Crystallographic snapshots of the bond-breaking isomerization reactions involving nickel(II) complexes with hemilabile ligands

Article abstract of DOI:10.1002/anie.201107620

Freeze-frame: Octahedral and square-planar structural isomers, representing the two "end states" in a hemilabile ligand bond-breaking isomerization reaction, have been characterized in solution by spectroscopic methods and in the solid state by X-ray crystallography (see picture: Nigreen, Cgray, Porange, Nblue, Syellow). Copyright

Full text of DOI:10.1002/anie.201107620

Angewandte
Chemie
DOI: 10.1002/anie.201107620
Coordination Chemistry
Crystallographic Snapshots of the Bond-Breaking Isomerization
Reactions Involving Nickel(II) Complexes with Hemilabile Ligands**
Charles W. Machan, Alejo M. Lifschitz, Charlotte L. Stern, Amy A. Sarjeant, and
Chad A. Mirkin*
Hemilabile ligands have been used to prepare a wide variety
of complexes, which are important in many fields ranging
from catalysis to biomimetic chemistry.[1–5] These ligands
allow coordination complexes to be prepared where one end
of the hemilabile ligand is anchored to a metal center, while
the other often is involved in fluxional processes. Indeed, in
certain cases, researchers refer to these structures as “wind-
shield-wiper” ligands,[6–9] since the weakly binding portion of
the ligand can often dissociate and recoordinate to the metal
Scheme 1. Synthesis of complexes 2a–c. a) 2 equiv NaSCN in 1:1
center, undergoing exchange in the presence of a coordinating
solvent. Consequently, many researchers employ hemilabile
ligands to stabilize what effectively are highly reactive solvent
adducts or coordinatively unsaturated species.[10–20] As a
result, complexes with certain types of hemilabile ligands
have been made and characterized in both monodentate and
multidentate states with different ancillary ligands.[21–26] To
our knowledge, however, there is no example where a pair of
isomeric structures, which represent the two “end states” that
define a hemilabile ligand exchange process (where no
displacement reaction has taken place), have been crystallo-
graphically characterized. The reason for this observation is
that invariably in such systems one state is significantly more
stable than the other in a given environment, or the exchange
process occurs dynamically under rapid exchange conditions
and the identities of the complexes exchanging can only be
inferred through in situ spectroscopic means. Herein, we
report the synthesis, isolation, and crystallographic character-
ization of a series of octahedral and square planar structurally
isomeric nickel(II) complexes 2a–2c,[27] which effectively
define the two key “end states” in a well-defined bond-
breaking isomerization reaction and a likely intermediate
complex. While these structures are stable in the solid state,
they rapidly interconvert in solution at room temperature.
The reaction of complex 1 with two equivalents of sodium
thiocyanate (NaSCN) in a mixture of CH2Cl2 and EtOH
yields a mixture of complexes 2a–2c (Scheme 1). In solution
at room temperature, these complexes are rapidly intercon-
CH2Cl2/EtOH for 1 h.
verting, as evidenced by a single broad 31P{1H} NMR reso-
nance observed at d = 32.6 in CD2Cl2. Variable-temperature
(VT) 31P{1H} NMR spectroscopy confirmed that this broad-
ening is a result of rapid exchange between multiple
complexes. When the temperature of the solution was
lowered to 213 K in a stepwise manner, the spectrum changed
significantly. At 238 K, a sharp resonance at d = 32.6 ppm
assigned to 2a is observed along with a broad one at d =
8.7 ppm assigned to 2c (Supporting Information, Figure S1).
The assignments for these resonances are based upon
literature precedent[28,29] involving model complexes and
crystallographic data (see below).
Interestingly, when red-orange CH2Cl2 solutions contain-
ing these rapidly interconverting complexes were layered with
diethyl ether, both purple and orange block-like crystals
formed, along with red plates. Single-crystal X-ray diffraction
studies of the purple, orange, and red crystals showed that
they were 2a, 2b, and 2c, respectively (Figure 1).[30] The
structure of 2a consists of a nickel(II) center in an octahedral
À
geometry, defined by two chelating P,S Me ligands coordi-
nated in a trans configuration to form the equatorial plane.
