Multiple weak supramolecular interactions stabilize a surprisingly twisted As2L3 assembly

Article abstract of DOI:10.1039/b806958a

A combined crystallographic, DFT and NMR spectroscopic study of a flexible As₂L₃ assembly reveals temperature dependent conformational behavior in solution and a highly asymmetric structure stabilized by As-π and edge-to-face aromati

Full text of DOI:10.1039/b806958a

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Multiple weak supramolecular interactions stabilize a surprisingly twisted
As2L3 assemblyw
b
Melanie A. Pitt,a Lev N. Zakharov,a Kumar Vanka,z Ward H. Thompson,b
Brian B. Lairdb and Darren W. Johnson*a
Received (in Austin, TX, USA) 24th April 2008, Accepted 22nd May 2008
First published as an Advance Article on the web 9th July 2008
DOI: 10.1039/b806958a
A combined crystallographic, DFT and NMR spectroscopic study
of a flexible As2L3 assembly reveals temperature dependent con-
formational behavior in solution and a highly asymmetric structure
stabilized by As–p and edge-to-face aromatic interactions.
Supramolecular chemistry based on metal–ligand coordina-
tion is a rapidly advancing field that has provided access to a
wide variety of structures through information programmed
into the coordination preferences of metal ions.1,2 These
coordination assemblies frequently display emergent proper-
ties such as guest encapsulation,3 with some metal–ligand
assemblies exhibiting changes in size through a non-dissocia-
tive mechanism that has been described as ‘‘breathing’’.4
Interest in main-group supramolecular chemistry has flourished
in recent years, as the main group elements provide access to
unique coordination geometries otherwise unavailable to re-
searchers preparing novel structures with new properties.5 In this
Scheme 1 Synthesis of As2L3 assembly; the included numbering
scheme will be used later in discussing the structure.
communication we report a novel arsenic-based supramolecular
assembly whose structure is determined by a combination of
As–p6 and edge-to-face aromatic interactions. The NMR spectra
of this assembly are highly temperature dependent, suggesting
that the ligand scaffold expands and contracts in response to
temperature changes, in effect ‘‘breathing.’’
of flexibility.12 The corresponding As2L3 assembly was then
prepared (Scheme 1) by slow addition of AsCl3 in benzene or
THF to a solution of the ligand and KOH in methanol.13
Slow evaporation of either chloroform or dichloromethane
solutions of As2L3 yielded clear, colorless blocks whose identity
was confirmed through X-ray crystallography.y The As2L3 struc-
ture (Fig. 1a) features four close contacts of varying distances
between As(III) and the nearest carbon on the phenyl rings of
Previous work in this laboratory has focused on the prepara-
tion of dinuclear arsenic assemblies bridged by rigid ligands, such
as 1,4-dimercaptomethylbenzene, which accommodates the pre-
ference of trivalent arsenic for trigonal pyramidal coordination.7,8
This assembly was found to be thermodynamically and kinetically
stable under a variety of harsh conditions and in the presence of
competing metal ions and ligands, presumably due to the strength
of the As–S bonds (B81 kcal molꢀ1).9 Furthermore, the endo-
hedrally-directed lone pairs and short As–p contacts suggested the
potential for unique host–guest interactions.10
A longer ligand containing two phenyl rings, 4,40-dimercapto-
methyldiphenylmethane (H2L), was prepared through modifica-
tions of literature methods.11 Diphenylmethane was chosen as a
spacer due to its ubiquity in supramolecular systems—it provides
an expanded cavity to host structures and has a moderate degree
a Department of Chemistry and Materials Science Institute, University
of Oregon, Eugene, OR 97403-1253 and The Oregon Nanoscience
and Microtechnologies Institute (ONAMI) USA. E-mail:
dwj@uoregon.edu
b Department of Chemistry, University of Kansas, Lawrence, KS
66045, USA
w Electronic supplementary information (ESI) available: Crystallo-
graphic data and NMR spectral data. CCDC 685997. For ESI and
crystallographic data in CIF format, see DOI: 10.1039/b806958a
z Current address: Physical Chemistry Division, National Chemical
Laboratory, Pune 411008, India
Fig. 1 Three-dimensional structure of As2L3. (a) Crystal structure shown
as wireframe; spheres represent arsenic atoms. (b) Crystal structure,
looking down Asꢁ ꢁ ꢁAs axis. Note the D axial chirality (clockwise twist)
of As(III) facing the viewer. (c) Cutaway view of two lower strands
showing centroid–centroid distances. (d) Schematic of equilibrium be-
tween skew (crowded) and gable (remote) geometries in diphenylmethane.
