Reagent-installed capsule network: Selective thiocarbamoylation of aromatic amines in crystals with preinstalled CH3NCS

Article abstract of DOI:10.1002/anie.201107190

Crystalline reagent capsules were prepared by installing CH₃NCS into networked molecular capsules. While the tight encapsulation completely prevented leaching of reagent molecules into the supernatant, introduction of amines into the interstitial pores triggered reagent delivery. As a result, enhanced substrate selectivity was observed in crystalline-state thiocarbamoylation (see picture; 86:14 in favor of 2- vs. 1-naphthylamine). Copyright

Full text of DOI:10.1002/anie.201107190

Angewandte
Chemie
DOI: 10.1002/anie.201107190
Molecular Capsules
Reagent-Installed Capsule Network: Selective Thiocarbamoylation of
Aromatic Amines in Crystals with Preinstalled CH3NCS**
Yasuhide Inokuma, Guo-Hong Ning, and Makoto Fujita*
Porous coordination networks[1] have recently served as
crystalline molecular flasks in which chemical reactions
occur between incoming reagents and preinstalled substrates
within the pores. Since the substrates are in most cases
covalently or noncovalently fixed on the network frame-
work,[2,3] the products cannot be isolated unless the complex is
demolished. With a view to practical applications in organic
synthesis, we had the idea of “reagent-installed porous
networks” in which common organic substrates are converted
to desired products by simply passing them through the pores.
Such reagent-installed networks are particularly useful if
troublesome reagents (e.g., highly volatile, unstable, toxic, or
explosive ones) are stably preinstalled.
voids are fluid, and their reactivity is enhanced. Given these
properties of 1, we envisioned stable storage of reagents in the
capsule units and triggering their delivery by introduction of
substrates into the voids. Furthermore, the reagent installed in
the complex may show unusual reactivity and selectivity in
crystalline-state reactions. We have now synthesized network
complex 1 encapsulating highly volatile CH3NCS (3) by single
crystal-to-single crystal (SCSC) reagent loading into as-
synthesized 1. Reagent 3 is strongly trapped in the capsule
units and does not leach out of the capsule even if the network
crystals are exposed to a solution of the substrate. We show
that networked capsules 1ꢀ3 can act as a crystalline reagent
for thiocarbamoylation of aromatic amines with enhanced
substrate selectivity and that, after the reaction, network 1 can
be reused by reloading the reagent.
We recently synthesized porous coordination network
1 composed of two compartments: infinitely arrayed M6L4
capsules and the remaining interstitial voids (Scheme 1).[4]
Within the capsule units, even reactive molecules are stably
stored. In contrast, incoming substrates in the interstitial
Capsule network 1 was prepared from tris(3-pyridyl)tria-
zine (2) and Co(NCS)2 in thiophene/MeOH.[4] CH3NCS (3)
was easily introduced into the capsules by soaking the as-
synthesized crystals of 1 in a solution of 3 in hexadecane at
room temperature for 3 d, during which the supernatant was
replaced three times with fresh guest solution. Treating the
crystals with fresh hexadecane for 4.5 h washed out the
surplus of 3 remaining in the interstitial pores to give reagent-
installed capsule network 1ꢀ3.
Reagent installation proceeded in an SCSC fashion, and
the structure of 1ꢀ3 was clearly analyzed by X-ray crystallog-
raphy (Figure 1). Each capsule is occupied by four molecules
of 3, which are tightly packed through CH–p and p–p
interactions with ligand 2.[5] Importantly, no electron density
assignable to 3 is found in the interstitial pores. These
Scheme 1. Preparation of reagent-installed capsule network 1ꢀ3.
[*] Dr. Y. Inokuma, G.-H. Ning, Prof. Dr. M. Fujita
Department of Applied Chemistry, School of Engineering
The University of Tokyo
Hongo, Bunkyo-ku, Tokyo 113-8656 (Japan)
E-mail: mfujita@appchem.t.u-tokyo.ac.jp
[**] This research was supported by the CREST project of the Japan
Science and Technology Agency (JST), for which M.F. is the principal
investigator, and also in part by KAKENHI (20044006), JSPS, and
Global COE Program (Chemistry Innovation through Cooperation
of Science and Engineering), MEXT (Japan). Y.I. thanks Kurata
Grants for younger scientists.
Figure 1. X-ray crystal structure of 1ꢀ3: a) Network structure. b) One
capsule subunit with guests shown as CPK models. c) Ball-and-stick
representation of the four packed guest molecules 3. (C light blue, S
yellow, N blue, and H white).
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201107190.
Angew. Chem. Int. Ed. 2012, 51, 2379 –2381
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2379

.
Angewandte
Communications
observations are in good accordance with inductively coupled
plasma (ICP) analysis, which determined a Co/S atomic ratio
of 1/3. Elemental analysis agreed with the formula {[(Co-
(NCS)2)3(2)4]·(3)3·(C16H34)x}n (x ꢁ 5), which indicates the
presence of four molecules of 3 per capsule (i.e., 7 wt%
inclusion; see Supporting Information).[6]
Leaching tests on 1ꢀ3 confirmed how strongly the guest
reagents are held within the networked capsules. 1) Guest 3
did not leach out into the supernatant after washing crystals of
1ꢀ3 with hexadecane (Figure S1 in the Supporting Informa-
tion). 2) Exposure of crystals of 1ꢀ3 (ca. 150 mg, containing
ca. 10 mg of 3) to air at room temperature for 3 d caused only
a very small change (< 1%) in Co/S ratio. The suppressed
volatility of encapsulated 3 is in striking contrast to the highly
volatile nature of neat 3: 10 mg of solid 3 sublimed within
30 min under the same conditions. 3) Thermogravimetric/
differential scanning calorimetry/mass spectrometry (TG-
DSC-MS) analysis indicated that guest 3 remained encapsu-
lated up to 2008C; after a gentle loss of 32 wt% due to
evaporation of the hexadecane solvent in the pore, a steep
weight loss with a sharp endothermic peak was recorded at
2058C, where encapsulated guest 3 (ca. 7 wt%) was released
from network 1 (Figure S2 in the Supporting Information).
Introduction of aromatic amines into the interstitial pores
triggered delivery of encapsulated reagent 3, and thus net-
work complex 1ꢀ3 was shown to be a useful reagent capsule
that is switched on by substrate uptake. From incoming amine
guests, thioureas were synthesized in situ. When crystals of
1ꢀ3 were soaked in a hexadecane solution of aniline (4a)
(0.24m, 0.75 equiv relative to CH3NCS), 4a was quickly
enclathrated into the pores of 1, concomitant with crystal
color change from pale red to deep orange. After 72 h, the
resultant crystals were collected by filtration, digested with
HCl(aq.), and extracted with CH2Cl2 to give 1-methyl-3-
phenylthiourea (5a) in 91% yield. In the supernatant, neither
reagent 3 nor product 5a was detected by 1H NMR spectros-
copy. Thus the reaction took place only in the network
crystals.
Figure 2. a) Synthesis of thiourea in network crystals of 1ꢀ3. Product
yield versus reaction time profiles b) in network crystals 1ꢀ3 and c) in
solution. Initial conditions: [4]0 =0.24m and [3]0 =0.36m (for solution
reactions). For crystalline-state reactions, 1.5 equiv of CH3NCS was
added as its inclusion complex 1ꢀ3. The yield was determined by
1H NMR spectroscopy.
Networked reagent capsules 1ꢀ3 were also able to
discriminate 2- and 1-naphthylamine (6a and 6b) in thio-
carbamoylation (Scheme 2). When crystals of 1ꢀ3 were
immersed in an equimolar solution of 6a and 6b, correspond-
ing thioureas 7a and 7b were formed in 84:16 ratio over the
initial 10 h. In contrast, the selectivity of the thiocarbamoy-
lation in solution was very poor (7a:7b = 58:42).
Scheme 2. Selective formation of thiourea 7a in network 1.
Encapsulated 3 is a much milder reagent than neat 3, and
we observed enhanced substrate selectivity. A remarkable
decrease in reaction rate was observed with a bulky aniline
derivative. The bulky amine 2,6-dimethylaniline (4b) was
enclathrated into network 1ꢀ3 as quickly as 4a. However,
thiourea 5b was formed only in 11% yield after 72 h. The plot
of reaction time versus conversion allowed us to estimate the
k4a/k4b ratio to be 10.9 (Figure 2b). This value is considerably
larger than that in a standard solution reaction (k4a/k4b = 4.3;
Figure 2c),[7] that is, networked capsules 1ꢀ3 notably dis-
criminate the difference in steric bulk around the amino
groups of the substrates. Clearer results were obtained by
a competition experiment: when a 1:1 mixture of 4a and 4b
was treated with crystals 1ꢀ3, thioureas 5a and 5b were
initially formed in 91:9 ratio, in contrast to a control experi-
ment in solution, which resulted in a 73:27 mixture (Figure S5
in the Supporting Information). Since inclusion of amines into
the interstitial pores of 1 is nonselective,[8] the observed
substrate selectivity is attributable to steric protection of 3 by
the capsules.
Finally, we emphasize that capsule network 1 is reusable.
After extraction of the products with thiophene/MeOH,
reagent 3 could be reloaded into the capsule by the same
procedure, and almost the same product selectivity was
obtained with the recycled reagent capsules.
In summary, we have achieved SCSC installation of
reagent 3 into networked molecular capsules 1 by simply
soaking crystals in a reagent solution. While the reagent
molecules were firmly encapsulated by capsule units, intro-
duction of amines into the interstitial pores triggered reagent
delivery. Moreover, we demonstrated substrate-selective
thiocarbamoylation of amines in crystals 1ꢀ3. Given strong
guest binding and stabilization effects, networked capsules
1 would provide significant benefits when the reagents are
tedious to handle (e.g., toxic or explosive), because these
guests can be safely encapsulated into the network with ease.
We believe that 1 can accommodate a variety of reagents and
thus expand the scope of crystalline-state reactions.
2380 www.angewandte.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 2379 –2381

