Rhenium hydride/boron lewis acid cocatalysis of alkene hydrogenations: Activities comparable to those of precious metal systems

Published on Web 12/08/2010

Rhenium Hydride/Boron Lewis Acid Cocatalysis of Alkene

Hydrogenations: Activities Comparable to Those of Precious

Metal Systems

Yanfeng Jiang, Jeannine Hess, Thomas Fox, and Heinz Berke*

Anorganisch-Chemisches Institut, UniVersita¨t Zu¨rich, Winterthurerstrasse 190,

CH-8037 Zu¨rich, Switzerland

Received August 10, 2010; E-mail: hberke@aci.uzh.ch

Abstract: Dibromonitrosyl(dihydrogen)rhenium(I) complexes [ReBr₂(NO)(PR₃)₂(η²-H₂)] (1; R ) iPr, a; Cy,

b) and Me₂NH·BH₃(DMAB) catalyze at either 90 °C or ambient temperature under 10 bar of H₂the

hydrogenation of various terminal and cyclic alkenes (1-hexene, 1-octene, cyclooctene, styrene, 1,5-

cyclooctadiene, 1,7-octadiene, R-methylstyrene). Maximum turnover frequency (TOF) values of 3.6 × 10⁴

h^-1at 90 °C and 1.7 × 10⁴h^-1at 23 °C were achieved in the hydrogenation of 1-hexene. The extraordinary

catalytic performance of the 1/DMAB system is attributed to the formation of ﬁve-coordinate rhenium(I)

hydride complexes [Re(Br)(H)(NO)(PR₃)₂] (2; R ) iPr, a; Cy, b) and the action of the Lewis acid BH₃

originating from DMAB. The related 2/BH₃·THF catalytic system also exhibits under the same conditions

high activity in the hydrogenation of various alkenes with a maximum turnover number (TON) of 1.2 × 10⁴

and a maximum TOF of 4.0 × 10⁴h^-1. For the hydrogenations of 1-hexene with 2a and 2b, the effect of

the strength of the boron Lewis acid was studied, the acidity being in the following order: BCl₃> BH₃>

BEt₃≈ BF₃> B(C₆F₅)₃> BPh₃. B(OMe)₃. The order in catalytic activity was found to be B(C₆F₅)₃> BEt₃

≈ BH₃·THF > BPh₃. BF₃·OEt₂> B(OMe)₃. BCl₃. The stability of the catalytic systems was checked

via TON vs time plots, which revealed the boron Lewis acids to cause an approximate inverse order with

the Lewis acid strength: BPh₃> BEt₃≈ BH₃·THF > B(C₆F₅)₃. For the 2a/BPh₃system a maximum TON

of 3.1 × 10⁴and for the 2a/B(C₆F₅)₃system a maximum TOF of 5.6 × 10⁴h^-1were obtained in the

hydrogenation of 1-hexene. On the basis of kinetic isotope effect determinations, H₂/D₂scrambling, halide

exchange experiments, Lewis acid variations, and isomerization of terminal alkenes, an Osborn-type catalytic

cycle is proposed with oleﬁn before H₂addition. The active rhenium(I) monohydride species is assumed to

be formed via reversible bromide abstraction with the “cocatalytic” Lewis acid. Homogeneity of the

hydrogenations was tested with ﬁltration and mercury poisoning experiments. These “rhenium(I) hydride/

boron Lewis acid” systems demonstrate catalytic activities comparable to those of Wilkinson- or

Schrock-Osborn-type hydrogenations accomplished with precious metal catalysts.

general, the 18 e^-rule, which limits ligand exchange or the

adoption of low coordination numbers. This presumably is the

Introduction

Rhenium as a border element to precious metals is expected

to display in its chemistry at least some of the features of its

precious metal neighbors.¹For instance, like precious metals,

rhenium is able to undergo facile redox changes,²which is for

many cases an important prerequisite to achieve proper catalytic

performance. However, in contrast to precious metals, rhenium

tends to obey, like middle transition element compounds in

main reason for the as yet restricted applicability of low-valent

rhenium in homogeneous catalysis.³In this context, it is worth

mentioning that rhenium compounds in higher oxidation states

often possess 16 or 14 e^-counts, and hence, these reveal higher

propensity for catalytic transformations.⁴We thus realized that

for the exploitation of low-valent rhenium complexes in catalysis

a tuning of the coordination sphere for ligand lability would

become necessary using ancillary ligand effects, such as the

trans effect,⁵the cis labilization effect,⁶and effects from ligands

with variable electron counts,⁷or eventually utilizing the help

(1) (a) Crabtree, R. H. Acc. Chem. Res. 1979, 12, 331. (b) Crabtree, R. H.

Acc. Chem. Res. 1990, 23, 95. (c) Crabtree, R. H.; Demou, P. C.;

Eden, D.; Mihelcic, J. M.; Parnell, C. A.; Quirk, J. M.; Morris, G. E.

J. Am. Chem. Soc. 1982, 104, 6994. (d) Lightfoot, A.; Schnider, P.;

Pfaltz, A. Angew. Chem., Int. Ed. 1998, 37, 2897. (e) Esteruelas, M. A.;

Oro, L. A. Chem. ReV. 1998, 98, 577. (f) Herrmann, W. A.; Cornils,

B. Angew. Chem., Int. Ed. 1997, 36, 1049. (g) Trost, B. M.; Toste,

F. D.; Pinkerton, A. B. Chem. ReV. 2001, 101, 2067. (h) Osborn, J. A.;

Jardine, F. H.; Young, J. F.; Wilkinson, G. J. Chem. Soc. A 1966, 12,

1711. (i) Schrock, R. R.; Osborn, J. A. J. Am. Chem. Soc. 1976, 98,

2134. (j) Schrock, R. R.; Osborn, J. A. J. Am. Chem. Soc. 1976, 98,

2143. (k) Schrock, R. R.; Osborn, J. A. J. Am. Chem. Soc. 1976, 98,

4450. (l) Wilson, M. E.; Whitesides, G. M. J. Am. Chem. Soc. 1978,

100, 306.

(3) (a) Seidel, F.; Gladysz, J. A. Synlett 2007, 986. (b) Ouh, L. L.; Muller,

T. E.; Yan, Y. K. J. Organomet. Chem. 2005, 690, 3774. (c) Kuninobu,

Y.; Tokunaga, Y.; Kawata, A.; Takai, K. J. Am. Chem. Soc. 2006,

128, 202. (d) Kuninobu, Y.; Nishina, Y.; Matsuki, T.; Takai, K. J. Am.

Chem. Soc. 2008, 130, 14062. (e) Kuninobu, Y.; Matsuki, T.; Takai,

K. J. Am. Chem. Soc. 2009, 131, 9914. (f) Kuninobu, Y.; Yu, P.; Takai,

K. Org. Lett. 2010, 12, 4274. (g) Kusama, H.; Yamabe, H.; Onizawa,

Y.; Hoshino, T.; Iwasawa, N. Angew. Chem., Int. Ed. 2005, 44, 468.

(h) Arceo, E.; Ellman, J. A.; Bergman, R. G. J. Am. Chem. Soc. 2010,

132, 11408.

(2) Casey, C. P. Science 1993, 259, 1552.

9

10.1021/ja107187r  2010 American Chemical Society

J. AM. CHEM. SOC. 2010, 132, 18233–18247 18233

A R T I C L E S

Jiang et al.

of additional reagents functioning as cocatalysts. In analogy to

octahedral ruthenium systems,⁸which possess a priori higher

ligand lability and therefore often great catalytic potential, we

wanted to approach homogeneous rhenium hydrogenations

exploiting among other effects the unique capability of the

nitrosyl ligand to exert the trans effect.⁹Also by the isoelectronic

analogy of Re-NO (NO is a 3 e^-donor; NO⁺is a 2 e^-donor)

with the a Ru-L (L ) 2 e^-donor) unit, we predicted a

respective impact of rhenium compounds on the hydrogenations

close to that of ruthenium.¹⁰

ligand replacements through the cocatalysis with Lewis acids.

Thus, novel catalytic systems for highly efﬁcient hydrogenations

of alkenes were created consisting of the rhenium(I) bromo

hydride complexes 1-4 and boron Lewis acids (BR′₃; R′ ) H,

Et, Ph, C₆F₅). Detailed mechanistic studies were carried out to

shed light on the reaction courses of such new catalyses with

emphasis on the special role of the boron Lewis acid.

Rhenium(I) complexes containing the nitrosyl and halide

ligands have recently been shown to display high afﬁnity to

hydrogen and oleﬁns and were found active in hydrogenation

(of oleﬁns and imines)¹¹and dehydrogenation catalysis (of

amine boranes).¹²Related catalyses of hydrosilylations and

dehydrogenative silylations could also be established.¹³There-

fore, we thought we could “tune” Re(I) complexes for improved

performance in hydrogen catalysis by providing assistance in

Results and Discussion

(4) (a) Mol, J. C. Catal. Today 1999, 51, 289. (b) Herrmann, W. A.;

Fischer, R. W.; Rauch, M. U.; Scherer, W. J. Mol. Catal. 1994, 86,

243. (c) Herrmann, W. A.; Wagner, W.; Flessner, U. N.; Volkhardt,

U.; Komber, H. Angew. Chem., Int. Ed. Engl. 1991, 30, 1636. (d)

Herrmann, W. A.; Fischer, R. W.; Marz, D. W. Angew. Chem., Int.

Ed. Engl. 1991, 30, 1638. (e) Romao, C. C.; Kuhn, F. E.; Herrmann,

W. A. Chem. ReV. 1997, 97, 3197. (f) Herrmann, W. A.; Kuhn, F. E.

Acc. Chem. Res. 1997, 30, 169. (g) Kirillov, A. M.; Haukka, M.;

Kirillova, M. V.; Pombeiro, A. J. L. AdV. Synth. Catal. 2005, 347,

1435.

Exploration of “Re(I)/Boron Lewis Acid” Cocatalyzed

Hydrogenations of Alkenes. Rhenium(I) nitrosyl complexes

[ReBr₂(NO)(PR₃)₂(η²-H₂)] (1; R ) iPr, a; Cy, b) bearing two

trans-phosphine ligands were previously reported to catalyze

the dehydrogenation of Me₂NH·BH₃(DMAB) and transfer

hydrogenations of oleﬁns using DMAB as a hydrogen donor.¹²

In extension of this work we realized that the combination of 1

with DMAB led to an efﬁcient catalytic system for hydrogena-

tions of alkenes. For instance, 1a alone showed only moderate

activity in hydrogenations. When a mixture of 5 mL of 1-hexene

and 0.03 mol % 1a was treated with 10 bar of H₂, a 60%

conversion was achieved within 3 h at 90 °C, giving a turnover

frequency (TOF) value of 803 h^-1(Table 1). In comparison,

when 0.15 mol % DMAB was added as a cocatalyst, 82%

conversion was achieved within only 5 min, giving a turnover

number (TON) of 2987 and a TOF of 3.5 × 10⁴h^-1, which

means a 43 times acceleration compared to the reaction of 1a

alone. Similar results were obtained using the 1b/DMAB (1:5)

system under the same conditions, where a TON of 3014 and

a TOF of 3.6 × 10⁴h^-1were obtained within 5 min. The active

species stayed “alive” after catalysis, since readdition of 5 mL

of 1-hexene to the resultant solution still afforded a TON of

3213 and TOF of 1.3 × 10⁴within 15 min. With other alkenes,

such as 1-octene, cyclooctene, styrene, and also dienes, such

as 1,5-cyclooctadiene and 1,7-octadiene, the 1/DMAB (1:5)

system showed excellent performance in the hydrogenations,

leading to satisfactory TONs and TOFs under the condition of

full conversions. Only in the case of 1,1-disubstituted alkenes,

such as R-methylstyrene, did the hydrogenations proceed slower.

Generally, the pursuit of high TONs worked at the expense of

the performance in TOFs. For instance, despite the fact that a

high TON (8291) was achieved in the hydrogenation of styrene

when 1a was loaded at 0.011 mol % into the catalytic system,

the TOF was 2 times less than that with 0.02 mol % 1a/DMAB.

This was even more pronounced in the case of 1,7-octadiene,

where the reaction with 0.06 mol % 1a was nearly 8 times faster

than that with 0.03 mol % 1b. It is also important to note that

not only were full conversions accomplished in all cases, but

also the catalyses could be carried out under solvent-free

conditions. Thus, the hydrogenated products could be easily

isolated by distillation.

(5) (a) Coe, B. J.; Glenwright, S. J. Coord. Chem. ReV. 2000, 203, 5. (b)

Hartley, F. R. Chem. Soc. ReV. 1973, 2, 163. (c) Crabtree, R. H. The

Organometallic Chemistry of the Transition Metals, 4th ed.; Wiley-

Interscience: Hoboken, NJ, 2005.

(6) (a) Kovacs, A.; Frenking, G. Organometallics 2001, 20, 2510. (b)

Flood, T. C.; Lim, J. K.; Deming, M. A.; Keung, W. Organometallics

2000, 19, 1166. (c) Macgregor, S. A.; MacQueen, D. Inorg. Chem.

1999, 38, 4868. (d) Atwood, J. D.; Brown, T. L. J. Am. Chem. Soc.

1976, 98, 3160.

(7) Llamazares, A.; Schmalle, H. W.; Berke, H. Organometallics 2001,

20, 5277.

(8) (a) Beach, N. J.; Blacquiere, J. M.; Drouin, S. D.; Fogg, D. E.

Organometallics 2009, 28, 441. (b) Chan, W.-C.; Lau, C.-P.; Chen,

Y.-Z.; Fang, Y.-Q.; Ng, S.-M.; Jia, G. Organometallics 1997, 16, 34.