ꢂc
This journal is The Royal Society of Chemistry 2008
3936 | Chem. Commun., 2008, 3936–3938

View Article Online
ligand strands L1 and L2. This observation is in agreement with
calculations showing that the geometry of the As–p interaction is
strong (at least 7.4 kcal molꢀ1) yet has variable directionality.7
The solid-state structure of this complex is surprisingly distorted
in comparison to geometries predicted by MM3 calculations and
the majority of self-assembled metal–ligand complexes that ex-
hibit a high degree of symmetry in both solution and the solid
state,2 including those containing the diphenylmethane spacer.14
Instead, the three ligands are unevenly distributed about the metal
centers—an arrangement best described as a distorted mesocate,
as the –CH2S– groups maintain opposite twists about the two
metal centers. For the purposes of this study, the counterclock-
wise twisted As(III) center will be described as L, while the
clockwise As(III) center will be described as D. We will therefore
refer to the As2L3 assembly as a DL mesocate; since the space
group P21/n is centrosymmetric, equal amounts of the structure
shown in Fig. 1 and its mirror image are present. In the view
along the Asꢁ ꢁ ꢁAs axis in Fig. 1b, the D twist faces the reader. The
angles between the mean planes of the phenyl rings about the D
metal center are 48.5(1), 49.8(1), and 85.2(1)1, while those about
the L center are 35.1(1), 46.7(1), and 80.9(1)1. The rings are
twisted so as to direct the edges toward the interior of the
complex, effectively filling any empty space inside.
Fig. 2 (a) Overlay of DFT optimized geometry (black) and crystal
structure (gray) of As2L3. (b) Overlay of higher-energy DFT geometry
(blue) and crystal structure.
from solution with APCI-MS. The 1H-NMR spectrum is surpris-
ingly uncomplicated at room temperature—singlets are observed
for both H1 and H6 (Fig. 3; see Scheme 1 for numbering). In
addition, there is an accidental chemical shift degeneracy for H3
and H4 as evidenced by the presence of a singlet resonance in the
aromatic region instead of the expected set of two doublets. The
13C-NMR spectrum showed a similar degeneracy for C3 and C4.
These uncomplicated spectra indicate dynamic solution beha-
vior on the NMR timescale. The simple aromatic region indicates
that the phenyl rings rotate rapidly, and the singlet observed for
H6 shows that the solution structure cannot be a static DL
mesocate at the As(III) centers. There must therefore be fast
interconversion between D1L2 and L1D2 isomers on the NMR
timescale, either by simultaneous interconversion or through a
D1D2 or L1L2 intermediate with a lifetime too short to observe.
In order to resolve the discrepancy between the solid and
solution phase data, variable temperature NMR experiments
were carried out. No resolution of either set of methylene protons
was observed even at ꢀ80 1C (in contrast to other diphenyl-
methane bridged supramolecular systems14), but a strong upfield
shift was observed for one aromatic proton, lifting the accidental
degeneracy of H3 and H4. H1 also displayed a noticeable upfield
shift as the temperature was decreased, consistent with the weak
contact with a neighboring aromatic ring observed in the crystal
structure. The accidental degeneracy in carbon signals C3 and C4
was also lifted as the temperature decreased. Returning the
sample to room temperature reversed the chemical shift change
and returned the aromatic protons to their coincidentally degen-
erate state. In contrast, VT-NMR experiments with the original
phenylene spaced assembly7 showed no appreciable changes at
lowered temperature. This is a strong indication that the changes
in the NMR spectra are related to changes in conformation along
the diphenylmethane spacer rather than activity about the As(III)
centers.
