Solid-State 17O NMR Study of Carboxylic Acid Dimers: Simultaneously Accessing Spectral Properties of Low- And High-Energy Tautomers

Article abstract of DOI:10.1021/acs.jpca.9b07224

We report a solid-state ¹⁷O NMR study of several crystalline carboxylic acids. We found that, while each of these compounds forms discrete hydrogen-bonded dimers in the crystal lattice, their solid-state ¹⁷O magic-angle spinning (MAS) NMR spectra display quite different features and different temperature dependencies. We showed that experimentally observed ¹⁷O NMR spectral behaviors can be explained as being due to thermal averaging between the two tautomers that are produced as a result of concerted double-hydrogen hopping dynamics within each dimer. In general, the two tautomers have different energies due to intramolecular interactions and crystal packing. From an analysis of variable-temperature ¹⁷O MAS NMR spectra, energy asymmetry between the two tautomers was experimentally determined for each of the carboxylic acid compounds studied. The same data analysis also offers an opportunity to simultaneously assess ¹⁷O NMR parameters in both low- and high-energy tautomers. We concluded that the periodic plane-wave density functional theory (DFT) calculations can produce reliable ¹⁷O NMR parameters (chemical shift and quadrupolar coupling tensors) for both tautomers. The same periodic DFT calculations have also produced reasonable energy asymmetry values for the studied carboxylic acid dimers. We have also observed substantial H/D isotope shifts in solid-state ¹⁷O NMR.

Full text of DOI:10.1021/acs.jpca.9b07224

Subscriber access provided by Macquarie University
A: Spectroscopy, Molecular Structure, and Quantum Chemistry
17
A Solid-State O NMR Study of Carboxylic Acid Dimers: Simultaneously
Accessing Spectral Properties of Low- and High-Energy Tautomers
Gang Wu, Ivan Hung, Zhehong Gan, Victor V. Terskikh, and Xianqi Kong
J. Phys. Chem. A, Just Accepted Manuscript • DOI: 10.1021/acs.jpca.9b07224 • Publication Date (Web): 27 Aug 2019
Downloaded from pubs.acs.org on August 30, 2019
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 46
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
1
7
A Solid-State O NMR Study of Carboxylic Acid Dimers:
Simultaneously Accessing Spectral Properties of Low- and
High-Energy Tautomers
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
b
b
a,c
a
Gang Wu, * Ivan Hung, Zhehong Gan, Victor Terskikh, and Xianqi Kong
aDepartment of Chemistry, Queen’s University, 90 Bader Lane,
b
Kingston, Ontario, Canada K7L 3N6; Center of Interdisciplinary Magnetic Resonance,
National High Magnetic Field Laboratory, 1800 East Paul Dirac Drive, Tallahassee,
c
Florida 32310, USA; Department of Chemistry, University of Ottawa, Ottawa, Ontario,
Canada K1N 6N5
*Corresponding author: wugang@queensu.ca
1
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 2 of 46
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Abstract
1
7
We report a solid-state O NMR study of several crystalline carboxylic acids. We found
that, while each of these compounds forms discrete hydrogen-bonded dimers in the
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
17
crystal lattice, their solid-state O magic-angle spinning (MAS) NMR spectra display
quite different features and different temperature dependences. We showed that
1
7
experimentally observed O NMR spectral behaviors can be explained as due to thermal
averaging between the two tautomers that are produced as a result of concerted double
hydrogen hopping dynamics within each dimer. In general, the two tautomers have
different energies due to intramolecular interactions and crystal packing. From an
1
7
analysis of variable-temperature O MAS NMR spectra, energy asymmetry between the
two tautomers was experimentally determined for each of the carboxylic acid compounds
1
7
studied. The same data analysis also offers an opportunity to simultaneously assess O
NMR parameters in both low- and high-energy tautomers. We concluded that the
periodic plane-wave DFT calculations can produce reliable 17O NMR parameters
(chemical shift and quadrupolar coupling tensors) for both tautomers. The same periodic
DFT calculations have also produced reasonable energy asymmetry values for the studied
carboxylic acid dimers. We have also observed substantial H/D isotope shifts in solid-
1
7
state O NMR.
2
ACS Paragon Plus Environment