Furthermore, it has two apical nitrogen-bound SCNÀ ligands,
which are bent relative to the equatorial plane (150–1608;
Figure 1a and Supporting Information, Table S1). The struc-
[*] C. W. Machan, A. M. Lifschitz, C. L. Stern, Dr. A. A. Sarjeant,
Prof. Dr. C. A. Mirkin
Department of Chemistry, Northwestern University
2145 Sheridan Rd, Evanston, IL 60208 (USA)
E-mail: chadnano@northwestern.edu
Homepage: http://chemgroups.northwestern.edu/mirkingroup/
[**] We thank Northwestern IMSERC for analytical instrumentation. We
are grateful to Prof. Dr. Gregory L. Hillhouse for helpful suggestions
concerning this manuscript. C.A.M. is grateful to NSF, ARO, and
AFOSR-MURI for generous funding.
Figure 1. X-ray crystallographically determined structures for com-
plexes 2a’’ (a), 2b’’ (b), and 2c (c). Ni green, C gray, P orange, N blue,
S yellow. Ellipsoids set at 50% probability; hydrogen atoms omitted
for clarity.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201107620.
Angew. Chem. Int. Ed. 2012, 51, 1469 –1472
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1469

.
Angewandte
Communications
ture of 2b is nearly identical to complex 2a, with the
exception of the orientation of the apical SCNÀ ligands, which
are bound to nickel perpendicular to the equatorial plane
formed by the P,S chelates (ca. 1708, Figure 1b). The structure
of 2c is square planar in geometry around the nickel center
richness of the thioether,[6] this ligand should favor an “open”
monodentate coordination mode. Indeed, complex 4a was
synthesized from a complex (3a) prepared from NiCl2·6H2O
and two equivalents of Ph2PCH2CH2SPh under similar
conditions to those used to prepare 2a–c (see the Supporting
Information). As well as being characterized in the solid state
by a single-crystal X-ray diffraction study (Figure 2a), 4a was
À
with two monodentate P,S Me ligands (bound only through
P) and two nitrogen-bound SCNÀ ligands in a near-linear
configuration with nickel (ca. 1708; Figure 1c, Supporting
Information, Table S1).
1
characterized in solution by 31P{1H}, H, and 13C{1H} NMR
spectroscopy.[30] All data are consistent with the solution and
solid-state structures being very similar.[28,29]
Polymorphism, linkage isomerism, and the ability of
coordination complexes to crystallize as different geometric
isomers under identical conditions have been reported
previously,[21,23–26] but to our knowledge there are no examples
of the solid-state characterization of some of the key
participants in a hemilabile ligand exchange process, espe-
cially the dissociated state where no displacement reaction by
solvent molecule or coordinating ligand has taken place.
These structures are most aptly described as “bond-breakage
isomers” rather than polymorphs. The observation of their
existence is also dissimilar to bond-stretch isomerism, since
the complexes in question (2a, 2b, and 2c) differ in ligand–
metal connectivity.[22,25b,31,32] Linkage isomerism can also
result in isomerization reactions that may be characterized
in the solid state, but these products differ in ligand
connectivity and not coordination number.[26] The fact that
these crystal structures have “open” and “closed” coordina-
tion sites under the same solvent conditions lends credence to
the conclusion that they represent crystallographic snapshots
of an in situ bond-breaking isomerization reaction (see
below).
