Spectroscopic and computational investigations of the thermodynamics of boronate ester and diazaborole self-assembly

Article abstract of DOI:10.1021/acs.joc.5b02548

The solution phase self-assembly of boronate esters, diazaboroles, oxathiaboroles, and dithiaboroles from the condensation of arylboronic acids with aromatic diol, diamine, hydroxythiol, and dithiol compounds in chloroform has been investigated by ¹H NMR spectroscopy and computational methods. Six arylboronic acids were included in the investigations with each boronic acid varying in the substituent at its 4-position. Both computational and experimental results show that the para-substituent of the arylboronic acid does not significantly influence the favorability of forming a condensation product with a given organic donor. The type of donor, however, greatly influences the favorability of self-assembly. ¹H NMR spectroscopy indicates that condensation reactions between arylboronic acids and catechol to give boronate esters are the most favored thermodynamically, followed by diazaborole formation. Computational investigations support this conclusion. Neither oxathiaboroles nor dithiaboroles form spontaneously at equilibrium in chloroform at room temperature. Computational results suggest that the effect of borylation on the frontier orbitals of each donor helps to explain differences in the favorability of their condensation reactions with arylboronic acids. The results can inform the use of boronic acids as they are increasingly utilized in the dynamic self-assembly of organic materials and as components in dynamic combinatorial libraries.

Full text of DOI:10.1021/acs.joc.5b02548

Article
pubs.acs.org/joc
Spectroscopic and Computational Investigations of The
Thermodynamics of Boronate Ester and Diazaborole Self-Assembly
Alexander R. Goldberg and Brian H. Northrop*
Department of Chemistry, Wesleyan University, Middletown, Connecticut 06459, United States
S
* Supporting Information
ABSTRACT: The solution phase self-assembly of boronate
esters, diazaboroles, oxathiaboroles, and dithiaboroles from the
condensation of arylboronic acids with aromatic diol, diamine,
hydroxythiol, and dithiol compounds in chloroform has been
investigated by 1H NMR spectroscopy and computational
methods. Six arylboronic acids were included in the
investigations with each boronic acid varying in the substituent
at its 4-position. Both computational and experimental results
show that the para-substituent of the arylboronic acid does not significantly influence the favorability of forming a condensation
1
product with a given organic donor. The type of donor, however, greatly influences the favorability of self-assembly. H NMR
spectroscopy indicates that condensation reactions between arylboronic acids and catechol to give boronate esters are the most
favored thermodynamically, followed by diazaborole formation. Computational investigations support this conclusion. Neither
oxathiaboroles nor dithiaboroles form spontaneously at equilibrium in chloroform at room temperature. Computational results
suggest that the effect of borylation on the frontier orbitals of each donor helps to explain differences in the favorability of their
condensation reactions with arylboronic acids. The results can inform the use of boronic acids as they are increasingly utilized in
the dynamic self-assembly of organic materials and as components in dynamic combinatorial libraries.
Scheme 1. General Representation of (a) Boronate Ester
Formation from the Reversible Condensation of Boronic
Acids with Organic Diols and of (b) Boroxine Anhydride
Formation from the Self-Condensation of Three Equivalents
of Boronic Acids
INTRODUCTION
■
Dynamic covalent reactions involving boronic acids have long
been recognized for their utility in the design of receptors and
sensors for compounds such as saccharides1 and anions.2 More
recently the self-assembly of boronic acids has proven to be a
powerful means of preparing complex macrocycles,3 cages,4 and
functional polymers5 and has given rise to the field of covalent
organic frameworks (COFs).6 Much of the unique and
advantageous chemistry of boronic acids arises from the low
valency and the empty p orbital of their constituent boron
atom(s), which enables them to accept electron density from a
wide variety of species. The condensation of boronic acids with
1,2-diols results in the formation of cyclic boronate esters
(Scheme 1a), while the self-condensation of boronic acids gives
boroxine anhydrides (Scheme 1b).7−11 Both boronate ester and
boroxine anhydride formations are thermodynamically rever-
sible despite their formation of strong B−O covalent bonds and
the entropic favorability of liberating two and three molecules
of water, respectively. The equilibria shown in Scheme 1 can be
driven toward products by a variety of means (azeotropic
removal of water, dehydrating agents, product precipitation,
etc.), however the hydrolysis of boronate esters and boroxine
anhydrides is a common concern and varies significantly with
the functionality of boronic acid and diol components as well as
the choice of solvent.7
nonaqueous solutions have been reported in the literature.16−18
This is despite over two decades of broad interest in the use of
boronic acids in synthesis, self-assembly, and materials
chemistry.8,9 It is known that the favorability of forming esters
and boroxines from boronic acids depends on the functionality
of the acid, the solvent, and the presence of coordinating
donors. Many boronic acids readily dehydrate in nonpolar
solvents (e.g., CH2Cl2 and CHCl3) to form their less polar
boroxine analogues. Tokunaga and co-workers have inves-
The self-assembly of boronate ester and boroxine anhydride
species in aqueous solutions has been thoroughly inves-
tigated.9,12−15 By contrast, few investigations of the thermody-
namics and dynamic assembly processes of boronic acids in
Received: November 5, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/acs.joc.5b02548
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
tigated the free energy of boroxine formation from para-
substituted aryl boronic acids in chloroform and found the self-
condensation to be entropically driven and generally ender-
gonic.17 Electron-donating substituents have been shown to
increase the favorability of boroxine formation by decreasing
the electrophilicity of boron, thereby decreasing their
susceptibility to hydrolysis. Electron-withdrawing substitutents
have the opposite effect. Computational investigations by Kua
and Iovine generally support the experimental observations.19
Dehydration of boronic acids to boroxines is facilitated by the
presence of nitrogen donor compounds (e.g., tertiary amines or
pyridine), resulting in the formation of boroxine donor adducts
that have been found to be thermodynamically more favorable
than uncoordinated boroxines.10,19−24 In fact, the ligand-
facilitated trimerization of boronic acids can be used to
promote the formation of boroxines.
Herein we present combined spectroscopic and computa-
tional investigations of the self-assembly of para-substituted aryl
boronic acids with a series of aryl donors bearing alcohol,
amine, and thiol functionalities. We find that the para-
substituent of the aryl boronic acid does not significantly
influence the thermodynamics of their self-assembly with the
different organic donors. The functionality of the organic
donor, by contrast, is found to influence the favorability of
forming stable condensation products considerably.
RESULTS AND DISCUSSION
■
Shown in Figure 1a are the six phenylboronic acids investigated
in the present study: 4-methoxyphenylboronic acid (1), 4-tert-
The formation of boronate esters from the condensation of
boronic acids with diols is also influenced by the nature of the
solvent and the functionality of the starting compounds. As
with boroxines, boronate esters are less polar than their starting
boronic acids, and this difference in polarity can be used to shift
their equilibria toward esters in nonpolar solvents. The stability
of boronate esters to hydrolysis is highly influenced by the
nature of their component diols. Sterically hindered boronates
(e.g., pinacolboranes) are known to be relatively stable in protic
solvents, while less hindered boronates (e.g., ethylene glycol
esters) are quite susceptible to hydrolysis.7 As noted earlier,
condensations of arylboronic acids and catechol derivatives
have been recently and widely used in the dynamic covalent
assembly of boronate ester-based organic materials, such as
boronate ester macromolecules5 and COFs.6 The majority of
such organic materials are prepared using organic solvents.
Despite this growing area of research, and to the best of our
knowledge, no experimental or computational studies focusing
on the thermodynamics of boronate ester self-assembly in
nonaqueous environments have been reported.
Similarly, very little is known of the thermodynamics of
related dynamic assembly processes between boronic acids and
ortho-phenylenediamines to give diazaboroles.25 The first
example of a diazaborole condensation, reported by Letsinger
over 50 years ago, was formed by the reaction of ortho-
phenylenediamine with the ethyl tartarate ester of phenyl-
boronic acid.26 More recently diazaboroles have been shown to
function as blue emissive materials,27 photoluminescent
polymers,28 and p-type semiconductors in organic field-effect
transistors.29 Diazaboroles are known to hydrolyze rapidly in
mildly acidic conditions, however they are generally stable in
neutral solutions. Dithiaboroles, another class of related boron
heterocycles, have been reported, though they are typically
synthesized from 2-chloro-1,3,2-dithiaborole derivatives30−32
rather than through the condensation of dithiols with boronic
acids. A more quantitative understanding of the influence of
boronic acid and organic donor functionality on the favorability
of their self-assembly processes will likely (i) provide access to
new boron-based materials from dynamic covalent and dynamic
combinatorial assembly of different secondary building units
and (ii) enable predictable exchange processes when
thermodynamics sufficiently favor the formation of one type
of assembly over another. The design of new boronic acid-
derived sensors, polymers, COFs, and other functional
materials will benefit from such an increased understanding
of the thermodynamics of their assembly processes.
Figure 1. Chemical structures of the (a) aryl boronic acids and (b)
organic donor compounds investigated in the current study.
butylphenylboronic acid (2), phenylboronic acid (3), 4-
fluorophenylboronic acid (4), 4-methoxycarbonylphenylbor-
onic acid (5), and 4-cyanophenylboronic acid (6). Para-
substituted arylboronic acids were chosen because experimental
investigations have demonstrated that ortho-substituted bor-
onic acids are less favored to assemble into boroxines or
boronate esters, largely due to steric effects.7 Para-substituted
acids allow the influence of electronic effects on assembly
processes to be studied independent of the size of a given
substituent. Furthermore, the symmetry of para-substituted
acids aids significantly in investigating their assembly processes
1
by H NMR spectroscopy. The chemical structures of four
different organic donors are shown in Figure 1b: 4-tert-
butylcatechol (7), 4-tert-butyl-ortho-phenylenediamine (8), 2-
hydroxybenzenethiol (9), and toluene-3,4-dithiol (10).
