Iridium(0) nanocluster, acid-assisted catalysis of neat acetone hydrogenation at room temperature: Exceptional activity, catalyst lifetime, and selectivity at complete conversion

Article abstract of DOI:10.1021/ja0437813

Acetone hydrogenation catalysis is important in applications such as heat pumps and fuel cells or in fulfilling the sizable demand for the product of selective acetone hydrogenation, 2-propanol. Reported herein is the discovery of a superior acetone hydrogenation catalyst- superior in terms of activity at low temperature, selectivity at complete conversion, and total catalyst lifetime. The new catalyst system consists of Ir(0)_n nanoclusters plus HCl easily and reproducibly formed from commercially available [(1,5-COD)-IrCl] ₂ under H₂. The resultant, room temperature, high activity, and highly selective (2/n)Ir(0)_n plus 2HCl catalyst system hydrogenates acetone at 22 °C and 40 psig of H₂ pressure to 95% 2-propanol and the rest diisopropyl ether at 100% conversion with 16400 total catalytic turnovers and with an initial turnover frequency of 1.9 s^-1 at 22 °C. When molecular sieves are added, the catalyst system becomes even more selective and long-lived, providing the complete and selective conversion of acetone to 100% 2-propanol with 188000 total turnovers. Also reported are initial kinetic, D-labeling and other mechanistic studies, a summary section detailing the four main findings, the green chemistry aspects, and the current main drawback (a limited catalytic lifetime due to nanocluster precipitation) of the present invention. A review of the extensive literature of acetone hydrogenation is also tabulated as part of the Supporting Information.

Full text of DOI:10.1021/ja0437813

Published on Web 03/08/2005
Iridium(0) Nanocluster, Acid-Assisted Catalysis of Neat
Acetone Hydrogenation at Room Temperature: Exceptional
Activity, Catalyst Lifetime, and Selectivity at Complete
Conversion
†
,‡
Saim O¨ zkar and Richard G. Finke*
Contribution from the Departments of Chemistry, Middle East Technical UniVersity,
0
6531 Ankara, Turkey, and Colorado State UniVersity, Fort Collins, Colorado 80523
Received October 12, 2004; E-mail: rfinke@lamar.colostate.edu
Abstract: Acetone hydrogenation catalysis is important in applications such as heat pumps and fuel cells
or in fulfilling the sizable demand for the product of selective acetone hydrogenation, 2-propanol. Reported
herein is the discovery of a superior acetone hydrogenation catalysts superior in terms of activity at low
temperature, selectivity at complete conversion, and total catalyst lifetime. The new catalyst system consists
of Ir(0)
IrCl] under H
catalyst system hydrogenates acetone at 22 °C and 40 psig of H
diisopropyl ether at 100% conversion with 16400 total catalytic turnovers and with an initial turnover frequency
n
nanoclusters plus HCl easily and reproducibly formed from commercially available [(1,5-COD)-
. The resultant, room temperature, high activity, and highly selective (2/n)Ir(0) plus 2HCl
pressure to 95% 2-propanol and the rest
2
2
n
2
-
1
of 1.9 s at 22 °C. When molecular sieves are added, the catalyst system becomes even more selective
and long-lived, providing the complete and selective conversion of acetone to 100% 2-propanol with 188000
total turnovers. Also reported are initial kinetic, D-labeling and other mechanistic studies, a summary section
detailing the four main findings, the “green chemistry” aspects, and the current main drawback (a limited
catalytic lifetime due to nanocluster precipitation) of the present invention. A review of the extensive literature
of acetone hydrogenation is also tabulated as part of the Supporting Information.
Introduction
A range of heterogeneous or homogeneous catalysts have
been used to hydrogenate acetone, commonly at temperatures
The catalytic hydrogenation of acetone is an important
reaction, one used to produce two industrial chemicals, 2-pro-
in the 100-300 °C range for heterogeneous catalysts; see Table
S-1 of the Supporting Information, where the results and
conditions for 14 of the best, prototype prior acetone hydro-
genation catalyst systems are tabulated from among >340
citations resulting from a SciFinder literature search of “acetone
hydrogenation”: Raney Ni, [Ir(CO)(PPh ) ]ClO , [Rh (CO) -
1
panol and, at higher temperature (as a result of acetone
condensation, dehydration, and subsequent hydrogenation),
methyl isobutyl ketone ((CH3)2CHCH2C(O)CH3). In addition,
acetone hydrogenation to 2-propanol is of interest for use in
chemical heat pumps or fuel cells, applications which require
a higher selectivity to 2-propanol at high conversion than is
2
3
3
4
2
2
3
4
+
(dppm) ] , 8-15% Ni/SiO , 0-50% Ni/Al O , Pt/zeolite, Pt/
3
2
2
3
SiO , 10-50% Ni or Co or Fe/Al O , Tc or Tc-M (M ) Pt,
2
2
3
3
presently available. The demand for 2-propanol is also increas-
ing due to its transfer hydrogenation reactions, chemistry which
Pd, Rh, Ru, Ni, Re, Co)/Al2O3 (or supported on SiO2 or MgO),
Ni/â-zeolite (or zeolite-Y), CoMgAl or NiMgAl layered double
hydroxides, CuCr2O4, Cu6O8Ln(NO3), and Pt/actived carbon.
The resultant products are typically 2-propanol and methyl
isobutyl ketone plus other products (methyl isobutyl carbinol,
5
exploits 2-propanol’s relatively easy, selective reoxidation to
6
acetone.
†
Middle East Technical University.
Colorado State University.
‡
(5) Reviews: (a) Birch, A. J.; Williamson, D. H. Org. React. (N. Y.) 1976, 24,
1-186. (b) Matteoli, U.; Frediani, P.; Binachi, M.; Botteghi, C.; Gladiali,
S. J. Mol. Catal. 1981, 12, 265-319. (c) Zassinovich, G.; Mestroni, G.;
Gladiali, S. Chem. ReV. 1992, 92, 1051-1069. (d) de Graauw, C. F.; Peters,
J. A.; van Bekkum, H.; Huskens, J. Synthesis 1994, 1007-1017. (e)
Gladiali, S.; Mestroni, G. Transition Met. Org. Synth. 1998, 2, 97-119.
(f) Palmer, M. J.; Wills, M. Tetrahedron: Asymmetry 1999, 10, 2045-
2061. (g) Wills, M.; Palmer, M.; Smith, A.; Kenny, J.; Walsgrove, T.
Molecules 2000, 5, 4-18. (h) Wills, M.; Gamble, M.; Palmer, M.; Smith,
A.; Studley, J.; Kenny, J. J. Mol. Catal. A 1999, 146, 139-148. (i) Fache,
F.; Schulz, E.; Tommasino, M. L.; Lemaire, M. Chem. ReV. 2000, 100,
2159-2231. (j) Casey, C. P.; Johnson, J. B. J. Org. Chem. 2003, 68, 1998.
(6) (a) Wilds, A. L. Org. React. (N. Y.) 1944, 2, 178-223. (b) Djerassi, C.
Org. React. (N. Y.) 1951, 6, 207-272. (c) Namy, J. L.; Souppe, J.; Collin,
J.; Kagan, H. B. J. Org. Chem. 1984, 49, 2045-2049. (d) Hallman, P. S.;
McGarvey, B. R.; Wilkinson, G. J. Chem. Soc. A 1968, 3143-3150.
(
1) Anderson, L. C.; MacNaughton, N. W. J. Am. Chem. Soc. 1942, 64, 1456.
2) Methyl isobutyl ketone is used as a solvent for vinyl, epoxy, and acrylic
resins as well as cellulose-based coatings. (a) Gandia, L. M.; Montes, M.
Appl. Catal., A: General 1993, 101, L1. (b) Reith, W.; Dettmer, M.;
Widdecke, H.; Fleischer, B. Stud. Surf. Sci. Catal. 1991, 59, 487.
(
(
3) Note that the preferred operating temperature in the heat-pump application
is around 200 °C: (a) Wojcik, A. M. W.; van der Koi, H. J.; Maschmeyer,
T. J. Phys. D: Appl. Phys. 2001, 34, 660-666. (b) Gandia, L. M.; Montes,
M. Int. J. Energy Res. 1992, 16, 851-864. (c) Meng, N.; Shinoda, S.;
Saito, Y. Int. J. Hydrogen Energy 1997, 22, 361-367. (d) Chung, Y.; Kim,
B. J.; Yeo, Y. K.; Song, H. K. Energy 1997, 22, 525-536.
(4) (a) Ando, Y.; Tanaka, T.; Doi, T.; Takashima, T. Energy ConVers. Manage.
2001, 42, 1807-1816. (b) Pardillos-Guindet, J.; Metais, S.; Vidal, S.; Court,
J.; Foulloux, P. Appl. Catal., A 1995, 132, 61-75. (c) Pardillos-Guindet,
J.; Vidal, S.; Court, J.; Foulloux, P. J. Catal. 1995, 155, 12-20.
4800
9
J. AM. CHEM. SOC. 2005, 127, 4800-4808
10.1021/ja0437813 CCC: $30.25 © 2005 American Chemical Society

Iridium Catalysis of Neat Acetone Hydrogenation
A R T I C L E S
diacetone alcohol, mesityl oxide, diisobutyl ketone, propane,
and 2-methylpentane) in ratios which depend on the reaction
lifetime catalyst yet reported for acetone hydrogenation to
2
1,22
2-propanol.
Our iridium(0) nanoclusters are easily and
7
-15
conditions.