To fully characterize this system, we studied the crystalline
products by FTIR spectroscopy, focusing on the SCNÀ regions
since this functional group exhibits highly diagnostic stretches
dependent upon coordination mode to the metal center.[33]
Interestingly, examination of the bulk powder consisting of
2a–c with an optical microscope shows a microcrystalline
powder that contains the same three crystal types obtained by
solvent layering (Supporting Information, Figure S2), with
the majority product being 2a. Owing to difficulties associ-
ated with the mechanical separation of 2c from the other
components of this mixture (2a and 2b), we were unable to
isolate adequate quantities of 2c for study by FTIR spectros-
copy. Since we could obtain macroscopic quantities of crystals
of pure 2a and 2b, respectively, we could measure their FTIR
spectra. Each exhibits a nCN stretch diagnostic of non-bridging
N-bound thiocyanate (Supporting Information, Figure S3,
Table S2).[33,34] Importantly, when these crystals of 2a or 2b
were dissolved in CD2Cl2 and characterized by 31P{1H} NMR
spectroscopy, only a single broad resonance at d = 32.6 ppm
was observed, which is consistent with the 31P{1H} NMR data
for the mixture of isomers in solution. Based on the
characterization of three distinct isomers in the solid state,
we decided to determine if the three structures exist in
solution as well. To do so, it was necessary to synthesize two
analogue complexes, 4a and 4b, which have weakly chelating
and non-chelating phosphine ligands, respectively, and can
provide representative and meaningful comparative spectro-
scopic data (see Scheme 2). Since substitution of the methyl
group on the thioether with a phenyl ring reduces the electron
Figure 2. X-ray crystallographically determined structures for com-
plexes 4a (a) and 4b (b). Ni green, C gray, P orange, N blue, S yellow.
Ellipsoids set at 50% probability; hydrogen atoms omitted for clarity.
The solid-state structure of 4a consists of a square-planar
nickel center with two equivalents of N-bound SCNÀ ligands
À
and two equivalents of P-bound monodentate P,S Ph ligands
arranged in trans configurations. Importantly, the room
temperature 31P{1H} NMR spectrum of 4a in CD2Cl2 exhibits
a broad resonance at d = 10.9 ppm, which sharpens with
decreasing temperature (Supporting Information, Fig-
ure S1B), consistent with a monodentate coordination mode
for the phosphine ligand.[28] Similarly, model compound 4b,
which contains a non-chelating phosphine ligand, was synthe-
sized from complex 3b under nearly identical chloride
abstraction conditions to those used for 3a (see the Support-
ing Information; Scheme 2, Figure 2a). As in the case of 4a, a
1
single-crystal X-ray diffraction study of 4b and 31P{1H}, H,
and 13C{1H} NMR spectroscopy indicated similar structures in
solution and the solid state (Figure 2b).[30] The solid-state
structure of 4b showed a square-planar nickel center, much
like 4a, with two equivalents of N-bound SCNÀ ligands and
two equivalents of the non-chelating phosphine each
arranged in a trans configuration. Significantly, room-temper-
ature 31P{1H} NMR spectroscopy of 4b in CD2Cl2 showed a
single sharp resonance at d = 16.7 ppm that remains
unchanged with decreasing temperature. This complex,
Scheme 2. Synthesis of complexes 4a and 4b from 3a and 3b.
a) 2 equiv NaSCN in 1:1 CH2Cl2/EtOH for 1 h.
1470
www.angewandte.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 1469 –1472

Angewandte
Chemie
which does not have thioethers, does not undergo the
fluxional exchange processes observed with 4a and 2a–c.
The chemical shift of the resonance is similar to that observed
for 4a, consistent with the assignment of this complex as one
with monodentate phosphinoalkylthioether ligands (Support-
ing Information, Figure S1B). Based on the spectroscopic
data of model complex 4a, we assign the 31P{1H} NMR
resonance observed at d = 8.7 ppm in the mixture of 2a–c at
low temperature in CD2Cl2 to the “open structure” 2c.
