A Fourier Transform Carbon-13 NMR Study of the Electronic Effects

А Fourier Transform Carbon-13 NMR Study of the Electronic Effects of Phosphorus, Arsenic, and Antimony Ligands in Transition-Metal Carbonyl Complexes 13С NMR spectra have been oЬtained for 116 LNi(CO)3 complexes of phosphorus, arsenic, and antimony ligands as well as analogous LCr(CO)5, LMo(CO)s, and (776-C6H6)Cr(CO)2L derivatives. Analysis of the carbonyl chemical shift data yields electronic effect parameters which characterize the ratio of the и basicity to 1r acidity of the ligands. These parameters are in close agreement with the electronic parameters for phosphorus ligands derived Ьу Tolman and Kabachnik. The donor/acceptor ratio for phosphorus ligands increases in the order: РС\3 < P(OPh)3 < P(SR)3 < P(OR)3 < PPh3 < P(NRi)3 :S PR3. The donor/acceptor ratio for phosphite ligands increases (1) with the replacement of aryl Ьу alkyl substituents, (2) with the replacement of caged phosphites Ьу open-chained analogues, and (3) with increasing chain length for alkyl substituents. The donor/acceptor ratio for phosphines increases (1) with decreasing unsaturation in alkyl substituents, (2) with increasing alkyl chain length, and (3) with increasing suЬstitution at the а carЬon. Comparison of data for phosphine, arsine, and stiblne ligands suggests that the range of donor/acceptor ratios is largest for phosphines and smallest for stiЬines. Furthermore, variations in the donor /acceptor ratio due to changes in the R group in ER3 ligands are significantly larger than variations due to changes in the nature of the group SA element. Our data suggest that the donor/acceptor ratio for ER3 ligands increases in the order: AsR3 < PR3 < SbR3• The magnitude of the 2/"сэ�р through-metal coupling to the trans carЬonyl in LMo(CO)5 complexes appears to increase with increasing electron-withdrawing power for the suЬstituents on the phosphorus ligand, а Ьehavior more commonly associated with directly Ьound coupling constants such as the 1Jэ1р1аэw coupling in L W(CO)5 derivatives. lntroduction А linear correlation between the 13С NMR carbonyl chemical shift and the infrared stretching force constant for the carbonyl mode was apparent from the earliest studies of the 13С NMR spectra of transition-metal carbonyl complex­ es.1·2 An increase in the stretching force constant appears to Ье reflected Ьу а shielding of the carЬonyl resonance, а shift to higher field strengths. This correlation has now been ex­ tended to а wide variety of metal-carbonyl systems and ap­ pears valid for most complexes where calculations of the in­ frared stretching force constants are not contentious.з-7 The Darensbourgs8 have argued that an increase in the carЬonyl stretching force constant in metal carЬonyl complexes may result from either an increase in the и basicity or а de­ crease in the 1r acidity of the carЬonyl. Therefore they suggest that an increase in the stretching force constant must mirror an increase in the positive charge on the carbonyl carbon. If variations in the 13С NMR carbonyl chemical shift are dom­ inated Ьу contributions from either local diamagnetic shield­ ing9 or the dependence of the local paramagnetic shielding l Q on electron density, we should expect а deshielding of the carbonyl resonance with increasing force constant rather than the shielding that is observed. Our studies of (71 6-C6H 5X)Cr(C0) 3 complexes suggested а net donation of electron density from the arene ring to the Cr(C0)3 fragment 11 and а deshielding of the carbonyl reso­ nance with increasing donation to the Cr(C0)3 moiety.4 Since the extent of transition metal - carbonyl 11" back-donation is expected to increase with increasing electron density on the transition metal, we4-7 and others 12 have attributed this de­ shielding of the carbonyl resonance to а decrease in the magnitude of the separation between the electronic ground state and the lowest lying excited states with increased back-Ьonding. This hypothesis is supported Ьу the observation that the carbonyl resonance in (17 5-С 5 Н 5)М(СО) 3 complexes (М == Cr-, Mn°, Ре+) is deshielded with increasing negative charge, 4 in accord with the results of molecular orbltal caI­ culations which suggest an increase in the metal-+carbonyl 13 This 1r back-donation with increasing negative charge. hypothesis is further supported Ьу the excellent correlation, (r = 0.991) between the carbonyl chemical shift and the Cr-C bond lengths in Cr(C0)6 , LCr(C0)5, and L 2Cr(C0) 4 complexes.4 If the carbonyl resonance is deshielded with increasing electron density at the transition metal, then the carbonyl chemical shift should reflect the electronic effects of ligands introduced into these complexes. Since the electron density at фе metal in LNi(C0)3 complexes should increase as the и basicity of L increases and decrease as the 1r acidity in­ creases, the carbonyl chemical shift i_n LNi(C0)3 complexes should Ье а measure of the ratio of the u-donor to 1r-acceptor character of the ligand. Ву subtracting the chemical shift of the parent Ni(C0)4 complex from the chemical shift of LNi(C0)3 derivatives, we can define an electronic effect pa­ rameter in which L is compared to the СО ligand. The sign convention is such that an increase in the magnitude of this parameter implies an increase in the ratio of the и basicity to 1r acidity or а net increase in the electron-donor ability of the ligand. We wish to report herein the results of а study of 116 different LNi(C0)3 derivatives, thereby significantly ex­ panding our earlier study. 6 We have investigated the extent to which these electronic effect parameters can Ье transferred to other systems such as LCr(C0)5 , LMo(C0) 5 , and (r,6 • C 6 H 6)Cr(C0) 2 L complexes. Experimental Section P(OR)3 ligands (R = Pr, i-Pr, Bu) were synthesized Ьу the reaction (1) О. А. Gansow, В. У. Kimura, G. R, Dobson, and R. А. Brown, J. Ат. Chem. Soc., 93, 5922 (1971). (2) О. А. Gansow, D. А. Schexnayder, and В, У, Кimura, J. Ат. Chem. Soc,, 94, 3406 (1972). (3) G. М. Bodner, S. В. Kahl, К. Bork, В. N. Storhoff, J. Е. Wuller, and L. J. Todd, Inorg. Chem., 12, 1071 (1973). (4) G. М. Bodner aJid L. J. Todd, Inorg. Chem., 13, 1335 (1974). (5) G. М. Bodner, Inorg. Chem., 13, 2563 (1974), (6) G. М. Bodner, Inorg. Chem., 14, 1932 (1975). (7) G. М. Bodner, Inorg. Chem., 14, 2694 (1975). (8) D. J. Darensbourg and М. У. Darensbourg, Inorg. Chem., 9, 1691 (1970), (9) W. Е. Lamb, Phys. Rev., 60, 817 (1941). (10) М. Karplus and J. А. Pople, J. Chem. Phys., 38, 2803 (1963). (11) G. М. Bodner and L. J. Todd, Inorg, Chem., 13, 360 (1974). (12) Р. S. Braterman, D. W. Milne, Е. W. Randall, and Е. Rosenberg, J. Chem. Soc., Dalton Trans., 1027 (1973). (13) К. G. Caulton and R. F. Fenske, Inorg. Chem., 7; 1273 (1968). of phosphorus trichloride with the corresponding alcohol in the presence of а slight excess of dimethylaniline. 