319
Journal of Organometallic Chemistry, 153 (1978) 319-326 @ EIsevier Sequoia S-A., Lausanne L Printed in The Netherlands
A “F NMR STUDY OF TFIE TRANSMISSION OF ELECTRONIC EFFECTS IN TRIARYLPHOSPHINES
S-1. POMBRIK, and E-1. FEDIN Institute I I7S13, (Received
V.F.
IVANOV,
of Organo-Element (U.S.S.R.}
AS.
PEREGUDOV,
Compounds
D.N.
of the Academy
KRAVTSOV of Sciences
*, A.A.
FEDOROV,
of the U_~_S.R.,
nf0sc0w
January
20th,
1978)
summary The nature and efficiency of transmission of electronic effects of substituents through the bridging phosphorus atom in systems of the Ar,PC,H,F-4 type have been investigated by “F NMR. Correlation of the obtained data on the fluorine chemical shifts in triarylphosphines with those in the corresponding triarylmethanes and tetraarylsikmes indicated that the bridging phosphorus atom shows a higher transmitting ability than carbon and silicon. On the basis of the data on two-parameter correlation of the fluorine chemical shifts with the inductive (or) and resonance (a,) parameters of the substituents in the aromatic rings for systems with bridged phosphorus, carbon, silicon, tin, nitrogen, and bismuth atoms it has been established that phosphorus occupies an intermediate position between nitrogen and bismuth as regards the nature of transmission of electronic effects, i.e. interactions through the P-C,, bonds occur both by an inductive and, to some extent, a resonance mechanism.
The distinguishing feature of the electron structure of the elements of Group VB in the trivalent state is the presence of a lone electron pair in their valency shells which can, in principle, undergo resonance interactions with adjacent n-electron systems and impart to these atoms properties of increased electron transmittance. Judging by the available data, in the case of diary1 and triaryl derivatives of Group VB elements this property is indeed manifested by nitrogen, an element of the second Period [1,2], and is entirely absent in the case of antimony and bismuth, “heavy” elements of Periods IV and V [2,3], which behave similarly to the carbon atom [1,2].
Therefore it seemed of great interest to investigate the nature of transmission
of electronic effects of substituents through the bridging trivalent phosphorus atom in the triarylphosphine system, for which purpose it was proposed to apply “F NMR, as before [1,3].
:_
::. .*
.’
_TAiiE
I- :
..FLUORkE &E&iiCAL TO INTERNAL CgHgF
SHIFTSIN
DIARYL-~-FLU~RCPHENYL~HOSP~INCHCI;REL_~~TIVE ..lgF chemical 1.2 --0.5 --O.-r -1.0
:
shift (ppm)
Q
-1.3 -1.9 -1.7 -2.4 -3.3 4.4
For this purpose me synthesized a number of model compounds of the type Ar,PC,H,F-4 (where _Ar= 4(CH,),NC,H,, 4CH,C,H,, 3-CH3C,Hd, C,H,,
4-F&HA, 3-FCsHI, 4-CICAHA,3-Cl&Ha, 3.4-CI,CaH3, 3,4,5-Cl&H,) and determined the fluorine chemical shifts in these relative to internal fluorobenzene in 0.2 M solutions in CHC13 (Tabie I)_ Consideration of the da+a obtained shows that the fluorine chemical shifts in the dieryl-p-fluorophenylphosphines investigated substantially depend on the nature of the substituents in the aromatic rings. The range of variation of the chemical shifts in going from acceptor.substituents to donor ones is 5.6 ppm.[ To establish the nature of transmission of electronic effects through the bridging fluorine atom we correlated the fluorine chemical shifts with the polar constants of the substituents in the aromatic rings, which characterize the electronic effect of a.mono- and polysubstituted atomatic E&G-~ in the presence (o [4]) OTin the absence (a” 151) of resonance interactions of the substituent in the
