Effects of co-solvents on reactions run under micellar catalysis conditions

Article abstract of DOI:10.1021/acs.orglett.6b03468

The impact of varying percentages of an organic solvent added to reactions run in aqueous nanomicelles as the reaction medium has been investigated. Issues such as rates of reaction, percent conversion, and yield, as well as various practical aspects (e.g

Full text of DOI:10.1021/acs.orglett.6b03468

Letter
pubs.acs.org/OrgLett
Effects of Co-solvents on Reactions Run under Micellar Catalysis
Conditions
Christopher M. Gabriel,† Nicholas R. Lee,† Florence Bigorne,§ Piyatida Klumphu,† Michael Parmentier,‡
Fabrice Gallou,‡ and Bruce H. Lipshutz*,†
†Department of Chemistry and Biochemistry, University of California, Santa Barbara, California 93106, United States
§Chimie Paris Tech, Paris, France
‡Novartis Pharma AG, Basel, Switzerland
S
* Supporting Information
ABSTRACT: The impact of varying percentages of an organic
solvent added to reactions run in aqueous nanomicelles as the
reaction medium has been investigated. Issues such as rates of
reaction, percent conversion, and yield, as well as various
practical aspects (e.g., effect on stirring, etc.), are discussed,
leading to an operationally simple method for the general
improvement of potentially problematic systems across a broad
range of reaction types, in particular for reactions run at scale.
he expanding list of applications of designer surfactants
such as TPGS-750-M (1, Figure 1) and their in situ
T
derived nanomicelles as an alternative reaction medium to
organic solvents in which numerous synthetic transformations
can be carried out is now quite extensive.1 While several types
of transition metal-catalyzed processes are especially well-suited
to these conditions (e.g., Pd- and Ni-catalyzed cross-
couplings,1b,2 Au-catalyzed processes,3 etc.), many other
reaction types not only are amenable but also offer unique
opportunities to benefit from the presence of the surrounding
water (e.g., olefin metathesis in the presence of CuI4 or at
reduced pH5), along with elimination of egregious dipolar
aprotic and water-miscible organic solvents (e.g., in SNAr
reactions6). The broad generality of this reaction medium
allows for tandem processes to be developed, where sequences
of reactions that no longer require differing reaction solvents
can be run in a single pot, maximizing efficiency and
minimizing waste.6−13 While the notion of “getting organic
solvents out of organic reactions” remains our goal, we have
developed an appreciation for the vagaries associated with the
solubility properties of starting materials, most notably highly
crystalline solids that can be quite slow to gain entry into the
hydrophobic micellar core, making use of this technology
impractical. Flocculent solids can be tough to fully consume,
and in some cases, educts viewed as “brick dust” that are only
soluble in highly polar media may be outside the scope of this
green chemistry.
Figure 1. Structure of TPGS-750-M.
the workup stage. For example, materials at any stage may
adhere to the stirrer and form an intractable gum, preventing
continued stirring and/or full conversion. Extraction of the
product from such intractable solids can be time-consuming
and may require extensive handling, including solubilization in
copious amounts of undesirable organic solvents.
Fortunately, a straightforward modification to existing,
otherwise optimized conditions that serves to further generalize
this aqueous technology and overcome these limitations has
been found. In this report, therefore, we describe the influential
role that selected organic co-solvents can play in facilitating
many types of reactions run in water, and under very mild
conditions.
In screening an array of organic solvents,3a,9,10,15 added in the
range of 5−20% by volume, the most effective have been found
to be THF, acetone, and PEG-200.
Each of these is water-miscible, which has implications for
workup and recycling of the aqueous reaction mixtures (vide
infra). On occasion, toluene has also been found to be quite
useful. Several types of reactions were investigated to determine
Initial solubility, however, is but one of the potential
problems that may arise, especially with reactions run at
scale.14 Starting materials, and/or the desired product(s), may
present unpredictable practical issues during or post-reaction at
Received: November 22, 2016
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.6b03468
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
co-solvent effects on kinetics, yields, and physical properties of
reaction mixtures.
Suzuki−Miyaura (SM) Cross-Couplings.1a Figure 2 shows
the results of an SM reaction involving solids 2 and 3, in 2 wt %
Figure 4. Miyaura borylations with co-solvents.
Figure 2. Effects of THF and acetone on a SM coupling.
TPGS-750-M/H2O. For this case, it was found that while the
use of a co-solvent had essentially no effect on the physical
appearance of the reaction mixture run at 45 °C, the presence
of 10% acetone (by volume) increased the isolated yield from
68% to 85%. Further increases in the acetone content (e.g., to
50%) led to a dramatic decrease in isolated product (25%).
Earlier this year our group released a communication
introducing HandaPhos, a highly active ligand for SM couplings
enabled by only ppm levels of its 1:1 complex with palladium.7
Examination of THF, acetone, and toluene as co-solvents at the
10% level identified both THF and acetone as effective
additives, thereby enhancing the rate of the coupling between
solid 5 and liquid 6, shown in Figure 3. In comparison to the
control system, higher isolated yields were produced using 10%
co-solvent. And while the highest yield of 7 was achieved by
utilizing THF (81%), 10% toluene showed a more uniform and
stable suspension without precipitation.
Figure 5. Representative SNAr reaction.
Amide Bond Formation.17 As a representative case,
coupling of solid diphenylacetic acid, 13, with solid L-
tryptophan methyl ester, 14, under standard (co-solvent-free)
conditions led to conversion to amide 15. However,
precipitation of the accumulated product onto the stir bar
halted stirring of the reaction mixture after only 30 min (Figure
6). This type of precipitation can grossly effect the extent of
Figure 3. Impact of co-solvent on SM cross-coupling.
Miyaura Borylation.16 C-BPin bond-forming reactions may
also benefit by the addition of a co-solvent, due mainly to the
flocculent/crystalline nature of B2Pin2. This reagent has the
tendency to float upon the surface of the aqueous reaction
mixture, which may lead to increased reaction times and
undesirable behavior at scale. Use of either 10% THF or
acetone, where the B2Pin2 could be added as a predissolved
solution to solid educt 8, completely eliminated this
phenomenon affording product 9 (Figure 4).
Figure 6. COMU-mediated amide bond formation.
conversion, rendering these conditions unsuitable for scale-up.
On the other hand, use of 10% THF leads to a stable
suspension, thus allowing the reaction to stir freely and proceed
to a higher degree of conversion. Previous studies on amide
couplings in aqueous TPGS-750-M have also been shown to
benefit from the presence of PEG as co-solvent.15
SNAr Reactions.6When performing these substitution
reactions involving equimolar amounts of solid benzimidazole
10 and liquid 2,4,5-trichloropyrimidine, 11, use of co-solvent
led to a distinct difference in reaction appearance compared
with that seen in water alone (Figure 5). Under standard SNAr
conditions, aggregation of the product was observed, and while
only a modest improvement in reaction yield was achieved by
the use of 15% THF, the mixture was transformed into a stable
suspension, an important consideration when run at scale.
C−H Activation.18 Palladium-catalyzed C−H activation of
aryl urea solid 16, followed by coupling with solid 4-
iodoanisole, 17, afforded a higher yield in the absence of a
co-solvent (Figure 7). Nonetheless, utilizing 10% THF or
acetone led to freely flowing material in the reaction mixture,
which also produced heavy precipitation of the product and
silver salt.
Reductions of Nitroaromatics.9 Aromatic/heteroaromatic
nitro compounds are often challenging substrates in general
B
DOI: 10.1021/acs.orglett.6b03468
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Figure 9. ppm (HandaPhos)Au(I)-catalyzed ether formation.
33% to 91%, further exemplifying the potential impact of co-
solvent screening in these systems.
Notwithstanding the observed benefits associated with co-
solvent usage, the question remains as to whether nanomicelles
still exist in the presence of organic solvent. Insight was
obtained using dynamic light scattering data acquired for
samples containing TPGS-750-M in the presence of 5%, 10%,
and 15% THF (Table 1). As the data illustrate, particles of
Figure 7. Pd-catalyzed C−H activation.
due to their limited solubility in aqueous nanomicelles, a result
of their highly crystalline nature and relatively high melting
points. Recently, we have disclosed a new method for efficient
Fe/ppm Pd-catalyzed/NaBH4 nitro group reductions in water
that utilizes a co-solvent to assist with highly crystalline,
sparingly soluble substrates.9 For this system, the co-solvent not
only enhances solubilization of the starting material and
intermediates within the micellar core but also shows a
pronounced effect on the amount of foaming resulting from
the evolution of H2. An excellent example of this is the
reduction of (solid) nitroaromatic 19 to amine 20, where the
absence of co-solvent increases nucleation sites for gas
evolution leading to foaming, which can grow to occupy the
entire volume of the flask (Figure 8A). When carried out with
Table 1. Average Particle Size of Micelles + Co-solvent
entry
1
co-solvent (% v/v)
ave particle size (nm)
57
none (control)
THF
5
2
3
4
104
129
101
10
15
acetone
15
5
113
initially 57 nm increase in size to ca. 100 nm, with populations
becoming more diverse when compared to the control sample
measured in water only. The expansion of micelles in the
presence of organic solvent is expected, as solvent molecules
can organize themselves within the micellar framework, causing
particle growth until saturation, at which point no additional
swelling is observed.19 These observations are in agreement
with the DLS data collected (Figure 10). Likewise, particles of
this surfactant also increased in size upon introduction of 15%
acetone (Table 1, entry 5).
Figure 8. Fe/ppm-Pd NP-catalyzed reduction.
20% THF, foaming is negligible due to the decrease in surface
tension of the solvent, leading to rapid reduction of any foam,
along with greatly improved stirring (Figure 8B). Furthermore,
solid precipitation is not observed throughout the course of the
reaction, which in the composite affords a significant increase in
yield from 84% to 95%.
Cyclizations of Allenic Alcohols. As part of a more general
study on HandaPhos technology applied to ppm level Au-
catalyzed reactions run under these micellar catalysis
conditions, cyclization of liquid allenic alcohol 21 was effected
using (HandaPhos)Au (1000 ppm, or 0.10 mol %) to produce
cyclic ether 22 in good yield (Figure 9). In the same manner as
ppm Pd-catalyzed SM couplings with HandaPhos, a low
catalyst loading results from precomplexation in organic
solvent, which is then used as a means of transferring the
catalyst to the aqueous reaction mixture. For this trans-
formation, a distinct relationship between co-solvent selection
and yield was revealed. That is, the addition of 10% toluene
(but not acetone or THF) was found to increase the yield from
Figure 10. DLS data (particle size): TPGS-750-M/H2O + THF.
As previously reported, the use of TPGS-750-M for micellar
catalysis lends itself for the opportunity to recycle the reaction
medium by simply removing organic components from the
reaction via extraction with minimal EtOAc. As Scheme 1A
highlights, the aqueous medium used for the SNAr reaction to
produce 12 can be recycled two additional times for the same
reaction without effecting yield. Upon neutralization of the
remaining aqueous medium, a nearly quantitative yield of 15
from 13 and 14 under COMU-mediated peptide coupling
C
DOI: 10.1021/acs.orglett.6b03468
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 1. Recycling and E Factor Studies
ACKNOWLEDGMENTS
Financial support provided by the NSF (GOALI SusChEM
1566212) and Novartis is warmly acknowledged with thanks.
■
REFERENCES
■
(1) (a) Lipshutz, B. H.; Ghorai, S.; Abela, A. R.; Moser, R.; Nishikata,
T.; Duplais, C.; Krasovskiy, A. J. Org. Chem. 2011, 76, 4379.
(b) Lipshutz, B. H.; Gallou, F.; Handa, S. ACS Sustainable Chem. Eng.
2016, 4, 5838. (c) For a general overview on the use of surfactants in
synthesis, see: La Sorella, G.; Strukul, G.; Scarso, A. Green Chem. 2015,
17, 644.
(2) (a) Lipshutz, B. H.; Ghorai, S. Aldrichimica Acta 2012, 45, 3.
(b) Lipshutz, B. H.; Ghorai, S.; Leong, W. W. Y.; Moser, R.; Taft, B. R.
J. Org. Chem. 2011, 76, 5061.
(3) (a) Handa, S.; Lippincott, D. J.; Aue, D. H.; Lipshutz, B. H.
Angew. Chem., Int. Ed. 2014, 53, 10658. (b) Minkler, S. R. K.; Isley, N.
A.; Lippincott, D. J.; Krause, N.; Lipshutz, B. H. Org. Lett. 2014, 16,
724. (c) Minkler, S. R. K.; Lipshutz, B. H.; Krause, N. Angew. Chem.,
Int. Ed. 2011, 50, 7820.
(4) Voigtritter, K.; Ghorai, S.; Lipshutz, B. H. J. Org. Chem. 2011, 76,
4697.
(5) Lipshutz, B. H.; Ghorai, S.; Leong, W. W. Y.; Moser, R.; Taft, B.
R. J. Org. Chem. 2011, 76, 5061.
(6) Isley, N. A.; Linstadt, R. T. H.; Kelly, S. M.; Gallou, F.; Lipshutz,
B. H. Org. Lett. 2015, 17, 4734.
(7) Handa, S.; Andersson, M. P.; Gallou, F.; Reilly, J.; Lipshutz, B. H.
Angew. Chem., Int. Ed. 2016, 55, 4914.
(8) Isley, N. A.; Dobarco, S.; Lipshutz, B. H. Green Chem. 2014, 16,
conditions was obtained. And although this modification
requires the addition of organic solvent to the reaction mixture,
the amount of organic waste generated is still quite low when
calculating Sheldon’s E Factor.13,20 By means of a simple
filtration, 12 was isolated in 80% yield resulting in an
exceptionally low E Factor of 1.25. The low E Factor of 6.8
was achieved by extraction with minimal MTBE to afford 4.
In summary, use of a co-solvent as an additive to aqueous
solutions containing nanomicelles derived from TPGS-750-M
adds yet another dimension to micellar catalysis as an attractive
alternative approach to traditional organic solvents alone as
reaction media. The preferred co-solvents are THF, acetone,
and PEG-200, although toluene is occasionally advantageous.
These observations can help address those occasions where
substrate and/or catalyst solubility may be an issue, and when
various practical aspects, especially at scale,21 can otherwise be
problematic. Such aqueous solutions are amenable to recycling,
while DLS measurements have shown that added co-solvent
enlarges the size of the nanoreactors in which the chemistry is
taking place.
1480.
(9) Feng, J.; Handa, S.; Gallou, F.; Lipshutz, B. H. Angew. Chem., Int.
Ed. 2016, 55, 8979.
(10) Handa, S.; Wang, Ye; Gallou, F.; Lipshutz, B. H. Science 2015,
349, 1087.
(11) Bhattacharjya, A.; Klumphu, P.; Lipshutz, B. H. Org. Lett. 2015,
17, 1122.
(12) Handa, S.; Fennewald, J. C.; Lipshutz, B. H. Angew. Chem., Int.
Ed. 2014, 53, 3432.
(13) (a) Lipshutz, B. H.; Ghorai, S. Green Chem. 2014, 16, 3660.
(b) Lipshutz, B. H.; Isley, N. A.; Fennewald, J. C.; Slack, E. D. Angew.
Chem., Int. Ed. 2013, 52, 10911.
(14) Gallou, F.; Guo, P.; Parmentier, M.; Zhou, J. Org. Process Res.
Dev. 2016, 20, 1388.
(15) Parmentier, M.; Wagner, M. K.; Magra, K.; Gallou, F. Org.
Process Res. Dev. 2016, 20, 1104−1107.
(16) Lipshutz, B. H.; Moser, R.; Voigtritter, K. R. Isr. J. Chem. 2010,
50, 691.
(17) Gabriel, C. M.; Keener, M.; Gallou, F.; Lipshutz, B. H. Org. Lett.
2015, 17, 3968.
(18) Nishikata, T.; Abela, A. R.; Lipshutz, B. H. Angew. Chem., Int. Ed.
2010, 49, 781.
ASSOCIATED CONTENT
* Supporting Information
■
(19) Hoffmann, H.; Ebert, G. Angew. Chem., Int. Ed. Engl. 1988, 27,
902.
(20) Sheldon, R. A. Green Chem. 2007, 9, 1273.
(21) Lee, N.; Gallou, F.; Lipshutz, B. H. Org. Process Res. Dev.,
submitted.
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.6b03468.
1
Experimental procedures, characterization data, and H
and 13C NMR spectra (PDF)
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: lipshutz@chem.ucsb.edu.
ORCID
Bruce H. Lipshutz: 0000-0001-9116-7049
Notes
The authors declare no competing financial interest.
D
DOI: 10.1021/acs.orglett.6b03468
Org. Lett. XXXX, XXX, XXX−XXX