Self-Healing Molecular Crystals

Article abstract of DOI:10.1002/anie.201606003

One of the most inevitable limitations of any material that is exposed to mechanical impact is that they are inexorably prone to mechanical damage, such as cracking, denting, gouging, or wearing. To confront this challenge, the field of polymers has developed materials that are capable of autonomous self-healing and recover their macroscopic integrity similar to biological organisms. However, the study of this phenomenon has mostly remained within the soft materials community and has not been explored by solid-state organic chemists. The first evidence of self-healing in a molecular crystal is now presented using crystals of dipyrazolethiuram disulfide. The crystals were mildly compressed and the degree of healing was found to be 6.7 %. These findings show that the self-healing properties can be extended beyond mesophasic materials and applied towards the realm of ordered solid-state compounds.

Full text of DOI:10.1002/anie.201606003

Angewandte
Communications
Chemie
International Edition: DOI: 10.1002/anie.201606003
German Edition: DOI: 10.1002/ange.201606003
Crystal Engineering
Self-Healing Molecular Crystals
ˇ
Patrick Commins, Hideyuki Hara, and Pance Naumov*
Abstract: One of the most inevitable limitations of any
material that is exposed to mechanical impact is that they are
inexorably prone to mechanical damage, such as cracking,
denting, gouging, or wearing. To confront this challenge, the
field of polymers has developed materials that are capable of
autonomous self-healing and recover their macroscopic integ-
rity similar to biological organisms. However, the study of this
phenomenon has mostly remained within the soft materials
community and has not been explored by solid-state organic
chemists. The first evidence of self-healing in a molecular
crystal is now presented using crystals of dipyrazolethiuram
disulfide. The crystals were mildly compressed and the degree
of healing was found to be 6.7%. These findings show that the
self-healing properties can be extended beyond mesophasic
materials and applied towards the realm of ordered solid-state
compounds.
Having polymers as a source of inspiration, the design of
a self-healing crystal required the selection of a chemically
reactive reaction system that would be applicable to molec-
ular crystals. Microencapsulate[27] and nanoparticle-based
polymers[28] and other multicomponent materials were omit-
ted because they cannot be realized with molecular crystals.
Other self-healing polymers that are based on Diels–Alder
cycloadditions[29] or ligand-exchange reactions[30] were also
not applicable since they would require introduction of
supramolecular templates.[31] Instead, we drew ideas from
dynamic covalent chemistry, which utilizes molecules that
undergo rapid reversible bond formations.[32,33] From dynamic
molecular systems with identical reacting species, disulfides
were ultimately selected based on the low bond dissociation
energy and reversibility of the bond reformation close to
ambient conditions.[34–37] With this in mind, we decided to
study dipyrazolethiuram disulfide (1H-pyrazole-1-carbothioic
dithioperoxyanhydride) 1 (Figure 1). The ease of homolytic
dissociation of the disulfide bond in 1 and the delocalization
of the ensuing radicals across the two sulfur atoms should
allow the resulting radical to readily regenerate the disulfide
bridge with a nearby reactive neighbor. Additionally, radical-
shuffling-based self-healing polymers have been proven to be
quite robust and some can heal without external stimuli.[38–40]
Compound 1 was synthesized in one step using an adapted
procedure from a previous report[41] (Supporting Information,
Scheme S1). A shallow non-penetrating incision was man-
ually applied along the a axis on the {010} face of a pristine
single crystal (Figure 1A and B; defecting a face was
necessary as the defect guided the propagation of the
fracture). The crystal was then flipped over and adhered to
silicone pads, which were glued to a tensile tester. The crystal
was then cracked along its a axis by applying local pressure
(Figure 1C), and the two halves were separated (Figure 1D
and E). The bifurcated pieces were compressed under 1–3 N
of force and maintained at rest at room temperature, under
ambient light for 24 h (Figure 1F). After decompression, the
strength of adhesion between the pieces was analyzed. The
self-healing process was later repeated without adhering the
crystals to the pads or adhering the pads to the tester. The
crystals were considered healed if could be lifted from one
edge and support their own weight across the crack. The
healed crystals were not only able to resist minor forces such
as gravity (Figure 1G and H; Supporting Information,
Movie S1), but also gentle poking with a polyamide tip
(Figure 1I and J). While some of the crystals separated under
this stress, quite remarkably, others did not.
