Synthesis of Tungsten Carbonyl and Nitrosyl Complexes of Monodentate and Chelating Aryl-N-sulfonylphosphoramides, the First Members of a New Class of Electron-Withdrawing Phosphine Ligands. Comparative IR and 13C and 31P NMR Study of Related Phosphorus Complexes

Article abstract of DOI:10.1021/ic960256z

Reaction of N,N'-bis(tolylsulfonyl)-1,2-diaminoethane with PhPCl₂ gives in 62% yield the phosphonous diamide 2-phenyl-1,3-bis(p-tolylsulfonyl)-1,3,2-diazaphospholidine (4, TosL ) and with Ph₂PCl in 43% yield the diphosphinous amide N,N'-bis(diphenylphosphino)-N,N'-bis(p-tolylsulfonyl)-1,2-ethanediamine (5, diTosL ). Reaction of 4 with (THF)W(CO)₅ gives (TosL)W(CO)₅ (6) in 77% yield, and reaction of 5 with trans-BrW(CO)₄NO gives cis, cis, trans-(diTosL)W(CO)₂(NO)Br (8) in 86% yield. The IR, ¹³C NMR, and ³¹P NMR spectra of 4, 5, 6, and 8 are compared to those of a variety of compounds including LW(CO)₅ (L = PMe₃, PPh₃, PPh(NEt₂)₂, P(OMe)₃, P(CF₃)₃), L₂W(CO)₂(NO)Br (L₂ = Ar₂PCH₂CH₂PAr₂ (Ar = Ph (diphos), C₆F₅ (diphos-F₂₀)), (CH₃CN)₂), and the free ligands as appropriate. The IR data are interpreted to suggest a relative ordering of ligand acceptor ability of P(CF₃)₃ > 4 ≈ P(OMe)₃ > PPh₃ ≈ PPh(NEt₂)₂ and a relative ordering of ligand donor ability of PPh(NEt₂)₂ > P(OMe)₃ > PPh₃ > 4 > P(CF₃)₃. The chelating ligand diTosL is about as electron-withdrawing as diphos-F₂₀, on the basis of the IR data. The ³¹P NMR data qualitatively support the conclusion that TosL and diTosL are highly electron-withdrawing ligands, on the basis of ¹Jpw. The ¹³C data do not permit any such generalizations, although the spectra of the diphosphine ligands and adducts are of interest due to the observation of virtual coupling that surprisingly can be simulated only as ABX rather than AA'X spin-systems.

Full text of DOI:10.1021/ic960256z

Inorg. Chem. 1996, 35, 5453-5459
5453
Synthesis of Tungsten Carbonyl and Nitrosyl Complexes of Monodentate and Chelating
Aryl-N-sulfonylphosphoramides, the First Members of a New Class of
Electron-Withdrawing Phosphine Ligands. Comparative IR and C and P NMR Study of
Related Phosphorus Complexes
13
31
William H. Hersh,* Ping Xu, Bing Wang, Jong Won Yom, and Cheslan K. Simpson
Department of Chemistry and Biochemistry, Queens College of the City University of New York,
Flushing, New York 11367-1597
ReceiVed March 8, 1996X
Reaction of N,N′-bis(tolylsulfonyl)-1,2-diaminoethane with PhPCl2 gives in 62% yield the phosphonous diamide
2-phenyl-1,3-bis(p-tolylsulfonyl)-1,3,2-diazaphospholidine (4, “TosL”) and with Ph2PCl in 43% yield the
diphosphinous amide N,N′-bis(diphenylphosphino)-N,N′-bis(p-tolylsulfonyl)-1,2-ethanediamine (5, “diTosL”).
Reaction of 4 with (THF)W(CO)5 gives (TosL)W(CO)5 (6) in 77% yield, and reaction of 5 with trans-BrW(CO)4NO
gives cis, cis, trans-(diTosL)W(CO)2(NO)Br (8) in 86% yield. The IR, 13C NMR, and 31P NMR spectra of 4, 5,
6, and 8 are compared to those of a variety of compounds including LW(CO)5 (L ) PMe3, PPh3, PPh(NEt2)2,
P(OMe)3, P(CF3)3), L2W(CO)2(NO)Br (L2 ) Ar2PCH2CH2PAr2 (Ar ) Ph (diphos), C6F5 (diphos-F20)), (CH3CN)2),
and the free ligands as appropriate. The IR data are interpreted to suggest a relative ordering of ligand acceptor
ability of P(CF3)3 > 4 ≈ P(OMe)3 > PPh3 ≈ PPh(NEt2)2 and a relative ordering of ligand donor ability of
PPh(NEt2)2 g P(OMe)3 > PPh3 > 4 > P(CF3)3. The chelating ligand diTosL is about as electron-withdrawing
as diphos-F20, on the basis of the IR data. The 31P NMR data qualitatively support the conclusion that TosL and
1
diTosL are highly electron-withdrawing ligands, on the basis of JPW. The 13C data do not permit any such
generalizations, although the spectra of the diphosphine ligands and adducts are of interest due to the observation
of “virtual coupling” that surprisingly can be simulated only as ABX rather than AA′X spin-systems.
Introduction
the possibility existed that the nitrogen would not be nucleophilic
enough to displace chloride from phosphorus. A small number
of fully heteroatom-substituted phosphorus compounds such as
112,13 and 214-16 have been prepared in this manner and reported
Syntheses of transition metal ligands that mimic the strong
π-acidity of carbon monoxide, but are larger and so can play a
steric role in reactivity at the metal, have focused on fluorinated
phosphines,1-4 although a recent report describes the use of
N-pyrrolylphosphines.5 Another strongly electron-withdrawing
group that might affect phosphorus σ-donation and π-acidity is
the sulfonyl group. Rather than attempt to directly attach the
sulfonyl group to phosphorus, however, we envisioned the
presence of an intermediate nitrogen atom to give N-sulfo-
nylphosphoramides as our target ligands. While phosphor-
amides of the type RnP(NR′2)3-n (n ) 0,1,2) are well-known6-8
as is the use of primary sulfonamides to generate R3PdNSO2R
species,6 we required secondary sulfonamides to generate the
desired ligands. Reaction with phosphorus chlorides in a
manner analogous to secondary amines9-11 seemed straight-
forward, but since sulfonamides are relatively acidic compounds,
recently, but we are aware of only a single aryl-substituted
analog (3), which was prepared using p-tolylsulfonyl isocyan-
ate.17 None of these compounds have been considered for study
as ligands in transition metal compounds. In fact, we report
here that the synthesis of aryl-substituted N-sulfonylphosphor-
X Abstract published in AdVance ACS Abstracts, August 15, 1996.