Aryl boronic acids and aryl organic donors were chosen
because of their prevalent use in the design, synthesis, and self-
assembly of polymeric materials and COFs. Donation of
electron density from oxygen to boron in boronate esters can
increase their stability by disfavoring hydrolysis. In the case of
catechol derivatives, however, lone pair electrons of catechol
oxygen atoms are also able to conjugate with the aromatic ring,
weakening their interaction with boron. Catechol-based
boronate esters are therefore more susceptible to hydrolysis
than, for example, pinacolboranes. This hydrolytic susceptibility
is often considered a favorable attribute in the context of
dynamic covalent self-assembly33 where readily reversible
covalent bond formation is necessary to provide a means of
error correction en route to the most thermodynamically stable
structure(s). The role that similar electronic effects play in the
thermodynamic stability of diazaboroles and dithiaboroles is
currently unknown. Amine- and thiol-functionalized donors 8−
B
DOI: 10.1021/acs.joc.5b02548
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
1
10 were therefore chosen to gain insight into the stability of
their assemblies with boronic acids 1−6. Investigations of
donors beyond catechol also open prospects for elucidating
new exchange processes, constructing new dynamic combina-
torial libraries, and developing routes to new diazaborole and
dithiaborole materials.
signals could be observed after approximately 4 h. H NMR
spectra of equimolar mixtures of boronic acids 1−6 and tert-
butyl-ortho-phenylenediamine (8) were observed to evolve
more progressively over the course of approximately 24 h, after
which no appreciable change could be observed. Mixtures
containing boronic acids 1−6 and either hydroxybenzenethiol
(9) or toluene-3,4-dithiol (10) showed no discernible
oxathiaborole or dithiaborole formation. It could be concluded
from 1H NMR spectroscopy alone that the formation of
oxathiaboroles (13) and dithiaboroles (14) by the condensa-
tion of aryl boronic acids with 9 or 10 is thermodynamically
unfavored under these conditions, though similar structures
have been prepared by the reaction of hydroxythiol and dithiol
compounds with dichlorophenylboranes34 or by the afore-
mentioned routes using haloboranes.31
To evaluate the thermodynamics of boronate ester self-
assembly, we used a combination of experimental, spectro-
scopic, and computational analysis. Experimentally, pairwise
combinations of boronic acids 1−6 and organic donors 7−10
were mixed in 1:1 molar ratios in CDCl3 at room temperature
and allowed to reach equilibrium. The extent of assembly
formation was then determined by the relative integration
values of signals corresponding to acid, ester, and, in some
cases, boroxine species by 1H NMR spectroscopy. In
conjunction with spectroscopic analysis, the energetics of
boronate ester and boroxine formation were studied computa-
tionally. The use of a combination of computational and
spectroscopic investigations was motivated by the desire to
benchmark different computational methods against exper-
imental results and, ideally, to elucidate the electronic and/or
structural factors that underlie differences in the thermody-
namics of boronate ester, diazaborole, and dithiaborole
formation.
1
Relative integrations of H NMR spectroscopic signals were
used to calculate equilibrium constants (Keq) and free energies
(ΔG°) for the condensation reactions between boronic acids
1−6 with catechol 7 and ortho-phenylenediamine 8. Assign-
ments of boronic acid, boroxine anhydride, and boronate or
diazaborole species were made by comparison to pure samples
of boroxines and condensation products 11a−f and 12a−f. For
most compounds, the singlet of the 4-tert-butyl group of 7 and
8 was found to be particularly diagnostic as it appears at 1.26
ppm for the free donor species but undergoes a downfield shift
to 1.35−1.37 ppm (esters 11a−f) or 1.36−1.38 ppm
(diazaboroles 12a−f) upon condensation with boronic acids
1−6. Furthermore, the tert-butyl singlets of assembled and
unassembled species, integrating to 9 symmetry equivalent
protons, provide the greatest signal-to-noise of all spectroscopic
signals and therefore provide the most reliable means of
evaluating the percentages of product formation. Assemblies
containing tert-butylphenylboronic acid 2 were the exception
because tert-butyl peaks corresponding to several species (i.e.,
unassembled 7 or 8, unassembled 2, ester or diazaborole
products 11b or 12b, and the boroxine anhydride of 2) overlap
in their equilibrated 1H NMR spectra. When possible,
calculated ratios of condensation products were corroborated
by the integration of diagnostic signals in the aromatic region of
each spectrum. Figure 2 shows a representative example
corresponding to the equilibrium between 4-cyanophenylbor-
onic acid 6, tert-butylcatechol 7, and their condensation
product boronate ester 11f. Collected in Table 1 are the
calculated equilibrium constants and free energies of con-
densation reactions between each of the boronic acids shown in
Figure 1a and donors 7−10.
1H NMR Spectroscopy. In principle, boronic acids 1−6 are
able to condense with donors 7−10 to give boronate esters
(11a−f), diazaboroles (12a−f), oxathiaboroles (13a−f), and
dithiaboroles (14a−f), as shown in Scheme 2. Equimolar
Scheme 2. Dynamic Covalent Exchange Reactions between
Phenylboronic Acids 1−6 and Aryl Donors 7−10
1
Investigated by H NMR Spectroscopy
Boronic acids functionalized with electron-withdrawing
groups, namely 4-methylcarbonylphenyl boronic acid (5) and
4-cyanophenylboronic acid (6), gave the lowest yields of
boronate ester formation at equilibrium: 94% of 11e and 92%
of 11f. The remaining four boronic acids gave boronate esters
11a−d in 96−97% yield, which may be considered identical
within experimental error. Overall the formation of all six
boronate esters is predicted to be exergonic, with reaction free
energies ranging from ΔG° = −1.1 to −2.5 kcal/mol. The
results summarized in Table 1 suggest that electron-rich
boronic acids form more thermodynamically stable boronate
ester assemblies with catechol 7, while acids bearing electron-
withdrawing groups form less stable boronate esters, however
the differences in yield, Keq, and ΔG° are quite small
considering the large differences in Hammett parameters
among the six functional groups investigated: σpara ranges
from −0.27 (methoxy) to +0.66 (cyano). Overall, all six of the
solutions of each boronic acid and each donor were mixed in
CDCl3 (0.05 M) and allowed to reach equilibrium at room
temperature. Particular care was taken to dry all starting
materials and CDCl3 given the role that water plays in the
dynamic equilibrium of boronic acid condensation reactions
(full details are provided in the Computational and
Experimental Methods section). While assembly kinetics were
not a focus of the current investigation, it was readily observed
that the room temperature condensation of boronate esters
11a−f is more rapid than the condensation of diazaboroles
12a−f under identical conditions. 1H NMR spectra of mixtures
containing any of the six boronic acids and tert-butylcatechol
(7) showed nearly complete product formation within 30 min
of mixing, and no changes in the relative integrations of proton
C
DOI: 10.1021/acs.joc.5b02548
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Figure 2. Example equilibrated mixture of 4-cyanophenylboronic acid (6), 4-tert-butyl catechol (7), and their corresponding boronate ester 11f (0.05
M in CDCl3, 298 K, 300 MHz). Signals of both unassembled and assembled species are observed and were used for the calculation of the
1
equilibrium constant (Keq) and condensation free energy (ΔG°) of boronate ester formation. Full H NMR spectra for all condensations are
provided in Figures S7−S18 of the Supporting Information.
Table 1. Equilibrium Constants and Corresponding
(ΔΔG° = 1.4 kcal/mol), again suggesting that the para-
substituent of boronic acids 1−6 plays a relatively minor role in
determining the thermodynamic favorability of assembly
formation. It is interesting to note that the small influence
the para-substituent has on diazaborole formation is the
opposite of its influence in boronate ester formation. That is,
the assembly of ortho-phenylenediamine 8 with boronic acids 5
and 6, which bear electron-withdrawing groups, is found to be
slightly more favored than the assembly of electron-rich
Reaction Free Energies for the Formation of Boronate Esters
11a−f and Diazaboroles 12a−f as Determined from 1H NMR
a
Spectroscopy at 25 °C
b
c
d
R
% product
Keq
44.7
ΔG°
a
b
c
d
e
f
OMe
t-Bu
H
97
96
97
96
94
92
33
44
30
30
42
42
−2.3
−1.9
−2.5
−2.1
−1.4
−1.1
3.3
24.9
75.5
32.9
10.8
6.4
11
12
1
F
boronic acids with 8. Looking at the H NMR spectra of
CO2Me
CN
equilibrated mixtures of diazaboroles 12a−f provides some
insight into why this may be the case. Equilibrium mixtures of
diazaboroles 12a−c contain substantial quantities of boroxine
anhydride and little to no free boronic acid species (see Figures
S13−S15 of the Supporting Information). By contrast no
boroxine species are observed in the equilibrated mixtures of
diazaboroles 12d−f. These results suggest that boroxine
anhydride formation may compete with diazaborole formation
when electron-rich arylboronic acids (e.g., 1 and 2) are used,
while boroxine formation is not favored in the case of boronic
acids bearing electron-withdrawing groups (e.g., 4−6). As
mentioned earlier, electron-poor arylboroxine anhydrides are
known to hydrolyze more readily than electron-rich boroxine
species. The competitive formation of broxine species will
decrease the extent of diazaborole formation at equilibrium,
potentially explaining why the formation of electron-poor
diazaboroles is more favored than electron-rich diazaboroles.
The same is likely true in the equilibrium formation of boronate
esters 11a−f, however the influence of boroxine formation is
less pronounced given the greater favorability of boronate
formation relative to diazaborole formation.
Collectively, the spectroscopic results summarized above
indicate that the para-substituent of boronic acids 1−6 plays a
relatively small role in influencing the thermodynamics of
boronate ester or diazaborole formation. The choice of donor,
however, significantly influences the favorability of assembly
formation. All else being equal (boronic acid, concentration,
temperature, and solvent), boronate ester formation is favored
over diazaborole formation by 3.9−6.0 kcal/mol. Furthermore,
as noted above, no evidence of the self-assembly of
a
b
c
d
e
f
OMe
t-Bu
H
3.84 × 10−3
1.37 × 10−2
2.59 × 10−3
2.61 × 10−3
1.07 × 10−2
1.10 × 10−3
2.5
3.5
F
3.5
CO2Me
CN
2.7
2.7
a
The dynamic covalent assembly of oxathiaboroles (13a−f) and
dithiaboroles (14a−f) as outlined in Scheme 2 was not observed.