Compound 1 (which has two chloride ligands) serves as a
model complex for octahedral compounds with trans phos-
phine and trans thioether ligands, such as 2a and 2b. Indeed,
in CD2Cl2, complex 1[29] exhibits a downfield resonance at d =
32 ppm in its 31P{1H} NMR spectrum, which is highly
diagnostic of structures with five-membered P-containing
chelates and remarkably close to the second resonance at d =
32.6 ppm observed for the mixture of 2a–c at low temper-
ature.[28] We have not observed a third resonance at low
temperature, and therefore speculate that 2a and 2b rapidly
interconvert, even at 213 K on the NMR spectroscopic
timescale. Indeed, the interconversion between the two only
requires a reversible linear to bent transformation for the
SCNÀ ligands. Similar behavior is observed in the 13C{1H}
NMR spectra of the mixture of 2a–c, where only two
resonances (d = 129.8 ppm and d = 129.3 ppm) are observed
for the carbon atoms that comprise the SCNÀ ligands, even at
208 K.
We propose that 2b is an intermediate in the intercon-
version of 2a and 2c (Scheme 1). The basis for this conclusion
comes from analysis of the solid state X-ray crystallographic
data, the solution NMR data for the complexes, and DFT
calculations (Supporting Information, Figures S6–S8,
Table S3). Such calculations suggest that 2b is higher in
energy than 2a by about 13 kcalmolÀ1 and therefore can serve
as intermediate in the interconversion of 2a and 2c, which is
only about 6.6 kcalmolÀ1 higher in energy than 2a (Support-
ing Information, Table S3). Indeed, the single-crystal X-ray
data shows longer nickel–thioether bond distances for 2b
ing Information, Figure S9). From these data, one can
conclude that the reaction is enthalpically disfavored
(DH8 = 2.8 kcalmolÀ1) as written but entropically favored
(DS8 = 11.5 calmolÀ1KÀ1; Scheme 1). This conclusion is con-
sistent with the conversion of a relatively rigid octahedral
complex into a more flexible square-planar isomer through
À
two Ni S bond-breaking processes. At 298 K, the calculated
DG8 of this reaction is À0.6 kcalmolÀ1. The fact that the major
product observed when solvent is removed is 2a can be
attributed to the decreased solubility of this product causing a
correspondingly more rapid precipitation as the solution
becomes more concentrated. The stability of all three species,
however, is likely in part the consequence of an energetic
barrier between the “open” and “closed” type complexes
because of their respective low- and high-spin configura-
tions.[31,32] In fact, Evans method measurements of the meff
values for 2a–c at 304 K are equal to 2.18 BM (a weighted
average of S = 1 for 2a–b and S = 0 for 2c), indicating a
distribution of both low- and high-spin nickel(II), which is
consistent with the broad resonance observed in the 31P{1H}
NMR spectrum in CD2Cl2 of the mixture of 2a–c.[35,36] Cooling
the sample from room temperature to 195 K shows a linear
increase in meff values to 3.14 BM (S = 1) with a concomitant
shift of product distribution towards high-spin, octahedral
nickel(II) consistent with previously discussed 31P{1H} NMR
spectroscopy (Supporting Information, Figures S1 and S9).
This result is intuitive based upon the temperature-dependent
equilibrium involving high-spin complexes 2a and 2b and
low-spin complex 2c. At high temperatures, the equilibrium
mixture favors 2c, decreasing the observed meff (Scheme 1).
Both 4a and 4b were determined to be diamagnetic by the
same procedure.
This system is intriguing, since we can isolate three
isomers and structurally characterize them. Clearly the
barrier to interconversion in the solid state is much greater
than in solution, where 2a and 2b undergo rapid exchange
even at 208 K. This phenomenon seems to be unique to these
types of hemilabile ligand complexes formed with SCNÀ
ligands. Indeed, the ClÀ versions of these complexes (for
example 1 and 3a) also undergo dynamic exchange in solution
(but such processes involve the ClÀ ligand moving from the
inner to outer coordination sphere); however, only one
isomer is ever observed in the solid state.[29] The ability of the
thiocyanate ligand to adopt multiple coordination modes
seems to provide pathways to not only weaken other bonds
within the complex but also stabilize other isomers, including
the four-coordinate square-planar complex 2c.
À
than 2a (ca. 0.05 ꢀ, Table S1), suggesting weaker Ni S bonds.
