J. Chem. Inf. Model. 2005, 45, 965-970
Stability Studies of Transition-Metal Linkage Isomers Using Quantum Mechanical
Methods. Groups 11 and 12 Transition Metals
Corneliu Buda,† Abul B. Kazi,‡ Adriana Dinescu,† and Thomas R. Cundari*,†
Department of Chemistry, University of North Texas, Denton, Texas 76203, and, Department of Chemistry,
University of Arkansas Pine Bluff, Pine Bluff, Arkansas 71601
Received February 10, 2005
Several hypotheses to elucidate the linkage isomer preference of the thiocyanate (SCN-) ion have been
offered. For complexes with small coordination numbers (i.e., 1 and 2) and groups 11 (Cu-triad) and 12
(Zn-triad) metals, different levels of theory and a variety of basis sets have been employed to study linkage
isomerism. Similar results are obtained for all density functionals tested, pure and hybrid. Overall, good
agreement, vis-à-vis experimentally identified linkage isomers, is achieved for ab initio techniques, whereas
semiempirical quantum mechanical methods show a bias toward S-ligated isomers. Despite the seeming
ease for the a priori prediction of the most stable thiocyanate isomers using acid/base principles, this research
highlights the sensitivity of quantitative calculations of transition-metal linkage isomerism to the choice of
basis set and electron correlation, particularly with post-Hartree-Fock treatments.
Linkage isomerism is a type of isomerism that is distinctive
of transition-metal complexes, where ligands are capable of
bonding through different atoms. Jorgensen reported the first
example of linkage isomerism in 18931 for nitro (Co-NO2)
and nitrito (Co-ONO) cobalt(III) ammine complexes. A
large variety of linkage isomers involving ligands such as
NCS-, NCO-, CN-, NO-, NO2-, NCSe-, SO32-, CO2-, and
so forth were reported2 in subsequent years. From all possible
ligands that readily form linkage isomers, the thiocyanate
group, with two different isomers exemplified by SCN(S bonding) and NCS- (N bonding), was selected for the
present research as there is considerable experimental data
with which to compare theoretical results.
Turco and Pecile reported the first thiocyanate linkage
isomers3 in 1961. One of their hypotheses for linkage
preference was based on the electronic effect of ancillary
ligands. For example, they proposed that a phosphine ligand
trans to a thiocyanate would monopolize the π-bonding d
orbitals of the square-planar Pt(II) complex, yielding an
N-bonded thiocyanate in the trans position, as the Pt-N
linkage is preponderantly σ in character. In the absence of
competition for Pt dπ orbitals from the trans ligand (e.g.,
NH3), the S-bonded thiocyanate was proposed to be favored
because a Pt-S linkage can maximize π bonding (Figure
Using this hypothesis, Basolo et al.4 prepared and identified (e.g., through IR spectra) both thiocyanate linkage
isomers in the following complexes: Pd(AsPh3)2(SCN)2,
Pd(AsPh3)2(NCS)2, Pd(bipy)(SCN)2, and Pd(bipy)(NCS)2. In
all cases, the S-bonded isomer, which is stable at room
temperature, is converted to the N-bonded isomer upon
warming. Electronic effects in linkage isomerism were also
* Corresponding author e-mail: [email protected]
† University of North Texas.
‡ University of Arkansas Pine Bluff.
Figure 1. Ligand competition for Pt π orbitals
experimentally tested by Bertini and Sabatini,5 who stated
that thiocyanates are N-bonded in [Pd(5-nitrophen)(NCS)2]
and S-bonded in [Pd(1,10-phenanthroline)(SCN)2]. Their
explanation is based on the electronic properties of the
5-nitro-phenanthroline ligand, whereby the nitro group is
presumably π-bonded to the arene ring, inducing π character
to the bond between Pd and N, thus enforcing the σ-bonded
thiocyanate in the trans position. It was proposed that by
increasing the temperature, that π bonding between the nitro
group and ring would be removed, inducing the S-bonded
linkage isomer in the trans position, in accord with experimental observations.5
The combined influence of electronic and steric factors
upon thiocyanate linkage isomerism has been investigated
for bidentate ligands6 that have one σ-donor functionality
(e.g., an amine) and one π-bonding functionality (e.g.,
phosphine), namely, [Ph2P-(CH2)3-NMe2]Pd(NCS)(SCN),
where the S-bonded thiocyanate is trans to the amine and
the N-bonded isomer is trans to the phosphine (Figure 2).
Another electronic explanation of thiocyanate bonding
preferences was proposed by Jorgensen7 and is called
“symbiosis theory.” This theory is based on the “hard” or
“soft” acid/base character of the metal and ligands and has
been applied mainly to octahedral complexes. Symbiosis
theory asserts that hard ligands will make the metal cation
to which they are connected “harder,” which will, in turn,
10.1021/ci050050t CCC: $30.25 © 2005 American Chemical Society
Published on Web 04/29/2005
966 J. Chem. Inf. Model., Vol. 45, No. 4, 2005
Figure 2. Meek’s complex.6 (Amine and phosphine substituents
are shown in the wireframe.)
promote further coordination of additional hard ligands;
similarly, “soft” ligands will lead the metal ion to favor additional “soft” ligands. This theory was supported by Gutterman and Gray,8 who, on the basis of spectroscopic and
theoretical data, suggested that -SCN would prefer to bind
soft metals, whereas -NCS would prefer to bind hard metals.
