Charge-Enhanced Acidity and Catalyst Activation

Article abstract of DOI:10.1021/jacs.5b01805

Acidities are commonly measured in polar solvents but catalytic reactions are typically carried out in nonpolar media. IR spectra of a series of phenols in CCl₄ and 1% CD₃CN/CCl₄ provide relative acidities. Nonprotonated c

Full text of DOI:10.1021/jacs.5b01805

Communication
pubs.acs.org/JACS
Charge-Enhanced Acidity and Catalyst Activation
Masoud Samet, Jordan Buhle,† Yunwen Zhou, and Steven R. Kass*
Department of Chemistry, University of Minnesota, Minneapolis, Minnesota 55455, United States
S
* Supporting Information
Table 1. Hydroxyl Stretching Frequencies for Substituted
Phenols in CCl4 along with Their DMSO pKa and Gas-Phase
Acidity Values
ABSTRACT: Acidities are commonly measured in polar
solvents but catalytic reactions are typically carried out in
nonpolar media. IR spectra of a series of phenols in CCl4
and 1% CD3CN/CCl4 provide relative acidities. Non-
protonated charged substituents with an appropriate
counterion are found to enhance their Brønsted acidities
and improve catalyst performance by orders of magnitude.
ν (cm−1
)
cmpd
(XC6H4OH)
X =
a
1%
Δν
pKa
(DMSO)
ΔG°acid
b
CCl4 ACN
(cm−1
)
(kcal mol−1
)
m-(CH3)2N
m-H
3616
3611
3611
3611
3608
3605
3606
3599
3606
3616
3616
3613
3608
3607
3607
3599
3602
3600
3597
3594
3041
3576
3566
3464
152
157
163
165
185
190
191
212
218
148
153
160
166
172
192
192
196
210
217
221
−2
19.1
343.5 2.0
341.5 1.0
341.3 1.4
341.5 2.0
337.2 2.0
332.4 2.0
335.3 2.0
327.6 2.0
329.0 2.0
344.4 2.0
343.9 2.0
343.8 2.0
340.4 2.0
336.5 2.0
c
c
3454
3448
3446
3423
3415
3415
3387
3388
3468
3463
3453
3442
3435
3415
3407
3406
3390
3380
3373
3043
3247
3196
18.0
m-CH3
m-CH3O
m-F
18.2
cid−base reactions are among the most common and
18.2
A
fundamental transformations in all of chemistry. As a
result, the development of new Brønsted acids and bases are of
general interest and have been the subject of extensive research
efforts.1−5 The strengths of these reagents are most commonly
measured in water and/or dimethyl sulfoxide (DMSO), both of
which are very polar solvents with high dielectric constants.6
Substituent effects are also routinely studied in polar media,
whereas most organic transformations are carried out in less
polar solvents.7,8 Structure−reactivity insights from pKa and
substituent effect data, consequently, can be misleading. In this
work, IR spectroscopy is used to obtain relative acidities of a
series of m- and p-substituted phenols in a nonpolar solvent,
and these results are better fit by gas-phase acidities than the
corresponding DMSO pKa values. This observation led us to
examine charged substituents in nonpolar solvents, and
enhanced acidity and catalytic performance is reported.
15.8
m-CF3
m-Cl
15.6
15.8
m-NO2
m-CN
14.4
14.8
p-(CH3)2N
p-CH3O
p-CH3
p-F
19.8
19.1
18.9
18.0
p-Cl
16.7
p-Br
16.4
p-CH3CO
p-CF3
14.0
328.6 2.0
330.1 2.0
324.2 2.0
325.5 2.0
320.9 2.0
15.3
p-CH3SO2
p-CN
13.6
13.2
p-NO2
10.8
d
e
e
e
f
In a clever study, Reed et al. showed that IR spectroscopy can
be used to provide relative acidities of the strongest Brønsted
acids known to date.9 This was accomplished by comparing the
N−H stretching frequencies of a series of trioctylammonium
salts of deprotonated carboranes. It was found that the weaker
the interaction of the base, the higher the frequency for this
band. Inspired by this work and related studies,10,11 we
obtained the IR spectra of 20 m- and p-substituted phenols in
carbon tetrachloride in the presence and absence of
acetonitrile-d3 (Table 1).12−17 Dilute solutions (5 mM) of
the phenols in the latter case gave rise to a sharp band for the
“free” O−H stretch around 3600 cm−1 (Figure 1, solid line).