14 (RO)PC\ 2 and (RO) 2 PCI (R = Ме, Et, i-Pr, Bu, Ph) were synthesized Ьу using stoichiometric amounts of the alcohol in the presence of ап equimolar amount of dimethylaniline. 1 5 P(NR2 ) 3 (R = Ме, Et), PhP(NMe 2 ) 2 , and Ph2P(NMe2) were synthesized through the reaction of РС\ 3 , PhPCl 2 , or Ph2PCI with а twofold amount of dialkylamine. 1 6 PhP(OR) 2 and Ph 2P(OR) (R = Ме, Et, i-Pr, Bu, Ph) were syn thesized Ьу the alcoholysis of either PhP(NMe 2 ) 2 or Ph 2 P(NMe 2 ) . 17 ER 3 ligands (Е = Р, R = Pr, i-Pr, pentyl , hexyl, heptyl, octyl; Е = As, R = Et, Bu; Е = Sb, R = Et) were synthesized via the dropwise addition of ап ether solution of the trihalide to а slight excess of the Grignard reagent in ether solution . 1 8 PhSbC\ 2 апd Ph 2SbCI were synthesized Ьу the metathesis of stoichiometric mixtures of Ph 3 Sb and SbCl 3 • 19 PhER 2 and Ph 2ER (Е = Р, R = Ме, Et, Pr, Bu; Е = As, R = Ме, Et, Bu; Е = Sb, R = Ме, Et) were synthesized via reaction of the PhEC\2 or Ph2ECl intermediate with а Grignard reagent in ether solution. 20 These ligands were purified through vacuum distillation or fractional recrystallization, when possiЬ!e, and characterized Ьу infrared and proton magnetic resonance spectroscopy. Other ligands were purchased or obtained as gifts and were used without further purification . Ni(C0) 4 was purchased from Matheson Gas Products and used without further purification. Particular attention must Ье paid to the extreme toxicity of Ni(C0) 4 • All manipulations were carried out in а well-ventilated hood tested with а smoke bomb before use to ensure that по fumes leaked into other laboratories . For minimization of toxicity through а reduction in the vapor pressure, Ni(C0) 4 was dispensed d irectly into CDC\ 3 to form а 4 М solution which was handled Ьу syringe. All nickel carbonyl complexes were destroyed immediately after use Ьу reaction with bromine. LNi(C0) 3 complexes were prepared Ьу the addition of а 2-mL aliquot of 4 М Ni(C0) 4 in CDC\ 3 to an argon- or nitrogen-purged N M R tube containing 8 mmol of the ligand in I mL of CDC\ 3 . The samples were allowed to stand for 1 0- 1 5 min and then agitated until СО evolution ceased. Preliminary 1 3 С NMR spectra were obtained to determine the extent of complexation. If complexation appeared complete, ап 0.2-mL aliquot was removed, diluted with 1 5 mL of chloroform, and used for infrared characterization of the complex. Approximately 2 mmol of excess Ni(C0) 4 was then added to the sample, and the 1 3 С N M R spectrum w a s accumulated until the signal-to-noise ratio for the carЬonyl region was between 50 and 1 00 to 1 . The 1 3 С NMR chemical shifts are reported in ppm downfield from the internal �i(CO) 4 reference and are the result of at least two independent measurements with а precision of ±0.03 ppm. LCr(CO) 5 complexes were synthesized Ьу the reaction of ап excess of the hexacarbonyl with the ligand in refluxing diglyme solution 2 1 or Ьу photolysis of the hexacarbonyl with а high-pressure mercury lamp in the presence of а hexane solution of the ligand. 2 2 LMo(CO) 5 derivatives were synthesized Ьу the reaction of an excess of the hexacarЬonyl with the ligand in refluxing methylcyclohexane. 23 Excess metal hexacarЬonyl was removed Ьу vacuum suЬ!imation or fractional recrystallization, and the complex was characterized Ьу infrared spectroscopy. Preliminary 13С NMR spectra were recorded in CDC\ 3 solution, excess hexacarbonyl was dissolved in the solution, and the 13С N M R carbonyl chemical shifts were measured relative to the internal metal hexacarbonyl reference with а precision of ±0.06 ppm. (77 6- C6 H 6 )Cr(CO) 2 L complexes were synthesized Ьу photolysis of the parent (arene) tricarbonylchromium(0) complex with а high­ pressure mercury lamp in either benzene or hexane solution in the ( 1 4) G. М . Kosalopoff, J. Ат. Chem. Soc., 74, 4953 ( 1 9 5 2 ) . ( 1 5) G. Kamai and F. М. Kharrasova, J . Gen. Chem. USSR (Engl. Transl.), 27, 1 034 ( 1 957). ( 1 6) К. А. Petrov, Е. Е. Nifant'ev, Т. N . Lysenko, and V. Р. Evdakov, Zh . Obshch. Khim., 3 1 , 2377 ( 1 96 1 ) ; G . Ewart, D. S. Раупе, А. L. Porte, and А. Р. Lane, J. Chem. Soc. , 3984 ( 1 962). ( 1 7) L. Maier, Helv. Chim. Acta, 52, 8 5 8 ( 1 969) . ( 1 8) Р. Pfeiffer, !. Heller, and Н . Pietsch, Chem. Ber., 37, 4620 ( 1 904); Н. Hibbert, iЫd., 39, 1 60 ( 1 906) . ( 1 9) М and Т Chemicals, Netherlands Appl. , 6 505 2 1 6; Chem. Abstr., 64, 9766d ( 1 966) . (20) W, С. Davies and W. J. Jones, J. Chem. Soc., 33 ( 1 9 29), and Н . D. Kaesz and F. G. А. Stone, J. Org. Chem. , 24, 6 3 5 ( 1 959). ( 2 1 ) Т. А. Magee, С . N . Mathews, Т. S. Wang, and J . Н . Wotiz, J . А т . Chem. Soc., 83, 3 200 ( 1 96 1 ) . (22) R . J , Dennenberg and D . J , Darensbourg, lnorg. Chem. , 1 1 , 7 2 ( 1 972). (23) D. J. Darensbourg and Т. L. Brown, lnorg. Chem. , 7, 9 5 9 ( 1 968 ) . ]_ 6 .. Q. о '::. 4 z {,О '° 2 u z � O '----'---'--'---'--.I...---'--.,_____, 1 6.0 1 6.2 1 6 .4 L N i ( СО) 3 k co !6 6 1 6. 8 Figure 1 . А plot of the 13С N M R carbonyl chemical shifts in ppm downfield from Ni(CO) 4 for 40 LNi(CO) 3 complexes vs. the infrared stretching force constants in mdyn/ А for the carbonyl mode (r = 0 . 962) . Force constants were calculated from the data of С. А. Tolman, J . Ат. Chem . Soc., 92, 2953 ( 1 970) , оп the basis of the Cotton-Кraihanzel approximations, F. А. Cotton and С. S . Кraihanzel, J. Ат. Chem. Soc., 84, 44 3 2 ( 1 962), under an assumed local symmetry of С3 ,. presence of а slight excess of the respective ligand. 24 ( 776 -C6 H 6 ) ­ Cr(CO) 3 was synthesized Ьу t h e method o f Rausch e t al. 25 All complexes were characterized Ьу infrared spectroscopy . 1 3 С NMR spectra were obtained with а JEOL FX-60- FT spec­ trometer operating at 1 5 MHz, or а Varian XL- 1 00- FТ spectrometer operating at 25 MHz, using an internal deuterium lock. The technique of equilibrium saturation was used to overcome the long Т1 relaxation time for the carbonyl resonance. 26 We used pulse widths that cor­ respond to flip angles оп the order of 30-40 ° and а repetition rate оп the FX-60 of 2.2 seconds. For maximum precision in carbonyl chemical shift measurements, ап internal reference was used whose chemical shift was as close as possiЬ!e to that of the complex under investigation . These data may Ье converted to а Me4Si reference Ьу using the approximate con­ versions: = o1'i(CO), + 1 9 1 . 