Diisobutylaluminum hydride reductions revitalized: A fast, robust, and selective continuous flow system for aldehyde synthesis

Article abstract of DOI:10.1021/ol2031872

A continuous flow system for the multiparameter (flow rate, temperature, residence time, stoichiometry) optimization of the DIBALH reduction of esters to aldehydes is described. Incorporating an in-line quench (MeOH), these transformations are generally complete in fewer than 60 s. Mixing of the DIBALH and ester solutions was observed to be an exceptionally critical parameter for optimum results. This system thus provides general guidelines based on the structure of the ester for selective reduction of an ester without overreduction.

Full text of DOI:10.1021/ol2031872

ORGANIC
LETTERS
2012
Vol. 14, No. 2
568–571
Diisobutylaluminum Hydride Reductions
Revitalized: A Fast, Robust, and Selective
Continuous Flow System for Aldehyde
Synthesis
Damien Webb and Timothy F. Jamison*
Department of Chemistry, Massachusetts Institute of Technology, Cambridge,
Massachusetts 02139, United States
tfj@mit.edu
Received November 29, 2011
ABSTRACT
A continuous flow system for the multiparameter (flow rate, temperature, residence time, stoichiometry) optimization of the DIBALH reduction of
esters to aldehydes is described. Incorporating an in-line quench (MeOH), these transformations are generally complete in fewer than 60 s. Mixing
of the DIBALH and ester solutions was observed to be an exceptionally critical parameter for optimum results. This system thus provides general
guidelines based on the structure of the ester for selective reduction of an ester without overreduction.
Aldehydes play a prominent role in synthetic organic
chemistry, and reliable methods to prepare this versatile
functional group are consequently of great importance.
The partial reduction of esters using diisobutylaluminum
hydride (DIBALH) at low temperatures is particularly
appealing conceptually; it converts widely available and
inexpensive esters directly to the corresponding aldehyde,
often proceeding in high yield.1 However, the capricious-
ness often observed from experiment to experiment has
conferred such a notorious reputation upon this transfor-
mation that it is rarely used.2 Notably, only four examples
of selective DIBALH reduction of an ester to an aldehyde
are described in Organic Syntheses, and three of these
indicate that overreduction is a significant problem. Ad-
ditionally, the requirement for cryogenic temperatures and
a slow addition rate of DIBALH has deterred the use of
this method on a large scale.3 Accordingly, one often
resorts to a multistep alternative: Reduction of the ester
to a primary alcohol and oxidation to the aldehyde.
With these challenges in mind, and as part of our interest
in continuous flow synthesis,4 we envisaged that a contin-
uous flow method for the DIBALH reduction of esters
would provide an effective and reliable protocol for this
most useful of transformations (Figure 1).5 Although the
benefits of continuous flow methods and microreactor
(1) Zakharkin, L. I.; Khorlina, I. M. Tetrahedron Lett. 1962,
619–620.
(2) Burns, N. Z.; Baran, P. S.; Hoffmann, R. W. Angew. Chem., Int.
Ed. 2009, 48, 2854–2867.
(4) Bedore, M. W.; Zaborenko, N.; Jensen, K. F.; Jamison, T. F. Org.
Process Res. Dev. 2010, 14, 432–440. (b) Sniady, A.; Bedore, M. W.;
Jamison, T. F. Angew. Chem., Int. Ed. 2011, 50, 2155–2158. (c) Palde,
P. B.; Jamison, T. F. Angew. Chem., Int. Ed. 2011, 50, 3525–3528. (d)
Zhang, Y.; Jamison, T. F.; Patel, S. J.; Mainofli, N. Org. Lett. 2010, 13,
280–283. (e) Gutierrez, A. C.; Jamison, T. F. Org. Lett. 2011, 13,
6414–6417.
(5) For a previous evaluation of the performance of commercial
micromixers in a single specific example, see: (a) Ducry, L.; Roberge,
D. M. Org. Process Res. Dev. 2008, 12, 163–167. For a recent report of a
single substrate using a segmented flow approach, see: (b) Carter, C. F.;
Lange, H.; Sakai, D.; Baxendale, I. R.; Ley, S. V. Chem.;Eur. J. 2011,
17, 3398–3405.
(3) For representative examples in which the use of DIBALH was
explicitly stated to be unsuitable on large scale: (a) Liu, C.; Ng, J. S.;
Behling, J. R.; Yen, C. H.; Campbell, A. L.; Fuzail, K. S.; Yonan, E. E.;
Mehrotra, D. V. Org. Process Res. Dev. 1997, 1, 45–54. (b) Botteghi, C.;
Corrias, T.; Marchetti, M.; Paganelli, S.; Piccolo, O. Org. Process Res.
Dev. 2002, 6, 379–383. (c) Brock, S.; Hose, D. R. J.; Moseley, J. D.;
Parker, A. J; Patel, I.; Williams, A. J. Org. Process Res. Dev. 2008, 12,
496–502. (d) Bio, M. M.; Hansen, K. B.; Gipson, J. Org. Process Res.
Dev. 2008, 12, 892–895. (e) Haycock-Lewandowski, S. J.; Wilder, A.;
˚
Ahman, J. Org. Process Res. Dev. 2008, 12, 1094–1103.
r
10.1021/ol2031872
Published on Web 12/29/2011
2011 American Chemical Society