The selectivity at lower temperatures typically
reproducibly formed in situ in neat acetone by the hydrogen
reduction of the commercial (1,5-cyclooctadiene)chloroiridium-
favors 2-propanol, while higher temperatures favor methyl
8c,9a
isobutyl ketone;
however, an interesting reversal of selectiv-
(I) dimer precatalyst, [(1,5-COD)IrCl] , at 22 °C and 40 psig
2
ity from 100% 2-propanol at 90% conversion for 10% Ni/Al2O3
to 4% 2-propanol and 96% methyl isobutyl ketone for 50% Ni/
Al2O3 at 100 °C has been reported.9c The best selectivity at the
highest conversion previously reported is 100% 2-propanol at
of H pressure. Reduction of the [(1,5-COD)IrCl] by H yields
2
2
2
+
-
1 equiv of H Cl for each Ir(I) reduced to iridium(0); the result
is a highly efficient, acid-assisted, high-selectivity acetone
hydrogenation catalyst. Nanocluster catalysis of acetone hydro-
genation has not been previously reported prior to our disclosure
in 1999 in a footnote that we had observed this reaction using
9
0% conversion at 100 °C for the above-noted 50%Ni/Al2O3
9a,c
system; the second-best is 97% selectivity at 65% conversion
at 100 °C for a bimetallic 0.2% Rh/0.2% Tc/γ-Al2O3 catalyst.
12
21
Ir and Rh nanoclusters for the first time and our brief citation,
-1
Activities (turnover frequency (TOF), s ) are reported for only
out of the 14 examples in Table S-1 of the Supporting
Information, the highest previous reported activity being a TOF
in 2001, that the present study was in progress (see footnote 52
on p 5806 in ref 22). Small molecules that can store H2, such
as an acetone/2-propanol couple, are also of general interest
regarding a future H2 economy, especially if such couples
proceeed with 100% conversion and the absence of side
3
-
1
11
-1
)
8.5 s at 100 °C for Pt/activated carbon and 11 s at 200
8
c
°
C for Ni/SiO2. The general trend for acetone hydrogenation
activity (for at least γ-Al2O3-supported metals) is reported to
23
products.
12
be Pt > Tc ≈ Rh > Pd > Ru > Ni ≈ Re > Co. The very
important catalysis parameter of catalytic lifetime (i.e., total
turnover number, TTO) for acetone hydrogenation has not been
specifically reported in the prior literature. In short, without prior
precedent and thus of considerable interest are acetone hydro-
genation catalysts which operate at room temperature, with an
(
16) Chloride anion is a common nanocluster stabilizer, albeit not one of the
22
best: Schmid, G.; Harms, M.; Malm, J. O.; Bovin, J. O.; van Ruitenbeck,
J.; Zandbergen, H. W.; Fu, W. T. J. Am. Chem. Soc. 1993, 115, 2046.
17) Representative reviews or papers on nanocluster catalysis: (a) Templeton,
A. C.; Wuelfing, W. P.; Murray, R. W. Acc. Chem. Res. 2000, 33, 27. (b)
Rao, C. N. R.; Kulkarni, G. U.; Thomas, P. J.; Edwards, P. P. Chem. Soc.
ReV. 2000, 29, 27. (c) Schmid, G.; Baumle, M.; Geerkens, M.; Heim. I.;
Osemann, C.; Sawitowski, T. Chem. Soc. ReV. 1999, 28, 179. (d) Stein, J.;
Lewis, L. N.; Gao, Y.; Scott, R. A. J. Am. Chem. Soc. 1999, 121, 3693. (e)
Reetz, M. T.; Breinbauer, R.; Wedemann, P.; Binger, P. Tetrahedron 1998,
(
-
1
activity g8.5 s at e100 °C, with selectivites to 2-propanol
g95% for 100% conversions, and which exhibit extended
catalytic lifetimes.
54, 1233. (f) Schmidt, T. J.; Noeske, M.; Gasteiger, H. A.; Behm, R. J.;
Britz, P.; Brijoux, W.; B o¨ nnemann, H. Langmuir 1997, 13, 2591. (g)
Schmid, G.; Maihack, V.; Lantermann, F.; Peschel, S. J. Chem. Soc., Dalton
Trans. 1996, 589. (h) Reetz, M. T.; Breinbauer, R.; Wanninger, K.
Tetrahedron Lett. 1996, 37, 4499. (i) Reetz, M. T.; Quaiser, S. A.; Merk,
C. Chem. Ber. 1996, 129, 741. (j) B o¨ nnemann, H.; Braun, G. A. Angew.
Chem., Int. Ed. Engl. 1996, 35, 1992. (k) Lewis, L. N. Chem. ReV. 1993,
93, 2693.
1
6
Herein we report our discovery that a chloride-stabilized
Ir(0) nanocluster1
7-20
catalyst is the lowest temperature, most
active, highest selective at high (100%) conversion, and longest
(
7) Raney nickel as catalyst: (a) Freund, T.; Hulburt, H. M. J. Phys. Chem.
1
957, 61, 909. (b) Kishida, S.; Teranishi, S. J. Catal. 1968, 12, 90. (c)
(18) For two introductory reviews, lead references to earlier reviews, and needed
definitions (e.g., of modern transition-metal nanoclusters vs, for example,
classical colloids), see: (a) Aiken, J. D., III; Finke, R. G. J. Mol. Catal. A
1999, 145, 1. (b) Aiken, J. D., III; Lin, Y.; Finke, R. G. J. Mol. Catal. A
1996, 114, 29-51. (c) Finke, R. G. In Metal Nanoparticles: Synthesis,
Characterization and Applications; Feldheim, D. L., Foss, C. A., Jr., Eds.;
Marcel Dekker: New York, 2002; Chapter 2, pp 17-54 (Transition-Metal
Nanoclusters: Solution-Phase Synthesis, then Characterization and Mech-
anism of Formation, of Polyoxoanion- and Tetrabutylammonium-Stabilized
Nanoclusters).
(19) Reviews (see also ref 18): (a) B o¨ nnemann, H.; Richards, R. Eur. J. Inorg.
Chem. 2001, 2455. (b) Schmid, G.; Baumle, M.; Geerkens, M.; Heim, I.;
Osemann, C.; Sawitowski, T. Chem. Soc. ReV. 1999, 28, 179. (c) Schmid,
G.; Chi, L. F. AdV. Mater. 1998, 10, 515. (d) Fendler, J. H., Ed.
Nanoparticles and Nanostructured Films; Wiley-VCH: Weinheim, Ger-
many, 1998. (e) F u¨ rstner, A., Ed. ActiVe Metals: Preparation, Charac-
terization, and Applications; VCH: Weinheim, Germany, 1996. (f) Bradley,
J. S. In Clusters and Colloids. From Theory to Applications; Schmid, G.,
Ed.; VCH: New York, 1994; pp 459-544. (g) Schmid, G. Chem. ReV.
1992, 92, 1709. (h) A superb series of papers is available in Faraday
Discuss. 1991, 92, 1-300. (i) Schmid, G. In Aspects of Homogeneous
Catalysis; Ugo, R., Ed.; Kluwer: Dordrecht, The Netherlands, 1990;
Chapter 1. (j) Andres, R. P.; Averback, R. S.; Brown, W. L.; Brus, L. E.;
Goddard, W. A.; Kaldor, A.; Louie S. G.; Moscovits, M.; Peercy, P. S.;
Riley, S. J.; Siegel, R. W.; Spaepen, F.; Wang, Y. J. Mater. Res. 1989, 4,
704. (k) Henglein, A. Chem. ReV. 1989, 89, 1861. (l) Thomas, J. M. Pure
Appl. Chem. 1988, 60, 1517. (m) Jena, P.; Rao B. K.; Khanna, S. N. Physics
and Chemistry of Small Clusters; Plenum: New York, 1987.
Lemcoff, N. O. J. Catal. 1977, 46, 356-364.
(
8) Silica-supported Co, Ni, Cu, Rh, Pd, Pt, and Au catalysts: (a) Van Druten,
G. M. R.; Ponec, V. React. Kinet. Catal. Lett. 1999, 68, 15-23. (b) Van
Druten, G. M. R.; Aksu, L.; Ponec, V. Appl. Catal., A 1997, 149, 181-
1
87. (c) Gandia, L. M.; Diaz, A.; Montes, M. J. Catal. 1995, 157, 461-
71. (d) Nakamura, M.; Wise, H. In Proceedings of the 6th International
4
Congress on Catalysis, London, 1976; Bond, G. C., Wells, P. B., Tompkins,
F. C., Eds.; The Chemical Society: London, 1977; p 881. (e) Simoikova,
J.; Hillaire, L.; Panek, J.; Kochleofl, K. Z. Phys. Chem. FR 1973, 83, 287-
3
04. (f) Sen, B.; Vannice, M. A. J. Catal. 1988, 113, 52-71.
(
9) Alumina-supported Ni, Co, and Fe catalysts: (a) Narayan, S.; Unnikrishnan,
R. J. Chem. Soc., Faraday Trans. 1998, 94, 1123-1128. (b) Narayan, S.;
Unnikrishnan, R. Stud. Surf. Sci. Catal. 1998, 113, 799-807. (c) Narayan,
S.; Unnikrishnan, R. Appl. Catal., A 1996, 145, 231-236.
(
(
(
10) Layered double hydroxide containing Co, Ni, Mg, and Al as catalysts:
Unnikrishnan, R.; Narayan, S. J. Mol. Catal. 1999, 144, 173-179.
11) Activated carbon-supported Pt catalysts: Fuente, A. M.; Pulgar, G.;
Gonzalez, F.; Pesquera, C.; Blanco, C. Appl. Catal., A 2001, 208, 35-46.
2 3 2
12) Tc or Tc-M (M ) Pt, Pd, Rh, Ru, Ni, Re, Co) supported on Al O , SiO ,
or MgO as catalysts: (a) Rimar, N. N.; Pirogova, G. N. Russ. Chem. Bull.
1
998, 47, 398-401. (b) Pirogova, G. N.; Popova, N. N.; Voronin, Y. V.
Radiochemistry 2000, 42, 565-568.
(
13) Zeolites containing transition-metal catalysts: (a) Melo, L.; Magnoux, P.;
Giannetto, G.; Alvarez, F.; Gusinet, M. J. Mol. Catal., A 1997, 124, 155-
1
61. (b) Knifton, J. F.; Dai, P. S. E. Catal. Lett. 1999, 57, 193-197.
(
2 4 6 8 3
14) Copper compounds such as CuCr O and Cu O Ln(NO ) as catalysts: (a)
Yurieva, T. M. Catal. Today 1999, 51, 457-467. (b) Yurieva, T. M. J.