(1) King, R. B. Acc. Chem. Res. 1980, 13, 243-248.
(2) Ernst, M. F.; Roddick, D. M. Inorg. Chem. 1989, 28, 1624-1627.
(3) Brookhart, M.; Chandler, W. A.; Pfister, A. C.; Santini, C. C.; White,
P. S. Organometallics 1992, 11, 1263-1274.
(4) Kundig, E. P.; Dupre, C.; Bourdin, B.; Cunningham, A.; Pons, D.
HelV. Chim. Acta 1994, 77, 421-428.
(11) Ramirez, F.; Patwardhan, A. V.; Kugler, H. J.; Smith, C. P. J. Am.
Chem. Soc. 1967, 89, 6276-6282.
(5) Moloy, K. G.; Petersen, J. L. J. Am. Chem. Soc. 1995, 117, 7696-
7710.
(6) Fluck, E. Top. Phosphorus Chem. 1967, 4, 291-481.
(7) Frank, A. W. In Organic Phosphorus Compounds; Kosolapoff, G. M.,
Maier, L., Eds.; Wiley-Interscience: New York, 1973; Vol. 4; Chapter
10.
(8) Gerrard, W.; Hudson, H. R. In Organic Phosphorus Compounds;
Kosolapoff,G. M., Maier, L., Eds.; Wiley-Interscience: New York,
1973; Vol. 5; Chapter 13.
(9) Stuebe, C.; Lankelma, H. P. J. Am. Chem. Soc. 1956, 78, 976-977.
(10) Ewart, G.; Payne, D. S.; Dorte, A. L.; Lane, A. P. J. Chem. Soc. 1962,
3984-3990.
(12) Chen, R.; Liu, Z.; Li, C. Gaodeng Xuexiao Huaxue Xuebao 1989, 10,
655-657.
(13) Chem. Abstr. 1990, 112, 217071q.
(14) Jacob, P.; Richter, W.; Ugi, I. Liebigs Ann. Chem. 1991, 519-522.
(15) Hunsch, S.; Richter, W.; Ugi, I.; Chattopadhyaya, J. Liebigs Ann.
Chem. 1994, 269-275.
(16) Puri, N.; Hunsch, S.; Sund, C.; Ugi, I.; Chattopadhyaya, J. Tetrahedron
1995, 51, 2991-3014.
(17) Cragg, R. H. Chem. Ind. (London) 1967, 1751.
(18) Apel, J.; Bacher, R.; Grobe, J.; Le Van, D. Z. Anorg. Allg. Chem.
1979, 453, 39-52.
S0020-1669(96)00256-X CCC: $12.00 © 1996 American Chemical Society

5454 Inorganic Chemistry, Vol. 35, No. 19, 1996
Hersh et al.
remove Et3NH+Cl-. Solvent removal on a rotary evaporator again gave
a pale yellow powder (5.90 g) which was dissolved (in the air) in 20
mL of hot CH2Cl2 and then treated with 20 mL of hot anhydrous ether.
Cooling to room temperature gave white crystals, and crystallization
was completed at -10 °C overnight; filtration and rinsing with hexanes
gave 2.57 g (43% yield) of 5 as air-stable white crystals. 1H NMR
(CDCl3): δ 7.46 (d, J ) 8.2 Hz, 4H, Ts), 7.36 (m, 4H, Ph), 7.29-7.20
(m, 16H, Ph), 7.18 (d, J ) 8.2 Hz, 4H, Ts), 3.30 (br s, 4H, CH2), 2.41
(s, 6H, CH3). MS: 581 (1.5%, M+ - Ts), 396 (2%, M+ - Ts - PPh2),
183 (100%, TsNCH2+). Anal. Calcd for C40H38N2O4S2P2: C, 65.20;
H, 5.20; N, 3.80. Found: C, 64.84; H, 5.05; N, 3.75.
amides is simple, and describe the first examples of such a
phosphine, including a bis(sulfonamido)phenylphosphine ligand
and a bis((diphenylphosphino)sulfonamido)ethane ligand, the
synthesis of tungsten complexes of these ligands, and spectro-
scopic data in support of the thesis that they are indeed strongly
electron-withdrawing.
Experimental Section
General Data. All manipulations of air-sensitive compounds were
carried out either in a Vacuum Atmospheres inert atmosphere drybox
under recirculating nitrogen or by using standard Schlenk techniques.
1H, 13C, and 31P NMR spectra were recorded on an IBM/Bruker WP-
200SY spectrometer and a Bruker DPX-400 spectrometer; chemical
shifts are reported relative to TMS or residual hydrogens in CD2Cl2 (δ
5.32) or CDCl3 (δ 7.24), to CDCl3 at 77.0 ppm for 13C NMR, and to
external 85% H3PO4 at 0 ppm (positive values downfield) for 31P NMR.
13C NMR data are collected in Table 1 and 31P NMR data in Table 2.
Infrared spectra were obtained on a Mattson Galaxy 4020 FT-IR
spectrometer with 0.1 mm NaCl solution cells. Elemental analyses
were performed by Desert Analytics, Tucson, AZ, and Quantitative
Technologies, Inc., Whitehouse, NJ. Mass spectra (EI, 70 eV) were
obtained on an HP5988A spectrometer.
All solvents were treated under nitrogen. Benzene, diethyl ether,
and tetrahydrofuran were distilled from sodium benzophenone ketyl.
Hexane was purified by washing successively with 5% nitric acid in
sulfuric acid, water, sodium bicarbonate solution, and water and then
dried over calcium chloride and distilled from n-butyllithium in hexane.
Methylene chloride was distilled from phosphorus pentoxide; CDCl3
and CD2Cl2 were vacuum-transferred from phosphorus pentoxide.
Silica gel (200-400 mesh) was dried for several hours under vacuum
while heating with a heat gun and was transferred under vacuum into
the drybox. PhPCl2, Ph2PCl (Aldrich), 1,2-bis(dipentafluorophenyl-
phosphino)ethane (Strem), and W(CO)6 (Pressure Chemical) were used
as received, and trans-BrW(CO)4NO (contaminated by ∼26% by weight
of W(CO)6 on the basis of elemental analysis) was prepared as
previously described.19,20
2-Phenyl-1,3-bis(p-tolylsulfonyl)-1,3,2-diazaphospholidine (4, TosL).
Under a nitrogen atmosphere, a solution of 1.17 mL of PhPCl2 (8.62
mmol) in 40 mL of anhydrous ether was added dropwise over a 50
min period to an ice-cooled suspension of 3.18 g of CH3C6H4-
SO2N(H)CH2CH2N(H)SO2C6H4CH321 (8.63 mmol) and 3.0 mL of Et3N
(distilled under N2 from CaH2; 21.5 mmol) in 220 mL of anhydrous
ether. The mixture was then stirred at room temperature for an
additional 30 min, and then the solvent was removed on a rotary
evaporator. The resultant white powder was purified by dissolving
the residue in 50 mL of CH2Cl2 and passing the solution through a
∼100 mL pad of silica gel packed in CH2Cl2 on a 150 mL sintered
glass frit. The product was eluted with 300 mL more CH2Cl2, and the
solvent removed on a rotary evaporator to give 2.54 g (62% yield) of
4 as a spectroscopically pure white powder. Material submitted for
elemental analysis was crystallized from 3:1 CH2Cl2/hexane at -40
°C in the glovebox. 1H NMR (CDCl3): δ 7.69 (m, 2H, Ph), 7.63 (d,
J ) 8.2 Hz, 4H, Ts), 7.45 (m, 3H, Ph), 7.18 (d, J ) 8.2 Hz, 4H, Ts),
3.50 (m, 2H, C(Ha)HbC(Ha)Hb), 3.20 (m, 2H, C(Ha)HbC(Ha)Hb), 2.44
(s, 6H, CH3). MS: 410 (13%, M+ - SO2), 409 (17%, M+ - HSO2),
397 (2%, P(N(Ts)CH2CH2NTs)+), 255 (72%, M+ - Ts - SO2), 155
(14%, Ts+), 91 (100%, C7H7); the 20 eV MS of 4 was virtually the
same as that at 70 eV. Anal. Calcd for C22H23N2O4S2P: C, 55.69; H,
4.89; N, 5.90. Found: C, 55.44; H, 4.83; N, 5.94.
(TosL)W(CO)5 (6). Tungsten hexacarbonyl (1.10 g, 3.13 mmol)
and 80 mL of THF were placed in a 250 mL septum-capped flask with
a magnetic stirring bar, which was placed in a water-cooled water bath
and irradiated under a nitrogen atmosphere with a 450 W Hanovia
medium pressure mercury lamp for 3 h. The resultant yellow solution
was transferred via syringe into a solution of 4 (1.09 g, 2.30 mmol,
0.73 equiv based on W(CO)6) in 10 mL of THF. No color change
was observed, and the reaction mixture was stirred at room temperature
under a nitrogen atmosphere and monitored periodically by IR. After
2.5 h little further change was observed, and the THF was removed on
a vacuum line at room temperature. The resultant off-white powder
was then warmed at ∼45 °C on the vacuum line to facilitate sublimation
of unreacted W(CO)6, and 1.65 g (90% yield based on 4) of crude
product was obtained. Attempts to crystallize this material from
CH2Cl2, toluene, or mixtures of CH2Cl2/hexane and CH2Cl2/1,1,2-
trichlorotrifluoroethane all failed. Final purification was achieved by
filtration of a CH2Cl2 solution of the material through silica gel, eluting
first with CH2Cl2 and then 1:1 CH2Cl2/THF, to give 5 (∼85% recovery
after solvent removal and washing with hexane, 77% overall yield) as
a spectroscopically pure white powder. A small sample of analytically
pure material was obtained with difficulty by recrystallization from
3:1 hexane-CH2Cl2. IR (CH2Cl2): 2080 (m), 1952 cm-1 (s). 1H NMR
(CDCl3): δ 7.56, 7.43, 7.32 (m, 5H, Ph), 7.12, 7.06 (AB quartet, J )
8.5 Hz, 8H, Ts), 3.90 (m, (approximate quintet, J ) 5 Hz), 2H, C(Ha)-
HbC(Ha)Hb), 3.51 (m, (approximate quintet, J ) 5 Hz), 2H, C(Ha)-
HbC(Ha)Hb), 2.35 (s, 6H). MS: 410 (3%, M+ - SO2), 409 (5%, M+
- HSO2), 255 (30%, M+ - Ts - SO2), 155 (11%, Ts+), 91 (100%,
C7H7). Anal. Calcd for C27H23N2O9S2PW: C, 40.62; H, 2.90; N, 3.51.
Found: C, 40.76; H, 2.75; N, 3.44.
(bis(diethylamino)phenylphosphine)W(CO)5 (7). A solution of
0.88 g (2.50 mmol) of W(CO)6 in 80 mL of THF was photolyzed for
3 h as described above for 5. A solution of PPh(NEt2)2 (0.57 g, 2.26
mmol, 0.9 equiv; prepared10 in 52% yield as an air-stable spectroscopi-
cally pure cloudy white liquid, 1H NMR (CDCl3) δ 7.43, 7.3, 7.23 (m,
2H, 2H, 1H), 3.09 (dq, 3JPH ) 9.6 Hz, 3JHH ) 7.0 Hz, 8H), 1.11 (t, JHH
) 7.0 Hz, 12 H)) in 10 mL of THF was added by syringe to the
(THF)W(CO)5 solution, which was allowed to stir under a nitrogen
atmosphere overnight. After solvent removal (vacuum line), the
resultant yellow solid and clear oil was heated at 40 °C under vacuum
for 2 h in order to partially remove unreacted W(CO)6. The residue
was then taken up in ∼10 mL of hexane, and the resulting solution
filtered, concentrated, and filtered again to remove additional W(CO)6.
Attempted recrystallization at -40 °C gave ∼35 mg of a mixture of
the product and more W(CO)6. The remainder was applied to 15 mL
of silica on a frit and eluted with 60 mL of hexane followed by 100
mL of 10% ether in hexane. The solvent was removed from this second
fraction, and the resultant solid was taken up in the minimum amount
of hexane, and the resulting solution was filtered and stripped to give
0.32 g (25% yield) of product as a pale yellow powder. IR (hexane):
2069.6 (m), 1977.4 (w), 1941.7 (s), 1935.9 (vs) cm-1
.