T
he deterioration and accumulation of damage of materials
over time has been a commonly accepted maxim by people
for thousands of years. The discovery of the self-healing
materials has begun to challenge this idea by creating
materials that are able to autonomously repair themselves
after a mechanical defect has been introduced.[1–4] The field
gained its breakthrough in 2001 when White and co-workers
reported the first autonomously self-healing polymer,[5] and it
has been growing exponentially over the past 15 years.[6–8]
Self-healing polymers and composites have been created
using microencapsulated healing agents,[9] thermal reac-
tions,[10] supramolecular interactions,[11–13] and other mecha-
nisms.[14–16] However, the interest in this exciting area of
research has remained limited to the soft materials research
community and it has not been explored by solid-state
chemists, mainly because of the perceived rigidity, slow
diffusion, and chemical inertness of solids. However, recent
advancements in solid-state chemistry[17–23] have been aimed
towards dispelling these beliefs and, considering that the
migration of mass across the surface of molecular crystals can
be substantial[24–26] we presume the self-healing concept can
be extended to ordered molecular solids. Herein, we describe
the findings towards the first evidence of self-healing in
a molecular crystal.
[*] Dr. P. Commins, Prof. P. Naumov
New York University Abu Dhabi,
Abu Dhabi (United Arab Emirates)
E-mail: pance.naumov@nyu.edu
Dr. H. Hara
The percent healing is calculated by measuring the tensile
strength of the virgin material against the healed material.[6,7]
The tensile strength of 1 was acquired by pulling a pristine
crystal apart (Figure 2A–D). As the sample is pulled the line
shape shows a gradual increase in the load due to the elastic
Bruker Biospin K.K.
3–9, Moriya, Kanagawa, Yokohama, Kanagawa 221-0022 (Japan)
Supporting information and the ORCID identification number(s) for
the author(s) of this article can be found under:
http://dx.doi.org/10.1002/anie.201606003.
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

Angewandte
Communications
Chemie
Figure 2. Mechanical evidence of self-healing crystal. A)–D) Images
and tensile strength profile of a pristine crystal of 1. E)–H) Images and
tensile strength of a self-healed crystal of 1. The sharp decrease in
load before point f corresponds to separation of the adhered pieces.
The calculated healing versus the virgin material is 6.7%. I)–L) Images
and tensile strength profile of tetraethylthiuram disulfide (2), a crystal
that is not able to self-heal.
Figure 1. The procedure for cracking and healing a crystal of 1. A),B) A
shallow non-penetrating incision is made along the a axis on the {010}
face of the crystal by using a scalpel. C) The crystal is then flipped and
adhered onto the tensile tester. D) The crystal is cracked by applying
localized pressure. E),F) The two pieces are completely separated,
compressed and left to self-heal for 24 hours. G) An optical image of
a healed crystal of 1 mounted in putty. H) The same crystal is then
inverted. I),J) A healed crystal did not separate after being lifted from
the cracked side or even after being poked by a polyamide tip.
K) Chemical structures of compounds 1, 2, and 3.
normalized to the surface area, the healed crystal had a tensile
strength of 0.086 MPa, which corresponds to 7.8% healing.
The average tensile strength of two healed crystals was
0.074 MPa, which gives a healing of 6.7%.
component of the glue–crystal interface. The sample reaches
a point of fracture (Figure 2B), followed by a steep drop and
flattening of the curve after separation (Figure 2C). The
average load decrease at the separation point normalized by
the cross-sectional area between the two faces of two crystals
gives a tensile strength of 1.1 MPa. The tensile profile of
a healed crystal is displayed in Figure 2D–G. At a displace-
ment of 0.03 mm (point e in Figure 2H), the crystal is still
compressed by the pads and there is a rapid increase in force
as they are decompressed. After a displacement of 0.18 mm
(Figure 2F), the crystal–crystal interface becomes strained
and breaks, as shown by the sudden drop in load at 0.19 mm
(point f in Figure 2H). Based on the drop in the force
Inspection by SEM revealed the topographic details of the
healed crystal (Figure 3A–D). While the front face of the
crystal is well faceted (Figure 3C), relatively smooth, and the
appearance of the crack is not apparent, the kerf created by
the scalpel blade can be seen running across the middle of the
back face (Figure 3D). The zoomed-in images of the interface
between the two parts in Figure 3E and F show areas where
the two pieces are still separated. However, there are also
areas of attachment where the pieces are physically connected
(Supporting Information, Comment 1).
