Synthesis of a new class of 1,4-bis(diphenylphosphino)-1,3-butadiene bridged diphosphine, NUPHOS, via zirconium-mediated reductive coupling of alkynes and diynes: Applications in palladium-catalyzed cross-coupling reactions

Organometallics 2002, 21, 1383-1399

1383

Syn th esis of a New Cla ss of

1,4-Bis(d ip h en ylp h osp h in o)-1,3-bu ta d ien e Br id ged

Dip h osp h in e, NUP HOS, via Zir con iu m -Med ia ted

Red u ctive Cou p lin g of Alk yn es a n d Diyn es: Ap p lica tion s

in P a lla d iu m -Ca ta lyzed Cr oss-Cou p lin g Rea ction s

Simon Doherty,*^,†,‡Edward G. Robins,^†Mark Nieuwenhuyzen,^†

J ulian G. Knight,*^,§Paul A. Champkin,^§and William Clegg^§

School of Chemistry, David Keir Building, The Queen’s University of Belfast,

Stranmillis Roads, Belfast BT9 5AG, United Kingdom, and Department of Chemistry,

Bedson Building, University of Newcastle, Newcastle upon Tyne NE1 7RU, United Kingdom

Received December 11, 2001

Zirconium-mediated inter- and intramolecular reductive cyclization of alkynes and diynes has

been used to prepare a new class of bidentate phosphine, based on a four-carbon sp²-hybridized

tether. Intermolecular coupling of diphenylacetylene and but-2-yne with Negishi’s reagent followed

by transmetalation with copper chloride prior to quenching with chlorodiphenylphosphine affords

the corresponding acyclic diphosphines 1,4-bis(diphenylphosphino)-1,2,3,4-tetraphenyl-1,3-buta-

diene (2a ; 1,2,3,4-Ph₄-NUPHOS) and 1,4-bis(diphenylphosphino)-1,2,3,4-tetramethyl-1,3-butadiene

(2b; 1,2,3,4-Me₄-NUPHOS), respectively. A single-crystal X-ray analysis of the former has been

obtained. Surprisingly, 1-phenylpropyne undergoes a highly regioselective reductive cyclization

to afford 1,4-bis(diphenylphosphino)-1,3-diphenyl-2,4-dimethyl-1,3-butadiene (2c; 1,3-Ph₂-2,4-

Me₂-NUPHOS). Similarly, transmetalation of the zirconacyclopentadiene generated from 3,9-

dodecadiyne and 1,8-diphenyloctadiyne followed by electrophilic liberation of the resulting copper

diene reagent with chlorodiphenylphosphine gave 1,2-bis(1-(diphenylphosphino)prop-1-ylidene)-

cyclohexane (2d ; 1,4-Et₂-2,3-cyclo-C₆H₈-NUPHOS) and 1,2-bis(1-(diphenylphosphino)benzylidene)-

cyclohexane (2e; 1,4-Ph₂-2,3-cyclo-C₆H₈-NUPHOS), respectively. This methodology provides a

convenient and versatile one-pot synthesis of a wide range of 1,3-diene bridged diphosphines.

Single-crystal X-ray analyses of [(1,2,3,4-Ph₄-NUPHOS)PdCl₂], [(1,3-Ph₂-2,4-Me₂-NUPHOS)PdCl₂],

and [(1,4-Ph₂-2,3-cyclo-C₆H₈-NUPHOS)PtCl₂] reveal that these new phosphines coordinate to

palladium and platinum in much the same manner as BINAP and dpbp, with a significant

torsional twist about the C(2)-C(3) bond of the backbone. The copper diphosphine intermediate

[Cu(1,4-Et₂-2,3-cyclo-C₆H₈-NUPHOS)Cl]₂(1d ) has also been isolated and characterized by single-

crystal X-ray analysis and exists as the chloro-bridged dimer in which the 1,2-bis(1-(diphen-

ylphosphino)prop-1-ylidene)cyclohexane coordinates in a bidentate manner. Palladium complexes

of these new diphosphines are highly active for the cross-coupling of bromobenzene and sec-

butylmagnesium bromide. Catalyst mixtures based on 1,2,3,4-Ph₄-NUPHOS are far superior to

those based on BINAP, with activities of 6900 and 260 (mol of product) (mol of palladium)^-1h^-1

,

respectively. In fact, catalysts based on 1,2,3,4-Ph₄-NUPHOS are ∼30 times more active than

the most active catalyst reported to date for this coupling. In comparison, the selectivity of the

corresponding cross-coupling with 2-bromopropene depends markedly on the nature of the

NUPHOS derivative. In general, those based on NUPHOS derivatives with acyclic tethers, namely

2a -c, are highly selective for the formation of 2,3-dimethylpentene, while those formed from

2d ,e gave a mixture of 2,3-dimethylpentene, 2-methylhexene, and 2,3-dimethylbutadiene. The

initial TOF, measured after 20 min, also shows a marked variation on the nature of the phosphine

and while all NUPHOS-based catalysts outperform those based on BINAP, catalyst mixtures

based on dppf showed both high selectivity (>99%) and impressive activity. Mixtures of Pd₂-

(dba)₃and the NUPHOS derivatives 2a -e also catalyze the Suzuki cross-coupling of bromobenzene

and 4-bromoacetophenone with phenyl boronic acid, with conversions up to 100% at catalyst

loadings as low as 0.0001 mol % Pd (TON ) 1 × 10⁶).

In tr od u ction

in homogeneous catalysis with immense potential for

performing key steps in the synthesis of important or-

ganic compounds.¹Transfer of chirality in BINAP-based

transformations, such as stereoselective hydrogenation,

relies on the formation of a highly skewed seven-mem-

bered chelate ring. The atropisomerism of BINAP fixes

the conformation of this ring and determines the spatial

arrangement of the chiral P-phenyl rings, which ulti-

The discovery that transition-metal complexes of 2,2′-

bis(diphenylphosphino)-1,1′-binaphthyl (BINAP) cata-

lyze a wide range of asymmetric transformations, often

with spectacular efficiency, was a landmark discovery

^†The Queen’s University of Belfast.

^‡E-mail: s.doherty@qub.ac.uk.

^§University of Newcastle. E-mail (J .G.K.): j.g.knight@ncl.ac.uk.

Publication on Web 03/07/2002

1384 Organometallics, Vol. 21, No. 7, 2002

Doherty et al.

of a more rigid and slightly larger chiral pocket, as

compared to that formed by MeO-BIPHEP. In contrast

to BINAP, MeO-BIPHEP is also capable of acting as a

six-electron donor to Ru(II), via coordination of one of

the biaryl double bonds, and these complexes catalyze

the regiospecific addition of carboxylic acids to terminal

alkynes, to give the corresponding enol esters.⁶Varia-

tion of the dihedral angle in a series of closely related

MeO-BIPHEP type diphosphines, TunaPhos, has re-

vealed that the ruthenium-catalyzed enantioselective

hydrogenation of â-keto esters depends markedly on the

bite angle and reaches a maximum at 88°.⁷The enan-

tioselectivities obtained in the ruthenium-catalyzed

asymmetric hydrogenation of â-keto esters using the

atropisomeric 3,3′-dipyridyldiphosphine 2,2′,6,6′-tet-

ramethoxy-4,4′-bis(diphenylphosphino)-3,3′-bipyridine

(P-phos) compare favorably with Ru(BINAP) systems,

with the advantage of facile catalyst separation by

extraction with aqueous hydrochloric acid.⁸

For reactions that do not involve an asymmetric

transformation, it would be more practical to use an

alternative, less expensive, but structurally similar

diphosphine: for example, 2,2′-bis(diphenylphosphino)-

1,1-biphenyl (BIPHEP).⁹In this regard, Hayashi and

co-workers have recently shown that the selectivity of

BIPHEP-based complexes in the palladium-catalyzed

cross-coupling of sec-BuMgCl with vinyl halides is

higher than that achieved with palladium catalysts of

1,2-bis(diphenylphosphino)ethane (dppe) and compa-

rable to that obtained using 1,1′-bis(diphenylphosphino)-

ferrocene (dppf).¹⁰This ligand has also been applied to

the rhodium-catalyzed Michael addition of boronic acids

to enones and the palladium-catalyzed amination of aryl

bromides, the former showing activities far superior to

those of the corresponding catalyst based on 1,4-bis-

(diphenylphosphino)butane (dppb). In an independent

report, Noyori has shown that activation of ruthenium

complexes of BIPHEP, with an appropriate chiral amine

such as (S,S)-1,2-diphenylethylenediamine ((S,S)-DPEN),

results in the highly enantioselective hydrogenation of

carbonyl compounds,¹¹further highlighting the potential

utility of this ligand type.

mately exert steric influence on the substrate binding

sites.

The success of BINAP and its derivatives has prompted

investigations into the development and applications of

alternative structurally related biaryl atropisomeric

diphosphines,²often based on a 6,6′-substituted biphen-

yl tether such as (R)-MeO-BIPHEP and (R)-BIPHEMP

(the 6,6′-substituents are OMe and Me, respectively).³

These ligands have several features in common with

BINAP in that they are bidentate triarylphosphines

with two diphenylphosphino groups bridged by an sp²-

hybridized four-carbon tether which form a skewed

seven-membered chelate ring when coordinated to a

metal center. Schmid and co-workers have prepared

over 60 such derivatives and investigated the influence

of steric and electronic properties on the enantioselec-

tivity and rates of hydrogenation of unconventional

substrates, including R-pyridyl ketones, γ-keto olefins,

an R-pyrone, and an enol ether of a â-keto lactam.⁴MeO-

BIPHEP-based diphosphines with R-furyl and 3,5-di-

tert-butylphenyl substituents attached to phosphorus

were shown to form particularly effective catalysts,

giving ee’s in excess of 95%. Pregosin has since shown

that the 3,5-di-tert-butylphenyl-substituted derivative

affects the asymmetric palladium-catalyzed Heck reac-

tion and allylic alkylation with enantioselectivities of

>98% and 90%, respectively,⁵and has attributed this

efficiency to the 3,5-dialkylmetal effect, the combination

Given the effectiveness of biphenyl-based diphos-

phines and their potential impact in platinum group

catalysis, there is likely to be considerable interest in

the development of new methodology for the synthesis

of related diphosphines. In this regard, we have re-

cently utilized the zirconium-mediated reductive cycli-

zation of alkynes to prepare a new class of diphosphine,

NUPHOS.¹²Some time ago Fagan and co-workers

demonstrated that zirconacyclopentadienes react with

(1) (a) Noyori, R. Asymmetric Catalysis in Organic Synthesis;

Wiley: New York, 1994. (b) Noyori, R.; Takaya, H. Acc. Chem. Res.

1990, 23, 345. (c) Miyashita, A.; Yasuda, A.; Takaya, K.; Torumi, T.;

Ita, T.; Sauchi, T.; Noyori, R. J . Am. Chem. Soc. 1980, 102, 7933. (d)

Cho, S. Y.; Shibasaki, M. Tetrahedron Lett. 1998, 39, 1773. (e) Ozawa,

F.; Kubo, A.; Hayashi, T. J . Am. Chem. Soc. 1991, 113, 1417. (f)

Shibasaki, M.; Boden, C. D. L.; Kojima, A. Tetrahedron 1997, 53, 7371.

(g) Wolfe, T. P.; Wagaw, S.; Buchwald, S. L. J . Am. Chem. Soc. 1996,

118, 7215.

(2) For selected examples see: (a) Andersen, N. G.; McDonald, R.;

Keay, B. A. Tetrahedron: Asymmetry 2001, 12, 263. (b) Tietze, L. F.;

Thede, K.; Schimpf, R.; Sannicolo, F. Chem. Commun. 2000, 583. (c)

Tietze, L. F.; Thede, K.; Sannicolo, F. Chem. Commun. 1999, 1811. (d)

Benincori, T.; Brenna, E.; Sannicolo, F.; Trimarco, L.; Antognazza, P.;

Cesarotti, E.; Demartin, F.; Pilati, T. J . Org. Chem. 1996, 61, 6244.

(e) Berens, V.; Brown, J .; Long, J .; Selke, R. Tetrahedron: Asymmetry

1996, 7, 285. (f) Cereghetti, M.; Arnold, W.; Broger, E. A.; Rageot, A.

Tetrahedron Lett. 1996, 37, 5347. (g) Heisser, B.; Broger, E. A; Crameri,

Y. Tetrahedron: Asymmetry 1991, 2, 51.

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

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

(6) den Reijer, C. J .; Drago, D.; Pregosin, P. S. Organometallics 2001,

20, 2982.

(7) Zhang, Z.; Qian, H.; Longmire, J .; Zhang, X. J . Org. Chem. 2000,

65, 6223.

(8) Pai, C.-C.; Lin, C.-W.; Lin, C.-C.; Chen, C.-C.; Chan, A. S. C. J .

Am. Chem. Soc. 2000, 122, 11513.

(9) (a) Desponds, O.; Schlosser, M. J . Organomet. Chem. 1996, 507,

257. (b) Desponds, O.; Schlosser, M. Tetrahedron Lett. 1996, 37, 47.

(10) Ogasawara, M.; Yoshida, K.; Hayashi, T. Organometallics 2000,

19, 1567.

(11) (a) Mikami, K.; Korenaga, T.; Terada, M.; Ohkuma, T.; Pham,

T.; Noyori, R. Angew. Chem., Int. Ed. 1999, 38, 495. (b) Badur-Rashid,

K.; Faatz, M.; Lough, A. J .; Morris, R. H. J . Am. Chem. Soc. 2001,

123, 7473.

(12) Doherty, S.; Knight, J . G.; Robins, E. G.; Scanlan, T. H.;

Champkin, P. A.; Clegg, W. J . Am. Chem. Soc. 2001, 123, 5110.

(3) (a) Schmid, R.; Firicher, J .; Cereghetti, M.; Schonholzer, P. Helv.

Chim. Acta 1991, 74, 370. (b) Schmid, R.; Cereghetti, M.; Schonholzer,

P.; Hansen, H.-J . Helv. Chim. Acta 1988, 71, 897.

(4) (a) Fehr, M. J .; Consiglio, G.; Scalone, M.; Schmid, R. J . Org.

Chem. 1999, 64, 5768. (b) Schmid, R.; Broger, E.; Cereghetti, M.;

Crameri, Y.; Foricher, J .; Lalonde, M.; Muller, R. K.; Scalone, M.;

Schoettel, G.; Zutter, U. Pure Appl. Chem. 1996, 68, 131.

