Simple and efficient catalytic reaction for the selective deuteration of alcohols

Article abstract of DOI:10.1021/cs400092p

A highly efficient system for the catalytic deuteration of α and β CH bonds of primary and secondary alcohols has been developed. The deuterium source is D₂O. Together with the low catalyst loadings and the simple experimental setup, the reaction has direct application to the synthesis of bioactive isotopologues and the direct synthesis of fully deuterated substrates, such as ethanol-d₆. The current system represents a significant advance in practicality for homogeneous metal catalyzed systems that carry out H/D exchange in organic substrates with water.

Full text of DOI:10.1021/cs400092p

Letter
pubs.acs.org/acscatalysis
Simple and Efficient Catalytic Reaction for the Selective Deuteration
of Alcohols
Eugene Khaskin and David Milstein*
Department of Organic Chemistry, Weizmann Institute of Science, Rehovot, Israel.
S
* Supporting Information
ABSTRACT: A highly efficient system for the catalytic
deuteration of α and β CH bonds of primary and secondary
alcohols has been developed. The deuterium source is D2O.
Together with the low catalyst loadings and the simple
experimental setup, the reaction has direct application to the
synthesis of bioactive isotopologues and the direct synthesis of
fully deuterated substrates, such as ethanol-d6. The current
system represents a significant advance in practicality for
homogeneous metal catalyzed systems that carry out H/D
exchange in organic substrates with water.
KEYWORDS: H/D exchange, Ru pincer catalyst, alcohols, D2O, selective deuteration, homogeneous catalysis
elective deuteration of alcohols is of importance for a range
out under hydrogen atmosphere. Baratta reported an effective
Sof uses, such as the preparation of deuterated pharmaceut- homogeneous system that utilizes osmium or ruthenium pincer
catalysts7a at only 1% catalyst loadings, but requires deuterated
icals and other biologically active compounds that have
isopropyl alcohol as solvent. Takahashi and Matsubara reported
a homogeneous Ru-based system that under microwave
irradiation gave good and selective deuterium incorporation
at 5% catalyst loadings for alcohols.7d A molybdenum-based
system shows poor activity for nonbenzylic substrates, and the
benzylic ones require large amounts (5% and more) of
catalyst.7e,f Yamada8 reported a cobalt-catalyzed (1 mol %
catalyst loading) reduction of aldehydes with NaBD4 that leads
to entantioselective deuteration at the α carbon position of the
product alcohol. Using a homogeneous Ru system (3 mol %
catalyst), deuteration exclusively at the β position of the
substrate was recently reported.9 All these reports concern
monofunctional substrates, except for a recent report by Sajiki
on the stereoselective α-deuteration of sugars7b using a
heterogeneous Ru-based system (5 mol % loading) under
hydrogen atmosphere.7c
properties similar to their nondeuterated analogues, but
would differ in rates of metabolism.1 The alcohol functionality
reacts with a broad range of enzymes, including dehydrogenase
enzymes in the liver, producing aldehydes with often substantial
kinetic isotope effects. Although the greatest use of selectively
deuterated alcohols might be in biochemical enzymatic rate
studies,1a uses in new materials may also become important.1a
A quick survey of the literature shows that even in recent
articles,2 authors use the reaction of esters or aldehydes with
alkali metal deuteride salts to obtain the desired labeled alcohol.
The prices of α-deuterated alcohols, relatively simple substrates
such as benzyl alcohol-d3, and even deuterated ethanol, can
become prohibitive on large scales, but deuterated water, the
original source of deuterium in all deuterated products,
including metal deuteride reactants, remains inexpensive.3
Although catalytic deuteration of simple hydrocarbons by
exchange with D2O is an established reaction,1 direct and
selective catalytic deuteration of alcohols is more rare. Koch
and Stuart reported on stereoselective deuteration of
carbohydrate derivatives with large amounts of hot Raney
nickel (>100 mol %), with some substrates being isomerized or
remaining unreacted under the reaction conditions.4 Cioffi
reported a milder method of deuteration with Raney nickel by
utilizing microwave and sonication methods.5,6 However,
because of the difficulty of preparing deuterated Raney nickel
and the necessity of using >100 mol % metal with respect to the
substrate, the method still suffers from serious drawbacks.
