Solid state selection between nearly isoenergetic tautomeric forms driven by right hydrogen-bonding pairing

Article abstract of DOI:10.1021/acs.cgd.8b01158

We have prepared new hydroxyphenyl derivatives of [1,2,4]triazolo[3,2-c][1,2,4]triazole and investigated their tautomeric behavior, for the neutral and the monoprotonated forms, either computationally and experimentally in solution and the solid state. The results of our analysis indicate that the tautomeric behavior, in particular for the monoprotonated forms, is strongly dependent on the position of the OH group in the phenyl ring. The different position of the group (ortho, meta, para) has an effect on the relative energy of the various tautomeric forms of the cations but also on their ability to get an optimum packing, which is driven by directional hydrogen bonds. In particular, with the OH in para position, the two tautomers 1H-3H and 2H-3H of the singly protonated form have a similar energy and can be selectively isolated in the solid state with different counterions; with OH in the meta position, though the energy difference between the tautomers is even smaller, only the 2H-3H is present in the solid state, because of optimum packing interactions (hydrogen-bonding pairing); finally, with ortho OH, only the 1H-3H is isolated in the solid state because it has energy significantly lower than 2H-3H.

Full text of DOI:10.1021/acs.cgd.8b01158

Article
pubs.acs.org/crystal
Cite This: Cryst. Growth Des. XXXX, XXX, XXX−XXX
Solid State Selection between Nearly Isoenergetic Tautomeric
Forms Driven by Right Hydrogen-Bonding Pairing
Sandra Fusco,*,† Emmanuele Parisi,† Antonio Carella,† Amedeo Capobianco,*,‡ Andrea Peluso,‡
Carla Manfredi,† Fabio Borbone,† and Roberto Centore*,†
†Department of Chemical Sciences, University of Naples Federico II, Via Cintia, I-80126 Naples, Italy
‡Department of Chemistry and Biology, University of Salerno, Via Giovanni Paolo II, 132, I-84084 Fisciano, Salerno, Italy
S
* Supporting Information
ABSTRACT: We have prepared new hydroxyphenyl derivatives of
[1,2,4]triazolo[3,2-c][1,2,4]triazole and investigated their tautomeric
behavior, for the neutral and the monoprotonated forms, either computa-
tionally and experimentally in solution and the solid state. The results of
our analysis indicate that the tautomeric behavior, in particular for the
monoprotonated forms, is strongly dependent on the position of the OH
group in the phenyl ring. The different position of the group (ortho, meta,
para) has an effect on the relative energy of the various tautomeric forms of
the cations but also on their ability to get an optimum packing, which is
driven by directional hydrogen bonds. In particular, with the OH in para
position, the two tautomers 1H-3H and 2H-3H of the singly protonated
form have a similar energy and can be selectively isolated in the solid state
with different counterions; with OH in the meta position, though the
energy difference between the tautomers is even smaller, only the 2H-3H is present in the solid state, because of optimum
packing interactions (hydrogen-bonding pairing); finally, with ortho OH, only the 1H-3H is isolated in the solid state because it
has energy significantly lower than 2H-3H.
different outcomes for the crystallization from solution:
crystallization of only one tautomeric form (likely, but not
necessarily, the most stable in solution), cocrystallization of
different tautomers in the same lattice, concomitant formation
of crystals of different tautomeric forms. A recent detailed
analysis has shown that crystallization of different tautomers of
the same compound is a rather rare occurrence.13 In particular,
only about 0.5% of molecules able to tautomerize and archived
in the Cambridge Structural Database CSD14 are actually
observed in different tautomeric forms in the solid state. This
can be related, at least in part, to the fact that for many
potentially tautomeric systems, the energy difference between
the tautomers is rather high (>3 kcal/mol), and so only one
form is prevailing by far in solution and found in the crystals.
However, there are counterexamples to this simple explan-
ation, i.e., tautomeric couples with a higher energy difference
that are isolated in the solid state. At any rate, a reasonable
condition for the observation of different tautomers in the solid
state is to reduce their molecular energy difference, a target
that can be achieved by introducing suitable substituent groups
within the compound.
INTRODUCTION
■
Tautomers are structural isomers in ready equilibrium between
each other.1 In the most common case of tautomerism, i.e.,
prototropism, tautomers differ in the position of a hydrogen
atom inside the molecule and in the distribution of π electrons.
A remarkable example of the importance of tautomers in
chemistry is provided by the landmark second paper of Watson
and Crick on the structure of DNA.2 In that paper, the authors
proposed the hydrogen-bonding pairing scheme between
purine and pyrimidine bases, which is rooted in the most
probable tautomeric forms of the bases. They also raised the
hypothesis that less probable (i.e., noncanonical) tautomeric
forms of the bases could lead to wrong pairings and so to
mutations (see Supporting Information). This “rare tautomer”
hypothesis for spontaneous mutagenesis has been confirmed
over the years.3,4
In a certain sense, tautomers can be considered as “living
molecules”,5 because of the thermodynamic equilibrium they
undergo. In fact, the relative amounts of the different forms can
be altered by physical or chemical factors6−10 (temperature,
solvent, pH, mechanochemical processes, etc.), and so a
tautomeric system can face a change in ambient conditions by
favoring one form over the others. This responsiveness of
tautomeric systems is potentially appealing for the develop-
ment of smart materials and single-molecule-based devices.11,12
The dynamic equilibrium between the different tautomeric
forms of a compound in solution can lead, in principle, to
In the realm of fused-ring heteroaromatic systems that we
have studied over the years,15−17 we have found in the
Received: July 31, 2018
Revised: September 8, 2018
Published: September 11, 2018
© XXXX American Chemical Society
A
DOI: 10.1021/acs.cgd.8b01158
Cryst. Growth Des. XXXX, XXX, XXX−XXX

Crystal Growth & Design
Article
[1,2,4]triazolo[3,2-c][1,2,4]triazole, Chart 1, a heterocyclic
RESULTS AND DISCUSSION
■
system with a rich tautomeric behavior.18−20
Computational and Experimental Analysis in Sol-
ution. We have performed a preliminary computational
analysis of the energy of the tautomers of TT6 and TT7 in
the neutral and monoprotonated forms, to be compared with
the results for TT5 that we have already published.20 While for
TT5 a single conformer was found for the neutral and
monoprotonated species, in the case of TT6 and TT7 four
different conformers were found, differing for the orientation
of the hydroxy-phenyl ring and of the methyl group. The
conformers are indicated by a three-letter symbol, where each
letter can be c or t (e.g., tcc). The explanation of the symbol is
given in Chart 3 below.
a
Chart 1. Neutral and Singly Protonated Triazolotriazoles
Chart 3. Definition of the Conformers for TT6 and TT7
a
(a) The three canonical tautomeric forms of neutral [1,2,4]triazolo-
[3,2-c][1,2,4]triazole, with atom numbering shown for the 2H
tautomer. (b) The two lowest energy tautomers of monoprotonated
[1,2,4]triazolo[3,2-c][1,2,4]triazole (only one resonance form is
shown).
