Nanoindentation as a probe for mechanically-induced molecular migration in layered organic donor-acceptor complexes

Article abstract of DOI:10.1002/asia.201200224

Nanoindentation and scratch experiments on 1:1 donor-acceptor complexes, 1 and 2, of 1,2,4,5-tetracyanobenzene with pyrene and phenanthrene, respectively, reveal long-range molecular layer gliding and large interaction anisotropy. Due to the layered arrangements in these crystals, these experiments that apply stress in particular directions result in the breaking of interlayer interactions, thus allowing molecular sheets to glide over one another with ease. Complex 1 has a layered crystal packing wherein the layers are 68° skew under the (002) face and the interlayer space is stabilized by van der Waals interactions. Upon indenting this surface with a Berkovich tip, pile-up of material was observed on just one side of the indenter due to the close angular alignment of the layers with the half angle of the indenter tip (65.35°). The interfacial differences in the elastic modulus (21 %) and hardness (16 %) demonstrate the anisotropic nature of crystal packing. In 2, the molecular stacks are arranged in a staggered manner; there is no layer arrangement, and the interlayer stabilization involves C-H...N hydrogen bonds and π...π interactions. This results in a higher modulus (20 %) for (020) as compared to (001), although the anisotropy in hardness is minimal (4 %). The anisotropy within a face was analyzed using AFM image scans and the coefficient of friction of four orthogonal nanoscratches on the cleavage planes of 1 and 2. A higher friction coefficient was obtained for 2 as compared to 1 even in the cleavage direction due to the presence of hydrogen bonds in the interlayer region making the tip movement more hindered. Copyright

Full text of DOI:10.1002/asia.201200224

DOI: 10.1002/asia.201200224
Nanoindentation as a Probe for Mechanically-Induced Molecular Migration
in Layered Organic Donor–Acceptor Complexes
Sunil Varughese,[a] Mangalampalli S. R. N. Kiran,[b] Upadrasta Ramamurty,*[b] and
Gautam R. Desiraju*[a]
Abstract: Nanoindentation and scratch
experiments on 1:1 donor–acceptor
complexes, 1 and 2, of 1,2,4,5-tetracya-
nobenzene with pyrene and phenan-
threne, respectively, reveal long-range
molecular layer gliding and large inter-
action anisotropy. Due to the layered
arrangements in these crystals, these
experiments that apply stress in partic-
ular directions result in the breaking of
interlayer interactions, thus allowing
molecular sheets to glide over one an-
other with ease. Complex 1 has a lay-
ered crystal packing wherein the layers
are 688 skew under the (002) face and
the interlayer space is stabilized by van
der Waals interactions. Upon indenting
this surface with a Berkovich tip, pile-
up of material was observed on just
one side of the indenter due to the
close angular alignment of the layers
with the half angle of the indenter tip
(65.358). The interfacial differences in
the elastic modulus (21%) and hard-
ness (16%) demonstrate the anisotrop-
ic nature of crystal packing. In 2, the
molecular stacks are arranged in a stag-
gered manner; there is no layer ar-
rangement, and the interlayer stabiliza-
À
tion involves C H···N hydrogen bonds
and p···p interactions. This results in
a higher modulus (20%) for (020) as
compared to (001), although the aniso-
tropy in hardness is minimal (4%). The
anisotropy within a face was analyzed
using AFM image scans and the coeffi-
cient of friction of four orthogonal
nanoscratches on the cleavage planes
of 1 and 2. A higher friction coefficient
was obtained for 2 as compared to
1 even in the cleavage direction due to
the presence of hydrogen bonds in the
interlayer region making the tip move-
ment more hindered.
Keywords: charge-transfer
complexes · layered compounds ·
mechanical anisotropy · molecular
migration · nanoindentation
Introduction
small-range molecular rearrangements to long-range diffu-
sion, as seen in the formation of molecular complexes, such
as the one formed between picric acid and naphthalene.[5]
Such migrations can also be artificially induced by applying
mechanical stress using indentation or nanoscratching.[6]
While techniques such as X-ray diffraction and spectroscopy
generally fail to perceive the resulting anisotropic migra-
tions, atomic force microscopy (AFM)[7] and grazing inci-
dence diffraction (GID)[8] can detect such effects unambigu-
ously. Nanoindentation is an effective technique for the pre-
cise mechanical characterization of small-volume materials
such as thin films and single crystals.[9] Whereas its utility in
the characterization of inorganic and engineering materials
is ubiquitous, only a limited number of studies have been re-
ported for molecular crystals.[10,11] These include our recent
report on the mechanical anisotropy in saccharin[12] and
a study on interaction anisotropy and shear instability of as-
pirin polymorphs.[13]
Molecular migration within crystalline solids is of topical
concern and it underlies phenomena such as solid-state reac-
tivity, cocrystal formation, and defect transport.[1] Although
classical topochemistry envisages minimum atomic or mo-
lecular mobility in crystalline solids,[2] Kaupp has suggested
that long-range molecular migrations are the reason for the
nearly 100% yields achieved in many solid-state reactions
that are supposedly (or nominally) governed by topochemis-
try.[3] Solid-state chemical reactions and the consequent
changes in molecular geometry induce molecular move-
ments that release the internal pressure created during the
initial reaction.[4] These molecular transits extend from
[a] Dr. S. Varughese, Prof. G. R. Desiraju
Solid State and Structural Chemistry Unit
Indian Institute of Science
In layered structures, strong interactions constitute the
layers while weak interactions in the interlayer region bind
the layers together in a stack so that the packing is distinctly
anisotropic. Such materials range from structures with
atomic thickness, like in graphite, to macrostructures such as
lamellar clays. Layered compounds are of interest due to
their utility as intercalates, lubricants, and conducting mate-
rials.[14] Mechanical shearing and bending of their crystals is
an important consequence of the packing anisotropy and
Bangalore 560 012 (India)
Fax : (+91)080-23602306
E-mail: desiraju@sscu.iisc.ernet.in
[b] Dr. M. S. R. N. Kiran, Prof. U. Ramamurty
Department of Materials Engineering
Indian Institute of Science
Bangalore 560 012 (India)
E-mail: ramu@materials.iisc.ernet.in
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/asia.201200224.
