Electronic Effects in Asymmetric Catalysis: Structural Studies of Precatalysts and Intermediates in Rh-Catalyzed Hydrogenation of Dimethyl Itaconate and Acetamidocinnamic Acid Derivatives Using C2-Symmetric Diarylphosphinite Ligands

J . Org. Chem. 1999, 64, 3429-3447

3429

Electr on ic Effects in Asym m etr ic Ca ta lysis: Str u ctu r a l Stu d ies of

P r eca ta lysts a n d In ter m ed ia tes in Rh -Ca ta lyzed Hyd r ogen a tion of

Dim eth yl Ita con a te a n d Aceta m id ocin n a m ic Acid Der iva tives

Usin g C₂-Sym m etr ic Dia r ylp h osp h in ite Liga n d s

T. V. RajanBabu,* Branko Radetich, Kamfia K. You,^†Timothy A. Ayers,^‡

Albert L. Casalnuovo,^§and J oseph C. Calabrese

Department of Chemistry, The Ohio State University, Columbus, Ohio 43210, and DuPont Central

Research and Development, Experimental Station, Wilmington, Delaware 19880

Received J anuary 22, 1999

Enantioselectivity of Rh(I)-catalyzed asymmetric hydrogenation of dehydroamino acid derivatives

and dimethyl itaconate can be enhanced by the appropriate choice of substituents on the aromatic

rings of vicinal diarylphosphinites derived from carbohydrates as well as trans-cyclohexane-1,2-

diol. For example, the use of phosphinites with electron-donating bis(3,5-dimethylphenyl) groups

at phosphorus provide high ee’s in these reactions whereas electron-withdrawing aryl substituents

decrease the enantioselectivity. In this paper, an attempt is made to clarify the origin of these

remarkable electronic effects at two levels. First, crystal structures of a number of precatalysts

([phosphinite]₂Rh⁺[diolefin]X^-) were determined and their structures were studied in detail to

examine the electronic effects, if any, on the ground-state conformations of these molecules. A study

of six of these complexes reveals that the gross conformational features of these precatalysts are

largely unaffected by electronic effects, which suggests that other explanations have to be sought

for the electronic amplification of enantioselectivity. One possibility is a change in the diastereomeric

equilibrium between the initially formed [substrate]Rh⁺[phosphinite] complexes as a function of

electronic effect of the ligand. In the Rh-catalyzed hydrogenation of dimethyl itaconate, we have

examined this equilibrium between the major and minor complexes by ³¹P NMR. There is a clear

difference in the ratio of these two diastereomers when 3,5-dimethylphenylphosphinite vis-a`-vis

the unsubstituted diphenylphosphinite is used. Electron-deficient ligands such as 1,2-bis-3,5-

diflurophenylphosphinite and 1,2-bis-3,5-bis-trifluromethylphenylphosphinite appear to form these

diastereomers more readily at room temperature, even though the exact ratio of the diastereomers

could not be established with any certainty.

In tr od u ction

rather than the thermodynamic stabilities of the diaster-

eomeric intermediates play a major role. In these in-

stances, it is impossible to predict, even qualitatively, the

selectivity of the reaction without knowing the details of

the kinetics and the position of the equilibrium involving

the relevant diastereomeric intermediates. In most cases,

a combination of both of these factors operates. As in the

case of the well-studied Rh-catalyzed hydrogenation of

acetamidoacrylic acid derivatives,⁴the reaction could be

under Curtin-Hammett conditions and the structure of

Strategies for controlling the regio- and stereoselec-

tivity of reactions is at the heart of modern organic

chemistry research. In asymmetric catalysis using orga-

nometallic reagents, where the origin of selectivity

depends on the properties of a chiral ligand, two principal

levels of induction can be recognized.¹(a) Catalyst-

substrate interactions lead to diastereomeric intermedi-

ates of widely different thermodynamic stabilities, each

reacting at comparable rates (or the more stable diaste-

reomer reacting faster), resulting in a reaction whose

selectivity can be predicted with reasonable certainty if

the diastereomeric composition is known. In this case, if

the exact structure of the major diastereomer can be

deduced, for example, from X-ray crystallography or from

solution NMR studies, there is also the possibility of

tuning the ligand to maximize the enantioselectivity.^2,3

(b) A second limiting scenario is when the reactivities

(2) For some recent examples, see: (a) Trost, B. M.; Van Vranken,

D. L. Chem. Rev. 1996, 96, 395 and references therein. (b) Bao, J .;

Wulff, W. D.; Rheingold, A. L. J . Am. Chem. Soc. 1993, 115, 3814. (c)

Corey, E. J .; Noe, M. C. J . Am. Chem. Soc. 1993, 115, 12579.

(3) Pd-catalyzed allylation of stabilized nucleophiles provides one

of the best examples of the use of structural information of diastero-

meric complexes in the design of a catalyst. For selected recent

references, see: (a) Trost, B. M.; Van Vranken, D. L.; Bingel, C. J .

Am. Chem. Soc. 1992, 114, 9327. (b) Allen, J . V.; Coote, S. J .; Dawson,

G. J .; Frost, C. G.; Martin, C. J .; Williams, J . M. J . J . Chem. Soc.,

Perkin Trans. 1 1994, 2065. (c) Sprinz, J .; Kiefer, M.; Helmchen, G.;

Reggelin, M.; Huttner, G.; Walter, O.; Zsolnai, L. Tetrahedron Lett.

1994, 35, 1523. (d) Brown, J . M.; Hulmes, D. I.; Guiry, P. J .

Tetrahedron 1994, 50, 4493. (e) Togni, A.; Breutel, C.; Schnyder, A.;

Spindler, F.; Landert, H.; Tijani, A. J . Am. Chem. Soc. 1994, 116, 4062.

(f) von Matt, P.; Lloyd-J ones, G. C.; Minidis, A. B. E.; Pfaltz, A.; Macko,

L.; Neuburger, M.; Zehnder, M.; Ru¨egger, H.; Pregosin, P. S. Helv.

Chim. Acta 1995, 78, 265. (g) Seebach, D.; Devaquet, E.; Ernst, A.;

Hayakawa, M.; Ku¨hnle, F. N. M.; Schweizer, W. B.; Weber, B. Helv.

Chim. Acta 1995, 78, 1636. (h) Pen˜a-Cabrera, E.; Norrby, P.-O.;

Sjo¨gren, M.; Vitagliano, A.; De Felice, V.; Oslob, J .; Ishii, S.; O’Neill,

D.; A° kermark, B.; Helquist, P. J . Am. Chem. Soc. 1996, 118, 4299.

^†Current address: BASF Corp., 1609 Biddle Ave., Wyandotte, MI

48192.

^‡Current address: Hoechst Marion Roussel, 2110 East Galbraith

Rd, Cincinnati, OH 45215.

^§Current address: DuPont Agricultural Products, Stine-Haskell

Research Center S300/124, Newark, DE 19714.

Questions regarding X-ray structures should be addressed to J .C.C.

at DuPont Central Research.

(1) (a) Bosnich, B. Asymmetric Catalysis; Martinus Nijhoff Publish-

ers: Dordrecht, 1986. (b) For a recent incisive analysis, see: Heller,

D.; Buschmann, H. Top. Catal. 1998, 5, 159.

Published on Web 04/24/1999

3430 J . Org. Chem., Vol. 64, No. 10, 1999

RajanBabu et al.

Ta ble 1. Electr on ic Effects in Asym m etr ic Ca ta lysis^a

a

F igu r e 1. Diastereomeric P ₂Rh(1) complexes of acetami-

doacrylates. When the ligand is C₂-symmetric, 1 and 3 (also 2

and 4) are identical.

For a more complete list see refs 5b and 6b.

the catalytically relevant diastereomer may not be known

with any certainty. To obtain high selectivities from such

reactions, an empirical approach to ligand design remains

a powerful option. In our previous work on the enanti-

oselective hydrocyanation (eq 1)⁵and hydrogenation (eq

2) reactions,⁶we have shown that the selectivity and

tioselectivity, whereas in Rh(I)-catalyzed hydrogenation,

electron-rich phosphinites are the most ideal.

While the sugar-derived phosphinites are among the

most easily accessible ligands, and their use results in

the highest enantioselectivities for the hydrogenation and

hydrocyanation reactions for several substrates, the non-

C₂-symmetric nature of the backbone leaves several

unanswered questions about the role of the unsym-

metrical backbone. For example, consider the potential

intermediates 1-4 shown in Figure 1. In a truly C₂-

symmetric ligand there will be only two square-planar

Rh(I) complexes: 1 will be equivalent to 3, and 2 will be

equivalent to 4. But the unsymmerical backbone of 9 (or

10) breaks such symmetry, making 1 and 3 marginally

different in these cases, even though both lead to addition

of H₂from the Si face of the prochiral olefin. The role of

each of these intermediates was not obvious from our

previous studies. For example, the reactivities of 1 and

3 could be different in the subsequent rate-determining

oxidative addition to form the octahedral Rh(III) hydride.

This difference would have an effect on the selectivity,

which may be superimposed on the electronic effects we

observed. To eliminate the potential contributions to

selectivity by such intermediates and truly gauge the

electronic effects in the absence of any other ligand

biases, we undertook a study of the hydrogenation of

acetamidocinammic acid derivatives using truly C₂-

symmetric diarylphosphinites from optically pure trans-

1,2-cyclohexanediol. The results of these studies are

reported in this paper. Having experimentally confirmed

the electronic effects in this C₂-symmetric system, we

decided to examine the ground-state conformations of

several precatalysts [(L)Rh(I)(diolefin)]⁺X^-by X-ray crys-

tallography. In the case of C₂-symmetric (and a few non-

C₂-symmetric) ligands that form five- and seven-mem-

bered Rh-chelates, some degree of success has been

achieved in relating the conformation of the precatalyst

to the sense of asymmetric induction (vide infra, ref 27).

Thus, the first question is, can electronic effect of a ligand

alone bring about a conformational change in the pre-

catalyst? If so, does this fit the already established

empirical relationships?

efficiency of a catalyst system can be optimized by the

choice of a tunable ligand system whose steric and

electronic properties can be systematically varied, whether

the effects of such variations are predictable. As shown

in Table 1, one of the important outcomes of these early

studies was the emergence of electronic effects of ligands

as a control element in asymmetric catalysis.⁷Using

sugar phosphinites (L*) as ligands, we have shown that

in the Ni(0)-catalyzed asymmetric hydrocyanation of

vinyl arenes electron-deficient ligands enhance the enan-

(4) (a) Landis, C. R.; Halpern, J . J . Am. Chem. Soc. 1987, 109, 1746.

(b) Brown, J . M. Chem. Soc. Rev. 1993, 25. (c) Bircher, H.; Bender, B.

R.; von Philipsborn, W. J . Magn. Reson. Chem. 1993, 31, 293.

(5) (a) RajanBabu, T. V.; Casalnuovo, A. L. J . Am. Chem. Soc. 1992,

114, 6265. (b) Casalnuovo, A. L.; RajanBabu, T. V.; Ayers, T. A.;

Warren, T. H. J . Am. Chem. Soc. 1994, 116, 9869.

(6) (a) RajanBabu, T. V.; Ayers, T. A.; Casalnuovo, A. L. J . Am.

Chem. Soc. 1994, 116, 4101. (b) RajanBabu, T. V.; Ayers, T. A.;

Halliday, G. A.; You, K. K.; Calabrese, J . C. J . Org. Chem. 1997, 62,

6012.

(7) Several leading references to the role of electronic effect as a

control element in asymmetric catalysis can be found in refs 5 and 6.

For a more recent report, see: Palucki, M.; Finney, N. S.; Pospisil, P.

J .; Gu´ller, M. L.; Ishida, T.; J acobsen, E. N. J . Am. Chem. Soc. 1998,

120, 948.

Two other important questions related to the origin of

electronic amplification of enantioselectivity in the Rh-

catalyzed asymmetric hydrogenation are as follows: (i)

How are the relative reactivities of the two diastereomers

affected by ligand electronics? (ii) What is the effect of

Electronic Effects in Asymmetric Catalysis

J . Org. Chem., Vol. 64, No. 10, 1999 3431

Sch em e 1. Syn th esis of

1,2-Dia r ylp h osp h in oxycycloh exa n e

ligand electronics on the ratio of the major and minor

diastereomers? Of these effects, the former is more

intractable.⁸However, the effect of ligand electronics on

the ratio of the major to minor diastereomers, in prin-

ciple, is more discernible, especially in the Rh-catalyzed

hydrogenation of dimethyl itaconate^13,14(eq 3), a reaction

7

6

5

that shares important mechanistic similarities with the

acetamidoacrylate reductions. In the itaconate system,

the diastereomeric intermediates can be more easily

characterized by low-temperature NMR spectroscopy.¹³

Here we report our attempts to probe the ligand elec-

tronic effects on the equilibrium between the major and

minor diastereomers. An added impetus for the study of

itaconate hydrogenation is the importance of succinnic

acid derivatives as valuable intermediates for several

pharmaceutical preparations including potent renin in-

hibitors.¹⁵The results of these studies are also reported

in this paper.

