Ranking relative hydrogen-bond strengths in hydroxybenzoic acids for crystal-engineering purposes

Article abstract of DOI:10.1002/chem.201301402

Systematic co-crystallizations resulting in a total of six new crystal structures involving either 3-hydroxy- or 4-hydroxybenzoic acid, complemented by calculated molecular electrostatic potential surfaces and existing structural data, have shown that in a competitive molecular recognition situation, the -OH moiety is a more effective hydrogen-bond donor than the -COOH moiety which, in turn, highlights that electrostatic charge can offer more useful guidance than acidity for predicting competitive hydrogen-bond preferences.

Full text of DOI:10.1002/chem.201301402

FULL PAPER
DOI: 10.1002/chem.201301402
Ranking Relative Hydrogen-Bond Strengths in Hydroxybenzoic Acids for
Crystal-Engineering Purposes
Christer B. Aakerçy,* Kanishka Epa, Safiyyah Forbes, Nathan Schultheiss, and
John Desper[a]
Abstract: Systematic co-crystallizations resulting in a total of six new crystal struc-
tures involving either 3-hydroxy- or 4-hydroxybenzoic acid, complemented by cal-
Keywords: co-crystals · crystal en-
culated molecular electrostatic potential surfaces and existing structural data, have
gineering · hierarchy · hydrogen
À
shown that in a competitive molecular recognition situation, the OH moiety is a
bonding · intermolecular interac-
tions
À
more effective hydrogen-bond donor than the COOH moiety which, in turn,
highlights that electrostatic charge can offer more useful guidance than acidity for
predicting competitive hydrogen-bond preferences.
Introduction
thetic strategies that have a reasonable chance of success
can be built.[13]
The deliberate and targeted synthesis of co-crystals[1] with
desired stoichiometries and intermolecular connectivities
relies on our ability to apply the concepts of tectons[2] and
synthons[3] in a rational and reproducible manner. The task
of identifying synthons in the first place is facilitated greatly
through careful analysis[4] of the structural information con-
tained in the CSD.[5] However, it is not always possible to
find enough data to allow a reliable ranking of the relative
importance of different synthons, especially in structurally
competitive situations. It is undoubtedly highly desirable to
establish robust guidelines for supramolecular synthesis, be-
cause multicomponent solid-state architectures have found
applications in areas involving pharmaceuticals[6], agrochem-
icals[7], nonlinear optics,[8] explosives[9], and organic semicon-
ductors[10]. In principle, co-crystals can offer a wide selection
of solid forms of a particular active ingredient, which, in
turn, improves our chances of optimizing physical properties
of a solid, without tampering with the chemical nature of
the key component.
As a starting point, it is known that certain molecules and
functional groups exhibit a higher propensity to form co-
crystals than others.[14] The hydrogen-bond-based selectivity
found in molecular solids was rationalized by Etter: “The
best hydrogen-bond donor and the best hydrogen-bond ac-
ceptor will preferentially form hydrogen bonds to one an-
other”.[15] A consequence of this statement is that we need
to acquire reliable means for ranking hydrogen-bond donor
and acceptor groups to develop transferable robust and
high-yielding[16] supramolecular design strategies based on
hydrogen bonds.
Several approaches have been proposed for quantifying
hydrogen-bond strength on a thermodynamic basis,[17,18,19]
and, in some systems, a successful ranking has been ach-
ieved by using pKa/pKb values of the participating compo-
nents.[20,21] However, it is important to note that the latter
quantities can only provide a useful guide when a series of
molecules carrying the same chemical functionality is being
examined. A more general strategy for ranking and compar-
ing hydrogen-bond donor strength across a broad spectrum
of chemical functionalities based on electrostatic charge has
been developed by Hunter and co-workers.[22,23] This ap-
proach utilizes calculated molecular electrostatic potential
surfaces (by using AM1 or DFT) around the molecule, in
which potential maxima and minima correspond to hydro-
gen-bond donor and acceptor sites, respectively.
