Effect of Coordination Geometry on the Gas-Phase Reactivity of Four-Coordinate Divalent Metal Ion Complexes

Article abstract of DOI:10.1021/jp0373954

The gas-phase reactions of ammonia with M(EN-(py)₂) ²⁺, M(en)(phen)²⁺, and M(phen)₂²⁺, where M is Mn-(II), Fe(II), Co(II), Ni(II), Cu(II), or Zn(II), EN-(py) ₂ is l,6-bis(2- pyridyl)-2,5-triazahexane, en is ethylene-diamine, and phen is 1,10-phenanthroline, have been studied in a quadrupole ion trap mass spectrometer. The trends in reactivity as a function of the metal are noticeably different for each complex despite the similarity of the ligand donor groups. These results indicate that coordination geometry and electronic structure have a significant impact on the gas-phase reactivity of four-coordinate metal complex ions. Specifically, for M(EN-(py) ₂)²⁺ the equilibrium constants for the reactions with ammonia follow the trend Mn ≈ Fe ≈ Co > Zn ? Cu > Ni. The trend for the equilibrium constants of the M(en)(phen)²⁺ complexes is Mn ≈ Co ? Ni > Cu > Zn, and for the M(phen)₂²⁺ complexes, the trend is Ni > Mn ≈ Fe > Co > Cu > Zn. The most notable changes in reaction equilibrium constants occur for the Ni and Zn complexes, which increase and decrease, respectively, as the ligands are changed from EN-(py)₂ to (en)(phen) to (phen)₂. Molecular orbital stabilization energy (MOSE) and density functional theory (DFT) calculations are used to explain the experimental trends. Calculations indicate that the changes in reactivity observed for the Ni complexes are a result of different complex geometries; Ni(EN-(py)₂)²⁺ is a low-spin square-planar complex, and Ni(phen)₂²⁺ is a high- spin tetrahedral complex. The decreased reactivity of the Zn complexes is due to the formation of a less distorted tetrahedral complex upon going from Zn(EN-(py)₂)²⁺ to Zn(phen)₂²⁺. In addition, calculations show that the reactivities of the Mn, Fe, and Co complexes are consistent with slightly distorted tetrahedral structures, and the reactivities of the Cu complexes indicate that the complexes of this metal are close to square planar.

Full text of DOI:10.1021/jp0373954

J. Phys. Chem. A 2004, 108, 1757-1763
1757
Effect of Coordination Geometry on the Gas-Phase Reactivity of Four-Coordinate Divalent
Metal Ion Complexes
Marianny Y. Combariza and Richard W. Vachet*
Department of Chemistry, UniVersity of Massachusetts, Amherst, Massachusetts 01003
ReceiVed: NoVember 7, 2003; In Final Form: January 13, 2004
The gas-phase reactions of ammonia with M(EN-(py)2)2+, M(en)(phen)2+, and M(phen)22+, where M is Mn-
(II), Fe(II), Co(II), Ni(II), Cu(II), or Zn(II), EN-(py)2 is 1,6-bis(2-pyridyl)-2,5-triazahexane, en is ethylene-
diamine, and phen is 1,10-phenanthroline, have been studied in a quadrupole ion trap mass spectrometer. The
trends in reactivity as a function of the metal are noticeably different for each complex despite the similarity
of the ligand donor groups. These results indicate that coordination geometry and electronic structure have a
significant impact on the gas-phase reactivity of four-coordinate metal complex ions. Specifically, for M(EN-
(py)2)2+ the equilibrium constants for the reactions with ammonia follow the trend Mn ≈ Fe ≈ Co > Zn .
Cu > Ni. The trend for the equilibrium constants of the M(en)(phen)2+ complexes is Mn ≈ Co . Ni > Cu
2+
> Zn, and for the M(phen)2 complexes, the trend is Ni > Mn ≈ Fe > Co > Cu > Zn. The most notable
changes in reaction equilibrium constants occur for the Ni and Zn complexes, which increase and decrease,
respectively, as the ligands are changed from EN-(py)2 to (en)(phen) to (phen)2. Molecular orbital stabilization
energy (MOSE) and density functional theory (DFT) calculations are used to explain the experimental trends.
Calculations indicate that the changes in reactivity observed for the Ni complexes are a result of different
2+
complex geometries; Ni(EN-(py)2)2+ is a low-spin square-planar complex, and Ni(phen)2 is a high-spin
tetrahedral complex. The decreased reactivity of the Zn complexes is due to the formation of a less distorted
tetrahedral complex upon going from Zn(EN-(py)2)2+ to Zn(phen)22+. In addition, calculations show that the
reactivities of the Mn, Fe, and Co complexes are consistent with slightly distorted tetrahedral structures, and
the reactivities of the Cu complexes indicate that the complexes of this metal are close to square planar.
Introduction
Several means of gathering gas-phase structural information
for metal complexes have been developed. Often these methods
rely on a synergy between experimental results and theoretical
calculations to draw conclusions. Experimentally, dissociation
techniques are commonly applied because mass spectrometers
are very good at measuring mass changes that accompany ion
dissociations. Gas-phase ion dissociations are typically induced
by collisional activation or photoactivation. With collision-
induced dissociation (CID), the magnitude of the mass change
is used to identify functional groups in metal complexes.28-33
Alternatively, the energetics of the dissociation process can be
used to draw conclusions about metal complex structure,34-36
especially when used in conjunction with calculations. In some
photoactivation methods, the kinetics and energetics of ion
dissociation provide insight into gas-phase structure,37,38 again
with help from theoretical calculations. In other photoactivation
techniques, the wavelengths required to dissociate ions are
measured in a manner very similar to solution-based optical
spectroscopic methods.39-46 Such photodissociation methods can
provide very detailed structural information, but these methods
are experimentally challenging and are usually limited to
relatively small complexes.
