OMS, OM(η2-SO), and OM(η2-SO) (η2-SO2) molecules (M = Ti, Zr, Hf): Infrared spectra and density functional calculations

Article abstract of DOI:10.1021/ic3008987

Infrared spectra of the matrix isolated OMS, OM(η²-SO), and OM(η²-SO)(η²-SO₂) (M = Ti, Zr, Hf) molecules were observed following laser-ablated metal atom reactions with SO₂ during condensation in soli

Full text of DOI:10.1021/ic3008987

Article
pubs.acs.org/IC
OMS, OM(η2-SO), and OM(η2-SO)(η2-SO2) Molecules (M = Ti, Zr, Hf):
Infrared Spectra and Density Functional Calculations
Xing Liu,† Xuefeng Wang,*,† Qiang Wang,† and Lester Andrews*,‡
†Department of Chemistry, Tongji University, Shanghai, 200092, China
‡Department of Chemistry, University of Virginia, Charlottesville, Virginia 22904, United States
ABSTRACT: Infrared spectra of the matrix isolated OMS, OM(η2-SO), and
OM(η2-SO)(η2-SO2) (M = Ti, Zr, Hf) molecules were observed following
laser-ablated metal atom reactions with SO2 during condensation in solid
argon and neon. The assignments for the major vibrational modes were
confirmed by appropriate S18O2 and 34SO2 isotopic shifts, and density
functional vibrational frequency calculations (B3LYP and BPW91). Bonding
in the initial OM(η2-SO) reaction products and in the OM(η2-SO)(η2-SO2)
adduct molecules with unusual chiral structures is discussed.
favor Cs structures for group 6 metals while the SUO2 molecule is
stabilized in a C2v structure, which reflects the difference between
d and f orbital participation in bonding and polarization of the
outermost core shells.
In this paper we present a matrix infrared spectroscopic study
of laser-ablated titanium, zirconium, and hafnium atom reaction
products with SO2 in solid argon and neon. The new metal
sulfido oxides OM(η2-SO) and its SO2 adduct OM(η2-SO)-
(η2-SO2), and the simple sulfide-oxide OMS (M = Ti, Zr, Hf)
are identified from matrix isolation infrared spectroscopy with
isotopic substitution and theoretical vibrational frequency
calculations.
INTRODUCTION
■
Conversion of SO2 to SO3 is an important issue for the sulfuric
acid industry. Recent study shows that oxidation of SO2 to form
SO3 on vanadium oxide clusters reduces SO2 to SO first, which
then reacts with O2 to give SO3. Certain oxygen-rich metal oxide
clusters provide one oxygen atom through an oxygen-detaching
process followed by the reaction of O with SO2 to give SO3.1,2
A
systematic study on the adsorption and dissociation of SO2 on
different metal-based surfaces has been performed.3−15 Both the
pure metal (Zn, Mo, Mg, Sn)12−15 and alkali or transition metal
doped metal oxide surfaces3−8,10,11 are higher in efficiency for
S−O bond cleavage, while in contrast, the MxOy (M = Sc, Ti, V,
Cr, Fe, Ni, Cu, Zn, Ce, Ba) surfaces are almost inactive
instead.3−8,10,11 Such observations combined with subsequent
calculations confirm that effective S−O bond breaking only
occurs in a proper electronic state on the target surface that is
near the lowest unoccupied molecular orbital (LUMO) of SO2;
thus, the charge transfer between the metal surface and SO2 plays
vital role.
Coordination chemistry in and of sulfur dioxide have been
investigated, and different bonding fashions of SO2 coordination
to transition metal center have been found.16 Several transition
metal cations and p-block metals coordinate SO2 via oxygen to
form η1-O- complexes16b,c while for alkali metal−sulfur dioxide
complexes O, O′-bridging SO2 ligands have been character-
ized.16a,d
There has been extensive investigation of metal oxide
molecules MOx and oxygen complexes M(O2)n by matrix
infrared and electron spin resonance (ESR) spectroscopy;17−24
however, mixed sulfur and oxygen analogues of these complexes
have rarely been reported. Only recently laser-ablated uranium
and group 6 metal atom reactions with SO2 have been
investigated,25,26 which produced very stable SMO2 (Cr, Mo,
W, U) molecules that have been identified in neon and argon
matrixes by infrared spectroscopy and reproduced by theoretical
vibrational frequency calculations. Notice that SMO2 molecules
EXPERIMENTAL AND COMPUTATIONAL
METHODS
■
The experimental apparatus for investigating laser-ablated transition metal
atom reactions with small molecules during condensation in excess argon or
neon at 4 K has been described in our previous papers.27,28 The Nd:YAG
laser fundamental (1064 nm, 10 Hz repetition rate with 10 ns pulse width)
was focused onto a rotating metal target, which gave a bright plume
spreading uniformly to the cold CsI window. The titanium, zirconium, and
hafnium targets (Alfa Aesar) were polished to remove oxide coating and
immediately placed in the vacuum chamber. A mixture of S18O2 and
S
16,18O2 with a trace of S16O2 was prepared by tesla coil discharge of 18O2
(>99%, SRICI) (about 2 Torr) in a 0.5 L pyrex bulb containing 10 mg of
sulfur powder (99.5%, Alfa Aesar) sublimed onto the walls heated by
external hot air (>450 °C). The sample of 34SO2 was similarly prepared
from sulfur-34 (98.8% 34S, CIL) and oxygen (99.999%, BOC). The laser
energy was varied about 10−20 mJ/pulse. FTIR spectra were recorded at
0.5 cm−1 resolution on a Bruker Vertex 80 V or Nicolet 750 instruments
with 0.1 cm−1 accuracy using an MCTB detector. Matrix samples were
annealed at different temperatures, and selected samples were subjected to
photolysis by a medium pressure mercury arc lamp (Philips, 175W) with
globe removed.