(c) Yi, C. S.; Lee, D. W. Organometallics 1999, 18, 5152. (d) Lee,

H. M.; Smith, D. C.; He, Z. J.; Stevens, E. D.; Yi, C. S.; Nolan, S. P.

Organometallics 2001, 20, 794. (e) Blaser, H. U.; Malan, C.; Pugin,

B.; Spindler, F.; Steiner, H.; Studer, M. AdV. Synth. Catal. 2003, 345,

103. (f) Kubas, G. J. Chem. ReV. 2007, 107, 4152. (g) Harmon, R. E.;

Gupta, S. K.; Brown, D. J. Chem. ReV. 1973, 73, 21. (h) Chen, B.;

Dingerdissen, U.; Krauter, J. G. E.; Rotgerink, H.; Mobus, K.; Ostgard,

D. J.; Panster, P.; Riermeier, T. H.; Seebald, S.; Tacke, T.; Trauthwein,

H. Appl. Catal., A 2005, 280, 17. (i) Kubas, G. J. Catal. Lett. 2005,

104, 79. (j) Lee, J. P.; Ke, Z. F.; Ramirez, M. A.; Gunnoe, T. B.;

Cundari, T. R.; Boyle, P. D.; Petersen, J. L. Organometallics 2009,

28, 1758. (k) Sabo-Etienne, S.; Chaudret, B. Coord. Chem. ReV. 1998,

178, 381. (l) Dharmasena, U. L.; Foucault, H. M.; dos Santos, E. N.;

Fogg, D. E.; Nolan, S. P. Organometallics 2005, 24, 1056. (m) Dinger,

M. B.; Mol, J. C. Eur. J. Inorg. Chem. 2003, 2827.

(9) (a) Berke, H.; Burger, P. Comments Inorg. Chem. 1994, 16, 279. (b)

Machura, B. Coord. Chem. ReV. 2005, 249, 2277.

(10) (a) Naota, T.; Takaya, H.; Murahashi, S. I. Chem. ReV. 1998, 98, 2599.

(b) Noyori, R.; Hashiguchi, S. Acc. Chem. Res. 1997, 30, 97. (c)

Noyori, R.; Ohkuma, T. Angew. Chem., Int. Ed. 2001, 40, 40. (d)

Noyori, R. Angew. Chem., Int. Ed. 2002, 41, 2008. (e) Ito, M.; Ikariya,

T. Chem. Commun. 2007, 5134. (f) Dragutan, V.; Dragutan, I.;

Delaude, L.; Demonceau, A. Coord. Chem. ReV. 2007, 251, 765.

(11) (a) Berke, H. ChemPhysChem 2010, 11, 1837. (b) Choualeb, A.;

Maccaroni, E.; Blacque, O.; Schmalle, H. W.; Berke, H. Organome-

tallics 2008, 27, 3474.

(12) (a) Jiang, Y.; Berke, H. Chem. Commun. 2007, 3571. (b) Jaska, C. A.;

Manners, I. J. Am. Chem. Soc. 2004, 126, 2698. (c) Clark, T. J.; Lee,

K.; Manners, I. Chem.sEur. J. 2006, 12, 8634. (d) Staubitz, A.;

Robertson, A. P.; Manners, I. Chem. ReV. 2010, 110, 4079.

(13) (a) Dong, H. L.; Berke, H. AdV. Synth. Catal. 2009, 351, 1783. (b)

Jiang, Y. F.; Blacque, O.; Fox, T.; Frech, C. M.; Berke, H. Chem.sEur.

J. 2009, 15, 2121.

Previously, we demonstrated that the reactions of 1a,b with

DMAB could generate the ﬁve-coordinate rhenium(I) hydrides

9

18234 J. AM. CHEM. SOC. VOL. 132, NO. 51, 2010

Cocatalysis of Alkene Hydrogenations

A R T I C L E S

a

Table 1. Hydrogenations of Alkenes Catalyzed by the 1/DMAB System at 90 °C under 10 bar of H₂

entry

alkene (vol, mL)

catalyst (amt, mmol)

T (h)

conversion (%)

TON

TOF (h^-1

)

isolated (%)

1^b

2

3

1-hexene (0.01)

1-hexene (5)

1-hexene (10)

1-octene (10)

cyclooctene (10)

styrene (10)

R-methylstyrene (5)

cyclooctadiene (5)

1,7-octadiene (5)

1,7-octadiene (10)

1a/DMAB (0.010)

1a (0.010)

1/12

3

30

60

82

83

88

100

89

100

97

3

2410

2987

3014

3213

4005

5178

5818

6769

5117

8291

4365

1933

2041

1680

3360

803

1a/DMAB (0.011)

1b/DMAB (0.011)

1a/DMAB (0.020)

1b/DMAB (0.020)

1a/DMAB (0.011)

1b/DMAB (0.011)

1a/DMAB (0.011)

1b/DMAB (0.015)

1a/DMAB (0.010)

1a/DMAB (0.020)

1b/DMAB (0.020)

1a/DMAB (0.020)

1b/DMAB (0.020)

1a/DMAB (0.020)

1b/DMAB (0.020)

1a/DMAB (0.020)

1b/DMAB (0.020)

1/12

0.25

0.5

1.0

0.5

1.5

1.2

6

1.5

24

6

3

6.5

3

3.5 × 10⁴

3.6 × 10⁴

1.3 × 10⁴

8010

4

5^c

6

93

92

7

8

9

4005

1.1 × 10⁴

3879

87

93

86

95

10

11

12

13

14

15

16

17

18^d

19^d

20^e

21^e

4848

1128

3411

345

727

1455

297

644

2041

816

8400

1120

100

95

100

78

88

89

86

93

88

85

1.0

2.5

0.2

3.0

^aReactions were performed using 10 or 5 mL of alkene, 0.01-0.02 mmol of 1, and 3.0-6.0 mg of DMAB and monitored with a Buechi Pressﬂow

controller under 10 bar of H₂at 90 °C. ^bThe reaction was carried out in toluene-d₈without H₂. ^cSecond run of entry 3 with 5 mL of 1-hexene.

^dHydrogenated to cyclooctane (2 equiv of H₂was consumed). ^eHydrogenated to octane (2 equiv of H₂was consumed).

[Re(Br)(H)(NO)(PR₃)₂] (2; R ) iPr, a; Cy, b) via heterolytic

cleavage of the H-H bond.¹⁴The catalytic activity of 2a in

the hydrogenation of 1-hexene was referenced to the activity

of 1a. 2a proved to be 2 times as active as 1a, but both catalysts

were much less active than the 1a/DMAB system. Furthermore,

addition of DMAB to 2a resulted in the same catalytic activity

as found for the 1a/DMAB system in the hydrogenation of

1-hexene and 1-octene. Not only was a high TOF of 3.1 × 10⁴

h^-1obtained in the case of 1-hexene (5 mL) using a 0.025 mol

% 2a/DMAB (1:5) mixture, but also the catalytic species was

revealed to be “living”, affording a TOF of 1.5 × 10⁴h^-1within

15 min after readdition of 5 mL of 1-hexene. Replacement of

DMAB with BH₃·THF resulted in a slightly higher activity.

When 5 mL of 1-hexene was treated with 0.03 mol % 2a and

0.15 mol % BH₃·THF, a 92% conversion was achieved at 90

°C within 5 min corresponding to a TON of 3083 and a TOF

of 3.7 × 10⁴h^-1. The active species of the 2a/BH₃·THF system

has a longer lifetime than that of the 2a/DMAB system since

Figure 1. Repeated hydrogenations of 1-hexene (5 mL × 4) catalyzed by

the 2a/BH₃·THF (7 mg/50 µL) system at 90 °C under 10 bar of H₂: 9,

ﬁrst run, 5 min, 92%, TON ) 3083, TOF ) 3.7 × 10⁴h^-1; b, second run,

10 min, 98%, TON ) 3350, TOF ) 2.0 × 10⁴h^-1; 2, third run, 15 min,

95%, TON ) 3240, TOF ) 1.3 × 10⁴h^-1; 1, fourth run, 24 min, 89%,

readdition of 5 mL of 1-hexene results in a 98% conversion

within 10 min corresponding to a TON of 3350 and TOF of

2.0 × 10⁴h^-1. This procedure was repeated twice, leading to

some decrease in activity, but in absolute terms still high TOFs

of 1.3 × 10⁴h^-1(third run) and 7558 h^-1(fourth run) were

determined (Figure 1). The rate of the catalysis seemed to

correlate with the alkene concentration. Due to the living

character of the 2/BH₃·THF system, the loading of 2a,b could

be decreased to a value of 0.006 mol % (60 ppm), affording a

TON of 1.2 × 10⁴and a TOF of 4.0 × 10⁴h^-1within 18 min

(2a/BH₃·THF) or a TON of 9942 and a TOF of 2.5 × 10⁴h^-1

within 24 min (2b/BH₃·THF). With other substrates such as

1-octene, cyclooctene, R-methylstyrene, 1,5-cyclooctadiene, and

3,3-dimethylbutene, the 2/BH₃·THF system performed also

extremely well, resulting in both high TONs and high TOFs,

TON ) 3023, TOF ) 7558 h^-1

.

as depicted in Table 2. Strangely enough, the hydrogenations

of allyl derivatives, such as allylbenzene and allyltrimethylsilane,

were slow, which may point to the formation of allyl complexes

which block the catalyses. It should be noted that under a H₂

atmosphere at 90 °C, only trace amounts of isomerization

products of the terminal alkenes could be observed after catalysis

1

using H NMR spectroscopy.

1. Inﬂuence of the Type of Boron Lewis Acid. Under the same

conditions other boron Lewis acids were tested as cocatalysts

in the hydrogenations with 2a,b as well. The Lewis acidity of

the employed boron Lewis acid was initially tested by the

(14) Jiang, Y. F.; Blacque, O.; Fox, T.; Frech, C. M.; Berke, H.

Organometallics 2009, 28, 5493.

9

J. AM. CHEM. SOC. VOL. 132, NO. 51, 2010 18235

A R T I C L E S

Jiang et al.

Table 2. Hydrogenations of Alkenes Catalyzed by 2 and Various Boron Lewis Acids at 90 °C^a

entry

alkene (vol, mL)

catalyst (x; amt, mmol)

T (h)

conversion (%)

TON

TOF (h^-1

)

1

2

3

4

5

6

7

8

1-hexene (5)

1-octene (5)

1-hexene (5)

1-hexene (10)

cyclooctene (5)

1-octene (10)

2a (0.030; 0.012)

1.0

0.11

0.5

1/12

0.3

0.4

1/3

0.4

1/3

6.0

7.0

0.5

2.0

0.6

0.1

1.0

1.5

0.3

1.0

0.25

0.3

0.06

1/12

4.0

8.0

1/3

8.0

1.0

1.25

56

86

94

92

73

62

75

90

83

81

96

68

99

0

50

97

47

88

87

93

88

56

65

51

84

35

52

62

100

72

1860

3455

5897

1860

2a/DMAB (0.025; 0.010)

2a/DMAB (0.016; 0.005)

2a/BH₃THF (0.030; 0.012)

2a/BH₃THF (0.006; 0.005)

2b/BH₃THF (0.006; 0.005)

2a/BH₃THF (0.026; 0.010)

2b/BH₃THF (0.026; 0.010)

2a/BH₃THF (0.015; 0.010)

2b/BH₃THF (0.015; 0.010)

2a/BH₃THF (0.024; 0.010)

2b/BH₃THF (0.025; 0.010)

2a/BF₃OEt₂(0.030; 0.012)

2a/BCl₃(0.030; 0.012)

3.1 × 10⁴

1.2 × 10⁴

3.7 × 10⁴

4.0 × 10⁴

2.5 × 10⁴

7165

3083

1.2 × 10⁴

9942

2866

3468

5525

5401

4000

2700

3331

0

1659

3227

3777

3517

1.0 × 10⁴

1.4 × 10⁴

1.6 × 10⁴

667

385

6662

9^b

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32^d

33^d

34^d

1-octene (1 0)

cyclooctadiene (5)

R -methylstyrene (5)

1-hexene (5)

1-hexene (20)

0

2765

2a/B(OMe)₃(0.030; 0.012)

c

2a/B(hexyl)₃(0.030; 0.012)

3.2 × 10⁴

3.8 × 10⁴

3.5 × 10⁴

2.9 × 10⁴

2.1 × 10⁴

3.7 × 10⁴

1.4 × 10⁴

1.3 × 10⁴

8480

2a/BEt₃(0.015; 0.006)

2a/BPh₃(0.025; 0.010)

2a/BPh₃(0.003; 0.005)

2b/BPh₃(0.003; 0.005)

2a/BPh₃(0.008; 0.005)

2b/BPh₃(0.004; 0.005)

2a/BPh₃(0.020; 0.010)

2b/BPh₃(0.020; 0.010)

2a/B(C₆F₅)₃(0.025; 0.010)

2a/BH₃THF (0.026; 0.010)

2a/BPh₃(0.026; 0.010)

2a/BPh₃(0.066; 0.010)

2a/B(C₆F₅)₃(0.04; 0.005)

2a/B(C₆F₅)₃(0.14; 0.012)

2a/B(C₆F₅)₃(0.03; 0.005)

2a/B(C₆F₅)₃(0.025; 0.01)

2a/B(C₆F₅)₃(0.020; 0.01)

2a/B(C₆F₅)₃(0.19; 0.016)

2.9 × 10⁴

3.1 × 10⁴

1.1 × 10⁴

1.4 × 10⁴

3260

2544

3343

1348

2000

946

2500

517

1-hexene (20)

1-octene (10)

1-octene (20)

cyclohexene (5)

1-hexene (5)

3,3-dimethylbutene (5)

allylbenzene (2)

allyltrimethylsilane (2)

1-methyl- 1-cyclohexene (1)

2,3-dimethyl-2-butene (2)

1-hexene (5)

5.6 × 10⁴

1.6 × 10⁴

2.4 × 10⁴

237

312

1551

50

7

36

2013

384

190

2013

384

152

cyclohexene (5)

1-methyl-1-cyclohexene (1)

^aReactions were performed using 1-20 mL of the alkene, x mol % 2 (0.005-0.016 mmol), and 5x mol % cocatalyst and monitored by a Buechi

Pressﬂow controller under 10 bar of H₂at 90 °C. ^bWith 10 equiv of BH₃THF relative to 2b. ^cIn situ prepared. ^dUnder 1 bar of H₂at 90 °C.