Chem. Asian J. 2012, 00, 0 – 0
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
&
&
&
These are not the final page numbers! ÞÞ

FULL PAPERS
the nonspecific nature of interactions in the interlayer
region.[15] These anisotropic effects can also lead to poly-
morphic forms with diverse mechanical properties, enabling
their mechanical sorting.[16] Pyrazine-2-carboxamide, venla-
faxine hydrochloride, and 6-chloro-2,4-dinitroaniline are lay-
ered compounds with distinct mechanical properties among
the polymorphs.[17] Further to mechanical evaluation, the di-
rectional nature of thermal expansion and high-pressure
studies can also provide information regarding the anisotro-
py in molecular crystals.[18]
Results and Discussion
Pyrene Complex
The 1:1 TCNB-pyrene complex (1) crystallizes in the mono-
clinic system and the molecules make a layered arrange-
ment;[20] (100) and (002) constitute the significant faces with
(002) being the major one (Figure 1). While the mean plane
of the molecular layers makes an angle of 258 with (100),
Charge-transfer complexes (CT complexes) or electron
donor–acceptor complexes often have layered structures in
which planar aromatic donors and acceptors are involved.
Molecules are stabilized through electrostatic attractions
wherein fractional charge is transferred between the molec-
ular entities.[19] This attraction, generally created by an elec-
tronic transition to an excited electronic state, is much
weaker than covalent bonding. Since the excitation energy
of this resonance occurs frequently in the visible region of
the electromagnetic spectrum, it usually imparts an intense
color to the complexes. 1,2,4,5-Tetracyanobenzene (TCNB)
is well known for its ability to make CT complexes with aro-
matic donors. In this study, 1:1 complexes of TCNB with
pyrene (1) and phenanthrene (2) were used to evaluate the
mechanical properties of anisotropic layered solids. Indenta-
tion experiments were carried out using a sharp-tipped Ber-
kovich indenter to characterize a range of mechanical prop-
erties including elastic modulus, hardness, fracture behavior,
and molecular migration under mechanical stress.
Abstract in Sanskrit:
Figure 1. Crystal packing of 1. a) Layer arrangement. b) Interactions in
an individual layer.
they are 688 skew under (002). The interplanar spacings, d,
are 7.138 and 7.747 ꢁ for the (100) and (002) orientations,
respectively. Within an individual layer, TCNB and pyrene
À
interact through several C H···N hydrogen bonds with an
average H···N distance of 2.72 ꢁ. The average interlayer
(p···p) separation is 3.45 ꢁ. Significant interaction anisotro-
py manifests itself as differences in the attachment energies,
Eatt, for the (100) and (002) faces as À33.699 and
À23.947 kJmolÀ1, respectively.
Both the load–penetration depth curves (P–h curves) and
the H and E values obtained for the (100) and (002) faces of
1 indicate significant mechanical anisotropy in the crystals
examined. The P–h curves show large residual depths upon
unloading, which indicates that the crystal undergoes signifi-
cant plastic deformation upon indentation (Figure 2). Inter-
estingly, both curves show similar penetration depths; how-
ever, the loading part of the curves is distinct in nature:
while the loading part of the P–h curve obtained on the
(002) face is smooth, several distinct displacement bursts are
observed on (100). The average values of H and E are
0.120Æ0.001 and 3.92Æ0.054 GPa for (002), and 0.143Æ
&
&
&2
&
www.chemasianj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 0000, 00, 0 – 0
ÝÝ These are not the final page numbers!

Molecular Migration in Layered Organic Donor–Acceptor Complexes
layers, they get compressed in the other orientations, thus
resulting in pile-up in one orientation only.