technology have been well documented,6 increasing the
knowledge base for flow-based chemical transformations
is imperative to maximize the impact of this potentially
disruptive technology.7
In this letter, we disclose the use of a simple flow
reactor;easily assembled from commercially available
parts;for a rapid, multiparameter investigation leading
to an improved protocol for the synthesis of aldehydes
from a wide variety of esters. Our results clearly demon-
strate the benefits of this approach and identify the crucial
experimental parameters required for optimum results:
flow rate (A), residence time (tR), temperature (T), and
stoichiometry.
conversion and selectivity.10 This is especially pertinent
when the transformation being investigated has a high
reaction rate and is thus likely to be influenced by the
mixing process. Therefore, our initial continuous flow
experiments were designed to investigate and understand
the effect of varying the tR as both a function of R1 volume
(constant flowrate) and asafunction of flowrate(constant
R1 volume), at a constant temperature (ꢀ78 °C). It was
quickly discovered that a key component for evaluation of
the system was the incorporation of an in-line quench (neat
methanol11) that was necessary to avoid overreduction to
the alcohol, even at ꢀ78 °C (Table 1, entry 4).
With this important modification the flow system al-
lowed for the selective DIBALH reduction of ethyl hydro-
cinnamate 1 at ꢀ78 °C (Table 1).12 Interestingly, at a
higher flow rate (at a constant R1 volume) higher conver-
sion and yield of the desired aldehyde was observed,
despite the shorter residence time (for example compare
Table 1, entries 1, 2 and 3). This observation indicates that
the reaction is very fast and that the mixing heavily
influences the outcome. As the flow rate is increased,
additional energy is provided for mixing, thus explaining
the higher conversion observed at shorter residence times.13
At very fast flow rates the outcome of the reaction was
independent of residence time (compare Table 1, entries 3,
6 and 9), indicating that mixing was very fast under these
conditions. Remarkably, even at very short residence times
(<50 ms), essentially full conversion and complete selec-
tivity was obtained (Table 1, entry 3). The extrapolated
throughput using the fastest flow rate examined is 10.4
mols (>1.8 kg) of starting material per day using the 23 μL
reactor.14
Our simply configured continuous system comprised
three precooling loops (P1, P2 and P3) and two reactors
(R1 and R2), each constructed from standard PFA
(perfluoroalkoxy alkane, 0.0300 inner diameter8) tubing
(Table 1).9 T-shaped mixers (M1 and M2, Tefzel, 0.0200
innerdiameter) wereused tocombinethe streamsthatwere
introduced by syringe pump devices, and the entire assem-
bly was submerged in a cooling bath held at the desired
reaction temperature.
To further our understanding of this transformation, we
conducted a systematic study of the interdependence of
three reaction variables: flow rate, R1 volume, and reac-
tion temperature. Due to the speed of the reaction and
flexibility of the system, this investigation required fewer
than 5 h, and the results of the 45 experiments were best
analyzed with contour plots (Figure 2).
Figure 1. Continuous DIBALH reduction of esters to aldehydes
using a continuous flow system.
As expected, the selectivity for partial reduction of 1 to
aldehyde 2 decreases with increasing reaction temperature
(Figures 2aꢀc). However, the degree of overreduction is
significantly reduced when compared to the corresponding
batch reaction at the same temperature. For example,
overreduction is negligible at ꢀ42 °C at all flow rates,
An often-overlooked aspect of flow chemistry is the
impact of the flow rate on the mixing and consequently
(6) For recent general reviews on continuous flow chemistry see: (a)
Wirth, T. Microreactors in organic synthesis and catalysis; Wiley-VCH:
Weinheim, 2008. (b) Hartman, R. L.; Jensen, K. F. Lab Chip 2009, 9, 2495.
(c) Mason, B. P.; Price, K. E.; Steinbacher, J. L.; Bogdan, A. R.;
McQuade, D. T. Chem. Rev. 2007, 107, 2300. (d) Geyer, K.; Gustafsson,
T.; Seeberger, P. H. Synlett 2009, 2382. (e) Kirschning, A.; Solodenko,
W.; Mennecke, K. Chem.;Eur. J. 2006, 12, 5972. (f) Wiles, C.; Watts, P.
Eur. J. Org. Chem. 2008, 10, 1655. (g) Ley, S. V.; Baxendale, I. R. Proc.
(11) It is essential to use pre-dried methanol to avoid clogging of the
system. Methanol was dried according to: Williams, D. B. G.; Lawton,
M. J. Org. Chem. 2010, 75, 8351–8354.
(12) As expected, the batch reduction of ethyl hydrocinnamate
(2 mmol scale) to hydrocinnamaldehyde, using one equivalent of DIBALH
in a variety of solvents, is strongly dependent on the reaction tempera-
ture and often results in some overreduction to hydrocinnamyl alcohol,
even at ꢀ78 °C. At elevated temperatures (ꢀ42 °C and above) over-
reduction is unavoidable with 3 becoming the major product of the
reaction.
€
€
Bosen Symp., Syst. Chem. 2008, 65. (h) Jahnisch, K.; Hessel, V.; Lowe,
H.; Baerns, M. Angew. Chem., Int. Ed. 2004, 43, 406.
(7) Hartman, R. L.; McMullen, J. P.; Jensen, K. F. Angew. Chem.,
Int. Ed. 2011, 50, 2–20.
(13) When a Y-shaped mixer (M1) was employed a notable decrease
in conversion was observed compared to the use of the T-shaped mixer.
It is known that Y-shaped mixers afford poorer mixing than T-shaped
mixers in these kind of flow systems (see ref 10).
(14) For a review on the large-scale use of organometallic reagents in
microreactors under flow conditions, see: ref 6a, p 211, and references
cited therein. For a recent example see: Browne, D. L.; Baumann, M.;
Harji, B. H.; Baxendale, I. R.; Ley, S. V. Org. Lett. 2011, 13, 3312–3315.
(8) Increasing the inner diameter of the R1 reactor to 0.0400 had a
negligible effect on the outcome under otherwise identical conditions.
(9) See the Supporting Information for full experimental details.
(10) For example see: Dolman, S. J.; Nyrop, J. L.; Kuethe, J. T.
J. Org. Chem. 2011, 76, 993–996. For a recent report comparing the
performance of different micromixers see: Falk, L.; Commenge, J. M.
Chem. Eng. Sci. 2009, 65, 405–411.
Org. Lett., Vol. 14, No. 2, 2012
569