Knowledge of the gas-phase structures of metal complexes
is important for analytical, as evidenced by several reviews,1-4
but also fundamental reasons. An enormous amount of funda-
mental information has been gathered about specific bond
activation by metal ions as a function of electronic structure,
ligand chemistry, and thermodynamic factors via gas-phase
studies.5-12 Indeed, recent experiments have drawn very direct
comparisons to catalytic species involved in, for example, olefin
metathesis,13-16 C-H bond activation,17-19 epoxidation reac-
tions,20,21 and alcohol oxidation reactions.22 These investigations
benefit from the ability to evaluate catalytically active species
in the gas phase that are too transient to study in solution. The
otherwise short lifetimes of such species in solution do not pose
a problem in the high-vacuum environment of a mass spec-
trometer. Furthermore, intermediates or catalytically active
species of interest can readily be isolated and studied without
interferences from counterions, solvent, or additional complexes
that are usually present in solution. Studies of metal complexes
in the gas phase have also provided insight into metal ion and
metal complex microsolvation.23-27 Again, the controlled
environment of a mass spectrometer allows measurements of
solvation enthalpies and can provide insight into the details of
the first and second solvation shells around metal ions and metal
ion complexes. In all of these studies, knowing the gas-phase
structure of the metal complexes of interest enhances our
understanding of the underlying chemistry.
The most readily accessible and experimentally straightfor-
ward technique for attempting to gather structural information
for metal complexes in the gas phase is CID. This method,
however, has some limitations. One drawback is that upon
energy deposition the structure of the ion might rearrange so
that the product ions are not necessarily indicative of the initial
parent ion structure. For organic ions, this is often not a problem.
For metal-ligand complexes in which the metal-ligand bonds
* Corresponding author. E-mail: rwvachet@chem.umass.edu. Phone:
(413) 545-2733. Fax: (413) 545-4490.
10.1021/jp0373954 CCC: $27.50 © 2004 American Chemical Society
Published on Web 02/14/2004

1758 J. Phys. Chem. A, Vol. 108, No. 10, 2004
Combariza and Vachet
are usually weaker than the intraligand bonds, rearrangement
of the coordination structure before dissociation can possibly
occur. Also, upon dissociation it is often difficult to determine
if a lost fragment was originally bound to the metal or not.
Consequently, we have begun to investigate another means of
gathering metal-complex structural information using a more
gentle approach that involves ion-molecule (I-M) reactions.
I-M reactions maintain much of the simplicity of CID
experiments and can often be performed on the same instru-
mentation used to carry out CID experiments.
Figure 1. Complexes examined in this work: 1 ) M(EN-(py)2)2+, 2
Recently, we have demonstrated that I-M reactions can
provide some insight into the coordination structure of metal-
complex ions. These reactions can indicate the coordination
number of a metal complex when the appropriate reagent gas
is chosen either to react selectively with complexes of a given
coordination number47-49 or to “titrate” the open coordination
sites.50 The kinetics and thermodynamics of these reactions have
also been shown to be very sensitive to the functional groups
bound to the metal in the complex.50 To further assess the
sensitivity of gas-phase I-M reactions to metal-complex
coordination structure, we present here a study of the reactions
of four-coordinate metal-complex ions with ammonia.
2+
) M(en)(phen)2+, and 3 ) M(phen)2
.
Pd on carbon). The mixture was then filtered, and the final
product was recovered as a yellow oil after allowing the solvent
to evaporate. Bidentate ligands ethylenediamine (en) and
phenanthroline (phen), for complexes 2 and 3 in Figure 1, were
purchased from Sigma-Aldrich (St. Louis, MO) and used
without further purification.
Metal complexes were prepared by mixing MnCl2, FeCl2,
CoCl2, NiCl2, CuCl2, or Zn(CH3COO)2 solutions with the
appropriate stoichiometric amount of the ligand of interest in
100% methanol or water/methanol (50:50) to a final concentra-
tion of 50 µM. Complex ions were transferred to the gas phase
by electrospray ionization using a needle voltage of about 4
kV and a flow rate of 1.0-2.0 µL/min. Typically, a capillary
temperature of 150 °C and a capillary exit offset voltage of
20-30 V were used.
Electronic Structure Calculations. Geometry optimizations
were performed on the various metal-ligand complexes using
the Gaussian suite of computational chemistry programs.54 The
calculations were carried out using the B3LYP method55-57 with
a basis set containing effective core potentials (ECPs). The
presence of transition-metal centers with d orbitals and generally
high-spin electronic configurations makes the use of an all-
electron basis set impractical for complexes of the size shown
in Figure 1. ECPs were chosen instead to treat the heavy atoms
in our systems. In calculations involving ECPs, only valence
electrons and some outer-core electrons are considered because
they are mainly responsible for bonding interactions. Electrons
in the atomic core regions, presumed to be chemically inert
because they are not significantly perturbed upon bonding, are
represented by an ECP. Thus, the total number of electrons
treated explicitly was reduced, and the calculations for systems
with a large number of atoms were much more reasonable
computationally. The basis set denoted as LANL2DZ in the
Gaussian program was employed.58-61 LANL2DZ contains a
combination of Gaussian orbital valence basis sets and ab initio
ECPs to replace the innermost core electrons for third-, fourth-,
and fifth-row atoms.
Experimental Section
Instrumentation. Ion-molecule reactions were performed
on a modified quadrupole ion trap mass spectrometer (Bruker
ESQUIRE-LC, Billerica, MA). Instrument modifications and
experimental conditions have been previously described,50 but
a brief overview is provided below.
Reagent and buffer gases were introduced into the ion trap
through a set of precision leak valves obtained from MDC
Vacuum Products Corporation (Hayward, CA). An ion gauge
was used to monitor the pressure inside the mass spectrometer.
To obtain an accurate gas pressure, a calibration of the ion gauge
was performed using the known deprotonation reaction rates
of the 13+, 12+, and 11+ charge states of ubiquitin with NH3
(99.99%, Matheson Tri-gas, Parsippany, NJ).51 The correction
factor obtained from the calibration was then used, together with
the ion gauge sensitivity to NH3, CH3CN, and He,52 to calculate
the real pressures of the gases inside the mass spectrometer.
He, NH3, and CH3CN pressures used throughout the experiments
were (1.0 ( 0.3) × 10-4, (2.0 ( 0.1) × 10-6 and (3.2 ( 0.3)
× 10-7 Torr, respectively. The vacuum system temperature was
monitored and kept at 300 ( 2 K using a heating blanket.