Addition of a small amount of CD3CN (1% v/v) led to a large
150−220 cm−1 frequency reduction (i.e., a red shift) and a
broadening of the band due to the formation of an ArOH···
NCCD3 hydrogen bond (Figure 1, dotted line).18,19
A plot of the experimental pKa values in DMSO versus the
observed frequency shifts for both the meta and para isomers is
reasonably well fit by a single line in which the p-nitro and p-
acetyl derivatives are omitted from the least-squares analysis to
improve the correlation coefficient from 0.89 to 0.94 (Figure
2).20 A similar correlation between the gas-phase acidities
(ΔG°acid) of the phenols and the change in their O−H
1
12.5 1.0
12.5 1.0
12.4 1.1
261.4
g
f
2
329
261.4
h
i
3
370
231.1
a
Equilibrium determinations from ref 16 unless otherwise noted; some
b
values are the average of two similar results. 1% ACN = 1% CD3CN/
c
99% CCl4. Measured by threshold collision-induced dissociation (ref
d
e
+
17). 1 = p-HOC6H4N(n-C8H17)2CH3 I−. Measured by bracketing
f
using two colored indicators. B3LYP/6-31+G(d,p) computations on
g
+
+
p-HOC6H4N(n-C8H17)2CH3 . 2 = p-HOC6H4N(n-C8H17)2CH3
BArF − where ArF stands for tetrakis(3,5-bis(trifluoromethyl)phenyl)-
4
h
i
borate. 3 = 3-hydroxy-N-octylpyridinium BArF −. This calculated
4
value is for 3-methylpyridinium phenol.
frequency shifts is obtained for all of the compounds including
the p-NO2 and p-COCH3 derivatives, but in this case, the data
are best fit by separate lines for the meta and para isomers
(Figure 3). These results suggest that resonance delocalization
is not as effective in carbon tetrachloride as it is in dimethyl
sulfoxide because this is a completely stabilizing mechanism for
solvent-separated ion pairs (DMSO) but not for tight ion pairs
Received: February 17, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/jacs.5b01805
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
those in nonpolar media better than the DMSO pKa values.
Consequently, they may be more useful in structure−reactivity
relationships.
Gas-phase substituents with the largest impact are charged
groups, whereas in polar media they are much less effective.21,22
For example, p-COCH3, p-CN, and p-NO2 substituted phenols
are all more acidic than the p-N(CH3)3+ derivative in DMSO,13
but the trimethylammonium ion is predicted to be at least 45
pKa units (i.e., 62.7 kcal mol−1) more acidic than the other
species in the gas phase. This suggests that there should be an
acidity reversal for these compounds in nonpolar solvents and a
new strategy of enhancing Brønsted acidities and catalytic
efficiencies in organic transformations.23−26 To test this
hypothesis, p-dioctylaminophenol was synthesized and sub-
sequently methylated with methyl iodide; octyl groups were
used to enhance the salt’s solubility in carbon tetrachloride. The
resulting methyl(dioctyl)ammonium ion (i.e., p-HOC6H4N(n-
C8H17)2CH3+ I−, 1) has a broad OH band at 3041 cm−1, which
indicates that there is a OH···I− hydrogen bond even though
iodide is a weakly basic anion. Not surprisingly then, the IR
spectrum of this salt is essentially unchanged upon addition of
1% CD3CN. These results imply that 1 is not particularly acidic
in CCl4, and so tetrakis(3,5-bis(trifluoromethyl)phenyl)borate
Figure 1. Representative IR spectrum of phenol in CCl4 (solid line)
and in 1% CD3CN/CCl4 (dotted line).
(BArF −)) was used to replace the iodide counterion. The
4
resulting phenol (p-HOC6H4N(n-C8H17)2CH3 BArF −, 2) has
+
4
an OH stretch at 3576 cm−1, which is red-shifted down to 3247
cm−1 in the presence of acetonitrile-d3. This decrease of 329
cm−1 is 50% larger than for p-nitrophenol, the compound with
the biggest shift that we previously observed, and indicates that
2 is more acidic in CCl4 than the other phenols that were
examined.
Figure 2. DMSO pKa values vs m- and p-substituted phenol O−H
frequency shifts in CCl4 upon addition of CD3CN. A linear least-
squares fit of the data affords y (pKa) = −0.0822 (Δν) + 31.6, r2 =
0.936, where the open diamonds are for the p-NO2 and p-COCH3
derivatives, which were excluded from the analysis.