64 ppm Oмe,Si = ос,(СО) 6 + 2 1 1 . 2 8 ppm O м e,Si = Омо(СО), + 200.77 ppm oм e,si Oмe,Si = О ( с 6н 6)Сr(СО)з + 2 3 3 . 3 4 ppm Results The 1 3 С N M R spectra of LN i ( C0) 3 complexes in CDCl 3 solution exhiЬit а single sharp resonance at low field assigned to the geometrically equivalent carbonyls and а pattern of resonances at high field analogous to the free ligand. For the vast maj ority of the complexes studied, the magnitude of the 2J1зсз1р through-metal nuclear spin-spin coupling constant was less than 1 Hz and could not Ье resolved. There is no obvious relationship between the magnitude of this coupling constant, when observed , and the nature of the ligand. The carbonyl chemical shift (Ьс0 ) and coupling constant data for 1 1 6 LNi(C0) 3 complexes of phosphorus, arsenic, and antimony ligands are given in ТаЫе I . If w e derive substituent constants for the Х groups i n X 3 ENi(C0) 3 complexes Ь у dividing Ь со Ь у three, w e can use these substituent constants to predict the carbonyl chemical shift in X 2YENi (C0) 3 derivatives . The correlation between the predicted and experimental values of Ь со for 4 9 ( 24) W .. Strohmeier, Chem. Ber., 94, 2490 ( 1 96 1 ); G . Jaouen a n d R. Dabard, J. Organomet: Chem ., 72, 377 ( 1 974) . (25) М . D. Rausch, G. А. Moser, Е . J , Zaiko, and А. L. Lipman, J . Orga­ nomet. Chem. , 23, 1 8 5 ( 1 970). (26) Т. С. Farrar and Е. D. Becker, "Pulse and Fourier Transform NMR", Academic Press, New York, 1 97 1 , р 79; D. Shaw, "Fourier Transform N.M.R. Spectroscopy", Elsevier, Amsterdam, 1 976, р 1 7 9. (27 ) С. А. Tolman, J . А т . Chem. Soc., 92, 2953 ( 1 970) , ТаЬ\е I. С NMR Chemical Shifts for LNi(C0) 3 Complexes, in Ppm Downfield from Ni(C0) 4a ligand chem shift ligand по. по. 13 l 2 3. 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 Р(ОМе) 3 P(OCH,) 3CEt P(OEt), P(OCH,CH,Cl), P(OPr) 3 P(O-i-Pr) 3 P(O-allyl) 3 P(OBu), P(OPh) 3 P(OPh-p-Cl), P(O-lauryl) 3 P(S-lauryl) 3 P(NMe,) 3 P(NEt,) 3 РС1 3 РМе 3 PEt 3 P(CH,CH,CN) 3 PVy,C PPr 3 P-i-Pr 3 P(allyl), PBu 3 P-i-Bu 3 P-s-Bu 3 P-t-Bu 3 P(SiMe 3 ) 3 P(pentyl) 3 P(hexyl) 3 P(heptyl) 3 P(octyl), Р(с-Нх),С РР11 3 P(benzyl), P(нaphthyl) 3 P(Ph-p-CH,) 3 P(Ph-p-OCH 3 )3 P(Ph-p-F) 3 P(Ph-p-Cl), P(Ph-p-NMe,) 3 P(Ph-o-CH 3 ) 3 P(Ph-m-CH 3) з P(Ph-o-OCH,) 3 (Ме0 )РС1 2 (EtO)PCl, (i-Pr0)PC1 2 (BuO)PCl 2 (PhO)PC1 2 (Ме0)2 РС1 (Et0) 2PC1 (i-Pr0) 2 PC1 (Bu0) 2PC1 (Ph0) 2PC1 (PhO)P(OCH 2 ), EtP(OMe) 2 PhP(OMe)2 PhP(OEt), PhP(O-i-Pr), 3.18 2.60 3.61 2.76 3.94 3.90 3.43 3.95 (1.4)ь 1.69 1.32 3.98 2.50 (3.0) 5.68 5.68 -1.41 (12.0) 5.05 5.54 3.47 4.22 5.62 6.20 4.92 5.69 5.40 6.23 (1.2) 6.37 (1.2) 5.67 5.70 5.70 5.69 5.66 (1.0) 6.32 (1.5) 4.30 3.98 4.33 4.50 4.43 3.77 3.54 5.35 3.67 4.48 5.37 0.17 0.30 0.38 0.40 -0.33 1.69 2.00 2.15 2.20 0.74 2.08 4.36 3.48 4.04 4.23 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 PhP(OBu) 2 PhP(OPh), PhP(NMe2 ), PhPC1 2 PhPH 2 PhPMe, PhPEt, PhPVy, PhPPr 2 PhPBu 2 PhP(c-Hx), PhP(c-C,H 10 ) Ph,P(OMe) Ph,P(OEt) Ph 2P(O-i-Pr) Ph 2P(OBu) Ph,P(OPh) Ph 2 P(NMe,) Ph,PCl Ph,PH Ph,PMe Ph,PEt Ph,PVy Ph 2PPr Ph 2PBu Ph,P-t-Bu Ph 2 PC 6 F S Ph 2 P(c-Hx) (NCCH 2CH 2)PH2 (NCCH 2 CH,) 2 РН (с-Нх)РН, (с-Нх) 2 РН Ph 2P(CH 2 ) 2PPh 2 Ph 2 P(CH,),PPh, Ph,P(CH,) 4PPh, (Ph,P),CH, cis-Ph 2PCH=CHPPh, d Ph 2 P(CH,) 2 AsPh, е Ph,As(CH 2 )2 AsPh 2 (Ph,As) 2CH, AsMe 3 AsEt 3 AsBu 3 AsPh 3 PhAsMe, PhAsEt 2 PhAsBu 2 Ph2 AsMe Ph2AsEt Ph 2 AsBu SbMe, SbEt 3 SbBu 3 SbPh, PhSbMe 2 PhSbEt 2 Ph 2 SbMe Ph2 SbEt chem shift 4.20 2.54 4.69 0.69 3.54 4.76 5.36 4.25 5.29 5.29 5.80 4.94 3.96 4.27 4.26 4.30 3.46 4.34 2.49 3.93 4.53 4.78 4.26 4.76 4.74 4.80 3.13 5.07 3.13 (1.7) 3.33 (1.5) 4.07 5.35 4.39 4.29 4.60 3.52 7.79 4.15, 4.36 4.19 3.33 4.46 5.33 5.26 4.16 4.34 4.89 4.90 4.19 4.47 4.48 4.99 5.89 5.85 4.86 4.94 5.52 4.89 5.16 а Chemical shifts were measured in CDC1 3 solution and are reported in ppm downfield f,9m Ni(C0) 4 • А11 data are the result of duplicate measurements with а precision of ±0.03 ppm. Part of these data were reported Ьу G. М. Bodner, Inorg. Chem., 14, 1932 (1975). ь 2J1з0з1р through-metal nuclear spin-spin coupling constant in Hz, ±0.2 Hz. с Vy = vinyl, с-Нх = cyclohexyl. d This is the chemical s11ift for the LNl(C0) 2 complex; no evidence was found in the 13С NMR spectruш for the LNl(C0) 3 derivative. е Excess Ni(C0) 4 was added to en• sure complexation at both ends of thi� Ыdentate ligand. Thus, two nonequivalent LNl(C0) 3 environme11ts were observed. X 2YENi(C0)3 complexes is excellent (r, linear correlation coefficient, = 0.997). Tolman27 has derived analogous electronic effect parameters for phosphorus ligands in LNi(C0)3 complexes Ьу measuring the А1 infrared stretching frequency. It is significant that а linear correlation (r = 0.962) has been observed between 500 and both the А1 stretching frequencies and the stretching force constants calculated from the data of Tolman (Figure 1). It should Ье noted, however, that the ratio of the range of the carbonyl chemical shifts to the precision of measurement is 2½ times larger than the equivalent ratio for the infrared stretching frequencies. А linear correlation has been reported between 500 for LNi(C0)3 complexes and the cis carbonyl chemical shift (500ci8) in analogous LM(C0)5 complexes of Cr, Мо, and W.7 Data for an extended set of LCr(C0)5 and LMo(C0)5 complexes are given in ТаЬ!е II. The excellent correlations shown in Figures 2 and 3 between 500 for LNi(C0)3 complexes and 500cis f9r LCr(C0)5 (r = 0.983) and LMo(C0)5 (r = 0.993) complexes suggest that electronic effect parameters obtained data are reported in ТаЫе III. The correlation between these data and Ь со for analogous LNi(C0) 3 complexes is relatively good (r = 0 . 9 5 6) . 6 , z 4 о ( p p m) 2 '· ... ./ C,Q z Carbonyl Chemical Shifts for Metal Carbonyl Complexes of Phosphorus Ligands _J C,Q о 8 LCr O L Cr ( С О ) 5 - 8 C r ( С О ) 6 -2 о 2 4 6 ( ppm) 1со1, - 8 c r ( COl6 T R ANS - CARBONYL f1..?. - C A R B O N Y L в 10 12 Figure 2 . А plot o f the 1 3 С !'. M R carbonyl chemical sh i fts i n ppm downfield from Cr(CO) 6 for the cis and trans carbonyl resonances for 42 LCr(CO) 5 complexes vs. the 1 3 С NMR carbonyl chemical shifts in ppm downfield from Ni(CO) 4 for the analogous LNi(CO) 3 com­ plexes . Correlation coefficients are 0 . 9 8 3 and 0 . 8 56 for the cis and trans carbonyl resonances, respectively . 6 / 4 .1 , . . . ..· .. . z (ppm) C,Q 'о z _J о C,Q 3,_ M o( C OI , - 8 м о ( С О), O L M o (СО1 5 - 8 М о !