There are only a few reports on metal-catalyzed selective
deuteration of alcohols.7 The interesting heterogeneous system
described by Sajiki7c uses high catalyst loadings, especially for
primary alcohols (20 wt % Ru) and the reaction must be carried
We have previously reported on Ru-based pincer systems
that catalyze transformation of alcohols and amines to esters,
amides, imines, and other substrates in open, organic solvent
10
systems while releasing H2 gas
or perform the reverse
reaction in a closed vessel under mild hydrogen pressure.11
Here, we report on a selective deuteration system that utilizes
D2O as a deuterium source and an air-stable Ru precatalyst 1
(Figure 1) at low loadings (typically 0.2 mol %, but activity can
be seen at 0.01 mol % loadings) in the presence of ∼20 mol %
NaOH. The experimental setup is a simple closed system that is
heated to 50−120 °C, depending on the substrate, under
Received: February 3, 2013
Revised: February 10, 2013
© XXXX American Chemical Society
448
dx.doi.org/10.1021/cs400092p | ACS Catal. 2013, 3, 448−452

ACS Catalysis
Letter
a
Table 1. Optimization with 1-Hexanol
deuteration (% yield) α; β
entry
conditions
(max % yield)
1
2
3
1 mol % 2, 120 °C, 24 h
10 mol % 2, 120 °C, 24 h
0.2 mol % 1, 110 mol % NaOH,
120 °C, 24 h
0; 0 (96.5)
80; 0 (96.5)
b
94; 20 (94.1)
Figure 1. Ru-pincer complexes relevant to the current study.
4
5
6
7
8
0.2 mol % 1, 10 mol % NaOH,
120 °C, 24 h
1 mol % 1, 300 mol % NaOH,
120 °C, 24 h
no catalyst, 110 mol % NaOH,
120 °C, 24 h
0.2 mol % 1, 20 mol % NaOH,
120 °C, 24 h
as in entry 7 under air, 0.1 mol % 1
78; 0 (96.5)
nitrogen atmosphere. We find, however, that air can be used
with no detrimental effects.
b
94; 14 (94.1)
The crux of the discovery lies in the fact that unlike earlier
systems reporting metal-catalyzed H/D exchange, our system is
very practical. It can be used without large amounts of air-
sensitive or expensive metal catalyst and can be envisaged as
being adoptable for large scale synthesis of industrially relevant
substrates. Upon complete deuteration, most alcohols can be
easily separated from the reaction medium and the catalyst by
neutralizing the reaction mixture and extracting the relevant
alcohol in an organic solvent, such as diethyl ether. Smaller
alcohols that form azeotropes with water, such as ethanol, need
to be separated from water and dried using well established
industrial and household methods.
no deuteration
92; 10 (94.1)
b
90; 10 (94.1)
a
Yields obtained by NMR via integration of unreactive CH3 groups or
dioxane internal standard. In parentheses is the theoretical maximum/
equilibrium deuteration possible with 0.4 mL of D2O. Trace (∼1%)
of sodium hexanoate was also formed.
b
showed that it is possible to get significant deuteration at the β
position, albeit with a side reaction producing ∼1% sodium
hexanoate. In an open system, with evolution of hydrogen gas,
sodium hexanoate would be the major product as per our
recent report (Scheme 1).12 It is easy to separate out this
byproduct by performing a simple extraction in an organic
solvent, such as diethyl ether, of the alcohol substrate, with the
acid salt remaining in the aqueous layer. However, in most
cases when the base loading is less than 100 mol %, a very small
trace of acid salt is obtained, or it is not detected.
Significantly, similar conditions were utilized in our very
recently reported catalytic transformation of primary alcohols
to carboxylic acid salts, utilizing water as an oxygen source, with
concomitant evolution of hydrogen gas (Scheme 1).12 In the
Scheme 1. Catalytic Transformation of Alcohols to
Carboxylic Acids
For most applications, the presence of deuterium in the β
position is not deleterious, and thus, we decided to utilize the
optimal 1-hexanol conditions (Table 1; entry 7) for most of the
primary alcohols. These conditions led to the full deuteration of
ethanol in both the α and β positions, which is particularly
interesting if the desired product is ethanol-d6 (Table 2; entry
3). A number of functional groups proved to be tolerant to the
deuteration conditions. For example, the amine substrate
(Table 2; entry 7) showed no deuteration at the carbons
next to the amine moiety, as checked by deuterium NMR,
although the overall deuteration at the alcohol was limited to
only ∼35%, meaning that for the amine entry, the catalytic
efficiency was compromised.