The topology of the healed crystals was further inspected
using confocal fluorescence microscopy. The fluorescence
2
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
These are not the final page numbers!

Angewandte
Communications
Chemie
Figure 4. A CT scan of the healed crystal. The cross-sections are
presented below the image, showing the disappearance of the crack.
Figure 3. Imaging of healed crystals of 1. A),B) SEM image of the front
and back face of the healed crystal. C),D) Zoomed-in images of the
highlighted sections in (A) and (B). Blue and red arrows highlight
areas of separation and attachment, respectively. E) A healed crystal
coated with a fluorescent dye was imaged using confocal fluorescence
microscopy and the material spanning the cracked region (red arrow)
is fluorescent. F) The dye is found at both sides of the attached region
running along the walls signifying the disconnected areas (blue
arrows).
analysis (highlighted by the red bar in Figure 4). As the
scan proceeds along the long axis of the crystal the green
coloration from the crack begins to dissipate and a much more
homogenous blue color appears, indicating uniform density
across the sample (see the cross-sectional images in Figure 4).
The homogeneity of the blue coloration across the entirety of
the crystal substantiates the attachment of the two pieces in
the crystal interior. Furthermore, the color gradient in the
region of the crack indicates that there is a change in density
along the recess, in support of the concept according to which
there are areas where the crystal is separated, partially healed,
and fully healed.
images overlaid against the optical images of a partially
healed crystal at two different depths are shown in Figure 3E
and F, and a movie with the 3D scan is available in the
Supporting Information, Movie S2. All of the exposed
surfaces of the crystal were coated with the dye, and the
area of attachment between the two halves was also
fluorescent. When the crystal is imaged at a lower depth the
disconnected portions come into focus. The dye runs along the
walls of these areas of separation from the top to the bottom.
However, the attached areas only show fluorescence at the
surface of the crystal, and the dye was not found at other focal
depths. These images provide qualitative
We postulated that the self-healing is caused by shuffling
À
of S S bonds across the cracked interfaces. In the crystal, the
À
S S bond would homolytically break and the radical would be
delocalized across the two sulfur atoms. The radical could
À
then interact with a nearby sulfur atom, forming a new S S
bond, and the reaction would propagate (Figure 5). To
substantiate the possibility of disulfide shuffling, mixed
evidence that the two pieces are connected.
Because the SEM was unable to acquire
information on the internal structure of the
crystal, we turned to X-ray computed
tomography (CT) to visualize the internal
features of the healed crystal by probing the
spatially resolved density using X-ray
absorption (Figure 5; Supporting Informa-
tion, Movies S3 and S4). The facets, angles,
and features from the electron and optical
micrographs are reproduced in the CT
images, and the crack is visible as a light
green line running across the center of the
crystal. The crack between the two sides of
the crystal can also clearly be seen by the
bright green line between the two pieces on
Figure 5. Crystal structure and the postulated mechanism of healing. During the compres-
sion the radicals are able to shuffle and bond to nearby molecules, whereby the two pieces
À
the far left side in the cross-sectional are rejoined. The S S distances are rounded; hydrogen atoms were omitted for clarity.
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

Angewandte
Communications
Chemie
solutions of 1 and the tetraethylthiuram disulfide 2 (Support-
clear whether this contributes to the healing. To assess the
possibility of pressure-induced transition, the phase of the
solid was investigated by powder XRD after successive
grinding durations, and no phase transitions were observed
(Supporting Information, Figure S4).
The shuffling of the disulfide bonds across the contact
surface would require significant mass transport at the
interface. To study the molecular migration at the crystal
surface of 1, the {010} face was indented with a diamond-
tipped AFM cantilever. The resulting canyon and pileup
features were monitored using AFM (Figure 6; Supporting
ing Information, Figure S1) in [D6]benzene were studied
1
using H NMR spectroscopy (Supporting Information, Fig-
ure S2). In principle, the two compounds should be able to
shuffle, and if equimolar solutions of 1 and 2 are mixed,
a 1:2:1 mixture of 1, 3, and 2 will form, as shown previously.[37]
Not surprisingly, after 24 hours of mixing, the peak of the
heterodimer ethylpyrazolethiuram disulfide (3) begins to
appear as a small shoulder at 8.06 ppm in the 1H NMR
spectrum. As time progresses, the intensity of the peak from 3
increases and the peak at 7.98 ppm decreases. After
149 hours, the reaction has progressed to a 1:1:1 ratio. The
slow reaction progression in solution can be rationalized by
À
the ESR spectra (described below), which show the R SC
radical signal is below the limit of detection in solutions of
benzene; thus, it is reasonable to assume the reaction may
proceed slower than previously reported.[37] However, the
reaction proceeds in solution and it can occur at ambient
conditions and under visible light. We assume this reaction is
equally possible in the solid and the disulfide bonds between
adjacent molecules can shuffle and reform to heal the two
halves of the crystal.
