Rate acceleration of organic reaction by immediate solvent evaporation

Article abstract of DOI:10.1039/b609567d

Several types of organic reactions were accelerated by immediate evaporation of solvents because of remarkable enhancement of molecule-to-molecule contacts between reactants. The Royal Society of Chemistry.

Full text of DOI:10.1039/b609567d

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Rate acceleration of organic reaction by immediate solvent
evaporation{
Akihiro Orita, Genta Uehara, Kai Miwa and Junzo Otera*
Received (in Cambridge, UK) 5th July 2006, Accepted 31st August 2006
First published as an Advance Article on the web 29th September 2006
DOI: 10.1039/b609567d
method (HCSM) which had been reported to be more efficient than
Several types of organic reactions were accelerated by imme-
diate evaporation of solvents because of remarkable enhance-
ment of molecule-to-molecule contacts between reactants.
the normal solution reaction for synthesis of [2]rotaxanes4 was
invokedasacontrolexperiment. Both1a(2mmol)and2(2.2mmol)
were dissolved in a minimum amount of CH2Cl2 (usually 1 mL).
The solution was stirred at rt for 5 min, but 3 was formed only in
56% yield. The reaction finished after 2 h. A more marked
difference was observed with 3,5-dichlorobenzaldehyde (1b). Upon
evaporation for 5 min, a trace amount of 1b (1%) was detected, and
hence the resulting solid film was kept standing under ambient
conditions for additional 30 min to complete the reaction (.99%
yield based on NMR) (entry 2). The conversion of the correspond-
ing reaction in highly concentrated solution after 5 min was only
17% and it took 24 h for the reaction to finish. p-Nitrobenzaldehyde
(1c) and 2-naphthaldehyde (1d), also gave comparable results
(entries 3 and 4) indicative of superiority of ISEM to HCSM.5
Wittig reaction is one of the most useful reactions in organic
synthesis,6 but a long reaction time is demanded occasionally, in
particular for stabilized ylides. It has turned out that reaction
between solid aldehydes and ylides undergoes a great deal of
acceleration by immediate evaporation (Table 2). In a reaction of
1c with ylide 4a (R9 = COOEt), only a weak spot of 1c was
detected by TLC monitoring after evaporation (y2% by 1H
NMR). The resulting solid film was kept standing for 1 h at rt to
bring the reaction to completion (entry 1). The desired olefin 5a
was isolated in 93% yield by column chromatography. On the
other hand, the ratio 5a/1c was only 70 : 30 after 5 min according
to HCSM. It took 24 h to give a 92% yield, but a small amount of
1c still remained nevertheless. Since the initial stage of reaction
between 1c and 4b (R9 = COPh) proceeded more slowly, the
Organic reactions are usually conducted in solution, partly because
better mixing is attainable when reactants are solid or immiscible
with each other. Although the reactions finish quickly on occasion,
a long time is required frequently. Curtailment of the reaction time
is desirable from both operational and economical points of view.
The reaction rate can be accelerated by elevating the reaction
temperature. Another way is to increase concentration of the
reaction solution so long as heat evolution is not violent. However,
it is not common to condense the solution beyond the concentration
level at which precipitation is triggered, because the heterogeneous
reaction is conventionally regarded as being slower than the
homogeneous one. Contrary to such general acceptance, we earlier
disclosed dramatic rate acceleration in solventless reactions. Simple
grinding of the solid reactants led to supramolecular self-assembling
much faster than the normal solution reaction.1 Rotaxanes were
fabricated even without grinding once a solid film of reactants had
been formed.2 Namely, the solid reactants were dissolved in a
solvent and the resulting solution was immediately evaporated to
give a solid film. Merely standing the film under ambient conditions
drove the reaction at a rate faster than in the case where the original
solution was kept stirring.3 During these experiments, we observed
in some cases that the reaction had already occurred to a certain
extent just when the evaporation was over. This suggested that a
reaction might proceed faster by immediate evaporation rather
thanbycontinuingthereaction insolution and, hence, we presumed
that the reaction time could be shortened by taking advantage of
such rate acceleration. This is indeed the case, and we herein
exemplify the validity with imine synthesis, Wittig reaction, and
quaternization of tertiary phosphine and pyridine.