Received: May 3, 2012
Published: June 20, 2012
© 2012 American Chemical Society
7415
dx.doi.org/10.1021/ic3008987 | Inorg. Chem. 2012, 51, 7415−7424

Inorganic Chemistry
Article
Table 1. Infrared Absorptions (cm−1) Observed for Products of the Reaction of Ti, Zr, Hf Atoms and SO2 Molecule
SO2/Ar
S18O2/Ar
34SO2/Ar
SO2/Ne
S18O2/Ne
34SO2/Ne
assignment
Ti
Zr
Hf
953.2
913.1
953.2
978.0
987.4
987.4
749.2
1015.8
973.5
961.9
613.7
536.1
978.0
STiO
−
SO2
970.9
747.3
1008.6
930.8
717.3
966.8
970.9
744.3
1008.6
944.3
719.5
985.6
939.6
925.6
583.0
987.4
746.3
1015.8
965.4
946.8
611.3
536.3
OTi(η2-SO)
OTi(η2-SO)
OTi(η2-SO)(η2-SO2)
OTi(η2-SO)(η2-SO2)
OTi(η2-SO)(η2-SO2)
OTi(η2-SO)(η2-SO2)
OTi(η2-SO)(η2-SO2)
612.4
582.0
610.2
877.4
889.1
702.7
553.2
915.5
835.1
846.4
674.3
528.8
871.7
877.4
889.1
698.3
SZrO
905.3
696.8
557.3
925.3
978.4
963.1
886.1
591.6
860.8
905.3
691.8
554.6
925.3
968.8
952.1
878.0
587.5
OZr(η2-SO)
OZr(η2-SO)
OZr(η2-SO)
915.5
881.1
943.8
933.3
852.8
563.5
OZr(η2-SO)(η2-SO2)
OZr(η2-SO)(η2-SO2)
OZr(η2-SO)(η2-SO2)
OZr(η2-SO)(η2-SO2)
OZr(η2-SO)(η2-SO2)
586.2
561.7
876.9
882.7
688.9
903.8
831.4
836.7
658.9
857.2
876.9
882.7
685.3
903.8
SHfO
897.0
685.9
913.4
973.5
960.6
875.7
608.0
850.2
651.1
866.3
938.2
927.2
841.2
577.2
897.0
683.2
913.4
960.3
951.6
869.7
605.3
OHf(η2-SO)
OHf(η2-SO)
OHf(η2-SO)(η2-SO2)
OHf(η2-SO)(η2-SO2)
OHf(η2-SO)(η2-SO2)
OHf(η2-SO)(η2-SO2)
OHf(η2-SO)(η2-SO2)
607.4
563.3
602.8
Figure 1. Infrared spectra for the titanium atom and SO2 reaction products in solid argon at 4 K. (a) Ti + SO2 deposition for 60 min; (b) after annealing
to 20 K; (c) after >220 nm irradiation; (d) after annealing to 35 K; (e) Ti + 34SO2 deposition for 60 min; (f) after annealing to20 K; (g) after >220 nm
irradiation; (h) after annealing to 35 K; (i) Ti+ S16,18O2 deposition for 60 min; (j) after annealing to 20 K; (k) after >220 nm irradiation; (l) after
annealing to 35 K.
Complementary density functional theory (DFT) calculations were
RESULTS
■
performed using the Gaussian 03 program,37 and the B3LYP functional, the
Infrared spectra of laser ablated titanium, zirconium, and hafnium
6-311++G(3df, 3pd) basis set for sulfur and oxygen atoms, and the SDD
pseudo potential for titanium, zirconium, and hafnium38−40 were employed.
All of the geometrical parameters were fully optimized, and the harmonic
vibrational frequencies were obtained analytically at the optimized structures.
atom reaction products with SO2 in excess argon or neon during
condensation at 4 K will be presented in turn, and the new
product absorptions are listed in Table 1. Density functional
7416
dx.doi.org/10.1021/ic3008987 | Inorg. Chem. 2012, 51, 7415−7424

Inorganic Chemistry
Article
Figure 2. Infrared spectra for the titanium atom and SO2 reaction products in solid neon at 4 K. (a) Ti + SO2 deposition for 45 min; (b) after annealing to
8 K; (c) after >220 nm irradiation; (d) after annealing to 10 K; (e) Ti + 34SO2 deposition for 45 min; (f) after annealing to 8 K; (g) after >220 nm
irradiation; (h) after annealing to 10 K; (i) Ti + S16,18O2 deposition for 45 min; (j) after annealing to 8 K; (k) after >220 nm irradiation; (l) after
annealing to 10 K.
Figure 3. Infrared spectra for the zirconium atom and SO2 reaction products in solid argon at 4 K. (a) Zr + SO2 deposition for 60 min; (b) after annealing
to 20 K; (c) after >220 nm irradiation; (d) after annealing to 35 K; (e) Zr + 34SO2 deposition for 60 min; (f) after annealing to20 K; (g) after >220 nm
irradiation; (h) after annealing to 35 K; (i) Zr + S16,18O2 deposition for 60 min; (j) after annealing to 20 K; (k) after >220 nm irradiation; (l) after
annealing to 35 K.
calculations were performed to support the identification of new
reaction products. Common species in these experiments, such as
SO, SO2− and metal oxide absorptions, have been identified in
previous papers.17,24,29−31 Experiments were also done with S18O2
and 34SO2, and the isotopic product frequencies are listed in Table 1.
Ti + SO2. Infrared spectra of laser-ablated titanium atoms co-
deposited with SO2 in argon are shown in Figure 1. Strong bands
at 970.9 cm−1 with matrix sites at 973.6 and 976.1 cm−1 and at
917.0 cm−1, two medium intensity bands at 953.2, 864.5 cm−1, and a
weak band at 946.9 cm−1 were observed in the TiO stretching
region after deposition. These bands increased on annealing to 20 K
by 50% and broadened by full-arc irradiation, which, additionally,
produced a new absorption at 1008.6 cm−1 in the same region.
Further annealing to 35 K increased the 970.9, 953.2, 946.9, and 917.0
cm−1 bands dramatically. When neon was emloyed as a host matrix
for similar reactions, new bands were observed at 987.4, 978.0, 936.5,
and 879.3 cm−1 (Figure 2). Another three additional bands were
observed at 1015.8, 973.5, and 961.9 cm−1 after deposition, which
exhibited little changes on annealing, but increased markedly at the
expense of band at 970.9 cm−1 on irradiation.
Zr + SO2. Figure 3 shows laser ablated Zr atom reactions with
SO2 in excess argon. New absorptions were observed centered at
889.1 and 886.3 cm−1 in the ZrO stretching region on deposition.
These bands increased dramatically on annealing to 20 K, broadened
7417
dx.doi.org/10.1021/ic3008987 | Inorg. Chem. 2012, 51, 7415−7424

Inorganic Chemistry
Article
Figure 4. Infrared spectra for the zirconium atom and SO2 reaction products in solid neon at 4 K. (a) Zr + SO2 deposition for 45 min; (b) after annealing
to 8 K; (c) after >220 nm irradiation; (d) after annealing to 10 K; (e) Zr + 34SO2 deposition for 45 min; (f) after annealing to 8 K; (g) after >220 nm
irradiation; (h) after annealing to 10 K; (i) Zr + S16,18O2 deposition for 45 min; (j) after annealing to 8 K; (k) after >220 nm irradiation; (l) after
annealing to 10 K.