Calculations and long-range NMR couplings have shown that
there are two predominant solution conformations of diphenyl-
methane, defined by the pair of dihedral angles relating the
bridging methylene protons and the ortho aromatic protons. These
conformations have been described as skew and gable15 (Fig. 1d)
and are thought to exist in dynamic equilibrium in solution.
Crystalline diphenylmethane, however, is exclusively skew—this
conformation suggests an attractive intramolecular edge-to-face
aromatic interaction. Two ligand strands (L1 and L2) exhibit this
type of interaction in which the centroids of the rings are near the
ideal offsetz of 5.025 A,16 (Fig. 1c) in nearly perpendicular skew
conformations. The third ligand strand (L3) is also skew, but not
to the same degree as the other two strands. This is a relatively rare
example of an ‘‘imploded’’ supramolecular structure similar to the
imploded cryptophane observed by Holman et al.,17 which con-
tracts in the absence of an appropriate guest.
DFT optimization of As2L3 (with a 6-31+G* basis set for all
atoms and the B3LYP functional18) produces two remarkably
similar minima which differ in energy by only 4.1 kcal molꢀ1 (Fig.
2a and b). Neither structure maintains the high degree of sym-
metry predicted by MM3 calculations. Instead, the phenyl rings
are twisted such that the edges point toward the interior of the
complex, effectively filling any empty interior space. These calcu-
lations predict the ‘‘meso’’ As2L3 crystal structure with a surprising
degree of accuracy: the Asꢁ ꢁ ꢁAs distance is underestimated by
only 0.11 A, while the average As–p distance is overestimated by
0.19 A (Fig. 2a). The calculated structure places L1 and L2 in
nearly perpendicular skew arrangements with centroid offsets of
4.96 and 5.01 A, in nearly perfect agreement with the observed
distances. The strong agreement between the calculated structure
and the crystal structure is a powerful indicator that the distorted
structure is not merely an artifact of crystal packing forces, but is
instead the result of a combination of supramolecular interac-
tions.19
The identity of the upfield shifted aromatic proton was con-
firmed as H3 by low-temperature HMBC and HMQC spectra
As2L3 is remarkably stable in solution: no ligand exchange is
observed with free ligand and the intact assembly can be ionized
Fig. 3 VT-NMR of As2L3 in CD2Cl2. Spectra are referenced to
residual solvent signal and are taken at 10 1C intervals.
ꢂc
This journal is The Royal Society of Chemistry 2008
Chem. Commun., 2008, 3936–3938 | 3937

View Article Online
(see ESIw). Similar flexible molecules capable of intramolecular
edge-to-face aromatic interactions are known to interconvert
rapidly between ‘‘crowded’’ and ‘‘remote’’ limiting geometries
on the NMR timescale at experimentally accessible temperatures.
In that analysis, lower temperatures favor a more crowded
geometry, enhancing edge-to-face aromatic interactions and a
contracted structure. At higher temperatures, entropy apparently
disfavors the edge-to-face interaction, promoting a more remote
geometry and an expanded structure; intermediate conformations
would be expected over the observed temperature range. In either
case, the conformations exist in a fast equilibrium; the observed
NMR shift is a weighted average of the populations of all
conformers present.20 For As2L3, the remote geometry corres-
ponds to a gable-like conformation of the diphenylmethane
spacer at room temperature, while the crowded geometry ob-
served at lowered temperatures corresponds to a more skew-like
conformation. As the structure of more expanded conformations
is not yet understood, it would be premature to calculate
thermodynamic parameters from this data.
1 S. Leininger, B. Olenyuk and P. J. Stang, Chem. Rev., 2000, 100,
853.