Complete details and 1H NMR spectra used to determine Keq
b
contained in the Supporting Information. R indicates the substituent
c
at the para-position of the boronic acid. Keq = [ester][H2O]2/
d
([acid][donor]). kcal/mol.
arylboronic acids investigated readily assemble with catechol 7
in chloroform, and the equilibrium formation of boronate esters
11a−f lies heavily toward products (>90% assembly
formation).
The dynamic assembly of ortho-phenylenediamine 8 with
boronic acids 1−6 results in significantly less diazaborole
formation as compared to boronate ester formation. Substantial
quantities of uncondensed boronic acid and diamine species
1
can be observed in the H NMR spectra of each equilibrated
mixture. Diazaborole formation was found to be endergonic in
all cases (ΔG° = 2.5−3.5 kcal/mol) with each mixture resulting
in <50% product formation at equilibrium for 0.05 M solutions.
The range of reaction free energies observed for the formation
of diazaboroles 12a−f (ΔΔG° = 1.0 kcal/mol) is somewhat
narrower than that for the formation of boronate esters 11a−f
D
DOI: 10.1021/acs.joc.5b02548
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
oxathiaboroles or dithiaboroles could be observed by 1H NMR
spectroscopy. To gain additional insight into the influence of
both the organic donor and para-substituted boronic acid
components on the favorability of boronate ester and
diazaborole formation as well as the unfavorability of
oxathiaborole and dithiaborole formation, we explored the
thermodynamics of their assembly using computational
methods.
B3LYP, CBS-QB3, and MP2 levels give the same trend (see
Table S1 of the Supporting Information), and overall results are
in line with earlier computational studies of aliphatic boronic
acids.40,41,48 The nature of the para-substituent has a small but
discernible influence on the structures of arylboronic acids 1−6.
For example, the O−H···O intramolecular hydrogen bond
present in endo−exo conformers varies with the electron-
donating or -withdrawing strength of each para-substituent.
Donors increase electron density around the boron atom and
lead to shorter hydrogen bond distances, for example, 2.392 Å
for acid 1 (R = OMe). Withdrawing groups do the opposite by
drawing electron density away from the boron atom, leading to
an increase in oxygen−boron conjugation and a corresponding
lengthening of the intramolecular hydrogen bond, for example,
2.407 Å for acid 6 (R = CN). Similar trends can be seen in the
variations in the C−B and B−O bond lengths of acids 1−6 as
well as their O−B−O bond angles (see expanded Table S1 of
the Supporting Information). Overall the donating or with-
drawing nature of the para-substituent can be observed to
influence the structural parameters of boronic acids 1−6.
Interestingly, however, no overall trend could be observed
relating the nature of the para-substituent and the relative free
energies of syn or anti conformations. It is also interesting to
note that while the endo−exo and syn conformations are flat,
the boronic acid moiety of each anti conformation is observed
to twist 25−32° out-of-plane relative to the aryl ring.
Table 3 summarizes the thermochemistry of forming
boronate esters 11a−f, diazaboroles 12a−f, oxathiaboroles
13a−f, and dithiaboroles 14a−f via the condensation of
boronic acids 1−6 with donors 7−10, respectively. Before
examining the specific influences of different functional groups,
it is worthwhile to note four general trends that emerge from
the data presented in Table 3: (i) All but three of the 96
condensation reactions (24 reactions at four levels of theory)
are predicted to be endothermic, and in all cases the
condensation free energies are predicted to be entropy driven.
Furthermore, calculated reaction entropies (TΔS°) are very
similar across nearly all condensation reactions at each level of
theory, with MP2 calculations of dithiaboroles 14a−f
representing the one exception to this observation. Excluding
this one exception, the average TΔS° for all reactions is 8.1
kcal/mol with a standard deviation of 0.26 kcal/mol. (ii) The
choice of donor significantly impacts the free energy of
condensation reactions between donors 7−10 and boronic
acids 1−6. As a general trend boronate esters 11a−f are
predicted to be the most favorable, followed by diazaboroles
12a−f, oxathiaboroles 13a−f, and dithiaboroles 14a−f. This
trend is predicted to be almost entirely enthalpy driven as
reaction entropies remain relatively constant. Calculations
performed at the MP2 level, which predict the energetics of
boronate ester and diazaborole formation to be equal within
error, represent the one exception to this trend. (iii) In contrast
to the choice of donor, the para-substituent of boronic acids 1−
6 has little impact on calculated reaction free energies. Within
each set of condensation products 11−14, the difference
between the most and least favored assembly is ≤0.9 kcal/mol
at the CBS-QB3 level. (iv) A clear trend exists between the
reaction energetics predicted by different levels of theory. For
any given condensation reaction MP2 calculations predict the
most favorable reaction free energies, followed by CBS-QB3,
then B3LYP, and finally M06-2X with successive differences of
roughly 2−4 kcal/mol between each different level of theory.
Computational Investigations. The energetics of forming
condensation products 11−14 were studied at four levels of
theory, in particular: B3LYP/6-311+G(d,p),35 M06-2X/6-
31+G(d,p),36 CBS-QB3,37 and MP2/aug-cc-pVDZ.38,39 Several
computational investigations of boronic acid-derived assemblies
have been carried out previously using a variety of theoretical
methods.24,40−43 However, investigations of the thermody-
namics of boronate ester, diazaborole, or dithiaborole self-
assembly in nonaqueous solutions have not yet been reported.
Kua has previously investigated the thermodynamics of
boroxine formation and their amine adducts at the B3LYP
level, providing insight into the influence of para-substitution
and solvent on boroxine and boroxine−amine complex
formation.24a,c Bock and co-workers, however, have cautioned
against the use of B3LYP to quantitatively describe boroxine
thermochemistry,40 recommending MP2 as a more suitable
method. B3LYP was still included in the current study to
investigate its applicability to modeling boronate thermochem-
istry and because it is capable of handling large chemical
systems. The M06-2X functional has been shown to be broadly
applicable for modeling main-group thermochemistry,36,44
noncovalent interactions,45 and excited-state chemistry.46
Furthermore, the M06-2X functional allows for large systems
to be handled at reasonable computational cost, making it a
good candidate for modeling boronate and boroxine chemistry.
CBS-QB3 has been widely shown to give high-accuracy
thermochemical results,47 though calculations at both the
CBS-QB3 and MP2 levels have high computational cost.
Independent of the method chosen, computational studies have
highlighted41 the need to include diffuse functions to
appropriately describe changes in bonding upon boronic acid
self-assembly. Basis sets containing diffuse functions were
therefore utilized for all levels of theory.
The three most favored conformations of arylboronic acids
1−6, which differ in the relative orientations of their two
hydroxyl groups, are shown in Figure 3. All four levels of theory
Figure 3. Relative conformations of the three lowest free energy
structures of boronic acids 1−6: endo−exo, syn, and anti. The dashed
line present in the endo−exo conformation indicates an intramolecular
O−H···O hydrogen bond.
predict the endo−exo conformation to be the most favored, as
may be expected given that it is the only conformation that
allows for the formation of an intramolecular hydrogen bond.
Syn conformers of 1−6 are predicted to be slightly less stable
than endo−exo global minima by 0.7−1.3 kcal/mol, while the
anti conformers are 2.4−3.0 kcal/mol less stable (Table 2) at
the M06-2X/6-31+G(d,p) level of theory. Results at the
E
DOI: 10.1021/acs.joc.5b02548
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Table 2. Intramolecular Hydrogen-Bonding Distances Observed in endo−exo Conformations of Acids 1−6 and Relative Free
a
Energies of syn and anti Conformations
1
2
3
4
5
6
(R = OMe)
(R = t-Bu)
(R = H)
(R = F)
(R = CO2Me)
(R = CN)
endo−exo
syn
O−H···O
ΔG°
2.392
1.0
2.393
1.0
2.393
1.3
2.396
0.9
2.401
0.9
2.407
0.7
anti
ΔG°
2.4
2.5
2.6
2.8
2.7
3.0
a
Distances are given in Å. Free energies are calculated relative to the endo−exo conformation of each boronic acid and are given in kcal/mol.
Distances and energies are reported from M06-2X/6-31+G(d,p) calculations.
Table 3. M06-2X, B3LYP, CBS-QB3, and MP2 Calculated Free Energies (ΔG°), Enthalpies (ΔH°), and Entropies (TΔS°) for
a
Each Condensation Reaction Outlined in Scheme 2
M06-2X/6-31+G(d,p)
B3LYP/6-311+G(d,p)
CBS-QB3
MP2/aug-cc-pVDZ
exp.