1H NMR
N,N′-Bis(diphenylphosphino)-N,N′-bis(p-tolylsulfonyl)-1,2-ethanedi-
amine (5, diTosL). Under a nitrogen atmosphere, 3.59 g (16.3 mmol)
of Ph2PCl was added dropwise to a solution of 3.00 g (8.14 mmol) of
CH3C6H4SO2N(H)CH2CH2N(H)SO2C6H4CH3 and 2.06 g (20.35 mmol)
of Et3N in 50 mL of THF, and the mixture was heated at reflux for
21.5 h. After solvent removal on a rotary evaporator, the resultant
pale yellow solid was taken up in 100 mL of benzene and filtered to
(CDCl3): δ 7.52, 7.45, 7.36 (m, 5H), 3.36 (m, 4H), 3.23 (m, 4H), 1.16
(t, J ) 7.0 Hz, 12H). MS: 576 (M+). Anal. Calcd for C19H25N2O5-
PW: C, 39.60; H, 4.37; N, 4.86. Found: C, 39.77; H, 4.20; N, 4.77.
cis, cis, trans-(diTosL)W(CO)2(NO)Br (8). In the glovebox a
solution of 226 mg of trans-BrW(CO)4NO (0.412 mmol) and 206 mg
of 5 (0.280 mmol) in 4 mL of CHCl3 was allowed to stand in a screw-
capped 1 dram vial at room-temperature for 24 h. The solvent then
was removed on a vacuum line to gave 320 mg of crude product as a
yellow solid. Purification was carried out by column chromatography
in the glovebox, and eluting with 9:1 CH2Cl2/hexane through silica
gel to gave 261 mg (86% yield) of analytically pure 8 as a yellow
(19) Colton, R.; Commons, C. J. Aust. J. Chem. 1973, 26, 1493-1500.
(20) Bonnesen, P. V.; Puckett, C. L.; Honeychuck, R. V.; Hersh, W. H. J.
Am. Chem. Soc. 1989, 111, 6070-6081.
(21) Berkowitz, W. F. J. Chem. Educ. 1970, 47, 536.

Tungsten Sulfonylphosphoramides
Inorganic Chemistry, Vol. 35, No. 19, 1996 5455

5456 Inorganic Chemistry, Vol. 35, No. 19, 1996
Table 2. 31P NMR Data
Hersh et al.
chemical shift (ppm)a
compound
free ligand
complex
∆δP (ppm)b
1JPW (Hz)
JPP (Hz)c
d,18
PMe3W(CO)5
PPh3W(CO)5
-62.0
-4.82
97.67
-2.5
91.28
141.61
-12.03
-43.59
60.37
-39.8
21.43
105.41
55.4
122.06
141.02
32.74
7.75
22.2
26.3
7.7
57.9
30.8
-0.6
44.8
51.3
16.7
230
244.1
289.5
300
329.5
386.5
249.1
262.2
270.8
PPh(NEt2)W(CO)5 (7)
d,18
P(CF3)3W(CO)5
(TosL)W(CO)5 (6)
P(OMe)3W(CO)5
(diphos)W(CO)2(NO)Br (9)
(diphos-F20)W(CO)2(NO)Br (10)
(diTosL)W(CO)2(NO)Br (8)
3c
5c
77.11
a CDCl3 solution except as noted. b δ(complex) - δ(free ligand). c Derived from analysis of the 13C NMR spectra; for free diphos, 3JPP ≈ 35 Hz
5
and for free diTosL, JPP ) 0 Hz (see text). d C6D6 solution.
Scheme 1
Scheme 2
powder. IR (CH2Cl2): 2036 (s), 1968 (s), 1650 (br, m), 1598 (w) cm-1
.
1H NMR (CDCl3): δ 7.91 (m, 4H, Ph), 7.69 (m, 4 H, Ph), 7.47 (m,
12H of Ph), 7.07 (d, J ) 8.2 Hz, 4H, Ts), 6.82 (d, J ) 8.2 Hz, 4H,
Ts), 4.45 (m, 2H, C(Ha)HbC(Ha)Hb), δ 4.03 (m, 2H, C(Ha)HbC(Ha)-
Hb), δ2.38 (s, 6H, CH3). Anal. Calcd for C42H38N3O7P2S2WBr: C,
46.43; H, 3.52; N, 3.87. Found: C, 46.50; H, 3.85; N, 3.94.
cis, cis, trans-(1,2-bis(dipentafluorophenylphosphino)ethane)-
(CO)2W(NO)Br (10). In the glovebox a solution of 227 mg of trans-
BrW(CO)4NO (0.414 mmol) and 210 mg of 1,2-bis(dipentafluoro-
phenylphosphino)ethane (diphos-F20, 0.277 mmol) in 12 mL of CHCl3
was stirred at room temperature for 23 h. The reaction mixture was
filtered through a pad of Celite, and the solvent was removed on a
vacuum line to gave 390 mg of crude product as a yellow solid.
Purification was accomplished by flash chromatography on an 18 × 1
cm silica gel column eluting with CH2Cl2 to give 304 mg of yellow
powder, followed by washing with hexane (in which 10 is slightly
soluble) to give 186 mg (61% yield) of 10 as an analytically pure yellow
powder. IR (CH2Cl2): 2054 (s), 1992 (s), 1650 (br, s), 1522 (s), 1483
(s) cm-1. IR of diphos-F20 (CH2Cl2): 1642 (w), 1520 (s), 1482 (s)
on the pyramidal phosphorus and the other due to the two
hydrogen atoms trans to the phenyl. In 5, these four hydrogens
give rise to a singlet, as expected. Otherwise the spectral
characterization of 4 and 5 is straightforward.
Synthesis of Tungsten Adducts. Preparation of the tungsten
pentacarbonyl adduct of 4 is conveniently carried out on a 1-g
cm-1 1H NMR (CDCl3): δ 3.33 (m, 2H, C(Ha)HbC(Ha)Hb), 3.02 (m,
.