Ions of interest were selected and reacted with either NH3 or
CH3CN for different periods of time. Reaction times and ion
isolation processes were controlled using ESQUIRE-LC soft-
ware. Kinetic data was obtained by monitoring the change in
the intensity of the product and parent ions for up to 4000 ms.
Rate constants were determined by fitting the experimental
kinetic data to a series of differential equations using the KinFit
program.53 Equilibrium constants were calculated as the ratio
of the forward and back rate constants or as the ratio of the
product and parent ion abundances at equilibrium in cases where
the fitting process failed to provide meaningful values for the
back rate constants.
Results and Discussion
Coordinatively unsaturated divalent metal complexes readily
react in the gas phase to add reagent gases. The number of
reagent gas molecules added and the kinetics and thermodynam-
ics of these reactions are found to be dependent upon the nature
of the coordination sphere around the metal.47-50,62-65 This work
deals with the gas-phase reactivity of four-coordinate species
formed by the complexation of metal ions from the first
transition series with either one tetradentate ligand (EN-(py)2)
or two bidentate ligands (en and/or phen). In general, these four-
coordinate complexes add one reagent molecule to become five-
coordinate when reacted with NH3. For some of the complexes,
the addition of a second NH3 molecule is also observed,
Ligand and Metal-Complex Synthesis. The synthesis of the
tetradentate ligand (EN-(py)2) for complex 1 in Figure 1 was
described previously.47 In short, 2-pyridinecarboxaldehyde was
mixed with ethylenediamine in a 2:1 (aldehyde/ethylenediamine)
ratio in anhydrous methanol. The mixture was reacted for 24 h
in a Parr shaking hydrogenator under a 60-psi atmosphere of
H2 (99.99%, Merriam-Graves, Springfield, MA) over Pd (10%

Reactivity of Divalent Metal Ion Complexes
J. Phys. Chem. A, Vol. 108, No. 10, 2004 1759
2+
TABLE 1: Rate and Equilibrium Constants for the Reactions of M(EN-(py)2)2+, M(en)(phen)2+, and M(phen)2 Complexes
with NH3
a
b
rate constant (× 10-11 cm3 molecule-1 s-1
)
equilibrium constant (× 109 atm-1
)
metal
EN-(py)2
(en)(phen)
(phen)2
EN-(py)2
(en)(phen)
(phen)2
Mn
Fe
Co
Ni
Cu
Zn
17 ( 3
12 ( 1
15 ( 4
13 ( 4
12 ( 5
14 ( 7
6 ( 0.7
11 ( 4
0.02 ( 0.01
0.10 ( 0.05
4 ( 1
30 ( 20c
50 ( 30c
50 ( 20c
3 ( 1
d
d
13 ( 1
7 ( 3
4.8 ( 0.5
19 ( 6c
0.3 ( 0.1
0.10 ( 0.02
0.07 ( 0.03
0.5 ( 0.2
0.4 ( 0.1
3 ( 1
2.2 ( 0.7
0.8 ( 0.1
0.8 ( 0.1
24 ( 8
120 ( 40c
e
e
0.02 ( 0.01
0.005 ( 0.002
a These values were obtained by fitting eq 1 to the experimental data. b Equilibrium constants were obtained from the equation K ) k1/k2, where
k1 and k2 are the values obtained by fitting eq 1 to the experimental data. c The fitting for these data failed to provide a value for k2, so the
equilibrium constant was determined directly from the mass spectrum taken under equilibrium conditions. d The Fe(II) complex of (en)(phen) was
difficult to generate by electrospray. e No significant addition of NH3 to the Cu(II) complex of (phen)2 was observed.
k1
k3
ML2+ + R [k
\
] [ML + R]2+
[
k ] [ML + 2R]2+
(1)
2
4
The data in Table 1 clearly show that the reactivity of the
four-coordinate complexes is dependent on the nature of the
ligand(s) and the metal center. When EN-(py)2 is the ligand,
the following reactivity trend is observed as a function of the
metal: Mn, Co, Fe > Zn > Cu, Ni. The trend changes slightly
when the two bidentate ligands en and phen are bound to the
metal (i.e., M(en)(phen)2+ complexes) so that the observed trend
is Mn, Co > Ni > Cu, Zn. An even more significant change in
the reactivity is observed when two phen ligands are bound to
2+
the metal (i.e., M(phen)2 complexes) such that the trend in
reactivity is Ni > Mn, Fe > Co > Cu, Zn.
Unique Reactivity of the Ni(II) and Zn(II) Complexes. The
most notable variations in reactivity as the ligand is changed
are observed for the Ni(II) and Zn(II) complexes. The equilib-
rium constants for the reaction of NH3 with the Ni(II) complexes
increase from 0.02 to 0.3 to 120 as the ligand(s) around it is-
(are) changed from EN-(py)2 to (en)(phen) to (phen)2. In
contrast, the equilibrium constants for the Zn(II) complexes
decrease from 4 to 0.07 to 0.005 as the ligands are changed
from EN-(py)2 to (en)(phen) to (phen)2. The cause of the
reactivity change for the Ni(II) and Zn(II) complexes is likely
due to changes in the electronic nature of each metal as the
ligand is varied. Recent work showed that as the functional
groups bound to a given divalent metal ion are changed from
nitrogen- to sulfur- to oxygen-containing groups the gas-phase
reactivity of five-coordinate metal complexes with CH3CN
changed significantly.50 The reactivity of CH3CN with the
complexes in that study provided some insight into the electronic
tuning of the metal by the different heteroatoms in the
coordination sphere. In the present study, the ligand donor
groups are not varied as much. Each ligand consists of four
nitrogen donors, at least two of which are in aromatic rings.
Furthermore, the reactivity of Mn, Fe, Co, and Cu complexes
does not change significantly as the ligand sphere is changed.
These results suggest that slight changes to the ligand donor
groups are not responsible for the significant differences in
reactivity observed for the Ni(II) and Zn(II) complexes.