We have found that the relative energy of the three
tautomers of the neutral molecule, Chart 1a, and of the two
most stable tautomers of the monoprotonated species, Chart
1b, can be significantly modulated by acting on the
substituents, mostly in position 7, and on the polarity of the
solvent. For the neutral compounds, in all the cases
investigated, the energy trend of the tautomers is E(2H) <
E(3H) ≪ E(5H); in particular, the energy of the 5H tautomer
is always prohibitive (+10.8 kcal/mol with respect to 2H in the
most favorable case),18 while, for the 3H tautomer, with R = p-
nitrophenyl and R′ = methyl, an energy difference of +0.9
kcal/mol with respect to 2H has been calculated in water.18
For the monoprotonated species, Chart 1b, the pattern is more
intriguing. In fact, the energy difference between the two
tautomers is small (1.9 kcal/mol in the worst case);20
moreover, electron withdrawing groups in position 7 make
the tautomer 1H-3H more stable, while with electron donor
groups the tautomer 2H-3H is more stable.19 In the case of the
compound with R = 4-hydroxyphenyl and R′ = methyl, TT5 of
ref 20, the energy difference between the two tautomers is only
0.6 kcal/mol (1H-3H more stable) so they are present in a
comparable amount in solution and were selectively
precipitated by using different counterions (chloride for 1H-
3H and sulfate for 2H-3H).20
Neutral Molecules. The relative energies of tautomers
(2H, 3H, 5H) of neutral TT6 follow a quite similar trend to
TT5, see Table 1: 2H is the most stable, both in the gas phase
Table 1. Computed Relative Energies (kcal/mol), in
Reference to 2H, of Tautomers 3H and 5H of Neutral TT6,
a
TT7, and TT5
gas
water
3H
9.5
8.8 (8.2)
6.2 (6.0)
5H
23.0
21.8 (21.1)
19.3 (18.7)
3H
3.2
2.7 (2.7)
1.0 (0.9)
5H
13.9
13.4 (12.9)
12.1 (11.8)
b
c
TT5
TT6
TT7
c
a
Zero point vibration energy (ZPVE) corrected values are reported in
b
c
parentheses. Data for TT5 are taken from ref 20. The energy of the
most stable conformer of each tautomer is reported.
In the present paper we further extend the study of the
tautomeric behavior of derivatives of [1,2,4]triazolo[3,2-
c][1,2,4]triazole, by considering the two structural isomers of
TT5 in which the hydroxy group is in the meta (TT6) and
ortho (TT7) position, Chart 2.
and in polar medium; 3H is significantly stabilized in polar
medium, but its energy, as compared to 2H, is still high (2.7
kcal/mol) in order that it be found in appreciable amount.
On the other hand, for neutral TT7 the 3H tautomer, in
polar medium, is less stable than 2H by 1.0 kcal/mol only.
In Table 2 are reported the computed energies of the
different conformers of tautomers 2H of neutral TT6 and TT7
(data for conformers of TT7/3H are reported in Supporting
Information). It should be noted that the full counting of the
conformers gives eight outcomes (23 = 8). However, we have
found that the conformation of the methyl group is always c,
both for TT6 and TT7. In fact, even starting from methyl in
conformation t, the density functional theory (DFT) geometry
Chart 2. Chemical Diagrams of the Studied Compounds
(Only 2H Tautomer Shown)
B
DOI: 10.1021/acs.cgd.8b01158
Cryst. Growth Des. XXXX, XXX, XXX−XXX

Crystal Growth & Design
Article
Table 2. Predicted Relative Energies (kcal/mol) of the
Possible Conformers of TT6/2H and TT7/2H
Table 4. Predicted Relative Energies (kcal/mol) of the
a
Conformers of the Most Significant Tautomers of
a
Protonated TT6 and TT7 in Water
TT6/2H gas
TT6/2H water
TT7/2H gas
TT7/2H water
TT6/1H-3H
TT6/2H-3H
TT7/1H-3H
TT7/2H-3H
Eel
Eel
Eel
Eel
Eel
Eel
Eel
Eel
ccc
ctc
tcc
ttc
0.0 (0.0)
0.8 (0.7)
0.3 (0.3)
0.7 (0.6)
0.0 (0.0)
0.0 (0.1)
0.0 (0.2)
0.0 (0.2)
2.5 (2.4)
9.3 (8.8)
0.0 (0.0)
8.4 (8.0)
2.6 (2.2)
5.4 (5.0)
0.0 (0.0)
5.3 (5.0)
ccc
ctc
tcc
ttc
0.1 (0.0)
0.0 (0.0)
0.3 (0.4)
0.1 (0.1)
0.1 (0.3)
0.0 (0.1)
0.0 (0.1)
0.0 (0.0)
5.3 (5.0)
4.8 (4.5)
goes to ttc
0.0 (0.0)
3.8 (3.5)
3.3 (2.9)
0.0 (0.0)
3.6 (3.4)
a
Zero point vibration energy (ZPVE) corrected values are reported in
parentheses. The most stable conformation is highlighted in bold.
a
Zero point vibration energy (ZPVE) corrected values are reported in
parentheses. The most stable conformation is highlighted in bold.
optimization changes its conformation in c. Therefore, only
four possible conformers are predicted by computations.