The reversible nature of intermolecular interactions and
the fact that they are relatively weak compared to most co-
valent bonds means that we will not always be able to pro-
duce a strict hierarchy of hydrogen-bond-based supramolec-
ular synthesis that completely avoids “synthon crossover”[11]
or “synthon polymorphism”.[12] Instead, the goal is to estab-
lish guidelines that provide a framework around which syn-
Herein, we examine two common functional groups, a car-
boxylic acid and a phenolic moiety in the context of hydro-
gen-bond strength and their ability to compete for hydro-
gen-bond acceptors with differing strengths.
Carboxylic acids are frequently assumed to be “better” or
more effective hydrogen-bond donors than phenols. When
considering 4-hydroxybenzoic acid, for example, pKa values
[a] Dr. C. B. Aakerçy, K. Epa, Dr. S. Forbes, Dr. N. Schultheiss,
Dr. J. Desper
Department of Chemistry, Kansas State University
213 CBC Building, Manhattan, KS 66506-0401 (USA)
Fax : (+1)785-532 6666
E-mail: aakeroy@ksu.edu
À
alone would suggest that the COOH group (pKa =4.48) is
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201301402.
À
a far superior hydrogen-bond donor to the OH group
Chem. Eur. J. 2013, 00, 0 – 0
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
1
&
ÞÞ
These are not the final page numbers!

(pKa =9.32). Molecular electrostatic potential calculations
acidity offer a better method for predicting synthon prefer-
ence in this family of ditopic hydrogen-bond donors.
À
on the other hand, show that the OH group could be
viewed as a highly competitive, or possibly even better
À
donor compared to the COOH group (Scheme 1). One
Results and Discussion
Formation of a co-crystal was readily determined by IR
À
spectroscopy through the appearance of an O H···N stretch
or by shifts to the carbonyl band. The data listed in Table 1
Table 1. Relevant IR spectroscopic data from all 16 solids obtained in
this study.
Acceptor
3-HBA
4-HBA
À
À
O H···N
C==O
O H···N
C==O
–
–
1681
1698
1692
1687
1665
1693
1694
1692
1693
–
–
1669
1691
1681
1666
1686
1666
1667
1667
1671
À
Scheme 1. MEP surface calculations show the OH group to be the best
1
2
3
4
5
6
7
8
À
donor D1 and the COOH moiety to be the second-best donor D2.
1932
1935
1920
1928
1926
1935
1932
1904
1940
1877
1864
1929
1912
1912
way to probe the practical consequences of intermolecular
interactions is to perform systematic co-crystallizations, in
which molecules carrying specific functional groups are con-
fronted with a variety of potential “partners”. By examining
binding preferences by using crystallographic data, it may
be possible to begin to formulate guidelines that will allow
us to predict which synthons are most likely to form in a
competitive situation. To determine how to rank the relative
hydrogen-bond capability of OH and COOH groups
when they are attached to the same molecular backbone, 3-
hydroxybenzoic acid, 3-HBA, and 4-hydroxybenzic acid, 4-
HBA, was co-crystallized with a series of ditopic molecules
each containing two hydrogen-bond acceptor sites of differ-
ent strength, Scheme 2. The fundamental question that we
want to address is whether a ranking based on charge or on
indicate that a co-crystal was formed in each of the sixteen
cases. Although vibrational spectroscopy offers an unambig-
uous assessment of reaction outcome, it is not possible to
elucidate the precise binding modes and the presence of
specific synthons in this family of compounds; this required
single-crystal diffraction data, which were obtained for six
compounds.
Despite considerable efforts, we were unable to grow suit-
able single crystals for any additional co-crystals; the 3-
HBA based co-crystals were particularly troublesome, be-
cause most of them produced oils upon re-crystallization.
The crystal structure determination of 4-HBA:1 shows
that in the resulting 1:1 co-crystal the best donor (as was de-
À
À
À
termined by MEP values), the OH moiety, forms a hydro-
gen bond to the N-oxide oxygen atom, the best acceptor,
À
(O24···O11 2.6442(11) ꢁ, O24 H24···O11 1.763(17) ꢁ), and
the second-best donor, the carboxylic moiety, engages in a
hydrogen bond with the pyridyl nitrogen atom, the second-
À
best acceptor (O21···N14 2.7203(11) ꢁ, O21 H21···N14
1.756(16) ꢁ; Figure 1).