aromatic ri_ng6ith the indicator centre (Table 2). F’rom the data of Table 2 it follows that in passing +om the constants IZooto
Eo the quality of the correlation dependences obtained improves slightly, but the difference detected is not so pronounced as in the case of diaryl-p-fluoro$henylmethSnes, for which the fluorine chemical shifts yield appreciably better correlation with the constants &I” than with the constants Zo, or.ln the case of diar$l-p-fluorophenylamines, for which the situation is reversed 121. Therefore k this case it cannot be asserted with confidence that the constants ITreflect best the electronic effects of the substituents on the fluorine-chemical shifts in the indicetor fluorophenyl grouping and, hence, the transmission of electronic effects through the phosphorus bonds - aromatic carbon occurs both by an inductive and a resonance mechanism-_ To verify this hypothesis we compared with the aid of the statistical criterion of the equality of two values [ 6],- the transmitting abilities of the bridging phosphorus, carbon, end silicon atoms in diary&p-fluorophenylphosphines end the previously-investigated diaryl-p-fluorophenylmethanes [2] and-in triaryl-jAluoro&enylsilanes- [-‘73(Table 2)_ -In comparing the values of p characterizing the ease of transmissionsof electronic effects in organophosphorus and organosilicon
6p(II)
6P
13
10 10 11 10
-2.68
-3.66 -3.36 -3807 0.90 (1.3G) b O.lG
0.18 0.12 0.22 0.00
0,3G
0.41 0.28 0.49 0,14 (0,20) b
AP
13
11 8 13 1B
Aplb
352
-0.78 4x94 -2.Gl 1,32
c
0.06
0008 0,OG 0.09 0.22
SC
-
0424
0,22 0.16 0,30 0,30
S
&I(I) 10 1.28 0.10 0,23 18 -G.4G 0.12 0,36 ,__,____________,_______ ,__.. __________________.__.__” . . ____..____,_.___,____ ..._..-. ._._......__._...-...._...._ -...-...-......._....._- .-----.-
Ca
Ca fiI(I) \
-
co
COG
&II,
;F
6F
SP
_~_~.~__~~~_~~~__.~-~__
0.9GG --......__ -__-_-
0,977
0.990 0,997 0,981 0.983
R
I
G G G 6 6
3.71 3.31 2.21 B.74 3.13
2,08 1,92 l-28 6.09 2,80
OJG 0.58 O.G8 I.,00 0.89
0.19 0.03 0.13 0.44 0.07
0.14 0.10 0.09 0.35 O.OG
-2.20 -l,T,l 0.23 G-G9 -0.95
0.14 0.08 0.08 0.29 0.06
0.904 0.997 0.993 0,994 0,999
_____ -.-
.._ .._
. -
-
-.
__
_
.
. _
-
-
.__..
-
’ n = the number of points; S = the stundard deviation of the Iwints from the I)lnnc: SpI = the standard dovintion of the cocfficicnt PI; Spit = tllc sf,andmd deviation of the coefficient /JR; R = the corrclntion coefficient.
(~.XCGI~~)~SIC~H~P-~ (~*XCGH~)~SI~C~H~F-~ (~.XCGI~~)~B~C~II~F.~ (4-XC,jII&NC6H$-4 (~-XCGII~)~PCC~I-I~~-~
---_.-____-~ ..--__..-. --. s..I-...e..m_.I.. S R t1 hzpR/pI -p1 -PI1 SPI SPIL c . __________,________._____________________._.___.__.___..,_________.._..._. _._____. _.._-.-_.--.. ..- ..----- .--.--.-- ------.--.-- .. ~-.----.----..___(4-XCbrr4)2CaC6H4F-4 6 2.64 l,GO O.G7 0.12 0.16 3.76 0,14 0.986
PARAMETERS OF THE CORRELATION EQUATION 6~19 = pIoI + pRflR + c fl _--___-.-_.___.-_---._..__-____.___...________~~__.,________~_______-__.__-..__.-
TABLE 3
’ )I = the numbor of points: s = tlrc standard deviation of tho points from the striabht Ilnc; S, = the stnndnrd deviation of the coefflcicnt c; Sp = the standnrd dcviatlon of tho cocffictont p: Ap = the absolute error of detcrminution of the cocfficlent /J for the 9G% confidence Icvcl; Ap9b = the rclativo error of dctcrmlnlng of the coefficient p for the 969b confidcncc Ievcl; R = the corrclrltion cocfficicnt, !J The vaIuc of p cnlculatcd from the additive schcmc,
Ar2CIIC&I4F.4 AK2PC&$-4 (I) AqCHC6H,jF*4 (II)
Ar3SlCgHqF-4 (II)
&PC,jI&$-4 hK3$IC,jHqP-4 /k2PCGk@-4 (I)
l\r2Pc#-‘i
PARAMETERS OF CORRELATION EQUhTlON y = ps + c a .__-_“_----.___---------~ System x n Y P
TABLE 2
322 :.-..-.