The discrete displacement bursts in the P–h curve ob-
tained on (100) can be correlated with the crystallographic
features. As discussed by us earlier, plastic deformation in
layered organic crystals can be attributed to slipping along
a crystallographic plane when sheets of molecules glide
across one another, similar to a stack of playing cards in
a deck.[16b] Generally, the primary slip plane in organic mate-
rials is assumed to be the weakest bound plane in that the
attachment energy for this plane is the least. Accordingly,
slipping is preferred along this plane. By this token, the
pop-ins observed in the (100) orientation are surprising be-
cause it corresponds to the plane with the largest attach-
ment energy. However, the presence of molecular layers
with weakly bound interlayer spaces normal to the indenta-
tion direction makes such a slip possible. In these cases the
slip occurs for the planes that are the most widely spaced
with the smallest Eatt and in the directions that are closest
packed so that the slip vector is the smallest. Hence, the
possible slip system in 1 could be {100}<002>. The absence
of any directional interactions along <002> makes gliding
of the planes without much shear resistance viable. This ob-
servation agrees with the proposal by Olusanmi et al.[22]
that, for aspirin, low attachment energy alone is not a suffi-
ciently reliable indication of a cleavage plane. The absence
of material pile-up in the case of indentation on the (100)
face can be due to the effective compressible deformation
nature of the p···p interactions as the indentation is per-
formed along the interaction directions.
Figure 2. Representative P–h curves obtained on two faces of a crystal of
1.
0.01 and 4.98Æ0.249 GPa for (100), respectively. The error
bars correspond to the standard deviations obtained for 15
measurements made on each crystal face. The large differ-
ence (21%) in the moduli for the crystal faces suggests sig-
nificant dissimilarities in the interaction characteristics. Fur-
ther, the large degree of anisotropy in H (16%) suggests
possible differences in the micromechanisms of plasticity.
On the P–h curve of (100) of 1, the first significant displace-
ment burst (pop-in) was consistently observed to occur at
1.05Æ0.1 mN with a magnitude, hpopÀin, of about 21 nm. In-
creasing the load results in several pop-ins with hpopÀin of 7
and 13 nm, with the largest one being 21 nm. It is clear that
hpopÀin is in multiples of 7 nm which in turn is an integral
multiple of d(100) (7.138 ꢁ). In an earlier study on crystalline
saccharin, we have also noted such a correspondence be-
tween the pop-in magnitude and the underlying crystal
length scale.[12]
The AFM images of the in-
dents on (002) show a significant
pile-up of material along one of
the faces of the indenter (Fig-
ure 3a), whereas no pile-up is
observed in the case of indenta-
tion on the (100) face (Fig-
ure 3b). The quantity and
shape of the profiles of the pile-
up strongly depends on the
layer orientation and the tip ge-
ometry. The indenter tip used
in the present study has a total
included angle of 142.38 with
a half angle of 65.358. The mo-
lecular layers at an angle of 688
with the (002) face and the in-
denter half angle are aligned
closely (Figure 3c and 3d),
thereby enabling the layers to
slide over the edge of the in-
denter tip. This is analogous to
the sliding of the tectonic plates
Figure 3. 3D representation of AFM images of the residual indent impressions on a) (002) and b) (100) of crys-
tals of 1. c) Representation of the angular alignment of the layers with the half angle of the Berkovich tip. The
grey curve is the line profile drawn at the middle of the residual indent impression of (002). d) Schematic rep-
during convergent continental
collision.[21] However, due to
the slant arrangement of the resentation of the layer sliding along the side of the indenter tip.
Chem. Asian J. 2012, 00, 0 – 0
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
3
&
&
&
These are not the final page numbers! ÞÞ

Upadrasta Ramamurty, Gautam R. Desiraju et al.
FULL PAPERS
À
These observations provide a possible explanation for
pop-ins only on the (100) face. On indenting along [100],
the layers are compressed due to the presence of weak van
der Waals interactions until they come close enough to acti-
vate a repulsion. This permits the broken/stretched layers to
rearrange themselves so that some of the stored elastic
strain energy is released, which in turn results in the gliding
of the layers, the energetically most viable option available
to the system. All this affects the sudden penetration of the
indenter tip into the sample and is observed as pop-ins on
the P–h curve. These pop-ins therefore reveal a release of
internal pressure. However, along [002], the tip penetrates
the crystal perpendicular to the stacking direction, thus ena-
bling the molecular layers, which are stabilized through
weak non-directional p···p interactions, to slide past one an-
other with ease; this appears as a smooth P–h curve. To put
these results in perspective, it may be stated that the ab-
sence of pop-ins in the P–h curve corresponding to (002) is
due to the availability of more slip planes.
structure is not observed. The C H···N hydrogen bonds are
inclined to the layer and contribute along with the p···p in-
teractions to the stabilization of the stacking arrangement of
the molecules. The d-spacing for (001) and (020) orienta-
tions is 7.157 and 6.517 ꢁ, respectively, with corresponding
attachment energies of À34.828 and À23.705 kJmolÀ1.
The observed dissimilarity in the crystal packing in 2, with
respect to 1, is reflected in the mechanical behavior as well.
The maximum penetration depth at a peak load of 6 mN for
2 is less than in 1 (Figure 5). The P–h curve obtained on
Phenanthrene Complex
Figure 5. Representative P–h curves obtained on (020) and (001) of crys-
tals of 2.