9

7

10

of the reactions in which they were used were low. We

prepared both the (+) and the (-) ligands from com-

mercially available 1,2-cyclohexanediols (99% ee) and the

corresponding diarylchlorophosphines^5bby procedures

reported previously (Scheme 1).¹⁷Likewise, the cationic

Rh-complexes were prepared by reacting stoichiometric

amounts of Rh⁺(diolefin)₂X^-[diolefin ) norbornadiene,

1,4-cyclooctadiene; X ) OTf, SbF₆, or BF₄) and the

diarylphosphinite in methylene chloride, acetone, or

Resu lts a n d Discu ssion

Syn th esis of th e C₂-Sym m etr ic Liga n d s a n d Th eir

Com p lexes: tr a n s-1,2-Dia r ylp h osp h in oxycycloh ex-

a n e. (S,S)-(+)-trans-1,2-bis(diphenylphosphinoxy)cyclo-

hexane was one of the first phosphinites used in Rh-

catalyzed asymmetric hydrogenations.¹⁶The enantiomeric

purity of the ligand and, therefore, the enantioselectivity

THF.¹⁸These compounds were fully characterized by H

1

and ³¹P NMR and in selected cases (vide infra) by X-ray

crystal structure determination. The ³¹P NMR of the

ligands and the complexes are recorded in Table 2.

Hyd r ogen a tion of Dim eth yl Ita con a te. The hydro-

genation of dimethyl itaconate using various cationic Rh-

complexes were carried out at 30-40 psi of hydrogen in

a Fischer-Porter tube. The results of hydrogenation

using complexes of 7a -e are shown in Table 3. For

comparison, the results of the reductions with two C₁

symmetric ligands derived from carbohydrates⁶(9 and

10) are also included in Table 3. In each case, the

enantiomeric excess was determined by chiral GC on a

Chiraldex GTA column. The absolute configuration of the

product was assigned on the basis of a comparison of

elution properties of an authentic sample prepared by

hydrogenation of dimethyl itaconate using (2S,4S)-BPPM-

(8) The origin of the difference in the reactivities of the two

diastereomeric Rh(I) complexes has never been fully understood despite

the extensive structural information that has become available.^4b,c,9

For example, consider the following: Brown suggested that the major

diastereomer produced upon coordination of acetamidocinnamate to

cationic Rh(1)P ₂is the more stable (and therefore less reactive), and

this complex leads to a sterically more conjested octahedral Rh(III)

complex. This makes the catalytically relevant minor isomer more

reactive, as observed experimentally. In sharp contrast, a more recent

molecular mechanics calculation¹⁰suggests that the diastereomeric

octahedral Rh(III) complexes could not be reliably differentiated in

energy from any other diastereomers and, in any case, cannot be viable

on energry considerations alone. A highly distorted octahedral geom-

etry is nonetheless possible, and tentative experimental support for

this idea has emerged from a recent para-hydrogen induced polaraza-

tion NMR (PHIP NMR) study of a related Rh-catalyzed hydrogena-

tion.¹¹A second explanation for the different reactivities is suggested

by the different ¹⁰³Rh chemical shifts (and the attendant difference in

electron density at Rh) in each of the two diastereomers.¹²How this

might be related to the reactvities of the two diastereomers remains

an open question.

(9) (a) McCulloch, B.; Halpern, J .; Thompson, M. R.; Landis, C. R.

Organometallics 1990, 9, 1392. (b) Chan, A.; Pluth, J .; Halpern, J . J .

Am. Chem. Soc. 1980, 102, 5952.

(10) Giovannetti, J . S.; Kelly, C. M.; Landis, C. R. J . Am. Chem.

Soc. 1993, 115, 4040.

(11) Harthun, A.; Kadyrov, R.; Selke, R.; Bargon, J . Angew. Chem.,

Int. Ed. Engl. 1997, 36, 1103.

(12) Bender, B. R.; Koller, M.; Nanz, D.; von Philipsborn, W. J . Am.

Chem. Soc. 1993, 115, 5889.

(Rh⁺)(COD) SbF₆catalyst, which is known to produce

-

predominantly the (S)-isomer.¹⁵As can be seen from

Table 3, the enantioselectivity depends on the electronic

nature of the P-Ar substituent, with the electron-deficient

phosphinites giving poor selectivity. As the electron

density of the aromatic ring, and presumably that on Rh,

is reduced the ee falls of precipitously. The ∑σ values,

which are a measure of the combined electronic effect of

the groups, parallel the enantioselectivity of the reaction

in each case. The case of the p- vs m-fluorine substitituent

is particularly interesting. The σ value for the p-F

substituent is 0.06 while that for m-fluoro substituent is

0.34, clearly showing the diminished electron-withdraw-

ing effect of a fluorine in the para position, presumably

due to the resonance-donating effect.¹⁹From a synthetic

(13) (a) Brown, J . M.; Parker, D. J . Org. Chem. 1982, 47, 2722. (b)

Brown, J . M.; Parker, D. J . Chem. Soc., Chem. Commun. 1980, 342.

(c) Kadyrov, R.; Freier, T.; Heller, D.; Michalik, M.; Selke, R. J . Chem.

Soc., Chem. Commun. 1995, 1745. (d) Heller, D.; Kadyrov, R.; Michalik,

M.; Freier, T.; Schmidt, U.; Krause, H. W. Tetrahedron Asymmetry

1996, 7, 3025.

(14) In principle, the equilibration of intermediates in the Rh-

catalyzed hydrogenation of acetamidocinnamate could also be studied

by the type of experiments described in refs 4c and 12.

(15) (a) Inoguchi, K.; Morimoto, T.; Achiwa, K. J . Organomet. Chem.

1989, 370, C9-C12. (b) J endralla, H. Tetrahedron Lett. 1991, 32, 3671.

(c) Morimoto, T.; Chiba, M.; Achiwa, K. Tetrahedron Lett. 1989, 30,

735.

(16) Tanaka, M.; Ogata, I. J . Chem. Soc., Chem. Commun. 1975,

735.

(17) (a) Selke, R.; Pracejus, H. J . Mol. Catal. 1986, 37, 213. (b)

Unruh, J . D.; Christenson, J . R. J . Mol. Catal. 1982, 14, 19.

(18) Schrock, R. R.; Osborn, J . A. J . Am. Chem. Soc. 1971, 93, 2397.

3432 J . Org. Chem., Vol. 64, No. 10, 1999

RajanBabu et al.

Ta ble 2. ³¹P NMR Sp ectr a of P h osp h in ite Liga n d s a n d Th eir Cor r esp on d in g Rh (1) Com p lexes

entry

Ar in 7

ligand

³¹P NMR (solvent)

111.5 (C₆D₆)

complex

³¹P NMR (solvent)

-

1

2

3,5-Me₂C₆H₃

Ph

7a

7b

Rh⁺[7a ][COD]SbF₆

133.4 (d, 178) (CD₂Cl₂)

131.0 (d, 177) (CDCl₃)

129.7 (d, 178) (THF-d₈)

133.3 (d, 179) (CD₂Cl₂)

127.9 (d, 178) (C₆D₆)

123.4 (d, 183) (CD₂Cl₂)

120.7 (d, 183) (CD₃COCD₃)

131.5 (d, 219) (CD₂Cl₂)

129.17 (d, 193) (THF-d₈)

-

108.7 (C₆D₆)

Rh⁺[7b][COD]Sbf₆

3

4-F-C₆H₄

3,5-F₂C₆H₃

7c

7d

105.3 (C₆D₆)

103.9 (C₆D₆)

Rh⁺[7c][COD]SbF₆_-

Rh⁺[7d ][COD]SbF₆

Rh⁺[7d ][COD]OTf^-

[7d ]Rh(µ-Cl)/2

-

4

3,5-(CF₃)₂C₆H₃

7e

102.3 (C₆D₆)

Rh⁺[7e][COD]SbF₆

99.6 (THF-d₈)

-

Rh⁺[7e][COD]SbF₆_-

124.8^a(d, 178) (CD₂Cl₂) (??)

127.25 (d, 182) (CD₃COCD₃)

119.0 (d, 182) (CDCl₃)

130.17 (d, 218) (CD₂Cl₂)

127.67 (d, 218) (CD₃COCD₃)

104.96 (d, 201) (CD₂Cl₂)

102.00 (ddm, 211, 28); 133.28

(dd, 168, 28) (CD₂Cl₂)

Rh⁺[7e][COD]SbF₆

Rh⁺[7e][COD]OTf^-

[7e]Rh(µ-Cl)/2

-

5

6

C₆F₅

C₆F₅/3,5-Me₂C₆H₃

7f

7g

85.0 (q, 41) CDCl₃

86.72 (q, 39); 108.3 (CD₂Cl₂)

Rh⁺[7f][COD]SbF₆

Rh⁺[7g][COD]SbF₆

-

7

C₆H₅/3,5-CF₃C₆H₃

7h

Rh⁺[7h ][COD]SbF₆

125.5 (dd, 268, 36); 126.8 (dd,

243, 36) (CD₃COCD₃)

-

99.5, 109.4 (C₆D₆)

a

Uncertain because the major product is [7e]Rh(µ-Cl)/2 in this solvent.

Ta ble 3. Electr on ic Effects in Rh (1)-Ca ta lyzed

pared to the former. Thus, both enantiomers of the

itaconates could, in principle, be obtained using ligands

readily available from natural sources. The unnatural

sugar ^L-glucose is more expensive than D-glucose, but

nonethelss could also be used for the production of the

(S)-isomer.

Hyd r ogen a tion of Dim eth yl Ita con a te^a

H yd r ogen a t ion of (Z)-r-N-Acet ylcin n a m ic Acid

Der iva tives Usin g C₂-Sym m etr ic Bis-p h osp h in ites.

Hydrogenation of dehydroamino acid derivatives (eq 2)

using cationic Rh(I) complexes of the C₂-symmetric

ligands 7 were carried out in THF at 35-40 psi of

hydrogen, and the results are tabulated in Table 4. All

experiments were carried out with (1S,2S)-1,2-bis-

arylphosphinoxycyclohexane or its optical antipode,

(1R,2R)-bis-phosphinite. Several important experimental

observations warrant comments here. The most impor-

tant conclusion is the remarkable electronic effect seen

across the board for these C₂-symmetric ligands. Results

for the prototypical dehydroamino acid ester, (Z)-methyl

R-N-acetylcinnamate (entry 1), are illustrative. In com-

parison with the unsubstituted diphenylphosphinite 7b

(74.2% ee), the electron-deficient ligands 3,5-difluorophen-

yl (7d ) and 3,5-bis-trifluoromethylphenyl (7e) phosphinite

gave the lowest ee’s in this series (49.6% ee and 25.9%

ee, respectively). A more electron-rich 3,5-dimethylphen-

ylphosphinite 7a gave the highest ee (82.0% ee) of all the

phosphinites under study. Most interestingly, pen-

taflurophenylphosphinite gave predominantly the op-

posite enantiomer of the amino acid ester.