1
conversion could be quantified more clearly by H NMR, which
revealed the ratio 5b/1c to be 73 : 27 after evaporation according to
ISEM and 53 : 47 after 5 min by HCSM (entry 2). The reaction
finished in 1 h and 3 h by ISEM and HCSM, respectively. The
reaction of p-chlorobenzaldehyde (1e) took place similarly but
exhibited more distinct difference in the reaction rate between both
protocols (entry 3). According to ISEM, the conversion was
already 99% after evaporation and the reaction finished in 15 min
upon additional standing. By contrast, a small amount of 1e
remained even after 24 h according to HCSM. A similar advantage
of ISEM was apparent with sterically demanding 2,3-dimethoxy-
benzaldehyde 1f (entry 4). Since the reaction of 9-anthraldehyde 1g
was rather slow under both conditions at rt (entry 5), evaporation
by rotary evaporator was carried out at 40 uC followed by
pumping in vacuo at rt. The ratio of the product olefin vs.
unreacted aldehyde was found to be 80 : 20 (based on 1H NMR) at
this stage, and the reaction finished in 3 h while it took 24 h for
completion by HCSM at 40 uC (entry 6). The utility of the higher
3,5-Dibromobenzaldehyde (1a) (2 mmol) and p-toluidine (2)
(2.2 mmol) were dissolvedinCH2Cl2 (4mL). Then, thesolution was
immediately evaporated by means of rotary evaporator followed
by vacuum pumping at rt (immediate solvent evaporation method:
ISEM). It took about 5 min for complete removal of the solvent.
The resulting white film was already constituted of pure imine 3a
(.99%) and a small amount of 2 on the basis of 1H NMR
spectroscopy,noaldehydebeingdetectedonTLC(Table1,entry1).
To assess this new protocol, the high concentration solution
Department of Applied Chemistry, Okayama University of Science,
Ridai-cho, Okayama, 700-005, Japan. E-mail: otera@high.ous.ac.jp;
Fax: +81 86 256 4292; Tel: +81 86 256 9525
{ Electronic supplementary information (ESI) available: Experimental
details and characterization of all products. See DOI: 10.1039/b609567d
This journal is ß The Royal Society of Chemistry 2006
Chem. Commun., 2006, 4729–4731 | 4729

View Article Online
after evaporation according to ISEM. The reaction had already
Table 1 Synthesis of imines: RCHO 1 + H2NC6H4CH3 2 A
RCHLN–C6H4CH3
ISEMa
3
finished at this stage while HCSM afforded a 92% yield after 24 h
(entry 8). Notably, electron-rich aldehyde which usually reacts very
slowly also enjoyed the acceleration. Thus, reaction of anisalde-
hyde (1i) was almost over in 7 h by ISEM whereas HCSM failed to
drive the reaction to finish after 7 h (entry 9). The double-bond
geometry of the resulting olefins deserves further comments.
Nearly the same E/Z ratio was revealed for the products from both
methods, thereby suggesting that the reaction mode in terms of
olefin geometry is not significantly altered by suction of solvent
from the conventional reaction system.
HCSMb
3 : 1c
(after evpn., after standing after stirring after stirring
3 : 1c
3 : 1c
3 : 1c
Entry 1 5 min)
(t/min)
(5 min)
56 : 44
17 : 83
30 : 70
46 : 54
(t/h)
1
2
3
1a .99 : 1
—
.99 : 1 (2)
.99 : 1 (24)
99 : 1 (24)
.99 : 1 (24)
1b
1c
99 : 1
98 : 2
.99 : 1 (15)
.99 : 1 (15)
.99 : 1 (120)
Transformation of triphenylphosphine (6) to phosphonium salts
8 was also accelerated.8 As shown in Table 3, smooth alkylation
occurred with 2-(bromomethyl)naphthalene (7a) by ISEM in
contrast to unsatisfactory outcome by HCSM. On the other hand,
reaction with 4-(bromomethyl)acetophenone (7b) required 20 h for
completion (.99% yield) by standard ISEM. However, upon
evaporation at 40 uC as described above, the reaction finished in
3 h. Notably, the reaction did not come to an end in 7 h in a highly
concentrated solution at 40 uC.