The crystal structure determination of 4-HBA:3 showed
À
that the outcome is a 1:1 co-crystal, in which the OH
moiety forms a hydrogen bond to the N-oxide site, the best
À
acceptor (O34···O21 2.590 ꢁ, O34 H34···O21 1.703 ꢁ). This
Scheme 2. A1 and A2 are assigned based on calculated AM1 charges and
Figure 1. Primary hydrogen-bond interactions in the crystal structure of
refer to best- and second-best acceptor, respectively.
4-HBA:1.
&
2
&
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Hydrogen-Bond Strengths in Hydroxybenzoic Acids
FULL PAPER
À
leaves the second-best donor, the COOH group, free to
form a hydrogen bond with the second-best acceptor, the
nitrogen atom (O43···N13 2.746(2) ꢁ) and the second-best
À
donor COOH forms a hydrogen bond with the pyridyl ni-
À
pyridyl nitrogen atom (O31···N11 2.616(3) ꢁ, O31
trogen atom, the second-best acceptor (O48···N31
H31···N11 1.57(3) ꢁ; Figure 2).
2.683(3) ꢁ; Figure 5).
Figure 2. Primary hydrogen-bond interactions in the crystal structure of
4-HBA:3.
The structure determination of 4-HBA:5 revealed a 1:1
Figure 5. Primary hydrogen-bond interactions in the crystal structure of
3-HBA:7.CH3CN.H2O.
À
co-crystal, in which the OH moiety forms a hydrogen bond
À
to the benzimidazole site (O34···N13 2.7238(13) ꢁ, O34
H34···N13 1.819(18) ꢁ). Similar to the previous two struc-
À
tures, the COOH group forms a hydrogen bond to the
second-best acceptor, the pyridyl nitrogen atom (O31···N21
Finally, the structure determination of 4-HBA:6 showed a
À
1:1 monohydrate co-crystal, in which the OH moiety forms
À
2.6627(13) ꢁ, O31 H31···N21 1.748(17) ꢁ; Figure 3).
a hydrogen bond to the benzimidazole site (O44···N13
À
À
2.683(3) ꢁ, O44 H44···N13 1.78(3) ꢁ), and the COOH
group interacts with the pyridyl nitrogen atom (O41···N31
À
2.659(3) ꢁ, O41 H41···N31 1.66(3) ꢁ; Figure 6). The water
molecule does not interfere with any of the postulated pri-
À
mary O H···N interactions, which is quite unusual in hy-
drates.[24]
Figure 3. Primary hydrogen-bond interactions in the crystal structure of
4-HBA:5.
Changing the donor molecule from 4-HBA to 3-HBA
while keeping the acceptor 5 the same also gave a 1:1 co-
À
crystal, in which the OH moiety forms a hydrogen bond to
the benzimidazole site (O33···N13 2.6778(16) ꢁ, O33-
À
H33···N13 1.77(2) ꢁ), and the COOH engages in a hydro-
gen bond with the pyridyl nitrogen atom (O31···N21
À
2.6266(17) ꢁ, O31 H31···N21 1.65(2) ꢁ; Figure 4).
In the crystal structure of 3-HBA:7·CH3CN·H2O, the two
principle components present in a 1:1 ratio are joined by
water and acetonitrile molecules. Despite the potentially dis-
ruptive influence that included solvent molecules can have
(especially, water molecules), the best donor on 3-HBA, the
Figure 6. Primary hydrogen-bond interactions in the crystal structure of
4-HBA:6.H2O.
In all six of the crystal structures that were obtained from
À
À
OH group, forms a hydrogen bond to the benzimidazole
this series of co-crystallizations, the OH group formed a
hydrogen bond with the acceptor atom associated with the
highest negative electrostatic potential, leaving the carboxyl-
ic acid to bind to the second-best acceptor (ranking deter-
À
À
mined by charge). If the OH and COOH groups had
been ranked according to acidity (which would reverse D1
and D2), then the commonly observed “best donor/best ac-
ceptor” behavior in hydrogen-bonded molecular solids
would not have held up in a single case. To place our obser-
vations in the context of other structural data, we also exam-
ined relevant crystallographic information from the existing
literature.
Figure 4. Primary hydrogen-bond interactions in the crystal structure of
3-HBA:5.
Chem. Eur. J. 2013, 00, 0 – 0
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
3
&
ÞÞ
These are not the final page numbers!

C. B. Aakerçy et al.
A previously reported structure of 4-HBA:4[28] exhibits
Scheme 3; 14:28) when a preference can be unambiguously
À
À
the same behavior: the OH group forms a hydrogen bond
established, there is about a 2:1 advantage in favor of the
OH group.
À
to the best acceptor, the imidazole moiety, and the COOH
À
À
group interacts with the pyridine nitrogen atom (Figure 7).
If the two donors, OH versus COOH, are ranked by
using
a pKa-based argument (in which the former is
105 times weaker), one should expect a far higher “winning
À
percentage” for the COOH than what is actually observed.
If the results from our current study are added to the struc-
tures found in the CSD, the bias is even more strongly in
À
favor of the OH being a more effective hydrogen-bond
À
donor than the COOH moiety in hydroxybenzoic acids
(Table 3).
Table 3. Combined results from current study and the CSD.
À
À
OH “wins”
Both groups bind
COOH “wins”
15:34
44%
14:34
41%
5:34
15%
Figure 7. Primary hydrogen-bond interactions in the crystal structure of
4-HBA:4.[28]
We subsequently carried out a search of the Cambridge
Structural Database (CSD) on all co-crystals of 3-HBA and
4-HBA with geometrically unbiased acceptors and obtained
a total of 28 hits and analyzed them in the context of a best
donor/best acceptor approach (Table 2). The results can be
The combined data show that when ditopic asymmetric
À
acceptors are employed, the OH group binds to the best
À
acceptor, and the COOH group binds to the second-best
acceptor in all seven cases. When symmetric ditopic accept-
À
ors are employed (19 structures), the OH group formed
hydrogen bonds with both acceptor groups (motif A) on
4:19 occasions. In 11:19 cases, both donor groups formed hy-
drogen bonds with the acceptor groups (motif B). Finally, in
Table 2. Analysis of 28 co-crystals of 3-HBA and 4-HBA with geometri-
cally unbiased hydrogen-bond acceptors.
À
4:19 cases, the COOH group formed hydrogen bonds with
both acceptors (motif C; Scheme 3).
À
9:28
À
5:28
OH “wins”
Both groups bind
COOH “wins”
14:28
50%
In the eight known co-crystals of 3-HBA and 4-HBA with
32%
18%
À
À
monotopic acceptors, the OH group wins four times, the
COOH moiety wins once, and in the three remaining cases,
the result is a 2:1 co-crystal with both donors participating
À
split into three different outcomes. In the first case, the
OH “wins” the competition for the donor (motif A in
À
in an O H···N hydrogen bond.
À
À
Scheme 3). In the second case, both the COOH and the
A closer examination of the distribution of outcomes with
symmetric ditopic donors as a function of donor molecule,
shows that for 3-HBA, both moieties bind to the available
equivalent acceptors in 8:9 cases (Table 4). However, in the
ten known co-crystals of 4-HBA, there are four instances of
(motif A) and three each of motif B and motif C, respective-
ly (Table 4).
OH moiety form hydrogen bonds to equivalent acceptor
À
sites (motif B in Scheme 3). In the third case, the COOH
moiety wins the competition for the hydrogen-bond accept-
or (motif C in Scheme 3).
In the majority of cases, both donors engage in hydrogen
bonding to equivalent acceptor sites, which indicates that
these two moieties are comparable strength and ability
when it comes to finding an acceptor in a competitive chem-
ical system. In the remaining cases (motif A and motif C in
The fact that the two probe molecules, 3-HBA and 4-
HBA, display somewhat different pattern preferences may
be attributed to the difference in donor strengths of the two
donors in each molecule. As shown in Scheme 1, the differ-
ence in molecular electrostatic potential (MEP) values be-
tween the two hydrogen-bond donor sites of 4-HBA
(53 kJmolÀ1) is much greater than in the case of 3-HBA
(31 kJmolÀ1). It is therefore reasonable to infer that the dif-
ference in charge between the two donors of 3-hydroxyben-
zoic acid is not sufficiently high to impart a more pro-
Table 4. Distribution of outcomes with symmetric ditopic acceptors.