Ar in ligands

∑σ

entry

7, 9, and 10

(σ_por 2 × σ_m)^b

7

9

10

1

2

3

4

5

3,5-Me₂C₆H₃

C₆H₅

-0.14

0.0

0.062

0.67

0.86

79.2 (R) 90.8 (R) 75.3 (S)

67.7 (R) 73.4 (R) 59.4 (S)

57.7 (R) 58.3 (R) 36.4 (S)

25.1 (R) 25.2 (R) 5.7 (S)

-

4-F-C₆H₄

3,5-F₂-C₆H₄

3,5-(CF₃)₂C₆H₃

-

0

5.5 (R)

a

Reaction done in THF with 0.16 mol % catalyst at rt and 45

psi initial H₂pressure. Isolated yield >99% in all cases. Ee

measured by chiral GC on Chiraldex GTA column (see Experi-

b

mental Section for details). See ref 19.

perspective, the synthesis of the (R)-methylsuccinic acid

using the ligand 9 in 90.8% ee represents one of the

highest ee’s obtained for this important class of com-

pounds via hydrogenation using gaseous hydrogen. 4-Hy-

droxyproline-derived BPPM gives >97% (S) ee under

transfer hydrogenation conditions.²⁰However, the op-

posite enantiomer of this catalyst is not readily available

for the production of the (R)-methylsuccinic acid deriva-

tives. Ru(allyl)(acac-F₆)(BINAP) or Ru₂(BINAP)₂Cl₄(NEt₃)

can be used for the synthesis of the (R)-isomer in ee’s

comparable to what is reported here (93 and 88% ee,

respectively).²¹

Another key feature of the reaction is the dependence

of the sense of asymmetric induction on the backbone

chirality of the ligand. Thus, the bis-2,3-diarylphosphinite

derived from a ^D-glucopyranoside (9) gives the (R)-

enantiomer while the bis-3,4-diarylphosphinite (10) gives

the (S)-enantiomer, albeit in lower selectivity, as com-

Another interesting observation is related to the effect

of an electronically unsymmetrical ligand. Two cases are

reported in Table 4. Note that in entry 1 (Table 4) the

bis-pentaflurophenylphosphinite (7f) gave 37.9% ee of the

(R)-enantiomer and the bis-3,5-dimethylphenylphosphin-

ite (7a ) gave 82.0% ee of the (S)-enantiomer in this

reduction. A mixed ligand 7g, which carries one each of

these diarylphosphinites, gave an ee of 79.3% (S), almost

as good as the best observed selectivity.²²A similar result

is obtained with a mixed ligand with phenyl and 3,5-bis-

trifluoromethylphenylphosphinite (last column in Table

(19) Carroll, F. A. Perspectives on Structure and Mechanism in

Organic Chemistry; Brookes/Cole: Pacific Grove, CA, 1998; p 384.

(20) Leitner, W.; Brown, J . M.; Brunner, H. J . Am. Chem. Soc. 1993,

115, 152.

(21) (a) Brown, J . M.; Brunner, H.; Leitner, W.; Rose, M. Tetrahe-

dron: Asymmetry 1991, 2, 331. (b) Kawano, H.; Ikariya, T.; Ishii, Y.;

Saburi, M.; Yoshikawa, S.; Uchida, Y.; Kumobayashi, H. J . Chem. Soc.,

Perkin Trans. I 1989, 1571. (c) Burk et al. reported the use of the

DuPhos series of ligands (>99% ee for dimethyl itaconate) while this

publication was under review: Burk, M. J .; Bienewald, F.; Harris, M.;

Zanotti-Gerosa, A. Angew. Chem., Int. Ed. Engl. 1998, 37, 1931.

(22) The earliest known example of a related enhancement of

enantioselectivity using electronically unsymmetrical ligand was

recorded by Pracejus. Pracejus, H.; Pracejus, G.; Costisella, B. J . Prakt.

Chem. 1987, 329, 235.

Electronic Effects in Asymmetric Catalysis

J . Org. Chem., Vol. 64, No. 10, 1999 3433

Ta ble 4. Electr on ic Effects on Rh -Ca ta lyzed Hyd r ogen a tion of r-N-Acetylcin n a m ic Acid Ester s^a

a

Quantitative yield, reactions done at rt and 35-40 psi H₂pressure in THF. Ee’s determiend by GC (Chiracel-L-Val). See Experimental

b

Section for details. Studies done with (1R,2R)-bis-phosphinites. Sign inverted in table for comparison.

Ta ble 5. Electr on ic Effects in Ni(0)-Ca ta lyzed

Hyd r ocya n a tion of MVN^a

a

b

For experimental procedure, see ref 5b. Measured by chiral

HPLC on a Chiracel OJ column.

4). We had previously seen related stereoelectronic effects

on enantioselectivity in the Ni-catalyzed hydrocyanation

of vinyl arenes. The highest enantioselctivity of asym-

metric hydrocyanation was obtained for a ligand with an

electronically unsymmetrical chelating bis-phosphinite.²³

The ee’s were determined by gas chromatography, and

they are highly reproducible. To verify the generality of

these results, Rh-catalyzed hydrogenation of two other

substrates, methyl esters of R-N-acetyl-4-fluorocinnamic

acid (entry 2, Table 4) and of R-N-acetyl-3,5-bis(trifluo-

romethyl)cinnamic acid (entry 3, Table 4) were also

carried out. As shown in Table 4, these substrates behave

as expected, with the electron-withdrawing ligands giving

the lowest enantioselectivity.

Ni(0)-Ca ta lyzed Asym m etr ic Hyd r ocya n a tion of

2-Meth oxy-6-vin yln a p h th a len e (MVN) Usin g tr a n s-

(1S,2S)-Dia r ylp h osp h in oxycycloh exa n e. In 1994, we

reported the most extensive studies of electronic effects

on a carbon-carbon bond-forming reaction, viz. the Ni-

catalyzed asymmetric hydrocyanation of vinyl arenes.^5b

The 1,2-bis-diarylphosphinoxycyclohexane system now

allows the examination of these effects in a truly C₂-

symmetric ligand. Asymmetric hydrocyanation of MVN

was carried out in toluene at room temperature as

described in our earlier paper,^5band the enantioselec-

tivities obtained are recorded in Table 5. As compared

to the sugar-derived ligands (see Table 1, column 3) the

ee’s are lower; nonetheless, the amplification of selectivity

by an electron-withdrawing group such as m-fluoro- or

F igu r e 2. Metal complexes of bis-1,2-diarylphosphinites.

m-trifluoromethyl is irrefutable from these data.²⁴Thus,

the electronic effect appears to be a more general

phenomenon, not unique to the sugar-derived C₁-sym-

metric bisphosphinites.

Str u ctu r es of th e C₂-Sym m etr ic P h osp h in ites a n d

Th eir Rh (I) a n d Ni(II) Com p lexes. As pointed out

earlier, the most convincing examples of amplification of

enantioselectivity by electronic tuning of ligands em-

ployed C₁- or pseudo-C₂-symmetric ligands.^23,26While it

was abundantly clear that the electronic effects of ligands

played the most decisive role, a quantitative measure of

this contribution, unaffected by backbone effects, was not

possible because of the lack of C₂symmetry (or another

higher symmetry element). With the C₂-symmetric bis-

(diaryl)phosphinites from cyclohexane-1,2-diol we can

begin to address this question. In this connection, the

structural aspects of the catalysts and intermediates are

discussed at two levels: (i) First, solution- and solid-state

structures of the precatalyst complexes 11-16 (Figure

(23) RajanBabu, T. V.; Casalnuovo, A. L. J . Am. Chem. Soc. 1996,

118, 6325.

(24) C₂-Symmetric bis-diarylphosphinites derived from (S,S)-tar-

tranil²⁵are other examples where similar electronic effects on asym-

metric hydrocyanation can be observed; see ref 23.

(25) Bourson, J .; Laureano, O. J . Organomet. Chem. 1982, 229, 77.

(26) (a) Asymmetric hydroformylation: RajanBabu, T. V.; Ayers, T.

A. Tetrahedron Lett. 1994, 35, 4295. (b) Pd-catalyzed asymmetric

allylation: Nomura, N.; Mermet-Bouvier, Y. C.; RajanBabu, T. V.

Synlett 1996, 745.

3434 J . Org. Chem., Vol. 64, No. 10, 1999

RajanBabu et al.

F igu r e 3. Diastereomers of [itaconate]Rh(I)P₂complex.

2) will be discussed. The relevance of such ground-state

structures to the catalytic process itself is, admittedly,

of questionable value. Yet, conformations of ground-state

precatalysts have been correlated with the sense of

asymmetric induction²⁷(vide infra), and as in any discus-

sion of electronic effects on enantioselectivity, structural

consequences of the electronic effects of ligands for the

conformation of the catalyst should be considered. (ii)

Next, our efforts to probe the effect of the P-aryl substit-

uents on the diastereomeric equilibration of (S,S)-1,2-

bis-diarylphosphenoxycylohexane]Rh(+)[[dimethyl ita-

-

conate]SbF₆(Figure 3) and the attendant effect on

enantioselectivity of dimethyl itaconate hydrogenation

will be presented.

The itaconate reduction was chosen as an example of

a prototypical Rh-catalyzed hydrogenation reaction be-

cause of its mechanistic similarity to the more important

reduction of the dehydroamino acids and the ease of

detection of the diastereomeric Rh(I) complexes (Figure

3) under the reaction conditions. Recall that in the case

of the dehydroamino acid reductions these intermediates

are more difficult to observe because the minor diaste-

reomer has broad ³¹P peaks.^13,28

The ³¹P NMR spectra of the ligands and their Rh⁺-

(COD)LX^-complexes (L ) bisphosphinite) clearly show

the C₂-symmetric nature of these compounds (Table 2).

For the ligands 7a -f (Scheme 1), as expected only one

³¹P signal is observed. Upon coordination to Rh, the shifts

move downfield. In addition, the coupling to ¹⁰³Rh is

observed in each case. In two cases, in addition to the

cationic complexes, we also prepared their neutral coun-

terparts, {[7d ][Rh(µ-Cl)]}₂and {[7e][Rh(µ-Cl)]}₂and

recorded their ³¹P NMR. These complexes are also C₂-

symmetric in solution as judged by the presence of a

single ³¹P peak coupled only to ¹⁰³Rh. For a given complex,

while there are considerable differences in the chemical

shifts depending on which solvent is used, the multiplici-

F igu r e 4. (A and B) Conformations of seven-membered

cationic Rh chelates. (C and D) λ- and δ-conformations of C₂-

symmetric vicinal [diarylphosphinite]Rh(I) complexes leading

to S and R amino acids.

nature of the latter in solution. The ¹H NMR shows broad

peaks for the allylic and olefinic hydrogens of the 1,5-

cyclooctadiene (COD) ligand, suggesting slow rotation

around the Rh-COD axis.²⁹

1

ties and the J _Rh-Pgenerally remain constant as the

solvent is varied. The occurrence of P-P coupling (²J _P-P),

which is seen in complexes of C₁-symmetric phosphinites

such as 9, but not of 7, clearly indicates the C₂-symmetric

The ³¹P chemical shifts are dramatically different

between complexes of the electron-rich and electron-poor

lignds. For example, in CD₂Cl₂, complexes of the electron-

(27) (a) Kagan, H. B. Asymmetric synthesis using organometallic

catalysts. In Comprehensive Organomeallic Chemistry; Wilkinson, G.,

Stone, F. G. A., Abel, E. W., Eds.; Pergamon Press: Oxford, 1982; Vol.

8, p 463. (b) Knowles, W. S. Acc. Chem. Res. 1983, 16, 106. (c) Oliver,

J . D.; Riley, D. P. Organometallics 1983, 2, 1032. (d) Bogdan, P. L.;

Irwin, J . J .; Bosnich, B. Organometallics 1989, 8, 1450. (e) Brown, J .

M.; Evans, P. L. Tetrahedron 1988, 44, 4905. (f) Pavlov, V. A.;

Klabunovskii, E. I.; Struchkov, Y. T.; Voloboev, A. A.; Yanovsky, A. I.

J . Mol. Catal. 1988, 44, 217. (g) Seebach, D.; Plattner, D. A.; Beck, A.

K.; Wang, Y. M.; Hunziker, D.; Petter, W. Helv. Chim. Acta 1992, 75,

2171. (h) Sakuraba, S.; Morimoto, T.; Achiwa, K. Tetrahedron: Asym-

metry 1991, 2, 597. For a highly pertinent, critical evaluation of these

various models, see ref 10.

(28) (a) Chan, A. C. S.; Pluth, J . J .; Halpern, J . J . Am. Chem. Soc.

1980, 102, 5952. (b) Brown, J . M.; Chaloner, P. A. J . Am. Chem. Soc.

1980, 102, 3040. (c) Brown, J . M.; Chaloner, P. A. Tetrahedron Lett.

1978, 1877. See, however, ref 4c.

(29) For other related systems, see: (a) Michalik, M.; Freier, T.;

Schwarze, M.; Selke, R. Magn. Reson. Chem. 1995, 33, 835. (b) Brown,

J . M.; Chaloner, P. A. J . Am. Chem. Soc. 1978, 100, 4307. (c) Berger,

H.; Nesper, R.; Pregosin, P. S.; Ru¨egger, H.; Wo¨rle, M. Helv. Chim.

Acta 1993, 76, 1520.

rich ligands Rh⁺[7a][COD]SbF₆^-and Rh⁺[7b][COD] SbF₆

-

have ³¹P shifts of δ 133.4 (d, J ) 178 Hz) and 133.3 (d, J

) 179 Hz). The complexes of electron-poor ligands, on

the other hand, have the following chemical shifts: 7d ,

δ 123.4 (d, J ) 183 Hz); 7e, δ 124.8 (d, J ) 178 Hz); 7f,

δ 104.96 (d, J ) 201 Hz). The chemical shifts and J _Rh-P

coupling constants for Rh⁺[7x][CD₃OD]₂SbF₆are as

-

follows: 7a , δ 143.90 (J ) 230 Hz); 7b, δ 144.60 (J )

230 Hz); 7d , δ 136.99 (J ) 230 Hz); 7e, δ 132.69 (J )

194 Hz). Phosphorus chemical shifts are dependent on

both steric as well as electronic effects and the back-

bonding properties of the metal.³⁰At this stage, no

(30) (a) Tolman, C. A. Chem. Rev. 1977, 77, 313. (b) White, D.;

Coville, N. J . Adv. Organomet. Chem. 1994, 36, 95.