Synthesis of a New Class of Bridged Diphosphine

Organometallics, Vol. 21, No. 7, 2002 1385

Sch em e 1

electrophilic reagents to liberate the corresponding

main-group heterocycle, in the case of dichlorophen-

ylphosphine to give the 1-phenyl-2,3,4,5-tetrasubsti-

tuted phosphole.¹³We have been investigating the

synthesis of bidentate phosphines using this route and

have recently reported that treatment of the zircona-

cyclopentadiene formed from diphenylacetylene with 2

equiv of chlorodiphenylphosphine affords 1,4-bis(di-

phenylphosphino)-1,2,3,4-tetraphenyl-1,3-butadiene

(1,2,3,4-Ph₄-NUPHOS) in high yield, noting that the

success of this reaction relied on the presence of a

stoichiometric amount of copper chloride. In comparing

the structures of NUPHOS, BINAP, BIPHEP, and

MeO-BIPHEP, there is a clear similarity between the

basic skeletal frameworks of these diphosphines in that

they all contain two diphenylphosphino groups con-

nected by an sp²-hybridized four-carbon tether. We see

this as a particularly attractive methodology for the

synthesis of diphosphines since (i) it is a relatively

straightforward procedure that can be carried out in a

single pot, (ii) it offers immense scope for structural

variation, and hence catalyst tuning, since a wide range

of alkynes and diynes are known to undergo reductive

cyclization by Negishi’s reagent, and (iii) the ready

availability of chiral diynes lends this methodology to

the synthesis of chiral diphosphines.

Sch em e 2

Preliminary catalyst testing has revealed that pal-

ladium complexes of this new class of diphosphine are

highly active for the cross-coupling of sec-butylmagne-

sium bromide with bromobenzene. Herein we describe

full details of the use of inter- and intramolecular

reductive cyclization of alkynes and diynes to synthesize

an entirely new class of 1,3-butadiene bridged diphos-

phine, the structures of several palladium and platinum

complexes of these diphosphines, and the results of

palladium-catalyzed Grignard cross-coupling and Su-

zuki cross-coupling reactions. The results of our pre-

liminary studies have appeared in the form of a com-

munication.¹²

intensely colored copper complex. The ³¹P{¹H} spectra

of these intermediate copper complexes contain a single

broad resonance at δ -4.3 and -0.9 for 2a and 2b,

respectively, with a half-height line width (∆ν_1/2) of ca.

250 Hz.

To fully characterize this new class of diphosphine,

examine their coordination chemistry, and undertake

catalyst testing, 2a ,b must first be liberated from the

copper, a process that is readily accomplished by re-

peated extraction of a dichloromethane or diethyl ether/

tetrahydrofuran solution with either aqueous ammonia

or ethylenediamine, until the washings no longer exhibit

a blue coloration on exposure to oxygen (Scheme 2). In

each case the diphosphine is isolated as an off-white,

relatively air stable solid which can be recrystallized

from an appropriate solvent. The ³¹P{¹H} NMR spectra

of 2a and 2b contain a single sharp resonance at δ 1.1

and -4.2, respectively, the former shifted slightly down-

field of that corresponding to its copper complex.

Transmetalation of the zirconium-carbon(sp²) bond

to copper is essential for the formation of these diphos-

phines. In this regard, Takahashi has recently shown

that the R-substitiuted zirconocenes [Cp₂Zr(OSiMe₃)-

{C(R)dC(R)CH₂CHdCH₂}] (R ) Pr, Ph) do not react

with chlorodiphenylphosphine, even at elevated tem-

peratures and that addition of copper chloride to the

reaction mixture is necessary to ensure formation of the

tethered olefin phosphine complex [Cu{PR₂C(H)dC(H)-

CH₂CHdCH₂}Cl]₂via P-C bond formation.¹⁴While

transfer of an sp²carbon from zirconium to an electro-

Resu lts a n d Discu ssion

Syn th esis a n d Ch a r a cter iza tion of NUP HOS-

Typ e Dip h osp h in es (2a -e). Some time ago Fagan

demonstrated that zirconacyclopentadienes, generated

either via intermolecular coupling of alkynes or in-

tramolecular coupling of diynes, readily react with

dichlorophenylphosphine via metallacycle transfer from

zirconium to afford the corresponding 1-phenylphos-

phole (Scheme 1a). We reasoned that treatment of such

zirconacyclopentadienes with chlorodiphenylphosphine

should liberate the corresponding substituted 1,4-bis-

(diphenylphosphino)-1,3-dienes according to Scheme 1b.

Unfortunately, we have been unable to isolate any

phosphorus-containing compounds from the reaction of

chlorodiphenylphosphine with a deep red solution of the

zirconacyclopentadiene, generated in situ from Negishi’s

reagent and 2 equiv of diphenylacetylene or but-2-yne.

Remarkably, though, transmetalation of the zirconacy-

clopentadiene with copper chloride prior to addition of

chlorodiphenylphosphine resulted in a rapid and clean

reaction to give the desired diphosphine 2a ,b, as an

(13) (a) Fagan, P. J .; Nugent, W. A. J . Am. Chem. Soc. 1988, 110,

2310. (b) Fagan, P. J .; Nugent, W. A.; Calabrese, J . C. J . Am. Chem.

Soc. 1994, 116, 1880.

1386 Organometallics, Vol. 21, No. 7, 2002

Doherty et al.

Ch a r t 1

philic aryl-substituted chlorophosphine appears to re-

quire a copper-mediated transmetalation, zirconacyclo-

pentanes and zirconacyclopentenes have been shown to

react with chlorodiphenylphosphine to liberate alkyl-

phosphines and homoallylic phosphines, respectively,

via selective activation of the Zr-C(sp³) bond.¹⁵In the

case of zirconacyclopentanes, the first Zr-C(sp³) bond

reacts rapidly with PPh₂Cl to give the corresponding

monophosphine as the major product together with

minor amounts of the desired diphosphine. Selectivity

for the monophosphine was attributed to the different

reactivity of Zr-C(sp³) bonds between monoalkyl zir-

conocenes and zirconacyclopentadienes.

The regioselectivty of this coupling was clearly ap-

parent from the ³¹P{¹H} NMR spectrum of 2c, which

contains two singlets at δ -2.4 and -6.4 ppm of equal

intensity and a pair of singlets at δ -3.4 and -2.3

(<5%), the latter of which most likely corresponds to

minor amounts of the other two possible regioisomers

Transmetalation of Zr-C(sp²) to copper is a relatively

recent development in organozirconium chemistry, and

is proving to be a highly effective methodology for per-

forming a range of carbon-carbon bond formation reac-

tions with zirconcyclopentadienes. For instance, recent

applications of copper-mediated or -catalyzed reactions

of zirconcyclopentadienes include the stereoselective

double allylation of zirconacyclopentadienes to afford

stereodefined 1,4,6,9-tetraenes,^16athe synthesis of fused

aromatics via the coupling of zirconacyclopentadienes

with dihaloaromatics,^16bthe highly selective one-pot

formation of benzene derivatives by intermolecular

coupling of three different alkynes,^16cthe formation of

cyclopentenones via regioselective intermolecular coup-

ling of trisubstituted alkenes, alkynes, and isocya-

nates,^16dformation of fully substituted cyclooctatetra-

enes,^16e,fthe synthesis of vinylcyclohexadienes and

methylenecycloheptadienes via tandem inter- and then

intramolecular allylation of zirconacyclopentadienes,^16g

a new pathway to eight- and nine-membered-ring

derivatives involving the copper-catalyzed intermolecu-

lar [4 + 4] and [4 + 5] coupling of zirconacyclopenta-

dienes with bis(halomethyl)aromatics,^16hthe formation

of cyclopentadiene derivatives and spirocyclic com-

pounds by reaction of zirconacyclopentadienes with 1,1-

dihalo compounds and enones,^16i,jand the preparation

of benzocycloheptene derivatives.^16k

1

(Chart 1). The H spectrum of 2c supports this assign-

ment and contains a singlet at δ 2.02 and a doublet of

doublets at δ 1.60 (J _PH) 3.1 Hz, J _PH) 1.3 Hz), which

correspond to the methyl groups at the 4- and 2-posi-

tions of the butadiene tether, respectively. These methyl

groups appear as a triplet at δ 25.1 (J _PC) 8.1 Hz) and

a doublet at δ 17.2 (J _PC) 3.8 Hz) in the ¹³C{¹H} NMR

spectrum. The alternative explanation, that the former

resonances correspond to a 1:1 mixture of regioisomers

2c′ and 2c′′ and that the minor signals belong to 2c,

was eliminated, since these latter signals clearly had

vastly disparate intensities. Crystallization of the crude

reaction product by slow diffusion of hexane into a

toluene solution at room temperature gave 2c as the

sole product (isolated yield 66%), and the substitution

pattern of the 1,3-diene tether was finally unequivocally

established by a single-crystal X-ray structure deter-

mination of its palladium dichloride derivative (vide

infra).

The intermolecular coupling of unsymmetrical inter-

nal alkynes has not been widely investigated as a

synthetic methodology, since the factors that control the

regiochemical outcome of this reaction are not well

understood. The generation of Negishi’s reagent in the

presence of an unsymmetrical internal alkyne would be

expected to give a statistical mixture of the three

possible regioisomers. However, under certain circum-

stances the regioselectivity of coupling appears to be

predictable, on the basis of the nature of the substitu-

ents. For example, coupling of the zirconocene complex

of 1-hexyne with a 1-trimethylsilyl-substituted internal

alkyne, such as 1-(trimethylsilyl)-1-propyne, gives a 1:1

mixture of regioisomers which after heating isomerize

to give the metallacycle with the trimethylsilyl sub-

stituent in the R-position, as the sole product (Scheme

3).¹⁷Both electronic and steric factors were believed to

be responsible for the regiochemical outcome of this

reaction. The reductive coupling of phenylacetylene in

the presence of Cp₂Ti(PMe₃)₂has been reported to result

in the selective formation of the R,R′-diphenyltitanacy-

clopentadiene, although no reasons were presented for

the regiochemical outcome of this transformation.¹⁸On

the basis of the above reports we might have expected

the zirconium-mediated coupling of 1-phenyl-1-propyne

to give the R,R-diphenylzirconcyclopentadiene, since the

Surprisingly, the reductive coupling of 1-phenyl-1-

propyne occurs with remarkable regioselectivity to give,

after transmetalation and reaction with chlorodiphen-

ylphosphine, the corresponding diphosphine 1,4-bis-

(diphenylphosphino)-1,3-diphenyl-2,4-dimethyl-1,3-buta-

diene (2c; 1,3-Ph₂-2,4-Me₂-NUPHOS).

(14) Miyaji, T.; Xi, Z.; Nakajima, K.; Takahashi, T. Organometallics

2001, 20, 2859.

(15) (a) Nishihara, Y.; Aoyagi, K.; Hara, R.; Suzuki, N.; Takahashi,

T. Inorg. Chim. Acta 1996, 252, 91. (b) Aoyagi, K.; Kasai, K.; Kondakov,

D. Y.; Hara, R.; Suzuki, N.; Takahashi, T. Inorg. Chim. Acta 1994,

220, 319. (c) Mirza-Aghayan, M.; Boukherroub, O. G.; Manuel, G.;

Koenig, M. J . Organomet. Chem. 1998, 564, 61. (d) Mirza-Aghayan,

M.; Boukherroub, O. G.; Etemad-Moghadam, G.; Manuel, G.; Koenig,

M. Tetrahedron Lett. 1996, 37, 3109.

(16) For a relevant review see: Takahashi, T.; Kotora, M.; Hara,

R.; Xi, Z. Bull. Chem. Soc. J pn. 1999, 72, 2591. (a) Takahashi, T.;

Kotora, M.; Kasai, K.; Suzuki, N.; Hakajima, K. Organometallics 1994,

13, 4183. (b) Takahashi, T.; Hara, R.; Nishihara, Y.; Kotora, M. J . Am.

Chem. Soc. 1996, 118, 5154. (c) Takahashi, T.; Xi, Z.; Akiko, Y.; Liu,

Y.; Nakajima, K.; Kotora, M. J . Am. Chem. Soc. 1998, 120, 1672. (d)

Takahashi, T.; Li, Y.; Tsai, F.-Y.; Nakajima, K. Organometallics 2001,

20, 595. (e) Takahashi, T.; Sun, W.-H.; Nakajima, K. Chem. Commun.

1999, 1595. (f) Yamamoto, Y.; Ohno, T.; Itoh, K. Chem. Commun. 1999,

1543. (g) Kotora, M.; Umeda, C.; Ishida, T.; Takahashi, T. Tetrahedron

Lett. 1997, 38, 8355. (h) Takahashi, T.; Sun, W.-H.; Liu, Y.; Nakajima,

K.; Kotora, M. Organometallics 1998, 17, 3841. (i) Duan, Z.; Sun, W.-

H.; Liu, Y.; Takahashi, T. Tetrahedron Lett. 2000, 41, 7471. (j) Xi, C.;

Kotora, M.; Nakajima, K.; Takahashi, T. J . Org. Chem. 2000, 65, 945.

(k) Takahashi, T.; Sun, W.-H.; Duan, Z.; Shen, B. Org. Lett. 2000, 2,

1197.

(17) Buchwald, S. L.; Neilsen, R. B. J . Am. Chem. Soc. 1989, 111,

2870.

(18) Alt, H. G.; Engelhardt, H. E.; Rausch, M. D.; Kool, L. B. J . Am.

Chem. Soc. 1985, 107, 3717.

Synthesis of a New Class of Bridged Diphosphine

Organometallics, Vol. 21, No. 7, 2002 1387

Sch em e 3

Sch em e 5

the copper-coordinated diphosphine derived from 3,9-

dodecadiyne by slow diffusion of methanol into a chlo-

roform solution at room temperature, and a single-

crystal X-ray study has been undertaken (vide infra).

The ³¹P{¹H} spectra of 1d and 1e contain quadrupole-

broadened resonances at δ -3.0 and -2.6, respectively,

with a half-height line width (∆ν_1/2) similar to those of

1a -c. Although the ³¹P NMR resonance of 1d is

broadened by quadrupolar effects (⁶³Cu, I ) ⁹/₂), the

Sch em e 4

1

resonances in the H NMR spectrum are fairly narrow,

since the line widths depend on the distance from the

1

copper nucleus. The H NMR spectrum of 1d contains

a distinctive triplet at δ 0.38 (J ) 7.6 Hz) and a broad

unresolved multiplet at δ 1.25 which correspond to the

ethyl group attached to the diphenylphosphino carbon.

The resonances associated with the methylene groups

of the cyclohexane ring appear as four separate sets of

signals between δ 1.4 and 2.1 ppm.

R-position is expected to be less hindered than the

â-position. At this stage it is impossible to provide an

explanation for the observed regioselectivity, and we

cannot rule out initial formation of a mixture of regio-

isomers followed by equilibration to a single isomer.

While it will not be possible to prepare regioisomers

2c′ and 2c′′ using this methodology, we note that it

should be possible to prepare a variety of unsymmetri-

cally substituted NUPHOS derivatives by exploiting the

facile C_â-C_â′bond cleavage reaction of appropriately

substituted zirconacyclopentenes¹⁹and zirconacyclopen-

tadienes²⁰in the presence of an alkyne to give the

corresponding unsymmetrical zirconacyclopentadiene

selectively, prior to quenching with chlorodiphenylphos-

phine, according to the procedure shown in Scheme 4.