To assess this result further, the Friedel−Crafts reaction
between β-nitrostyrene and N-methylindole was studied in
chloroform (eq 1).This transformation was chosen because it is
an acid-catalyzed process and its rate should serve as an
indicator of the phenol acidity.24,27 As expected, 1 is a very poor
catalyst, p-nitrophenol is about 7 times better, and 2 is 200-fold
more reactive (Table 2, entries 1−3). These results are in
accord with the acidity reversal of p-nitrophenol and 2 in going
Table 2. Kinetic Results for a Friedel−Crafts Reaction (eq
a
1)
Figure 3. Gas phase acidities (kcal mol−1) vs O−H frequency shifts for
m- (triangles) and p-substituted (diamonds) phenols. Linear least-
squares analyses give y (pKa) = −0.241 (Δν) + 380.3, r2 = 0.950
(meta) [dotted line] and y (pKa) = −0.324 (Δν) + 393.2, r2 = 0.970
(para) [solid line].
entry
catalyst
[nitrostyrene] (mM)
t1/2 (h)
t1/2 (rel)
1
2
3
4
5
6
7
8
9
p-HOC6H4NO2
83
1200
8400
42
1
1
83
0.14
29
750
1
2
83
3
83
1.6
(CCl4). That is, in the latter case charge dispersal diminishes
the electrostatic stabilizing interaction between oppositely
charged ions, and this results in a delicate balance between
charge delocalization and Coulombic attraction. In the gas
phase, counterions are absent and charge delocalization is of
paramount importance, so it is not surprising that the behavior
of the para derivatives diverge from the meta isomers. It was
unexpected, however, that the gas-phase acidities correlate with
p-HOC6H4NO2
235
235
29
640
37
2
17
1
p-HOC6H4NO2
4100
64
2
3
29
64
2000
29
2.1
a
Three equivalents of N-methylindole and a fixed amount of catalyst
(8.3 mM) were used in each case.
B
DOI: 10.1021/jacs.5b01805
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
(6) Bordwell, F. G. Acc. Chem. Res. 1988, 21, 456−463.
from DMSO to CCl4. The relative rates of this transformation,
however, are expected to be concentration-dependent since the
reactants are polar compounds and the relative acidity of p-
nitrophenol and 2 is sensitive to the polarity of the medium.
This hypothesis was borne out in that when the concentrations
of the two reactants were tripled (entries 5 and 6), k2/
kp‑HOC H NO decreased from 29 to 17. Likewise, when the
(7) Charton, M. In Progress in Physical Organic Chemistry; Taft, R. W.,
Ed.; Interscience: New York, 1981; Vol. 13, p 119−251.
(8) Hansch, C.; Leo, A.; Taft, R. W. Chem. Rev. 1991, 91, 165−195.
(9) Juhasz, M.; Hoffmann, S.; Stoyanov, E.; Kim, K. C.; Reed, C. A.
Angew. Chem., Int. Ed. 2004, 43, 5352−5355.
(10) Abkowicz-Bienko, A. J.; Latajka, Z. J. Phys. Chem. A 2000, 104,
1004−1008.
(11) Hussein, M. A.; Millen, D. J.; Mines, G. W. J. Chem. Soc.,
Faraday Trans. 2 1976, 72, 686−692.
(12) All acidities come from refs 6 and 13−17.
(13) Bordwell, F. G.; et al. University of Wisconsin Madison’s
Bordwell pKa table (http://www.chem.wisc.edu/areas/reich/
pkatable).
6
4
2
original concentrations were reduced by a factor of 3 (entries 7
and 8) the ratio increased from 29 to 64. This latter difference
is even larger (i.e., 180) when the reaction is run in toluene-d8.
Additional gas-phase computations revealed that 3-hydroxy-
N-methylpyridinium ion is 22 pKa units (30 kcal mol−1) more
acidic than p-methyl(dipentyl)ammonium phenol, so the
(14) Bordwell, F. G.; McCallum, R. J.; Olmstead, W. N. J. Org. Chem.
1984, 49, 1424−1427.
octylammonium BArF − salt of 3-hydroxypyridine (i.e., 3) was
4
synthesized. Its IR spectra in CCl4 and 1% CD3CN/CCl4 have
bands at 3566 and 3196 cm−1, respectively. This red shift of
370 cm−1 indicates that 3 is more acidic than 2 in carbon
tetrachloride as predicted by the B3LYP/6-31+G(d,p)
calculations. The pyridinium ion 3 was also found to be a
more active catalyst than p-nitrophenol and 2 by factors of up
to 2000 and 30, respectively (entries 7−9).
(15) Bordwell, F. G.; Cheng, J. P. J. Am. Chem. Soc. 1991, 113, 1736−
1743.