Page 3 of 46
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
1
. Introduction
Simple carboxylic acids are known to have great tendency to form hydrogen-bonded
dimers in the solid state.1-2 Within each dimer, the two protons can undergo concerted
double-jumps to produce two tautomers or configurations, A and B, as illustrated in
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 1.
From the carbon-oxygen bonding perspective, the consequence of
interchanging the A and B tautomers is to interchange the CO with C=O bonds. For an
isolated carboxylic acid dimer, the two tautomers would have identical energies. In the
solid state, however, the two tautomers will generally have different energies due to
crystal packing. In some extreme cases, both intra- and inter-molecular interactions
strongly favour one tautomer over the other. While in principle the double proton
transfer occurs in a two-dimensional (2D) potential energy surface, it is often instructive
to think of the problem by using an over-simplified 1D double-well potential energy
curve as shown in Figure 1. The energy barrier is defined as V and the energy
0
asymmetry between the two tautomers is usually denoted as E (it is sometimes defined
as A). In the past several decades, a large number of investigations have been conducted
to probe the dynamics of the concerted double hydrogen hopping process in this type of
hydrogen-bonded systems.3-40 At high temperatures (e.g. > 50 K), the double proton
transfer can be considered to be a classic Arrhenius-type over-the-barrier process. At low
temperatures, however, the proton transfer dynamics is largely due to incoherent quantum
tunnelling between the two vibrational ground states, A(0) and B(0). Thus, at low
8
–1
temperatures below 50 K, the rate of proton transfer is still on the order of 10 s and
does not display any temperature dependence. This proton jump rate can be as high as
1
012 s–1 at high temperatures. Because the proton jump rate is almost always much faster
3
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 4 of 46
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
than the NMR timescale, any observed NMR spectral parameter corresponds to the
averaged value between those in the two tautomers. That is, the two tautomers are not
accessible separately in systems where E is small. While most previous studies have
focused on the quantum tunnelling nature of the proton transfer dynamics in carboxylic
acid dimers, little attention has been paid to the evaluation of the observed NMR
properties and their links to those in both low- and high-energy tautomers. The main
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
17
objective of this study is to investigate whether we can infer the O NMR properties of
1
7
the two tautomers by examining the temperature dependence of the experimental O
NMR parameters, which are only the averaged results between the two tautomers. It
17
should be noted that early O nuclear quadrupole resonance (NQR) studies by Smith and
co-workers5, 9 provided perhaps the strongest evidence to support the double proton
1
7
transfer model in carboxylic acid dimers. The advantage of solid-state O NMR over
NQR is that it can produce information about both 17O magnetic shielding and
quadrupole coupling (QC) tensors.
1
7
In this work, we conducted a comprehensive solid-state O NMR study of several
carboxylic acid dimers where the carboxylate groups are labeled by 17O: [1,2-
17
17
17
O ]benzoic acid (BA), [1,2- O ]salicylic acid (SA), [1,2- O ]Aspirin (ASA), [1,2-
2
2
2
1
7
17
O ]cinnamic acid (CA), [1,2- O ]-4-methoxycinnamic acid (MCA). The crystal
2
2
structures of these compounds indicate that they all form the same basic carboxylic acid
dimer structure in the solid state.41-45 While solid-state O NMR has been used to study
many organic compounds in the past 20 years,46-50 only some preliminary solid-state O
NMR data were reported previously for carboxylic acid dimers such as BA, SA, ASA,
17
17
ibuprofen, and terephthalic acid.51-54 In particular, Kong et al. found that the O MAS
53
17
4
ACS Paragon Plus Environment

Page 5 of 46
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
NMR spectra of ASA and SA exhibit very different temperature dependencies. In the
present work, we further examined the temperature dependence of 17O MAS NMR
spectra for this class of compounds. To aid spectral analysis, we employed periodic
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
17
plane-wave DFT computations to evaluate the O NMR parameters in both low- and
high-energy tautomers. Finally, we will address how the experimental variable-
1
7
temperature O NMR data can be properly interpreted to obtain new information about
17O NMR spectral properties in low- and high-energy tautomers.
2. Experimental section
2
.1 Synthesis
Preparations of [1,2- O ]BA-H, [1,2- O ]SA, [1,2- O ]ASA, and [1,2- O ]MCA
1
7
17
17
17
2
2
2
2
were described previously.51, 53, 55 [1,2- O ]BA-D was obtained by dissolving [1,2-
17
2
1
7
17
O ]BA in an acetone/D O mixture. [1,2- O ]CA was prepared in the following fashion.
2
2
2
[17
O ]Malonic acid (250 mg, 18% O-enrichment, prepared by heating malonic acid in
4
17
1
7
4
0% O-enriched water) was mixed with benzaldehyde (252 mg) and piperidine (two
drops) in 4 mL pyridine. The mixture was heated at reflux (oil bath temperature 90 ᵒ C)
for 12 h. After the solution was cooled to room temperature, solvent was then evaporated
on a rotary evaporator to give a paste. This material was dissolved in 1.5 mL ethanol.
Upon addition of cold 0.5 M HCl (10 mL), precipitates formed, which were collected by
filtration, first washed with cold 0.5 M HCl (8 mL) and then with water (2  5 mL). The
solids were initially air dried and then further dried on vacuum. The final product was
1
7
obtained as a white solid with a distinct honey-like smell (270 mg). The O-enrichment
5
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 6 of 46
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
17
17
level in [1,2- O ]CA was estimated to be 18% based on the O NMR signal integral
2
with respect to that of the solvent (acetone-d6).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
.2 Solid-state NMR experiments
1
7
Preliminary O MAS NMR experiments were performed at 14.1 T (Queen’s
University) and 21.1 T (National Ultrahigh Field NMR Facility for Solids, Ottawa,
1
7
Canada). Variable-temperature (VT) O MAS NMR spectra were obtained at 21.1 T at
the National High Magnetic Field Laboratory (Tallahassee, Florida, USA) with a 900
1
MHz Bruker Avance console and a 3.2 mm home-built MAS H-X transmission line
probe. Sample temperatures were adjusted for rotor frictional heating under the 20-kHz
56-58
MAS condition by using solid Pb(NO ) for temperature calibration.
The actual
3
2
sample temperature range of the VT MAS probe was between 363 and 129 K. A liquid
1
7
H O sample was used for both RF power calibration and O chemical shift referencing
2
 = 0 ppm). All spectral simulations were performed with DMfit.59
(
2.3 Data analysis
Analysis of the isotropic 17O chemical shifts. As mentioned earlier, the proton
1
0
12 –1
dynamics in crystalline carboxylic acid dimers occurs on the order of 10 -10 s within
the temperature range examined in this study (between 129 and 363 K). As a result,
17
NMR observables such as O chemical shift and QC tensors will be the averaged values
between the two possible tautomers A and B. This situation is distinctly different from
1
7
other cases where solid-state O NMR was used to extract energetic information about
dynamic processes.60-66 Following a simple “Arrhenius approach” developed in previous
6
ACS Paragon Plus Environment