To develop chiral versions of NUPHOS, it will be

necessary to apply the methodology described above to

chiral diynes. With this in mind, we have shown that

1,8-diphenyloctadiyne and 3,9-dodecadiyne undergo a

reductive cyclization transmetalation sequence to afford

the corresponding four-carbon-bridged diphosphines

1,2-bis(1-(diphenylphosphino)benzylidene)cyclohexane

(2d ; 1,4-Ph₂-2,3-cyclo-C₆H₈-NUPHOS) and 1,2-bis(1-(di-

phenylphosphino)prop-1-ylidene)cyclohexane (2e; 1,4-

Et₂-2,3-cyclo-C₆H₈-NUPHOS), respectively (Scheme 5).

We have successfully isolated a crystalline sample of

X-r a y Str u ctu r e of Bis[ch lor o{1,2-bis(1-(d ip h en -

ylp h osp h in o)p r op -1-ylid en e)cycloh exa n e}cop p er ]

(1d ) a n d 1,4-Bis(d ip h en ylp h osp h in o)-1,2,3,4-t et -

r a p h en yl-1,3-bu ta d ien e (2a ). As the synthetic meth-

odology used to prepare 2a -e involves the generation

of an intermediate copper complex, we have undertaken

a single-crystal X-ray structure determination of one of

these, 1d , to establish the mode of coordination and

extent of aggregation. Single crystals of 1d ‚2CHCl₃‚

2MeOH suitable for X-ray analysis were grown from a

chloroform solution layered with methanol. A perspec-

tive view of the molecular structure of 1d together with

the atomic numbering scheme is shown in Figure 1; a

selection of bond lengths and angles is listed in Table

1, and details of the crystal data are given in Table 8.

The structure reveals that 1d exists as a halide-bridged

dimer, a feature common to many copper phosphine

complexes, including [Cu(PPh₂CH₂CHEtOPPh₂)Cl]₂,

bis[chloro{o-phenylenebis(diisopropylphosphine)}cop-

per], and bis[bromo((2,3-dimethyl-2-silabut-3-enyl)di-

phenylphosphine)copper].^21-23The copper atoms are

bridged asymmetrically by the two chlorine atoms

(Cu(1)-Cl(1) ) 2.4564(6) Å, Cu(2)-Cl(1) ) 2.3491(6) Å),

with the difference in the Cu-Cl bond lengths of 0.1073

Å being similar to that recently reported for [Cu(PPh₂-

CH₂CHEtOPPh₂)Cl]₂.²¹As expected, the coordination

(19) Takahashi, T.; Kegeyama, M.; Denisov, V.; Hara, R.; Negishi,

E. Tetrahedron Lett. 1993, 34, 687. (b) van Wagenen, B. C.; Living-

house, T. Tetrahedron Lett. 1989, 30, 3495. (c) Xi, Z.; Hara, R.;

Takahashi, T. J . Org. Chem. 1995, 60, 4444.

(20) Hara, R.; Xi, Z.; Kotora, M.; Xi, C.; Takahashi, T. Chem. Lett.

1996, 1004.

(21) Bayler, A.; Schier, A.; Schmidbaur, H. Inorg. Chem. 1998, 37,

4353.

(22) Baker, R. T.; Calabrese, J . C.; Westcott, S. A. J . Organomet.

Chem. 1995, 498, 109.

(23) Alyea, E. C.; Meehan, P. R.; Ferguson, G.; Kannan, S. Polyhe-

dron 1997, 16, 3479.

1388 Organometallics, Vol. 21, No. 7, 2002

Doherty et al.

F igu r e 2. Molecular structure of 1,4-bis(diphenylphos-

phino)-1,2,3,4-tetraphenyl-1,3-butadiene (2a ). Hydrogen

atoms have been omitted for clarity. Ellipsoids are at the

40% probability level.

F igu r e 1. Molecular structure of [CuCl(1,4-Et₂-2,3-cyclo-

C₆H₈-NUPHOS)]₂(1d ). Hydrogen atoms and CHCl₃and

MeOH molecules of crystallization have been omitted for

clarity. Ellipsoids are at the 50% probability level. Ring

atoms are numbered sequentially in all structures.

Ta ble 2. Selected Bon d Dista n ces (Å) a n d An gles

(d eg) for 2a

P(1)-C(7)

1.8265(13)

1.8334(13)

1.518(2)

P(1)-C(13)

C(13)-C(20)

C(13)-C(14)

1.8436(12)

1.3505(16)

1.4954(16)

P(1)-C(1)

C(20)-C(20A)

C(20)-C(21)

1.4933(16)

Ta ble 1. Selected Bon d Dista n ces (Å) a n d An gles

(d eg) for 1d

C(7)-P(1)-C(1)

C(7)-P(1)-C(13)

106.37(6) C(1)-P(1)-C(13)

100.25(5)

99.68(6) C(20)-C(13)-P(1) 117.41(9)

Cu(1)-P(1)

Cu(1)-Cl(1)

P(1)-C(22)

C(22)-C(1)

C(6)-C(7)

2.2721(6)

2.4564(6)

1.846(2)

1.347(3)

1.349(3)

Cu(1)-P(2)

Cu(1)-Cl(2)

P(2)-C(7)

C(1)-C(6)

2.2613(6)

2.3491(6)

1.836(2)

1.501(3)

C(20)-C(13)-C(14) 124.31(10) C(14)-C(13)-P(1) 117.89(8)

C(21)-C(20)-C(20A) 114.17(10) C(13)-C(20)-C(21) 124.07(10)

C(13)-C(20)-C(20A) 121.57(11)

approximate tetrahedral geometry. The bonding in the

four-carbon tether is highly localized; the C(1)-C(22)

and C(6)-C(7) bond lengths of 1.347(3) and 1.349(3) Å,

the planar nature of C(6), C(7), C(1), and C(22), and the

dihedral angles of 2.4 and 7.0° between the pairs of

planes C(6)C(5)C(1)/C(7)P(2)C(8) and C(1)C(6)C(2)/

C(22)P(1)C(23), respectively, are consistent with C(sp²)-

C(sp²) double bonds, while the C(1)-C(6) bond length

of 1.501(3) Å is close to that expected for a C(sp²)-C(sp²)

single bond. The C(1)-C(22) and C(6)-C(7) bond lengths

(average 1.347 Å) are similar to the carbon-carbon

double bond length of 1.337(12) Å, which bridges the

two diphenylphosphino groups in [Cu{1,2-bis(diphen-

ylphosphino)ethylene}₂][PF₆].²⁴The internal angles of

the cyclohexane ring are all close to the expected value

of 109.5°, including those of the exocyclic double bond,

C(1) ( C(2)C(1)C(6) ) 110.39(19)°) and C(6) ( C(1)-

C(6)C(5) ) 111.93(19)°), which are sp²-hybridized. As a

result, there is a marked deviation of the three inter-

bond angles at C(6) and C(1) from the regular angle of

120°, presumably due to the angular constraints im-

posed by the chairlike six-membered ring.

A single-crystal X-ray analysis of diphosphine 2a has

also been carried out to establish precise details of the

conformational arrangement of the 1,3-diene tether.

Crystals of 2a suitable for X-ray analysis were grown

by slow diffusion of hexane into a toluene solution at

room temperature. The molecular structure together

with the atomic numbering scheme is shown in Figure

2, and a selection of bond lengths and angles is given

in Table 2. The 1,3-diene skeleton is highly skewed

toward the s-transoid conformation rather than s-cisoid

with a torsion angle of 113.3° from the central single

bond C(13)C(20)C(20A)C(13A). This angle is signifi-

P(1)-Cu(1)-P(2)

99.11(2) Cl(1)-Cu(1)-Cl(2) 91.12(2)

P(1)-Cu(1)-Cl(1) 126.87(2) P(1)-Cu(1)-Cl(2) 114.95(2)

P(2)-Cu(1)-Cl(1) 109.23(2) P(2)-Cu(1)-Cl(2) 116.88(2)

C(6)-C(7)-P(2)

C(8)-C(7)-C(6)

C(5)-C(6)-C(7)

122.55(18) C(8)-C(7)-P(2)

114.99(17)

111.93(19)

125.0(2)

122.0(2)

122.7(2)

C(1)-C(6)-C(5)

C(1)-C(6)-C(7)

C(2)-C(1)-C(6)

C(2)-C(1)-C(22) 123.1(2)

C(6)-C(1)-C(22) 126.3(2)

110.39(19)

C(23)-C(22)-C(1) 121.2(2)

C(23)-C(22)-P(1) 118.03(16) C(3)-C(4)-C(5)

111.1(2)

110.5(2)

C(4)-C(5)-C(6)

C(1)-C(2)-C(3)

111.5(2)

C(2)-C(3)-C(4)

110.30(19)

sphere around Cu(1) is close to tetrahedral with a

dihedral angle of 82.3° between the planes containing

P(1), Cu(1), P(2) and Cu(1), Cl(1), Cl(2). The Cu-P bond

lengths of 2.2721(6) and 2.2613(6) Å for Cu(1)-P(1) and

Cu(1)-P(2), respectively, are similar to those reported

for copper complexes of bidentate diphosphines such as

[Cu{1,2-bis(diphenylphosphino)ethylene}₂][PF₆] (2.276-

(2)-2.289(2) Å, mean 2.285(2) Å)²⁴and [Cu(dppf)I]₂

(mean Cu-P distance 2.286(6) Å)²⁵but significantly

shorter than those in [Cu(PPh₃)₄][PF₆] (mean Cu-P

distance of 2.54(6) Å).²⁶The natural bite angle

P(1)-Cu(1)-P(2) of 99.11(2)° is similar to that of 98.98-

(3)° in the recently reported [Cu{(S)-BINAP}-

(MeCN)₂][ClO₄]²⁷and significantly larger than that of

92.066(15)° in the palladium NUPHOS derivative

[(1,2,3,4-Ph₄-NUPHOS)PdCl₂], presumably due to its

(24) Berners-Price, S. J .; Colquhoun, L. A.; Healy, P. C.; Byriel, K.

A.; Hanna, J . V. J . Chem. Soc., Dalton Trans. 1992, 3357.

(25) Neo, S. P.; Zhou, Z.-Y.; Mak, T. C. W.; Hor, A. T. S. J . Chem.

Soc., Dalton Trans. 1994, 3451.

(26) Bowmaker, G. A.; Healy, P. C.; Engelhardt, L. M.; Kildea, J .

D.; Skelton, B. W.; White, A. H. Aust. J . Chem. 1990, 43, 1697.

(27) Ferraris, D.; Young, B.; Cox, C.; Drury, W. J .; Dudding, T.;

Lectka, T. J . Org. Chem. 1998, 63, 6090.

Synthesis of a New Class of Bridged Diphosphine

Organometallics, Vol. 21, No. 7, 2002 1389

cantly larger than the corresponding angle of 35.7° in

1,1′-bis(3,3,4,4-tetrafluoro-2-(diphenylphosphino)cy-

clobutene).²⁸The observed conformation reduces steric

repulsions between the bulky diphenylphosphino groups

and the phenyl rings as well as lone-pair-lone-pair

repulsions, and 2a is therefore not preorganized to

coordinate, since the corresponding angle in [(1,2,3,4-

Ph₄-NUPHOS)PdCl₂] is 47.7°, which requires a rotation

of 27.5° about the central carbon-carbon bond of the

butadiene tether. For comparison, the dihedral angle

between the two naphthyl rings in free BINAP²⁹is 88.3°,

while the values typically found for chelating BINAP

are 65-77°. The bonding in the butadiene tether is

highly localized, and the C(13)-C(20) bond length of

1.3505(16) Å is in the region expected for a C(sp²)-

C(sp²) double bond, similar to that reported for octa-

phenyl[4]radialene³⁰and the diphenylphosphino-sub-

stituted alkenes (E)-[Ph₂P(Ph)CdC(H)Ph],³¹(Z)-[Ph₂P-

(Ph)CdC(H)Ph],³¹[(Ph₂P)₂CdC(H)Me],³²and (Z)-[Ph₂P-

(H)CdC(H)PPh₂],²⁴while the C(20)-C(20A) bond length

of 1.518(2) Å is slightly longer than that of the 1.48 (

0.01 Å expected for a single bond between sp²-hydridized

carbon atoms. Although the sums of the bond angles at

C(13) and C(20) are 359.61 and 359.81°, respectively,

close to the value expected for an sp²-hybridized carbon,

the interbond angles deviate significantly from the

regular sp²value of 120°. Similar distortions have been

noted in diphenylphosphino-substituted alkenes, includ-

ing those in (E)- and (Z)-[Ph₂P(Ph)CdCHPh].³¹

and, in the case of 3b,c and 4e, by single-crystal X-ray

crystallography. For each of the compounds 3a -e, the

³¹P{¹H} NMR spectrum contains a single sharp reso-

nance shifted downfield relative to that of the uncoor-

dinated diphosphine, while that of 4e contains a singlet

at δ 1.2 ppm flanked by platinum satellites with J _Pt-P

) 3054 Hz, in the region expected for phosphorus trans

1

to chloride. In the H NMR spectrum of 4e four distinct

sets of resonances between δ 1 and 2 ppm correspond

to the axial and equatorial protons of the cyclohexane

ring, while the low-field region contains numerous well-

resolved multiplets and a distinctive exchange-broad-

ened signal at δ 8.57. A single low-field exchange-

broadened resonance is a feature common to the ¹H

NMR spectra of compounds 3a -e. Although seemingly

uncomplicated, the presence of a broad low-field signal

prompted us to undertake a variable-temperature ¹H

NMR study of 4e, the result of which is shown in Figure

3. As the temperature is lowered, the resonance at δ

8.57 ppm broadens, disappears into the baseline (243

K), and eventually reappears as two well-separated

multiplets at 9.19 and 7.86 ppm (( in Figure 3), each

corresponding to 2H. Within the same temperature

range the resonance at δ 7.79 broadens and ultimately

gives rise to two ill-defined triplets at δ 7.91 and 6.69

ppm (b in Figure 3), also of intensity 2H, while two

triplets at δ 6.75 and 6.72 ppm (9 in Figure 3), each of

intensity 4H, begin to broaden and decoalesce. Even at

203 K these last two signals are not resolved and appear

as broad resonances. The remaining signals are tem-

perature independent and appear as sharp well-resolved

multiplets over the entire temperature range (1 in

Syn th esis a n d Ch a r a cter iza tion of P a lla d iu m

a n d P la tin u m Com p lexes of 2a -e. Dropwise addition

of a dichloromethane solution of 2a -e into a dichlo-

romethane solution of [(cycloocta-1,5-diene)PdCl₂] re-

sulted in a gradual color change from pale yellow to

yellow-orange with the formation of [(NUPHOS)PdCl₂]

(3a -e) in yields of up to 73%. In a similar manner, the

platinum dichloride complex of 2e, [(1,4-Ph₂-2,3-cyclo-

C₆H₈-NUPHOS)PtCl₂] (4e), was also prepared in near-

quantitative yield (eq 1).