Both in the gas phase and in water, the four conformers of
TT6/2H are almost isoenergetic. Instead, in the case of TT7/
2H (and TT7/3H, see Supporting Information) the two
conformers that allow the formation of the intramolecular
hydrogen bond with the ortho OH are by far more stable. In
particular, the conformer tcc, in which the intramolecular
hydrogen bond involves N1 as the acceptor, is predicted to be
more stable than ccc, in which N5 is the acceptor, by about 2.5
kcal/mol, as shown in Chart 4.
For singly protonated TT7, on the other hand, the tautomer
1H-3H is predicted to be far more stable than the others, and
for this tautomer, only one conformer is populated: ttc (Chart
5). Remarkably, in this tautomer, the intramolecular hydrogen
a
Chart 5. Two Conformers of TT7[1H-3H]+
Chart 4. Two Conformers of TT7/2H That Allow the
Formation of the Intramolecular Hydrogen Bond
a
Only one resonance form is shown.
bond between OH donor and N1 acceptor of the neutral
molecule (Chart 4) is no longer present, because the H atom
bonded to O is on the opposite side; there is, instead, an
intramolecular hydrogen bond between N−H donor and OH
acceptor, Chart 5.
For protonated TT7[1H-3H]+, a further conformation with
an O−H···N intramolecular hydrogen bond could be devised,
in which N5 would be the acceptor (conformer ccc of Table 4,
see also Chart 5). However, that conformation turns to be
energetically disfavored because, according to computations,
the positive charge of the molecule-ion would be mostly
localized on N6 (Chart 5).
Neutral TT6 and TT7 (H2L) can accept a proton, forming
the cationic species H3L+, and can release up to two protons,
forming the species HL− and L2−. In particular, the most acidic
hydrogen atom of the neutral molecules is N−H, and,
therefore, in the monoanionic HL− species, the phenolic O−
H hydrogen is retained.20 The acid−base equilibria in solution
have been studied at 25 °C by potentiometric-spectrophoto-
metric titrations in constant ionic medium NaCl 0.5 M/
ethanol 4% (v/v). UV−vis absorption spectra of TT6 recorded
at different pH and constant total concentration of TT6 are
reported in Figure 1a, and they show a remarkable dependence
on pH. In Figure 1b are reported the correspondent
absorbance vs pH data at four different wavelengths (similar
data for TT7 are reported in Supporting Information).
The calculated conditional equilibrium constants are
reported in Table 5.
Of course, the conformational analysis that we have
performed is only applicable to dilute solutions, because the
possible formation of dimers or n-mers with intermolecular
hydrogen bonds has not been considered.
Singly Protonated TT6 and TT7. Calculations for singly
protonated TT6 and TT7 have been performed in water. Out
of the possible six tautomers of singly protonated species,
tautomers 1H-3H and 2H-3H, see Chart 1b, are expected to be
the most stable, on the basis of our previous studies.19,20 The
tautomer 1H-2H has been tested for comparison.
Singly protonated TT6 and TT7 have several populated
conformers. For data of Table 3 the energy of the most stable
conformer for each tautomer has been considered.
In the case of TT6, singly protonated 1H-3H and 2H-3H
tautomers are predicted almost isoenergetic in solution, and for
both four conformers are populated (Table 4).
Table 3. Computed Relative Energy (kcal/mol, not
Including ZPVE) of the Most Relevant Tautomers of Singly
Protonated TT5, TT6, and TT7 in Water
a
TT5
TT6
TT7
1H-2H
1H-3H
2H-3H
13.8
0.0
0.6
14.6
0.0
0.2
13.6
0.0
2.5
The most remarkable difference in the data of Table 5 is in
the pKa3 values of TT5 and TT7, which differ by about 2 units.
This can be an electrostatic effect related to the space
separation between the negative charges of the dianion L2−,
a
Data for TT5 are taken from ref 20.
C
DOI: 10.1021/acs.cgd.8b01158
Cryst. Growth Des. XXXX, XXX, XXX−XXX

Crystal Growth & Design
Article
Figure 2. Distribution diagram of TT6, drawn using the equilibrium
constants of Table 5.
Crystallographic Analysis. The crystallographic analysis
has been performed on the singly protonated forms H3L+ of
TT6 and TT7, that will be indicated as TT6H+ and TT7H+. As
we have shown above, the computational analysis suggests that
for TT6H+ the two most stable tautomers, i.e., 1H-3H and 2H-
3H, are almost isoenergetic, and so, both being present in
solution, in principle both could be isolated in the crystals. For
TT7, on the other hand, only the 1H-3H is predicted to be
populated in solution and should be found in the crystals. In
order to assess the propensity of the different tautomers to
crystallize, in particular for TT6 for which computations
suggest nearly isoenergetic structures, we have prepared several
salts of the monocations, using different anions, both
monatomic (chloride, bromide) and molecular (perchlorate,
hydrogensulfate, sulfate).
Figure 1. (a): UV−vis absorption spectra of TT6 at constant total
concentration, cM = 3.5 × 10−5 M, in NaCl 0.5 M/ethanol 4% (v/v)
recorded at 0.3 ≤ pH ≤ 14; (b) absorbance/pH data, at four different
wavelengths, taken from panel a; the continuous curves were obtained
by fitting the experimental points using the Bouger−Lambert−Beer
equation (see Supporting Information).
The results for TT6H+ are shown in the Ortep diagrams of
Figures 3, 4, 5, 6, and 7.
Table 5. Acid−Base Reactions and Equilibrium Constants
(in NaCl 0.5 M/ethanol 4%, with Estimated Standard
Deviations in Parentheses) at 25 °C for TT5, TT6, and TT7
a
reaction
TT5
TT6
=
TT7
H3L+ + H2O =
pKa1
=
pKa1
pKa1 = 2.21(2)
H2L + H3O+
2.02(5)
2.45(3)
H2L + H2O =
HL− + H3O+
pKa2
=
pKa2
=
pKa2 = 6.55(3)
6.97(5)
6.62(4)
HL− + H2O =
pKa3
9.00(5)
=
pKa3
9.67(7)
=
pKa3
=
L2− + H3O+
10.98(6)
a
Data for TT5 are taken from ref 20.
that, for TT7, is considerably reduced, this possibly leading to
a diminution of the acid strength of HL−.
Figure 3. Ortep diagram of TT6·HCl·H2O. Thermal ellipsoids are
drawn at the 30% probability level. Selected hydrogen bonds are
indicated by dashed lines.
The behavior of TTn compounds in solution is complex.
Indeed, the final composition of the solution results from three
acid base reactions involving four different species and
regulated by relatively close values of pKa, as indicated by
the distribution diagram of TT6 reported in Figure 2.
Things are further complicated by tautomeric equilibria
affecting the singly protonated species. As an example, at pH =
2.45 and T = 298 K, TT6 exists as a 50% mixture of the neutral
and monoprotonated forms; the monoprotonated form, in
turn, is predicted to be an equilibrium mixture of the 1H-3H
(∼60%) and 2H-3H (∼40%) tautomers (see Figure S22 in
Supporting Information).
In all the cases, the 2H-3H tautomer is observed, in the
conformation ttc. The selective crystallization of this tautomer
seems independent of the anion (monatomic, molecular,
mononegative, dinegative) and of the presence of water
molecules in the lattice, so it cannot be happenstance.
Completely different results were obtained for TT5. In that
case, the two tautomers 1H-3H and 2H-3H, predicted to differ
in energy by 0.9 kcal/mol by DFT computations, were
selectively precipitated, respectively, as chloride and sulfate.20
We think that the persistent selection of the 2H-3H tautomer
D
DOI: 10.1021/acs.cgd.8b01158
Cryst. Growth Des. XXXX, XXX, XXX−XXX

Crystal Growth & Design
Article
in crystals of TT6 is related with a better pairing of cation
molecules through hydrogen bonding, as compared with the
1H-3H tautomer. This is illustrated in Chart 6, showing that
the formation of H-bonded dimers through symmetry
operations allowed in common space groups, for instance,
crystallographic inversion centers or binary rotation axes, is
accomplished through optimum hydrogen-bonding geometry
only in the case of the 2H-3H tautomer. In fact the two
possible dimers corresponding to 1H-3H tautomer (see Chart
6) are penalized by the lower acceptor capability of N2 atom in
the case of R22(8) and by the geometry, in the case of dimer
R22(14).
Actually, the R22(16) dimer of Chart 6 is present in all the
structures investigated of TT6H+ salts, with the only exception
of the bromide salt, so in four structures out of five (80%),
Figure 8 (see also Supporting Information for full packing
diagrams); moreover, if we consider that in some crystal
structures the independent unit contains more than one cation,
the R22(16) dimer is observed in seven cases out of eight
(88%). This suggests it is a highly robust synthon.21
Figure 4. Ortep diagram of TT6·HBr·2H2O. Thermal ellipsoids are
drawn at the 30% probability level. Selected hydrogen bonds are
indicated by dashed lines.
These arguments have been checked by energy calculations
on the dimers. In Table 6 are reported the interaction energies
of the dimers, Eint, defined as Eint = E(A2) − 2E(A), where
E(A2) is the energy of the optimized dimer A2 and E(A) is the
energy of the optimized monomer A.
Interaction energies are positive because the DFT
optimization of the dimers has been performed without
counterions, and so there is repulsion between the two cations
forming the dimer. However, the occurrence of zero first
derivatives for the nuclear configurations corresponding to the
optimized dimers and the concomitant occurrence of real and
positive vibrational frequencies indicate that the optimized
dimers do correspond to minima of energy. Of course they are
relative minima of the hypersurface of the potential energy
V(x1,y1,z1, ···, xN,yN,zN) [xi are the nuclear coordinates].
That said, the results of Table 6, that have been obtained
without imposing any constraint during the optimization,
indicate that for the dimer 2H-3H R22(16) the repulsion is
considerably lower than 1H-3H R22(14) (by about 6 kcal/mol)
and lower by far than 1H-3H R22(8) (by about 14 kcal/mol). If
we consider that the difference in latttice energy of polymorphs
is generally within 1−2 kcal/mol,22 we can exclude that
tautomer 1H-3H of TT6, though having energy 0.2 kcal/mol
lower than 2H-3H (see Table 3), be found in crystals, and this
because of packing interactions (H bonding pairing). More-
over, the analysis of the geometry of the optimized dimers (see
Figures S13−S15 in Supporting Information) shows that in the
case of 2H-3H R22(16) the dimer has C2h symmetry, as basically
found in the crystals. In the other two cases, repulsions
between nonbonded hydrogen atoms of the two monomers
induce a nonplanar geometry of the dimer that only retains Ci
symmetry. All the above arguments could also suggest that the
2H-3H R22(16) dimers are already present in concentrated
solutions of the salts23,24 and that crystallization of salts of the
TT6H+ cation can be actually regarded as crystallization of the
dimers.
Figure 5. Ortep diagram of TT6·HClO4. Thermal ellipsoids are
drawn at the 30% probability level. Selected hydrogen bonds are
indicated by dashed lines.
Figure 6. Ortep diagram of (TT6H)(HSO4). Thermal ellipsoids are
drawn at the 30% probability level. Selected hydrogen bonds are
indicated by dashed lines.
Figure 7. Partial Ortep diagram of 2[(TT6H)(HSO4)]·(TT6H)2SO4·
4H2O, in which only the sulfate ion and the two cation molecules
hydrogen-bonded to it are shown. Thermal ellipsoids are drawn at the
30% probability level. Only selected atoms are numbered for clarity.
Selected hydrogen bonds are indicated by dashed lines.
Ortep diagrams of the crystal structures of salts of TT7
(chloride, bromide, and perchlorate) are reported in Figures 9,
10, and 11. In all the cases the 1H-3H tautomer is present, as
expected on the basis of the energy computations. Interest-
ingly, the conformation of the cation detected in all crystals is
always ttc, in line with theoretical results, finding ttc more
E
DOI: 10.1021/acs.cgd.8b01158
Cryst. Growth Des. XXXX, XXX, XXX−XXX

Crystal Growth & Design
Article
Chart 6. Possible Pairing of TT6H+ Cations (ttc Conformer) through Hydrogen Bonds in Tautomers 2H-3H and 1H-3H
Figure 8. Partial packing diagrams of some salts of the TT6H+ cation. (a) TT6·HCl·H2O; (b) (TT6H)(HSO4); (c) TT6·HClO4; (d)
2[(TT6H)(HSO4)]·(TT6H)2SO4·4H2O. In some cases, water molecules are not shown for clarity. Selected hydrogen bonds are indicated by
dashed lines.