Page 7 of 46
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
1
13
4, 6, 22
17
H and C NMR studies,
we can treat the averaging effect on the O isotropic
chemical shifts as a simple “thermal averaging” between the two tautomers:
퐴
+(1 ― 푃 )훿퐵
훿 = 푃 훿
(1)
(2)
1
퐴 퐶 = 푂
퐴
퐶 ― 푂퐻
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
퐴
+(1 ― 푃 )훿퐵
퐶 = 푂
훿 = 푃 훿
2
퐴 퐶 ― 푂퐻
퐴
where
Δ퐸
푒푅푇
Δ퐸
+ 푒푅푇
푃퐴 =
(3)
1
In the above equation, E refers to the energy asymmetry of the double-well potential
curve between tautomers A and B. A schematic diagram showing this averaging is
depicted in Figure 2. In this study, we are particularly interested in the temperature
1
7
dependence of the averaged O isotropic chemical shift values because they can be
determined precisely from the experimental spectra. As also illustrated in Figure 2, there
are two general types of temperature dependencies. In the first type, as seen in cases (i)-
1
7
(iii), the separation between the two averaged O NMR signals increases as the
temperature is decreased. In the second type, the signal separation initially decreases as
the temperature is decreased, and then increases after a “cross-over” point where the two
signals merge into one; see case (iv) in Figure 2. We should note that, while the two
types of temperature dependencies shown in Figure 2 are produced with the same set of
17O isotropic chemical shifts but different E values, it is also possible to generate these
1
7
two types with the same value of E but different sets of O isotropic chemical shifts for
the two tautomers (vide infra).
17
17
Analysis of the O QC tensors. Unlike the aforementioned treatment of O isotropic
chemical shifts, averaging of the 17O QC tensors involved in low- and high-energy
7
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 8 of 46
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
tautomers is not so straightforward. This is because one has to consider the tensor
orientations in the molecular frame of reference between the two tautomers. Fortunately,
1
7
48, 67-69
in the case of carboxylic acids, the O QC tensor orientations are well established.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
17
Figure 3 shows typical orientations of the O QC tensors found for the C=O and C-OH
1
7
groups of carboxylic acids. For the C=O group, the largest O QC tensor component,
zz, is perpendicular to the C=O bond and lies within the O=C-OH plane, whereas the
smallest component,  , is perpendicular to the plane. In contrast, while  of the C-OH
xx
zz
group also lies within the O=C-OH plane, it is along the C-O bond. In addition,  is
xx
now in-plane and perpendicular to the C-O bond. Another important feature to consider
1
7
is the sign of each O QC tensor component. For carboxylic acids, it is also well known
1
7
that the sign of C ( O) for the C=O group is positive, which means  > 0,  < 0,  <
Q
zz
yy
xx
17
0
. For the C-OH group, C ( O) is known to be negative, so  < 0,  > 0,  > 0. It
Q
zz
yy
xx
should be noted that, while the sign of C cannot be easily determined by solid-state
Q
NMR, it can be reliably accessed by quantum chemical computations.
It is interesting to note that, for carboxylic acid dimers, the directions of the three
1
7
principal axes of the O QC tensor are entirely defined by the C-O (or C=O) bond and
the O=C-OH plane; see Figure 3. As a result, averaging between A and B tautomers will
not change the directions of the principal axes. Using this feature, we can define the
three principal axes of the averaged QC tensor within the molecular frame as 휒 (in-
∥
plane and along to the C-O bond), 휒 (in-plane but perpendicular to the C-O bond), and
⊥
휒휋 (perpendicular to the O=C-OH plane or out-of-plane). As seen from Figure 3(a), the
averaged tensor components for A(C=O)  B(C-OH) can then be readily calculated:
퐴
퐵
휒 = 푃 휒 +(1 ― 푃 )휒
푧푧
(4)
∥
퐴
푦푦
퐴
8
ACS Paragon Plus Environment

Page 9 of 46
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
퐴
퐵
휒 = 푃 휒 +(1 ― 푃 )휒
(5)
(6)
⊥
퐴
푧푧
퐴
푥푥
퐴
퐵
휒 = 푃 휒 +(1 ― 푃 )휒
휋
퐴
푥푥
퐴
푦푦
A similar set of equations can be obtained for A(C-OH)  B(C=O). So, in general, two
averaged QC tensors will be produced.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 3(b) also shows an example to illustrate the temperature dependence of the
17
17
averaged O QC tensor. In this case, we assumed the following O QC tensors: A,
tautomer, C = 7.00 MHz,  = 0.5 for C=O ( = 7.00,  = –5.25,  = –1.75 MHz)
Q
Q
zz
yy
xx
and C = –6.00 MHz,  = 0.1 for C-OH ( = –6.00,  = 3.30,  = 2.70 MHz); B
Q
Q
zz
yy
xx
tautomer, C = 8.00 MHz,  = 0.3 for C=O ( = 8.00,  = –5.20,  = –2.80 MHz)
Q
Q
zz
yy
xx
and C = –7.00 MHz,  = 0.2 for C-OH ( = –7.00,  = 4.20,  = 2.80 MHz). As
Q
Q
zz
yy
xx
seen from Figure 3(b), the averaging of the QC tensors between the two tautomers will
lead to two averaged QC tensors. At the high-temperature limit, both averaged QC
tensors display   1.0. This can be understood by examining the individual tensor
Q
components. For example, let us look at 휒 of the averaged tensor for A(C=O)  B(C-
휋
퐴
퐵
OH) at the high-temperature limit. Because 휒 = –1.75 MHz and 휒 = 2.80 MHz nearly
푥푥
푦푦
cancel each other when P  0.5 at T  , 휒 would be very small, resulting in an
A
휋
averaged QC tensor with   1.0. At the low-temperature limit, however, the averaged
Q
tensors have   0 and   0.5, which resemble those in the A tautomer alone. This is
Q
Q
because P  1 when T  0.
A
2
.4 Quantum Chemical Computations
9
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 10 of 46
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Plane-wave pseudo-potential DFT calculations of the NMR magnetic shielding and
quadrupole coupling parameters were performed using Materials Studio CASTEP
7
0
software version 4.4 (Accelrys) on a HP xw4400 workstation with a single Intel Dual-
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Core 2.67 GHz processor and 8 GB DDR RAM. The Perdew, Burke, and Ernzerhof
(
PBE) functionals were used in all calculations in the generalized gradient approximation
GGA) for the exchange correlation energy. On-the-fly pseudo potentials were used as
(
supplied with NMR CASTEP with a plane wave basis set cut-off energy of 550 eV and
the Monkhorst-Pack k-space grid sizes of 4 × 4 × 4 (16 k-points used). The most recent
crystal structures of BA,41 SA,42 ASA,43 CA,44 and MCA45 were used as starting
structures and then geometry optimization was performed for both low- and high-energy
tautomers using the BFGS method without the cell optimization. Following convergence
tolerance parameters were used in the geometry optimization process: total energy 10–5
eV/atom, maximum displacement 0.001 Å, maximum force 0.03 eV/Å, and maximum
stress 0.05 GPa.
3
. Results and discussion
1
7
Figure 4 shows the O MAS NMR spectra obtained for several carboxylic acid dimers
at 21.1 T. While these carboxylic acid dimers have very similar hydrogen bonding
1
7
geometries, their O MAS NMR spectra are quite different. For example, two well-
1
7
17
resolved O NMR signals are observed for [1,2- O ]SA, but the two signals from [1,2-
2
17
O ]BA are nearly indistinguishable. Further examination of the spectral details reveals
2
1
7
more differences in the O quadrupole parameters among these carboxylic acid dimers.
1
7
17
For example, the two signals in the O MAS NMR spectrum of [1,2- O ]SA have the
2
1
0
ACS Paragon Plus Environment