Motif A
Motif B
Motif C
3-HBA
4-HBA
0
4
8
3
1
3
Scheme 3. Three plausible outcomes when forming co-crystals of 3-HBA
or 4-HBA with symmetric ditopic acceptors.
&
4
&
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Hydrogen-Bond Strengths in Hydroxybenzoic Acids
FULL PAPER
nounced “winner” resulting in a higher percentage for
motif B in co-crystals involving 3-HBA.
proach indicates that a hydroxylic group is a highly competi-
tive, and in the majority of cases, a more dominant hydro-
gen-bond donor than a carboxylic moiety, as long as both
groups are attached to the same conjugated backbone. If the
two components are competing for two hydrogen-bond ac-
ceptor sites of different strengths (again ranked based on
In this study, we have focused exclusively on molecules
that contain hydroxylic and carboxylic groups attached to
the same backbone. Because the two moieties are present
on a single conjugated frame, there is a cooperative effect in
place that attenuates the charge advantage that the ÀOH
group has over the ÀCOOH moiety (Scheme 4).
À
MEPs values), the OH group is more likely to bind to the
À
best acceptor, leaving the COOH group to bind to the
À
second-best acceptor. This means that the OH group is a
more effective hydrogen-bond donor for crystal-engineering
purposes than is the carboxylic acid group in co-crystals of
hydroxybenzoic acids. This study has focused on geometri-
cally unbiased acceptors, because there is no doubt that
steric factors could influence the outcome. If a potential ac-
ceptor site also contained an auxiliary group that could act
as a powerful hydrogen-bond donor for the C=O moiety of
À
the acid, then the COOH group is favored to win. For ex-
ample, 2-aminopyridine is more likely to bind to a carboxyl-
À
À
Scheme 4. Effect on electrostatic potentials of OH and COOH moiet-
ies when both are attached to the same aromatic scaffolding.
À
À
ic acid through a pair of O H···N/N H···O=C hydrogen
À
The interdependence of charge on different functionalities
on the same backbone also means that the results may be
slightly different if a molecular-recognition competition is
bonds than to opt for a phenol through a single O H···N in-
teraction.
Although the overall conclusion reached in this systematic
structural study may challenge conventional wisdom, it is
certainly widely accepted that thiols, which are significantly
more acidic than phenols, are inferior hydrogen-bond
donors (an observation, which again is consistent with a
ranking based upon electrostatic charge). The fact that the
charges on the two sites, OH and COOH, are quite com-
parable and that these interactions are weak and reversible
mean that we cannot expect these results to hold up under
any and all conditions. However, it is unrealistic to expect
synthetic strategies based on intermolecular interactions to
always lead to a specific outcome that can be traced back to
a single molecular property. The mere existence of poly-
morphism is certainly a manifestation thereof. However,
even though the number of data points examined herein is
quite small, the combination of new and existing structural
data indicate that the balance of intermolecular power is
À
À
arranged, in which the OH and COOH moieties are lo-
cated on different molecules or on one backbone that pre-
cludes electrostatic cooperation.
Because hydrogen bonds are relatively weak and reversi-
ble, one should not expect that supramolecular guidelines
can offer any guarantees as far the specific structural out-
come of any co-crystallization goes, as was evidenced by
conventional polymorphism and synthon crossover. Howev-
er, it may be useful to employ a MEP-based view of hydro-
gen-bond interactions to rationalize somewhat unexpected
structural behavior. For example, the four symmetric ditopic
acceptors in Table 5 are known to exhibit synthon crosso-
ver.[12a,25] The charges on the carbonyl group on 4-HBA and
3-HBA are À285 and À267 kJmolÀ1, respectively, and are
comparable or in some cases greater than the charges of the
acceptors in Table 5, and this could explain the appearance
À
À
À
À
À
of the alternate synthon, in which the OH group binds to
the carbonyl group on the acid.
likely to favor OH over COOH in hydroxybenzoic acids,
especially when facing a molecule with two different binding
sites. It is certainly conceivable that with a much larger data-
set, an adjustment to these conclusions may be necessary
but, in the meantime, we hope that the results presented
herein can serve as a starting point for re-examinations of
existing structural data, as well as for new studies[26] involv-
ing tailor-made molecules that can further advance our abili-
ty to rank effective hydrogen-bond strength for crystal-engi-
neering purposes.
Table 5. Symmetric hydrogen-bond acceptors known to display synthon
crossover.
Acceptor
MEP-AM1 [kJmolÀ1
]
1,2-bi(4-pyridyl)ethane
2,3,5,6-tetramethylpyrazine
4,4’-bipyridine
À285
À273
À269
À224
pyrazine
Experimental Section
Conclusion
Synthesis of pyrazine mono N-oxide (1):[27] A solution of hydrogen per-
oxide (30%; 1.42 g, 0.042 mol) in acetic acid (10 mL) was added dropwise
by using a drop funnel over a period of 2.5 h to a solution of pyrazine
(1.00 g, 0.013 mol) in acetic acid (12.5 mL) at 70–808C. Heating was con-
tinued for about 5 h. Acetic acid was removed on a rotary evaporator,
and then water (10 mL) was added followed by evaporation. The residue
Our results emphasize first of all that a reliable ranking of
hydrogen-bond donor strength can be achieved by using mo-
lecular electrostatic potential surfaces obtained by using
low-level semi-empirical methods. Furthermore, this ap-
Chem. Eur. J. 2013, 00, 0 – 0
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
5
&
ÞÞ
These are not the final page numbers!

C. B. Aakerçy et al.
was dissolved in hot chloroform (50 mL) and dried with a mixture of
sodium sulfate and sodium carbonate, and the solvent removed on a
rotary evaporator. The residue was chromatographed on silica with
chloroform/methanol (9:1) as the eluent. Product was isolated as a white
powder and recrystallization from methanol gave needle-like crystals
(1.2 g, 72%). M.p.: 113–1158C (lit. m.p.: 113–1148C);[11] 1H NMR
(400 MHz, [D6]DMSO): d=8.66 (d, 2H, J=4 Hz), 8.36 ppm (d, 2H, J=
Acknowledgements
We are grateful for financial support from the NSF (CHE-0957607) and
to Curtis Moore, Wichita State University, for collecting X-ray single-
crystal diffraction data on 4-HBA:3.
4 Hz); 13C NMR (200 MHz, [D6]DMSO): d=158, 144 ppm; IR (KBr
[1] G. R. Desiraju, CrystEngComm 2003, 5, 466–467; C. B. Aakerçy,
D. J. Salmon, CrystEngComm 2005, 7, 439.
[2] M. Simard, D. Su, D. Wuest, J. Am. Chem. Soc. 1991, 113, 4696–
4698.
[3] G. R. Desiraju, Angew. Chem. 1995, 107, 2541; Angew. Chem. Int.
Ed. Engl. 1995, 34, 2311.
[4] V. R. Thalladi, B. S. Goud, V. J. Hoy, F. H. Allen, J. A. K. Howard,
G. R. Desiraju, Chem. Commun. 1996, 401–402.
[5] ConQuest, version 1.14; Cambridge Structural Database: Cam-
bridge, UK.
À
pellet): u˜ =3430, 2914, 1596, 1470, 1433, 1312 (N+ O ), 1213, 1005, 861,
À
540, 477 cmÀ1
.
Synthesis of tetramethylpyrazine mono N-oxide (2): 2,3,5,6-Tetramethyl-
pyrazine mono-N-oxide was synthesized following the procedure for 1.
Yield: 61%. M.p.: 98–1008C; 1H NMR (200 MHz, [D6]DMSO): d=2.43
(d, 6H), 2.32 ppm (d, 6H); 13C NMR (200 MHz, [D6]DMSO): d=150,
138, 21, 12 ppm; IR (KBr pellet): u˜ =3462, 2914, 1572, 1472, 1320 (N+
À
OÀ), 1138, 1004, 922, 691 cmÀ1
.
4,4’-Bipyridyl mono N-oxide (3): mixture of 4,4’-bipyridine (2.00 g,
12.82 mmol), hydrogen peroxide (30%; 1.33 g, 39 mmol), and glacial
acetic acid (8 mL) was stirred in a round-bottom flask for 18 h at 708C.
