Correlation of Intermolecular Acyl Transfer Reactivity with Noncovalent Lattice Interactions in Molecular Crystals: Toward Prediction of Reactivity of Organic Molecules in the Solid State

Article abstract of DOI:10.1021/acs.joc.8b00293

Intermolecular acyl transfer reactivity in several molecular crystals was studied, and the outcome of the reactivity was analyzed in the light of structural information obtained from the crystals of the reactants. Minor changes in the molecular structure resulted in significant variations in the noncovalent interactions and packing of molecules in the crystal lattice, which drastically affected the facility of the intermolecular acyl transfer reactivity in these crystals. Analysis of the reactivity vs crystal structure data revealed dependence of the reactivity on electrophile···nucleophile interactions and C-H···π interactions between the reacting molecules. The presence of these noncovalent interactions augmented the acyl transfer reactivity, while their absence hindered the reactivity of the molecules in the crystal. The validity of these correlations allows the prediction of intermolecular acyl transfer reactivity in crystals and co-crystals of unknown reactivity. This crystal structure-reactivity correlation parallels the molecular structure-reactivity correlation in solution-state reactions, widely accepted as organic functional group transformations, and sets the stage for the development of a similar approach for reactions in the solid state.

Full text of DOI:10.1021/acs.joc.8b00293

Article
pubs.acs.org/joc
Cite This: J. Org. Chem. XXXX, XXX, XXX−XXX
Correlation of Intermolecular Acyl Transfer Reactivity with
Noncovalent Lattice Interactions in Molecular Crystals: Toward
Prediction of Reactivity of Organic Molecules in the Solid State
Shobhana Krishnaswamy† and Mysore S. Shashidhar*
Division of Organic Chemistry, CSIR-National Chemical Laboratory, Pune 411008, India
S
* Supporting Information
ABSTRACT: Intermolecular acyl transfer reactivity in several molecular
crystals was studied, and the outcome of the reactivity was analyzed in the
light of structural information obtained from the crystals of the reactants.
Minor changes in the molecular structure resulted in significant variations
in the noncovalent interactions and packing of molecules in the crystal
lattice, which drastically affected the facility of the intermolecular acyl
transfer reactivity in these crystals. Analysis of the reactivity vs crystal
structure data revealed dependence of the reactivity on electrophile···
nucleophile interactions and C−H···π interactions between the reacting molecules. The presence of these noncovalent
interactions augmented the acyl transfer reactivity, while their absence hindered the reactivity of the molecules in the crystal. The
validity of these correlations allows the prediction of intermolecular acyl transfer reactivity in crystals and co-crystals of unknown
reactivity. This crystal structure−reactivity correlation parallels the molecular structure−reactivity correlation in solution-state
reactions, widely accepted as organic functional group transformations, and sets the stage for the development of a similar
approach for reactions in the solid state.
molecules, while their reactivity in solids is determined largely by
packing of (neighboring) molecules. The current article attempts
to arrive at parameters that can be used to recognize the ease of
acyl transfer reactivity in molecular crystals by a comparison of
the crystal structures of several O-acylated-myo-inositol 1,3,5-
orthoesters and the facility of acyl transfer reactivity in these
crystals.
The acyl group transfer between oxygen atoms of the same or
different molecules is basically a transesterification reaction
which involves the addition of the hydroxyl (−OH) group (a
nucleophile, Nu) to an ester carbonyl group (an electrophile, El)
followed by elimination of a different alcohol (−OH group)
leading to the formation of a new ester. If the −OH group being
added is that of a water molecule or another alcohol, the reaction
is referred to as hydrolysis or alcoholysis, respectively. Acyl
transfer reactions have been studied extensively in the solution
state since ester hydrolysis and transesterification reactions are
ubiquitous in flasks and living cells. Examples of the same
reaction are however rare in the crystalline state.7 We had earlier
reported8 the synthesis and structures of several molecular
crystals and co-crystals that exhibit efficient intermolecular acyl
transfer reactions (Scheme 1).
INTRODUCTION
■
Crystal engineering is the design and construction of molecular
solids with desired physical and/or chemical properties, utilizing
the principles of molecular recognition, through noncovalent
intermolecular interactions.1 Noncovalent interactions such as
hydrogen bonding,2 halogen bonding,3 C−H···π,4 halogen···
halogen,5 and π···π6 interactions between the appropriate
molecular building blocks help in the creation of solids which
possess the desired properties or structures for a designated
function. However, these interactions are energetically weak and
hence competitive, resulting in the formation of other molecular
assemblies of comparable energies. The concomitant crystal-
lization of polymorphs and solvates is a consequence of various
competitive solute−solute and solute−solvent interactions
during the process of nucleation and crystal growth. Therefore,
routes for the construction of molecular assemblies of desired
structure and/or properties cannot be visualized and realized
routinely. This is in contrast to the construction of single
molecular entities wherein covalent linkages which are far more
stable (compared to noncovalent interactions) at ambient
temperature are made. This is obvious when we try to identify
the structure of a molecule that is expected to undergo a
particular kind of a reaction (the “organic functional group”
approach in solution-state reactions) as compared to the
identification of a molecular crystal that is capable of undergoing
a reaction (in the crystalline state). In other words, it is far more
difficult to recognize the chemical reactions that a molecule can
undergo in its crystal as compared to its reactions in solution.