Another possible explanation for the trends observed for the
Ni(II) and Zn(II) complexes is changes in the metal electronic
structure or electronic density caused by changes in the
complexes’ coordination geometry. Our recent results with five-
coordinate complexes show that gas-phase reactivity is affected
by coordination geometry65 and that coordination geometry
might be expected to affect the reactivity of four-coordinate
complexes as well. In general, four-coordinate complexes adopt
structures that are close to one of two ideal geometries, square
planar (D4h) or tetrahedral (Td). In our experiments, upon the
addition of a reagent molecule four-coordinate complexes
Figure 2. Typical kinetic plots obtained from the reactions of NH3
with (a) Zn(EN-(py)2)2+ and (b) Ni(EN-(py)2)2+
.
indicating the formation of a six-coordinate complex, but the
abundance of this ion is generally very low. The signal
corresponding to the second addition of NH3 never exceeds a
relative abundance of 10%, which is consistent with previous
studies that demonstrated the selective reactivity of NH3 with
four-coordinate complexes.47
By monitoring the reactions of M(EN-(py)2)2+, M(en)-
(phen)2+, and M(phen)2 complexes as a function of time,
2+
kinetic plots can be created. Figure 2 shows typical kinetic plots
for the reactions of Zn(EN-(py)2)2+ and Ni(EN-(py)2)2+ with
NH3. From these plots, it is evident that the reactions achieve
equilibrium after about 2000 ms. The establishment of an
equilibrium can be confirmed by isolating the product (or parent)
ion after a reaction time of 2000 ms and allowing it to react for
another 2000 ms. Upon doing so, the same product/parent ion
abundance ratio is reestablished. Table 1 shows that significant
differences in the kinetics and thermodynamics of these reactions
exist for different complexes. The kinetic behavior of the
reactions follows eq 1, and the reaction rate and equilibrium
constants can be determined by fitting the experimental data to
this equation.

1760 J. Phys. Chem. A, Vol. 108, No. 10, 2004
Combariza and Vachet
TABLE 2: Molecular Orbital Stabilization Energies (MOSE) Calculated from the Angular Overlap Model for High- and
Low-Spin Square Planar (D4h), Tetrahedral (Td), Square Pyramidal (C4W), and Trigonal Bipyramidal (D3h) Geometries,
Considering Only σ Contributions from the Ligands and NH3
MOSE (eσ)a
∆MOSE (eσ)
D4h
Td
C4V
D3h
D4h f C4V
D4h f D3h
Td f C4V
Td f D3h
High-Spin ML4 to High-Spin ML5
Mn
Fe
Co
Ni
Cu
Zn
-4
-4
-4
-4
-3
0
-4
-5
-5
-5
-5
-3
0
-5
-1
-1
-1
-1
0
-1
-1
-1
-1
-1
-4
-5
-1
-1
-4
-5
-1
-1
-2.67
-1.33
0
-3.88
-2.75
0
0.125
-2.33
-1.67
0
-1.21
-1.42
0
0.25
0
0
Low-Spin ML4 to Low-Spin ML5
Mn
Fe
Co
Ni
Cu
Zn
-8
-8
-7
-6
-3
0
-6.67
-5.33
-4
-2.67
-1.33
0
-10
-10
-8
-6
-3
0
-8.88
-7.75
-6.63
-5.5
-2.75
0
-2
-2
-1
0
0
0
-0.88
0.25
0.37
0.5
0.25
0
-3.33
-4.67
-4
-3.33
-1.67
0
-2.21
-2.42
-2.63
-2.83
-1.42
0
Low-Spin ML4 to High-Spin ML5
Mn
Fe
Co
Ni
Cu
Zn
-8
-8
-7
-6
-3
0
-6.67
-5.33
-4
-2.67
-1.33
0
-5
-5
-5
-5
-3
0
-5
3
3
2
1
0
0
3
3
2
2.12
0.25
0
1.67
0.33
-1
-2.33
-1.67
0
1.67
0.33
-1
-1.21
-1.42
0
-5
-5
-3.88
-2.75
0
High-Spin ML4 to Low-Spin ML5
Mn
Fe
Co
Ni
Cu
Zn
-4
-4
-4
-4
-3
0
-4
-10
-10
-8
-6
-3
0
-8.88
-7.75
-6.63
-5.5
-2.75
0
-6
-6
-4
-2
0
-4.88
-3.75
-2.63
-1.5
0.25
-6
-4.88
-3.75
-2.63
-2.83
-1.42
0
-4
-6
-4
-4
-2.67
-1.33
0
-3.33
-1.67
0
0
0
a MOSE values are given in terms of σ overlap integrals. The values of the overlap integrals are assumed to be the same for each metal in each
coordination environment.
become five-coordinate and can have trigonal-bipyramidal (D3h)
or square-pyramidal structures (C4V). The preference for these
geometric arrangements depends on the structural constraints
of the ligand, the electronic nature of the donor atoms (i.e., σ
or π donors), and the metal’s number of d electrons. When the
ligands are the same for a series of complexes, any differences
in geometric arrangements are usually caused by the number
of d electrons on the metal. In such cases, the angular overlap
model (AOM)66-69 can be used to obtain molecular orbital
stabilization energies (MOSE), which can be useful as a
qualitative tool in examining the effects of different geometries
on d-orbital splittings.
Angular Overlap Model Calculations. To understand better
the experimental trends observed for the four-coordinate com-
plexes in this study, the AOM was used to calculate MOSE
values for the reactant and product structures while considering
all possible geometries, metal ion spin states, and σ- and π-donor
contributions from the ligands. The MOSE values indicate the
degree to which the d orbitals on a given metal are stabilized
(or destabilized) relative to the free metal ion by a particular
geometric arrangement of ligand donor groups. The degree of
energetic stabilization (or destabilization) for each d orbital is
typically given in terms of σ-orbital overlap integrals (eσ), whose
magnitudes are related to the σ-bonding capacity of the ligand
donor groups. The difference in stabilization energy (∆MOSE)
between a product and reactant should be related to the reaction
∆G and thus the experimental equilibrium constant, assuming
that the σ-orbital overlap integrals (eσ) are similar for each metal
complex. This assumption should be fairly valid given that the
same ligand is bound to each metal and any systematic
differences between metals cancel upon subtracting the MOSE
value for the reactant from that of the product. Negative ∆MOSE
values indicate a greater degree of stabilization in the product
ion complex relative to the parent ion complex and thus a more
favorable reaction or higher reaction equilibrium constant (i.e.,
-∆G).