Page 11 of 46
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
1
7
following O NMR parameters: signal-1,  = 281 ppm, C = 7.0 MHz,  = 0.45;
iso
Q
Q
signal-2, iso = 164 ppm, C = 7.5 MHz,  = 0.16. The corresponding parameters for
Q
Q
17
[
1,2- O ]BA are: signal-1,  = 234 ppm, C = 5.8 MHz,  = 0.8; signal-2,  = 227
2
iso
Q
Q
iso
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
7
ppm, C = 5.9 MHz,  = 0.8. From the O MAS NMR spectra shown in Figure 4, two
Q
Q
general features are note worthy. First, the C values are considerably larger in SA than
Q
in BA. Second, while the two signals in SA have very different  values, both of which
Q
are less than 0.5, the two signals in BA have the same high  value. In general, as seen
Q
from Figure 4, the  values of both signals increase from SA, ASA, CA, to MCA and
Q
BA.
To better understand the origin of the spectral differences observed for different
17
carboxylic acid dimers, we obtained VT O MAS NMR spectra for these compounds.
1
7
17
17
Figure 5 shows the results for [1,2- O ]ASA, [1,2- O ]CA, and [1,2- O ]MCA. We
2
2
2
1
7
17
53
should note that the O MAS NMR spectra from [1,2- O ]SA, as reported previously,
2
17
did not display any temperature dependence. As seen from Figure 5, the O MAS NMR
1
7
spectra of [1,2- O ]ASA change as a function of temperature in three aspects. First, the
2
separation between the two signals decreases as the temperature of the sample is
increased. Second, the values of  appear to increase with the temperature for both
Q
signals. Third, the C values for the two signals also decrease slightly as the temperature
Q
1
7
is increased. The spectra of [1,2- O ]CA display similar temperature dependencies as
2
17
described for [1,2- O ]ASA. In contrast, a quite difference temperature dependence is
2
1
7
observed for [1,2- O ]MCA. That is, the separation of the two signals increases with the
2
temperature, and the values of neither C nor  appear to be sensitive to the
Q
Q
temperature. Thus, in terms of the signal separation, the general spectral behaviors
1
1
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 12 of 46
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
17
17
observed for [1,2- O ]ASA and [1,2- O ]CA are similar to those in cases (i)-(iii) shown
2
2
1
7
in Figure 2, but [1,2- O ]MCA displays a temperature dependence similar to case (iv).
2
1
7
Now in order to interpret the VT O NMR results shown in Figure 5, it is necessary to
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
17
have some general ideas about the O NMR parameters in both low- and high-energy
tautomers. To this end, we decided to perform periodic plane-wave DFT computations to
obtain information about both energetic and spectral properties of the two tautomers. For
each compound, we started from the best crystal structure available in the literature.
Then we obtained the optimized structures for both A and B tautomers. In this step, we
1
7
obtained the energy asymmetry between the two tautomers. Finally, we computed O
NMR tensors for both tautomers. Table 1 shows the computational results. In general,
17
the computations suggest that the O NMR parameters for the carboxylic acid functional
group in the two tautomers are quite similar, even when there is a large energy
1
7
asymmetry between the two tautomers. This is not a total surprise because the O NMR
parameters of the -COOH group should be largely determined by the hydrogen bonding
interaction within the dimer, whereas the energy asymmetry is mainly influenced by
crystal packing. Further, the computations have also confirmed the 17O QC tensor
1
7
orientations depicted in Figure 3 as well as the aforementioned signs of C ( O) for both
Q
C=O and C-OH groups.
Using the computational results listed in Table 1 as initial inputs, we were able to
1
7
obtain best-fit results for the experimental VT O NMR data ( , C and  ) as well as
iso
Q
Q
for the energy asymmetry, E. Figure 6 shows the comparison between the experimental
and best-fit data for ASA, CA, and MCA. The best-fit parameters are summarized in
Table 2. As seen from Figure 6, the overall agreement between the experimental and
1
2
ACS Paragon Plus Environment