Secondary alcohols proved much more active (Table 3). At
room temperature, some deuteration at the α position was
already in evidence after 3 h; however, most of the precatalyst
remains insoluble and, thus, unactivated at room temperature.
Heating to 50 °C overnight allowed for full catalyst activation
and subsequent full deuteration of both the α and β positions.
Thiols and very electron-poor alcohols, such as CF3CH2OH,
showed almost no activity in deuteration, probably because of a
stronger Ru−O or Ru−S bond that stabilizes an intermediate
species. In contrast, electron-rich alcohols, including electron-
rich benzyl alcohol and 1,3-bis(hydroxymethyl)benzene, gave
full deuteration after just 18 h of heating under the standard
primary alcohol conditions.
current work, the reaction is performed in a closed system,
preventing dihydrogen removal from the reaction vessel, and
often requires less than stoichiometric amounts of base and
lower catalyst loadings, especially for secondary alcohols.
In addition to low catalyst loadings and relative simplicity of
use, our system is tolerant of a number of functional groups
that do not react with aqueous warm base, including amines,
unactivated double bonds and CF bonds.13 Impurities are also
well tolerated, as all substrates tested were used “as is” from the
manufacturer’s bottle without further purification. In a typical
reaction, 0.2% of the catalyst and 20% NaOH was put into a 2.5
mL NMR Young tube, to which the substrate and 0.4 mL of
D2O were added (note: higher loadings of D2O lead to higher
deuteration percentages). The reaction conditions were
optimized for the substrate 1-hexanol, for which the optimal
conditions led to full (highest percentage possible) deuteration
at the α carbon and small amounts of side deuteration at the β
position (Table 1). Temperature and amount of base play a
significant role in the selectivity, with the optimal conditions
differing depending on the substrate. For 1-hexanol, 10%
NaOH loadings gave the best selectivity, but reactivity was
slightly compromised. Really high base loadings (≥110%)
Base is essential for the reaction to proceed under low
catalyst loading, although it was found that the activated
complex 3, prepared by the reaction of dearomatized 2 with
D2O,14 could catalyze the deuteration of 1-hexanol at 10 mol %
449
dx.doi.org/10.1021/cs400092p | ACS Catal. 2013, 3, 448−452

ACS Catalysis
Letter
a
used, the catalyst loading can be substantially reduced when
working on larger scales without sacrificing reactivity. Thus, a
reaction with 0.01 mol % of 1 carried out on a large amount of
cycloheptanol with 2 mL of D2O, both of which enabled an
amount of solid catalyst to be weighed out that equaled 0.01
mol %, gave full deuteration at both the α and β positions after
36 h of heating at 120 °C.
Table 2. Deuteration of Primary Alcohols
By utilizing a number of different substrates, we gained
insight into the possible catalytic mechanism. When carrying
out the catalytic reaction of hexanol in a vessel with a large
amount of headspace volume (i.e., “open system”) with H2O, a
large sample of the gaseous phase showed the presence of
hydrogen in GC/TCD. However, the open system led to large
amounts of carboxylic acid salt being observed, probably
because of the inability of hydrogen to recombine with
intermediate aldehyde before its capture by OH− and further
dehydrogenation.
Although no trace of aldehyde was detected with 1-hexanol,
systems that produce relatively stable conjugated aldehydes that
have time to react via a different pathway proved the
intermediacy of aldehydes. Thus, crotyl alcohol decomposed
into the retro-aldol condensation product acetaldehyde under
the reaction conditions, which was detected by NMR.
Cinnamic alcohol underwent 50% decomposition to give
cinnamic aldehyde and 3-phenyl-propanol as major products
(Scheme 2) and the minor products benzaldehyde and benzyl
alcohol.15 When D2O was utilized, the remaining non-
decomposed cinnamic alcohol was found to be fully deuterated
at the α position.
a
See Table 1, footnote a.
a
Table 3. Deuteration of Secondary Alcohols
Scheme 2. Decomposition of Cinnamic Alcohol under
Catalytic Reaction Conditions
The major products of decomposition suggest that the
mechanism proceeds through dehydrogenation to an aldehyde
and H2, with the latter reacting via the complex with cinnamic
alcohol to form 3-phenylpropanol. In contrast, when cinnamic
aldehyde was used as a starting material under the standard
reaction conditions, benzaldehyde was formed as the only
major product with acetaldehyde detected by NMR and no
further minor products.