With this hypothesis, when the thiocyanate ligands rearrange
from bent to linear coordination modes in the conversion of
À
2a to 2b, a concomitant weakening of the Ni S bonds occurs,
preparing for the formation of open complex 2c. Consistent
with the conclusion that 2b is the intermediate rather than 2a,
DFT calculations suggest greater ring strain in its five-
membered chelates (the ligand conformation in 2b is higher
in energy than that of 2a and 2c by about 3.3 and
4.0 kcalmolÀ1,
respectively;
Supporting
Information,
The reason why the interconversion of complexes 2a and
2c does not involve the dissociation of the isocyanate ligand
similar to the ClÀ dissociation event involving 1 most likely
Table S3). Interestingly, the SOMO orbitals of the high-spin
2a are partially located on the metal center, but those of 2b
are predominantly centered on the SCNÀ ligands with little
nickel center contribution, suggesting that the delocalized
electronic character of the metal–ligand interaction may be
important to the exchange mechanism (Supporting Informa-
tion, Figures S6,S7). Complex 2c is calculated to be low spin,
consistent with a square-planar nickel(II) complex.
À
À
involves the higher Ni N bond strength as compared to Ni
Cl. N-bound SCNÀ is a much stronger field ligand than ClÀ
according to the spectrochemical series.[37] Consistent with
this conclusion, the UV/Vis data show a shift in the d–d
transition from 512 nm in 1 to 485 nm in 2a–c (Supporting
Information, Figure S11), which is indicative of a higher-
energy transition. Therefore, the only weak metal–ligand
The thermodynamic parameters for the interconversion of
2a and 2c can be determined by the construction of a Vanꢁt
Hoff plot from the VT 31P NMR spectroscopic data (Support-
À
interactions in this system are the Ni S bonds.
Angew. Chem. Int. Ed. 2012, 51, 1469 –1472
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
1471

.
Angewandte
Communications
[18] C. W. Rogers, M. O. Wolf, Coord. Chem. Rev. 2002, 233 – 234,
341.
[19] Z. Weng, S. Teo, T. S. A. Hor, Acc. Chem. Res. 2007, 40, 676.
[20] T. J. M. de Bruin, L. Magna, P. Raybaud, H. Toulhoat, Organo-
metallics 2003, 22, 3404.
[21] B. M. Foxman, P. L. Goldberg, H. Mazurek, Inorg. Chem. 1981,
20, 4368.
[22] G. Parkin, Chem. Rev. 1993, 93, 887.
These computational and experimental results are con-
sistent with the observed properties of these complexes being
À
directly affected by the presence and nature of Ni S(R)2 and
À
À
Ni SCN interactions both in solution and the solid state.
This interplay has allowed for the unprecedented solid-state
characterization of some important species from a bond-
breaking isomerization exchange.
[23] J. Bernstein, R. J. Davey, J.-O. Henck, Angew. Chem. 1999, 111,
3646; Angew. Chem. Int. Ed. 1999, 38, 3440.
Received: October 28, 2011
[24] D. Braga, F. Grepioni, Chem. Soc. Rev. 2000, 29, 229.
[25] a) D. Braga, L. Maini, M. Polito, L. Scaccianoce, G. Cojazzi, F.
Grepioni, Coord. Chem. Rev. 2001, 216 – 217, 225; b) J. A.
Labinger, C. R. Chim. 2002, 5, 235.
[26] a) T. P. Brewster, W. Ding, N. D. Schley, N. Hazari, V. S. Batista,
R. H. Crabtree, Inorg. Chem. 2011, 50, 11938; b) L. E. Hatcher,
M. R. Warren, D. R. Allan, S. K. Brayshaw, A. L. Johnson, S.
Fuertes, S. Schiffers, A. J. Stevenson, S. J. Teat, C. H. Woodall,
P. R. Raithby, Angew. Chem. 2011, 123, 8521; Angew. Chem. Int.
Ed. 2011, 50, 8371; c) M. R. Warren, S. K. Brayshaw, A. L.
Johnson, S. Schiffers, P. R. Raithby, T. L. Easun, M. W. George,
J. E. Warren, S. J. Teat, Angew. Chem. 2009, 121, 5821; Angew.
Chem. Int. Ed. 2009, 48, 5711.
Revised: November 21, 2011
Published online: December 23, 2011
Keywords: bond-breaking isomers · coordination modes ·
.
exchange processes · hemilabile ligands · nickel
[1] E. Poverenov, M. Gandelman, L. J. W. Shimon, H. Rozenberg, Y.
Ben-David, D. Milstein, Organometallics 2005, 24, 1082.