Page 13 of 46
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
17
17
best-fit curves is satisfactory. Because the O MAS NMR spectrum of [1,2- O ]SA
2
does not show any temperature dependence, as mentioned earlier, we estimated that E
–
1
in SA must be greater than 15 kJ mol . We will further discuss the results shown in
Table 2 in a later section.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
For now we turn our attention to a different aspect of carboxylic acid dimers.
Previous studies have shown significant H/D isotope effects in these systems.12, 19, 21, 23, 32
In particular, many studies have established that the H hopping dynamics within the
hydrogen-bonded carboxylic acid dimer is significantly slowed down when H is replaced
by D.19, 32 In this study, we set out to investigate whether there is any H/D isotope effect
17
on the O NMR parameters as well as their temperature dependence. Figure 7 shows the
VT 17O MAS spectra of [1,2- O ]BA-H and [1,2- O ]BA-D; the detailed spectral
17
17
2
2
fittings for BA-H and BA-D are given in the Supporting Information (Figures S1 and S2).
Immediately, one sees that the two compounds display quite different spectral behaviors
with BA-D being more sensitive to the temperature than BA-H. At 357 K, the separation
between the two signals is about 3 ppm for BA-H. In comparison, for BA-D at 349 K,
the signal separation is about 5 ppm. This H/D isotope effect gets amplified at low
temperatures. For example, a signal separation of 34 ppm is observed for BA-H at 129
1
7
K, which is compared with 55 ppm for BA-D at 137 K. A close inspection of the O
isotropic chemical shifts observed for BA-H and BA-D at the lowest temperatures reveals
that the high- and low-frequency signals display an apparent H/D isotope shift of –4 and
1
7
17
1
7 ppm, respectively. Note that the isotope shift is defined as  O(H/D) =  O(H) –
1
7
 O(D). In the present case, these isotope shifts are the averaged values between the two
tautomers. Nonetheless, these H/D isotope shifts are quite substantial as compared with
1
3
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 14 of 46
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
those reported for other hydrogen-bonded systems.71-72 Figure 8 shows the experimental
1
7
and best-fit O NMR results for BA-H and BA-D. Again, the general trends are similar
1
7
to those shown in Figure 6 for other carboxylic acid dimers. The best-fit O NMR
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
17
parameters are also listed in Table 2. Indeed, BA-H and BA-D do exhibit different O
NMR parameters in both A and B tautomers. Again, substantial isotope shifts,
1
7
 O(H/D), are seen for both the C=O and C-OH groups; however, we caution that the
17
uncertainties in the best-fit O NMR parameters shown in Table 2 are quite large. For
this reason, we will not further discuss these 17O isotope shifts. Another interesting
finding with regard to the H/D isotope effect is that the energy asymmetry in BA-D is
nearly twice that in BA-H. Since the error in E is relatively small, the observed
difference between BA-H and BA-D is significant (vide infra).
After analysing all the VT 17O NMR data, we can now compare the best-fit data
reported in Table 2 with those from DFT computations summarized in Table 1. Figure 9
shows the direct comparisons for both 17O chemical shift/shielding constant and
quadrupole parameters. The general agreement is consistent with other studies dealing
1
7
with O NMR parameters in organic solids. Furthermore, the periodic plane-wave DFT
computations appear to have treated the two tautomers equally well. This seems
reasonable since the two tautomers indeed have very similar ground-state energies in the
first place. This is an important conclusion because the two tautomers in carboxylic acid
dimers cannot be separately assessed by NMR. Thus, this study represents a very rare
case where information about 17O NMR parameters from high-energy tautomers are
indirectly evaluated.
1
4
ACS Paragon Plus Environment

Page 15 of 46
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
Finally we will discuss the energy asymmetry E values found for carboxylic acid
dimers. Table 3 shows a partial compilation of E values determined for carboxylic acid
1
2
dimers by various techniques including NMR relaxation measurement ( H and H),
neutron/X-ray diffraction data analysis, IR spectral analysis, NQR, and solid-state NMR
measurement. The hydrogen bonding lengths, dO···O, for the carboxylic acid dimers listed
in Table 3 fall in a narrow range between 2.570 and 2.654 Å. In general, the E values
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
are not related to the hydrogen bonding lengths. Rather, E should depend on crystal
packing and inramotlecular interactions other than the hydrogen bonding. For example,
the compound with the largest E value among those listed in Table 3 is SA. In the A
tautomer of SA, the 2-OH group forms an intramolecular hydrogen bond of the type O–
H···O=C. This makes the C=O group to be the common acceptor of two hydrogen
bonds. In the B tautomer, however, the hydrogen bonding arrangement becomes O–
H···O–H··· where the central OH group serves as both acceptor and donor. Apparently,
the A tautomer has significantly lower energy than the B tautomer in SA. It is interesting
to note that, in the case of ASA, where the 2-OH group is modified to lose the capability
–
1
as a hydrogen bond donor, the E value, 3 kJ mol , is considerably smaller than that in
–
1
SA, > 15 kJ mol . Another interesting ortho-substituted BA derivative where a
–
1
relatively large E was found is 2NBA (E = 5.8 kJ mol ). The crystal structure of
2
NBA shows that both the nitro and carboxylic acid functional groups are significantly
out of the benzene plane to avoid O···O contact.73
For the H/D isotope effect, the E value generally increases on going from H to D,
except for the case of BA. As seen from Table 3, BA is the most studied carboxylic acid
dimer. However, there are some discrepancies among the literature data. As mentioned
1
5
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 16 of 46
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
–
1
earlier, we determined the E values for BA-H and BA-D to be 0.6 and 1.2 kJ mol ,
respectively. Our results are in agreement with those reported by Agaki et al.,12 by
Stöckli et al, and by Takeda and Tsuzumitani. In contrast, Heuer and Haeberlen21
1
9
28
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
–
1
32
reported that E = 0.72 kJ mol for BA-D. Xue et al. also found that BA-H and BA-D
–
1
exhibit the same E value, 0.71 kJ mol . The exact reason for these discrepancies is
unclear to us at this time.
4. Conclusions
17
We have recorded VT solid-state O NMR spectra at 21.1 T for several carboxylic
1
7
acid dimers. One important feature found in these compounds is that their O MAS
NMR spectra exhibit very different temperature dependences. This temperature effect
1
7
can be rationalized by considering the thermal averaging of O NMR tensors between
the two tautomers associated with double proton hopping within the hydrogen-bonded
carboxylic acid dimer. Since the proton transfer dynamics in carboxylic acid dimers is
much faster than the NMR timescale, only the averaged results are experimentally
accessible. We showed that the best-fit 17O NMR parameters obtained for the two
tautomers are in reasonable agreement with those from plane-wave DFT calculations.
–
1
Since energy asymmetry values are usually less than a few kJ mol in most systems
studied, quantum chemical computations to this degree of accuracy are not to be expected
from the current DFT methodologies. Nonetheless, we found that the periodic DFT
calculations are in qualitative agreement with the experimental results. We have also
1
7
presented an interesting case of H/D isotope effects on both O NMR parameters and
1
6
ACS Paragon Plus Environment