Angewandte
Chemie
M. B. Dewal, D. Sobransingh, M. C. Paderes, A. C. Wibowo,
Received: October 11, 2011
Published online: January 19, 2012
M. D. Smith, J. A. Krause, P. J. Pellechia, L. S. Shimize, J. Am.
Chem. Soc. 2011, 133, 7025 – 7032; d) J. M. Falkowski, C. Wang, S.
Liu, W. Lin, Angew. Chem. 2011, 123, 8833 – 8837; Angew. Chem.
Int. Ed. 2011, 50, 8674 – 8678.
Keywords: host–guest systems · molecular capsules ·
.
porous coordination networks · solid-state reactions ·
synthetic methods
[4] Y. Inokuma, S. Yoshioka, M. Fujita, Angew. Chem. 2010, 122,
9096 – 9098; Angew. Chem. Int. Ed. 2010, 49, 8912 – 8914.
[5] Crystallographic data for 1ꢀ3: C26H16CoN10S2·0.36C2H3NS, Mr =
3
ꢀ
617.86, cubic Fd3c, a = b = c = 64.058(3) ꢂ, V= 262853(21) ꢂ ,
T= 90(2) K, Z = 192, 1calcd = 0.749 gcmÀ3, 8153 unique reflections
out of 11287 with I > 2s(I), 390 parameters, final R1 = 0.1424 [I >
2s(I)]. CCDC 846321 contains the supplementary crystallo-
graphic 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.. Note that the refined
occupancy of guest 3 is lower than that expected from ICP and
elemental analyses, since it is disordered within the capsule, and
was not fully resolved because of the high symmetry of the lattice.
[6] The unit lattice of 1 contains 64 M6L4 subunits, sixteen of which
are open-mouthed M6L4 cages in which the guest molecules were
not restrained.
[7] The reactions in solution followed second-order kinetics [first-
order in CH3NCS (3) and in aniline (4)], and thus the kinetic rate
constants were calculated according to the reported procedure:
a) Chemical Kinetics, 3rd ed. (Ed.: K. J. Laidler), Harper and
Row, New York, 1987; b) E. Dyer, J. F. Glenn, J. Am. Chem. Soc.
1957, 79, 366 – 369.
[1] For reviews, see: a) S. R. Batten, R. Robson, Angew. Chem. 1998,
110, 1558 – 1595; Angew. Chem. Int. Ed. 1998, 37, 1460 – 1494;
b) O. M. Yaghi, M. OꢀKeeffe, N. W. Ockwig, H. K. Chae, M.
Eddaoudi, J. Kim, Nature 2003, 423, 705 – 714; c) S. Kitagawa, R.
Kitaura, S. Noro, Angew. Chem. 2004, 116, 2388 – 2430; Angew.
Chem. Int. Ed. 2004, 43, 2334 – 2375; d) M. Kawano, M. Fujita,
Coord. Chem. Rev. 2007, 251, 2592 – 2605; e) G. Fꢁrey, Chem. Soc.
Rev. 2008, 37, 191 – 214; f) B. Moulton, M. J. Zaworotko, Chem.
Rev. 2001, 101, 1629 – 1658.
[2] Examples of crystalline-state reactions within porous coordina-
tion networks in which substrates or catalysts are embedded
before the reaction: a) Z. Wang, S. M. Cohen, Chem. Soc. Rev.
2009, 38, 1315 – 1329; b) Y. Inokuma, M. Kawano, M. Fujita, Nat.
Chem. 2011, 3, 349 – 358; c) J. S. Seo, D. Whang, H. Lee, S. I. Jun,
J. Oh, Y. J. Jeon, K. Kim, Nature 2000, 404, 982 – 986; d) L. Ma, C.
Abney, W. B. Lin, Chem. Soc. Rev. 2009, 38, 1248 – 1256; e) T.
Kawamichi, T. Haneda, M. Kawano, M. Fujita, Nature 2009, 461,
633 – 635; f) S. C. Jones, C. A. Bauer, J. Am. Chem. Soc. 2009, 131,
12516 – 12517.
[8] Nonselective inclusion of 4a and 4b into network 1 was confirmed
with crystals of 1ꢀ3: when coordination network 1ꢀ3 was soaked
in a 1:1 solution of 4a and 4b, the system reached equilibrium
within 1 h, and more than 80% of the substrates was enclathrated
in almost 1:1 ratio.
[3] a) K. Ohara, Y. Inokuma, M. Fujita, Angew. Chem. 2010, 122,
5639 – 5641; Angew. Chem. Int. Ed. 2010, 49, 5507 – 5509; b) T.
Uemura, K. Kitagawa, S. Horike, T. Kawamura, S. Kitagawa, M.
Mizuno, K. Endo, Chem. Commun. 2005, 5968 – 5970; c) S. Dawn,
Angew. Chem. Int. Ed. 2012, 51, 2379 –2381
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
2381