Mol. Catal., A 1996, 113, 455-468. (c) Sakata, Y.; Nobukuni, S.;
Kikumoto, E.; Tanak, K.; Imamura, H.; Tsuchiya, S. J. Mol. Catal., A 1999,
(20) See ref 18 for a review of nanocluster catalysis which includes necessary
key terms and definitions of nanoclusters vs traditional (nano-) colloids,
monodisperse ((0% size distribution) and near-monodisperse (e(15% size
distribution) nanoparticles, and “magic number” (i.e., full shell and thus
enhanced stability) nanoclusters, Schwartz’s updated definition of homo-
geneous vs heterogeneous catalysts, and definitions of inorganic (“charge”)
and organic (“steric”) stabilization mechanisms for colloids and nano-
particles.
1
41, 269-276.
(
15) Homogeneous catalysts, including Noyori’s transfer-hydrogenation cat-
alysts,1
5e-i
are listed below. References 38 and 39 provided additional
systems, many of which probe the underlying mechanisms of ketone
hydrogenation. (a) Chin, C. S.; Park, S. C. Bull. Korean Chem. Soc. 1988,
9
1
1
4
, 260-261. (b) Geraty, S. M.; Harkin, P.; Vos, J. G. Inorg. Chim. Acta
987, 131, 217-220. (c) Harman, W. D.; Taube, H. J. Am. Chem. Soc.
990, 112, 2261. (d) Shafiq, F.; Eisenberg, R. J. Organomet. Chem. 1994,
72, 337-345. (e) Haack, K. J.; Hashiguchi, S.; Fuji, A.; Ikeriya, T.; Noyori,
(21) Acetone as well as acetonitrile hydrogenation by both Ir and Rh nanoclusters
was first noted in footnote 27 and Table B in Aiken, J. D., III; Finke, R.
G. Chem. Mater. 1999, 11, 1035.
R. Angew. Chem., Int. Ed. Engl. 1997, 36, 285-288. (f) Noyori, R.,
Hashiguchi, S. Acc. Chem. Res. 1997, 30, 97. (g) Matsumura, K.;
Hashiguchi, S.; Noyori, R. J. Am. Chem. Soc. 1997, 119, 8738. (h)
Yamakawa, M.; Ito, M.; Noyori, R. J. Am. Chem. Soc. 2000, 122, 1466.
(22) (a) O¨ zkar, S.; Finke, R. G. J. Am. Chem Soc. 2001, 124, 5796. (b) O¨ zkar,
S.; Finke, R. G. Langmuir 2002, 18, 7653-7662.
(23) Basic Research Needs For the Hydrogen Economy; Report of the Basic
Energy Sciences Workshop on Hydrogen Production, Storage and Use, May
13-15, 2003; Office of Science, U.S. Department of Energy: Washington,
DC, 2003; www.sc.doe.gov/bes/ hydrogen.pdf.
(
i) Noyori, R., Yamakawa, M.; Hashiguchi, S. J. Org Chem. 2001, 66,
7
931.
J. AM. CHEM. SOC.
9
VOL. 127, NO. 13, 2005 4801

A R T I C L E S
O¨ zkar and Finke
Results and Discussion
Acetone Hydrogenation Products, Selectivity, Conversion,
and Yield. In what we will hereafter term a standard condi-
2
4
tions nanocluster formation and concomitant acetone hydro-
genation experiment [(1,5-COD)IrCl]2 (3.6 µmol Ir) was placed
in neat acetone (0.5 mL, 6.7 mmol) in a Fischer-Porter (F-P)
pressure bottle along with a stir bar, all while working inside a
e1 ppm O2 inert-atmosphere N2 drybox. The F-P bottle was
then sealed, removed from the drybox, and attached to a
computer-interfaced pressure transducer apparatus that has been
described in detail previously.2 The F-P bottle was thermo-
stated at 22 ( 0.1 °C, and an Ir(0) nanocluster formation and
acetone hydrogenation reaction was started with the addition
of 40 psig of H2 under vortex stirring. We know that H2 mass-
transfer limitations (MTLs) are not introducing a significant error
in this work since, for our apparatus with acetone as the solvent
and under the vigorous stirring conditions employed, the H2
MTL limit is gca. 25 mmol of H2/h26 while the fastest
hydrogenation rate we observe is ca. 6.9 mmol of H2/h, Figure
S-1 of the Supporting Information.
2,25
Equation 1 shows the stoichiometry for the formation of the
Ir(0) nanoclusters, (2/n)Ir(0)n per 2HCl; the predicted H2 uptake,
Ir(0) formation, and cyclooctane evolution have been quanti-
tatively verified before for our other, closely analogous Ir(0)
nanocluster formation reactions.24 As a control, the expected
evolution of 2.0 equiv of cyclooctane (1.0 equiv per Ir) was
verified in the present case by quantitative GLC. Note the
+
-
Figure 1. TEM images (340 K) of 18 ( 5 Å Ir(0) nanoclusters formed by
the hydrogen reduction of [(1,5-COD)IrCl]2 (3.6 µmol Ir) during the
hydrogenation of neat acetone at 22 ( 0.1 °C. (a) Nanoclusters dispersed
on a silicon monoxide coated copper grid by dipping it into solution
immediately after the hydrogenation (all of which was done under N2 in a
drybox). (b) TEM of a solution of nanoclusters diluted with acetonitrile
and then dispersed on a chloroform-cleaned copper grid just before the
measurement. The sample was harvested after 0.7 h of hydrogenation (i.e.,
after ca. 70% conversion of the acetone) since the nanoclusters aggregate
to larger particles and ultimately into bulk metal as the hydrogenation
proceeds.
production of 2.0 equiv of H Cl (1.0 equiv for each Ir(I)
reduced2 ), eq 1. Confirmation of the presence of 18 ( 5 Å
Ir(0) nanoclusters is provided by their visualization by TEM,
Figure 1.
2b
During the nanocluster formation and concomitant hydro-
genation reaction at 22 ( 0.1 °C, NMR and GC/MS analyses
show that the neat acetone undergoes 95% conversion in 2 h
and has reached 100% conversion in 9 h with a selectivity of
0
.95 equiv (95%) 2-propanol and 0.025 equiv ()5% of the
converted acetone) of diisopropyl ether (plus 0.025 equiv of
Overall, the (2/n)Ir(0)n plus 2HCl nanocluster system exhibits
water by mass balance), eq 2. The diisopropyl ether is the
2
8
-1
a record initial activity (TOF(initial, minimum ) ) 1.92 s at
2 °C and 40 psig () 2.7 atm of H2)), the highest selectivity
95% 2-propanol) at the highest conversion (100%) for any
catalyst reported to date for acetone hydrogenation to 2-pro-
panol. Note that the true TOF may be closer to 11-fold larger
+
expected side product from the H -catalyzed condensation of
2
(
two isopropyl alcohol product molecules.27 Further details
regarding the composite reaction stoichiometry due to the
consecutive reactions of isopropyl alcohol and then diisopropyl
ether formation, plus use of that balanced reaction to calculate
properly the product yields, are provided in the Supporting
Information.
(
27) For a review of Williamson ether synthesis, see: Staude, E.; Patat, F. In
The Chemistry of the Ether Linkage; Patai, R., Ed.; Wiley: New York,
1967; pp 22-46.
(
28) (a) The TOFs and TTOs reported herein are those typically reported [TOF
)
(mol of product)/[(mol of total catalyst loading) s]; TTO ) TOF × time
)
(mol of product)/(mol of total catalyst loading)]. That is, the reported
TOFs and TTOs are not corrected for the amount of metal that is on the
surface of the catalyst and/or the actual number of active sites. In only one
case to date have the quantitative catalyst poisoning studies needed for
2
8b
such active site correction been done for a nanocluster. In that study,
9
-
+
(
(
(
24) Lin, Y.; Finke, R. G. J. Am. Chem. Soc. 1994, 116, 8335.
the P W15Nb O62 polyoxoanion- and Bu N -stabilized, 40 ( 6 Å Rh(0)
2
3
4
28b
25) Lin, Y.; Finke, R. G. Inorg. Chem. 1994, 33, 4891.
nanoclusters reported elsewhere have ∼31% of the Rh on the surface,
but only ∼(1.8X)% of that total Rh is catalytically active as determined by
2
CS poisoning experiments (where X is the number of active Rh sites
28b
26) (a) Experiments performed with increasing catalyst concentrations, Figure
S-2 of the Supporting Information, show the highest observed hydrogenation
rate under our specific conditions, apparatus, and vigorous stirring of ca.
poisoned by a single CS
2
, a working value for which is X ≈ 5). Applying
2
5 mmol of H
of 1.9 mM; hence, the H
mmol of H /h for our current system. For the related system of cyclohexene
hydrogenation in acetone as solvent in the same apparatus with vigorous
2
/h for a [(1,5-COD)IrCl]
2
nanocluster precursor concentration
those data to the present case for the sake of illustration, if this same ∼(1.8
× 5) ) 9% number is roughly applicable to the present case, results in the
reported TOF and TTO values herein being larger by a factor of 100/9 )
2
(gas) to H
2
(solution) MTL rate must be g25
2
-1
11. The resultant values would then be TOF ≈ 1.92 × 11 ) 21 s and
TTO ≈ 188000 × 11 ) ∼2000000. (b) Hornstein, B. J.; Aiken, J. D., III;
Finke, R. G. Inorg. Chem. 2002, 41, 1625.
2
6b
stirring, the H
Finke, R. G. J. Am. Chem. Soc. 1998, 120, 9545.