-
: .
_
-_
-.
_:.
systems vkt(iok,into~ consideration that the .bridgi$ silicon atom, -asdistinguis&d ffoin phosphorus, transmits the effect of three+ &d not tk, varied .. substituents.~T@efo~e, if the &&tivity pi%i&ple is valid and &o if the. transmitting abilities of. the phosphorus and silicon atoms had- been..equal, the value of-t& coefficient p for the-former should have been Z/3 of the silicon coefficient,namely -2.45~-~ O-16, The substantiallyhigbervalue of-the slope@ = -3.35 f O-28)-of the correlation line actually obtained, seems to indicate that the phosphorus atom posse&es a higher efficiency with respect to transmission of electrome effedtsof the substituents than the silicon atom. This is also supported by the larger-than-unity value of the coefficient p = 1.35_* 0.14 (due allowance being made for the correction of’1.5 introduced according to the additivity principle) for the correlation dependence, which establishes the relationship between the fluorine chemical shifts in triarylphosphines and tetraarylsilanes. In addition, on the basis of the tabulated quantitative data.(Table 2) it can be asserted with 95% confidence that the bridging phosphorus atom (p(-P-) = -3.35 t- 0.28) exceeds the methine group (P(-CH-) = -2.68 t 0.35) as well in ita transmitting ability. S-kce transmission of the electronic effects through the aliphatic carbon-aromatic carbon bonds occurs mainly by an inductive mechanism [Z,S], the better transmission properties of phosphorus as compared with the methine group may be attributed to the contribution of the resonance interactions to the overall mechanism of transmission of electronic effects through the bridging phosphorus atom. The possibility of partial transmission of electronic effects in the triarylmethane series by direct interaction between the electron systems of the aromatic rings [9] and the absence of such interaction in compounds of the triphenylphosphine type [IO] must result in a still greater difference of the true transmitting abilities of the aliphatic carbonaromatic carbon and phosphorusaromatk carbon bonds. Proceeding from the fact that phosphorus and silicon atoms are the closest neighbours in the Period III and possess identical vacant 3d-orbitals, which (as shown for tetraarylsilanes 171) do not take any substantial part in transmission of electronic effects, it may be assumed that in the case of the bridging phosphorus atom the lone pair takes some part in resonance interactions. At the same time, in discussing the possible causes for the increased transmitting~ability of phosphorus as compared with silicon and carbon one must take into account the fact that the polarizability of the electron shell of phosphorus is slightly higher, as witnessed by the atomic refraction data [ ll]. Thus it is impossible to give preference exclusively either to the contribution of the resonance interactions, or to the increased polarizability of the electron shell, which explain the behaviour of the bridging phosphorus atom, on the basis of the data of one-parameter correlation treatment. Therefore we applied the method of two-parameter correlation of fluorine chemical shifts with the inductive and resonance parameters of the substituents in the aromatic rings jl2] for the systems with bridging atoms of phosphorus, carbon,-silicon, tin [13], nitrogen, and bismuth (Table 3). ln applying this method we found an interesting regularity in the values of A, which characterize the relative contribution of the resonance component to the overall transmission
323