Palenik et al.9 proposed an analysis of linkage isomerism
based solely on steric effect for Pd complexes of the type
[Ph2P-(CH2)n-PPh2]Pd(NCS)2. The electronic environment
at the phosphorus is essentially constant throughout this
series, and thus, by adding a CH2 group, only the steric effect
is changed, which should make it possible to delineate steric
and electronic influences upon linkage isomerism. By varying
the number of CH2 groups from 1 to 3, the P-Pd-P bite
angle increases, and both thiocyanate groups, thus, change
from both being S-bonded (n ) 1) to a mixture of S-bonded
and N-bonded (n ) 2) to only N-bonded (n ) 3) because
the linearly coordinated NCS is sterically less hindering than
the bent SCN isomer.
At first glance, “symbiosis theory” and “π-bonding theory”
appear to contradict each other. In addition, the possibility
that multiple factors (e.g., both electronic and steric) are
involved in determining linkage isomer preferences seems
plausible. To incorporate all effects, Pearson10 proposed that
two soft ligands connected to a soft transition metal would
destabilize each other. He also indicated that symbiosis
theory is predominant in octahedral complexes, whereas
π-bonding theory is more applicable to square-planar species.
Furthermore, small energy differences between both linkage
isomers have been identified in solutions, where the bonding
mode is decided by the nature of the solvent.11 For the
majority of transition metals, thiocyanate complexes are
preponderantly S-bonded in solvents with a high dielectric
constant, whereas in solvents with a low dielectric constant,
N-bonded isomers are preferred.11
To better understand linkage isomer complexity for the
thiocyanate ion, and to fill in existing gaps in experimental
data, we performed theoretical calculations to elucidate
factors that influence linkage to a transition metal through
the S or N atom. In many cases, experimental methods
encounter difficulties in differentiating between the S and
N linkage isomers of thiocyanate. Furthermore, small energy
differences between both possible coordination isomers will
present an extremely challenging task to identifying the most
stable linkage isomer with theory. Epps et al.12 prepared and
investigated [Co(DH)2(NCS)2]-, DH is the monoanion of
dimethylglyoxime, using IR spectral methods and molecular
orbital calculations; they observed small energy differences
(less than 1 kcal/mol) between S- and N-bonded linkage
isomers. Moreover, the authors stated that π-bonding effects
will be more pronounced for square-planar complexes than
for octahedral complexes and that S-bonded isomers are more
likely to form a π bond with fourth- and fifth-row transition
metals than with third-row transition metals. Small energy
differences between S- and N-bonded linkage isomers were
also identified for [Co(NH3)5(NCS)]2+ and [Ru(NH3)5(NCS)]+ by Tuan et al.13 using extended Hückel calculations,
consistent with the experimental observation of both linkage
isomers resulting from small temperature variations. The
electronic structure of the thiocyanate ion was investigated
by Di Sipio et al.14 using SEQM (semiempirical quantum
mechanics) methods in order to determine bonding preference. The ambidentate nature of the thiocyanate ion was
explained as arising from the almost equally distributed
charge over the terminal atoms, and thus, formation of Nor S-bonded linkage isomers was mainly attributed to the
establishment of the most stable electron configuration during
bond formation with the transition metal.
Reiher15 evaluated the performance of calculations [Hartree-Fock (HF) and density functional theory (DFT) methods] to determine spin-crossover and vibrational effects for
Fe(phen)2(NCS)2, phen ) phenanthroline. Reliable energy
splittings compared with experimental results were provided
by hybrid DFT functionals, particularly B3LYP, when one
of the three parameters was modified (i.e., c3 ) 0.15 instead
of the original c3 ) 0.20). The same DFT functional was
then used by Tercero et al.16 to investigate the contribution
of the thiocyanate ion to the magnetic coupling of Cu(II)Ni(II) supramolecular compounds. Theoretical results suggested that coupling through the thiocyanate ion is very small
and antiferromagnetic. The stability of group 12 transition
metals (Zn, Cd, and Hg) as four-coordinate complexes where
all ligands are either N- or S-bonded thiocyanate was
investigated by Fukushima et al.17 using DFT methods. For
the first- and second-row metals Zn and Cd, respectively,
the N-bonded isomer was found to be more stable, whereas
for the heavier Hg, the S-bonded isomer is preferred; for
the Cd complex, the difference in energy between the two
linkage isomers was relatively small.
In this research, the accuracy of the “symbiosis theory”
proposed by Jorgensen7 is investigated for a series of
transition-metal thiocyanate complexes using several theoretical methods.