Figure 5. Infrared spectra for the hafnium atom and SO2 reaction products in solid argon at 4 K. (a) Hf + SO2 deposition for 60 min; (b) after annealing
to 20 K; (c) after >220 nm irradiation; (d) after annealing to 35 K; (e) Hf + 34SO2 deposition for 60 min; (f) after annealing to 20 K; (g) after >220 nm
irradiation; (h) after annealing to 35 K; (i) Hf + S16,18O2 deposition for 60 min; (j) after annealing to 20 K; (k) after >220 nm irradiation; (l) after
annealing to 35 K.
on full-arc irradiation, and further increased on annealing to 30 and
35 K. In addition, a photosensitive band at 915.5 cm−1 was also
observed. Similar experiments in neon matrix are illustrated in
Figure 4. Strong new bands at 978.4, 963.1, 925.3, 905.3, and 886.1 cm−1
appear on initial deposition and the sharp absorption at 905.3 cm−1
increased upon annealing and decreased on subsequent irradiation.
These new absorptions are listed in Table 1.
Hf + SO2. The reaction product spectra of hafnium with SO2
in solid argon and neon are shown in Figures 5 and 6, and
absorptions are listed in Table 1. One strong band at 882.7 cm−1
tracking with a weak peak at 688.9 cm−1 were the major product
absorptions in solid argon, which appeared on deposition and
increased markedly on annealing. New features were found at 973.5,
960.6, and 913.4 cm−1 in solid neon, which doubled on broadband
irradiation but a 897.0 cm−1 band shows a substantial decrease
(Figure 6).
Calculations. Figure 7 illustrates the structures calculated
with B3LYP and BPW91 functionals for the insertion products
OMS and OM(η2-SO) (M = Ti, Zr, Hf), which were optimized
to singlet ground states. The MO and M−S bond lengths
for OMS molecules increase while the OMS and OMO bond
angles decrease from titanium to zirconium to hafnium. These
7418
dx.doi.org/10.1021/ic3008987 | Inorg. Chem. 2012, 51, 7415−7424

Inorganic Chemistry
Article
Figure 6. Infrared spectra for the hafnium atom and SO2 reaction products in solid neon at 4 K. (a) Hf + SO2 deposition for 45 min; (b) after annealing to 8 K;
(c) after >220 nm irradiation; (d) after annealing to 10 K; (e) Hf + 34SO2 deposition for 45 min; (f) after annealing to 8 K; (g) after >220 nm irradiation; (h)
after annealing to 10 K; (i) Hf + S16,18O2 deposition for 45 min; (j) after annealing to 8 K; (k) after >220 nm irradiation; (l) after annealing to 10 K.
Figure 7. Optimized structures for SMO, OM(η2-SO), and OM(η2-SO)(η2-SO2) molecules based on B3LYP and BPW91 functional calculations (bond
lengths in angstrom and bond angles in degree).
calculated structures are very similar to group 4 MO3 molecules,
but different pyramidal structures were observed in reactions
of group 6 metal atoms with SO2.25,32 For the MO(η2-SO)
molecules the SO subunit is coordinated to MO, and the
calculated trends of bond lengths and bond angles are the same as
for the SMO molecules. One metal atom reaction with two SO2
molecules gave the OM(η2-SO)(η2-SO2) adducts, which were
calculated to have triplet ground electronic states. Calculated
frequencies for the OMS, OM(η2-SO), and OM(η2-SO)(η2-
SO2) reaction products are listed in Tables 2−4, and natural
7419
dx.doi.org/10.1021/ic3008987 | Inorg. Chem. 2012, 51, 7415−7424

Inorganic Chemistry
Article
a
Table 2. Calculated Frequencies for OM(η2-SO) (1A) Isotopic Molecules (M = Ti, Zr, Hf)
B3LYP
18-O
BPW91
16-O
16-O/34-S
16-O
18-O
mode assignment
OTi(η2-SO)
OZr(η2-SO)
OHf(η2-SO)
149.3(17.5)
237.5(30.9)
421.3(39.6)
582.3(14.2)
770.8(108.1)
1040.9(338.1)
144.5(16.2)
226.9(28.2)
418.9(38.7)
561.3(12.4)
740.2(98.7)
997.1(338.1)
148.1(17.4)
237.5(30.9)
417.0(39.1)
578.2(14.4)
767.8(108.2)
1040.9(338.0)
140.4(15.4)
228.1(27.6)
416.7(34.7)
570.0(14.2)
747.6(88.7)
1002.7(275.2)
135.8(14.3)
217.9(25.1)
412.8(33.9)
549.6(12.7)
718.0(80.8)
960.5(258.7)
S−Ti−O bend
O−Ti−O bend
Ti−SO str
STiO def
S−O str,
TiO str
139.6(15.8)
211.0(28.7)
375.7(35.4)
552.4(27.3)
708.2(95.8)
920.9(260.7)
134.7(14.3)
200.6(26.0)
372.3(34.2)
529.4(25.3)
678.5(85.9)
875.5(239.5)
138.4(15.8)
210.8(28.6)
368.3(34.5)
549.2(26.9)
704.0(96.3)
920.1(259.8)
137.7(14.4)
207.0(24.1)
375.1(31.3)
539.4(24.5)
693.9(80.0)
894.6(215.1)
132.7(13.0)
196.8(21.8)
371.8(30.2)
516.8(23.0)
665.1(71.5)
850.6(197.6)
S−Zr−O bend
O−Zr−O bend
Zr−SO str
SZrO def
S−O str
ZrO str
140.2(15.9)
200.9(26.4)
358.5(30.0)
541.8(24.9)
682.5(86.9)
888.8(193.9)
134.8(14.1)
190.7(23.7)
355.9(29.5)
519.7(22.1)
651.9(77.6)
842.4(176.4)
139.2(16.0)
201.0(26.4)
351.3(29.1)
538.1(25.2)
679.7(87.5)
889.1(194.4)
136.5(15.0)
200.2(21.4)
358.0(27.0)
527.7(20.8)
666.5(72.4)
861.5(157.5)
131.0(13.4)
190.1(19.2)
355.5(26.5)
506.1(18.6)
636.9(64.5)
816.6(143.2)
S−Hf−O bend
O−Hf−O bend
Hf−SO str
SHfO def
S−O str
HfO str
a
Frequencies and intensities (in parentheses) are in cm−1 and km mol−1.