2
MTL rate is ∼17 mmol of H
2
/h. (b) Aiken, J. D., III;
4802 J. AM. CHEM. SOC.
9
VOL. 127, NO. 13, 2005

Iridium Catalysis of Neat Acetone Hydrogenation
A R T I C L E S
due to the amount of active Ir being estimated as ca. 9% from
Catalyst Lifetime. A catalyst lifetime experiment starting
with 0.6 mg of [(1,5-COD)IrCl]2 (i.e., 0.6 mM in Ir) in 3.0 mL
of acetone at 22 ( 0.1 °C and at a constant 40 ( 1 psig of H2
reveals 16400 TTOs of acetone hydrogenation over 32 h before
deactivation by aggregation into bulk metal occurs. The aVerage
TOF during this lifetime is, therefore, 0.14 s . Comparison of
this TOF to the ca. 4 times faster initial TOF provided above
makes it clear that the nanoclusters are deactivating as the
catalysis proceeds over its 32 h total lifetime. The clear yellow
solution turns to brown as the iridium(0) nanoclusters are first
formed, followed by the formation of a black, bulk iridium(0)
metal precipitate which becomes visible to the naked eye after
a few hours along with an eventually clear, colorless (i.e., Ir(0)-
free) solution. Hence, the catalytic lifetime is limited to 16400
TTOs in this case, with some of this TTO value being due to
bulk Ir(0) metal.
28
CS2 poisoning studies on a related Ir nanocluster system, TOF-
-
1
(
true, estimated) ≈ 21 s .
A control experiment was performed with the original intent
of seeing if adding molecular sieves to remove any water present
would influence the 5% diisopropyl ether formation. (The water
content of the Burdick & Jackson acetone employed is <0.2%,
which corresponds to 15.4 equiv of H2O per iridium used in a
standard conditions nanocluster formation and acetone hydro-
genation experiment; 1 equiv of water is also formed for every
equivalent of diisopropyl ether produced, eq 2.) However, an
interesting result was obtained which cannot be explained by
an effect of the molecular sieves on the water content but which,
instead, implicates a support interaction between the molecular
sieves and the Ir(0)n nanoclusters. Specifically, a standard
conditions hydrogenation of acetone at 22 ( 0.1 °C, starting
with [(1,5-COD)IrCl]2 (3.6 µmol Ir) in neat acetone (6.7 mmol)
with 0.2 g of predried 5 Å molecular sieves added, gave a
sigmoidal-shaped curve29 (Figure S-8 in the Supporting Infor-
mation). A 100% conversion of acetone to 2-propanol in 6 h is
-
1
Interestingly, a catalyst lifetime experiment starting with 0.6
mg of [(1,5-COD)IrCl] (i.e., 0.6 mM in Ir) in 27 mL of acetone
2
with 0.5 g of 5 Å molecular sieves added, at 22 ( 0.1 °C and
at a constant 40 ( 1 psig of H , reveals 188000 TTOs of acetone
2
-
1
seen, eq 3, with an initial TOF of 0.32 ( 0.02 s , as
hydrogenation selectively to 100% 2-propanol over 110 h before
deactivation by aggregation into bulk metal occurs. Again, due
to the formation of bulk metal, this TTO value is an upper limit
3
1
to the true nanoparticle TTO value. Nevertheless, this is the
highest reported value to date that we have been able to find as
well as a value which is 115-330-fold higher than those
demonstrated by H and 13C NMR spectra (Figure S-9 in the
Supporting Information). Note that the NMR spectra were
obtained from the raw hydrogenation product (i.e., without
purification); neither spectrum reveals any impurities or side
products, such as the previously observed diisopropyl ether
1
38f,g
reported in two recent patents (for ketones other than acetone).
Again, using the estimate of ∼9% of the total Ir being active,
the true TTO (estimated) per active Ir is likely something
28
approaching ∼2000000.
+
1
Requirement for Acid and Kinetic Evidence for a [H ]
Dependence. To test the hypothesis that the 1 equiv of H /
(
estimated detection limit 1%). The complete conversion of
+
acetone to one product, 2-propanol, at room temperature and
e3 atm of H2 in the presence of molecular sieves is an important
result. Additional control experiments were performed to
confirm that the molecular sieves were not just adsorbing the
diisopropyl ether (and thereby obscuring its detection; see the
Experimental Section for details). On the basis of kinetic studies
of the nanocluster formation in the presence of molecular sieves
Ir(0) formed, eq 1, is crucial for the reaction, the nanocluster
formation and acetone hydrogenation reaction were repeated,
but now with 1 equiv of added Proton Sponge (1,8-bis-
32
(dimethylamino)naphthalene), a strongly basic (conjugate acid
aqueous pKa ) 12.3) but weakly coordinating, preferred
+
22b
scavenger of H in such nanocluster syntheses. As expected,
.6 mM [(1,5-COD)IrCl]2 (i.e., 1.2 mM in Ir) and 1.2 mM (1
0
(Figure S-8 of the Supporting Information), the main effect of
equiv per Ir) Proton Sponge in acetone at 22 ( 0.1 °C exhibits
no detectable hydrogen uptake over 8 h. As a control to show
that an active catalyst had been formed, the same experiment
was performed (0.6 mM [(1,5-COD)IrCl]2 (i.e., 1.2 mM in Ir)
and 1.2 mM (1 equiv per Ir) Proton Sponge in acetone at 22 (
adding molecular sieves appears to be on the nanocluster
formation reaction (and then subsequent catalysis), presumably
by supporting the Ir(0)n nanoclusters on the molecular sieves.
An additional control experiment also produced an unexpected
result: deliberately added water (0.4 wt %, i.e., double the 0.2
wt % in Burdick & Jackson acetone) also led to a 100%
selectivity to 2-propanol, the amount of diisopropyl ether side
product being reduced to below our detection limit (e1%).
Apparently, added water leads to the precedented30 hydrolysis
of any diisopropyl ether present back to 2-propanol. However,
the reaction rate was slowed 2-fold and the catalyst stability
reduced, with bulk Ir(0)n metal precipitating from this control
reaction.
0
.1 °C), except with a 1.6 M concentration of the more easily
reduced substrate, cyclohexene. The resultant (2/n)Ir(O)n nano-
cluster catalyst was quite active, exhibiting the normal, sigmoi-
2
4
dal-shaped cyclohexene loss vs time curve (Figure S-5 of the
Supporting Information). No acetone hydrogenation was ob-
served in this control experiment as expected (and in which 1
equiv of Proton Sponge was present).
(
31) One caveat on the lifetime measurements: when the presence of a bulk
metal precipitate is noted, the TTOs given in Table 1 or in the main text
are an upper limit to the TTOs due to nanoclusters alone and as indicated
by placing the TTO number in brackets, [TTOs]. See also Table 1, p 4900,
in ref 25 for data showing that any bulk metal present generally has a
considerably lower surface area in comparison to that of the nanoclusters
and, therefore, a significantly slower rate of hydrogenation than the
nanoclusterssa fortunate situation that helps limit (the still nonzero,
however) contribution of bulk metal to the observed TTO number.
(
29) For a mechanism of transition-metal nanocluster formation, consisting of
slow, continuous nucleation (A f B, rate constant k ) and then fast,
autocatalytic surface growth (A + B f 2B, rate constant k ), a mechanism
1
2
that is proving to be quite general for transition-metal particle formation
under reducing conditions, see: (a) Watzky, M. A.; Finke, R. G. J. Am.
Chem. Soc. 1997, 119, 10382 and references therein. (b) Watzky, M. A.;
Finke, R. G.; Chem. Mater. 1997, 9, 3083. (c) Aiken, J. D., III; Finke, R.
G. J. Am. Chem. Soc. 1998, 120, 9545 and references therein to diffusive
agglomeration of nanoparticles. (d) Widegren, J. A.; Aiken, J. D., III; O¨ zkar,
S.; Finke, R. G. Chem. Mater. 2001, 13, 312, and references therein.
30) Adelman, R. L. J. Am. Chem. Soc. 1953, 75, 2678.
(32) Proton Sponge: (a) Brzezinski, B.; Schroeder, G.; Grech, E.; Malarski,
Z.; Sobczyk, L. J. Mol. Struct. 1992, 274, 75. (b) Brzezinski, B.; Glowiak,
T.; Grech, E.; Malarski, Z.; Sobczyk, L. J. Chem. Soc., Perkin Trans. 1991,
2, 1643.
(
J. AM. CHEM. SOC.
9
VOL. 127, NO. 13, 2005 4803

A R T I C L E S
O¨ zkar and Finke
+
1
order dependence on acid, [H ] , as well as a first order
1
dependence on the Ir precursor, [[{(1,5-COD)IrCl}2]initial] .
Consistent with this interpretation and assuming a general rate
law of the form shown in eq 4 for the sake of illustration, eq 5
is just a restatement of the stoichiometry back in eq 1.
Substitution of eq 5 into eq 4 followed by the appropriate algebra
yields eq 6 (the full details of the kinetic derivations implied
by eqs 4-9 are available in the Supporting Information).
Next, one can write the appropriate mass balance equation
(7a), which simplifies to eq 7b for the case where the nanocluster
formation reaction is fast
[
{(1,5-COD)IrCl}2]initial ) [{(1,5-COD)IrCl} ] + n[Ir(0) ]
2
n
(7a)
[{(1,5-COD)IrCl}2]initial ≈ n[Ir(0)n]
(7b)
Figure 2. Acetone hydrogenation ln(steady-state rate) vs ln [[(1,5-COD)-
IrCl]2] for the catalytic hydrogenation of acetone at 22 ( 0.1 °C. The linear
plot and slope of 1.9 ( 0.2 reveal an apparent second-order dependence on
the initial concentration of [(1,5-COD)IrCl]2, which is, however, as discussed
in the text, due to a first-order dependence on [H] plus a first-order
dependence on [[(1,5-COD)IrCl]2] plus their stoichiometric relationship,
and most of the initial [(1,5-COD)IrCl]2 is present as Ir(0)n
nanoclusters; that is, eq 7b results when n[Ir(0)n] . [{(1,5-
COD)IrCl}2]. Recall the GLC cyclooctane evolution experiment
(vide supra) which demonstrated that the Ir(0)n catalyst does,
indeed, evolve in e0.6 h according to eq 1; that is, the
simplification of eq 7a to eq 7b has experimental support.
Inserting eq 7b into eq 6, then, yields the desired eqs 8 and 9.
+
vide infra.