This is because the reactivity of small organic molecules in fluid
phases depends largely on electronic and steric factors within
There are very few molecular crystals that exhibit efficient
reactions (as compared to corresponding reactions in the
solution state), hence, they evoke intense interest among
chemists.9 This is mainly because the reactions in crystals exhibit
selectivities (regio- as well as stereo-) that are often not possible
Received: January 31, 2018
Published: March 14, 2018
© XXXX American Chemical Society
A
DOI: 10.1021/acs.joc.8b00293
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Scheme 1. Examples of Intermolecular Acyl Transfer
Reactions in the Solid State
Scheme 2. Synthesis of O-acylated-myo-inositol 1,3,5-
orthoesters 8−11
these crystals in the presence of solid sodium carbonate were
carried out by experimental procedures developed earlier in our
laboratory.8a
Assembly of the Molecules in Crystals of 8−11. Glide-
related molecules in the crystals of 8 form chains along the ac-
diagonal through O−H···O hydrogen-bonding interactions
which are connected along the b-axis by a set of four C−H···O
interactions, forming a molecular layer (Figure S21, Table S2,
SI). Scrutiny of the molecular assembly revealed that the distance
between the nucleophilic (−OH group) and electrophilic (ester
CO group) of adjacent molecules along the b-axis is 5.945(4)
Å, and the angle between them is 86.63° (Figure 1a). The El···Nu
distance is longer than the values observed in crystals that
supported acyl transfer reactivity (see below).
Analysis of the organization of molecules in crystals of the p-
iodo (9) and the p-methyl (10) dibenzoates (Figures S22 and
S23, SI) reveal that the hydroxyl group in 9 (and 10) is engaged
in hydrogen bonding with the orthoester oxygen (O4−H4A···
O1), forming a homochiral molecular chain along the a-axis.
Neighboring homochiral molecular chains are linked by C5−
H5···O8 contacts and a pair of centrosymmetric C−H···O
interactions (C7−H7···O8, C6−H6···O5), forming bilayers
(Figures S22a and S23a, Table S2, SI).16 The bilayers are linked
by halogen bonding (C12−I1···O7C8− of equatorial p-
iodobenzoate group) interactions in crystals of 9 (Figure S22b,
Table S2, SI) while centrosymmetric C−H···π interactions link
adjacent bilayers along the bc-diagonal in crystals of 10 (Figures
S23b, Table S2, SI). Hence, there exists a one-dimensional iso-
structurality in these crystal structures. Preference for molecular
organization through intermolecular interactions as described
above does not bring the hydroxyl group and the ester carbonyl
group (El-Nu) in the proximity required for the acyl transfer
reaction, in crystals of 9 and 10 (Figure 1b, 1c).
In crystals of the p-toluate 11, adjacent glide related molecules
are bound by O4−H4A···O7 (of the p-toluate CO) hydrogen
bonds to form a chain (Figure S24a, Table S2, SI). Adjacent
antiparallel chains are linked by C13−H13···O1 contacts
between the aromatic proton H13 of the p-toluate ester and
orthoester oxygen O1 along the ab-diagonal. The oxygen atom of
the second hydroxyl group (O6) forms short O···CO contacts
- El···Nu interactions (O6···C8O7: 2.909(3) Å, ∠O6···C8
O7: 87.1°) and also engages in C−H···O interactions (C3−H3···
O6) with neighboring molecules along the c-axis (Figure S24b,
Table S2, SI). The promising criteria for acyl transfer reaction in
11 are exhibited by unit translated molecules along the c-axis
to achieve in solution. The selectivities observed during reactions
in crystals are comparable to those observed in enzyme-catalyzed
reactions.10 This is also illustrated by complete enantioselectivity
during product formation in crystals comprised solely of achiral
molecules.11 Furthermore, the study of structure of reactive
crystals aids in understanding of the corresponding reaction
mechanisms and can also help in identification of other crystals
that can facilitate similar reactions. The latter possibility, in fact,
amounts to the prediction of reactivity of molecular crystals.
Reactions in the solid state also have implications for the stability
of APIs and drug formulations.12 The present manuscript
describes the correlation of the acyl transfer reactivity with crystal
structure and inherent noncovalent interactions of several
molecular crystals. This is perhaps the first attempt at systematic
correlation of structure and reactivity in molecular crystals for
reactions other than addition to C−C multiple bonds.13 The fact
that conclusions arrived at during this analysis allowed us to
identify molecular crystals (from a survey of the Cambridge
Crystal Structure Database, CSD), which can support
intermolecular acyl transfer reactions in them,8a establishes the
validity of this approach. In a wider perspective, prediction of the
structure of possible products in organic reactions is of current
interest and relevance to a wide cross section of researchers
ranging from chemists to engineers involved in data mining and
machine learning.14
RESULTS AND DISCUSSION
■
The racemic tosylate 8 was prepared by sequential reaction of the
triol 7 with tosyl chloride and benzoyl chloride (Scheme 2). The
racemic hydroxyl esters 9, 10 and 11 were prepared by the O-
acylation of myo-inositol orthoformate 7 with an appropriate acyl
chloride. The outcome of these O-substitution reactions in the
solution state could be predicted based on prior art.15 All these
hydroxyl esters were crystallized from common organic solvents;
the crystals obtained were suitable for single crystal X-ray
diffraction experiments. Acyl transfer reactivity experiments in
B
DOI: 10.1021/acs.joc.8b00293
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Figure 1. Electrophile-nucleophile geometry observed between neighboring molecules in crystals of 8−11. These diagrams were generated using the
corresponding packing diagrams (Figure S25, SI) obtained with the help of the program ‘Mercury 3.8’. Hydrogen atoms other than those of the hydroxyl
groups are omitted for clarity.