The results of the faster PM3(tm) semiempirical method
are compared with HF and DFT methods (e.g., pure BLYP
and BP86 and hybrid B3LYP and B3PW91 functionals). The
“B” functionals include Slater18 local and Becke19 nonlocal
exchanges, whereas the “B3” functionals are formed from
Slater + HF local and Becke19 nonlocal exchanges, along
with the Vosko, Wilk, and Nusair20 local correlation. The
“LYP” functionals contain the Lee, Yang, and Parr21 local
+ nonlocal correlation functional; the designation “PW91”
denotes the 1991 nonlocal correlation functional developed
by Perdew and Wang,22 and the “P86” functional designates
Perdew 8123 local correlation and Perdew 8624 nonlocal
In cases in which DFT-calculated energy differences
between linkage isomers are small, other geometry optimizations (e.g., MP2) and single point [CCSD(T)//MP2] calculations are performed to further differentiate isomer stability.
The large majority of theoretical calculations employed the
CEP-31G and CEP-121G effective core potential basis sets25
with s and p diffuse and d polarization functions obtained
from the 6-31+G* basis set for all p-block elements.26 The
LANL2DZ27 scheme employs the Dunning/Huzinaga valence
double-ζ basis set D95V28 for the first-row main-group
elements and the Los Alamos ECP/valence basis set for all
elements between Na and Bi.
For some species, calculations with additional basis
functions are performed to determine the most stable linkage
isomer; in particular, the 3f and 2g polarization functions
from the Ahlrichs coulomb fitting basis set29 are added to
all group 11 and 12 metals of the periodic table.26 Moreover,
extra diffuse s and p and polarization d and f functions are
added to C, N, and S atoms, these being obtained from the
aug-cc-pVTZ diffuse basis set.30 To this are added additional
polarization functions: 3d1f for S [from the cc-pV-(T+d)Z
basis set31] and 2d1f for C and N atoms (from the cc-pVTZ
basis set32). Adding these diffuse and polarization basis
functions to the existing CEP-121G yields a basis set
Some discrepancies between results provided by the
preceding “b0” basis set and experimental data are observed
for mercury bis(thiocyanate) complexes, and thus, four
additional smaller basis sets are employed for further
calculations. First, “b1” is obtained from “b0” by removing
polarization functions for N, C, and S atoms that came from
the cc-pV-(T+d)Z basis set.31 Second, “b2” is obtained from
“b0” by removing diffuse functions for N, C, and S atoms
that came from the aug-cc-pVTZ diffuse basis set30 and then
adding the polarization and diffuse functions from the
cc-pVTZ basis sets32 for N, C, and S atoms. Third, “b3” is
similar to “b2” except for the S atom, for which the extra
polarization and diffuse basis functions came from cc-pV(T+d)Z31 instead of cc-pVTZ. The last, and largest, basis
set employed in the study of linkage isomers, “b4”, included
the same extra basis functions for N, C, and S atoms as “b3”
and included the complete Ahlrichs coulomb fitting basis
set29 for the Hg atom.
This research employs the PM3(tm) method as implemented in the Spartan 0433 package, whereas the Gaussian
98 and Gaussian 0334 programs are used for all ab initio and
DFT calculations. Reported energies quoted below are
differences in electronic energies; the inclusion of zero-point,
enthalpy, and entropic corrections does not affect the linkage
RESULTS AND DISCUSSION
The preference for S- versus N-bonded linkage isomers
for thiocyanate complexes of different transition metals, with
various coordination numbers and ligand environments, is
investigated using theoretical calculations. To determine the
accuracy of each individual method, theoretical results are
compared with crystallographic data when available.
When the principle first proposed by Jorgensen is considered,7 it is expected that transition metals will favor the
J. Chem. Inf. Model., Vol. 45, No. 4, 2005 967
coordination of the N or S atom of thiocyanate on the basis
of complementarity to the metal's own acid/base properties.
Thus, first-row transition metals are expected to be preponderantly “hard” and should, thus, prefer to be ligated to the
“hard” N atom of the thiocyanate ion, whereas third-row
transition metals are primarily “soft” and would favor a “soft”
S atom. For second-row transition metals, both isomers may
be feasible with small energy differences between them.17
Of course, metal cation softness and hardness is also a
function of formal oxidation state.
Analysis of the Cambridge Structural Database (CSD35)
reveals that transition metals exhibit a strong preference for
the formation of N-ligated thiocyanates, in a ratio of 7:1
(1782 samples) as compared to S-ligated thiocyanates.
Surprisingly, even though Jorgensen’s symbiosis theory
predicts that third-row transition metals will form predominantly S-ligated thiocyanates, the CSD shows an approximately equal distribution between the two isomers (N:S
) 108:133 ≈ 0.81). Thus, only the first and second-row
transition metals follow the expected acid/base prediction;
the ratios between the two isomers are ∼42:1 (N:S )
1271:30) and ∼3:1 (N:S ) 176:64), respectively, favoring
the N-ligated thiocyanates over the S-ligated thiocyanates.