Table 6. Interaction Energies of Dimers of Chart 6, in kcal/
mol
dimer
Eint
2H-3H R22(16)
1H-3H R22(8)
1H-3H R22(14)
18.0
32.3
23.8
stable by ≈5 kcal/mol than the other conformations (see Table
4).
In particular, it is confirmed that the intramolecular
hydrogen bond between O−H donor and N1 acceptor,
present in the neutral molecule, is no longer present in the
protonated 1H-3H tautomer. This features a formal conforma-
tional switching of the OH hydrogen upon protonation,
simultaneous to the tautomeric 2H → 1H-3H switching, as
shown in Scheme 1.
Figure 9. Ortep diagram of TT7·HCl. Thermal ellipsoids are drawn at
the 30% probability level. Selected hydrogen bonds are indicated by
dashed lines.
EXPERIMENTAL SECTION
■
Materials and Methods. All reagents were analytical grade and
were used without further purification. Melting points were
determined by temperature controlled optical microscopy (Zeiss
Axioskop polarizing microscope equipped with a Mettler FP90
heating stage). NMR spectra were recorded with Bruker or Varian
spectrometers operating at 400 MHz, in CDCl3 or DMSO-d6. ESI
mass spectrometric analyses were recorded with an Applied
Biosystems API 2000 mass spectrometer equipped with an electro-
spray source used in the positive mode.
F
DOI: 10.1021/acs.cgd.8b01158
Cryst. Growth Des. XXXX, XXX, XXX−XXX

Crystal Growth & Design
Article
diaminotriazole with acetic anhydride at reflux gives the diacetylated
triazolo-triazole that, after basic hydrolysis and acid recovery, affords
pure TT6 and TT7. Single crystals of salts of monoprotonated TT6
and TT7 cations were prepared by slow evaporation of solutions
obtained by solving 10 mg of the neutral compound in 20 mL of 1 M
solutions of the inorganic acid (HCl, HBr, HClO4, H2SO4).
Acid−Base Equilibria. The protolytic equilibria of TT6 and TT7
were studied by UV−vis absorption spectroscopy in 0.5 M NaCl−4%
ethanol (v/v), as the ionic medium, following the same procedure
described in ref 20. The experiments were performed as acid−base
titrations at constant total concentration of TT6/TT7 without varying
the 0.5 M concentration of Cl−. The investigated pH range extends
from 0.3 to 14. For each experimental point, the equilibrium free
proton concentration was evaluated from the measured electromotive
force at the ends of the galvanic cell GE/TS/RE, where TS indicates
the test solution, GE is the glass electrode, and RE is a reference
electrode (0.5 M NaCl|Hg2Cl2|Hg(Pt)) placed outside but electrically
connected to TS through a salt bridge. All the experiments were
carried out in air in a thermostat, at 25.00 0.03 °C. Potentiometric
experimental data were collected by means of an automatic data
acquisition system based on Hewlett-Packard (HP) instrumentation.
Coulometric variations of the solution composition were carried out
using a Hewlett-Packard “DC Power Supply”. Absorption spectra
were recorded with a Varian Cary 50 UV−vis spectrophotometer
using 1 cm cell. The primary spectrophotometric data (A, pH, λ) were
numerically analyzed by the computerized program HYPERQUAD.26
A detailed description of the mathematical procedures is given in
Supporting Information.
Figure 10. Ortep diagram of TT7·HBr·3H2O. Thermal ellipsoids are
drawn at the 30% probability level. Selected hydrogen bonds are
indicated by dashed lines.
Crystallography. All data for crystal structure determinations
were measured on a Bruker-Nonius KappaCCD diffractometer
equipped with Oxford Cryostream 700 apparatus, using graphite
monochromated MoKα radiation (λ = 0.71073 Å). Reduction of data
and semiempirical absorption correction were done using SADABS
program.27 The structures were solved by direct methods (SIR97
program28) and refined by the full-matrix least-squares method on F2
using SHELXL-2016/6 program29 with the aid of the program
WinGX.30 H atoms bonded to C were generated stereochemically and
refined by the riding model. After having placed C-bound H atoms,
those bonded to O and N, that are essential in the identification of
tautomers, were clearly found in difference Fourier maps as the first
maxima and their coordinates were refined with some restraints on
bond length. For all H atoms Uiso was set at 1.2 times Ueq of the
carrier atom (1.5 in the case of methyl group). The analysis of the
crystal packing was performed using the program Mercury.31 CCDC
1857839−1857846 contain the supplementary crystallographic data
for this article.
̅
TT6·HCl·H2O: C10H12ClN5O2, Mr = 269.70, triclinic, P1, Z = 2,
Figure 11. Ortep diagram of TT7·HClO4. Thermal ellipsoids are
drawn at the 30% probability level. Selected hydrogen bonds are
indicated by dashed lines. Only half of the crystallographic
independent unit is shown. The solvent water molecule is not shown.
colorless, a = 5.7230(10) Å, b = 7.365(2) Å, c = 13.889(4) Å, α =
88.539(10)°, β = 88.740(10)°, γ = 82.39(2)°, V = 580.0(3) Å3, T =
293 K, 4809 reflns collected, 2605 unique (Rint = 0.0231), R = 0.0418
(I > 2σ(I)), wR (all) = 0.1164, GOF = 1.066, final max/min peak
0.243/−0.230 e/Å3. CCDC 1857839.
2[TT6·HBr]·3H2O: C20H26Br2N10O5, Mr = 646.32, monoclinic,
C2/c, Z = 4, colorless, a = 23.746(7) Å, b = 6.751(3) Å, c = 16.917(5)
Å, α = 90°, β = 102.32(2)°, γ = 90°, V = 2649.5(16) Å3, T = 293 K,
12724 reflns collected, 2965 unique (Rint = 0.0374), R = 0.0470 (I >
2σ(I)), wR (all) = 0.1047, GOF = 1.115, final max/min peak 0.611/−
0.391 e/Å3. CCDC 1857840.
Scheme 1. Formal Conformational and Tautomeric
Switching of TT7 upon Protonation
TT6·HClO4: C10H10ClN5O5, Mr = 315.68, monoclinic, C2/m, Z =
4, colorless, a = 13.107(5) Å, b = 7.004(3) Å, c = 13.851(6) Å, α =
90°, β = 100.36(2)°, γ = 90°, V = 1250.8(9) Å3, T = 293 K, 6522
reflns collected, 1547 unique (Rint = 0.0403), R = 0.0416 (I > 2σ(I)),
wR (all) = 0.1049, GOF = 1.045, final max/min peak 0.311/−0.400
e/Å3. CCDC 1857841.
(TT6H)(HSO4): C10H11N5O5S, Mr = 313.30, monoclinic, C2/c, Z
= 8, colorless, a = 17.195(5) Å, b = 6.242(3) Å, c = 24.702(6) Å, α =
90°, β = 107.03(3)°, γ = 90°, V = 2535.1(16) Å3, T = 293 K, 4992
reflns collected, 2764 unique (Rint = 0.0243), R = 0.0414 (I > 2σ(I)),
wR (all) = 0.1116, GOF = 1.057, final max/min peak 0.240/−0.367
e/Å3. CCDC 1857842.
General Synthetic Procedures. The synthesis of TT6 and TT7
was performed according to the procedure already described by us,20
and that is shortly outlined here (full details are given in Supporting
Information). Starting from 3-hydroxybenzoic acid (for TT6) or
salicylic acid (for TT7), reaction with diaminoguanidine monohydro-
chloride in polyphosphoric acid (PPA), at 150 °C overnight, affords
5-(3-hydroxyphenyl)-3,4-diamino-1,2,4-triazole and 5-(2-hydroxy-
phenyl)-3,4-diamino-1,2,4-triazole, respectively.25 Reaction of the
G
DOI: 10.1021/acs.cgd.8b01158
Cryst. Growth Des. XXXX, XXX, XXX−XXX

Crystal Growth & Design
Article
1
Detailed synthetic procedures, including H NMR, 13C
NMR and mass spectra of final compounds and
intermediates; detailed description of the UV−vis and
electrochemical methods used in the study of acid−base
properties; geometry of DFT optimized H-bonded
dimers; tables of H-bonding distances from single crystal
X-ray analysis; full packing diagrams. Observed and
calculated UV−vis spectrum of TT6 at very acidic pH
(PDF)
2[(TT6H)(HSO4)]·(TT6H)2SO4·4H2O: C40H50N20O20S3, Mr =
̅
1227.18, triclinic, P1, Z = 2, colorless, a = 12.997(3) Å, b = 14.596(4)
Å, c = 14.977(7) Å, α = 77.73(7)°, β = 78.870(13)°, γ = 64.30(8)°, V
= 2485(2) Å3, T = 173 K, 42899 reflns collected, 10903 unique (Rint
=
0.0632), R = 0.0562 (I > 2σ(I)), wR(all) = 0.1395, GOF = 1.062, final
max/min peak 0.575/−0.640 e/Å3. CCDC 1857843.
TT7·HCl: C10H10ClN5O, Mr = 251.68, monoclinic, P21/c, Z = 4,
colorless, a = 7.044(3) Å, b = 8.934(4) Å, c = 18.301(7) Å, α = 90°, β
= 105.253(10)°, γ = 90°, V = 1111.1(8) Å3, T = 293 K, 6662 reflns
collected, 2458 unique (Rint = 0.0790), R = 0.0593 (I > 2σ(I)), wR
(all) = 0.1303, GOF = 1.033, final max/min peak 0.284/−0.340 e/Å3.
CCDC 1857844.
Accession Codes
̅
TT7·HBr·3H2O: C10H16BrN5O4, Mr = 350.19, triclinic, P1, Z = 2,
CCDC 1857839−1857846 contain the supplementary crys-
tallographic data for this paper. These data can be obtained
free of charge via www.ccdc.cam.ac.uk/data_request/cif, or by
emailing data_request@ccdc.cam.ac.uk, or by contacting The
Cambridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK; fax: +44 1223 336033.
colorless, a = 6.7830(12) Å, b = 9.6050(15) Å, c = 12.7370(11) Å, α
= 106.651(9)°, β = 96.355(10)°, γ = 102.920(11)°, V = 761.2(2) Å3,
T = 293 K, 8939 reflns collected, 3407 unique (Rint = 0.0393), R =
0.0503 (I > 2σ(I)), wR (all) = 0.1252, GOF = 1.082, final max./min
peak 0.479/-0.482 e/Å3. CCDC 1857845.
̅
2[TT7·HClO4]·H2O: C20H12Cl2N10O11, Mr = 649.38, triclinic, P1,
Z = 2, colorless, a = 8.0260(5) Å, b = 13.3910(13) Å, c = 13.8430(10)
Å, α = 112.786(7)°, β = 100.887(7)°, γ = 96.787(7)°, V =
1316.57(19) Å3, T = 293 K, 22073 reflns collected, 5988 unique (Rint
= 0.0316), R = 0.0459 (I > 2σ(I)), wR(all) = 0.1283, GOF = 1.055,
final max./min peak 0.647/−0.353 e/Å3. CCDC 1857846.
AUTHOR INFORMATION
Corresponding Authors
*(S.F.) E-mail: sandra.fusco@unina.it.
*(A.C.) E-mail: acapobianco@unisa.it.
*(R.C.) E-mail: roberto.centore@unina.it.
■
Computational Details. Quantum chemical computations were
carried with the Gaussian 09 package by using density functional
theory (DFT).32 The BMK meta functional33 was employed
throughout in conjunction with the 6-31+G** basis set. Owing to
its high fraction (42%) of HF exchange which mitigates the self-
interaction problem,34 BMK has proven to give excellent perform-
ance, nearly reproducing experimental electrical and optical properties
for donor−acceptor systems,35 matching the quality of highly
parametrized functionals, correlated post-Hartee-Fock and ad hoc
parametrized methods.36−39 Solvent (water) effects were included by
the polarizable continuum model (PCM).40 Starting geometries of the
dimeric patterns reported in Chart 6 were generated in planar
configuration. Optimized structures for single molecules and hydro-
gen-bonded dimers were obtained by DFT computations carried out
without imposing any geometrical constraint. The nature of located
stationary points was verified by checking the eigenvalues of the
Hessian matrix; all the minimum energy structures have positive
eigenvalues.