(Figure 1d). The distance between the potential reaction centers,
the nucleophilic O6 and the electrophilic C8O7, and the
corresponding angle of approach are both marginally lesser in
magnitude (hence marginally better) than the values observed in
reactive crystals.17 The reactive molecular partners exhibit weak
C−H···π interactions between the toluoyl group and the inositol
ring H atoms (Figure S24b, SI).18
The p-toluate diol 11 reacted in the solid state in the presence
of sodium carbonate, yielding the diester 10, the triester 20, and
the triol 7 (Scheme 3). The crystal structure of 11 revealed that
Scheme 3. Acyl Transfer Reactivity in Crystals of 11
Solid-State Reactivity−Structure Correlation. Heating a
mixture of 8 and sodium carbonate below the melting point of
racemic 8 led largely to charring of racemic 8. Analysis of the
resulting solid indicated the presence of 2-O-tosyl-4,6-di-O-
benzoyl-myo-inositol 1,3,5-orthoformate, 2-O-tosyl-myo-inositol
1,3,5-orthoformate,19 and the starting tosylate 8 in low yields. In
crystals of 8, the distance between the nucleophilic (Nu) −OH
(O4−H4A) group and the electrophilic ester carbonyl group of
the axial benzoate of adjacent molecules along the b-axis is
(5.945(4) Å) much longer than that observed in the reactive
crystals,17 leading to lower yield of the products, although, the
angle of approach, ∠O4···C8O7 (86.63°) is in the favorable
range for acyl transfer reactivity (see below).
While the crystals of the di-p-iodobenzoate 9 reacted in the
solid state yielding a mixture of products, the di-p-methyl-
benzoate 10 was unreactive under similar conditions. The lack of
reactivity in these crystals is consistent with molecular
organization which does not bring the El-Nu in proximity
required for the acyl transfer reaction.
the O6−H6A hydroxyl group makes a short O···CO contact
with the carbonyl carbon C8 of a unit translated molecule along
the c-axis which is in fact an El−Nu interaction with the distance
separating them, 2.909(3) Å, and the angle of approach of Nu
toward El, ∠O6···C8O7, 87.1° (Table 1). These values
correspond well with those observed for the reactive dibenzoates
18 and 19,17 and the distance between the electrophilic and
nucleophilic centers is lesser than the sum of their van der Waals
radii. p-Toluoyl group transfer is initiated in the presence of the
base, and the equatorial ester migrates to the axial position of the
next molecule, thereby forming the diester 10 and the triol 7.
C
DOI: 10.1021/acs.joc.8b00293
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Table 1. Summary of Acyl Transfer Reaction Experiments in Crystals
a
El (OC)···(OH) Nu distance O4−C15 (Å)/∠O4···C15O8 (deg) for all crystals except 9−11. O4···C8 (Å)/∠O4···C8O7 (deg) for 9 and
b
10; O6···C8 (Å)/O6···C8O7 (deg) in 11, see Figure 1. C−H···π interaction between R6 (or R5) and a C−H in the adjacent molecule, dist. (Å)/
angle (deg). Crystals in entries 1 and 12−15 had good C−H···π interactions and a well-defined reaction channel for the propagation of the
intermolecular acyl transfer reaction. Molecular assembly in other crystals did not result in the formation of well-defined reaction channels. For
c
details of noncovalent interactions in crystals of 8−11 see the SI, and 12, 14−17·19 see refs 8 and 17. Mixture of products, not separated. For
d
e
experimental details of solid-state reactivity of 12, 14−17·19 see refs 8 and 17. No reaction. El···Nu geometry between the equatorial ester CO
group and hydroxyl group of neighboring molecules (“nearest neighbors” in the crystal structure). In all other crystals El···Nu geometry is between
the axial ester CO group and hydroxyl group of neighboring molecules.
Since molecular packing in crystals of 11 does not support
propagation of the reaction in the crystal (as was observed in
crystals of the racemic dibenzoate 19),17 it appears the reaction
occurs in localized pockets in crystals leading to low conversion
of reactants to products. It is therefore reasonable to conclude
that the lower reactivity observed is a consequence of the absence
of a proper channel for the reaction to propagate through the
crystal (in spite of excellent El-Nu geometry between the
reacting groups in crystals of 11).
In solid-state reactions, two parameters can be used to judge
the efficacy of a reaction: (i) reactivity, the amount of the starting
material consumed (more the better), and (ii) selectivity, the
number of products formed (lesser the better). We have
attempted to correlate both these parameters (see below) with
the crystal structures of hydroxyl esters shown in Scheme 4. The
extent of acyl transfer reactivity (Scheme 1) in the solid state of
compounds shown in Scheme 2 and their previously reported
analogues (Scheme 4) is summarized in Table 1. These
compounds were chosen for the study to have a wide variety in
molecular structures as well as crystal (supramolecular)
structures. These crystalline compounds could be broadly
grouped into three categories of reactivity: (a) those that
exhibited good and selective intermolecular acyl transfer
reactivity; (b) those that exhibited intermolecular acyl transfer
reactivity but led to a mixture of products (i.e., nonselective); and
(c) those that did not exhibit considerable intermolecular acyl
transfer reactivity (i.e., unreactive). Correlation of the crystal
structure parameters with the experimentally observed reactivity
of the hydroxyl esters in their crystals and co-crystals is presented
below.