Group 11 Thiocyanate Complexes. Thiocyanate complexes of group 11 ions (Cu+, Ag+, and Au+) are chosen on
the basis of their reduced isomeric complexity, resulting from
the low coordination numbers (CN ) 1 and 2). For onecoordinate complexes, MN (N-thiocyanato) and MS
(S-thiocyanato) linkage isomers are considered (M ) Cu,
Ag, Au; N f NCS, S f SCN), whereas for two-coordinate
complexes, M(thiocyanate)2-, MN2, MNS, and MS2 are
investigated at disparate levels of theory.
The performance of different DFT functionals (e.g., pure
BP86 and BLYP and hybrid B3LYP and B3PW91) is
investigated for both one- and two-coordinate thiocyanate
complexes of group 11 ions to establish the sensitivity of
the results to functional choice. All DFT functionals tested
are associated with two pseudopotential basis sets,
CEP-121+G(d)25 and LANL2DZ(d)27 (ξspS ) 0.0405, ξspC
) 0.0438, ξspN ) 0.0639, ξdS ) 0.65, and ξdC,N ) 0.80).26
Less expensive methods exemplified by HF with the same
basis sets and PM3(tm) semiempirical quantum mechanics
are also assessed. More expensive MP2 and CCSDT//MP2
calculations are carried out with the CEP-121+G(d) basis
set. Calculated bond lengths and bond angles for optimized
geometries are consistent with available experimental data.
Table 1 organizes the relative energies of the different linkage
isomers at the different levels of theory investigated.
For one-coordinate species, complete agreement is found
among all DFT functionals using both CEP-121+G(d) and
LANL2DZ(d) basis sets; the linkage isomers calculated to
be most stable are CuN, AgS, and AuS, Table 1. The
S-bonded linkage isomer is always found to be more stable
by PM3(tm). When HF/CEP-121+G(d) and HF/LANL2DZ(d)
methods are used, the DFT results are reproduced, although
the smaller basis set predicts the CuS linkage isomer to be
more stable than the CuN isomer by 0.3 kcal/mol.
MP2/CEP-121+G(d) and CCSDT/CEP-121+G(d)//MP2/
CEP-121+G(d) predicted CuN, AgS, and AuS to be the more
stable linkage isomers, supporting the consensus DFT and
HF predictions, Table 1.
968 J. Chem. Inf. Model., Vol. 45, No. 4, 2005
Table 1. Relative Energies (kcal/mol) of the Linkage Isomers Investigateda
Group 11 Metals
For the group 11 monothiocyanates, the energy difference between MN and MS (e.g., S-N) linkage isomers is given; a negative number
indicates the EMS < EMN. For group 11 and group 12 bis(thiocyanate) complexes, the energy differences are relative to the MN2 (e.g., NS-N2 and
S2-N2) linkage isomer. Basis sets are described in the Computational Methods section.
The similarity of results among DFT functionals suggests
that this parameter is not of overriding significance in
determination of the preferred linkage isomer; therefore,
B3LYP is chosen for further DFT investigations on the basis
of its larger utilization for similar complexes in the literature.15,16 In combination with the larger b0 basis set (see
Computational Methods), the B3LYP functional reproduces
the results provided by all DFT methods using the
CEP-121+G(d) and LANL2DZ(d) basis sets. Thus, in the
main, theoretical calculations predict CuN, AgS, and AuS
to be the most stable linkage isomers, a result in accord with
Jorgensen’s hard/soft theory of linkage isomerism.7
For anionic group 11 bis(thiocyanate) complexes, M(thiocyanate)2-, all functionals and HF methods used in conjunction with both the CEP-121+G(d) and LANL2DZ(d) basis
sets indicated N-linked thiocyanates (i.e., CuN2, AgN2, and
AuN2) to be more stable than the corresponding MNS and
MS2 linkage isomers, Table 1. Conversely, S-bonded linkage
isomers are predicted to be more stable with the PM3(tm)
method. Post-HF methods MP2 and CCSDT//MP2 coupled
with the CEP-121+G(d) basis set provided predictions identical to those of the DFT and HF methods, the N-linked linkage isomer being more stable than the other possible isomers.
When the CSD35 was searched for bis(thiocyanate) complexes, five copper complexes corresponding to the CuN2 isomer,
one AgS2 isomer, and one AuS2 isomer were identified.