Previous ESR measurements on similar thiuram disulfides
have shown that an air-stable radical is present in the solid
and in solution.[42,43] Once the radicals are generated, whether
by mechanical force, heat, or light irradiation, they are
expected to be able to propagate within the crystalline lattice.
Past research in this field performed by the McBrideꢀs group
has demonstrated propagation of radical chain reactions
throughout molecular layers.[44] Both solids and solutions of
1 were studied by ESR spectroscopy (Supporting Informa-
tion, Figure S3). At 298 K, in the dark, solution of 1 shows
a signal at g = 2.004. When the sample is irradiated with
visible light, an additional signal at g = 2.009 is observed, and
when irradiated with UV light, a third signal occurs at g =
2.024. When a crystalline powder is used, an identical
spectrum is obtained regardless of irradiation, and the three
signals (g = 2.004, 2.009, and 2.026) are present. This result
confirms that the radical species necessary for shuffling are
present at ambient temperature and even without irradiation
(Supporting Information, Comment 2).[45]
Figure 6. Mass transport at the surface of the crystal of 1 observed by
AFM. The images of the indented sample over 62.5 h. The pileup
features gradually disappear over time, indicating significant mass
transfer.
Information, Movie S5). After 7.5 hours, the pileup features
have noticeably decreased in height. After 62.5 hours almost
all of the features have dissipated and only the initial indent is
visible. Each of the peaks decreases fairly linearly with
respect to time but at different rates (Supporting Information,
Figure S6). This observation can be rationalized by the peak
morphology; wider, more blunt features are expected to
decrease slower than sharp, narrow peaks. For instance,
peak 1 (Supporting Information, Figure S5, blue line) with
initial height of 0.49 mm slowly recedes into the baseline
features after 37 hours. The average rate of decrease of the
three features was 9.4 nmhÀ1. Assuming that this rate is
indicative of the mass transfer at the surface, it implies that
when the sample is compressed for 24 hours, the material may
move up to 225 nm during the compression time to achieve
contact and heal.
À
The shuffling mechanism requires proximity of the S S
bonds in the crystal. The X-ray crystal structure of 1[46] shows
there are three prominent sulfur–sulfur contacts of 3.6375(8),
3.7898(7), and 4.1107(8) ꢁ (Figure 5). The 3.6375(8) and
3.7898(7) ꢁ contacts appears to be ideal for shuffling, as the
atoms are perfectly positioned to form a new bond. However,
experimentally the split crystal cannot be easily compressed
along this axis because of its rhombohedral morphology; it
has two sloped sides of 1148 and when the pieces are
compressed, they slide and eventually slip. Indeed, the crystal
could only be compressed along the sides that were flat, which
required splitting along the a axis. Along the c axis there is
a close contact between two sulfur atoms of 4.1107(8) ꢁ. This
interaction should be able to create new covalent bonds
between the two cracked halves and may be responsible for
the healing (Supporting Information, Comment 3). A degree
In conclusion, we have obtained evidence for the first
example of a self-healing organic molecular crystal. The
crystals exhibit about 6.7% healing at ambient conditions,
À
of S S bond shuffling may also occur between the two sulfur
atoms along the a axis of the crystal (Figure 5), but it is less
4
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
These are not the final page numbers!

Angewandte
Communications
Chemie
Takeda, S. Kang, S. Kanegawa, O. Sato, Nat. Commun. 2015, 6,
8810.
