Cyclometallation of trimesitylphosphine

119

Journal of Organometallic Chemistry, 340 (1988) 119-126 Elsevier Sequoia S.A., Lausanne - Printed in The Netherlands

Cyclometallation

of trimesitylphosphine

Elmer C. Alyea * and John Malito Guelph- Waterloo Centre for Graduate Work in Chemistry, Ontario, NIG 2 WI (Canada) (Received

Guelph Campus,

Unioersrty of Guelph, Guelph,

April 15th, 1987)

Abstract Synthetic and spectral (‘H, i3C and 3’P NMR) details are presented for the formation of cyclometallated palladium(I1) and platinum(I1) dimers by reaction of trimesitylphosphine with MCl, and MCl,Y, (M = Pd, Pt; Y = CH,CN, PhCN, iCOD> and for the corresponding halide metathesis products. These dimers are very stable with a strong M-C bond. There were no non-cyclometallated products isolated and cyclopalladation is definitely more facile than cycloplatination. The latter two observations are unusual with respect to the known reactivity of phosphines with palladium(I1) and platinum(I1) centres. Geometric isomerism is observed for the Pt species.

Introduction The relevance of internal metallation of various organic ligands to homogeneous catalysis has prompted many studies, including several involving tertiarlr phosphine ligands [1,2]. Shaw and co-workers [3] have clearly shown that bulky substituents on phosphorus increase the tendency for cyclometallation although rigorous reaction Possible mechanisms and the various factors conditions are usually required. influencing cyclometallation of tertiary phosphines have been discussec’ !3,4] but are incompletely understood. Nevertheless, it has been generally accepted that cyclometallation is favoured for platinum(I1) over palladium(I1). Our structural studies of steric effects in tertiary phosphine complexes led us to investigate the chemistry of trimesitylphosphine in several environments [5]. The question of whether or not this very bulky ligand [2,6] would undergo internal metallation with palladium(I1) and platinum(I1) centres increased in interest when we found no evidence for the internal metallation of tri-o-tolylphosphine in Pt(P-otolyl,),X, (X = Cl, I) [7]. A preliminary communication of the present study has appeared elsewhere [ 81. 0022-328X/88/$03.50

0 1988 Elsevier Sequoia

S.A

121 Table 1 Best yield and analytical M

X

data for the [M(P-C)X]

2 U complexes

Yield

Dec.

Analysis

(W)

(“CJb

C

H

X

61.38 (61.26) 56.66 (56.52) 52.41 (52.24) 53.23 (52.47) 48.89 (48.95) 46.05 (45.71)

6.02 (6.09) 5.12 (5.62) 5.04 (5.20) 5.14 (5.22) 5 .oo (4.87) 4.81 (4.55)

6.23 (6.70) 14.06 (13.93) 20.16 (20.44) 5.97 (5.74) 12.06 (12.06) 18.12 (17.89)

Pd

Cl

98.1

285

Pd

Br

44.0

283

Pd

I

83.3

284

Pt

Cl

49.2

216

Pt

Br

86.2

285

Pt

I

82.2

280

LIP-C = Pmes,C,H,(CH,),CH,. values shown.

’ Decomposition

(Found

prepared

occurs

(calcd.) (%))

over

a temperature

v(M-Cl) (cm-‘) 279(s), 249(m)

282(s), 248(m)

range

starting

at the

Any change in solvent, reflux time, starting material and/or the amount of triethylamine added led to lower yields (< 40%). There was no reaction for PtCl 2 (COD). [M(Pmes,C,H,(CH,),CH,)(X)l 2. For M= Pd, Pt and X = Br, I, the appropriate chloro-bridged complex was treated with either KX or NaX in acetone in l/2 molar ratios at ambient temperatures. The end of reaction was signalled by precipitation of product. Reaction times were longer for X = I than for X = Br and for M = Pt than for M = Pd. Best yield and analytical data for these complexes and their chloro analogues appear in Table 1. Results and discussion In contradiction to the usual trend observed [3,4], our results indicate that for the bulky trimesitylphosphine ligand (Pmes,), cyclopalladation is favoured over cycloplatination. The former proceeds smoothly at ambient temperatures with marked colour changes in relatively short periods of time (0.5 h), whereas the latter process requires both reflux conditions and the presence of a stoichiometric amount of base. We view this reversal as a fairly clear indication that the intimate mechanisms for cyclometallation (i.e., M-C bona formation) are different for palladium(I1) and platinum(I1) and separate studies are pursuing this further. Although either cyclopalladation or cycloplatination does occur faster or slower and with varying yields in a given solvent, there is no real solvent dependence on the final product distribution. In every case where reaction occurs, only one metal-phosphine product can be isolated. This is unlike the situation for the bulky trialkyl ligand, tri-t-butylphosphine [4]. This and the observed lack of reaction with MCl,(COD) (M = Pd, Pt) corroborating some earlier work by Clark et al. [lo] serve to demonstrate the steric effects of the large mesityl substituents. Clearly. Pmes, is too large to displace the coordinated diolefin ligand.

122

Table 2 : I1

and ~ P NMR data f o l [ M ( P -.(-}X~,,~ "'e'

M

X

~( ~1 p) {ppmt

i'5(I H) (ppm)

( :j(

( 4,,r( :~: P - : i | ) i~{z

:v~

Pt -- '~ : P ) t] z )

Pd

('I.

28.39.

Pd

Br

29.04

Pd Pt

1 (i

IX

Br

30.69 8,35{ 5l 18) 9.2()(519~ ) 1 l. 17(5059) l 1.97{5t35)

Pl

l

15.3444748}

15.64(4839)

.

~", 782 . 8%6 77 (; 48 . ...:, . : a ..... I ) , 4 6.79{2.6(}) 6.71 674 3.35 236 2.2(, 2.23 I {~4 ' ~'~.73(br) 6 , 6 1 6 , 3 8 3 ~6 2,3i ~" 19 .7. i4 I 5s /" 6.84 6.77(3.(?2"~ ( 4 6 ) ' , ~7 ?.?2 2 !6 2 24 { ~ 7 ; < 8 4 6.77(br) 6.45 :..i5 .'..3',~ 2.2S 2,2b i.e,.; " (',.77(2,97) 6.70 {~.4,4 5.~S 2, L2 2.25 2_23 t#,5 '7 6.81 t,.76{hr! 6.4) 3. t 4 2.34 2.23 2. [ t 62 .... 6.80(2 1(,) ,%77 6,t2 ~12 2.:L5 226 2.2] . 6 7 ' 6 83(3,(t0) 677 6 4 ] 3 i f , 235 2.25 2~'~ ; t 5

" Pmes~ (CDCI~} 61 :~p) 35.77 ppm; 8 ( ! H ) 6.7g (2.82 }tz~ 2,25. 2.~}4 rmm {tcl:~fi~c ilv.e~r:,~ion 6/9/t8}_, :' br = broad s i n g i e t . Rdafixe h~tegr::io~ : ; ' t ,', .*', :,. 3 3. ': Reid.+',: :ntcera~,~on.. J / 4 / I / 2 / 1 2 / 6 . " 3 / 3 ~' Majo~ i~
Thus, from the standpoint of the minimization of steric crowding, it is reasonable that upon reaction with palladium(lI) or platinum(lI). Pines:, should favour both dimerization and internal metatlation. Our analytical and more detailed spectrai data support our earlier structural assignment [81 abbreviated bek',~, l'hi~ structure as drawn represents the t r a n s or a,~m configuration wMch place, the wv~ P-donors in p o s i t i o n s t r a n s t o e a c h o t h e r r e l a t i v e t o t h e M M v e c t e r . Fb.i~; a r r a n g e m e n t s h o u l d e.-

P

/

~

C

\ \ /'

(

M C

~, C'L

I P

( P~C = P m e s 2 C s H 2 ( C H ~ ) ; ! C H ; , M=

Pd,

Pt)

aid in further minimizing the steric interactions bet~een substituents on P but our h i g h f i e l d N M R d a t a for t h e M Pt c a s e i n d i c a t e t h a t :~ s t a b l e c i . s ' - i s o m e ~ n : a 5~ also b e f o r m e d . W e h a d n o t o b s e r v e d this p r e v i o u s l y [8] o ' M n g to ~he i n s e n s i t i x i ~ y o f i n f r a r e d m e a s u r e m e n t s for t h e i d e n t i f i c a t i o n o f g e o m e t r i c { s e i n e r s a n d o, u r e a r l i e r u s e o f a w e a k e r m a g n e t i c tick1 for t h e N M R m e a s u r e m e n t s . T h e 1H N M R s p e c t r a for {he [ P d ( P - C T ) X } : (X = ( t . B r , 1} c o m p l e x e s { F a b l e 2) a r e r e l a t i v e l 3 u n c o m p l i c a t e d , s h o w i n g in e a c h c a s e lhe p r e s e n c c o f o : "~ o n e s p e c i e s t o w h i c h w e a s s i g n a H ' a t ~ v - c o n f i g u r a t i o n on. {he b a s i s o [ I h e a . , g u m e n t s g i v e n a b o , , e . F o r t h e X = Cf c a s e . t h e r e s o n a n c e at ?; 6.78 p p m c o r r e s p o n d s :o fl~c ."q u p" , a J e t t t

me/a-protons of the non-meiatlated mesit>[ rings. Theh' position :.rod ~oupling 1o phosphorus are very., similar to those observed for the cocresl,mdMo..=, proteins of free Pines> The meta-proton signals, of the metallated ring at ,'~ o.~"~..:. and 6.4g pp n are magnetically inequivatent as ,expected. show m; coupling to phosphon>, illld aFe s h i e l d e d . T h e p e a k at 8 3.27 p p m is a s s i g n e d to t h e 'moo pr~. {c,n- or~ tile m e t a l l a t i n g c a r b o n a t o m a n d is d i a g n o s t i c o f c y c i o m e m i i a { i o n . 'i'i~ese pr,,top,:, a r c d e s i f i e i d e d

123

(b)

Fig. 1. 3’P NMR

spectra

for [Pt(P-C)CI],

at (a) low and (b) high magnetic

fields.

with respect to the other methyl protons especially those on the same ring. In the aliphatic region, the two downfield peaks at S 2.35 and 2.28 ppm are assigned to the ortho- and para-methyl protons, respectively, of the non-metallated rings and are only slightly deshielded from free ligand positions. The final two resonances at 6 2.23 and 1.64 ppm are assigned to the para- and or&o-methyl protons respectively of the metallated ring and are significantly shielded. Assignments for the analogous bromo and iodo complexes are made in a similar way. The ‘H NMR spectra for [Pt(P-C)X12 (X = Cl, Br, I) are assigned as above but based on a mixture of coexisting geometric isomers. The major isomer is assigned the rruns-configuration on the basis of steric arguments and solution studies. The ratio of major/minor isomer was observed to decrease with increasing dielectric constant of solvent. Thus, the major isomer can be assigned a [runs-geometry since this geometry, with an expected negligible dipole moment, would be less favoured than the c&geometry by polar solvents [13]. A similar study using ‘iP NMR corroborated this conclusion. The major/minor isomeric ratio also decreases as Cl > Br > I (i.e. 3.7 > 3.2 > 2.6). This may be a steric effect but electronic effects cannot be completely ruled out. The low field (24.288 MHz) 31P NMR spectrum for the platinum(H) species, shown in Fig. 1, is a single line with a set of symmetric 195Pt satellites (S 5.13 ppm, 1J(195Pt-3’P) 5120 Hz). At higher field, two central lines in the ratio 3/l and two sets of ‘95Pt satellites are observed. These high field spectra change in position and/or coupling constant values only, with change in temperature. This suggests that the two sets of signals observed for the platinum(I1) case really do correspond to the two geometric isomers and are not due to a mixture of rotational isomers arising from restricted rotations within the P-C moiety. The coupling constants observed for the [Pt(P-C)X], (X = Cl, Br, I) complexes

Y

IO

II 12 13 14 Ii

125

phosphite, the positions for the carbons of the non-metallated rings do not change appreciably relative to the free phosphite positions whereas positions for carbons of the metallated ring can shift in either direction by up to 50 ppm. Our assignments appear in Table 3. The positions for carbon atoms of the non-metallated rings are either unshifted or slightly deshielded relative to free Pmes, in either the aromatic or aliphatic regions. Coupling to phosphorus except for the ipso carbon is decreased. The effects for carbons of the metallated ring are much more pronounced. Here, all the carbon atoms are magnetically inequivalent and different from their counterparts on the non-metallated rings. C(15) is severely deshielded as expected [14] but disappointingly shows no measurable coupling to phosphorus. In the case of the platinum(I1) complex, which is complicated by the presence of the c&isomer, a 1J(195Pt-i3C) value could not be measured even though these values are reportedly high (400-2000 Hz) [12]. The C(12) and C(11) positions are deshielded relative to either C(8) and C(9) or C(2) and C(3). The para positions, C(10) and C(14), are also deshielded relative to C(4) and C(6) but to a much lesser extent. C(8), C(9) and C(13) are each more shielded relative to C(2), C(3) and C(5) positions. These 13C NMR spectral assignments are wholly consistent with the ‘H and 31P NMR assignments. The observed shift of electron density from one side of the metallated ring to the other could be envisaged as occurring through the C(lS)-M-P u bonds without the need to invoke M-P a-bonding. All the species discussed here proved to be stable either as solids or in solution over very long periods of time. There were no reactions observed with HX (X = F, Cl) or X, (X = H, Cl, Br) suggesting that the M-C bond remains intact even though one might expect the metallating carbon to be susceptible to electrophilic attack. The M-C bond is broken however under reduction conditions. These reduction reactions and a wide range of bridge-cleavage reactions are still under investigation. Conclusions Cyclometallated dimers are formed as the only products in the reaction of the extremely bulky phosphine, trimesitylphosphine, with divalent Pd and Pt centres. The observation that cyclopalladation occurs much more readily than cycloplatination is unusual and suggests a different mechanism. Acknowledgements The award of an Imperial gratefully acknowledged.

Oil Limited

(Canada)

University

Research

Grant

is

References 1 J. Dehand and M. Pfeffer, Coord. Chem. Rev., 18 (1976) 327; M.1. Bruce, Angew. Chem. Int. Ed. Engl., 16 (1977) 73; I. Omae, Coord. Chem. Rev., 32 (1980) 235. 2 CA. Tolman, Chem. Rev., 77 (1977) 313. 3 A.J. Cheney, B.E. Mann, B.L. Shaw and R.M. Slade, J. Chem. Sot., A, (1971) 3833; A.J. Cheney and B.L. Shaw, J. Chem. Sot., Dalton Trans., (1972) 860; H.D. Empsall, P.H. Heys and B.L. Shaw, ibid., (1978) 257; B.L. Shaw, J. Organomet. Chem., 200 (1980) 307.

126 4 R.G. Goel and R.G, M o n t e m a y o r . I n o r g Chem., 16 (19771 2183: H.C. Clark. A.B. G o d , R.(L G o e l . S. Goel and W.O. Ogini. I n o r g C h i m , Aeta. 31 ( I 9 7 8 ) IA41: H ( ' ('lark. A.B. ( ; o e l ~md S. G,;el. J. O r g a n o m c t . Chem.. 166 (1979) (729. 5 E.C. Alyea. S.A. Dias, G. Fergus()n and M. Parvez. lnorg. Chin/. Acta. 37 (t9771 45: L.(. Alvea. S.*\, l)ias and S. Stevens. Inorg. C h i m . Acta I~ett. 44 (1~80) 207: E.( Ai'~ea. (}. [:erguson and *\. Somog?vari, lnorg. Chem., 2! t]~)82) 1369: E.('. Alyea, S.A. I)ia~,, G. Ferguson and P . Y Sie~., ('a,a~. C h e m . Acta, 5 7 ( 1 9 8 4 ) 457: [ i . ( . ,Myea. G. Ferguson. J M a l i t o a n d B. R~H~] h~o:~:: < ! e m . . 2 4 ~ ; ~ ) 8 5 ) 371~L 6 .I.F. Blount, C.A. M a r y a n o f f and K Mislov~. Tetrahedr,',n LctL. i 19"75} 9ii:. 7 E.('. Al>ea. S.A. Dias. (]. l-ergtisun and PJ. Roberts. J. ( h e m . S~;,:, i)a]!~,~} | ' I a t > . . {!!:}7q) g.~-S. 8 S.A. Dias and I-{.(7. Alyea. Transition Met. ( ' h e m . . 4 t 1<~79i 2U5. 9 F.R. Hartley, O r g a n o m e t . ('her,:. Rev. 4, 6 ( i 9 7 0 ) 119: ( i Beech. (i. M m n and .I &sllcroft. J ( . h e m . S(~c. A. (1970) 2903: J.R. Do',le, P 1£. Slade and 1 |.B. Jollassei?. [not g. Svnl.h.. 6 i 19(,I}} 216: [). I)rew and J.R. Doyle. ibid., 1"~ (1~721 I4: M.S. Khai-asch, 3. Am. C h u m S ~ . . ,~11 (11:1381 £81 10 G.K. Anderson. f1.C. (;lark and .!A. I)ax, ies, Inorg. C h e m . 20 I IOSl* 36(i7, l l ('. Eaborn. K..!. ()dell and A. Pidcock..l. ( ' h e m . Sot., Dalt(m lian>.,il07Y,) 1288. 12 P.S. Pregosm and R . W K u n z . ?',/MR!6 Springer-Verlag, Berlin. tteideJhcrg. New 5~,lk. i~-170. 13 G.K. Anderson, R.J. Cross. L Manc!il~>xic-Muir, K W . Muh-~md !" S,~!or~;un. ! ('~rg.H/,mleL ( ' h e m . 170(1979) 385: P.[L Garr()tL ( ' h e m Rex.. 81 11!481} 22t) 14 M.H. Chisholm and S. (i(~desk . Pr<~gr. l m ~ r g (7hem.. 2i) (i'97(,) 2'48. 15 F.W.B. Finstein, A B , Gi!chri,q amJ G.\V. R a y n e > ( a n h a n l . J.-M~I. ( ' h e m . Soc. 94 (1,)72) ~'~45 16 P..P. Stewa~t,.l.J. Benedict. I. I , h r a n d t and R.S. Ampulski, lnorg. ( ' h e m . . !4 ( I 8 7 % 293 ~,