a
Table 3. Calculated Frequencies for OM(η2-SO)(η2-SO2) (3A) Isotopic Molecules (M = Ti, Zr, Hf)
B3LYP
18-O
BPW91
16-O
16-O/34-S
OTi(η2-SO)(η2-SO2)
16-O
18-O
mode assignment
370.4(32.6)
535.1(112.4)
612.1(171.0)
907.0(30.0)
953.9(150.8)
973.8(143.4)
1072.9(221.5)
360.0(29.0)
518.8(104.3)
582.4(161.4)
872.3(28.7)
918.9(138.9)
937.5(135.9)
1027.8(206.0)
368.6(31.6)
534.6(108.8)
610.3(176.8)
898.4(29.2)
944.0(148.3)
964.2(138.3)
1072.8(222.7)
373.6(14.2)
468.9(22.1)
552.6(15.1)
816.0(165.3)
924.2(90.2)
1016.2(185.8)
1020.5(109.9)
369.8(13.8)
451.0(20.3)
528.4(14.1)
785.1(150.8)
887.2(82.1)
973.6(176.8)
984.5(99.7)
Ti−SO str
Ti−O str
OSO bend
S−O str
SO str
SO str
TiO str
OZr(η2-SO)(η2-SO2)
327.5(26.4)
465.6(106.9)
590.3(113.2)
896.2(36.2)
939.4(247.3)
956.6(138.1)
977.0(140.1)
316.3(30.2)
445.6(104.5)
560.7(97.4)
861.9(34.4)
894.1(220.3)
920.9(127.4)
940.7(136.7)
325.9(23.4)
464.8(105.0)
588.2(116.4)
887.6(35.2)
936.7(245.1)
948.2(142.7)
967.4(130.9)
327.3(20.2)
451.0(85.8)
559.4(94.9)
865.2(33.2)
898.3(167.1)
913.7(133.3)
930.1(110.3)
316.4(21.0)
431.6(84.0)
531.2(82.2)
832.1(32.8)
859.9(169.3)
874.6(96.8)
895.8(110.7)
Zr−SO str
Zr−O str
OSO bend
S−O str
ZrO str
SO str
SO str
OHf(η2-SO)(η2-SO2)
325.9(5.3)
308.6(5.7)
325.9(5.3)
320.9(4.7)
304.1(5.3)
Hf−SO str
Hf−O str
OSO bend
S−O str
448.9(75.2)
599.5(96.3)
885.0(41.1)
905.8(178.4)
955.2(150.5)
970.4(141.1)
426.5(71.7)
569.2(81.6)
850.2(58.8)
859.8(139.4)
920.3(136.4)
934.4(136.4)
449.1(75.4)
597.4(99.3)
876.6(37.5)
905.9(183.6)
945.2(148.9)
960.8(134.2)
431.8(57.4)
567.2(77.0)
847.7(38.8)
871.3(126.3)
904.8(107.6)
921.6(117.0)
410.2(54.8)
538.5(65.3)
813.5(52.7)
828.2(95.1)
871.0(97.1)
888.1(113.8)
HfO str
SO str
SO str
a
Frequencies and intensities (in parentheses) are in cm−1 and km mol−1.
charge and Mulliken atomic spin densities for these molecules
are given in Tables 5 and 6.
during condensation at 4 K will be presented. Isotopic
substitution and theoretical calculations were performed to
support the identifications of new metal sulfide oxides.
MO(η2-SO) (M = Ti, Zr, Hf). The laser-ablated Ti atom
reaction with SO2 in solid argon produced a stronger band at
970.9 cm−1 and a weaker band at 747.3 cm−1, which show the
DISCUSSION
■
Infrared spectra of products formed in the reactions of laser
ablated Ti, Zr, and Hf atoms with SO2 in excess argon and neon
7420
dx.doi.org/10.1021/ic3008987 | Inorg. Chem. 2012, 51, 7415−7424

Inorganic Chemistry
Article
a
Table 4. Calculated Frequencies of Isotopic SMO(1A′) Molecules (M = Ti, Zr, Hf)
B3LYP
BPW91
16−32
18−32
16−34
16−32
18−32
mode assignment
STiO
SZrO
SHfO
247.4(6.5)
239.4(6.0)
245.6(6.3)
248.2(4.8)
240.1(4.4)
STiO bend
TiS str
577.6(71.5)
1034.5(308.0)
577.2(70.2)
990.5(286.9)
567.9(69.8)
1034.5(308.3)
559.9(58.7)
999.1(246.2)
559.6(57.6)
956.6(229.4)
TiO str
219.9(5.8)
211.9(5.2)
218.3(5.8)
218.8(4.1)
210.9(3.6)
SZrO bend
ZrS str
501.3(66.5)
917.5(235.2)
500.3(65.7)
872.3(214.2)
489.9(64.0)
916.7(235.0)
493.2(54.8)
893.4(191.8)
492.3(54.1)
849.4(174.7)
ZrO str
206.6(6.9)
198.7(6.0)
205.4(6.9)
205.6(4.7)
198.0(4.0)
SHfO bend
HfS str
464.7(41.6)
889.7(175.7)
465.0(41.5)
843.4(158.5)
453.6(40.0)
890.0(176.3)
459.3(34.6)
863.8(139.8)
459.6(34.5)
818.9(126.1)
HfO str
a
Frequencies and intensities (in parentheses) are in cm−1 and km mol−1.
Table 5. Natural Charge from NPA Analysis
c
molecule
atom
Ti
Zr
Hf
SO2
a
OM(η2-SO) (1A)
M
S
1.49124
0.05813
1.91799
2.04379
−0.02200
−0.96166
−0.93433
−0.04032
−1.04265
−0.96082
1.62123
−0.81062
−0.81062
b
O(MO)
−0.73346
−0.81591
b
O(S−O)
a
b
c
:M represents corresponding metal atom in each column. Represents corresponding atom in bracket. NPA for SO2 reactant.
Table 6. Mulliken Atomic Spin Densities for the 3A OM(η2-
SO)(η2-SO2) Molecules
previously,30 which overlaps with the TiO stretching mode of
OTi(η2-SO) in our experiment. Isotopic substituted precursors
confirmed these assignments. With 18O enriched samples, two
bands shift to 944.3 and 719.5 cm−1 while in 34S enriched samples,
no shifts were observed for the Ti−O stretching mode but the
S−O stretching mode shifted to 746.3 cm−1.