22
scale by reaction with (THF)W(CO)5 to give 6 in 77% yield
2H, C(Ha)HbC(Ha)Hb). Anal. Calcd for C28H4NO3F20P2BrW: C, 30.35;
as a white powder (Scheme 2). Coordination of 4 to tungsten
via the phosphorus atom is clear on the basis of the observed
chemical shifts and the tungsten-phosphorus satellites (due to
the 14% natural abundance of 183W) in the 31P NMR (4, 91.3
H, 0.36; N, 1.26. Found: C, 30.57; H, 0.52; N, 1.24
Results and Discussion
Synthesis of N-Sulfonylphosphoramides. Reaction of N,N′-
bis(tolylsulfonyl)-1,2-diaminoethane with PhPCl2 or Ph2PCl
in the presence of NEt3 in ether or THF gave the desired
phosphonous diamide “TosL” (4, 2-phenyl-1,3-bis(p-tolylsul-
fonyl)-1,3,2-diazaphospholidine) in 62% yield and the diphos-
phinous amide “diTosL” (5, N,N′-bis(diphenylphosphino)-N,N′-
bis(p-tolylsulfonyl)-1,2-ethanediamine) in 43% yield, both as
air-stable white crystalline solids (Scheme 1). Like the starting
bis(sulfonamide), 4 and 5 are apparently polar materials as
judged by their insolubility in ether, but other solubility
properties are peculiar: 4 is benzene insoluble and slightly
soluble in acetone, while 5 is insoluble in acetone and ethanol
1
ppm; 6, 122.1 ppm, JPW ) 329 Hz) and phosphorus-carbon
coupling in the 13C NMR spectrum (JPC ) 37 and 8 Hz for the
trans and cis carbonyl ligands, respectively). In addition, a non-
sulfonyl analog, (Et2N)2PhPW(CO)5 (7), was prepared by the
same route and found to exhibit NMR and IR spectra that were
sufficiently comparable to conclude that the compounds are both
P-ligated LW(CO)5 adducts.
In order to prepare a chelated complex from 5, reactions were
carried out with trans-BrW(CO)4NO,19,20 since the resultant
adduct would be the precursor for the desired Lewis acid Diels-
Alder catalyst.20 Using our standard method (30 min reflux in
THF)20 or letting the reactants stand in CHCl3 for 1 day gave
a high yield of the desired complex 8 (Scheme 2). During the
1
yet soluble in benzene. The H NMR spectra of 4 and 5 are
consistent with the cyclic and acyclic structures shown. In 4
the signals for the CH2CH2 ring hydrogens are split into two
multiplets, one due to the two hydrogen atoms cis to the phenyl
(22) Strohmeier, W.; Mu¨ller, F. J. Chem. Ber. 1969, 102, 3608-3612.

Tungsten Sulfonylphosphoramides
Inorganic Chemistry, Vol. 35, No. 19, 1996 5457
of combinations of coupling constants.26-28 For 8 and 9, the
signals with JPC ≈ 6-20 Hz all give three-line signals, where
the separation of the two outer lines is equal to |JPC + JP′C|;
since in most cases the longer range coupling is zero, the
separation of the outer lines is equal in magnitude to JPC. The
height of the middle line depends on the relative size of JPP,
allowing its magnitude to be approximated by simulation of
the observed spectrum. The multiplets with JPC ≈ 40-50 Hz
would not be expected to yield a detectable central line in the
multiplet. Simulations of the observed multiplets were carried
out, and while no attempt was made to carry out a completely
rigorous analysis, coupling contstants for 8 and 9 that can
generate the observed multiplets were found with JPP ≈ 5 Hz
for 8 and 3 Hz for 9. Representative NMR signals are shown
in Figure 1, illustrating the dependence of the appearance of
these multiplets on JPP. The value for 9 is consistent with a
literature value of 5.5 Hz for (diphos)W(CO)4.25 The value for
8 might be compared to values25 of -15.0 Hz for a complex
with a 7-member chelate ring, (Ph2P(CH2)4PPh2)W(CO)4, and
of 2.9 Hz for a complex with P-N bonds in the chelating ligand,
(Ph2P(NH)PPh2)W(CO)4, but no conclusions can be drawn from
these particular numbers.
Figure 1. 13C NMR spectra. The meta PPh carbon atoms provide
representative AA′X 13C NMR signals (A, A′ ) 31P, X ) 13C),
3
illustrating the effect of JPP (JP′C ≈ 0 Hz): (a) 9, 129.03 ppm, JPC
)
3
10.5 Hz, JPP ) 3 Hz; (b) 8, 128.40 ppm, JPC ) 9.7 Hz, JPP ≈ 5 Hz;
3
(c) diphos, 128.44, JPC ) 6.4 Hz, JPP ) 35 Hz.
course of this work it was found that if 5 was contaminated
with any Et3NH+Clj remaining from its synthesis, the reaction
with BrW(CO)4NO gave a new compound that is spectroscopi-
cally similar to 8 and presumed to be the chloride analog. The
facility of this side-reaction is unexpected and these results will
be reported separately.23 The known Ph2PCH2CH2PPh2 (diphos)
adduct (9) was prepared for comparison to 8,24 as was adduct
10 of the electron-withdrawing diphosphine ligand (C6F5)2PCH2-
CH2P(C6F5)2 (diphos-F20). Unfortunately 10 was too insoluble
1
to acquire a 13C NMR spectrum, although the IR, H, and 31P
The 13C NMR spectra of diTosL and diphos were recorded
as well. The spectrum of diTosL apparently is first-order
NMR spectral features are comparable to those of 8 and 9.
The IR and NMR spectra of 8 are similar to those of known
chelate nitrosyl bromide analogs20,24 and so it is presumed to
be isostructural. Each of 8, 9, and 10 exhibit two carbonyl bands
and a nitrosyl band in the IR, although 8 reproducibly exhibits
a small splitting of this band (∼4 cm-1) suggestive of two
conformational isomers either due to the 7-member chelate ring
or more likely due to rotation about the N-S bonds of the tosyl
moieties. In addition, 10 exhibits bands at 1522 and 1483 cm-1
that are even stronger than the carbonyl and nitrosyl bands, but
these are assigned to C6F5 ring modes since they are virtually
unchanged from bands observed in the IR spectrum of the free
ligand. The molecular symmetry of 8 is confirmed by the fact
that 8 and 9 each exhibit a singlet in their 31P NMR spectrum
while 10 exhibits a single fluorine-coupled multiplet, each with
the expected 183W satellites; furthermore, each of 8-10 exhibits
5
because JPP ) 0, but the complete absence of visible phos-
2
3
phorus coupling to the backbone CH2 (i.e. JPC ) JPC ) 0) is
unexpected. An alternative explanation is that |JPC + JP′C| )
0, where 2JPC ) -3JPC. This equality is nearly reached in diphos
(Vide infra), so this possibility cannot be dismissed, and in fact
other workers have reached this conclusion previously.28,29
The 100 MHz 13C NMR spectrum of diphos was consistent
with that previously reported at 25 MHz,29 but could not be
simulated as an AA′X spin-system. In particular, the center
line of the putative AA′X CH2 multiplet is split by 2.5 Hz, and
the pattern is readily modeled as an ABX system with ∆νPP′
)
3 Hz; the chemical shift equivalency of the phosphorus atoms
is broken by the proximity of one to 13C and the other to 12C.