С О 1 6 T R A N S - C A R BO N Y _ C I S - CARBONYL -z о 4 6 (ppm) 1О 12 Figure 3 . А plot o f the 1 3 С N M R carbonyl chemical shifts in ppm downfield from Мо(СO) 6 for the cis and trans carbonyl resona nces for 30 LMo(CO) 5 complexes vs. the 13С NMR carbonyl chemical shifts in ppm downfield from N i(CO) 4 for the analogous LNi(CO) 3 complexes. Correlation coefficients are 0 . 9 9 3 and 0 . 8 5 2 for the cis and trans carbonyl resonances, respectively. from the study of LNi(C0) 3 complexes may Ье applied to other metal-carbonyl systems. The analogous correlations for the trans carbonyl chemical shift (ocotrans ) are not as good. We have reported difficulty with the analysis of oco trans in LM(C0) 5 complexes of Cr, Мо and W, 7 Webb and Graham28 have reported similar proЬ!ems with ocotra ns in XRe(C0) 5 derivatives, and Woodard, Angelici, and Dombek 29 have re­ ported difficulties with the interpretation of both ocotrans and ocstrans in LW(COMCS) complexes. It is therefore interesting to note that predictions of о са for X 2YPCr(C0) 5 derivatives from data for X 3PCr(C0) 5 and У 3PCr(C0) 5 complexes lead to linear correlations between the calculated and experimental chemical shifts for both the cis (r = 0.980) and trans carbonyl (r = 0.964) resonances. А linear correlation has also been observed between оса far analagaus LCr(C0) 5 and LMo(C0) 5 camplexes far bath the cis (r = 0.993) and trans carbanyl (r = 0.9 32) resanances. Substituent canstants from LNi(C0) 3 complexes can also Ье applied to (17 6 -C 6 H 6 )Cr(C0) 2 L derivatives far which о са (28) М . J. Webb and W. А. G. Graham, J, Organomet. Сhет. , 93, 1 1 9 ( 1 97 5 ) . ( 2 9 ) S . S. Woodard, R. J. Angelici, a n d В. D . Dombek, lnorg . Chem . , 1 7 , 1 634 ( 1 97 8 ) . The replacement of one or more carbonyls in M ( CO) m complexes with trivalent phosphorus ligands leads ta а de­ shielding of the carbanyl resonance for the remaining СО ligands and therefore an increase in о со relative to the parent M(CO) m complex. Since there is an apparent deshielding of the carbonyl resanance with increasing metal - carbonyl 1r back-bonding, and back-bonding increases as electran density on the metal increases, the carbonyl chemical shift in these complexes should reflect the danar/ acceptar character of the phosphorus ligand. Data for LNi(C0) 3 complexes suggest a n increase in the donor/ acceptor ratio for phospharus ligands in the order: РС1 3 < P(0Ph) 3 < P(SR) 3 < P(OR) 3 < PPh 3 < P(NR 2 ) 3 $ PR 3 ; the net donation of electron density increases as the elec­ tron-withdrawing power of the substituents on phosphorus decreases. 30 Within the phosphite ligands, the danor/acceptor ratio increases in the order: P(O-p-C!Ph) 3 < P(0Ph) 3 < P( O C H 2) 3 C R < Р(ОСН 2 СН 2 С1) 3 < Р(ОМе) 3 < Р(ОС Н 2 С­ Н=СН 2 ) 3 < P(OEt) 3 < P(OR) 3 in accord with the relative basicities af phosphite ligands as determined from studies af 1 J1 нз1р coupling constants, the vвн stretching frequency in borane adducts, CNDO calculations, and studies of the v:-;a stretching frequency in Ni(NO)L/ complexes. 3 1 - 3 5 The donor/acceptor ratio far phosphine ligands increases in the order: P(CH 2 CH 2 CN) 3 < P(CH=CHJ 3 < PPh 3 < Р( СН 2СН=СН 2 ) 3 < РМе 3 < P-i-Bu 3 < PEt 3 < PPr 3 < PR 3 (R = butyl, pentyl, hexyl, heptyl, actyl) < P-i-Pr 3 $ P-s-Bu 3 < Р(с- Н х) 3 $ P-t-Bu 3 • This ratio appears to increase as the u nsaturation of the alkyl substituent decreases, as the alkyl chain length increases, and as substitution at the a-carbon atom increases. These data are in agreement with the pre­ dictions of studies of gas-phase basicity, NMR band-shape analyses of exchange in borane adducts, and molecular orbltal calculations 3 6-38 as well as with classical assumptions of the relative electron-donor character of alkyl substituents obtained from studies of carbonium ion stabllity. То investigate the transmission af aromatic substituent effects through metal carbonyl complexes, we have analyzed the о са data far (p-XPh) 3 PNi(C0) 3 complexes Ьу using the field (F) and resonance (R) substituent parameters of Swain and Lupton. 39 The оса data were fit Ьу using the equation fхнСО = fF + rR (30) We have used the field (F) and resonance (R) substituent parameters of Swain and Lupton 39 as а rough estimate of the e1ectron-withdrawing power of substituents on phosphorus. А mu1tiple linear regression ana]ysis of й со for eight X 3 PNi(CO) 3 complexes led to а linear corre­ lation (r = 0.965) between experimenta1 and ca]culated chemical shifts when the contribution to оса from field or inductive effects (69. 3%) was assumed to dominate resonance effects (30.7%) . (3 l ) D. W. White апd J. G. Verkade, Phosp horus, 3, 9 ( 1 9 7 3 ) . (3 2) Е. L . Lines, L. F. Centofani, and D. А. Hafler, Phosphorus, 5, 5 ( 1 974). ( 3 3 ) J. О. Albright, F. L. Tanzella, and J. G . Verkade, J. Coord. Chem. , 5, 225 ( 1 976). (3 4 ) D . S. Milbrath, J. Р. Springer, J. С. Clardy, and J. G. Verkade, J. Ат. Chem. Soc., 98, 5493 ( 1 976). ( 3 5 ) L. J. VanderGriend, J. G. Verkade, J. F. М. Pennings, and Н . М. Buck, J. Ат. Chem. Soc. , 99, 2 4 5 9 ( 1 97 7 ) . (3 6 ) R. Н. Stahley a n d J. L. Beauchamp, J . А т . Chem . Soc., 9 6 , 6 2 5 2 ( 1 974 ). (3 7 ) К. J. Alford, Е. О. Bishop, and J. D . Smith, J. Chem. Soc. , Dalton Trans. , 920 ( 1 976) . (3 8 ) М. Graffeui1, J. F . Labarre, М . F . Lappert, С. LeiЬcvici, and О. Stelzer, J. Chim. Phy s. Phy s.-Chim. Вiо/. , 72, 799 ( 1 97 5 ) . ТаЫе II. 13 С NMR Chemical Shifts in Ppm for the Cis and Trans Carboнyl Resonances in LCr(C0)5 and LMo(C0) 5 Complexesa L Cr(CO), по. 1 2 3 4 6 7 9 10 13 15 16 17 19 20 22 23 24 26 32 33 61 62 63 64 65 66 69 76 77 78 79 80 81 86 89 90 99 101 102 109 111 112 LMo(CO), ligand cis СО trans СО Р(0Ме) 3 P(OCH2 )3 CEt P(OEt) 3 P(OCH2 CH 2 Cl) 3 P( O-i-Pr ) 3 P(O-allyl), P(OPh )3 P(OPh-p-Cl)3 P(NMe 2 ) 3 РН3 PCl 3 РМе, PEt, PVy, PPr, P(ally1) 3 PBu 3 P-i-Bu, P-t-Bu 3 · Р(с-Нх) 3 PPh, PhP(NMe 2 )2 PhPCl 2 PhPH2 PhPMe, PhPEt 2 PhPVy, PhP(c-Hx) 2 Ph2P(NMe 2 ) Ph 2PCl Ph2 PH Ph 2PMe Ph 2PEt Ph 2 PVy Ph 2 P(c-Hx) (с-Нх)РН, (с-Нх)2 РН AsMe, AsBu, AsPh, PhAsC1 2 Ph 2 AsO SbMe, SbBu, SbPh 3 C 6 H 11NH2 C,H,N C(OMe )Ph C(NH 2 )Ph 3.89 (21)ь 2.88(22.0) 4.28 (21.3) 3.31 (20. 7) 4.67 (20.8) 3.90(20.7) 2.59(20) 2.29(20.1) 6.17(17.1) 3.96(13.5) 0.18(16.4) 5.50(14.0) 5.96 (14.5) 4.92(13.4) 6.03(14.0) 5.40(13.4) 6.08(14.5) 6.05(13.4) 6.66(11.0) 7.19(13.4) 5.25 (13.5) 5.29(15.2) 1.71(15.3) 4.29 (13.7) 5.26 (13.7) 5.59 (13.7) 5.00(13.4) 7.05(12.8) 5.29(13.9) 3.59(14.0) 4.81 (13.7) 5.20 (13.7) 5.33(13.7) 5.14 (12.8) 5.50(12.8) 4.80(13.4) 6.03(12.8) 5.44 6.06 5.37 1.26 3.23 6.34 6.89 5.74 2.67 2.77 4.83 5.80 4.88 7.90 (4) 7.43с 8.24 7.13(3.6) 9.22 7.76 6.93 6.01 9.67(7.4) 8.43(7) 5.66(4.5) а· d cis СО 3.60 (14) 2.62 (14.0) 3.92 (13.3) 3.04 (13.3) 4.25 (13.3) 3.66 (12.9) 2.32 (13) 5.52 (11.6) -0.01 (11) trans СО 7 .39 (40) d 7.85 (39.9) 6.52 (41.5) 8.38 (38.7) 7.31(40.4) 6.15 (46) 9.34 (29.9) 5.25(66) 9.89(8) 10.07 9.83(7.3) 9.24(6. 7) 9.83(8) 9.77(7.9) 10.36(4.9) 10.28 (6.1) 10.06 (7) 5.35 (9.4) 8.86(20.7) 4.92 (9.5) 5.55 (9) 8.40 (23.2) 8.88(21) 6.22 (7.4) 6.37 (8.6) 4.89(9) 9.74(22.0) 9.32 (21.5) 9.44 (23) 6.39 9.04(7.3) 10.01 (8.6) 10.01 (7.8) 9.87(7.3) 9.91 (6.1) 2.01 (10.8) 3.68 (9.1) 4.73 (9.1) 5.00 (9.1) 4.61 (9.1) 7.24 (46.5) 7.87(24.1) 9.28 (21.6) 9.06 (21.6) 8.95 (21.6) 7.40 9.71(6.8) 9.95(6.9) 9.97(6.8) 9.97(6.7) 10.25 (6.7) 9.16(6.7) 9.74(6.1) 10.89 10.89 10.76 7.96 9.34 11.94 12.03 11.11 8.61 8.79 12.87 12.27 12.31 3.37(10.0) 4.17(9.1) 4.72 (9.1) 4. 76(8.3) 8.34 (31.5) 8.67(23.2) 9.20(22.4) 9.13(22.0) 4.05(9.1) 5.32(8.3) 4.69 5.25 4.52 7.96 (23.2) 8.62 (21.6) 9.83 9.56 9.49 5.86 4.85 9.49 8.79 d d а Chemical shifts were measured in CDCI, solution and are reported in ppm downfield from the parent hexacarbonyl resonance . All data are the result ofduplicate measurementswith а precision of at least ±О.Об ppm. Part of these data were reported Ьу G. М. Bodner, Inorg. Chem., 14, 2694(1975). ь 21,зсзtр through-metal nuclear spin-spin coupling constant in Hz, ±0.6 Hz. с Тhе magnitude of the 'J"c'tp coupling constant to the trans carbonyl in LCr(CO), derivatives can become vanishingly small . d Data for the trans carbonyl chemical shift for samples in which proЫems with either solubility or sample size lead to а trans carbonyl resonance with insufficient signal-to-noise ratio are not included . where f х"СО represents the carbonyl chemical shift for (p­ XPh) 3PNi(C0) 3 relative to Ph3PNi(C0) 3 . А multiple linear regression analysis of the data yields а linear correlation (r = 0.997) between experimental and calculated chemical shifts for values of/and r corresponding to 47.6% and 52.4%, re­ . spectively. If 5 со data for X 3PNi(C0) 3 and Y 3 PNi(C0) 3 complexes can Ье used to predict 5 со for X 2YPNi(C0) 3 derivatives, we should Ье аЫе to reverse this process to predict the chemical shift for complexes such as H 3 PNi(C0) 3 which are difficult to synthesize. Analysis of 5 со data for six primary and sec­ ondary phosphine nickel tricarbonyl complexes yields а pre­ dicted 5со for H 3PNi(C0) 3 of 3.17 ::1: 0.07 ppm. Extrapolation of the linear correlation between 5c0cis for LCr(C0) 5 com­ plexes and 5 со for analogous LNi(C0) 3 derivatives yields а predicted value of 5 со for the H 3 PNi(C0) 3 complex of 3.18 ppm. Тhese data suggest that the donor/acceptor ratio for the Н 3 Р ligand is smaller than any PR 3 ligand studied, ap­ proximately equal to the donor/acceptor ratio of Р(0Ме) 3 • The 5со data for X 3PNi(C0) 3 complexes can Ье transposed into а set of substituent parameters for the Х groups on phosphorus, as described in ТаЫе IV. An excellent linear correlation (r = 0.972) exists between these substituent con­ stants and the и parameters for phosphorus ligands derived Ьу Kabachnik from studies of the рКа data for phosphoric acids of the type R 1 R 2P(O)OH. 40 (39) С, G. Swain and Е. С. Lupton, Jr., J. Ат. Chem. Soc., 90, 4328 (1968). (40) М.1. Kabachnik, Ви/1, Acad. Sci. USSR, Div. Chem. Sc/., 495 (1962). ТаЫе III. 1 3С NMR Oiemical Shifts for (ri 6-C 6 H 6 )Cr(CO) 2 L Complexes, in Pp m Do\vnfield fro m (ri 6-C 6 H 0 )Cr(CO) , a no. ligand chem sh ift 1 4.90 Р(OМе) 3 5 . 34 3 P(OEt) 3 4 P(OCH 2 CH 2 Cl) , 4 . 2 8 6.36 6 P(O-i- Pr) 3 9 P(OPh ) , 3.48 7.89 2 3 PBu , 1 0.40 3 2 Р(с-Нх ) 3 3 3 PPh , 7.88 ( 3 2.4 ) ь (3 1 . 6 ) (3 1 . 6 ) (32.0) (32. 7) (2 1 . 6 ) ( 1 9 . 9) (21.6) по. ligand chcm shift 63 69 78 80 86 89 90 Ph PH , Ph P(c-Hx) 2 Ph \ PH Ph , PEt Ph , P(c-Hx) (с-Нх )РН, (с-Нх ) , РН 5.65 (1 9.9) 9 . 76 ( 1 9 . 8 ) 6. 1 2 (20.6) 7 . 9 3 (20 . 6 ) 9.07 (19.8) 6.32 (19.8) 7.95 (19.9) а Chemical shifts \vere measured i n CD , Cl , solution and are re­ ported in ppm downfield from (ri 6 ·C " H 6 )Cr(CO) , . All data a re thc result of dup licate measuremcnts with а prccision o f at least ± О . Об ppm. ' 'С NM R carb onyl chemical shift data for (ri 6 ·C,,H , X)Cr (CO) , deriva tivcs h ave been repo rted p rcviously ; scc G . М . &o dner анd L. J. Todd, Jnorg. Chem. , 1 3, 360, 1 3 3 5 ( 1 9 7 4 ) . ь 'J " C " P througl1-metal nuclear spin-spin cou p ling const,шts in Hz , ± 0. 6 Hz . ТаЫе IV. Substituent Parameters for Х Substituents in Trivalent Phosplюrus Compounc:s х t- Bu с- Н х s- Bu i- Pr Bu NMe , SiMe , Pr Et i- Bu P!1-p - NMc 2 сн, allyl Ph-p-CH , Ph-p- OC H , naphthyl Ph Vy х а - 1 . Оба - 1 .05 - 1 .02 - 1.01 - 0. 84 - 0. 8 3 - 0. 8 3 - 0. 8 1 - 0. 7 9 - 0. 7 4 - 0. 7 2 -0.6 2 -0.58 - 0. 4 4 - 0. 4 2 - 0. 3 8 - 0. 3 7 - 0. 3 5 benzyl O-lauryl O Bu OPr 0-i- Pr Рh-р - Г OEt Ph-p- 0 CH,CH , CN O-allyl оме н осн,сн,о S-lauryl OPh OPh-p- Cl о а -0.27 -0.27 - 0. 26 -0.25 - 0 . 24 -0.20 -0. 14 - 0. 1 2 -0. 1 0 - 0. 0 8 0 . 00 0.00 0. 1 4 0.23 0.50 0 . 62 1.53 а The substituent p arameters were calculated Ь у dividin g 8 с о in the X , PNi (CO) , com p lex Ь у three , reversing the sign of the p aram­ eter so that substituents which are electron withdrawing relative to Н have p ositive values of а, and then scalin g t!1ese data so tl1at tl1e sub stituent parameter for Н is equal to zero. There has been some concern about the analysis of o co trans data for LM(CO) 5 complexes of Cr, Мо, and \V 7 • 29 and XRe(CO) 5 complexes. 28 Correlations between ocotrans and the trans carbonyl stretching force constant (k 1 ) have met with only marginal success, perhaps reflecting assumptions inherent in the calculation of k 1 • The data in ТаЫе II suggest several conclusions. First, predictions of 0cotrans on the basis of the additivity of substituent effects lead to а correlation between calculated and experimental chemical shifts (r = 0.964) es­ sentially as good as the correlation for /Jcoci, data (r = 0.980). Second, the mean value of /Jcotrans (9.47) is almost exactly twice the mean value of /Jcocis ( 4 . 7 7 ) for LCr(CO) 5 complexes. Third, variations in either 0co cis or oc 0 trans are significantly larger than variations in 0cotran, - ococis , Fourth, linear cor­ relations exist between the u substituent constants in ТаЫе IV and both /Jco1'""' (r = -0.930) and 0cocis (r = -0.9 1 8) for X 3 PCr(CO) 5 complexes. Finally, there is а reasonaЫe cor­ relation between ocotrans and ococis (r = 0.923) for phosphine complexes of LCr(CO) 5 , /Jcotrans is relatively small when /Jcocis is small; however, oc01'"ns does not increase as rapidly as Бсосi,. Comparison of the Donor/Acceptor Character of PR 3, AsR3, and SbR 3 Ligands Analysis of /Jco data for R 3 ENi (CO) 3 complexes (Е = Р, As, Sb; R = Ме, Et, Bu, Ph) yield several generalizations. First, the effect of variation of the R substituent on the do­ nor /acceptor ratio of ER 3 ligands decreases in the order: PR 3 > AsR 3 > SbR 3 . Second, the donor/acceptor ratio decreases in the order: Bu ""' Et > Ме > Ph, regardless of Е. Third, variations in the donor / acceptor ratio of ER 3 ligands due to changes in the R group are significantly larger than variations due to changes in the nature of the group 5А element. Finally, the donor / acceptor ra tio for these ligands а ppears to decrease in the order: SbR 3 > PR 3 > AsR 3 . Gas-phase proton affinity measurements suggest that the intrinsic basicity of РН 3 is greater than that of AsH 3 . 41 Enthalpy data for the gas-phase reaction of group 5А ligands with ВХ 3 Lewis acids suggest а decrease in the basicity of these ligands in the order: PR 3 > AsR 3 > SbR 3 , 42 as have studies of the photoelectron spectra of the free ligands. 43 These data may explain the relative donor /acceptor ratios for PR 3 and AsR 3 ligands, but they are inconsistent with the observation that the donor/acceptor ratio for SbR 3 is larger than for either PR 3 or AsR 3 ligands. It might Ье tempting to invoke either steric effects, or through-space shielding effects, to explain this apparent anomalous behavior. No steric effect on о со has been seen, however, even in the sterically crowded (17 6 -C 6 H 6 )Cr(CO) 2 P­ (c-Hx) 3 complex. Furthermore, neighboring diamagnetic screening would shield the carbonyls in stiЬine complexes relative to their arsine and phosphine analogues, just as о са for W(CO) 6 is shielded relative to Мо(СO) 6 or Cr(CO) 6 . 7 It is therefore of interest to note that X-ray crystallographic studies of Ph 3 ECr(CO) 5 complexes (Е = Р, As, Sb, Bi) imply an increase in both the bond order and the s character of thc group 5А element in the Cr-E bond in these complexcs in the order: Р < As < Sb < Bi. 4 The a nomalously large donor / acceptor ratio for stiЬine ligands may therefore Ье due to the strength of thc u interaction between the group 5А element and the transition metal. It should also Ье notcd that Brill has found an increase in the donor /acceptor ratio for E(OR) 3 ligands in (17 5 -C 5 H 5 ) M n(CO) 2 L complexes in thc order: Sb­ (OR) 3 > P(OR) 3 > As (OR) 3 . 45 2 Jвсэ•р Through-Metal Nuclear Spin-Spin Coupling Constants The magnitudc of the 21,зс"Р through-metal nuclear spin­ spin coupling to the trans СО in LM(CO) 5 complexes of Мо and W is significantly larger than that of thc analogous cou­ pling to the cis СО, whereas in LCr(CO) 5 complexes the converse is true, coupling to the cis СО cxceeds the coupling to the trans СО. 7 Similar behavior has becn reported for the 2Jэ1pJJ p coupling for cis- and trans-L 2 M (C0) 4 complexes of the group 6В metals . 46 Thc data in ТаЫе I reinforced our initial suggestion that 2J1эсз1 р coupling in LM(CO) 5 complexes of Cr, Мо, and W mirrors the behavior obscrved for the 2Jз1рз1р coupling in L 2 M (CO) 4 complcxes of these metals. The magnitude of the 2J"c"P coupling to the trans СО in LMo(CO) 5 derivatives increases in the order: PR 3 ::. PPh 3 < P(OR) 3 < P(OPh) 3 < РС\ 3 , the order of increasing elec­ tron-withdrawing power of the substituents on phosphorus. А similar observation has been reported for both the 1 Jз1р9s,91м0 and 1 Jз1 p 1sзw coupling in LM(CO) 5 complexes, where а linear correlation has been observed between the magnitude of these coupling constants and the electronegativity of the substituents on. phosphorus . 46-48 (4 1 ) R. У. Hodg es and J. L, Beauchamp , Inorg, Chem. , 1 4, 2887 ( 1 975). (42) D , С. Mente , J. L. Mills , and R. Е. Mitchell , Inorg. Chem., 14, 1 23 ( 1 975 ) , and D. С. Mente and J. L. Mills , Inorg. Chem., 14, 1 862 ( 1 975 ) . ( 4 3 ) Т. Р, DeЬies and J. W. Rabalais , Inorg, Chem. , 13, 308 ( 1 974 ) , (44 ) А. J. Carty , N, J. Ta ylor , А. W. Coleman , and М, F. La ppert, J. Chem, Soc., Chem. Соттип., 639 ( 1 979 ) , (45 ) Т. В. Brill, J. Organomet. Chem., 40, 373 ( 1 97 2 ) . (46) J. G. Verkade, Coord. Chem. Rev., 9, 1 ( 1 97 2 ). The approximate theory for directly bound coupling49 1 lxy сх: 1 д Eax 2 av2 1Фx(0)l 2 1Фv(0)l 2 suggests that directly bound coupling constants should increase with increasing s character (а 2 ) of the hybrid orbltals used to form the Х-У bond or with increasing electron density at the nucleus (l,,t,(O)1 2 ). 50 There are two factors which can lead to an increase in the magnitude of 1Jрм coupling with in­ creasing electron-withdrawing power for the substituents on phosphorus. An increase in the electron-withdrawing power of the substituents leads to а decrease in the electron density on the phosphorus and therefore an increase in the effective nuclear charge on the phosphorus atom. The increased nuclear charge draws electrons in the metal-phosphorus bond toward the nucleus of the phosphorus atom, thereby increasing lfp(O)/ 2, leading to an increase in the magnitude of the phosphorus­ metal coupling. Bent has argued that the introduction of an electron-withdrawing substituent concentrates s character in orЬitals directed toward the groups of lowest electronegativ­ ity.51 Thus, as the electron-withdrawing power of the sub­ stituents on phosphorus increases, ар2 for the metal-phosphorus bond increases, and the coupling constant increases as well. The linear correlations between the trans 2J,зс"Р coupling constants in LMo(C0)5 complexes and both the 1Jз1р9s,91м0 coupling in LMo(C0)5 complexes (r = 0.987) and the 1 Jз1p1sзw coupling in LW(C0)5 complexes (r = 0.907) would suggest that the metal-mediated trans 2Jвсз1р coupling behaves much as one might expect directly bound 1J1зсз1р coupling to behave. Acknowledgment. We gratefully acknowledge the financial support provided Ьу а Cottrell College Science Grant from the Research Corp. as well as grants from Sigma Xi and the Missouri Academic of Sciences which made this work possiЬ!e. We are indebted to D. Е. Troutner and М. S. Feather of the University of Missouri at ColumЬia for providing access to infrared and nuclear magnetic resonance spectrometers which greatly facilitated completion of this work. We also ac­ knowledge our debt to С. А. Tolman, Е. 1. du Pont de Nem­ ours and Со., D. Forster, Monsanto Со., and S. Searles, Jr., University