ON THE KINETICS AND CHEMISTRY OF SOME REACTIONS OF PHOSPHONITRILIC DERIVATIVES by C J . Stewart B.ScfHons.), Univers i ty of B r i t i s h Columbia, 1967 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE i n the Department of Chemistry We accept th i s thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA October, 1970 In p resent ing t h i s t h e s i s in p a r t i a l f u l f i l m e n t o f the requirements f o r an advanced degree at the U n i v e r s i t y of B r i t i s h Columbia, I agree that the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r reference and study. I f u r t h e r agree t h a t permiss ion fo r e x t e n s i v e copying o f t h i s t h e s i s f o r s c h o l a r l y purposes may be granted by the Head of my Department or by h i s r e p r e s e n t a t i v e s . I t i s understood that copying or p u b l i c a t i o n o f t h i s t h e s i s f o r f i n a n c i a l ga in s h a l l not be a l lowed without my w r i t t e n p e r m i s s i o n . Depa rtment The U n i v e r s i t y o f B r i t i s h Columbia Vancouver 8, Canada ABSTRACT The k i n e t i c parameters i n a c e t o n i t r i l e of the nuc leophi l i c subs t i tu t ion reac t ion: N_P_C1. + KCNS -> N_P_C1CNCS + KC1 5 6 b 5 5 b have been determined. Tlie Arrhenius ac t iva t ion energy i s 15.5 ± 5 Kcal mole" 1 and the common logarithm of the pre-exponential factor i s 10.2 ± 3. The react ion i s f i r s t order i n each reagent and i s probably bimolecular . Comparison with s i m i l a r reactions indicates a lone pa i r e lectron donation from the nitrogen to the phosphorus atoms of the r i n g . The compound, N^P^(NMe2)gHCuClg was prepared by the react ion of tetrameric phosphon i t r i l i c dimethylamide with copper (II) chlor ide i n butanone. The inf ra - red spectrum indicates that the copper atom i s bound to one r ing nitrogen atom, and the proton to the opposite nitrogen atom. The r ing i s found to be too small to allow chelat ion of the copper atom. The s a l t , N 6 P 6 ( N M e 2 ) 1 2 C u C l + C u C l 2 " was prepared by the dehydro-halogenation of N ^ P ^ ( N M e 2 ) H C l C ^ C l g . The hydrochloride was prepared by the react ion of hexameric phosphon i t r i l i c dimethylamide (H.P.D.) with ' copper (II) chlor ide i n the reducing solvent butanone. The sa l t was also produced by the react ion of H.P.D. with an equimolar mixture of copper (I) chlor ide and copper (II) ch lor ide i n a c e t o n i t r i l e . Chemical and magnetic studies on the s a l t showed i t to have one copper (II) atom per molecule. The x-ray c r y s t a l s tructure showed the copper (II) atom to be i n a d is tor ted square pyramidal environment, bonded to four r ing nitrogen and one chlor ine atom. The sa l t contains the f i r s t known example of the C u C ^ " anion. o This i s l inear with a Cu-Cl bond length of 2.11A. The anion i s also one of the few examples of a f i n i t e species containing a two co-ordinate copper (I) atom. The inf ra - red spectrum of the s a l t was very s i m i l a r to that of the parent H.P.D. The main difference was i n the frequencies of the s t retching modes of the phosphorus nitrogen r ing bonds; which i s consistent with copper che la t ion . Conductiometric and a n a l y t i c a l studies showed that the sa l t does not r e t a in the form N..P,.(NMe„), 0CuCl+CuCl„ i n a c e t o n i t r i l e so lu t ion ; O O I. 1/ References 90 - V -LIST OF TABLES I K ine t i c Parameters of Subst i tu t ion Reactions of 4 Phosphonitr i les II The Second Order Rate Constants at Various Temperatures 28 I I I Phosphon i t r i l i c Addi t ion Compounds 37 IV pKa Values for Some Phosphonitr i les 3 9 - 4 0 V Infra-Red Spectra of N ^ f N M e ^ g and (NMe 2 ) g HCuCl 3 43 VI Infra-Res Spectra of N 6 p 6 ( N M e 2 ) 1 2 and N f iP 6(NMe 2) 1 2 C u 2 C l 3 54 VII Elemental Analysis of H.P.D. Copper Complex 58 o VIII Bond Length (A) and Valency Angles (Degrees), with Standard Deviations i n Parenthesis 73 IX Actual and Predicted Bond Lengths 74 X Intra Molecular Distances . 7 8 XI L imi t ing Conductance of Some Sal ts i n A c e t o n i t r i l e at 25°C 84 XII Diamagnetic S u s c e p t i b i l i t i e s 88 - v i -LIST OF FIGURES 1. Conductivi ty C e l l 9 2. Graph of Temporary Thermal Effect at 8.8°C 12 3. Graph for Ca lcu la t ion of Subst i tu t ion Rate Constant at 25°C 14 4 - 1 1 Plo ts of Isothiocyanate Subst i tu t ion Reactions at 8.8°C 1 5 - 2 2 12. Graph for Ca lcu la t ion of Subst i tu t ion Rate Constant at 8.8°C 23 13. Graph for Ca lcu la t ion of Subst i tu t ion Rate Constant at 0.9°C 24 14. Graph for Ca lcu la t ion of Subst i tu t ion Rate Constant at 40.2°C 25 15. Arrhenius Ac t iva t ion Energy Graph . 2 6 16. The Structure of P^MegHCuClg (Exocyclic Groups Omitted) 38 17(a). The Infra-Red Spectrum of N.P.(NMe„) 0 ' 44 - 45 . 4 4 z o 17(b). The Infra-Red Spectrum of N^P^(NMe^)gHCuCl^ 4 6 - 4 7 18(a). The Infra-Red Spectrum of N,P . (NMe_) 1 0 5 5 - 5 6 D O £. 1/ 18(b). The Infra-Red Spectrum of (NMe2) 1 2 C u 2 C 1 3 5 7 " 5 8 19. Phosphoni t r i l i c Ring of N 6 P 6 ( N M e 2 ) 1 2 C u C l + Anion 71 20. General View of the N 6 P 6 ( N M e 2 ) 1 2 C u C l + i on , Showing the Copper (II) Atom Co-Ordination 72 21. Conductance Plot of N.P,(NMe„),„Cu„Cl_ 85 - V l l -ACKNOWLEDGEMENT I wish to record my grati tude to Professor N . L . Paddock under whose guidance and ins t ruc t ion I completed th i s work. C H A P T E R O N E GENERAL INTRODUCTION Phosphon i t r i l i c der ivat ives contain the formally unsaturated I I repeating, un i t - ^ P ^ ) - for which several types of chemical reactions are poss ib le . Among these are nuc leoph i l i c subs t i tu t ion of the ligands (X) which has been extensively studied, mainly from a preparative point of view. Addi t ion reactions e i ther to the double bond or of a donor-acceptor type have also been studied, but not so extensively . Since nitrogen i s more electronegative than phosphorus, binding electrons tend to accumulate near i t . This causes the phosphorus and nitrogen atoms to be respect ively e l e c t r o p h i l i c and nuc l eoph i l i c . Of the various types of reactions at the phosphorus and nitrogen centres which could occur, two are studied i n t h i s thes is : (1) acceptor propert ies t y p i f i e d by nuc leoph i l i c subst i tu t ion ' at phosphorus by the thiocyanate anion, (2) and donor proper t ies , through the chela t ion of copper (II) by ni t rogen. L i t t l e information exis ts about the k ine t i c s of nuc leoph i l i c subs t i tu t ion at phosphorus i n phosphoni t r i les . This may, i n par t , be due to the d i f f i c u l t y of studying react ion rates with a molecule having many - 2 -poten t ia l attack s i t e s . Therefore, i t was f e l t that a k i n e t i c study would be worthwhile, because i t would give some information i n a f i e l d where few facts are known. A l s o , i t might throw some further l i g h t on the nature of the ir-bonding systems i n phosphoni t r i l i c compounds. For t h i s work the i n i t i a l react ion of potassium thiocyanate with t r imer ic phosphoni t r i l i c chlor ide was studied. N,P_C1, + KCNS -> KC1 + N_P_C1_NCS 5 5 o 5 5 b The react ion was found to be bimolecular and i t s rate constant and ac t iva t ion energy were determined. The study of the preparation of t r a n s i t i o n metal complexes of phosphonitr i les i s also a f i e l d i n which very l i t t l e information i s ava i l ab l e . Metal halides have been made to react with phosphoni t r i les , but the r e su l t i ng complexes have, at most, one r ing nitrogen bonded to the metal . In th i s work, two copper complexes have been prepared; ' "N 4 P 4 (NMe 2 ) 8 HCuCl s and N 6 P 6 ( N M e 2 ) 1 2 Cu(I I )Cl C u ( I ) C l 2 . The former probably has a structure s i m i l a r to N 4P 4(Me)gH CuCl^ with one nitrogen bound to a proton and one bound to the C u C l ^ - group. Determination of the c r y s t a l s tructure of the complex of the hexameric dimethylamide showed i t to contain a unique porphyr in- l ike framework i n which copper (II) i s co-ordinated to four endocyclic nitrogen atoms i n the same molecule with the phosphonitr i le acting as a macrocyclic l igand . The de ta i led geometry shows evidence of competitive donor-acceptor in teract ions which are compatible with ir -e lec t ron theory. - 3 -C H A P T E R T W 0 KINETIC STUDIES OF NUCLEOPHILIC SUBSTITUTION IN PHOSPHONITRILES Phosphoni t r i l i c hal ides undergo many nuc leophi l i c subs t i tu t ion react ions . Their reactions with primary and secondary amines, a lcohols , phenols, ha l ide ions, and the thiocyanate ion have a l l been reported. These reactions are among the most important and in te res t ing i n phospho-n i t r i l i c chemistry, and i t i s therefore surpr i s ing that there exis t s almost no quant i ta t ive information concerning t h e i r k i n e t i c s . Since the rates of subs t i tu t ion reactions would be expected, as a f i r s t approximation, to depend on ir-electron density at the phosphorus atom, an increase i n k i n e t i c information should therefore lead to a greater understanding of the ir-molecular o r b i t a l s . Some quant i ta t ive work has already been done^ ^ \ and the resu l t s are summarized i n Table I . However the mass of k i n e t i c information i s q u a l i t a t i v e , dealing mainly with the types of observed or ien ta t ion patterns. The object of th i s work has been to obtain quant i ta t ive k i n e t i c data on nuc leoph i l i c subs t i tu t ion i n t r imer ic phosphoni t r i l i c ch lo r ide . The react ion studied was that producing the mono-substituted de r iva t ive . I t should be noticed that phosphoni t r i l i c halides offer many dif ferent react ion s i t e s to subs t i tu t ion and therefore, for any react ion there ex is t several steps. At tent ion was concentrated on the f i r s t step, because a l t e rna t ive paths are open to the second subst i tuent . Because of the Compound (NPC1 2 ) 3 (NPC1 2 ) 4 C N P C 1 2 ) S (NPC1 2 ) 6 (NPC1 2 ) 3 Table I Kine t ic Parameters of Subst i tu t ion Reactions of Phosphonitr i les Parameters Reagent C l ~ A n i l i n e Ethanol Piper idine Solvent A c e t o n i t r i l e Ethanol-Benzene Toluene Temp. E • (Kcal /mole)" 1 l o g 1 Q A Ref 0-35°C 18.3 12.1 (3) I I 16.3 12.0 i t i t 17.0 11.9 t i i t 16.3 11.2 t i 34.5°C k 2 = 148 x 10" 3 t o o l e " 1 s e c " 1 (1) i t k 2 = 0.01 x 10 ~ 3 £mole~ 1 sec~ 1 i t 0°C k 2 = 2.2 x 10" 3Jlmole" 1sec~ 1 (2) I I - 5 -equivalence of the chlor ine atoms i n the parent phosphoni t r i le , the f i r s t step can only go to form one de r iva t ive . P + N J \ However, the mono-substituted der iva t ive has three non-equivalent s i t es for nuc leoph i l i c attack which cannot be discriminated i n a s ingle experiment. N P N N Geminal Cis non-geminal Trans non-geminal The react ion used was the nuc leoph i l i c displacement of a chlor ine atom on phosphorus by a thiocyanate i o n . P_N,C1, + NCS" + p N C1CNCS + CI ' The thiocyanate ion was chosen because i t reacts more rap id ly than many other nucleophi les . A l s o , while the mono-substituted der iva t ive has not - 6 -been i s o l a t e d , the r e a c t i o n does go to produce the completely substituted N^P^CNCS)^ which has been prepared and characterized . Preliminary experiments showed that the r e a c t i o n proceeds at a s u i t a b l e rate. Also, there i s a p o s s i b l e i n t e r e s t in the thiocyanate group because i t has a ir-system which could i n t e r a c t with that of the r i n g and modify the rate of r e a c t i o n . F i n a l l y , a convenient method of following the r e a c t i o n i s p o s s i b l e because of the conductivity of the thiocyanate ion. M a t e r i a l s : Trimeric p h o s p h o n i t r i l i c c h l o r i d e was p u r i f i e d by re-c r y s t a l l i z a t i o n from benzene followed by sublimation and r e c r y s t a l l i z a t i o n to a constant melting point. (113.5°C, l i t e r a t u r e 112.8 ^ and 114°C (-10-)) . The reagent was examined by i n f r a - r e d and mass spectrometry and found to contain no tetrameric c h l o r i d e . Before the reactions, the t r i m e r i c p h o s p h o n i t r i l i c c h l o r i d e was dried i n vacuo for three hours at 50°C. Reagent grade potassium thiocyanate was p u r i f i e d by repeated r e c r y s t a l l i z a t i o n from 95% ethanol and then dri e d i n vacuo at 1 0 0 ° C ^ ^ . Potassium c h l o r i d e was d r i e d by repeated washings with reagent grade acetone and stored at 150°C. Reagent grade a c e t o n i t r i l e was d r i e d over calcium oxide by heating i t under r e f l u x f o r several hours. This was followed by d i s t i l l a t i o n under nitrogen on to phosphoric anhydride. A f t e r two d i s t i l l a t i o n s over the acid anhydride, the a c e t o n i t r i l e was transferred to a dry-box. It was found important to exclude oxygen from the solutions as well as moisture. At t h i s stage, i t was checked for dryness by e l e c t r o c o n d u c t i v i t y measurement on a Wayne Kerr Universal Bridge B221A. I f the spec i f i c conductance was above the a r b i t r a r i l y set l i m i t of 1.9 x 10~ 6 Ohm - 1 ( l i t e r a tu re spec i f i c conductance IO" 8 Ohm" 1 ^ 1 2 - ' ) , i t was r e d i s t i l l e d . (7*) Due to the warnings of Audr i e th v ' about the i n s t a b i l i t y of the isothiocyanate de r iva t ives , a l l the p u r i f i e d materials and dr ied apparatus were handled i n the dry box. Preparation: The preparation of the react ion solut ions was done by clean dry pipettes i n the dry box. A quantity of reagent was placed i n a weighing bo t t l e which was then removed from the box and weighed. This was returned to the box and a su i tab le quantity of the reagent was added to a conical f l a sk . Both the f lask and bo t t l e were weighed, and the f lask was returned to the dry box. A c e t o n i t r i l e was added and the f lask was removed and weighed again. The concentration was then calcula ted i n uni ts of moles per gram of s o l u t i o n . This procedure was used because the pipettes d id not de l ive r reproducible volumes. A l s o , i t was more accurate to use weight measure-ments which do not vary with temperature. The resu l t s were changed l a t e r in to moles per l i t r e using the dens i t ies of solvent and solute . ' Apparatus: 1 Two constant temperature baths were used during the experiment. The one for 25°G var ied no more than ± 0.01C 0 , and the other by ± 0.02C° . - 8 -An e lec t roconduct iv i ty c e l l was constructed of the type shown i n Figure 1. Because the electrodes were held i n place only by t h e i r connecting wires , utmost caution had to be exercised so as not to a l t e r the c e l l constant. The c e l l was ca l ib ra ted with potassium thiocyanate at 25°C. F i r s t the background conductance due to potassium chlor ide and a c e t o n i t r i l e was measured. Next, an amount of thiocyanate so lu t ion with a known concentration was added to the tared c e l l with a syringe, (a pipet te gave non-reproducible r e s u l t s ) . The t o t a l conductance was measured and the background conductance was subtracted from i t . The c e l l was then weighed and the concentration was ca lcu la ted . The p lo t of thiocyanate concentration against thiocyanate conductance gave a curve upon which three excel lent s t ra ight l ines could be f i t t e d : C = " 0.5307 8 ^ r k = 1.8 - 6 . 4 . ' - (2-1) C s k ' o . 4 4 9 7 6 3 for k = 6.4 - 10.8 (2-2) _ k - 2.974 0.3733 for k = 10.8 - 14.5 (2-3) C = concentration KCNS x 10 6 -g , mole" 1 k = conductance x 103*0hms No s ign i f icance i s attached to these l ines other than that they made the c a l c u l a t i o n of the concentration of thiocyanate easier and more accurate. The background conductance was small and never exceeded 1% of the lowest measured conductance. I t was none the l e s s , not quite steady Figure 1 . Conduct ivi ty C e l l - 10 -and a correct ion for i t was always appl ied . Procedure: The d i f ferent k i n e t i c runs were car r ied out using the same chemical procedures. For a l l of them th i s meant making reagent solut ions with known concentrations, equ i l i b r a t i ng these so lu t ions , mixing them, and then fol lowing the decrease i n thiocyanate concentration as a function of time. The reactant solut ions were prepared i n the dry-box. A few ml . of one stock so lu t ion were added to the c e l l which was then weighed. A 50 ml . f lask was treated i n the same way, using the other so lu t ion . The f lask was then replaced i n the dry-box and pure a c e t o n i t r i l e was added. A l l reactions were car r ied out with 30 ml . of so lu t ion , and the f i n a l concentrations were var ied by varying the amount of stock so lu t ions . Once the c e l l had been f i l l e d , i t could not be put back into the dry-box because i t could not be sealed wel l enough. The two reactant solut ions were then equi l ibra ted i n the constant temperature bath, (usually 30 minutes). The f lask was then removed from the bath and i t s contents poured into the c e l l . At th i s point , the clock and conduct iv i ty runs were s tar ted. In a l l cases the react ion was followed u n t i l 20% of the thiocyanate had been used. Once the react ion was f in i shed , the f u l l c e l l was weighed. This gave the amount of both react ion solut ions present, and from th i s the concentrations could be ca lcu la ted . 11 -When the two solut ions were mixed, there was an unavoidable increase i n the temperature of one of the solut ions which was caused by removing the so lu t ion from the bath by hand i n order to pour the so lu t ion . This caused the i l l u s i o n of a sharp decrease i n thiocyanate ion concentration. That t h i s was an i l l u s i o n was i l l u s t r a t e d by taking the conductance of an equi l ibra ted KCNS so lu t ion then pouring the so lu t ion into an equi l ibra ted f l a sk . After t h i s had equ i l ib ra ted , i t was poured back into the c e l l and conductance was measured as a function of time (Figure 2) . A downward curye l eve l ing o f f at the o r i g i n a l value wi th in one or two minutes, showed that what appeared to be a decrease i n concentration was i n fact a temperature ef fec t . Because many of the experiments have concentration readings wi th in one minute of addi t ion , th i s curve i s superimposed on t h e i r concentration graphs. K i n e t i c s : The reac t ion: — P 3 N 3 C 1 6 + 6KCNS -> 6KC1 + P 3 N 3 (NCS) 6 f71 has been reported by R . J . A . Otto and L . F . Audr ie th v . I t has been assumed for the present work that the react ion takes place by a series of i n d i v i d u a l and separate subs t i tu t ion steps, therefore, the f i r s t reac t ion w i l l be: P,N_C1 '+ KCNS -* P_N_C1CNCS + KC1 I f t h i s react ion i s f i r s t order i n each reagent, then: 76.6 7 6.5 O > >-t CO 3 rt n co n o 3 o CD 3 rt >-i ja rt H ' o 3 O 3 o 7 6.4 76 5 7 6.2 76 7 - 76.6 0 75.9 O O o o I o o o o ° o o o o o o o o 0 2 4 5 6 7 8 9 Time*(Minutes - 1) Figure 2. Graph of the Temporary Thermal Effect at 8.8°C 7 0 77 12 - 13 -- d [ K C N S ] = k [ K C N S ] [ p 3 N 5 C l 6 ] ( 2 - 4 ) T h e i n i t i a l r a t e , r " d [ K C N S ] ^ wm t h e n b e d i r e c t l y p r o p o r t i o n a l t o t h e p r o d u c t o f t h e i n i t i a l c o n c e n t r a t i o n s o f t h e two r e a c t a n t s , [ P „ N . C l J . a n d [ K C N S ] . . 3 3 6 1 L J 1 r - d [ K C N S 3 ^ s k r p 3 N 5 C l 6 ] . [ K C N S ] . ( 2 - 5 ) (The s u b s c r i p t * i * w i l l b e o m i t t e d f r o m a l l f u t u r e e q u a t i o n s . ) I f t h e r e a c t i o n i s o f t h e s e c o n d o r d e r , t h e n a p l o t o f " d ^ K C N S ^ - [ K C N S ] ~ 1 a g a i n s t [ P , N , C l J w i t h [KCNS] c o n s t a n t w i l l b e l i n e a r . - d [ K C N S j d t L J . . . ^ ^ j T h i s w i l l a l s o b e t r u e f o r a p l o t o f " d [ ^ N S ^ [p N c i 6 ] - 1 a g a i n s t [ K C N S ] when [ P ^ N g C l g ] i s c o n s t a n t . T h i s r e s u l t s i n two e q u a t i o n s w h i c h m u s t b e s a t i s f i e d f o r a r e a c t i o n o f s e c o n d o r d e r . - d [ K C N S ] [ K C N S ] - 1 = k [ P 3 N 3 C l 6 ] ( 2 - 6 ) - d [ K C N S ] a k [ K C N g ] ( 2 _ y ) T h e r e f o r e , e q u a t i o n s ( 2 - 6 ) a n d ( 2 - 7 ) w i l l p r o d u c e two c o i n c i d e n t a l s t r a i g h t l i n e s w i t h s l o p e . k o r i g i n a t i n g a t t h e o r i g i n . T h e r e a c t i o n r a t e 1 ^ ^ t ^ ^ — C a n ^ e d e t e r m : ' - n e d f r o m a p l o t o f [KCNS] a g a i n s t t i m e . T h i s p l o t s h o u l d r e s u l t i n a g o o d a p p r o x i m a t i o n o f a s t r a i g h t l i n e a t s m a l l t i m e . The i n i t i a l c o n c e n t r a t i o n s o f b o t h r e a g e n t s c a n b e c a l c u l a t e d k n o w i n g t h e q u a n t i t i e s o f r e a g e n t s u s e d i n t h e i r p r e p a r a t i o n . O n c e k h a s b e e n d e t e r m i n e d f o r s e v e r a l t e m p e r a t u r e s , t h e n t h e A r r h e n i u s ( c o n t i n u e d o n p a g e 27) 73 Time-(minutes)~ 1 Figure 5. Plot of isothiocyanate subs t i tu t ion react ion at 8.8°C 13 12 11 O O I n i t i a l KCNS cone. 14.314 x 10" 6 moles-g" 1 I n i t i a l P 3 N 3 C 1 6 cone. 13.32(5 x 10" b moles-g' o o 00 I 0 3 8 Time-(minutes)" 1 Figure 7. Plot of isothiocyanate subs t i tu t ion react ion at 8.8°C 10 11 12 Time.(minutes)~ Figure 8. Plot of isothiocyanate subs t i tu t ion react ion at 8.8°C Figure 9. Plot of isothioc/anate subs t i tu t ion react ion at 8.8°C Time-(minutes) _ 1 Figure 10. Plot of isothiocyanate subs t i tu t ion react ion at 8.8°C Figure 12. Graph for ca l cu l a t i on of subs t i tu t ion rate constant at 8.8°C - 27 -ac t iva t ion energy, E , can be calculated from equation (2-8). Si E l n k = - + l n A ( 2 " 8 ^ where A i s the frequency or pre-exponential factor . Equations (2-6) and (2-7) were thoroughly checked at 25°C (Figure 3) . This graph gave a good approximation to a s t ra ight l i n e beginning at the o r i g i n . This proves that the reac t ion: KCNS + P,NC1, -> P,N,Cl r NCS + KC1 6 o 6 5 b i s second order i n both reagents and, therefore, that: " d [ K ^ S ] = k[KCNS] [P 3 N 3 C1 6 ] (2-4) Charac te r i s t i c p lo t s of thiocyanate ion concentration against time i n minutes are shown i n Figures 4 - 11. At small time, a s t ra ight l i n e approximation was found which was used to ca lcula te —ft{rcCNS]—, A p lo t of - d £ K ^ s l against [P 3 N 3 C1 6 ] [KCNS] i s shown on Figure 12 for the react ion at 8.8°C. Figures 3, 13 and 14 are p lo ts of equations (2-6) and (2-7) for 25° , 0.9° and 40.2°C. The study showed the react ion to be second order, being f i r s t order i n each reagent. I ts second order rate constant (k) at various temperatures i s given i n Table I I . The Arrhenius ac t iva t ion energy was 15.5 ± 0.5 Kcal/mole and the log^g of the pre-exponential factor was 10.2 ±0 . 3 (Figure 15). - 28 -TABLE II THE SECOND ORDER RATE CONSTANTS AT VARIOUS TEMPERATURES Temperature ° C _ 1 k • [fc/mole-sec.] 1 0.9 6.1 ± 0.3 x 10" 3 8.8 11.9 ± 0.4 x IO" 3 25 67 ± 5 x IO" 3 40.2 191 ± 10 x IO" 3 Discussion: The thiocyanate group i s an ambidentate l igand and can bond to an e lec t roph i le through e i ther of i t s nuc leophi l i c ends; i n fac t , the thiocyanate anion reacts with t r imer ic phosphoni t r i l i c chlor ide to produce the f u l l y substi tuted isothiocyanato de r iva t ive . Several factors cha rac te r i s t i c of the e lec t rophi le determine which atom w i l l bond. These factors are mainly determined by the e lec t ron ic environment of the e l ec t roph i l e , and i t s stereochemistry. Two sets of antibonding TT o r b i t a l s are l o c a l i z e d on the (131 sulphur atom , and these can accept electrons from the e l ec t roph i l e . This resu l t s i n addi t ional s t a b i l i t y of the sulphur-e lec t rophi le bond. The thiocyanate der iva t ive i s formed i n groups that are able to back bond i n th i s manner. With the metals, the second h a l f of the second and t h i r d t r a n s i t i o n ser ies such as Rh, Pd, Ag, Cd, I r , P t , Au, Hg, Te and (14") Pb a l l form metal-sulphur bonds . However, with smaller atoms such as the metals of the f i r s t t r a n s i t i o n se r ies : Cr, Mn, Fe, N i , Cu and Zn, the b a s i c i t y of the nucleophile becomes increas ingly important, and - 29 -they form metal-nitrogen bonds. No nuc leophi l i c subs t i tu t ion react ion of an a l k y l ha l ide with a thiocyanate has apparently ever produced an isothiocyanato der iva t ive and so r e l a t i v e l y small atoms can f ind the p o l a r i z a b i l i t y of sulphur more a t t r ac t ive than the high b a s i c i t y of ni t rogen. However, with perita-valent phosphorus, the bonding does occur through the n i t r o g e n . P 2 0 3 C 1 4 + KSCN CS l 4 P0C12(NCS) Knowing the element and i t s oxidat ion number i s not i n i t s e l f su f f i c i en t information for p red ic t ion of the type of bond which w i l l be formed. The nature of the surrounding ligands has been shown to be a (17) determining factor j ir-electron withdrawing ligands reduce the electron density at the e l ec t roph i l e , thus making i t a weaker nr-electron donor. The next determining factor has been shown to be s t e r i c . The + isothiocyanato group can become l inea r by assuming the form (-NsC-S~). The thiocyanate on the other hand, i s always bent/S-C=N . An example of t h i s i s found when the non u-bonding l igand diethylenetr iamine, NH 2 C 2 H 4 NHC 2 H 4 NH 2 , gives the S-bonded complex [Pd(dien)SCN] + . The extremely bulky te t rae thyl substi tuted l igand , ( C 2 H 5 ) 2 N C 2 H 4 N H C 2 H 4 N ( C 2 H 5 ) 2 , gives the N-bonded complex [Pd(Et 4 dien)(NCS)] + . For the phosphoryl der iva t ives i t seems u n l i k e l y that s t e r i c factors could have any effect. on the subs t i tu t ion , and i n the f u l l y subst i tuted isothiocyanato-(23) phosphoni t r i le the substituent groups are wel l separated. K i n e t i c studies have been done on phosphoryl compounds, but none have been reported which use the thiocyanate anion as the nucleophile. This l i m i t s the extent to which v a l i d comparisons can be made between -. 30 -the k ine t i c s of the phosphoryl and the phosphoni t r i l i c series of compounds. However, i t i s s t i l l possible to compare the base strength preferences of the phosphoryl groups with that of the phosphoni t r i le . S i m i l a r l y , for the phosphonitr i les only very l imi ted information i s ava i lab le which can be used as a comparison with th i s work. Nucleophi l ic r e a c t i v i t i e s are mainly determined by the e l ec t ron ic , s t e r i c , and so lva t ion charac te r i s t i c s of the reactants. I f s t e r i c and solvent effects are ignored, then the r e a c t i v i t y of a ser ies of s i m i l a r nucleophiles with one e lec t rophi le can be correlated from the equation: log Ji_ - a H + BP (2-9) o where: k = rate constant for subs t i tu t ion with water o k = rate constant for subs t i tu t ion with the nucleophile P . = a function of the p o l a r i z a b i l i t y of the nucleophile H = a function of the pKa of the conjugate acid of the nucleophi le; a and g are parameters associated en t i r e ly with the e l e c t r o p h i l i c sub-(19) s t r a t e v . I f , as i n the case of saturated carbon, p o l a r i z a b i l i t y i s more important than b a s i c i t y , then the value of B w i l l be higher than that of a. The r e a c t i v i t y w i l l then fol low the p o l a r i z a b i l i t y funct ion, P, of the nucleophi le . The opposite has been found for the phosphorochloridates where:'- 2 0 ' ' F" > HO" > C\.H c0" > EtOH > C £ H r S " > CH_C0 " 6 5 6 5 3 2 Here r e a c t i v i t y follows the b a s i c i t y of the nucleophi le , and g can be assumed to be smal l . The base preference of the phosphorochloridates i s - 31 -also evident i n the formation of the isothiocyanato de r iva t ive , C^PCCONCS, i n preference to the corresponding thiocyanato de r iva t ive . This allows equation (2-9) to be expressed i n the form: log k = apKa + constant (2-10) Therefore, where the r e a c t i v i t y of the e lec t rophi le follows the p o l a r i z a b i l i t y of the nucleophile 3 i s predominant i n equation (2-9). On the other hand, where the base strength i s followed a i s predominant and equation (2-10) i s the r e s u l t . Furthermore, i n compounds with saturated carbon, where $ i s important, the thiocyanato der iva t ive i s formed. Conversely, with the phosphorochloridates, where a i s important, the isothiocyanato der iva t ive i s formed. The formation of the isothiocyanato phosphoni t r i l i c der iva t ive i s then strongly suggestive of a r e a c t i v i t y dependent upon a. This can be shown by a comparison of the rate constant, k ^ , for the r e a c t i o n ^ N„P 7 C1, + C l " * -r N P 7 C1 C C1* + C l " with the rate constant k for the thiocyanate subs t i tu t ion reac t ion . * * • k c l i s lower, 5.01 x 10" 2 than k, 6.7 x 10" 2 . Therefore, the r e a c t i v i t y does fol low the b a s i c i t y . These two rate constants allow a value of a to be ca lcu la ted : HCl l o g i n k c l = a P K a + constant (2-11) l o g 1 Q k = ctpKa + constant (2-12) i n uni t s of £ m o l e _ i s e c - 32 -log k - log k c l = a ( p K a H C N S - p K a H C 1 ) (2-13) , v HCNS „ HC1 . (19) where pKs - pKa = 4 v J (2-14) k _ log y- ~ 4 a (2-15) CI T l o g f n j r = a ( 2 - 1 6 ) a = 0.0317 (2-17) The pKa value i s an approximation, the accuracy of which determines to a large extent the accuracy of the a value. This approximation, has been used to corre la te base strength and r e a c t i v i t y i n s i m i l a r reactions and therefore, i t would seem to be an acceptable value to use here. Furthermore, the in teres t i n a l i e s not i n i t s absolute value, but i n comparisons with a values for s i m i l a r compounds. Such comparisons do show a sensible difference between the a value of the phosphonitr i le and a values for compounds containing penta-valent phosphorus atoms. a Coeff ic ients for the Reaction of Nucleophiles with Phosphoryl Compounds J Compound a Et 2N(OEt)P(0)CN 0.50 (EtO) 2 P(0)OP(0)(0Et) 2 0.70 PrO'(Me)'P(0)F 0.82 - 33 -As the pos i t i ve charge on the phosphorus atom increases, the rate constant and the value of a should also increase. Therefore, i f a l l other factors are constant, the value of a should follow the electron withdrawing effect of the l igands . The small value of a for the phosphonitr i le compared with those for the phosphoryl compounds i s then an ind ica t ion of the low pos i t i ve charge on the phosphon i t r i l i c phosphorus atom. This would be due to electron donation from the lone pai rs of the r i ng nitrogens to the phosphorus atom. This effect i s already known; recent studies of (22) the i o n i z a t i o n potent ia ls of phosphonitr i les have indicated lone pa i r donation. I t should be observed from equation (2-15) that any factor which would increase k would a lso increase a. So decreasing the pos i t i ve charge on the phosphorus atom decreases a. However, other factors could be important a l so . In the isothiocyanate the lone pa i r on the nitrogen could donate to the phosphorus atom. The corresponding increase i n the exocycl ic P-N bond energy would be expected to decrease the ac t iva t ion energy and, thus, increase k. Such donation would be evident (23 i n a shortened exocycl ic P-N bond length; however, recent x-ray studies show only a r e l a t i v e l y small shortening and therefore indicate a weak exocycl ic donation. Two types of nitrogen lone p a i r donation are now evident: . (1) donation from the r ing nitrogen atoms which decreases a, (2) and donation from the isothiocyanate nitrogen atom which increases k by s t a b i l i z i n g the t r ans i t i on s ta te . Because the ct value for the phosphonitr i le i s smaller than the - 34 -a values for the phosphorochloridates, endocyclic donation would appear to be dominant. This i s consistent with the x-ray structure of the f u l l y subst i tuted de r iva t ive , where the endocyclic shortening of the P-N bond i s greater than the exocyc l i c . The E & value of the thiocyanate react ion i s 15.5 ± 0.5 Kcal/mole. This i s lower than the value of 18.3 Kcal/mole found by f31 Sowerbyv J when using chlor ine as the nucleophi le . The Log^A or pre-exponential factor was 10.2 ± 0.3 compared with 12.1 for the ch lor ine . The difference i n ac t iva t ion energies i s compatible with the fact that subs t i tu t ion by isothiocyanate i s faster than by ch lo r ine . The increase i n rate i s then due to the s t a b i l i z a t i o n of the t r a n s i t i o n step. In conclusion i t can be stated that the work i n th i s sect ion has provided information on several cha rac te r i s t i c s of phosphoni t r i le chemistry. • The reac t ion: N 3 P 3 C 1 6 + 6KCNS •* N 3 P 3 ( N C S ) 6 + 6KC1 s ta r t s with the separate and ind iv idua l step: N . P , C L + KCNS '-»• N_P,C1CNCS + KC1 o o o 5 o b which i s followed by a ser ies of other steps to give the f u l l y subst i tuted d e r i v a t i v e . The f i r s t step i s second order, being f i r s t order i n each reagent and i s most probably bimolecular . This would involve a penta-co-ordinate phosphorus i n the t r a n s i t i o n s ta te . - 35 -Comparison of the k ine t i c s of th i s f i r s t step with those of the s i m i l a r chlor ine exchange reac t ion , enables the value of the a factor to be ca lcu la ted . Further comparison of th i s value with a factor values for phosphoryl compounds indicates that the phosphorus atom i n the phosphoni t r i le has a higher electron density than the phosphorus atom i n the phosphoryl groups. - 36 -C H A P T E R T H R E E METAL HALIDE DERIVATIVES OF PHOSPHONITRILES Introduction: In the study of the chemistry of phosphoni t r i l i c der ivat ives a major objective i s the discovery of experimental evidence relevant f24) f25) to theore t ica l expectations. .Cra ig and Paddock and others have, on the basis of Huckel M.O. theory, been able to explain many of the empir ical resu l t s of such studies i n terms of n-electron systems covering the whole molecule. The structures of the tetrameric and hexameric phosphon i t r i l i c dimethylamide der ivat ives are known&^>^) a nd part of the object of the present work has been to make and study t h e i r complexes with metals, so as to throw more l i g h t on the factors inf luencing ir-electron d r i f t i n the phosphoni t r i les . Phosphonitr i les are known to form add i t ion . . compounds with Lewis ac ids . Some of the reported compounds are l i s t e d i n Table I I I . Recent x-ray determination of the c r y s t a l stucture of . N^P^MegH C u C l , ^ ^ shows the copper bound to only one r i ng ni t rogen, and the proton bound to the opposite nitrogen (Figure 16). In the f39) metal carbonyl de r iva t ive , N^P^MegMo(Co)^ , in f ra - red evidence suggest a molybdenum atom co-ordinated to two opposing r i ng ni trogens. However, - 37 -Table I I I Phosphoni t r i l i c Addi t ion Compounds N_P_C1 "HCIO. (28) (29,30) (29) (28) N 3 P 3 (NHC 6 H 5 ) 6 'HC1 (28) N_P_C1 *3S0_ j o o 6 N 3 P 3 C 1 6 ' N ° 2 3 3 6 5 6 4 N 3 P 3 C1 4 (NHCH 3 ) 2 'HC10 4 (28) N 3 P 3 C1 4 (NHC 2 H 4 NH 2 ) 2 'HC1 (28) N 3 P 3 ( N H 2 ) 6 - ( H C 2 H 3 0 2 ) 3 (30) N,P,Me 6 RI (R=Me, Et) (31) N 3 P 3 C l 2 ( N H P r ) 4 - H C l (32) [N 3 P 3 (NMe 2 ) 4 (NMe 3 ) 2 ] ' 2BF 4 (33) N 3 P 3 C1 6 -2A1C1 3 (34) N 3 P 3 C l 6 - A l B r 3 (35) N 3 P 3 B V A 1 B r 3 ( 3 5 ) N 3 P 3 B r 6 ' 2 A l B r 3 (35) N 3 P 3 Me 6 'MCl 4 (M=Sn,Ti) (36) N P3(NHR) -HCl(R=Et, n-Propyl , i - P r o p y l , n - B u t y l , i - B u t y l ) N 4 P 4 C l g - 2 H C 1 0 4 (28) N 4 P 4 Me 8 Mo(C0) 4 (39) N 4 P 4 ( N H 2 ) g ( C 2 H 3 0 2 ) (30) N 4 P 4 Me g H CuCl^ (40) (N 4 P 4 Me 8 H) 2 CoCl 4 (41) N 4 P 4 Me g RI (R=Me, Et) (31) (NPF 2 ) n -2SbF 5 (n=3-6) (42) - 38 / ® • Figure 16. The Structure of P.N.Me 0HCuCl 4 4 o (Exocycl ic Groups Omitted) - 39 -since i t has not been made i n c r y s t a l l i n e form, no x-ray determination of i t s structure i s ava i l ab l e . [ (NPMe2)4H] 2 C o C l 4 , which i s s to i ch iomet r i ca l ly s i m i l a r to N^P^MegH CuCl j , has an ion ic structure composed of two s t e r i c a l l y d i s s i m i l a r N^P^MegH"1" cations and one CoCl^ anion. To date, no phosphonitr i le has been bonded with a t r a n s i t i o n metal hal ide i n such a way as to form bonds from more than one r i ng ni t rogen. A few examples ex is t where the acceptor atom probably : nitrogf (29,30) bonds to one en, e .g. (Me 2 PN) 3 SnCl 4 > ( M e 2 P N ) 3 T i C l 4 ( ' 3 6 ' ) , and N^PjCl^'SSO^ ' . The evidence for t h i s type of bonding comes from inf ra- red and nuclear magnetic resonance s tudies . Trimeric and tetrameric der ivat ives are probably too crowded •-• to allow the metal to s i t as c lose ly to the r ing plane as would be necessary i n a multi-dentate l igand. P a r t i a l l y o f f se t t ing t h i s , however i s the b a s i c i t y of the r ing ni trogens. The more basic they are, the less important w i l l be the s t e r i c ef fec ts . The suggested structure of (39) N 4P 4MegMo(Co) 4 J indicates the p o s s i b i l i t y of forming an N 4 P 4 RgCuCl 2 molecule with several r ing nitrogens donating to copper. However, as t h i s does not happen for R = Me, a more e lectron-releasing exocycl ic group may be necessary. Information on the b a s i c i t y of the phosphoni t r i l i c der iva t ives i s ava i l ab le , and some i s presented i n Table IV. TABLE IV pKa VALUES FOR SOME PHOSPHONITRILES* Group Trimer Tetramer pKa P^ a 2 P ^ a P^ a 2 Et (CH ) ( 4 3 ' 4 4 ) 6.4 7,6 0.2 - 40 -Group Trimer ^ Tetramer pKa pK&2 pKa P K a 2 N M e 2 ( 4 5 ) 7.6 -3.3 8.3 0.6 * determined i n nitrobenzene ** refers to addi t ion of a second proton In none of the der iva t ives studied d id the pKa values change (44) by more than 1.2 units when varying r ing s ize from trimer to tetramer J The effect of v a r i a t i o n of r i ng s ize on base strength i s evidently much smaller than that of v a r i a t i o n of the exocycl ic groups. I t should also be noticed that pKa values for phosphon i t r i l i c amides are very s i m i l a r to those for the free amines; dimethylamine has (45") a pKa value of 7.5 . Knowing the ease with which amines form complexes with a high copper co-ordinat ion number, i t might be expected that phosphonitr i les would do the same. The tetrameric phosphon i t r i l i c dimethylamide der iva t ive i s more basic than the methyl der iva t ive both to f i r s t and second protonation, and therefore, i t i s more l i k e l y to form a chelate complex with copper. This would involve a structure s imi l a r to that suggested for P 4 N 4 ( M e ) g M o ( C o ) 4 ( 3 9 ) . However, s t e r i c interference would be greater for the amide groups than the methyl groups. This would offset some, i f not a l l , of the increased b a s i c i t y . S t e r i c in teract ions would be reduced i n the larger r ings without appreciable change i n base strength. In th i s work the react ion of copper (II) chlor ide with hexameric phosphon i t r i l i c dimethylamide der iva t ives was invest igated. - 41 -With phosphoni t r i l i c amides, addi t ion reactions can occur using the lone pa i r electrons at the r i n g , or exocycl ic nitrogens or the molecular o r b i t a l s of the r ing system. A number of a r t i c l e s suggest that the r i ng nitrogen atoms are the centres of b a s i c i t y . One of these deals with the s t ruc tura l determination of N ^ P ^ C ^ (NHPr*) ^HCl . The proton i s bonded to a r ing nitrogen which indicates that th i s s i t e i s the centre of b a s i c i t y . The structure of [N^P^MegHjCuCl^^ 0 ' ' a lso supports th i s conclusion. Never the l e s s , i t i s not inconceivable that, i n hindered molecules, co-ordinat ion of the exocycl ic groups might (33) occur, and in teract ions of t h i s sort have been suggested In the course of t h i s work, the tetrameric phosphoni t r i l i c dimethylamide der iva t ive was made to react with copper(II) chlor ide i n an attempt to produce a chelate complex of copper. However, the stoichiometry, N^P^(NMe2)gHCuCl,j, of the r e su l t ing complex and i t s inf ra- red spectra suggest a structure with only one copper-nitrogen covalent bond and one protonated r ing ni t rogen. THE REACTION OF TETRAMERIC PHOSPHONITRILIC DIMETHYLAMIDE (T.P.D.) WITH COPPER (II) CHLORIDE. The Preparation of T . P . D . : T .P .D . i s formed according to the equation: P 4 N 4 C l g + 16NHMe2 -*• P 4 N 4 ( N M e 2 ) g + 8 NH 2Me 2Cl The tetrameric phosphon i t r i l i c ch lor ide was prepared by the method of (47) Stokes and the dimethylamine was the reagent grade chemical obtained - 42 -from Eastman Kodak. At -78°C, 80 ml (1.21 moles) of dimethylamine i n 20 ml of d i e thy l ether were slowly mixed with 30.2 g (0.0652 moles) of tetrameric phosphoni t r i l i c chlor ide in 150 ml of d i e thy l ether. A dense white p rec ip i t a t e formed immediately. When th is formation had stopped the react ion mixture was allowed to reach room temperature where i t remained for two hours. The react ion mixture was then heated under ref lux for nineteen hours. The white p rec ip i t a t e was f i l t e r e d off , leaving a c lear so lu t i on . Evaporation of th i s so lu t ion l e f t a white hard powder (m.pt. 220°C) assumed to be the crude product (A). The res idual dimethylamine and d ie thy l ether were removed under vacuum from the f i r s t white residue. Co-precipi ta ted product was then extracted from i t using d i e thy l ether as the solvent . Evaporation of t h i s solvent l e f t a gummy white residue which was reacted for a further t h i r t y - f o u r hours with dimethylamine. The crude product r e su l t i ng from th i s reac t ion , combined with that previously obtained, (A), weighed 16 g (40% y i e l d based on the amount of tetrameric phosphon i t r i l i c chlor ide added). I t was p u r i f i e d by repeated r e - c r y s t a l l i z a t i o n from d ie thy l ether to produce white c r y s t a l l i n e needles of about 2 mm i n length, m.pt. 238°C, (N 4P 4(NMe 2)g m.pt. 2 3 8 ° C ^ ^ ) . The inf ra- red spectrum of the sample i n a potassium bromide p e l l e t i s shown i n Table V and Figure 17(a). A q u a l i t a t i v e chlor ine analysis showed the sample to be free from contamination by hydrogen ch lo r ide . Quanti tat ive micro analysis gave N, 31.88; C, 35.79; H, 9.55%; (formula N 4 P 4 ( N M e 2 ) g requires N, 31.60; C, 36.10; H, 9.02%. - 43 -TABLE V Infra-red Spectra* of N 4 P 4 ( N M e 2 ) 8 and (NMe 2 ) g HCuCl 3 N 4 P 4 (NMe 2 ) 8 HCuCl 3 Peaks i n cm" 1 3100 s 3000 - 2750 s v b 2600 m 1900 -,2400 p r w bands 1440, 1460, 1480, 1495 p r s 1300, 1250, 1220, 1170 s 1060 m 980 s 920 s 860 m 785 w 750s, 745m p r 660s, 665s 495 m 450 m N 4 P 4 ( N M e 2 ) g •1 Peaks i n cm" 3010m, 2980s, 2860s, 2780s p r 1870, 1980, 2020, 2070 p r w 1430, 1450, 1465, 1480 p r s 1400 - 1230 p r s 1175, 1145 s 1060 m 1000 - 950 b s 910 s 820 w 725 s 630 s 520 s 470 s s - strong; m - medium; w - weak; v b - very broad; b - broad; p r - poorly resolved *Taken on a Perkin-Elmer 457 Grating Infra-Red Spectrophotometer 3000 2500 2000 : T8DTJ (Wave Number)•(cm) • Figure 17(a). Infra-red Spectrum, of N P (NMe_)ft (Wave Number)•(cm) Figure 17(a). Continued 3000 2500 2000 TBOU (Wave Number)•(cm) Figure 17(b). Infra-red Spectrum of N 4 P 4 (NMe 2 ) g HCuCl 3 - 48 -The Preparation of N^P^(NMe 2)gHCuCl^: Anhydrous copper (II) chlor ide was obtained by dehydrating f49) the hydrate i n b o i l i n g butanone . 0.3306 g (4.3 x 10" 4 moles) of T .P .D. i n 100 ml of a c e t o n i t r i l e were added to 0.0743 g (5.53 x 10" 4 moles) of copper (II) chlor ide i n 200 ml of a c e t o n i t r i l e . The brown colour of the copper (II) chlor ide so lu t ion immediately changed to a red-brown tea colour . The react ion mixture was then heated under ref lux for t h i r t y minutes, cooled to -4°C, and l e f t for fourteen hours. Flash evaporation of the so lu t ion l e f t a yellow-brown residue. This was dissolved i n d ie thy l ether [copper (II) ch lor ide i s insoluble i n th i s solvent] and f i l t e r e d to give a golden wheat-coloured so lu t ion . Upon concentration of the so lu t ion and standing, there formed 0.0872 g of non-homogeneous primrose-yellow c r y s t a l s , (m.pts. 103 - 105°C, 112°C). The previous react ion was repeated using 3.4889 g (4.55 x 10~ 3 moles) of T .P .D . and 0.6679 g (4.97 x 1 0 - 3 moles) of copper (II) ch lo r ide . However, th i s time the yellow-brown residue was washed i n 200 ml of d i e thy l ether and f i l t e r e d . The undissolved product was extracted with d ie thy l ether i n a Soxhlet extractor for two days. During the ex t rac t ion , yellow rhombohedral needles formed i n the f l a sk . These were separated from the solvent and dr ied (m.pt. 114°C). The amount of product recovered was: rhombohedral needles from extrac t ion 0.6980 g material dissolved i n solvent 0.1557 g residue l e f t i n thimble after ext rac t ion of main product 1.2006 g Total 1.9543 g - 49 -The yellow product i n the thimble was unstable to the atmosphere, turning green upon prolonged exposure. Except for a small amount of insoluble red gum the yellow product was dissolved i n benzene. This was f i l t e r e d and evaporated to give a brown o i l containing some benzene and some seemingly colourless c r y s t a l s . The o i l was d i lu ted with 150 ml of carbon te t rachlor ide and the c rys ta l s f i l t e r e d o f f as a gummy yellow powder. The c rys ta l s formed i n the f lask during extract ion (m.pt. 114°C) were r e - c r y s t a l l i z e d from d ie thy l ether forming yellow rhombohedral c r y s t a l s , (m.pt. 114 - 116°C); micro-analysis found N, 23.94; H, 7.04; C, 27.3; C l , 15.45%. Formula ( P ^ ( N M e 2 ) g H ) C u C l 3 requires N, 23.8; H, 6.95; C, 27.3; C l , 15.1%. The inf ra- red spectrum i n potassium bromide i s shown i n Table V and Figure 17 (b). Discussion: N^P 4(NMe 2)gHCuCl 3 bears an obvious s toichiometr ic s i m i l a r i t y to N^P^MegHCuCl^, both ligands being tetrameric r ings with bulky e lect ron-re leas ing groups attached and each molecule having three chlor ine atoms and one copper atom. A corresponding s t ruc tura l s i m i l a r i t y might therefore be expected. The sturcture of N^P^MegHCuCl^ has been determined by x-ray d i f f r a c t i o n ' ' 4 0 ^ and the r i n g shown to be covalent ly bonded through one nitrogen to the copper atom of the C u C l ^ - group, and through the opposing nitrogen to the proton (Figure 16). A s i m i l a r pos i t i on for the proton i n the amide der iva t ive i s suggested by the inf ra- red spectrum. - 50 -Therefore, C u C l ^ - i s most probably bound to the opposing r i ng ni t rogen. The main differences i n the spectra between the octa-amino der iva t ive and i t s copper complex are the appearance of new bands at 3100, 2600, and 920 cm" 1 , and the hypsochromic sh i f t of the peaks at 725 and 630 c m - 1 to 750 and 660 c m - 1 , and the peaks at 1175 and 1145 cm"1 to 1220 and 1170 c m - 1 . Moeller and Kokal is^ 5 0 -* observed new bands at 2300 - 2650 c m - 1 on going from the hexa-amino der ivat ives to t he i r hydrochlorides. On the basis of s i m i l a r resul ts found upon the hydrohalogenation of pyr id ine d e r i v a t i v e s ^ ' * ^ , he assigned these bands to S-H s t re tching modes. (52") In several pyridinium der ivat ives Evans assigned peaks at 3100 -3374 cm" 1 and 2800 - 2273 c m - 1 to the same stre tching modes. S imi la r bands (at 2660 and 3080 cm - 1 ) are found i n the infra- red spectrum of N^P^MegHCuClg, which i s known to be protonated on the nitrogen a tom^ 4 " ' . Peaks at 3100 and 2600 cm" 1 i n the spectrum of N 4 P 4 (NMe 2 ) g HCuCl 3 are most probably due to a s i m i l a r protonation of one of the ni trogens. However, whether protonation i s exocycl ic or endocyclic cannot be determined from these bands. Stahlberg *-54-) studied the spectrum of N 3 P 3 C 1 2 [N(CH3) 2 ] 4 and assigned peaks at 686 and 684 c m - 1 to symmetric PN 2(exo) s t re tching modes, and peaks at 755 and 751 cm - 1 to asymmetric PN 2(exo) s t re tching modes. S imi la r peaks at 630 and 725 cm" 1 are found with N.P.(NMe_) 0 4 4 Z o and may be assigned to the analogous s tretching modes (these may be unresolved doublets) . The spectrum of N 4P 4(NMe 2)gHCuClj shows a hypsochromic sh i f t of these peaks to 665, 660 c m - 1 and 750, 740 c m - 1 . This i s consistent -51 -with the presence of ir-electron withdrawing groups bonded to the r i n g . The P-N(exo) bond i n N 4P 4(NMe2)g has considerable double bond character caused by the d e r e a l i z a t i o n of the lone pa i r electrons on nitrogen into the ir-r ing bonding s y s t e m . Any bonding of the r ing through i t s nitrogen lone pa i r s to an electron withdrawing group would cause an increase i n the double bond character of the P-N(exo) bond and a r e su l t i ng hypsochromic sh i f t of the frequency of the v ib ra t i ona l modes. Furthermore, no difference i s noticed i n the frequency of the N-C s t re tching mode (1060 cm - 1 ) between the two compounds. The bonding of an acceptor to an exocycl ic group would have been expected to change th i s frequency; p a r t i c u l a r l y i f the acceptor had been a bulky copper anion. The proton and the copper atom must, therefore, be bound to the r ing ni t rogens. It seems l i k e l y that , analogous to the case of N 4P 4(NMe 2)gHCuCl the anion i s the group CuCl^" . This would be expected on e l ec t ro s t a t i c grounds to have an antipodal o r ien ta t ion r e l a t i v e to the proton. Evidence for t h i s i s found i n the structure of N^P^MegHCuCl^. Because protonation of the r ing reduces the b a s i c i t y of the nitrogen atoms, t h i s o r ien ta t ion would eliminate any che la t ion . So the copper i s most probably bound to only one r ing ni trogen. This would involve a structure analogous to that of N.P.Me„H"CuCl_. - 52 -C H A P T E R F O U R THE PREPARATION OF A PHOSPHONITRILIC COMPLEX CONTAINING A CHELATED COPPER Both s i ze and base strength should determine whether or not a phosphoni t r i le can act as a multidentate l igand. The f a i l u r e of tetrameric phosphoni t r i l i c dimethylamide to act as such a l igand indicates that a phosphonitr i le with a larger r i ng and high b a s i c i t y i s necessary for che la t ion . The hexameric phosphon i t r i l i c dimethylamide meets both (45 55) these requirements ' J and, therefore, i t s reactions under varying conditions with copper (I) chlor ide and copper (II) chlor ide have been studied. In the course of t h i s work the complex N^P^(NMe 2 ) 1 2 Cu(II )Cl C u ( I ) C l 2 has been prepared and character ized. I t i s the f i r s t phosphon i t r i l i c der iva t ive to be shown to have a chelated metal atom, and also the f i r s t compound to contain the CuCl 2 ~ i o n ^ ^ . Several attempts were made to produce complexes containing copper i n only one oxidat ion s ta te . However, only poorly defined so l ids were obtained. THE PREPARATION OF HEXAMERIC PHOSPHONITRILIC DIMETHYLAMIDE (H.P.D.) H.P.D, i s produced according to the equation: N - P . C l . ^ + 24HNMe„ + N..P,(NMe„).