Electronic Effects in Asymmetric Catalysis

J . Org. Chem., Vol. 64, No. 10, 1999 3435

discernible correlation between enantioselectivity or even

reactivity can be drawn³¹except to suggest that the

precatalyst Rh(I) complexes that produce higher enan-

tioslectivity have ³¹P chemical shifts in the more deshield-

ed region in both the COD and bis-CD₃OD complexes.

Significantly, in all cases the complex seems to maintain

the C₂symmetry in solution.

Cr ysta l Str u ctu r es of th e P r eca ta lysts w ith Elec-

tr on ica lly Differ en t Dia r ylp h osp h in ites. A number

of studies aimed at clarifying the origin of enantioselec-

tivity in Rh-catalyzed hydrogenation of acetamidoacrylic

acid^4,9,12and itaconic acid derivatives^11,13,20have been

carried out. Details of kinetics and thermodymamic

aspects of key steps under a variety of reaction conditions

are well-known. In the case of five- and seven-membered

Rh-phosphine chelates, there exists a good empirical

correlation between the λ- and δ-conformations of the

starting Rh-catalyst and the sense of optical induction

in the resulting amino acid, the λ-conformer of the

precatalyst leading to the S-isomer and the δ-conformer

to the R-isomer of the amino acid (Figure 4).²⁷These

conformations of the precatalyst, in the case of ideal C₂-

symmetric ligands, present an asymmetric, alternating

pseudoaxial and pseudoequatorial array of P-aryl groups

around the metal, defining a chiral space. This leads to

clockwise or anticlockwise skewing of the substrate

providing kinetically and/or thermodynamically different

diastereomeric intermediates, which ultimately results

in enantioselective hydrogenation. The effect of such an

asymmetric disposition of a coordinated ligand can be

seen even in coordinated NBD or COD.^27c,32,33The forces

that are responsible for the skewing of the dienes are

thought to be important in the diastereoselective recogni-

tion and the subsequent enantioselective hydrogenation

of prochiral olefins.³³While the empirical relationship of

the λ- vs δ-conformation of the precatalyst seems to hold

good for five-membered chelates, it is more tenuous for

seven-membered complexes. Because of the conforma-

tional mobility of seven-membered rings, the λ- and

δ-conformations are sometimes poorly defined. Even for

some five-membered Rh-chelates, several conformations

are possible in the solid state, especially if different

polymorphs are present. For example, Oliver and Riley^27c

have described one case where the Rh⁺-complex of (R)-

1,2-bis(diphenylphosphino)-1-cyclohexylethane ((R)-cyc-

phos) exists in two different polymorphic forms and one

of the polymorphs has two conformations for the complex

(Figure 5). Nonetheless, all three arrangements of the

precatalyst can be approximately defined as belonging

to the λ-conformation shown in Figure 4 (ω and ω′, as

defined in Figure 7, are -6.7/-7.7, -22/-3, and -14.2/-

11.2, respectively, for these cationic structures). Accord-

ingly, very high enantioselectivity for the formation of

the (S)-isomer is observed in the asymmetric hydrogena-

tion of acetamidoacrylic acid derivatives. As these au-

thors have correctly pointed out, these results clearly

show that the coordinated ligands need not form a rigid

complex. There is often considerable flexibility in the

conformations of the ligand backbone, and this could have

important implications for enantioselectivity.

-

F igu r e 5. Cations of Rh⁺[(R)-cycphos][NBD]ClO₄from dif-

ferent polymorphs.^27cAll structures are roughly of the λ-con-

formation shown in Figure 4. NBD and the anions are omitted

for clarity.

In light of the above discussion, it is imperative that

any discussion of electronic effects of phosphorus ligands

should first address the effect of electronic changes on

the ground-state conformations of the catalyst. For

example, could one trace the difference in the behavior

of electron-rich and electron-poor phosphinites to a

difference in the ground-state conformations of the

respective Rh chelates? To explore this possibility, we

conducted a crystallographic analysis of a number of Rh-

phosphinite complexes. After repeated attempts we were

able to grow good-quality crystals of compounds 11, 12,

15, and 16 (Figure 2). The structures of these complexes

were solved by standard techniques.³⁴While our study

was in progress, Selke et al. reported³⁵the structure of

(31) In an NMR study of the diastereomeric Rh-complexes of

acetamidoacrylates Philipsborn also came to the same conclusions. See

ref 12 and the corresponding Supplementary Material.

(32) Armstrong, S. K.; Brown, J . M.; Burk, M. J . Tetrahedron Lett.

1993, 34, 879.

(33) Kyba, E. P.; Davis, R. E.; J uri, P. N.; Shirley, K. R. Inorg. Chem.

1981, 20, 3616.

(34) See the Supporting Information for details. The authors have

deposited atomic coordinates for the structures of 11, 12, 15, and 16

with the Cambridge Crystallographic Data Centre. The coordinates

can be obtained, upon request, from the Director, CCDC, 12 Union

Road, Cambridge CB 1EZ, U.K.

(35) Kempe, R.; Schwarze, M.; Selke, R. Zeitschrift. Kristal. 1995,

210, 555.

3436 J . Org. Chem., Vol. 64, No. 10, 1999

RajanBabu et al.

conformity with the generalizations,²⁷gives (S) amino

acids in typical hydrogenation experiments (Table 4,

under 7a and 7b).

Substituting the auxiliary 1,4-cyclooctadiene ligand in

the complex 11 by norbornadiene brings about substan-

tial changes in the solid-state structure of the resulting

complex (Figure 8). The crystal has two distinct confor-

mations for the cationic Rh complex. Cation 1, 12(Rh ₁),

has a λ-twist-chair conformation, roughly comparable to

the COD complex 11, while cation 2, 12(Rh ₂), has a twist-

boat conformation difficult to define by the λ/δ convention.

The λ-twist-chair conformation for a metal complex of

the trans-(1S,2S)-1,2-di-O-diphenylphosphinocyclohexane

appears to be general. For example, an almost perfectly

C₂-symmetric complex 16 is formed upon reaction of the

ligand with NiBr₂‚DME complex. As judged from the

structural parameters R, â, γ, and ω (Tables 6 and 7,

entries 1 and 8), the Rh(I)-complex 11 and the Ni(II)-

complex 16 (Figure 9) have identical conformations for

the seven-membered ring.

The twist-boat conformation 12(Rh ₂) of the L₂Rh⁺-

(NBD) complex also appears in catalytically relevant

complexes 13 and 14 (Figure 10). A careful inspection of

the geometric parameters associated with structures of

13 and 14 (Tables 6 and 7) reveal that the values of R, â,

δ, and γ and the positioning of the P-aryl groups in both

of these complexes are comparable. Both of these com-

plexes derived from C₁-symmetric ligands prepared from

D-glucose and are known to give very high enantioselec-

tivities for the production of S-phenylalanine methyl

ester from the corresponding dehydroamino acids. It is

interesting to note that the alignment of the sugar

backbone is clearly different in each case. To maintain a

nearly identical seven-membered chelate conformation

for 13 and 14, C₁of the sugar is kept to the right of the

viewer in the former, whereas C₄is to the right of the

viewer in 14. Since both of these complexes give excellent

enantioselectivity³⁶(95% and 90% ee’s respectively), one

may reasonably conclude that the chirality of the carbons

on the sugar frame to which the phosphinoxy groups are

attached, rather than the rest of the sugar backbone, is

the more important control element, an idea that we have

since used in the synthesis of both (S)- and (R)-amino

acids from ^D-glucose.⁶Yet another aspect of the structure

of the 3,5-dimethylphosphinite is the possible hindered

rotation around the two P-aryl groups on the â-face (Ar₁

and Ar₂in 14). An examination of the model suggests

that the m-methyl groups from these two aryl groups

could approach within van der Waals radii in certain

rotamers. No two other aryl groups have this problem.

Pregosin recently compiled a number of asymmetric

reactions where the 3,5-dimethylphenyl motif in the

metal catalysts have been shown to have a beneficial

effect for high enantioselectivity.³⁷It would be interesting

to explore what role, if any, restricted rotation plays in

these instances.

F igu r e 6. Definitions of angles and distances used in Table

6. (a) R and â represent lower of the two angles between the

plane defined by P₁, Rh, and P₂and the P-Ar bond. The front

(f)/rear (r) designations refer to the positioning of the P-aryl

groups with respect to a plane perpendicular to the P₁-Rh-

P₂plane that passes through P₁and P₂. (b) Angles δ and γ

are formed between the P-Ar bond and the line connecting

P₁and P₂.

F igu r e 7. Definition of ω and ω′. The dihedral angle between

O-P₁and Rh-P₂is ω, and the corresponding angle between

O-P₂and Rh-P₁is ω′. When viewed from the front, if the

smallest angle rotation from P₂-Rh to P₁-O is clockwise ω is

positive and if it is anticlockwise ω is negative. For the λ

conformation both ω and ω′ will be negative and for the

δ-conformation both will be positive.

a related complex, 13. The structure of 14 was not fully

solved because of major crystal disorders; nonetheless,

the parameters we were able to extract enabled us to

compare conformational features of this complex with

others in the series.

Chem3D versions (created from a PDB format) of the

crystal structures of 11-16 are shown in Figures 8-13.

Selected interatomic distances and intramolecular angles

are shown in Tables 6 and 7. In reporting these data, we

have used the conventions used by Pavlov^27fand

Seebach^27g(Figures 6 and 7).

Complexes 11-14 form discrete Rh⁺chelates in the

solid state, with no interaction showing between the

counteranion and Rh. As expected, all are nearly square

planar with P₁-Rh-P₂near 90° in all cases.

As shown in Figure 8, the conformation of the cationic

Rh⁺-complex of trans-(1S,2S)-1,2-di-O-diphenylphosphi-

nocyclohexane depends on the ancillary ligand present.

With 1,5-cyclooctadiene as the ancillary ligand the com-

plex exists in an idealized λ-twist-chair conformation with

the endocylic ω and ω′ angles -19°, -36° and -22°, -30°,

respectively. Note that the complex 11 shows only a

single ³¹P peak in the NMR, indicating its near-C₂-

symmetric nature. In the solid-state structure of 11 the

COD ligand shows a clockwise skew rotation with respect

to the P₁-Rh-P₂plane.^32,33The Rh-complex 11, in

Finally, the solid-state structure of a precatalyst from

1,2-bis-3,5-bis-trifluoromethylphenylphosphinite (15) is

shown in Figure 11. We were unable to grow good quality

crystals of the corresponding COD or NBD complexes.

Conditions under which the COD and NBD complexes

11, 12, and 14 were formed (addition of stoichiometric

amount of a CH₂Cl₂solution of the ligand to (COD)₂Rh⁺

(37) Trabesinger, G.; Albinati, A.; Feiken, N.; Kunz, R. W.; Pregosin,

P. S.; Tschoerner, M. J . Am. Chem. Soc. 1997, 119, 6315.

(36) Selke, R. J . Organomet. Chem. 1989, 370, 249.

Electronic Effects in Asymmetric Catalysis

J . Org. Chem., Vol. 64, No. 10, 1999 3437

F igu r e 8. Crystal structures of 11 and 12 (Figure 2). 11 and 11′ are two perspectives. 12(Rh ₁) and 12(Rh ₂) are two distinct

conformations in the crystal.

with those of 13 and 14 is especially revealing. For 13,

14, and 15 these values (Table 6, entries 4, 5, 6, and 7;

Table 7, entries 4, 5, 6, and 7) are as follows:

13: 2 (f), 56 (r), 88 (-), 26 (-), 11, 63, 51, 17, 63, -27

14: 8 (f), 70 (r), 71 (-), 52 (-), 11, 57, 40, 29, 53, -3

15: 2 (f), 63 (r), 87 (-), 30 (-), 12, 58, 44, 21, 59, -24

Recall that 13 and 14, which incorporate relatively

electron-rich P-aryl groups, are excellent precatalysts for

hydrogenation of dehydroamino acids. In sharp contrast,

Rh complex 15 with its eletron-deficient CF₃groups is a

very poor precatalyst. Yet they are structurally similar

in the ground state. Therefore, to a first approximation,

it is reasonable to conclude that the electronic enhance-

ment of enantioselectivity is not due to changes in the

ground-state conformations.