OM(η2-SO)(η2-
SO2)
atom
Ti
Zr
Hf
subunit: M
M
−0.037569
−0.002099
−0.006522
subunit:
O(M
0.024632
0.017472
0.027941
a
OM(η2-SO)
O)
In experiments for Zr reactions with SO2 in argon (Figure 3),
one strong absorption at 889.1 cm−1 in the ZrO stretching
region and a weak band at 702.7 cm−1 in the S−O stretching
region track very well, which can be assigned to the OZr(η2-SO)
molecule. The 889.1 cm−1 band shows a 42.7 cm−1 S18O2 isotopic
shift, but no shift with 34SO2 sample. The 16O/18O isotopic
frequency ratio 1.0504 suggests that this band is due to the
ZrO stretching vibration. The 702.7 cm−1 band exhibits a larger
28.4 cm−1 oxygen-18 shift and smaller 4.4 cm−1 sulfur-34 isotopic
shift. The 16O/18O isotopic ratio 1.0421 and 32S/34S isotopic ratio
1.0063 indicate that this band is due to the S−O stretching
vibration. The ZrO and S−O stretching modes in solid neon
were found at 905.3 and 696.8 cm−1, respectively (Figure 4).
In 18O enrichment experiments, the former shifts to 860.8 cm−1
while the latter, unfortunately, is covered by absorptions due to
CO2. With the 34SO2 sample the 905.3 cm−1 band shows no shift
but the 696.8 cm−1 band shifts to 691.8 cm−1.
S
0.746739
0.282476
0.677629
0.162586
0.162586
0.718337
0.284760
0.666074
0.169409
0.146047
0.711242
0.288648
0.670879
0.160212
0.147600
a
O(S−O)
S
subunit: SO2
O
O
a
Represents corresponding atom in bracket.
same behaviors on deposition, annealing, and irradiations, and
therefore suggest that these two modes are due to the same
molecule. With 18O enriched sample these bands shifted to 930.8
and 717.3 cm−1 giving 1.0431 and 1.0418 16O/18O isotopic
frequency ratios. As shown in Figure 1 where there is about
10% SO2 + 30% S16,18O2 + 60% S18O2) in the sample, two band,
that is, doublet oxygen, isotopic distributions were observed
for these two absorptions, which indicates the participation of
one oxygen atom in each mode. With 34S enriched sample
(30% 32SO2 and 70% 34SO2), the 970.9 cm−1 band exhibits
very little shift (to 970.7 cm−1), suggesting very little S
involvement in this vibration, while the 747.3 cm−1 band shifts
to 744.3 cm−1 as appropriate for the S−O stretching mode.
These two modes are characteristic of the OTi(η2-SO)
molecule.
In solid neon as shown in Figure 2, the Ti−O and S−O
stretching modes for the OTi(η2-SO) molecule are found at
987.4 and 749.0 cm−1, respectively. These bands appeared on
deposition, increased on annealing to 8 K, decreased after
irradiation (>220 nm) but recovered on further annealing to
10 K in concert. Notice that the photosensitive band at 987.4 cm−1
is in agreement with the ν1 fundamental of SO2− reported
In the case of hafnium (Figure 5), the strong band at
882.7 cm−1 is associated with the 688.9 cm−1 band on annealing
and photolysis throughout our experiments, which can be
assigned to the OHf(η2-SO) molecule. Such assignment is
further substantiated by observation of two fundamentals at
836.7 and 658.9 cm−1 in the 18O enriched sample and at 897.0
and 685.3 cm−1 in the 34SO2 sample. The 16O/18O isotopic ratios
of 1.0550 and 1.0455 for both modes and 32S/34S isotopic ratio
1.0051 for the lower mode are very close to typical values for
HfO and S−O stretching vibration modes. The behavior of
the OHf(η2-SO) molecule in excess neon is somewhat similar to
that in solid argon (Figure 6). The absorptions due to the
OHf(η2-SO) molecule are identified as a HfO stretching mode
7421
dx.doi.org/10.1021/ic3008987 | Inorg. Chem. 2012, 51, 7415−7424

Inorganic Chemistry
Article
at 897.0 cm−1 and a S−O stretching mode at 685.9 cm−1. With
S16O2/ S18O2 mixture, both modes are found in a doublet
distribution which suggests only one oxygen atom participated in
each vibration. With the 34S enriched sample, the S−O stretching
mode shows a red shift to 683.2 cm−1 giving the 1.0040 32S/34S
ratio, which is very close to the analogous mode ratios for
OTi(η2-SO) and OZr(η2-SO).
argon to neon for the OTiS molecule. However strong
absorption at 905.3 cm−1 for OZr(η2-SO) would cover this band.
The HfO stretching mode was found at 876.9 cm−1 for the
OHfS molecule in solid argon (Figure 5), which is near the band
at 877.4 cm−1 for OZrS, suggesting lanthanide contraction and
relativistic effects for hafnium atom, as reported earlier for the
dioxide molecules.17 Diagnostic information from isotopic
experiments further confirms the assignment; the HfO
stretching mode was found at 831.4 cm−1 with S18O2, giving an
isotopic 16O/18O ratio of 1.0547, but this mode exhibited no shift
with 34SO2. We did not observe the absorption of OHfS in solid
neon, and the reasonable explanation could be made on the fact
that absorptions due to HfO stretching mode for both
OHf(η2-SO) and OHfS are quite close with each other.
DFT frequency calculations support these OMS assignments.
With the BPW91 functional the MO stretching modes of the
OMS molecules were calculated at 999.1 cm−1 (Ti), 893.4 cm−1
(Zr), and 863.8 cm−1 (Hf), which are overestimated by 4.8%,
1.8%, and −1.5%, respectively. The B3LYP functional gave
slightly higher frequencies, but they are still in good agreement
with group 4 metal oxide frequency calculations. Notice the
predicted absorption of the MS vibration is only 20%−25% of
the MO mode intensity for all three metals, and it was not
observed in our experiments.
DFT calculations support these assignments. With the B3LYP
functional the TiO and S−O stretching modes for the OTi(η2-
SO) molecule are predicted at 1040.9 and 770.8 cm−1, which are
overestimated by 7.2% (Ar), 5.4%(Ne) and 3.1% (Ar),
2.8%(Ne), respectively. The predicted TiO stretching mode
shows no shift while the S−O mode shifts red about 3.0 cm−1
upon 34S substitution. With 18O substitution the calculated
TiO and S−O stretching modes give 1.0431(Ar), 1.0456(Ne)
and 1.0418(Ar), 1.0413(Ne) 16O/18O isotopic frequency ratios,
respectively, which match the observed values very well. When
the BPW91 functional is used, these modes are calculated at
1002.7 and 747.6 cm−1, which give very similar predictions.
The ZrO stretching frequency of OZr(η2-SO) is calculated at
920.9 cm−1 (B3LYP) and 894.6 cm−1 (BPW91) with the S−O
stretching mode at 708.2 cm−1 (B3LYP) and 693.9 cm−1
(BPW91). For all of above calculated frequencies, the maximum
error is only 3.6%. Very similar frequency calculations were
performed for OHf(η2-SO) molecule, and as shown in Table 2,
the calculations reproduced experimental values very well.