ORCID
Amedeo Capobianco: 0000-0002-5157-9644
Andrea Peluso: 0000-0002-6140-9825
Fabio Borbone: 0000-0001-7433-9267
Roberto Centore: 0000-0002-2797-0117
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
Thanks are due to COST association for support and critical
discussions within the COST Action CM-1402-Crystallize and
to the CRdC NTAP of Regione Campania (Italy) for the X-ray
facility.
REFERENCES
■
CONCLUSIONS
■
(1) Entry “tautomerism” in IUPAC, Compendium of Chemical
Terminology, 2nd ed. (the “Gold Book”); Compiled by McNaught,
A. D.; Wilkinson, A., Blackwell Scientific Publications: Oxford, 1997.
(2) Watson, J. D.; Crick, F. H. C. Genetical implications of the
structure of deoxyribonucleic acid. Nature 1953, 171, 964−967.
(3) Goodman, M. F. Mutations caught in the act. Nature 1995, 378,
237−238.
(4) Wang, W.; Hellinga, H. W.; Beese, L. S. Structural evidence for
the rare tautomer hypothesis of spontaneous mutagenesis. Proc. Natl.
Acad. Sci. U. S. A. 2011, 108, 17644−17648.
(5) Martin, Y. C. Let’s not forget tautomers. J. Comput.-Aided Mol.
Des. 2009, 23, 693−704.
(6) Desiraju, G. R. Intramolecular Proton Transfers in the Solid
State. Conversion of the Hydroxyazo into the Quinone Hydrazone
Tautomer of 2-Amino-3-hydroxy-6-phenylazopyridine. X-Ray Crystal
Structures of the Two Forms. J. Chem. Soc., Perkin Trans. 2 1983,
1025−1030.
(7) Hanus, M.; Ryjacek, F.; Kabelac, M.; Kubar, T.; Bogdan, T. V.;
Trygubenko, S. A.; Hobza, P. Correlated ab Initio Study of Nucleic
Acid Bases and Their Tautomers in the Gas Phase, in a
Microhydrated Environment and in Aqueous Solution. Guanine:
Surprising Stabilization of Rare Tautomers in Aqueous Solution. J.
Am. Chem. Soc. 2003, 125, 7678−7688.
(8) Zhang, C.; Xie, L.; Ding, Y.; Sun, Q.; Xu, W. Real-Space
Evidence of Rare Guanine Tautomer Induced by Water. ACS Nano
2016, 10, 3776−3782.
Within the class of 4-methyl-7-hydroxyphenyl-[1,2,4]triazolo-
[3,2-c][1,2,4]triazoles, the placement of the OH group in
different positions on the phenyl ring (ortho, meta, para) has
an effect on the energy of the single different tautomer
molecules. However, we have shown that it also affects the
energy of the modes of pairing of the different tautomer
molecules via strong hydrogen bonding, and hence the
selective crystallization of the tautomeric forms. A particularly
striking evidence of the latter issue is observed, for the
monoprotonated triazolo-triazolium cations, in the case of the
para and meta positions. In fact, even if the energy difference of
the tautomers 1H-3H and 2H-3H is small for singly protonated
TT5 (para OH) and even smaller for singly protonated TT6
(meta OH), the outcome of the crystallization experiments is
completely different: only 2H-3H for TT6, independent of the
counterion, either 1H-3H or 2H-3H for TT5 depending on the
counterion.
́
̌
́
̌
̌
ASSOCIATED CONTENT
■
S
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.cgd.8b01158.
H
DOI: 10.1021/acs.cgd.8b01158
Cryst. Growth Des. XXXX, XXX, XXX−XXX

Crystal Growth & Design
Article
(9) Marshall, M. G.; Lopez-Diaz, V.; Hudson, B. S. Single-Crystal X-
ray Diffraction Structure of the Stable Enol Tautomer Polymorph of
Barbituric Acid at 224 and 95 K. Angew. Chem., Int. Ed. 2016, 55,
1309−1312.
(10) Ladenthin, J. N.; Frederiksen, T.; Persson, M.; Sharp, J. C.;
Gawinkowski, S.; Waluk, J.; Kumagai, T. Force-induced tautomeriza-
tion in a single molecule. Nat. Chem. 2016, 8, 935−940.
(11) Bussetti, G.; Campione, M.; Riva, M.; Picone, A.; Raimondo,
L.; Ferraro, L.; Hogan, C.; Palummo, M.; Brambilla, A.; Finazzi, M.;
(30) Farrugia, L. J. WinGX and Ortep for Windows: an update. J.
Appl. Crystallogr. 2012, 45, 849−854.
(31) Macrae, C. F.; Bruno, I. J.; Chisholm, J. A.; Edgington, P. R.;
McCabe, P.; Pidcock, E.; Rodriguez-Monge, L.; Taylor, R.; van de
Streek, J.; Wood, P. A. Mercury CSD 2.0 − new features for the
visualization and investigation of crystal structures. J. Appl. Crystallogr.
2008, 41, 466−470.
(32) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci,
B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H.
P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.;
Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.;
Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.;
Montgomery, J. A., Jr.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.;
Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Kobayashi,
R.; Normand, J.; Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar,
S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, J. M.; Klene, M.; Knox,
J. E.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.;
Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.;
Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.;
Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A.
D.; Farkas, O.; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J.,
Gaussian 09, Revision A02; Gaussian Inc.: Wallingford, CT, 2009.
(33) Boese, A. D.; Martin, J. M. L. Development of Density
Functionals for Thermochemical Kinetics. J. Chem. Phys. 2004, 121,
3405−3416.
(34) Peach, M. J. G.; Benfield, P.; Helgaker, T.; Tozer, D. J.
Excitation Energies in Density Functional Theory: An Evaluation and
a Diagnostic Test. J. Chem. Phys. 2008, 128, 044118−8.