A possible catalytic cycle for the deuteration reaction is
shown in Scheme 3, and a mechanistic scheme for deuterium
scrambling into all positions of the complex is provided in
Scheme 4, where an intermediate Ru(0) complex may be
responsible for deuterium scrambling into the metal hydride
position.
a
See Table 1, footnote a.
loading in the absence of added base. Introducing 20 mol % of
base to 0.2 mol % of 3 allowed deuteration to take place as well,
although slightly less efficiently than by utilizing precatalyst 1.
We believe that although 3 and no other complex is always
visible in 1HNMR throughout the reaction, no matter if 1 or 3
is used as the initial catalytic species, the difference in reaction
outcome is due to the fact that water-insoluble 1 is activated
slowly over time with NaOH, thus creating a steady amount of
active catalyst. However, we do not have evidence that 3
decomposes to any other inactive species.
1H NMR of the ethanol deuteration taken during the
reaction clearly indicated the presence of geminal coupled H
peaks (JHD ∼ 2.5) in both the α and β positions. Thus, the
substrate was sequentially deuterated at each position and
showed the presence of CHD groups. However, 4-amino-
butanol, which did not reach the theoretical deuteration
maximum, displaying only 35% of deuteration in both the α
and β positions after 40 h at 120 °C even though 3 was still
present in solution, showed only the presence of either CH2 or
Although 0.2 mol % catalyst was used for convenience,
because it is difficult to weigh out very small amounts of the
catalyst and it is insoluble in water so a stock solution cannot be
2
CD2 groups, but never CHD. D NMR showed the presesnce
450
dx.doi.org/10.1021/cs400092p | ACS Catal. 2013, 3, 448−452

ACS Catalysis
Letter
different mechanism, but that would be difficult to implement
practically, despite its heterogeneous nature, because of high Ru
loading and a requirement for a hydrogen atmosphere.
Stereochemical information is also lost in the α position in β-
selective deuteration in the system developed by Jia and co-
workers.9
Scheme 3. Possible Catalytic Cycle
In conclusion, we have developed a catalytic deuteration
reaction of primary and secondary alcohols under simple
conditions with low catalyst loadings and utilizing D2O as the
deuterium source. The reactions can be run under air. We
believe this reaction will be of significant utility because it
represents a simple way to produce substrates deuterated at the
α CH2 position, useful for kinetic studies. For a particular
substrate that is desired in large amounts, it should be
optimized with regard to the amount of base and reaction time
because these were found to vary in optimality depending on
the substrate. Simple substrates that can be further used as
NMR solvents, such as ethanol and isopropyl alcohol, are also
efficiently deuterated. We believe that complex 1 is a highly
practical and a highly active green catalyst for H/D exchange
and alcohol to carboxylic acid salt transformation catalysis.12
Scheme 4. Pathway for Deuterium Scrambling into Complex
2
ASSOCIATED CONTENT
* Supporting Information
■
S
General experimental procedures, synthesis details of 3, sample
relevant NMR spectra. This material is available free of charge
via the Internet at http://pubs.acs.org. A typical catalytic
procedure is available in ref 16.
AUTHOR INFORMATION
Corresponding Author
*E-mail: david.milstein@weizmann.ac.il.
■
Notes
The authors declare no competing financial interest.
of only two aliphatic peaks corresponding to the α and β CH2
groups.
ACKNOWLEDGMENTS
■
The cycle in Scheme 3 can help explain the preference for
either CHD intermediates or for the direct CD2 formation. The
intermediacy of a coordinated aldehyde and its lifetime will
directly affect the presence of either one species or another. A
short-lived aldehyde-bound complex that rapidly undergoes β
hydride insertion to form the parent alcohol and is then quickly
released will give one deuterium at the α position. Depending
on whether this alcohol is quickly released or reverts back to
the bound aldehyde, CHD can be observed as an intermediate
outcome, or CD2 is directly formed after just one catalytic cycle
in the α position. The exact result will depend on the equilibria
between the species.