The insertion reactions of M (M = Ti, Zr, Hf) atoms into SO2
to give OM(η2-SO) are exothermic (ZPE correction included)
by 112.0 kcal/mol (Ti), 139.5 kcal/mol (Zr), and 129.4 kcal/mol
(Hf), respectively, based on B3LYP calculations.
OM(η2-SO)(η2-SO2) (M = Ti, Zr, Hf). As shown in Figure 4,
new group of bands was observed at 978.4, 963.1, 925.3, 886.1,
and 591.6 cm−1 in laser-ablated Zr atom reactions with SO2 in
solid neon where more diffusion takes place on sample
deposition, and these bands are appropriate for the OZr(η2-
SO)(η2-SO2) adduct molecule. Irradiation (>220 nm) increased
these bands 2-fold while the bands due to OZr(η2-SO) at 905.3
and 696.8 cm−1 decreased by 50%, suggesting this enhancement
comes at the expense of OZr(η2-SO). In 18O isotopic
experiments the counterpart bands were found at 943.8, 933.3,
881.1, 852.8, and 563.5 cm−1 with 16O/18O isotopic ratios
1.0367, 1.0319, 1.0502, 1.0390, and 1.0499, respectively. Notice
that the 925.3 cm−1 band shows the largest 1.0502 16O/18O
isotopic ratio, which is a typical heavy metal−oxygen stretching
vibration mode, while the remaining bands with smaller 16O/18O
isotopic ratios give evidence of sulfur−oxygen stretching and
bending vibrations. The mixed 16O/18O sample (5% SO2, 20%
S16,18O2 and 75% S18O2) gave doublet oxygen isotopic
distributions for the upper four bands along with triplet
distribution for the bond at 563.5 cm−1 which indicates the
involvement of one oxygen atom for each of the upper four
modes and two oxygen atoms for 563.5 cm−1 band. With 34S
enriched sample, bands at 978.4, 963.1, 886.1, and 591.6 cm−1
shift to 968.8, 952.1, 878.0, and 587.5 cm−1, giving 1.0099,
1.0116, 1.0092, and 1.0070 32S/34S isotopic ratios, The 978.4 and
963.1 cm−1 bands can be assigned to SO stretching modes
while 886.1 and 591.6 cm−1 bands are due to the S−O stretching
and OSO bending modes, respectively. The 925.3 cm−1
band shows no 34-S shift because of the ZrO stretching mode
barely perturbed by SO and SO2 subunits. In solid argon very
weak group bands at 966.9, 946.5, 915.4, and 877.2 cm−1
appeared on broadband irradiation and are due to the OZr(η2-
SO)(η2-SO2) molecule.
M + SO2 → OM(η2‐SO)
OMS (M = Ti, Zr, Hf). The strong band at 953.2 cm−1 in solid
argon increased on annealing to 20 K after deposition (Figure 1),
and annealing to 35 K doubled the absorbance for this band. The
953.2 cm−1 band shows no shift with 34SO2 but shifts to 913.1
cm−1 with S18O2 sample, which is typical for a TiO stretching
mode. This band appears lower than the TiO diatomic mode at
987.7 cm−1 but higher than the TiO2 modes at 946.9 and 917.0
cm−1. The triatomic OTiS molecule comes into mind for the
953.2 cm−1 band assignment. With the B3LYP functional TiO
and TiS stretching modes of OTiS molecule were calculated
at 1034.5 and 577.6 cm−1 (Table 4). However, the 577.6 cm−1
band is predicted to have only one-fifth the intensity of the
1034.5 cm−1 band, which suggests that this mode is too weak to
be observed here. Analogous experiments were done in a neon
matrix, and representative spectra are shown in Figure 2. A weak
feature at 978.0 cm−1 is proper for TiO stretching mode of
STiO, but the 18O counterpart was not observed because of
overlap by a very strong TiO2 band at 936.5 cm−1. Similar
neon matrix shifts were observed for the stretching modes
of TiO2.24
Laser-ablated Zr atom reactions with SO2 in argon produced a
new ZrO stretching frequency at 877.4 cm−1, which was very
weak on deposition and increased markedly on annealing to 35 K
(Figure 3). This band shows no shift with enriched 34SO2 sample,
but shifts to 835.1 cm−1 with S18O2, giving the 16O/18O isotopic
ratio 1.0507, which is appropriate for a ZrO stretching
mode of the OZrS molecule. A complementary experiment was
performed using a neon matrix, but appropriate absorptions for
OZrS were not observed. The ZrO mode for OZrS in neon
could locate around 903 cm−1 if we take a 25 cm−1 blue-shift from
Our B3LYP functional calculation for OZr(η2-SO)(η2-SO2)
molecule gave a strong ZrO stretching vibration at 939.4 cm−1,
two SO stretching modes at 977.0 and 956.6 cm−1, an S−O
vibration at 896.2 cm−1 and an OSO bending mode at
590.3 cm−1, which reproduced our observed values very well.
The predicted 16O/18O isotopic ratios for ZrO and SO
7422
dx.doi.org/10.1021/ic3008987 | Inorg. Chem. 2012, 51, 7415−7424

Inorganic Chemistry
Article
stretching modes are essentially the same as our experimental
which is consistent with the reaction energy predicted and
reflects the relativistic effects.17,41
values. The optimized structure for the OZr(η2-SO)(η2-SO2)
3
molecule with the A ground state is very similar to OZr(η2-
For the complex OM(η2-SO)(η2-SO2), DFT calculations
indicate that all the three molecules are found in the triplet
ground state, which are more stable than the corresponding
singlet by energies of 22 (Ti), 23 (Zr), and 25 (Hf) kcal/mol. In
addition, formation of the complex OM(η2-SO)(η2-SO2) (3A) is
energetically preferred. Our calculated results suggest that once
another SO2 coordinates to OM(η2-SO), the energy will be
lowered by 45 to 56 kcal/mol from titanium to hafnium.
However, the new adducts are all increased under λ > 220 nm
irradiation as shown in Figures 4, 5, and 6. Noticing the ground
state for SO2 and OM(η2-SO) are both singlet while that for
OM(η2-SO)(η2-SO2) is triplet. It seems reasonable that
irradiations induced a charge transfer between two reactants
leading to the reaction to proceed through an energy preferred
channel. From Table 6, we also learn that the coordinated
subunit SO along with SO2 in OM(η2-SO)(η2-SO2) (3A)
molecule are both found to concentrate spin densities. This
distribution facilitates the activation of SO2 and brings about
two elongated S−O (1.54/1.55 Å) bonds as shown in Figure 7.