The variation in the reactivity of hydroxyl esters as a function
of the El···Nu geometry is shown in Figure 2. The reactivity is
D
DOI: 10.1021/acs.joc.8b00293
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
a
Scheme 4. Molecular Structure of Hydroxyl Esters Investigated for Intermolecular Acyl Transfer Reactivity in their Crystals
a
The reactivity of 4·5, 17 (Form II crystals), 18, 19, and 17·19, was good; reactivity of 8, 9, 11, and 14−17 (Form I crystals) was not good, while the
esters 10, 12, 1320 were unreactive. (Camph = camphanoyl).
geometry deviates from the optimum distance and angle
mentioned above (filled rectangles). The exception of 2-p-
toluate suggests that El···Nu geometry is not the sole factor that
determines the reactivity and selectivity of intermolecular acyl
group transfer in crystals of these hydroxyl esters.
The variation in the reactivity of the hydroxyl esters as a
function of the C−H···π interactions in their crystals (Figure S26,
SI) clearly suggested that crystals which exhibit good reactivity
also showed good C−H···π interactions (shorter and linear).
This is further supported by a plot of the C−H···π interactions vs
the yield of the expected products (Figure 3). This plot shows
that the specificity and yield of the acyl transfer reaction is good
in crystals that have good C−H···π interactions. These results
imply that C−H···π interactions contribute to both reactivity as
well as specificity of acyl transfer in crystals of hydroxyl esters.
Figure 3 reveals that reactivity of crystals of the 2-p-toluate 11 is
not as good as that of dibenzoates 18 or 19 due to the weaker C−
H···π interactions (Table 1). A similar effect is also seen by
comparison of the reactivity of dimorphs of the orthoacetate
Figure 2. Variation of the acyl transfer reactivity as a function of the El···
Nu geometry. Numbers refer to the compounds listed in Table 1.
dibenzoate 17.17d
A third factor which contributes to the acyl transfer reactivity
of crystals under investigation is the presence of reaction
channels which help in the propagation of the reaction
throughout the crystal in a domino fashion. Such reaction
channels are essential since the reaction is initiated by solid
sodium carbonate at the surface of crystals, but the acyl transfer
reaction must take place throughout the crystal to have high
conversion of reactant to products.17b
This is clearly established by a comparison of the molecular
packing in reactive crystals of dibenzoates 19 and the diol 11
(Figure 4).17 The p-toluate 11 exhibited a molecular assembly
where the reactive centers of unit translated molecules possessed
excellent geometry for an acyl transfer reaction. However, a
reaction channel for the propagation of the reaction through the
very good (>90% conversion of hydroxyl ester to expected
products by the selective intermolecular migration of the C6−O-
acyl group to the C4-hydroxyl group, see cartoon in Table 1),
when the El···Nu distance and the angle are close to 3 Å and 85−
90° respectively (filled rectangles, Figure 2).13b The only
exception to this is the 2-p-toluate 11 (circle with cross-wires
close to the El···Nu angle-axis of the graph) for which the yield of
the product (10) based on El···Nu interaction is much <50%
(maximum possible for a disproportionation reaction) even
though the El···Nu geometry is the best among all the
compounds tested. For all other compounds (circles with
cross-wires and triangles), the reactivity is low since the El···Nu
E
DOI: 10.1021/acs.joc.8b00293
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
and angle in the neighborhood of 3.2 Å and 90°, respectively; (b)
good C−H···π interaction which contributes to maintain the
topochemical control of the reaction; and (c) presence of
channels for the progress of the reaction in the crystal in a
domino fashion, exhibits good intermolecular acyl transfer
reactivity. In fact, we demonstrated the use of these parameters to
identify other crystals from the CSD and verified their reactivity
experimentally.8a Hence the results presented here show that the
crystal (supramolecular) structure−reactivity correlation based
on noncovalent interactions presented above is causative and not
merely statistical, since absence of any one structural criterion
(a−c) in the crystal leads to absence of reactivity.
CONCLUSIONS
■
Investigation of reactions in molecular crystals is a contemporary
area of research which has banked heavily on serendipity for the
discovery of crystals in which chemical reactions are facile. Solid-
state reactions have not been developed to a great extent as a tool
in synthetic organic chemistry (as compared to solution-state
reactions) barring the exception of photochemical reactions in
olefinic crystals, presumably due to the complexity involved in
obtaining a suitably preorganized reaction system in the
crystalline state.13b,21 Identification or recognition of reactive
molecular crystals is in fact the rate-determining step for the
progress in this area of research. Although crystal engineering
shows promise of being one of the ways of preparing a reactive
molecular crystal, availability of custom designed crystals is still
not a reality. The second-best alternative to designing crystals
with desired properties (in the present case, reactivity) is to be
able to identify reactive crystals from a library of crystal
structures. The study and correlation between the solid-state
reactivity and the crystal structures of O-acyl-myo-inositol 1,3,5-
Figure 3. Variation of yield of the triester (diester in case of 11 and 15)
in the solid-state reactions as a function of the C−H···π interactions in
the crystals. Numbers refer to the compounds listed in Table 1.
crystal was absent, and hence, the solid-state reaction proceeded
with less specificity as well as relatively lower yield of the
products. The presence of a well-defined reaction channel in a
crystal which lacks one or more of the other parameters discussed
above, also results in poorer acyl transfer reactivity, as seen in
Form I crystals of 17.17 It is clear from a comparison of the
reactivity and crystal structures of all the hydroxyl esters studied
thus far that a crystal which satisfies all the three conditions, viz.
(a) the relative geometry of the El and Nu groups with distance
Figure 4. (a) The helical organization of molecules in crystals of racemic 19 provides channels for the propagation of acyl transfer reaction through each
helix, resulting in clean reaction and good yield of the products. Sodium carbonate initiates the reaction at one end of each helix; the reaction progresses
(as indicated by arrows) along the helix due to successive intermolecular benzoyl group transfer and intramolecular proton transfer. Starred (*) and
unstarred molecules represent enantiomers. (b) Unit-translated molecules in crystals of 11 where channels suitable for the propagation of the acyl
transfer reaction are not available. These diagrams were generated using the packing diagrams obtained with the help of the program Mercury 3.8.
Hydrogen atoms other than those of the hydroxyl groups are omitted for clarity.
F
DOI: 10.1021/acs.joc.8b00293
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
241.5 °C; IR (Nujol, cm−1) υ 3468, 1716; 1H NMR (200 MHz, CDCl3)
orthoesters aided us in identifying key features in the crystal
lattice that would support a facile intermolecular acyl transfer
reaction. The results presented here demonstrate that analysis of
intermolecular interactions and prior assessment of the
molecular organization can be used as criteria for the
rationalization and prediction of solid-state reactivity. Hence,
we have been able to identify and predict the acyl transfer
reactivity in co-crystals of naphthalene derivatives8a from a survey
of the CSD, based on the results presented in this paper. It is
important to note that the molecular structure of compounds
used to arrive at the supramolecular structural criteria is very
different from the constituents of the reactive crystal identified by
survey of the CSD. The ability to identify molecular crystals
capable of undergoing chemical reactions provides a practical
alternative to designing reactive molecular crystals. The ability to
predict the facility of reactions in molecular crystals is also
important in view of the stability of crystals which play an
important role in pharmaceutical solids,12,22 synthetic inter-
mediates,23 and other supramolecular functional assemblies. This
work also illustrates that an understanding of the structure of
supramolecular assemblies is more relevant to explain observed
reactivities and/or predict reactivity of small molecules rather
than single molecular structures (as witnessed in ‘organic reactive
functional group’ approach). This is evident from the fact that,
although the single molecular structures of compounds in the
present study varied widely, solids with comparable reactivity
possess similar supramolecular features essential for facile solid-
state acyl transfer reactivity.
̅
δ 2.39 (1H, br s) 4.45−4.51 (1H, m), 4.55−4.65 (2H, m), 4.71−4.79
(1H, m), 5.60−5.66 (2H, m), 5.78−5.84 (1H, td, J = 4 and 1.7 Hz),
7.71−7.78 (2H, m), 7.80−7.88 (6H, m) ppm; 13C NMR (50.3 MHz,
CD3COCD3 + CD3SOCD3) δ 65.6, 67.5, 69.8, 69.9,70.4, 72.9, 102.2,
103.8, 130.3, 130.5, 132.2, 132.4, 138.9, 139.0, 165.6, 166.1 ppm;
Elemental analysis calcd for C21H16O8I2: C, 38.80; H, 2.48. Found: C,
39.03; H,2.40%.
Racemic 2,4(6)-Di-O-(p-toluoyl) myo-inositol 1,3,5-Orthofor-
mate (10) and 2-O-(p-Toluoyl)-myo-inositol 1,3,5-Orthoformate
(11). myo-Inositol orthoformate (0.950 g, 5 mmol) was acylated as
described in the general procedure A with p-toluoyl chloride (∼11
mmol) in dry pyridine (12 mL). The products were separated by silica
gel column chromatography to obtain racemic 10 (0.842 g, 39%) and 11
(0.630 g, 41%). Racemic 10 when crystallized from acetonitrile,
acetonitrile-DCM, methanol, ethyl acetate, chloroform, dioxane, or
nitromethane consistently yielded thin rectangular plates. Mp 227−229
°C; IR (Nujol, cm−1) υ 3473, 1720, 1716; 1H NMR (200 MHz, CDCl3)
̅
δ 2.41 (3H, s), 2.42 (3H, s) 2.59 (1H, br s), 4.45−4.53 (1H, m), 4.56−
4.64 (2H, m), 4.68−4.77 (1H, m), 5.60−5.68 (2H, m), 5.83 (1H, td, J =
4 and 1.6 Hz), 7.20−7.31 (4H, m), 7.87−7.97 (2H, m), 7.99−8.08 (2H,
m) ppm; 13C NMR (50.3 MHz, CD3COCD3) δ 21.7, 65.0, 67.9, 69.5,
70.0, 70.7, 73.0, 103.9, 128.0, 128.3, 130.1, 130.2, 130.6, 130.8, 145.1,
166.4 ppm; Elemental analysis calcd for C23H22O8: C, 64.78; H, 5.20.
Found: C, 64.99; H, 4.80%.
The diol 11 when crystallized from acetone, ethyl acetate,
chloroform, or dichloromethane yielded hexagonal plates. Mp 148−
1
150 °C; IR (Nujol, cm−1) υ 3550−3350, 1703; H NMR (200 MHz,
̅
CDCl3): δ 2.05 (2H, br s), 2.42 (3H, s), 4.34−4.43 (1H, m), 4.43−4.53
(2H, m), 4.60−4.72 (2H, m), 5.50−5.62 (2H, m), 7.21−7.32 (2H, m),
7.97−8.08 (2H, m) ppm; 13C NMR (50.3 MHz, CDCl3): δ 21.7, 63.6,
68.0, 68.5, 71.9, 102.4, 126.4, 129.2, 130.0, 144.5, 166.9 ppm; Elemental
analysis Calcd for C15H16O7: C, 58.44; H, 5.23. Found: C, 58.52; H,
5.20%.