Table 1. Evaluation of the Effect of Residence Time on the
Continuous DIBALH Reduction of Ethyl Hydrocinnamate 1
using a Continuous Flow Systema at ꢀ78 °C
selectivityb
entry
R1 (μL)
A1 (mL minꢀ1
)
tR (s)
2
1
3
3
1
2
3
4c
5
6
7
8
9
23
23
5
10
30
5
0.23
0.11
0.04
2.28
1.14
0.38
6.84
3.42
1.14
36
50
96
43
61
97
57
85
96
64
50
4
0
0
0
0
0
0
0
0
1
23
228
228
228
684
684
684
57
39
3
10
30
5
43
15
3
10
30
a P1, P2, P3: precooling loops; M1, M2: T-shaped mixers; R1, R2:
reactors; A = flow rate (mL minꢀ1). b Selectivity detetermined by GC.
3
c >5% 3 was formed if the in-line methanol quench was ommitted.
which is not the case in the batchwise format of this
reaction.15 This effect is most pronounced at the faster
flow rates (Figure 2c) where even at room temperature the
yield of aldehyde is 80% with less than 10% overreduction
to alcohol 3. In contrast, at room temperature in batch,
only 14% of aldehyde 2 was obtained, along with 43%
overreduction to alcohol 3.
The increased selectivity at higher flow rates is most
likely a result of improved mixing and more rapid quench-
ing of the organoaluminum intermediate, thereby prevent-
ing aldehyde release and overreduction. At ꢀ20 °C and
higher, the experimental outcome is largely independent of
residence time, indicating that the reaction is very fast at
these temperatures and that higher flow rates are neces-
sary for selectivity.5a Our results are a further indication
that mixing devices that approach ideal mixing16 could
improve selective transformations in synthetic organic
chemistry.
Figure 2. Contour plots showing the effect of reaction temperature
(T) and residence time (tR) on the amount of aldehyde 2at different
flow rates: (a) A1 =5mL minꢀ1, (b) A1 =10mL minꢀ1, (c) A1 =
3
3
30 mL minꢀ1, using the continuous flow apparatus depicted in
3
Table 1. Numbers in parentheses are the amount of alcohol 3.
(15) In batch at ꢀ42 °C, 7% of alcohol 3 was formed.
(16) For a review see: Yoshida, J. Flash Chemistry. Fast Organic
Synthesis in Microsystems; Wiley-Blackwell: New York, 2008; Chapter 6,
pp 69ꢀ104 and references cited within.
To the best of our knowledge this is the first systematic
investigation of these reaction parameters for the venerable
570
Org. Lett., Vol. 14, No. 2, 2012