The acid-assisted nature of the acetone hydrogenation reaction
is further, quantitatively supported by initial kinetic studies
varying the [(1,5-COD)IrCl]2 from 0.2 to 1.9 mM. An ln/ln plot
of the acetone hydrogenation steady-state rate (i.e., the constant
rate post the induction period, Figure S2 of the Supporting
Information) vs the initial [(1,5-COD)IrCl]2 concentration,
Figure 2, is linear and shows a slope of 1.9 ( 0.2. That is, an
apparent second-order dependence on [(1,5-COD)IrCl]2 con-
d[acetone] d[H
]
2
-
)
)
dt
dt
k2,obs
a
b
]c+1 (8)
2 initial
[
acetone] [H ] [[{1,5-CODIrCl} ]
2
n
d[acetone]
d[H2]
]c+1 (9)
2 initial
-
)
∝ [[{1,5-CODIrCl} ]
2
centration is observed, -d[acetone]/dt ∝ [[(1,5-COD)IrCl]2] .
dt
dt
At first sight the (apparent) second-order dependence upon
the initial [(1,5-COD)IrCl]2 concentration is perplexing. How-
ever, a bit of reflection reveals what the apparent second-order
dependence on the [{(1,5-COD)IrCl}2]initial is telling us. To
understand this, first one must be aware that the measured
hydrogenation rate is the steady-state rate past a short, ca. 0.2
From eq 8, it is apparent if the reaction is first-order in [(1,5-
COD)IrCl]2 (i.e., for c ) 1 and with fast nanocluster formation
so that eq 7b is in effect), then the reaction will appear to be
second-order in [(1,5-COD)IrCl]2, eq 9, as is observed experi-
mentally.
(
0.1 h induction period, Figure S2. Second, TEM and a
In short and although a full kinetic study remains to be
performed, the key point of these initial kinetic as well as added
Proton Sponge studies is that H is essential; that is, the present
Ir(0) nanocluster catalyzed acetone hydrogenation reaction is
strongly acid-assisted.
22,29
cyclooctane evolution experiment
using GLC confirm that
+
the Ir(0)n catalyst has fully evolved, in e0.6 h, and thus by the
time the hydrogenation rate is measured, according to eq 1 so
that [{(1,5-COD)IrCl}2](initial) ≈ n [Ir(0)n]. Third, we know from
+
the Proton Sponge scavenging experiments that the evolved H
Demonstration of Reproducible Reaction Rates. Because
33
(as HCl, eq 1) is crucial for the facile catalysis. It follows,
Bradley has shown that the uncontrolled production (and loss)
+
-
therefore, and as shown in the kinetic derivations in eqs 4-9,
of volatile H Cl in nanoparticle synthesis reactions yields
nanoparticles with catalytic rates variable by up to (670%, it
was important for us to establish if our catalysis is reproducible
or not. We expected that it would be, since our system is a
closed system (a sealed Fischer-Porter pressure bottle) while
Bradley’s was by design an open system, so that the volatile
HCl could be lost. However, it was still important to test
experimentally the reproducibility of our system.
d[acetone] d[H
]
2
a
b
c
+ d
-
)
) k2,obs[acetone] [H ] [Ir(0) ] [H ]
2
n
dt
dt
(4)
2
n
+
[
Ir(0) ] ) 2[H ]
(5)
n
d[acetone]
k
2,obs
(33) (a) The difference between nanoclusters and the historically better known
a
b
c+1
-
)
[acetone] [H ] [Ir(0) ]
33
2
n
nanocolloids is convincingly illustrated by Bradley’s seminal paper
dt
n
showing that irreproducibility in the colloid’s composition of surface-bound
-
+
-
Cl /H Cl , the latter being the byproduct of the nanocluster formation
reaction, is the origin of the up to 670% irreproducibility in the rate of
(
for the d ) 1 case) (6)
+
-
catalysis by Pt/PVP/H Cl nanocolloids: K o¨ hler, J. U.; Bradley, J. S. Catal.
Lett. 1997, 45, 203. (b) Five-fold (i.e., 500%) rate variations are seen for
that the apparent second-order dependence on the Ir(I) precursor
is reflecting the concomitant, stoichiometric production of 1
equiv of H per 1 equiv of Ir(I) reduced, eq 1, and then a first-
2
the photoreduction of CO catalyzed by a series of 10 different batches of
the Pd colloids: Willner, I.; Mandler, D. J. Am. Chem. Soc. 1989, 111,
+
1330.
4804 J. AM. CHEM. SOC.
9
VOL. 127, NO. 13, 2005

Iridium Catalysis of Neat Acetone Hydrogenation
A R T I C L E S
In four repeat acetone hydrogenations under standard condi-
tions, we find a reproducible rate of 6.9 ( 0.2 mmol of H2/h
(3%) for the acid-assisted, iridium(0) nanocluster catalyzed
hydrogenation of acetone (and over the use of several batches
of the commercially available, well-known, pure precatalyst
(1,5-COD)IrCl]2, to answer the query of a reviewer). The
again with 100% conversion: after 7.5 h, NMR and GLC/MS
analyses show that the deuterated acetone is completely
converted to 2-propan-1,1,1,3,3,3-d6-ol (95%) plus diisopropyl
ether and, by mass balance, water (0.025 equiv each). Signifi-
cantly, the MS analysis is definitive in showing that both the
2-propanol and diisopropyl ether have all the D atoms in the
methyl groups, eq 10, and as shown by the data supplied in the
(
[
reasons that our nanocluster catalysis is 230-fold more repro-
ducible than Bradley’s important, illustrative system are (i) our
closed, Fischer-Porter bottle reaction system noted above which
+
-
retains all the volatile H Cl , (ii) our use of a structurally and
compositionally well characterized precursor, a known require-
ment for reproducible nanocluster syntheses and catalytic
rates,1 and (iii) the fact that we carry out the reaction under
exactly the same conditions each time (the same catalyst and
substrate concentrations, temperature, pressure, and stirring rate,
for example; see the Experimental Section for further details).
Attempted Extension of the Catalytic Lifetime of the
Iridium(0) Nanoclusters by Using the Polar Solvent Pro-
pylene Carbonate To Extend the Nanocluster Stability.
8c
Experimental Section (Complete Hydrogenation of Deuterated
Acetone). The enol of acetone is not involved in the Ir(0)
nanocluster mechanism under our catalytic conditions.36
A
Because we3 and others have shown that the polar, non-
coordinating solvent propylene carbonate is as good as any
presently available to improve the lifetime, isolabilty, and
4
35
literature study of a C-supported Ru catalyst has shown that a
mixture of 2-propanol and deuterated acetone also gives
37
2-propan-1,1,1,3,3,3-d -ol and acetone, with no enolization.
6
-
stability of anionically (e.g., Cl ) stabilized transition-metal
Proposed Minimum Mechanism for the Acid-Assisted
Catalytic Hydrogenation of Acetone. There is excellent
product, structural, kinetic, mechanistic, and computational work
in the literature of homogeneous, organometallic complex
catalysis of ketone and other polar group hydrogenations (i.e.,
CdN, imines) from the groups of Vos, Harman/Taube, Eisen-
nanoclusters, experiments examining the hydrogenation of
acetone in propylene carbonate were undertaken (i.e., rather than
in neat acetone as has been done up to this point). The key
result of these experiments (which are descried in detail in the
Supporting Information) is that a more stable, clear-brown
solution of 21 ( 3 Å nanoclusters (Figure S-3 in the Supporting
Information) is produced which remains stable and catalytically
active at room temperature under N2 for even 3-4 weeks after
the initial acetone hydrogenation reaction. However, the maxi-
mum rate of hydrogenation (e2 mmol of H2/h) is 3.5-fold
slower than that in neat acetone (6.9 ( 0.2 mmol of H2/h).
Furthermore, a catalyst lifetime experiment starting with 0.6
mg of [(1,5-COD)IrCl]2 (i.e., 0.6 mM in Ir) and 3.0 mL of
acetone in 2.5 mL of propylene carbonate at 22 ( 0.1 °C and
1
5,38,39
berg, Bullock, Norton, Casey, and Noyori.
The essence
of that combined literature that is relevant to the present work
is (i) that “ionic hydrogenations” of polar groups are facile when
+
a M-H and H component are both present, and (ii) that
mechanisms ranging from those involving a prior protonation
+
of acetone, to a concerted M-H and H addition to ketones in
a 1,2 fashion, are well documented.
Given the complex, always challenging mechanistic issue of
the timing of steps where two or more species are involved
-
+
40 ( 1 psig of H2 shows that the Ir(0) nanoclusters in propylene
(i.e., M-H /H in the present case), the most plausible,
deliberately minimal (Occam’s razor) mechanism in our system
can be written concisely as shown in eqs 11a and 11b.
carbonate afford only 4800 TTOs of acid-assisted acetone
hydrogenation over 122 h before deactivation by aggregation
into bulk metal. This catalytic lifetime in propylene carbonate
is 3.4-fold lower, despite the 3.8-fold longer reaction time, in
comparison to the TTOs observed for the superior, neat acetone
system. The catalysis of neat acetone hydrogenation is, therefore,
4
3
an important as well as “green” component of the present
4
3
invention (“green” since cosolvent is not only not needed, it
is actually a disadvantage).
This deliberately brief mechanistic scheme is consistent with
and supported by the experiments using D-labeled acetone, the
kinetic studies implicating a first-order dependence on both the
Evidence Regarding the Catalytic Mechanism Using
+
D-Labeled Acetone. Given that H is a requirement in our
+
acetone reduction system, a reasonable question is whether the
enol of acetone is the actual species being reduced. The
prediction from the literature here is clear: a direct 1,2 addition
of H2 (D2) across the ketone is expected, with the enolization
[Ir] and [H ], as well as the above-noted, excellent precedent
(
36) Our results require that the enolization of acetone with dry H Cl- is slow
+
under our conditions given the D-isotope is involved, as otherwise the
+
+
following reversible reaction would have scrambled D /H in the acetone
+
+
methyl groups: CD
3
C(O)CD
3
+ H f CD
2
+d C(OH)CD
3 2
+ D , then CD d
1
mechanism becoming important only at higher temperatures.