Page 17 of 46
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
17
energy asymmetry. Further studies on the H/D isotope effects on O NMR parameters in
hydrogen-bonded systems are underway in this laboratory.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Supporting Information Available. Two figures showing experimental and simulated
17O MAS spectra for BA-H and BA-D at different temperatures.
Acknowledgements
This work was supported by the Natural Sciences and Engineering Research Council
(
NSERC) of Canada. Access to the 900 MHz NMR spectrometer and CASTEP software
was provided by the National Ultrahigh Field NMR Facility for Solids (Ottawa, Canada),
a national research facility funded by a consortium of Canadian universities, National
Research Council Canada and Bruker BioSpin and managed by the University of Ottawa
(http://nmr900.ca). A portion of this work was performed at the National High Magnetic
Field Laboratory, which is supported by the National Science Foundation Cooperative
Agreement No. DMR-1644779 and the State of Florida.
1
7
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 18 of 46
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Figure Captions
Tautomer A
Tautomer B
O
O
H···
O
···H
O
O
C
C
C
C
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
·
· · H
O
O
H···
O
A
B
νB(0)
νA(0)
V0

E
Figure 1. (Top) Illustration of the two tautomers generated from concerted double proton
hopping within a hydrogen-bonded carboxylic acid dimer in the solid state. (Bottom)
Simplified 1D potential curve where the energy difference between the two energy
minima is defined as the energy asymmetry (E).
1
8
ACS Paragon Plus Environment

Page 19 of 46
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
(a)
C=O
C-OH
Tautomer A
1
7
δ( O)
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
C=O
C-OH
Tautomer B
1
7
δ( O)
Averaged
1
7
δ( O)
δ1 δ2
3
00
50
(b)
(
i)
(
ii)
2
(iii)
(
iv)
200
150
2
4
6
8
10
12
1
000/T (1/K)
1
7
Figure 2. (a) Illustration of the averaging effect for O isotropic chemical shifts between
tautomers A and B. (b) General types of temperature dependence of the averaged O
isotropic chemical shift. The following O isotropic chemical shifts are used in the
calculations: A tautomer, 280 ppm (C=O) and 180 ppm (C-OH); B tautomer, 300 ppm
C=O) and 150 ppm (C-OH). As examples, calculations using four different E values
are shown: (i) 10; (ii) 5; (iii) 2; (iv) 0.5 kJ mol . Calculations were based on Eqs. (1)-(3).
See text for discussion.
1
7
17
(
–
1
1
9
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 20 of 46
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
7
Figure 3. (a) O QC tensor orientations in the molecular frame of reference in tautomers
1
7
A and B. (b) General temperature dependence of the averaged O QC tensor components
solid lines for the  signal and dashed lines for the  signal). The following
quadrupole parameters are used in the calculations: A tautomer, C = 7.0 MHz,  = 0.5
for C=O and C = –6.0 MHz,  = 0.1 for C-OH; B tautomer, C = 8.0 MHz,  = 0.3
17
(
O
1
2
Q
Q
Q
Q
Q
Q
for C=O and C = –7.0 MHz,  = 0.2 for C-OH. The E value used in the calculations
Q
Q
–1
was 2 kJ mol . Calculations were based on Eqs. (4)-(6). See text for discussion.
2
0
ACS Paragon Plus Environment

Page 21 of 46
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
(
(
(
a)
b)
c)
*
*
*
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
*
* *
*
*
*
*
(
d)
e)
*
*
*
*
*
(
*
*
5
00
450
400
350
300
250
200
150
100
50
0
1
7
δ( O)/ppm
1
7
Figure 4. Experimental (black trace) and simulated (colored traces) O MAS NMR
spectra of (a) [1,2- O ]BA, (b) [1,2- O ]MCA, (c) [1,2- O ]CA, (d) [1,2- O ]ASA,
and (e) [1,2- O ]SA. All spectra were obtained at 295 K and at 21.1 T. The sample
spinning frequencies were: (a, b) 15 kHz; (c) 18 kHz; (d, e) 20 kHz. The signals marked
with * are spinning sidebands.
1
7
17
17
17
2
2
2
2
17
2
2
1
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 22 of 46
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
ASA
CA
MCA
3
43 K
351 K
319 K
295
2
2
95
63
327
295
239
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
263
231
231
1
1
99
30
199
159
167
131
3
00
250 200 150
100 300
250
200
150 300
250
200
150
17
17
17
δ( O)/ppm
δ( O)/ppm
δ( O)/ppm
1
7
Figure 5. Experimental (blue trace) and simulated (colored traces) O MAS NMR
spectra of (a) [1,2- O ]ASA, (b) [1,2- O ]CA, and (c) [1,2- O ]MCA at different
temperatures. All spectra were obtained at 21.1 T.
1
7
17
17
2
2
2
2
2
ACS Paragon Plus Environment