EXPERIMENTAL SECTION
■
Racemic 2-O-Tosyl-4-O-benzoyl-myo-inositol 1,3,5-Orthofor-
mate (8). Tosyl chloride (0.382 g, 2.0 mmol) was added to a solution of
myo-inositol 1,3,5-orthoformate24 (0.380 g, 2.0 mmol) in dry pyridine (5
mL), and the mixture was stirred for 12 h at room temperature. Benzoyl
chloride (0.281 g, 2.0 mmol) was then added and stirring continued for
another 12 h. Pyridine was evaporated under reduced pressure. Usual
work up of the gum obtained in ethyl acetate followed by purification by
silica gel column chromatography (gradient elution with light petroleum
- ethyl acetate) gave the racemic 8 (0.584 g, 65%) as a colorless solid; it
2,4,6-Tri-O-(p-toluoyl)-myo-inositol 1,3,5-Orthoformate (20).
myo-Inositol orthoformate (0.190 g, 1 mmol) was acylated as described
in the general procedure A with p-toluoyl chloride (∼5 mmol) in dry
pyridine (5 mL). The tri-p-toluate 20 (0.140 g, 26%) was isolated by
column chromatography (silica gel). Mp 156−158 °C; IR (Nujol, cm−1)
υ 1738, 1721; 1H NMR (200 MHz, CDCl3) δ 2.36 (6H, s), 2.44 (3H, s),
̅
4.62−4.71 (2H, m), 4.94−5.03 (1H, m), 5.66−5.71 (1H, m),5.74 (1H,
d, J 1.3 Hz), 5.78−5.86 (2H, m), 6.97 (4H, d, J = 8 Hz), 7.29−7.33 (2H,
m), 7.66−7.77 (4H, m), 8.01−8.11 (2H, m) ppm; 13C NMR (50.3
MHz, CDCl3) δ 21.6, 63.6, 66.9, 68.2, 69.4, 103.3, 125.7, 126.5, 128.9,
129.2, 129.8, 129.9, 144.1, 144.4,165.1, 166.2 ppm; Elemental analysis
Calcd for C31H28O9: C, 68.38; H, 5.18. Found: C, 68.07; H, 5.26%. The
di-p-toluate 10 and p-toluic acid were also isolated as minor products.
Transesterification of Racemic 2-O-Tosyl-6-O-benzoyl-myo-
inositol 1,3,5-Orthoformate (8) in its Crystals. The racemic
benzoate 8 (0.200 g, 0.44 mmol) and Na2CO3 (0.379 g, 3.58 mmol,
previously activated at 270 °C for 12 h) were ground together, and the
mixture heated at 100 °C for 192 h. The reaction mixture turned black
on heating. Analysis of this solid indicated the presence of 2-O-tosyl-4,6-
di-O-benzoyl-myo-inositol 1,3,5-orthoformate, 2-O-tosyl-myo-inositol
1,3,5-orthoformate,19 and the starting tosylate 8. No attempt was
made to purify these products since the yield was low.
was crystallized from chloroform-light petroleum mixture. Mp 140−142
1
°C; IR (CHCl3) υ 1726, 3210−3472 cm−1; H NMR (200 MHz,
̅
CDCl3) δ 2.36 (s, 3H), 2.77−2.80 (d, J = 5.3 Hz, 1H, D2O
exchangeable), 4.12−4.25 (m, 1H), 4.33−4.45 (m, 1H), 4.45−4.55
(m, 1H), 4.60−4.75 (m, 1H), 5.03−5.13 (m, 1H), 5.56 (s, 1H), 5.60−
5.71 (m, 1H), 7.15−7.33 (m, 2H), 7.42−7.57 (m, 2H), 7.58−7.71 (m,
1H), 7.77−7.82 (d, J = 8.3 Hz, 2H), 7.92−7.98 (d, J = 8.3 Hz, 2H); 13C
NMR (CDCl3, 125 MHz) δ 21.6, 67.3, 68.3, 69.4, 69.5, 72.1, 102.7,
127.8, 128.7, 129.8, 130.0, 133.8, 145.4, 164.7. Elemental analysis calcd
for C21H20O9S; C 56.25%, H 4.50%, found C 55.85%; H 4.56%.
General Procedure A: Acylation. Freshly distilled acid chloride
was added to an ice-cooled solution of myo-inositol 1,3,5-orthofor-
mate24 in dry pyridine, with constant stirring. The reaction mixture was
brought to room temperature, stirred for 18−20 h, and quenched with
ice. Pyridine was removed under reduced pressure by co-evaporation
with toluene (3 × 10 mL), and the residue was diluted with ethyl acetate
and washed successively with water, 2% aqueous HCl, water, saturated
sodium bicarbonate solution, and water followed by brine. The organic
layer was dried over anhydrous sodium sulfate, and the solvent was
evaporated under reduced pressure. The residue was flash chromato-
graphed on silica using light petroleum-ethyl acetate (gradient elution).
Racemic 2,4(6)-Di-O-(p-iodobenzoyl)-myo-inositol 1,3,5-Or-
thoformate (9). myo-Inositol orthoformate (0.190 g, 1 mmol) was
acylated as described in the general procedure A with 4-iodobenzoyl
chloride (∼2 mmol) in dry pyridine (7 mL). The mixture of products
obtained after workup was chromatographed to isolate racemic 9 (0.150
g, 23%). Crystallization of 9 from acetone, chloroform, nitromethane, 2-
propanol or ethyl acetate yielded thin plate like crystals. Mp 240.3−
Transesterification of Racemic 2,4(6)-Di-O-(p-iodobenzoyl)-
myo-inositol 1,3,5-Orthoformate (9) in its Crystals. Crystals of 9
(0.0195 g, 0.03 mmol) and activated sodium carbonate (0.0254 g, 0.24
mmol) were ground together into a powder which was heated at 130 °C
for 72 h. TLC analysis indicated a mixture of products, including 9 which
were not separated.