„ + 12NH_Me„Cl D O 12 2 6 6 2 l z 2 2 - 53 -The hexameric phosphon i t r i l i c chlor ide was already ava i lab le and the dimethylamine was the reagent grade chemical obtained from Eastman Kodak. At -78°C 50 ml (0.75 moles) of dimethylamine were slowly added to 11.9 g (0.0171 moles) of hexameric phosphon i t r i l i c chlor ide i n 75 ml of benzene. Upon warming to room temperature, the react ion mixture produced a dense white p r e c i p i t a t e . After several hours at t h i s temperature, the mixture was heated to ref lux and kept there for ten hours. F i l t r a t i o n and evaporation of the react ion mixture l e f t 7 g of a s t i c k y white powder which was assumed to be the desired product. Washing the f i l t r a t e with hot benzene removed several more grams of co-p rec ip i t a t ed product. The product was r e - c r y s t a l l i z e d repeatedly from benzene to give 6.5 g (48% of theory) of white powder, m.pt. 260CC ( l i t e r a t u r e 256°C ) . A q u a l i t a t i v e chlor ine analysis was negative, ind ica t ing complete subs t i tu t ion and the absence of the hydrochloride, NgP^fNN^) C l . The infra- red spectrum, which has not as yet been published, i s given i n Table VI and Figure 18(a). THE REACTION OF H.P.D. WITH COPPER (II) CHLORIDE Reactions i n Butanone: The butanone used was d i s t i l l e d and the f rac t ion b o i l i n g between 79 - 81°C was co l l ec ted and passed through a column containing 400 mm of alumina and 200 mm of s i l i c a g e l . Anhydrous copper (II) chlor ide was obtained from the Fisher S c i e n t i f i c Company. TABLE VI Infra-red Spectra* of H.P.D. and NgPgfNMe^ 1 2 C u 2 C l 3 3010m, 2970m, 2860bs, 2790s; p r t 2150 b w 1480s, 1460, 1450; s 1400-- 1250 v b s 1183 s 1145 m 1070, 1063; p r m 990, 970; p r s 890 m 780 m 718, 708; p r s 650 m 590 s 3010m, 2890s, 2850s, 2800s; p r 1490w, 1480m, 1460m, 1455m, 1440w, 1425w; p r 1290 s 1250 s 1200, 1175; p r s 1135 s 1062 m 980 s 859 m 795 m 768 m 743m, 730m 568 m 545 m On 515w, 509m, 490m; p r * Taken on a Perkin-Elmer 457 Grating Infra-Red Spectrophotometer t Symbols Defined on TABLE V (Wave Number)•(cm) Figure 18(a). Infra-red Spectrum of N-P,(NMe_) VI ON (Wave Number)•(cm) Figure 18(a). Continued A clear so lu t ion of 3.0054 g (3.76 x 10~ 3 moles) of H.P.D. p a r t i a l l y dissolved i n 325 ml of butanone was added to a brown so lu t ion of 1.0306 g (7.66 x 10~ 3 moles) of copper (II) ch lor ide . A br ight red colour was instantaneously formed. Shaking the react ion vessel for several seconds dissolved the H.P.D. leaving a c lear red so lu t ion with approximately 30 mg of high density dark powder. Upon standing for an hour, none of the powder v i s i b l y went in to so lu t ion , though some of i t dissolved on b o i l i n g . Concentrating, f i l t e r i n g , and cooling the so lu t ion gave 2.6758 g of rust - red f lakes , m.pt. 174°C. Further c r y s t a l l i z a t i o n from butanone yie lded 2 g of f lakes (m.pt. 175°C) and some insoluble brown powder. A second crop of c rys ta l s was recovered from the mother l iquor of the second c r y s t a l l i z a t i o n . A f i n a l r e - c r y s t a l l i z a t i o n was accomplished by slow evaporation of a butanone so lu t ion with p u r i f i e d ni t rogen, (obtained from the Canad-i a n Liqu id A i r Co , ) , and 0.79 g of small red-brown f lakes (m.pt. 180°C) were obtained. After drying i n vacuo these c rys ta l s were analyzed; the resu l t s are shown i n Table V I I . TABLE VII Elemental Analysis of Copper Complex Obtained Calculated for N 6 P 6 ( N M e 2 ) 1 2 C u 2 C l 4 Calculated for N 6 P 6 ( N M e 2 ) 1 2 C u 2 C l 3 - H C l %N %Cu %C %H 23.3 11.40 26.8 6.80 %C1 13.21 13.25 23.62 11.90 26.99 6.75 13.29 23.59 11.88 26.96 6.83 13.28 - 59 -The inf ra - red spectrum i s : 2900*, 2800*, 2750*, p . r . s , ; 1725* v .w . ; 1485*, 1460*, 1450*, p . r . s . ; 1420* w.; 1280 s; 1250 s; 1185 +, 1170 f 11451", p . r . s . ; 1050+m; 850+m; 795+m; 766+m; 745+m; 725+m i n uni ts of cm - 1 , (abbrevia t ions defined i n Table V.) The absence of any peaks i n the N-H s t re tching region i s not conclusive evidence for the absence ei ther of the hydrochloride, N^P^. (NMe2) 1 2 C u ^ l ^ - H C l , or of free hydrogen ch lo r ide . The formula, N g P ^ f N f ^ ) ^ ^ U 2 C 1 4 w a s n o t c o n s i - s t e n ' t with the propert ies of the compound. When a small sample was heated to 153°C i n vacuo the chlor ine analysis dropped to 10.5%, ind ica t ing a loss of one chlor ine atom per molecule. This could be due to the loss of eas i ly removable hydrogen chlor ide from a complex having the formula, N,P.. (NMe„).. _ Cu„Cl„-HCl. Table VII shows that the two formulas have 6 6 2 12 2 3 almost i d e n t i c a l percentage compositions and are both consistent with the ana ly t i c a l r e s u l t s . The hydrogen chlor ide could come from the oxidat ion of the solvent by the copper (II) ch lor ide , e .g . : -CH 2 C0- + 2CuCl 2 •> -CHC1C0- + 2CuCl + HC1 I t i s important to note that the formula, N^P^(NMe 2)^ 2 ( A ^ C I ^ ' H C I , requires the presence of a copper (I) atom which i s also produced i n the redox reac t ion . Succeeding work showed that the complex does i n fact contain two oxidat ion states of copper. I t might, therefore, be expected that copper (II) chlor ide and H.P.D. would react d i f f e r en t ly i n .a non-reducing solvent , a c e t o n i t r i l e , than i n the reducing solvent * Taken from halocarbon o i l mull t Taken from nujol mull ^ - 60 -butanone. This react ion and others s i m i l a r to i t are studied next. REACTIONS IN ACETONITRILE Reaction of Copper (II) Chloride with H . P . D . ; 0.98909 g (1.24 x I O - 3 moles) H.P.D. p a r t i a l l y dissolved i n 200 ml o f a c e t o n i t r i l e , were added to a murky green solu t ion of 0.3356 g (2.49 x 1 0 - 3 moles) of copper (II) ch lor ide i n 200 ml of a c e t o n i t r i l e . One minute after mixing, a red-tea colour appeared and a l l but a small amount of H.P.D. went in to so lu t ion . After standing for ten hours, the react ion mixture was f i l t e r e d to give a c lear red so lu t ion . Removing the solvent under reduced pressure at 50°C l e f t 1.2 g of a dark red powder, m.pt. 169°C. This was c r y s t a l l i z e d from an acetonitr i le-benzene mixture to produce 0.6932 g of brown c r y s t a l l o i d needles, m.pt. 170°C. The inf ra- red spectrum taken i n potassium bromide wafer was: 2950, 2900, 2850, 2825, p . r . s . ; 1500 b.m.; 1300s; 1260; 1190, 1170, 1130 p . r . m . ; 1065m; 985 b . s . ; 860s; 800m; 750m; 730m; i n uni ts of c m - 1 , (Abbreviations defined i n Table V . ) , and i s d i f ferent from that of N.P , (NMe„) 1 0 C u 0 C l , . D O Z 1Z Z J Microscopic examination showed the c rys ta l s to be of two types,-transparent, ruby red needles and opalescent brown needles. The former were mechanically separated and then analyzed; (N, 22.51; H, 6.96; C, 29.42%). Further attempts at r e - c r y s t a l l i z a t i o n from a c e t o n i t r i l e f a i l e d , and produced a decomposition of the product, evident by a depressed melting point of 153 - 156°C. Under vacuum at 92°C, t h i s material - 61 -decomposed into a white sublimate and a dark red powder. Repeated r e - c r y s t a l l i z a t i o n from chloroform and carbon te t rachlor ide solutions gave a golden-yellow powder, m.pt. 171 - 173°C. The inf ra- red spectrum was i d e n t i c a l to that previously obtained. Analysis of th i s material gave: N, 20.29; P, 12.81; C, 23.88; H, 5.82; CI , 20.97; Cu, 15.09; 0, 0.96%. Assuming a molecule having eighteen nitrogen atoms, th i s analys corresponds to: N 6 P S . 1 3 ( N 1 2 C 2 4 . 7 2 H 7 2 . 2 ) C u 2 . 9 5 C 1 7 . 3 ° 0 . 7 5 which i s most probably: N 6 P 6 (NMe 2 ) 1 2 HC1(CuCl 2 ) 3 (H 2 0) Further p u r i f i c a t i o n by th in layer chromatography f a i l e d owing to the decomposition of the sample. No further p u r i f i c a t i o n was attempted. Reaction of Copper (I) Chloride with H . P . D . : Copper (I) chlor ide was prepared by the reduction of an aqueous so lu t ion of copper (II) ch lor ide with sulphur d i o x i d e v . The product was washed with sulphurous ac id and g l a c i a l ace t ic ac id , then dr ied i n vacuo at 100°C. The pure white powder produced by th i s method was stored under nitrogen i n a stoppered and wax sealed v i a l . The a c e t o n i t r i l e used i n th i s experiment was dr ied by the method described i n Chapter 2. A l l oxygen was removed by heating the - 62 -so lu t ion under ref lux and bubbling i n p u r i f i e d nitrogen for two hours. The H.P.D. was heated at 90°C i n vacuo for two hours. The melting point was then 260°C. A l l reactions were car r ied out under p u r i f i e d ni t rogen. A c lea r , colourless so lu t ion of 1.98 g (20.0 x 1 0 - 3 moles) of copper (I) chlor ide i n 500 ml of a c e t o n i t r i l e was added to a c lea r , colourless so lu t ion of 4.65 g (5.82 x 1 0 - 3 moles) of H.P.D. partly-dissolved i n 100 ml of a c e t o n i t r i l e . A further 5.2 g (6.5 x 10" 3 moles) of H.P .D. was added to the react ion mixture. No colour change, was noticed during ei ther add i t ion . Warming of the reac t ion mixture dissolved a l l the remaining H.P.D. During th i s warming, the so lu t ion turned a pale yel low. Concentration of the so lu t ion produced a dark yellow so lu t ion and a brown-white p r ec ip i t a t e . Flash evaporation revealed 5.41 g of dark powder. Extract ing th i s powder with benzene for eight hours gave a clear so lu t ion and l e f t insoluble brown powder i n the Soxhlet thimble. Flash evaporation of the so lu t ion gave 0.3331 g of H.P.D. (m.pt. 243°C; i n f r a -red spectrum as i n Table V I ) . The benzene insoluble mater ial was washed with dry chloroform to produce a red so lu t ion with a green re f lex and leaving an insoluble black powder. The black powder was insoluble i n a l l the common organic solvents , but dissolved i n sulphuric and hydrochloric ac ids , as we l l as ammonium hydroxide. I ts in f ra - red spectrum was a blank; a q u a l i t a t i v e copper test performed on i t was p o s i t i v e ; and i t s melting point was i n excess of 320°C. A l l of t h i s i s consistent with the black powder being - 63 -me ta l l i c copper. This , i n turn suggests disproport ionat ion during the react ion according to the equation: 2CuCl -> Cu° + C u C l 2 Because copper ( I I ) , un l ike copper ( I ) , forms coloured compounds, disproport ionat ion would also explain the presence of a coloured product. When i t was cooled to -4°C, the chloroform so lu t ion formed 1.19 g o f c rys ta l s having the appearance of c lo t ted blood. These were a i r s ens i t i ve , turning green upon prolonged exposure to the atmosphere. Upon drying i n vacuo, the compound had a melting point of 191 -193°C and i n nujol and halocarbon o i l mul ls , i t s inf ra- red spectrum was: 3012*m, 2930*s, 2900*s, 2850*s, 2810*s, 1835*w, 1800*m, 147*s, 1465*s, 1455*s, 1445*s, 1435*m, 1300* +s, 1255 +s, 1210 +s, 1190 +s, 1175 +s,. 1150 +s, 1070+m, 980 + s, 863+m, 800+m, 773+m, 750 + s, 735 +s, i n uni ts o f cm" 1 , (abbreviations defined i n Table V ) . They were soluble i n chloroform and a c e t o n i t r i l e , but insoluble i n pe t ro l ether, carbon te t rach lor ide and benzene. Further attempts at r e - c r y s t a l l i z a t i o n from a c e t o n i t r i l e and chloroform resul ted i n adul terat ion of the product evident by a depressed melt ing poin t , 190 - 191°C. Attempted p u r i f i c a t i o n by th in layer chromatography was s i m i l a r l y unsuccessful. * Taken from halocarbon o i l mull t Taken from nujol mull - 64 -Reaction of Copper (I) Chloride and Copper (II) Chloride with H . P . D . : The H . P . D . , copper (I) chlor ide and anhydrous, oxygen free a c e t o n i t r i l e used i n th i s experiment were prepared i n an i d e n t i c a l manner to those used i n the previous experiment. The anhydrous copper chlor ide was obtained from the Fisher S c i e n t i f i c Company. 0.1761 g (1.31 x 10~ 3 moles) of copper (II) chlor ide were added to 300 ml of warm a c e t o n i t r i l e under p u r i f i e d ni t rogen. There was an immediate colour change, g iv ing a yellow so lu t ion . To th i s so lu t ion 0.1335 g (1.34 x 10~ 3 moles) of copper (I) chlor ide powder was added. F i n a l l y , 300 ml of a c e t o n i t r i l e containing 1.0235 g (1.28 x 10 3 moles) of p a r t i a l l y dissolved H.P.D. was added. There was an immediate colour change after the f i n a l addi t ion r e su l t i ng i n a weak tea-coloured so lu t i on . The H.P.D. dissolved completely upon b o i l i n g the react ion mixture. During th i s heating, the so lu t ion became appreciably darker and assumed a dark red colour . Concentration of the so lu t ion yielded 0.8 g of yellow f lakes , m.pt. 179 - 181°C. The c rys ta l s were reported to be c ry s t a l l og raph i ca l l y i d e n t i c a l to those of N^P^(NMe 2 ) . ^Cu^ l^ as prepared by the dehydro-f58") halogenation of the product prepared previously i n butanone . The analysis was: N, 24.15; P, 18.22; C, 28.18; H, 6.86; Cu, 12.33; CI , 10.28% calcula ted for N 6 P 6 ( N M e 2 ) 1 2 C u 2 C l 3 : N, 24.4; P, 18.05; C, 27.9; H, 6.96; Cu, 12.3; C I , 10.2%. The inf ra- red spectrum taken i n a potassium bromide wafer i s i d e n t i c a l to the spectrum of N f iP f i(NMe_).. „Cu»Cl_ as shown i n Table V I . - 65 -Discussion: The react ion between hexameric phosphoni t r i l i c dimethylamide and copper (II) ch lor ide i n butanone produces c rys ta l s of N ^ P ^ f M ^ ) C u C l^ -HCl which dehydrohalogenates upon warming to form the compound, N^P^(NMe 2)j2 C i ^ C l g . This contains two oxidat ion states of copper and can be produced by the react ion of H.P.D. with a mixture of copper (II) chlor ide and copper (I) ch lor ide i n the non-reducing solvent, a c e t o n i t r i l e form hydrochlorides i n the presence of secondary amines. However, i f a ir-electron withdrawing group were to complex strongly to the r ing nitrogens of such a hydrochloride, there would be a considerable weakening of the forces holding the hydrogen chlor ide to the molecule. For example, i f the system: Phosphoni t r i l i c amides are such strong bases that they often \ / H C complexes with a n-electron to form: \ / i =P-N H - 66 -there would be a reduction i n density of the lone pa i r on N(2). The + N(2)-H bond would be weakened and hydrogen chlor ide could be more r ead i ly removed from the molecule. I f the complexing were strong enough, hydrogen chlor ide could be r ead i ly removed. Hydrogen chlor ide must, therefore, have been present at some time during the reac t ion . This would occur i f copper (II) ch lor ide were reduced by the solvent . Ketones are known reducing agents, the fol lowing react ion occuring for acetone: -CH 2 -CO- + 2CuCl 2 -> -CHCl-CO- + 2CuCl + HCl S imi la r reduction could also explain the presence of hydrogen chlor ide i n N 4 P 4 M e g H C u C l 3 ^ 4 9 \ (N 4 P 4 Me g H)£oCl 4 ( ' 4 9 - ) , and (NMe2) g H C u C l 3 . Each of these complexes was prepared using metal chlorides previously dehydrated i n butanone. The formula, N^P^fNMe,,)^ C u ^ l ^ H C l suggests that copper i s present as copper (I) and copper ( I I ) . This , and the presence of hydrogen chlor ide could be explained by a react ion path s i m i l a r to : N 6 P 6 ( I W e 2>12 + C U C 1 2 * N 6 P 6 ^ 1 2 C U C 1 2 CH 3 CH 2 'CO-CH 3 + 2CuCl 2 -* CH3CHC1-CO-CH3 + 2CuCl + HCl ' N 6 P 6 ( N M e 2 ) 1 2 , C u C 1 2 + C U C 1 " N 6 P 6 ^ 1 2 C u 2 C 1 3 , ( N 6 P 6 ( N M e 2 ) 1 2 C u 2 C l 3 + HCl ( N 6 P 6 ( N M e 2 ) 1 2 C u ^ l ^ H C l However, i t remains to be shown that the complex does i n fact contain copper (I) and copper (II) atoms. To do t h i s , purely chemical evidence - 67 -i s the most s a t i s f y i n g . A c e t o n i t r i l e i s a non-reducing solvent, therefore, copper (II) chlor ide w i l l not be reduced i f dissolved i n i t . I f N^P^NN^) - ]^ ^U2^^3 could be produced by the react ion of copper (II) chlor ide with H.P.D. i n a c e t o n i t r i l e , then i t must contain only copper ( I I ) . On the other hand, i f i t contains both oxidat ion states, then i t cannot be produced by th i s reac t ion . The react ion was invest igated, and while an unstable product was i so l a t ed , inf ra- red spectroscopy and chemical analysis showed i t was not N ,P , (NMe_).. _ Cu„Cl„. 6 6 2 1 2 2 3 S i m i l a r l y i f the complex could be produced by the react ion of copper (I) chlor ide with H.P.D. i t would be pos i t ive evidence that i t contained only copper ( I ) . This react ion was invest igated, and d id not produce the required product. Furthermore, ti^P^QMe^)Ch^Cl^ i s a coloured compound and there i s no known coloured complex containing only copper (I) as an acceptor, (except when the colour resu l t s from charge transfer bands) . These experiments showed that N^P^fNMe^)^ ^u2^3 c a n n o t be produced from the react ion of H.P.D. with ei ther copper (I) chlor ide or copper (II) chlor ide i n a non reducing solvent such as a c e t o n i t r i l e . This indicates that the metal i n the complex does not belong exc lus ive ly to one oxidat ion s ta te . And the presence of one atom each of copper (I) and copper (II) i n the complex was proven by the react ion of H.P.D. with an equimolar mixture of copper (I) chlor ide and copper (II) chlor ide i n a c e t o n i t r i l e which produces N ^ P ^ N N ^ ) - ^ C ^ C l ^ . Because no reduction of the copper (II) ch lor ide could have occured, and because s i m i l a r solut ions containing e i ther of the copper chlorides do not produce the - 68 -desired product, N^P^NMe,,).^ ("u2<"^3 m u s t > therefore, contain both oxidat ion states of copper. The absence of hydrogen chlor ide from the product obtained from the a c e t o n i t r i l e so lu t ion shows that i t s presence on the product from the butanone so lu t ion i s a r e su l t of a redox react ion and i s not a necessary part of the copper.complexing process. I f i t were, then N^P^(NMe 2 ) 1 2 C u 2 C l 3 could not be produced except through the hydrochlorid intermediate. In conclusion, the fol lowing facts are now known about N^P^(NMe 2 )^ 2 Cu^jCl^. I t can be produced by the react ion of H.P.D. with e i ther copper (II) ch lor ide i n butanone or with a mixture of copper (I) chlor ide and copper (II) chlor ide i n a c e t o n i t r i l e . The mono-hydro-chlor ide of the complex i s r ead i ly made and loses hydrogen chlor ide upon heating. F i n a l l y , the complex contains both copper (I) and copper ( I I ) . The actual formation of the N^P^(NMe 2)^ 2 C u ^ l ^ i s , however, s t i l l unknown, although many p o s s i b i l i t i e s are evident. The number of these i s somewhat l imi t ed by the known co-ordinat ion numbers of copper (I) and ( I I ) C 5 6 ' 5 9 ) . I f the compound i s a true complex, then i t can only be: N 6 P 6 ( N M e 2 ) 1 2 Cu(I) C u ( I I ) C l 3 , N 6 P 6 ( N M e 2 ) 1 2 Cu(I)Cl C u ( I I ) C l 2 , or N 6 P 6 ( N M e 2 ) 1 2 C u ( I ) C l 2 C u ( I I ) C l . On the other hand, i f i t i s i n fact a s a l t then i t can only be: N 6 P 6 ( N M e 2 ) 1 2 C u + 2 C u C l 3 " 2 , or N^P^(NMe 2 )^ 2 CuCl + CuCl 2 ~.• The choice waits upon evidence presented i n the fol lowing chapter and here no dec is ion regarding the true formation of N ^ P ^ ( N M e 2 ) 1 2 C u 2 C l 3 can be made. - 69 -C H A P T E R F I V E PHYSICAL STUDIES OF N,P,(NMe 0),„Cu„Cl„ D O I iZ I 5 Introduction arid Summary: An x-ray c rys t a l structure determination done on the complex shows that of the p o s s i b i l i t i e s referred to i n the previous chapter, the structure i n the s o l i d i s ion ic being N.P,(NMe„)..„CuCl+CuCl ~. The D O I 1.1 Z cat ion i s the only example of a phosphon i t r i l i c der iva t ive which forms a chelate complex with a t r a n s i t i o n metal . The deta i led structure found i n the c rys t a l does not pe r s i s t i n so lu t ion . The existence i n so lu t ion of the ca t ion , N.P,(NMe_) n „Cu + + D O l i-l i s suggested by evidence obtained from the react ion of the complex with s i l v e r n i t r a t e . This cat ion would have less in te rna l crowding than the corresponding N ^ P ^ ( N M e 2 ) c a t i o n and might, therefore, be favoured i n so lu t ion . Electrochemical measurements support t h i s supposi t ion. The s a l t i s not a 1:1 e l ec t ro ly te i n a c e t o n i t r i l e so lu t ion , the l i m i t i n g conductance being higher than has been reported for any such e l e c t r o l y t e . The existence of the anion, C u C ^ " i s proven by the molecular s t ruc ture . CuCl^" i s l inea r and has bond lengths corresponding to the sum of the covalent r a d i i for Cu (I) and CI atoms. A possible path for producing anions of the type Cu(NO„) ~ i s suggested by the s i l v e r n i t r a t e 70 -experiment. Magnetochemical studies on the s o l i d complex showed that i t contained only one copper (II) atom per molecule. The inf ra- red spectrum of the complex i s very s i m i l a r to. that of the parent H.P.D. This i s a t t r ibuted to the basic s i m i l a r i t y of the two molecules. Due to the lack of information concerning the inf ra- red spectra of phosphoni t r i l i c de r iva t ives , very few assignments of the peaks i n e i ther spectrum could be made. The Crys t a l .S t ruc tu re .o f N 6 P 6 ( N M e 2 ) 1 2 C u 2 C l 3 : The c r y s t a l structure of N^P^(NMe 2 )^ 2 C u 2 C l 3 was determined (581 i n t h i s department J and shown to be formed from two discre te ions, N^P^(NMe 2 ) j 2 CuCl + and C u C l 2 . The cat ion contains a chelated copper with a d is tor ted square pyramidal environment. The anion i s l i n e a r . The formula N^P^(NMe 2 )^ 2 CuCl + CuCl 2 _ can be proven by consideration of s t ruc tura l evidence alone, (Figures 19,20; Table V I I I . The chelated copper, Cu(2), i s penta co-ordinate, and while such co-ordinat ion i s we l l known for copper (II) atoms f (for example, dimethylglyoximatocopper ( I I ) ) , t e t ra co-ordinat ion i s the highest observed for copper ( I ) , s i l v e r ( I ) , or gold (I) atoms . Thus, i t can be presumed that th i s copper atom i s d iva len t . o Furthermore, the Cu(2)-Cl(2) bond length (2.28A) i s better approximated by the sum of the covalent r a d i i for Cu (II) and C l , o o (2.34A), than for Cu(I) and C l , (2.17A). Thus, the phosphon i t r i l i c de r iva t ive i s far more l i k e l y to be the ca t ion , N^P^(NMe 2 )^ 2 Cu(II)C1 + Figure 19. Phosphoni t r i l i c Ring of N.P.(NMe ) C u C l + Anion atom c o o r d i n a t i o n (HMe 2 groups o m i t t e d f o r c l a r i t y ) . - 73 -TABLE VII I o Bond lengths (A) and valency angles (degrees), with (791 standard deviations i n parentheses. Cu(2)-N(l) Cu(2)-N(2) 0u(2)-Cl(2) 2.03(2) 2.11(2) 2.28(1) Cu(l)-Cft(l) 2.06(1) [2.11(1) corrected f o r l i b r a t i c N ( l ) - P ( l ) 1 . 6 2 ( 2 ) N ( 4 ) - 0 ( 1 ) 1 . 4 6 ( 3 ) N ( l ) - P ( 3 ' ) 1 . 6 5 ( 2 ) N(4) -C(2) 1 . 4 3 ( 3 ) N ( 2 ) - P ( l ) 1 . 6 0 ( 2 ) N ( 5 ) -C(3) 1 . 4 5 ( 3 ) N ( 2 ) - P ( 2 ) 1 . 6 1 ( 2 ) N ( 5 ) -0(4) 1 . 4 5 ( 3 ) N ( 3 ) - P ( 2 ) 1 . 5 3 ( 2 ) . N ( 6 ) - C ( 5 ) 1 . 4 6 ( 3 ) N ( 3 ) - P ( 3 ) 1 . 5 7 ( 2 ) N(6) - C ( 6 ) 1 . 4 9 ( 3 ) P ( l ) - N ( 4 ) 1 . 6 2 ( 2 ) N ( 7 ) - C ( 7 ) 1 . 5 5 ( 3 ) P ( l ) - N ( 5 ) 1 . 6 4 ( 2 ) N ( 7 ) - 0 ( 8 ) 1 . 4 4 ( 3 ) P(2)-N(6) 1 . 6 5 ( 2 ) N ( 8 ) - 0 ( 9 ) 1 . 4 6 ( 3 ) P ( 2 ) - N ( 7 ) 1 . 6 7 ( 2 ) N ( 8 ) - 0 ( 1 0 ) 1 . 5 0 ( 3 ) P ( 3 ) - N ( 8 ) 1 . 6 7 ( 2 ) N ( 9 ) - C ( l l ) 1 . 5 0 ( 3 ) P ( 3 ) - N ( 9 ) 1 . 6 8 ( 2 ) N ( 9 ) - 0 ( 1 2 ) 1 . 4 7 ( 3 ) 7 1 . 2 ( 7 J N { 1 ) - r U ;-ra^ ; 9 7 . 2 ( 1 0 N ( 2 ) - C u ( 2 ) - N ( l » ) 9 9 . 1 ( 8 ) N(l) - P ( l ) - N ( 4 ) 1 1 8 . 2 ( 1 1 N(2)-Cu(2)-N(2» ) 1 2 0 . 5 ( 1 1 ) N(l) - P ( D - N ( 5 ) 1 0 8 . 2 ( 1 2 N(l )-Cu(2J-N(l» ) 1 6 0 . 9 ( 1 2 ) N(2 ) - P ( l ) - N ( 4 ) 1 1 1 . 3 ( 1 1 C H ( 2 ) - C u ( 2 ) - N ( 2 ) 1 1 9 . 8 ( 5 ) N(2 ) - P ( D - N ( 5 ) 1 1 7 . 9 ( 1 1 CA(2)-Cu(2)-N(l) 9 9 . 6 ( 6 ) N(4) - P ( D - N ( 5 ) 104.5(12 CA(1)-Cu(l)-Cjt(l») 1 7 9 . 5 ( 9 ) N(2) -P(2)-N(3) 1 1 0 . 0 ( 1 0 N ( 2 ) - P ( 2 ) - N ( 6 ) 1 1 8 . 5 ( 1 2 N(2 ) - P ( 2 ) - N ( 7 ) 1 0 4 . 3 ( 1 1 P(l)-N(l)-P(3') 1 3 0 . 7 ( 1 3 ) N ( 3 ) - P ( 2 ) - N ( 6 ) 1 0 3 . 8 ( 1 2 p ( i j- ; j ( i j-'Ju(2j 9 6 . 9 ( 1 0 ) N ( 3 ) - P ( 2 ) - N ( 7 ) 1 1 5 - 5 ( 1 2 P ( 3 r )-N(l J - C u ( 2 ) 1 1 8 . 4 ( 1 1 ) N ( 6 ) - P ( 2 ) - K ( 7 ) 1 0 5 . 2 ( 1 2 P ( 1 ) ~ N ( 2 ) - P ( 2 ) 1 3 7 . 6 ( 1 3 ) N ( 3 ) -P(3)-N(l») 1 1 5 . 4 ( 1 0 P ( l ) - N ( 2 ) - C u ( 2 ) 9 4 . 5 ( 9 ) N ( 3 ) -P(3)-N(S) - 1 0 6 . 0 ( 1 1 P(2)-N(2)-Cu(2) 1 2 7 . 9 ( 1 1 ) N ( 3 ) - P ( 3 ) - M ( 9 ) 1 0 9 . 6 ( 1 1 P ( 2 ) - N ( 3 ) - P ( 3 ) 1 3 2 . 4 ( 1 3 ) N(l» ) - P ( 3 ) - N ( 8 ) 1 1 0 . 7 ( 1 0 N(l» ) - P ( 3 ) - N ( 9 ) 1 0 5 . 3 ( 1 1 N ( 8 ) - P ( 3 ) - N ( 9 ) 1 0 9 . 8 ( 1 0 C-N-C 1 1 1 - 1 1 8 , mean 1 1 4 P-N-C 1 1 4 - 1 2 7 , mean 1 2 0 74 -than the uncharged species, N^P^. (NMe 2)22 <" uCI)C1. The second group found i n the c r y s t a l s tructure i s then the anion C u C l 2 . This anion has been postulated to ex is t i n solution(64,65) but the evidence has been mainly electrochemical and i t s v a l i d i t y has been se r ious ly ques t ioned^ 2 - ' . Sui table analogues for CuCl 2 ~ do e x i s t , however, i n the anions A g C l 2 " , A u C ^ " ^ ^ , and A u B r 2 _ . A l l of these are l i n e a r , as i s the postulated C u C l 2 _ anion. Furthermore, a co-ordinat ion number of two has never been found for e i ther Cu (II) or Ag (II) atoms. I t should also be noted that the difference between the observed C u ( l ) - C l ( l ) bond length, and that predicted from the sums o f the covalent r a d i i of the atoms ' , i s 0.06A. This difference i s of the same order as those found for A g C l 2 ~ , AuCl 2 ~ and AuBr 2~ (Table IX) . TABLE IX ACTUAL AND PREDICTED BOND LENGTH ANION . C u C l 2 " A g C l 2 " A u C l 2 - AuBr 2 sum of co- o valent r a d i i (A) 2.17 2.38 2.33 ,2.45 o bond length (A) 2.11 2.36 2.31 2.35 o difference (A) .06 .02 .01 .10 A l l of these facts point to the existence of the anion, C u C l , , - . This , together with the evidence for the ca t ion , N ^ P ^ N M e ^ ^ C u C l * , i s su f f i c i en t proof that the complex, N^P^(NMe 2 ) 1 2 C u 2 C l 3 i s i n fact the s a l t , N ^ P 6 ( N M e 2 ) 1 2 C u C l + C u C l 2 - . I t should be noticed i n passing that except for the structure of copper (I) oxide , there i s almost no other - 75 -s t ruc tura l evidence for the existence of the two co-ordinate copper (I) atom. The structure and the conformation of the 'N,P,(NMe„).„CuCl + o fa z 11 cat ion are in t e res t ing , espec ia l ly when compared to those pf the (271 parent N^P^CNN^)-^ . One of the in te res t ing features of the H.P.D. s tructure i s the effect -rr-bonding has on the exocycl ic P-N bond lengths and on the p lana r i ty of the P-N system. X M e The dimethylamide group formally has a lone pa i r of electrons on the nitrogen atom which i n the phosphon i t r i l i c system i s donated to the adjacent phosphorus atom. This causes a shortening of the exocycl ic P-N bond length from a calculated s ingle bond length value of 1.74A^ ' ° (27) to 1.675 and 1.663A k . The most obvious effect of th i s d e r e a l i z a t i o n i s seen i n the sums of the angles around the exocycl ic nitrogen atom. In the H.P.D. molecule, they are 357.5° and 348.7°, i nd ica t ing a f l a t t en ing ^ M e • of the P-N group. This i s the resu l t of the loss of lone pa i r electron Nle density on the nitrogen atom which allows the methyl groups and the phosphorus atom to move away from each other i n the d i r ec t i on of the lone p a i r . The r ing ir-bonding system involves the donation of the lone pa i r on the endocyclic nitrogen atom to a su i tab le , vacant d -o rb i t a l on the phosphorus atom. However, back bonding from the exocycl ic nitrogen increases the ir-electron density of the phosphorus atom. As i t does so, lone pa i r electrons are increas ingly l o c a l i z e d on the r ing nitrogen atoms, so lengthening the endocyclic P-N bond length. When the lone pa i r on an exocycl ic nitrogen donates to the adjacent phosphorus atom, the fol lowing effects occur: - 76 -1) the exocycl ic P-N bond length i s shortened, 2) the sum of the angles around the exocycl ic nitrogen atom increases, 3) the endocyclic P-N bond length increases. Ident ica l effects would be noticed i f the r ing nitrogen lone pa i rs were removed from the tT-bonding system by complexing, i . e . the r e su l t ing decrease i n the u-electron density on the phosphorus atom would resu l t i n a-larger n-electron donation from the exocycl ic nitrogen atoms. In N^P^. (NMe2) . ^CuCl* the copper atom removes -rr-electron density from four r ing nitrogen atoms, and the expected effects are a l l observed. o o The average endocyclic P-N bond length i s 1.60A compared to 1.563A for o the parent, and the average exocycl ic P-N bond length i s 1.652A as o compared with 1.669A for the parent. While these differences are not large i n themselves, they are i n the correct d i r e c t i o n . Comparisons between the i n d i v i d u a l bond lengths wi th in the cat ion are also in t e re s t ing . There are two Cu-N bond lengths, one of o o 2.03A and one of 2.11A. This difference could eas i ly be due to the effect of s t e r i c factors on the alignment of the nitrogen atom lone p a i r s , i . e . as the cat ion i s very crowded, optimum bonding overlap cannot be expected for a l l the Cu-N bonds. The ir-electron donation from N(l) and N(2) atoms to the central copper atom lowers the bond order of the corresponding N-P bonds. This i s evident i n the increased bond lengths of the ca t ion , ( P ( 3 ' ) - N ( l ) , o o o o 1.65A; N ( l ) - P ( l ) , 1.62A; N ( 2 ) - P ( l ) , 1.60A; N(2)-P(2), 1.61A), compared to o the parent N 6 P 6 (NMe 2 ) (1.563A). The least affected are the N(3 )-P(3) o o bond length, (1.57A), and the N(3)-P(2) bond length, (1.53A). In both cases, the nitrogen lone pa i r electrons donate f u l l y in to the r ing - 77 -iT-system. I t would be expected that P ( l ) have a lower -rr-electron density than e i ther P(3 ') or P(2) . In other words, P ( l ) i s joined to two complexing nitrogen atoms and P(3') and P(2) to only one. This resu l t s i n the shortest exocycl ic bond lengths occuring at P(l)-N(4) and P ( l ) -N(5 ) , (mean, 1.63(1)1). Because the Cu(2)-N(l) bond i s shorter than the Cu(2)-N(2) bond, i t would be reasonable to assume that the lone pa i r density would be lower on N( l ) than on N(2). However, the sum of the angles around N(l ) i s 346°, while around N(2) i t i s 360°. This apparent anomaly can be explained by assuming that s t e r i c factors have caused a difference i n the effectiveness of the Tr-system at N(l) and N(2). Therefore, N(2) lone pa i r electrons are more l i k e l y to form bonds to phosphorus than are those of N ( l ) . This i s evident i n the shortness of the N(2)-P(l) o o and N(2)-P(2) bonds, (1.60A and 1.61A respec t ive ly ) , compared to the N ( l ) - P ( l ) and N( l ) -P(3«) bonds, (1.62A and 1.65A re spec t ive ly ) . Again, the difference i s small but i n the correct d i r e c t i o n . The configurat ion of the r i ng i s mainly caused by chela t ion which r e s t r i c t s the s i ze of the r i n g , p u l l i n g atoms N ( l ) , N(2), N ( l ' ) and N(2') towards the centre, so g iv ing an average endocyclic PNP angle f 271 of 134° compared to 147.5°for the parent phosphonitr i le . The s t r a i n r e su l t i ng from th i s crowding i s p a r t i a l l y offset by chela t ion i t s e l f which lowers the electron density of the r i ng bonds. An i l l u s t r a t i o n of t h i s i s the N ( l ) - P ( l ) and P(l ) -N(2) bonds which are the most affected by electron withdrawal, and the N(1)P(1)N(2) angle which i s the lowest of t h i s type found i n phosphoni t r i les . The angles at P(2) and P(3) , (110.0° and 1 1 5 . 4 ° ) , are also smaller than that found for the parent, (120 - 78 The angle P(3)N(3)P(2) i s 132.4° which i s s i m i l a r to the same type of angle i n N ^ f N M e j g , 1 3 3 ° ^ 2 6 ^ ; and N ^ M e g , 1 3 2 ° ( 2 6 \ but f 27) s i g n i f i c a n t l y smaller than that i n N^P^(NMe 2 ) 1 2 , 147.5° . This large » angle i s an outstanding feature of the parent and i s a t t r ibuted to s t e r i c ef fec ts . Wagner and Vos suggest that for N^P^(NMe 2 )^ 2 > no angle less than 148° could accomodate the bulky dimethylamide groups without at least one unacceptably short C . . . C distance. Because a smaller angle i s found for the N^P^ (NMe2) . ^CuCl* ca t ion , high s t e r i c crowding should be evident i n the intramolecular distances. In the N^P^(NMe 2 )^ 2 CuCl + ca t ion , there are three types of intramolecular distances of in t e res t . These are: 1) the distance between the carbon atoms of the lower exocycl ic groups and the chlor ine atom, CI (2); , 2) the distance between the carbons of the upper and lower exocycl ic group on the same phosphorus atom; and 3) the distance between the carbon atoms of the upper exocycl ic group on di f ferent phosphorus atoms. Some of these distances are shown i n Table X for N,P,. (NMe„), „ and o b z u N 6 P 6 ( N M e 2 ) 1 2 C u C l + . TABLE X INTRAMOLECULAR DISTANCES : - Distance i n Distance in Sum o f Van der Waals Type of Distance Species N 6 P 6 ( N M e 2 ) 1 2 N^P^NMe^ 1 2 C u C l + Rad i i o o 1) Smallest , 3.56A 3.8A o o o 2) Smallest 3.52A 3.23A 3.8A o o o 3) Smallest 3.77A 3.689A 3.8A - 79 -In each of these types of intramolecular distances, there are distances smaller than the sum of the i nd iv idua l Van der Waals r a d i i , which i s a d i r ec t i nd ica t ion of s t e r i c crowding due to the pos i t i on of the chlor ine atom. This atom forces the lower exocycl ic methyl' groups to bend upwards, thus•increasing the crowding of the exocycl ic group on the same phosphorus atom, which i n turn, increases the crowding of the methyl groups on di f ferent phosphorus atoms. The chlor ine has therefore pushed a l l the methyl groups up and away from i t s e l f , thus the smallest C . . . C bond distances are found i n the N,P,(NMe„),~CuCl + ion and not i n o o Z 1z the parent N^P^fNMe^) A further effect of crowding i s the d i s t o r t i o n of the Cu(2) atom from square pyramidal towards a t r igona l bipyramidal environment. -D i s t o r t i o n raises the four r ing nitrogens out of a planar environment, pushing up the whole r ing and the exocycl ic methyl groups with i t , and i s so obvious i t becomes u n l i k e l y that any smaller phosphoni t r i l i c amide could chelate i n the same manner; a further ind ica t ion that N.P.(NMe ) c 4 4 2. 8 would not chelate with copper. The Infra-Red Spectrum of N,P,(NMe„).„Cu„Cl„: O O Z YZ Z o The infra- red spectra of H^P^(W[e^)^^Cu^Cl^ and the parent ^6^6(^e2-' 12 a r e v e r y s i w i l a r (Figure 18, Table V I ) . This i s reasonable because the complex has retained both the phosphoni t r i l i c r ing and the exocycl ic groups of the parent. Moreover, the spectra were taken i n a region where Cu-N or Cu-Cl v i b r a t i o n a l modes would not produce peaks, so only the r i n g and exocycl ic groups would contribute to the spectrum. However, differences between spectra would be expected due to electron - 80 -withdrawal from the r ing i n the complex and changes i n symmetry. Unfortunately, very l i t t l e information exis ts about the assignment of bands i n the spectra of phosphon i t r i l i c de r iva t ives , so that a complete in te rpre ta t ion of the spectrum of NgPgCNN^).^ ^u^Cl j must await a bet ter understanding of the ent i re f i e l d . However, . (50,54) , « . •« . , . • • <-previous work does allow some tentat ive assignments. -The peaks between 3010 and 2790 cm" 1 i n both spectra may safely be assigned to asymmetric and symmetric C-H stre tching modes and overtones of CH^ bonding v ib ra t ions . The s i m i l a r i t y of the spectra i n t h i s region indicates that there i s l i t t l e difference between the environments of the CH^ groups i n the complex and the parent.-' This , i n turn, i s consistent with co-ordinat ion through the rings rather than the exocycl ic groups. The peak at 2150 cm - 1 present i n the parent, yet absent i n the complex, i s most probably an overtone of the peak at 1070 cm" 1 . The peaks between 1490 - 1425 cm - 1 for the complex and between 1480 - 1450 cm"1 for the parent can be assigned to asymmetric C-H bending v ib ra t i ons . These are e s sen t i a l l y s i m i l a r i n both compounds; again i nd i ca t i ve of bonding between the copper and the endocyclic nitrogen atoms. There i s no doubt that the main v (P-N-P) band i s centered as near 1270 cm 1 i n the parent, but extensive s p l i t t i n g and reduction of frequency occurs i n the complex ind ica t ing strong in te rac t ion with the copper atom. I t should be noticed that chela t ion of the copper w i l l have great ly increased the r i g i d i t y of the r i ng r e su l t ing i n a more complex v i b r a t i o n a l pat tern . - 81 -The doublet at 1070 and 1063 cm"1 i n the spectrum of the parent molecule corresponds to the asymmetric band at 1062 c m - 1 i n the spectrum of the complex, and i s to be a t t r ibuted to C-N s t re tch ing . Any further in te rpre ta t ion of the spectra past th i s point becomes f u t i l e because of the increasing number and complexity of the peaks. The main point of in teres t i n the spectra i s the s i m i l a r i t y between the spectrum of the complex and that of the parent. This i s consistent with a l l the known s t ruc tura l information on the two molecules. Complexing does, however, have a large effect on the v (P-N-P) bands, causing them to s p l i t . 3-S Therefore, the dominant features of the spectra are due to the various modes of the phosphon i t r i l i c system; only as copper chelat ion affects these does i t have any effect on the spectra. F i n a l l y , the fact that only the v (P-N-P) bands are affected i s then cha rac t e r i s t i c 3.S of endocyclic bonding to copper. Studies of N,P,(NMe_)._Cu 0 Cl_ i n Solu t ion : 6 6 2 12 2 3 The s t e r i c interference evident i n the structure of N..P,(NMe„),„-o o 2 12 Q ^ C l g could be re l ieved by the loss of the Cl(2) atom from the ca t ion . This would change the environment of Cu(2) from d is tor ted square pyramidal to square planar. The l a t t e r i s a wel l known type of environment for Cu(II) atoms, e.g. CuO, [ C u ( P y ) 4 ] 2 + ( 5 6 ) . I f the chlor ine were los t as the anion C l ~ , than the double charged ca t ion , N^P^OMe^)^2^n^+ w o u l d be l e f t i n so lu t ion . The other - 82 -species, CuCl 2 ~ could also break down, though no information exis t s concerning i t s behaviour i n so lu t i on . Nyho ' lm^^ used preparative evidence to suggest the existence of the sa l t s [Cu(AsMecj>2)4] [CuX2] . These, however, were non e lec t ro ly tes of undetermined molecular weight and probably do not contain the anion CuCl^" . No work has been done on a s a l t d e f i n i t e l y known to contain the anion CuCl,,"". With maximum ion i za t i on th i s anion would produce one Cu cat ion and two CI anions, so that the highest possible degree of d i s soc i a t i on which can be envisaged for NgPg(NMe 2)^ 2 ^ u2^*3 * s ' N 6 P 6 ( N M e 2 ) 1 2 C u C l + C u C l 2 " > HP(NMe2) 1 2 C u + + + Cu + + 3Cl" I f d i s soc i a t i on occurs i n so lu t i on , then an analysis for i on ic chlor ine w i l l show three chlor ines .per molecule. Quanti tat ive Ionic Chlorine Ana lys i s : A gravimetric chlor ine analysis with s i l v e r n i t r a t e was employed. The complex was dissolved i n pure, dry a c e t o n i t r i l e and a so lu t ion of s i l v e r n i t r a t e added. The r e su l t i ng dense white p rec ip i t a t e was f i l t e r e d of f i n a tared s intered glass f i l t e r , d r i ed , and weighed. 90.1 mg of the complex gave 0.0381 g of s i l v e r ch lor ide : three chlor ine ions would give 0.0376 g. Thus, under the conditions employed, both cat ion and anion d issocia te completely. Conductance Studies; The poss ible i on i za t i on path of N ^ P ^ C N M e ^ ^ C u ^ l ^ i n so lu t ion i s : - 83 -N 6 P 6 ( N M e 2 ) 1 2 C u 2 C l 3 > N^P^(NMe 2 )^CuCl* + CuCl 2 " N 6 P 6 ( N M e 2 ) 1 2 C u C l + ) N 6 P 6 ( N M e 2 ) 1 2 C u + + + C l CuCl 2 ~ ——>CuCl + C l ' CuCl > Cu + + C l -The ion ic chlor ine analysis showed that a l l of these steps do occur, but d id not show the i r r e l a t i v e importance. Electrochemical measurements were done on a Wayne-Kerr Universal Bridge B221A. The solvent used was a c e t o n i t r i l e dr ied by the method out l ined i n Chapter 2. The c e l l used (Figure 1) had a constant, 0.19249, determined using an aqueous so lu t ion with a known concentration of potassium ch lo r ide . The complex studied had been analyzed previously as N ^ P ^ ( N M e 2 ) 1 2 C u 2 C l 3 . The constant temperature bath was held at 25°C ± 0.02C°, A standard so lu t ion of N^P^(NMe 2 ) 1 2 Cu 2 Cl^ i n 9 ml of a c e t o n i t r i l e was transferred i n 1 ml quant i t ies into the c e l l containing 25 ml of a c e t o n i t r i l e . After each addi t ion conductance readings were taken. The molecular conductance was then calcula ted using the formula weight of N , P , ( N M e J . _ C u 0 C l _ . D O 2 12 2 o A l l simple e l e c t r o l y t e s ^ 9 ' ' i n a c e t o n i t r i l e and s i m i l a r solvents obey the equation: A A-A/C t5.1} where: _/\_ = molecular conductance l i m i t i n g molecular conductance - 84 -TABLE XI L imi t ing Molecular Conductances of Some Sal ts i n A c e t o n i t r i l e at 25°C Sal t A .(Ohms) Reference O LiC10 4 183.25 (70) NaClO. 192.40 " 4 KCIO. 208.92 " 4 RbC104 203.24 " CsCKK 207.63 » 4 BuNH3C104 194 (71) BuNH 3Pic. 167 " BuoNH„C10, 185 " Bu 2 NH 2 Pic . 158 " Bu3NHC104 177 . . " Bu 3NHPic. 149 " Et.NCIO, 189 " 4 4 E t 4 N S a l i c y l a t e 176 " PyHC104 202 » E t 4 NCl 176.6 " E t . C l O . 188.9 " 4 4 E t . N B C l . 180.9 " 4 4 . ( C 6 H 5 ) 4 P 2 N 3 H 4 S b C l 6 162.5 380 / C x 1 0 2 - A m o l e s ~ J 2 Figure 21. Conductance Plot of NLP^fNMe 1 Cu - 86 -(2, = molar i ty / \ = constant dependant on solvent and The l i m i t i n g conductance, i . e . the conductance at i n f i n i t e d i l l u t i o n , i s a function of the solvent and the number and types of ions i n so lu t ion . For a given solvent the A values for any 1:1 e l ec t ro ly te w i l l a l l be very s i m i l a r . In a c e t o n i t r i l e the A -values for 1:1 e lec t ro ly tes are between 209 - 155 Ohms"1 (Table X I ) . . Therefore, any A. Q h i g h e r than 209 Ohms"1 demonstrates the presence of an e l ec t ro ly te possessing more than two univalent ions. Figure 21 shows the p lo t of t /(T 1 for N,P, (NMe 0 ) .„Cu 0 Cl„. The l i m i t i n g conductance i s 377 ± 10 Ohms"1 which i s much too large for a 1:1 e l e c t r o l y t e . The resul ts are compatible with various d i s soc ia t ion patterns, t y p i c a l l y wi th : N 6 P 6 (NMe 2 ) 1 2 CuCl + CuCl 2 " ^ N 6 P 6 ( N M e 2 ) 1 2 C u + + + C l " + CuC^" Further studies are necessary to es tab l i sh th i s pattern d e f i n i t e l y . Magnetic Measurements of N^P^(NMe 2 ) ^ 2 Cu 2 Cl 3 : In the c r y s t a l s tructure of N-P,(NMe„) n„Cu-Cl_ the i n t r a -6 6 2 1 2 2 3 • molecular distance between the two copper atoms i s too large to allow any appreciable in t e rac t ion . The c r y s t a l should therefore have a magnetic moment corresponding to one copper (II) ion per molecule. This i s indeed the case and therefore i s further evidence that the molecule contains two oxidat ion states of the copper atom. The magnetic moment, u, of a metal ion may be approximated by - 87 -a calculated value, u obtained from. ' s y s = / 4 s ( s + i ) ( 5 _ 2 ) where s i s the t o t a l spin quantum number. In p rac t i ce , however, the observed magnetic moment, u, i s usual ly higher than For the copper (II) atom i s 1.73 B.M. (Bohr (72 73) Magnetons), but the usual experimental value i s 1.7 - 2.2 B.M. 5 . This i s due to incomplete o r b i t a l quenching which i n the extreme case would give a u value approximated by: y s + L = /4s (s + 1) + L(L + 1) (5-3) where L i s the t o t a l o r b i t a l quantum number. The experimental magnetic moment, y, i s calculated from the equation: y = 2.83/X'T (5-4) m J where X^ = molar paramagnetic s u s c e p t i b i l i t y of the metal ion T = temperature. The molar s u s c e p t i b i l i t y of the whole molecule, X , i s found by the (74) Faraday method . This value has to be corrected for the diamagnetism of every species i n the molecule, which i s done using the equation: X = X ' + E X d i a (5-5) m m . A v J A cLxci where X^ = diamagnetic molar s u s c e p t i b i l i t y of a l l species A. d i e t Values of X^ can be found i n the l i t e r a t u r e for various atoms and (72,73) ions . The magnetic s u s c e p t i b i l i t y measurements were done i n t h i s - 88 -f 75") department using the Faraday method . An Alpha Model 9500 water-cooled 6 inches electromagnet equipped with pole t i p s of Heyding design, (1 1/2 inches i n pole gap), was used. Samples, (approx. 5 mg), were suspended i n a quartz bucket from a Cahn Rg electrobalance. Measurements were done under a nitrogen atmosphere at 295°K. The magnetic s u s c e p t i b i l i t i e s were f i e l d independent. C a l i b r a t i o n was achieved using HgCofCNS)^ The molar s u s c e p t i b i l i t y of N,P,(MNe„)„Cu„Cl„, X , was r ' 6 6 2 12 2 3 m' 845.83 x 10~ 6 c.g.s. u n i t s . The diamagnetic molar s u s c e p t i b i l i t y of N,P, (NMe„)., „ has not been determined and therefore had to be calcu l a t e d . 6 6 2 12 d is. The hexameric p h o s p h o n i t r i l i c c h l o r i d e d e r i v a t i v e has a X m of -300.5 x 10~ 6 c.g.s. u n i t s ( 7 7 ^ } therefore, taking X^* a of a chl o r i n e atom as -20.1 x 10~ 6 c.g.s. u n i t s ^ 7 2' 7 3"* the x ^ a of the P,N, r i n g i s b m 6 6 " -59.3 x 10~ 6 c.g.s. u n i t s . . The diamagnetic c o r r e c t i o n necessary f o r N.P.(NMe 0).„Cu_Cl, i s -573.3 x 10~ 5 c.g.s. u n i t s . This i s cal c u l a t e d using the constants Table XII < 7 2 ' 7 3 \ i n Table XII Diamagnetic S u s c e p t i b i l i t i e s Species ^A^ X l ^ ) 6 , (c.g.s. u n i t s ) " 1 N 6P 6 r i n g -59.3 N i n amine - 5.6 C - 6.0 H - 2.9 CI" - 2.6 C u + + -11 Cu + -12 - 89 -Therefore, as: X' = X - E X ^ i a (5-6) m m . A A The molar suscep t ib i l i t y .due to the paramagnetism of the Cu (II) cat ion i s 1, 419.13 x 10~ 6 c . g . s . u n i t s . Therefore, the magnetic moment i s 1.83 B.M. A magnetic moment of 1.7 - 2.2 B.M. i s i nd ica t ive of one copper (II) atom per molecule. The value here of 1.83 B.M. i s then further proof that a molecule of N ^ P ^ ( N M e 2 ) ^ C i ^ C l ^ contains one copper (II) atom and one copper (I) atom. - 90 -REFERENCES 1. J . V . Bai ley and R .E . Parker, Chem. and Ind . , 1823 (1962). 2. B. Capon, K. H i l l s and R.A. Shaw, Proc. Chem. S o c , 390 (1962). 3. D.B. Sowerby, J . Chem. S o c , 1396 (1965). 4. C D . Schmulbach and V .R . M i l l e r , Inorg. Chem. 7, 2189 (1968). 5. B. Capon, K. H i l l s and R.A. Shaw, J . Chem. S o c , 4059 (1965). 6. T. Moel ler , S.G. Koka l i s , J . Inorg. Nucl . Chem., 25, 1397 (1963). 7. R . J . A . Otto and L . F . Audrie th , J . Amer. Chem. Soc. , 80, 5894 (1958). 8. G. T e s i , R . J . A . Otto, F .G. Sher i f and L . F . Audrieth, Ib id , 82^ , 528 (1960) 9. L . C Lund, N . L . Paddock, J . E . Proctor , H.T. Searle, J . Chem. S o c . , . ' 2542 (1960). 10. Handbook of Chemistry arid Physics, The Chemical Rubber Co, Cleveland, 49th Ed. (1968-69). 11. J . B . Faught, J . Moel le r , J . C Paul , Inorg. Chem., 9_, 1656 (1970). 12. A . C Harkness, H.M. Daggett, J r . , Canad. J . Chem., 4_3, 1215 (1965). 13. A. Turco and C. P e c i l e , Nature, 191, 66 (1961). 14. K. Nakamoto, Infrared Spectra of Inorganic and Cooridinat ion Compounds, Wiley-Interscience, New York, 2nd Ed. (1963). 15. C R . N o l l e r , Chemistry of Organic Compounds, W.B. Saunders Co . , . P h i l i d e l p h i a (1965). 16. R. Schmitt, K. Dehnecke, Z. Anorg. A l l g . Chem., 358, 38 (1968). 17. J . L . Burmeister and F . Basolo, Inorg. Chem., 3, 1587 (1964). 18. F. Basolo, W.H. Baddley and J . L . Burmeister, Inorg. Chem., 3, 1202 (1964) 19. J .O . Edwards and R.G. Pearson, J . Amer. Chem. S o c , 84, 16 (1962). F. Basolo, R.G. Pearson, Mechanisms of Inorganic Reactions, John Wiley $ Sons, Inc . , New York (1965). - 91 -20. J . Dostrovosky and M. Halmann, J . Chem. Soc, 508 (1953). 21. R .F . Hudson, Advances i n Inorganic Chemistry and Radiochemistry, V o l . _5, 347 (1963). 22. G.R. Branton, C . E . Br ion , D.C. Frost , K .A.R . M i t c h e l l and N . L . Paddock, J . Chem. Soc. (A), 151 (1970). 23. J . B . Faught, T. Moeller and I . C . Paul , Inorg. Chem., 9, 1656 (1970). 24. D.P. Craig and N . L . Paddock, Nature, 181, 1052 (1958). 25. M . J . S . Dewar, E .A . Lucken and M.A. Whitehead, J . Chem. S o c , 2423 (1960) 26. G . J . Bu l len , J . Chem. S o c , 3193 (1962). 27. A . J . Wagner and A. Vos, Acta C r y s t . , B24, 1423 (1968). 28. H. Bode, K. Butow, G. Lienau, Ber. Dtsch. Chem. Ges., 81_, 547 (1948). 29. A. Besson and M. Rosset, C R . Acad. S c i . Pa r i s , 143, 13 (1906). 30. D.B. Sowerby and L . F . Audrie th , Ber. Dtsch. Chem. Ges., 94_, 2670 (1961). 31. G. A l l e n , J . Dyson and N . L . Paddock, Chem. and Ind. , 1832 (1964). 32. N .V. Mani and A . J . Wagner, Chem. Comm.,658 (1968). 33. J . N . Rapko and G.R. F e i s t e l , Chem. Comm., 474 (1968). 34. H. Bode and H. Bach, Chem. Ber . , 75B, 215 (1942). 35. G.E. Coxon and D.B. Sowerby, J . Chem. Soc. (A), 3012 (1969). 36. M.F. Lappert and G. Sr ivastava, J . Chem. Soc. (A), 210 (1966). 37. S.K. Ray and R.A. Shaw, Chem. and Ind . , 1173 (1961). 38. D. Feakins, W.A. Last and R.A. Shaw, Chem. and Ind . , 510 (1962). 39. J . Dyson and N . L . Paddock, Chem. Comm., 191 (1966). 40. J . Trot ter and S.H. Whitlow, J . Chem. Soc. (A), 455 (1970). 41. J . Trot ter and S.H. Whitlow, J . Chem. Soc. (A), 460 (1970). 42. T. Chivers and N . L . Paddock, J . Chem. Soc. (A), 1687 (1969). 43. D. Feakins, W.A. Last , S.N. Nabi, R.A. Shaw and P. Watson, J . Chem. S o c (A), 196 (1969). - 92 -44. D. Feakins, W.A. Last , N . Neemuchwala and R.A. Shaw, J . Chem. S o c , 2804 (1965). 45. D. Feakins, W.A. Last and R.A. Shaw, J . Chem. S o c , 4464 (1964). 46. N . L . Paddock, Quart. Rev., 168 (1964). 47. H. Stokes, J . Amer. Chem. S o c , 19_, 782 '(1897). 48. S.K. Ray and R.A. Shaw, Chem. and Ind. , 53 (1959). 49. J . Dyson, Ph.D. D i se r t a t ion . 50. T. Moeller and S.G. Koka l i s , J . Inorg. Nucl . Chem., 25, 875 (1963). 51. D. Cook, Canad. J . Chem., 39_, 2009 (1961). 52. R .F . Evans and W. Kynaston, J . Chem. S o c , 1005 (1962). 53. R .C. Lord and R .E . M e r r i f i e l d , J . Chem. Phys., 21_, 1 6 6 (1953). 54. R. Stahlberg and E. Steger, Spectrochim. Acta , 23A, 2005 (1967). 55. A . J . Wagner, A. Vos, Acta C r y s t . , B24, 1423 (1968). 56. F . A . Cotton and G. Wilkinson, Advanced Inorganic Chemistry, Wiley-Interscience, New York (1962). 57. G. Brauer, Handbook of Preparative Inorganic Chemistry, Academic Press, New York, V o l . 2, 2nd Ed. (1965). 58. W.C. Marsh and J . Tro t te r , Pr ivate communication. 59. A . F . Wells , S t ruc tura l Inorganic Chemistry, Clarendon Press, Oxford, 3rd Ed. (1962). 60. E. Frassen, R. Bardi and S. Bezz i , Acta C r y s t . , 1_2 201 (1959). 61. A. Vaciaga and L . Zambonelli, J . Chem. S o c (A), 218 (1970). 62. A . F . Wells , S t ruc tura l Inorganic Chemistry, Clarendon Press, Oxford, 3rd Ed. (1962). 63. L . Paul ing, The Nature of the Chemical Bond, Cornel l Univers i ty Press, I t h i ca , 2nd Ed. (1942). , - 93 -64. A. Kabesh and R.S. Nyholm, J . Chem. S o c , 38 (1951). 65. R.S. Nyholm, J . Chem. S o c , 1257 (1952). 66. N. E l l i o t and L . Paul ing, J . Am. Chem. S o c , 60_, 1846 (1938). 67. P.T. Beurskens, H . J . A . Blaauw, J . A . Cras and J . J . Steggerda, Inorg. . . Chem., 7, 805 (1968). 68. L . Paul ing, The Nature of the Chemical Bond, The Cornel l Univers i ty Press, I t h i c a , 3rd Ed. (1960). 69. S. Glasstone, The Electrochemistry of Solut ions , D. Van Nostrand Co. Inc . , New York, 2nd Ed. (1937). 70. S. Mine and L . Werblan, Electrochim. Acta , 7_, 257 (1962). 71. J . F . Coetzee and G.P. Cunningham, J . Amer. Chem. S o c , 87_, 2529 (1965). 72. J . Lewis and R.G. Wi lk ins , Modern Coordination Chemistry, Interscience, London (1960). 73. P.W. Selwood, Magnetochemistry, Interscience, London (1956). 74. B.N. F iggis and J . Lewis, Progress i n Inorganic Chemistry, V o l . 6, 37 (1964) . 75. R.C. Thompson, personal communication. 76. B .N. F iggis and R.S. Nyholm, J . Chem. S o c , 4190 (1958). 77. D.P. Cra ig , M.L . Heffernan, R. Mason and N . L . Paddock, J . Chem. S o c , 1376 (1961). 78. G. A l l e n , D . J . O l d f i e l d , N . L . Paddock, F . R a l l o , J . Serreqi and S.M. Todd, Chem. and Ind . , 24_, 1032 (1965). 79. W.C. Marsh, Ph.D. Dise r t a t ion , Univers i ty of B r i t i s h Columbia, Department of Chemistry (1970).