R =
ΔG
ΔH
TΔS
ΔG
ΔH
TΔS
ΔG
ΔH
TΔS
ΔG
ΔH
TΔS
ΔG
a
b
c
d
e
f
OMe
t-Bu
H
−2.1
−2.0
−1.7
−1.6
−1.7
−1.2
2.1
6.1
6.4
8.2
8.5
8.1
8.1
8.5
8.1
8.2
8.5
8.6
8.2
8.5
8.3
8.2
8.4
8.2
8.2
8.5
8.1
6.4
7.5
9.9
8.8
9.3
9.5
−4.2
−4.0
−3.8
−3.8
−3.5
−3.4
−0.6
−1.9
−1.1
−0.9
−1.3
−1.4
2.7
3.7
4.0
7.9
8.0
7.8
7.8
7.8
7.8
8.3
9.4
8.3
8.3
8.1
8.1
8.1
8.2
8.1
8.1
8.0
8.1
7.6
7.2
7.2
7.1
7.5
7.2
−6.5
−6.7
−6.4
−6.5
−6.1
−6.2
−3.2
−3.0
−3.3
−3.3
−3.3
−3.7
−0.3
−0.6
−0.3
−0.3
0.0
1.2
1.3
7.7
7.9
7.8
7.8
7.6
7.8
8.2
7.9
8.0
8.0
7.7
8.0
7.8
8.1
8.0
8.0
7.8
8.0
7.2
8.0
7.3
7.7
7.7
−8.2
−8.1
−7.9
−7.8
−7.8
−7.7
−8.1
−8.1
−8.1
−7.5
−8.4
−8.4
−5.0
−4.7
−4.6
−4.6
−4.5
−4.5
−4.1
−3.7
−2.4
−4.6
−3.7
−4.0
−0.1
0.1
0.2
0.2
0.3
0.4
0.2
0.0
0.0
0.1
−0.3
−0.5
3.1
3.3
3.5
3.5
3.5
3.6
1.5
1.6
1.3
2.0
1.9
2.0
8.1
8.1
8.1
8.0
8.1
8.1
8.3
8.1
8.1
7.6
8.1
8.0
8.1
8.0
8.1
8.1
8.1
8.0
5.6
5.3
3.7
6.6
5.5
5.9
−2.3
−1.9
−2.5
−2.1
−1.4
−1.1
3.3
6.5
4.0
1.4
11
12
13
14
F
6.5
4.1
1.4
CO2Me
CN
6.8
4.3
1.5
6.9
4.4
1.6
a
b
c
d
e
f
OMe
t-Bu
H
10.3
10.2
10.0
10.1
9.6
7.7
5.0
1.6
7.5
4.8
2.5
1.4
7.3
4.6
3.5
F
1.9
7.4
4.7
3.5
CO2Me
CN
1.2
6.8
4.3
2.7
1.4
9.6
6.7
4.2
2.7
b
a
b
c
d
e
f
OMe
t-Bu
H
5.2
13.4
13.8
13.9
13.9
14.2
14.3
18.8
19.7
20.3
19.8
20.8
20.8
10.8
11.0
11.1
11.2
11.4
11.4
16.4
16.6
16.7
16.8
16.8
16.9
7.5
b
b
b
b
b
b
b
b
b
b
b
5.4
2.9
7.6
5.7
3.0
7.7
F
5.7
3.1
7.7
CO2Me
CN
5.7
3.4
7.8
6.2
3.4
−0.2
4.6
7.9
a
b
c
d
e
f
OMe
t-Bu
H
12.4
12.2
10.4
11.1
11.4
11.4
8.8
11.8
11.7
11.9
11.9
11.9
12.1
9.4
3.7
9.5
4.6
F
9.7
4.3
CO2Me
CN
9.3
4.2
9.6
3.8
8.3
b
a
All values are given in kcal/mol at 298.15 K and 1.0 atm in a PCM implicit solvent model for CHCl3. No equilibrium product formation by 1
H
NMR spectroscopy.
Upon comparing experimental results presented in Table 1
with computational results of Table 3, it is clear that
computational results at the M06-2X/6-31+G(d,p) level are
the most consistent with experimental data. Computations at
the M06-2X level predict the equilibrium free energy of forming
boronate esters 11a−f to range from ΔG° = −1.2 to −2.1 kcal/
mol. This range is in reasonable agreement with experimental
results (ΔG°exp = −1.1 to −2.5 kcal/mol). The average
difference between experimental and computational M06-2X
results for individual boronate ester condensation reactions is
0.3 kcal/mol with a maximum difference of 0.8 kcal/mol
(product 11c). It is also observed that M06-2X results are in
the best agreement with experimental free energies of
diazaborole formation, though the agreement between
experimental and computational results is not as good. The
calculated free energies for the formation of diazaboroles 12a−f
range from ΔG° = 1.2−2.1 kcal/mol, while spectroscopic
results give condensation free energies between ΔG° = 2.5−3.5
kcal/mol. In the case of diazaborole formation, the average
difference between experimental and computational results is
1.5 kcal/mol with a maximum deviation of 2.1 kcal/mol
(product 12c).
B3LYP, CBS-QB3, and MP2 calculations predict reaction
free energies for boronate ester and diazaborole formation to be
more negative, i.e., more favorable, than 1H NMR spectroscopic
results reveal. B3LYP/6-311+G(d,p) calculations, for example,
predict reaction free energies between ΔG° = −3.4 to −4.2
kcal/mol for the formation of boronate esters 11a−f. Given the
reaction stoichiometry and concentrations of reactants, such
reaction free energies would predict that all six boronate esters
would be be formed in 98−99% yield at equilibrium. B3LYP
predictions are, therefore, relatively consistent with spectro-
scopic results for the formation of boronate esters 11a−d given
1
the detection limits of H NMR spectroscopy, however, the
agreement is considerably less consistent than calculations
performed at the M06-2X level. Calculations performed at the
CBS-QB3 and MP2/aug-cc-pVDZ levels suggest that all
boronate ester condensation reactions investigated have
reaction free energies that are exergonic by at least −6.1
kcal/mol, which would correspond to ≥99.7% formation of
F
DOI: 10.1021/acs.joc.5b02548
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
boronate esters 11a−f at equilibrium. This computational
prediction is clearly not consistent with 1H NMR spectroscopic
results that all show measurable amounts of unassembled
starting materials present.
differences are very small. This trend likely arises because para
donating substituents are able to increase the electron density
around the boron atom, therefore shortening and strengthening
the C−B bond while at the same time reducing electron
donation from the oxygen atoms of the catechol moiety, leading
to longer B−O bond lengths. Para withdrawing groups decrease
electron density at the boron atom, accounting for the longer
C−B bonds and shorter B−O bonds as catechol oxygen atoms
donate electron density to the electron-poor boron atom. For
all esters 11a−f the C−O bond lengths fall within the tight
range of 1.379−1.381 Å, which is longer than the average C−O
bond length in catechol itself (1.369 Å), further supporting the
notion that catechol behaves as a donor in the boronate ester
condensation reactions. Computed electrostatic potential maps
shown in Figure 4 further support these observations.
In comparing computational and experimental results for
diazaborole condensation reactions, it is again clear that B3LYP,
CBS-QB3, and MP2 calculations predict reaction free energies
that are too favorable to be within reasonable agreement with
spectroscopic results. Formation of diazaboroles 12a−f from
the condensation of ortho-phenylenediamine with boronic acids
1−6 is predicted to be exergonic by ΔG° = −0.6 to −1.9 kcal/
mol at the B3LYP/6-311+G(d,p) level. If the condensation
reactions were that favorable then diazaboroles 12a−f would be
present in 80−92% yield at equilibrium, in clear contrast to
results obtained spectroscopically. Again, CBS-QB3 and MP2
calculations predict the condensation reactions to be even more
exergonic (ΔG°CBS‑QB3 = −3.0 to −3.7 kcal/mol and ΔG°MP2
=
−7.5 to −8.4 kcal/mol), suggesting >97% diazaborole
formation at equilibrium. Lastly, no oxathiaborole (13a−f)
1
formation is observed by H NMR spectroscopy, yet computa-
tional results at the B3LYP, CBS-QB3, and MP2 levels each
predict equilibrium formation of oxathiaborole products
ranging from 20 to 30% (B3LYP) to >99% (MP2).
Computational results obtained at the M06-2X level are again
more consistent with experimental results, predicting the
formation of oxathiaboroles to be endergonic by 5.2−6.2
kcal/mol, which would correspond to a maximum of <8%
oxathiaborole formation at equilibrium. An 8% yield of
oxathiaborole species would be detectable by 1H NMR
spectroscopy, which suggests the M06-2X results likely
overestimate the favorability of oxathiaborole formation,
however M06-2X computations are still the most consistent
with experimental results.
Figure 4. Electron density maps of boronate esters 11a−f (isovalues:
−0.02 to 0.02) calculated at the B3LYP/6-311+G(d,p) level.
Progressing from more donating substituents to more with-
drawing substituents results in an easily observable decrease in
electron density at the boron atom of esters 11a−f as well as a
decrease in electron density around the periphery of the
catechol moiety.
Electronic and Structural Effects. Both experimental and
computational investigations reveal that the choice of donor
(7−10) plays a primary role in determining the favorability of
forming a condensation product with boronic acids 1−6, while
the para-substituent of the boronic acids plays a much more
secondary role. Still, the subtle influence of the para-substituent
of boronic acids 1−6 can be observed both structurally and
electronically when comparing results obtained from compu-
tations. For example, a general trend is observed in both
experimental and computed free energies of boronate ester
condensations wherein electron-donating substituents tend to
increase the favorability of forming boronate esters, while
electron-withdrawing substituents decrease the favorability of
their formation. This trend correlates reasonably well with
differences in B−O and C−B bond lengths measured in the
optimized structures of boronates 11a−f as shown in Table 4.