[20] J.-K. Sun, W. Li, C. Chen, C. X. Ren, D. M. Pan, J. Zhang,
Angew. Chem. Int. Ed. 2013, 52, 6653 – 6657; Angew. Chem. 2013,
125, 6785 – 6789.
with no other external stimuli other than moderate mechan-
ical compression. We postulate the disulfide shuffling mech-
anism is responsible for the self-healing property. The crystal
À
structure of 1 contains three different S S contacts that are
proximal and they may be capable of forming new bonds. The
radicals proposed for the shuffling mechanism are also
observed, and a degree of mass transfer necessary to span
the interstitial gaps in between the two interfaces was also
found. We believe other molecular crystals may also be
capable of this phenomenon.
[21] A. G. Shtukenberg, Y. O. Punin, A. Gujral, B. Kahr, Angew.
Chem. Int. Ed. 2014, 53, 672 – 699; Angew. Chem. 2014, 126, 686 –
715.
[22] M. K. Panda, S. Ghosh, N. Yasuda, T. Moriwaki, G. D. Mukher-
jee, C. M. Reddy, P. Naumov, Nat. Chem. 2015, 7, 65 – 72.
[23] S. Hayashi, T. Koizumi, Angew. Chem. Int. Ed. 2016, 55, 2701 –
2704; Angew. Chem. 2016, 128, 2751 – 2754.
[24] G. Kaupp, M. R. Naimi-Jamal, CrystEngComm 2005, 7, 402 –
410.
Acknowledgements
[25] S. Varughese, M. S. R. N. Kiran, U. Ramamurty, G. R. Desiraju,
Angew. Chem. Int. Ed. 2013, 52, 2701 – 2712; Angew. Chem.
2013, 125, 2765 – 2777.
[26] S. Varughese, M. S. R. N. Kiran, U. Ramamurty, G. R. Desiraju,
Chem. Asian J. 2012, 7, 2118 – 2125.
The experiments were partially carried out using Core
Technology Platform resources at NYU Abu Dhabi. We
thank Dr. James Weston for assistance with the CT and Dr.
Gijo Raj for help with the AFM.
[27] R. S. Trask, G. J. Williams, I. P. Bond, J. R. Soc. Interface 2007, 4,
363 – 371.
[28] J. Y. Lee, G. A. Buxton, A. C. Balazs, J. Chem. Phys. 2004, 121,
5531 – 5540.
Keywords: computed tomography · crystal engineering ·
[29] Y. Zhang, A. A. Broekhuis, F. Picchioni, Macromolecules 2009,
42, 1906 – 1912.
disulfides · self-healing · smart materials
[30] Y. Shi, M. Wang, C. Ma, Y. Wang, X. Li, G. Yu, Nano Lett. 2015,
15, 6276 – 6281.
ˇˇ ´
[31] T. Friscic, L. R. MacGillivray, Chem. Commun. 2003, 1306 –
1307.
[1] C. Edvardsen, ACI Mater. J. 1999, 96, 448 – 454.
[2] H. M. Jonkers, A. Thijssen, G. Muyzer, O. Copuroglu, E.
Schlangen, Ecol. Eng. 2010, 36, 230 – 235.
[3] H. J. Yang, Y. T. Pei, J. C. Rao, J. T. M. De Hosson, J. Mater.
Chem. 2012, 22, 8304 – 8313.
[4] G. Q. Xu, M. J. Demkowicz, Phys. Rev. Lett. 2013, 111, 145501.
[5] S. R. White, N. R. Sottos, P. H. Geubelle, J. S. Moore, M. R.
Kessler, E. N. Brown, S. Viswanathan, Nature 2001, 409, 794 –
797.
[6] D. Y. Wu, S. Meure, D. Solomon, Prog. Polym. Sci. 2008, 33,
479 – 522.
[7] E. B. Murphy, F. Wudl, Prog. Polym. Sci. 2010, 35, 223 – 251.
[8] Y. Yang, M. W. Urban, Chem. Soc. Rev. 2013, 42, 7446 – 7467.
[9] E. N. Brown, S. R. White, N. R. Sottos, J. Mater. Sci. 2004, 39,
1703 – 1710.
[10] X. Chen, M. A. Dam, K. Ono, A. Mal, H. Shen, S. R. Nutt, K.
Sheran, F. Wudl, Science 2002, 295, 1698 – 1702.
[11] A. Phadke, C. Zhang, B. Arman, C. C. Hsu, R. A. Mashelkar,
A. K. Lele, M. J. Tauber, G. Arya, S. Varghese, Proc. Natl. Acad.
Sci. USA 2012, 109, 4383 – 4388.