The backbone CH2 groups of 9 also exhibit such splitting of
what in this case is a weak central line, and in general the spectra
for 8 and 9 were better fit using ∆νPP′ ) 1-3 Hz. Other workers
have commented on the possibility that the rigorously ABX
nature of this type of spin-system can affect the observed
spectrum,28-31 but examples that necessitate the ABX formula-
tion, including (η3-2-methylallyl)2Ru(PMe3)2, trans-(MeO)2P-
(O)CHdCHP(O)(OMe)2, Ph(H)PCH2CH2P(H)Ph, and some
Ph2P(CH2)3PPh2 platinum complexes, are rare.32-35 The likely
1
two multiplets in the H NMR spectrum for the two pairs of
chelate backbone hydrogens that are syn to the NO and Br
ligands, respectively. In addition, 8 exhibits a single tosyl
methyl singlet, so if there is any conformational isomerism
involving the relative orientations of the tolyl moieties, it is
1
fast on the H NMR time scale. The 13C NMR spectra of 8
and 9 similarly conform to the proposed symmetry, with a single
signal for the chelate backbone carbon and two sets of signals
for the phenyl carbons (two phenyl rings syn to NO, two to Br)
and for 8 a single set of tolyl carbons, again indicating the
absence of conformational differences on the NMR time scale.
Simulation of the 13C NMR Spectra. While the number
of signals in the 13C NMR spectra were consistent with the
proposed structures, the individual signals merit scrutiny, since
many exhibit additional lines due to “virtual coupling” to the
distant phosphorus atom (Table 1, Figure 1). Each 13C nucleus
is nominally part of an AA′X system where the 13C X nucleus
is coupled to the magnetically inequivalent AA′ phosphorus
nuclei. The observed multiplets will depend on the sign and
relative magnitudes of 2JPP (coupling though tungsten), 3JPP or
5JPP (coupling through the chelate backbone for 9 and 8
respectively), and the various carbon-phosphorus coupling
constants.25 Simulated spectra have been published for a variety
(26) Redfield, D. A.; Nelson, J. H.; Cary, L. W. Inorg. Nucl. Chem. Lett.
1974, 10, 727-733.
(27) Redfield, D. A.; Cary, L. W.; Nelson, J. H. Inorg. Chem. 1975, 14,
50-59.
(28) Verstuyft, A. W.; Nelson, J. H.; Cary, L. W. Inorg. Chem. 1976, 15,
732-734.
(29) King, R. B.; Cloyd, J. C. J. J. Chem. Soc., Perkin Trans. 2 1975,
938-941.
(30) Pregosin, P. S.; Kunz, R. HelV. Chim. Acta 1975, 58, 423-431.
(31) Pregosin, P. S.; Kunz, R. W. 31P and 13C NMR of Transition Metal
Phosphine Complexes; Springer-Verlag: Berlin, 1979; Vol. 16.
(32) Jolly, P. W.; Mynott, R. In AdVances in Organometallic Chemistry;
Stone, F. G. A., West, R., Eds.; Academic Press: San Diego, CA,
1981; Vol. 19, pp 257-304. A second example (a nickel bis-
(dicyclohexyl)phosphinoethane complex) seems apparent in published
spectra in this review (for the cyclohexyl carbon bound to phosphorus,
Figure 14, p 300), but it is not discussed.
(33) Harris, R. K.; Jones, M. S.; Kenwright, A. M. Magn. Reson. Chem.
1993, 31, 1085-1087.
(34) Kimpton, B. R.; Mcfarlane, W.; Muir, A. S.; Patel, P. G.; Bookham,
J. L. Polyhedron 1993, 12, 2525-2534.
(23) Xu, P.; Hersh, W. H. Unpublished results.
(24) Connelly, N. G. J. Chem. Soc., Dalton Trans. 1973, 2183-2188.
(25) Andrews, G. T.; Colquhoun, I. J.; McFarlane, W. Polyhedron 1983,
2, 783-790.
(35) Andrews, M. A.; Voss, E. J.; Gould, G. L.; Klooster, W. T.; Koetzle,
T. F. J. Am. Chem. Soc. 1994, 116, 5741-5746 (Supporting Informa-
tion, S-1).

5458 Inorganic Chemistry, Vol. 35, No. 19, 1996
Hersh et al.