Boronate esters with donating substituents are found to exhibit
longer B−O bond lengths and shorter C−B bond lengths
relative to boronate esters with withdrawing groups, though the
Similar trends in electron density and C−B and B−N bond
lengths can be observed for diazaboroles 12a−f (see Figure S23
and Table S2 of the Supporting Information). Computations
predict greater electron density around the boron atom of each
diazaborole 12a−f relative to its corresponding boronate ester
analogue 11a−f. As mentioned earlier, the hydrolytic stability of
boronate esters correlates well with the electron density at their
boron atom. Diazaboroles are known to be less hydrolytically
stable than boronate esters, however their susceptibility to
hydrolysis results not from differences in electron density but
rather from general differences in the strengths of a typical B−
O versus B−N bond. No clear trend could be observed
between the nature of the para-substituent of boronic acids 1−
6 and the favorability of their condensation with ortho-
phenylenediamine 8. Both experimental and computational
results suggest that the least favored (most endergonic)
diazaborole condensation reactions involve the formation of
methoxy-substituted diazaborole 12a and fluoro-substituted
diazaborole 12d. Interestingly, the most favored (least ender-
gonic) diazaborole condensations are those that form
diazaboroles 12e and 12f bearing withdrawing methyl ester
and cyano functionalities, respectively. When discussing
experimental diazaborole results we had postulated that the
favorability of forming boroxine anhydride species may account
for differences in the equilibrium constants of diazaborole
formation, noting that the boroxine of para-methoxyphenylbor-
onic acid 1 is more thermodynamically stable17 than that of
para-cyanophenylboronic acid 6. Computational results suggest
that boroxine formation may not be the only influence, as
Table 4. Bond Lengths (Å) of B−O and C−B Bonds in
Boronate Esters 11a−f along with Hammett σ Constants for
Each para-Substituent
11a
11b
11c
11d
11e
11f
R
OMe
t-Bu
H
F
CO2Me
CN
σp
−0.27
1.397
1.533
−0.20
1.396
1.537
0.00
0.06
0.45
0.66
B−O (Å)
C−B (Å)
1.395
1.539
1.394
1.539
1.392
1.543
1.391
1.545
G
DOI: 10.1021/acs.joc.5b02548
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Figure 5. Energies (eV, B3LYP/6-311+G(d,p)) and electron density maps of the HOMOs and LUMOs of (a) catechol, (b) ortho-
phenylenediamine, and (c) benzene-1,2-dithiol along with the corresponding molecular orbitals upon condensation with phenylboronic acid.
calculations at the M06-2X/6-31+G(d,p) level are in line with
the experimental observation that the formation of cyano-
substituted diazaborole 12f is more favored than methoxy-
substituted diazaborole 12a even though computational results
do not take boroxine formation into account. It should be
stressed, however, that both computationally predicted and
experimentally observed differences between the condensation
free energies of diazaboroles 12a−f are so similar that they are
within error, and it is likely not possible to definitively
determine what accounts for the small differences between
condensation free energies.
Significantly greater differences are observed when compar-
ing condensation reactions involving different donors, i.e., 7−
10. To better understand these differences in reactivity, we
investigated the effects of borylation on catechol, ortho-
phenylenediamine, and benzene-1,2-dithiol, which serve as
model systems of tert-butylcatechol (7), tert-butyl-ortho-phenyl-
enediamine (8), and toluene-3,4-dithiol (10). More specifically,
natural bond order49 calculations were carried out to examine
the influence of borylation on the frontier orbitals of the three
different classes of donors. Figure 5 shows the HOMO and
LUMO orbitals of catechol (Figure 5a), ortho-phenylenedi-
amine (Figure 5b), and benzene-1,2-dithiol (Figure 5c) along
with the corresponding orbitals of their boronate ester,
diazaborole, and dithiaborol products formed upon condensa-
tion with phenylboronic acid 3. Catechol and ortho-phenyl-
enediamine each have a nondegenerate HOMO that
corresponds to the π bond between the 4 and 5 carbons
(C4−C5) of their respective aromatic rings. The LUMO orbitals
of catechol and ortho-phenylenediamine are both doubly
degenerate and correspond to π* antibonding orbitals between
C2−C3 and C6−C1. Condensation of catechol or ortho-
phenylenediamine with phenylboronic acid results in the
stabilization both of these sets of frontier orbitals. For example,
the HOMO of catechol (Figure 5a) is lowered by 0.16 eV upon
condensation to the corresponding boronate ester, while the
LUMO is 0.10 eV more stable. Borylation of ortho-phenyl-
enediamine to give the corresponding diazaborole also results
in a stabilization of the ortho-phenylenediamine HOMO and
LUMO orbitals, though to a lesser extent (0.04 and 0.03 eV,
respectively) as shown in Figure 5b.
The frontier orbitals and effects of borylation are notably
different in the case of benzene-1,2-dithiol as compared to
catechol and ortho-phenylenediamine. The HOMO of benzene-
1,2-dithiol is doubly degenerate and corresponds to the lone
pair Pz orbital of each of the two sulfur atoms (Figure 5c). The
LUMO is nondegenerate and corresponds to the C1−C2
antibonding π* orbital. Borylation of benzene-1,2-dithiol to
give the corresponding dithiaborole results in a 0.02 eV
stabilization of the C1−C2 π* antibonding orbital but a 0.40 eV
destabilization of the degenerate lone pair Pz orbitals of the
sulfur atoms. This destabilization mirrors the overall trend
observed throughout this study, namely that boronate ester
formation is somewhat more thermodynamically favorable than
diazaborole formation, while both boronate and diazaborole
formations are notably more favorable than dithiaborole
formation. This general trend can again be observed when
comparing the lengths of B−X bonds (where X = O, NH, or S),
wherein stronger bonds are typically shorter than the sum of
the covalent radii of their constituent atoms. Indeed B−O and
B−N(H) bond lengths obtained from calculations (Table 5)
are both 7% shorter than the sum of the covalent radii50 of their
constituent atoms. The calculated length of the B−S bond, by
contrast, is only 4% shorter than the sum of boron and sulfur
covalent radii. This observation further supports the conclusion
that B−S bonds of dithiaboroles are weaker than the B−O and
B−(NH) bonds of boronate esters and diazaboroles,
respectively. Collectively the computationally observed differ-
ences in bond lengths and the effects of borylation on the
frontier molecular orbitals of catechol, ortho-phenylenediamine,
and benzene-1,2-dithiol help rationalize the large differences
between reaction enthalpies (ΔH°, Table 3) for the formation
of aryl boronate ester, diazaborole, and dithiaborole con-
densation products.
H
DOI: 10.1021/acs.joc.5b02548
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
a
Table 5. Predicted and Calculated Bond Lengths of B−O,
B−(NH), and B−S Bonds in Boronate Ester 11c,
Diazaborole 12c, and Dithiaborole 14c
COMPUTATIONAL AND EXPERIMENTAL
METHODS
■
Computational Details. All calculations were performed with the
Gaussian09 suite of programs.51 Initial structures were built within the
GaussView interface and then conformationally searched by scanning
all easily rotating dihedral angles at the HF/6-31G(g) level of theory.
Approximate global energy minima obtained from conformational
searches were then optimized to full convergence and subjected to
vibrational and thermal analysis at each of four levels of theory:
B3LYP/6-311+G(d,p), M06-2X/6-31+G(d,p), CBS-QB3, and MP2/
aug-cc-pVDZ. Geometry optimizations and frequency analysis were
performed in an implicit solvent model for chloroform (ε = 4.7113)
using the PCM reaction field model.52 Stationary points were
confirmed as minima by the lack of any imaginary vibrational
frequencies. All reaction enthalpies (ΔH°), entropies (ΔS°), and free
energies (ΔG°) reported in this manuscript were calculated at 1.0 atm
pressure, 298.15 K, and in a solvent model for chloroform.
Materials. Chemicals and reagent-grade solvents were obtained
from commercial sources and used as purchased. Deuterated
chloroform was obtained from Cambridge Isotope Laboratories and
dried by passing a freshly opened bottle of CDCl3 over a glass column
packed with a small pad of anhydrous Na2SO4 on top of basic Al2O3
using a positive pressure of dry N2 gas. The resulting anhydrous
CDCl3 was collected in a flame-dried Schlenk flask containing 4 Å
molecular sieves and stored under nitrogen.
General Procedure for the Synthesis of Boronate Esters
11a−f. Analytical samples of boronate ester compounds 11a−f were
prepared for spectroscopic comparison with equilibrated mixtures
described in Scheme 2. Into a small round-bottom flask were added
200 mg (1.2 mmol) of tert-butyl catechol (7) and 1.0 molar equiv of a
given boronic acid (1−6). The solids were taken up in 12.0 mL of
CHCl3 and stirred at 50 °C for 3 h in the presence of a catalytic
amount of DOWEX. The resulting solution was cooled down to room
temperature, dried over anhydrous MgSO4, filtered, and concentrated
under reduced pressure. The resulting solids were washed with
hexanes and ether and then dried under high vacuum to afford pure
samples of esters 11a−f.
11a. Reaction Scale. 7 (200 mg, 1.2 mmol) and 1 (184 mg, 1.2
mmol), yield 295 mg (87%). The product was isolated as white solid.
Mp = 123−126 °C. TOF MS CI+ (m/z) [MH]+ calcd for
C17H2011BO3, 283.1506; found 283.1516. 1H NMR (300 MHz,
CDCl3, 298 K) δ = 8.05 (2H, d, J = 8.7 Hz), 7.38 (1H, d, J = 1.8
Hz), 7.22 (1H, d, J = 8.1 Hz), 7.15 (1H, dd, J = 8.1, 1.8 Hz), 7.03 (2H,
d, J = 8.7 Hz), 3.88 (3H, s), 1.39 (9H, s). 13C NMR (125 MHz,
CDCl3) δ = 163.0, 148.5, 146.4, 146.3, 136.8, 119.3, 113.9, 111.4,
109.8, 55.2, 34.8, 31.8.
bond
B−O
B−(NH)
B−S
predicted (Å)
calculated (Å)
difference
1.50
1.55
1.89
1.39
1.44
1.81
0.11 (7.3%)
0.11 (7.1%)
0.08 (4.2%)
a
Bond lengths are predicted from covalent radii of boron (0.84 Å),
oxygen (0.66 Å), nitrogen (0.71 Å), and sulfur (1.05 Å).50 Computed
bond lengths are reported at the M06-2X/6-31+G(d,p) level, however
they are identical to within 3 significant figures to bond lengths
computed using DFT and CBS-QB3. MP2 calculated bond lengths
were each 0.01 Å longer.
CONCLUSIONS
■
The results presented herein provide a greater understanding of
the energetics of dynamic covalent condensation reactions
between para-substituted arylboronic acids with diol, diamine,
mercaptophenol, and dithiol donors to give boronate esters,
diazaboroles, oxathiaboroles, and dithiaboroles, respectively.