The significant red-shift for the SO stretching mode shows
activation of the SO bond through this molecule.
O2)(η2-O3) molecules obtained from reactions of Zr and O2, for
which the ZrO stretching mode was observed at 912.9 cm−1 in
solid argon.33 This is very close to the same mode at 915.4 cm−1
for the OZr(η2-SO)(η2-SO2) molecule in our argon matrix.
Notice that the optimized structure shows two unequal SO
bond lengths: one is 0.01 Å longer than the other. The predicted
isotopic 16,18-O distributions for the SO stretching and O
SO bending modes show doublet and triplet absorptions,
respectively, suggesting that two SO stretching modes vibrate
separately while three atoms bends together. This calculated
distribution is in excellent agreement with our experimental
values. Notice that our OZr(η2-SO)(η2-SO2) complex with η2-
O,O-coordination of SO2 toward the metal center was stabilized
(Figure 7). However two coordination modes, η1-S and η2-S,O,
of SO2 ligand to transition metal center were found in
M(SO2)2(PPh3)2 (M = Ni, Pt) and Rh(NO)(SO2)(PPh3)2
complexes42 and η1-O-coordination mode of SO2 was charac-
terized in transition metal cation complexes (e.g., [M(SO2)x-
(AsF6)2] M = Mg, Mn, Fe, Co, Ni, Zn).16
In the hafnium atom reaction with SO2 in solid neon (Figure 6),
new bands at 973.5, 960.6, 913.4, 875.7, and 608.0 cm−1 were
observed, which are due to the OHf(η2-SO)(η2-SO2) molecule.
The effect of 18-O and 34-S substitution on the spectra are very
similar to the Zr case, so the assignment is straightforward. The
913.4 cm−1 band with the 1.0545 16O/18O isotopic ratio is due to
the HfO stretching mode while the 973.5 and 960.6 cm−1
bands are appropriate for SO stretching, and the 608.0 cm−1
band should be assigned to an OSO bending mode.
Similar assignments are made for the OTi(η2-SO)(η2-SO2)
molecule as shown in Table 1 and Figure 2. Two SO stretching
modes appeared at 973.5 and 961.9 cm−1, which are very close to
the same modes observed for OZr(η2-SO)(η2-SO2) and OHf(η2-
SO)(η2-SO2); however, the TiO stretching mode was
observed higher at 1015.8 cm−1. The S−O stretching mode
was not observed because of band weakness. The 18-O and 34-S
experiments gave similar isotopic shifts and distribution as shown
in Figure 6.
More interestingly, both OM(η2-SO) and OM(η2-SO)(η2-
SO2) molecules exhibit chirality. Optimized molecules shown in
Figure 8 are enantiomers with chiral metal centers. Origins for
The OM(η2-SO)(η2-SO2) molecules can in principle be
formed by adding SO2 to OM(η2-SO). Our B3LYP calculations
predict the reactions to be exothermic by 45 kcal/mol (Ti),
51 kcal/mol (Zr), and 56 kcal/mol (Hf). Notice that the
OM(η2-SO)(η2-SO2) molecules are calculated to have triplet
ground states, and their product yields were increased 2-fold by
uv irradiation.
Figure 8. OM(η2-SO) reaction with SO2 (λ > 220 nm) giving OM(η2-
SO)(η2-SO2) with chiral molecules.
hυ
SO2 + OM(η2‐SO) → OM(η2‐SO)(η2‐SO2)
Bonding. As has been discussed, insertion of M(Ti, Zr, Hf)
into SO2 to give the OM(η2-SO) (1A) molecule is a highly
exothermic reaction, and the energy released is predicted to be
112(Ti), 140 (Zr), and 129 (Hf) kcal/mol. Reasons for the
differences can be attributed to the core-like and less reactive
nature of 3d orbitals.17 NPA analysis summarized in Table 5
further confirm that charges centered on M and other atoms
increase in series Ti−Zr−Hf, suggesting more electrons are
participating in bonding and hence the more ionic bonds are
formed. Bond component analysis for OM(η2-SO) (1A) based
on NBO give an average d orbitals participation in bonding as
64.2% (Ti), 69.1% (Zr) and 64.4% (Hf). It is clear that d orbital
participation increases from Ti to Zr and decreases from Zr to Hf,
the formation of such molecules can be expected from the
insertion reaction of M to SO2 to produce different OM(η2-SO)
(1A) molecules. Although the combination of quantum chemical
calculations and vibrational absorption and circular dichroism
spectra (VCD spectra) proved to be powerful in determining the
absolute configuration of such molecules,34 we cannot identify
chiral molecules under our current experimental conditions.
Since catalysts containing group 4 atoms have been applied
extensively in enantioselective synthesis, asymmetric catalysis,
and many other related areas,35,36 the above mentioned
mechanism for the formation of new adducts may be involved
in one of the indispensible processes which are of prime
importance to those areas.
7423
dx.doi.org/10.1021/ic3008987 | Inorg. Chem. 2012, 51, 7415−7424

Inorganic Chemistry
Article
(18) Chertihin, G. V.; Bare, W. D.; Andrews, L. J. Phys. Chem. A 1997,
101, 5090.
(19) Gong, Y.; Zhang, Q.; Zhou, M. J. Phys. Chem. A 2007, 111, 3534.
(20) Gong, Y.; Zhou, M. J. Phys. Chem. A 2008, 112, 9758.
(21) Gong, Y.; Zhou, M.; Tian, S. X.; Yang, J. J. Phys. Chem. A 2007,
111, 6127.
(22) Thompson, K. R.; Weltner, W. J. Phys. Chem. 1971, 75, 3243.
(23) Gong, Y.; Zhou, M. J. Phys. Chem. A 2007, 111, 8973.
(24) Gong, Y.; Zhou, M.; Andrews, L. Chem. Rev. 2009, 109, 6765.
(25) Wang, X.; Andrews, L.; Marsden, C. J. Inorg. Chem. 2009, 48,
6888.
CONCLUSIONS
■
Laser-ablated group 4 atoms react with SO2 during condensation
in excess argon and neon at 4 K, and the new products are
presented here. Infrared absorption bands for OMS and OM(η2-
SO) (M = Ti, Zr, Hf) are observed, and isotopic substitution and
calculated frequencies of optimized structures further confirm
the existence of these products. In the softer neon matrix new
bands are assigned to adduct OM(η2-SO)(η2-SO2) molecules. A
bonding motif is also proposed for the OM(η2-SO)(η2-SO2)
molecules. More interestingly, optimized OM(η2-SO) and
OM(η2-SO)(η2-SO2) molecules are enantiomers with chiral
centers on M (Ti, Zr, Hf) which can find a meaningful role in
enantioselective synthesis, asymmetric catalysis, and other
related areas.