Page 23 of 46
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
(
a)
(b)
(c)
300.0
250.0
200.0
150.0
3
2
2
1
00.0
50.0
00.0
50.0
300.0
O1
O1
O2
O1
O2
250.0
200.0
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
O2
150.0
2.00
4.00
6.00
8.00
8.00
8.00
10.00
2.00
4.00
6.00
8.00
10.00
2.00
4.00
6.00
8.00
10.00
1
000/T (1/K)
1000/T (1/K)
1000/T (1/K)
1
0.0
1.0
10.0
1.0
0.0
1.0
O1
O1
O1
8
.0
0.8
0.6
0.4
0.2
0.0
8.0
6.0
4.0
2.0
0.0
0.8
0.6
0.4
0.2
0.0
-2.0
0.8
0.6
0.4
0.2
0.0
6.0
-4.0
ηQ
ηQ
ηQ
4
2
.0
.0
-6.0
-8.0
0.0
2
-10.0
.00
4.00
6.00
10.00
2.00
4.00
6.00
8.00
10.00
2.00
4.00
6.00
8.00
10.00
1
000/T (1/K)
1000/T (1/K)
1000/T (1/K)
0
.0
1.0
0.8
0.6
0.4
0.2
0.0
0.0
-2.0
1.0
0.8
0.6
0.4
0.2
0.0
0.0
1.0
0.8
0.6
0.4
0.2
0.0
O2
O2
O2
-
2.0
4.0
6.0
-2.0
-4.0
-
-
-4.0
ηQ
ηQ
ηQ
-6.0
-6.0
-8.0
-8.0
-8.0
-
10.0
-10.0
-10.0
2
.00
4.00
6.00
10.00
2.00
4.00
6.00
8.00
10.00
2.00
4.00
6.00
8.00
10.00
1
000/T (1/K)
1000/T (1/K)
1000/T (1/K)
1
7
Figure 6. Experimental (data symbols) and best-fitted (solid lines) O isotropic chemical
shift and quadrupole parameters (C and  ) for (a) [1,2- O ]ASA, (b) [1,2- O ]CA,
and (c) [1,2- O ]MCA as a function of temperature. The calculations were based on Eqs.
1)-(6) with parameters given in Table 2. The error bars for experimental  ( O) values
( 1 ppm) are not shown for clarity.
17
17
Q
Q
2
2
1
7
2
1
7
(
iso
2
3
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 24 of 46
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(a)
(b)
3
3
57 K
49
349 K
*
*
*
*
321
281
249
225
191
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
325
2
95
2
1
29
81
1
1
53
29
157
137
4
00
300
200
δ(17O)/ppm
100
0 400
300
200
δ(17O)/ppm
100
0
17
Figure 7. Experimental O MAS NMR spectra of (a) BA-H and (b) BA-D at different
temperatures. All spectra were obtained at 21.1 T. The signals marked with * are
spinning sidebands. Simulated spectra are not shown for clarity; they are given in the
Supporting Information.
2
4
ACS Paragon Plus Environment

Page 25 of 46
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
(a)
3
00.0
O1
250.0
200.0
O1 BA-H
O2
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
BA-D
O2
1
50.0
2
.00
4.00
6.00
8.00
10.00
1
000/T (1/K)
(
b)
BA-H
ηQ
BA-D
ηQ
0
.0
1.0
0.8
0.6
0.4
0.2
0.0
0.0
-2.0
1.0
-2.0
-4.0
-6.0
-8.0
0.8
0.6
0.4
0.2
0.0
-4.0
-6.0
-8.0
O1
O1
-10.0
-10.0
2
.0 4.0 6.0 8.0 10.0
2.0
4.0
6.0
8.0 10.0
1000/T (1/K)
1000/T (1/K)
ηQ
ηQ
0.0
1.0
0.0
-2.0
1.0
0.8
0.6
0.4
0.2
0.0
-2.0
-4.0
-6.0
-8.0
0.8
0.6
0.4
0.2
0.0
-4.0
-6.0
-8.0
O2
O2
-10.0
-10.0
2
.0 4.0 6.0 8.0 10.0
2.0
4.0
6.0
8.0 10.0
1000/T (1/K)
1000/T (1/K)
1
7
Figure 8. Experimental (data symbols) and best-fitted (solid lines) O (a) isotropic
17
chemical shift and (b) quadrupole parameters (C and  ) for [1,2- O ]BA-H and [1,2-
Q
Q
2
17
O ]BA-D as a function of temperature. The calculations were based on Eqs. (1)-(6) with
2
1
7
parameters given in Table 2. The error bars for experimental  ( O) values ( 1 ppm)
are not shown for clarity.
iso
2
5
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 26 of 46
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(
a)
3
50.0
00.0
A tautomer
B tautomer
3
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
250.0
200.0
150.0
y = -1.36x + 266
R = 0.986
2
100.0
-
100.0
0.0
100.0
1
7
Computed σ ( O) (ppm)
iso
(b)
10.0
y = 0.954x – 0.02
2
R = 0.987
5.0
0.0
A tautomer
B tautomer
-
5.0
-
10.0
-
10.0 -5.0
0.0
5.0
10.0
17
Computed χ ( O) (MHz)
ii
1
7
Figure 9. Comparison between experimental and computed O NMR parameters for
tautomers A (open circles) and B (filled circles) in several carboxylic acids. (a)
Experimental  versus computed  . (b) Experimental  versus computed  where ii
iso
iso
ii
ii
=
xx, yy, and zz.
2
6
ACS Paragon Plus Environment