of electronic effects through the element--aromatic carbon bonds. Thus, for all the bridging atoms (C, Si, Sn, Bi) through which (as was established previously) the electronic effects are transmitted mainly by an inductive mechanism, the value of X remains practically unchanged and is equal to 0.56-0.58. The value of h found exceeds unity (h = 1.06) for the bridging nitrogen atom, in the case of which the resonance interactions due to conjugation of the lone pair with the aromatic rings are of great importance for the transmission of the electronic effects_ Finally, for phosphorus the value of X is approximately equal to the arithmetical mean of the two indicated values (h = 0.89). Form the above data it can be assumed that as to the character of transmission of electronic effects the bridging phosphorus atom in triarylphosphines occupies an intermediate position between nitrogen, on the one hand;.and bismuth, on the other, and interactions through the phosphorus-aromatic carbon bonds proceed both by an inductive and,-to some extent, by a resonance mechanism. It is probably due to this that phosphorus shows an increased transmissive ability as compared with the nontransition elements of Periods III and higher of Groups IVB and VB_ In conclusion it should be noted that in order to improve the statistical validity of our conclusions it would undoubtedly be of interest to investigate the transmission of the electronic effects in systems of the triarylboron type and in aryl derivatives of germanium, arsenic, and lead. Experimental 19F NMR spectra were taken on an R-20 Hitachi-Perkin-Elmer spectrometer (56.4 MHz) at 34°C. All measurements were made for dilute solutions at a concentration not exceeding 0.2 M_ The fluorine chemical shifts were determined with the use of the substitution method [ 141. The experimental error in determining the fluorine chemical shift did not exceed +O_l ppm. The solvent was chloroform, which was purified by the standard procedure and distilled over phosphorus pentoxide. Tris(p-fluorophenyl)phosphine was obtained by reacting p-fluorophenylmagnesiumbromide with phosphorus trichloride [ 151. Unsymmetric diaryl-p-fluorophenylphosphine and, in particular, the diphenyl- and bis(m-fluorophenyl)-pfluorophenylphosphine described in the literature [ 161, were obtained by the reaction of the corresponding arylmagnesiumbromides or aryllithium with p-fluorophenyldichlorophosphine, synthesized from phosphorus trichloride and fluorobenzene in the presence of AlCl, [ 171. The purity of the compounds investigated and the intermediate products was controlled with the aid of fluorine NMR, mass spectroscopy, and thin-layer chromatography on aluminium oxide. Below we give typical examples of syntheses of diaryl-p-fluorophenylphosphines. The analytical data and the physical characteristics of the compounds obtained are given in Table 4. Bis(p-dimethyIaminophenyI)-p-fluorophenylphosphine
A solution of 1.97 g (0.01 mol) p-fluorophenyldichlorophosphine in 20 ml abs. ether was added in an atmosphere of dry argon, on cooling, to a solution
4
WJ)
208-13/2 03-4 (methanol) 130-2 (propnnol)
77 46 31 76 50
(s.cligC6i~q)2FC6Ei4F.‘I ,
(4'clc6H4)~Pc~II4I~~4
(~.CICGII~)~PC~I.I~F.~
(3,4~~12CgPI3)~PCgN4F~4
(3,4,3.dI3C6Ii2)2PC~I~~I~.4
0.7
0.6
0,7
0,6
0.7
086
--._--_l
RIO
____.._.__,_.__,.._.___.,___.___.__” .....__.. ___..-_-.-._-.---._----F-e....--eN....-a Eluont, petroleum ether : otIiylacatate = GO : 1. ’ Analysis for F.
200--6/l
31
(~.CI~gCg114)21’C61~4F.4
136-138 (methanol) 67-D (mcthnnol) 220--30/l
31
--_--.-_-_-_._
ILp.(°C/mmlfg)
ot m,g. (%)
YIcld
[~.(CII~)~NC~RI)]~I’CGII~~-~
------_--.
Compound
*
,.
-
71.76 (12Sl) 77,G4 (77,OO) 77,26 (77,DO) 61 ,D’l (61,Dl) 02,35 (Gl,Dl) 61,7D (81,71) 43,86 (44638)
C
(5.83) : 5.78 ((i,BR) 3.68 (3.48). 3,D2 (3048) 2053 (2,4i) lo74 (1.65)
,:
‘,
-Be-6.W (6,60) 6.30
II
hnalysleFound(calc.)(Ib) -I_-_-.._--~
.-..I..-..--..~-----------------
460
418
340
349
308
308
366
Ill/l?
ANALYTICAL DATA AND PHYSICAL PROPERTIES OF DIARYL~~.I~LUOROPII~~YLPIiOSPHINES -__1--_-----I______-_.--I_._.--_.-_-___._-_.____II_.__-C___-~__II_-_
T@LE
I,
‘,
‘.