Do Electr on ic Effects of Liga n d s Alter th e Ra tio

of Ma jor to Min or Dia ster eom er s? Halpern’s classic

studies^4aon hydrogenation have shown that two factors

F igu r e 9. Crystal structure of 16 (Figure 2).

that are most important in asymmetric induction in Rh-

catalyzed hydrogenation are the ratio of the initially

formed diastereomers (Figure 1) and the reactivity

(oxidative addition of hydrogen, which converts these Rh-

or (NBD)₂Rh⁺) produced only the chloride-bridged dimer

15. Highly electron-deficient Rh abstracts a chlorine from

the solvent to produce the bridged dimer. We were able

to prepare the cationic complex in nonchlorinated sol-

vents (see Table 2), but this complex could not be

crystallized. The structure of 15 consists of two almost

identical RhP₂fragments as shown in Figure 11. A close

examination of the seven-membered chelate (15B) shows

that the ring has approximately the same twist-boat

conformation of 12(Rh ₂), 13, and 14. Comparison of the

structural parameters R, â, R′, â′, γ, δ, γ′, δ′, ω, and ω′

(I) complexes into octahedral Rh(III) intermediates) of

these intermediates. As we have reasonably concluded

in the previous section, if the electronic effects do not

alter the ground-state conformations of the precatalysts,

the next logical question is how electronic variations in

a ligand affect these two factors. Here we describe our

attempts to provide partial answers to these questions.

We decided to examine the ratio of the two diastereomers

as a function of electronic effects of the ligand in the

3438 J . Org. Chem., Vol. 64, No. 10, 1999

RajanBabu et al.

F igu r e 10. Crystal structures of complexes 13 (from Selke et al.³⁵) and 14. COD, counterion, and the H’s are omitted for clarity.

F igu r e 11. Crystal structure of 15 (Figure 2). The conformation of both halves of the dimer are approximately the same. B

shows the details of the seven-membered chelate.

hydrogenation of dimethyl itaconate (eq 3). The itaconate

system, rather than the R-acetamidocinnamate, was

chosen for this study since the two reactions are mecha-

nistically similar, yet the Rh(I)-substrate complexes in

the former are easier to observe by ³¹P NMR spectros-

copy.¹³

itaconate (Scheme 2). The ³¹P NMR spectra were recorded

at various temperatures between 303 and 208 K. These

are shown in Figures 12-15.

The unsubstituted bis-diphenylphosphinite 7b at 303

K shows a clean doublet (δ 144.6, J _Rh-P) 230 Hz)

characteristic of [CD₃OD]₂Rh⁺[7b] except for a very broad

small peak at δ 125.5-128.5.^13cJ udging from the chemi-

cal shift of this minor peak and changes it undergoes

upon cooling (vide infra), it is conceivable that this signal

is due to all the phosphorus atoms from the diasteromeric

complexes (Figure 3) that are under rapid equilibrium.

We prepared the [dimethyl itaconate]Rh⁺[phosphinite]

complexes with ligands 7a , 7b, 7d , and 7e (Scheme 1)

by removal of the COD ligand by hydrogenation of the

appropriate precatalyst from [COD]Rh⁺[phosphinite] in

CD₃OD followed by addition of an excess of dimethyl

Electronic Effects in Asymmetric Catalysis

J . Org. Chem., Vol. 64, No. 10, 1999 3439

F igu r e 12. Temperature-dependent ³¹P NMR spectra of diastereomers of [dimethyl itaconate]Rh⁺[7b].

Upon cooling, in accordance with the observations of

Brown^13band Selke,^13csignals due to the two distinct

phosphorusatomsofthediastereomersof[dimethylitaconate]-

Rh⁺[7b] begin to appear (Figure 12b-e). At 253 K (Figure

12b), two broad doublets of multiplets (of equal intensity)

are seen at δ 108.5-111.5 (J _Rh-P) 170 Hz) and 137.5-

140.5 (J _Rh-P) 208 Hz). The broadness of these peaks and

the absence of discernible P-P coupling at this temper-

ature is probably due to the fast intermolecular exchange

between the major and minor diastereomers (Figure 3).^13c

Upon cooling to 238 K (Figure 12c), the high-field signals

become sharp (δ 109.9, J _Rh-P) 169 Hz, J _PP) 49 Hz)

while the low-field signals still remain as a doublet of

multiplets (δ 137.0-141.0, J _Rh-P) 177). The ratio of

intensities of the signals due to [CD₃OD]₂Rh⁺[7b] and

those of the itaconate adducts at this temperature is

approximately 2:1. At 223 K (Figure 12d), the proportion

of the itaconate adducts increases at the expense of the

3440 J . Org. Chem., Vol. 64, No. 10, 1999

RajanBabu et al.

F igu r e 13. Temperature-dependent ³¹P NMR spectra of diastereomers of [dimethyl itaconate]Rh⁺[7a ].

solvate complex (the ratio of the solvate/itaconate adducts

1:2), and the low-field phosphorus signals begin to further

broaden and resolve (δ 136.0-142.0, with a doublet of

multiplets centered around δ 140 (J _Rh-P) 164 Hz)). The

high-field signal around δ 110, possibly the phosphorus

trans to the olefin,^4cbroadens, indicating that indeed this

could arise from a major and a minor peak. At 208 K

(Figure 12e), only 20% of the ³¹P intensity is due to the

[CD₃OD]₂Rh⁺[7b]. As observed by Selke et al.,^13cthe high-

field phosphorus signal shows no change except for

further line broadening. At this temperature, the signals

between δ 137 and 142 are fairly well resolved into two

doublets of doublets, presumably arising from the major

and minor diastereomers of the [itaconate]Rh⁺[7b ]

complex,^13cand we can now estimate the ratio of the

major to minor diastereomer as 2.6:1. Finally, upon

warming the solution to 303 K, the original ratio of the

Rh-complexes is restored (Figure 12f).

The [itaconate]Rh⁺[7a ] complex was prepared in the

same fashion, and the temperature-dependent ³¹P NMR

Electronic Effects in Asymmetric Catalysis

J . Org. Chem., Vol. 64, No. 10, 1999 3441

Ta ble 6. Selected In ter a tom ic Dista n ces (Å) a n d An gles (d eg) for P h osp h in ite Com p elxes^a

entry

complex

R

â

R′

â′

γ

δ

γ′

δ′

P-P

P₁-Rh-P₂

Rh-P₁

Rh-P₂

1

2

3

4

5

6

7

8

11

75 (r)

52 (-)

8.3 (f)

2 (f)

8 (f)

2 (f)

47 (f)

74 (-)

72 (r)

56 (r)

70 (r)

63 (r)

62 (r)

45 (f)

90 (r)

72 (r)

64 (-)

88 (-)

71 (-)

87 (-)

88 (-)

85 (r)

31 (f)

11 (f)

41

29

15

11

12

11

44

29

41

58

63

57

58

60

27

48

54

37

51

40

44

47

46

21

17

31

17

29

21

20

22

3.179

3.243

3.184

3.150

3.171

3.017

3.025

3.165

90.2

91.8

90.2

88.2

89.8

88.6

88.8

95.2

2.239

2.247

2.238

2.250

2.207

2.161

2.156

2.133

2.250

2.270

2.258

2.276

2.283

2.157

2.165

2.153

12 (Rh₁)

12 (Rh₂)

13^b

54 (-)

26 (-)

52 (-)

30 (-)

26 (-)

37 (f)

14

15 (Rh₁)

15 (Rh₂)

16^c

1 (f)

79 (r)

a

b

See Figures 6 and 7 for description of angles and distances: (f) front, (r) rear, (-) position uncertain. From Selke.^{35 c}NiBr₂complex.

Ta ble 7. Dih ed r a l An gles^aω a n d ω′ in Rh P h osp h in ite

of the catalytically relevant minor diastereomer at equi-

librium. This could at least in part explain the pro-

Com p lexes

entry

complex

ω

ω′

nounced enantioselectivity with this ligand.

Further verification of this hypothesis was sought with

a temperature-dependent ³¹P NMR study of the diaster-

eomers of [dimethylitaconate]₂Rh⁺[7d ] and [dimethylita-

conate]₂Rh⁺[7e], each with an electron-deficient ligand.

The results for the 3,5-difluorophenylphosphinite are

shown in Figure 14. The ³¹P signal for [CD₃OD]₂Rh⁺[7d ]

appears at δ 137.0 (J _Rh-P) 230 Hz). In sharp contrast

to the more electron-rich phosphinite, 7a , the electron-

deficient 7d appears to induce coordination of the ita-

conate to Rh even at 303 K. A very broad peak at δ 111-

114 was assigned to fast equilibrating mixtures of the

itaconate complexes based on the effect of temperature

on this signal. Upon cooling first to 253 K, and subse-

quently to 238 and 223 K (Figure 14b-d), this broad peak

is replaced by four sets of peaks at δ 96-101, 105-108,

112-115, and 120-127. Unfortunately, assignment of

these peaks to the major and minor diastereomers has

not been possible because of the complicated coupling

patterns (some of them undoubtedly due to the long-

range P-F couplings). Another complication is the ap-

pearance of peaks at δ 134-138 under the signals due

to the solvate complex. Even though the changes in the

spectrum are intractable, the reversible changes are

highly reproducible, as shown by the last spectrum

(Figure 14f), obtained upon warming the solution to 303

K. No reliable quantitative comparison to the 3,5-di-

methylphosphinite system ([dimethylitaconate]₂Rh⁺[7a ])

is possible.

1

2

3

4

5

6

7

8

11

-19

55

7

63

53

59

56

-22

-36

1

-57

-27

-3

-24

-29

-30

12 (Rh₁)

12 (Rh₂)

13^b

14

15 (Rh₁)

15 (Rh₂)

16^c

See Figure 7 for definitions (after Pavlov et al.^27f). From

a

b

Selke.³⁵^cNiBr₂complex.

Sch em e 2. Syn th esis of

[P h osp h in ite]Rh (I)-Ita con a te Com p lexes

spectra obtained are shown in Figure 13. At 303 K, only

signals due to [CD₃OD]₂Rh⁺[7a ] that appear as a doublet

at δ 143.9 (J _Rh-P) 230 Hz) are visible. Cooling the

methanolic solution to 253 K results in the formation of

three sets of new peaks. The highest field signal is a

broad doublet of multiplets reminiscent of the high-field

signal from the complex [itaconate]Rh⁺[7b]. At 238 K,

the spectrum is well resolved (Figure 13c), and the major

and minor diastereomers can be easily identified. The

major diastereomer has a doublet of doublets at δ 111.9

(J _Rh-P) 168 Hz, J _PP) 47 Hz) and another poorly

resolved doublet of doublets at δ 139.0 (J _Rh-P) 166 Hz),

whereas the minor diastereomer has two doublet of

doublets at δ 129.1 (J _Rh-P) 157 Hz, J _PP) 35 Hz) and

140.9 (J _{J Rh-P}) 157 Hz, J _PP) 35 Hz). For this ligand,

the ratio of the major to minor diastereomer at this

temperature as determined from the respective ³¹P

signals is approximately 1:1. Further cooling results in

the appearance of even more peaks (Figure 13c-e), and

one can only speculate the origin of these peaks. As

pointed out earlier, the 3,5-dimethylphenylphosphinite

has the possibility of some restricted rotation around the

P-aryl bond, and these rotomers could be responsible for

the new peaks. At 208 K, where one can reliably estimate

the major/minor ratio of diastereomers for the [itaconate]-

Rh⁺[7b] complex () 2.6:1), the corresponding ratio for

the [itaconate]Rh⁺[7a ] complex is at least 1:1 (Figure

13e). In any case, there is a clear difference in the relative

proportion of the major and minor diastereomers of the

P ₂Rh-itaconate complex, as a function of the P-aryl

substitituent of the ligand. For the 3,5-dimethylphen-

ylphosphinite, 7a , which gave the highest ee, there is more

Finally, the itaconate complexes with the most electron-

deficient ligands, viz., the 1,2-(bis-bis-(3,5-bis-trifluoro-

methylphenyl)phosphinoxy)cyclohexane (7e) were stud-

ied. Recall that this ligand gave essentially 0% ee in the

hydrogenation. The most striking feature of the spectrum

at 303 K (Figure 15a) is the large proportion of the

dimethyl itaconate complex present in solution vis-a´-vis

the spectra when the ligands 7a and 7b (or even 7d ) are

employed. The solvate complex [CD₃OD]₂Rh⁺[7e] appears

at δ 132.7 (J _Rh-P) 194 Hz), and the broad doublet of

multiplets due to the fast-exchanging mixture of itacon-

ate complexes appears at δ 122-126 (Figure 15a).