(16) Bartmess, J. E. NIST Chemistry WebBook, NIST Standard
Reference Database Number 6; Mallard, W. G., Lustrum, P. J., Eds.;
National Institute of Standards and Technology: Gaithersburg, MD
(http://webbook.nist.gov).
(17) Angel, L. A.; Ervin, K. M. J. Phys. Chem. A 2006, 110, 10392−
10403.
Protonated catalysts have been successfully employed in
organic transformations24,28−31 but provide an additional
hydrogen bond donor site that can be actively involved in
the reaction. Nonprotonated ions eliminate this concern and
can enhance acidities of ionizable groups in nonpolar
environments. This effect undoubtedly can be reversed to
increase the basicity of basic sites, and the exploitation of
electrostatics is an exciting avenue for further exploration.21
Studies along these lines are in progress, and even more
dramatic effects may be obtained by incorporating the
counterion into the reagent at a remote noninteracting location.
(18) One hydrogen bond criterion typically observed is a red shift in
the IR Arunan, E.; et al. Pure Appl. Chem. 2011, 83, 1637−1641.
(19) Similar results are observed in CDCl3 and 1% CD3CN/99%
CDCl3 (i.e., p-XC6H4OH, X = H (3598 and 184 cm−1), Br (3596 and
188 cm−1), CN (3582 and 227 cm−1), NO2 (3580 and 240 cm−1), and
2 (3555 and 320 cm−1) where the values in parentheses are ν and Δν,
respectively.
(20) Separate plots for the meta and para derivatives do not lead to
improved data fits.
(21) Patrick, J. S.; Yang, S. S.; Cooks, R. G. J. Am. Chem. Soc. 1996,
118, 231−232.
(22) Strittmatter, E. F.; Wong, R. L.; Williams, E. R. J. Am. Chem. Soc.
2000, 122, 1247−1248.
ASSOCIATED CONTENT
* Supporting Information
(23) For related studies employing protonated catalysts, charged
metal-templated hydrogen bond donors, and pyridinium cations as
anion-binding catalysts, see refs 24−26, respectively.
(24) Ganesh, M.; Seidel, D. J. Am. Chem. Soc. 2008, 130, 16464−
16465.
■
S
Experimental and computational sections including synthetic
procedures, NMR spectra, and calculated structures and
energies along with the complete citation to ref 18 are
provided. This material is available free of charge via the
Internet at http://pubs.acs.org.
(25) (a) Scherer, A.; Mukherjee, T.; Hampel, F.; Gladysz, J. A.
Organometallics 2014, 33, 6709−6722. (b) Mukherjee, T.; Ganzmann,
C.; Bhuvanesh, N.; Gladysz, J. A. Organometallics 2014, 33, 6723−
6737.
AUTHOR INFORMATION
Corresponding Author
*kass@umn.edu
(26) Berkessel, A.; Das, S.; Pekel, D.; Neudorfl, J.-M. Angew. Chem.,
̈
■
Int. Ed. 2014, 53, 11660−11664.
(27) Shokri, A.; Wang, X. B.; Kass, S. R. J. Am. Chem. Soc. 2013, 135,
9525−9530.
Present Address
(28) Huang, J.; Corey, E. J. Org. Lett. 2004, 6, 5027−5029.
(29) Bolm, C.; Rantanen, T.; Schiffers, I.; Zani, L. Angew. Chem., Int.
Ed. 2005, 44, 1758−1763.
†Ripon College, Ripon, Wisconsin 54971, United States.
Notes
(30) Takenaka, N.; Sarangthem, R. S.; Seerla, S. K. Org. Lett. 2007, 9,
2819−2822.
The authors declare no competing financial interest.
(31) Auvil, T. J.; Schafer, A. G.; Mattson, A. E. Eur. J. Org. Chem.
2014, 2633−2646.
ACKNOWLEDGMENTS
■
Generous support from the National Science Foundation and
the Minnesota Supercomputer Institute for Advanced Compu-
tational Research are gratefully acknowledged.
REFERENCES
■
(1) Reed, C. A. Acc. Chem. Res. 1998, 31, 133−139.
(2) Akiyama, T. Chem. Rev. 2007, 107, 5744−5758.
(3) Yamamoto, H.; Boxer, M. B. Chimia 2007, 61, 279−281.
(4) Tian, Z.; Fattahi, A.; Lis, L.; Kass, S. R. J. Am. Chem. Soc. 2009,
131, 16984−16988.
(5) Parmar, D.; Sugiono, E.; Raja, S.; Rueping, M. Chem. Rev. 2014,
114, 9047−9153.
C
DOI: 10.1021/jacs.5b01805
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX