Structural considerations for charge-enhanced Br?nsted acid catalysts

Article abstract of DOI:10.1002/poc.4069

All three N-methylated and N-protonated hydroxypyridinium BAr^F₄^– salt isomers were synthesized and their hydrogen bond donating abilities were investigated. DFT and G4 theory computations along with IR spectroscopic measurements were found to be effective methods for predicting the catalytic activities of these O–H and N–H Br?nsted acids. A UV-vis titration approach for rapidly quantifying hydrogen bond donating ability revealed that carbon-hydrogen bonds also can participate in electrostatic interactions, but the presence of multiple equilibrium complexes results in a limitation of this method. In the methylated series of hydroxypyridines, the ortho and para isomers displayed modest rate enhancements relative to the meta derivative. Protonation introduces a new acidic site and the ortho hydroxypyridinium ion salt is a significantly more active catalyst than all of the other species examined. This is indicative of bidentate activation by the N–H and O–H acidic sites, and suggests a new design strategy for improving charge-enhanced catalysts.

Full text of DOI:10.1002/poc.4069

Received: 3 January 2020
DOI: 10.1002/poc.4069
Revised: 18 February 2020
Accepted: 24 February 2020
R E S E A R C H A R T I C L E
Structural considerations for charge-enhanced Brønsted
acid catalysts
Curtis Payne
| Steven R. Kass
Department of Chemistry, University of
Minnesota, Minneapolis, MN, USA
Abstract
All three N-methylated and N-protonated hydroxypyridinium BArF – salt iso-
4
Correspondence
mers were synthesized and their hydrogen bond donating abilities were inves-
tigated. DFT and G4 theory computations along with IR spectroscopic
measurements were found to be effective methods for predicting the catalytic
activities of these O–H and N–H Brønsted acids. A UV-vis titration approach
for rapidly quantifying hydrogen bond donating ability revealed that carbon-
hydrogen bonds also can participate in electrostatic interactions, but the pres-
ence of multiple equilibrium complexes results in a limitation of this method.
In the methylated series of hydroxypyridines, the ortho and para isomers dis-
played modest rate enhancements relative to the meta derivative. Protonation
introduces a new acidic site and the ortho hydroxypyridinium ion salt is a sig-
nificantly more active catalyst than all of the other species examined. This is
indicative of bidentate activation by the N–H and O–H acidic sites, and sug-
gests a new design strategy for improving charge-enhanced catalysts.
Steven R. Kass, Department of Chemistry,
University of Minnesota, 207 Pleasant
Street SE. Minneapolis, Minnesota 55455,
USA.
Email: kass@umn.edu
Funding information
National Science Foundation, Grant/
Award Number: CHE-1665392
K E Y W O R D S
Brønsted acid catalysis, charge-activated acids, computations, IR and UV-vis spectroscopy,
kinetics
1 | INTRODUCTION
electron-withdrawing substituents are commonly
employed to improve reactivity.[5] For example, phenyl
Enzymes exploit hydrogen bonds to control their three-
dimensional structures, facilitate substrate molecular rec-
ognition and binding, and accelerate reaction pro-
cesses.[1] Their catalytic ability is intertwined with
numerous chemical mechanisms, but often involves elec-
trophilic activation of a substrate toward nucleophilic
attack by lowering the energy of its lowest unoccupied
molecular orbital (LUMO).[2] As a result, the develop-
ment of small metal-free Brønsted acids and hydrogen
bond donors to catalyze a wealth of organic transforma-
tions has emerged as an attractive research area in the
past several decades.[3] From these studies it has been
found that for the same class of compounds with the
same binding motifs, that the acidity of the catalyst often
correlates with its activity.[4] It is for this reason that
rings are routinely replaced by bis(3,5-trifluoromethyl)
phenyl groups, and this latter framework is viewed as
occupying a privileged position.[6]
Brønsted acids are routinely characterized by their
pKa values which typically are measured in water,
dimethyl sulfoxide, or acetonitrile.[7] These solvents have
high dielectric constants which facilitate the acidity
determinations and lead to solvent separated ion pairs.
Brønsted acids and hydrogen bond donating
organocatalysts, however, are usually employed in non-
polar media such as toluene, chloroform, and dic-
hloromethane where intimate ion pairs and aggregates
are formed. Substituent effects and catalyst activities,
consequently may be poorly represented by routinely
determined pKa values. To address this issue, Kozlowski
J Phys Org Chem. 2020;e4069.
wileyonlinelibrary.com/journal/poc
© 2020 John Wiley & Sons, Ltd.
1 of 16
https://doi.org/10.1002/poc.4069

2 of 16
PAYNE AND KASS
et al. developed a UV-vis spectroscopy approach in which
a colorimetric sensor, 7-methyl-2-phenylimidazo[1,2-a]
pyrazine-3(7H)-one (1) is titrated with a hydrogen bond
donor (HX) in dichloromethane (Scheme 1).[8] The recip-
rocals of the observed blue-shifts of the bound complexes
(1••HX) were found to correlate with the logarithms of
the association binding constants (K) and the rate con-
stants (k) for the Diels Alder reaction of cyclopentadiene
with methyl vinyl ketone.
structural framework on catalyst activity, we report
herein on all three isomers of N-methylated and N-
protonated hydroxypyridinium BArF – salts along with
4
the two parent pyridinium ions (Figure 1). These com-
pounds were probed by IR and UV-vis spectroscopy, DFT
and G4 theory computations,[19] and a pseudo-first-order
rate study.
More recently, we made use of IR spectroscopy to
determine the relative acidities of substituted phenols in
carbon tetrachloride by examining the change in their O–
H stretching frequencies (Δν) upon addition of a small
amount of a hydrogen bond acceptor (i.e., CD3CN).[9] It
was found that Δν correlates with the gas-phase acidity
more strongly than DMSO pKa values. This led to the
suggestion and observation that charged substituents are
much more effective at enhancing acidities and catalytic
abilities than neutral polar groups such as CF3, CN, and
NO2. For example, it was found that the Friedel-Crafts
reaction between N-methylindole and trans-β-
nitrostyrene in chloroform occurs ~103 times more rap-
idly when N-octyl-3-hydroxypyridinium tetrakis[3,5-bis
2 | EXPERIMENTAL
2.1 | General
All reaction glassware (vials, NMR tubes, flasks) and stir
bars were dried in an oven (120ꢀC) for at least 16 h, and
glass syringes and volumetric flasks were stored in a vac-
uum desiccator over Drierite for at least 16 h. Alumina
(neutral, Brockman I, standard grade, 150 mesh, 58 Å)
and 3 Å molecular sieves were activated at 300ꢀC for at
least 24 h. Calcium chloride was obtained from Fischer
Scientific and was dried at 300ꢀC for at least 30 min
(note: after 24 h at this temperature, the CaCl2 becomes
tan and the use of this discolored material led to poor
results).
(trifluoromethyl)phenyl]borate (BArF ) rather than when
4
4-nitrophenol is used as the catalyst (equation 1).
Anhydrous acetonitrile and iodomethane were
obtained from Acros Organics and used without further
purification. 1,2-Dichloroethane (DCE) was purchased
from Fischer Scientific and was dried with a column of
activated alumina and stored over 3 Å molecular sieves
under inert atmosphere for at least 24 h prior to use. N-
Methylindole was used as received from Alfa Aesar. Con-
centrated HCl (37%) was obtained from VWR Interna-
tional. Sodium tetrakis(3,5-bis (trifluoromethyl)phenyl)
Since this discovery, achiral and chiral thioureas[10]
and phosphoric acids[11] have been investigated with this
charge-activated design. In these reports, the charged
borate (NaBArF ) was obtained as a 2.5 hydrate (based on
4
its 1H NMR spectrum) from AK Scientific. The water was
removed by dissolving the material in anhydrous metha-
nol (2 g in 2 mL) and running the light-yellow mixture
through two pipets of activated alumina (1.25 g/pipet,
each pipet was rinsed with an additional 2 mL of metha-
nol) and then through a 0.45 μm syringe filter. The sol-
vent was removed with a rotary evaporator and the
remaining colorless solid was ground into a fine powder
and heated at 150ꢀC at 0.1 torr for at least 16 h[20] in a
glovebox (less than 0.06 H2O remained based upon the
1H NMR spectrum). N-Methyl-2-pyridone was acquired
center was introduced most commonly with
a
3-substituted N-alkylated pyridine ring.[12] Other cationic
compounds such as amidinium,[13] guanidinium,[14]
ammonium,[15] pyridinium,[16] and quinolinium[17] ions
also have been shown to accelerate a variety of chemical
transformations.[18] These protonated compounds are
more acidic than their neutral conjugate bases and have
an additional hydrogen bond donating site that can alter
the binding motif of the catalyst, and enhance its associa-
tion with the substrate. To address the impact of the
SCHEME 1 UV-vis sensor 1 and its bound complex with a
Brønsted acid
FIGURE 1 Charged catalysts screened in this work

PAYNE AND KASS
3 of 16
from ArkPharm, Inc. as a dark brown oil that was dis-
solved in anhydrous ethyl acetate, passed through a col-
umn of activated alumina (1.25 g), and concentrated to
N2 for 10 min. High resolution electrospray ionization
mass spectra (HRMS-ESI) were obtained in methanol
with
a
Bruker ESI-BioTOF instrument, and
afford
a
light-yellow oil. In
a
similar manner,
tetramethylammonium and tetraethylammonium salts
were used as mass calibrants. Due to the low masses of
the protonated salts (~96 m/z), protonated 2,6-lutidine
was used as a third calibration point. UV spectra were
collected with a Lambda XLS spectrometer from CH2Cl2
solutions in a 10 mm quartz cell that was sealed with a
PTFE septum.
2-methoxypyridine (Oakwood Chemical) was dried with
an activated alumina column in anhydrous CH2Cl2.
All other chemicals and anhydrous solvents were sup-
plied by Sigma-Aldrich and were used as received unless
described below. Pentane was stored over 3 Å molecular
sieves under argon for 24 h prior to use. Pyridine was
dried with a short column of activated alumina and
stored over 3 Å molecular sieves under inert atmosphere
for 2 days prior to use.[21] 2- and 3-Hydroxypyridines
were dried by dissolving 300 mg samples in 2 mL of
anhydrous methanol and passing the yellow mixture
2.1.1 | N-Methyl-4-pyridone[23]
In a 25 mL round-bottomed flask fitted with a reflux
condenser, 4-hydroxypyridine (200 mg, 2.10 mmol) and
toluene (5 mL) were stirred and the atmosphere was
purged with argon for 15 min. Iodomethane (700 μL,
1.60 g, 11.2 mmol) was then added in one portion by
syringe and the resulting mixture was refluxed over-
night. Upon cooling to room temperature an orange
precipitate and a light-yellow supernatant remained.
Hexanes (15 mL) were added to the suspension and the
solid was isolated via vacuum filtration. The collected
material (~460 mg) was placed in a 6-dram vial, dis-
solved in methanol (5 mL), and K2CO3 (300 mg,
2.17 mmol) was added with vigorous stirring. After 2 h
the reaction mixture was filtered, and the solid material
was rinsed with 5 mL of methanol. Removal of the sol-
vent under reduce pressure afforded a yellow oily solid
which was dissolved in 5 mL of CH2Cl2. A white pre-
cipitate was filtered away and the light-yellow filtrate
was concentrated with a rotary evaporator to give
146 mg of an oily solid. Purification of this material by
column chromatography on alumina with 95/5
MeOH/CH2Cl2 (Rf = 0.3) afforded 104 mg (45%) of a
colorless powder (mp 93 – 96ꢀC) that was stored in a
glovebox. 1H NMR (500 MHz, CD2Cl2) δ 7.23 (d,
J = 7.2 Hz, 2H), 6.20 (d, J = 7.3 Hz, 2H), 3.58 (s, 3H).
13C NMR (125 MHz, CD2Cl2) δ 178.5, 141.1, 118.8, 43.9.
IR (ATR source) 1674, 1637, 1555 cm–1.
3
through a Pasteur pipet 4 full of activated alumina
(~1.25 g). After rinsing the column with additional sol-
vent (2 mL), the colorless solution was concentrated with
a rotary evaporator and then a mechanical pump
(0.1 torr), and the remaining white solid was stored in a
glovebox. 4-Hydroxypyrdine was dried in a similar man-
ner to its isomers but was melted at 150ꢀC while under
reduced pressure (0.1 torr) for 5 min to remove any resid-
ual water (30 min under these conditions led to discolor-
ation and presumed decomposition).
Deuterated solvents came from Cambridge Isotope
Laboratories and were stored over 3 Å molecular sieves
for at least 24 h prior to use. Bruker Avance III HD
400 or 500 MHz instruments were used to collect 1H, 13C,
and 19F NMR spectra. All chemical shifts are reported in
ppm (δ) and H and 13C spectral data are referenced as
1
follows: δ 5.32 and 54.0 (CD2Cl2); 1.94 and 118.3
(CD3CN); 2.50 and 39.5 (d6-DMSO). Fluorobenzene was
used as an internal standard for all 19F spectra (δ -113.78
in CD2Cl2; -114.81 in CD3CN).[22]
Thin layer chromatography was carried out with
Macherey-Nagel precoated polyester sheets (0.2 mm alu-
mina with a fluorescent indicator) and visualized using a
UV lamp. Purification by MPLC was carried out with a
CombiFlash® Rf automated flash chromatography system
from Teledyne Isco. Inc. on alumina (neutral,
Brockman I, standard grade, 150 mesh, 58 Å) columns.
Uncorrected melting points were observed with a
Thomas Hoover Uni-Melt apparatus in unsealed capil-
laries. FT-IR data for characterization were collected with
a Thermo Scientific Nicolet iS5 spectrometer with an iD5
ATR source and solution measurements were observed
with a liquid cell (1.0 mm path length) with NaCl win-
dows. In all cases, ~5 mM phenol solutions were made in
both dry CD2Cl2 and dry 1% CD3CN/CD2Cl2 (v:v).[9]
Backgrounds of the solvent mixtures were used to correct
the corresponding spectra. Between each run, the cell
was rinsed with CH2Cl2 and then purged with a flow of
2.1.2 | N-Methyl-3-hydroxypyridinium
iodide
In a 25 mL round-bottomed flask fitted with a reflux con-
denser, 3-hydroxypyridine (300 mg, 3.15 mmol) and ace-
tonitrile (6 mL) were stirred and the atmosphere was
purged with argon for 10 min. Iodomethane (200 μL,
456 mg, 3.21 mmol) was rapidly added and the resulting
solution was refluxed overnight. Upon cooling to room
temperature, the reaction mixture was concentrated with

4 of 16
PAYNE AND KASS
a rotary evaporator and subsequently dissolved in 1 mL
of anhydrous methanol. Dropwise addition of the yellow
material into 20 mL of anhydrous ethyl acetate in a
6-dram vial with vigorous stirring under an inert atmo-
sphere afforded a white precipitate. The yellow superna-
tant was removed via syringe and the solid residue was
washed with ethyl acetate in 10 mL portions until the sol-
vent became colorless. A rotary evaporator followed by a
mechanical pump (0.1 torr) were used to dry the product
and 668 mg (89%) of an off-white fluffy solid (mp 112 –
113ꢀC) was obtained. This material was stored in a
2.1.4 | N-Methyl-2-hydroxypyridinium
trifluoromethanesulfonate[25]
In a 6-dram vial under inert atmosphere, N-methyl-
2-pyridone (100 μL, 111 mg, 1.02 mmol) was dissolved
in
1
mL
of
anhydrous
CH2Cl2.
Neat
trifluoromethanesulfonic acid (90 μL, 153 mg,
1.02 mmol) was then added dropwise over 2-3 min at
room temperature. A white precipitate was observed
after 10 min and the suspension was stirred for a total
of 1 h. The reaction mixture was diluted with 10 mL of
pentane and the supernatant was removed via syringe.
The solid residue was washed twice with 10 mL of pen-
tane, and then dried with a rotary evaporator followed
by a mechanical pump to afford 252 mg (95%) of a
shiny, colorless solid (mp 109 – 110ꢀC) that was stored
1
glovebox. H NMR (500 MHz, CD3CN) δ 9.64 (OH, s,
1H), 8.50 (s, 1H), 8.24-8.23 (m, 2H), 7.81 (t, J = 7.8 Hz,
1H), 4.25 (s, 3H). 13C NMR (125 MHz, CD3CN) δ 157.6,
137.7, 134.7, 132.6, 129.2, 49.4. IR (ATR source) 2963,
1581, 1506, 1492 cm–1. HRMS-ESI: calcd for C6H8NO
(M – I–)+ 110.0606, found 110.0586.
1
in a glovebox. H NMR (500 MHz, CD3CN) δ 12.05 (bs,
OH, 1H), 8.20 (t, J = 8.1 Hz, 1H), 8.12 (d, J = 6.6 Hz,
1H), 7.43 (d, J = 8.7 Hz, 1H), 7.29 (t, J = 6.9 Hz, 1H),
3.91 (s, 3H). 13C NMR (125 MHz, CD3CN) δ 161.1,
2.1.3 | N-Methyl-3-hydroxypyridinium
tetrakis(3,5-bis (trifluoromethyl)phenyl)
borate (3Me)
1
147.9, 143.2, 121.5 (q, JF-C = 319 Hz), 118.7, 115.3,
41.6. 19F (376 MHz, CD3CN) δ -79.4. IR (ATR source)
2932, 1644, 1601, 1520, 1287, 1217, 1156, 1023 cm–1.
– +
In
a
6-dram vial under argon, N-methyl-
HRMS-ESI: calcd for C6H8NO (M
110.0606, found 110.0581.
– CF3SO3 )
3-hydroxypyridinium iodide (31.9 mg, 0.135 mmol) and
sodium tetrakis(3,5-bis (trifluoromethyl)phenyl)borate
(132 mg, 0.148 mmol) were added to 3 mL of anhy-
drous CH2Cl2 and vigorously stirred overnight. A white
precipitate was filtered with a 0.45 μm syringe filter
and then washed with 2 mL of additional CH2Cl2. The
combined liquid material was concentrated with a
rotary evaporator and the resulting solid residue was
dissolved in 2 mL of CH2Cl2. This solution was added
dropwise to 15 mL of anhydrous pentane in a 6-dram
vial with stirring. A white solid formed immediately
and was allowed to settle to the bottom of the vial
before the supernatant was removed via syringe. The
solid product was resuspended in 15 mL of pentane,
collected by vacuum filtration, and washed with 15 mL
of pentane. It was then dissolved in 4 mL of CH2Cl2,
dried over MgSO4, and concentrated under reduced
pressure to afford 82.6 mg (63%) of a white solid
2.1.5 | N-Methyl-2-hydroxypyridinium
tetrakis(3,5-bis (trifluoromethyl)phenyl)
borate (2Me)
In
trifluoromethanesulfonate (25.9 mg, 0.100 mmol) and
sodium tetrakis(3,5-bis(trifluoromethyl)phenyl)borate
a
6-dram vial, N-methyl-2-hydroxypyridinium
(99.9 mg, 0.113 mmol) were added and the atmosphere
was purged with argon. Anhydrous CH2Cl2 (3 mL) was
then added and the resulting suspension was vigorously
stirred for 5.5 h. The reaction mixture was filtered
through a 0.45 μm syringe filter and then washed with
2 mL of additional CH2Cl2. The solvent was removed
with a rotary evaporator and the solid reside was dis-
solved in 1 mL of CH2Cl2 and the filtration procedure
was repeated. Further drying was performed with a
mechanical pump to afford 96.2 mg (99%) of a colorless
1
(mp 179 – 180ꢀC). H NMR (500 MHz, CD2Cl2) δ 8.05
(s, 1H), 8.01 (d, J = 8.0 Hz, 1H), 7.89 (d, J = 7.9 Hz,
1H), 7.81-7.80 (m, 1H), 7.76 (s, 8H), 7.59 (s, 4H) 6.39
(OH, s, 1H), 4.26 (s, 3H). 13C NMR (125 MHz, d6-
1
solid (mp 105 – 106ꢀC). H NMR (500 MHz, CD2Cl2) δ
11.62 (bs, OH, 1H), 8.15 (dt, J = 8.1 and 1.4 Hz, 1H), 7.93
(dd, J = 6.6 and 1.3 Hz, 1H), 7.73 (s, 8H), 7.56 (s, 4H),
7.28 (m, 2H), 3.97 (s, 3H). 13C NMR (125 MHz, CD2Cl2) δ
162.3 (q, 1JB-C = 49.5 Hz), 161.0, 147.6, 141.6, 135.4, 129.5
1
DMSO) δ 161.0 (q, JB-C = 50.0 Hz), 156.7, 136.4, 134.1,
2
133.8, 130.9, 128.5 (qq, 3JB-C = 2.7 Hz and JF-
1
= 31.6 Hz), 128.3, 124.0 (q, JF-C = 273 Hz), 117.7
C
(sept, JF-C = 4.1 Hz),[24] 47.8. 19F (376 MHz, CD2Cl2) δ
(qq, JB-C = 2.7 Hz and JF-C = 31.6 Hz), 125.2 (q, JF-
3
3
2
1
3
-62.6. IR (ATR source) 3593, 3510, 1611, 1598, 1516,
= 273 Hz), 118.4, 118.1 (sept, JF-C = 4.1 Hz),[26] 115.5,
C
1354, 1275, 1104 cm–1. HRMS-ESI: calcd for C6H8NO
41.7. 19F (376 MHz, CD2Cl2) δ -62.7. IR (ATR source)
– +
(M – C32H12BF24
)
110.0606, found 110.0583.
3695, 3617, 1646, 1606, 1521, 1353, 1273, 1105 cm–1.

PAYNE AND KASS
5 of 16
– +
1
3
HRMS-ESI: calcd for C6H8NO (M – C32H12BF24
110.0606, found 110.0587.
)
162.3 (q, JB-C = 50.3 Hz), 142.7, 135.4, 129.5 (qq, JB-
2
1
= 2.5 Hz and JF-C = 31.4 Hz), 125.2 (q, JF-
C
C
= 273 Hz), 118.1 (sept, JF-C = 3.8 Hz),[26] 116.1. 19F
3
(376 MHz, CD2Cl2) δ -62.6. IR (ATR source) 3555, 3488,
2.1.6 | 4-Hydroxypyridinium
trifluoromethanesulfonate
3403, 1649, 1608, 1519, 1354, 1274, 1114 cm–1. HRMS-
– +
ESI: calcd for C5H6NO (M – C32H12BF24
)
96.0449,
found 96.0438.
In
a
6-dram vial, 4-hydroxypyridine (52.5 mg,
0.552 mmol) was suspended in 1 mL of anhydrous aceto-
nitrile and the atmosphere was purged with argon for
15 min. Neat trifluoromethanesulfonic acid (100 μL,
170 mg, 1.13 mmol) was then added dropwise over
2-3 min at room temperature resulting in a colorless clear
solution that was stirred for an additional 30 min. The
reaction mixture was diluted with 10 mL of anhydrous
diethyl ether and then 10 mL of pentane was added to
form a white precipitate. The supernatant was removed
by syringe and the solid residue was washed twice with
10 mL of pentane. The product was dried on a rotary
evaporator followed by a mechanical pump at 110ꢀC for
4 h to afford 101 mg (75%) of a light tan solid (mp 111 –
113ꢀC) which was stored in a glovebox. 1H NMR
(500 MHz, CD3CN) δ 12.30 (bt, NH, JN-H = 56.0 Hz, 1H),
10.78 (bs, OH, 1H), 8.39 (d, J = 7.3 Hz, 2H), 7.31 (d,
J = 7.5 Hz, 2H). 13C NMR (125 MHz, CD3CN) δ 172.8,
2.1.8 | N-Methylpyridinium iodide
Following a literature procedure,[27] pyridine (200 μL,
196 mg, 2.48 mmol) was dissolved in 10 mL of acetoni-
trile under argon in a 25 mL round-bottomed flask
equipped with
a
reflux condenser. Iodomethane
(600 μL, 1.37 g, 9.64 mmol) was then added by syringe
and the reaction mixture refluxed overnight. Upon
cooling to room temperature, the yellow solution was
concentrated with a rotary evaporator. The remaining
solid residue was then dissolved in 3 mL of acetonitrile
and added dropwise into stirring anhydrous diethyl
ether under argon (15 mL). A white precipitate formed
and was allowed to settle to the bottom before removing
the yellow supernatant via syringe. The solid was
washed with additional diethyl ether until the superna-
tant was colorless (4 × 10 mL) and then was dried on a
rotary evaporator followed by a mechanical pump to
afford 545 mg (99%) of a light-yellow solid (mp 108 –
110ꢀC). 1H NMR (500 MHz, d6-DMSO) δ 9.00 (d,
J = 5.4 Hz, 2H), 8.59 (t, J = 7.8 Hz, 1H), 8.14 (t, 6.7 Hz,
2H), 4.37 (s, 3H). 13C NMR (125 MHz, d6-DMSO) δ
145.5, 145.0, 127.6, 48.0.
1
143.7, 121.6 (q, JF-C = 319 Hz), 115.1. 19F (376 MHz,
CD3CN) δ -79.4. IR (ATR source) 3245, 3101, 1648, 1612,
1513, 1271, 1249, 1211, 1190, 1171, 1024 cm–1. HRMS-
– +
ESI: calcd for C5H6NO (M – CF3SO3 ) 96.0449, found
96.0437.
2.1.7 | 4-Hydroxypyridinium
tetrakis(3,5-bis (trifluoromethyl)phenyl)
borate (4H)
2.1.9 | N-Methylpyridinium
tetrakis(3,5-bis (trifluoromethyl)phenyl)
borate (PyrMe)
In
a
6-dram
vial,
4-hydroxypyridinium
trifluoromethanesulfonate (20.5 mg, 0.0836 mmol) and
sodium tetrakis(3,5-bis (trifluoromethyl)phenyl)borate
(85.0 mg, 0.0959 mmol) were added and the atmosphere
was purged with argon. Anhydrous CH2Cl2 (2 mL) was
then added and the resulting suspension was vigorously
stirred for 4.5 h. The reaction mixture was filtered
through a 0.45 μm syringe filter and then washed with
2 mL of additional CH2Cl2. The solvent was removed
with a rotary evaporator and the solid reside was dis-
solved in 1 mL of CH2Cl2 and the filtration procedure
was repeated. Further drying was performed with a
mechanical pump to afford 74.5 mg (93%) of a colorless
In a 6-dram vial under argon, N-methylpyridinium iodide
(31.2 mg, 0.141 mmol) and sodium tetrakis(3,5-bis
(trifluoromethyl)phenyl)borate (134 mg, 0.151 mmol)
were added to 3 mL of anhydrous CH2Cl2 and vigorously
stirred for 1 h. A white precipitate was filtered with a
0.45 μm syringe filter and then washed with 2 mL of
additional CH2Cl2. The combined liquid material was
concentrated with a rotary evaporator and the resulting
solid residue was dissolved in 2 mL of CH2Cl2, dried over
MgSO4, and filtered again. Removal of the solvent under
reduced pressured afforded 96.2 mg (71%) of a colorless
solid (mp 205 – 206ꢀC) that had spectra consistent with a
previous report.[28] 1H NMR (400 MHz, CD2Cl2) δ
8.52-8.47 (m, 3H), 8.03 (t, J = 7.0 Hz, 2H), 7.73 (s, 8H),
7.56 (s, 4H), 4.37 (s, 3H). 13C NMR (125 MHz, CD2Cl2) δ
1
solid (mp 162 – 164ꢀC). H NMR (500 MHz, CD2Cl2) δ
10.82 (t, NH, JN-H = 68.0 Hz, 1H), 8.35 (t, J = 6.9 Hz, 2H),
8.26 (bs, OH, 1H), 7.73, (s, 8H), 7.57 (s, 4H), 7.37 (d,
J = 6.4 Hz, 2H). 13C NMR (125 MHz, CD2Cl2) δ 172.4,

6 of 16
PAYNE AND KASS
162.4 (q, 1JB-C = 50.0 Hz), 147.0, 144.8, 135.4, 129.7, 129.5
2.3.1 | Pyridinium tetrakis(3,5-bis
(trifluoromethyl)phenyl)borate (PyrH)[30]
3
2
1
(qq, JB-C = 2.7 Hz and JF-C = 31.6 Hz), 125.2 (q, JF-
3
= 273 Hz), 118.1 (sept, JF-C = 4.1 Hz), 49.7. 19F
C
(376 MHz, CD2Cl2) δ -62.7.
Pyridine (18.0 μL, 17.7 mg, 0.223 mmol) and sodium
tetrakis(3,5-bis (trifluoromethyl)phenyl)borate (202 mg,
0.228 mmol) in 2 mL of CH2Cl2 were used. Following
the general procedure led to 200 mg (95%) of a white
2.2 | Description of HCl gas apparatus[29]
1
solid (mp 194 – 196ꢀC). H NMR (500 MHz, CD2Cl2)
A picture of the reaction setup is given in Figure S1. An
argon line was connected to the top of an addition funnel
(A) where concentrated HCl (1.0 – 2.0 mL) was stored
prior to its slow dropwise addition onto CaCl2 (1 g for
every 1 mL of conc. HCl) in a 3-necked round-bottomed
flask (B). A small drying tube containing 0.5 – 1.0 g of
additional CaCl2 (C) was used to further dry the HCl
prior to entering the 6-dram reaction vial (E) via a 4” nee-
dle (D) that could be inserted into the reaction medium;
bubbling should be visible when an argon flow is applied
to the system. The effluent was then passed through a
neutralization column (F) filled with NaOH pellets to
scavenge any residual HCl. Finally, an oil-containing
bubbler (G) was connected to the end of the apparatus
and was useful for identifying leaks and establishing suit-
able HCl flows.
δ
12.25 (bs, 1H), 8.69-8.64 (m, 3H),[31] 8.13 (t,
J = 7.0 Hz, 2H), 7.73 (s, 8H), 7.57 (s, 4H). 13C NMR
1
(125 MHz, CD2Cl2) δ 162.3 (q, JB-C = 49.9 Hz), 149.9,
3
2
141.4, 135.4, 129.5 (qq, JB-C = 3.0 Hz and JF-
1
= 31.2 Hz), 129.4, 125.2 (q, JF-C = 272 Hz), 118.1
C
(sept, JF-C = 4.0 Hz).[26] 19F (376 MHz, CD2Cl2) δ -62.6.
3
IR (ATR source) 3379, 1609, 1541, 1354, 1274,
1115 cm–1.
2.3.2 | 3-Hydroxypyridinium
tetrakis(3,5-bis (trifluoromethyl)phenyl)
borate (3H)
3-Hydroxypyridine (15.1 mg, 0.159 mmol) and sodium
tetrakis(3,5-bis (trifluoromethyl)phenyl)borate (141 mg,
0.159 mmol) in 1.5 mL of CH2Cl2 were used. Follow-
ing the general procedure led to 137 mg (91%) of an
1
2.3 | General one-pot protonation and
off-white solid (mp 178 – 180ꢀC). H NMR (500 MHz,
anion exchange procedure
CD2Cl2) δ 11.61 (NH, tt, JN-H = 65.0 Hz and
J = 7.2 Hz, 1H), 8.29-8.27 (m, 2H), 8.14-8.12 (m, 1H),
8.00 (t, J = 7.4 Hz, 1H), 7.73 (s, 8H), 7.56 (s, 4H), 6.48
(s, OH, 1H). 13C NMR (125 MHz, d6-DMSO) δ 161.0
In a 6-dram vial, the starting pyridine or pyridone
(1.0 eq.), NaBArF4 (1.0 – 1.2 eq.), and a stir bar were
placed. This reaction vessel was connected to the HCl
generation apparatus (Figure S1) and the entire system
was purged for 5 min with argon. The desired solvent
was then introduced and the resulting suspension was
stirred for an additional 5 min. Insertion of the 4” needle
into the liquid medium (bubbling should be observed)
was followed by charging the addition funnel with con-
centrated HCl and its subsequent slow dropwise addition
onto the CaCl2. Bubbling in the 3-necked round-
bottomed flask was observed immediately and after
2-3 min, the suspension in the 6-dram vial became less
cloudy but did not clear up entirely. Upon completion of
the HCl generation (no further bubbling in the 3-necked
flask), the reaction mixture was stirred for an additional
5 min before it was passed through a 0.45 μm PTFE
syringe filter which was then washed with 2 mL of the
reaction solvent (DCE or CH2Cl2). Removal of the vola-
tiles under reduced pressure afforded a solid residue that
was dissolved in CH2Cl2 and the filtration procedure was
repeated. Thereafter, the product was dried under high
vacuum (0.1 torr) for at least 6 h and then it was stored
in a glovebox.
1
(q, JB-C = 49.5 Hz), 156.2, 134.1, 133.5, 131.5, 130.6,
3
2
128.5 (qq, JB-C = 2.9 Hz and JF-C = 31.5 Hz), 127.7,
124.0 (q, 1JF-C 273 Hz), 117.7 (sept, 3JF-
= 3.6 Hz).[24] 19F (376 MHz, CD2Cl2) δ -62.7. IR
=
C
(ATR source) 3582, 3527, 3379, 1610, 1559, 1355, 1274,
1104 cm–1. HRMS-ESI: calcd for C5H6NO (M
–
– +
C32H12BF24
) 96.0449, found 96.0436.
2.3.3 | 2-Hydroxypyridinium
tetrakis(3,5-bis (trifluoromethyl)phenyl)
borate (2H)
2-Hydroxypyridine (11.9 mg, 0.125 mmol) and sodium
tetrakis(3,5-bis (trifluoromethyl)phenyl)borate (112 mg,
0.126 mmol) in 1.5 mL of CH2Cl2 were used. Follow-
ing the general procedure led to 91.4 mg (76%) of a
1
white powder (mp 113 – 115ꢀC). H NMR (500 MHz,
CD2Cl2) δ 10.77 (NH, t, JN-H = 60 Hz, 1H), 8.86 (bs,
OH, 1H), 8.40 (t, J = 7.8 Hz, 1H), 8.09 (bs, 1H), 7.73
(s, 8H), 7.56 (s, 4H), 7.50 (t, J = 7.0 Hz, 1H), 7.37 (d,
J = 7.4 Hz, 1H). 13C NMR (125 MHz, CD2Cl2) δ 162.3

PAYNE AND KASS
7 of 16
1
(q, JB-C = 49.9 Hz), 161.9, 151.4, 137.4, 135.3, 129.5
2.4 | General procedure for dimerization
determinations with 1H NMR spectroscopy
3
2
1
(qq, JB-C = 3.0 Hz and JF-C = 31.4 Hz), 125.1 (q, JF-
= 273 Hz), 120.2, 118.1 (sept, JF-C = 4.0 Hz),[26]
3
C
115.4. 19F (376 MHz, CD2Cl2) δ -62.7. IR (ATR source)
A solution of known concentration consisting of 3Me,
3H, 4H, and 2MeOPyrH was made in CD2Cl2 and an
aliquot (0.5 mL) was transferred into a 9” NMR tube. An
initial spectrum was obtained before additional CD2Cl2
(0.1-0.5 mL) was added and mixed by inverting the NMR
tube twice. Another spectrum was recorded and this pro-
cess was repeated multiple times before carrying out a
non-linear fit of the data using the BindFit program (see
below) to obtain the corresponding dimerization
constants.
3542, 3369, 1649, 1633, 1610, 1543, 1354, 1273,
1112 cm–1. HRMS-ESI: calcd for C5H6NO (M
–
– +
C32H12BF24
) 96.0449, found 96.0440.
2.3.4 | N-Methyl-4-hydroxypyridinium
tetrakis(3,5-bis (trifluoromethyl)phenyl)
borate (4Me)
N-Methyl-4-pyridone (11.3 mg, 0.104 mmol) and sodium
tetrakis(3,5-bis (trifluoromethyl)phenyl)-borate (93.3 mg,
0.105 mmol) in 2 mL of DCE were used. Following the
general procedure led to 82.3 mg (82%) of a white powder
(mp 148 – 150ꢀC); residual DCE was removed by
redissolving the material in 1 mL of CH2Cl2 and remov-
ing the solvent with a rotary evaporator (20 torr), and
then repeating this process at least 3 times. 1H NMR
(500 MHz, CD2Cl2) δ 8.80 (OH, bs, 1H), 8.10 (d,
J = 6.6 Hz, 2H), 7.73 (s, 8H), 7.56 (s, 4H), 7.25 (d,
J = 6.5 Hz, 2H), 4.10 (s, 3H). 13C NMR (125 MHz,
2.5 | General kinetic study procedure for
the Friedel-crafts reaction of N-
methylindole with trans-β-nitrostyrene
In a 1 mL volumetric flask, 74.6 mg (0.500 mmol) of
trans-β-nitrostyrene, 6.2 μL (6.5 mg, 0.050 mmol) of N-
methylindole, and the catalyst (0.005 mmol) were added
and dissolved in CD2Cl2 to a total volume of 1 mL. The
flask was stoppered, and inverted twice to ensure good
mixing, and then the entire contents were transferred to
an NMR tube which was capped and sealed with electri-
cal tape. A 1H NMR spectrum was taken as soon as possi-
ble (within 10 minutes of mixing) and when the sample
was not in the NMR spectrometer, it was submerged in a
27ꢀC water bath. Reaction conversions were calculated
using the signals at 6.47 and 5.20 – 4.96 ppm for N-
methylindole and the product, respectively, and a
pseudo-first-order kinetic model was employed.
1
CD2Cl2) δ 170.5, 162.3 (q, JB-C = 49.5 Hz), 146.1, 135.3,
3
2
129.5 (qq, JB-C = 3.2 Hz and JF-C = 31.6 Hz), 125.1 (q,
1JF-C = 273 Hz), 118.1 (sept, JF-C = 4.1 Hz),[26] 116.4,
3
47.7. 19F (376 MHz, CD2Cl2) δ -62.7. IR (ATR source)
3562, 3495, 1649, 1610, 1533, 1502, 1353, 1272, 1104 cm–1.
– +
HRMS-ESI: calcd for C6H8NO (M – C32H12BF24
)
110.0606, found 110.0587.
2.3.5 | 2-Methoxypyridinium
tetrakis(3,5-bis (trifluoromethyl)phenyl)
borate (2MeOPyrH)
2.6 | General procedure for UV
titrations8, 10d
2-Methoxypyridine (12.5 μL, 13.0 mg, 0.119 mmol) and
sodium tetrakis(3,5-bis (trifluoromethyl)phenyl)borate
(115 mg, 0.130 mmol) in 2.0 mL of CH2Cl2 were used.
Following the general procedure led to 112 mg (97%) of a
2.6.1 | Solution preparation
In a 1 mL volumetric flask, 1.0-1.5 mg of the sensor was
dissolved in CH2Cl2 to the line in the glassware. In a
5 mL volumetric flask, 25 μL of the 1 mg/mL sensor solu-
tion was added and then was diluted to the line with
CH2Cl2 (solution A). In a separate 5 mL volumetric flask
containing 20-30 equivalents of the Brønsted acid,
another 25 μL of the 1 mg/mL sensor solution was added
and diluted to the line with CH2Cl2 (solution B).
1
white powder (mp 163 – 165ꢀC). H NMR (500 MHz,
CD2Cl2) δ 8.46 (ddd, J = 9.2, 7.4, and 1.8 Hz, 1H), 8.10
(dd, J = 6.2 and 2.1 Hz, 1H), 7.73 (s, 8H), 7.56 (s, 4H),
7.49 (t, J = 6.8 Hz, 1H), 7.35 (d, J = 9.0 Hz, 1H), 4.21 (s,
3H), missing NH. 13C NMR (125 MHz, CD2Cl2) δ 162.3
1
(q, JB-C = 50.0 Hz), 161.2, 151.4, 138.4, 135.4, 129.5 (qq,
2
1
3JB-C = 3.2 Hz and JF-C = 31.6 Hz), 125.1 (q, JF-
= 273 Hz), 119.9, 118.1 (sept, JF-C = 4.1 Hz),[26] 111.3,
3
C
59.7. 19F (376 MHz, CD2Cl2) δ -62.7. IR (ATR source)
2.6.2 | Titration procedure
3372, 1644, 16118, 1545, 1353, 1273, 1113 cm–1. HRMS-
– +
ESI: calcd for C6H8NO (M – C32H12BF24
)
110.0606,
In a 10 mm cuvette with a screwcap top, 2 mL of
solution A was added and the vessel was sealed with a
found 110.0585.