Table 2 shows the MOSE values for ideal high- and low-
spin square-planar, tetrahedral, square-pyramidal, and trigonal-
bipyramidal complexes and the ∆MOSE values for all of the
possible geometric transformations starting as four- and going
to five-coordinate complexes. Only the σ-bonding contributions
have been included in the MOSE values shown in Table 2
because including π contributions does not lead to a better fit
of the experimental data, which suggests that π interactions have
a minimal effect on the reactivity of these complexes. For all
of the geometries considered, metals with between five and eight
d electrons (i.e., Mn(II) through Ni(II)) can form either high-
or low-spin complexes. Four-coordinate complexes of these
transition metals can either be high or low spin depending on
the electronic structure of the metal center and the ligand’s
characteristics.70 Five-coordinate complexes of the same metals,
however, are usually high spin when complexed to ligands
having N-containing functional groups. Typically, ligands with
weakly electronegative π donors (e.g., P-containing functional
groups) are required to obtain low-spin five-coordinate
complexes.70-72 Nonetheless, both high-spin and low-spin
possibilities have been considered for the five-coordinate
complexes of Mn, Fe, Co, and Ni, and the MOSE values for
both cases are listed in Table 2.
The ∆MOSE values suggest that the differences in reactivity
between complexes, as seen in Table 1, can be explained by
differences in the geometries of the starting structures. Figure
3 shows a few combinations that help explain the experimental
trends. Figure 3 and Table 2 indicate that the relatively high

Reactivity of Divalent Metal Ion Complexes
J. Phys. Chem. A, Vol. 108, No. 10, 2004 1761
Figure 3. ∆MOSE values for selected transitions from four-coordinate
square-planar (D4h)/tetrahedral (Td) reactants to five-coordinate trigonal-
bipyramidal (D3h)/square-pyramidal (C4V) products.
equilibrium constants for the Mn, Fe, Co, and Zn complexes
can be explained by either D4h or Td geometries for the reactants
and either D3h or C4V geometries for the products because these
transitions lead to stabilized products (i.e., negative ∆MOSE
values). The reactivity of the four-coordinate Mn, Fe, and Co
complexes is not explained well, though, if they are considered
to be low-spin D4h complexes because the resulting ∆MOSE
values are positive. The low equilibrium constants for Ni(EN-
(py)2)2+, Ni(en)(phen)2+, and all of the Cu complexes, however,
can be explained only by D4h reactant geometries because
transitions from this geometry lead to less-stabilized products
(i.e., positive ∆MOSE values). In fact, four-coordinate com-
plexes of d8 metal ions such as Ni(II) are well known to be
stable as square-planar structures.73-75 The very high equilibrium
constant for Ni(phen)22+, however, can be understood by
realizing that a tetrahedral Ni complex is expected to be more
reactive than a square-planar Ni complex because of the greater
stabilization obtained upon the addition of a fifth donor group
to its coordination sphere. This result suggests that the two phen
ligands enforce a geometry close to tetrahedral for Ni whereas
EN-(py)2 and (en)(phen) allow Ni to adopt a more square-planar
geometry. Ligand-induced variations between tetrahedral and
square planar in Ni(II) complexes are well known. Holm et al.
reported geometric fluctuations in Ni(II) complexes as a function
of the ligand’s size. Basically, the introduction of bulky
functional groups into ligands around Ni can enforce a high-
spin tetrahedral structure whereas less-bulky functional groups
on the ligands around Ni usually result in square-planar,
diamagnetic geometries.76 Evidently, the aromatic rings of the
Figure 4. Kinetic plots for the reactions of (a) Mn(EN-(py)2)2+ and
(b) Ni(EN-(py)2)2+ with CH3CN.
in the gas-phase by the addition of a reagent molecule, then
this five-coordinate complex should readily react to form a six-
coordinate complex. Because NH3 is fairly unreactive with five-
coordinate complexes in the gas phase,47 another reagent is
needed to test this supposition. We observed previously that
CH3CN reacted readily with five-coordinate complexes,50 so
the EN-(py)2 complexes of each of the metals were reacted with
this reagent gas. Figure 4 shows the kinetic plots for the
reactions of Mn(EN-(py)2)2+ and Ni(EN-(py)2)2+ with CH3CN.
The main difference between the two plots is the relative
abundance of the species [M(EN-(py)2) + CH3CN]2+, which is
the five-coordinate complex that is necessarily formed before
the six-coordinate complex with two CH3CN molecules is
formed. In the case of the Mn complex (Figure 4) and the Fe
and Co complexes (data not shown), the five-coordinate complex
reaches almost 100% relative abundance before it starts forming
the product with two CH3CN molecules. The Ni complex,
however, behaves differently. The species [Ni(EN-(py)2) + CH3-
CN]2+ does not reach a maximum before adding the second
CH3CN molecule. In fact, the Ni complex adds the second
molecule of CH3CN at a faster rate than it adds the first.
2+
phen ligands provide enough steric bulk to force the Ni(phen)2
complex into a high-spin tetrahedral geometry, which explains
its high reactivity. In contrast, the ligands in the Ni(EN-(py)2)2+
and Ni(en)(phen)2+ complexes are less bulky and allow Ni to
adopt a more favorable low-spin square-planar geometry.
Furthermore, as mentioned above, five-coordinate metal com-
plexes of first-row transition metals with amine-based ligands
are usually high-spin species, and Table 2 shows that the
conversion from a low-spin D4h complex to either a high-spin
C4V or D3h complex is relatively unfavorable with positive
∆MOSE values. Thus, the reaction does not proceed very
extensively.