+
C(OH)CD
3
+ H f HCD C(O)CD + H .
2
3
(
37) Yamashita, M.; Kawamura, T.; Suzuki, M.; Saito, M. Bull. Chem. Soc.
Jpn. 1991, 64, 272.
To test the mechanism, D-labeled acetone-d6 and H2 were
employed. In an experiment starting with 0.6 mM [(1,5-COD)-
IrCl]2 (i.e., 1.2 mM in Ir) in deuterated acetone, (CD3)2CO, at
(
38) (a) Bullock, R. M.; Song, J.-S. J. Am. Chem. Soc. 1994, 116, 8602. (b)
Bullock, R. M.; Luan, L.; Song, J.-S. J. Org. Chem. 1995, 60, 7170. (c)
Bullock, R. M.; Voges, M. H. J. Am. Chem. Soc. 2000, 122, 12594. (d)
Voges, M. H.; Bullock, R. M. J. Chem. Soc., Dalton Trans. 2002, 759. (e)
Song, J.-S.; Szalda, D. J.; Bullock, R. M. Organometallics 2001, 20, 3337.
(f) Bullock, M. R.; Schlaf, M.; Hauptman, E. M. US Patent 6,462,206,
Oct 8, 2002. (g) Bullock, M. R.; Kimmich, B. F.; Fagan, P. J.; Hauptman,
E. M. US Patent 6,613,923, Sept 2, 2003.
22 ( 0.1 °C, the same product distribution is obtained as before,
(
34) Hornstein, B. J.; Finke, R. G. Chem. Mater. 2003, 15, 899-909.
(
35) Reetz, M. T.; Lohmer, G. J. Chem. Soc., Chem. Commun. 1996, 1921-
1
922.
J. AM. CHEM. SOC.
9
VOL. 127, NO. 13, 2005 4805

A R T I C L E S
O¨ zkar and Finke
from the mechanistic organometallic literature. That extant
literature of ketone and other, polar functional group hydrogena-
(3) The most probable mechanism of these ionic hydrogena-
tions3
8,39
is that anticipated from the literature,
1,38,39
a 1,2
1
5,38,39
38,39
+
tions
(a) effectively predicts that the mechanism of
addition of the expected
M-H/H components across the
+
nanocluster catalysis in the present work should be H -assisteds
polar CdO bond. Strong support for this minimal mechanism
is provided by labeling studies using acetone-d6, kinetic studies
as it is; and (b) therefore offers strong support for the findings
+
+ 1
herein of H -assisted acetone hydrogenation. Moreover, the pKa
revealing a [H ] dependence, complete inhibition experiments
+
-
+
values of H Cl and protonated acetone, (CH3)2CdOsH , are
using Proton Sponge which demonstrate the requirement for
acid, and the excellent mechanistic precedent in the organome-
40
similar (pKa ≈ -7), indicating that a substantial fraction of
the acetone should be protonated under our (highly acidic)
reaction conditions (and assuming that sizable differential
3
8,39
tallic literature.
(4) The present system is “relatively green” in its environ-
mental impact in that it satisfies 7 of the 12 proposed principles
solvent effects on these pKa values are not present). The reader
interested is referred to the cited literature1
5,38,39
for further
43
of green chemistry detailed elsewhere including that (i) it is
details of the most probable, additional mechanistic steps
underlying the present Ir-H- and H -assisted acetone hydro-
genation.
100% selective and minimizes any byproducts or waste, (ii) it
exhibits atom economy by incorporating all the reactants into
the products, (iii) it is solventless (i.e., uses neat acetone as the
substrate/solvent), (iv) it has relatively low energy requirements
due to its room temperature and e3 atm pressure conditions,
+
Summary
The main findings of this work, including the limitations, as
well as the implications or predictions of our findings, can be
summarized as follows.
(v) it is catalytic not stoichiometric, (vi) it does not use any
derivatization (blocking or protecting/deprotecting groups; in
the present case this is effectively a restatement of the fact that
the reaction is accomplished catalytically), and (vii) real-time
monitoring is easy by following the H2 pressure loss or the
(1) A system for the room temperature, acid-assisted hydro-
genation of acetone by Ir(0) nanoclusters has been developed,
one which provides exceptional (record) activity, up to 100%
selectivity at 100% conversion, and up to 188000 TTO catalytic
lifetime. The system is easily and reproducibly prepared from
commercially available [(1,5-COD)IrCl]2 and H2; the use of neat
1
products by GLC or H NMR, for example.
Some implications or as of yet untested predictions of this
work are also apparent, as listed below.
(
i) The extension to other, polar functional group selective
+
-
acetone, and the 1 equiv of H Cl generated per Ir reduced
that is retained in the closed system, is an important component
of the present discovery.
39d
reductions is expected, since as McGee and Norton have noted
“
ionic hydrogenations lend themselves to selecting polar CdX
bonds (for reduction) over CdC bonds”. Imine, CdN reductions
(2) The addition of molecular sieves yields an enhanced
+
via M-H/H ionic hydrogenations are of course already
catalytic lifetime of 188000 TTOs, perhaps because the nano-
clusters become supported on the molecular sieves, thereby
reducing the nanocluster’s tendency to agglomerate. Use of the
polar, nanocluster-stabilizing solvent propylene carbonate did
yield more stable, albeit less catalytically active, Ir(0) nano-
clusters. The current Achilles heel of this system then, as with
most soluble nanoclusters, is their limited catalytic lifetime and
1
5f,39d,e
known,
and we have already observed CH3CN reduction
21
to CH3CH2NH2 by our Ir(0)/HCl catalyst system. Also of
interest, but not established by this first report, are the scope of
higher ketones, R-C(O)-R′, that can (or cannot) be reduced,
and the functionality that can (or cannot) be tolerated in more
complex CdX bond reductions (i.e., acid-sensitive groups are
expected to prove problematic).
1
8,19
instability at raised temperatures.
for this reason, expected to have any impact on the acetone to
-propanol heat pump problem, since temperatures in the 200
The present system is not,
(ii) The present system should be immediately extendable to
Rh(0)/HCl nanoclusters since we have shown that nanoclusters
of other metals are obtained from analogous, commercially
available, [(1,5-COD)RhCl]2 and [(1,5-COD)MCl2] systems
2
3
°
C range are preferred there. However, the factors determining
the stability of modern transition-metal nanoclusters are only
(
M(II) ) Pd, Pt, Ru) if appropriate conditions are employed in
22
just now becoming better understood, so that improvements
44
the different cases. Studies are in progress to see if the general
reactivity trend found for γ-Al2O3-supported metals12 of Pt >
Tc ≈ Rh > Pd > Ru > Ni ≈ Re > Co is the same or different
35,41
in this area can be anticipated (and are beginning to appear
).
And simply supporting the nanoclusters, for example, by
performing the reaction with the mesoporous silica SBA-15
present, may lead to improved catalysis.42
4
4
for nanocluster catalysts.
(
iii) Last, the present results may be of at least general interest
+
-
(
39) (a) Tooley, P. S.; Ovalles, C.; Kao, S. C.; Darensbourg, D. J.; Darensbourg,
M. Y. J. Am. Chem. Soc. 1986, 108, 5465. (b) Song, J.-S.; Szalda, D. J.;
Bullock, R. M.; Lawrie, C. J. C.; Rodkin, M. A:; Norton, J. R., Angew.
Chem., Int. Ed. Engl. 1992, 31, 1233. (c) Smith, K.-T.; Norton, J. R.; Tilset,
M. Organometallics 1996, 15, 4515. (d) McGee, M. P.; Norton, J. R. J.
Am. Chem. Soc. 2001, 123, 1778. (e) Casey, C. P.; Singer, S. W.; Powell,
D. R.; Hayashi, R. K.; Kavana, M. J. Am. Chem. Soc. 2001, 123, 1090.
40) Gordon, A. J.; Ford, R. A. The Chemist’s Companion, A Handbook of
Practical Data, Techniques and References; Wiley-Interscience: New York,
to other problems, notably the need for H X -tolerant, expanded
lifetime catalysts for C-X + H2 f C-H + H X hydro-
+
-
dehalogenation catalysis important in environmental remedia-
(42) The catalytic activity of SBA-15-supported Ir(0) nanoclusters for acetone
hydrogenation was tested in an initial experiment as follows: a standard
conditions acetone hydrogenation was performed by starting with 0.6 mM
(
(
1
972; p 60.
2 2
[(1,5-COD)IrCl] in acetone (3.0 mL) at 22.0 ( 0.1 °C and 40 psig of H
41) (a) Aiken, J. D., III; Finke, R. G. J. Am. Chem. Soc. 1999, 121, 8803.
in the presence of SBA-15 (50 mg, corresponding to 1.4% (by weight) Ir
on the support); we thank Prof. T. D. Tilley at the University of California
at Berkeley for providing us the sample of SBA-15. The resulting Ir(0)
Note that the true active-site-corrected TOF and TTO for the polyoxoanion-
+
and Bu
4
N -stabilized nanoclusters in that study, based on CS
2
poisoning
-
1
+
-
studies (the first kind for a nanocluster system), are TOF ≈ 380900 h
nanoclusters/H Cl /SBA-15 system converted neat acetone to 95% 2-pro-
panol and diisopropyl ether plus water by mass balance (0.025 equiv each),
34
and TTO ) 2142000. (b) Li, Y.; El-Sayed, M. A. J. Phys. Chem. B 2001,
1
1
05, 8938. (c) Narayanan, R.; El-Sayed, M. A. J. Am. Chem. Soc. 2003,
25, 8340. (d) Interesting early work, which provides high TOF and TTO,
2
at 12.1 ( 2 mmol of H /h, ca. 2-fold faster than in the absence of SBA-15.
(43) Poliakoff, M.; Fitzpatrick, J. M.; Farren, T. R.; Anastas, P. T. Science 2002,
3
3
but in an ill-characterized colloidal system where the true catalyst is not
known for certain, is Beller, M.; Fischer, H.; K u¨ hlein, K.; Reisinger, C.-
P.; Herrmann, W. A. J. Organomet. Chem. 1996, 520, 257.
297, 807.
(44) Finney, E. E.; O¨ zkar, S.; Finke, R. G. Unpublished results and experiments
in progress.
4806 J. AM. CHEM. SOC.
9
VOL. 127, NO. 13, 2005

Iridium Catalysis of Neat Acetone Hydrogenation
A R T I C L E S
tion.45 Sonochemically prepared Mo2C and W2C catalysts, for
assistance of Dr. Eric Schabtach or JoAn Hudson, using the sample
preparation procedure and Philips CM-12 TEM instrument with a 70
µm lens operating at 100 kV and with a 2.0 Å point-to-point resolution,
as described in detail previously.2
Sample Preparation for TEM. The solutions used for the TEM
analysis were prepared by a standard conditions hydrogenation of
example, operate at 225-325 °C with catalyst half-lives as long
as 600 h and catalytic activities of 101 10 molecules‚g ‚s ,
7-
20
-1 -1 46
2,24,25
so that nanoclusters of greatly enhanced stability would be
needed before any such application would become possible.
4
7
Other possible applications can be imagined as well.
acetone (0.5 mL, 6.7 mmol) starting with 1.2 mg of [(1,5-COD)IrCl]
3.6 µmol of Ir) either in neat acetone or in propylene carbonate (2.5
2
(
Experimental Section
mL) at 22.0 ( 0.1 °C. In neat acetone the TEM sample was harvested
from the solution after 0.7 h of hydrogenation because the nanoclusters
aggregate to larger particles and ultimately into bulk metal as the
hydrogenation proceeds. In propylene carbonate the TEM sample was
harvested from the solution after 11 h, a time when cyclooctane
Materials. All commercially obtained compounds were used as
received unless indicated otherwise: acetone was purchased from
Burdick & Jackson (water content <0.2%) and was purged with argon
and transferred into a nitrogen atmospheres drybox before use. It is
2
evolution experiments show that all the [(1,5-COD)IrCl] has been
known that the source and H
reproducible nanocluster syntheses.
9.7%, anhydrous, water content <0.002%, packaged under N
transferred into the drybox and used as received. Cyclohexene (Aldrich,
9%) was purified by distillation over sodium under argon and stored
in the drybox. Deuterated NMR solvents CDCl , (CD CO, and CD Cl
Cambridge Isotope Laboratories) were received in 1 mL glass ampules
2
O content of the acetone both matter for
2
4,25
converted to Ir(0) nanoclusters. (This is imporant, since it has recently
Propylene carbonate (Aldrich,
) was
n
been shown that TEM of samples still containing organometallic
precursors can lead to TEM-induced nanoparticle formation, thereby
9
2
48
introducing artifacts into the TEM analysis. ) After the hydrogenation
was stopped, the F-P bottle was detached from the hydrogenation line
via its quick connects and brought back into the drybox, and its solution
was quantitatively transferred with a disposable polyethylene pipet into
a clean, 5 mL screw-capped glass vial. The vial was sealed and sent to
the University of Oregon for TEM investigation. There, one drop of
the sample solution was added to 1 mL of acetonitrile, in air, just before
a TEM image was obtained, to yield a clear amber, homogeneous
solution (in general, no bulk metal was visible by the naked eye at any
time unless otherwise indicated). A drop of this solution was then
dispersed on a chloroform-cleaned, carbon-coated Cu TEM grid.
Particle Size Measurements. Particle size analysis was performed
using the public domain NIH Image 1.62 program developed at the
U.S. National Institutes of Health, available on the Internet at http://
9
3
)
3 2
2
2
(
which were transferred into a Vacuum Atmospheres drybox for NMR
sample preparations therein.
Analytical Procedures. Unless otherwise reported all reaction
solutions were prepared under oxygen- and moisture-free conditions
using a Vacuum Atmospheres nitrogen drybox (<1 ppm O
2
). High-
resolution solution NMR spectra were taken on a Varian INOVA-300
1
13
31
instrument ( H, 300.115 MHz; C, 75.472 MHz; P, 121.489 MHz).
Gas chromatography-mass spectrometry (GC-MS) was performed
using a Hewlett-Packard 5890 series II GC instrument with an MSD
5
0
970 B. The GC instrument was equipped with a 30 m (0.25 mm i.d.,
.25 µm film thickness) Supelco SPB-1 column. The ionizing voltage
4
9
rsb.info.nih.gov/nih-image/, and as described in detail elsewhere.
was 70 eV. The GC parameters were as follows: initial temperature,
0 °C; initial time, 3 min; solvent delay, 2 min; temperature ramp, 10
C/min; final temperature, 270 °C; final time, 5 min; injector port
temperature, 280 °C; detector temperature, 290 °C; injection volume,
5
°
Hydrogenations. All the nanocluster formation and hydrogenation
reactions were carried out on the previously described custom-built
pressurized hydrogenation apparatus.2
2,25,29
0
.1 µL. The mass marker calibration of the GC-MS instrument was
Curve Fits of the Hydrogen Uptake Data and Data Handling.
performed using perfluorotributylamine.
2
Curve-fitting of the H pressure (or, equivalently, the cyclohexene, or
Transmission Electron Microscopy (TEM). TEM analyses were
performed as before at the University of Oregon with the expert
acetone) vs time data was performed, as described previously in
detail,2
2,25,29
using the software package Microcal Origin 3.5.4, which
is a nonlinear regression subroutine (RLIN) and uses a modified
Levenberg-Macquard algorithm.50
General Procedure for Nanocluster Formation and Acetone
Hydrogenation (Standard Conditions) Experiments. These were
(
45) A few lead references to the more extensive area of C-X bond hydro-
dehalogenation/remediation: (a) Richmond, T. G. Top. Organomet. Chem.
1
999, 3, 243-269. (b) Schrick, B.; Blough, J. L.; Jones, A. D.; Mallouk,
T. E. Chem. Mater. 2002, 14, 5140-5147. (c) Oxley, J. D.; Suslick, K. S.
Presented at the 225th ACS National Meeting, New Orleans, LA, 2003;
Paper INOR-072. See also: Oxley, J. D.; Mdleleni, M.; Suslick, K. S.
Presented at the 224th ACS National Meeting, Boston, MA, 2002; Paper
INOR-100. (d) Yu, H.; Kennedy, E. M.; Azhar U., M.; Dlugogorski, B. Z.
Appl. Catal., B 2003, 44 (3), 253-261. (e) Park, C.; Menini, C.; Valverde,
J. L.; Keane, M. A. J. Catal 2002, 211 (2), 451-463.
generally done by our established protocol, but with acetone as the
25,29a
hydrogenation substrate.
Information.
Details are given in the Supporting
Acetone Hydrogenation Experiments. Additional Details. In a 2
dram glass vial, 1.2 mg (3.6 µmol of Ir) of the precatalyst [(1,5-COD)-
IrCl] was weighed and dissolved in 0.5 mL (6.7 mmol) of acetone,
2
(46) (a) Oxley, J. D.; Mdleleni, M. M.; Suslick, K. S. Catal. Today 2004, 88
(3-4), 139-151. (b) Sonochemical preparation of the metal carbides:
Suslick, K. S.; Hyeon, Fang, M. Chem. Mater. 1996, 8, 2127.
47) (a) Benzene is commercially converted to phenol, plus the coproduct
acetone, by a current process that involves the alkylation of benzene with
propene to give cumene, cumene autoxidation to cumene hydroperoxide,
and then acid-catalyzed decomposition of cumene hydroperoxide to phenol
and acetone.4 A problem with this coproduct process is that the demand
for phenol is outstripping the demand for acetone, and for such a large
volume of commodity chemicals, even a small difference in demand leads
to a glut of the coproduct in smaller demand, in this case acetone. While
added with a 1.0 mL gastight syringe, to yield a clear, yellow solution.
The solution was then transferred via a disposable polyethylene pipet
into a new 22 × 175 mm Pyrex culture tube containing a new 5/16 ×
5/8 in. Teflon-coated stir bar. The culture tube was then sealed inside
the F-P pressure bottle and brought outside the drybox, and the
hydrogenation was carried out exactly as described in the previous
section and its Supporting Information. The hydrogenation was
continued until the complete conversion of acetone was observed as
judged by monitoring the hydrogen uptake (Figure S-1 in the Supporting
Information). When no more hydrogen uptake was observed, the
experiment was stopped and the F-P bottle was sealed, disconnected
(
7b
direct routes for benzene to phenol using, for example, N
2
O and Fe catalysts
are being developed, one can at least conceive of a route that recycles
the acetone by its H reduction to 2-propanol, and then dehydration of the
-propanol to propene which could be reused. This would be in essence a
/O process, analogous to the one that is being explored for other
4
7c
2
2
H
2
2
oxygenations (e.g., propene to propylene oxide) but using Au/TiO
2
catalysts.4 However, at present such a recycled acetone and propene
process is not of commercial interest, it appears, due to the fact that the
acetone can be sold as long as the price is low enough. This situation could
change in the future, however, making a process that recycles some of the
acetone of future interest. (b) Kirk Othmer Encyclopedia of Chemical
Technology, 4th ed.; Wiley: New York, 1996; Vol. 18, pp 592-602. (c)
Panov, G. I.; Uriarte, A. K.; Rodkin, M. A.; Sobolev, V. I. Catal. Today
7d
(48) (a) Jaska, C. A.; Manners, I. J. Am. Chem. Soc. 2004, 126, 9776. (b) Hagen,
C. M.; Widegren, J. A.; Maitlis, P. M.; Finke, R. G. Is It Homogeneous or
Heterogeneous Catalysis? Compelling Evidence for Both Types of Catalysts
5
Derived from [Rh(η -C
5
Me
5
)Cl
2
]
2
as a Function of Temperature and
Hydrogen Pressure. J. Am. Chem. Soc. 2005, in press.