After cooling the reaction mixture, the solvent was removed by a rotary
evaporator and diluted with water (20 mL). The solution was basified
with excess sodium carbonate (2 g) and extracted with chloroform (3ꢂ
50 mL). The organic layers were combined and then concentrated under
reduced pressure by using a rotary evaporator. The product was further
purified by column chromatography with an ethyl acetate/methanol mix-
ture (3:1) producing an off-white solid (1.1 g, 52%). M.p.: 170–1718C;
1H NMR (400 MHz, [D6]DMSO): d=8.70 (d, 2H, J=12 Hz), 8.36 (d,
2H, J=12 Hz), 7.94 (d, 2H, J=12 Hz), 7.83 ppm (d, 2H, J=21 Hz);
[6] M. L. Cheney, D. R. Weyna, N. Shan, M. Hanna, L. Wojtas, M. J. Za-
worotko, Cryst. Growth Des. 2010, 10, 4401–4413.
[7] E. Nauha, E. Kolehmainen, M. Nissinen, CrystEngComm 2011, 13,
6531–6537.
[8] C. B. Aakerçy, G. S. Bahra, P. B. Hitchcock, Y. Patell, K. R. Seddon,
J. Chem. Soc. Chem. Commun. 1993, 152–156.
[9] D. I. A. Millar, H. E. Maynard-Casely, D. R. Allan, A. S. Cumming,
A. R. Lennie, A. J. Mackay, I. D. H. Oswald, C. C. Tang, C. R.
Pulham, CrystEngComm 2012, 14, 3742–3374.
[10] G. Wang, Y. Huang, J. Phys. Chem. Solids 2007, 68, 2003–2007.
[11] C. B. Aakerçy, P. D. Chopade, J. Desper, Cryst. Growth Des. 2011,
11, 5333–5336.
13C NMR (400 MHz, [D6]DMSO): d=150, 142, 139, 133, 124, 120 ppm;
[12] a) A. Mukherjee, G. R. Desiraju, Chem. Commun. 2011, 47, 4090–
4092; b) B. Sarma, P. Sanphui, A. Nangia, Cryst. Growth Des. 2010,
10, 2388.
À
IR (KBr pellet): u˜ =3222, 2910, 1600, 1515, 1482, 1410, 1253 (N+ O ),
À
1228, 1191, 1029, 851, 821, 714, 651, 580 cmÀ1
.
[13] C. B. Aakerçy, I. Hussain, S. Forbes, J. Desper, CrystEngComm
2007, 9, 46–64; C. B. Aakerçy, I. Hussain, J. Desper, Cryst. Growth
Des. 2006, 6, 474–480.
[14] T. A. Galek, L. Fꢃbiꢃn, W. D. S. Motherwell, F. H. Allen, N. Feeder,
Acta Crystallogr. Sect. B 2007, 63, 768–782.
[15] M. C. Etter, Acc. Chem. Res. 1990, 23, 120.
[16] C. B. Aakerçy, A. M. Beatty, B. A. Helfrich, J. Am. Chem. Soc.
2002, 124, 14425–14432.
Ligands 4–7 were synthesized according to previously published meth-
ods.[28,29]. 4-((2-Phenyl-1H-imidazol-1-yl)methyl)pyridine (4):[28] m.p. 35–
398C (lit m.p. 33–388C). 1-(Pyridin-3-ylmethyl)-1H-benzo[d]imidazole
(5):[29] m.p. 50–558C (lit m.p. 48–518C). 5,6-Dimethyl-1-(pyridin-4-ylmeth-
yl)-1H-benzo[D]imidazole (6):[28] m.p.: 185–1908C (lit. m.p. 182–1908C).
5,6-Dimethyl-1-(pyridin-3-ylmethyl)-1H-benzo[D]imidazole (7):[28] m.p.
147–1508C (lit. m.p. 150–1538C).
Synthesis of 1-(pyridin-4-ylmethyl)-1H-benzo[D]imidazole (8): Benzimi-
dazole (0.5 g, 4.23 mmol) was dissolved in acetonitrile (50 mL). Crushed
NaOH (0.508 g 12.7 mmol) was added to the solution and was stirred for
3 h. 4-Picolylchloride hydrogen chloride (0.69 g, 4.23 mmol) was dissolved
in acetonitrile (50 mL) and added to the benzimidazole solution and
stirred for 6 h. Once the absence of the picolyl chloride was confirmed by
TLC, the acetonitrile was removed under reduced pressure. The resulting
oil was dissolved in ethyl acetate and washed with NaOH (1 n), distilled
water, and brine. The solution was dried over MgSO4. Ethyl acetate was
removed under reduced pressure to give a brown solid (5.40 g, 69.4%).