The decreasing reactivity of the Zn complex upon changing
the ligand is more difficult to explain using MOSE calculations
because the d10 configuration of Zn(II) prevents any ligand-
field stabilization. Instead, the reduced reactivity probably has
something to do with the tendency of four-coordinate Zn(II)
complexes to adopt tetrahedral geometries. The EN-(py)2 ligand
is certainly more strained than (en)(phen) or (phen)2 when trying
to adopt a tetrahedral geometry. The result is that the donor
atom orbitals of EN-(py)2 do not overlap as effectively with
Zn’s d orbitals, which leads to a more electropositive Zn center
that is more reactive. The two bidentate ligands in both (en)-
(phen) and (phen)2 provide greater flexibility, and its donor
The low equilibrium constants for Ni(EN-(py)2)2+ and Ni-
(en)(phen)2+ are also the consequence of the low stability of
five-coordinate Ni(II) complexes when compared to that of five-
coordinate complexes of Mn(II), Fe(II), and Co(II) (Table 1).
Indeed, MOSE calculations indicate that five-coordinate com-
plexes of Ni(II) should undergo facile reagent addition because
six-coordinate complexes of Ni(II) are stabilized to the same
extent as six-coordinate complexes of Mn(II), Fe(II), and Co-
(II).65 Given this observation, if a square-planar four-coordinate
Ni(II) complex can be converted to a five-coordinate complex

1762 J. Phys. Chem. A, Vol. 108, No. 10, 2004
Combariza and Vachet
TABLE 3: Geometric Descriptors (â, ω)a from Optimized
interactions between the pyridine rings. The distorted square-
planar structures for Ni(EN-(py)2)2+ and Ni(en)(phen)2+ are
consistent with their low reactivity, as suggested by the MOSE
calculations. Mn, Fe, Co, and Zn complexes are closer to
tetrahedral geometries, as the â and ω values indicate. Calcula-
Structures at the DFT/LANL2DZ Level
2+
M(EN-(py)2)2+
M(en)(phen)2+
M(phen)2
metalb
â
ω
â
ω
â
ω
Mn
Fe
Co
Ni
Cu
Zn
61.0
61.7
62.4
76.3 62.1
74.7 61.8
73.1 61.3
73.7
74.5
75.6
4.1
62.8 72.0
59.3 80.2
61.8 74.5
64.7 68.0
2+
tions of the M(phen)2 complexes lend support to the sup-
2+
position that Ni(phen)2 is tetrahedral and not square planar,
which explains its increased reactivity as noted from the MOSE
calculations. The hydrogen atoms on the 2 and 9 positions of
the phen ligands seem to provide steric interactions that prevent
Ni from adopting a preferred square-planar geometry. Interest-
ingly, though, the Cu complex of (phen)2 maintains a structure
that is close to square planar. The ionic radius of Ni(II) in
square-planar complexes is 0.49 Å, and for Cu(II), it is 0.57
Å.78 Evidently, the larger radius of Cu allows it to position the
phen rings further from each other so that the steric interactions
between the rings are reduced and the complex can attain a
(S ) 0) 79.6 29.1 (S ) 0) 88.6
77.5
62.8
34.8 87.0
72.1 61.4
8.61 76.2 38.3
75.4
62.0 73.9
a In degrees. b High-spin state unless indicated otherwise. S ) /2
5
3
for Mn(II), S ) 2 for Fe(II), S ) /2 for Co(II), S ) 1 for Ni(II), S )
1/2 for Cu(II), and S ) 0 for Zn(II).
2+
square-planar structure. The low reactivity of the Cu(phen)2
complex supports this idea. The MOSE calculations indicate
that square-planar Cu(II) complexes should have low reactivity
and tetrahedral Cu(II) complexes should be more reactive.
Conclusions
In summary, the results presented here indicate that I-M
reactions are sensitive to the geometry of certain four-coordinate,
gas-phase divalent metal-complex ions. For metals in which a
different degree of stabilization is expected depending upon
whether the complex is square planar or tetrahedral (e.g., Ni-
(II) and Cu(II)), the kinetics and thermodynamics of reagent
addition can indicate the geometry of the complex. Furthermore,
because four-coordinate Zn(II) complexes typically adopt
tetrahedral geometries, I-M reactions may be able to provide
geometric information for complexes of this metal because Zn
complexes that have difficulty adopting tetrahedral geometry
are likely to be more reactive than Zn complexes that can readily
adopt this geometry. Currently, our results appear to be limited
to simply distinguishing between the two extreme four-
coordinate geometries, but future work may make it possible
for I-M reactions to determine the degree of distortion in
tetrahedral or square-planar structures.
Determining the gas-phase coordination geometry of metal
complexes should provide some fundamentally useful informa-
tion. For example, a comparison of gas-phase structures with
solution-phase or solid-state structures might lend insight into
the effects of solvent or crystal-packing forces on metal-complex
structure. In addition, because the gas phase provides a unique
environment in which short-lived (e.g., reaction intermediates)
metal complexes can be studied, the ability to gather detailed
coordination structure information for such complexes might
provide some interesting insight into the chemical transforma-
tions that these species undergo.
Figure 5. Geometry-optimized structure of Cu(EN-(py)2)2+ indicating
the two angles used to describe â and ω. â is the average of θ1/2 and
θ2/2, and ω is defined by eq 2.
atoms can more easily orient themselves toward Zn, which leads
to reduced reactivity.
Density Functional Theory Calculations. To further confirm
these conclusions, density functional theory (DFT) calculations
were also performed on each of the complexes in Figure 1.
Because of the complexity of transition-metal-containing spe-
cies, effective core potentials (ECP) were used to perform
geometry optimizations. For complexes in which high-spin and
low-spin states are possible, both were calculated. In all cases
but two (i.e., Ni(EN-(py)2)2+ and Ni(en)(phen)2+), the high-
spin complexes were found to be considerably lower in energy.