Page 27 of 46
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
1
7
Table 1. Periodic plane-wave DFT computed O NMR parameters (magnetic shielding
constant, quadrupole coupling constant, and asymmetry parameter) for low- and high-
energy tautomers (A and B) and energy asymmetry (E) in several crystalline carboxylic
acids.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Tautomer A
Tautomer B
E/kJ mol–1
≈ 0
Compound
BA-H
 /ppm C /MHz 
 /ppm C /MHz 
iso
Q
Q
iso
Q
Q
C=O
–29 7.616
0.44
–24
7.477 0.47
–7.252 0.07
COH
79
–7.336 0.10
75
MCA
CA
0.8
1.3
3.7
18.0
C=O
COH
–9
66
7.171
0.61
–44
88
7.556 0.51
–7.507 0.18
–7.331 0.14
C=O
COH
–15 7.278
0.57
–49
83
7.644 0.45
–7.476 0.11
61 –7.175 0.08
ASA
SA
C=O
COH
–35 7.682
0.45
–49
74
7.802 0.40
–7.489 0.10
66
–7.402 0.08
C=O
COH
–3
86
6.628
0.78
–14
77
7.074 0.62
–7.224 0.23
–7.576 0.17
2
7
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 28 of 46
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
7
Table 2. Experimental O NMR parameters (chemical shift, quadrupolar coupling
a
constant, and asymmetry parameter) for low- and high-energy tautomers (A and B) and
energy asymmetry (E) values used in the analysis of O VT NMR data obtained for
several crystalline carboxylic acids.
b
17
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Tautomer A
Tautomer B
E/kJ mol–1
0.6
Compound
BA-H
 /ppm C /MHz 
 /ppm C /MHz 
iso
Q
Q
iso
Q
Q
C=O
300 7.5
0.3
0.0
314
8.0
0.3
COH
158 –6.5
152
–7.3 0.3
BA-D
MCA
CA
1.2
0.5
2.5
3.0
>15
C=O
COH
290 7.0
0.5
0.0
355
140
8.5
0.0
145 –6.5
–7.0 0.4
C=O
COH
278 6.5
0.6
0.0
315
150
7.0
0.1
180 –7.5
–6.0 0.2
C=O
COH
292 6.5
0.5
0.0
330
155
7.7
0.1
170 –6.2
–7.5 0.4
ASA
C=O
COH
302 7.8
0.2
0.0
340
160
6.0
0.1
170 –7.4
–7.5 0.60
SA
C=O
COH
280 7.1
0.5
0.2
—
—
—
—
—
—
165 –7.4
a
17
The uncertainties in the O isotropic chemical shifts and QC tensor components are
estimated to be  10 ppm and  1.0 MHz, respectively. Note that, because these
quantities were extracted from the best-fit plots shown in Figures 6 and 8, the
uncertainties were rather large.
b
–1
The uncertainty in E is estimated to be  0.2 kJ mol , except for the case of SA.
2
8
ACS Paragon Plus Environment

Page 29 of 46
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
Table 3. Summary of experimental hydrogen-bond OO distances determined by X-
ray/or neutron diffraction and energy asymmetry (E) values determined by various
techniques in crystalline carboxylic acid dimers.
Compounda,b dO…O/Å
E/kJ mol-1
0.42 0.04
0.79  0.06
Method
Ref.
4
19
22
39
26
28
32
This work
12
19
21
28
32
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
3
1
BA-H
2.608
C NMR and H T
1
1H T
1
0.418  0.128 2D spin diffusion
0
0
0
0
.50  0.04
Neutron diffraction
1
1
1
1
7
.2
O NQR
H T1
H T
.75
.71  0.02
1
7
0.6  0.2
1.3  0.1
O SSNMR
BA-D
IR
2
1
0
3
.5  0.1
H T
1
2
2
1
1
2
2
2
1
.72
H T1
H T1
.0
0.71
H T
O SSNMR
1
7
1
0.66
.2  0.2
This work
21
IA-D
2.623
2.631
2.634
2.630
2.608
2.69
H T1
H T1
H T1
H T1
DA-D
FA-D
TA
4-ClBA
GA-H
GA-D
1.31
0.12
1.0  0.2
1.66  0.12
3.3
5.8
1.4
2.8
2.0
6
34
12
Neutron diffraction
1
H T1
H T1
H T1
H T1
2
1
4
4
4
-FormBA-H 2.57
2
-FormBA-D
-DABA
2.622
2.653
2.634
2.674
2.609
2.608
2.580
2.610
2.661
2.634
2.601
2.581
2.614
2.630
2.642
2.654
X-ray diffraction
17O NQR
35
26
3,5-DNBA
0.8
4
4
4
4
-HBA
-NBA
-FBA
-ClBA
0.16  0.02
1.25  0.02
1.25  0.17
1.55  0.04
1.41  0.08
1.41  0.04
5.8
2.63  0.84
0.83
0.22
1H T1
30
4-BrBA
4
2
-IBA
-NBA
17O NQR
27
37

-9AC
X-ray diffraction
1
35
4A35ClBA
4A35IBA
MCA
H T and Cl NQR 39
1
17O SSNMR
This work
0.5  0.2
2.5  0.2
3.0  0.2
> 15
CA
ASA
SA
2
9
ACS Paragon Plus Environment