of Missouri at ColumЬia, for providing generous samples of various ligands. Finally we acknowledge our debt to the School of Chemical Sciences at the University of Illinois for the use of their FT NMR spectrometers. Registry No. P(OMe)3Ni(CO)3 , 17099-58-0; P(OCH2) 3CEtNi­ (CO)3 , 72868-01-0; P(OEt) 3 Ni(CO)3 , 18558-97-9; P(OCH 2CH2Cl)3 Ni(CO)3 , 18559-00-7; P(OPr) 3 Ni(CO) 3 , 72868-02-1; P(O-i­ Pr) 3 Ni(CO) 3, 55333-41-0; P(O-allyl) 3 Ni(CO) 3 , 72868-03-2; P­ (OBu)3 Ni(CO)3 , 18558-98-0; P(OPh) 3Ni(CO) 3 , 18474-92-5; P­ (OPh-p-Cl)3Ni(CO)3 , 72868-04-3; P(O-lauryl)3Ni(CO)3, 72868-05-4; P(S-lauryl)3Ni(CO)3 , 72868-06-5; P(NMe2) 3 Ni(CO) 3 , 30512-95-9; P(NEt 2 ) 3 Ni(CO) 3 , 72868-07-6; PC1 3 Ni(CO) 3 , 18474-97-0; PMe 3Ni(CO)3 , 16406-99-8; PEt 3 Ni(CO)3 , 18555-45-8; P(CH2C­ H 2CN)зNi(CO)з, 72881-54-0; P(Vy)3 Ni(CO) 3, 72881-55- 1; PPr 3 Ni(CO)3 , 72881-56-2; P-i-Pr3 Ni(CO)3 , 18475-09-7; P(ally1)3Ni(CO)3 , 72881-57-3; PBu3Ni(CO)3 , 15698-54-1; P-i-Bu3 Ni(CO)3 , 72881-58-4; P-s-Bu 3Ni(CO) 3 , 18475-00-8; P-t-Bu3 Ni(CO) 3, 22955-41-5; P­ (SiMe 3)3 Ni(CO)3 , 16010-24-5; P(penty1)3Ni(CO)3, 72881-59-5; P(hexyl)3 Ni(CO) 3 , 72868-08-7; P(heptyl) 3 Ni(CO) 3 , 72868-09-8; P(octyl)3 Ni(CO)3 , 72868-10-1; P(c-Hx) 3 Ni(CO) 3, 18475-08-6; PPh 3Ni(CO) 3 , 14917-13-6; P(benzyl)3Ni(CO)3 , 18558-95-7; P­ (naphthyl)3 Ni(CO)3 , 72868-11-2; P(Ph-p-CH3) 3Ni(CO)3 , 52729-09-6; P(Ph-p-OCH3 ) 3Ni(CO)3 , 52729-08-5; P(Ph-p-F)3 Ni(CO)3 , 25493(47) В. О. Fischer, R. L. Keiter, L. Knauss, and J. G. Verkade, J. Organo­ met. Chem., 37, С7 (1972). (48) D. S. Milbrath, J. G. Verkade, and R. J. Clark, Inorg. Nuc/. Chem. Lett,, 12, 921 (1976). (49) С. Juan and Н. S. Gutowsky, J, Chem. Phys,, 37, 2198 (1962), (50) W, McFarlane, Q. Rev., Chem, Soc,, 23, 187 (1969), (51) Н, А. Bent, J, Chem, Educ., 37, 616 (1960), 52-1; P(Ph-p-Cl)3 Ni(CO) 3, 52151-06-1; P(Ph-p-NMe 2) 3Ni(CO)3, 72868- 1 2-3; P(Ph-o-CH 3 ) 3 Ni(CO) 3, 28796-13-6; P(Ph-m­ CH 3) 3Ni(CO) 3, 72868-13-4; P(Ph-o-OCH3) 3 Ni(CO) 3, 62945-86-2; (MeO)PC1 2Ni(CO)3 , 72868-14-5; (EtO)PC1 2Ni(CO)3 , 72868-15-6; (i-PrO)PC1 2Ni(CO)3 , 72868-16-7; (BuO)PCl 2Ni(CO)3 , 72868-17-8; (PhO)PC1 2Ni(CO) 3 , 72868-18-9; (MeO) 2PC1Ni(CO)3 , 72868-19-0; (Et0)2PC1Ni(CO) 3 , 72868-20-3; (i-PrO) 2PC1Ni(CO)3 , 72868-21-4; (BuO)2PC1Ni(CO)3, 72868-22-5; (PhO) 2PC1Ni(CO) 3, 72868-23-6; (PhO)P(OCH 2) 2Ni(CO) 3 , 72868-24-7; EtP(OMe) 2Ni(CO) 3 , 72868-25-8; PhP(OMe)2Ni(CO)3 , 55333-40-9; PhP(OEt)2Ni(CO)3 , 62945-82-8; PhP(O-i-Pr)2Ni(CO)3, 72868-26-9; PhP(OBu)2Ni(CO)3, 72868-27-0; PhP(OPh)2Ni(CO)3 , 55333-54-5; PhP(NMe 2)2Ni(CO)3, 72868-28-1; PhPCl 2Ni(CO)3, 18474-95-8; PhPH2Ni(CO) 3 , 7286829-2; PhPMe2Ni(CO) 3, 55333-45-4; PhPEt 2Ni(CO) 3 , 62945-89-5; PhPVy 2Ni(CO) 3 , 72868-30-5; PhPPr 2Ni(CO) 3 , 72868-31-6; PhPBu 2Ni(CO) 3, 18555-44-7; PhP(c-Hx)2 Ni(CO) 3, 72868-32-7; PhP(c-C5H 10)Ni(CO)3 , 72868-33-8; Ph 2P(OMe)Ni(CO)3, 55333-43-2; Ph2P(OEt)Ni(CO)3 , 72868-34-9; Ph2P(O-i-Pr)Ni(CO) 3, 72868-35-0; Ph 2P(OBu)Ni(CO)3 , 72868-36-1; Ph2P(OPh)Ni(CO) 3, 55333-39-6; Ph 2P(NMe 2)Ni(CO) 3 , 72868-37-2; Ph 2PCINi(CO) 3, 18559-01-8; Ph 2 PHNi(CO) 3 , 72868-38-3; Ph 2PMeNi(CO) 3 , 55333-44-3; Ph 2 PEtNi(CO) 3 , 69970-14-5; Ph 2PVyNi(CO) 3 , 72868-39-4; Ph 2PPrNi(CO) 3 , 72868-40-7; Ph2PBuNi(CO) 3 , 18705-97-0; Ph 2P­ t-BuNi(CO)3 , 72868-41-8; Ph 2PC6 F 5 Ni(CO)3 , 72868-42-9; Ph 2P(c­ Hx)Ni(CO)3, 72868-43-0; (NCCH2CH 2)PH2 Ni(CO)3, 72868-44-1; (NCCH 2CH 2 ) 2PHNi(CO)3, 72868-45-2; (c-Hx)PH2 Ni(CO)3 , 72868-46-3; (c-Hx)2PHNi(CO) 3 , 72868-47-4; Ph2P(CH 2) 2PPh 2Ni­ (CO)3 , 62945-84-0; Ph 2P(CH2) 3PPh2 Ni(CO)3 , 62945-85-1; Ph 2P­ (CH2)4PPh2Ni(CO)3, 72868-48-5; (Ph2P)2CH2Ni(CO)3 , 62945-83-9; cis-Ph 2 PCH=CHPPh 2 Ni(CO) 2 , Ph 2 P­ 21044-41-7; (CH 2 ) 2AsPh 2Ni(CO) 3 (Р bonded), 72868-49-6; Ph 2As­ (CH2) 2AsPh 2Ni(CO) 3 , 72868-50-9; (Ph 2As) 2CH2Ni(CO)3 , 728685 1 -0; AsMe 3 Ni(CO) 3, 16407-00-4; AsEt 3 Ni(CO) 3, 72868-52-1; AsBu3 Ni(CO) 3 , 72868-53-2; AsPh 3 Ni(CO) 3 , 37757-32-7; PhAsMe 2Ni(CO) 3 , 72868-54-3; PhAsEt2Ni(CO) 3 , 72868-55-4; PhAsBu 2Ni(CO) 3 , 72881-60-8; Ph 2AsMeNi(CO)3 , 72868-56-5; Ph 2AsEtNi(CO) 3 , 72868-57-6; Ph 2AsBuNi(CO) 3 , 72868-58-7; SbMe 3 Ni(CO) 3 , 72868-59-8; SbEt 3 Ni(CO) 3 , 16787-29-4; SbBu 3 Ni(CO) 3 , 72868-60-1; SbPh 3 Ni(CO) 3 , 52151-12-9; PhSbMe 2Ni(CO) 3 , 72868-61-2; PhSbEt 2Ni(CO) 3 , 72868-62-3; Ph 2SbMeNi(CO) 3 , 72868-63-4; Ph2SbEtNi(CO) 3 , 72868-64-5; P­ (OMe) 3Cr(CO)5, 18461-34-2; P(OCH 2)3CEtCr(CO) 5 , 42194-79-6; P(OEt)3Cr(CO)5, 18461-32-0; P(OCH2CH2Cl)3Cr(CO)5, 18461-35-3; P(O-i-Pr)3Cr(CO)5 , 65774-61-0; P(O-allyl)3Cr(CO) 5, 72868-65-6; P(OPh)3Cr(CO)5 , 18461-39-7; P(OPh-p-Cl) 3Cr(CO) 5, 72868-66-7; P(NMe 2) 3Cr(CO)5 , 15137-66-3; PH 3Cr(CO) 5 , 18116-53-5; PCl 3 Cr(CO) 5 , 18461-41-1; PMe 3Cr(CO) 5, 26555-09-9; PEt 3Cr(CO) 5, 21321-30-2; PVy3Cr(CO) 5, 65338-52-5; PPr3Cr(CO)5 , 72868-67-8; P(allyl) 3 Cr(CO) 5 , 72868-68-9; PBu 3 Cr(CO) 5 , 18497-59-1; P-i­ Bu 3Cr(CO) 5 , 72868-69-0; P-t-Bu 3Cr(CO) 5 , 23046-54-0; P(c­ Hx) 3Cr(CO) 5, 15603-93-7; PPh 3 Cr(CO) 5 , 14917-12-5; PhP­ (NMe 2) 2Cr(CO)5 , 72868-70-3; PhPCl2Cr(CO) 5 , 18461-40-0; PhPH 2Cr(CO) 5 , 72868-71-4; PhPMe 2Cr(CO) 5 , 72881-35-7; PhPEt2Cr(CO) 5 , 72868-72-5; PhPVy2Cr(CO)5 , 72868-73-6; PhP(c­ Hx)2Cr(CO) 5 , 72868-74-7; Ph 2 P(NMe 2)Cr(CO) 5, 72881-36-8; Ph 2 PC1Cr(CO) 5 , 18461-36-4; Ph 2 PHCr(CO) 5 , 18399-60-5; Ph 2PMeCr(CO)5 , 18497-53-5; Ph 2PEtCr(CO) 5 , 18497-54-6; Ph2PVyCr(CO) 5 , 72868-75-8; Ph 2P(c-Hx)Cr(CO)5 , 72868-76-9; (c-Hx)PH2Cr(CO) 5 , 72868-77-0; (c-Hx)2PHCr(CO)5 , 72868-78-1; AsMe 3 Cr(CO) 5 , 31238-08-1; AsBu 3 Cr(CO) 5 , 72868-79-2; AsPh 3 Cr(CO) 5 , 29742-98-1; PhAsC1 2Cr(CO) 5 , 58448-94-5; Ph 2 AsC1Cr(CO) 5 , 72881-61-9; SbMe 3 Cr(CO) 5 , 59136-89-9; SbBu 3Cr(CO)5, 72868-80-5; SbPh 3Cr(CO) 5, 29985-15-7; C6H 11 N­ H2Cr(CO)5 , 15134-57-3; C5H 5 NCr(CO)5, 14740-77-3; C(OMe)­ PhCr(CO)5, 27436-93-7; C(NH 2)PhCr(CO)5 , 32370-44-8; ClCr(C­ O) 5 , 14911-56-9; Р(ОМе) 3 Мо(СO) 5 , 15631-20-6; P­ (OCH 2)3CEtMo(CO)5 , 15665-65-3; P(OEt) 3Mo(CO) 5 , 15603-75-5; Р(ОСН 2 СН 2С1) 3Мо(СO) 5, 72868-83-8; P(O-i-Pr) 3 Mo(CO) 5 , 34369-19-2; P(O-allyl)3Mo(CO)5 , 72868-84-9; P(OPh)3Mo(CO) 5 , 15711-66-7; P(NMe2) 3 Mo(CO) 5 , 14971-43-8; РС1 3 Мо(СО) 5, 19212-18-1; PEt 3Mo(CO)5 , 19217-79-9; P(ally1)3 Mo(CO)5 , 7286885-0; PBu 3Mo(CO) 5 , 15680-62-3; P-t-Bu 3Mo(CO) 5 , 22955-40-4; Р(с-Нх) 3 Мо(СO) 5 , 15603-94-8; PPh 3 Mo(CO) 5, 14971-42-7; PhPC1 2Mo(CO) 5, 21485-20-1; PhPH 2Mo(CO) 5 , 72868-86-1; PhPMe 2Mo(CO) 5 , 24554-78-7; PhPEt 2Mo(CO) 5 , 72868-87-2; PhPVy 2Mo(CO) 5 , 72868-88-3; Ph 2PC1Mo(CO) 5, 23581-74-0; Ph 2 PHMo(CO) 5 , 1 8 3 9 9 - 6 1 - 6 ; Ph 2 