Gutmann-Beckett method¹⁵and was found to follow the order

BCl₃(129.2%) > BH₃(124.8%) > BEt₃(104.4%) ≈ BF₃

(104.4%) > B(C₆F₅)₃(100%) > BPh₃(77.8%) . B(OMe)₃

(4.2%) (in parentheses is the Gutmann-Beckett value (%) with

reference to B(C₆F₅)₃). Weaker Lewis acids such as B(OMe)₃

did not reveal any rate enhancement with respect to the catalyses

of the rhenium complexes alone. Strong Lewis acids, such as

BF₃·OEt₂, did show an effect but were found less active than

BH₃·THF. In the case of the stronger Lewis acid BCl₃,

hydrogenations even failed. The weak Lewis acid BPh₃turned

out to be the most efﬁcient cocatalyst, also demonstrating the

longest lifetime of the active species. When 5 mL of 1-hexene

was treated with 0.025 mol % 2a and 0.125 mol % BPh₃under

10 bar of H₂, a TON of 3517 and a TOF of 3.5 × 10⁴h^-1were

obtained at 90 °C within 6 min. Repeated addition of 10 mL of

1-hexene (three times) to the resultant solution still afforded

TONs of 6849, 7085, and 6652 corresponding to TOFs of 2.3

× 10⁴, 2.1 × 10⁴, and 1.3 × 10⁴h^-1, respectively. Thus, the

loading of 2 could be further decreased to 0.003 mol % in the

hydrogenation of 1-hexene. In the case of 1-octene, a low

catalyst loading of 0.004-0.008 mol % resulted in both high

TONs and high TOFs (Table 2). The relatively bulky Lewis

acid B(C₆F₅)₃of medium strength afforded the highest TOF (3.6

min, 84%, 5.6 × 10⁴h^-1) in the hydrogenation of 1-hexene at

90 °C. However, readdition of 1-hexene resulted in a much

slower reaction rate, meaning that the catalytic system had a

limited stability. The hydrogenation of the more sterically

hindered alkenes, such as 1-methyl-1-cyclohexene, proceeded

fast with the “2a/B(C₆F₅)₃” system (0.14 mol %, 1:5), affording

a relatively good TOF of 1551 h^-1within 20 min corresponding

to 72% conversion (Table 2, entry 30). In the case of the most

hindered alkene 2,3-dimethyl-2-butene, unfortunately no reaction

occurred at all. The 2a/B(C₆F₅)₃system is so active that the

catalysis could even be carried out under 1 bar of H₂. For

example, a TON of 2013 and a TOF of 2013 h^-1were achieved

within a reaction time of 1 h in the hydrogenation of 1-hexene

under 1 bar of H₂at 90 °C (Table 2, entry 32), and a TON of

384 and a TOF of 384 h^-1within 1 h were obtained in the case

of cyclohexene under the same conditions (Table 2, entry 33).

Even the trisubstituted oleﬁn 1-methyl-1-cyclohexene furnished

an acceptable TON of 190 and TOF of 152 h^-1within 75 min

under 1 bar of H₂at 90 °C (Table 2, entry 34). No catalytic

activity was observed in the hydrogenation of tetrasubstituted

oleﬁn 2,3-dimethyl-2-butene under the same conditions.

2. Ambient Temperature Hydrogenations. We next compared

the catalytic activities of different systems for the hydrogenation

of 1-hexene (5 mL) at ambient temperature (23 °C). With a

(15) (a) Beckett, M. A.; Brassington, D. S.; Coles, S. J.; Hursthouse, M. B.

Inorg. Chem. Commun. 2000, 3, 530. (b) Gutmann, V. Coord. Chem.

ReV. 1976, 18, 225.

9

18236 J. AM. CHEM. SOC. VOL. 132, NO. 51, 2010

Cocatalysis of Alkene Hydrogenations

A R T I C L E S

(1:1, 20 min), 1571 h^-1(1:2, 1 h), 5807 h^-1(1:3, 18 min), 7978

h^-1(1:4, 48 min), and 1.6 × 10⁴h^-1(1:5, 24 min), respectively.

There is a sharp increase in the performance from the 1:3 ratio

(46%) to the 1:4 ratio (over 80%). The presence of a Re-H

bond proved to be crucial to the catalyses, since the hydrogena-

tion of 1-hexene at 23 °C catalyzed by the 1a/BEt₃system is

much slower (TOF of 200 h^-1, 30 min) than with the 2a/BEt₃

system. The 1a/B(C₆F₅)₃combination showed no reaction at

room temperature at all. Other alkenes, such as 1-octene,

cyclohexene, cyclooctene, 1,5-cyclooctadiene, styrene, and 3,3-

dimethylbutene, could also be hydrogenated at ambient tem-

perature, affording satisfactory TOFs using 1a/DMAB, 2a/

B(C₆F₅)₃, or 2a/BH₃·THF as the catalyst system, as depicted

in Table 3. It should be noted that isomerization of alkenes was

not observed in all these cases when the reaction was performed

under a H₂atmosphere and at ambient temperature. Similarly,

applying the most active catalytic system of our series, 2a/

B(C₆F₅)₃, we found hydrogenations to perform surprisingly good

at ambient temperature under 1 bar of H₂. A TON of 2875 and

a TOF of 1725 h^-1were achieved within 100 min in the

hydrogenation of 1-hexene, and a TON of 230 and a TOF of

230 h^-1were obtained within 1 h in the case of cyclohexene.

Mechanistic Studies. 1. Deuterium Isotope Studies. Deute-

rium isotope kinetics were pursued with 10 bar of H₂or D₂in

the hydrogenation of 1-hexene with 2a/BH₃·THF (0.030 mol

%, 1:5) at different temperatures. At 23 °C, a kinetic isotope

effect (KIE) value of 1.11 was determined with initial TOFs of

6107 h^-1(H₂) and 5521 h^-1(D₂) within 18 min. The reactions

at 60 and 90 °C proceeded much faster (1.5-1.6 × 10⁴h^-1

within 9-10 min at 60 °C and 3.1-3.7 × 10⁴h^-1within 5-6

min at 90 °C), giving similar but on an absolute scale small

KIE values of 1.07 (60 °C) and 1.19 (90 °C). In general, very

small kinetic isotope effects were seen, which might imply that

hydrogen-related transformations such as H-H splitting, ꢀ

hydride shift (oleﬁn insertion), and reductive elimination are

not rate-determining steps, since KIEs with rate-determining H₂

splitting are generally high,¹⁷and those of reductive eliminations

of alkanes are often inverse.¹⁸The KIEs of ꢀ hydride shifts

Figure 2. Ambient temperature hydrogenation of 1-hexene (5 mL)

catalyzed by 0.012 mol % 2a/boron Lewis acid: pink solid squares, 2a/

BH₃·THF; yellow solid circles, 2a/DMAB; green solid triangles, 1a/DMAB;

blue inverted solid triangles, 2a/BF₃·OEt₂; purple solid tilted squares, 2a/

B(C₆F₅)₃(1:5); brown left-pointing solid triangles, 2a/BPh₃; green right-

pointing solid triangles, 2a/BEt₃; aqua solid circles, 2a/B(C₆F₅)₃(1:4); blue

solid stars, 2a/B(C₆F₅)₃(1:3); orange solid circles, 2a/B(C₆F₅)₃(1:2); purple

solid circles, 2a/B(C₆F₅)₃(1:1).

catalyst loading of 0.025 mol % 1a or 2a, the hydrogenation

did not occur at all at 23 °C, but could be initiated at high

temperature, probably inducing phosphine dissociation or an

isomerization to a geometry with cis alignment of the alkene

and the hydride.¹⁶In the presence of 0.06 mol % boron Lewis

acid, the hydrogenations catalyzed by 2a (0.012 mol %) occurred

with reaction rates in the order of B(C₆F₅)₃> BEt₃≈ BH₃·THF

> BPh₃> DMAB > BF₃·OEt₂, as depicted in Figure 2. The

2a/DMAB system operated at an extremely slow rate, giving a

TOF of 321 h^-1within 1 h. In contrast, the 2a/BH₃·THF system

showed a satisfying activity with a TOF of 4066 h^-1(61%)

within 1 h. This can be interpreted in terms of the fact that

BH₃·THF is a better Lewis acid than DMAB and that the BH₃

group dissociates more easily from the Me₂NH Lewis base than

from THF. The 1a/DMAB (0.012 mol %) system worked even

better than 2a/BH₃·THF at 23 °C, giving a TON of 2803 and

a TOF of 1.7 × 10⁴h^-1within 10 min, as BH₃can be freed in

the process from 1a to 2a.¹⁴The disadvantage of the 1a/DMAB

system is its drop-off in catalysis, with a tremendously slowed

reaction after 35% conversion. The weak Lewis acid BPh₃

performed not as efﬁciently at 23 °C as at 90 °C, giving a TOF

of 812 h^-1. The better Lewis acid BEt₃showed an improved

activity with a TOF of 4901 h^-1within 1 h. The most active

cocatalyst for the hydrogenation of 1-hexene at 23 °C proved

to be the medium Lewis acid B(C₆F₅)₃, which afforded a TON

of 6321 and a TOF of 1.6 × 10⁴h^-1within 24 min,

corresponding to 79% conversion. In contrast, the stronger Lewis

acid BF₃·OEt₂afforded at this temperature the lowest TOF of

107 h^-1within 1 h, which might be due to complex irreversible

reactions with 2a or halide exchange with ﬂuoride, among

others.

L_nM(H)(CH₂d CHR¹) f L_nM(CH₂s CH₂R¹)

are in most cases small,¹⁹which leaves some ambiguity of

whether the oleﬁn insertion is rate-determining in the catalyses

or other steps which do not involve X-H bond breakages.

Ligand replacements, for instance, are in hydrogenations not

expected to be associated with any primary KIE₁, but since the

(17) (a) Parkin, G. Acc. Chem. Res. 2009, 42, 315. (b) Sandoval, C. A.;

Ohkuma, T.; Muniz, K.; Noyori, R. J. Am. Chem. Soc. 2003, 125,

13490. (c) Iuliis, M. Z. D.; Morris, R. H. J. Am. Chem. Soc. 2009,

131, 11263.

(18) (a) Jones, W. D. Acc. Chem. Res. 2002, 36, 140. (b) Churchill, D. G.;

Janak, K. E.; Wittenberg, J. S.; Parkin, G. J. Am. Chem. Soc. 2003,

125, 1403. (c) Buchanan, J. M.; Stryker, J. M.; Bergman, R. G. J. Am.

Chem. Soc. 1986, 108, 1537. (d) Periana, R. A.; Bergman, R. G. J. Am.

Chem. Soc. 1986, 108, 7332. (e) Gould, G. L.; Heinekey, D. M. J. Am.

Chem. Soc. 1989, 111, 5502. (f) Bullock, R. M.; Headford, C. E. L.;

Hennessy, K. M.; Kegley, S. E.; Norton, J. R. J. Am. Chem. Soc. 1989,

111, 3897.

(19) (a) Mitchell, G. P.; Tilley, T. D. J. Am. Chem. Soc. 1998, 120, 7635.

(b) Vela, J.; Vaddadi, S.; Cundari, T. R.; Smith, J. M.; Gregory, E. A.;

Lachicotte, R. J.; Flaschenriem, C. J.; Holland, P. L. Organometallics

2004, 23, 5226. (c) van der Boom, M. E.; Higgitt, C. L.; Milstein, D.

Organometallics 1999, 18, 2413. (d) Evans, J.; Schwartz, J.; Urquhart,

P. W. J. Organomet. Chem. 1974, 81, C37. (e) Ikariya, T.; Yamamoto,

A. J. Organomet. Chem. 1976, 120, 257. (f) Burger, B. J.; Thompson,

M. E.; Cotter, W. D.; Bercaw, J. E. J. Am. Chem. Soc. 1990, 112,

1566. (g) Berke, H.; Hoffmann, R. J. Am. Chem. Soc. 1978, 100, 7224.

3. Inﬂuence of the Cocatalyst Loading. The inﬂuence of the

cocatalyst loading was examined exemplarily for the 2a/B(C₆F₅)₃

system. In the range of 1:1 to 1:5 ratios of 2a and B(C₆F₅)₃, the

reaction proceeded with increasing rates with TOFs of 507 h^-1

(16) Jiang, Y. F.; Blacque, O.; Fox, T.; Frech, C. M.; Berke, H.

Organometallics 2009, 28, 4670.

9

J. AM. CHEM. SOC. VOL. 132, NO. 51, 2010 18237

A R T I C L E S

Jiang et al.