Although caution is needed in making conclusions about
the stability of isomers on the basis of the existence or
absence of structurally characterized complexes, particularly
for small sample sizes, the structurally characterized bis(thiocyanate) complexes are consistent with simple acid/base
considerations. For copper bis(thiocyanate), the CuN2 linkage
isomer is always predicted to be more stable. However, an
interesting trend is observed when comparing the energy
differences between AgS2-AgN2 and AuS2-AuN2 linkage
isomers using the CEP-121+G(d) basis set: B3LYP (6.7
and 6.2 kcal/mol for the silver and gold complexes,
respectively) > MP2 (3.6 and 3.9 kcal/mol, respectively) >
CCSDT//MP2 (1.6 and 1.2 kcal/mol, respectively). When
more elaborate levels of electron correlation are incorporated,
the S-linked isomer becomes more energetically competitive,
consistent with the crystallographic data and the predictions
of acid/base theory. It is remarkable that such extensive
theoretical treatments are needed to properly describe what,
a priori, would have been expected to be “simple” acid/base
trends in linkage isomerism stability! A possible explanation
for the need of such highly correlated calculations may result
from an insufficient representation of relativistic effects for
fourth- and fifth-row transition metals. Support for this
inference arises from the method sensitivity encountered
below for mercury complexes, for which relativistic effects
are known to be substantial.
Group 12 Thiocyanate Complexes. For group 12 metals
(Zn, Cd, and Hg) only neutral, two-coordinate thiocyanate
complexes are investigated. On the basis of the agreement
among DFT functionals for similar calculations performed
on group 11 species, the B3LYP functional is used to model
Figure 3. Resonance structures of N- and S-ligated thiocyanates
with transition metals; bond lengths (Å) were estimated using the
Pauling covalent radii.38
In the CSD,35 one CdN2 (refcode ZIFKUU36) and one HgS2
(refcode REJNAV37) (recall that N f N-ligated thiocyanate
and S f S-ligated thiocyanate) bis(thiocyanate) complex are
identified. No data were found for Zn, but “symbiosis
theory”7 leads to the prediction that the “hard” N-bonded
linkage isomer will be energetically preferred by the “hard”
At all levels of theory, the predicted energy ordering of
the three possible linkage isomers for zinc and cadmium is
MN2 < MNS < MS2, Table 1. Interestingly, the mercury
complexes are much more sensitive to both the basis set and
the electron correlation treatment. For example, the HgS2
linkage isomer is calculated as the most stable species when
the MP2/CEP-121+G(d) level of theory is used, Table 1.
When the larger b0 basis set is used, maintaining the MP2
method, the HgN2 linkage isomer is now lowest in energy.
For the B3LYP functional with these same basis sets, different ground state predictions are again obtained: HgS2
[B3LYP/CEP-121+G(d)] and HgNS (B3LYP/b0)! At the
highest level of electron correlation employed in this study,
the HgS2 isomer is predicted to be the most stable linkage
isomer. The expense of the CCSD(T) method precluded its
use with augmented basis sets. However, B3LYP calculations
and the augmented basis sets (b2, b3, and b4) display a convergence to the HgN2 complex as the most stable linkage isomer,
being ca. 3 kcal/mol more stable than the HgS2 isomer,
Resonance Structures. Two resonance structures seem
plausible for the N-ligated thiocyanates (Figure 3, A and
B); bond lengths were estimated using the Pauling covalent
radii.38,39 Investigation of the Cambridge crystallographic
database35 indicates that the average distance (537 samples)
between N and C is 1.15 ( 0.01 Å (triple bond), whereas
the average distance (547 samples) between C and S for
N-ligated thiocyanates is 1.62 ( 0.01 Å (double bond).
Moreover, the average transition-metal-N-C bond angle
(558 samples) is 165.6 ( 7.9°, suggesting that both the A
and B resonance structures contribute to the molecular
structure of N-ligated thiocyanates, with perhaps a slightly
higher percentage of A than B.
Likewise, when the sulfur of thiocyanate ligates the
transition metal, two plausible resonance structures are
expected (Figure 3, C and D). In this case, the experimental
geometries show that the average distance (71 samples)
between S and C for S-ligated thiocyanates is 1.66 ( 0.02
Å (double bond), whereas the average bond length (74
samples) between C and N is 1.14 ( 0.02 Å (triple bond).
Again, both C and D resonance structures contribute to the
molecular structure of the S-ligated thiocyanates, but the
J. Chem. Inf. Model., Vol. 45, No. 4, 2005 969
average transition-metal-S-C bending angle (75 samples)
of 103.0 ( 4.5° suggests that D is the dominant contributor.