PMeMo(CO) 5 , 1 8 497-6 2 - 6 Ph 2 PEtMo(CO) 5 , 1 8 497-63-7; (с- Н х ) Р Н 2 Мо(СO) 5 , 72 868- 89- 4 (с-Нх) 2РНМо(СО) 5 , 7 2 8 6 8 -90- 7 ; AsMe 3 Mo(CO) 5 , 7 2 8 6 8 - 9 1 - 8 As B u 3 M o ( C O ) 5 , 7 2 8 6 8 - 9 2 - 9 ; A s P h /-..1 o ( C O ) 5 , 1 9 2 1 2- 2 2 - 7 Sb8u 3 Mo(CO) 5 , 72868-93-0; S b Ph 3 Mo(CO) 5 , 1 92 1 2- 2 1 -6; (77 6 C 6 H 6 ) Cr( CO)2P(O Me) 3 , 3 1 8 5 2-04- 7; (77 6 - C 6 H 6 )Cr(CO) 2 P( OEt) 3 , 5 3 6 1 1 -65-7; (77 6 -C 6 H 6 ) Cr(CO) 2 P(OCH 2 CH 2 Cl) 3 , 7 2 86 8 -94- 1 ; ( 77 6 C 6 H 6 ) Cr(CO) 2 P(O-i-Pr) 3 , 7 2 8 6 8 - 9 5 - 2 ; ( 77 6-C 6 H 6 ) Cr(CO) 2 P(OPh) 3 , 7 2 8 6 8-96- 3 ; ( 77 6 -C 6 H 6 ) Cr(CO) 2 PBu 3 , 1 2 27 8-02- 3 ; (17 6 - C 6 H 6 )Cr­ (CO) 2 P(c-Hx) 3 , 1 2278-74-9; (17 6 -C 6 H 6 )Cr(CO) 2 PPh 3 , l 227 8-67-0; ( 776-C6H 6)Cr( CO) 2PhPH 2, 72868-97-4; ( 776-C6 H6) Cr( CO) 2PhP(с-Нх) 2, 7 2 8 6 8 - 9 8 - 5 ; (17 6 - C 6 H 6 )Cr(CO) 2 Ph 2 P H , 72868 -99-6; (17 6 -C 6 H 6 )Cr­ (CO) 2 Ph 2 PEt, 72869-00-2; (17 6 -C 6 H 6 ) Cr(CO) 2 Ph 2 P(c-Hx), 728 690 1 - 3 ; (11 6 -C 6 H 6 )Cr(CO) z(c- Hx)PH 2 , 728 69-02-4; (17 6 - C 6 H 6 )Cr­ (CO) z(c-Hx) 2 PH, 7 2 8 69-0 3 - 5 ; Ph 2 P(CH 2 ) 2 AsPh 2 Ni(CO) 3 (As bonded ) , 72869-04-6. Infrared Metal-Ligand Vibrations of Hexaaquametal (III) Ions in Alums The infrared spectra of mulls of the cesium $-alums Сsм ш (SO4 ) 2· 1 2Н 2 O (М ш = Al, Ti, V, Cr, Fe) and of the ammonium and potassium a-alums of Аl ш and Сr ш are reported between 300 and 1 200 cm- 1 at amЬient temperature, together with that of the a-alum CsCo(SO 4 ) 2• 1 2 Н 2 O bet,veen 450 and 1 200 cm- 1 at about 80 К. Only two bands are observed to shift appreciaЬly with change in the trivalent metal ion ; these are assigned to the antisymmetric metal-ligand stretch v 3 and the antisymmetric metal-ligand bend v4 of the М(Н 2 O) 6 з+ complex. These assignments are confirmed Ьу the observation of metal isotope shifts of 5 and l cm- 1 , respectively, between cesium chromium alums enriched in 5°Cr and 53 Cr. Introduction There have been remarkaЬly few vibrational studies of the ш М(Н 2 0) 6 з+ species 1 -7 and assignments of the 2 stretching frequencies remain elusive. Most reports are on individual compounds, and, although some assignments have been aided Ьу deuteration , few comparative studies have been reported. The alums, of composition М 1 (Н 2 0) 6 М ш ( Н 2 0k (ХО4 ) 2 , provide а series of compounds in which the trivalent metal ion can Ье systematically varied, usually in isomorphous structures . This advantage is somew hat offset, however, Ьу the presence of the М ( Н 2 0) 6 + species , which contributes additional vibrations due to the coordination sphere of the monovalent cation. This study reports the infrared mull spectra of the cesium sulfate alums of Al, Ti, V, Cr, Fe, and Со. Comparison of these spectra allows а clear assignment of the two infrared active ш 2 vibrations. The cesium alums ,vere chosen because it is possiЫe to obtain staЫe specimens containing the Со ( Н 20) 6 з + species 8 and because crystal structure determi­ nations of all of the compounds are availaЫe. 9- 1 1 This series м -он м -он ( 1 ) Campbell, J. А.; Ryan, D. Р.; Simpson, L. М . Spectrochim. Acta, Part А 1970, 26а, 235 1 . (2) Petrov, К . !.; Bol'shakova, N . К.; Kravchenko, V . У.; lskhakova, L . D. Russ. J. Inorg. Chem . (Engl. Transl.) 1970, 1 5, 1 529. (3) Strupler, N . ; Guillermet, J . ; Hoffelt, J.; Froment, F. Ви/1. Soc. Chim . Fr. 1974, 1 8 30. (4) Eysel, Н. Н.; Eckert, J. Z. Anorg. Allg. Chem. 1 976, 424, 6 8 . (5) Eckert, J.; Eysel, Н. Н.; Kampfmeyer, G. L. Z . Anorg. Allg. Chem. 1 976, 424, 8 1 . (6) Adams, D . М.; Hills, D . J . J. Chem . Soc., Dalton Trans. 1978, 782. Note that the labels р, and P w should Ье interchanged in the tаЫе оп р 787. (7) James, D. W.; Whitnall, J . М . J. Raman Spectrosc. 1 978, 7, 225. (8) Johnson, D. А.; Sharpe, А. G . J. Chem. Soc. А 1966, 7 9 8 . ( 9 ) Cromer, D. Т.; Кау, М . !.; Larsen, А. С . Acta Crystallogr. 1 966, 2 1 , 383. ( 1 0) Sygusch, J. Acta Crystallogr., Sect. В 1 974, 30, 662. ( 1 1 ) Best, S. Р.; Beattie, J. К.; Skelton, В. W.; White, А. Н., to Ье submitted for puЬlication. does not contain the Sc alum, and we were unaЫe to obtain reliaЫe spectra of the Mn alum, which decomposed under the conditions used in recording the other spectra. Experimental Section The cesium alums were prepared from reagent ,grade chemicals either Ьу using methods described in the literature (AJ, 9 V, 8 Cr, 8 Fe8 ) or Ьу the procedures described below. They were recrystallized from 1 М H 2S04 unless stated otherwise and characterized Ьу their infrared spectra, Ьу single-crystal X-ray structure determinations, and in some cases Ьу their melting points. The ammonium and potassium alums were recrystallized commerical samples. The isotope-enriched chromium alums were prepared Ьу oxidizing enriched chromium(III) oxide obtained from Oak Ridge National Laboratory with perchloric acid to chromium(VI) oxide and then proceeding as above. The cesium titanium alum was prepared under an inert atmosphere. Cesium sulfate ( 1 g) was dissolved in 30 mL of deoxygenated 1 М sulfuric acid. Addition of 7 mL of а deoxygenated 1 5% w /v solution of titanium trichloride gave а violet solution. Storage under refrig­ eration gave violet crystals which were collected, washed with 1 М sulfuric acid, and recrystallized from 1 М sulfuric acid. The crystals are unstaЫe in air and were stored under а nitrogen atmosphere. The cesium manganese alum 1 2 was prepared Ьу adding with stirring а solution of 3 . 6 g of cesium sulfate in 5 mL of 30% sulfuric acid to а solution of 5 . 3 g of freshly prepared manganese(III) acetate 1 3 in 30 mL of 30% sulfuric acid cooled to О 0 С. The mixture was kept at 3 °С for 3 days while crystallization occurred. The product is very susceptiЫe to decomposition Ьу heat and Ьу dehydration but сап Ье kept in the mother liquor at low temperatures . The cesium cobalt alum8 was prepared Ьу electrolysis of cobalt(II) sulfate for 5 h at 0.2 А, producing а large deposit of Ыuе microcrystals. Aging these under refrigeration for 1 w eek produced larger crystals which were collected Ьу filtration and washed with 1 7 М acetic acid followed Ьу dry ether. Infrared spectra were recorded on а Perkin-Elmer Model 580 spectrometer, generally using cesium iodide plates. With the man­ ganese and cobalt alums, oxidation of the iodide in the plates occurs. ( 1 2) Christensen, О, Т. Z. Anorg. Chem. 190 1 , 27, 329. ( 1 3) Brauer, G. "Handbook of Preparative Inorganic Chemistry", 2nd ed.; Academic Press: New York, 1 963.