Finally, quaternization of pyridine proved to experience rate
enhancement as well (Table 4). Bipyridine derivative 9 was
alkylated quantitatively in 3 h by ISEM whereas HCSM provided
only a 61% conversion after 24 h.
4
1d
97 : 3
a
Dissolve
1
(2 mmol) and 2 (2.2 mmol) in CH2Cl2 (4 mL);
evaporation by rotary evaporator followed by pumping in vacuo;
b
standing at rt. Stirring a solution of 1 (2 mmol) and 2 (2.2 mmol)
in CH2Cl2 (1 mL) at 27 uC. Determined by 1H NMR.
c
Table 2 Witting reaction: RCHO 1 + Ph3PLCHR9 4 A RCHLCHR9 5
Yield (%)a of 5
ISEMb
HCSMc
5 : 1d
after
Yield
(time/h;
E/Zd)
5 : 1d
after
Yield
(time/h;
Entry 1
4
evpn.
5 min E/Zd)
70 : 30 92 (24; 97 : 3)e
53 : 47 93 (3; 98 : 2)
62 : 38 91 (24; 96 : 4)e
Recently, Sharpless et al. put forth the ‘‘on water’’ method, by
which various reactions take place very quickly at the macroscopic
phase boundary between water and insoluble oils.9 We found a
considerable effectiveness of this on-water protocol for the reaction
1
2
3
1c 4a 5a 98 : 2 93 (1; 99 : 1)
1c 4b 5b 73 : 27 92 (1; 99 : 1)
1e 4a 5c .99 : 1 93 (15 min;
95 : 5)f
1f 4a 5d 99 : 1 93 (10 min;
92 : 8)
4
12 : 88 89 (24; 91 : 9)e
5
6
7
1g 4a 5e
1g 4a 5e
1d 4a 5f
63 : 37 94 (24; 97 : 3)e 11 : 89 89 (24; 96 : 4)e
80 : 20 90 (3; 98 : 2)g 33 : 67 91 (24; 95 : 5)h
91 : 9 93 (10 min;
93 : 7)g
Table 3 Syntheses of phosphonium salts
63 : 37 94 (5; 98 : 2)h
8
9
a
1h 4a 5g .99 : 1 93 (0; 94 : 6)i ,1 : 99 92 (24; 95 : 5)
1i 4a 5h 32 : 68 90 (7; 96 : 4)e ,1 : 99 81 (7; 94 : 6)e
b
Isolated yield after column chromatography. Dissolve 1 (2 mmol)
and
4 (2.2 mmol) in CH2Cl2 (4 mL); evaporation by rotary
c
evaporator followed by pumping in vacuo; standing at rt. Stirring a
solution of 1 (2 mmol) and 4 (2.2 mmol) in CH2Cl2 (1 mL) at 27 uC.
Yield/%
d
f
e
Determined by NMR. Aldehyde remained by TLC monitoring.
g
Reaction was almost complete after evaporation. Evaporation at
ISEM
HCSM
h
i
40 uC. Stirring at 40 uC. After evaporation.
8:7 after
evaporation
8:7 after
5 min
Yield (time/h)
Yield (time/h)
temperature was also seen in the reaction between 1d and 4a
(entry 7). The reaction was nearly over after evaporation at 40 uC
(5f/4a = 91 : 9) and came to an end after standing for additional
10 min. In contrast, much slower reaction occurred according to
HCSM. It should be noted, however, that the elevation of
temperature is not always effective since in some cases, too quick
evaporation hampers homogeneous mixing of reactants which is
necessary for efficient contacts between reactant molecules in the
solid film. Since Wittig reaction without solvent is known,7 it may
be of interest to compare ISEM with the typical solventless method
in which the reactants are ground on the mortar. Although the
reactions of entries 1 and 3 exhibited comparable rates in both
protocols, the reaction timeofotherreactionsin ISEM wasroughly
half of that in the grinding method except for the reaction of 1g with
which no reaction occurred in the latter method.
8a
8b
a
91 : 9a
91b (2)
.99a,d (3)
b
12 : 88a
22 : 78a
87b,c (34)
91a,c,e (7)
89 : 11a
Determined by 1H NMR.