Unlike the pyrene complex, the TCNB complex with phen-
anthrene (2) is composed of trimer units (one TCNB and
two phenanthrene molecules) stacked in
a staggered
manner.[23] Crystals of 2 are mainly constituted of two sets
of faces: (001) and (020), of which the latter is the major
face. The molecules stack down <001> and adjacent stacks
(020) exhibits some serrations while pop-ins are observed on
(001). The crystal undergoes substantial plastic deformation
as evident from the large residual depth upon unloading.
The average H and E values obtained on (020) are 0.140Æ
0.001 and 5.76Æ0.115 GPa, while that for (001) are 0.146Æ
0.001 and 4.59Æ0.081 GPa, respectively. Thus, (020) is stiffer
by 20% while (001) is harder by 4%. Hence, a significant
anisotropy in E exists although
À
are linked with C H···N interactions (Figure 4). While mo-
lecular stacks make an angle of 218 with (001), they are per-
pendicular to the (020) face. However, due to the staggered
arrangement of the molecules, the formation of a layered
H is comparable. The latter can
be due to the effect of the stag-
gered arrangement of the layers
and the slanted orientation of
À
C H···N hydrogen bonds, thus
allowing the interactions to
make contributions in both
faces. The first significant dis-
placement burst in the P--h
curve obtained on (001) is
13 nm in magnitude and the
load at which it occurs is consis-
tent at 0.7Æ0.05 mN. In addi-
tion, a few small pop-ins with
a magnitude of 7 nm are seen
on the loading part of the P–h
curve and it should be noted
that the pop-in magnitude is
again an integral multiple of
d(001) (7.157 ꢁ).
The AFM images of the
indent impressions on (020)
show possible slip steps on the
surface of the crystal (Fig-
Figure 4. Crystal structure of 2. a) Molecular arrangement with respect to the (020) and (001) plane. b) The
trimer arrangement of the complex. c) The molecular arrangement down [001]. d) The molecular arrangement
down [020].
&
&
&4
&
www.chemasianj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 0000, 00, 0 – 0
ÝÝ These are not the final page numbers!

Molecular Migration in Layered Organic Donor–Acceptor Complexes
Figure 7. a) AFM images of a nanoscratch on (002) of crystals of 2.
b) Schematic representation of the tip movement during the nanoscratch
experiments. c) The coefficient of friction at 08 and 908.
Figure 6. AFM image of the residual indent obtained on a) (020) of a crys-
tal of 2. b) 3D representation. The arrow shows the presence of slip lines
in various directions.
tion that is more evident towards the right hand side (Fig-
ure 7a). A significant molecular pile-up was noted on both
the front and on either side of the scratch when the tip tra-
verses at 08 and 1808. At 2708 the molecules migrate only to
the left hand side (scratch images of 1808 and 2708 are
shown in Figure S1 in the Supporting Information). At 08
the tip moves along the skew direction of the layers and
thus the layers are open for the release of material on either
side of the scratch; at 1808 the tip traverses against the skew
direction and thus experiences the maximum friction, and
moreover makes the largest material pile-up on both sides
of the scratch as well as in the front (Figure 7b). This postu-
lation is further substantiated by comparing the coefficient
ure 6a), possibly due to the broken layer arrangement of
the molecular units and its deformation under the compres-
sive stress brought about by the pyramidal geometry of the
indenter tip. This deformation of layers is analogous to the
distortion generated upon inserting a pencil point between
the pages of a book. Furthermore, it is evident from the
scan images that the indentation results in ꢂsinking inꢃ of the
material (Figure 6b). By contrast, indenting on (100) results
in fracture; this is not surprising due to its greater hardness.
of friction values; 08 and 1808 experience the lowest (mavg
=
0.4) and highest friction (mavg =0.5), respectively (Figure 7c).
In the 908 and 2708 settings, the tip movement is relatively
featureless, even without much variation in the coefficient
of friction, and such a movement is due to the weak nature
of the interlayer interactions making the slip rather smooth.
Such anisotropy can be observed in organic molecular crys-
tals with cleavage planes; good examples of molecular mi-
gration that takes place under mechanical stress are provid-
ed by thiohydantoin and anthracene.[3a] Because of the flat
orientation of the molecular layers with respect to (100),
scratch experiments on this face yielded only abrasions with
small material pile-up at the end of the scratch, irrespective
of the scratch direction. Such behavior is due to the restrict-
ed molecular migration within molecular layers upon
scratching, and the pile-up could be due to the shifting of
some fragments of the anisotropic monolayers in front of
the tip. The migration distance (the distance affected from
the middle of the scratch to the side) due to the indenter
movement at 908 and 2708 is 3.8 mm and the distance be-
Nanoscratch Experiments
Nanoscratching is an effective probe for the molecular pack-
ing in various faces in different orientations and can further
provide insights into mechanically induced molecular migra-
tions that occur without chemical transformations.
In 1 the layered structure is such that intra- and interlayer
interactions are quite distinct. This anisotropy means that
the crystal can be cleaved easily along the layers. The layers
are 688 skew under (002). Such an arrangement can yield
distinct molecular migration events upon scratching at vari-
ous orientations. In our experiments, four orthogonal direc-
tions yielded distinct scratch profiles and coefficient of fric-
tion, m (ratio of lateral force to normal force) values, and
the results may be correlated with the crystal packing.