The low-spin and high-spin states of Ni(EN-(py)2)2+ differed
by only 3.5 kJ/mol, which is within the error of the calculation,
whereas the low-spin state of Ni(en)(phen)2+ was found to be
67.3 kJ/mol lower in energy than the high-spin state. The
geometric characteristics of all of the complexes calculated are
displayed in Table 3. The square-planar or tetrahedral character
of a four-coordinate complex can be determined from the
parameters â and ω (Figure 5). â is defined as half of the
average angle formed by the central metal atom and the two
trans donor groups (i.e., N1-M-N3 and N2-M-N4 in Figure
5) and is equal to 90° for pure D4h geometry and 54.7° for pure
Td geometry.77 ω is the dihedral angle between the two planes
that contain the central metal ion and the two cis donor groups
and is equal to 0° for pure D4h geometry and 90° for pure Td
geometry.77 The relationship between â and ω is given by eq
2:
Acknowledgment. M.Y.C thanks the Schering-Plough Re-
search Institute for a summer research fellowship. Partial support
for this work came from the Office of Naval Research under
award no. N000140010796.
1 - 3 cos2 â
1 + cos2 â
cos ω ) (
(2)
References and Notes
(1) Colton, R.; D’Agostino, A.; Traeger, J. C. Mass Spectrom. ReV.
1995, 14, 79-106.
(2) Gatlin, C. L.; Turecˇek, F. In Electrospray Ionization Mass
Spectrometry; Cole, R. B., Ed.; John Wiley: New York, 1997; pp 527-
570.
DFT calculations of M(EN-(py)2)2+ and M(en)(phen)2+ show
that the Ni(II) and Cu(II) complexes are the closest to square-
planar geometries with the highest â values and the lowest ω
values (Table 3). The deviations from planarity in these
complexes are caused mainly by ligand strain and steric
(3) Stewart, I. I. Spectrochim. Acta, Part B 1999, 54, 1649-1695.
(4) Fisher, K. J. Prog. Inorg. Chem. 2001, 50, 343-432.

Reactivity of Divalent Metal Ion Complexes
J. Phys. Chem. A, Vol. 108, No. 10, 2004 1763
(5) Gas-Phase Inorganic Chemistry; Russell, D. H., Ed.; Plenum
Press: New York, 1989.
(6) Armentrout, P. B.; Beauchamp, J. L. Acc. Chem. Res. 1989, 22,
315-321.
(46) Duncan, M. A. Int. J. Mass Spectrom. 2000, 200, 545-569.
(47) Vachet, R. W.; Hartman, J. R.; Callahan, J. H. J. Mass Spectrom.
1998, 33, 1209-1225.
(48) Vachet, R. W.; Callahan, J. H. J. Mass Spectrom. 2000, 35, 311-
(7) Eller, K.; Schwarz, H. Chem. ReV. 1991, 91, 1121-1177.
(8) Weisshaar, J. C. Acc. Chem. Res. 1993, 26, 213-219.
(9) Eller, K. Coord. Chem. ReV. 1993, 126, 93-147.
(10) Freiser, B. S. Acc. Chem. Res. 1994, 27, 353-360.
(11) Armentrout, P. B. Acc. Chem. Res. 1995, 28, 430-436.
(12) Plattner, D. A. Int. J. Mass Spectrom. 2001, 207, 125-144.
(13) Hinderling, C.; Adlhart, P.; Chen, P. Angew. Chem., Int. Ed. 1998,
37, 2685-2689.
320.
(49) Vachet, R. W.; Hartman, J. R.; Gertner, J. W.; Callahan, J. H. Int.
J. Mass Spectrom. 2001, 204, 101-112.
(50) Combariza, M. Y.; Vachet, R. W.; J. Am Soc. Mass Spectrom. 2002,
13, 813-825.
(51) Zhang, X.; Cassady, C. J. J. Am. Soc. Mass Spectrom. 1996, 7,
1211-1218.
(52) Bartmess, J. E.; Georgiadis, R. M. Vacuum 1983, 33, 149-153.
(53) Shen, N. Z.; Pope, R. M.; Dearden, D. V. Int. J. Mass Spectrom.
2000, 196, 639-652.
(14) Adlhart, C.; Volland, M. A. O.; Hofmann, P.; Chen, P. HelV. Chim.
Acta 2000, 83, 3306-3311.
(15) Volland, M. A. O.; Adlhart, C.; Kiener, C. A.; Chen, P.; Hofmann,
P. Chem.sEur. J. 2001, 7, 4621-4632.
(54) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A., Jr.;
Stratmann, R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels, A.
D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, V.; Cossi,
M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford, S.;
Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.; Malick,
D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Cioslowski, J.;
Ortiz, J. V.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi,
I.; Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.;
Peng, C. Y.; Nanayakkara, A.; Gonzalez, C.; Challacombe, M.; Gill, P. M.
W.; Johnson, B. G.; Chen, W.; Wong, M. W.; Andres, J. L.; Head-Gordon,
M.; Replogle, E. S.; Pople, J. A. Gaussian 98, revision A.3; Gaussian,
Inc.: Pittsburgh, PA, 1998.
(16) Adlhart, C.; Chen, P. HelV. Chim. Acta 2003, 86, 941-949.
(17) Hinderling, C.; Plattner, D. A.; Chen, P. Angew. Chem., Int. Ed.
Engl. 1997, 36, 243-244.
(18) Hinderling, C.; Feichtinger, D.; Plattner, D. A.; Chen, P. J. Am.
Chem. Soc. 1997, 119, 10793-10804.
(19) Gerdes, G.; Chen, P. Organometallics 2003, 22, 2217-2225.
(20) Feichtinger, D.; Plattner, D. A. Angew. Chem., Int. Ed. Engl. 1997,
36, 1718-1719.
(21) Plattner, D. A.; Feichtinger, D.; El-Bahraoui, J.; Wiest, O. Int. J.
Mass Spectrom. 2000, 195, 351-362.
(22) Waters, T.; O’Hair, R. A. J.; Wedd, A. G. J. Am. Chem. Soc. 2003,
125, 3384-3396.
(55) Pople, J. A.; Head-Gordon, M.; Fox, D. J.; Raghavachari, K.;
Curtiss, L. A. J. Chem. Phys. 1989, 90, 5622-5629.
(56) Curtiss, L. A.; Jones, C.; Trucks, G. W.; Raghavachari, K.; Pople,
J. A. J. Chem. Phys. 1990, 93, 2537-2545.
(23) Kebarle, P. Annu. ReV. Phys. Chem. 1977, 28, 445-476.