2 D. L. Caulder and K. N. Raymond, J. Chem. Soc., Dalton Trans.,
1999, 1185.
3 F. Hof, C. Nuckolls, J. Rebek, Jr and S. L. Craig, Angew. Chem.,
Int. Ed., 2002, 41, 1488.
4 A. V. Davis, R. M. Yeh and K. N. Raymond, Proc. Natl. Acad.
Sci. U. S. A., 2002, 99, 4793; M. D. Pluth and K. N. Raymond,
Chem. Soc. Rev., 2007, 36, 161.
5 M. A. Pitt and D. W. Johnson, Chem. Soc. Rev., 2007, 36, 1441.
6 H. Schmidbaur, W. Bublak, B. Huber and G. Muller, Angew.
Chem., 1987, 99, 248; H. Schmidbaur, R. Nowak, O. Steigelmann
and G. Muller, Chem. Ber., 1990, 123, 1221.
7 W. J. Vickaryous, R. Herges and D. W. Johnson, Angew. Chem.,
Int. Ed., 2004, 43, 5831.
8 W. J. Vickaryous, E. R. Healey, O. B. Berryman and D. W.
Johnson, Inorg. Chem., 2005, 44, 9247; T. A. Shaikh, S. Parkin
and D. A. Atwood, J. Organomet. Chem., 2006, 691, 4167.
9 This value is estimated from As–S bond strengths in AsxSy poly-
hedra: D. Babıc, S. Rabii and J. Bernholc, Phys. Rev. B: Condens.
Matter Mater. Phys., 1989, 39, 10831. Bond strengths are known
for other As–E bonds but are not reported for E = S in Chemistry
of Arsenic Antimony and Bismuth, ed. N. C. Norman, Blackie
Academic & Professional, London, 1988.
10 R. W. Alder and S. P. East, Chem. Rev., 1996, 96, 2097; V. M.
Cangelosi, L. N. Zakharov, S. T. Fontenot, M. A. Pitt and D. W.
Johnson, Dalton Trans., 2008, 3447.
11 D. J. Cram and H. Steinberg, J. Am. Chem. Soc., 1951, 73, 5691; H.
Steinberg and D. J. Cram, J. Am. Chem. Soc., 1952, 74, 5388.
12 J. W. Steed and J. L. Atwood, Supramolecular Chemistry, Wiley,
Chichester, 2000.
13 As2L3: degassed KOH (0.1023 N in MeOH) was added to H2L
under N2. The solution was heated to 50 1C and AsCl3 in benzene
was added dropwise. Heating was continued for 3 h; the solution
was then filtered to remove white precipitate. The solvent was
removed, yielding 12% of the desired product. 1H-NMR (500
MHz; CD2Cl2; ꢀ20 1C): d 7.03 (d, J = 7.5 Hz, 12 H), 6.98 (d, J
= 7.5 Hz, 12 H), 4.01 (s, 6H), 3.62 (s, 12H). 13C-NMR (126 MHz;
CD2Cl2; ꢀ20 1C): d 140.45, 138.00, 129.71, 129.56, 41.28, 35.94;
m/z (APCI) found 924.8 (As2L3+, C45H42As2S6 requires 924.00).
14 M. Albrecht and R. Frohlich, Bull. Chem. Soc. Jpn., 2007, 80, 797.
15 T. Schaefer, W. Niemczura, W. Danchura and T. A. Wildman, Can.
J. Chem., 1979, 57, 1881; T. Straßner, Can. J. Chem., 1997, 75, 1011.
16 E. A. Meyer, R. K. Castellano and F. Diederich, Angew. Chem.,
Int. Ed., 2003, 42, 1210.
In summary, weak forces such as edge-to-face aromatic and
arsenic–p interactions play an important role in determining the
solution and solid-state structures of supramolecular complexes.