Scratching along the long crystal edge (08) results in molecu-
lar migration on both sides together with a small pile-up of
material at the end of the scratch. At 908, molecules migrate
only to the right hand side; AFM images show layer migra-
Chem. Asian J. 2012, 00, 0 – 0
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
5
&
&
&
These are not the final page numbers! ÞÞ

Upadrasta Ramamurty, Gautam R. Desiraju et al.
FULL PAPERS
tween slip lines is estimated as 700 nm, which is an integral
multiple of d(100) as shown in Figure S2 in the Supporting In-
formation.
ure S3 in the Supporting Information). The friction coeffi-
cient for 908 and 2708 indicates the presence of several
uneven events that are due to the tip movement across the
cleavage plane. The distances between the two consecutive
troughs, where the friction coefficient falls sharply, are 210,
260, and 340 nm, respectively, and are closely related to the
multiples of d(020) (6.517 ꢁ).
For 2 in which the layers are oriented perpendicular to
(020), a different scenario exists. The discontinuous layers
with a staggered arrangement could result in interlocking
which in turn would lead to reduced indenter movement.
On (020) at 08, when the tip traverses along the cleavage
plane, no material pile-up was noticed although AFM
images provide evidence of a limited layer movement (Fig-
ure 8a). At 908, the movement is across the cleavage plane
and the tip experiences uneven friction (evident from the
jagged profile) brought about by the vertical layer arrange-
ment (Figure 8b). This results in material pile-up at the end
of the scratch. The tip again makes a smooth movement
upon scratching the surface at 1808 since it moves along the
cleavage plane. Similar to the scratch at 908, the tip move-
ment at 2708 resulted only in abrasion with a small pile-up
(AFM scratch images for 1808 and 2708 are shown in Fig-
In both 08 and 908 scratch orientations, the tip experiences
uniform friction and this is due to smooth sliding through
the cleavage plane. However, the friction coefficient ob-
tained from 08 and 908 in crystals of 2 is higher than the
values obtained for 1 (90 and 2708), although the tip move-
ments in both cases are along the cleavage plane. This
higher magnitude could be the collective outcome of the
presence of interlayer hydrogen bonds together with the
broken layers in 2, which hinder free movement of the in-
denter. On (001), scratching along orthogonal directions
yielded only abrasions due to the flat orientation of the mo-
lecular layers, and the pyramidal tips efficiently obstruct
upward migration of flat layers (Figure S4 in the Supporting
Information). The migration distance by indenter movement
on 2 at 08 and 1808 is higher (i.e., 5 mm) and is brought
about by the staggered arrangement of the discontinuous
layers. The distance between slip lines is 350 nm, which is an
integral multiple of d(001) (Figure S5 in the Supporting Infor-
mation). While scratching at 908, the indenter encounters
a slip–slide mechanism (Figure S6 in the Supporting Infor-
mation). The line profile on the slip–slide portion revealed
that the height of the pile-up increases with increasing
scratch load together with a periodic increment and decre-
ment in the friction coefficient. As the indenter passes in
a direction perpendicular to the layer arrangement, a few
layers collectively create a resistance against indenter move-
ment; however, with increasing lateral force they break up
to release their stored elastic strain energy, thus resulting in
a sudden fall in the friction coefficient. Since this process is
continuous, a wavy movement of the indenter is observed.
Structure–Property Correlations
Due to the layered crystal structures, 1 and 2 exhibit large
anisotropies in their mechanical properties, which can be
correlated well with their crystal structures. Although the
maximum penetration depths observed for both compounds
are similar, the slope of the unloading segment is unique to
each crystal facet, which indicates significant elastic aniso-
tropy. This markedly dissimilar modulus values signify
a strong correlation between stiffness and the underlying
crystalline structure. The maximum change in modulus in
1 is 21% with (100) being the stiffest, while for 2 (020) is
stiffer by 20%. This high degree of elastic anisotropy, as
compared to that in our earlier reports, is due to the layered
molecular arrangement. However, the magnitude of aniso-
tropy is modest as compared to that seen in organic–inor-
ganic hybrid layered compounds wherein the anisotropy is
Figure 8. a) AFM images of the nanoscratch on (020). b) The coefficient
of friction at 08 and 908 (the inset shows the analysis of the distance be-
tween the two consecutive troughs where the friction coefficient sharply
decreases). c) Schematic representation of the tip movement during the
nanoscratch experiments.
of the order of 77%.[24] This distinction is due to the fact
À1
À
that hydrogen bonds such as C H···N (5 kJmol ), within
&
&
&6
&
www.chemasianj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 0000, 00, 0 – 0
ÝÝ These are not the final page numbers!

Molecular Migration in Layered Organic Donor–Acceptor Complexes
the layers of 1 and 2, are much weaker than most of the co-
ordination bonds (50–200 kJmolÀ1) in the hybrid com-
pounds. The interlayer bonds consisting of van der Waals
forces (<5 kJmolÀ1) are relatively weak and hence there is
a lower modulus in 1. However, the absence of a continuous
region, which make the indenter movement more demand-
ing. Thus, further to the mere evaluation of the crystal ani-
sotropy in terms of structure, nanoindentation can be used
as a tool to study layered crystal packing and molecular mi-
gration in molecular crystals. This study adds a new dimen-
sion to the understanding of structure–property relationships
such as softness of layered crystals and deformation mecha-
nisms, thus quantifying mechanical anisotropy and molecu-
lar migration in terms of crystal packing. In conclusion,
nanoindentation of molecular crystals is a promising tech-
nique with significant implications in various aspects of
solid-state chemistry and crystal engineering.
À
layer and the presence of C H···N hydrogen bonds in addi-
tion to p···p interactions in the interlayer space make (020)
of 2 stiffer.
Unlike E, H is a function of test methods and various
other parameters such as indenter tip geometry, applied
load, and penetration depth; hence, it cannot be considered
as an intrinsic material property. The plastic deformation is
understood in terms of nucleation, rapid multiplication, and
propagation of dislocations in metals and semiconductors,
while in amorphous materials it arises due to shear bands.
However, in molecular crystals, molecular sheets glide past
one another along specific crystallographic planes resulting
in plastic deformation. For 1, the anisotropy in H is 16%
and for 2 it is only 4%. The continuous layer structure in
1 and the absence of directional interaction in the interlayer
region make the slip more facile. In 2, the layers in [020] are
Experimental Section
Complex Preparation
A 1:1 mixture of TCNB and pyrene or phenanthrene was ground in the
presence of a few drops of methanol (solvent-drop grinding). The pyrene
mixture turned deep red while the phenanthrene mixture became orange.
The respective complexes were dissolved in various solvents and solvent
mixtures. While an acetone–ethyl acetate mixture yielded blocks of the
pyrene complex, small blocks of the phenanthrene complex were ob-
tained from acetone. These crystals were of a quality suitable for nanoin-
dentation; in other words, they were large enough and with well-devel-
oped faces.
À
stabilized through both C H···N and p···p interactions and
this makes the slip comparatively demanding. In addition,
À
the breaking of directional C H···N hydrogen bonds, which
are responsible for interlayer binding, appears as serrations
Nanoindentation
on the P–h curves.
The crystals were firmly mounted on a stud using cyanoacrylate glue
such that two different crystallographic faces can be indented. Indenta-
tion experiments were performed on these facets using a Triboindenter
(Hysitron, Minneapolis, USA) with in situ AFM imaging capability. The
machine continuously monitors the load, P, and depth of penetration, h,
of the tip with force and displacement resolutions of 1 nN and 0.2 nm, re-
spectively. A Berkovich diamond indenter with a tip radius of 100 nm
was used. Before the indentations, crystal surfaces were imaged in the
Conclusions
Large anisotropy in interaction characteristics and long-
range molecular layer gliding in organic charge-transfer
complexes of 1,2,4,5-tetracyanobenzene (TCNB) with
pyrene (1) or phenanthrene (2) have been unambiguously
established using nanoindentation and scratch experiments.
The layered crystal packing in 1 with layers that are 688
skew under the (002) face yields material pile-up in just one
orientation. This is due to the large structural anisotropy in
the intra- and interlayer regions along with a close align-
ment in the angular orientation with the half angle of the
Berkovich indenter tip. The interaction anisotropy and the
layered nature of crystal packing are evident from the large
anisotropy existing between (100) and (002) in the elastic
modulus (21%) and hardness (16%). The discontinuous
molecular stacks, arranged in a staggered manner, result in
a higher modulus of (020) (20%) as compared to (001),
AFM mode in order to find relatively smooth regions. A peak load, Pmax
,
of 6 mN, with loading and unloading rates of 0.6 mNsÀ1 and a hold time
(at Pmax) of 30 s, was employed. Post-indentation images of the impres-
sions were captured immediately to avoid any time-dependent elastic re-
covery. A minimum of fifteen indentations were performed in each case
and the average of them is reported. The P–h curves were analyzed using
the Oliver–Pharr method[25] to extract the elastic modulus, E, of the crys-
tal. However, this method was not employed for estimating the hardness,
H, as pile-up of material against the indenter faces (which is due to the
plastic flow) can lead to an overestimation of H. Hence, H was deter-
mined as Pmax/A, where A is the contact area estimated from AFM
images of the indentation impressions. Nanoscratch experiments were
performed with a ramping force (normal force increasing with time) of
3.33 N msÀ1. In all cases, 10 mm long scratches were made. The experi-
mental methodology has been already reported.[26]
X-Ray Crystallography
À
since both C H···N hydrogen bonds and p···p interactions
Single-crystal X-ray diffraction data were collected at low temperature
(150 K) on a Rigaku Mercury 375R/M CCD (XtaLAB Mini) diffractome-
ter using graphite monochromated MoKa radiation, equipped with
a Rigaku low-temperature gas spray cooler. The detailed description of
the experiment is provided in the Supporting Information. CCDC 869505
(1) and CCDC 869504 (2) contain the supplementary crystallographic
data for this paper. These data can be obtained free of charge from the
Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_-
request/cif.