Further cooling of this mixture to 253 K, and subse-

quently to 238 K, results in what appears to be a

symmetrical movement of the ³¹P signals to either side

of the broad peak seen at room temperature. Even at 208

K, the spectrum is poorly resolved, and the intermediate

complexes appear to be in fast equilibrium. In sharp

contrast to the electron-rich systems, no discernible P-P

or even Rh-P couplings can be extracted from these

spectra. Another significant difference between the elec-

tron-rich and electron-poor systems is the relatively large

3442 J . Org. Chem., Vol. 64, No. 10, 1999

RajanBabu et al.

F igu r e 14. Temperature-dependent ³¹P NMR spectra of diastereomers of [dimethyl itaconate]Rh⁺[7d ].

proportion of the unreacted solvate complex [CD₃OD]₂Rh⁺-

[7x] still present in the case of the latter, even at 208 K

(compare Figures 12e and 13e with Figures 14e and 15e).

The relevance of these observations to the selectivity

must await further studies.

ligands are affected by substituents on the P-aryl groups.

The electron-rich 3,5-dimethylphenyl substituent gives

the highest ee, whereas the corresponding 3,5-bis-tri-

fluoromethylphenyl derivative gives the lowest. The same

substituent effects are seen with 2,3-diarylphosphinites

derived from phenyl-â-^D-glucopyranoside. In this work,

we have tried to clarify these remarkable effects at two

levels. First, we obtained crystal structures of a number

of relevant precatalysts ([phosphinite]₂Rh⁺[diolefin]X^-)

and studied in detail their ground-state conformations.

These conformations were compared among themselves

Con clu sion s

Enantioselectivities of the Rh(I)-catalyzed hydrogena-

tions of dimethyl itaconate and methyl R-N-acetylcin-

namate using (1,2-diarylphosphinoxy)cyclohexane as

Electronic Effects in Asymmetric Catalysis

J . Org. Chem., Vol. 64, No. 10, 1999 3443

F igu r e 15. Temperature-dependent ³¹P NMR spectra of diastereomers of [dimethyl itaconate]Rh⁺[7e].

and with those of other related structures with the goal

of testing whether electronic effects alone can alter the

ground-state conformation of a precatalyst. The impor-

tance of such a study is underscored by the highly

predictable empirical correlations that have evolved,

which relate the ground-state conformations of the pre-

catalyst ([phosphine]Rh⁺[diolefin]) to the sense of asym-

metric induction. In the event, a study of six complexes

with different phosphinites reveals that the gross con-

formational features of the precatalysts are largely

unaffected by electronic effects, which means that one

has to look elsewhere (to the stability and/or reactivity

of other intermediates along the reaction path) for an

explanation of the electronic amplification. In the ita-

conate reduction,where the diastereomers ofthe [itaconate]-

Rh⁺[phosphinite] complex are detectable by ³¹P NMR,

there is a clear difference in the ratio of the major to

minor diastereomer when the 3,5-dimethylphenylphos-

phinite (7a ) is employed as the ligand, as compared to

when the unsubstituted diphenylphosphinite 7b is used.

3444 J . Org. Chem., Vol. 64, No. 10, 1999

RajanBabu et al.

repeated in acetone and the spectrum was recorded in CD₂-

Cl₂: ³¹P NMR (CD₂Cl₂) 133.3 (d, J ) 179 Hz). The other peak

δ 122.4 (dd J ) 173, 6 Hz) accounted for <9% area.

A sample of this compound was recrystallized from benzene

by slow evaporation, and X-ray structure analysis was done

to confirm the structure (see ref 6b for details).

We have not been able to establish the diastereomer ratio

in the more electron-deficient 3,5-difluorophenyl (7d ) and

3,5-bis-trifluoromethylphenyl (7e) systems. In the case

of 7d , the isomers could not be reliably identified. With

7e, even at the coldest temperature (208 K) the isomers

appear to be in fast equilibrium. Further insight into the

diasteromer equilibria may be gained by looking at

(³¹P,³¹P)-{¹H]}-2D EXSY NMR of the exchanging system.^4c

We plan to pursue these studies in the future.

[7b]Rh ⁺(NBD)SbF ₆^-. The title compound was prepared by

the above procedure, except that Rh⁺(norbornadiene)₂SbF₆

-

was substituted for Rh(COD)₂⁺SbF₆

.

X-r a y Cr ysta llogr a p h y. Crystal data: SbRhP₂F₆O₂C₃₇H₃₈

,

from benzene, red-orange, irregular block, ∼0.40 × 0.30 × 0.37

mm, monoclinic, C2 (No. 5), a ) 38.889(4) Å, b ) 10.617(2) Å,

c ) 19.134(4) Å, â ) 114.27(1)°, from 25 reflections, T ) -70

°C, V ) 7201.9 Å³, Z ) 8, FW ) 915.32, D_c) 1.688 g/cm³,

Exp er im en ta l Section

µ(Mo) ) 13.55 cm^-1

.

Gen er a l Meth od s. Manipulations of air- and moisture-

sensitive materials were conducted in a nitrogen atmosphere

by using either Schlenk techniques or a Vacuum Atmospheres

drybox. Tetrahydrofuran (THF), ether, dimethoxyethane (DME),

hexane, and pentane were distilled from sodium benzophenone

ketyl under nitrogen. Benzene and toluene were distilled from

lithium aluminum hydride and stored over activated 4 Å

molecular sieves. Methylene chloride, dimethylformamide

(DMF), and acetonitrile were distilled from calcium hydride

and stored over activated 3 Å, molecular sieves. Flash column

chromatography was carried out on 230-400 mesh (0.040-

0.063 mm) silica (EM Reagents). Gas chromatographic (GC)

analysis was performed with a Hewlett-Packard (HP) model

5890 GC fitted with a HP3396A integrator. Conversions were

determined using an HP cross-linked methyl silicone capillary

column (30 m × 0.530 mm). GC separations of the amino acid

ester enantiomers were accomplished using Chirasil ^L-Val on

WCOT fused silica (25 m × 0.25 mm, 0.12 mm film thickness)

capillary GC column purchased from Chrompack (1130 Route

202 South, Raritan, NJ 08869). The succinate products were

separated on a Chiraldex GTA capillary column (10 m × 0.25

mm) purchased from Bodman (P.O. Box 2421, Aston, PA

19014). HPLC separations were carried out on Chiralcel OJ

or Chiralcel OB HPLC columns (25 cm × 4.6 mm) manufac-

tured by Daicel (available from Chiral Technologies, Inc., 730

Springdale Drive, Drawer 1, Exton, PA 19341). Preparation

of the sugar-derived ligands, Rh-complexes, and the acetami-

docinnamate substrates have been described before.^6b

Da ta collection : Enraf-Nonius CAD4 diffractometer, Mo

KR radiation, 7856 data collected, 2.3° e 2θ e 52.0°, maximum

h, k, l ) 47, 13, 23, data octants ) +++, -++, ω scan method,

scan width ) 1.20-1.50°ω, scan speed ) 1.50-6.70°/min,

typical half-height peak width ) 0.13° ω, two standards

collected 32 times, adjusted for a 6% decrease in intensity,

9.1% variation in azimuthal scan, no absorption correction,

226 duplicates, 1.7% R-merge, 5278 unique reflections with I

g 3.0σ(I).

Solu tion a n d Refin em en t. Solved by direct methods

(SHELX). The asymmetric unit consists of two ion pairs in

general positions related by a pseudomirror at (0, 0.47, -0.50).

This results in near C2/c space-group symmetry and high

correlations in the refinement. Hydrogen atoms were included

as fixed atoms in idealized positions. Both SbF₆anions show

high thermal motion or disorder. The acentric space group

indicates an optically pure enantiomer, here chosen to be C₁₁

/

S, C₁₂/S based on the lowest R value for anomalous terms (the

centric positioning of the heavy atoms and the oscillation in

the refinement limits the sensitivity of the anomalous disper-

sion test), refinement by full-matrix least-squares on F,

scattering factors from International Tables for X-ray Cystal-

lography, Vol IV, including anomalous terms for Sb, R,

biweight [σ²(1) + 0.0009(1)²]^-1′2(excluded 1), refined aniso-

tropic: Sb, Rh, P, F, isotropic: F, C, fixed atoms: H; 846

parameters, data/parameter ratio ) 6.24, R ) 0.049, W )

0.048, error of fit ) 1.75, max ∆/σ ) 1.32. The disordered

fluorine atoms of one SbF₆were modeled with two-half-

weighted orientations, while the second SbF₆was modeled

with normal anisotropic thermal tensors. Due to high correla-

tions, several atoms could not be refined anisotropically (Table

1 in the Supporting Information). Refinement in C2/c (with

Liga n d s a n d Ca ta lysts fr om (1S,2S)-tr a n s-1,2-Cyclo-

h exa n ed iol. (1S,2S)-bis-Dip h en ylp h osp h in oxycycloh ex-

a n e (7b). This experiment was done inside the drybox. To 200

mg (1.72 mmol) of azeotropically dried(1S,2S)-trans-1,2-cyclo-

hexanediol and 11 mg (0.05 equiv) of 4-(dimethylamino)-

pyridine dissolved in 2 mL of CH₂Cl₂and 1 mL of pyridine

was added, at 0 °C, 0.836 g (3.78 mmol) of chlorodiphenylphos-

phine in 3 mL of methylene chloride. An additional 1 mL of

methylene chloride was used to facilitate transfer of the

chlorophosphine quantitatively. The mixture was warmed to

room temperature and was stirred for 18 h. The precipitated

solids were filtered off with the aid of a cotton-plugged

disposable pipet. The NMR spectrum of this solid showed that

it was pyridine hydrochloride. The vials were washed with 2

mL of 1:1 ether/hexanes, and the solid in the disposable pipet

was washed with this solvent. The filterate was concentrated

to get a white solid. The product was recrystallized from

benzene (95% yield): ¹H NMR (C₆D₆) 0.70-0.95 (m, 2 H),

1.10-1.45 (m, 4 H), 1.90-2.05 (m, 2 H), 4.05 (m, 2 H), 6.70-

7.15 (m, aromatic), 7.45-7.70 m, aromatic); ³¹P NMR 108.7.

omission of ∼150 reflections, resulted in R ) 0.082, R_w

)

0.081), largest residual density ) 2.52 e/Å³. There are four

intense residual peaks near two phenyl rings and midway

between the PO₂C₂bridge on each cation. These may result

from refinement correlations or crystal imperfections.

-

Analysis confirms the structure of the complex as the SbF₆

salt. Although the bidentate ligand is enantiomeric, the

relative orientation of the phosphine phenyl rings results in

two coordination modes creating a pseudoinversion center.

Accurate assessment of bond parameters is limited by the

disorder and the high least-squares correlations (highest

correlation ) 0.89).

Fractional coordinates, isotropic and anisotropic themal

parameters, and complete listing of bond angles and bond

lengths are given in the Supporting Information.

X-r a y Cr ysta llogr a p h y of [7b]Ni(Br )₂. This complex was

prepared by dissolving 0.066 g (0.137 mmol) of NiBr₂‚DME^5b

and 0.042 g (0.137 mmol) of 7b in 3 mL of THF. The mixture

was filtered, and to the filtrate were added DME and then

hexane to induce crystallization.

[7b]Rh ⁺(COD) SbF ₆^-. This complex was prepared by

adding a CH₂Cl₂solution of the bis-phosphinite to a solution

of (COD)₂Rh⁺SbF₆in CH₂Cl₂at room temperature and

-

stirring the solution overnight. The low-boiling components

were removed, and the NMR spectra were recorded in CDCl₃:

¹H NMR δ 0.70-1.15 (m, 4 H), 1.35-1.60 (d, br, 2 H), 1.78 (d,

br, 2 H), 2.10-2.30 (m, 4 H), 2.32-2.60 (m, 4 H), 3.82-4.02

(m, 2 H), 4.42 (m, br, 2 H), 4.90 (s, br, 2 H), 7.00-7.80 (m,

aromatic); ³¹P NMR (CDCl₃) δ 131.0 (d, J _Rh-P) 177). Minor

peaks (together <2%) at δ 122.4 (dd J ) 173, 6). The

experiment was repeated in THF, and the ³¹P spectrum was

recorded in THF-d₈: 129.7 (d, J ) 178 Hz), no minor peaks

were visible under these conditions. The experiment was

Cr ysta l d a ta : Br₂NiP₂O₂C₃₀H₃₀, red, rod, from DME/

hexane, ∼0.50 × 0.10 × 0.05 mm, orthorhombic, P2₁2₁2₁(No.