Table 3. Hydrogenation of Alkenes Catalyzed by Re/Boron Lewis Acid at Ambient Temperature^a

entry

alkene (vol, mmol)

catalyst (amt, mmol)

T (min)

TON

TOF (h^-1

)

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21^c

22^c

1-hexene (40)

1-octene (10)

cyclohexene (10)

cyclohexene (5)

cyclohexene (10)

1,5-cyclooctadiene (3)

cyclooctene (10)

styrene (10)

3,3-dimethylbutene (10)

1-hexene (5)

cyclohexene (5)

1a (0.010)

2a (0.010)

2a/DMAB (0.005)

2a/BH₃THF (0.005)

2a/BEt₃(0.005)

60

10

3

24

3

5

15

10

5

0

321

4066

4901

812

2803

357

6321

1026

1750

1964

464

0

321

4066

4901

812

2a/BPh₃(0.005)

1a/DMAB (0.005)

1b/DMAB (0.005)

2a/B(C₆F₅)₃(0.005)

2b/B(C₆F₅)₃(0.005)

1a/DMAB (0.005)

2a/B(C₆F₅)₃(0.005)

1a/DMAB (0.005)

2a/BH₃THF (0.010)

2a/B(C₆F₅)₃(0.005)

2a/BH₃THF (0.005)

1a/DMAB (0.005)

2a/B(C₆F₅)₃(0.005)

1a/DMAB (0.005)

2a/B(C₆F₅)₃(0.005)

2a/B(C₆F₅)₃(0.012)

1.7 × 10⁴

7140

1.6 × 10⁴

2.0 × 10⁴

2.1 × 10⁴

7856

2784

200

2400

5

1196

491^b

1200

161

1.4 × 10⁴

20

30

5

12

5

1473

2400

1932

200

1000

1339

2875

230

1.6 × 10⁴

100

60

1725

230

^aReactions were performed using 3-40 mmol of alkene, x mmol of Re, and 5x mmol of cocatalyst under 10 bar of H₂at ambient temperature. ^bA 2

equiv amount of H₂was consumed to produce cyclooctane. ^cUnder 1 bar of H₂at 23 °C.

determined KIEs are slightly greater than 1.0, we might be

inclined to state that the oleﬁn insertion into Re-H could to a

very small extent participate in the rate-determining step(s) of

the described catalyses and cause a small KIE₁.

h^-1within 5 min and full conversion within 12 min, which

further supported the hypothesis of a homogeneous reaction

course.

Mercury poisoning tests were also carried out on the “2a/

BPh₃” and “2a/B(C₆F₅)₃” catalytic systems. In the presence of

50 equiv of Hg (per Re atom), the hydrogenation of 1-hexene

using the 2a/B(C₆F₅)₃cocatalytic system proceeded at 23 °C

under either 1 or 10 bar of H₂as fast as the reference system

with no mercury added. No reduction of the activity was noticed

in the catalytic hydrogenation with the 2a/BPh₃system at 90

°C under 10 bar of H₂in the presence of excess mercury. The

ﬁltration and mercury poisoning tests were thus all negative,

ruling out colloid and amalgam formation. This is consistent

with the fact that rhenium metal has a very high atom binding

energy (second largest in the Periodic Table of Elements) with

low propensity for colloid and amalgam formation.²¹

2. Filtration and Mercury Poisoning Experiments. To ex-

clude any heterogeneous reaction course in the hydrogenation

catalyses, ﬁltration experiments were carried out.²⁰At 50%

conversion (TON of 2000) of the hydrogenation of 1-hexene

(5 mL) catalyzed by the 2a/BH₃·THF system at 90 °C under

10 bar of H₂, the reaction solution was found to be of a light

brown color and clear without any visible precipitate. Repeated

trials showed no black precipitate or dark solutions expected

for metallic or colloidal rhenium. The ³¹P NMR spectrum of

the ﬁltrate indicated that 3a remained in solution as the major

rhenium-containing species (66% by integration) along with

10% free PiPr₃and a 24% concentration of one yet unknown

organometallic species showing a singlet at δ 55.4 ppm. The

solution was ﬁltered through Celite into a new vessel with a

new stirring bar, and 2 mL of additional 1-hexene was added.

The catalysis was conducted again at 90 °C under 10 bar of

H₂, revealing a high TOF value of 1.2 × 10⁴h^-1within 10

min, which indeed implies a homogeneous reaction course. The

slight reduction of the originally higher activity is most probably

due to the loss of the active rhenium species via yet unidentiﬁed

side reactions during the ﬁltration process. The catalytic potential

of the Celite after the ﬁltration step was examined in the

hydrogenation of 1-hexene at 90 °C under 10 bar of H₂. It

proved to be catalytically inactive, implying the absence of

catalytically active rhenium deposits. Similar results were

obtained under the same conditions in the hydrogenation of

1-hexene using the 2a/BPh₃system. At 53% conversion within

3 min, the reaction solution was ﬁltered through Celite, and 2

mL of 1-hexene was added. The ﬁltrate still showed quite

acceptable catalytic performance, affording a TOF of 1.5 × 10⁴

3. Boron Lewis Acids as Halide Abstractors. In this context

the function of the boron Lewis acid is more closely examined.

A few respective observations are summarized in the following:

(1) The stoichiometric reaction between 2a,b and BH₃·THF in

1-hexene immediately afforded the 18 e^-rhenium(I) hydride

complex [Re(Br)(H)(NO)(PR₃)₂(η²-CH₂dCHBu)] (3; R ) iPr,

a; Cy, b). (2) Formation of P-B adducts, such as iPr₃P·BH₃

and iPr₃P·BCy₃, was not observed within the NMR detection

limits.²²(3) After the autoclave was ﬁlled with 10 bar of H₂,

hydrogenation took place at ambient temperature without any

induction period. The courses of the reactions were examined

by ³¹P NMR spectroscopy at 50% conversion, providing

evidence that 3a and 3b are the major organometallic species

stable in solution. (4) The ¹¹B NMR spectra indicated that the

cocatalyst BH₃·THF underwent hydroboration during catalysis

and that this species was completely transformed into trihexy-

(21) Brewer, L. Science 1968, 161, 115.

(22) Yi, C. S.; Lee, D. W.; He, Z.; Rheingold, A. L.; Lam, K.-C.; Concolino,

T. E. Organometallics 2000, 19, 2909.

(20) Widegren, J. A.; Finke, R. G. J. Mol. Catal. A 2003, 198, 317.

9

18238 J. AM. CHEM. SOC. VOL. 132, NO. 51, 2010

Cocatalysis of Alkene Hydrogenations

A R T I C L E S

lborane (s, 92 ppm). A mixture of 3a and B(hexyl)₃was then

separately prepared and tested for hydrogenation of 1-hexene,

showing activities similar to those of the 2a/BH₃·THF and 2a/

BEt₃systems (entry 16, Table 2). Similarly, the hydrogenation

of cyclohexene catalyzed by the 2a/BH₃·THF system also

furnished initial hydroboration and formation of Cy₃B (s, 81.5

ppm, in the ¹¹B NMR spectrum). The hydroboration of cyclo-

hexene is thus faster than the hydrogenation reaction. Using

BPh₃as a cocatalyst provided no evidence that phosphine

abstraction from rhenium could have taken place either at 23

°C or at 90 °C. In separate experiments it was also conﬁrmed

that BPh₃does not form adducts with PiPr₃, also not in the

presence of H₂or alkene. All these results point to primary

generation of alkylboranes, when possible phosphine abstraction

could however not be induced by action of the cocatalyst and

is therefore not the activation process. It should be mentioned

at this point that the most active cocatalyst B(C₆F₅)₃caused

during catalysis insoluble precipitates of yet unidentiﬁed

structures. This precipitate turned out to be inactive in catalytic

hydrogenations of 1-hexene in THF solutions. However, yet

unknown homogeneous deactivation pathways may be another

cause for the shorter lifetimes of the B(C₆F₅)₃cocatalytic

systems.

The halide substitution proceeds in the order I f Br f Cl f

F, which is interpreted in terms of thermodynamics with an

increasing Re-X bond strength in this order. In the ³¹P NMR

spectra the chemical shifts of the phosphine ligands were found

shifted to low ﬁeld in the sequence of I, Br, Cl, and F, for

example, from δ 10.7 ppm (d, ²J_PP) 122 Hz) and 4.2 ppm (d,

2

²J_PP) 122 Hz) in 3a-I to 26.1 (dd, J_PP) 145 Hz, J_FP) 31

Hz) and 19.3 (dd, ²J_PP) 145 Hz, ²J_FP) 31 Hz) in 3a-F (Figure

3). This trend looks reasonable, since the electronegativity of

the halide increases from I to F, resulting in a decrease of the

electron density at the phosphorus atom.

The in situ prepared halide derivatives including 3a-I, 3a-

Cl, and 3a-F were tested as catalysts in combination with 5

equiv of BH₃·THF as a cocatalyst to hydrogenate 1-hexene at

90 °C under 10 bar of H₂. Their activities were indeed found to

increase with decreasing electronegativity of the halide X. For

example, a TON of 2982 and TOF of 1.8 × 10⁴h^-1were

achieved with the 3a-Cl/BH₃·THF (1:5) system within 10 min,

corresponding to a 78% conversion, which is merely half as

active as the 2a/BH₃·THF system. The iodide case in the form

of the 3a-I/BH₃·THF (1:5) system digressed from the given

trend of the halide order. A TON of 4600 and a TOF of 2.8 ×

10⁴h^-1were achieved within 10 min. Although the 3a-I/

BH₃·THF system was expected to be superior to the 3a/

BH₃·THF system, a lower activity was observed for the earlier.

This observation could not be attributed to the presence of small

amounts of a 3a-F contaminant; rather it is assumed that the

relatively hard Lewis acids BH₃and BR′₃are only moderate

abstractors for the soft iodide ion. The 3a-F/BH₃·THF (1:5)

system constitutes the worst case within this halide catalysis

series. A TOF of 2937 h^-1was accomplished within 1 h. The

Lewis acid is anticipated to be an effective ﬂuoride abstractor.

However, the Re-F bond is very strong, as we derived from

the thermodynamically determined exchange experiments. The

superior role of the bromide as the “best” halide in this series

of “Re(I) complex/boron Lewis acid” systems is interpreted in

terms of a proper balance between the Re-Br bond strength

and the still relatively high afﬁnity of BH₃or BR′₃species for

the Br^-ion. These results suggest that the reversible abstraction

of the bromide ligand by the boron Lewis acid is most probably

the initiation step in these hydrogenation processes to generate

the catalytic species driving the cycle. The cocatalyst determines

the initiation process of the catalyses. In many types of

organometallic catalyses the initiation processes are kinetically

relevant. For the actual case, we can state that the initiation

rate is high when the Lewis afﬁnity of the cocatalyst is high.

However, for the beneﬁt of high TON values or the longevity

of real catalytic species, the halide afﬁnity of the boron Lewis

acid should be intermediate. Strong Lewis acids, such as BCl₃

and BF₃, were also suspected to interact with the substrate

(alkene) or with the hydride ligand.²³For our systems these

interactions seemed kinetically not relevant, since the halide

interactions were expected to dominate. The interaction between

the boron Lewis acid and the bromide ligand in 3 could however

not be traced. Variable temperature (VT) NMR measurements

in the temperature range of -80 to +23 °C of 3a with B(C₆F₅)₃,

BH₃·THF, BEt₃, or BPh₃in excess 1-hexene did not provide

any evidence for the formation of a new species in concentra-

tions above the NMR detection limit.

4. Halide Abstraction as the Crucial Mechanistic Step. We

next examine whether the dissociation of the bromide ligand

might get crucially involved in the catalytic pathway. In the

presence of 5 equiv of nBu₄NBr, the hydrogenation of 1-hexene

catalyzed by the 2a/BH₃·THF (1:10) system proceeded rapidly,

showing an initial TOF value of 1.6 × 10⁴h^-1within 8 min.

However, the reaction decelerated after a TOF of 8235 h^-1was

reached within 20 min. A similar rate drop was observed in the

presence of an excess of various other halide salts nBu₄NX (X

) I, Cl, F). In general, the extent of the inhibitation was found

to correlate with increasing electronegativity of the halide in

the order F > Cl > Br ≈ I. For example, the rate inhibitation

was minor in the presence of excess nBu₄NI so that a TOF of

2.3 × 10⁴h^-1was reached within 5 min and a TOF of 7698

h^-1was reached within 20 min. In the case of nBu₄NCl, the

reaction rate was diminished to a great extent, showing TOFs

of 1.8 × 10⁴h^-1within 4 min and 2866 h^-1within 1 h. The

catalysis in the presence of excess nBu₄NF turned out to be

very sluggish, showing a TOF of 539 h^-1within 6 h, which is

even slower than the hydrogenation catalyzed by “plain” 1a or

2a.

This led us to investigate stoichiometric transformations of

the rhenium catalyst in the presence of halide salts nBu₄NX.

The reaction of 2a with nBu₄NCl (5 equiv) and 1-hexene (5

equiv) in benzene afforded at 85 °C within 5 min the 18 e^-

rhenium(I) monochloro hydride complex [Re(Cl)(H)(NO)-

(PiPr₃)₂(η²-CH₂dCHBu)] (3a-Cl) in over 99% in situ yield. The

reverse reaction from 3a-Cl to 3a by chloride replacement with

bromide failed even with a large excess of nBu₄NBr and

prolonged reaction times. Addition of nBu₄NF to the solution

of 3a-Cl or 3a in benzene instantaneously afforded at room

temperature the corresponding 18 e^-ﬂuoride derivative [Re(F-

)(H)(NO)(PiPr₃)₂(η²-CH₂dCHBu)] (3a-F) in over 99% in situ

yield. Although the direct substitution of bromide with iodide

via the reaction of 3a with nBu₄NI failed, the iodide analogue

[Re(I)(H)(NO)(PiPr₃)₂(η²-CH₂dCHBu)] (3a-I) could be ac-

cessed by treatment of 3a with a small excess of AgSbF₆

followed by reaction with nBu₄NI, although this product was

contaminated by a small amount of 3a-F formed through

substitution of iodide with ﬂuoride from the AgSbF₆reagent.