A sterically hindered or otherwise long-lived aldehyde may
also dissociate and undergo H/D exchange at the β positions
via keto/enol tautomerism. An example is 4-aminobutanol,
which shows only the presence of CD2 at the β carbon, or more
sterically demanding secondary alcohols, which are deuterated
nonselectively. Decoordination of the aldehyde intermediate
may also be promoted by a higher concentration of NaOH
because the selectivity in deuteration of 1-hexanol between the
α and β positions is diminished upon reaction with ≥110 mol
% of base.
This research was supported by the European Research Council
under the FP7 framework (ERC No 246837), by the Kimmel
Center for Molecular Design, and by the Bernice and Peter
Cohn Catalysis Research Fund. D.M. is the Israel Matz
Professorial Chair of Organic Chemistry.
REFERENCES
■
(1) For a review, see: (a) Atzrodt, J.; Derdau, V.; Fey, T.;
Zimmerman, J. Angew. Chem., Int. Ed. 2007, 46, 7744−65. For a
recent example using Ru, see: (b) Prechtl, M. H. G.; Hoelscher, M.;
Ben-David, Y.; Theyssen, N.; Milstein, D.; Leitner, W. Eur. J. Inorg.
Chem. 2008, 22, 3493−500. (c) For a recent example with 5% Ir
catalyst loading, see: Yung, C. M.; Skaddan, M. B.; Bergman, R. G. J.
Am. Chem. Soc. 2004, 126, 13033−43.
(2) Wang, Q.; Sheng, X.; Horner, J. H.; Newcomb, M. J. Am. Chem.
Soc. 2009, 131, 10629−36.
(3) http://www.isotope.com/cil/index.cfm (accessed February 3,
2013).
(4) (a) Koch, H. J.; Stuart, R. S. Carbohydr. Res. 1977, 59 (1), C1−
C6. (b) Koch, H. J.; Stuart, R. S. Carbohydr. Res. 1978, 64, 127−34.
(c) Koch, H. J.; Stuart, R. S. Carbohydr. Res. 1978, 67, 341−8.
(5) (a) Cioffi, E. A.; Prestegard, J. H. Tetrahedron Lett. 1986, 27,
415−18. (b) Cioffi, E. A.; Alston, K. E.; Patel, A. M. Tetrahedron Lett.
2002, 43, 8985−8987. (c) Cioffi, E. A.; Bell, R. H.; Le, B. Tetrahedron
Assymetry 2005, 16, 471.
(6) For reviews on Raney-nickel catalyzed deuteration, see:
(a) Bokatzian-Johnson, S. S.; Maier, M. L.; Bell, R. H.; Alston, K.
E.; Le, B. Y.; Cioffi, E. A. J. Labelled Compd. Radiopharm. 2007, 50,
These results mean that in all cases, deuteration proceeds via
a temporary transformation of the alcohol to a carbonyl species,
and stereochemical information would thus be lost. Therefore,
the current method is not appropriate for chiral alcohols, unlike
the Sajiki mothod,7b which proceeds by a fundamentally
451
dx.doi.org/10.1021/cs400092p | ACS Catal. 2013, 3, 448−452

ACS Catalysis
Letter
380−3. (b) Heys, J. R. J. Labelled Compd. Radiopharm. 2010, 53, 716−
21.
(7) (a) Bossi, G.; Putignano, E.; Rigo, P.; Baratta, W. Dalton Trans.
2011, 40, 8986−8995. (b) Fujiwara, Y.; Iwata, H.; Sawama, Y.;
Monguchi, Y.; Sajiki, H. Chem. Commun. 2010, 46, 4977−9.
(c) Maegawa, T.; Fujiwara, Y.; Inagaki, Y.; Monguchi, Y.; Sajiki, H.
Adv. Synth. Catal. 2008, 350, 2215−8. (d) Takahashi, M.; Oshima, K.;
Matsubara, S. Chem. Lett. 2005, 34, 192−3. (e) Balzarek, C.; Weakley,
T. J. R.; Tyler, D. R. J. Am. Chem. Soc. 2000, 122, 9527−34. (f) Breno,
K. L.; Tyler, D. R. Organometallics 2001, 20, 3864−8.
(8) Miyazaki, D.; Nomura, K.; Ichihara, H.; Ohtsuka, Y.; Ikeno, T.;
Yamada, T. New J. Chem. 2003, 27, 1164−6.