(35) Capobianco, A.; Centore, R.; Noce, C.; Peluso, A. Molecular
hyperpolarizabilities of push−pull chromophores: A comparison
between theoretical and experimental results. Chem. Phys. 2013,
411, 11−16.
(36) Suponitsky, K. Y.; Tafur, S.; Masunov, A. E. Applicability of
Hybrid Density Functional Theory Methods to Calculation of
Molecular Hyperpolarizability. J. Chem. Phys. 2008, 129, 044109−15.
(37) Capobianco, A.; Centore, R.; Fusco, S.; Peluso, A. Electro-
optical properties from CC2 Calculations: A comparison between
theoretical and experimental results. Chem. Phys. Lett. 2013, 580,
126−129.
̀
Duo, L.; Sassella, A.; Ciccacci, F. Stable Alignment of Tautomers at
Room Temperature in Porphyrin 2D Layers. Adv. Funct. Mater. 2014,
24, 958−963.
(12) Horiuchi, S.; Kobayashi, K.; Kumai, R.; Ishibashi, S. Proton
tautomerism for strong polarization switching. Nat. Commun. 2017, 8,
14426.
(13) Cruz-Cabeza, A. J.; Groom, C. R. Identification, classification
and relative stability of tautomers in the cambridge structural
database. CrystEngComm 2011, 13, 93−98.
(14) Allen, F. H. The Cambridge Structural Database: a quarter of a
million crystal structures and rising. Acta Crystallogr., Sect. B: Struct.
Sci. 2002, 58, 380−388.
(15) Centore, R.; Concilio, S.; Panunzi, B.; Sirigu, A.; Tirelli, N.
Nonlinear Optical Properties of Some Side Chain Copolymers Based
on Benzoxazole Containing Chromophores. J. Polym. Sci., Part A:
Polym. Chem. 1999, 37, 603−608.
(16) Carella, A.; Centore, R.; Mager, L.; Barsella, A.; Fort, A.
Crosslinkable organic glasses with quadratic nonlinear optical activity.
Org. Electron. 2007, 8, 57−62.
(17) Fusco, S.; Centore, R.; Riccio, P.; Quatela, A.; Stracci, G.;
Archetti, G.; Kuball, H.-G. NLO-active polymers containing triazolo-
thiadiazole segments. Polymer 2008, 49, 186−191.
(18) Centore, R.; Fusco, S.; Capobianco, A.; Piccialli, V.; Zaccaria,
S.; Peluso, A. Tautomerism in the Fused N-Rich Triazolotriazole
Heterocyclic System. Eur. J. Org. Chem. 2013, 2013, 3721−3728.
(19) Centore, R.; Manfredi, C.; Fusco, S.; Maglione, C.; Carella, A.;
Capobianco, A.; Peluso, A.; Colonna, D.; Di Carlo, A. Proton induced
tautomeric switching in N-rich aromatics with tunable acid-base
character. J. Mol. Struct. 2015, 1093, 119−124.
(20) Centore, R.; Manfredi, C.; Capobianco, A.; Volino, S.; Ferrara,
M. V.; Carella, A.; Fusco, S.; Peluso, A. Solid State Separation and
Isolation of Tautomers of Fused-Ring Triazolotriazoles. J. Org. Chem.
2017, 82, 5155−5161.
(21) Allen, F. H.; Motherwell, W. D. S.; Raithby, P. R.; Shields, G.
P.; Taylor, R. Systematic analysis of the probabilities of formation of
bimolecular hydrogen-bonded ring motifs in organic crystal
structures. New J. Chem. 1999, 23, 25−34.
(22) Cruz-Cabeza, A. J.; Reutzel-Edens, S. M.; Bernstein, J. Facts
and fictions about polymorphism. Chem. Soc. Rev. 2015, 44, 8619−
8635.
(23) Davey, R. J.; Allen, K.; Blagden, N.; Cross, W. I.; Lieberman, H.
F.; Quayle, M. J.; Righini, S.; Seton, L.; Tiddy, G. J. T. Crystal
engineering − nucleation, the key step. CrystEngComm 2002, 4, 257−
264.
(38) Castaldo, A.; Centore, R.; Peluso, A.; Sirigu, A.; Tuzi, A.
Structure and Electronic Properties of Extended Chromophores for
Applications in Second-Order Nonlinear Optics. Struct. Chem. 2002,
13, 27−36.
(39) Capobianco, A.; Esposito, A.; Caruso, T.; Borbone, F.; Carella,
A.; Centore, R.; Peluso, A. Tuning Wavefunction Mixing in Push−
Pull Molecules: From Neutral to Zwitterionic Compounds. Eur. J.
Org. Chem. 2012, 2012, 2980−2989.
̌
(40) Miertus, S.; Scrocco, E.; Tomasi, J. Electrostatic Interaction of a
Solute with a Continuum. A Direct Utilization of Ab Initio Molecular
Potentials for the Prevision of Solvent Effects. Chem. Phys. 1981, 55,
117−129.
̀
(24) Causa, M.; Centore, R. Actual and virtual structures in
molecular crystals. CrystEngComm 2017, 19, 1320−1327.
(25) Centore, R.; Carella, A.; Fusco, S. Supramolecular synthons in
fluorinated and nitrogen-rich ortho-diaminotriazoles. Struct. Chem.
2011, 22, 1095−1103.
(26) Gans, P.; Sabatini, A.; Vacca, A. Investigation of equilibria in
solution. Determination of equilibrium constants with the HYPER-
QUAD suite programs. Talanta 1996, 43, 1739−1753.
(27) SADABS; Bruker-Nonius: Delft, The Netherlands, 2002.
(28) Altomare, A.; Burla, M. C.; Camalli, M.; Cascarano, G. L.;
Giacovazzo, C.; Guagliardi, A.; Moliterni, G. G.; Polidori, G.; Spagna,
R. SIR97: a new tool for crystal structure determination and
refinement. J. Appl. Crystallogr. 1999, 32, 115−119.
(29) Sheldrick, G. M. Crystal structure refinement with. Acta
Crystallogr., Sect. C: Struct. Chem. 2015, 71, 3−8.
I
DOI: 10.1021/acs.cgd.8b01158
Cryst. Growth Des. XXXX, XXX, XXX−XXX