19), a ) 17.593(1) Å, b ) 34.680(2) Å, c ) 9.565(1) Å, T ) -75

°C, V ) 5835.9 Å³, Z ) 8, FW ) 703.04, D_c) 1.600 g/cm³,

µ(Mo) ) 35.14 cm^-1

.

Da ta collection : Rigaku RU300, R-AXIS image plate area

detector, Mo KR radiation, filament size ) 0.5 mm × 1.0 cm,

anode power ) 55 kV × 200 ma, crystal to plate distance )

Electronic Effects in Asymmetric Catalysis

J . Org. Chem., Vol. 64, No. 10, 1999 3445

79.1 mm, 210 u pixel raster, number of frames ) 45, oscillation

range ) 2.0°/frame, exposure ) 2.0 min/frame, box sum

integration, 11055 data collected, 2.3° e 2θ e 49.7°, maximum

h, k, l ) 20, 40, 11, corrected for absorption (DIFABS), range

of transmission factors ) 0.29-1.05, 2013 duplicates, 4.1%

R-merge, 3583 unique reflections with I g 3.0σ(I).

Solu tion a n d Refin em en t. Structure solved by automated

Patterson analysis (PHASE). The asymmetric unit consists of

two independent molecules in general positions. Hydrogen

atoms were idealized with C-H ) 0.95 Å. Refinement by full-

matrix least squares on F, scattering factors from International

Tables for X-ray Crystallography, Vol IV, including anomalous

terms for Br, Ni, P, biweight [σ²(1) + 0.0009(1)²]^-1/2(excluded

4), refined anisotropic: Br, Ni, P,O; isotropic: C; fixed atoms:

H; 662 parameters, data/parameter ratio ) 5.41, R ) 0.063,

R_w) 0.052, error of fit ) 1.27, max ∆/σ ) 0.02. [The acentric

space group indicates an optically pure enantiomorph. The

absolute configuration was determined from an R-value test

(R ) 8.30 and 7.79 vs 7.23 and 6.64)].

The same experiment done in acetone-d₆gave the following

peaks (inter alia): ³¹P (acetone-d₆) δ 127.67 (d, J ) 218 Hz).

The experiment was repeated as follows in a different

procedure. To a solution of 103 mg (0.10 mmol) of the

diphosphinite 7e in 10 mL of benzene was added chloro(1,5-

cyclooctadiene)rhodium(1) dimer (30 mg, 0.6 mmol), and the

mixture was brought to reflux. To this solution was added 35

mg (0.10 mmol) of silver hexafluoroantimonate, and the

solution was filtered while hot. Concentration and reprecipi-

tation yielded a solid whose ³¹P NMR (CD₂Cl₂) (δ 129.4, d, J

) 217 Hz), and the corresponding minor peaks looked remark-

ably like the one from previous experiment with the major

product as the chloride-bridged dimer.

Rea ction of [7e] w ith Rh ⁺(COD)₂SbF ₆^-. To a solution of

56 mg of Rh⁺(COD)₂SbF₆^-in 1 mL of acetone, was added 103

mg (0.1 mmol) of the bis-phosphinite in 2 mL of acetone and

the mixture was stirred for 3 h. The reaction mixture was

concentrated, and excess hexane was added. The precipitated

salt was collected and analyzed. ³¹P NMR in CD₂Cl₂was

exceptionally clean in ³¹P NMR δ 129.30 (d, J ) 216). The

major compound was identified as the bridged complex pre-

pared by the independent routes described above. A single

crystal obtained from the NMR solution by slow evaporation

was subjected to X-ray analysis, and the Cl-bridged structure

was confirmed.

The analysis confirms the structure of the [7b]NiBr₂com-

plex. Fractional coordinates, isotropic and anisotropic themal

parameters, and a complete listing of bond angles and bond

lengths are given in the Supporting Information.

Syn t h esis of Ot h er P h osp h in it es a n d Th eir R h -

Com p lexes. The following diarylphosphinites and related

complexes were prepared by similar routes:

A minor component (∼6%) whose ³¹P signal appeared at δ

124.8 (J _RhP) 178 Hz) was identified as the cationic complex

[7e]Rh⁺[COD]SbF₆^-by comparison of spectral data, especially

the J _Rh-Pcoupling constant, with those of a complex made in

a choride-free medium (see below).

Repetition of the reaction in CH₂Cl₂gave essentially the

same compound as the major product: ³¹P NMR (CD₂Cl₂) δ

129.4 (d, J _Rh-P) 217 Hz) + other peaks; ³¹P NMR (CDCl₃) δ

119.6 (J _Rh-P) 178 Hz) (cationic complex?); 125.2 (J _Rh-P) 230

Hz) (dimer?).

(1S,2S)-1,2-[Bis-(3,5-d im eth yl)p h en ylp h osp h in oxy]cy-

1

cloh exa n e (7a ): H NMR (C₆D₆) δ 0.90 (m, 2 H), 1.30 (m, 2

H), 1.42 (m, 2 H), 1.85-2.05 (m, 2 H), 1.95 (s, 12 H), 2.00 (s,

12 H), 4.21 (m, 2 H) 6.60 (s, br, 2 H), 6.69 (s, br, 2 H), 7.38 (d,

J ) 8 Hz, 2 H), 7.45 (d, J ) 8, 2H): ³¹P NMR δ 111.5 (s).

-

[7a ]Rh ⁺(COD) SbF ₆

178 Hz).

: )

³¹P NMR(CD₂Cl₂) δ 133.4 (d, J _RhP

(1S,2S)-1,2-[Bis-(3,5-d iflu r o)p h en ylp h osp h in oxy]cyclo-

1

h exa n e (7d ): H NMR (C₆D₆) δ 0.65 (m, 2 H), 0.80-1.20 (m,

4 H), 1.55 (m, 2 H), 3.60 (m, 2 H), 6.25 (m, br, aromatic), 6.75

(m, br, aromatic), 6.85 (m, br, aromatic); ³¹P NMR (C₆D₆) δ

103.9.

The component showing δ 119.6 is relatively insoluble in

benzene, supporting the cationic nature.

When the reaction is carried out in CH₂ClCH₂Cl as the

solvent only the cationic complex was obtained, presumably

because ClCH₂CH₂Cl is a poor Cl^-donor, and when the

spectrum was run in THF-d₈, the following clean spectrum was

observed: ³¹P NMR (THF-d₈) δ 129.17 (d, J _Rh-P) 193 Hz).

It is believed that the highly electrophilic Rh complex

abstracted the Cl from the methylene chloride solvent in the

above experiments.

[7d ]Rh ⁺(COD) OTf^-. This complex was prepared by mixing

an acetone solution of the ligand and Rh⁺(COD)₂OTf^-in

stoichiometric proportions at room temperature: ¹H NMR

inter alia δ 4.80 (s, br, 2 H), 5.05 (m, 2 H), 5.38 (m, 2 H); ³¹P

NMR (CD₃COCD₃) δ 120.7 (d, J _RhP) 183 Hz); ¹⁹F NMR (CD₃-

COCD₃) δ -79.05 (s, 3 F), -107.8 (d, br, 8 F).

[7d ]Rh ⁺(COD)SbF ₆^-. To a solution of 227 mg (0.41 mmol)

of Rh⁺(COD)₂SbF₆^-in 1 mL of acetone was added 270 mg (0.44

mmol) of the bis-phosphinite in 2 mL of acetone at room

temperature. The mixture was stirred for 10 min, and 5 mL

of CH₂Cl₂was added followed by 20 mL of hexane. The

precipitated solid was isolated and identified as the desired

complex from its characteristic NMR spectra: ¹H NMR (CD₂-

Cl₂) δ 1.00-1.45 (m, 4 H), 1.60 (m, 2 H), 1.90 (d, br, J ) 12

Hz, 2 H), 2.30-2.65 (m, 8 H), 4.15 (s, br, 2 H), 4.75 (m, 2 H),

5.08 (m, 2 H), 7.00-7.20 (m, aromatic); ³¹P NMR (CD₂Cl₂) δ

123.4 (d, J _RhP) 183 Hz); (THF-d₈) δ 119.1 (J _RhP) 178 Hz).

The choride-bridged dimer ([7d ]Rh(µ-Cl)/2) was prepared

according to the procedure given below for 7e: (CD₂Cl₂) δ 131.5

(d, J ) 219 Hz).

X-r a y Cr yst a llogr a p h y of Ch lor id e-Br id ged Dim er

([7e]Rh (µ-Cl)/2). Cr ysta l d a ta : Rh₂Cl₃P₄F₄₈O₄C₇₇H₄₇, from

CH₂Cl₂, gold, plate, ∼0.51 × 0.20 × 0.45 mm, monoclinic, P2₁

(No. 4), a ) 14.576(4) Å, b ) 21.770(10) Å, c ) 15.185(4) Å, â

) 108.16(1)°, T ) -70 °C, V ) 4578.5 Å³, Z ) 2, FW ) 2378.29,

D_c) 1.725 g/cm³, µ(Mo) ) 6.52 cm^-1

.

Da ta collection : Enraf-Nonius CAD4 diffractometer, Mo

KR radiation, 10613 data collected, 1.9° e 2θ e 54.0°,

maximum h, k, l ) 18, 27, 19, data octants ) +++, -++, ω

scan method, scan width ) 1.20-2.20° ω, scan speed ) 1.50-

5.00°/min, typical half-height peak width ) 0.14° ω, two

standards collected 49 times, adjusted for a 5% decrease in

intensity (251 omitted), 9.6% variation in azimuthal scan, no

absorption correction, 375 duplicates, 1.8% R-merge, 6855

unique reflections with I g 3.0σ(I).

(1S,2S)-[Bis-1,2-bis-(3,5-bis-tr iflu r om eth ylph en yl)ph os-

p h in oxy]cycloh exa n e (7e): ¹H NMR (C₆D₆) δ 0.50-0.70 (m,

2 H), 0.80-1.10 (m, 4 H), 1.35-1.50 (d, br, 2 H), 3.70 (m, 2

H),), 7.42 (s, 2 H), 7.52 (s, 2 H), 7.75 (d, J ) 7 Hz, 4 H), 7.90

(d, J ) 7 Hz, 4 H); ¹⁹F NMR δ 101.9; ¹H NMR (CD₂Cl₂) δ 1.20-

2.00 (4 sets of m, 8 H), 4.20 (m, 2 H), 7.60 (s, 2 H), 7.70 (d, J

) 5 Hz, 4 H), 7.81 (s, 2 H), 7.87 (d, J ) 7 Hz, 4 H); ³¹P NMR

(CD₂Cl₂) δ 102.3; ³¹P (THF-d₈) δ 99.64). Anal. Calcd for

Solu tion a n d Refin em en t. Structure solved by automated

Patterson anaylsis (PHASE). The asymmetric unit consists of

one dimer and a partially occupied methylene chloride solvate.

Hydrogen atoms were idealized with C-H ) 0.95 Å, refine-

ment by full-matrix least squares on F, scattering factors from

International Tables for X-ray Cyrstallography, Vol. IV, includ-

C

₃₈H₂₂O₂P₂F₂₄: C, 44.38; H, 2.16. Found: C, 44.37; H, 2.53.

ing anomalous terms for Rh, Cl, biweight [σ²(l) + 0.00091²]^-1/2

,

P r ep a r a tion of Ch lor id e-Br id ged Dim er [7e]Rh (µ-Cl)/

(excluded 13), refined anisotropic: Rh, Cl, P, F, O, C, isotro-

pic: C, fixed atoms: H; 1246 parameters, data/parameter ratio

) 5.49; final R ) 0.053, R_w) 0.047, error of fit ) 1.50, max

∆/σ ) 1.32. The multiplicity of the solvate molecule was

adjusted to give a reasonable thermal parameter. No attempt

was made to model the disordered CF₃groups. Atom C47′ was

refined isotropically due to nonpositive thermal matrix, largest

residual density ) 0.70 e/Å³, near Cl3.

2. To a solution of 13 mg (0.02 mmol) of chloro-bis-(cycloocten-

e)rhodium(1) dimer in ∼1 mL of CD₂Cl₂was added 44 mg (0.04

mmol) of the bis-phosphinite in ∼1 mL of CD₂Cl₂, and the

mixture was stirred for 30 min at room temperature. The

solvent and excess cyclooctene were pumped off, and ³¹P NMR

was recorded in CD₂Cl_2.A major doublet, presumably due to

the bridged dimer, appeared at δ 130.17 (d, J ) 218 Hz).

3446 J . Org. Chem., Vol. 64, No. 10, 1999

RajanBabu et al.