8.70.
.(7.41) 6.M (6.36),:
(8.84): G.64 ,(5;44) 7.40
‘.
~
::
‘G,. ..:
,,
,,
!”
I
,I
,;
,I’.
1;
I’
,. ,I,,,’
;,,.$,i
‘,i
‘_““I, :, ,.’
.I ,,‘,,I / ,w:
.)
,.
.,” ‘,,
I,
,.;
‘I
‘2
,,‘,I
” ‘,:‘)r:.:‘.,’ ..,:,
,.,,I;;,,,;:; :
./ “, (’ /. ,,. ‘,. ‘( ‘,” : ,, ,, :.;;
I
q.
;#I’,‘. ,; :j. ,‘I ,,,” ,;:, //’ ,‘, : ,’ .;’ : ,::“: m,,:‘,.
,I,, ,;I;” : ‘,’ ,, I;,,‘, ,‘,‘:::‘, ,;; 1’. ‘S’, ;.:
((’ : ,’ :
I ‘,” ‘. ,.;;‘,/ ‘!, 8, .,,, (/.’ .,,,‘,,’ ,:; _,. ‘,‘. ‘., > ,*,,,? )(. .:., ‘, . )I, : (8 ~.,I/‘,: .!, : ,,‘, ,’ ,i.: .(
:
“1 ; ,,,‘,
:
;,
,t, ;: ‘;:,
,:_:
.I
.,,’
,I
,‘.
,
,’
,.,‘,,,
(,‘>., .?{,,, ,, )I,,‘: ,I,:
,‘,’
;,
,,,‘;‘,,
,’
‘.I’
,.,.
.’ ‘,
,,
‘/: ”‘,
‘.
;
,*.I,
: ,,I’.
,‘6,8, b
II
“(:
‘,
.:’
(6.y) \., '9.94 ‘y;)
”
.,’
,,,,.
.,‘,S,’
,':(10;04)
,
,’
325
of pYdjmethyla+nophenyllithium in 50 ml abs. ether obtained from 0.45 g (0.064 g-at) Li-and_6.4-g (0.032 mol) p-bromodimethylaniline. After stirring for 40 min at room temperature the reactionmixture was decomposed with a saturatid aqueous solution of NH&l. The ether layer was separated and extracted with an excess of dilute HCI (1 : 1). The hydrochloric acid extract was neutrahzei with concentrated NH40H and repeatedly extracted with chloroform, and the organic layer was separated and dried with Na,SO,. On removal of the solvent a green oil was obtained which crystallized on treatment. with MeOH. Upon crystallization 1.1 g of a white powderlike substance with m-p. 136-38°C (MeOH) was isolated. Bis(p-folyl)-p-fluorophenyiphosphine
A solution of 1.97 g (0.01 mol) p-fluorophenyldichlorophosphine in 20 ml abs. ether was added in an argon atmosphere to a solution of p-tolylmagnesiumbromide in 50 ml abs. ether obtained from 5.13 g (0.03 mol) p-bromotoluene and 0.72 g (0.03 mol) magnesium and cooled to -5O“C. The temperature of the reaction mixture was gradually brought up to room temperature, and stirring was continued for 1 h. After boiling for 2 h the reaction mass was decomposed with a saturated aqueous solution of NH&l, the ether layer was separated and dried with Na2S04, the ether was removed, and the residue was distilled in vacua in an atmosphere of dry argon. 0.95 g light yellow oii with a b-p. of 217-27”C/ 2 mmHg was obtained, which crystallized after chromatographing in a thick layer of A1103 (eluent, petroleum ether : ethylacetate = 50.1, _R, = 0.6). Bis(ptolyl)-p-fluorophenylphosphine was isolated as white crystals with a m-p_ of 67-9”C (MeOH). Bis( 3,4,5-trichlorophenyl)-p-fluorophenylphosphine, bis(m-chlorophenyl)p-fluorophenylphosphine, and bis( 3,4-dichlorophenyl)-p-fluorophenylphosphine were obtained in a similar way. Bis(m-tolyl)-p-fiuorophenylphosphosphine