Table 3. Carbonyl Stretching Frequencies, Force Constants,42 and σ and π-Bonding Parameters38 for LW(CO)5 Complexesa
ν(CO) cm-1
FCO (md/Å)
L
A12 (w)
A11 (m)
E (s)
k1
k2
ki
∆d
∆π
PPh(NEt2)2
PPh3
2069.6
2071.2
2078.9
2081.1
2101
1941.7
1942.0
1962.2
1960.0
2006
1935.9
1942.0
1947.8
1958.4
1998
15.46
15.45
15.79
15.72
16.38
15.78
15.87
15.96
16.10
16.53
0.32
0.31
0.32
0.30
0.27
-0.1346
-0.0131
-0.0934
0.0787
0.1789
0.1583
0.3324
0.2967
0.6268
P(OMe)3
TosL (4)b
c
P(CF3)3
0.1848
2
1
a Solvent is hexane except as noted. b ν(CO) in CH2Cl2: 2079.9 (w, A1 ), 1992.6 (w, shoulder, B1) 1951.6 cm-1 (s, overlapping A1 and E),
giving k1 ) 15.59, k2 ) 16.03, and ki ) 0.31 mdyn/Å. The IR in hexane was taken by FT-IR with nitrogen purging of the sample chamber,
acquiring 250 scans at 1 cm-1 resolution. The saturation solution (baseline at 99.90% transmittance (%T) with a noise level of (0.03 %T) gave
2
1
for the A1 band an intensity of 99.7 %T and for the overlapping A1 and E bands an intensity of 98.4 %T. c Cyclohexane solution.18
Table 4. Carbonyl and Nitrosyl Stretching Frequencies and Force
reason for our observation, at least in the case of diphos, is the
Constants44 for cis-L2W(CO)2(NO)(Br) Complexesa
1
higher field-strength used (400 MHz for H), due not to the
ν(CO) cm-1
A2 (m) B (m)
FCO/NO (md/Å)
kNO kCO ki
ν(NO) cm-1
higher dispersion in the 13C NMR where the observed splittings
are merely due to spin-spin coupling, but rather to the higher
dispersion in the 31P NMR, since this will increase ∆νPP′. These
results will be described in detail elsewhere.36
b
L2
A1 (m)
c
(CH3CN)2 2015
1930
1630
11.82 15.66 0.62
11.78 15.96 0.53
12.11 16.14 0.51
12.03 16.54 0.46
diphos 2025.8 1954.4
diTosL (5) 2035.0 1966.6
diphos-F20 2055.3 1994.4
1630.0
1652.4d
1649.2
Infrared Spectra. Ligand σ-donor and π-acid strength is
most simply assessed by infrared CO stretching frequencies.37-39
While the separation of these effects by IR has been ques-
tioned,40,41 comparisons that involve a constant donor atom (here
phosphorus) may be made with relative confidence. Infrared
data, calculated Cotton-Kraihanzel CO force constants,42 and
derived parameters are collected in Table 3 for W(CO)5 adducts
of several phosphorus ligands including 4. Consideration of
the CO stretching frequencies themselves or the trans and cis
CO stretching force constants k1 and k2 clearly shows that while
4 is not as good a π-acceptor nor as poor a σ-donor as P(CF3)3,
it is somewhat comparable to P(OMe)3 despite the presence of
the phenyl group in place of a third heteroatom. Consideration
of Dobson’s ∆d and ∆π parameters38 suggests than 4 is
comparable in π-acceptor ability to P(OMe)3 but is a much
weaker σ-donor. The effect of the sulfonyl groups is enormous,
since PPh(NEt2)2 is a much stronger σ-donor and weaker
π-acceptor than 4. The data suggests a relative ordering of
ligand acceptor ability of P(CF3)3 > 4 ≈ P(OMe)3 > PPh3 ≈
a Solvent is benzene except as noted. b Diphos and diphos-F20
)
Ar2PCH2CH2PAr2 (Ar ) Ph and C6F5, respectively). c Solvent is
CH2Cl2.43 d Average of reproducibly observed splitting of peak (1654.1,
1650.8 cm-1).
data suggest that the sulfonyl moiety is quite an effective
electron-withdrawing group.
NMR Spectra. Numerous groups have attempted to correlate
electronic properties of ligands with their NMR chemical shifts
and/or coupling constants,39,45-49 but it is fair to say that none
of the results have led to the type of widespread use enjoyed
by the above IR methods. For phosphorus tungsten penta-
carbonyl complexes, there appears to be a positive correlation
1
between JPW and the electron-withdrawing ability of the
ligand.45,46,50,51 In order to determine whether or not the NMR
data for the N-sulfonylphosphoramide ligands fits this correlation
and/or is unusual, 13C and 31P NMR spectra were collected for
PPh3, PPh(NEt2)2, P(OMe)3, and their W(CO)5 adducts for
comparison to those of 4 and 6, along with data for the chelating
phosphines (Tables 1, 2).
PPh(NEt2)2, and of ligand donor ability of PPh(NEt2)2
g
P(OMe)3 > PPh3 > 4 > P(CF3)3. On the basis of the IR data,
the N-sulfonylphosphoramide ligand is second only to fluorin-
ated phosphines in electron-deficiency.
1
The order of JPW increases as PMe3 < PPh3 < PPh(NEt2)2
< P(CF3)3 < TosL < P(OMe)3, and for the chelating ligands
as diphos < diphos-F20 < diTosL, again supporting the notion
that the phosphoramides are highly electron-withdrawing,
although the precise ordering on this basis is suspect. The
change in phosphorus chemical shift upon coordination, ∆δP,
varies from -0.6 ppm for P(OMe)3 to +57.9 ppm for P(CF3)3,
and does not appear to be of any utility. For instance, P(OMe)3
and P(CF3)3 are both electron-withdrawing ligands (on the IR
scale) and both exhibit large tungsten-phosphorus coupling
constants of 386 and 300 Hz, respectively, yet their ∆δP values
fall at the two extremes. Of the remaining ligands, data for all
except PPh(NEt2)2 (for which ∆δP ) +7.7) fall in the relatively
narrow range of 24 ( 7 ppm after one takes account of the
well-documented “ring contribution” ∆R for chelating phos-
A smaller set of data was collected for comparison of the
diTosL (5) and Ar2PCH2CH2PAr2 (Ar ) Ph, C6F5) ligands using
the nitrosyl bromide adducts 8-10 as well as the analogous
bis(acetonitrile) adduct43 cis,cis,trans-(CH3CN)2(CO)2(NO)WBr
(Table 4). As before Cotton-Kraihanzel force constants were
calculated,44 but no attempt was made to compare σ and π
effects for this limited data set. The data show that the electron-
withdrawing diTosL and fluorinated ligands are qualitatively
quite different from the stronger donor diphos and acetonitrile
ligands; 8 actually exhibits the highest ν(NO) and kNO of the
group. What makes this particularly interesting is that the
substitution of two perfluorophenyl groups for the phenyl rings
of diphos has a comparable effect to substitution of one
sulfonamide for the CH2 linker of diphos. Once again, the IR
(36) Hersh, W. H. Submitted for publication in J. Chem. Educ.
(37) Graham, W. A. G. Inorg. Chem. 1968, 7, 315-321.
(38) Brown, R. A.; Dobson, G. R. Inorg. Chim. Acta 1972, 6, 65-71.
(39) Grobe, J.; Le Van, D. Z. Anorg. Allg. Chem. 1984, 518, 36-54.
(40) Pacchioni, G.; Bagus, P. S. Inorg. Chem. 1992, 31, 4391-4398.
(41) Davies, M. S.; Pierens, R. K.; Aroney, M. J. J. Organomet. Chem.
1993, 458, 141-146.
(45) Grim, S. O.; Wheatland, D. A.; McFarlane, W. J. Am. Chem. Soc.
1967, 89, 5573-5577.
(46) Keiter, R. L.; Verkade, J. G. Inorg. Chem. 1969, 8, 2115-2120.
(47) Mathieu, R.; Lenzi, M.; Poilblanc, R. Inorg. Chem. 1970, 9, 2030-
2034.
(48) Guns, M. F.; Claeys, E. G.; Van der Kelen, G. P. J. Mol. Struct. 1979,
54, 101-109.