[12] I. Jeon, J. Cui, W. R. K. Illeperuma, J. Aizenberg, J. J. Vlassak,
Adv. Mater. 2016, 28, 4678 – 4683.
[13] Q. Wang, J. L. Mynar, M. Yoshida, E. Lee, M. Lee, K. Okuro, K.
Kinbara, T. Aida, Nature 2010, 463, 339 – 343.
[32] Y. Jin, C. Yu, R. J. Denman, W. Zhang, Chem. Soc. Rev. 2013, 42,
6634 – 6654.
[33] A. Herrmann, Chem. Soc. Rev. 2014, 43, 1899 – 1933.
[34] R. Steudel, Y. Steudel, A. M. Mak, M. W. Wong, J. Org. Chem.
2006, 71, 9302 – 9311.
[35] M. M. Coleman, J. R. Shelton, J. L. Koenig, Macromolecules
2011, 44, 2536 – 2541.
[36] J. Canadell, H. Goossens, B. Klumperman, Macromolecules
2012, 45, 142 – 149.
[37] Y. Amamoto, H. Otsuka, A. Takahara, K. Matyjaszewski, Adv.
Mater. 2012, 24, 3975 – 3980.
[38] K. Imato, M. Nishihara, T. Kanehara, Y. Amamoto, A. Takahara,
H. Otsuka, Angew. Chem. Int. Ed. 2012, 51, 1138 – 1142; Angew.
Chem. 2012, 124, 1164 – 1168.
[39] C. E. Yuan, M. Z. Rong, M. Q. Zhang, Z. P. Zhang, Y. C. Yuan,
Chem. Mater. 2011, 23, 5076 – 5081.
[40] Y. Amamoto, J. Kamada, H. Otsuka, A. Takahara, K. Matyjas-
zewski, Angew. Chem. Int. Ed. 2011, 50, 1660 – 1663; Angew.
Chem. 2011, 123, 1698 – 1701.
[41] L. N. Sobenina, A. P. Demenev, A. I. Mikhaleva, V. N. Elokhina,
A. G. Malꢀkina, O. A. Tarasova, I. A. Ushakov, B. A. Trofimov,
Synthesis 2001, 293 – 299.
[42] P. J. Nichols, M. W. Grant, Aust. J. Chem. 1983, 36, 1379 – 1386.
[43] M. Coleman, J. R. Shelton, J. L. Koenig, Rubber Chem. Technol.
1973, 46, 957 – 980.
[44] K. L. Pate, J. M. McBride, Helv. Chim. Acta 2000, 83, 2352 –
2362.
[45] O. Ito, K. Nogami, M. Matsuda, J. Phys. Chem. 1981, 85, 1365 –
1368.
[14] B. Ghosh, M. W. Urban, Science 2009, 323, 1458.
[15] P. Cordier, F. Tournilhac, C. Souliꢂ-Ziakovic, L. Leibler, Nature
2008, 451, 977 – 980.
[16] J. J. Cash, T. Kubo, A. P. Bapat, B. S. Sumerlin, Macromolecules
2015, 48, 2098 – 2106.
[17] P. Naumov, S. Chizhik, M. K. Panda, N. K. Nath, E. Boldyreva,
Chem. Rev. 2015, 115, 12440 – 12490.
[18] C. S. Vogelsberg, M. A. Garcia-Garibay, Chem. Soc. Rev. 2012,
41, 1892 – 1910.
[46] F. K. Keter, M. J. Nell, I. A. Guzei, B. Omondi, J. Darkwa, J.
Chem. Res. 2009, 322 – 325.
[19] S. Q. Su, T. Kamachi, Z.-S. Yao, Y.-G. Huang, Y. Shiota, K.
Yoshizawa, N. Azuma, Y. Miyazaki, M. Nakano, G. Murata, S.
Received: June 21, 2016
Published online: && &&, &&&&
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!

Angewandte
Communications
Chemie
Communications
Crystal Engineering
P. Commins, H. Hara,
P. Naumov*
&&&& — &&&&
Self-Healing Molecular Crystals
Healing crystals: Single crystals of dipyr-
azolethiuram disulfide display self-heal-
ing properties in air when using only
moderate mechanical compression. The
compound shows that the self-healing
phenomenon found in polymers can be
extended to crystalline materials.
6
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
These are not the final page numbers!