Both experimental and computational results show that the
electron-donating or -withdrawing nature of the para-
substituent of the arylboronic acid plays a relatively small role
in influencing the favorability of forming a condensation
product with any of the donors. The choice of donor, on the
other hand, significantly influences the favorability of forming a
condensation product. Only two of the four classes of donors
undergo spontaneous self-assembly with the para-substituted
arylboronic acids investigated: 4-tert-butylcatechol (7) and 4-
tert-butyl-ortho-phenylenediamine (8). Catechol 7 readily
assembles with the arylboronic acids to give boronate ester
products in high yields (>90%) at equilibrium. The assembly of
diazaboroles from condensation of the ortho-phenylenediamine
8 and aylboronic acids is less favorable thermodynamically,
giving diazaborole products in more moderate yields (30−44%)
at equilibrium. Neither of the remaining two donors, 2-
hydroxybenzenethiol (9) and toluene-3,4-dithiol (10), could be
observed to undergo dynamic covalent self-assembly with any
of the arylboronic acids in chloroform at room temperature.
Computational results carried out at four different levels of
theory generally support the experimental observations that
para-substituents have little influence on the energetics of self-
assembly, while the donor has a large influence. While the
different computational methods agree on these overall trends,
they differ significantly in their predictions of reaction free
energies and reaction enthalpies for boronate ester, diazaborole,
oxathiaborole, and dithiaborole formation. Calculations carried
out at the M06-2X /6-31+G(d,p) level were found to be the
most consistent with experimental results, while B3LYP, CBS-
QB3, and MP2 calculations each predicted reaction energetics
that were notably more favorable than those found
experimentally. Given the increasing use of boronic acids in
the dynamic covalent self-assembly of polymers, COFs, and
other organic materials, it is important to understand how
different functionalities influence the favorability of assembly
formation. This study of the self-assembly of boronate esters,
diazaboroles, oxathiaboroles, and dithiaboroles in an organic
solvent helps provide that understanding, allows for direct
comparisons across different classes of donors and/or
arylboronic acids, and gives insight into which computational
methods are most appropriate for modeling these important
classes of compounds.
11b. Reaction Scale. 7 (200 mg, 1.2 mmol) and 2 (213 mg, 1.2
mmol), yield 336 mg (91%). The product was isolated as white solid.
Mp = 97−100 °C. TOF MS CI+ (m/z) [MH]+ calcd for C20H2611BO2,
1
309.2026; found 309.2030. H NMR (300 MHz, CDCl3, 298 K) δ =
8.09 (2H, d, J = 8.4 Hz), 7.57 (2H, d, J = 8.4 Hz), 7.43 (1H, d, J = 1.2
Hz), 7.27 (1H, d, J = 8.1 Hz), 7.19 (1H, dd, J = 8.1, 1.2 Hz), 1.42
(18H, s). 13C NMR (125 MHz, CDCl3) δ = 155.7, 148.5, 146.5, 146.4,
135.0, 126.3, 119.4, 111.5, 109.9, 35.1, 34.9, 31.8, 31.2.
11c. Reaction Scale. 7 (200 mg, 1.2 mmol) and 3 (148 mg, 1.2
mmol), yield 240 mg (79%). The product was isolated as white solid.
Mp = 49−50 °C. TOF MS EI+ (m/z) [M]+ calcd for C16H1711BO2,
252.13217; found 252.13228. 1H NMR (300 MHz, CDCl3, 298 K) δ =
8.15 (2H, d, J = 8.1 Hz), 7.62−7.50 (3H, m), 7.43 (1H, d, J = 1.5 Hz),
7.27 (1H, d, J = 8.7 Hz), 7.20 (1H, dd, J = 8.7, 1.5 Hz), 1.43 (9H, s).
13C NMR (125 MHz, CDCl3) δ = 148.5, 146.6, 146.3, 135.0, 132.3,
128.3, 119.5, 111.6, 110.0, 34.9, 31.9.
11d. Reaction Scale. 7 (200 mg, 1.2 mmol) and 4 (171 mg, 1.2
mmol), yield 276 mg (85%). The product was isolated as white solid.
Mp = 106−107 °C. TOF MS EI+ (m/z) [M]+ calcd for C16H1611BO2F,
270.12274; found 270.12250. 1H NMR (300 MHz, CDCl3, 298 K) δ =
8.08 (2H, t, J = 6.3 Hz), 7.36 (1H, d, J = 1.2 Hz), 7.26−7.15 (4H, m),
1.37 (9H, s). 13C NMR (125 MHz, CDCl3) δ = 166.7, 164.7, 148.4,
146.5, 138.0, 137.2, 119.5, 115.4, 111.5, 109.8, 34.8, 31.7.
I
DOI: 10.1021/acs.joc.5b02548
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
1
11e. Reaction Scale. 7 (200 mg, 1.2 mmol) and 5 (217 mg, 1.2
mmol), yield 340 mg (91%). The product was isolated as white solid.
Mp = 155−157 °C. TOF MS CI+ (m/z) [MH]+ calcd for
C18H2011BO4, 311.1455; found 311.1454. 1H NMR (300 MHz,
CDCl3, 298 K) δ = 8.14 (4H, s), 7.38, (1H, d, J = 1.8 Hz), 7.23
(1H, d, J = 8.1 Hz), 7.17 (1H, dd, J = 8.1 Hz, 1.8 Hz), 3.95 (3H, s),
1.36 (9H, s). 13C NMR (125 MHz, CDCl3) δ = 166.8, 148.4, 146.9,
146.1, 134.8, 133.4, 129.0, 119.7, 111.6, 110.0, 52.1, 34.8, 31.7.
11f. Reaction Scale. 7 (200 mg, 1.2 mmol) and 6 (177 mg, 1.2
mmol), yield 273 mg (82%). The product was isolated as white solid.
Mp = 174−176 °C. TOF MS EI+ (m/z) [M]+ calcd for
C18H2111BO2N2, 308.16962; found 308.16974. H NMR (300 MHz,
CDCl3, 298 K) δ = 8.14 (4H, br), 7.38 (1H, d, J = 1.8 Hz), 7.23 (1H,
d, J = 8.1 Hz), 7.17 (1H, dd, J = 8.1, 1.8 Hz), 3.95 (3H, s), 1.36 (9H,
s). 13C NMR (125 MHz, CDCl3) δ = 167.2, 143.2, 136.1, 133.9, 133.0,
130.9, 129.1, 116.8, 110.7, 108.6, 52.2, 34.5, 32.0.
12f. Reaction Scale. 8 (150 mg, 0.91 mmol) and 6 (134 mg, 0.91
mmol), yield 168 mg (67%). The product was isolated as a light-
brown oil that gradually solidifies upon sitting. Mp = 65−69 °C. TOF
MS EI+ (m/z) [M]+ calcd for C17H1811BN3, 275.15938; found
1
275.15974. H NMR (300 MHz, CDCl3, 298 K) δ = 7.79 (2H, d, J =
8.4 Hz), 7.69 (2H, d, J = 8.4 Hz), 7.21 (1H, s), 7.08−7.06 (2H, m),
6.82 (2H, br, −NH−), 1.37 (9H, s). 13C NMR (125 MHz, CDCl3) δ =
143.6, 136.2, 134.0, 133.5, 131.6, 125.5, 119.2, 117.2, 112.8, 111.0,
108.9, 34.6, 32.1.
C17H1611BO2N, 277.12741; found 277.12761. H NMR (300 MHz,
1
CDCl3, 298 K) δ = 8.16 (2H, d, J = 8.4 Hz), 7.76 (2H, d, J = 8.4 Hz),
7.39 (1H, d, J = 1.2 Hz), 7.22 (1H, d, J = 8.1 Hz), 7.20 (1H, dd, J =
8.1, 1.2 Hz), 1.36 (9H, s). 13C NMR (125 MHz, CDCl3) δ = 148.2,
147.3, 146.0, 135.2, 131.6, 1120.0, 118.4, 115.7, 111.8, 110.1, 34.9,
31.7.
General Procedure for the Equilibration of Boronate Esters
11a−f and Diazaboroles 12a−f. When setting up equilibration
reactions involving boronic acids, it is important to ensure that water
moisture is excluded from starting materials, glassware, and solvents so
that excess water does not unduly influence the resulting equilibrium.
Therefore, all NMR tubes, round-bottom flasks, and glass vials used
were oven-dried overnight at >115 °C. A freshly opened bottle of
CDCl3 was dried as described in the above Materials section. Stock
solutions of 4-tert-butylcatechol (7) and 4-tert-butyl-ortho-phenyl-
enediamine (8) were prepared by weighing out 108 mg of 7 and 106.8
mg of 8 (6.5 × 10−4 mol in each case) into separate 25 mL round-
bottom flasks. The flasks were then capped with a rubber septum,
placed under vacuum, briefly flame-dried to further remove trace
moisture, and placed under dry N2(g). Once cooled down to room
temperature, 13 mL of dry CDCl3 was added to each sample of 7 and
8 via syringe, resulting in 0.05 M stock solutions of each donor. Into
separate 2 dram vials were weighed 0.05 mmol of each boronic acid
1−6. Each vial was capped with a septum and placed under vacuum for
10 min (Note: placing the boronic acids under vacuum for longer
periods of time, oven drying, or exposing them to flame drying, results
in significant to quantitative formation of their boroxine anhydrides.
This should be avoided as it changes the nature, and therefore
energetics, of the dynamic covalent reactions leading to 11a−f and
12a−f because the starting materials are no longer boronic acids 1−6).
A 1.0 mL sample of the stock solution of 7 or 8 was added to each
boronic acid vial giving a 1:1 molar ratio of donor and boronic acid
(0.05 M). Each mixture was briefly agitated, transferred to an oven-
dried NMR tube via syringe, and capped, and the caps were wraped
General Procedure for the Synthesis of Diazaboroles 12a−f.
Analytical samples of diazaboroles 12a−f were prepared by suspending
150 mg (0.91 mmol) of 4-tert-butyl-ortho-phenylenediamine (8) and
1.0 molar equiv of a given boronic acid (1−6) in 9.1 mL of toluene in
the presence of 4 Å molecular sieves. The mixtures were stirred at 80
°C for 8 h and then filtered hot, washing with hot (∼80 °C) toluene.