Isolated yield by column
A small amount of 7 remained on TLC.
e
c
chromatography.
d
Evaporation at 40 uC. Stirring at 40 uC.
Table 4 Synthesis of pyridinium salt
10 : 9a
ISEM
HCSM
After evaporation
After standing (3 h)
After 5 min
After 7 h
61 : 39
The present protocol is not restricted to solid aldehydes. When
liquid benzaldehyde (1h) and 4a were combined directly, no
homogeneous mixture emerged causing incomplete reaction.
However, a smooth but somewhat wet solid film was obtained
80 : 20 .99 : 1
Determined by 1H NMR.
a
1 : 99
4730 | Chem. Commun., 2006, 4729–4731
This journal is ß The Royal Society of Chemistry 2006

View Article Online
a solution reaction. It is fully recognized that solvation heavily
decreases the reaction rate.13 Furthermore, smaller entropy change
during the reaction in the solid state may serve for lowering the
activationenergy.Since0.4mLofCH2Cl2isequivalentto6.7mmol,
each substrate (2.1 mmol in total) is estimated to be solvated at most
by three or four molecules of CH2Cl2 on average in the saturated
solution. It follows that removal of the final portions of the solvent
molecules may give rise to a substantial leap of rate increase. The
downhill slope under heterogeneous conditions also suggests the
effectiveness of ISEM, by which the substrates are once dissolved so
as to be completely intermingled atthe molecular level. On the other
hand, under the above heterogeneous conditions or normal
solventless conditions (grinding method), the contacts between the
reactants are achievable only at the level of mass particles.
Fig. 1 Concentration effect: Yields of Wittig reaction of 1f with 4a in
HCSM after 5 min.
It is now apparent that immediate evaporation of the solvent
can conduct reaction faster than running it in solution. The
reaction is completed immediately after solvent evaporation in
some cases, but not always. If not, standing the resulting film
drives the reaction to completion again faster than the solution
reaction. Such a simple operation is of great use from the synthetic
point of view. Since the rate acceleration by ISEM is deviated from
the simple extrapolation of normal solution reaction, the present
protocol provides a new facet on organic reaction mechanisms. In
conclusion, we propose to reconsider the reactions in the light of
the immediate evaporation method when their reaction rate is too
slow in solution. The reaction time in solution may well be
shortened. Of course, this protocol is not always effective for any
reactions, yet shortening of reaction time is basically feasible for a
number of reactions which require a long time in solution.
of 4a with 1h, yet its reaction rate (75% and .99% yields of 5g
1
based on H NMR after 5 min and 1.5 h, respectively, at 27 uC)
was slower than that obtained with ISEM.10
To get further insight into characteristics of ISEM, the
concentration effect of the reaction of entry 4 in Table 2 was
scrutinized. The reaction between 1f (1.0 mmol) and 4a (1.1 mmol)
1
in various amounts of CH2Cl2 at 27 uC was monitored by H
NMR spectroscopy.11 The yields 5 min after mixing of the two
reactants were determined (Fig. 1) because the solvent evaporation
in ISEM finishes within 5 min or it may be that the evaporation
has finished actually within 2–3 min. The reaction mixture was
homogeneous when the amount of the solvent was between 2.00–
0.40 mL, while a part of the reactants precipitated in less than
0.40 mL of the solvent. As expected from the concentration effect,
the relationship of yield vs. solvent volume exhibited an uphill
slope under homogeneous conditions, and the slope was reversed
downhill under heterogeneous conditions. Extrapolation of the
uphill slope of the homogeneous region (concentration between
1.00 mmol/0.55 mL and 1.00 mmol/0.40 mL) by use of the least
square method12 to the putative solventless extreme leads to the
75% yield (dotted line), an outcome suggesting that the rate
enhancement in ISEM does not simply result from the concentra-
tion effect. The downhill slope indicates that the concentration
effect does not hold in the region below 0.40 mL. Therefore, it is
not reasonable to expect such a big leap in the yield (from 60 to
99%) as found in ISEM on the basis of the concentration effect.
The present outcome, if being qualitative, clearly shows that ISEM
gives rise to the higher yield than that expected to be attained at
the ultimate high concentration in solution. Practically, the
advantage of ISEM over HCSM is apparent if the limitation of
HCSM due to the saturated solubility is taken into account as is
evident from the 60% maximum yield in the present case.
Notes and references
1 A. Orita, L. Jiang, T. Nakano, N. Ma and J. Otera, Chem. Commun.,
2002, 1362.
2 A. Orita, J. Okano, Y. Tawa, L. Jiang and J. Otera, Angew. Chem., Int.
Ed., 2004, 43, 3724.
3 For solventless reactions in general: (a) F. Toda, Synlett, 1993, 303; (b)
F. Toda, Acc. Chem. Res., 1995, 28, 480; (c) K. Tanaka and F. Toda,
Chem. Rev., 2000, 110, 1025; (d) G. W. V. Cave, C. L. Raston and
J. L. Scott, Chem. Commun., 2001, 2159; (e) K. Tanaka, Solvent-free
Organic Synthesis, Wiley-VCH, Weinheim, 2003.
4 K. Nikitin, B. Long and D. Fitzmaurice, Chem. Commun., 2003, 282.
5 Imine formation without solvent: (a) J. Schmeyers, F. Toda, J. Boy and
G. Kaupp, J. Chem. Soc., Perkin Trans. 2, 1998, 989; J. Schmeyers,
F. Toda, J. Boy and G. Kaupp, J. Chem. Soc., Perkin Trans. 2, 2001, 132;
(b) G. Kaupp, J. Schmeyers and J. Boy, Tetrahedron, 2000, 56, 6899.
6 (a) B. E. Maryanoff and A. B. Reitz, Chem. Rev., 1989, 89, 863; (b)
O. I. Kolodiazhnyl, Phophorus Ylides, Chemistry and Application in
Organic Synthesis, Wiley-VCH, Weinheim, 1999.
7 (a) F. Toda and H. Akai, J. Org. Chem., 1990, 55, 3446; (b) W. Liu,
Q. Xu, Y. Ma, Y. Liang, N. Dong and D. Guan, J. Organomet. Chem.,
2001, 625, 128; (c) V. P. Balema, J. W. Wiench, M. Pruski and
V. Pecharsky, J. Am. Chem. Soc., 2002, 124, 6244; (d) T. Thiemann,
M. Watanabe, Y. Tanaka and S. Mataka, New J. Chem., 2004, 28, 578.
8 Mechanochemical synthesis of phosphonium salts without solvent:
V. P. Balema, J. W. Wiench, M. Pruski and V. K. Pecharsky, Chem.
Commun., 2002, 724.
Notably, ISEM works on a large scale as well. When the same
reaction was carried out on a 20 mmol scale, the solvent (20 mL)
was removed at 27 uC in 10 min to leave a dry film which consisted
of 5d : 1f in a 91 : 9 ratio. The reaction was complete after 15 min
upon standing the resulting film at room temperature.
9 S. Narayan, J. Muldoon, M. G. Finn, V. V. Fokin, H. C. Kolb and
K. B. Sharpless, Angew. Chem., Int. Ed., 2005, 44, 3275.
10 More recently, Wittig reaction in water only has been reported:
J. Dambacher, W. Zhao, A. El-Batta, R. Anness, C. Jiang and
M. Bergdahl, Tetrahedron Lett., 2005, 46, 4473.
It is not clear enough at present what plays a pivotal role for the
acceleration of the reaction rate by ISEM. It can be said that
molecule-to-molecule contacts between the respective reactants are
made easier by the evacuation of solvent molecules which have
joined reactants through solvation. In addition, the overall reaction
time is saved by quick reaction in the solid film.2 Solventless
reactionsarefreefromthede-solvationprocesswhichisinevitablein
11 For detailed experiments, see ESI{.
12 The results of 3–5 experiments were used for respective concentrations.
13 C. Reichardt, Solvents and Solvent Effects in Organic Chemistry, Wiley-
VCH, Weinheim, Third, Updated and Enlarged Edition, 2004, ch. 5.
This journal is ß The Royal Society of Chemistry 2006
Chem. Commun., 2006, 4729–4731 | 4731