5. H₂/D₂Scrambling Catalyzed by 2a. To gain some insight

into the interaction between the rhenium complex and hydrogen,

(23) Kawata, N.; Maruya, K.; Mizoroki, T.; Ozaki, A. Bull. Chem. Soc.

Jpn. 1974, 47, 413.

9

J. AM. CHEM. SOC. VOL. 132, NO. 51, 2010 18239

A R T I C L E S

Jiang et al.

Figure 3. Halide exchange experiments based on 3a monitored by ³¹P NMR spectroscopy at room temperature or 85 °C in benzene. The AB coupling

pattern of the ³¹P NMR spectrum presumably originates from chemically different phosphorus ligands caused by binding of 1-hexene with parallel alignment

with the P-Re-P axis and hindered rotation.^13b

H₂/D₂isotope scrambling experiments were carried out. Previ-

ously we reported that addition of H₂to 2 instantaneously

afforded the unstable rhenium(I) dihydrogen hydride complexes

[Re(Br)(H)(NO)(PR₃)₂(η²-H₂)] (4; R ) iPr, a; Cy, b).¹⁴When

a solution of 2a and 5 equiv of B(C₆F₅)₃in toluene in an NMR

tube was purged with a 1:1 mixture of H₂and D₂(each 600

mbar, 23 °C), an instantaneous H₂/D₂scrambling was observed

displaying singlets at 41.72, 41.66, 41.53, and 41.44 ppm. With

1

the aid of a 2D H,³¹P correlation spectrum, the downﬁeld

signals at 41.72 and 41.66 ppm could be assigned to Re-H/

H₂, Re-H/HD or Re-H/D₂species, whereas the signals at 41.53

and 41.44 ppm were attributed to the Re-D/H₂and Re-D/HD

species, showing no ³¹P,¹H cross-peak.

The reaction of the 2a/B(C₆F₅)₃system with D₂in toluene

led to the spontaneous formation of a D_Resignal. Similarly, VT

²H NMR spectroscopy showed a broad coalescence signal at

3.40 ppm at room temperature, which was resolved by decoa-

lescence in the temperature range of -40 to -60 °C into a

singlet of free H₂at 4.50 ppm, a small HD doublet at 4.55 ppm,

and two broad signals at 3.56 and 3.13 ppm, which are assigned

to various Re-D (T₁) 22 ms) and Re-D₂(T₁) 24 ms)

isotopomers. The quadrupolar deuterium relaxation of Re-D₂

is obviously reduced by rotation and, therefore, is slightly less

efﬁcient than in the case of Re-D.²⁴The Re-H, Re-H₂, and

Re-HD assignments of the ¹H NMR spectrum at -50 °C could

also be conﬁrmed by ¹H T₁relaxation time measurements. Here

1

using VT H NMR measurements in the temperature range of

1

+23 to -50 °C. At room temperature, the H NMR spectrum

showed coalescence of the free H₂and the rhenium-bound H₂

signal merging into a broad singlet at 3.83 ppm, indicative of

fast exchange between free H₂and coordinated H₂in 4. When

the temperature was decreased to -50 °C, not only was the

free hydrogen signal observed as a singlet at 4.50 ppm, but also

the hydrogen scrambling product HD was detected, displaying

the well-known triplet resonance at 4.46 ppm (J_HD) 43 Hz, T₁

) 827 ms). The ¹H NMR spectrum (500.25 MHz) showed two

Re-H triplets at 3.44 ppm (J_HP) 26 Hz, T₁) 264 ms) and

3.42 ppm (J_HP) 25 Hz, T₁) 260 ms). Rhenium-coordinated

H-H and H-D moieties were observed as a broadened

multiplet at 3.07 ppm (H₂, T₁) 116 ms) and a broad (>80 Hz)

signal at 2.73 ppm (HD, T₁) 501 ms), respectively. The ³¹P

NMR spectrum revealed the formation of four H,D isotopomers,

1

the H relaxation in Re-H₂is very efﬁcient (T₁) 116 ms),

(24) Facey, G. A.; Fong, T. P.; Gusev, D.; Macdonald, P. M.; Morris, R. H.;

Schlaf, M.; Xu, W. Can. J. Chem. 1999, 77, 1899.

9

18240 J. AM. CHEM. SOC. VOL. 132, NO. 51, 2010

Cocatalysis of Alkene Hydrogenations

A R T I C L E S

Scheme 1. Proposed Reaction Path for the H₂/D₂Scrambling with the 2a/B(C₆F₅)₃System

because it is rotation accelerated and based on mutual H-¹H

dipolar interaction. In contrast, the ¹H relaxation in the Re-HD

unit is slow (T₁) 501 ms) due to a weak proton-deuterium

dipolar interaction, whereas the relaxation time of Re-H is in

the range of about 260 ms, which is typical for M-H bonds.

Under a D₂atmosphere, ³¹P NMR shows a prominent singlet

of the Re-D/D₂unit at 41.40 ppm, which corresponds to the

high-ﬁeld signal at 41.44 ppm under a H₂/D₂atmosphere.

Besides the given one, at least two more isotopomers are present

in solution, displaying weak resonances at 41.47 and 41.63 ppm.

In total, these NMR studies support a rhenium trihydride

exchange mechanism²⁵induced by bromide abstraction assisted

by the boron Lewis acid, as depicted in Scheme 1.²⁶2a alone

is also a catalyst in the H₂/D₂scrambling at room temperature,

although it is less efﬁcient than the 2a/B(C₆F₅)₃system,

temperature for 24 h, the solution turned deep brown with

formation of the internal alkene isomer (E/Z)-2-hexene in over

90% yield, showing a characteristic multiplet signal at 5.3 ppm

1

in the H NMR spectrum. Unfortunately, the 2-hexene isomer

could not be distinguished as either (E)-2-hexene or (Z)-2-

hexene due to the overlap of the C-H protons in the region of

5.3 ppm. However, in a ¹³C, ¹H correlation spectrum the scalar

couplings between the multiplet H_alkenesignals and the three

C_alkeneatoms (δ ) 132.0, 130.6, 124.7 ppm) were observable,

which indicated the presence of both the (E)- and (Z)-2-hexene

isomers. It should also be added that 2-hexene coordinates rather

weakly to the rhenium center in comparison with 1-hexene;

therefore, the ﬁve-coordinate Re(I) species 2a was partially

recovered in solution, as indicated by ³¹P NMR spectroscopy.

In contrast, isomerization of 1-hexene did not take place at all

with 2a alone at room temperature, even at prolonged reaction

times. Nevertheless, a nearly complete conversion to the

2-hexenes could be achieved by 2a alone at 85 °C within 5

min, which implied that the required trans/cis rearrangement

of the Re(H)(oleﬁn) unit could also be thermally induced. These

results all point to a reversible alkene migratory insertion of

the alkenes into the Re-H bond involved in catalyses, as

depicted in Scheme 2. The key species are 6 and 6′ containing

a cis-aligned alkene and hydride ligand. The complex salt 6 is

formed via the boron Lewis acid induced bromide abstraction,

whileitsneutralcongener6′isformedbyaBr^-abstraction-trans/

cis rearrangement-Br^-recoordination sequence supported by

the Lewis acid. Then the migratory insertion of the oleﬁn into

the Re-H bond takes place with different orientations of the

oleﬁnic ligand. The formed primary or secondary alkyl species

undergo ꢀ-hydride shift, leading to the 1-alkene or to the

2-alkene.

1

furnishing a much slower H₂/D₂conversion to HD in the H

NMR spectrum. This can be interpreted in terms of an only

slow intramolecular rearrangement from the trans-Re(H)(H₂)

to the cis-Re(H₃) transition state, in which the scrambling can

take place. Exactly this geometrical rearrangement is anticipated

to be promoted by Br^-abstraction with the Lewis acid.

6. Catalysis of the Terminal Alkene Isomerization. Besides

H₂/D₂scrambling, the 2a/B(C₆F₅)₃system was also found to

be active in the isomerization of terminal alkenes to internal

ones, the so-called one-carbon migration process:²⁷

When 10 equiv of 1-hexene was added to a solution of 2a/

B(C₆F₅)₃(1:5) in toluene, the purple solution instantaneously

turned light yellow, forming 3a as the only organometallic

species (NMR monitoring). After being allowed to stand at room

Besides the boron Lewis acid promoted activation of the

rhenium catalysts, another standard protocol was applied to

activate these complexes, replacing the halide ligand with a

(25) (a) Kubas, G. J. Acc. Chem. Res. 1988, 21, 120. (b) Luo, X.; Crabtree,

R. J. Am. Chem. Soc. 1990, 112, 6912.

(27) (a) Gauthier, D.; Lindhardt, A. T.; Olsen, E. P. K.; Overgaard, J.;

Skrydstrup, T. J. Am. Chem. Soc. 2010, 132, 7998. (b) Donohoe, T. J.;

O’Riordan, T. J. C.; Rosa, C. P. Angew. Chem., Int. Ed. 2009, 48,

1014.

(26) (a) Shapley, J. R.; Osborn, J. A. Acc. Chem. Res. 1973, 6, 305. (b)

Rossi, A. R.; Hoffmann, R. Inorg. Chem. 1975, 14, 365.

9

J. AM. CHEM. SOC. VOL. 132, NO. 51, 2010 18241

A R T I C L E S

Jiang et al.

Scheme 2. Proposed Reaction Pathway for the Isomerization of 1-Hexene to (E)- and (Z)-2-Hexene Catalyzed by the 2a/B(C₆F₅)₃System or

2a Alone

Table 4. Hydrogenation of 1-Hexene at 90 °C Using Catalyst 2a

weakly coordinating anion.²⁸When the hydrogenation of

1-hexene was carried out at 90 °C under 10 bar of H₂with a

mixture of 2a and a slight excess of AgSbF₆, no reaction

occurred at all. In contrast, the in situ ﬁltered solution of 2a

and AgSbF₆(2 equiv) in 1-hexene exhibited excellent catalytic

activity, affording a satisfactory TON of 3446 and a TOF of

1.4 × 10⁴h^-1within 15 min, which is 16 times faster than that

with 2a alone. This not only suggests that a “catalytic” vacant

site is created via bromide release, but also implies that even a

slight excess of silver ions could completely block the catalysis.

This is most probably due to the loss of Re(I) active species

via the oxidation of Re(I) to Re(II) by Ag(I) ions. Repeated

addition of 5 mL of 1-hexene to the resultant solution afforded

only moderate TONs and TOFs, indicating degradation of the

catalyst. The stoichiometric reaction of 2a with 2 equiv of

1-hexene and 1 equiv of AgSbF₆was carried out in benzene-d₆

at room temperature. The reaction yielded a mixture of rhenium

species as indicated by ³¹P NMR spectroscopy, which could

not be identiﬁed. The formation of a small amount of (E)- and

with Various Non-Boron Lewis Acids^a

entry

Lewis acid (amt, equiv)

T (min)

conversion (%)

TON

TOF (h^-1

)

1

2

3

4

5

6

7

8

9

AlBr₃(5)

AlBr₃(3)

AlBr₃(1)

AlCl₃(5)

AlCl₃(1)

AlF₃(5)

AlF₃(1)

InCl₃(1)

InCl₃(5)

4

10

8

26

53

89

29

1

83

95

19

27

76

89

58

919 1.4 × 10⁴

1861 1.1 × 10⁴

3102 2.3 × 10⁴

1000 3.0 × 10⁴

50 50

2

60

10

7

40

7

2910 1.7 × 10⁴

3308 2.8 × 10⁴

665 998

946 1419

10

11

12

Yb(OTf)₃(5)

Yb(OTf)₃(1)

Zn(OTf)₂(5)

2647 2.3 × 10⁴

3116 2.7 × 10⁴

2022 8665

7

14

^aReactions were performed using 5 mL of 1-hexene (40 mmol), 7

mg of 2a (0.0113 mmol), and an appropriate amount of non-boron

Lewis acid under 10 bar of H₂at 90 °C.

catalytic performance with AlF₃with respect to AlBr₃may be

related to the higher Lewis afﬁnity of AlF₃over AlBr₃.²⁹

Mechanism of the Re(I)/Boron Lewis Acid Cocatalyzed

Hydrogenations of Alkenes. A mechanism including reversible

bromide abstraction by the Lewis acid is proposed for the

1/DMAB and 2/BR′₃(R′ ) Et, Ph, C₆F₅, “H”) catalyzed

hydrogenations of alkenes, as depicted in Scheme 3. Two

catalytic cycles are proposed depending on whether the Lewis

acid promoted reversible bromide abstraction is involved as an

initiation step (A) and the cycle is turning over “bromide-free”

or whether the bromide is coming back during the course of

the cycle (B). Invariably the alkene-coordinated 18 e^-rhenium(I)

hydride 7 is ﬁrst generated either directly from 2 or indirectly

from 1 with DMAB and the alkene. Then the Lewis acidic BR′₃

component reversibly abstracts the bromide ligand, affording

Re(I) species 8 bearing one vacant site and the bromo borate

counteranion [BBrR′₃]^-. This step serves as the initial step for

1

(Z)-2-hexene was also observed in the H NMR spectrum.

7. Non-Boron Lewis Acids as Cocatalysts. Besides boron

Lewis acids, we also tested other types of Lewis acids as

cocatalysts in 2a catalyzed hydrogenations of 1-hexene under

10 bar of H₂at 90 °C, as listed in Table 4. In general, most of

these performed not comparable to boron Lewis acids with

respect to high conversions, good activities, and longevity of

the catalytic system, except for AlF₃, AlBr₃, and Yb(OTf)₃.