Application of the natural bond order (NBO40) analysis
to the equilibrium geometry of the NCS anion at the B3LYP/
aug-ccpVTZ level of theory reveals a triple bond (1.17 Å)
between nitrogen and carbon and a single bond (1.67 Å) between carbon and sulfur. The NBO calculation indicates that
nitrogen and carbon are s-p mixing, whereas the sp
hybridization for sulfur is very small (20% s, 80% p); s
character is primarily located in a sulfur lone pair. The NBO
results are consistent with the experimental metric data. The
predominant p character of sulfur with little s mixing is
consistent with the value (103.0 ( 4.5°) for the bond angle
of the S-ligated thiocyanates, which is greater than 90°
(expected in the complete absence of s-p mixing) and
smaller than 109° (expected for sp3 hybridization). A major
reason for the S-C bond shortening (1.67 Å calculated
compared to 1.81 Å expected38) in thiocyanate is not multiple
bonding, but rather, the nature of the C atom, which is sp
hybridized. B3LYP/aug-ccpVTZ geometry optimizations of
acetylene thiol (H-S-CtC-H) and thiocyanic acid (HSCN)
systems show C-S single bond lengths of 1.69 and 1.70 Å,
respectively, that are comparable to the values obtained for
To assess different theoretical methods, a series of calculations are performed for thiocyanate complexes of group 11
(Cu-triad) and group 12 (Zn-triad) metals. The PM3(tm) semiempirical parametrization systematically predicted S-bonded linkage isomers to be more stable than the N-ligated
counterparts. Thus, prediction of the stability of linkage
isomers with PM3(tm) calculations is not always in agreement with experimental data. In general, predictions with
ab initio methods are found to be reliable for neutral complexes in that the predicted ground-state linkage isomer did
not change with the level of theory, although the magnitude
of the energy differences is more sensitive to the basis set
and the correlation method. For anionic complexes of group
11 bis(thiocyanates), the demands of the basis sets are more
stringent and extensive correlation at the post-HF levels of
theory is required to bring the calculations into correspondence with crystallographic data. However, it is remarkable
to observe that the quantitative prediction of linkage isomer
preferences for transition-metal thiocyanates, which can be
done reliably using “symbiosis theory”, requires extended
basis set augmentation and high levels of electron correlation.
The authors acknowledge the Office of Energy Sciences
and the Office of Science, United States Department of
Energy, for support of this research through Grants DEFG02-97ER14811 and DE-FG02-03ER15490, respectively.
REFERENCES AND NOTES
(1) Huheey, J. E. Inorganic Chemistry, 3rd ed.; Harper and Row: New
York, 1983; pp 513-524.
(2) (a) Basolo, F.; Hammaker, G. S. Synthesis and Isomerization of
Nitritopentammine Complexes of Rhodium(III), Iridium(III), and
Platinum(IV). Inorg. Chem. 1962, 1, 1-5. (b) Burmeister, J. L. Linkage
isomerism in metal complexes. Coord. Chem. ReV. 1968, 3, 225245. (c) Norbury, A. H.; Sinha, A. I. P. Coordination of Ambidentate
Ligands. Q. ReV., Chem. Soc. 1970, 24, 69-4. (d) Norbury, A. H.
Coordination Chemistry of the Cyanate, Thiocyanate, and Selenocyanate Ions. AdV. Inorg. Chem. Radiochem. 1975, 17, 231-386.
970 J. Chem. Inf. Model., Vol. 45, No. 4, 2005
(3) Turco, A.; Pecile, C. Coordination of the Thiocyanate Group in
Inorganic Compounds. Nature 1961, 191, 66-67.
(4) Basolo, F.; Burmeister, J. L.; Poe, A. J. Linkage Isomerism: Synthesis
of Thiocyanato and Isothiocyanato Isomers of Some Palladium(II)
Complexes. J. Am. Chem. Soc. 1963, 85, 1700-1701.
(5) Bertini, I.; Sabatini, A. Infrared Spectra of Substituted Thiocyanate
Complexes. The Effect of the Substituent on Bond Type. II. Inorg.
Chem. 1966, 5, 1025-1028.
(6) (a) Meek, D. W.; Nicpon, P. E.; Meek, V. I. Mixed Thiocyanate
Bonding in Palladium(II) Complexes of Bidentate Ligands. J. Am.
Chem. Soc. 1970, 92, 5351-5359. (b) Clark, G. R.; Palenik, G. J.
Crystal and Molecular Structure of Isothiocyanatothiocyanato-(1diphenylphosphino-3-dimethylaminopropane) Palladium(II). Inorg.
Chem. 1970, 9, 2754-2760.
(7) Jorgensen, C. K. “Symbiotic” Ligands, Hard and Soft Central Atoms.
Inorg. Chem. 1964, 3, 1201-1202.
(8) Gutterman, D. F.; Gray, H. B. Electronic Structures of Thiocyanatopentacyanocobaltate(III) and Related Complexes. J. Am. Chem. Soc.
1971, 93, 3364-3371.
(9) Palenik, G. J.; Steffen, W. L.; Mathew, M.; Li, M.; Meek, D. W. Steric
Control of Thiocyanate Coordination in Palladium(II)-diphosphine
Complexes. Inorg. Nucl. Chem. Lett. 1974, 10, 125-128.
(10) Pearson, R. G. Antisymbiosis and the Trans Effect. Inorg. Chem. 1973,
(11) Burmeister, J. L.; Hassel, R. L.; Phelan, R. J. Solvent-induced Linkage
Isomerizations. Inorg. Chem. 1971, 10, 2032-2038.
(12) Epps, L. A.; Marzilli, L. G. Preparation of a Mixed N-bonded-Sbonded trans-Dithiocyanato Complex. The Three Isomers of Tetraphenylarsonium trans-(Dithiocyanato)bis(dimethylglyoximato)cobaltate(III). Inorg. Chem. 1973, 12, 1514-1517.