M.p. 105–1108C; 1H NMR ([D6]DMSO, 400 MHz): d=8.51 (d, J=6.2 Hz,
1H), 8.42 (s, 1H), 7.68 (dd, J=9.0, 3.5 Hz, 1H), 7.46 (dd, J=9.4, 3.5 Hz,
1H), 7.21 (dd, J=9.0, 3.5 Hz, 1H), 7.18 (d, J=5.5 Hz, 1H), 5.58 ppm (s,
1H).
[17] C. Laurence, M. Berthelot, Perspect. Drug Discovery Des. 2000, 18,
39–60.
[18] O. A. Raevsky, J. Phys. Org. Chem. 1997, 10, 405–413.
[19] M. H. Abraham, J. A. Platts, J. Org. Chem. 2001, 66, 3484–3491.
[20] C. B. Aakerçy, A. M. Beatty, B. A. Helfrich, Angew. Chem. 2001,
113, 3340; Angew. Chem. Int. Ed. 2001, 40, 3240.
[21] C. B. Aakerçy, J. Desper, D. J. Salmon, M. M. Smith, CrystEng-
Comm 2009, 11, 439–443.
[22] C. A. Hunter, Angew. Chem. 2004, 116, 5424–5439; Angew. Chem.
Int. Ed. 2004, 43, 5310–5324.
[23] D. Musumeci, C. A. Hunter, R. Prohens, S. Scuderi, J. F. McCabe,
Chem. Sci. 2011, 2, 883–890.
ˇˇ ´
[24] S. Karki, T. Friscic, W. Jones, W. Motherwell, Mol. Pharm. 2007, 4,
347–354.
[25] T. R. Shattock, K. K. Arora, P. Vishweshwar, M. J. Zaworotko,
Cryst. Growth Des. 2008, 8, 4533–4545.
[26] C. B. Aakerçy, K. N. Epa, S. Forbes, J. Desper, CrystEngComm
2013, 15, 5946–5949.
[27] C. F. Koelsch, W. H. Gumprecht, J. Org. Chem. 1958, 23, 1603.
[28] C. B. Aakerçy, J. Desper, M. M. Smith, Chem. Commun. 2007,
3936–3938.
Molecular structures for 3-HBA, 4-HBA, and 1–8 were constructed by
using Spartan 06 (Wavefunction, Inc. Irvine, CA). All molecules were op-
timized by using AM1, with the maxima and minima in the electrostatic
potential surface (0.002 ea.u.À1 isosurface) determined by using a positive
point charge in vacuum as a probe.
Solvent-assisted grinding was carried for combinations of 3-HBA or 4-
HBA with each of the eight acceptors in a 1:1 ratio with methanol as the
solvent. The resulting sixteen solids were characterized by IR spectrosco-
py, and five of them produced crystals (slow evaporation from methanol)
of sufficient quality to enable single-crystal X-ray diffraction analysis to
be carried out.
[29] C. B. Aakerçy, J. Desper, J. F. Urbina, Chem. Commun. 2005, 2820–
2822.
Received: April 12, 2013
Revised: July 26, 2013
Published online: && &&, 0000
&
6
&
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Hydrogen-Bond Strengths in Hydroxybenzoic Acids
FULL PAPER
Crystal Engineering
C. B. Aakerçy,* K. Epa, S. Forbes,
N. Schultheiss, J. Desper . . . . &&&& — &&&&
Ranking Relative Hydrogen-Bond
Strengths in Hydroxybenzoic Acids for
Crystal-Engineering Purposes
À
Intermolecular interactions: A ranking
based on electrostatic charge and
tested against structural data indicate
that the OH moiety is a better hydro-
À
gen-bond donor than the COOH site
in hydroxybenzoic acids (see scheme).
Chem. Eur. J. 2013, 00, 0 – 0
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
7
&
ÞÞ
These are not the final page numbers!