A solution of 4.9 g (0.031 mo:) m-bromoSoluene in 20 ml abs. ether was added in an atmosphere of dry argon to a suspension of 0.43 g (0.062 g-at) Li in 30 ml abs. ether. After stirring for 40 min the unreacted lithium was filtered and a solution of 1.97 g (0.01 mol) p-fiuorophenyldichlorophosphine in 20 ml abs. ether was added to the filtrate cooled to 0°C. After boiling for 2 h the reaction mixture was decomposed with a saturated aqueous solution of NH,Cl, and the ether layer was dri,edwith Na2S04. Upon removal of the solvent and distillation in vacua in an atmosphere of dry argon at 220-3O”C/X mmHg 2.4 g of bis(m-tolyl)-p-fluorophenylphosphine was obtained, which after chromatographing on A1203 (eluent, petroluem ether: ethylacetate = 50 : 1; R, = 0.7), appeared as a light yellow transparant oil. Bis(p-chlorophenyl)-p-fluorophenylphosphine
A solution of 5.75 g (0.03 mol) p-chlorobromobenzene in 20 ml abs. ether was added gradually in an argon atmosphere to a solution of butyllithium in 50 ml abs. ether obtained from 3.9 g (0.042 mol) butyl chloride and 0.7 g (0.1 g-at) Li and cooled to -5O-70°C. After stirring for 1 h at the same temperature a solution of 1.97 g (0.01 mol) p-fluorophenyldichlorophosphine in 20 ml abs.
I- :
Ack&wcedgement ,The authors e&e3 their profound thanks to Dr_ B.A. Kvasov for his help in correlation treatment of the results obt&ed. References 1 L.ti. &vinenko. R-S_ Popova and A.F. Popov. Uspekhi Kbimii. XLIV (9) (1975) 1594. 2 S-1. Pomba q_N_ Kravtsov. B-4. Kvasov and E_I. Fedin. J. Orzanometal. Chem.. 136 (1977) 185 3 I).N. Kravtsov. B.A. Kvasov, SJ. Pombrik and E-I_ Fedin;Izv. Akad_ Nauk SSSR. Ser. Khim.. (1974) 234. -4 H.H. Jaffe. Chem:Rev.. 53 (1953) 191. 5 R-P. WelIs. S.E..Ebrenson and R.F. Taft. Prog. Phys. Org. Chem.. 6 (1968) 147. 6 S-A. _Ai~azyao. Statisticheskoye issledovaniye zavisim ostei (Statistical Investigation of Dependences). M&Jhugiya. Moscow. 1968_ ? SJ. Pombrik. D_N_ e&ov. AS. Peregudov and E-1. Fedin. J. Organometi. Chem.. 131 (lSi7) 355. ar,d R;W_ Taft. J_.Amer_ Chem. Sot.. 94 (1972) 9113. 8 S-R. DayaL S. Ehzvn J. Organometal. Chem-. 104 (1976) 173. 9 G. Distefano. S. Pignatato. I_._Szepes and J. Borossau. L_ Szepes and J_ Borossay. J_ O~anometal. Chem.. 102 (1975) 313. 10 G. Distefano. S. ~taro. 11 C-K_ Ingold; Structure and Mechanism in Organic Chemistry. Corn~il Univ. Ren. Ithaca. New York, l953.p.&21. Phys. Org. Chem-. 10 <1973) 1. 12 S. Ehrenson. R.T.C_ Brownlee and R-W_ Taft. Roz 13 fi.N_ Kravtsov. B.A. Kvasov. T.S. K&.anova and E.I. Fedin. J. OrganometaI. Chem.. 61 (1973) 21914 -4-N. Nesmeyanov. D.N. Kravtsov. E.I. Fedin. B.A. Kvasov. V-M. Pachevskaya and L.S. Golovchenko. DokI. Akad. Nauk SSSR. 183 (1968) 1098. E. Clayer and G-P. Van Der Kellen. Bull. Sot. Chim. 15 R. De Ketelaere. E. MuyeIIe. W. Vandermen. 16 17
eelges. 78 il969) Y_ Schindlb&er. EI. SchimiIbauer.
219. C&m. Be-r.. 100 (1967) 3432. Monatsch. Cbem.. 96 (1965) 1936.