[2] V. V. Grushin, Organometallics 2001, 20, 3950.
[3] P. J. W. Deckers, B. Hessen, J. H. Teuben, Angew. Chem. 2001,
113, 2584; Angew. Chem. Int. Ed. 2001, 40, 2516.
[4] G. L. Moxman, H. E. Randell-Sly, S. K. Brayshaw, R. L. Wood-
ward, A. S. Weller, M. C. Willis, Angew. Chem. 2006, 118, 7780;
Angew. Chem. Int. Ed. 2006, 45, 7618.
[5] R. Lindner, B. van der Bosch, M. Lutz, N. H. J. Reek, J. I.
van der Vlugt, Organometallics 2011, 30, 499.
[6] A. M. Spokoyny, C. W. Machan, D. J. Clingerman, M. S. Rosen,
M. J. Wiester, R. D. Kennedy, A. A. Sarjeant, C. L. Stern, C. A.
Mirkin, Nat. Chem. 2011, 3, 590.
[7] M. S. Rosen, A. M. Spokoyny, C. W. Machan, C. L. Stern, A. A.
Sarjeant, C. A. Mirkin, Inorg. Chem. 2011, 50, 1411.
[8] O. T. Summerscales, F. G. N. Cloke, P. B. Hitchcock, J. C. Green,
N. Hazari, Science 2006, 311, 829.
[27] J. Bernstein, Cryst. Growth Des. 2011, 11, 632.
[28] P. E. Garrou, Chem. Rev. 1981, 81, 229.
[29] C. W. Machan, A. M. Spokoyny, M. R. Jones, A. A. Sarjeant,
C. L. Stern, C. A. Mirkin, J. Am. Chem. Soc. 2011, 133, 3023.
[30] CCDC 851134 (4b), 851135 (2b), 851136 (3a), 851137 (2c),
851138 (2a), and 851139 (4a) 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.
[31] U. Kçlle, J. Kossakowski, N. Klaff, L. Wesemann, U. Englert,
G. E. Heberich, Angew. Chem. 1991, 103, 732; Angew. Chem. Int.
Ed. Engl. 1991, 30, 690.
[9] M. D. Fryzuk, W. E. Piers, Polyhedron 1988, 7, 1001.
[10] L. Kovbasyuk, R. Kramer, Chem. Rev. 2004, 104, 3161.
[11] H. J. Yoon, C. A. Mirkin, J. Am. Chem. Soc. 2008, 130, 11590.
[12] M. Takeuchi, M. Ikeda, A. Sugasaki, S. Shinkai, Acc. Chem. Res.
2001, 34, 865.
[13] H. J. Yoon, J. Kuwabara, J.-H. Kim, C. A. Mirkin, Science 2010,
330, 66.
[14] A. Bader, E. Lindner, Coord. Chem. Rev. 1991, 108, 27.
[15] C. S. Slone, D. A. Weinberger, C. A. Mirkin, Prog. Inorg. Chem.,
1999, 48, 233.
[32] U. Kçlle, Angew. Chem. 1991, 103, 970; Angew. Chem. Int. Ed.
Engl. 1991, 30, 956.
[33] R. A. Bailey, S. L. Kozak, T. W. Michelsen, W. N. Mills, Coord.
Chem. Rev. 1971, 6, 407.
[34] J. Coates, Encyclopedia of Analytical Chemistry (Ed.: R. A.
Meyers), Wiley, Chichester, 2000, p. 10815.
[35] G. A. Bain, J. F. Berry, J. Chem. Educ. 2008, 85, 532.
[36] E. M. Schubert, J. Chem. Educ. 1992, 69, 62.
[37] P. Atkins, T. Overtin, J. Rourke, M. Weller, F. Armstrong,
Inorganic Chemistry, Oxford University Press, Oxford, England,
2006.
[16] M. J. Wiester, P. A. Ulmann, C. A. Mirkin, Angew. Chem. 2011,
123, 118; Angew. Chem. Int. Ed. 2011, 50, 114.
[17] P. Braunstein, J. Organomet. Chem. 2004, 689, 3953.
1472
www.angewandte.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 1469 –1472