(49) Honeychuck, R. V.; Hersh, W. H. Inorg. Chem. 1987, 26, 1826-
1828.
(42) Cotton, F. A.; Kraihanzel, C. S. J. Am. Chem. Soc. 1962, 84, 4432-
4438.
(43) Hersh, W. H. Inorg. Chem. 1990, 29, 713-722.
(44) Cotton, F. A. Inorg. Chem. 1964, 3, 702-711.
(50) Verkade, J. Coord. Chem. ReV. 1972/73, 9, 1-106.
(51) Schumann, H.; Kroth, H.-J. Z. Naturforsch., B 1977, 32B, 768-770.

Tungsten Sulfonylphosphoramides
Inorganic Chemistry, Vol. 35, No. 19, 1996 5459
phines.52 That is, five-member chelate rings typically are
deshielded by ∼30 ppm compared to the related nonchelated
adducts; for instance ∆R for (diphos)W(CO)4 is +27.3 ppm.
The limited data available for larger rings suggests that ∆R )
0 for the seven-member ring of 8. After this adjustment, then,
all three of the chelating adducts would have ∆δP values near
+20 ppm.
effect was proposed to be due to rotation about the P-N bonds,
while the large disparity in coupling constants despite the near-
identity of attached atoms was presumed to be due to hybridiza-
tion changes at phosphorus as well as fixing the P-N lone pair
dihedral angles. This would suggest that for PPh(NEt2)2 the
steric and electronic consequences of coordination to tungsten
are simply larger than for the other compounds, while the cyclic
TosL ligand perhaps undergoes a somewhat different hybridiza-
tion change upon coordination than do the other (acyclic)
compounds.
2
In the 13C NMR, JPC for the trans carbonyl ligand (22, 26,
37, and 38 Hz, respectively for PPh3, PPh(NEt2)2, 4, and
1
P(OMe)3) follows the same pattern as JPW (2JPC for the cis-
Further close inspection of the data reveals other curious
pointssfor instance 2JPC is larger than 1JPC for free PPh3, PPh-
(NEt2)2, diphos, and diTosL, but smaller for TosL and all of
the complexed ligands. One-bond phosphorus-carbon coupling
constants in particular are well-known to be quite variable,54
so this switching in absolute magnitude of one- and two-bond
coupling constants is not unusual. This comparison depends
on the validity of the assignments of the ortho and meta phenyl
carbons, but phosphorus coupling to ortho and meta phenyl
carbons is known to follow the order JPC > JPC. Further
confidence in the ortho/meta assignments arises from the internal
consistency of the chemical shift data: all of the ortho carbons
are downfield of the meta carbons. Nonetheless, the limited
data set does not permit further generalizations.
CO’s follows the same trend but lies in a range from 7.3 to
10.9 Hz). A unique feature of this set of phosphines is the
presence of the phenyl group on phosphorus, so in addition to
the comparisons of 1JPW obtained from the 31P NMR, compari-
sons of phenyl coupling constant and chemical shift data
obtained from the 13C NMR can be made. Another unique
feature is that in all cases for the nonchelating set, the 13C NMR
spectra were measured at 50 and 100 MHz, so assignments
inVolVing doublets due to coupling to phosphorus are unam-
biguous. While the chemical shifts do not change much from
compound to compound or upon complexation, coupling
constants of the phosphorus-phenyl carbons do vary although
at this point the differences are better described as a curiosity
than an illuminating point of comparison. Thus, PPh3 exhibits
2
3
54
an increase in the one-bond P-Cipso coupling constant (∆Jipso
)
Conclusions
of 31 Hz upon coordination (from 10.7 to 41.6 Hz), while 4
exhibits a decrease of 12 Hz (from 32.1 to 20.1 Hz). While
PPh(NEt2)2 exhibits an increase like PPh3, the actual values are
quite different, namely a 67 Hz increase from 3.4 to 70.2 Hz.
The chelating ligands exhibit ∆Jipso values (26 and 29 Hz for 8
and 9) that are similar to that of PPh3, so TosL is anomalous in
The demonstration of the ease of synthesis of TosL and
diTosL, the first arylsulfonylphosphoramide ligands, opens up
a variety of research opportunities. The IR and to a lesser extent
the NMR data suggest that the sulfonamide moiety is strongly
electron-withdrawing in character, on a par with perfluorinated
alkyl groups. Work in progress includes the use of these ligands
in tungsten nitrosyl Diels-Alder catalysts and the synthesis of
chiral analogs55 that may allow catalytic asymmetric induction
of Diels-Alder reactions.
the sign and PPh(NEt2)2 in the magnitude of ∆Jipso
.
An obvious question is whether or not PPh(NEt2)2 is an
appropriate model compound for TosL. We chose not to
examine the more appropriate cyclic analog, 1,3-dimethyl-2-
phenyl-1,3,2-diazaphospholidine (i.e. the analog of TosL in
which the tosyl moieties are replaced by methyls), because of
a warning about acute nausea and vomiting caused by several
Acknowledgment. We thank Prof. Paul Pregosin and Dr.
Mark Andrews for bringing work in refs 32-35 to our attention.
Financial support for this work from The City University of
New York PSC-CUNY Research Award Program and the
Howard Hughes Undergraduate Biological Science Program
(J.W.Y. and C.K.S.), and the National Science Foundation
(Grant CHE-9408535) for funds for the purchase of the 400
MHz NMR spectrometer is gratefully acknowledged.
1
related cyclic compounds.11 In fact, JPC for PPh(NEt2)2 has
been reported to vary with temperature, rising from 0 Hz in the
20-45 °C range to 2 Hz at 85 °C (we have no explanation for
the discrepancy between these values and our room-temperature
value of 3.4 Hz; the chemical shifts are comparable). In
contrast, in 1,3-dimethyl-2-phenyl-1,3,2-diazaphospholidine, 1JPC
) 42 Hz and is temperature independent.53 The temperature
IC960256Z
(54) Gorenstein, D. G. Prog. NMR Spectrosc. 1983, 16, 1-98.
(55) Hersh, W. H.; Xu, P.; Simpson, C. K.; Wood, T.; Rheingold, A. L.
Manuscript in preparation.
(52) Garrou, P. E. Chem. ReV. 1981, 81, 229-266.
(53) Gray, G. A.; Nelson, J. H. Org. Magn. Reson. 1980, 14, 8-13.