In particular, the arsenic–p interaction appears to persist in the
presence of flexible ligand scaffolds, showing that it is an im-
portant component of arsenic ligand design and can itself be
considered an emerging supramolecular interaction. The varying
strength and geometry in the crystal structure and the tempera-
ture dependence of these weak aromatic interactions in solution
have been shown to dramatically affect the structure of a new
As2L3 assembly with a flexible ligand scaffold. In particular, we
have observed expansion and contraction or ‘‘breathing’’ of the
host in solution over a range of temperatures. This may have
significant implications in the design of expanded arsenic-contain-
ing host structures—the addition of conformationally mobile
spacer units may play a role in determining the size, shape, and
exchange mechanism of guests.
17 S. T. Mough, J. C. Goeltz and K. T. Holman, Angew. Chem., Int.
Ed., 2004, 43, 5631.
The University of Oregon and an NSF-CAREER award
(CHE-0545206) are gratefully acknowledged for their support.
D.W.J is a Cottrell Scholar of Research Corporation. M.A.P
acknowledges the National Science Foundation (NSF) for an
Integrative Graduate Education and Research Traineeship. K.V.,
W.H.T. and B.B.L. gratefully acknowledge support from the
National Science Foundation through a SBIR/STTY supplement
to grant EEC-0310689 providing for the University of Kansas
Center for Environmentally Beneficial Catalysis.
18 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A.
Robb, J. R. Cheeseman, J. A. Montgomery, T. Vreven, K. N.
Kudin, J. C. Burant, J. M. Millam, S. S. Iyengar, J. Tomasi, V.
Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G. A.
Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R.
Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O.
Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian, J.
B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E.
Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J.
Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth, P. Salvador, J.
J. Dannenberg, V. G. Zakrzewski, S. Dapprich, A. D. Daniels, M.
C. Strain, O. Farkas, D. K. Malick, A. D. Rabuck, K. Raghava-
chari, J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S.
Clifford, J. Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko, P.
Piskorz, I. Komaromi, R. L. Martin, D. J. Fox, T. Keith, M. A.
Al-Laham, C. Y. Peng, A. Nanayakkara, M. Challacombe, P. M.
W. Gill, B. G. Johnson, W. Chen, M. W. Wong, C. Gonzalez and
J. A. Pople, GAUSSIAN 03 (Revision C.02), Gaussian, Inc.,
Wallingford, CT, 2004.
Notes and references
y Crystal data for As2L3: C45H42As2S6, Mr = 942.99, 0.18 ꢃ 0.14 ꢃ
0.08 mm, monoclinic, P21/n, a = 14.5184(16) A, b = 12.9661(14) A,
c = 23.094(3) A, b = 106.773(2)1, V = 4162.3(8) A3, Z = 4, rcalcd
=
1.476 g cmꢀ3, m = 1.939 mmꢀ1, 2ymax = 54.001, T = 173(2) K, 37 996
measured reflections, 9082 independent reflections [Rint = 0.0644], 478
parameters, R1 and wR2 = 0.0414 and 0.0835 (I 4 2s(I)); 0.0678 and
0.0954 (all), GOF = 1.013 for all 9082 reflections, max/min residual
electron density +0.483/ꢀ0.340 e Aꢀ3. CCDC 685997.
19 It is important to note that the weak edge-to-face aromatic
interactions evidenced in both the DFT geometries and crystal
structures are likely not dispersion-based CH–p interactions, since
DFT does not properly describe such forces; see, e.g.: M. O.
Sinnokrot and C. D. Sherrill, J. Phys. Chem. A, 2006, 110, 10656.
20 W. B. Jennings, B. M. Farrell and J. F. Malone, Acc. Chem. Res.,
2001, 34, 885.
z There are two common methods for quantifying aromatic edge-to-
face contacts. Meyer et al.16 refer to centroidꢁ ꢁ ꢁcentroid distances,
while Jennings et al.20 refer to Ar–Hꢁ ꢁ ꢁcentroid distances. We utilize
the former method.
ꢂc
This journal is The Royal Society of Chemistry 2008
3938 | Chem. Commun., 2008, 3936–3938