are present in the interlayer region. This difference in crys-
tal packing is further demonstrated by AFM image scans
and by analyses of the friction coefficients of four orthogo-
nal nanoscratches on the cleavage planes of 1 and 2. De-
pending on the orientation of the tip movement with respect
to the orientation of the cleavage plane, the indenter experi-
ences distinct friction coefficient and layer migration. How-
ever, for crystals of 2, the tip experiences a higher friction
coefficient with respect to 1, even in the cleavage direction,
due to the presence of hydrogen bonds in the interlayer
Chem. Asian J. 2012, 00, 0 – 0
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
7
&
&
&
These are not the final page numbers! ÞÞ

Upadrasta Ramamurty, Gautam R. Desiraju et al.
FULL PAPERS
Attachment Energy Calculations
Mitchell, M. J. Snowden, Powder Technol. 2004, 143, 179–185;
f) L. J. Taylor, D. G. Papadapoulos, P. J. Dunn, A. C. Bentham, N. J.
Dawson, J. C. Mitchell, M. J. Snowden, Org. Process Res. Dev. 2004,
8, 674–679.
Potential slip planes for the complexes were determined from the attach-
ment energy of various predicted crystallographic planes. The attachment
energy (Eatt) is defined as the energy released on attachment of a growth
slice to a growing crystal face. Thus, Eatt =ElattÀEslice, where Elatt is the lat-
tice energy of the crystal and Eslice is the energy that is released on the
formation of a growth slice of a thickness equal to the d-spacing for the
crystallographic plane that represents a face. The Dreiding 2.21 (Materi-
als Studio 4.4) force field was used to calculate the attachment energies
of various faces of the complexes. The complexes consist of layers of mol-
ecules stabilized through hydrogen bonds and stacking interactions.
[11] a) W. C. Duncan-Hewitt, D. L. Mount, A. Yu, Pharm. Res. 1994, 11,
616–623; b) K. J. Ramos, D. F. Bahr, J. Mater. Res. 2007, 22, 2037–
2045; c) K. J. Ramos, D. E. Hooks, D. F. Bahr, Philos. Mag. 2009, 89,
2381–2402; d) R. J. Roberts, R. C. Rowe, P. York, Int. J. Pharm.
1995, 125, 157–162; e) S. Chattoraj, L. Shi, C. C. Sun, CrystEng-
Comm 2010, 12, 2466–2472.
[12] M. S. R. N. Kiran, S. Varughese, C. M. Reddy, U. Ramamurty, G. R.
Desiraju, Cryst. Growth Des. 2010, 10, 4650–4655.
[13] S. Varughese, M. S. R. N. Kiran, K. A. Solanko, A. D. Bond, U.
Ramamurty, G. R. Desiraju, Chem. Sci. 2011, 2, 2236–2242.
[14] a) M. B. Dines, J. Chem. Educ. 1974, 51, 221; b) H. Hu, J. C. Martin,
M. Xiao, C. S. Southworth, Y. Meng, L. Sun, J. Phys. Chem. C 2011,
115, 5509–5514; c) F. J. Di Salvo, R. Schwall, T. H. Geballe, F. R.
Gamble, J. H. Osiecki, Phys. Rev. Lett. 1971, 27, 310–313.
[15] a) C. Sun, D. J. W. Grant, Pharm. Res. 2001, 18, 274–280; b) C. G.
Pierpont, Proc. Indian Acad. Sci. 2002, 114, 247–254; c) R. Bandyo-
padhyay, D. J. W. Grant, Pharm. Res. 2002, 19, 491–496; d) C. Sun,
D. J. W. Grant, Pharm. Res. 2004, 21, 382–386; e) Y. Feng, D. J. W.
Grant, Pharm. Res. 2006, 23, 1608–1616; f) C. C. Sun, H. Hou,
Cryst. Growth Des. 2008, 8, 1575–1579.
Acknowledgements
S.V. thanks the Department of Science and Technology for a Young Sci-
entist Fellowship. M.S.R.N.K thanks The University Grants Commission
for a Dr. D. S. Kothari Post-Doctoral Fellowship. G.R.D thanks DST for
a J. C. Bose Fellowship. The authors thank Mr. V. S. S. Pavan Kumar
Hari (Electrical Engineering, IISc) for help in preparing the abstract in
Sanskrit.
[16] a) C. M. Reddy, S. Basavoju, G. R. Desiraju, Chem. Commun. 2005,
2439–2441; b) C. M. Reddy, M. T. Kirchner, R. C. Gundakaram,
K. A. Padmanabhan, G. R. Desiraju, Chem. Eur. J. 2006, 12, 2222–
2234.
[17] a) C. M. Reddy, R. C. Gundakaram, S. Basavoju, M. T. Kirchner,
K. A. Padmanabhan, G. R. Desiraju, Chem. Commun. 2005, 3945–
3947; b) C. M. Reddy, K. A. Padmanabhan, G. R. Desiraju, Cryst.
Growth Des. 2006, 6, 2720–2731; c) C. M. Reddy, G. R. Krishna, S.