Transesterification of Racemic 2,4(6)-Di-O-(p-methylbenzo-
yl)-myo-inositol 1,3,5-Orthoformate (10) in its Crystals. Crystals
of 10 (0.149 g, 0.35 mmol) and activated sodium carbonate (0.297 g, 2.8
mmol) were ground together into a powder which was initially heated at
110 °C for 17 h. TLC indicated the presence of only 10. The
temperature was increased up to 130 °C, and heating was continued for
47 h, at the end of which 10 was recovered quantitatively.
G
DOI: 10.1021/acs.joc.8b00293
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Transesterification of 2-O-(p-Methylbenzoyl)-myo-inositol
1,3,5-Orthoformate (11) in its Crystals. Crystals of 11 (0.047 g,
0.152 mmol, mp. 148−150 °C) were heated with Na2CO3 (0.150 g, 1.4
mmol) at 126 °C for 20 h; TLC of the reaction mixture showed the
presence of the diester 10 and the triester 20. The reaction mixture was
loaded on a short silica gel (100−200 mesh) column and eluted with
ethyl acetate (5 × 10 mL) and methanol (2 × 5 mL). The filtrate was
concentrated, and the residue was chromatographed (230−400 mesh
silica gel) by gradient elution with ethyl acetate−light petroleum mixture
to obtain the triol 21 (0.012 g, 39%), the diester 10 (0.020 g, 29%), and
the triester 20 (0.007 g, 8%). Other products present in trace amounts
were not isolated.
(7) (a) Vyas, K.; Mohan Rao, V.; Manohar, H. Acta Crystallogr., Sect. C:
Cryst. Struct. Commun. 1987, 43, 1197−1200. (b) Vyas, K.; Manohar, H.;
Venkatesan, K. J. Phys. Chem. 1990, 94, 6069−6073. (c) Vyas, K.;
Manohar, H. Mol. Cryst. Liq. Cryst. 1986, 137, 37−43.
(8) (a) Tamboli, M. I.; Krishnaswamy, S.; Gonnade, R. G.; Shashidhar,
M. S. Chem. - Eur. J. 2013, 19, 12867−12874. (b) Tamboli, M. I.;
Shashidhar, M. S.; Gonnade, R. G.; Krishnaswamy, S. Chem. - Eur. J.
2015, 21, 13676−13682 and references cited therein..
(9) (a) Ramamurthy, V.; Gupta, S. Chem. Soc. Rev. 2015, 44, 119−135.
(b) Friscic, T.; MacGillivray, L. R. In Making Crystals by Design -
̌ ̌ ́
Methods, Techniques and Applications; Braga, D., Grepioni, F., Eds.;
Wiley-VCH: Weinheim, Germany, 2007; Chapter 2.3. (c) Huang, S.-L.;
Hor, T. S. A.; Jin, G.-X. Coord. Chem. Rev. 2017, 346, 112−122.
(d) Garcia-Garibay, M. A. Acc. Chem. Res. 2003, 36, 491−498.
(e) Suzuki, M.; Fujii, T.; Naito, Y.; Yamoto, K.; Matsuoka, S.-I.;
Takagi, K.; Sugiyama, H.; Uekusa, H. Bull. Chem. Soc. Jpn. 2018, 91,
343−348.
(10) (a) Blow, D. Nature 1990, 343, 694−695. (b) Brannigan, J. A.;
Dodson, D.; Duggleby, H. J.; Moody, P. C. E.; Smith, J. L.; Tomchick, D.
R.; Murzin, A. G. Nature 1995, 378, 416−419.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.joc.8b00293.
■
S
Characterization data and 1H and 13C NMR spectra for all
new products; single crystal X-ray diffraction data for 8−
11, ORTEPs and packing diagrams for 8−11 (PDF)
Crystallographic data (CIF)
(11) Yang, C.; Xia, W. Chem. - Asian J. 2009, 4, 1774−1784.
(12) (a) Troup, A. E.; Mitchner, H. J. Pharm. Sci. 1964, 53, 375−379.
(b) Jacobs, A. L.; Dilatush, A. E.; Weinstein, S.; Windheuser, J. J. J.
Pharm. Sci. 1966, 55, 893−895. (c) Koshy, K. T.; Troup, A. E.; Duvall, R.
N.; Conwell, R. N.; Shankle, L. L. J. Pharm. Sci. 1967, 56, 1117−1121.
(d) Galante, R. N.; Visalli, A. J.; Patel, D. M. J. Pharm. Sci. 1979, 68,
1494−1498.
AUTHOR INFORMATION
Corresponding Author
*E-mail: ms.shashidhar@ncl.res.in. Fax: 91-20-25902629. Tel:
91-20-25902055.
■
(13) (a) Biradha, K.; Santra, R. Chem. Soc. Rev. 2013, 42, 950−967.
(b) Burgi, H. B.; Dunitz, J. D. Acc. Chem. Res. 1983, 16, 153−161.
̈
ORCID
(14) Coley, C. W.; Barzilay, R.; Jaakkola, T. S.; Green, W. H.; Jensen, K.
F. ACS Cent. Sci. 2017, 3, 434−443.
Shobhana Krishnaswamy: 0000-0003-2479-2826
Mysore S. Shashidhar: 0000-0002-8477-1149
(15) Devaraj, S.; Shashidhar, M. S.; Dixit, S. S. Tetrahedron 2005, 61,
529−536.