(26) Wang, X.; Andrews, L.; Marsden, C. J. J. Phys. Chem. A 2009, 113,
8934.
(27) Andrews, L. Chem. Soc. Rev. 2004, 33, 123.
(28) Andrews, L.; Citra, A. Chem. Rev. 2002, 102, 885.
(29) Lo, W. J.; Wu, Y. J.; Lee, Y. P. J. Chem. Phys. 2002, 117, 6655.
(30) Forney, D.; Kellogg, C. B.; Thompson, W. E.; Jacox, M. E. J. Chem.
Phys. 2000, 113, 86.
(31) Chen, L. S.; Lee, C. I.; Lee, Y. P. J. Chem. Phys. 1996, 105, 9454.
(32) Wang, X.; Andrews, L. J. Phys. Chem. A 2009, 113, 8934.
(33) Chertihin, G. V.; Andrews, L. J. Phys. Chem. 1995, 99, 6356.
(34) Martin, P. R.; Forsythe, I. D. J. Phys. Chem. 1994, 98, 11623.
(35) Wood, M. C.; Leitch, D. C.; Yeung, C. S.; Kozak, J. a.; Schafer,
L. L. Angew. Chem., Int. Ed. 2007, 46, 354.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: xfwang@tongji.edu.cn (X.W.), lsa@virginia.edu (L.A.).
Notes
The authors declare no competing financial interest.
(36) Brunel, J. M.; Holmes, I. P. Angew. Chem., Int. Ed. 2004, 43, 2752.
(37) (a) Becke, A. D. J. Chem. Phys. 1993, 98, 5648. (b) Lee, C.; Yang,
Y.; Parr, R. G. Phys. Rev. B 1988, 37, 785.
(38) Frisch, M. J.; Pople, J. A.; Binkley, J. S. J. Chem. Phys. 1984, 80,
3265.
ACKNOWLEDGMENTS
■
We gratefully acknowledge financial support from NSFC Grant
(20973126, 21173158) and STCSM Grant (10PJ1409600) to
X.W, and DOE Grant DE-SC0001034 and NCSA computing
Grant CHE07-0004N to L.A.
(39) Andrae, D.; Haeussermann, U.; Dolg, M.; Stoll, H.; Preuss, H.
Theor. Chim. Acta 1990, 77, 123.
(40) (a) Becke, A. D. Phys. Rev. A 1988, 38, 3098. (b) Perdew, J. P.;
Burke, K.; Wang, Y. Phys. Rev. B 1996, 54, 16533 , and references therein.
(c) See also: Becke, A. D. J. Chem. Phys. 1997, 107, 8554.
(41) Pyykko, P. Chem. Rev. 2012, 112, 371−384.
(42) (a) Kubas, G. J. Inorg. Chem. 1979, 18, 182. (b) Ryan, R. R.;
Kubas, G. J.; Moody, D. C.; Eller, P. G. Struct. Bonding (Berlin) 1981,
46, 47.
REFERENCES
■
(1) Jakubikova, E.; Bernstein, E. R. J. Phys. Chem. A 2007, 111, 13339.
(2) He, S. G.; Xie, Y.; Dong, F.; Heinbuch, S.; Jakubikova, E.; Rocca,
J. J.; Bernstein, E. R. J. Phys. Chem. A 2008, 112, 11067.
(3) Zhao, X.; Hrbek, J.; Rodriguez, J. A.; Pe, M. Surf. Sci. 2006, 600,
229.
(4) Rodriguez, A.; Jirsak, T.; Hrbek, J. J. Phys. Chem. B 1999, 103, 1966.
(5) Rodriguez, J. A.; Jirsak, T.; Gonzal
J. Chem. Phys. 2001, 115, 10914.
(6) Rodriguez, J. A.; Perez, M.; Evans, J.; Liu, G.; Hrbek, J. J. Chem.
Phys. 2005, 122, 241101.
́ ́
ez, L.; Evans, J.; Perez, M.
́
(7) Rodriguez, J. a.; Jirsak, T.; Freitag, A.; Larese, J. Z.; Maiti, A. J. Phys.
Chem. B 2000, 104, 7439.
(8) Rodriguez, A.; Liu, P.; Pe, M.; Liu, G.; Hrbek, J. J. Phys. Chem. A
2010, 114, 3802.
(9) Rodriguez, A.; Dvorak, J.; Jirsak, T. J. Phys. Chem. B 2000, 104,
11515.
(10) Schlereth, T. W.; Hedhili, M. N.; Yakshinskiy, B. V.; Gouder, T.;
Madey, T. E. J. Phys. Chem. B 2005, 109, 20895.
(11) Rodriguez, A.; Liu, G.; Jirsak, T.; Hrbek, J.; Chang, Z.; Dvorak, J.;
Maiti, A. J. Am. Chem. Soc. 2002, 124, 5242.
(12) Chaturvedi, S.; Rodriguez, A.; Jirsak, T.; Hrbek, J. J. Phys. Chem. B
1998, 102, 7033.
(13) Rodriguez, A.; Jirsak, T.; Chaturvedi, S.; Hrbek, J.; York, N. J. Am.
Chem. Soc. 1998, 11149.
(14) Rodriguez, J. A.; Ricart, J. M.; Clotet, A.; Illas, F. J. Chem. Phys.
2001, 115, 454.
(15) (a) Rodriguez, A.; Hrbek, J.; Jirsak, T. Surf. Sci. 1999, 426, 319.
(b) Rodriguez, J. A.; Jirsak, T.; Chaturvedi, S.; Dvorak, J. J. Mol. Catal. A:
Chem 2001, 167, 47.
(16) (a) Mews, R.; Lork, E.; Watson, P. E.; Gortler, B. Coord. Chem.
Rev. 2000, 197, 277. (b) Akkus,̧ O. N.; Decken, A.; Knapp, C.; Passmore,
J. J. Chem. Crystallogr. 2006, 36, 321−329. (c) Decken, A.; Knapp, C.;
Nikiforov, G. B.; Passmore, J.; Rautiainen, J. M.; Wang, X.; Zeng, X.
Chem.−Eur. J. 2009, 15, 6504. (d) Derendorf, J.; Kebler, M.; Knapp, C.;
Ruhle, M.; Schulz, C. Dalton Trans. 2010, 39, 8671.
(17) Chertihin, G. V.; Andrews, L. J. Phys. Chem. 1995, 99, 6356.
7424
dx.doi.org/10.1021/ic3008987 | Inorg. