5.0 (4) 2.4 (4) 0.7(3) 0 .3(3) 1. 1 (3) Al(3) 3. 2(3) 5.6(4) 6.5 (4) -0.4 (3) 1 .6 (3) -2. 0 (4) (b) Isotropic thermal parameters A torn B (A2) Atom B (A2) C (1) 3.3 (4) C(23) 5.0(5) C(2) 3.3 (4) C (24) 4.2 (5) C (3) 3.4 (4) C(25) 3.1(4) C(4) 7.1 (6) C (31) 3.4 (4) C (5) 5.5 (5) C(32) 5.4(6) C(11) 2.6 (4) C (33) 6.5 (6) C (12) 4.5 (5) C(34) 6.8(6) C(13) 4.4 (5) C (35) 5.0 (5) C (14) 4.9 (5) C (41) 5.8(6) C(15) 3.4 (4) C (42) 4.6 (5) C (2 1) 2.7 (4) C(43) 5.8(6) C(22) 3.9(4) C (44) 4.5 (5) C(45) 5.1 (5) ( 111 Table 42 (a) Calculated hydrogen atom positions* (fractional x 10*) and assigned isotropic temperature factors Atom X 2 z B (A *) H (1 A) 3164 9078 1488 5.0 H (1B) 3971 9244 1440 5.0 H(1C) 3646 8379 616 5.0 H (2A) 2290 7572 7195 5.0 H (2B) 2475 6554 7715 5.0 H (2C) 2808 7417 8529 5.0 H {3A) 4507 6213 6454 5.0 H (3B) 4256 6531 80 4 4 5.0 H (3C) 3892 5687 7242 5.0 H (4A) 5718 8229 720 7.5 H (4B) 5413 7349 -98 7.5 H (4C) 4938 8221 201 7.5 H(5A) 6135 6817 3719 7.5 H (5B) 5533 6179 4328 7.5 H(5C) 5852 5987 2743 7.5 B (12) 4965 7499 60 12 5.5 H (13) 5300 8922 7400 5.5 H (14) 5177 10220 5704 5.5 H(15) 4817 9712 3216 5.5 B (22) 3605 6870 349 5.5 H (23) 3 051 5296 617 5. 5 B (24) 3399 4547 3149 5.5 H(25) 4243 5719 4232 5.5 H (31) 2527 8758 61 10 6.5 H (32) 2917 9764 3930 6.5 fl (33) 3856 10854 4843 6.5 B (34) 3945 10705 7501 6.5 B (35) 3 175 9405 8432 6.5 H (41) 1723 5446 3483 6.5 B (42) 1808 6742 5264 6.5 H (43) 2233 8144 4034 6.5 B (44) 2277 7747 1212 6.5 H (45) 1935 6046 1038 6.5 continued... 112 (b) Cyclopentadienyl ring centroid coordinates (fractional x 104) Ring x y z R (1) R (2) R(3) R (4) 4990 3743 3280 1999 8840 5908 9889 6834 5109 2192 6162 3002 * The hydrogen atoms are labelled as fellows: the cyclopentadienyl hydrogens have the same number as the carbon atom to which it is bonded, e.g. H(12) is bonded to C(12); the methyl hydrogens are denoted by a numeral referring to the carbon atom to which it is bonded and by A, E, or C to distinguish between the three different hydrogens associated with each carbon. 113 Table 43 Results of Hamilton's Test Parameter compared Value for enantiomorph Sig. (A) (B) (E/A) level* Conventional R (3<r data) 6.56 6 .58 1. 0030 97. 5 Conventional B (all F) 14.34 14 .39 1. 0035 99. 5 Weighted R (3<rdata) 6.31 6 .32 1. 0022 95.0 Weighted 1 (all F) 6.51 6 .52 1. 0021 99. 5 1 This is the % probability that enantiomorph (A) is the correct absolute configuration. 1 1 4 units with normal van der Waals contacts between units. The closest intermolecular contacts, including those fer hydrogen atoms in calculated positions, are listed in Table 47. Figure 13 A stereoscopic view of the C25H3^Al3Mo2 molecule. 50% probability ellipsoids are shown for Mo and Al atoms. Carbon atoms are represented by equivalent spheres. The molecular structure exhibits several unusual and novel features. The three aluminum atoms in the molecule are of different structural types, one of them, Al(2), was of the predictable dimethylaluminum type, bridging two molybdenum atoms. The distances Mo(1)-Al(2) and Mo(2)-Al(2) are 2.944(6) and 3.003(6) A. The Mo (1)-Al (2) - Mo (2) angle is 106.2(2)° while the opposite angle C (2)-Al (2)-C (3) is 103.9(7)°. Thus the coordination about Al(2) is that of a distorted tetrahedron. Other angles at Al(2) range from 110.5 to Figure 14 The structure viewed down c. Figure 15 The structure viewed along b. 116 Table 44 Bond lengths (A) with estimated standard deviations in parentheses Atoms distance Atoms distance MO (1 ; -Al(1) 2. 662 (6) Al(1) -C (1). 1.98 (2) Mo (2) -Al{1) 2.655 (5) Al (1) -C (11) 2.30 (2) MO (1 ] -Al (2) 2. 944 (6) Al (1) -C (21) 2.35 (2) Mo (2 -Al (2) 3.003 (6) Al (2) -c (2) 2.00 (2) MO (1) -C (11) 2. 35(2) Al (2) -c (3) 1.99 (2) Mo (1 ] -C (12) 2. 25 (2) Al (3) -c (4) 2.01 (3) MO Mi -C (13) 2.28 (2) Al (3) -c (5) 2.00 (2) MO (11 -C (14) 2. 23 (2) Al (3) -c (11) 2.04 (2) Mo -C (15) 2.28 (2) Al (3) -c (21) 2.05 (2) Mo -C (31) 2. 34 (2) C (1 1] -c (12) 1.41 (3) Mo (i: -C (32) 2.28 (2) C (12] -c (13) 1.37 (3) Mo (1 i -C (33) 2. 23 (2) C(13) -c (14) 1.31 (3) Mo (1) -C (34) 2.31 (2) C (14] -c (15) 1.36 (3) MO -C (35) 2. 38 (2) C (15] -c (11) 1.46 (2) Mo (2) -C (21) 2.34 (2) C (21 ) -c (22) 1.45 (2) Mo (2 l-C (22) 2. 28 (2) C(22) -c (23) 1.41 (3) Mo (2) -C (23) 2.25 (2) C (23) -c (24) 1.45 (3) Mo <2' l-C (24) 2. 29 (2) C (24) -c (25) 1.42 (2) Mo (2) -C (25) 2.25 (2) C (25] -c (21) 1.39 (2) Mo (2 |-C (41) 2. 26 (2) C(31) -c (32) 1.40 (2) Mo (2) -C (42) 2.31 (2) C (32] -c (33) 1.4 1 (3) MO (21 l-C (43) 2.28 (2) C (33) -c (34) 1.30 (3) Mo (2] -C (44) 2.26 (2) C (34] -c (35) 1.41 (3) Mo (2 l-C (45) 2. 25 (2) C (35) -c (31) 1.44 (3) Mo (1) -R (1) 1.95 C (41] -c (42) 1.35 (3) Mo (2 |-R (2) 1.94 C (42] -c (43) 1.32 (3) Mo (1] -R (3) 1.98 C (43] -c (44) 1.43 (3) Mo (2 |-R (4) 1. 96 C (44] -c (45) 1.40 (2) C (45' -c (41) 1.28 (3) 1 1*7 Table 45 Bond angles (deg) with estimated standard deviations in parentheses atoms angle atoms angle ai(i) -Mc(1) -Al(2) 62. 9(2) C 15) -C (11) -C (12) 101 2) al (1) -Mo (1) -E (1) 85. 9 C i 15) -C (11) -Al (1) 105 D ai(i) -Mo (1) -R (3) 110. 4 C i 15) -C (11) -Al (3) 132 ( D Al(2) -MO (1) -B(1) 107. 3 c (12) -C (11) -Al (1 ) 1 18 D Al (2) -Mo (1) -R (3) 105. 5 C ( 12) -C (11) -Al (3) 1 19 ( 1) R(1)- Mo (1)- R(3) 147. 1 Al (D -C (11) -Al (3) 81. 3 (6) Al (1 ) -Mo (2) -Al(2) 62. 1 (2) C ( 25) -C (21) -C (22) 104 ( 2) aid) -Mo (2) -R (2) 87. 5 C [25] -C (21) -Al (1 ) 1 19 1) Al{1) -MO (2) -R (4) 108. 5 C | 25) -C (21) -Al (3) 121 1) Al(2) -Mo (2) -R (2) 108. 6 c I 22) -C (21) -Al (1) 10 1 | 1) ai(2) -MO (2) -R(4) 106. 6 c (22] -C (21) -Al (3) 129 D H (2)-Mo (2)- R (4) 144. 8 Al (D -C (21) -Al (3) 79. 6 (6) C{1)- A1(D- Mo (1) 1 14. 2 (5) C (11] -c (12) -C (13) 1 14 2) c (1)- Al (1)-•Mo (2) 111. 5 (5) c I 12) -c (13) -C (14) 105 ( 2) C(1)- al(D- C(11) 105. 4 (7) c 13] -c (14) -C (15) 1 14 2) C (1 )- Al (1 )- C{21) 104. 1 (7) c ( 14] -c (15) -C (11) 107 2) Ho (1) -Al(1) -Mo (2) 126. 9 (2) c I 21) -c (22) -C (23) 107 | 2) Mo(1) -Al (1) -C(11) 56. 0 (4) c (22] -c (23) -C (24) 112 (2) Ho (1) -Al (1) -C (21) 131. 8 (5) C ( 23) -c (24) -C(25) 100 2) Mo (2) -Al (1) -C (11) 131. 9 (5) c (24] -c (25) -C (21) 1 17 (2) Mo (2) -Al (1) -C (21) 55. 2 (4) c I [35) -c (31) -C (32) 106 2) C (1 1) -Al (1) -C(21) 87. 1 (6) c 31] -c (32) -C (33) 109 2) MO (1 ) -Al (2) — Mo (2) 106. 2 (2) c ( 32) -c (3 3) -C (34) 108 2) Ho(1) -Al (2) -C(2) 112. 0 (6) c [33] -c (34) -C (35) 112 2) MO (1) "Al (2) -C(3) 112. 3 (5) c (34] -c (35) -C (31) 105 (2) Ho (2) -Al (2) -C (2) 110. 5 (6) c 45) -c (41) -C (42) 108 2) Mo{2) -Al (2) -C(3) 112. 1 (6) c 41 ] -c (42] -C (43) 112 (2) C (2)- Al (2)- C (3) 103. 9 (7) c I 42] -c (4 3) -C (44) 105 2) C (4)- Al(3)- C(5) 114. 8 (10) c (43] -c (4 4) -C (45) 104 (2) C (4)- Al (3) -•C(11) 1 12. 6 (9) c I 44) -c (45) -C (41) 111 ( 2) C(4)- Al(3)- C(21) 112. 6 (9) C (5)- Al (3)- C(11) 106. 2 (8) C(5)- Al (3)- C(21) 107. 1 (8) C (11) -Al (3) -C(21) 102. 7 (7) 118 112,3°, the mean angle at Al(2) being 109,5°. The two remaining aluminum atoms are involved in the unique structural feature of this system. Instead of a second bridging AIMe2 unit an AlMe group bridges the two molybdenum atoms and at the same time is involved in a novel multicentre ; bonding arrangement with the two unique carbon atoms of the C^Hjj, groups, C(11) and C(21), and the remaining aluminum atom, Al(3), which occurs as an AlMe2 unit. The two Al(1)-Mo o distances, 2. 662 (6) and 2.657 (5) A, are equal to within experimental error. The fact that these distances are more 0 than 0.3 A shorter than the corresponding Al(2)-Mo bonds has interesting structural implications which will be discussed. The Al2Mo2 bridging system is significantly non-planar (see Table 46). The angle between the normals to the two AlAlMo planes is 168.9°. The Al(1)-Al(3) and Mo(1)-Mo{2) distances are 2.935 (8) and 4.757 (2) A respectively, neither of which represents any direct interaction. The remaining angles in this system are Mo (1)-Al (1)-Mo (2) , 129.9(2), Al(1)-Mo(1)-Al(2), 62.9(2), and Al (1)-Mo (2)-Al (2) , 62.1(2)°. Bond angles at Al(1) involving the two molybdenum atoms, C(1), and Al (3) have a mean value of 108.2°. This is indicative that Al(1) is sp3 hybridized with three hybrids nearly parallel to the two Al(1)-Mo and Al(1)-C(1) bonds, and the remaining hybrid, which is involved in the multicentre bonding, directed toward Al(3). The Al (1) , C(11), C(21), Al{3) multicentre bonding arrangement resembles that in the trimethylaluminum dimer (68), although closer examination 119 reveals unique differences. The [(CH^^Alj^ structure is centre-symmetric with a planar bridging arrangement; the two independent Al-C (bridge) distances are 2.125 and 2.123 A, and a Al-C(terminal) are 1.949 and 1.956 A. The angles in the bridge portion are 75.7° at C and 104.3° at Al. The bridge system in the present structure is non-planar (see Table 46), the angle between the two AlAlC planes is 149.7°, and also asymmetric with short bonds to Al(2), mean Al (2)-C (bridge) = 2.05 A, and long bonds to Al(1), mean Al (1) -C (bridge) •= 2.33 A. The angles in the bridge are 87.1(6) at Al(1), 102.7(7) at Al(3), 79.7(6) at C(21), and 81.3(6)° at C (11) . Figure 16 shows a schematic representation of the atomic orbitals believed to be involved in the multicentre bonding: one sp_2 hybrid orbital from each of C(11) and C(21), one s_p3 hybrid orbital from Al(1) and two S£3 hybrid orbitals from Al(3), Note that Al(1) lies twice as far from the mean planes of the C5 Hjj. rings (represented by the horizontal dotted lines in Figure 16) as does Al(3). The bonding scheme represented by Figure 16 is adeguate to explain the observed geometry of the system, particularly the difference between the Al(1)-C (bridge) and Al (3)-C (bridge) distances. The coordination about Al(3) is a somewhat distorted tetrahedron, with the angle C (4)-Al (3)-C (5) expanded to 114.8(10)° corresponding to the contraction of the opposite angle, C (11)-Al (3)-C (21), to 102.7(7)°. Other angles at Al(3) range from 106.2 to 112.6°, and the mean of all angles at Al(3) is 109.3°. None of the five Al-C (terminal) distances o differs significantly from the mean value of 2.00(1) A, which 120 Aid) C(R1) AK3) Figure 16 A representation of the bonding in the Al (1) -C (1 1) -C (21)-A1 (3) bridging system. Mean bond distances are shown. is equal to the sum of covalent radii. The two C5H5 aDC^ two C^H^ groups are all fientahapto to the molybdenum atoms, and, assuming that one hydrogen atom is also coordinated to each of the molybdenum atoms, the latter obey the 18-electron rule. If the Ccj H«j and C^H^ groups are regarded as formally negatively charged and occupying three coordination sites at the metal atom, the complex may be regarded as a nine-cccrdinate complex cf Mo (II) (assuming the H atom is a one electron donor). The recently reported structure of the niobccene dimer [ (C^H^) (C^H^JNbHJg (69) also contains IHonoha^to and pentahapto Cc Hu ligands. The present 121 Table 46 Weighted least-squares mean planes 0 (a) Distances (A) of relevant atoms from the mean planes Atom d Atom d d/<r Plane 1 : C (11 )-C (15) Plane 3: C (31) -C (35) C(11) 0.011 0.7 C(31) 0.012 0. 6 C(12) -0.019 1.0 C (32) -0.024 1. 1 C(13) 0.011 0.6 C(33) 0. 028 1. 1 C(14) 0.000 0.0 C (34) -0.013 0. 6 C (15) -0.008 0.4 C(35) -0.005 0.2 Mo(1) -1.951 1296.3 Mo (1) 1.976 1304.8 Plane 2: C (21)-C (25) Plane 4: C(41)-C(45) C(21) 0.008 0.5 C (41) -0.024 1. 2 C (22) -0.014 0.8 C (42) 0.024 1. 2 C(23) 0.016 0.8 C(43) -0.023 1.0 C{24) -0.006 0.3 C (44) 0.007 0. 3 C (25) -0.003 0. 2 C(45) 0.008 0.4 Mo (2) - 1.934 1278.7 Mo (2) -1.956 1339.1 Plane 5: Mo(1S2), Al(1S2) Plane 6: Al(1&3), C(11&12) Ho (1) -0.017 11.5 Al(1) 0.033 5.7 Mo (2) -0.017 11.3 Al (3) 0.064 . 9. 4 Al(1) 0.240 49.4 C(11) -0.390 22.2 Al(2) 0. 176 30.8 C (21) -0.375 22. 3 continued. 122 (b) Equations of planes: /X + mY + nZ = £, where 1, Y, and Z are orthogonal angstrom coordinates derived as follows: IH = LZJ r a 0 | 0 b L 0 0 0 i TXT 0 I m c J Plane / m n 1 0.9.605 -0.1070 -0.2570 6.7463 2 0.7566 -0.4539 -0. 4707 0.6892 3 0.7270 -0.6674 -0.1612 -5.8012 4 -0.9584 0.2596 -0. 1189 - 1. 4746 5 0.8852 -0.4608 -0.0635 0. 6 188 6 -0.0060 0.5164 -0.8563 -3. 3572 (c) Angles between plane normals Planes angle Planes angle Planes angie (1)-(2) 154 (2)-(3) 158 (3)- (5) 164 (1)~(3) 144 (2)- (4) 142 (3)- (6) 102 <1)~(4) 157 (2)-(5) 155 (4)- (5) 164 <D-(5) 156 (2)-(6) 81 (4)- (6) 104 (D-(6) 81 (3)-(4) 148 (5)- (6) 101 123 Table 47 Selected intra- and intermolecular contacts Intramolecular Intermolecular* atoms distance Atoms distance MO (1) • • .Mo (2) 4.757 (2) C(13) .. .C( 14) l 3.29(3) AKD • * -Al (2) 2.935 (8) C(1). • . H (3 5) 2 3.02 Al{1) • • . Al (3) 2.831 (8) C (2). . . H (22) 3 3.14 C (11) • • .C (21) 3.20 (2) C(3) . . .H (14) * 2.87 C(15) • * .C (33) 3.25 (3) C (23) ...H (41) s 2.98 C (23) .C (45) 3.00 (3) C (4 1) ...H(23) 6 2.93 C(33) • • .C (14) 3. 16 (3) H (2B) ...H (24) 6 2.36 C (34) • • .C (14) 3. 28 (3) H (2C) .. .H(22) 3 2.39 C(2) . • • 8(31) 2.66 H (3C) . . . H (4 1 ) 6 2.31 C (2). • • H (42) 2.73 C(3) . • • H(12) 2.60 C (3) . • • H (25) 2.73 C(31) * * .H(43) 2.77 H (1B) • « . H (15) 2.39 H(1C) • * .H (22) 2.19 H (2A) w m .H (31) 2.07 H(2A) w m .H (42) 2.31 H (3a) • • .H (12) 2. 10 H (3a) • • .H (25) 2.19 H (31) • • . H (43 2. 15 •Superscripts refer to atoms at positions: 1 1/2+x 3/2-2 1"2 * 1-x .X-1/2 3/2-z 2 x j z-1 s 1/2-x 1-2 z-1/2 3 x 2 z+1 6 1/2-x 1-2 1/2+z 124 Table 48 Structural data for some molybdenum cyclopentadienyl complexes Compound Mo-C (Cp) C-C (Cp) Ko-centroid a 2.285 1.389 1.96 b 1.40-1.44 c 2. 333 1.413 d 2.333 1.412 e 2. 310 1.378 f 2.32-2.68 1.347-1.427 g 2.34 1.41 2.08 h 2.30 1.97 i 2. 35 1.42 2.01 j 2.338 1.418 2.00 k 2. 324 1.421 1 1.39 2.02 m 2.289 1. 425 1. 94 n 2.345 1.416 o 2. 38 1.43 2.04 P 2.329 1.391 2.00 g 2.253-2.368 1. 385 1. 999, 1.993 r 2.32-2.39 1.405 s 2.229-2.388 1.396 1.976,2.002 t 1.41 u 2.244-2.396 1. 394 1.980,1.981 V 2.21-2.42 1.40 1.96-2.01 w 2.27-2.36 1.27-1.42 1.986,1.993 x 2.26-2.40 1.40 1.962,1.991 a. C25H35AI3M02' tnis work b. (C^Hj^) (C5H5) (CO) MoMn (CO)^ , ref. 72 c. (C5H5) Mo (CO) (PPh^)2 (NCO) , ref. 75 d. (C5K5) no (CO)2 (PPI13) I# ref. 76 e. Ko(CO) (Ph2PCB2CH2PPh2)Cl, ref. 76 f. (C5H5) 3M0 (NO) , ref. 77 g. Mo (C5H5) (CO) 2 (CH2SCH^) , ref. 78 h. [ Mo (C5H5) (SCH3) 2 12 ' ref« 79 continued... (C^I!^) Mo (CO)3CH2COOH, ref. 80 [PPhk ]+ [ (C5H5) Mo{ S2C2 (CN)2 }2 ]-, ref. 81 (C5H5) Mo (CO) (Ph2PCH2)2Cl, ref. 82 (C^H5)2MoS2C^Hk, ref. 83 (C^H^)2MoH2, ref. 71 [ Mo (CO) ^ ]2 , ref. 84 (C5H5 ) Mo (CO)3C2He,, ref. 85 [ (C5H5) Mo(CO)2 ]2 (H)[ P (CH3)2 ], ref. 86 (C5H5)2MoS2C6H3CH3 , ref. 87 C5H5 (CO) 2MoN (H) NC (C02C2H5) COH, ref. 88 (C5H5) 2MoS (CH2) 2NH2I, ref. 89 (C5H5 (CO) 2MoN. N (CH3) . C (C02C2H5) COH) PFg, ref. H[ (C5H5) 2MoNH2CH (CH2S) COO]Clf ref. 91 H[ (C5H5) 2MoNH2CH (CH2S) COO ]PF£, ref. 91 [(C5H5)2MoNH2CH2COO]Cl.H20, ref. 91 [ (C^H^)2MoHN(CH3)CH2COO ]Cl.CH3OH, ref. 91 126 structure again demonstrates the versatility of the C5H5 ligand in that the C^Hk groups derived therefrom are £§Si3liapto to a molybdenum atom and are involved via the unique carbon atom in multicentre bonding to aluminum atoms. The mean Mo-C distance is 2.285 A with individual 0 distances ranging from 2.23 to 2.38 A and the mean Mo-R (ring centroid) distance is 1.96 A. The four cyclopentadienyl rings are all planar to within experimental error (see Table 46). o The mean C-C bond length in the rings is 1.389 A and the mean C-C-C angle is, as expected, 108°. The Mo-C, C-C(cyclopentadienyl), and Mo-R distances are in good agreement with those of related compounds which are compiled in Table 48. The structure may be interpreted in terms of valence bond theory in a manner analogous to that described for the niobocene dimer (69). The latter approach views structures of this type of bis(cyclopentadienyl)-transition metal complex as having canted rings with three hybrid orbitals in the horizontal mirror plane (70) as shown in Figure 17. Some structures which can be rationalized by this scheme are given by Guggenberger (69). Both molybdenum atoms in the present molecule have Al (2) in the ^2 position and Al(1) in the position, the hydrogen atom is assumed to be in the ^3 position. The angles between the C^H^ and C^H^ planes are 32.9° at Mo(1) and 35.2° at Mo(2), which are similar to those in other molybdenum complexes, e.g. 34° in CpgMoHg (71) and 35° in (Cp) (CO) Mo (C^fy) Mn (CO)^ (72). The length of the two 127 Figure 17 Idealized structure of bis (cyclopentadienyl)-transition metal complexes with canted Cp rings. e Mo-Al(2) bonds (0.3 A longer than the Mo-Al(1) bonds) suggests the possibility of a M'o-H-Al(He)2-H-Mo bridging system analogous to the Ti-H-AlEt2 system in [ (Cp) (C^H^JTiHAlEtglg (73,74). The three aluminum and five methyl carbon atoms are approximately coplanar. The halves of the molecule with respect to this plane are not equivalent, the most interesting difference is that the C^H5 an(^ groups associated with Mo(1) are staggered while those at Mo (2) are eclipsed. This results from steric interactions between the Ring 3 and Ring 4 hydrogen atoms. The distance between 128 calculated positions for H(31) and H (43) is 2.15 A which is less than the sum of van der Waals radii. If the conformation of the rings were the same at both molybdenum atoms there would be even greater steric interference. Inspection of bond lengths and angles shows other small differences between the two halves of the molecule, some of which are significant. o The mean Mo-C (C5 Hij.) distances are the same, 2.28 A for each molybdenum atom, while the Mo-Al(2) distances differ by 10 standard deviations being 3.003 (6) A for Mo (2) and 2.944 (6) 0 for Mo(1). The mean Mo-C (Cp) also differ, being 2.31 A at Mo(1) and 2.27 at Mo (2). The corresponding angles at the two molybdenum atoms show significant differences as well, and may be caused by a small energy difference between the staggered and eclipsed conformations of the C^H^ and C^H^ rings. 129 PART 6 THE COMPUTER PROGRAM "SIGCOR" 130 INTRODUCTION This section of the thesis describes a computer program which calculates estimated valence bond orders given a set of atomic coordinates. The work is not yet completed but an operational version, which gives satisfactory results when the hybridization at both atoms in the bond involves cnly s and 2 orbitals, will be described. Instructions for the use of the program are given and the source deck may be obtained from the author. The bond order is derived from the fractional difference between the observed interatomic distance and the calculated single bond distance. The calculated value is based on the sum of covalent radii (92) corrected for (T hybridization effects (93-95) and in some cases for electronegativity effects (96). The dependence on electronegativities has not yet been completely worked out. The program provides qualitatively accurate information about the bonding and electron distribution in the structure. It is intended to serve as an aid in the comparison and analysis of structural information obtained from diffraction and spectroscopic experiments. GENERAL DESCRIPTION The program, written in FORTRAN 17, is divided into subroutines to facilitate modification and expansion. The main program performs most of the basic operations, while 131 subroutine PABSET is a library of covalent radii and electronegativities for a number of commonly occurring atoms. A list is included in the set of instructions. Subroutine ENCOR applies corrections to the covalent radius for electronegativity differences and for formal charges where appropriate. Subroutine BOND is the function which relates bond order to the relative contraction of the bond distance from its calculated value. Finally, subroutine ANGLE is optionally called to calculate and print both the observed valence angles and the calculated angles between the appropriate hybrid orbitals. The sequence of operations begins with reading the input information. The general atomic coordinates are transformed to orthogonal angstrom coordinates. The covalent radii and electronegativities are assigned by subroutine PARSET for atoms in the library or are read from cards with the atomic positions for atoms not in the library. The internal values may be altered for any particular atom (see instructions) . These values are stored in the arrays BAD (i) and CA (i) . Bonding information is read in and stored in the form of a symmetric connectivity matrix (97) KB(i,j) where KB(i,j) = 1 if the atoms i and j are bonded to one another and KB(i,j) = 0 otherwise. The hybridization states of "terminal" atoms (those which are univalent) cannot be calculated and are assumed to be S£3 . If the hybridization state of a terminal atom is known to be different from s£3, as in the case of carbonyl groups, this information is read from an optional terminal atom card. 132 The bond distances and direction cosines of the bonds are then calculated and stored in arrays DB, DL, CM, and DH. The next step is the determination of the fractional s characters, SF(i,j), for the hybrid orbital at atom i involved in the bond between atoms i and j. In general, an orthogonal set of non-equivalent hybrid orbitals which follow the bond directions cannot be constructed from s and 2 atomic orbitals only. Since the orthogonality conditions must be met, there are usually differences between the interhybrid angles and the observed valence angles if the hybrids are constructed only from s and 2 atomic orbitals. The general orthogonality conditions may be expressed as: aj_aj + b^bjcos = 1 where O^j is the angle between the non-eguivalent hybrids a^s • b^2 anfl aji? + b-jjO. It ^s assumed each of the functions is normalized. This requires: a z + b 2 = 1 in which case the fractional s and jg characters are simply a2 and b2. For divalent atoms it is assumed that the two bonding hybrids are equivalent. In this case the orthogonality condition becomes: 133 a2 • b2 (cos 9) = 0 where 9 is the bond angle. Since the function is normalized the fractional s character (a2) is given by: SF = cos 9/ (cos 9 - 1) trigonometric identities transform this expression to the equivalent form: SF = 1 - 0.5[csc2 (9/2) ] which is used in the program. The resulting hybrid orbitals follow the bond directions. For trigonally coordinated atoms the values of SF(i,j) are calculated using the same formula as for divalent atoms. In this case 9 is taken as the mean valence angle at atom i involving the bond i-j. This approach yields non-equivalent hybrid orbitals which satisfy the orthogonality conditions, implying that the total s character at a given centre must equal 1. This includes vacant, lone-pair, or Tf bonding ' orbitals for which the s character is not explicitly calculated but may be deduced. The calculated interhybrid angles are not generally the same as the observed angles, but deviations from ideal geometry are always in the same direction. The approach which gives the best agreement between 134 calculated interhybrid angles and bond angles for four-coordinate atoms is based on an initial assumption of threefold symmetry. The hybrid for which the s character is being calculated is assumed to be the unigue hybrid and the remaining three are treated as if they were equivalent. Let 9 be the mean bond angle at atom i not involving atom j. The s character in each of the artificially equivalent hybrid orbitals is given by the equation derived above. If we denote this quantity by x, orthogonality requires that: SF(ifj) = 1 - 3x and substitution of the value of x in the above expression gives: SF(i,j) = 1.5[csc2 (9/2) ] - 2 Except in cases of extreme steric distortion, application of the above equation to each of the bonds at a four-coordinate atom yields a set of orthogonal hybrid orbitals. When the sum of the s characters in the four hybrids differs by more than 2% from unity, a message to that effect is printed by the program. The most probable causes for such deviations are severe steric distortions and possible involvement of d (or f) orbitals in the makeup of the bonding hybrids, the latter being most likely for atoms beyond the first row of the periodic table. As for trivalent atoms, the hybrids generally do not follow the bond directions. 135 The dependence of the single bond distance on the amount of s character in the hybrid orbitals involved is a geometric factor, independent of the types of atoms involved. Since the greatest amount of experimental information is available for C-C bonds, they will be used as a standard. A plot of the percent contraction of the single bond distance (relative to an sp3-sp3 bond) versus the fractional s character of the bonding hybrids is shown in Figure 18. The data points correspond to single C-C bond distances of 1.537, 1.486, and 1.379 A for jsp3-sj33, sp2-sp2, and sp-sp bonds; giving the following relationship between the fractional s character, SF(i,j), and the fractional contraction of the covalent radius of atom i in the bond i-j, DELTA (i,j): DELTA (i, j) = 0.4112 (SF (i, j) - 0. 2500) The corrected single bond distance for the bond i-j is given by: SIGC0R(i,j) = (1 - DELTA (i, j) ) RAD (i) + (1 - DELTA (j , i) ) RAD (j) where RAD(i) and RAD (j) have been corrected for formal charges (such as quaternary N and B atoms) and electronegativity differences. The general relationship between bond contraction and bond order has again been based on the behavior of C-C bonds. The values of 1.537, 1.394, 1.335, and 1.206 A were used for 116 Figure 18 A plot of % contraction of C-C single bond distances (relative to an sp3- s_n3 single bond) vs. % s character in the bonding orbitals. 1 37 Bond Order -5 0 5 % Contraction Figure 19 A plot of valence bond order vs. % contraction of the internuclear distance relative to the corrected single bond distance. formal bond orders of 1, 1. 5 (aromatic) , 2, and 3 respectively. These values were fitted to the following function: x = 2exp(0.613 TC - 0.693) {TC < 0.47} X = 1 + 1.891 TC2 - 1.790 TC« + 0.887 TC* {TC > 0.47} where x is the bond order and TC is 10 times the fractional contraction of the bond length. Figure 19 shows a plot of 138 bond order vs. percent contraction calculated from the above eguation. The contraction is relative to the calculated single bond distance. The printed output begins with a listing of the control parameters followed by the transformation matrix and its inverse. The original input coordinates and the transformed coordinates are listed next. For each chemical bond the following information is given: the observed interatomic distance, the fractional s characters for both of the hybrids involved, the calculated single bond distance, the absolute (A) and percent contractions of the bond relative to the calculated value, the covalent radius used for each of the atoms, the calculated bond order, and finally the derivative e of the bond order with respect to a 0.01 A change in bond length. In summary, the total calculated bond order for the molecule (excluding bonds involving H) and the number of bonds included in the sum are given. For each ncn-hydrogen atom the coordination number, sum of calculated s characters, and sum of bond orders are printed. The observed valence angles and calculated interhybrid angles are then (optionally) listed. A sample of the printed output follows the symbolic program listing. 139 SIGCOB CALCULATION OF SIGMA HYBBIDIZATION EFFECTS AND APPROXIMATE BOND ORDERS FROM MOLECULAR GEOMETRY INPUT Card Ii Title (20A4) 1-80 general title card Card 2i Control Card (615) 1-5 NA number of atoms to be read in (max. 100) 6-10 ND number of cards in bonding array 1 1-15 NOUT = 0 for normal output, 3 to include angles 16- 20 NC = 0 for one card/atom, 1 for two cards 21- 25 NTAC number of terminal atom cards l 3: Cell Dimensions in BUCILS format (6F10.5) 1-10 a 11- 20 b 21- 30 C (A) 31- 40 alpha 4 1-50 beta 51- 60 gamma (degrees) 140 Card 4j_ Atomic £ositionsx covalent radiix and Sl§£tronec£ativities (5A2,20X,5F10.6) If NC = 0 1 card per atom NC = 1 2 cards per atom Covalent radii and electronegativities (EN) are stored internally for the following atoms: atom r EN atom r EN B f (EN) 2.01 Br 1. 14 2.74 C 0.768 2.50 I 1.33 2. 21 H 0.23 2. 20 Si 1. 17 1.74 0 0.652 3.50 Sn 1.40 1.72 N 0.701 3.07 Ge 1. 22 2.02 P 1.069 2.06 Sb 1.41 1.82 As 1.21 2.20 S 1.04 2.44 F 0.64 4.10 Se 1. 17 2. 48 Cl 0.99 2.83 Te 1. 37 2.01 These are set by the program if the atomic s ymb right justified in columns 3 and 4 of the coordinate card. For atoms in the above list the values punched in the covalent radius and electronegativity fields of the coordinate card are added to the library values. The above values may therefore be changed by punching the desired increment in the appropriate fields (see below). Covalent radius and electronegativity values must be given if the chemical symbol is not in columns 3 and 4 or if the atom is not included in the above list. 141 The first card for each atom must contain: 1-10 atom i.d., chemical symbol in cols. 3 and 4 31-40 x 41-50 y 51-60 z (fractional coordinates) 61-70 covalent radius or change in covalent radius 71-80 electronegativity The second card may contain any form of information (e.g. temperature factors) and are ignored by the program. Card 5:_ Bonding array (1613) ND cards of the form: 1-3 I number of reference atom in atoms list 4-6 JB (n) , n = 1, 15 7-9 numbers in atoms list of atoms bonded to 10-12 reference atom, if this number is less 13-15 than I then it should be left out. Each 16-18 bond is included once. etc. This corresponds to a matrix B (i,j). If the element B(if j) is non-zero, then the atoms i and j are bonded to each other. Card i in the bonding array input gives the values of j which correspond to non-zero elements in row i of the matrix. 142 Card 6i_ Terminal atom corrections (F10.6,2013) (Omit if UTAC = 0) The program assumes sp3 hybridization for all terminal atoms. If this is not the case these cards are used to set the fractional s character of the atoms in question. SCT is added to .0.2500 to give the desired fractional s character. 1-10 SCT 11-13 numbers in atoms list of terminal 14-16 atoms with hybridization corresponding 17-19 to the value of SCT in col. 1-10 etc, (up to 20 atoms) For , example, if SCT = 0.08333 then the atoms corresponding to the numbers punched in columns 11-13 etc. will have sp2 hybridization states. Execution time on the IBM/370 are on the order of 0.5 s for a typical structure (30 atoms). Total formal bond orders will tend to be too high if a libration correction has not been made. In most cases this error will not exceed 5%. At present coordination numbers higher than four cannot be dealt with properly although atoms with higher coordination numbers can be input. SOURCE LISTING 143 C SIGCOR: A PROGRAM FOR CALCULATING ESTIMATED BOND LENGTHS AND BOND C ORDERS CORRECTED FOR SIGF1A HYBRIDIZATION EFFECTS C WRITTEN IN FORTRAN IV AT THE UNIVERSITY OF BRITISH COLUHBIA, 1972 C REVISED JOLT, 1973 COHHON NA DATA FC,PHRFO,FN,FP,FAS,FF,FCL,FBR,FI,FSI,FGE,FSN,FSB,FS,FSE,PTE/2 1H C.2H H,2H 0,2H N,2U P,2HAS,2H F,2HCL,2HBR,2H I,2HSI,2HGE,2HSN,2H 1SB,2H S,2HSE,2HTE/ DATA FB/2H B/ DIMENSION X (100),1 (100),Z (100),TITLE (20),SIGS (100) DIMENSION CA (100) ,RAD (100),A (5,100) DIMENSION KB (100,100),DL (100, 100),DM (100, 100),DN ( 100, 100) DIMENSION SF(100, 100),SIGCOR (100, 100) ,DELTA(100,100) DIMENSION TOR (100),DB (100,100),NTOT (100),NAT (20),SFT (100),PICOR (10 ?0,100),PHI (100,100) 90 FORMAT (• THIS PROGRAM CALCULATES ESTIMATED SIGMA HYBRIDISATION EFF 1ECTS AND PI CONTRACTIONS, BASED ON THE FOLLOWING ASSUMPTIONS:*/) 91 FORMAT (10X,• 1: HYBRIDISATION AT BOTH ATOMS INVOLVES ONLY S AND P 10BBITALS •) 92 FORHAT(10X,' 2: THE SINGLE BOND COVALENT RADII ARE THOSE GIVEN IN 1 THE ATOMS LIST BELOW') 93 FORMAT ('1') 94 FORHAT(10X, * 3: THE AT IONIC POSITIONS ARE CORRECT, I.E. CONTAIN NO SYSTEMATIC ERRORS.',/, 110X,' 4: THE HYDROGEN RADIUS HAS BEEN CONTRACTED TO BE CONSISTENT 1WITH XBAY DATA.',/,10X,' 5: RADII OF B AND N ATOMS ARE AUTOMATICAL 1LY ADJUSTED WHEN THESE ATOMS ARE FORMALLY CHARGED.*,/,10X, • 6: THE 1 RADIUS OF BOBON DEPENDS ON THE ELECTRON EGAT IV ITY OF ITS SUBSTITUE 1NTS.•,/,10X,' 7: BOND ORDER CALCULATIONS ARE ONLY APPROXIMATE AND 1 FOR BOND ORDERS >2 SMALL ERRORS IN THE BOND DISTANCE',/, 13X, 1 WIL 1L CAUSE LARGE ERRORS IN THE BOND ORDER.') 99 FORMAT (20AU) 100 FORMAT(6F10.5) 101 FORMAT (//,'OBTHOGONALIZATION MATRIX IS:',31X,'INVERSE MATRIX IS:', V) 102 FORMAT (3F15.6,15X,3F15.6,/) 103 FORMAT(//) 104 FORMAT (615) 105 FORHAT(5A2,20X,5F10.6) 106 FORMAT (/,'0BTH0G0NALISED COORDINATES: 34X,'FRACTIONAL COORDINATES 1:',//,' NO. ATOM ID X Y Z',23X 1,'X',10X,'Y',10X,'Z') 107 FORMAT(I5,5A2,3(5X,F10.6) ,10X,3 (3X,F8.4)) 108 FORMAT (1613) 118 FORMAT (' 1 *, ' ATOM (I) ATOM (J) LENGTH FS(I,J) FS(J,I) 1LC0RR PICON RAD (I) RAD (J) ICON ORDER DORD/DB'//) 119 FORMAT (10A2,5F10.tt,2F10.3,F8.1,2F8.2) 120 FORMAT (F10.6,2013) 124 FORMAT (/• END OF CALCULATION') 125 FORMAT ('1',' ATOM SUM FS NO OF BONDS,TOTAL BOND ORDER AT AT XOB COMMENTS') 126 FORMAT(5A2,F10.4,5X,13,F10.3) 130 FORMAT (5A2,F10.4,5X,13,F10.3,7X,'ASSUMPTION 1 MAY NOT HOLD AT THIS X ATOM') 131 FORMAT(30X.A30) 133 FORMAT('NA=',13,5X,' ND=*, 13,5X,•NOUTA'FI3,5X,•NC=•,13,5X,'NTAC=•,I 13,5X, 'NOPT= ',13) 134 FORMAT(/,• TOTAL BOND ORDER FOR THE ',13,' BONDS NOT INVOLVING H I IN THE ASSYHHETRIC UNIT IS: •,F5.2,/,'THE TOTAL NUMBER OF BONDS IS: 1 '.13) 999 FORMAT (• EXECUTION TERMINATED ON ERROR') PI=3.1415927 PA=0.4111906 144 TOBD=0.00 NBT=0 NBNH=0 WRITE (6,90) WRITE(6,91) WRITE (6,92) WRITE (6, 9U) WRITE (6,93) C READ IN AND PRINT OUT TITLE CARD READ (b,99) (TITLE (H),11=1,20) WRITE(6,99) (TITLE (rl) ,11=1, 20) C READ IN CONTROL CARD READ(S,10U) NA,ND,NOUT,NC,NTAC,NOPT WRITE (6,103) WRITE (6,133) NA,ND,NO0T,NC,NTAC,NOPT C READ IN CELL DIMENSIONS AND SET UP BETA AND INVERSE MATRICES DOUBLE PRECISION C(12),D,V READ(5,100) (C(J),J=1,6) DO 2 J=7,9 D = PI*C (0-3)/180 C(J)=CCOS(D) 2 C(J»3) =DSIN(D) V=DSQBT (1.0»2.0*C(7) *C (8) *C(9)-C (7) **2-C (8) **2-C (9) **2) V=C (1) *C (2) *C (3) *V BA1=C (1) BA2=C (2)*C (9) BA3=C (3) *C(8) BAU=0.0 BA5=C(2) *C(12) BA6=C (3)* (C (7)-C (8)*C (9) )/C (12) BA7=0.0 BA8=0.0 BA9*V/(C(1) *C(2) *C(12)) DA1=1.0/C (1) DA2=-C(9)/(C(1) *C(12) ) DA3 =(BA2*BA6-BA3*BA5)/V DAU-0.0 DA5=1 .0/ (C (2)»C (12)) DA6=-BA6/(BA5*BA9) DA7=0.0 DA8=0.0 DA9=1.0/BA9 WRITE (6, 101) WRITE (6,102) BA1,BA2,BA3,DA1,DA2,DA3 WRITE (6, 102) BAU,BA5,BA6,DA4,DA5,DA6 WRITE(6,102) BA7,BA8,BA9,DA7,DA8,DA9 NE=NA IF (NOPT.EQ.1) GO TO 1 WRITE (6,103) C CLEAR ARBATS 1 DO 8 1*1,BB DO 8 J=»1,HB SF(J,I)*0.0 8 KB(J,I) = 0 DO 9 1=1,NA TOR (I)=0.00 9 RAD(I)=0.00 C READ IN POSITIONS AND COVALENT RADII, IF NEEDED C PRINT ORTHOGONAL COORDINATES AND SINGLE BOND COVALENT RACII WRITE (6,106) DO 14 1*1,IA BEAD (5,105) (A(J,I),J=1,5),XP,YF,ZF,RAD(I),CA (I) IF(IC.EQ.O) GO TO 13 BEAD (5,131) SIGS(I) 13 X (I) =8AmF«BA2*YP*BA3*ZP I (I)=BA5»TP»BA6*ZP Z (I)=BA9*ZF C ASSIGN COVALENT RADII AND ELECTRONEGATIV IT IBS CALL PARSET (A (2, I) ,R,CX) 145 CA (I) =CA (I) + CX BAD (I)=RAD(I) *R WHITE (6,107) I, (A (J,I) ,J=1,5) ,X (I) ,Y (I) ,Z (I) ,XF,YF,ZF 14 CONTINUE WRITE (6,103) C READ IN BONDING INFORMATION AND SET UP CONNECTIVITY ARRAY (KB) 606 FOR NAT (' ATOH NUMBER ',15,' ON CARD 5( ',15,' ) EXCEEDS BONDING ARR X AY DIMENSIONS, I.E. IS GREATER THAN NB') DIMENSION JB(15) L = 0 17 READ (5,108) I, (JB (N),N=1,15) L = L*1 IF (I.GT.NA) GO TO 605 DO 18 N=1, 15 IF (JB (N) .GT. NA) GO TO 607 IF(JB (N) .EQ.O) GO TO 21 KB (JB (N) ,1) = KB (JB (N) , I) • 1 KB (I, JB (N) ) =KB (JB (N) ,1) 18 CONTINUE 21 IF (L. EQ.ND) GO TO 32 GO TO 17 605 WRITE(6,103) WRITE (6,606) I,L GO TO 998 607 WRITE (6,103) WRITE(6,606) JB(N),L GO TO 998 32 IF(NTAC.EQ.O) GO TO 26 C READ TERMINAL ATOH HYBRIDIZATION STATES IF DIFFERENT FROM SP3 DO 12 K=1,NTAC READ(5,120) SCT, (NAT (L),L= 1,20) DO 25 L=1,20 IF(NAT (L).EQ.O) GO TO 12 DO 31 J=1,NB 31 SF (J, NAT (L) ) = SCT 25 CONTINUE 12 CONTINUE C NTOT(M) IS NUMBER OF BONDS FORMED BY ATOH (H) 26 DO 215 1=1,NA N NB=0 DO 214 J=1,NA 214 NNB=NNB*KB(J.I) NTOT (I)=NNB 215 CONTINUE C CALCULATION OF AND STORAGE OF BOND LENGTHS IN ARRAY (DB) M=1 DO 28 1 = 1 ,NB M=H*1 IF (M . GT. NB) GO TO 39 DO 29 J=M,NB N = KB (J,I) IF (N.EQ.O) GO TO 29 DELX=X (J)-X (I) DELY=Y(J)-Y(I) DELZ=Z (J) -Z (I) D£(J,I)= SQRT(DELX**2»DELY**2*DELZ**2) DB (I , J)=DB (J,I) CL (J, I) =DELX/CB (J, I) DL (I,J)=-DL (J,I) DM (J, I) =DELY/DB (J, I) DM(I,J)=-DM (J,I) DN (J, I) =DELZ/CB (J, I) DN (I,J)=-DN (J,I) 29 CONTINUE 28 CONTINUE C CALCULATION OF BOND ANGLES AND S CHARACTER ARRAY ELEMENTS 39 NG= NB- 1 DO 42 J=1,NB DO 4 1 1=1 , NB 146 IF (KB (J, I) . EQ.O) GO TO 41 H = I+ 1 PHI (I,J)=0.00 NN=0 DO 40 K=1,NB IF (KB (K,J).EQ.O) GO TO 40 IF (K.EQ.I) GO TO 40 NN=NN*1 COSANG= (CL (J, 1) *DL (J,K) ) • (Dfl (J,I) *Dfl (J, K) ) • (DN (J, I) *CN (J,K) ) BNGLE = ARCOS (COSANG)*180/PI 44 PHI (I,J)=PHI(I,J) +BNGLE 40 CONTINUE IF(NN.NE.3) GO TO 66 PHI (I,J) = 656. 8-PHI (I* J) PHI(I,J)=PI*PHI(I,J)/(360»HN) GO TO 67 66 IF (NN.EQ.O) GO TO 64 PHI (I ,J) =PI*PHI (I,J)/ (360 »NN) GO TO 65 64 SF (I,J)=SF(I.J)+0.2500 GO TO 410 65 SF (I, J)* 1.0000- (0.5/ (SIN (PHI (I, J ) ) •*2)) GO TO 410 67 SF (I, J) = (1.5/ (SIN (PHI (I, J) ) **2)) -2. 000 410 DELTA (I,J)=PA*(SF(I,J)-0.25000) 41 CONTINUE 42 CONTINUE C CALCULATE CORRECTED SINGLE BOND LENGTHS AND APPROXIMATE BOND ORDER WRITE (6,118) H=1 DO 71 1=1,NB H=H*1 IF(H.GT.NB) GO TO 132 DO 70 J=H,NB IF(KB (J, I) . EQ.O) GO TO 70 CALL ENCOR (I,J,A,NTOT,RAD,CA) SIGCOR (J,I) = ( 1. 000-DELTA (I, J) ) •RAD (J) • (1 .000-DELTA (J,I))*RAD(I) PICOR (J,I)=SIGCOR(J,I)-DB(J,I) PC=100*PICOR(J,I)/SIGCOR(J,I) TC=PC/10 CALL BOND(I,J,SIGCOR,TC,BORD,DORD) IF (A (2,1) . EQ. FH) GO TO 78 IF(A(2,J) .EQ.PH) GO TO 78 TOB (I)=TOR(I)+BOBD TOR (J) =TOR (J) • BORD TORD=TORD*BORD NBNH=NBNH*1 78 PC=PC*0.05 BORD=BORD»0.005 DORD=DORD+0.005 WRITE (6,119) (A(K,I) ,K=1,5) , (A (K,J) ,K=1,5) ,DB (J,I) ,SF (J,I) ,SF (I, J) 1,SIGCOR (J,I),PICOR (J,I),RAD (I),RAD (J),PC,BORD,DORD NBT=NBT*1 70 CONTINUB 71 CONTINUE 132 TORD=TORD*0.005 WRITE(6,134) NBNH,TORD,NBT WRITE (6,125) DO 74 K=1,NA 74 SFT(K)=0.0 DO 73 1=1,NA IF (A (2,1).EQ.PH) GO TO 73 C Son UP BOND ORDER AND FRACTIONAL S CHARACTER AT EACH NON-HIDROGEN C ATOH. DO 72 J=1,NA IF (KB (J,I) . EQ.O) GO TO 72 SFT (I) *SFT (I) •SF(J.I) 72 CONTINUE IF(NTOT.(I) .LT.4) GO TO 75 147 IF (ABS (1.OOO-SFT (I)).LT.0.02) GO TO 75 WRITE (6, 130) (A(K,I) ,K=1,5) ,SFT(I) , NTCT (I) ,TOR (I) GO TO 73 75 WRITE(6,126) (A (K , I) , K= 1 , 5) , SFT (I) , NTOT (I) ,TOR (I) 73 CONTINUE C PRINT OUT BOND ANGLES AND INTERHYBRID ANGLES IF DESIRED IF (NOUT.NE.3) GO TO 997 CALL ANGLE(KB,DL,DM,DN.SF.A) 997 WRITE (6,12U) GO TO 1000 998 WRITE (6,999) 1000 STOP END SUBROUTINE PABSET (A,B,C) C LIBRARY OF COVALENT RADII AND ELECTRONEGATIVITIES DATA FC,FH,FO,FN,FP,FAS,FF,FCL, FBR, FI,FSI, FGE, FSN,FSB,FS,FSE,FTE/2 1B C,2H H,2H 0,2H N,2H P,2HAS,2H F,2HCL,2HBR,2H I,2HSI,2HGE,2HSN,2H 1SB,2H S,2HSE,2HTE/ IF (A. EQ.FC ) R=0. 768 IF (A. EQ.FH ) R=0. 23 IF (A. EQ.FN ) R=0. 701 IF (A. EQ.FO ) H=0. 652 IF ;A. EQ. FP ) R=1. 069 IF (A. EQ.FAS) H=1. 21 IF (A. EQ.FCL) R=0. 99 IF (A. EQ.FBR) R=1. 14 IF [A. EQ.FSI) 8=1. 17 IF (A. EQ.FSN) R=1. 40 IF (A. EQ.FGE) R=1 . 22 IF (A. EQ.FSB) R=1. 41 IF A. EQ.FS ) R=1. 04 IF [A. EQ.FF ) 8=0. 64 IF A. EQ.FI ) R=1. 33 IF A. EQ.FSB) R=1. 17 IF A. EQ.FTE) R=1. 37 IF A. EQ.FB ) C=2. 01 IF *• EQ.FC ) C=2. 50 IF A. EQ.FH ) C=2. 20 IF (A.EQ.FO ) C=3. 50 IF (A. EQ.FN ) C=3. 07 IF A. EQ. FP ) C=2. 06 IF A. EQ.FP ) C=4. 10 IF A. EQ.FI ) C=2. 21 IF :A. EQ.FS ) C=2. 44 IF (A. EQ.FAS) C=2. 20 IF ;A. EQ.FCL) C=2. 83 IF (A. EQ.FBR) C=2. 74 IF (A. EQ.FSI) C=1. 74 IF (A. EQ.FSN) C=1. 72 IF (A. EQ.FGE) C=2. 02 IF (A. EQ.FSB) C=1. 82 IF (A. EQ.FSB) C=2. 48 IF (A.EQ.PTE) C=2. 01 RETURN END SUBROUTINE ENCOR (I,J,A,NTOT,RAD,CA) C SUBROUTINE FOR ELECTRONEGATIVITY CORRECTIONS C AT A FB/2H B/ DATA FC,FH,FO,FN,FP,FAS,FP,FCL,FBR, PI,FSI,FGE.FSN.FSB.FS,FSE,FTE/2 1H C,2H H,2H 0,2H N.2H P,2HAS,2H F.2HCL.2HBR,2H I,2HSI,2HGE,2HSN,2H 1SB.2H S,2HSE,2HTE/ DATA FB/2H B/ DIMENSION A (5,100),RAD (100),CA (100),NTOT (100) i 148 IF(A(2,I) . EQ.FB) GO TO 68 IF (A (2,1) . EQ.FN) GO TO 69 GO TO 81 68 IF (NTOT (I) .EQ.3) RAD (I) = 0. IF (NTOT (I) •EQ.4) RAD (I) = 0. GO TO 81 69 IF (NTOT (I) .EQ.4) RAD (I) = 0. 81 IF (A (2.J). EQ.FB) GO TO 66 IF(A(2,J) . EQ.FN) GO TO 80 GO TO 67 66 IF (NTOT (J) .EQ.3) RAD (J) = 0. IF (NTOT (J) .EQ.U) RAD (J) = 0. GC TO 67 80 IF (NTOT (J) .EQ.U) RAD (J) = 0. 67 RETURN END 80 3-0. 0 75* A BS (2.01-CA (J) ) 918-0. 064*ABS(2.01-CA(J) ) 708 80 3-0.0 75*ABS (2.01-CA (I)) 918-0.064*ABS (2.01-CA (I) ) 708 SUBROUTINE BOND (I,J,SIGCOR,TC,BORD,DORD) C FUNCTION RELATING PERCENT CONTRACTION TO BOND ORDER DIMENSION SIGCOR (100,100) IF (TC.GE.O.47) GO TO 76 BORD=2.0*(EXP ( (0.612766*TC)-0.69 3)) D0RD=-(0.0612766*BORD)/SIGCOR (J,I) GO TO 77 76 TC2=TC**2 TC4=TC2**2 TC6=TC2*TC4 BORD=1.000* (1.89145589*TC2)- (1.79044767*TC4) •(0.88704 228*TC6) DORD=TC*(-0. 378291 18* (0.71617907*TC2)- (0.53222537*TC4))/SIGCOR (J,I D 77 RETURN END SDBROUTINE ANGLE (KB,DL,DM,DN,SF,A) C SUBROUTINE FOR CALCULATING BOND ANGLES COMMON NA DIMENSION A (5,100),KB (100,100),DL (100,100),DM (100,100),DN (100, 100) DIMENSION SF(100,100) PI=3. 141592 NB=NA 116 FORMAT (• 1 ', ' ATOH (I) ATOH (J) ATOM (K) OBS. ANGLE INTERHYBRID AN 1GLE',/) 117 FORHAT(12A2,5X,F7.3,5X,F5.1) WRITE (6,116) DO 42 J=1,NB DO 41 1=1,NB IF (KB (J , I) . EQ. 0) GO TO 41 H=I*1 DO 40 K=1,NB IF (KB (K,J).EQ.O) GO TO 40 IF(K.EQ.I) GO TO 40 COSANG= (DL(J,I)*DL (J,K) ) • (DM (J,I) *DH (J,K) ) • (Di (J, I) *DN (J , K) ) ANGLE=ARCOS(COSAHG)*180/PI C= (SF (I,J) *SF (K, J))/ ( (1.000-SF (I,J)) • (1.000-SF (K, J) ) ) IF (C.LT.0.000) GO TO 140 CC—SQRT(C) CANG*ARCOS (CC)*180/P1 CANG=CANG*0.05 GO TO 141 140 CANG^O.O 141 IF (K.LT.I) GO TO 40 WRITE (6,117) (A (N,I) ,N»1 ,4), (A (N,J) ,N«1,4), (A (B,K),N=1,4) ,ANGLE,CA ING 40 CONTINUB 41 CONTINUB 42 CONTINUE RETURN END B.B-r.IPHENYLBOROXAZOLIDINE BOND ORDRR CALCALATICN NA = 13 NE = )7 KOUT = STAC" NOPT= ORTHOGONALIZATION MATRIX IS: 13.8U0230 0.000003 -1.563807 0.0 8.916880 0.000004 0.0 0.0 10.048817 INVERSE MATRIX IS: 0.072253 0.0 0.0 -0.000000 0. 1121*7 0.0 0.011244 -O.COOCOO 0.099514 ORTHOGONAL COORDINATES: FRACTIONAL COORDINATES: ATOM ID X 1 Z X T Z 1 B 10.291018 2. 4 14442 3.727608 0.7855 0.2708 0.3709 2 0 9.467303 3.641119 3.591718 0.7244 0.4083 0.3574 3 N 9. 329347 1.340697 2.917914 0.7069 0.1504 0.2904 4 C 1 8. 120409 3.297132 3.341392 0.6243 0.3698 0.3325 5 C2 8. 18230 1 2. 126631 2.397346 0.6182 0.2385 0.2386 6 C3 11.710424 2.655144 3.002817 0.8799 0.2978 0.2988 7 C4 11.902396 2.599992 1. 6239 19 0.8782 0.2916 0. 1616 8 C5 13. 1 16244 2.946305 1.037087 0.9594 0. 3 304,, 0.1032 9 C6 14. 168496 3. 350659 1. 8044 15 1.0440 0.3758 0.1796 10 C7 14.019948 3.417439 3.172863 1.0487 0.3833 0.3157 1 1 C8 12.811918 3.075735 3.747806 0.9678 0.3449 0.3730 12 C9 10.410710 1.881695 5.241000 0.8111 0.2110 0.5216 13 CIO 11.037951 0.676410 5.552895 0.8600 0.0759 0.5526 14 Cl 1 1 1.151515 0.219884 6.860044 0.8829 0.0247 0.6827 15 C12 10.660658 0.968055 7.896218 0.8591 0.1086 0.7858 16 C 1 3 10.037719 2.156175 7.631754 0.8111 0.2418 0.7595 17 C1U 9.917236 2.596400 6.319348 0.7876 0.2912 0.6289 18 H (1, 1) 7.661106 3.079945 4.219457 0.6010 0.3454 0.4199 19 H(1,2) 7.640915 3. 998204 2.956743 0.5853 0.4484 0.2942 20 H (2 1) 8.46074 1 2.399273 1.525048 0.6285 0.2691 0.1518 21 H(22) 7. 385065 1.628079 2.327958 0.5598 0.1826 0.2317 22 H (NI) 8.992624 0.691648 3.582553 0.6900 0.0776 0.3565 23 H (N2) 9.706861 0.851982 2.332310 0.7276 0.0955 0.2321 24 H (4) 1 1 . 108064 2.335633 1.030716 0.8142 0.2619 0. 1026 25 H (5) 13.154810 2.897549 0.080189 0.9514 0.3250 0.0080 26 H (6) 15.023149 3.560367 1.404512 1. 1013 0.3993 0.1398 27 H (7) 14.759949 3.700006 3.759905 1.1087 0.4149 0.3742 28 H (8) 12.718121 3.138929 4.716502 0.9720 0.3520 0.4694 29 H (10) 1 1.448815 0.201407 4.860863 0.8819 0.0226 0.4837 30 H (11) 1 1 .614739 -0.625249 6.975526 0.9176 -0.0701 . 0.6942 31 H (12) 10.773760 0.663499 8.825694 0.8777 0.0744 0.8783 32 H (13) 9.652521 2.731019 8.318761 0.7910 0.3063 0.8278 33 H (14) 9.462642 3.419044 6. 157130 0.7529 0.3834 0.6127 to ATOM (I) ATOH (J) LENGTH FS(I.J) FS (J.I) LCORR PICON RAD (I) RAD (J) ICON ORDER DOBD/DB B B B B 0 H N N CI CI CI C2 C2 CI C3 Ct CU C5 C5 Cb C6 C7 C7 C8 C9 C9 CU) CIO C1 1 CI 1 C12 C12 CI 1 cn CI 4 N a a ci C2 H(N1) H (N 2) C2 H (1, 1) HO.2) H (21) H (22) CU CB CS H(1) C6 H (5) C7 H (6) C8 H (7) H (8) CIO C11 CI 1 H(10) C12 H(11) cn H (12) C11 H (13) H (11) 1.1838 1.6533 1.6118 1.6089 1.1125 1.1817 0.9881 0.8511 1.5050 1.0115 0.9321 0.9551 0. 9128 1. 393 3 1.3917 1.3920 1.0260 1.3636 0.9589 1.3781 0.9666 1.3808 0.9859 0. 9753 1. 391 1 1.3816 1. 3892 0.9315 1.3691 0. 9707 1. 3673 0.9816 1.3895 0.9751 0.9538 0. 1955 0.1707 0.3215 0.3080 0. 2561 0.2529 0. 1719 0.3082 0.2532 0.2570 0. 2580 0.2173 0.3115 0.3322 0. 3181 0.3381 0. 3269 0.3151 0.3512 0.3301 0. 3351 0.3278 0. 3313 0.3231 0. 3253 0.3261 0. 3389 0.3256 0.3165 0.3325 0.3323 0.3351 0. 3203 0.3339 0. 3238 0. 2561 0.2665 0. 3167 0.3175 0.2355 0.1815 0. 2500 0.2500 0. 2571 0.2500 0.2500 0.2500 0. 2500 0.3311 0. 3373 0.3228 0.2500 0.3312 0.2500 0.3378 0. 2500 0.3389 0.2500 0.2500 0. 3311 0.3387 0. 3200 0.2500 0.3325 0.2500 0. 3151 0.2500 0.3371 0.2500 0.2500 1.1915 1.5811 1.5980 1.6027 1.1229 1.1968 0.9607 0.9210 1.5326 0.9958 0.9955 1.0083 0.9691 1.1835 1.1869 1.1851 0.9737 1.1793 0.9721 1.1830 0.9710 1.1833 0.9711 0.9718 1.1857 1.1839 1.1858 0.9711 1.1795 0.9719 1.1800 0.9711 1.1863 0.9715 0.9717 0.0076 -0.0722 -0.0138 -0.0062 0.0105 0.0121 -0.0271 0.0700 0.0276 -0.0187 0.0631 0.0529 0.0262 0.0902 0.0922 0.0931 -0.0523 0. 1 156 0.0131 0.1019 0.0011 0.1025 -0.0115 -0.0001 0.0916 0.0992 0.0966 0.0396 1101 .0013 , 1126 .0135 .0968 -0.0036 0.0209 0.823 0.850 0.887 0.887 0.652 0.708 0.708 0.708 0.768 0.768 0.768 0.768 0.768 0.768 0.768 0.768 0.768 0.768 0.768 0.768 0.768 0.768 0.768 0.768 0.768 0.768 0.768 0.768 0.768 0.768 0.768 0.768 0.768 0.768 0.768 0.6S2 0.708 0.768 0.768 0.768 0.768 0.230 0.230 0.768 0.230 0.230 0.230 0.230 0.768 0.768 0.768 0.230 0.768 0.230 0.768 0.230 0.768 0.230 0.230 0.768 0.768 0.768 0.230 0.768 0.230 0.768 0.230 0.768 0.230 0.230 0.6 -1.5 -0.8 -0.3 0.8 0.9 -2.8 7.6 1.9 -1.8 6.1 5.3 2.8 6. 1 6.3 6.3 -5.3 7.9 1.1 7. 1 0.5 7.0 -1.1 0.0 6.2 6.7 6.6 1.1 7.5 0.2 7.7 -1.3 6.6 -0.3 2.2 1.01 C.76 0.95 0. 98 1.05 1. C6 0.81 67 12 90 1.53 1.11 1. 19 1.50 1.52 1.53 0.72 1.69 1.09 1.61 1.03 1.60 0. 92 1. C0 1.51 1.57 1.55 1.29 1.66 1.01 1.67 C.92 1.55 0. 98 1. 15 -0.01 -0.02 -0.03 -0.03 -0.01 -0.01 -0.05 -0.11 -0.01 -0.05 -0. 11 -0.11 -0.07 -0.07 -0.07 -0.07 -0.01 -0.07 -0.06 -0.C7 -0.06 -0.07 -0.05 -0.06 -0.07 -0.07 -0.07 -0.08 -0.07 -0.06 -0.07 -0.05 -0.07 -0.06 -0.07 TOTAL BOND ORDER FOR THE 19 BONDS NOT INVOLVING H IN THE ASSYRHETRIC UNIT IS: 25.61 THE TOTAL NUMBER OF BONDS IS: 35 ATOH SUM FS NO OF BONDS,TOTAL BOND ORDER AT ATOM COMMENTS B 0.9957 U 3.713 0 0.S122 2 2.078 N 0.9994 U 1.807 Cl 1.0036 4 2. 163 C2 0.9977 It 2.168 C3 0.9970 3 3.961 C4 0.9995 3 3.020 C5 0.9993 3 3.210 C6 0.9997 3 3.299 C7 0.9999 3 3.201 C8 0.9996 3 3.105 C9 0.9992 3 U.053 CIO 0.9986 3 3.056 Cl 1 0.9990 3 3. 198 C12 0.9999 3 3.318 C1 3 0.9993 3 3.215 C14 0.9997 3 3.115 ATOM (I) ATOM(J) ATOH(X) OBS. ANGLE INTERRTBRID ANGLE 0 B N 99.739 103.0 0 B C3 108.919 109.9 0 B C9 113.653 109.2 M B C3 112.892 108.3 N B C9 106.792 107.7 C3 E C9 114.011 1 17.4 B 0 Cl 110.136 1 10.2 B N C2 106.099 1 10.6 B N H(N1) 107. 181 106.0 B N H(»2) 116.867 113.8 C2 N H(N1) 108.679 105.4 C2 N H(N2) 113.917 112.9 H(N1) H H (N2) 10 3.697 107.8 0 C1 C2 105.150 108.9 0 Cl H(1. 1 109.298 109. 1 0 C1 H(1,2 112.357 109.1 C2 Cl H<1,1 113.268 1 10. 1 C2 Cl H(1,2 110.313 1 10.1 H (1, 1 Cl H(1,2 106.572 1 10. 3 N C2 C1 102.916 106.1 N C2 H<21) 104.242 104.4 N C2 H(22) 113.525 109.9 Cl C2 H(21) 111.269 108. 1 Cl C2 B(22) 115.001 115.1 H(21) C2 H(22) 109.277 112.4 B C3 CU 124.087 121.0 B C3 C8 120.022 119.9 C« C3 C8 115.579 118.8 C3 C4 C5 121.838 120.5 C3 CD H(«) 118.U06 119.6 C5 CU H(U) 1 19.700 1 19.9 CU C5 C6 120.626 120.1 CU C5 H(5) 116.302 1 19. 1 C6 C5 H(5) 123.063 120.8 C5 C6 C7 119.337 119.9 C5 C6 H(6) 120.918 120.3 C7 C6 H (6) 119.710 120.0 C6 C7 C8 119.721 120.0 C6 C7 H(7) 121.616 120.5 C8 C7 H(7) 118.662 119.7 C3 C8 C7 122.899 120.8 C3 C8 H(8) 118.303 1 19.6 C7 C8 H(8) 118.795 1 19.7 B C9 CIO 122.020 120.5 B C9 CIU 122. 35U 120.6 CIO C9 CIU 115.619 119.0 C9 CIO Cl 1 122.101 120.5 C9 CIO H(10) 118.137 119.5 C1 1 C10 H(10) 119.590 119.9 C10 C11 C12 120.21U 120.0 C10 C11 H(ll) 115.918 1 19.0 C12 C11 H(11) 123.8U0 121.0 cn C12 C13 119.460 119.9 Cl 1 C12 8(12) 120.281 120.1 C1 3 C12 H(12) 120.244 ' 120. 1 C12 C13 CIU 119.833 119.9 C12 C13 H(13) 123.777 121.0 Cl 4 C13 H(13) 116.387 119.1 C9 Cl u C13 122.762 120.7 C9 ciu H(1U) 118.856 119.7 C13 C1U H(14) 118.374 119.6 EM D OF CALCULATION STOP 0 EXECUTION TERMINATED SSIG 153 DISCUSSION The results of bond order calculations for <c6H5)2BOCH2CH2NH2 (*» Part 1)' (£_FC6H^) 2BOCH2CH2NH2 Part 2), and [ CH3N (CH2CH20) 2GaH ]2 (III, Part 4) are given as examples in Table 49. The reliability of the calculated bond orders depends on the accuracy of the structural data as well as on the errors inherent in the empirical method of calculation described above. For X-ray and neutron diffraction data the total bond order for a molecule calculated by the program tends to be too high if a libration correction has not been applied. This error is usually less than 5%, the actual magnitude depending on the degree of thermal motion in the sample and on the types of bonds present in the structure. Neglect of corrections for thermal motion leads to small errors in the individual bond orders for weak bonds, but becomes increasingly important as the bond order increases. This effect can be judged by the magnitude of the derivative of the bond order with respect to a 0.01 A change in the bond length which is included in the output for each chemical bond. The effect . of applying a libration correction on the total calculated bond order is shown in Table 49 for I and II. Bonds involving hydrogen atoms are not included in the total. The expected formal bond order for a molecule is figured with the assumption that all bonds involving hydrogen atoms have a bond order of 1. If the mean bond order for all such bonds in a structure is different from 1, then the 154 Table 49 Sample calculations using SIGCOR I II III Total bond order (excluding bonds involving H) expected (formal) 25 27 14 calculated (uncorrected) 25.84 28.37 partial libration correction 25.10 27.54 full libration correction — — 13.91 Mean deviation between valence and calculated interhybrid angles for 3 and 4 coordinate atoms 1.9 1.8 1.4 I Bond orders in the X-O-C-C-•N rings A B X-0 1.04 1.10 — 0-C 1.05 1.07 0.88 0.98 C-C 1. 12 1.28 1.12 1.03 C-N 1.06 1.08 0.98 1.09 N-X 0.76 0.73 — — Bond orders in the phenyl rings* x-o 1.46 1.44 o-m 1 .52 1 .49 m-p 1.59 1.66 mean 1 .52 1.53 *x refers to the atom bonded to B 155 expected value for the remaining bonds will change. For I and II the libration correction was not applied to bonds in the five-membered ring (see Parts 1 and 2 for details). as mentioned previously, the calculated interhybrid angles and the observed valence angles are generally not equivalent for 3 and 4 coordinate atoms. The magnitude of such deviations can be judged from the mean deviations for structures I-III given in Table 49. The orthogonality of the constructed hybrid orbitals is ensured if the sura cf the fractional s characters in all the hybrids at a given atom equals unity. In cases where the sum was calculable (tetrahedral or trigonal-planar coordination) the values ranged from 0.994 to 1.004 with a mean value of 1.00 for the three sample structures. The calculated bond orders for the X-O-C-C-N (X = B or Ga) rings in I-III and for the phenyl rings in I and II are also given in Table 49. The values indicate slightly different charge distributions in each of the four unique chelate rings as well as in the phenyl groups in I and II. This has been discussed in Parts 2 and 4. The bond orders give more information than can be deduced from a simple comparison of bond distances. Caution must be exercised when analysing data produced by the program. Inaccurate covalent radii and the neglect of electronegativity corrections (which are included only for boron atoms in the present version of the program) lead to incorrect bond orders. An example of this occurs in the bonds 156 involving the phenyl carbon atoms carrying the F substituents in II. It is clear that in this case the bond orders are too high since the mean bond order in the phenyl groups cf II should be less that for I (see Part 2). This is a result of not applying an electronegativity correction to the radius of the carbon atom bonded to the highly electronegative fluorine atom. ' The determination of the best values for the single bond covalent radii, electronegativity corrections, and the bond order - bond contraction relationship is a long and tedious process. Sellable and self-consistent parameters can only be obtained if a great deal of accurate experimental data is examined. This is complicated by the necessity that hybridization effects must first be accounted for and also by the fact that there is a high correlation between the parameters which are being derived. Work on the program will continue in the future, hopefully yielding an adeguate set of electronegativity corrections. It is also hoped that a method for dealing with hybrids involving d orbitals can be found in order that the program will work for atoms with coordination numbers greater than 4. 157 SUMMARY 158 The aim of this research has been to determine the structures of the five molecules previuosly described. The structures of the three boron compounds (Parts 1-3) have provided accurate geometric data for tetrahedral boron atoms in organic molecules. The analysis of BfB-diphenylboroxazolidine (Part 1) proved that the ethanolamine esters of diphenylborinic acid are intramolecular N—*B coordinated complexes. The p-fluorophenyl derivative (Part 2) was found to have a conformation different from that of the parent molecule, largely due to involvement of the fluorine atoms in hydrogen bonding. The two structures show small differences in the bond distances in the phenyl and five-membered BOCCN rings which indicate differences in charge distribution as a result of replacing the two hydrogen atoms by fluorine atoms. In Part 3 the compound C15HlgBNO2 was shown to be Ph 2BOCH 2NMe26 rather than Ph2BOCH2ONMe2, the latter analogous to the boroxazolidines in Parts 1 and 2. This compound has B-0 and B-C distances different from those in the other two boron compounds. This results from changing one of the substituents at the boron atom from nitrogen to oxygen. The bond distance alteration in the phenyl rings has a different pattern and the phenyl ring valence angles indicate that the boron atom in this case is less electron releasing than in the two boroxazolidines. The related gallium complex, [MeN (CH^CH^O^ GaH J,^ proved to be one of the first known crystallographic examples of 159 pentacoordinate gallium (Part 4). There are two distinct types of GaOCCN chelate rings in the molecule, both of which have 0-C, C-C, and C-N bond length patterns different from those in the related boron compounds. This can be seen by examination of the bond orders in Table 49, Both the gallium and hydridomolybdenum (Part 5) complexes are examples of compounds in which steric effects are the most probable cause of unusual geometries. The latter structure has provided an example of unique multicentre bonding involving aluminum atoms, Further research will include structural analyses of additional related compounds. Ethanol solutions of B,B-bis(p-fluorophenyl)boroxazolidine slowly decompose in light to give darkly colored crystals of unknown composition, the structure of which is now under study. Other structural studies planned for the near future are those of the monoethanolamine complex [Me2NCH2CH2OGaMe2]2, which may also feature a five-coordinate gallium atom, and ethylenediamine complexes of both boron and gallium. It may also be of interest to determine the structure of the boron analogue of the diethanolamine gallium compound. Research on compounds related to the hydridomolybdenum complex in Part 5 is being done by Prof. C. K. Prout and his associates at Oxford. 160 REFERENCES 1. G. H. Stout and L. H. Jensen. X-ray Structure Determination: A Practical Guide. The Macmillan Company, London. 1968. 2. H. Lipson and W. Cochran. The Crystalline State, vol. Ill: The Determination of Crystal Structures, 3rd. edn. G. Bell and Sons, Ltd., London. 1966. 3. M. J. Buerger. Crystal Structure Analysis. J. Wiley and Sons, Inc., New York. 1960. 4. M. J. Buerger. Vector Space. J. Wiley and Sons, Inc., New York. 1959. 5. M. M. Woolfson. X-ray Crystallography. Cambridge University Press. 1970. 6. International Tables for X-ray Crystallography, vols. I-III. Kynoch Press, Birmingham. Vol. I, 1952; vol. II, 1959; and vol. Ill, 1962. 7. R. L. Letsinger and I. Skoog. J. Am. Chem. Soc. 77* 2491 (1955) . 8. H. C. Brown and E. A. Fletcher. J. Am. Chem. Soc. 73, 2808 (1951). 9. H. K. Zimmermann. Adv. Chem. Ser. No. 42, 23 (1964). 10. Y. Rasiel and S. Rettig. Unpublished results. 11. R. E. Long. Ph. D. Thesis. University of California at Los Angeles. 1965. 12. Ref. 6, Vol. III. 13. R. F. Stewart, E. R. Davidson, and W. T. Simpson. J. Chem. Phys. 42, 3175 (1965). 14. V. Schomaker and K. N. Trueblood. Acta Cryst. B24, 63 (1968) . 15. D. W.J. Cruickshank. Acta Cryst. 9, 747 and 754 (1956). 16. D. W. J. Cruickshank. Acta Cryst. 14, 896 (1961). 17. J. B. Hendrickson. J. Am. Chem. Soc. 83, 4537 (1961); 85, 4059 (1963). 18. Chem. Soc. Special Publ. No. 11, 1958, and No. 18, 1965. 19. R. J. Sime, R. P. Dodge, A. Zalkin, and D. H. Templeton. Inorg. Chem. JO, 537 (1971). 161 20. Z. Taira and K. Osaki. Inorg. Nucl. Chem. letters 7, 509 (1971) . 21. J. Karle and I. L. Karle. Acta Cryst. 2J, 849 (1966). 22. J. Karle and H. Hauptman. Acta Cryst. 9, 635 (1956). 23. M. G. B. Drew. Private communication (1969). see eg. M. G. B. Drew, D. H. Templeton, and A. Zalkin. Acta Cryst. B25, 261 (1969). 24. W. C. Hamilton. Acta Cryst. 18, 502 (1965). 25. H. Hope, U. de la Camp, and H. E. Thiessen. Acta Cryst. A25, S78 (1969) . 26. W. R. Busing and H. A. Levy. Acta Cryst. 27, 142 (1964). 27. C. K. Johnson. In 'Crystallographic Computing,' pp. 207-226. Munksgaard, Copenhagen. 1970. 28. W. C. Hamilton and J. A. Ibers. Hydrogen Bonding in solids, pp.14-16, 214, 260-264. W. A. Benjamin Inc., New York. 1968. 29. G. C. Pimentel and A. L. McClellan. Annu. Rev. Phys. Chem. 22, 347 (1971) . 30. I. A. S. Edwards and H. P. Stadler. Acta Cryst. E26, 1905 (1970) . 31. 0. L. Carter, A. T. McPhail, and G. A. Sim. J. Chem. Soc. (A) , 822 (1966) . 32. R. Hoffmann. Advan. Chem. Ser. 42, 78 (1964). 33. J. Krausse and H. Dunken. Acta Cryst. 20, 67 ( 1966). 34. G. Zinner and w. Ritter. Angew. Chem. 74, 217 (1962). 35. W. Ritter. Dissertation. Marburg, 1963. 36. H. Weidmann and H. K. Zimmerman. Liebig's Ann. Chem. 6J9, 28 (1958. 37. H. Weidmann and H. K. Zimmerman. Liebig's Ann. Chem. 620, 4 (1958). 38. H. C. Fu, T. Psarras, H. Weidmann, and H. K. Zimmerman. Liebig's Ann. Chem. 641, 1 16 ( 1961). 39. G. Zinner, W. Ritter, and w. Kliegel. Pharmazie 20, 291 (1965). 40. G. Zinner and W. Kliegel. Arch. Pharmaz. 299, 166 (1966). 162 41. G. Zinner and W. Kliegel. Chem. Ber. 99, 895 (1966). 42. G. Zinner and R. Mcll. Chem. Ber. 99, 1292 (1966). 43. G. Zinner and E. Duerkop. Arch. Pharraaz. 30J, 776 (1968). 44. P. A. S. Smith. Open-chain Nitrogen Compounds, Vol. 2. P. 3 and p. 22, W. A. Benjamin, Inc., New York. 1966. 45. H. Arnold, N. Brock, and H. J. Hohorst. Arzneimittelforsch. 7, 735 (1957). 46. W. Kliegel. Z. Chem. 9, 112 (1969). 47. W. Kliegel. Z. Chem. 9, 454 (1969). 48. W. Kliegel. Liebig's Ann. Chem. 721, 231 (1969). 49. W. Kliegel. Tetrahedron Letters, 223 (1969). 50. W. Kliegel. Angew. Chem. 80, 614 ( 1968). 51. Molecular Structures and Dimensions, Vols. 1-4. Ed. 0. Kennard, D. G. Watson, and W. G. Town. A. Costhoek, Utrecht. Vols. 1 and 2, 1970; Vol. 3, 1971; Vol. 4, 1973. 52. A. Storr, B. S. Thomas, and A. D. Penland. J. Chem. Soc. Dalton, 328 (1972). 53. W. Harrison, A. Storr, and J. Trotter. J, Chem. Soc. Dalton, 1554 (1972). 54. D. F. Rendle, A. Storr, and J. Trotter. J. Chem. Soc. Dalton, 2252 (1973). 55. D. T. Cromer and J. B. Mann. Acta Cryst. A24, 321 (1968). 56. D. T. Cromer and D. Liberman. J. Chem. Phys. 53, 1891 (1970) . 57. G. E. Coates and R. G. Hayter. J. Chem. Soc, 2519 (1953) . 58. K. Dymock, G. J. Palenik, and A. J. Carty. Chem. Commun., 1218 (1972). 59. C. H. L. Kennard. Inorg. Chim. Acta 1, 347 (1967). 60. C. W. Heitsch, C. E. Nordman, and R. W. Parry. Inorg. Chem. 2, 508 (1963). 61. M. R. Churchill. Inorg. Chem. J2, 1213 (1973). 62. R. Juza and H. Hahn. Z. Anorg. Chem. 239, 282 (1938). 163 63. G. S. Smith and J. L. Hoard. J. Am. Chem. Soc. 8±, 3907 (1959) . 64. J. L. Caldercn, F. A. Cotton, and A. Shaver. J. Organometal. Chem. 37, 127 (1972). 65. R. Restivo and G. J. Palenik. Chem. Commun., 867 (1969). 66. G. Beran, A. J. Carty, H. A. Patel, and G. J. Palenik. Chem. Commun., 222 (1970). 67. A. Storr and B. S. Thomas. Can. J. Chem. 49, 2504 (1971). 68. J. C. Huffman and W. E. Streib. Chem. Commun., 911 (1971) . 69. L. J. Guggenberger. Inorg. Chem. Yl, 294 (1973). 70. C. J. Ballhausen and J. P. Dahl. Acta Chem. Scand. J5, 1333 (1961). 71. M. Gerlock and R. Mason. J. Chem. Soc, 296 (1965). 72. R. Hoxmeier, B. Deubzer, and H. D. Kaesz. J. Am. Chem. Soc. 93, 536 (1971) . 73. P. Corradini and A. Sirgu. Inorg. Chem. 6, 601 (1967). 74. F. N. Tebbe and L. J. Guggenberger. Chem. Commun., 227 (1973) . 75. A. T. McPhail, G. R. Knox, C. G. Robertson, and G. A. Sim. 3, Chem. Soc. (A), 205 (1971). 76. M. A. Bush, A. D. U. Hardy, Lj. Manojlovic-Muir, and G. A. Sim. J. Chem. Soc. (A), 1003 (1971). 77. J. L. Calderon, F. A. Cotton, and P. Legzdins. J. Am. Chem. Soc. SM, 2528 (1969). 78. E. Rodulfo de Gil and L. F. Dahl. J. Am. Chem. Sec. 9.1, 3751 (1969). 79. N. G. Connelly and L. F. Dahl. J. Am. Chem. Soc. 92, 7472 (1970). 80. J. K. P. Ariyaratno, A. M. Bierrum, M. L. H. Green, M. Ishag, C. K. Prout, and M. G. Swanwick. J. Chem. Soc. (A) , 1309 (1 969) . 81. M. R. Churchill and J. Cooke. J. Chem. Soc. (A), 2046 (1970) . 82. J. H. Cross and R. H. Fenn. J. Chem. Soc. (A), 3019 ( 1970) . 164 83. A. Kutoglu and H. Kopf. J. Organometal. Chem. 25, 455 (1970). 84. F. C. Wilson and D. P. Shoemaker. J. Chem. Phys. 27, 809 (1957). 85. M. J. Bennett and R. Mason. Proc. Chem. Soc, 273 (1963). 86. R. J. Doedens and L. F. Dahl. J. Am. Chem. Soc. 87, 2576 (1965). 87. J. R. Knox and C. K. Prout. Acta Cryst. B25, 2013 (1969). 88. J. R. Knox and C. K. Prout. Acta Cryst. B25, 1952 (1969). 89. J. R. Knox and C. K. Prout. Acta Cryst. B25, 2482 (1969). 90. C. K. Prout, T. S. Cameron, and A. R. Gent. Acta Cryst. B28, 3 2 (197 2). 91. C. K. Prout, G. B. Allison, L. T. J. Delbaere, and E. Gore. Acta Cryst . B28, 3043 (1972). 92. L. Pauling. The Nature of the Chemical Bond, 3rd edn. Cornell University Press, New York. 1960. 93. M. J. S. Dewar and H. N. Schmeising. Tetrahedron 1J.r 96 (1960). 94. D. P. Craig and N. L. Paddock. In 1 Nonbenzenoid Aromatics, ' vol. 2. Academic Press, New York. 1971. 95. C. A. Coulson. Valence, 2nd edn. Oxford University Press. 1961. 96. V. Schomaker and D. P. Stevenson. J. Amer. Chem. Soc. 63, 37 (1941). 97. F. H. Allen and D. Rogers. Acta Cryst. B25, 1326 (1969). PUBLICATIONS W. Harrison, S. Rettig, and j. Trotter, "Crystal and Molecular Structure of Hippuric Acid", J. Chem. Soc. Perkin II, 1036(1972). W. Harrison, S. Rettig, and J. Trotter, "Crystal and Molecular Structure of Tetra-u-o_-bromobenzoato-bis[aquocopper(II)", J. Chem. Soc. Dalton, 1852(1972). Ian W. Nowell, Steve Rettig, and James Trotter, "Disordered Crystal Structures of Six Complexes of the Type, Me2XCR1R2CF2XMe2' M(C0K (M=Mc or Cr; X - As or P; R1 = F or H; and R2 = H, CF3, or Cl)", J. Chem. Soc. Dalton, 2381(1972).-Steven J. Rettig and James Trotter, "Crystal and Molecular Structure of B_, B_-Diphenylboroxazol idine (2-Aminoethyl Diphenylborinate), Ph21&0(CH2)2NH2", Can. J. Chem., 51, 1288(1973). Steven J. Rettig and James Trotter, "Crystal and Molecular Structure of Hexakis(dimethylamino) cyclotriphosphazene, [NP(NMe2)2]3", Can. J. Chem., 51, 1295(1973). Steven J. Rettig and James Trotter, "Crystal and Molecular Structure of Potassium trans-1,2-Diaminocyclohexane-N,N-tetraacetatoman-ganate (III) Monohydrate, K(Mn(DCTA)]«H20", Can. J. Chem., 51, 1303(1973). Steven J. Rettig, Alan Storr, Brian S. Thomas, and James Trotter, "Crystal and Molecular Structure of (pentahaptocyclopentadienyl) hydri domolybdenum-u-di methyl a 1umi ni um-u-[methyla 1umi n i um-d i-(u-pentahapto(monohapto) cyclopentadienyl) dimethylaluminium]-(pentahaptocyclopentadienyl) hydridomolybdenum, [(C5H5)(C5H4) MoH]2Al3(CH3)5", Acta Cryst., B30, 666(1974). Alistair L. Macdonald, Steven J. Rettig, and James Trotter, "Crystal and Molecular Structure of 2-Deacylusnic Acid", Can J. Chem., 52, 723(1974). Steven J. Rettig, Alan Storr, and James Trotter, "Crystal and Molecular Structure of the N-Methyldiethanolaminogallane Dimer, [CH3N(CH2CH20)2GaH]2", Can. J. Chem., 52, in press. Steven J. Rettig and James Trotter, "Crystal and Molecular Structure of B_5B_-Bis(p_-fluorophenyl)boroxazolidine, (£-FC6H^)2 1k)(CH2)2NH2", Acta Cryst., B30, in press. Steven J. Rettig, James Trotter, and W. Kliegel, "Crystal and Molecular Structure of 4,4-Dimethyl-2,2-diphenyl-l,3-dioxa-4-azonia-2-boranatacyclopentane", Can. J. Chem., 52, in press. G.L. Hodgson, D.F. MacSweeney, T. Money, S.J. Rettig, and J. Trotter, "Crystal and Molecular Structure of (±)-7,7-(2,2'-Dimethyl) pentamethylene-1-methyl-norbornane-2-oxime", Can. J. Chem., to be published. S.J. Rettig, A. Storr, and J. Trotter, "Crystal and Molecular Structure of the N^N-Dimethylethanolaminodimethylgallane Dimer, [(CHahNCHzCHzOGatCHahL". in preparation. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Country Views Downloads
China 428 8
United States 12 2
Russia 7 0
United Kingdom 3 0
Turkey 2 0
France 2 0
Japan 2 0
Germany 2 0
Vietnam 2 0
India 1 0
Canada 1 0
Brazil 1 0
Ukraine 1 0
City Views Downloads
Hangzhou 228 0
Beijing 61 0
Unknown 32 0
Ürümqi 16 0
Guangzhou 16 0
Nanjing 11 0
Ashburn 9 0
Changsha 7 0
Nanchang 7 0
Shanghai 6 0
Zhengzhou 6 0
Dongguan 6 0
Xi'an 6 0

{[{ mDataHeader[type] }]} {[{ month[type] }]} {[{ tData[type] }]}
Download Stats

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0061057/manifest

Comment

Related Items