The analysis confirms the dimeric structure as a CH₂Cl₂

solvate. It is a neutral dimer with no cyclooctadiene. Fractional

coordinates, isotropic and anisotropic themal parameters, and

a complete listing of bond angles and bond lengths are given

in the Supporting Information.

on a vacuum pump. To the residue was added 1 mL of benzene,

and the precipitated solids were removed by filtration through

a cotton plug. Column chromatography inside the drybox using

10% ether/cyclohexane as the solvent gave pure 7h : ¹H NMR

(C₆D₆) δ 0.70 (m, br, 2 H), 1.00-1.20 (m, br, 4 H), 1.55 (d, br,

J ) 12 Hz, 1 H), 1.75 (d, J ) 12 Hz, 1 H), 3.85 (s, br, 2 H),

6.75-7.10 (m), 7.25 (m), 7.35-7.60 (m), 7.80-7.90 (q, J ) 4

Hz), (6.75-7.90 all aromatic); ³¹P NMR δ 99.5 (s, 1 P), 109.4

(s, 1 P). Two other minor peaks at δ 100.9 and 107.7 (together

4% signal intensity) have been tentatively identified as due

to the two symmetrical phosphinites formed by an exchange

reaction.

P r ep a r a tion of Bis-(p en ta flu or op h en yl)ch lor op h os-

p h in e. Pentafluorobromobenzene (8.20 mL, 66.0 mmol) in dry

THF (15 mL) was added dropwise to freshly crushed magne-

sium turnings (24.31 g, 69.0 mmol) in THF (60 mL). After 2

h, this solution was added slowly to a solution of (ⁱPr₂N)PCl₂

(5.76 g, 30.0 mmol) in THF (20 mL) at room temperature. The

reaction mixture was heated to reflux. After 13 h, the mixture

was cooled to room temperature, and saturated aqueous NH₄-

Cl (80 mL) was added. The aqueous layer was extracted with

hexane (80 mL × 3), and the combined organic extract was

dried (Na₂SO₄) and concentrated in vacuo to give (C₆F₅)₂PN-

(i-Pr) as a colorless oil. Crystallization with methanol gave

white crystals (11.1 g, 79.8%): mp ) 79-82 °C; ¹H NMR

(CDCl₃) δ 3.58 (m, 1H), 1.19 (d, J ) 6.1 Hz, 6 H); ³¹P NMR

(CDCl₃) δ 1.61 (quintet, J ) 29.8 Hz).

Dry HCl was passed through a solution of the aminophos-

phine (2.45 g, 5.29 mmol) in dry cyclohexane (40 mL) at room

temperature for 20 min. After filtration under argon atmo-

sphere and concentration, 1.97 g (93.1%) of (C₆F₅)₂PCl was

collected as a colorless oil: ³¹P (CDCl₃) δ 35.37 (quintet, J )

38.5 Hz) (lit.³⁸δ 37.1 (CH₂Cl₂)).

[(1S,2S)-[1-Diphenylphosphinoxy-2-bis-(3,5-bis-triflurometh-

ylph en yl)ph osph in oxy]cycloh exan e]Rh⁺(COD)SbF₆^-. This

complex was made in acetone by mixing stoichiometric amounts

of the reagents: ³¹P NMR (CD₃COCD₃) δ 125.5 (dd, J _PRh

)

268 Hz, J _PP) 36 Hz, 1 P), 126.8 (dd, J _PRh) 243 Hz, J _PP) 36

Hz, 1 P).

[(1S,2S)-[1-Diphenylphosphinoxy-2-bis-(3,5-bis-triflurometh-

ylp h en yl)ph osph in oxy]cycloh exa n e]Rh ⁺(COD)OTf^-. This

complex was prepared in DME: ³¹P NMR (CD₃COCD₃) δ 124.8

(dd, J _PRh) 195 Hz, J _PP) 36 Hz, 1 P), 126.6 (dd, J _PRh) 170

Hz, J _PP) 36 Hz, 1 P).

[(1R,2R)-[1-Bis(pentafluorophenyl)phosphinoxy-2-bis-(3,5-di-

-

m eth ylph en yl)ph osph in oxy]cycloh exan e]Rh ⁺(COD)SbF₆

(7g): ³¹P NMR (CD₂Cl₂) δ 133.28 (dd, 1 P, J _PRh) 168 Hz, J _PP

) 28 Hz), 102.00 (dquintet, 1 P, J _PRh) 211 Hz, J _PP) 28 Hz).

[Note that this complex has the (1R,2R) configuration. In

Table 4, appropriate changes in the absolute configuration of

the product have been made to account for this.]

P r ep a r a tion of (1R,2R)-1,2-bis-[(d ip en ta flu or op h en yl)-

p h osp h in oxy]cycloh exa n e: ¹H NMR (CDCl₃) δ 3.99 (m, 2

H), 2.07 (m, 2 H), 1.73-1.43 (m, 4 H), 1.27 (m, 2 H); ³¹P NMR

(CDCl₃) δ 84.95 (quintet, J ) 41.2 Hz).

Typ ica l Scou tin g P r oced u r e for Hyd r ogen a tion of

Aceta m id oa cr yla te Der iva tives (Eq 2). In the drybox, a

150 mL Fisher-Porter tube was loaded with 50 mg (0.23 mmol)

of acetamidoacrylate derivative, 1 mg (0.4 mol %) of 3b, and 2

mL of THF. After sealing, the tube was removed from the

drybox and placed behind proper shielding. With adequate

stirring of the solution at room temperature, the tube was

charged with 40 psi of H₂gas and vented. This procedure was

repeated twice more. The tube was charged with 30-40 psi of

H₂. After 3 h, the tube was vented. The conversion as estimated

by gas chromatography was quantitative in most instances.

When a methyl ester derivative of 1 was used, the crude

mixture was analyzed directly by GLPC (Chirasil S-Val, 170

°C/20 min, programmed for 10 °C/min to 195 °C, 195 °C/30

min. See ref 6b for retention times of the two enantiomers.

Typ ica l Scou tin g P r oced u r e for Rh -Ca ta lyzed Hyd r o-

gen a tion of Dim eth yl Ita con a te. Dimethyl itaconate (0.12

g, 0.77 mmol) was dissolved in 3 mL of THF and placed in a

Fisher-Porter apparatus in a drybox. The cationic Rh catalyst

(typically 0.01 mmol) dissolved in 2 mL of THF was added to

the itaconate solution. The apparatus was taken outside the

drybox, and the reaction was placed under 45 psi of hydrogen

after three evacuation-refill cycles as described above. The

reaction was allowed to stir overnight, and it was analyzed

by GC. The reaction was exceptionally clean and the yield

nearly quantitative. On a Chiraldex GTA column (initial

temperaure 85 °C, initial time 20 min, rate of heating 5 °C/

min, final temperature 95 °C), the (S) isomer of the product

appears at 6.30 min and the (R)-isomer at 6.60 min. Each

reaction was repeated twice, and the ee’s are recorded in Table

3. Absolute configurations of the product enantiomers were

determined by comparison of GC retention times with those

P r ep a r a t ion of [(1R,2R)-1,2-Bis-(d ip en t a flu or op h en -

ylp h osp h in oxy)cycloh exa n e]R h ⁺(COD)Sb F ₆^-. (1R,2R)-

1,2-Bis-(dipentafluorophenyl)phosphinoxycyclohexane (0.17 g,

0.20 mmol) in dichloroethane (1 mL) was added slowly to a

-

solution of Rh⁺(COD)₂SbF₆(0.11 g, 0.20 mmol) in dichloro-

ethane (1 mL) at room temperature. After 30 min, the reaction

mixture was concentrated to give the product as an orange

solid: ¹H NMR (CD₂Cl₂) 5.76 (s, br, 2 H), 4.94 (s, br, 2 H),

4.56 (q, J ) 7 Hz, 2 H), 3.63 (s, br, 2 H), 2.63-2.17 (m, 6 H),

0.91-1.83 (m, 8 H); ³¹P NMR (CD₂Cl₂) 104.96 (d, J ) 201 Hz).

[Note that this complex has the 1R,2R configuration. In

Table 4, appropriate changes in the absolute configuration of

the products obtained using this ligand have been made to

account for this.]

P r ep a r a tion of Un sym m etr ica l P h osp h in ites. (1S,2S)-

[1-diph en ylph osph in oxy-2-bis-(3,5-bis-tr iflu r om eth ylph en -

yl)p h osp h in oxy]cycloh exa n e (7h ). The following procedure

was performed inside a drybox. A solution of 58 mg (0.50 mmol)

of the diol and 5 mg of DMAP in 2 mL of 1:1 CH₂Cl₂and

pyridine was chilled inside a freezer at -33 °C, and to this

was added, over 5 min, a CH₂Cl₂solution of 232 mg (0.48

mmol) of bis-(3,5-bis-trifluromethylphenyl)chlorophosphine.

The mixture was allowed to warm to room temperature. After

3 h, the low-boiling solvents were removed on a high vacuum

pump. To the residue was added 1.5 mL of benzene, and the

precipitated solids were removed by filteration through a

cotton pad in a small disposable pipette. The benzene solution

was concentrated, and the product was chromatographed on

silica gel (dried at 200 °C and cooled under nitrogen) inside a

drybox using 10-15% ether/cyclohexane as the solvent. The

fastest moving band was collected and was identified as 7e

by NMR. The major product (190 mg, 66%) was identified as

the monophosphinite from the following data: ¹H NMR (C₆D₆)

δ 0.80-1.80 (m), 3.01 (m, 1 H), 3.20 (m, 1 H), 7.57 (d, J ) 8

Hz, 2 H), 7.86 (d, J ) 5 Hz, 2 H), 8.02 (d, J ) 5 Hz, 2 H); ³¹P

NMR δ 100.8.

of authentic materials obtained from a known hydrogenation

- 15

catalyst, (2S,4S)-[BPPM]Rh⁺(COD)SbF₆

.

A Typ ica l P r ep a r a tion of [Dim eth yl ita con a te]Rh ⁺-

[bis-diarylphosphinite]SbF₆^-.Thecomplex[(7a)Rh⁺(COD)]SbF₆

-

(0.045 g, 0.043 mmol) was dissolved in 2 mL of CD₃OD in a

150 mL Fischer-Porter tube. The solution was hydrogenated

at 45 psi of hydrogen until it turned brown-green. Subse-

quently, the Fisher-Porter tube was transferred back into the

drybox, the gas was vented, and to the reaction mixture was

added 0.040 g (0.25 mmol, 5.8 equiv) of dimethyl itaconate.

The mixture was carefully transferred into an NMR tube and

sealed, and the temperature-dependent ³¹P NMR spectra were

recorded at the various temperatures shown in Figure 13. At

The product from the previous reaction (190 mg, 0.35 mmol)

was dissolved in 1 mL of CH₂Cl₂and 1 mL of pyridine, and 3

mg of 4-(dimethylamino)pyridine was added. To this solution

was added a CH₂Cl₂solution of 72 mg (0.33 mmol) of diphen-

ylchlorophosphine at ∼0 °C. The reaction was allowed to warm

to room temperature. After 16 h, the mixture was concentrated

(38) Ali, R.; Dillon, K. B. J . Chem. Soc., Dalton Trans. 1990, 2593.

Electronic Effects in Asymmetric Catalysis

J . Org. Chem., Vol. 64, No. 10, 1999 3447

the end of the low-temperature runs, the solution was brought

back to room temperature and the spectrum was recorded.

Thus, we have confirmed that all changes are reversible.

for her help in obtaining X-ray data from Cambridge

Crystal Data Files. Dr. Charles Cottrell (OSU) helped

with the low-temperature NMR studies.

Su p p or tin g In for m a tion Ava ila ble: Tables of crystal

data, structure solution and refinement, atomic cordinates,

bond lengths and angles, and anisotropic thermal parameters

for 12, 14, 15, and 16. This material is available free of charge

via the Internet at http://pubs.acs.org.

Ack n ow led gm en t. We are grateful to the National

Science Foundation (CHE-9706766) and EPA/NSF Pro-

gram for Technology for Sustainable Environment (EPA

R826120-01-0) for financial support of this work. Fur-

ther assistance by the Petroleum Research Fund (ACS)

is also acknowledged. We thank Dr. J udy Galluci (OSU)

J O9901182

Article Doi

Electronic Effects in Asymmetric Catalysis: Structural Studies of Precatalysts and Intermediates in Rh-Catalyzed Hydrogenation of Dimethyl Itaconate and Acetamidocinnamic Acid Derivatives Using C2-Symmetric Diarylphosphinite Ligands

DOI: 10.1021/jo9901182

Source and publish data:

Authors:

Article abstract of DOI:10.1021/jo9901182

Full text of DOI:10.1021/jo9901182

Products guided by the article

R&D Labs maybe for 227463-34-5

Relevant to this article

Hot Product