Diazaboroles 12a, 12b, and 12e (R = OMe, t-Bu, and CO2Me,
respectively) crystallized directly from their toluene solutions upon
cooling, and the resulting crystals were washed with hexanes and dried
under vacuum to give pure diazaborole products. Additional quantities
of diazaboroles 12b and 12e could be obtained upon concentrating the
toluene filtrate. Diazaboroles 12c and 12d (R = F and Ph, respectively)
were isolated by concentrating their toluene filtrates and recrystallizing
from warm hexanes. Lastly, diazaborole 12f (R = CN) required
additional heating (refluxing toluene, Dean−Stark apparatus, 18 h)
followed by drying the resulting solution over MgSO4, filtering,
concentrating under reduced pressure, and washing with cold hexanes.
12a. Reaction Scale. 8 (150 mg, 0.91 mmol) and 1 (138 mg, 0.91
mmol), yield 242 mg (95%). The product was isolated as needle-like
white solid crystals. Mp = 253−255 °C. TOF MS EI+ (m/z) [M]+
calcd for C17H2111BON2, 280.17471; found 280.17469. 1H NMR (300
MHz, CDCl3, 298 K) δ = 7.66 (2H, d, J = 8.4 Hz), 7.15 (1H, s), 7.01−
6.96 (4H, m), 6.63 (2H, br, −NH−), 3.86 (3H, s), 1.36 (9H, s). 13C
NMR (125 MHz, CDCl3, poor solubility limited signal-to-noise of
quaternary carbon signals) δ = 134.5, 116.2, 113.9, 110.2, 108.2, 55.1,
32.0.
1
with Teflon tape. H NMR spectra were periodically taken of each
mixture until no changes in spectral signals or their integration could
be observed, indicating that equilibrium had been reached.
12b. Reaction Scale. 8 (150 mg, 0.91 mmol) and 2 (163 mg, 0.91
mmol), yield 245 mg (88%). The product was isolated as white
crystalline solid. Mp = 213−215 °C. TOF MS EI+ (m/z) [M]+ calcd
for C20H2711BN2, 306.22674; found 306.22709. H NMR (300 MHz,
1
ASSOCIATED CONTENT
* Supporting Information
■
CDCl3, 298 K) δ = 7.72 (2H, d, J = 8.1 Hz), 7.51 (2H, d, J = 8.1 Hz),
7.21 (1H, s), 7.10−7.04 (4H, m), 6.74−6.71 (2H, br, −NH−), 1.41−
1.40 (18H, s). 13C NMR (125 MHz, CDCl3) δ = 152.8, 142.8, 136.3,
134.2, 132.9, 125.2, 116.4, 110.4, 108.3, 34.8, 34.5, 32.0, 31.3.
12c. Reaction Scale. 8 (150 mg, 0.91 mmol) and 3 (111 mg, 0.91
mmol), yield 191 mg (84%). The product was isolated as a white solid.
Mp = 191−192 °C. TOF MS EI+ (m/z) [M]+ calcd for C16H1911BN2,
250.16414; found 250.16439. 1H NMR (300 MHz, CDCl3, 298 K) δ =
7.77 (2H, dd, J = 9.3, 3.3 Hz), 7.49−7.46 (3H, m), 7.22 (1H, s), 7.10−
7.07 (2H, m), 6.73 (2H, br, −NH−), 1.42 (9H, s). 13C NMR (125
MHz, CDCl3) δ = 152.8, 142.8, 136.3, 134.2, 132.9, 125.2, 116.4,
110.4, 108.3, 34.8, 34.5, 32.0, 31.3.
12d. Reaction Scale. 8 (150 mg, 0.91 mmol) and 4 (127 mg, 0.91
mmol), yield 214 mg (88%). The product was isolated as white solid
flakes. Mp = 186−187 °C. TOF MS EI+ (m/z) [M]+ calcd for
C16H1811BN2F, 268.15471; found 268.15459. 1H NMR (300 MHz,
CDCl3, 298 K) δ = 7.71 (2H, t, J = 6.3 Hz), 7.20 (1H, s), 7.14 (2H, t, J
= 6.3 Hz), 7.08−7.05 (2H, m), 6.67 (2H, br, −NH−), 1.40 (9H, s).
13C NMR (125 MHz, CDCl3) δ = 165.1, 163.1, 143.0, 136.2, 134.8,
134.0, 116.6, 115.2, 110.5, 108.4, 34.5, 31.9.
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.joc.5b02548.
1H and 13C NMR spectra of boronate esters 11a−f and
1
diazaboroles 12a−f, H NMR spectra of boronic acids
1
1−6 and their boroxine anhydrides, H NMR spectra of
equilibrated mixtures of all boronate ester and diazabor-
ole assemblies, expanded tables detailing the relative
energetics, bond lengths, and bond angles of different
boronic acid conformations, tables of calculated bond
distances of diazaboroles 12a−f, oxathiaboroles 13a−f,
and dithiaboroles 14a−f, calculated electron density
maps of diazaborole and dithiaborole species, Cartesian
coordinates of all stationary points reported in this
manuscript and their absolute free energies in hartrees,
and complete ref 51 (PDF)
AUTHOR INFORMATION
Corresponding Author
*E-mail: bnorthrop@wesleyan.edu.
■
12e. Reaction Scale. 8 (150 mg, 0.91 mmol) and 5 (164 mg, 0.91
mmol), yield 230 mg (82%). The product was isolated as a white solid.
Mp = 212−213 °C. TOF MS EI+ (m/z) [M]+ calcd for
J
DOI: 10.1021/acs.joc.5b02548
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Giovino, M. R.; White, S. L.; Dichtel, W. R. Chem. Sci. 2011, 2, 1588−
1593.
(17) Tokunaga, Y.; Ueno, H.; Shimomura, Y.; Seo, T. Heterocycles
Notes
The authors declare no competing financial interest.
2002, 57, 787−790.
ACKNOWLEDGMENTS
■
(18) Tokunaga, Y.; Ueno, H.; Shimomura, Y. Heterocycles 2007, 74,
219−223.
This research was supported by Wesleyan University. We thank
Wesleyan University for computer time supported by the NSF
under grant number CNS-0619508.
(19) Kua, J.; Iovine, P. J. Phys. Chem. A 2005, 109, 8938−8943.
(20) Fielder, W. L.; Chamberlain, M. M.; Brown, C. A. J. Org. Chem.
1961, 26, 2154−2155.
(21) (a) Das, M. K.; Mariategui, J. F.; Niedenzu, K. Inorg. Chem.
1987, 26, 3114−3116. (b) Mariategui, J. F.; Niedenzu, K. J. Organomet.
Chem. 1989, 369, 137−145.
(22) Beckett, M. A.; Strickland, G. C.; Varma, K. S.; Hibbs, D. E.;
Hursthouse, M. B.; Malik, K. M. A. J. Organomet. Chem. 1997, 535,
33−41.
(23) Perttu, E. K.; Arnold, M.; Iovine, P. M. Tetrahedron Lett. 2005,
46, 8753−8756.
(24) (a) Kua, J.; Gyselbrecht, C. R. J. Phys. Chem. A 2008, 112,
9128−9133. (b) Iovine, P. M.; Gyselbrecht, C. R.; Perttu, E. K.; Klick,
C.; Neuwelt, A.; Loera, J.; DiPasquale, A. G.; Rheingold, A. L.; Kua, J.
Dalton Trans. 2008, 3791−3794. (c) Kua, J.; Gyselbrecht, C. R. J. Phys.
Chem. A 2007, 111, 4759−4776. (d) Kua, J.; Fletcher, M. N.; Iovine, P.
M. J. Phys. Chem. A 2006, 110, 8158−8166.
(25) Substituted diazaboroles have more commonly been prepared
by reacting 2-bromo-1,3-di(alkyl)-1,3,2-diazaboroles with alkyl or aryl
lithium compounds than by boronic acid-diamine condensation
REFERENCES
■
(1) (a) Wu, X.; Li, Z.; Chen, X. X.; Fossey, J. S.; James, T. D.; Jiang,
Y. B. Chem. Soc. Rev. 2013, 42, 8032−8048. (b) Whyte, G. F.; Vilar, R.;
Woscholski, R. J. Chem. Biol. 2013, 6, 161−174. (c) Bull, S. D.;
Davidson, M. G.; Vanden Elsen, J. M. H.; Fossey, J. S.; Jenkins, A. T.
A.; Jiang, Y. B.; Kubo, Y.; Marken, F.; Sakurai, K.; Zhao, J. Z.; James, T.
D. Acc. Chem. Res. 2013, 46, 312−326. (d) Mader, H. S.; Wolfbeis, O.
S. Microchim. Acta 2008, 162, 1−34. (e) Cao, H. S.; Heagy, M. D. J.
Fluoresc. 2004, 14, 569−584. (f) Wang, W.; Gao, X. M.; Wang, B. H.
Curr. Org. Chem. 2002, 6, 1285−1317.
(2) (a) Martinez-Aguirre, M. A.; Yatsimirsky, A. K. J. Org. Chem.
2015, 80, 4985−4993. (b) Zhou, Y.; Zhang, J.; Yoon, J. Chem. Rev.
2014, 114, 5511−5571. (c) Wade, C. R.; Broomsgrove, A. E. J.;
Aldridge, S.; Gabbai, F. P. Chem. Rev. 2010, 110, 3958−3984.
(3) (a) Ito, S.; Ono, K.; Iwasawa, N. J. Am. Chem. Soc. 2012, 134,
13962−13965. (b) Christinat, N.; Scopelliti, R.; Severin, K. Angew.
Chem. 2008, 120, 1874−1878. (c) Iwasawa, N.; Takahagi, H. J. Am.
Chem. Soc. 2007, 129, 7754−7755.
reactions (see, e.g.: Weber, L.; Eickhoff, D.; Werner, V.; Bohling, L.;
̈
(4) (a) Nishimura, N.; Kobayashi, K. J. Org. Chem. 2010, 75, 6079−
6085. (b) Nishimura, N.; Yoza, K.; Kobayashi, K. J. Am. Chem. Soc.
Schwedler, S.; Chrostowska, A.; Dargelos, A.; Maciejczyk, M.;
Stammler, H.-G.; Neumann, B. Dalton Trans. 2011, 40, 4434−4446.