(24) Keesee, R. G.; Castleman, A. W. J. Phys. Chem. Ref. Data 1986,
15, 1011-1071.
(25) Kebarle, P. Int. J. Mass Spectrom. 2000, 200, 313-330.
(26) Stace, A. J. Phys. Chem. Chem. Phys. 2001, 3, 1935-1941.
(27) Stace, A. J. J. Phys. Chem. A 2002, 106, 7993-8005.
(28) Hu, P.; Gross, M. L. J. Am. Chem. Soc. 1993, 115, 8821-8828.
(29) Reiter, A.; Adams, J.; Zhao, H. J. Am. Chem. Soc. 1994, 116, 7827-
7838.
(57) Becke, A. D. J. Chem. Phys. 1993, 98, 5648-5652.
(58) Dunning, T. H., Jr.; Hay, P. J. In Modern Theoretical Chemistry;
Schaefer, H. F., III, Ed.; Plenum: New York, 1976; Vol. 3, pp 1-28.
(59) Hay, P. J.; Wadt, R. J. Chem. Phys. 1985, 82, 270-283.
(60) Hay, P. J.; Wadt, R. J. Chem. Phys. 1985, 82, 284-298.
(61) Hay, P. J.; Wadt, R. J. Chem. Phys. 1985, 82, 299-310.
(62) Hartman, J. R.; Vachet R. W.; Callahan, J. H. Inorg. Chim. Acta
2000, 297, 79-87.
(30) Hopfgartner, G.; Piguet, C.; Henion, J. D. J. Am. Soc. Mass
Spectrom. 1994, 5, 748-756.
(31) Hu, P.; Loo, J. A. J. Am. Chem. Soc. 1995, 117, 11314-11319.
(32) Gatlin, C. L.; Turecek, F.; Vaisar, T. J. Am. Chem. Soc. 1995, 117,
3637-3638.
(63) Hartman, J. R.; Vachet, R. W.; Pearson, W.; Wheat, R. J.; Callahan,
J. H. Inorg. Chim. Acta 2003, 343, 119-132.
(64) Hartman, J. R.; Combariza, M. Y.; Vachet, R. W. Inorg. Chim.
Acta 2004, 357, 51-58.
(33) Alvarez, E. J.; Wu, H.-F.; Liou, C. C.; Brodbelt, J. J. Am. Chem.
Soc. 1996, 118, 9131-9138.
(65) Combariza, M. Y.; Fermann, J. T.; Vachet, R. W. Submitted to
Inorg. Chem.
(34) Armentrout, P. B. Acc. Chem. Res. 1995, 28, 430-436.
(35) Rodgers, M. T.; Armentrout, P. B. Mass Spectrom. ReV. 2000, 19,
215-247.
(66) Gerloch, M.; Woolley, R. G. Prog. Inorg. Chem. 1983, 31, 371-
446.
(36) Armentrout, P. B. Int. J. Mass Spectrom. 2003, 227, 289-302.
(37) Rodriguez-Cruz, S. E.; Jockusch, R. A.; Williams, E. R. J. Am.
Chem. Soc. 1998, 120, 5842-5843.
(67) Schaffer, C. E.; Jørgensen, C. K. Mol. Phys. 1965, 9, 401.
(68) Woolley, R. G. Mol. Phys. 1981, 42, 703-720.
(69) Jørgensen, C. K. Modern Aspects of Ligand Field Theory; North-
Holland Publishing Company: Amsterdam, 1971.
(70) Sacconi, L.; Speroni, G. P.; Morassi, R. Inorg. Chem. 1968, 7,
1521-1525.
(71) Sacconi, L. J. Chem. Soc. A 1970, 248-256.
(72) Morassi, R.; Bertini, I.; Sacconi L. Coord. Chem. ReV. 1972, 8,
351-367.
(73) Weber, J. H. Inorg. Chem. 1967, 6, 258-262.
(74) Manson, W. R., III; Gray, H. R. J. Am. Chem. Soc. 1968, 90, 5721-
5729.
(75) Delong, Z.; Busch, D. H. Inorg. Chem. 1994, 33, 5138-5143.
(76) Holm, R. H.; Chakravorty, A.; Theriot, L. J. Inorg. Chem. 1966,
5, 625-635.
(38) Stevens, S. M.; Dunbar, R. C.; Price, W. D.; Sena, M.; Watson, C.
H.; Nichols, L. S.; Riveros, J. M.; Richardson, D. E.; Eyler, J. R. J. Phys.
Chem. A 2002, 106, 9686-9694.
(39) Ranasinghe, Y. A.; Surjasasmita, I. B.; Freiser, B. S. In Organo-
metallic Ion Chemistry; Freiser, B. S., Ed.; Kluwer Academic: Dordrecht,
The Netherlands, 1996; pp 229-258.
(40) Burns, T. D.; Spence, T. G.; Mooney, M. A.; Posey, L. A. Chem.
Phys. Lett. 1996, 258, 669-679.
(41) Duncan, M. A. Annu. ReV. Phys. Chem. 1997, 48, 69-93.
(42) Spence, T. G.; Burns, T. D.; Guckenberer, G. B.; Posey, L. A. J.
Phys. Chem. 1997, 101, 1081-1092.
(43) Husband, J.; Aguirre, F.; Ferguson, P.; Metz, R. B. J. Chem. Phys.
1999, 111, 1433-1437.
(77) Rybak-Akimova, E. V.; Nazarenko, A. Y.; Chen., L.; Krieger, P.
W.; Herrera, A. M.; Tarasov, V. V.; Robinson, P. D. Inorg. Chim. Acta
2001, 324, 1-15.
(44) Husband, J.; Aguirre, F.; Thompson, C. J.; Laperle, C. M.; Metz,
R. B. J. Phys. Chem. A 2000, 104, 2020-2025.
(45) Puskar, L.; Barran, P. E.; Wright, R. R.; Kirkwood, D. A.; Stace,
A. J. J. Chem. Phys. 2000, 112, 7751-7754.
(78) Shannon, R. D. Acta Crystallogr., Sect. A 1976, 32, 751-767.