(9) Tse, S. K. S.; Xue, P.; Lau, C. W. S.; Sung, H. H. Y.; Williams, I.
D.; Jia, G. Chem.Eur. J. 2011, 17, 13918−25.
(10) (a) Gnanaprakasam, B.; Milstein, D. J. Am. Chem. Soc. 2011,
133, 1682−5. (b) Gnanaprakasam, B.; Ben-David, Y.; Milstein, D. Adv.
Synth. Catal. 2010, 352, 3169−73. (c) Gunanathan, C.; Milstein, D.
Angew. Chem., Int. Ed. 2008, 47, 8661−4. (d) Gunanathan, C.; Ben-
David, Y.; Milstein, D. Science 2007, 317, 790−2. (e) Zhang, J.;
Gandelman, M.; Shimon, L. J. W.; Milstein, D. Dalton Trans. 2007, 1,
107−13. (f) Zhang, J.; Leitus, G.; Ben-David, Y.; Milstein, D. Angew.
Chem., Int. Ed. 2006, 45, 1113−5. (g) Gunanathan, C.; Milstein, D.
Acc. Chem. Res. 2011, 44, 588−602.
(11) (a) Balaraman, E.; Fogler, E.; Milstein, D. Chem. Commun. 2012,
48, 1111−3. (b) Balaraman, E.; Ben-David, Y.; Milstein, D. Angew.
Chem., Int. Ed. 2011, 50, 11702−5. (c) Balaraman, E.; Gnanaprakasam,
B.; Shimon, L. J. W.; Milstein, D. J. Am. Chem. Soc. 2010, 132, 16756−
8. (d) Balaraman, E.; Gunanathan, G.; Zhang, J.; Shimon, L. J. W.;
Milstein, D. Nat. Chem. 2011, 3, 609−14.
(12) Balaraman, E.; Khaskin, E.; Leitus, G.; Milstein, D. Nat. Chem.
2013, 5, 122−5.
(13) Very recently, selective deuteration of tertiary amines catalyzed
by the Shvo Ru catalyst (using a catalyst loading of 5% for the
optimized reaction conditions) has been reported: Norbert, L.;
Michalik, D.; Baehn, S.; Imm, S.; Neumann, H.; Atzrodt, J.; Derdau,
V.; Holla, W.; Beller, M. J. Am. Chem. Soc. 2012, 134, 12239−44.
(14) Characterization of complex 3 is available in the Supporting
Information. For another PNN−Ru(H)(OH) complex implicated in
water splitting, see: Kohl, S. W.; Weiner, L.; Schwartsburd, L.;
Konstantinovski, L.; Shimon, L. J. W.; Ben-David, Y.; Iron, M. A.;
Milstein, D. Science 2009, 324, 74−7.
(15) Some other minor products formed during the course of the
reaction included hydrocarbons that are identified only by their GC/
MS spectra, contain a C3 tail, and are as of yet unexplained, since they
cannot result from decarbonylation reactions; they may be of interest
as an example of CO cleavage in organic substrates. However, they
represent <2% of total products, and we cannot adequately explain the
mechanism by which they form, except to note their presence. It is
worthwhile to further explore the reactivity of complex 3 in this
direction at a later time.
(16) Typical catalytic procedure: In a typical experiment, 0.5 mg of
complex 1 (0.001 mmol) and 100 equiv of NaOH (4.2 mg; 0.1 mmol)
were added to a J. Young NMR tube; 500 equiv of substrate alcohol
(0.5 mmol) was subsequently added as well as 0.4 mL of D2O. The
Young tube was closed, leaving ∼2.5−3 mL of free headspace volume.
The tube was then immersed in an oil bath preheated to 120 °C and
left for 18 h. Before and after, 1H NMR spectra were used to compare
the level of deuteration if nondeuterable CH2 or CH3 groups were
present in the molecule. For substrates such as ethanol, 1,4 dioxane (5
μL) was added as an internal standard, and the intensity of the peaks
was compared with the standard.
452
dx.doi.org/10.1021/cs400092p | ACS Catal. 2013, 3, 448−452