Again the appropriate balance between the Re-Br bond strength

and the Lewis afﬁnity for Br of the Lewis acid is expected to

determine the performance of the catalyses. For example, InCl₃

was even found to inhibit the hydrogenation, which can be

interpreted in terms of InCl₃being such a soft Lewis acid that

replacement of the bromide ligand at rhenium by chloride could

take place, resulting in poorer activity. The slightly better

(28) (a) Strauss, S. H.; Shriver, D. F. Inorg. Chem. 1978, 17, 3069. (b)

Beck, W.; Sunkel, K. Chem. ReV. 1988, 88, 1405. (c) Lassauque, N.;

Francio, G.; Leitner, W. Eur. J. Org. Chem. 2009, 3199.

(29) (a) Branchadell, V.; Sbai, A.; Oliva, A. J. Phys. Chem. 1995, 99, 6472.

(b) Wilson, M.; Coolidge, M. B.; Mains, G. J. J. Phys. Chem. 1992,

96, 4851. (c) Ogawa, A.; Fujimoto, H. Inorg. Chem. 2002, 41, 4888.

9

18242 J. AM. CHEM. SOC. VOL. 132, NO. 51, 2010

Cocatalysis of Alkene Hydrogenations

A R T I C L E S

Scheme 3. Alternative Rhenium Hydrogenation Cycles: (A) Lewis Acid Promoted Bromide-Free Catalysis, (B) Lewis Acid Promoted Bromide

On-Off Catalysis

catalytic cycle A. The Osborn-type cycle begins operation with

the alkene before H₂addition, reaching transition state 12,

through which the alkane is eventually released by reductive

elimination, re-forming 8 to close the catalytic cycle. The

bromide may coordinate back to form the resting state 7. In

this mechanism, the initial step from 7 to 8 would be rate-

determining.³⁰

create a vacant site.³²Interaction of the O_NOatom with Lewis

acids would in principle increase metal to nitrosyl π-back-

donation, enforcing the tendency for NO bending. However,

no experimental evidence for NO bending in any species of

the catalytic cycle and eventual take-up of alkenes or H₂was

obtained. This possibility is currently still under investigation

by density functional theory (DFT) calculation.

An alternative catalytic cycle (cycle B) can be formulated.

Different from cycle A, the Lewis acid promoted reversible

bromide dissociation and association are involved as steps in

the catalytic cycle. Otherwise, the sequence of an Osborn-type

hydrogenation cycle is followed again with the alkene before

H₂addition. Then alkane reductive elimination occurs from

transition state 12′, being distinguished from 12 of cycle A by

a bromide in the coordination sphere. Cycle B is presumably

slower than cycle A, because the bromide on-off process would

in principle retard the reaction rate. It should be noted that the

H₂uptake from 10 (10′) to 11 (11′) could be to some extent

rate-determining, as the equilibria are dependent on the con-

centration of H₂. In this step, the alkene could in principle

occupy the vacant site rather than H₂, yielding the bisalkene-

coordinated rhenium(I) alkyl species as a resting state. As

discussed before, the migratory insertion of the alkene into

Re-H from 9 (9′) to 10 (10′) could also contribute to the overall

rate.

Conclusions

In conclusion, 1/DMAB was discovered in the course of

dehydrocoupling of DMAB to be a very efﬁcient catalytic

system for homogeneous hydrogenation of alkenes at both high

temperature and ambient temperature. Mechanistic studies

revealed that the high activity is derived from cooperation of

the ﬁve-coordinate rhenium(I) hydride 2 with Lewis acids, in

the best cases with boron Lewis acids as cocatalysts. The

inﬂuence of the boron Lewis acid on the total catalytic activity

was found to be in the qualitative order B(C₆F₅)₃> BEt₃≈

BH₃·THF > BPh₃, but the longevity of the catalytic system was

found to be of the reverse order, BPh₃> BEt₃≈ BH₃·THF >

B(C₆F₅)₃, so that activity and longevity have to be compromised

for the best choice of systems. Detailed mechanistic studies were

carried out. The boron Lewis acids assisted reversible bromide

abstraction, which is anticipated to contribute to the rate-limiting

step. It is the initiating step of the catalyses and inﬂuences the

catalytic performance. From the point of view of catalytic

activity and longevity of the catalysts, these rhenium hydrogena-

Another possible Lewis basic site where the Lewis acid could

dock on is the O_NOatom.³¹Nitrosyl ligands may be either linear

or bent at a metal center. The bending of a nitrosyl ligand would

(32) (a) Enemark, J. H.; Feltham, R. D. Coord. Chem. ReV. 1974, 13, 339.

(b) Enemark, J. H.; Feltham, R. D. J. Am. Chem. Soc. 1974, 96, 5002.

(c) McCleverty, J. A. Chem. ReV. 1979, 79, 53. (d) Collman, J. P.;

Farnham, P.; Dolcetti, G. J. Am. Chem. Soc. 1971, 93, 1788. (e)

Roncaroli, F.; Videla, M.; Slep, L. D.; Olabe, J. A. Coord. Chem.

ReV. 2007, 251, 1903. (f) Richter-Addo, G. B.; Wheeler, R. A.; Hixson,

C. A.; Chen, L.; Khan, M. A.; Ellison, M. K.; Schulz, C. E.; Scheidt,

W. R. J. Am. Chem. Soc. 2001, 123, 6314.

(30) (a) Dias, E. L.; Nguyen, S. T.; Grubbs, R. H. J. Am. Chem. Soc. 1997,

119, 3887. (b) Trnka, T. M.; Grubbs, R. H. Acc. Chem. Res. 2001,

34, 18. (c) Ulman, M.; Grubbs, R. H. Organometallics 1998, 17, 2484.

(31) (a) Sharp, W. B.; Legzdins, P.; Patrick, B. O. J. Am. Chem. Soc. 2001,

123, 8143. (b) Gusev, D.; Llamazares, A.; Artus, G.; Jacobsen, H.;

Berke, H. Organometallics 1999, 18, 75.

9

J. AM. CHEM. SOC. VOL. 132, NO. 51, 2010 18243

A R T I C L E S

Jiang et al.

Table 5. Comparison of the Catalytic Performance of Classic Precious Metal Catalysts with the “Rhenium(I) Hydride/Boron Lewis Acid”

System^1a

tion systems belong to the best transition-metal catalysts for

hydrogenations and are comparable in performance to most of

the reported hydrogenation catalyses involving precious metals,

e.g., RuHCl(CO)(PCy₃)₂(TOF of 1200 h^-1for 1-hexene at 25

°C and 1 bar of H₂)^8cand its derivatives,⁸RhCl(PPh₃)₃

(Wilkinson catalyst, TOF of 650 h^-1for 1-hexene at 25 °C and

1 bar of H₂),^1h[Rh(cod)(PPh₃)₂]PF₆(Schrock-Osborn catalyst,

TOF of 4000 h^-1for 1-hexene at 25 °C and 1 bar of H₂),^1i,k

and [Ir(cod)(py)(PCy₃)][PF₆] (Crabtree catalyst, TOF of 6400

h^-1for 1-hexene at 0 °C and 1 bar of H₂).^1aFurther details

concerning substrate selectivities of these catalysts are depicted

in Table 5. It is worth mentioning that the Crabtree catalyst is

remarkably active at room temperature, but is thermally less

stable and susceptible to deactivation through formation of an

inactive hydride-bridged trimer. This work provides great

potential for the application of rhenium in ﬁne-chemical

production, showing hydrogenation activities of the same order

of magnitude as those of precious metals. Despite the fact that

the current cost of rhenium is comparable to that of ruthenium,

it is still much cheaper than rhodium or iridium. We are very

optimistic that in the future rhenium will attain a more important

role in industrial processes. Studies of the scope of functional

group tolerance and of enantioselective hydrogenations are in

progress.

Experimental Section

General Procedures. All manipulations were carried out under

an atmosphere of dry nitrogen using standard Schlenk techniques

or in a glovebox (M. Braun 150B-G-II) ﬁlled with dry nitrogen.

Solvents were freshly distilled under N₂by employing standard

procedures and were degassed by freeze-thaw cycles prior to use.

The deuterated solvents were dried with sodium/benzophenone

(benzene-d₆) and vacuum transferred for storage in Schlenk ﬂasks

ﬁtted with Teﬂon valves. ¹H NMR, ¹³C{¹H} NMR, ³¹P{¹H} NMR,

and ¹¹B NMR data were recorded on a Varian Gemini-300, Varian

9

18244 J. AM. CHEM. SOC. VOL. 132, NO. 51, 2010

Cocatalysis of Alkene Hydrogenations

A R T I C L E S

Mercury 200, or Bruker DRX 500 spectrometer using 5 mm

diameter NMR tubes equipped with Teﬂon valves, which allow

degassing and further introduction of gases into the probe. Chemical

hydrogenation products from the reaction with H₂were character-

ized by H NMR spectroscopy in CDCl₃.

1

Lewis Acidity Tests. Gutmann-Beckett Method in Toluene-

d₈at 23 °C. The boron Lewis acid was coordinated to triethylphos-

phine oxide (Et₃PdO or TPO), and the Lewis acidity was compared

by the change of the chemical shift in the ³¹P{¹H} NMR spectra

(121.47 MHz, ppm): Et₃PdO, δ ) 45.9; (Et₃PdO)(B(C₆F₅)₃), δ

) 75.7, reference shift ∆δ ) 29.8; (Et₃PdO)(BF₃), δ ) 77.0,

reference shift ∆δ ) 31.1, Lewis acidity relative to B(C₆F₅)₃

104.4%; (Et₃PdO)(BH₃), δ ) 83.1, reference shift ∆δ ) 37.2,

Lewis acidity relative to B(C₆F₅)₃124.8%; (Et₃PdO)(BPh₃), δ )

69.1, reference shift ∆δ ) 23.2, Lewis acidity relative to B(C₆F₅)₃

77.8%; (Et₃PdO)(BEt₃), δ ) 77.0, reference shift ∆δ ) 31.1, Lewis

acidity relative to B(C₆F₅)₃104.4%; (Et₃PdO)(BCl₃), δ ) 84.4,

reference shift ∆δ ) 38.5, Lewis acidity relative to B(C₆F₅)₃

129.2%; (Et₃PdO)[B(OMe)₃], δ ) 47.1, reference shift ∆δ ) 1.2,

Lewis acidity relative to B(C₆F₅)₃4.2%.

1

shifts are expressed in parts per million. H and ¹³C{¹H} NMR

spectra were referenced to the residual proton or ¹³C resonances of

the deuterated solvent. All chemical shifts for the ³¹P{¹H} NMR

data are reported downﬁeld in parts per million relative to external

85% H₃PO₄at 0.0 ppm. Signal patterns are reported as follows: s,

singlet; d, doublet; t, triplet; m, multiplet. IR spectra were obtained

by using ATR methods with a Bio-Rad FTS-45 FTIR spectrometer.

Microanalyses were carried out at the Anorganisch-Chemisches

Institut of the University of Zurich. ESI-MS spectroscopic data were

obtained from an HCT Esquire Bruker Daltonics instrument.

Rhenium(I) complexes 1^31band 2^13bwere prepared according to

reported procedures. DMAB, BH₃· THF, BPh₃, BEt₃, BCl₃,

BF₃· OEt₂, B(OMe)₃, AlBr₃, AlCl₃, AlF₃, InCl₃, Yb(OTf)₃,

Zn(OTf)₂, and diverse alkenes were purchased from Aldrich and

Fluka and used without further puriﬁcation. 1-Hexene, 1-octene,

cyclooctene, allylbenzene, and 1,5-cyclooctadiene were puriﬁed by

distillation.

[Re(Br)(H)(NO)(PiPr₃)₂(η²-CH₂dCHBu)] (3a). In a 3 mL

Young NMR tube, 2a (31 mg, 0.05 mmol) and 1-hexene (6.3 µL,

0.05 mmol) were mixed in benzene-d₆(0.5 mL). Immediately a

light brown solution was formed. The NMR spectrum indicated

the formation of 3a in 100% in situ yield. The solvent was

evaporated in vacuo for 24 h to give a light brown residue. Yield:

35 mg, 99%. IR (ATR, cm^-1): ν(C-H) 2961, 2922, 2873, ν(NO)

General Procedure for Hydrogenation of Alkenes Cocata-

lyzed by Rhenium Complexes and Lewis Acids. In a 30 mL steel

autoclave equipped with a stirring bar, a certain amount of substrate

alkene (e.g., 1-hexene, 5 mL, 40 mmol; 1-octene, 10 mL, 64 mmol;

cyclooctene, 10 mL, 77 mmol; 1,5-cyclooctadiene, 5 mL, 41 mmol;

cyclohexene, 5 mL, 49 mmol; R-methylstyrene, 5 mL, 38 mmol; 3,3-

dimethylbutene, 5 mL, 39 mmol; styrene, 10 mL, 87 mmol; 1,7-

octadiene, 5 mL, 34 mmol; allylbenzene, 2 mL, 15 mmol; allyltrim-

ethylsilane, 2 mL, 12 mmol; 1-methyl-1-cyclohexene, 1 mL, 8.5 mmol;

2,3-dimethyl-2-butene, 2 mL, 17 mmol), the appropriate amount of

rhenium catalyst (e.g., 1a, 6.7 mg, 0.01 mmol; 1b, 9.6 mg, 0.01

mmol; 2a, 7.0 mg, 0.0113 mmol; 2b, 8.6 mg, 0.01 mmol), and 5

equiv of Lewis acid (e.g., DMAB, 3.0 mg, 0.05 mmol; BH₃·THF,

50 µL, 0.05 mmol; BPh₃, 12.2 mg, 0.05 mmol; BEt₃in hexane, 50

µL, 0.05 mmol; B(C₆F₅)₃, 12.8 mg, 0.025 mmol; BCl₃in hexane,

50 µL, 0.05 mmol; BF₃·OEt₂, 8 µL, 0.05 mmol; AlBr₃, 2.68 mg,

0.01 mmol; AlCl₃, 6.6 mg, 0.05 mmol; AlF₃, 4.2 mg, 0.05 mmol;