(13) Tuan, D. F.; Reed, J. W.; Hoffmann, R. Studies of the Linkage and
Bonding of Triatomics in Transition Metal Complexes: Part 2. NCSComplexes. THEOCHEM 1991, 232, 111-121.
(14) Di Sipio, L.; Oleari, L.; De Michelis, G. A Study of the Electronic
Structure of the Thiocyanate Ion. Coord. Chem. ReV. 1966, 1, 7-12.
(15) Reiher, M. Theoretical Study of the Fe(phen)2(NCS)2 Spin-Crossover
Complex with Reparametrized Density Functionals. Inorg. Chem.
2002, 41, 6928-6935.
(16) Tercero, J.; Diaz, C.; Ribas, J.; Ruiz, E.; Mahia, J.; Maestro, M. New
Oxamidato-Bridged CuII-NiII Complexes: Supramolecular Structures
with Thiocyanate Ligands and Hydrogen Bonds. Magnetostructural
Studies: DFT Calculations. Inorg. Chem. 2002, 41, 6780-6789.
(17) Fukushima, N.; Iisaka, G.; Saito, M.; Waizumi, K. Ab initio Density
Functional Studies on the Stability of Tetrathiocyanato Complexes of
Zn(II), Cd(II) and Hg(II). Inorg. Chim. Acta 1997, 255, 211-214.
(18) Slater, J. C. Quantum Theory of Molecules and Solids; McGraw-Hill:
New York, 1974; Vol. 4.
(19) Becke, A. D. Density-functional Exchange-energy Approximation with
Correct Asymptotic Behavior. Phys. ReV. A 1988, 38, 3098-3100.
(20) Vosko, S. H.; Wilk, L.; Nusair, M. Accurate Spin-dependent Electron
Liquid correlation Energies for Local Spin Density Calculations: A
Critical Analysis. Can. J. Phys. 1980, 58, 1200-1211.
(21) Lee, C.; Yang, W.; Parr, R. G. Development of the Colle-Salvetti
Correlation-energy Formula into a Functional of the Electron Density.
Phys. ReV. B 1988, 37, 785-789.
(22) (a) Perdew, J. P.; Chevary, J. A.; Vosko, S. H.; Jackson, K. A.;
Pederson, M. R.; Singh, D. J.; Fiolhais, C. Atoms, Molecules, Solids, and
Surfaces: Applications of the Generalized Gradient Approximation for
Exchange and Correlation. Phys. ReV. B 1992, 46, 6671-6687. (b)
Perdew, J. P.; Chevary, J. A.; Vosko, S. H.; Jackson, K. A.; Pederson,
M. R.; Singh, D. J.; Fiolhais, C. Erratum: Atoms, Molecules, Solids,
and Surfaces: Applications of the Generalized Gradient Approximation
for Exchange and Correlation. Phys. ReV. B 1993, 48, 4978.
(23) Perdew, J. P.; Zunger, A. Self-interaction Correction to Densityfunctional Approximations for Many-electron Systems. Phys. ReV. B
1981, 23, 5048-5079.
(24) Perdew, J. P. Density-functional Approximation for the Correlation
Energy of the Inhomogeneous Electron Gas. Phys. ReV. B 1986, 33,
(25) Stevens, W. J.; Krauss, M.; Basch, H.; Jasien, P. G. Relativistic
Compact Effective Potentials and Efficient, Shared-exponent Basis
Sets for the Third-, Fourth-, and Fifth-row Atoms. Can. J. Chem. 1992,
(26) (a) Hariharan, P. C.; Pople, J. A. Influence of Polarization Functions
on Molecular-orbital Hydrogenation Energies. Theor. Chim. Acta 1973,
28, 213-222. (b) Francl, M. M.; Petro, W. J.; Hehre, W. J.; Binkley,
J. S.; Gordon, M. S.; DeFrees, D. J.; Pople, J. A. Self-consistent
Molecular-orbital Methods. 23. A Polarization-type Basis Set for 2ndrow Elements. J. Chem. Phys. 1982, 77, 3654-3665. (c) Clark, T.;
Chandrasekhar, J.; Spitznagel, G. W.; Schleyer, P. v. R. Efficient
Diffuse Function-augumented Basis-sets for Anion Calculations. 3.
The 3-21+G Basis Set for 1st-row Elements, Li-F. J. Comput. Chem.
1983, 4, 294-301. (d) Krishnan, R.; Binkley, J. S.; Seeger, R.; Pople,
J. A. Self-consistent Molecular-orbital Methods. 20. A Basis Set for
Correlated Wave Functions. J. Chem. Phys. 1980, 72, 650-654. (e)
Gill, P. M. W.; Johnson, B. G.; Pople, J. A.; Frisch, M. J. The
Performance of the Becke-Lee-Yang-Parr (B-LYP) Density Functional Theory with Various Basis Sets. Chem. Phys. Lett. 1992, 197,
(a) Hay, P. J.; Wadt, W. R. Ab initio Effective Core Potentials for
Molecular CalculationssPotentials for the Transition-Metal Atoms Sc
to Hg. J. Chem. Phys. 1985, 82, 270-283. (b) Wadt, W. R.; Hay, P.