(49) Woehrle, G. H.; Hutchison, J. E.; O¨ zkar, S.; Finke, R. G. Computer-Based,
Statistically Valid Analysis of Nanoparticle Transmission Electron Mi-
croscopy Data Using a Public Domain Image-Processing Program, Image.
Mater. Charact., submitted for publication.
(50) Press, W. H.; Flannery, B. P.; Teukolsky, S. A.; Vetterling, W. T. Numerical
Recipes; Cambridge University: Cambridge, U.K., 1989.
1
998, 41, 365 and references therein, notably to the Solutia process for the
direct conversion of benzene to phenol using N O as the oxidant. (d)
2
Hayashi, T.; Tanaka, K.; Haruta, M. J. Catalysis 1998, 178, 566 and
subsequent papers in the area of Au/TiO
propene.
2
-catalyzed H
2
/O
2
epoxidation of
J. AM. CHEM. SOC.
9
VOL. 127, NO. 13, 2005 4807

A R T I C L E S
O¨ zkar and Finke
from the hydrogenation line, and transferred back into the drybox. After
those of the instrument’s library spectra (WILEY6N.L) and also those
available in the literature51 thereby unambiguously identifies the major
the H
withdraw a ca. 0.05 mL aliquot from the culture tube. This reaction
solution aliquot was then added to 1 g of CDCl in an individual glass
2
pressure was released, a 9 in. glass Pasteur pipet was used to
3 2 3 2 3 2
(CD ) CHOH, and minor (CD ) CH-O-CH(CD ) , products. Neither
3
NMR nor GC-MS showed the presence of any detectable unreacted
acetone (i.e., e1%).
ampule, mixed under agitation using the Pasteur pipet, and then
transferred into an NMR sample tube which was subsequently sealed
Catalyst Lifetime Experiments. The catalyst lifetime experiments
were performed by following our established protocol, but with acetone
as the hydrogenation substrate.25 Details are given in the Supporting
Information.
1
and then brought out of the drybox. The H NMR spectrum of this
solution showed that the acetone is completely converted to 2-propanol
(0.95 equiv, δ (ppm) 4.01 heptet, 1.92 broad, 1.20 doublet) and
diisopropyl ether (0.025 equiv, δ (ppm) 3.64 heptet, 1.12 doublet). The
C NMR spectrum verified the product identifications (δ (ppm) 64.57,
Control Experiment of Acetone Hydrogenation Starting with
[(1,5-COD)IrCl] Plus 1 equiv of Proton Sponge. The same acetone
2
13
2
5.51 for 2-propanol, 68.60, 23.05 for diisopropyl ether). Another 0.1
hydrogenation experiment as in the standard conditions hydrogenation,
but with 1 equiv of added Proton Sponge,2 was carried out to test the
role of protons in the catalytic hydrogenation of acetone. After 8 h,
there was no detectable hydrogen uptake as monitored by measuring
the hydrogen pressure.
Control Experiments of Cyclohexene Hydrogenation (Standard
2
Conditions) Starting with [(1,5-COD)IrCl] Plus Cyclohexene in the
Absence or Presence of 1 equiv of Proton Sponge. Two standard
conditions cyclohexene hydrogenation experiments were performed
2b
mL aliquot of the solution in a culture tube was used for GC-MS
analysis, which showed only 2-propanol, (CH CHOH, as the major
product and diisopropyl ether, (CH CHOCH(CH , as the minor
product. The comparison of the fragmentation pattern in the MS spectra
with those of the library spectra unambiguously identifies the major
and minor products. Neither NMR nor GC-MS showed the presence
of detectable, unreacted acetone. (The residual acetone detection limit
is estimated as e1%.)
3 2
)
3
)
2
3 2
)
Complete Hydrogenation of Acetone Selectively to 2-Propanol.
The same hydrogenation experiment as in the previous section was
carried out, except that 0.2 g of predried 5 Å molecular sieves were
added. An acetone loss vs time plot (Figure S-8 in the Supporting
Information) exhibits a sigmoidal shape with a short induction period
of 0.5 ( 0.1 h, after which a linear hydrogen uptake of 4.1 ( 0.2
2
starting with 1.2 mg (3.6 µmol) of [(1,5-COD)IrCl] plus 0.5 mL (4.9
mmol) of cyclohexene in 2.5 mL of acetone without or with 1 equiv
of Proton Sponge. In the experiment without Proton Sponge, the fast
cyclohexene hydrogenation was followed by the slower hydrogenation
of acetone (Figure S-4 in the Supporting Information), while in the
experiment with Proton Sponge there was no detectable hydrogen
uptake after the cyclohexene hydrogenation (Figure S-5 in the Sup-
porting Information).
mmol of H
B nucleation (rate constant k
surface-growth (rate constant k
2
/h is observed. The sigmoidal curve is well-fit by the A f
1
) and then A + B f 2B autocatalytic
2
) mechanism for nanocluster formation
Control Experiments with Added Molecular Sieves or Added
H O. Control experiments were performed addressing the effect of
2
2
8
-1
as first detailed elsewhere with k
1
) 0.066 ( 0.003 h and k
2
)
3
-1 -1
1
13
(
1.3 ( 0.1) × 10 M
h
. The H and C NMR spectra of the solution
added molecular sieves or water on the reaction’s selectivity (eq 2).
The first control experiment was performed by adding a known amount
of water to acetone to address the following question: Does a 2-fold
increase in the amount of water affect the catalytic activity or the
product selectivity of the system? A standard conditions hydrogenation
after reaction show 2-propanol as the only detectable product (Figure
S-9 in the Supporting Information); GC-MS also is unable to detect
any unreacted acetone or side products (estimated detection limits
e1%).
Complete Hydrogenation of Deuterated Acetone. The same
hydrogenation experiment as in the previous section, but with deuterated
acetone, was carried out to obtain mechanistic insight into the
hydrogenation reaction (Figure S-6 in the Supporting Information). The
2
of acetone starting with [(1,5-COD)IrCl] (3.6 µmol of Ir) in neat
acetone (6.7 mmol) containing 0.4% water (recall that the Burdick &
Jackson acetone contains <0.2% water) at 22 ( 0.1 °C completely
(100%) converts acetone to 2-propanol as the sole product in 11 h (eq
1
1
13
H NMR spectrum of this solution showed that the deuterated acetone
3, main text) as determined by H and C NMR spectra. This doubling
of the water content also reduced the rate of acetone hydrogenation by
2-fold vs that performed in the Burdick & Jackson acetone with its
water content of <0.2%. The hydrogenation with the added water starts
is completely converted to propan-1,1,1,3,3,3-d
6
-2-ol (0.95 equiv) or
signal was
observed. The C{ H} NMR spectrum gives a singlet for the CH carbon
diisoropyl ether (0.025 equiv); in both cases no CH
3
13
1
1
3
2
-1
and a heptet with J( C- D) ) 77.4 Hz for the CH
3
group, while the
with an initial TOF of 0.29 ( 0.02 s , and then slows as the reaction
1
3
coupled C NMR spectrum shows additional splitting only for the CH
proceeds (Figure S-10 in the Supporting Information). After 11 h of
hydrogenation, the solution is clear with black precipitates on the
coating; that is, nanoclusters are formed, but are not stable and aggregate
into big particles and, ultimately, into bulk metal.
1
3
1
signal to a doublet, J( C- H) ) 140 Hz. These results verify that the
-propanol product carries the deuterium atoms only on the terminal
2
carbons and the hydrogen only on the second carbon and the hydroxyl,
OH, group. Another 0.1 mL aliquot of the solution in a culture tube
A second control addressed the question of whether disiopropyl ether
could be adsorbed by added molecular sieves, thereby obscuring its
detection. This control was carried out simply by leaving a mixture of
0.95 equiv (95%) 2-propanol plus 0.025 equiv diisopropyl ether,
produced in a standard conditions hydrogenation of normal acetone,
was used for GC-MS analysis; only propan-1,1,1,3,3,3-d
CHOH, as the major product, and the minor product diisopropyl ether,
CD CHOCH(CD , were observed. The match of the observed
6 3 2
-2-ol, (CD ) -
(
3
)
2
3 2
)
fragmentations with the isotopic mass distribution from the library
spectra unambiguously demonstrates that both products have D atoms
1
over 0.2 g of molecular sieves. A H NMR spectrum of the mixture
only in the methyl groups. Notably, for the (CD
3
)
2
CHOH (m/z ) 66 )
after 8 h shows that the ratio of 2-propanol to diisopropyl ether does
not change; that is, the diisopropyl ether is not selectively adsorbed by
the added molecular sieves.
[
-
M]) major product one observes a peak at m/z ) 65 assignable to [M
+
H] ) (CD
3
)
2
CdOH and a base (100%) peak at m/z ) 48 assignable
+
to [M - CD
3
] ) 48; the m/z ) 18 peak for CD
3
was also observed).
Acknowledgment. This work was supported by the Depart-
ment of Energy, Office of Basic Energy Sciences, via DOE
Grant FG06-089ER13998, and by a Fulbright Scholar Fellow-
ship to S. O¨ .
Supporting Information Available: Kinetic derivations, treat-
ment of the composite reaction stoichiometry, experimental
details, including calculations, figures, and tables. This material
is available free of charge via the Internet at http://pubs.acs.org.
+
For the minor product, the molecular ion peak (M ) at m/ z ) 114 and
a CD (M - 18) fragment at m/z ) 96 are as expected for (CD CH-
O-CH(CD . By comparison, in the case of the minor product,
3
3 2
)
3 2
)
(
CH
3
)
2
CH-O-CH(CH
3
)
2
, obtained from the hydrogenation of undeu-
+
+
terated acetone the M and M - CH
1
comparison of the mass spectra and their fragmentation patterns with
3
fragment peaks are at m/z )
02 and m/z ) 87, respectively, also as expected. Overall, the
(
51) Aplin, R. T.; Budzikiewicz, H.; Djerassi, C. J. Am. Chem. Soc. 1965, 87,
3
180.
JA0437813
4808 J. AM. CHEM. SOC.
9
VOL. 127, NO. 13, 2005