Ghosh, CrystEngComm 2010, 12, 2296–2314.
[18] J. D. Bauer, E. Haussuhl, B. Winkler, D. Arbeck, Cryst. Growth Des.
2010, 10, 3132–3140.
[19] R. Foster in Organic Charge Transfer Complexes, Academic Press,
New York, 1968, pp. 216–251.
[1] a) K. Chadwick, R. Davey, W. Cross, CrystEngComm 2007, 9, 732–
734; b) R. Kuroda, K. Higashiguchi, S. Hasebe, Y. Imai, CrystEng-
Comm 2004, 6, 463–468; c) G. Kaupp, Curr. Opin. Solid State Mater.
Sci. 2002, 6, 131–138; d) G. Kaupp, J. Schmeyers, Angew. Chem.
1993, 105, 1656–1658; Angew. Chem. Int. Ed. 1993, 32, 1587–1589;
e) R. B. Godiksen, Z. T. Trautt, M. Upmanyu, J. Schiotz, D. J.
Jensen, S. Schmidt, Acta Mater. 2007, 55, 6383–6391.
[2] a) G. M. J. Schmidt, J. Chem. Soc. 1964, 2014–2021; b) G. M. J.
Schmidt, Pure Appl. Chem. 1971, 27, 647–678.
[3] a) G. Kaupp, M. R. Naimi-Jamal, CrystEngComm 2005, 7, 402–410;
b) G. Kaupp, Top. Curr. Chem. 2005, 254, 95–183.
ˇ
ˇ
ˇ ˇ
[4] C. Karunatilaka, D. K. Bucar, L. R. Ditzler, T. FriScic, D. C. Swen-
son, L. R. MacGillivray, A. V. Tivanski, Angew. Chem. 2011, 123,
8801–8805; Angew. Chem. Int. Ed. 2011, 50, 8642–8646.
[5] R. P. Rastogi, N. B. Singh, J. Phys. Chem. 1968, 72, 4446–4449.
[6] G. Kaupp, J. Schmeyers, U. D. Hangen, J. Phys. Org. Chem. 2002,
15, 307–313.
[20] C. K. Prout, T. Morley, I. J. Tickle, J. D. Wright, J. Chem. Soc. Perkin
Trans. 2 1973, 523–527.
[21] G. Toussaint, E. Burov, J. P. Avouac, Tectonics 2004, 23, TC6003.
[22] D. Olusanmi, K. J. Roberts, M. Ghadiri, Y. Ding, Int. J. Pharm.
2011, 411, 49–63.
[7] G. Kaupp, CrystEngComm 2003, 5, 117–133.
[8] A. Herrmann, G. Kaupp, T. Geue, U. Pietsch, Mol. Cryst. Liq. Cryst.
1997, 293, 261–275.
[9] A. Gouldstone, N. Chollacoop, M. Dao, J. Li, A. M. Mino, Y. L.
Shen, Acta Mater. 2007, 55, 4015–4039.
[10] a) M. S. R. N. Kiran, S. Varughese, U. Ramamurty, G. R. Desiraju,
CrystEngComm 2012, 14, 2489–2493; b) P. J. Halfpenny, K. J. Rob-
erts, J. N. Sherwood, J. Mater. Sci. 1984, 19, 1629–1637; c) D. H. C.
Wendy, G. C. Weatherly, J. Mater. Sci. Lett. 1989, 8, 1350–1352;
d) X. Liao, T. S. Wiedmann, J. Pharm. Sci. 2004, 93, 2250–2258;
e) L. J. Taylor, D. G. Papadapoulos, P. J. Dunn, A. C. Bentham, J. C.
[23] J. D. Wright, K. Yakushi, H. Kuroda, Acta Crystallogr. Sect. B 1978,
34, 1934–1938.
[24] J. C. Tan, A. K. Cheetham, Chem. Soc. Rev. 2011, 40, 1059–1080.
[25] W. C. Oliver, G. M. Pharr, J. Mater. Res. 1992, 7, 1564–1583.
[26] a) P. Kumar, M. S. R. N. Kiran, Nanoscale Res. Lett. 2010, 5, 1085–
1092; b) P. Kumar, M. S. R. N. Kiran, Sci. Technol. Adv. Mater. 2010,
11, 025003–025010.
Received: March 9, 2012
Published online: && &&, 0000
&
&
&8
&
www.chemasianj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 0000, 00, 0 – 0
ÝÝ These are not the final page numbers!

FULL PAPERS
Migration under stress: Indentation
and nanoscratch experiments on lay-
ered organic donor–acceptor com-
plexes result in breaking of interlayer
interactions, allowing molecular sheets
to glide over one another with ease.
Molecular Migration
Sunil Varughese,
Mangalampalli S. R. N. Kiran,
Upadrasta Ramamurty,*
Gautam R. Desiraju*
&&&& — &&&&
Nanoindentation as a Probe for
Mechanically-Induced Molecular
Migration in Layered Organic Donor–
Acceptor Complexes
Chem. Asian J. 2012, 00, 0 – 0
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
9
&
&
&
These are not the final page numbers! ÞÞ