J. Ab initio Effective Core Potentials for Molecular Calculationss
Potentials for Main Group Elements Na to Bi. J. Chem. Phys. 1985,
82, 284-298. (c) Hay, P. J.; Wadt, W. R. Ab initio Effective Core
Potentials for Molecular CalculationssPotentials for K to Au Including
the Outermost Core Orbitals. J. Chem. Phys. 1985, 82, 299-310.
Dunning, T. H., Jr.; Hay, P. J. Modern Theoretical Chemistry;
Plenum: New York, 1976; Vol. 3, pp 1-28.
(a) Eichkorn, K.; Treutler, O.; Ohm, H.; Haser, M.; Ahlrichs, R.
Auxiliary Basis Sets to Approximate Coulomb Potentials. Chem. Phys.
Lett. 1995, 240, 283-290. (b) Eichkorn, K.; Weigend, F.; Treutler,
O.; Ahlrichs, R. Auxiliary Basis Sets for Main Row Atoms and
Transition Metals and their use to Approximate Coulomb Potentials.
Theor. Chem. Acc. 1997, 97, 119-124.
(a) Kendall, R. A.; Dunning, T. H., Jr.; Harrison, R. J. Electronaffinities of the 1st-row Atoms RevisitedsSystematic Basis-sets and
Wave Functions. J. Chem. Phys. 1992, 96, 6796-6806. (b) Woon,
D. E.; Dunning, T. H., Jr. Gaussian Basis Sets for use in Correlated
Molecular Calculations. 4. Calculation of Static Electrical Response
Properties. J. Chem. Phys. 1994, 100, 2975-2988. (c) Woon, D. E.;
Dunning, T. H., Jr. Gaussian Basis Sets for use in Correlated Molecular
Calculations. 3. The Atoms Al through Ar. J. Chem. Phys. 1993, 98,
Dunning, T. H., Jr.; Peterson, K. A.; Wilson, A. K. Gaussian Basis
Sets for use in Correlated Molecular Calculations. X. The Atoms
Aluminum through Argon Revisited. J. Chem. Phys. 2001, 114, 92449253.
(a) Dunning, T. H., Jr. Gaussian Basis Sets for use in Correlated
Molecular Calculations. 1. The Atoms B through Ne and H. J. Chem.
Phys. 1989, 90, 1007-1023. (b) Wilson, A. K.; Woon, D. E.; Peterson,
K. A.; Dunning, T. H., Jr. Gaussian Basis Sets for use in Correlated
Molecular Calculations. IX. The Atoms Ga through Kr. Chem. Phys.
1999, 110, 7667-7676. (c) Koput, J.; Peterson, K. A. Ab Initio
Potential Energy Surface and Vibrational-Rotational Energy Levels
of X2∑+ CaOH. J. Phys. Chem. A 2002, 106, 9595-9599.
SPARTAN 04; Wavefunction, Inc.: Irvine, CA.
Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin,
K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone,
V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G.
A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.;
Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai,
H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.;
Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev,
O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P.
Y.; Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.;
Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas,
O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J.
B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.;
Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.;
Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.;
Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen,
W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. GAUSSIAN 03;
Gaussian, Inc.: Carnegie, PA.
Allen, F. H.; Kennard, O. 3D Search and Research using the Cambridge
Structural Database. Chem. Des. Autom. News 1993, 8, 31-37.
Mori, H.; Tanaka, S.; Mori, T.; Maruyama, Y. ZIFKUU [R-bis(Ethylenedithio)tetrathiafulvalenium bis(isothiocyanato)-cadmium(ii)]:
Crystal Structures and Electrical Resistivities of Three-component
Organic Conductors: (BEDT-TTF)2MM′(SCN)4 [M ) K, Rb, Cs; M′
) Co, Zn, Cd]. Bull. Chem. Soc. Jpn. 1995, 68, 1136-1144.
Bauer, V.; Lang, M.; Krieger, C.; Staab, H. A. REJNAV [21,24,45,48Tetramethoxy-2,5,8,11,14,17,26,29,32,35,38,41-dodecaoxa(18.18) metacyclophane bis(dithiocyanato-mercury)]. Private communication, 1997.
Pauling, L. The Nature of the Chemical Bond, 3rd ed.; Cornell
University Press: Ithaca, NY, 1960; Chapter 7, pp 224-288.
Valence bond calculations at the MP2/6-311+G(d,p) optimized
geometry of thiocyanate anion indicate that the lowest energy
resonance structure is [S-CtN]- followed by [SdCdN]- and then
[StC-N]-. Mo, Y. Western Michigan University. Personal communication.
Reed, A. E.; Curtiss, L. A.; Weinhold, F. Intermolecular Interactions
from a Natural Bond Orbital, Donor-Acceptor Viewpoint. Chem. ReV.
1988, 88, 899-926.