1

Figure 3). Analysis of the COSY H NMR spectrum at

room temperature and 193 K has been used to identify

and assign the aromatic protons. The intensity and

1

pattern of resonances in the room-temperature H NMR

spectrum of 4e is consistent with time-averaged C₂

symmetry with seven temperature-independent mul-

tiplets, each of intensity 2H, belonging to the protons

of two static phenyl rings and the para phenyl signals

of the PPh₂groups and four temperature-dependent

resonances, of total intensity 16H (4:4:4:4), which cor-

respond to the protons of two pairs of freely rotating

phenyl rings. On the basis of the multiplicity and

intensity of the former resonances, in particular the two

doublets at δ 6.21 and 5.94 (J ) 7.4 Hz), we are

confident that these signals correspond to the phenyl

rings attached to the four-carbon tether, while those

that show temperature-dependent ¹H NMR behavior

correspond to the axial and equatorial phenyl rings of

the diphenylphosphino groups. The low-temperature ¹H

NMR spectrum of 4e is consistent with nonequivalent

ortho and meta positions of the diphenylphosphino

phenyl rings as a result of restricted rotation about the

P-Ph bonds. As the temperature is raised, exchange

occurs between the ortho protons associated with signals

A and B to give X and between the meta protons C and

D to give Y. Examination of the space-filling model of

4e (Figure 4) reveals that it is the phenyl rings on the

four-carbon tether that are likely to experience the

greatest barrier to rotation, since these lie close and

Complexes 3a -e and 4e have been characterized

using standard spectroscopic and analytical methods

(28) Rettig, S. J .; Trotter, J . Can. J . Chem. 1977, 55, 3065.

(29) Deeming, A. J .; Speel, D. M.; Stchedroff, M. Organometallics

1997, 16, 6004.

(30) (a) Iyoda, M.; Otani, H.; Oda, M. J . Am. Chem. Soc. 1986, 108,

5371. (b) Ottersen, T.; J elinski, L. W.; Kiefer, E.; Seff, K. Acta

Crystallogr. 1974, B30, 960.

(31) Bookham, J . L.; Smithies, D. M.; Wright, A.; Thornton-Pett,

M.; McFarlane, W. J . Chem. Soc., Dalton Trans. 1998, 811.

(32) Bookham, J . L.; Conti, F.; McFarlane, C. E.; McFarlane, W.;

Thornton-Pett, M. J . Chem. Soc., Dalton Trans. 1994, 1791.

1390 Organometallics, Vol. 21, No. 7, 2002

Doherty et al.

1

F igu r e 3. Variable-temperature H NMR spectra of [(1,4-Ph₂-2,3-cyclo-C₆H₈-NUPHOS)PtCl₂] (4e) showing the pairwise

coalescence of ortho phenyl signals and meta phenyl signals.

perpendicular oriented equatorial P-Ph and an axial

hydrogen atom of the cyclohexyl ring. Restricted rota-

tion about the P-Ph bonds of a coordinated BINAP has

been reported, and it was assumed to be the pseudoaxial

phenyl rings, since they lie close and parallel to the

naphthyl groups and would thus experience the greatest

1

barrier.²⁹The temperature dependence of the H NMR

spectra of 3a -e bears a close similarity to that of 4e in

that the aromatic resonances of each, in particular those

corresponding to the ortho protons, are strongly tem-

perature dependent. In each case the chemical shift

difference is ca. 1 ppm or greater, suggesting that the

ortho sites are in very different environments in the

temperature-limiting conformation.

X-r a y Str u ctu r e of Dich lor o[1,4-bis(d ip h en yl-

F igu r e 4. Space-filling model of 4e. Rotation of the phenyl

p h osp h in o)-1,2,3,4-t e t r a m e t h yl-1,3-b u t a d ie n e ]-

p la tin u m (3b), Dich lor o[1,4-bis(d ip h en ylp h osp h i-

n o)-1,3-d im eth yl-2,4-d ip h en yl-1,3-bu ta d ien e]p la ti-

n u m (3c), a n d Dich lor o[1,2-bis(d ip h en ylp h osp h i-

n ob en zylid en e)cycloh exa n e]p la t in u m (4e). The

similarity between NUPHOS-type diphosphines and

other biaryl-based phosphines such as BINAP and MeO-

BIPHEP prompted us to undertake a single-crystal

X-ray analysis of 3b‚CHCl₃and 4e‚4CHCl₃to provide

precise details of the metal-ligand bonding and the

conformation adopted by the chelate ring. Since there

are three possible isomers of 2c that would result from

the nonselective intermolecular coupling of two unsym-

metrical alkynes, a single-crystal X-ray structure de-

termination of the palladium dichloride complex 3c‚

CH₂Cl₂was undertaken in order to unequivocally

establish the regioselectivity of coupling. Perspective

views of the molecular structures of 3b and 3c together

rings on the four-carbon tether is severely restricted, while

the PPh₂phenyl rings experience a limited barrier to

rotation.

parallel to the pseudoequatorial phenyl rings of the

PPh₂group and are sandwiched between those rings

and the equatorial hydrogen atom on the adjacent

methylene group of the cyclohexyl ring. Variable-tem-

1

perature H NMR studies have shown that this barrier

must be sufficient to restrict rotation, even at room

temperature on the NMR time scale. In contrast, the

barrier to P-Ph rotation of the remaining four phenyl

1

rings is significantly lower and can be monitored by H

NMR spectroscopy. The barrier to rotation of the pseu-

doequatorial phenyl ring presumably arises from its

close proximity to the phenyl ring on the four-carbon

tether, while the barrier to rotation of its axial coun-

terpart is most likely due to interactions between the

Synthesis of a New Class of Bridged Diphosphine

Organometallics, Vol. 21, No. 7, 2002 1391

Ta ble 3. Selected Bon d Dista n ces (Å) a n d An gles

(d eg) for 3b,c

3b

3c

Pd(1)-P(1)

Pd(1)-P(2)

Pd(1)-Cl((1)

Pd(1)-Cl(2)

P(1)-C(1)

P(2)-C(7)

C(1)-C(3)

C(3)-C(5)

C(5)-C(7)

2.2589(5)

2.2680(5)

2.3551(5)

2.3501(5)

Pd(1)-P(1)

Pd(1)-P(2)

Pd(1)-Cl(1)

Pd(1)-Cl(2)

2.2504(7)

2.2505(7)

2.3644(7)

2.3340(8)

1.842(3)

1.849(3)

1.343(4)

1.502(4)

1.352(4)

1.8350(19) P(1)-C(1)

1.841(2)

1.338(3)

1.489(3)

1.335(3)

P(2)-C(4)

C(1)-C(2)

C(2)-C(3)

C(3)-C(4)

P(1)-Pd(1)-P(2)

89.817(18) P(1)-Pd(1)-P(2)

90.37(3)

Cl(1)-Pd(1)-Cl(2) 89.651(19) Cl(1)-Pd(1)-Cl(2) 88.76(3)

P(1)-Pd(1)-Cl(2) 170.999(19) P(1)-Pd(1)-Cl(2) 171.04(3)

P(2)-Pd(1)-Cl(1) 167.424(19) P(2)-Pd(1)-Cl(1) 171.20(3)

P(1)-Pd(1)-Cl(1) 91.751(18) P(1)-Pd(1)-Cl(1) 89.59(3)

Pd(2)-Pd(1)-Cl(2) 90.747(18) Pd(2)-Pd(1)-Cl(2) 92.61(3)

P(1)-C(1)-C(3)

P(1)-C(1)-C(2)

C(2)-C(1)-C(3)

C(1)-C(3)-C(4)

C(1)-C(3)-C(5)

C(4)-C(3)-C(5)

C(3)-C(5)-C(7)

C(3)-C(5)-C(6)

C(6)-C(5)-C(7)

P(2)-C(7)-C(8)

P(2)-C(7)-C(5)

C(5)-C(7)-C(8)

122.11(15) P(1)-C(1)-C(17)

115.20(14) P(1)-C(1)-C(2)

116.38(19)

121.6(2)

122.34(18) C(2)-C(1)-C(17) 122.0(2)

123.0(2)

123.29(18) C(1)-C(2)-C(3)

C(1)-C(2)-C(18) 120.1(2)

125.2(2)

F igu r e 5. Molecular structure of [(1,2,3,4-Me₄-NUPHOS)-

PdCl₂] (3b). Hydrogen atoms and the CHCl₃molecule of

crystallization have been omitted for clarity. Ellipsoids are

at the 50% probability level.

113.70(18) C(3)-C(2)-C(18) 114.4(2)

122.30(18) C(4)-C(3)-C(24) 121.5(3)

114.63(18) C(2)-C(3)-C(24) 114.0(2)

122.9(2)

C(2)-C(3)-C(4)

124.5(2)

115.12(19)

124.1(2)

116.40(16) P(2)-C(4)-C(25)

121.76(15) P(2)-C(4)-C(3)

121.82(19) C(3)-C(4)-C(25) 120.6(2)

and 3c, respectively, are close to the ideal value of 90°

and are only slightly smaller than those in [(1,2,3,4-Ph₄-

NUPHOS)PdCl₂] (92.066(15)°) and [(BIPHEP)PdCl₂]

(92.24(4)°). The remaining cis angles in 3b ( Cl(1)-

Pd(1)Cl(2) ) 89.651(19)°, P(2)Pd(1)Cl(2) ) 90.747(18)°,

P(1)Pd(1)Cl(1) ) 91.751(18)°) and 3c ( Cl(1)Pd(1)-

Cl(2) ) 88.76(3)°, P(2)Pd(1)Cl(2) ) 92.61(3)°, P(1)-

Pd(1)Cl(1) ) 89.59(3)°) are all close to the expected value

of 90°. The corresponding trans angles in 3b ( P(2)-

Pd(1)Cl(1) ) 167.424(19)°, P(1)Pd(1)Cl(2) ) 170.999-

(19)°) and 3c ( P(2)Pd(1)Cl(1) ) 171.20(3)°, P(1)-

Pd(1)P(2) ) 171.04(3)°) are slightly smaller than is

commonly found in related palladium complexes. Simi-

lar distortions away from square-planar geometry in

[PdBr(p-NC-C₆H₄){(S)-MeO-BIPHEP}] have been de-

scribed as a rotation of the P-Pd-P and Cl-Pd-Cl

planes relative to one another, and not a tetrahedral

distortion.³⁴In comparison, the BINAP complex [PdBr-

(p-NC-C₆H₄){p-tol-BINAP}]³⁴showed only minor distor-

tions from a regular square-planar geometry. While the

Pd-Cl bond lengths in 3b are essentially the same

(Pd(1)-Cl(1) ) 2.3551(5) Å, Pd(1)-Cl(2) ) 2.3501(5) Å)

and are within the range reported for the structurally

related complexes [(BINAP)PdCl₂]³³and [(BIPHEP)-

PdCl₂],¹⁰those in 3c are markedly disparate (Pd(1)-

Cl(1) ) 2.3644(7) Å, Pd(1)-Cl(2) ) 2.3340(8) Å). In both

complexes, the 1,3-butadiene bridged diphosphine co-

ordinates to form a seven-membered chelate ring with

F igu r e 6. Molecular structure of [(1,3-Ph₂-2,4-Me₂-

NUPHOS)PdCl₂] (3c). Hydrogen atoms and the CH₂Cl₂

molecule of crystallization have been omitted for clarity.

Ellipsoids are at the 50% probability level.

with their atomic numbering schemes are shown in

Figures 5 and 6, respectively, and selected bond lengths

and angles for both molecules are listed in Table 3. Since

the structures of both compounds are based on pal-

ladium complexes of closely related four-carbon-bridged

diphosphines, they will be discussed in parallel. The

palladium atoms in 3b and 3c show a pronounced

distortion from square planar, as indicated by the

dihedral angles of 15.5 and 12.4°, respectively, between

the planes containing P(1)P(2)Pd(1) and Pd(1)Cl(1)Cl-

(2) and marked deviations of the chlorides from the

PdP₂coordination plane (Cl(1) -0.5087 Å, Cl(2) 0.3669

Å for 3b and Cl(1) 0.3618 Å, Cl(2) -0.3427 Å for 3c).

These distortions are comparable to those reported

for [(BINAP)PdCl₂],³³[(BIPHEP)PdCl₂],¹⁰and [(1,2,3,4-

Ph₄-NUPHOS)PdCl₂],¹²all of which are based on struc-

turally related four-carbon sp²-hybridized tethers. The

natural bite angles of 89.817(18) and 90.37(3)° for 3b

a

distorted-skew-boat conformation, in much the

same manner as BINAP, BIPHEP, MeO-BIPHEP, and

BIPHEMP. The dihedral angle of 63.0° between the

least-squares plane containing the two sp²carbon atoms

and their substituents C(1)C(2)C(3)C(4) and C(5)C(6)C-

(7)C(8) in 3b is slightly larger than the corresponding

(34) (a) Drago, D.; Pregosin, P. S.; Tschoerner, M.; Albinati, A. J .

Chem. Soc., Dalton Trans. 1999, 2279. (b) Magistrato, A.; Merlin, M.;

Pregosin, P. S.; Rothlisberger, U.; Albinati, A. Organometallics 2000,

19, 3591.

(33) Ozawa, F.; Kubo, A.; Matsumoto, Y.; Hayashi, T.; Nishioka, E.;

Yanagi, K.; Moriguchi, K. Organometallics 1993, 12, 4188.

1392 Organometallics, Vol. 21, No. 7, 2002

Doherty et al.

Ta ble 4. Selected Bon d Dista n ces (Å) a n d An gles

(d eg) for 4e

Pt(1)-P(1)

Pt(1)-Cl((1)

P(1)-C(2)

C(2)-C(3)

C(4)-C(5)

2.2453(18)

2.3507(18)

1.843(6)

1.352(8)

1.355(9)

Pt(1)-P(6)

Pt(1)-Cl(2)

P(6)-C(5)

C(3)-C(4)

2.2508(17)

2.3585(16)

1.860(7)

1.501(9)

P(1)-Pt(1)-P(6)

P(1)-Pt(1)-Cl(1)

P(1)-Pt(1)-Cl(2)

P(1)-C(2)-C(21)

C(3)-C(2)-C(21)

C(2)-C(3)-C(4)

C(3)-C(4)-C(5)

C(5)-C(4)-C(41)

C(4)-C(5)-C(51)

C(4)-C(41)-C(42)

92.25(6) Cl(1)-Pt(1)-Cl(2)

176.02(6) P(6)-Pt(1)-Cl(1)

91.47(6) P(6)-Pt(1)-Cl(2)

85.88(6)

90.64(6)

173.93(6)

125.8(5)

123.1(6)

113.3(5)

113.7(6)

116.0(5)

123.1(5)

114.3(2)

119.9(6)

123.5(6)

124.7(6)

121.5(6)

120.9(6)

109.8(6)

P(1)-C(2)-C(3)

C(2)-C(3)-C(31)

C(31)-C(3)-C(4)

C(3)-C(4)-C(41)

P(6)-C(5)-C(51)

C(4)-C(5)-P(6)

C(41)-C(42)-C(32) 110.2(5)

C(32)-C(31)-C(3) 112.1(6)

C(42)-C(32)-C(31) 111.6(6)

(92.3(1)°), [Pt{(R)-Tol-BINAP}(stilbene)]³⁷(94.41(3)°),

and [Pt{(R)-BINAP)(o-C₆H₅OMe)₂] (93.1°)³⁸and is char-

acteristic of diphosphines bridged by four-carbon-

atom sp²-hydridized tethers. The bonding in the 1,3-

diene tether is highly localized with C(2)-C(3) and

C(4)-C(5) bond lengths of 1.352(8) and 1.355(9) Å,

respectively, close to that expected for a C(sp²)-C(sp²)

double bond, and the C(3)-C(4) bond length of 1.501(9)

Å is typical of a single bond between sp²-hybridized

carbon atoms. There is a marked deviation of the inter-

bond angles at C(3) and C(4) from the expected angle

of 120° for an sp²-hybridized carbon. The angles of

113.3(5) and 113.7(6)° for C(4)C(3)C(31) and C(3)C(4)-

C(41), respectively, reflect a strong tendency for these

carbon atoms to adopt an angle close to 109.5°, required

by a cyclohexane ring. To compensate, the remaining

angles at C(3) and C(4) are much greater than the

ideal value of 120°. In contrast to the structures of

3b and 3c, in which the NUPHOS ligand is highly

distorted above the PdP₂plane, that of 4e is close to C₂

symmetry and the conformation of the four-carbon

tether bears a close resemblance to that of many BINAP

and BIPHEP derivatives. The dihedral angle between

the least-squares planes containing C(4)C(41)C(5)(C51)

and C(2)C(21)C(3)C(31) which defines the atropisom-

erism of NUPHOS is 53.3° and is similar to that in 3c.

The four rings of the diphenylphosphino groups are

arranged in an alternating edge/face manner, as is often

observed for atropisomeric diphosphines.³⁹The two

pseudoaxial phenyl rings expose their edges toward

Pt(1) and form torsional angles of 27.5 and 34.7° for the

C(61B) and C(11A) rings, and the pseudoequatorial

rings expose their faces toward platinum to form

torsional angles of 48.0 and 44.7° for C(61A) and C(11B),

respectively.

F igu r e 7. Molecular structure of [(1,4-Ph₂-2,3-cyclo-C₆H₈-

NUPHOS)PtCl₂] (4e). Hydrogen atoms and CHCl₃mol-

ecules of crystallization have been omitted for clarity.

Ellipsoids are at the 50% probability level.

angle of 52.1° between C(3)C(4)C(24)C(25) and C(1)-

C(2)C(7)C(8) in 3c. These angles are similar to that

reported for [Rh{(R)-1,11-bis(diphenylphosphino)-5,7-

dihydrodibenz[c,e]oxepin}(nbd)][BF₄]³(55.9°) and

[{1,2,3,4-Ph₄-NUPHOS)PdCl₂]¹²(47.7°). The pseudo-

axial P-phenyl rings C(15)-C(20) and C(27)-C(32) of

3b expose their edges toward the palladium atom to

form torsional angles of 12.5 and 20.8°, respectively,

while the spatial arrangement of the pseudoequatorial

P-phenyl rings is intermediate between edge and face

with torsional angles of 39.6 and 27.8° for the C(8)-

C(14) and C(21)-C(26) rings, respectively. The P-phenyl

rings in 3c also adopt similar orientations with torsional

angles of 21.1 and 9.0° for the pseudoaxial phenyl rings

C(5)-C(10) and C(31)-C(36), respectively.

Unfortunately, we have been unable to grow crystals

of the palladium complexes 3d ,e suitable for analysis

by single-crystal X-ray crystallography. However, crys-

tals of the platinum derivative of 2e, dichloro[1,2-bis-

(diphenylphosphinobenzylidene)cyclohexane]platinum

(4e), have been obtained from a saturated chloroform

solution at room temperature. A perspective view of the

molecular structure of 4e together with the atomic

numbering scheme is shown in Figure 7, and a selection

of bond lengths and angles is listed in Table 4. Details

of the crystal data are given in Table 8. The molecular

structure shows that the coordination sphere around

Pt(1) is close to square planar with a dihedral angle of

5.7° between the PtP₂and PtCl₂planes and deviations

of -0.1123 and 0.1969 Å for Cl(1) and Cl(2), respectively,

out of the PtP₂coordination plane. The Pt(1)-P(1) and

Pt(1)-P(6) bond lengths of 2.2453(18) and 2.2508(17)

Å are within the range reported for platinum diphos-

phine complexes such as [Pt(BIPHEP)(BINOL}]³⁵(aver-

age Pt-P distance ) 2.2443(18) Å), as are the Pt-Cl

bond lengths of 2.3507(18) and 2.3585(16) Å for

Pt(1)-Cl(1) and Pt(1)-Cl(2), respectively. The natural

bite angle of 92.25(6)° is comparable to those reported

for [Pt(BIPHEP)(BINOL}]³⁵(93.03(7)°), [Pt(BINAP)₂]³⁶

Ku m a d a Cou p lin g Rea ction s of Br om oben zen e

a n d 2-Br om op r op en e. Hayashi and co-workers have

recently reported the first use of dpbp (BIPHEP) for the

(35) Tudor, M. D.; Becker, J . J .; White, P. S.; Gagne, M. R.

Organometallics 2000, 19, 4376.

(36) Tominaga, H.; Sakai, K.; Tsubomura, T. J . Chem. Soc., Chem.

Commun. 1995, 2273.

(37) Wicht, D. K.; Zhuravel, M. A.; Gregush, R. V.; Glueck, D. S.;

Guzei, I. A.; Liable-Sands, L. M.; Rheingold, A. L. Organometallics

1998, 17, 1412.

(38) (a) Alcock, N. W.; Brown, J . M.; Perez-Torrente, J . Tetrahedron

Lett. 1992, 33, 389. (b) Brown, J . M.; Perez-Torrente, J . J .; Alcock, N.

W. Organometallics 1995, 14, 1195.

(39) Toriumi, K.; Ito, T.; Takaya, H.; Souchi, T.; Noyori, R. Acta

Crystallogr. 1982, B38, 807.

Synthesis of a New Class of Bridged Diphosphine

Organometallics, Vol. 21, No. 7, 2002 1393

Ta ble 5. Cr oss-Cou p lin g Rea ction s of

Br om oben zen e w ith sec-Bu tylm a gn esiu m

Ch lor id e^a

catalyst loading of 0.1 mol % Pd. The reaction mixture

begins to reflux vigorously within minutes of adding the

catalyst, and reaction is complete after only 30 min,

resulting in an initial TOF of 6900 (mol of product) (mol

of palladium)^-1h^-1. In contrast, under similar condi-

tions catalysts formed from dppf and BINAP are sig-

nificantly less active and require 5-7 h to reach

completion, with initial TOF values of 145 and 260 (mol

of product) (mol of palladium)^-1h^-1, respectively. The

initial TOF of 6900 (mol of product) (mol of palladium)^-1

h^-1achieved with 1,2,3,4-Ph₄-NUPHOS is over 30 times

greater than that of its dppf counterpart, which clearly

demonstrates the impressive performance of NUPHOS-

type diphosphines in palladium-catalyzed cross-cou-

plings.

In one of the first studies of the influence of the

diphosphine on the efficiency of palladium-catalyzed

cross-couplings, Hayashi showed that [(dppf)PdCl₂]

catalyzes the reaction of bromobenzene with sec-butyl-

magnesium bromide to give sec-butylbenzene in near-

quantitative yield (95%), while other phosphine-based

palladium and nickel complexes were much less active,

and in a number of cases less selective, generating

significant quantities of n-butylbenzene byproduct.⁴⁰In

a recent and systematic examination of the effect of the

bite angle of the diphosphine (ân)⁴¹on the activity and

selectivity of the palladium-catalyzed cross-coupling of

bromobenzene, Kamer has shown that both selectivity

and activity reach a maximum when the natural bite

angle is ca. 100°.⁴²In a comparative study of a range of

Xantphos⁴³type diphosphines, catalysts based on

DPEphos (ân ) 102.7°) were found to be more active

and selective than their dppf counterparts, with initial

TOF values of 181 and 79 (mol of product) (mol of

palladium)^-1h^-1, respectively.^42bThe dppf-based cata-

lyst also generated small amounts of biphenyl byproduct

(∼ 3%), not reported in the early studies of Hayashi,

whereas that based on DPEphos formed ca. 1% each of

biphenyl and n-butylbenzene. An increase in the natural

bite angle above 102.7° resulted in a significant reduc-

tion in activity and selectivity, with the amount of

biphenyl reaching a maximum of 40% for Sixantphos

(ân ) 110°). The high activity of dppf- and DPEphos-

based catalysts has been attributed to the large natural

bite angles of 99.07 and 102.7°, respectively, which

forces the R-Pd-R angle in the diorgano intermediate

to decrease, which effects an increase in the rate of

reductive elimination. Diphosphines with bite angles

greater than that of DPEphos result in a distortion

toward tetrahedral coordination, which increases the

R-Pd-R angle and results in a decrease in the rate of

reductive elimination.

entry

diphosphine

TOF^b

GC yield (%)^c

1

2

3

4

5

6

7

2a

2b

2c

2d

2e

BINAP

dppf

6900

270

880

295

450

260

145

100 (3)

78 (17)

100 (22)

64 (22)

a

Reaction conditions: 1.0 equiv of bromobenzene; 2 equiv of

sec-butylmagnesium chloride (2.0 mol dm^-3); 0.1 mol % [(diphos-

phine)PdCl₂]; 0.5 equiv of n-decane internal standard; diethyl ether

1 mL/5 mmol bromide; substrate (R-Br) bromobenzene; amount

b

of catalyst 0.1 mol %. Initial turnover frequency in units of (mol

of product) (mol of cat.)^-1h^-1, determined by GC analysis of the

reaction mixture after hydrolysis with 10% HCl, based on con-

sumption of starting material. ^cDetermined by GC analysis of the

reaction mixture, based on product. The reaction time (h) is given

in parentheses.

palladium-catalyzed coupling of aryl and alkenyl bro-

mides with sec-butylmagnesium bromide and the rhod-

ium-catalyzed conjugate addition of phenyl boronic acids

to enones.¹⁰In particular, they showed that the activity

and selectivity of catalysts based on BIPHEP compare

favorably with those formed from BINAP and dppf.

Since the four-carbon tether of NUPHOS-type phos-

phines resembles that of BIPHEP and BINAP, we have

examined the efficiency of these new diphosphines in

the palladium-catalyzed cross-coupling of bromobenzene

and 2-bromopropene (eqs 2 and 3). The coupling of

The activity of catalysts formed from NUPHOS type

diphosphines 2a -e show a dramatic dependence on the

substituents attached to the 1,3-butadiene tether. For

instance, substitution of the phenyl groups for methyl

bromobenzene, to give sec-butylbenzene (I), is commonly

used as a model reaction to compare catalyst activity

and selectivity; the possible byproducts of this reaction

being n-butylbenzene (II) and biphenyl (III) (eq 2). The

results of catalyst testing of bromobenzene and 2-bro-

mopropene are summarized in Tables 5 and 6, respec-

tively. In all cases, catalysts formed from NUPHOS

derivatives 2a -e either rival or are far superior to those

based on BINAP and dppf. Initial studies showed that

[(1,2,3,4-Ph₄-NUPHOS)PdCl₂] catalyzes the cross-cou-

pling of bromobenzene to give sec-butylbenzene at a

(40) (a) Hayashi, T.; Konishi, M.; Kumada, M.; Tetrahedron Lett.

1979, 21, 1871. (b) Hayashi, T.; Konishi, M.; Koboni, Y.; Kumada, M.;

Higuchi, T.; Hirotsu, K. J . Am. Chem. Soc. 1984, 106, 158.

(41) Casey, C. P.; Whiteker, G. T. Isr. J . Chem. 1990, 20, 299.

(42) (a) van Leeuwen, P. W. N. M.; Kamer, P. C. J .; Reek, J . N. H.;

Dierkes, P. Chem. Rev. 2000, 100, 2741. (b) Kranenburg, M.; Kamer,

P. C. J .; van Leeuwen, P. W. N. M. Eur. J . Inorg. Chem. 1998, 155.

(43) Kraneburg, M.; van der Burgt, Y. E. M.; Kamer, P. C. J .; van

Leeuwen, P. W. N. M.; Goubitz, K.; Fraanje, J . Organometallics 1995,

14, 3081.

1394 Organometallics, Vol. 21, No. 7, 2002

Doherty et al.

Ta ble 6. Cr oss-Cou p lin g Rea ction of 2-Br om op r op en e w ith sec-Bu tylm a gn esiu m Ch lor id e^a

a

Reaction conditions: 1.0 equiv of 2-bromopropene; 2 equiv of sec-butylmagnesium chloride (2.0 mol dm^-3); 0.1 mol % [(diphosphine)-

b

PdCl₂]; 0.5 equiv of n-decane internal standard, diethyl ether 1 mL/5 mmol bromide. Initial turnover frequency in units of (mol of product)

(mol of cat.)^-1h^-1, determined by GC analysis of the reaction mixture after hydrolysis with 10% HCl, based on consumption of starting

material.

Sch em e 6

groups results in a dramatic decrease in activity from

6900 to 880 and 270 (mol of product) (mol of cat.)^-1h^-1

for [(1,3-Ph₂-2,4-Me₂-NUPHOS)PdCl₂] and [(1,2,3,4,-

Me₄-NUPHOS)PdCl₂], respectively. Since the reductive

coupling of 1-phenyl-1-propyne occurs selectively to give

the 1,3-Ph₂-2,4-Me₂regioisomer as the sole product, it

will be impossible to examine the performance of the

other two regioisomers with 1,4-Me₂-2,3-Ph₂and 1,4-

Ph₂-2,3-Me₂substitution. However, because of the geo-

metrical constraints associated with intramolecular

reductive coupling, the use of diynes to prepare

NUPHOS derivatives incorporates the tether as part of

a carbocycle and provides a means of selectively varying

the substituents at the 1- and 4-positions. In the case

of 2d ,e, the C₄tether is comprised of two adjacent

exocyclic double bonds of a cyclohexane ring and the 1,4-

substituents are Et and Ph, respectively. Catalysts

all are typically close to 90°, and yet they all form

catalysts that are considerably more efficient for the

cross-coupling of bromobenzene. In fact, the bite angles

formed using ligands 2d ,e were found to be markedly

less active than that of 2a but comparable to those based

on 2b,c.

of 2a -e are similar to that of dppp, which has been

shown to form catalysts with low n/ iso selectivity and

particularly poor activity. The impressive activity of

NUPHOS-based catalysts and the pronounced influence

of the nature of the four-carbon tether on catalyst

performance requires further investigation in order to

fully understand the origin of these effects. Although

the reasons for the high activity of NUPHOS-based

catalysts are not well understood, it is clear that bite

angle effects alone are not responsible for the observed

trends in activity and selectivity and the efficiency of

catalysts based on 2a -e is more likely to be due to a

combination of factors.

In stark contrast to the high efficiency of catalysts

based on NUPHOS-type diphosphines for the cross-

coupling of bromobenzene, the performance of the same

catalysts for coupling 2-bromopropene is highly depend-

ent on the nature of the 1,3-diene tether. Specifically,

catalysts based on the acyclic NUPHOS derivatives

2a -c couple 2-bromopropane and sec-butylmagnesium

bromide to give 2,3-dimethyl-1-pentene (IV) with high

selectivity (92-98%) and with the formation of only

minor amounts of the isomerization (V) and homo-

coupled (VI) byproducts 2-methyl-1-hexene and 2,3-

Isomerization⁴⁰and homocoupling⁴⁴are often major

problems associated with cross-couplings involving sec-

ondary Grignard reagents, which in the case of sec-

butylmagnesium bromide results in the formation of

n-butylated byproduct, via a â-hydride elimination

pathway and biphenyl, respectively. Kamer has sug-

gested that the formation of n-butylbenzene results from

an increasing tendency for large natural bite angle

phosphines to stabilize the trigonal-bipyramidal hy-

drido-alkene intermediate (VIII), which increases the

rate of â-hydride elimination compared to reductive

elimination from [(diphos)Pd(R)(2-Bu)] (VII) (Scheme 6).

Catalysts based on diphosphines 2a -e are highly selec-

tive for cross-coupling sec-butylmagnesium bromide

with bromobenzene and give exclusively the sec-buty-

lated (I) product with no evidence for the formation of

its n-butylated isomer (II) or biphenyl (III). Since the

studies of Kamer involved Xantphos type diphosphines,

all of which have similar steric and electronic properties,

any effect on selectivity and activity was attributed to

the bite angle. In contrast, the natural bite angles of

NUPHOS type diphosphines are considerably smaller,

(44) van Asselt, R.; Elsevier, C. J . Organometallics 1994, 13, 1972.

Synthesis of a New Class of Bridged Diphosphine

Organometallics, Vol. 21, No. 7, 2002 1395

Ta ble 7. Su zu k i Cr oss-Cou p lin g Rea ction s w ith Ar yl Br om id es^a

entry

substrate (R-Br)

diphosphine

amt of cat. (mol %)

TOF^b

GC yield %

approx time

TON

750

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

bromobenzene

4-bromoacetophenone

2a

2b

2c

2d

0.1

0.06

0.0001

2250

1509

1230

1893

1458

2100

1800

8700

8300

7600

7900

8500

8600

8900

75

50

41

63

49

70

60

87

83

76

79

85

86

89

100

82

90

99

20 min

10 min

22 h

503

410

631

486

700

600

1450

1383

1267

1317

1417

1433

1483

2e

BINAP

dppf

2a

2b

2c

2d

2e

BINAP

dppf

2a

2b

2c

2d

2e

BINAP

dppf

1000000

820000

900000

990000

1000000

990000

1000000

43 h

20 h

21 h

100

99

100

a

Reaction conditions: 1.0 equiv of aryl bromide; 1.5 equiv of phenylboronic acid; 2 equiv of K₂CO₃; diphosphine:Pd₂(dba)₃1:2; DME (3

b

mL/mmol of aryl bromide); 0.43 equiv of n-decane internal standard; 85 °C. Reaction times have not been minimized. Initial turnover

frequency in units of (mol of product) (mol of cat.)^-1h^-1, determined by GC analysis of the reaction mixture after diluting with diethyl

ether and filtration thought Celite, based on consumption of starting material.

F igu r e 8. Graphical representation of the Grignard cross-coupling reaction between 2-bromopropene and sec-butylmag-

nesium bromide, illustrating the dramatic difference in selectivity for 2,3-dimethylpentene (I), 2-methylhexene (II), and

2,3-dimethylbutadiene (III) between catalysts based on 2a -c and 2d -f.

dimethylbutadiene, respectively (eq 3). The initial TOF

values of 2200 and 2900 (mol of product) (mol of cat.)^-1

h^-1for catalysts formed from 2a and 2c, respectively,

are substantially higher than that of 800 (mol of

product) (mol of cat.)^-1h^-1for 2b and comparable to

that of 2200 (mol of product) (mol of cat.)^-1h^-1obtained

for [(dppf)PdCl₂]. The high activity obtained using

[(dppf)PdCl₂] was somewhat unexpected, considering its

poor performance in the cross-coupling of bromobenzene,

for which a maximum TOF of 145 (mol of product) (mol

of cat.)^-1h^-1was obtained. Under the same conditions

the initial TOF of catalysts based on 2d ,e, as measured

by the consumption of 2-bromopropene, are noticeably

lower than those of 2a -c. Moreover, in contrast to the

high selectivity of catalysts based on 2a -c, those

employing 2d ,e show exceptionally poor selectivity and

both form significant amounts of isomerization (V) and

homocoupled (VI) byproduct (Table 7 and Figure 8). The

disparate selectivities of these catalysts was surprising

and somewhat unexpected, since diphosphines 2a -e all

form catalysts that are highly selective for the Grignard

cross-coupling of bromobenzene.

The most striking difference between NUPHOS de-

rivatives that are highly selective for the formation of

2,3-dimethylpentene and those that generate significant

amounts of byproduct is the structure of the 1,3-diene

tether. NUPHOS derivatives that form highly selective

catalysts contain a conformationally flexible acyclic 1,3-

butadiene tether, while those that form catalysts that

generate large amounts of byproduct contain a 1,3-diene

tether formed from two exocyclic double bonds of a

cyclohexane ring. This trend is further supported by the

poor selectivity of catalysts based on 2f,g (Chart 2),

which have been prepared as part of a continuing

program to develop and evaluate chiral versions of

NUPHOS. The 1,3-butadiene tethers of these NUPHOS

derivatives bear a clear structural similarity to those

of 2d ,e, in that they are comprised of adjacent exocyclic

double bonds on a cyclohexane ring. Most importantly,

the selectivity of catalysts based on 2f,g parallels that

1396 Organometallics, Vol. 21, No. 7, 2002

Doherty et al.

Ch a r t 2

form catalysts that are highly active for the low-

temperature Suzuki coupling of aryl chlorides and

bromides.⁴⁶Since the methodology outlined in Scheme

1 could be applied to the preparation of electron-rich

diphosphines by quenching the organocopper intermedi-

ate with R₂PCl (R ) alkyl), we thought it worthwhile

to examine the efficiency of diphosphines 2a -e in the

palladium-catalyzed Suzuki coupling reactions of aryl

bromides. Preliminary studies have been restricted to

the Pd₂(dba)₃/2a -e-catalyzed coupling of bromobenzene

and the electronically activated 4-bromoacetophenone

with phenylboronic acid (eq 4), the results of which are

obtained with 2d,e, all of which form significant amounts

of 2-methyl-1-hexene and 2,3-dimethylbutadiene. The

disparate performance of these catalyst systems is

clearly illustrated in Figure 8, which compares catalyst

selectivity as a function of diphosphine.

The precise origin of this dramatic change in selectiv-

ity is unclear, since 2a -e all form catalysts that are

highly selective for the cross-coupling of bromoben-

zene: that is, the selectivity is clearly dependent on the

nature of the substrate. We believe that the difference

in the 1,3-diene tethers of 2a -c and 2d -g most likely

manifests itself in the conformational flexibility of the

diphosphine, the former being considerably more flexible

than the latter, the C2-C3 bond of which is constrained

in a cyclohexane ring. Thus, one possible explanation

for the different selectivities of these catalysts could

involve the accessible range of bite angles for NUPHOS

derivatives and the role of this flexibility in stabilizing

and/or influencing the structure of catalytic intermedi-

ates. Kamer has related the formation of n-butylbenzene

and biphenyl in the cross-coupling of bromobenzene to

an increasing tendency to form trigonal-bipyramidal

intermediates of the type [(diphosphine)Pd(H)(2-Bu)R]

(VII) and [(diphosphine)Pd(R)₂(sec-Bu)] (IX) with in-

creasing natural bite angle (Scheme 6). However, any

explanation used to account for the selectivity of

NUPHOS-based catalysts may not be as straightforward

as that presented by Kamer for several reasons. First,

diphosphines 2d -g are unlikely to have sufficient

flexibility to increase their bite angle to the range

required to stabilize five-coordinate intermediates of the

type shown in Scheme 6, which is close to 110°,

assuming that the diphosphine coordinates in an equa-

torial-equatorial manner and not axial-equatorial.

Second, any explanation for the observed selectivity

must also account for the marked dependence on the

nature of the substrate in addition to the type of four-

carbon tether. Thus, the influence of the substrate on

the formation and/or fate of catalytic intermediates as

well as the possible effect of changes in natural bite

angle on their stability must be considered.

summarized in Table 7. It is well known that the

efficiency of palladium-catalyzed Suzuki coupling reac-

tions depends on a large number of variables, including

the palladium source, additives, solvents, temperature,

and base. However, because there is no generality to

the protocol, we have not made any attempt to optimize

reaction conditions but have relied on using a standard

set of reaction conditions. As expected, catalysts based

on 2a -e are more efficient at cross-coupling the elec-

tronically activated 4-bromoacetophone than bromoben-

zene. The results shown in Table 7 demonstrate that

the efficiency of catalysts supported by NUPHOS de-

rivatives 2a -e compare favorably with those obtained

using BINAP and dppf. In contrast to the Grignard

cross-coupling reactions, which showed a dramatic

dependence of catalyst performance on the nature of the

four-carbon tether, the efficiency of 2a -e in palladium-

catalyzed Suzuki cross-coupling reactions appear to be

largely independent of this variable.

Studies at low catalyst concentration were under-

taken to minimize the amount of catalyst. In all cases

reactions of 4-bromoacetephone proceeded to near comple-

tion with 0.0001 mol % Pd, giving TON values close to

1 000 000 (mol of product) (mol of cat.)^-1at 80 °C in ca.

20 h. High turnover numbers for Suzuki cross-coupling

reactions are now commonplace, with early reports

describing the use of cyclometalated tris(2-methylphen-

yl)phosphine and tris(2,4-di-tert-butylphenyl) phosphite

palladium complexes for the cross-coupling of aryl

bromides with TON of 74 000 and 1 000 000 (mol of

product) (mol of cat.)^-1, respectively.⁴⁷More recently,

Buchwald has reproducibly obtained a TON value of

10 000 000 (mol of product) (mol of cat.)^-1for the cross-

coupling of 4-bromoacetophenone with phenylboronic

Su zu k i Cr oss-Cou p lin g Rea ction s of Ar yl Br o-

m id es. The palladium-catalyzed coupling of aryl halides

with organoboron reagents is an effective method for

the formation of C(sp²)-C(sp²) bonds, which offers

immense potential in many areas of organic synthesis.⁴⁵

Intense interest in this area has resulted in a number

of recent noteworthy developments, including the dis-

covery that electron-rich dialkyl- and trialkylphosphines

(46) (a) Wolfe, J . P.; Singer, R. A.; Yang, B. H.; Buchwald, S. L. J .

Am. Chem. Soc. 1999, 121, 9550. (b) Wolfe, J . P.; Buchwald, S. L.

Angew. Chem., Int. Ed. 1999, 38, 2413. (c) Old, D. W.; Wolfe, J .;

Buchwald, S. L. J . Am. Chem. Soc. 1998, 120, 9722. (d) Littke, A. F.;

Fu, G. C. Angew. Chem., Int. Ed. 1998, 37, 3387. (e) Littke, A. F.; Dai,

C.; Fu, G. C. J . Am. Chem. Soc. 2000, 122, 4020. (f) Bei, X.; Crevier,

T.; Guram, A. S.; J andeleit, B.; Powers, T. S.; Turner, H. W.; Uno, T.;

Weinberg, W. H. Tetrahedron Lett. 1999, 40, 3855.

(47) (a) Beller, M.; Fischer, H.; Hermann, W. A.; Ofelen, K.;

Brossmer, Angew. Chem., Int. Ed. Engl. 1995, 34, 1848. (b) Albisson,

D. A.; Bedford, R. B.; Lawrence, S.; Scully, P. N. Chem. Commun. 1998,

2095.

(45) (a) Suzuki, A. In Metal-Catalyzed Cross-Coupling Reactions:

Diedrich, F., Stang, P. J ., Eds.; Wiley-VCH: Weinheim, Germany,

1998; Chapter 2. (b) Sturmer, R. Angew. Chem., Int. Ed. 1999, 28, 3307.

(c) Miyaura, N.; Suzuki, A. Chem. Rev. 1995, 95, 2457. (d) Suzuki, A.

J . Organomet. Chem. 1999, 576, 147. (e) Miyaura, N. In Advances in

Metal-Organic Chemistry; Liebeskind, L. S., Ed.; J AI: London, 1998;

Vol. 6, pp 187-243. (f) Stenforth, S. P. Tetrahedron 1998, 54, 263.

Synthesis of a New Class of Bridged Diphosphine

Organometallics, Vol. 21, No. 7, 2002 1397

acid in <24 h at 100 °C using a (o-biphenyl)P(^tBu)₂

catalyst system and Mathey has reported that octaeth-

yldiphosphaferrocene is an excellent ligand for this

Suzuki cross-coupling, giving a TON of 980 000 (mol of

product) (mol of cat.)^-1at 100 °C, which compares

favorably with those obtained using 2a -e.⁴⁸The en-

couraging performance of catalysts based on 2a -e

suggests that catalytic asymmetric Suzuki couplings

using chiral versions of NUPHOS are a worthwhile

target.

glovebox or using standard Schlenk line techniques under an

atmosphere of nitrogen or argon in oven-dried glassware.

Diethyl ether and hexane were distilled from potassium/

sodium alloy, tetrahydrofuran from potassium, dichloromethane

from calcium hydride, and methanol from magnesium. Unless

otherwise stated, commercially purchased materials were

purchased and used without further purification. Bis(cyclo-

pentadienyl)zirconium dichloride, diphenylacetylene, but-2-

yne, 1-phenylpropyne, butyllithium, copper(I) chloride, phen-

ylboronic acid, 4-bromoacetophenone, sodium carbonate, and

tris(dibenzylideneacetone)dipalladium were purchased from

Aldrich Chemical Co., and 3,9-dodecadiyne was purchased

from Acros Chemicals. Deuteriochloroform was predried with

calcium hydride, vacuum-transferred, and stored over 4 Å

molecular sieves. ¹H and ³¹P{¹H} and ¹³C{¹H} NMR spectra

were recorded on a J EOL LAMBDA 500 or Bruker AC 200,

AMX 300, and DRX 500 machines. GC analyses were con-

ducted on a Varian CP3800 connected to a Varian C8400 auto

sampler or a Perkin-Elmer 8700 gas chromatograph. Response

factors were determined by injection of samples containing

known quantities of authentic sample. The palladium and

platinum complexes [(cycloocta-1,5-diene)MCl₂] (M ) Pd,

Pt)^49,50and 1,8-diphenyloctadiyne⁵¹were prepared as previ-

ously described.

In summary, the zirconium-mediated inter- and in-

tramolecular coupling of internal alkynes and diynes

has been used to prepare an entirely new class of 1,3-

butadiene bridged diphosphine, by treating the resulting

zirconacyclopentadiene with chlorodiphenylphosphine in

the presence of copper chloride. Palladium catalysts

based on NUPHOS derivatives 2a -e are highly effective

for the Grignard cross-coupling of bromobenzene with

sec-butylmagnesium bromide and give exclusively sec-

butylbenzene. The activity of catalysts based on 1,2,3,4-

Ph₄-NUPHOS is particularly impressive, and the initial

TOF of 6900 (mol of product) (mol of cat.)^-1h^-1is more

than 30 times greater than that of the most active

catalyst reported to date for this coupling. In contrast,

the activity and selectivity of the palladium-catalyzed

coupling of 2-bromopropene depends markedly on the

nature of the 1,3-diene tether of the NUPHOS deriva-

tive, those based on an acyclic 1,3-diene forming cata-

lysts that are highly selective for the sec-butylated

product, while those that incorporate a fused cyclohex-

ane ring form catalysts that generate significant amounts

of n-butylated and homocoupled byproduct. On the basis

of the distinctive structural differences between these

two types of NUPHOS derivatives, we believe that the

accessible range of bite angles, i.e., ligand flexibility,

could influence the activity and selectivity of NUPHOS-

based catalysts. Although NUPHOS derivatives 2a -e

form catalysts that either rival or outperform those

based on more familiar diphosphines, such as dppe,

BINAP, BIPHEP, and dppf, delineation of the factors

that influence catalyst selectivity and activity is not

straightforward. However, even at this early stage it is

clear that the activity and selectivity of NUPHOS-based

catalysts is unlikely to be determined solely by the

natural bite angle of the diphosphine, as this parameter

varies only slightly and is close to 90°. NUPHOS

derivatives 2a -e and Pd₂(dba)₃also form highly effec-

tive catalysts for the Suzuki coupling of aryl bromides

and phenylboronic acid, and at low catalyst loading TON

values of 1 000 000 (mol of product) (mol of cat.)^-1have

been achieved. Since the results of our preliminary

catalytic studies have proven to be highly encouraging,

the effectiveness of this new class of diphosphine for a

range of palladium-catalyzed reactions, detailed mecha-

nistic investigations to determine the factors that influ-

ence/determine catalyst activity and selectivity, and the

synthesis of chiral versions of NUPHOS and their

applications in asymmetric transformations are cur-

rently underway.

Syn th esis of 1,4-Bis(d ip h en ylp h osp h in o)-1,2,3,4-tet-

r a p h en yl-1,3-bu ta d ien e (2a ). A Schlenk flask charged with

Cp₂ZrCl₂(4.10 g, 14.0 mmol), diphenylacetylene (5.0 g, 28.0

mmol), and THF (30 mL) was cooled to -78 °C, and nBuLi in

hexane (11.2 mL, 2.5 M, 28.0 mmol) was added with rapid

stirring. The solution was stirred for 20 min, after which time

it was warmed to room temperature and stirred for a further

2.5 h. The solution was then cooled to 0 °C, CuCl (2.8 g, 28.0

mmol) was added, and the reaction mixture was stirred for

20-30 min, after which time a solution of PPh₂Cl (5.0 mL,

28.0 mmol) in THF (15 mL) was added, while the reaction

mixture was maintained at 0 °C. The solution was stirred

overnight and filtered and the solvent removed under reduced

pressure to give a deep orange crystalline solid (³¹P{¹H} NMR

-5.1 ppm, br s). The solid was dissolved in CH₂Cl₂, the solution

was washed with water (3 × 10 mL) and extracted with

aqueous ammonia (3 × 60 mL), the extracts were separated,

dried over magnesium sulfate, and filtered, and the solvent

was removed to leave a yellow crystalline solid. Crystallization

by slow diffusion of n-hexane into a toluene solution at room

temperature gave 2a in 44% yield (4.4 g). ³¹P{¹H} NMR (202.0

1

MHz, CDCl₃, δ): 1.1 (s, PPh₂). H NMR (500.0 MHz, CDCl₃,

δ): 7.4 (td, J ) 6.4 Hz, 1.5 Hz, C₆H₅), 7.31-7.35 (m, C₆H₅),

6.96-7.04 (m, C₆H₅), 6.87 (t, J ) 7.0 Hz, C₆H₅), 6.74 (t, J )

8.2 Hz, C₆H₅), 6.61 (t, J ) 7.3 Hz, C₆H₅). ¹³C{¹H} NMR (125.65

MHz, CDCl₃, δ): 157.5 (dd, J _PC) 7.5, 40.3 Hz, C₆H₅), 143.2

(d, J _PC) 13.4 Hz, C₆H₅), 140.8 (s, C₆H₅), 139.1 (dd, J _PC) 20.6,

4.1 Hz, C₆H₅), 137.6 (d, J _PC) 12.2 Hz, C₆H₅), 135.4 (d, J _PC

14.5 Hz, C₆H₅), 134.3 (d, J _PC) 21.6 Hz, C₆H₅), 132.5 (d, J _PC

)

16.5 Hz, C₆H₅), 131.5 (s, C₆H₅), 129.7 (s, C₆H₅), 128.3 (s, C₆H₅),

127.6 (s, C₆H₅), 127.5 (s, C₆H₅), 127.4 (s, C₆H₅), 127.3 (s, C₆H₅),

217.2 (s, C₆H₅), 216.7 (s, C₆H₅), 125.8 (s, C₆H₅). Anal. Calcd

for C₅₂H₄₀P₂: C, 85.92; H, 5.55. Found: C, 85.55; H, 5.42.

Syn th esis of 1,4-Bis(d ip h en ylp h osp h in o)-1,2,3,4-tet-

r a m eth yl-1,3-bu ta d ien e (2b). Compound 2b was prepared

according to the procedure described above and was isolated

as yellow crystals in 55% yield by slow diffusion of a chloroform

solution layered with methanol. ³¹P{¹H} NMR (121.5 MHz,

CDCl₃, δ): -4.2 (s, PPh₂). ¹H NMR (500.0 MHz, CDCl₃, δ): 7.33

(t, J ) 7.0 Hz, C₆H₅), 7.2-7.25 (m, C₆H₅), 6.96-7.10 (m, C₆H₅),

2.20 (s, 6H, CH₃), 1.58 (s, 6H, CH₃). ¹³C{¹H} NMR (125.65

MHz, CDCl₃, δ): 155.2 (dd, J _PC) 10.4 Hz, 39.3, C₆H₅), 138.4

(d, J _PC) 14.5 Hz, CdC), 136.9 (d, J _PC) 12.3 Hz, C₆H₅), 133.1

Exp er im en ta l Section

Gen er a l P r oced u r es. All manipulations involving air-

sensitive materials were carried out in an inert-atmosphere

(49) Drew, D.; Doyle, J . R. Inorg. Synth. 1990, 28, 348.

(50) Drew, D.; Doyle, J . R. Inorg. Synth. 1990, 28, 346.

(51) Lucht, B. L.; Mao, S. S. H.; Tilley, T. D. J . Am. Chem. Soc. 1998,

120, 4354.

(48) Sava, X.; Ricard, L.; Mathey, F.; Le Floch, P. Organometallics

2000, 19, 4899.

1398 Organometallics, Vol. 21, No. 7, 2002

Doherty et al.

(d, J _PC) 16.5 Hz, C₆H₅), 132.4 (d, J _PC) 19.0 Hz, C₆H₅), 128.1

(d, J _PC) 5.2 Hz, C₆H₅), 127.9 (s, C₆H₅), 127.7 (d, J _PC) 5.2

Hz, C₆H₅), 127.2 (s, C₆H₅), 125.5 (d, J _PC) 7.3 Hz, CdC), 21.9

(t, J _PC) 9.3 Hz, CH₃), 16.2 (d, J _PC) 4.1 Hz, CH₃). Anal. Calcd

for C₃₂H₃₂P₂: C, 80.31; H, 6.73. Found: C, 79.99; H, 6.67.

Syn th esis of 1,4-Bis(d ip h en ylp h osp h in o)-1,3-d ip h en yl-

2,4-d im eth yl-1,3-bu ta d ien e (2c). Compound 2c was pre-

pared according to the procedure described above for 2a and

isolated as colorless crystals in 66% yield by slow diffusion of

a chloroform solution layered with hexane. ³¹P{¹H} NMR

(121.5 MHz, CDCl₃, δ): -2.4 (s, PPh₂), -6.4 (s, PPh₂). ¹H NMR

(500.0 MHz, CDCl₃, δ): 6.75-7.45 (m, C₆H₅, 28H), 6.5 (d, J )

diffusion of n-hexane into a dichloromethane solution at room

temperature gave 2a as deep red prisms in 85% yield (0.70 g).

³¹P{¹H} NMR (121.4 MHz, CDCl₃, δ): 19.6 (s, PPh₂). ¹H NMR

(500.0 MHz, CDCl₃, 193 K, δ): 9.57 (br, 2H, C₆H₅), 8.41 (br t,

2H, C₆H₅), 8.01 (br, 6H, C₆H₅), 6.76-7.12 (m, 14H, C₆H₅), 6.66

(t, J ) 7.3 Hz, 2H, C₆H₅), 6.63 (t, J ) 7.2 Hz, 4H, C₆H₅), 6.51

(t, J ) 7.3 Hz, 2H, C₆H₅), 6.18 (d, J ) 7.6 Hz, 4H, C₆H₅), 6.06

(d, J ) 7.6 Hz, 2H, C₆H₅), 5.94 (d, J ) 7.6 Hz, 2H, C₆H₅). Anal.

Calcd for C₅₂H₄₀P₂PdCl₂‚CH₂Cl₂: C, 64.43; H, 4.25. Found: C,

64.20; H, 4.28.

Dich lor o[1,4-b is(d ip h e n ylp h osp h in o)-1,2,3,4-t e t r a -

m eth yl-1,3-bu ta d ien e]p a lla d iu m (3b). Compound 2b was

prepared according to the procedure described above for 4a

and was isolated as yellow crystals in 78% yield by slow

diffusion of diethyl ether into a chloroform solution layered

at room temperature. ³¹P{¹H} NMR (121.4 MHz, CDCl₃, δ):

24.0 (s, PPh₂). ¹H NMR (500.0 MHz, CDCl₃, δ): 8.30 (br, C₆H₅),

7.50-7.57 (m, C₆H₅), 7.15-7.27 (m, C₆H₅), 1.20 (d, J ) 12.5

Hz, 6H, CH₃), 1.12 (s, 6H, CH₃). ¹³C{¹H} NMR (125.65 MHz,

CDCl₃, δ): 150.7 (s, C₆H₅), 135.7 (d, J _PC) 12.4 Hz, C₆H₅), 134.6

(d, J _PC) 10.3 Hz, C₆H₅), 131.9 (s, C₆H₅), 130.5 (s, C₆H₅), 130.3

(s, C₆H₅), 129.9 (d, J _PC) 26.0 Hz, CdC), 128.9 (d, J _PC) 10.4

Hz, C₆H₅), 127.6 (s, C₆H₅), 127.3 (d, J _PC) 40.0 Hz, CdC), 127.2

(s, C₆H₅), 20.4 (d, J _PC) 6.1 Hz, CH₃), 18.7 (d, J _PC) 12.3 Hz,

CH₃). Anal. Calcd for C₃₂H₃₂P₂PdCl₂‚CHCl₃: C, 51.13; H, 4.29.

Found: C, 51.23; H, 4.32.

Dich lor o[1,4-bis(d ip h en ylp h osp h in o)-1,3-d ip h en yl-2,4-

d im eth yl-1,3-bu ta d ien e]p a lla d iu m (3c). Compound 2c was

prepared according to the procedure described above for 2a

and was isolated as yellow crystals in 78% yield by slow

diffusion of methanol into a chloroform or dichloromethane

solution at room temperature. ³¹P{¹H} NMR (121.4 MHz,

CDCl₃, δ): 26.3 (d, J _PP) 6.1 Hz, PPh₂), 21.4 (d, J _PP) 6.1 Hz,

PPh₂). ¹H NMR (500.0 MHz, CDCl₃, δ): 8.65 (m br, 4H, C₆H₅),

6.80-7.50 (m, 24H, C₆H₅), 6.75 (dt, J ) 2.8 Hz, 7.6 Hz, 2H,

C₆H₅), 1.20 (d, J ) 12.5 Hz, 6H, CH₃), 0.77 (br, 6H, CH₃). Anal.

Calcd for C₄₂H₃₆P₂PdCl₂‚CH₂Cl₂: C, 59.72; H, 4.42. Found: C,

59.93; H, 4.21.

Dich lor o[1,2-bis((diph en ylph osph in o)eth ylm eth ylen e)-

cycloh exa n e]p a lla d iu m (3d ). Compound 2d was prepared

according to the procedure described above for 2a and was

isolated as yellow crystals in 78% yield by slow diffusion of

methanol into a chloroform solution at room temperature.

³¹P{¹H} NMR (121.4 MHz, CDCl₃, δ): 21.5 (s, PPh₂). ¹H NMR

(500.0 MHz, CDCl₃, δ): 8.3 (br, 4H, C₆H₅), 7.57 (m, 16H, C₆H₅),

2.2 (d, J ) 10.4 Hz, 2H, cy CH₂), 2.08 (8 line multiplet. 2H,

CH₂CH₃), 1.57 (br, 2H, cy CH₂), 1.51 (m, 2H, CH₂CH₃), 1.18

(br, 4H, cy CH₂), 0.21 (br t, J ) 7.0 Hz, 6H, CH₂CH₃). ¹³C{¹H}

NMR (125.65 MHz, CDCl₃, δ): 135.0-127.2 (m, C₆H₅, CdC),

34.2 (s, cy CH₂), 26.9 (s, cy CH₂), 24.1 (s, CH₂CH₃), 12.8 (s,

CH₂CH₃). Anal. Calcd for C₃₆H₃₈P₂PdCl₂: C, 60.93; H, 5.40.

Found: C, 60.43; H, 4.91.

6.7 Hz, 2H), 2.02 (s, 3H, CH₃), 1.60 (dd, J _PH) 3.1 Hz, J _PH

)

1.3 Hz, 3H, CH₃). ¹³C{¹H} NMR (75.4 MHz, CDCl₃, δ): 140.9-

124.5 (m, C₆H₅, CdC), 25.1 (t, J _PC) 8.1 Hz, CH₃), 17.26 (d,

J _PC) 3.8 Hz, CH₃). Anal. Calcd for C₄₂H₃₆P₂: C, 83.69; H, 6.02.

Found: C, 83.99; H, 6.37.

Syn th esis of 1,2-bis(1-(d ip h en ylp h osp h in o)p r op -1-

ylid en e)cycloh exa n e (2d ). To a Schlenk flask charged with

Cp₂ZrCl₂(5.40 g, 18.4 mmol), 3,9-dodecadiyne (3.6 mL, 3.0 g,

18.4 mmol), and THF (50 mL), cooled to -78 °C, was added

nBuLi (14.0 mL, 2.5 M, 35.0 mmol) with rapid stirring. The

solution was stirred for 20 min, after which time it was

warmed to room temperature and stirred for a further 2.5 h.

The solution was then cooled to 0 °C, CuCl (3.60 g, 36.3 mmol)

was added, and the reaction mixture was stirred for 20-30

min, after which time a solution of PPh₂Cl (6.77 mL, 37.0

mmol) in THF (15 mL) was added, while the reaction mixture

was maintained at 0 °C. The solution was stirred overnight

and filtered and the solvent removed under reduced pressure

to give a deep orange crystalline solid. The solid was extracted

with 150 mL of a diethyl ether/tetrahydrofuran mixture (v/v,

5:1), washed with water (3 × 10 mL), extracted with 80%

ethylenediamine (3 × 60 mL), separated, dried over magne-

sium sulfate, and filtered and the solvent removed to leave a

yellow crystalline solid. Crystallization by slow diffusion of

n-hexane into a toluene solution at room temperature gave

2d in 26% yield (2.60 g). X-ray-quality crystals of the inter-

mediate copper complex 1d were grown by slow diffusion of

methanol into a chloroform solution at room temperature prior

to washing the crude filtrate with 1,2-ethylenediamine solu-

tion. Compound 1d : ¹P{¹H} NMR (121.5 MHz, CDCl₃, δ) -3.0

(s, PPh₂); ¹H NMR (500.0 MHz, CDCl₃, δ) 7.77 (q, J ) 5.8 Hz,

4H, C₆H₅), 7.69 (m, 4H, C₆H₅), 7.16-7.36 (m, 12H, C₆H₅), 2.20

(d, J ) 9.4 Hz, 2H, C₆H₅), 1.76 (m, 2H, C₆H₅), 1.67 (br, 2H,

C₆H₅), 1.49 (m, 2H, C₆H₅), 1.30 (br, 4H, CH₂CH₃), 0.37 (t, J )

7.6 Hz, 6H, CH₂CH₃). Compound 2d : ³¹P{¹H} NMR (121.5

1

MHz, CDCl₃, δ) -5.5 (s, PPh₂); H NMR (500.0 MHz, CDCl₃,

δ) 7.00-7.75 (m, C₆H₅, 20H), 2.8 (d, J ) 11.3 Hz, 2H, cy CH₂),

2.21 (m, 4H, CH₂CH₃), 1.9 (br, 2H, cy CH₂), 1.50 (br t, 2H, cy

CH₂), 1,2 (br, 2H, cy CH₂), 0.35 (t, J ) 5.8 Hz, 6H, CH₂CH₃);

¹³C{¹H} NMR (75.4 MHz, CDCl₃, δ) 139.4-127.3 (m, C₆H₅, Cd

C), 35.7 (t, J _PC) 5.1 Hz, cy-CH₂), 28.9 (s, cy-CH₂), 23.7 (s,

CH₂CH₃), 14.4 (s, CH₂CH₃). Anal. Calcd for C₃₆H₃₈P₂: C, 81.17;

H, 7.19. Found: C, 80.91; H, 7.01.

Dich lor o[1,2-b is((d ip h e n ylp h osp h in o)p h e n ylm e t h -

ylen e)cycloh exa n e]p a lla d iu m (3e). Compound 2d was

prepared according to the procedure described above for 2a

and was isolated as yellow crystals in 76% yield by slow

diffusion of methanol into a chloroform solution at room

temperature. ³¹P{¹H} NMR (121.4 MHz, CDCl₃, δ): 18.9 (s,

PPh₂). ¹H NMR (500.0 MHz, CDCl₃, δ): 8.69 (br, 4H, C₆H₅),

7.57 (br, 8H, C₆H₅), 6.74-7.4 (m, 8H, C₆H₅), 6.60 (t, J ) 7.0

Hz, 2H, C₆H₅), 6.23 (br s, 4H, C₆H₅), 6.19 (d, J ) 7.6 Hz, 2H,

C₆H₅), 5.89 (d, J ) 7.6 Hz, 2H, C₆H₅), 2.82 (d, J ) 11.9 Hz,

2H, cy CH₂), 2.48 (d, J ) 9.45 Hz, 2H, cy CH₂), 1.98 (br d, J )

12.8 Hz, 2H, cy CH₂), 1.50 (br s, 2H, cy CH₂). Anal. Calcd for

Syn th esis of 1,2-bis(1-(d ip h en ylp h osp h in o)ben zyli-

d en e)cycloh exa n e (2e). 1,2-Bis(((diphenylphosphino)phe-

nyl)methylene)cyclohexane was isolated in 37% yield according

to the procedure described above for 2d . ³¹P{¹H} NMR (121.5

1

MHz, CDCl₃, δ): 1.2 (s, PPh₂). H NMR (500.0 MHz, CDCl₃,

δ): 6.70-7.60 (m, 30H, C₆H₅), 2.45 (br d, J ) 12.1 Hz, 2H, cy

CH₂), 1.90 (td, J ) 8.9 Hz, J ) 3.7 Hz, 2H, cy CH₂), 1.7 (br,

2H, cy CH₂), 1.48 (t, J ) 9.5 Hz, CH₂, cy CH₂). Anal. Calcd for

C

₄₄H₃₈P₂: C, 84.05; H, 6.09. Found: C, 84.44; H, 6.01.

C

₄₄H₃₈P₂PdCl₂: C, 65.59; H, 4.75. Found: C, 65.88; H, 4.97.

Dich lor o[1,4-b is(d ip h e n ylp h osp h in o)-1,2,3,4-t e t r a -

p h en yl-1,3-bu ta d ien e]p a lla d iu m (3a ). A solution of [(cy-

cloocta-1,5-diene)PdCl₂] (0.25 g, 0.88 mmol) in dichlorometh-

ane (8-10 mL) was treated with a dichloromethane solution

(7-10 mL) of 1a (0.64 g, 0.88 mmol) and stirred vigorously

for ca. 1-2 h. The reaction mixture was filtered and the solvent

removed to leave a yellow solid residue. Crystallization by slow

D i c h l o r o [1 ,2 -b i s ((d i p h e n y l p h o s p h i n o )p h e n y l -

m eth ylen e)cycloh exa n e]p la tin u m (4e). A solution of [(cy-

cloocta-1,5-diene)PtCl₂] (0.400 g, 1.07 mmol) in dichloromethane

(20 mL) was treated with a dichloromethane solution (10 mL)

of 1e (0.67 g, 1.07 mmol) and stirred vigorously for ca. 12 h.

The reaction mixture was filtered and the solvent removed to

Synthesis of a New Class of Bridged Diphosphine

Organometallics, Vol. 21, No. 7, 2002 1399

Ta ble 8. Su m m a r y of Cr ysta l Da ta a n d Str u ctu r e Deter m in a tion for Com p ou n d s 1d , 2a , 3b,c a n d 4e

1d

2a

3b

3c

4e

formula

C

₅₂H₄₀P₂

C₇₂H₇₆Cl₂Cu₂P₄‚

2CH₃OH‚2CHCl₃

1566.0

C₃₂H₃₂Cl₂P₂Pd‚

CHCl₃

775.2

C₄₂H₃₆Cl₂P₂Pd‚

CH₂Cl₂

864.9

C₄₄H₃₈Cl₂P₂Pt‚

4CHCl₃

1372.2

M_r

726.8

color

colorless

160

monoclinic

C2/ c

17.3356(9)

10.1305(5)

22.8891(12)

97.829(1)

3982.3(4)

4

1.45

28.7

16803

4723

0.0254

244

0.0350

0.0953

1.050

yellow

160

monoclinic

C2/c

29.012(2)

10.1870(7)

26.5126(19)

108.923(2)

7412.1(9)

4

9.94

28.5

31062

8796

0.0346

419

0.0407

0.1154

yellow

160

monoclinic

P2₁/c

12.7882(7)

16.2276(9)

16.8847(10)

108.070(1)

3331.1(3)

4

10.77

28.5

28070

7807

0.0232

374

0.0254

0.0679

red

160

monoclinic

P2₁/c

10.2320(8)

35.194(3)

11.4065(8)

108.964(11)

3884.6(5)

4

8.66

28.5

33131

9202

0.0270

453

0.0401

0.0889

1.074

colorless

153

monoclinic

P2₁/ n

10.762(2)

21.866(4)

23.807(5)

101.978(3)

5480.7(18)

4

33.35

28.6

26954

12276

0.0762

586

0.0547

0.1238

temp, K

cryst syst

space group

a, Å

b, Å

c, Å

â, deg

V, Å³

Z

µ, cm^-1

θ_max, deg

no. of rflns measd

no. of unique rflns

R_int(on F²)

no. of params

R(F²>2σ(F²))^a

R_w(all data)^b

GOF (S)^c

1.046

1.28, -0.89

1.068

1.06, -0.97

0.888

2.73, -1.25

max, min diff map, e Å^-3

0.30, -0.26

0.67, -1.06

Conventional R ) ∑||F_o| - |F_c||/∑|F_o| for “observed” reflections having F_o> 2σ(F_o²). R_w) [∑w(F_o- F_c²)²/∑w(F_o²)²]^1/2for all data.

a

2

b

2

^cGOF ) [∑w(F_o- F_c²)²/((no. of unique reflections) - (no. of parameters))]^1/2

.

2

leave a yellow solid residue. Crystallization from a dichlo-

romethane-chloroform mixture at room temperature gave 4e

as colorless prisms in 73% yield (0.70 g). ³¹P{¹H} NMR (121.5

MHz, CDCl₃, δ): 1.2 (t, J _PtP) 3054 Hz, PPh₂). ¹H NMR (500.0

MHz, CDCl₃, δ): 8.57 (br, 4H, C₆H₅), 7.76 (br, 6H, C₆H₅), 7.0

(t, J ) 7.5 Hz, 2H, C₆H₅), 6.96 (t, J ) 7.2 Hz, 2H, C₆H₅), 6.89

(t, J ) 7.4 Hz, 2H, C₆H₅), 6.78 (t, J ) 7.1 Hz, 2H, C₆H₅), 6.75

(m, 2H, C₆H₅), 6.25 (t, J ) 7.4 Hz, 2H, C₆H₅), 6.21 (d, J ) 7.3

Hz, 2H, C₆H₅), 5.94 (t, J ) 7.3 Hz, 2H, C₆H₅), 1.86 (d, J ) 13.4

Hz, 2H, cy CH₂), 1.56 (br, 2H, cy CH₂), 1.47 (br, 2H, cy CH₂),

min, after which time it was cooled and diluted with diethyl

ether and an aliquot extracted and filtered through Celite be-

fore being analyzed by gas chromatography. GC conditions:

initial temperature 80 °C, final temperature 170 °C, ramp rate

8 °C/min, injection temperature 200 °C, detector temperature

300 °C, carrier gas He 30 mL/min, column 25 m × 0.32 mm

i.d, WCOT fused silica CP-Chirasil-Dex CB stationary phase.

Cr ysta l Str u ctu r e Deter m in a tion s of 1d , 2a , 3b,c, a n d

4e. All measurements were made on Bruker AXS SMART 1K

CCD diffractometers using graphite-monochromated Mo KR

radiation (λ ) 0.71073 Å). Further information is given in

Table 8. The structures were determined and refined by

standard procedures, including constrained hydrogen atoms.

The programs used were Bruker AXS SMART (diffractometer

control), SAINT (integration),⁵²and SHELXTL (structure

determination and graphics),⁵³together with local programs.

1.19 (br t, J ) 9.0 Hz, 2H, cy CH₂). Anal. Calcd for C₄₄H₃₈

-

Cl₂P₂Pt‚4CHCl₃: C, 41.99; H, 2.79. Found: C, 41.23; H, 2.33.

P a lla d iu m -Ca ta lyzed Cr oss-Cou p lin gs. A Schlenk flask

was charged with [(diphosphine)PdCl₂] (16.6 µmol), n-decane

(1.7 mL, 8.3 mmol), and bromobenzene or 2-bromopropene

(16.6 mmol) under nitrogen. The solution was cooled to -78

°C, and sec-butylmagnesium chloride (16.6 mL, 2 M solution

in n-hexane, 16.6 mmol) was added. The reaction mixture was

warmed to room temperature with stirring and the reaction

monitored by gas chromatography until the starting material

had been completely consumed. Samples were hydrolyzed with

10% HCl and the ether layer separated and analyzed by gas

chromatography. GC conditions: initial temperature 50 °C for

5 min, final temperature 300 °C, ramp rate 10 °C/min to 200

°C for 5 min, ramp rate 20 °C/min to 300 °C for 5 min, injection

temperature 300 °C, detector temperature 300 °C, carrier gas

He 20.0 psig, RTX5 capillary column, 30 m × 0.25 mm i.d.

Gen er a l P r oced u r e for th e Su zu k i Cou p lin g of Ar yl

Br om id es. An oven-dried Schlenk flask was evacuated, back-

filled with argon, and charged with potassium carbonate (1.520

g, 11.00 mmol), phenylboronic acid (1.006 g, 8.25 mmol),

4-bromoacetophenone (1.095 g, 5.50 mmol), n-decane (0.5 mL,

2.37 mmol), and dioxane (2 mL). In a separate oven-dried flask,

a solution of ligand (0.0034 mmol) and [Pd₂(dba)₃] (0.0015 g,

0.0017 mmol) in ∼1 mL of dioxane was stirred for 15 min

before being transferred to the reaction mixture. The flask was

sealed, and the reaction mixture was stirred at 85 °C for 10

Ackn owledgm en t. We gratefully acknowledge fund-

ing from the Queen’s University of Belfast, the EPSRC

(E.G.R., P.A.C., and equipment to W.C.), and INEOS

Acrylics (P.A.C.) and thank J ohnson Matthey for loans

of palladium salts.

Su p p or tin g In for m a tion Ava ila ble: For 1d , 2a , 3b,c,

and 4e, tables giving details of the structure determination,

non-hydrogen atomic positional parameters, all bond distances

and angles, anisotropic displacement parameters, and hydro-

gen atomic coordinates. This material is available free of

charge via the Internet at http://pubs.acs.org. Observed and

calculated structure factor tables are available from the

authors upon request.

OM011049W

(52) SMART and SAINT software for CCD area detectors; Bruker

AXS Inc., Madison, Wi., 1997.

(53) Sheldrick, G. M. SHELXTL Manual; Bruker AXS: Madison,

WI, 1997.

Article Doi

Synthesis of a new class of 1,4-bis(diphenylphosphino)-1,3-butadiene bridged diphosphine, NUPHOS, via zirconium-mediated reductive coupling of alkynes and diynes: Applications in palladium-catalyzed cross-coupling reactions

DOI: 10.1021/om011049w

Source and publish data:

Authors:

Article abstract of DOI:10.1021/om011049w

Full text of DOI:10.1021/om011049w

Products guided by the article

R&D Labs maybe for 420120-83-8

Relevant to this article

Hot Product