(26) Letsinger, R. L.; Hamilton, S. B. J. Am. Chem. Soc. 1958, 80,
5411−5413.
(27) Maruyama, S.; Kawanishi, Y. J. Mater. Chem. 2002, 12, 2245−
2249.
2010, 132, 777−790. (c) Icli, B.; Christinat, N.; Tonnemann, J.;
̈
Schuttler, C.; Scopelliti, R.; Severin, K. J. Am. Chem. Soc. 2009, 131,
̈
3154−3155. (d) Takahagi, H.; Fujibe, S.; Iwasawa, N. Chem. - Eur. J.
2009, 15, 13327−13330. (e) Nishimura, N.; Kobayashi, K. Angew.
Chem., Int. Ed. 2008, 47, 6255−6258. (f) Kataoka, K.; James, T. D.;
Kubo, Y. J. Am. Chem. Soc. 2007, 129, 15126−15127.
(28) Yamaguchi, I.; Choi, B.-J.; Koizumi, T.-a.; Kubota, K.;
Yamamoto, T. Macromolecules 2007, 40, 438−443.
(29) Kojima, T.; Kumaki, D.; Nishida, J.-I.; Tokito, S.; Yamashita, Y.
J. Mater. Chem. 2011, 21, 6607−6613.
(5) See, for example: (a) Cromwell, O. R.; Chung, J.; Guan, Z. J. Am.
Chem. Soc. 2015, 137, 6492−6495. (b) Cai, M.; Daniel, S. L.; Lavigne,
J. J. Chem. Commun. 2013, 49, 6504−6506. (c) Roy, D.; Cambre, J. N.;
Sumerlin, B. S. Chem. Commun. 2009, 2106−2108. (d) Sanchez, J. C.;
Trogler, W. C. J. Mater. Chem. 2008, 18, 5134−5141. (e) Christinat,
(30) (a) Yamashita, M.; Yamamoto, Y.; Akiba, K.-Y.; Nagase, S.
Angew. Chem., Int. Ed. 2000, 39, 4055−4058. (b) Yamashita, M.;
Yamamoto, Y.; Akiba, K.-Y.; Hashizume, D.; Iwasaki, F.; Takagi, N.;
Nagase, S. J. Am. Chem. Soc. 2005, 127, 4354−4371.
(31) Goswami, A.; Maier, C.-J.; Pritzkow, H.; Siebert, W. Eur. J. Inorg.
Chem. 2004, 2004, 2635−2645.
(32) Solomon, S. A.; Del Grosso, A.; Clark, E. R.; Bagutski, V.;
McDouall, J. J. W.; Ingleson, M. J. Organometallics 2012, 31, 1908−
1916.
N.; Croisier, E.; Scopelliti, R.; Cascella, M.; Rothlisberger, U.; Severin,
̈
K. Eur. J. Inorg. Chem. 2007, 2007, 5177−5181. (f) Niu, W.; Smith, M.
D.; Lavigne, J. J. J. Am. Chem. Soc. 2006, 128, 16466−16467. (g) Niu,
W.; O’Sullivan, C.; Rambo, B. M.; Smith, M. D.; Lavigne, J. J. Chem.
Commun. 2005, 4342−4344. (h) Nakazawa, I.; Suda, S.; Masuda, M.;
Asai, M.; Shimizu, T. Chem. Commun. 2000, 881−882. (i) Mikami, M.;
Shinkai, S. Chem. Lett. 1995, 24, 603−604.
(6) For recent reviews of COFs see: (a) Xiang, Z.; Cao, D.; Dai, L.
Polym. Chem. 2015, 6, 1896−1911. (b) Colson, J. W.; Dichtel, W. R.
Nat. Chem. 2013, 5, 453−465. (c) Ding, S.-Y.; Wang, W. Chem. Soc.
Rev. 2013, 42, 548−568. (d) Feng, X.; Ding, X.; Jiang, D. Chem. Soc.
Rev. 2012, 41, 6010−6022. (e) Furukawa, H.; Yaghi, O. M. J. Am.
Chem. Soc. 2009, 131, 8875−8883.
(33) For reviews of dynamic covalent self-assembly see: (a) Jin, Y.;
Wang, Q.; Taynton, P.; Zhang, W. Acc. Chem. Res. 2014, 47, 1575−
1586. (b) Corbett, P. T.; LeClaire, J.; Vial, L.; West, K. R.; Wietor, J.-
L.; Sanders, J. M. K.; Otto, S. Chem. Rev. 2006, 106, 3652−3711.
(c) Rowan, S. J.; Cantrill, S. J.; Cousins, G. R. L.; Sanders, J. K. M.;
Stoddart, J. F. Angew. Chem., Int. Ed. 2002, 41, 898−952.
(34) Cabiddu, S.; Maccioni, A.; Mura, L.; Secci, M. J. Heterocycl.
Chem. 1975, 12, 169−170.
(35) (a) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B: Condens. Matter
Mater. Phys. 1988, 37, 785−789. (b) Becke, A. D. J. Chem. Phys. 1993,
98, 1372−1377. (c) Becke, A. D. J. Chem. Phys. 1993, 98, 5648−5652.
(36) (a) Zhao, Y.; Truhlar, D. G. Theor. Chem. Acc. 2008, 120, 215−
241. (b) Peverati, R.; Truhlar, D. G. Philos. Trans. R. Soc., A 2014, 372,
20120476.
(7) Hall, D. G. Boronic Acids, 1st ed; Wiley-VCH: Weinheim, 2006;
pp 1−26.
(8) Severin, K. Dalton Trans. 2009, 5254−5264.
(9) Fujita, N.; Shinkai, S.; James, T. D. Chem. - Asian J. 2008, 3,
1076−1091.
(10) Korich, A. L.; Iovine, P. M. Dalton Trans. 2010, 39, 1423−1431.
(11) Tokunaga, Y. Heterocycles 2013, 87, 991−1021.
(12) Lorand, J. P.; Edwards, J. O. J. Org. Chem. 1959, 24, 769−774.
(13) Springsteen, G.; Wang, B. Tetrahedron 2002, 58, 5291−5300.
(14) Arzt, M.; Seidler, C.; Ng, D. Y. W.; Weil, T. Chem. - Asian J.
2014, 9, 1994−2003.
(37) (a) Nyden, M. R.; Petersson, G. A. J. Chem. Phys. 1981, 75,
1843−1862. (b) Petersson, G. A.; Al-Laham, M. A. J. Chem. Phys.
1991, 94, 6081−6090. (c) Petersson, G. A.; Tensfeldt, T.;
Montgomery, J. A. J. Chem. Phys. 1991, 94, 6091−6101. (d) Mont-
gomery, J. A.; Ochterski, J. W.; Petersson, G. A. J. Chem. Phys. 1994,
101, 5900−5909.
(15) Marinaro, W. A.; Prankerd, R.; Kinnari, K.; Stella, V. J. J. Pharm.
Sci. 2015, 104, 1399−1408.
(16) Dichtel et al. have investigated dynamic exchange processes
between different boronate esters and boronic acids as part of a
mechanistic investigation of COF formation, see: Spitler, E. L.;
(38) Moler, C.; Plesset, M. Phys. Rev. 1934, 46, 618−622.
K
DOI: 10.1021/acs.joc.5b02548
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
(39) (a) Dunning, T. H., Jr. J. Chem. Phys. 1989, 90, 1007−1023.
(b) Woon, D. E.; Dunning, T. H., Jr. J. Chem. Phys. 1993, 98, 1358−
1371.
(40) Bock, C. W.; Larkin, J. D. Comput. Theor. Chem. 2012, 986, 35−
42.
(41) Bhat, K. L.; Markham, G. D.; Larkin, J. D.; Bock, C. W. J. Phys.
Chem. A 2011, 115, 7785−7793.
(42) Monajemi, H.; Cheah, M. H.; Lee, V. S.; Zain, S. M.; Abdullah,
W. A. T. W. RSC Adv. 2014, 4, 10505−10513.
(43) Smith, M. K.; Northrop, B. H. Chem. Mater. 2014, 26, 3781−
3795.
(44) Zhao, Y.; Truhlar, D. G. J. Chem. Theory Comput. 2011, 7, 669−
676.
(45) Burns, L. A.; Vazquez-Mayagoitia, A.; Sumpter, B. G.; Sherrill,
C. D. J. Chem. Phys. 2011, 134, 084107.
(46) Jacquemin, D.; Perpete, E. A.; Ciofini, I.; Adamo, C.; Valero, R.;
̀
Zhao, Y.; Truhlar, D. G. J. Chem. Theory Comput. 2010, 6, 2071−2085.
(47) (a) Ess, D. H.; Houk, K. N. J. Phys. Chem. A 2005, 109, 9542−
9553. (b) Gruner, V.; Khuong, K. S.; Leach, A. G.; Lee, P. S.;
Bartberger, M. D.; Houk, K. N. J. Phys. Chem. A 2003, 107, 11445−
11459.
(48) Larkin, J. D.; Bhat, G. D.; Markham, G. D.; Brooks, B. R.;
Schaefer, H. F.; Bock, C. W. J. Phys. Chem. A 2006, 110, 10633−
10642.
(49) Glendening, E. D.; Weinhold, F. J. Comput. Chem. 1998, 19,
610−627.
(50) Cordero, B.; Gom
Echeverría, J.; Cremades, E.; Barragan
2008, 2832−2838.
́
ez, V.; Platero-Prats, A. E.; Reves
́
, M.;
́
, F.; Alvarez, S. Dalton Trans.
(51) Frisch, M. J. et al. Gaussian 09, revision A.1, see Supporting
Information.
(52) (a) Miertus, S.; Scrosso, E.; Tomasi, J. Chem. Phys. 1981, 55,
117−129. (b) Cossi, M.; Barone, V.; Cammi, R.; Tomasi, J. Chem.
Phys. Lett. 1996, 255, 327−335.
L
DOI: 10.1021/acs.joc.5b02548
J. Org. Chem. XXXX, XXX, XXX−XXX