Present Address
(16) Krishnaswamy, S.; Gonnade, R. G.; Shashidhar, M. S.; Bhadbhade,
M. M. CrystEngComm 2010, 12, 4184−4197.
†Department of Chemistry, Indian Institute of Technology,
Madras, Chennai 600036
(17) (a) Praveen, T.; Samanta, U.; Das, T.; Shashidhar, M. S.;
Chakrabarti, P. J. Am. Chem. Soc. 1998, 120, 3842−3845. (b) Sarmah, M.
P.; Gonnade, R. G.; Shashidhar, M. S.; Bhadbhade, M. M. Chem. - Eur. J.
2005, 11, 2103−2110. (c) Murali, C.; Shashidhar, M. S.; Gonnade, R.
G.; Bhadbhade, M. M. Eur. J. Org. Chem. 2007, 2007, 1153−1159.
(d) Murali, C.; Shashidhar, M. S.; Gonnade, R. G.; Bhadbhade, M. M.
Chem. - Eur. J. 2009, 15, 261−269.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
Authors thank Manash P. Sarmah, Rajesh G. Gonnade, and
Mohan M. Bhadbhade for discussion and technical help. S.K.
thanks CSIR, New Delhi, for a fellowship.
(18) Nishio, M.; Umezawa, Y.; Honda, K.; Tsuboyama, S.; Suezawa, H.
CrystEngComm 2009, 11, 1757−1788.
(19) Sureshan, K. M.; Shashidhar, M. S.; Praveen, T.; Gonnade, R. G.;
Bhadbhade, M. M. Carbohydr. Res. 2002, 337, 2399.
REFERENCES
■
(20) Riley, A. M.; Mahon, M. F.; Potter, B. V. L Angew. Chem., Int. Ed.
(1) Desiraju, G. R. Angew. Chem., Int. Ed. 2007, 46, 8342−8356.
(2) (a) Li, P.; He, Y.; Guang, J.; Weng, L.; Zhao, J. C.-G.; Xiang, S.;
Chen, B. J. Am. Chem. Soc. 2014, 136, 547−549. (b) Black, H. T.;
Perepichka, D. F. Angew. Chem. 2014, 126, 2170−2174.
Engl. 1997, 36, 1472−1474.
(21) (a) MacGillivray, L. R. J. Org. Chem. 2008, 73, 3311−3317.
̌ ̌ ́
(b) MacGillivray, L. R.; Papaefstathiou, G. S.; Friscic, T.; Hamilton, T.
D.; Bucar, D.; Chu, Q.; Varshney, D. B.; Georgiev, I. G. Acc. Chem. Res.
2008, 41, 280−291. (c) Natarajan, A.; Bhogala, B. R. In Bimolecular
Photoreactions in the Crystalline State, in Supramolecular Photochemistry:
Controlling photochemical processes, Ramamurthy, V., Inoue, Y., Eds.;
John Wiley & Sons: Hoboken, NJ, 2011; pp 175−228.
(22) (a) Cheng, Y.-D.; Lin, S.-Y. J. Agric. Food Chem. 2000, 48, 631−
635. (b) Wang, S.-L.; Lin, S.-Y.; Chen, T.-F. Chem. Pharm. Bull. 2001,
49, 402−406. (c) Hsu, C.-H.; Lin, S.-Y. Thermochim. Acta 2009, 486, 5−
10.
(3) (a) Priimagi, A.; Cavallo, G.; Metrangolo, P.; Resnati, G. Acc. Chem.
Res. 2013, 46, 2686−2695. (b) Gilday, L. C.; Robinson, S. W.; Barendt,
T. A.; Langton, M. J.; Mullaney, B. R.; Beer, P. D. Chem. Rev. 2015, 115,
7118−7195. (c) Li, B.; Zang, S.-Q.; Wang, L.-Y.; Mak, T. C. W. Coord.
Chem. Rev. 2016, 308, 1−21.
(4) Nishio, M.; Umezawa, Y.; Suezawa, H.; Tsuboyama, S. In The
Importance of π-Interactions in Crystal Engineering: Frontiers in Crystal
Engineering; Tiekink, E. R. T., Zukerman-Schpector, J., Eds.; Wiley-
VCH: Chichester, UK, 2012; pp 1−39.
(23) Sardessai, R.; Krishnaswamy, S.; Shashidhar, M. S. CrystEngComm
2012, 14, 8010−8016.
(24) Uhlmann, P.; Vasella, A. Helv. Chim. Acta 1992, 75, 1979−1994.
(5) Mann, L.; Voßnacker, P.; Muller, C.; Riedel, S. Chem. - Eur. J. 2017,
̈
23, 244−249.
(6) (a) Chang, Y.; Chen, Y.; Chen, C.; Wen, Y.; Lin, J. T.; Chen, H.;
Kuo, M.; Chao, I. J. Org. Chem. 2008, 73, 4608−4614. (b) Burattini, S.;
Greenland, B. W.; Merino, D. H.; Weng, W.; Seppala, J.; Colquhoun, H.
M.; Hayes, W.; Mackay, M. E.; Hamley, I. W.; Rowan, S. J. J. Am. Chem.
Soc. 2010, 132, 12051−12058. (c) Bishop, R. In The Importance of π-
Interactions in Crystal Engineering: Frontiers in Crystal Engineering;
Tiekink, E. R. T., Zukerman-Schpector, J., Eds.; Wiley-VCH: Chi-
chester, UK, 2012; pp 41−77.
H
DOI: 10.1021/acs.joc.8b00293
J. Org. Chem. XXXX, XXX, XXX−XXX