cases. Additionally, a number of useful observations were
made during these experiments. Both ethyl- and methyl
hydrocinnamate 1 and 4a could be reduced selectively to
hydrocinnamaldehyde 2. However, the methyl ester was
more prone to overreduction. Methyl cyclohexanecarbox-
ylate 4c proved more prone to overreduction than 4a
and required a reaction temperature of ꢀ78 °C to enable
high selectivity. The synthetically useful lactate derivative
4d could be reduced selectively to the desired aldehyde,
even at elevated temperatures (up to ꢀ20 °C) and when
using an excess of DIBALH. Previously reported yields
vary greatly for this transformation.
Scheme 1. Representative Conditions for the Selective Reduc-
tion of Various Commonly Used Estersa
Similarly, propanoate derivative 4e could be trans-
formed to the desired aldehyde although this substrate
required the reaction temperature to be ꢀ78 °C, suggesting
that β-OꢀAl chelation is more easily ruptured than the
R-OꢀAl chelation involved for substrate 4d. Interesti-
ngly, using a slower flow rate for this substrate
(A1 = 0.1 mL minꢀ1) resulted in significant unwanted
3
overreduction of 4e(>5%) atincompleteconversion, even
at this cryogenic temperature, further emphasizing the
importance of understanding the mixing requirements of
this transformation.5b Finally, given the lack of operator
bias in executing these flow reactions, it seems this experi-
mental approach is ideal for delineating which substrates
are unsuitable for this transformation. For example, over-
reduction of ethyl benzoate could not be avoided under a
variety of conditions, indicating that simple aromatic
esters are unlikely candidates for a selective reduction until
a more efficient mixing device is developed.16
In conclusion, a simple continuous flow system has been
developed for the continuous DIBALH reduction of es-
ters. The flow format has facilitated the rapid optimization
of multiple reaction parameters and has been successfully
applied to yield a selective and reproducible synthesis of
aldehydes from esters. Additionally, the data generated in
our study should be of considerable use to those who wish
to conduct DIBALH reductions on larger scales and/or
wish to incorporate this transformation into a multistep
sequence.17 We are currently engaged in the development
one-flow multistep processes wherein the aldehyde is not
isolated or purified but undergoes a subsequent reaction in
a concatenated flow reactor.18
a Yield of aldehyde and alcohol determined by GC using an internal
standard.9 The yield of alcohol is 0% in all cases except 1 (1%), 4a (2%)
and 4b (2%).
DIBALH reduction and further demonstrates that flow
systems are ideally suited for such studies.
We also performed similar optimization studies (tempera-
ture, flow rate, residence time, stoichiometry) for a number
of substrates that are typical of the most commonly used
esters, and representative results and conditions are outlined
in Scheme 1.9 These reductions could be rapidly optimized
to afford >95% GC yield of the desired aldehyde 5 in all
Acknowledgment. We thank Novartis for its generous
financial support of the Novartis-MIT Center for Contin-
uous Manufacturing (CCM) and the members of the
CCM, particularly Kevin Nagy and Dr. Adam Sniady,
for stimulating and insightful discussions.
(17) For representative one-pot multistep transformations employ-
ing DIBALH see: (a) Takacs, J. M.; Helle, M. A.; Seely, F. L. Tetra-
hedron Lett. 1986, 27, 1257–1260. (b) Burke, S. D.; Deaton, D. N.; Olsen,
R. J.; Armistead, D. M.; Blough, B. E. Tetrahedron Lett. 1987, 28,
3905–3906. (c) Polt, R.; Peterson, M. A.; DeYoung, L. J. Org. Chem.
1992, 57, 5469–5480. (d) Dickson, H. D.; Smith, S. C.; Hinkle, K. W.
Tetrahedron Lett. 2004, 45, 5597–5599. (e) Sasaki, M.; Yudin, A. K.
Synlett 2004, 2443–2444. (f) Hoye, T. R.; Kopel, L. C.; Ryba, T. D.
Synthesis 2006, 1572–1574.
Supporting Information Available. Experimental pro-
cedures and analytical data. This material is available free
of charge via the Internet at http://pubs.acs.org.
(18) For a review see: Webb, D.; Jamison, T. F. Chem. Sci. 2010, 1,
675–680.
Org. Lett., Vol. 14, No. 2, 2012
571