8 of 16
PAYNE AND KASS
PTFE septum. After collecting the background of only
CH2Cl2, a spectrum of the free sensor from 300 to
950 nm was obtained and the λmax and the absorbance
value at λmax were recorded. Then an aliquot
(10-15 μL) of solution B was added by microsyringe to
the cuvette which was then shaken for 15-20 seconds,
and then another spectrum was collected. The new
λmax and its corresponding absorbance value were
recorded and cuvette was shaken again (15-20 seconds)
and the spectrum was recollected to ensure that λmax
and the absorbance at λmax were invariant. If signifi-
cant changes were observed (greater than 1 for λmax
3 | RESULTS AND DISCUSSION
3.1 | Model catalysts syntheses
N-Methyl-3-hydroxypyridinium BArF (3Me) was pre-
4
pared by reacting 3-hydroxypyridine with methyl iodide
to afford N-methyl-3-hydroxypyridinium iodide, which
was then converted to the BArF salt by anion exchange
4
with NaBArF in dichloromethane (Scheme 2).[36] Alkyl-
4
ation of 2- and 4-hydroxypyridine in a similar manner
afforded a mixture of N- and O-alkylated products due to
the tautomeric nature of the starting substrates,[37,38] and
we were unable to obtain dry samples of the separated
species. Protonation of the corresponding N-methyl-
pyridones, consequently was explored. Aqueous acids
(HI and HCl) led to the desired halide salts but problems
were encountered in trying to remove water from these
compounds. To circumvent this issue, a one pot proton-
ation and anion exchange procedure was developed.
Anhydrous HCl gas was produced in situ[29] and bubbled
through a dichloromethane or 1,2-dichloroethane solu-
or
0.010 for the absorbance), the mixture was
shaken again and another spectrum was collected until
the data was consistent (in general, no more than
4 repetitions were needed). Then another portion
(10-15 μL) of solution B was added and the collection
procedure was repeated. As the titration went along,
the amount of titrant (solution B) could be increased
(20 – 500 μL) until the observed λmax and the absor-
bance value did not change after
2 consecutive
additions.
tion containing the pyridone of interest and NaBArF .
4
This successfully led to 4Me but a minor byproduct that
resisted purification arose in the preparation of 2Me; we
speculate that this impurity is an isomer of 2Me and cor-
responds to N-protonation of N-methyl-2-pyridone. A sat-
isfactory sample of the latter compound was generated
via its triflate salt, which is easier to handle and purify
than the corresponding halides, by reacting N-methyl-
2-pyridone with neat triflic acid and taking advantage of
2.6.3 | Calculating the binding constants
Absorbance values from 4 different wavelengths and the
equivalents of Brønsted acid to sensor were used to calcu-
late the binding constant (K). The λmax of the free sensor
(~500 nm) and of the bound complex, and wavelengths
30 nm above (~530 nm) and below those values were
used in the analysis. These data points were nonlinearly
fit with the BindFit app (http://app.supramolecular.org/
bindfit/) for 1:1, 1:2, and 2:1 sensor to catalyst relation-
ships to obtain the binding constants and the statistical
errors.
the differential solubility of NaOTf and NaBArF in dic-
4
hloromethane. The one pot procedure with gaseous HCl
was also used to synthesize protonated pyridinium salts
2H, 3H, and PyrH, but not 4H since 4-hydroxypyridine
is insufficiently soluble in CH2Cl2. In this latter case, 4H
was prepared from its triflate salt in an analogous man-
ner to 2Me.
2.7 | Computations
DFT (B3LYP/6-31G(2df,p), B3LYP/cc-pVDZ, B3LYP/cc-
pVTZ//B3LYP/cc-pVDZ, M06-2X/cc-pVDZ and M06-2X/cc-
pVTZ//M06-2X/p-VDZ)[32,33] and G4-theory[19] computations
were carried out using Gaussian 16 at the Minnesota
Supercomputer Institute for Computational Chemistry or
on a desktop MacIntosh computer with GaussView 6.[34,35]
All structures were fully optimized and correspond to
minima on the potential energy surface (i.e., all of the
vibrational frequencies correspond to positive values).
Geometries, energies and select vibrational frequencies are
provided in the Supporting Information.
SCHEME 2 Synthetic routes for all three isomeric N-
methylhydroxypyridinium BArF4 salts

PAYNE AND KASS
9 of 16
3.2 | IR studies
positions are calculated to be the more acidic sites, but
the differences are small for 2H and 4H (i.e., 0.9 and
3.5 kcal mol–1, respectively) due to tautomer formation
upon O-deprotonation in these cases. In contrast, a
zwitterion results from 3H and the G4 acidity difference
is much larger (i.e., 15.1 kcal mol–1). A different acidity
ordering of 2H > 3H > 4H is found, and is surprising
since one would expect the relative stabilities of these
ions (i.e., 4H (0) > 2H (0.5) > 3H (6.0 kcal mol–1)) to be
inversely related to their acidities. 2-Hydroxypyridine is
9.5 kcal mol–1 more stable than 3-hydroxypyridine,
however, due to a favorable O–H••N electrostatic inter-
action. This energy difference is larger than that
between 2H and 3H, and is sufficient to reverse their
acidities and account for why 2H is the more acidic
compound.
Both the N–H and O–H stretching frequencies for
2H-4H in the presence and absence of CD3CN are given
in Table 1. Computed values for all of the acids listed in
entries 1-9 along with their acetonitrile complexes are
provided in the Supporting Information, and the former
and latter values are linearly correlated with the observed
stretching frequencies (Figures S2 and S3, respectively).
As expected, Δυ for both the N–H and O–H stretches cor-
relate with the gas-phase acidities of these positions
although the linear correlation is much tighter for the
more acidic site (Figure 2).
IR spectroscopy was used to assess the acidities and
hydrogen bond donating abilities of the N-methylated
phenols in a nonpolar solvent. These salts are insuffi-
ciently soluble in carbon tetrachloride, the solvent that
was previously utilized,[9] so dichloromethane-d2 was
used instead. The free and hydrogen bound O–H
stretching frequencies for 2Me, 3Me, 4Me, phenol, and
4-nitrophenol were obtained in CD2Cl2 and a 1% (v/v)
mixture of CD3CN in CD2Cl2 (Table 1). Gas-phase acidi-
ties (ΔGꢀacid) of these compounds were computed at
298 K using the highly accurate G4 composite approach
and are also given in Table 1.
As shown in entries 1-5 in Table 1, a larger Δυ was
observed for more acidic compounds with smaller
ΔGꢀacid values. For example, phenol has a gas-phase acid-
ity of 341.5 kcal mol–1 and afforded a difference in the O–
H
stretch of 170 cm–1 (entry 1), whereas
ΔGꢀ
= 227.0 kcal mol–1 for 4Me and a frequency
acid
change of 398 cm–1 (entry 4) was found. The same trend
was observed with the free O-H stretches with the possi-
ble exception of 2Me, but the range was approximately
four times smaller (i.e., 435 vs 102 cm–1 or ~4 vs
1 mol cm–1 kcal–1 given that the acidities span
114.5 kcal mol–1 from 341.5 to 227.0 kcal mol–1).[40]
Two acidic sites are present in the protonated series
of hydroxypyridines, and as expected, the O–H acidities
for 2H-4H follow the same trend as 2Me-4Me (i.e., 2Me
(most acidic) > 4Me > 3Me (least acidic)). The N–H
This is not surprising since one would expect the ini-
tial acetonitrile to bind at the more acidic N–H position
and only subsequently coordinate at the O–H site.
TABLE 1 IR O–H and N-H stretching frequencies and gas-phase acidities of a series of phenol and pyridinium derivativesa
Cmpd
υ (cm-1)b
Entry
CD2Cl2
3582
1% CD3CN
3412
Δυ (cm-1)b,c
170 [157]
229 [221]
524
ΔGꢀacid (kcal mol–1)b,d
341.7
1
2
3
4
5
6
7
8
9
C6H5OH
4-O2NC6H4OH
2Me
3559
<3501>e
3330
320.6
2977
215.5
3Me
3513
3179
334 [370f]
234.3
4Me
3480
3082
398
227.0
(214.9)h
PyrH
2Hg
3Hg
(3307)
(2903)
(306)
<3499>e(3316)
3516 (3308)
3477 (3348)
3071 (2962)
3193 (2902)
3249 (3107)
428 (354)
323 (300)
228 (241)
211.9 (211.0)
230.1 (215.0)
222.2 (218.7)
4Hg
aSpectra collected using ~5 mM solutions of each compound in a 1 mm liquid cell.
bParenthetical numbers are for N–H stretches or acidities.
cValues in brackets were obtained in CCl4 and taken from ref. [9].
dComputed G4 theory values. Experimental acidities of 341.5 (PhOH), 320.9 (4-O2NC6H4OH), 213.9 (2Me), 214.9 (PyrH), and 214.6 (3H) kcal mol–1 have been
reported; see ref. [39].
eThe O–H stretch was not observed and the given value is a scaled (0.923) B3LYP/6-31G(2df,p) vibrational frequency.
fThis value is for N-octyl-3-hydroxypyridinium BArF
.
4
gFree energies for the hydroxypyridines are computed to be lower in energy than the corresponding pyridones (a zwitterion in the case of 3H) by 0.93 (2H),
3.46 (4H), and 15.1 (3H) kcal mol–1
.

10 of 16
PAYNE AND KASS
behavior was found for 2Me, 2H, and 4H in that an
apparent isosbestic point is seen only when up to about
1 equivalent of the phenol is added (Figure 4 and Fig-
ures S14 and S16).[41] It disappears upon further addition
of the substrate due to a continuous blue shift that occurs
throughout the titration. This makes it difficult to obtain
λmax for the 1:1 complex as there is no clear endpoint in
the titration. A new band also grows in at higher concen-
trations of the hydrogen bond donor (e.g., the feature
between 450 and 600 nm in Figure 4), and this led to dif-
ferent colored solutions for 2Me and 2H.[42] Taken
together, these observations indicate that more than one
species is formed in these three cases, and two or more
equilibria are involved.
FIGURE 2 Changes in the N–H (open circles) and O–H (filled
circles) IR frequencies of pyridinium BArF – salts in CD2Cl2 upon
4
addition of CD3CN versus the corresponding gas-phase G4 acidities
of the free ions. Linear least squares fits provide the following
equations: Δυ (cm–1) = -14.7 × ΔGꢀacid + 3451, r2 = 0.991 (N–H
acids) and Δυ (cm–1) = -6.87 × ΔGꢀacid + 1939, r2 = 0.662 (O–H
acids with 4H omitted and not shown)
To account for all of the above results, 1:1, 1:2, and
2:1 interactions with 1 were considered for each of the
pyridinium salts (Scheme 3). In the 1:2 complex, the two
acids presumably interact with the carbonyl oxygen and
the β-nitrogen atom. B3LYP/cc-pVTZ//B3LYP/cc-pVDZ
and M06-2X/cc-pVTZ//M06-2X/cc-pVDZ computations
on 1•••PhOH are in accord with this view in that coordi-
nation at both of these sites are similar in energy. That is,
the enthalpy at 298 K for binding at the carbonyl oxygen
as opposed to the β-nitrogen is predicted to be favored by
2.11 and 1.37 kcal mol–1, respectively. Alternatively, two
molecules of 1 can interact with a single substrate
(i.e., 2:1 binding) if multiple hydrogen bonding sites are
present in the catalyst, such as the N-H and O-H bonds
in 2H. For 2Me, we speculated that the hydroxyl group
and carbon–hydrogen bonds of the methyl substituent
can serve as hydrogen bond donors.[43] To assess this pos-
3.3 | UV-Vis studies
An alternative approach for assessing acids in non-polar
solvents is to titrate them with a colorimetric hydrogen
bond accepting sensor (1), and monitor the resulting UV-
vis spectra to obtain changes in the absorption maxima
10d
along with 1:1 association binding constants.8,
An
updated procedure was developed involving a non-linear
fit of the binding isotherm data, and well-defined iso-
sbestic points were observed with 3Me, 4Me, 3H, and
PyrH. For example, over the course of adding 10 equiva-
lents of 3H to 1 (Figure 3) an isosbestic point at 484 nm
is observed and λmax of the bound complex at 471 nm is
readily obtained. These observations are consistent with
a single intermediate being formed between the Brønsted
acid and UV/vis sensor (i.e., 1:1 binding). Different
sibility, N- methylpyridinium BArF (PyrMe) was pre-
4
pared and examined. Its spectra were well behaved and a
clean isosbestic point was observed indicating a 1:1 asso-
ciation with 1. This strongly suggests that the carbon-
FIGURE 3 UV-vis spectra from 300 to 600 nm for the titration
of 1 with 3H. For clarity only half of the spectra are shown and
they correspond to the addition of 0.00, 0.35, 0.69, 1.02, 1.67, 2.89,
5.59, and 10.7 equivalents of 3H
FIGURE 4 UV-vis spectra from 300 to 600 nm for the titration
of 2Me with 1. For clarity only half of the spectra are shown and
they correspond to the addition of 0.00, 0.22, 0.44, 0.65, 0.76, 1.00,
1.50, 2.50, 5.80, 11.0, and 22.0 equivalents of 2Me

PAYNE AND KASS
11 of 16
SCHEME 3 Proposed binding modes for
Brønsted acids such as 2Me with 1
hydrogen bonds are sufficiently polarized to lead to a
binding interaction. Equilibrium association constants
were determined using the free online BindFit program
and are given in Table 2.[44,45] As expected, those com-
pounds that displayed clean isosbestic points throughout
the titration are well fit by a 1:1 association model. The
remaining three salts are best fit by 2:1 (2Me and 2H)
and 1:2 (4H) binding schemes, although the latter sce-
nario can be viewed as a 1:1 interaction of 1 with the
dimerization equilibrium constants of 3.3 × 102 and
1.2 × 102 M–1, respectively.[46] For both the methylated
and protonated series of pyridinium ions the binding
equilibrium constants follow the same order: 2 > 4 > 3
and ln K is found to linearly correlate with ΔGꢀ
for
acid
the six compounds that were fit by 1:1 or 1:2 binding
models (Figure 5). The two ortho species (2H and 2Me)
are best described by 2:1 associations and their K1:1
values are smaller than expected given their computed
acidities.
1
dimer of 4H. To explore this possibility, H NMR spectra
of 4H were recorded and the OH, NH, and aryl hydrogen
chemical shifts were found to move downfield with
increasing concentration. A nonlinear fit of these results
afforded a monomer-dimer equilibrium constant (KD) of
1.3 × 103 M–1. Given this value, >99.99% of 4H exists as a
monomer at the mid-concentration used in the UV-vis
titration with 1. In contrast, 3H and 3Me (which afforded
simple 1:1 behavior) were found to have smaller
3.4 | Kinetics
Rate constants and reaction half-lives for the Friedel-
Crafts alkylation of N-methylindole with trans-β-
nitrostyrene were obtained under pseudo-first order con-
ditions (Table 3).[47] In the absence of a catalyst (entry 1),
this transformation takes place extremely slowly and has
an estimated half-life of 3700 h or 155 days. Upon
TABLE 2 Wavelength shifts and binding constants from
UV-vis titrations with 1
Binding
model
b
Entry
Catalyst
2Mec
3Me
λmax1:1 (nm)a,b
410.7
K1:1 (M–1
)
1
2
3
4
5
6
7
8
2:1
3.47 × 104
2.01 × 104
1.23 × 105
6.50 × 102
1.78 × 105
4.71 × 104
6.05 × 104
8.51 × 104
1:1
472.1 (471.7)
466.5 (465.6)
474.9 (474.9)
470.3 (469.3)
408.9
4Me
1:1
PyrMe
PyrH
2Hd
1:1
1:1
2:1
3H
4He
1:1
470.6 (470.3)
463.5
1:2
aWavelengths for the absorption maxima of the 1:1 complexes were obtained
from a plot of χ/λmax at each titration point, where χ is the 1:1 mole fraction
resulting from the fit of the data; for additional details, see the Supporting
Information. Parenthetical values correspond to observed λmax endpoints.
bAverages of duplicate measurements.
FIGURE 5 Linear least squares fit of the logarithms of the 1:1
equilibrium association constants between 1 and the pyridinium
salts listed in Table 2 versus the G4 acidities of the pyridinium ions.
The equation for the line is ln K = -0.118 × ΔGꢀacid + 37.2,
r2 = 0.882, and the values for 2H and 2Me (open circles) were
omitted from the regression
cK2:1 = 1.44 × 104 M–1
dK2:1 = 8.42 × 103 M–1
eK1:2 = 1.23 × 104 M–1
.
.
.

12 of 16
PAYNE AND KASS
TABLE 3 Kinetic data for a Friedel-crafts reaction between N-Methylindole and trans-β-Nitrostyrene
Entry
Catalyst
t1/2 (h)a
3700
k (h–1 a
)
krel
1
2
-
1.86 × 10-4
2.73 × 10-1
0.0022
3.3
2.5
3
4
8.3
4.2
8.33 × 10-2
1.0
2.0
1.67 × 10-1
5
6
790
8.89 × 10-4
0.011
2.7
2.58 × 10-1
3.2
(Continues)
employing 10 mol% of 2Me–4Me (entries 2-4), rate con-
stants spanning from 0.083 to 0.27 h–1 with half-lives
between 2.5 and 8.3 h were observed. This leads to a reac-
tivity order of 2Me > 4Me > 3Me, which is in accord
with the IR data, the G4 acidities, and the UV-vis λmax
values for the 1:1 complexes noted above. The magni-
tudes of the 1:1 equilibrium binding constants for 2Me
and 4Me, however, are reversed. To verify that 2Me–

PAYNE AND KASS
13 of 16
TABLE 3 (Continued)
Entry
Catalyst
t1/2 (h)a
k (h–1 a
)
krel
7c
0.47
1.49
18
8
9
2.9
3.7
2.38 × 10-1
2.9
2.2
1.83 × 10-1
10
5.9
1.17 × 10-1
1.4
aAverage of duplicate trials.
bBackground corrected data.
cT = 25ꢀC.
4Me are functioning as O–H hydrogen bond donating
catalysts, PyrMe was also examined (entry 5) since the
UV-vis data revealed that it associates with 1 even though
the hydroxyl group is absent. This species does catalyze
the Friedel-Crafts reaction relative to the background
process even though all of the hydrogens are attached to
carbon atoms, but it is ~100–300 times less effective than
2Me–4Me. This is in accord with the UV-vis results
(i.e., λmax and K) and is consistent with the N-
methylhydroxypyridinium ions serving as O–H hydrogen
bond donors.
For the protonated series of hydroxypyridinium
BArF – salts (entries 6-9), the reactivity order is
4
2H > PyrH, 3-H > 4H and the half-lives range from
under 30 minutes to 4 hours. Both PyrH and 3H have
similar catalytic activities, and this suggests that the latter
species also acts as a N-H hydrogen bond donor rather
than an O–H Brønsted acid. The same conclusion can be

14 of 16
PAYNE AND KASS
drawn for 2H and 4H since the the protonated catalysts
lead to faster reactions than their corresponding N-
methylated analogues (i.e., 2H > 2Me, 3H > 3Me and
4H > 4Me). Interestingly, 2H is significantly more active
than all of the other catalysts (entry 7), which implies
that bidentate activation involving both the N–H and O–
H sites occurs in this case. To assess this possibility,
modes. Neither λmax or the 1:1 equilibrium binding con-
stants from the UV-vis titration experiments with 1, how-
ever, are good measures of the catalyst activities. This
break-down in the latter approach presumably is related
to the higher-order binding obtained with 2H and 4H. A
plot of the logarithm of the experimental reaction rate
constants versus the computed G4 gas-phase acidities of
the pyridinium ions provides a reasonable linear correla-
tion for all nine catalysts listed in Table 3 (Figure 6) even
though these compounds include N–H, O–H, and C–H
hydrogen bond donors and the counteranion is not
accounted for in the calculations.
2-methoxypyridinium BArF (2MeOPyrH) was prepared
4
from 2-methoxypyridine (Scheme 4)[48] following the one
pot protonation and anion exchange method described
above. This compound is electronically similar to 2H and
4H, but is missing the O–H hydrogen bond donating site.
Its catalytic ability (entry 10) is comparable to 4H (half-
lives of ~6 and 4 h, respectively), but is more than an
order of magnitude less than that for 2H. Given that the
dimerization equilibrium constant determined by 1H
NMR for 2MeOPyrH is only 22 M-1 and a factor of
10-100 times smaller than for 3H, 3Me, and 4H,[49] it is
likely that 2H employs both N–H and O–H hydrogen
bonds to activate trans-β-nitrostyrene for nucleophilic
attack by N-methylindole.
4 | CONCLUSIONS
All three isomers of both the N-methylated and N-
protonated hydroxypyridinium BArF – salts along with
4
several reference compounds were investigated. DFT and
G4 theory computations, IR, UV-vis, and kinetic studies
revealed the effects of changing the relative position of
the charged center to the ionization site, and the conse-
quences of having one (O–H) or two (O–H and N–H)
hydrogen bond donating groups. Calculated gas-phase
acidities of the corresponding cations without their
In an analogous fashion to the N-methylated
pyridininium ions, the reaction rate data for the N-
protonated catalysts correlate with the IR vibrational fre-
quency changes (Δυ) in the N–H and O–H stretching
BArF – counteranion were found to be good predictors of
4
the hydrogen bond donating abilities of these salts, and
are linearly correlated with IR and UV-vis spectroscopic
results and pseudo-first-order rate constants. As a result,
changes in the IR stretching frequencies of the O–H and
N–H bonds in the absence and presence of a hydrogen
bond acceptor (Δυ) are reliable indicators of their relative
reactivity orders (i.e., O–H acids and separately N–H
acids). In a similar manner, 1:1 equilibrium association
constants obtained by UV-vis monitored titrations with a
colorimetric sensor (1) are good measures of the catalytic
abilities of O–H (3Me, 4Me), N–H (PyrH, 3H, 4H) and
C–H (PyrMe) Brønsted acids. This correlation breaks
down, however, for 2Me and 2H, both of which failed to
give an isosbestic point and were found to associate with
two molecules of 1 (i.e., 2:1 binding).
SCHEME 4 Synthesis of control compound 2MeOPyrH
Alkyl pyridinium ions with the charge center meta to
the ionization site have been the most widely employed
charge-enhanced structures to come from our laboratory.
These species avoid introducing an additional O–H or N–
H hydrogen bond donating site or resonance delocaliza-
tion effects. The latter interactions as reflected by com-
parisons of 2Me and 4Me relative to 3Me were found to
lead to catalysts that are several times more reactive in
the Friedel-Crafts alkylation of N-methylindole with
trans-β-nitrostyrene. Introduction of the charged-center
by protonation leads to more acidic compounds, alters
the interaction site with the substrate, and affords
FIGURE 6 Logarithm of experimental rate constants vs
computed G4 gas-phase acidities. A linear least squares fit using all
9 catalysts in Table 3 gives ln k = –0.126 × ΔGꢀacid + 28.6,
r2 = 0.875 whereas omission of 2MeOPyrH leads to ln k =
–0 .131 × ΔGꢀacid + 29.9, r2 = 0.912 (not shown). Circles,
diamonds, and triangles are used for 2Me–4Me (O–H acids), 2H–
4H, PyrH, and 2MeOPyrH (N–H acids), and PyrMe (C–H acids),
respectively

PAYNE AND KASS
15 of 16
[7] a) F. G. Bordwell, Acc. Chem. Res. 1988, 21, 456. b)
S. Tshepelevitsh, A. Kütt, M. Lökov, I. Kaijurand, J. Saame,
A. Heering, P. G. Plieger, R. Vianello, I. Leito, Eur. J. Org.
Chem. 2019, 2019, 6735. c) M. Lökov, S. Tshepelevitsh,
A. Heering, P. G. Plieger, R. Vianello, I. Leito, Eur. J. Org.
Chem. 2017, 2017, 4475.
[8] a) P. N. H. Huynh, R. R. Walvoord, M. C. Kozlowski, J. Am.
Chem. Soc. 2012, 134, 15621. b) R. R. Walvoord,
P. N. H. Huynh, M. C. Kozlowski, J. Am. Chem. Soc. 2014, 136,
16055.
modestly more active catalysts. The one exception is 2H,
which due to the ortho arrangement of its two acidic sites
presumably leads to bidentate activation, and resulted in
a significantly larger rate enhancement. This type of
structural modification can lead to more rigid complexes
and transition state structures providing a means for
more closely mimicking enzymes and designing more
reactive charge-activated catalysts.
[9] M. Samet, J. Buhle, Y. Zhou, S. R. Kass, J. Am. Chem. Soc.
2015, 137, 4678.
NOTES
The authors declare no competing financial interest.
[10] a) Y. Fan, S. R. Kass, Org. Lett. 2016, 18, 188. b) Y. Fan,
S. R. Kass, J. Org. Chem. 2017, 82, 13288. c) Y. Fan, M. Tiffner,
J. Schorgenhumer, R. Robiette, M. Waser, S. R. Kass, J. Org.
Chem. 2018, 83, 9991. d) Y. Fan, C. Payne, S. R. Kass, J. Org.
Chem. 2018, 83, 10855.
[11] a) J. Ma, S. R. Kass, Org. Lett. 2016, 18, 5812. b) J. Ma,
S. R. Kass, Org. Lett. 2018, 20, 2689. c) J. Ma, S. R. Kass, J. Org.
Chem. 2019, 84, 11125.
[12] A para-N-alkylated thiourea has been studied (see ref. 10d)
and phosphonium phosphoric acids have been reported (see
refs. 11b and 11c).
[13] a) T. Schuster, M. Kurz, M. W. Gobel, J. Org. Chem. 2000, 65,
1697. b) D. Akalay, G. Durner, J. W. Bats, M. Bolte,
M. W. Gobel, J. Org. Chem. 2007, 72, 5618.
ACKNOWLEDGEMENTS
We would like to thank Ms. Grace Gast for her prelimi-
nary investigations on this project, Mr. Serdar Yalvac for
his contributions toward developing the UV-vis titration
procedure, and Prof. Phillipe Buhlmann for helpful dis-
cussions and his guidance regarding titrations that result
in multiple binding equilibria. Generous support from
the National Science Foundation (CHE-1665392) and the
Minnesota Supercomputer Institute for Advanced Com-
putational Research are also gratefully acknowledged.
[14] a) E. J. Corey, M. J. Grogan, Org. Lett. 1999, 1, 157. b)
M. Terada, H. Ube, Y. Yaguchi, J. Am. Chem. Soc. 2006, 128,
1454. c) M. Terada, M. Nakano, H. Ube, J. Am. Chem. Soc.
2006, 128, 16044. d) C. Uyeda, E. N. Jacobsen, J. Am. Chem.
Soc. 2008, 130, 9228. e) D. Leow, C.-H. Tan, Synlett 2010, 2010,
1589. f) C. Uyeda, E. N. Jacobsen, J. Am. Chem. Soc. 2011, 133,
5062. g) P. Selig, Synthesis 2013, 45, 703. (h) M. P. Coles, Chem.
Commun. 2009, 25, 3659.
[15] J. Huang, E. J. Corey, Org. Lett. 2004, 6, 5027.
[16] a) N. Takenaka, R. S. Sarangthem, S. K. Seerla, Org. Lett.
2007, 9, 2819. b) N. Takenaka, J. Chen, B. Captain,
R. S. Sarangthem, A. Chandrakumar, J. Am. Chem. Soc.
2010, 132, 4536. c) Y. Nishikawa, S. Nakano, Y. Tahira,
K. Terazawa, K. Yamazaki, C. Kitamura, O. Hara, Org. Lett.
2016, 18, 2004.
ORCID
Curtis Payne https://orcid.org/0000-0001-9100-2375
Steven R. Kass https://orcid.org/0000-0001-7007-9322
REFERENCES
[1] a) K. B. Schowen, H.-H. Limbach, G. S. Denisov,
R. L. Schowen, Biochim. Biophys. Acta 2000, 1458, 43. b)
L. Simon, J. M. Goodman, J. Org. Chem. 2010, 75, 1831. c)
D. Röthlisberger, O. Khersonsky, A. M. Wollacott, L. Jiang,
J. DeChancie, J. Betker, J. L. Gallaher, E. A. Althoff,
A. Zanghellini, O. Dym, S. Albeck, K. N. Houk, D. S. Tawfik,
D. Baker, Nature 2008, 453, 190.
[2] R. R. Knowles, E. N. Jacobsen, Proc. Natl. Acad. Sci. U. S. A.
2010, 107, 20678.
[3] a) A. G. Doyle, E. N. Jacobsen, Chem. Rev. 2007, 107, 5713. b)
F. Giacalone, M. Gruttadauria, P. Agrigento, R. Noto, Chem.
Soc. Rev. 2012, 41, 2406. c) R. C. Wende, P. R. Schreiner, Green
Chem. 2012, 14, 1821. d) R. J. Phipps, G. L. Hamilton,
F. D. Toste, Nat. Chem. 2012, 4, 603.
[4] a) J. Hine, S.-M. Linden, V. M. Kanagasabapathy, J. Am.
Chem. Soc. 1985, 107, 1082. b) K. H. Jensen, M. S. Sigman,
J. Org. Chem. 2010, 75, 7194. c) K. Kaupmees,
N. Tolstoluzhsky, S. Raja, M. Rueping, I. Leito, Angew. Chem.
Int. Ed. 2013, 52, 11569.
[17] a) B. M. Nugent, R. A. Yoder, J. N. Johnston, J. Am. Chem.
Soc. 2004, 126, 3418. b) A. Singh, R. A. Yoder, B. Shen,
J. N. Johnston, J. Am. Chem. Soc. 2007, 129, 3466. c)
A. Singh, J. N. Johnston, J. Am. Chem. Soc. 2008, 130,
5866. d) M. Ganesh, D. Seidel, J. Am. Chem. Soc. 2008, 130,
16464.
[18] C. Bolm, T. Rantanen, I. Schiffers, L. Zani, Angew. Chem. Int.
Ed. 2005, 44, 1758.
[19] L. A. Curtiss, P. C. Redfern, K. Raghavachari, J. Chem. Phys.
2007, 126, 084108.
[5] a) T. J. Auvil, A. G. Schafer, A. E. Mattson, Eur. J. Org. Chem.
2014, 2014, 2633. b) T. Akiyama, Chem. Rev. 2007, 107, 5744.
c) T. Akiyama, K. Mori, Chem. Rev. 2015, 115, 9277.
[6] a) P. R. Schreiner, A. Wittkopp, Org. Lett. 2002, 4, 217. b)
A. Wittkopp, P. R. Schreiner, Chem. A Eur. J. 2003, 9, 407. c)
K. M. Lippert, K. Hof, D. Gerbig, D. Ley, H. Hausmann,
S. Guenther, P. R. Schreiner, Eur. J. Org. Chem. 2012, 2012,
5919. d) Z. Zhang, Z. Bao, H. Xing, Org. Biomol. Chem. 2014,
12, 3151.
[20] N. A. Yakelis, R. G. Bergman, Organometallics 2005, 24, 3579.
[21] D. R. Burfield, R. H. Smithers, A. S. C. Tan, J. Org. Chem.
1981, 46, 629.
[22] C. P. Rosenau, B. J. Jelier, A. D. Gossert, A. Togni, Angew.
Chem. Int. Ed. 2018, 57, 9528.
[23] a) P. Beak, J. Bonham, J. Am. Chem. Soc. 1965, 87, 3365. b)
A. R. Katritzky, R. C. Patel, M. Shanta, J. Chem. Soc., Perkin
Trans. 1 1980, 11, 1888. c) P. Guerry, R. Neier, Synthesis 1984,
1984, 485.

16 of 16
PAYNE AND KASS
[24] This is given as a septet rather than a triplet based upon its
chemical assignment, despite the fact that not all of the peaks
are fully resolved.
[25] P. J. Stang, G. Maas, D. L. Smith, J. A. McCloskey, J. Am.
Chem. Soc. 1981, 103, 4837.
[26] This is given as a septet rather than a pentet based upon the
chemical assignment and the observed intensities even though
the outermost peaks are not visible. That is, the five lines bet-
R. J. Twieg, Tetrahedron Lett. 1999, 40, 8759. b) M. Breugst,
H. Mayr, J. Am. Chem. Soc. 2010, 132, 15380. c) D. L. Comins,
G. Jianhua, Tetrahedron Lett. 1994, 35, 2819. d) O. R. Ludek,
C. Meier, Synlett 2006, 2006, 324. e) M. Torres, S. Gil,
M. Parra, Curr. Org. Chem. 2005, 9, 1757. f) H. Liu, S. B. Ko,
H. Josien, D. P. Curran, Tetrahedron Lett. 1995, 36, 8917. g)
T. Sato, K. Yoshimatsu, J. Otera, Synlett 1995, 1995, 845.
[39] J. E. Bartmess, NIST Chemistry WebBook, NIST Standard Ref-
erence Database Number 6; W. G. Mallard, P. J. Lustrum,
National Institute of Standards and Technology: Gaithersburg,
MD (http://webbook.nist.gov).
[40] This trend was observed in our initial study (ref. 9) but the
range of the free O–H frequencies was only ~20 cm–1 for
20 substituted phenols.
[41] See Supporting Information for titration plots for all com-
pounds in this work.
ter fit to
a 1:2.5:3.3:2.5:1 ratio rather than a 1:4:6:4:1
distribution.
[27] C. Yin, K. Zhong, W. Li, X. Yang, R. Sun, C. Zhang, X. Zheng,
M. Yuan, R. Li, Y. Lan, H. Fu, H. Chen, Adv. Synth. Catal.
2018, 360, 3990.
[28] J. Mendez-Arroyo, J. Barroso-Flores, A. M. Lifschitz,
A. A. Sarjeant, C. L. Stern, C. A. Mirkin, J. Am. Chem. Soc.
2014, 136, 10340.
[29] F. J. Arnaiz, J. Chem. Educ. 1995, 72, 1139.
[30] C. M. McGuirk, J. Mendez-Arroyo, A. I. d'Aquino, C. L. Stern,
Y. Liu, C. A. Mirkin, Chem. Sci. 2016, 7, 6674.
[31] This feature is comprised of a triplet (8.67, 1H) overlapping
with a doublet (8.65, 1H).
[32] a) A. D. Becke, J. Chem. Phys. 1993, 98, 5648. b) C. Lee,
W. Yang, R. G. Parr, Phys. Rev. B 1988, 37, 785.
[42] See Supporting Information for pictures of titrant solutions.
[43] The α-carbon hydrogen bonds of tetraalkylammonium salts
have been reported to participate in hydrogen bonding, see a)
Y. Kumatabara, S. Kaneko, S. Nakata, S. Shirakawa,
K. Maruoka, Chem. Asian J. 2016, 11, 2126. b) S. Shirakawa,
S. Liu, S. Kaneko, Y. Kumatabra, A. Fukuda, Y. Omargari,
K. Maruoka, Angew. Chem. Int. Ed. 2015, 54, 15767. c)
T. C. Cook, M. B. Andrus, D. H. Ess, Org. Lett. 2012, 14, 5836.
d) C. E. Cannizzaro, K. N. Houk, J. Am. Chem. Soc. 2002, 124,
7163. Also, see e) C. T. Dewberry, J. L. Mueller,
R. B. Mackenzie, B. A. Timp, M. D. Marshall, H. O. Leung,
K. R. Leopold, J. Mol. Struct. 2017, 1146, 373. and refs. Therein
[44] a) P. Thordarson, Chem. Soc. Rev. 2011, 40, 1305. b)
B. D. Hibbert, P. Thordarson, Chem. Commun. 2016, 52,
12792. c) See also supramolecular.org.
[33] a) Y. Zhao, D. G. Truhlar, J. Phys. Chem. A 2008, 112, 1095. b)
Y. Zhao, D. G. Truhlar, Theor. Chem. Account 2008, 120, 215.
c) Y. Zhao, D. G. Truhlar, Acc. Chem. Res. 2008, 41, 157.
[34] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone,
G. A. Petersson, H. Nakatsuji, X. Li, M. Caricato,
A. V. Marenich, J. Bloino, B. G. Janesko, R. Gomperts,
B. Mennucci, H. P. Hratchian, J. V. Ortiz, A. F. Izmaylov,
J. L. Sonnenberg, D. Williams-Young, F. Ding, F. Lipparini,
F. Egidi, J. Goings, B. Peng, A. Petrone, T. Henderson,
D. Ranasinghe, V. G. Zakrzewski, J. Gao, N. Rega, G. Zheng,
W. Liang, M. Hada, M. Ehara, K. Toyota, R. Fukuda,
J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao,
H. Nakai, T. Vreven, K. Throssell, J. A. Montgomery Jr.,
J. E. Peralta, F. Ogliaro, M. J. Bearpark, J. J. Heyd,
E. N. Brothers, K. N. Kudin, V. N. Staroverov, T. A. Keith,
R. Kobayashi, J. Normand, K. Raghavachari, A. P. Rendell,
J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, J. M. Millam,
M. Klene, C. Adamo, R. Cammi, J. W. Ochterski, R. L. Martin,
K. Morokuma, O. Farkas, J. B. Foresman, D. J. Fox, Gaussian
16, Gaussian, Inc, Wallingford CT 2016.
[45] F. Ulatowski, K. Dabrowa, T. Balakier, J. Jurczak, J. Org.
Chem. 2016, 81, 1746.
[46] A monomer/dimer equilibrium process was not included in
binding isotherm fits, but modifying the initial concentrations
of 4H accordingly has no effect on the resulting association
constants. This is because at the employed concentrations for
the titrations, the dimer contribution is negligible (<< 0.01%).
[47] For the kinetic data, see Supporting Information.
[48] This compound was also employed in the UV-vis titration with
1. See the Supporting Information for more details.
[49] Under the reaction conditions, >99% of 2MeOPyrH is in the
monomeric form. For the dimerization data, see Supporting
Information.
[35] GaussView, Version 6, R. Dennington, T. Keith, J. Millam,
Semichem Inc., Shawnee Mission, KS, 2016.
SUPPORTING INFORMATION
Additional supporting information may be found online
in the Supporting Information section at the end of this
article.
[36] A similar procedure has been used before, see ref. 9.
[37] For some reports regarding the tautomeric equilibria of 2- and
4-hydroxypyridines, see a) J. Wang, R. J. Boyd, J. Phys. Chem.
1996, 100, 16141. b) L. Forlani, G. Cristoni, C. Boga,
P. E. Todesco, E. Del Vecchio, S. Selva, M. Monari, ARKIVOC
2002, 11, 198. c) J. Gao, L. Shao, J. Phys. Chem. 1994, 98,
13772. d) P. Beak, J. B. Covington, S. G. Smith, J. M. White,
J. M. Zeigler, J. Org. Chem. 1980, 45, 1354. e) N. Tsuchida,
S. Yamabe, J. Phys. Chem. A 2005, 109, 1974.
How to cite this article: Payne C, Kass SR.
Structural considerations for charge-enhanced
Brønsted acid catalysts. J Phys Org Chem. 2020;
e4069. https://doi.org/10.1002/poc.4069
[38] For some examples of N- and O- alkylation studies of 2- and
4-hydroxypyridines and related compounds, see a) F. You,