InCl₃, 11 mg, 0.05 mmol; Yb(OTf)₃, 31 mg, 0.05 mmol; Zn(OTf)₂,

18.2 mg, 0.05 mmol) were mixed. After being ﬂushed with 10 bar

(or 1 bar) of H₂one time, the system was charged with 10 bar (or

1 bar) of H₂and was heated in an oil bath with stirring at either

ambient temperature or 90 °C. The progress of the reactions was

monitored by a Buechi Pressﬂow gas controller, and the conversion

of the hydrogenations was calculated on the basis of the consump-

tion of H₂in the system. At the end of the catalyses, the autoclave

was cooled to ambient temperature and vented carefully. The

1

2

1662. H NMR (500.25 MHz, benzene-d₆, ppm): δ 3.54 (dd, J_HP

) 29.0 Hz, 38.0 Hz, 1H, ReH), 3.44 (br, 1H, CH₂dCH-), 3.12

(br, 1H, CH₂dCH-), 3.05 (br, 1H, CH₂dCH-), 2.83 (m, 3H,

PCH(CH₃)₂), 2.69 (m, 3H, PCH(CH₃)₂), 1.81 (m, 2H, CH₂), 1.59

3

(m, 4H, CH₂), 1.14-1.24 (m, 36H, PCH(CH₃)₂), 1.00 (t, J_HH

)

7.5 Hz, 3H, CH₃). ¹³C{¹H} NMR (50.28 MHz, benzene-d₆, ppm):

δ 53.2 (s, CH₂dCH-), 45.0 (s, CH₂dCH-), 39.4 (s, CH₂), 38.6

(s, CH₂), 25.8 (t, PCH(CH₃)₂), 22.7 (s, CH₂), 21.7 (s, PCH(CH₃)₂),

20.2 (s, PCH(CH₃)₂), 19.8 (s, PCH(CH₃)₂),19.5 (s, PCH(CH₃)₂),

14.5 (s, CH₃). ³¹P{¹H} NMR (80.94 MHz, benzene-d₆, ppm): δ

14.9 (d, J ) 128 Hz, 1P), 12.0 (d, J ) 128 Hz, 1P). Anal. Calcd

for C₂₄H₅₅BrNOP₂Re (701.76): C, 41.08; H, 7.90; N, 2.00. Found:

C, 41.26; H, 7.99; N, 1.83.

[Re(Br)(H)(NO)(PCy₃)₂(η²-CH₂dCHBu)] (3b). In a 3 mL

Young NMR tube, [Re(Br)(H)(NO)(PCy₃)₂] (2b; 34 mg, 0.04

mmol) and 1-hexene (5.0 µL, 0.04 mmol) were mixed in benzene-

d₆(0.5 mL). Immediately a light brown solution was formed. NMR

spectroscopy indicated the formation of 3b in 100% in situ yield.

The solvent was evaporated in vacuo for 24 h to give a light brown

residue. Yield: 36 mg, 99%. IR (ATR, cm^-1): ν(C-H) 2922, 2848,

1

ν(NO) 1661. H NMR (300.08 MHz, benzene-d₆, ppm): δ 3.82

(dd, ²J_HP) 27.0 Hz, 33.0 Hz, 1H, ReH), 3.42 (br, 2H, CH₂dCH-),

3.12 (br, 1H, CH₂dCH-), 1.31-2.86 (m, 66H, P(C₆H₁₁)₃), 1.86

(m, 2H, CH₂), 1.67 (m, 4H, CH₂), 1.02 (br, 3H, CH₃). ¹³C{¹H}

NMR (75.47 MHz, benzene-d₆, ppm): δ 52.7 (s, CH₂dbd)CH-),

45.2 (s, CH₂dCH-), 39.6 (s, CH₂), 39.1 (s, CH₂), 36.8 (d, J ) 22

Hz, PC), 35.7 (d, J ) 19 Hz, PC), 32.0 (s), 30.5 (s), 30.0 (s), 28.0

(s), 26.9 (s), 23.0 (s), 14.5 (s). ³¹P{¹H} NMR (121.47 MHz,

benzene-d₆, ppm): δ 4.9 (d, J ) 126 Hz, 1P), -0.4 (d, J ) 126

Hz, 1P). Anal. Calcd for C₄₂H₇₉BrNOP₂Re (942.14): C, 53.54; H,

8.45; N, 1.49. Found: C, 53.28; H, 1.44; N, 1.57.

1

hydrogenation products were characterized by H NMR spectros-

copy in CDCl₃. Some of the products were also characterized by

GC-MS analyses. The TONs and TOFs were determined on the

basis of the consumption of H₂. These data were found to be quite

comparable to the results determined according to the integrations

1

in the H NMR spectra. In the reactions, where full conversions

were obtained, the hydrogenated products were isolated via reduced

pressure distillation. The boiling points were comparable to the

reported values.

Preparation of the Rhenium(I) Chloro Hydride Complex

[Re(Cl)(H)(NO)(PiPr₃)₂(η²-CH₂dCHBu)] (3a-Cl) and Its Test

for Catalytic Activity. In a 3 mL Young NMR tube, 2a (12.3 mg,

0.02 mmol), 1-hexene (100 µL, 0.80 mmol), and nBu₄NCl (10 mg,

0.036 mmol) were mixed in benzene (0.5 mL). The mixture was

kept at 80 °C for 5 min. Complete formation of 3a-Cl was achieved

in over 99% yield as indicated by ³¹P NMR spectroscopy. The

solution was dried in vacuo, and the residue was extracted with

pentane (3 × 2 mL). The solvent was evaporated in vacuo, affording

a purple solid which was identiﬁed to be a mixture containing 19%

[Re(Cl)(H)(NO)(PiPr₃)₂] and 81% 3a-Cl by ¹H NMR and ³¹P NMR

spectroscopy. Yield: 5.8 mg. The following are data for 3a-Cl. IR

KIE Studies of 2a/BH₃·THF Catalyzed Hydrogenation of

1-Hexene at Different Temperatures. In a 20 mL vial in a

glovebox, 62 mg (0.10 mmol) of [Re(Br)(H)(NO)(PiPr₃)₂] (2a) was

dissolved in 2 mL of 1-hexene. The resultant solution was divided

into 10 portions ready for catalysis. In a 30 mL steel autoclave

equipped with a stirring bar, 1-hexene (5 mL) was mixed with one

portion of the prepared 1-hexene solution of 2a and 50 µL (0.05

mmol) of BH₃·THF. After being ﬂushed with 10 bar of H₂or D₂

once, the system was charged with 10 bar of H₂or D₂and was

heated in an oil bath with stirring at the required temperature (23,

60, and 90 °C). The reaction process was monitored by a Buechi

Pressﬂow gas controller. At the end of the catalyses, the autoclave

was cooled to ambient temperature and vented carefully. The

1

(ATR, cm^-1): ν(C-H) 2958, 2926, 2870, ν(NO) 1597. H NMR

2

(199.95 MHz, benzene-d₆, ppm): δ 3.92 (t, J_HP) 34.0 Hz, 1H,

9

J. AM. CHEM. SOC. VOL. 132, NO. 51, 2010 18245

A R T I C L E S

Jiang et al.

ReH), 3.38 (br, 1H, CH₂dCH-), 2.76-2.96 (m, 2H, CH₂dCH-),

2.64 (m, 6H, PCH(CH₃)₂), 1.84 (m, 2H, CH₂), 1.50 (m, 4H, CH₂),

CH₂dCH-), 3.13 (m, 2H, CH₂dCH-), 2.87 (m, 6H, PCH(CH₃)₂),

1.78 (m, 2H, CH₂), 1.54 (m, 4H, CH₂), 1.24 (m, 36H, PCH(CH₃)₂),

3

1.04 (t, J_HH) 6.0 Hz, 3H, CH₃). ³¹P{¹H} NMR (121.47 MHz,

1.22 (m, 36H, PCH(CH₃)₂), 1.03 (t, J_HH) 7.0 Hz, 3H, CH₃).

³¹P{¹H} NMR (80.94 MHz, benzene-d₆, ppm): δ 19.1 (d, J ) 130

Hz, 1P), 17.3 (d, J ) 130 Hz, 1P). The following are data for

[Re(Cl)(H)(NO)(PiPr₃)₂]. IR (ATR, cm^-1): ν(C-H) 2958, 2926,

benzene, ppm): δ 10.5 (d, J ) 120 Hz, 1P), 3.7 (d, J ) 120 Hz,

1P). ESI-MS spectroscopy conﬁrmed the composition of fragments

of 3a-I bearing a Re-I unit. MS (ESI): m/z 718.9 {[Re-

(I)(NO)(PiPr₃)₂(η²-CH₂dCHBu)] - Et}⁺. The solution was dried

in vacuo, and the residue was extracted with pentane (4 × 2 mL).

The solvent was evaporated, affording a light brown residue (5.6

mg, 0.006 mmol, of 3a-I). The obtained residue was dissolved with

5 mL of 1-hexene, to which 30 µL of BH₃·THF was added. The

mixture was charged with 10 bar of H₂and continuously stirred at

90 °C. The reaction process was monitored by a Buechi Pressﬂow

gas controller.

Halide Exchange Experiments As Traced by ³¹P NMR

Spectroscopy. In a 3 mL Young NMR tube, 2a (6.2 mg, 0.01

mmol), 1-hexene (10.0 µL, 0.08 mmol), and nBu₄NCl (14 mg, 0.05

mmol) were mixed in benzene (0.5 mL). The mixture was kept at

85 °C for 5 min. ³¹P NMR spectra indicated the formation of 3a-

Cl in stoichiometric yield. ³¹P{¹H} NMR (80.94 MHz, benzene,

ppm): δ 19.1 (d, J ) 130 Hz, 1P), 17.3 (d, J ) 130 Hz, 1P).

Furthermore, to the obtained mixture was added nBu₄NF in THF

(50 µL, 0.05 mmol). ³¹P NMR spectra indicated the immediate

formation of 3a-F at room temperature within 1 min. ³¹P{¹H} NMR

(80.94 MHz, benzene, ppm): δ 26.1 (dd, J_PP) 145 Hz, J_FP) 31

Hz, 1P), 19.3 (dd, J_PP) 145 Hz, J_FP) 31 Hz, 1P).

Similarly, in a 3 mL Young NMR tube, 2b (8.4 mg, 0.01 mmol),

1-hexene (10.0 µL, 0.08 mmol), and nBu₄NCl (14 mg, 0.05 mmol)

were mixed in benzene (0.5 mL). The mixture was kept at 85 °C

for 5 min, and the ³¹P NMR spectra indicated the formation of

[Re(Cl)(H)(NO)(PCy₃)₂(η²-CH₂dCHBu)] (3b-Cl) in stoichiometric

yield. ³¹P{¹H} NMR (80.94 MHz, benzene, ppm): δ 7.9 (d, J )

132 Hz, 1P), 4.8 (d, J ) 132 Hz, 1P). Subsequently, nBu₄NF in

THF (50 µL, 0.05 mmol) was added to the obtained mixture. The

³¹P NMR spectra indicated the immediate formation of [Re(F)-

(H)(NO)(PCy₃)₂(η²-CH₂dCHBu)] (3b-F). ³¹P{¹H} NMR (80.94

MHz, benzene, ppm): δ 13.8 (dd, J_PP) 146 Hz, J_FP) 34 Hz, 1P),

6.4 (dd, J_PP) 146 Hz, J_FP) 34 Hz, 1P). Detailed experimental

data for 3b-F, 3b-Cl, and 3b-I are given in the Supporting

Information.

VT ³¹P NMR Measurement of the Reaction between 3a and

Boron Lewis Acids in 1-Hexene. In a 3 mL Young NMR tube,

2a (6.2 mg, 0.01 mmol) and 0.05 mmol of each boron Lewis acid

(B(C₆F₅)₃, 26.0 mg; BH₃·THF, 50 µL; BEt₃, 50 µL; BPh₃,12.2 mg)

were mixed in 0.5 mL of 1-hexene. The immediate formation of

the 18 e^-1-hexene-coordinated rhenium(I) hydride 3a was observed

in quantitative yield, as checked by the ³¹P NMR spectra. The

mixture was monitored by ³¹P NMR spectroscopy in the temperature

range of +23 to -80 °C in intervals of 20 °C.

1

2870, ν(NO) 1687. H NMR (199.95 MHz, benzene-d₆, ppm): δ

2.80 (m, 6H, PCH(CH₃)₂), 1.19 (m, 36H, PCH(CH₃)₂), -17.94 (br,

1H, ReH). ³¹P{¹H} NMR (80.94 MHz, benzene-d₆, ppm): δ 46.8

(s, 1P). These data are quite comparable to those of the ﬁve-

coordinate rhenium(I) bromo hydride compound 2a. ESI-MS

spectroscopy conﬁrmed the composition of fragments of 3a-Cl

bearing a Re-Cl unit. MS (ESI): m/z 572.1 [Re(Cl)(NO)(PiPr₃)₂]⁺.

The obtained solid was dissolved with 5 mL of 1-hexene and

transferred to a 30 mL steel vessel. Afterward 50 µL of BH₃·THF

(5 equiv relatrive to Re) was added, and the mixture was charged

with 10 bar of H₂and continuously stirred at 90 °C. The reaction

process was monitored by a Buechi Pressﬂow gas controller.

Preparation of the Rhenium(I) Fluoro Hydride Complex

[Re(F)(H)(NO)(PiPr₃)₂(η²-CH₂dCHBu)] (3a-F) and Its Test

for Catalytic Activity. In a 3 mL Young NMR tube, 2a (26 mg,

0.04 mmol), 1-hexene (100 µL, 0.80 mmol), and nBu₄NF in THF

(80 µL, 0.08 mmol) were mixed in benzene (0.5 mL). The mixture

was kept at room temperature for 2 min. The formation of 3a-F

was achieved in over 99% in situ yield as indicated by ³¹P NMR

spectroscopy. The solution was dried in vacuo, and the residue was

extracted with cold pentane (2 × 2 mL). The solvent was evaporated

in vacuo, affording a light brown residue. Yield: 12.8 mg, 0.02

mmol, 50%. IR (ATR, cm^-1): ν(C-H) 2955, 2925, 2870, ν(NO)

1647. ¹H NMR (300.08 MHz, benzene-d₆, ppm): δ 4.58 (ddd, ²J_HP

2

) 33.0 Hz, J_HP) 30.0 Hz, J_HF) 8.0 Hz, J_HF) 6.0 Hz, 1H,

ReH), 3.38 (br, 1H, CH₂dCH-), 3.14 (m, 2H, CH₂dCH-), 2.74

(m, 3H, PCH(CH₃)₂), 2.57 (m, 3H, PCH(CH₃)₂), 1.89 (m, 2H, CH₂),

1.64 (m, 4H, CH₂), 1.19-1.37 (m, 36H, PCH(CH₃)₂), 1.07 (t, ³J_HH

) 7.5 Hz, 3H, CH₃). ¹³C{¹H} NMR (75.47 MHz, benzene-d₆, ppm):

δ 54.3 (s, CH₂dCH-), 39.9 (s, CH₂dCH-), 36.7 (s, CH₂), 30.2

(s, CH₂), 28.4 (d, J ) 21 Hz, PCH(CH₃)₂), 27.8 (d, J ) 21 Hz,

PCH(CH₃)₂), 23.0 (s, CH₂), 19.8 (s, PCH(CH₃)₂), 19.5 (s,

PCH(CH₃)₂),19.0 (s, PCH(CH₃)₂), 14.6 (s, CH₃). ³¹P{¹H} NMR

(121.47 MHz, benzene-d₆, ppm): δ 24.9 (dd, J_PP) 146 Hz, J_FP

)

30 Hz, 1P), 18.7 (dd, J_PP) 146 Hz, J_FP) 30 Hz, 1P). ¹⁹F NMR

(282.33 MHz, THF-d₈, ppm): -271.65 (s). Satisfactory elemental

analysis data were not obtained due to the presence of a trace

amount of free 1-hexene in the residue. ESI-MS spectroscopy

conﬁrmed the composition of fragments of 3a-F bearing a Re-F

unit. MS (ESI): m/z 556.4 [Re(F)(NO)(PiPr₃)₂]⁺. The obtained

residue was dissolved with 200 µL of 1-hexene, and half of it was

mixed with 50 µL of BH₃·THF and 5 mL of 1-hexene in a steel

vessel. The mixture was charged with 10 bar of H₂and continuously

stirred at 90 °C. The reaction process was monitored by a Buechi

Pressﬂow gas controller.

H₂/D₂Scrambling Experiments. In a 3 mL Young NMR tube,

2a (3.1 mg, 0.005 mmol) and B(C₆F₅)₃(12.8 mg, 0.025 mmol)

were dissolved in 0.5 mL of toluene-d₈. The nitrogen atmosphere

was replaced with 1100 mbar of H₂and D₂in a 1:1 ratio using a

freeze-pump-thaw cycle. The solution was immediately investi-

In Situ Preparation of the Rhenium(I) Iodo Hydride Com-

plex [Re(I)(H)(NO)(PiPr₃)₂(η²-CH₂dCHBu)] (3a-I) and Its Test

for Catalytic Activity. In a small vial in a glovebox, 2a (12.4 mg,

0.02 mmol) and 1-hexene (0.1 mL) were mixed in 0.5 mL of

benzene. To the light yellow solution was added AgSbF₆(10.3 mg,

0.03 mmol), and a brown precipitate was formed immediately. After

being stirred at room temperature for 5 min, the solution was ﬁltered

through Celite. The ﬁltrate was checked by ³¹P NMR spectroscopy,

indicating the quantitative formation of a new species which might

1

gated by H NMR spectroscopy in a temperature range of +23 to

-50 °C in intervals of 20 °C. At 296 K, H₂, HD, and rhenium-

coordinated H₂(or HD) cannot be observed due to a too fast

1

exchange between free H₂and coordinated H₂. H NMR (199.95

MHz, toluene-d₈, ppm, 296 K): δ 3.82 (br, ReH, free H₂, η²-H₂),

2.40 (m, 6H, PCH(CH₃)₂), 1.18 (m, 36H, PCH(CH₃)₂). At 243 K,

be assigned to

a

complex salt, [Re(H)(NO)(PiPr₃)₂(η²-

1

CH₂dCHBu)₂][SbF₆]. ³¹P{¹H} NMR (80.94 MHz, benzene, ppm):

δ 27.5 (d, J ) 112 Hz, 1P), 24.3 (d, J ) 122 Hz, 1P). To the

ﬁltrate was added nNBu₄I (11.1 mg, 0.03 mmol). The mixture was

kept at room temperature for 2 h, and ³¹P NMR spectroscopy

indicated the formation of iodide hydride analogue 3a-I in 81% in

situ yield along with 19% rhenium(I) ﬂuoro hydride byproduct 3a-

F. The following are data for 3a-I. ¹H NMR (300.08 MHz, benzene-

a broad triplet at around 4.5 ppm could be detected. H NMR

(199.95 MHz, toluene-d₈, ppm, 243 K): δ 4.48 (s, free H₂), 4.45 (t,

J ) 42 Hz, HD), 3.40 (t, J ) 26 Hz, 1H, ReH), 3.06 (br, 2H,

η²-H₂), 2.37 (m, 6H, PCH(CH₃)₂), 1.16 (m, 36H, PCH(CH₃)₂). At

223 K, the formation of HD was observed. ¹H NMR (199.95 MHz,

toluene-d₈, ppm, 223 K): δ 4.50 (s, free H₂), 4.46 (t, J ) 43 Hz,

HD), 3.47 (t, J ) 26 Hz, 1H, ReH), 3.08 (br, 2H, η²-H₂), 2.33 (m,

6H, PCH(CH₃)₂), 1.14 (m, 36H, PCH(CH₃)₂). From the NMR

2

d₆, ppm): δ 3.36-3.59 (t, J_HP) 30.0 Hz, 2H, ReH and

9

18246 J. AM. CHEM. SOC. VOL. 132, NO. 51, 2010

Cocatalysis of Alkene Hydrogenations

A R T I C L E S

results, it is concluded that the H₂/D₂scrambling occurs spontane-

Filtration Experiments. In a 30 mL steel autoclave equipped

with a stirring bar, 1-hexene (5 mL) and 2a/BH₃·THF (0.01 mmol/

0.05 mmol) were mixed. After one ﬂush with 10 bar of H₂, the

system was charged with 10 bar of H₂and was heated in an oil

bath with stirring at 90 °C. After 5 min, a TON of 2000 was

reached, corresponding to a 50% conversion. At this point, the

reaction was terminated manually by removing the H₂atmosphere

and cooling the reaction vessel to room temperature. Then the

autoclave was transferred to a glovebox, and the solution was found

to be light brown and clear without any precipitate or colloids

formed. The ﬁltrate was examined by ³¹P NMR spectroscopy,

indicating the presence of 3a (66%, δ 14.7 (d, J ) 127 Hz, 1P),

11.6 (d, J ) 127 Hz, 1P)), free PiPr₃(10%, δ 20.4 ppm), and an

unknown organometallic species (24%, δ 55.4 (s)). The solution

was ﬁltered through Celite into a new vessel with a new stirring

bar, to which 2 mL of 1-hexene was readded. No residue was visible

on the Celite surface. The catalysis batch was repeated at 90 °C

under 10 bar of H₂, and it was found that the high activity still

remained (TOF of 1.2 × 10⁴h^-1). The Celite was washed with 2

mL of 1-hexene and transferred into a autoclave vessel containing

5 mL of 1-hexene. No catalytic activity was observed at 90 °C

under 10 bar of H₂within 1 h. Similar manipulations were applied

to the 2a/BPh₃system except that 3 mL of 1-hexene was used and

the reaction was terminated at 53% conversion within 3 min.

Mercury Poisoning Experiments. In a 30 mL steel autoclave

equipped with a stirring bar, excess mercury (96-112 mg) was

added to the 1-hexene (5 mL) solution loaded with 2a/BPh₃(0.01

mmol/0.05 mmol) or 2a/B(C₆F₅)₃(0.01 mmol/0.05 mmol) before

the catalysis was carried out. The hydrogenation catalyzed by the

2a/B(C₆F₅)₃system was carried out at 23 °C under either 1 or 10

bar of H₂. The hydrogenation catalyzed with the 2a/BPh₃system

was carried out at 90 °C under 10 bar of H₂.

ously at room temperature with the 2a/B(C₆F₅)₃system.

1

The 500 MHz H NMR data are as follows. H NMR (500.25

MHz, toluene-d₈, 223 K, ppm): 4.46 (t, J_HD) 43 Hz, HD, T₁)

827 ms), 3.44 (t, J_HP) 26 Hz, ReH, T₁) 264 ms), 3.42 (t, J_HP

)

25 Hz, ReH, T₁) 260 ms), 3.07 (m, ReH₂, T₁) 116 ms), 2.73

(br, ReHD, T₁) 501 ms). ³¹P NMR (202.5 MHz, toluene-d₈, 223

K, ppm): 41.72 (s), 41.66 (s), 41.53 (s), 41.44 (s).

Reaction of the 2a/B(C₆F₅)₃System with D₂. In a 3 mL Young

NMR tube, 2a (12.4 mg, 0.02 mmol) and B(C₆F₅)₃(30 mg, 0.06

mmol) were mixed in toluene (0.5 mL). The nitrogen atmosphere

was replaced with 995 mbar of D₂using a freeze-pump-thaw

2

cycle. VT H NMR measurements were carried out immediately.

At 293 K, a broad signal was observed in the range of 3.00-5.00

ppm. ²H NMR (93.13 MHz, toluene-d₈, ppm, 293 K): δ 3.30 (free

D₂, ReD, η²-D₂). At low temperatures, such as 213 K, the formation

of the Re-D/D₂isotopomer was clearly observed, displaying a

broad singlet at 3.56 and 3.13 ppm. ²H NMR (93.13 MHz, toluene-

d₈, 213 K, ppm): δ 4.55 (s, HD), 4.50 (s, free D₂), 3.56 (br, ReD,

T₁) 22 ms), 3.13 (br, ReD₂, T₁) 24 ms). ³¹P NMR (202.5 MHz,

toluene-d₈, 213 K, ppm): 41.40 (s, ReD/D₂).

Isomerization of 1-Octene to (E/Z)-2-Octene Induced by

the 2a/B(C₆F₅)₃System. In a 3 mL Young NMR tube, 2a (12.4

mg, 0.02 mmol), B(C₆F₅)₃(30 mg, 0.06 mmol), and 1-octene (15

µL, 0.10 mmol) were dissolved in 0.5 mL of toluene-d₈. The

1

solution was immediately investigated by H NMR spectroscopy,

and the formation of (E)- or (Z)-2-octene was observed instantly

1

in 23% in situ yield. H NMR (199.95 MHz, toluene-d₈, ppm): δ

5.33 (m, -CHdCH-). In the case of the isomerization of 1-hexene

induced by the 2a/B(C₆F₅)₃system, the same conditions were

applied. In the reaction of 1-hexene with 2a alone, a high

temperature of 80 °C was applied to initiate the isomerization.

Hydrogenation of 1-Hexene Catalyzed by the in Situ Pre-

pared active Complex from 2a with AgSbF₆. In a small vial in

a glovebox, 2a (6.3 mg, 0.01 mmol) and 1-hexene (5.0 mL) were

mixed. To the light yellow solution was added AgSbF₆(6.9 mg,

0.02 mmol), and a brown precipitate was formed immediately. After

being stirred at room temperature for 10 min, the solution was

ﬁltrated through Celite, and the ﬁltrate was transferred to a steel

vessel. After being ﬂushed with 10 bar of H₂once, the system was

charged with 10 bar of H₂and was heated in an oil bath with stirring

at 90 °C. The progress of the reaction was monitored by a Buechi

Pressﬂow gas controller. At the end of the reaction, the autoclave

was cooled to ambient temperature and vented carefully. The

Acknowledgment. Financial support from the Swiss National

Science Foundation, Lanxess AG, Leverkusen, Germany, and the

Funds of the University of Zurich is gratefully acknowledged.

Supporting Information Available: Various experimental

details, ﬁgures including the KIE measurements, the ﬁltration

and mercury poisoning experiments, catalysis under 1 bar of

H₂, and the halide exchange experiments, and NMR spectra

including those of the VT ¹H NMR and VT ²H NMR

measurements for H₂/D₂scrambling and the isomerization

reactions of terminal alkenes. This material is available free of

charge via the Internet at http://pubs.acs.org.

1

hydrogenation products were characterized by H NMR spectros-

copy in CDCl₃, and the conversion of the hydrogenation was

calculated on the basis of the consumption of H₂.

JA107187R

9

J. AM. CHEM. SOC. VOL. 132, NO. 51, 2010 18247

Article Doi

Rhenium hydride/boron lewis acid cocatalysis of alkene hydrogenations: Activities comparable to those of precious metal systems

DOI: 10.1021/ja107187r

Source and publish data:

Authors:

Article abstract of DOI:10.1021/ja107187r

Full text of DOI:10.1021/ja107187r

Products guided by the article

R&D Labs maybe for 1260114-10-0

Relevant to this article

Hot Product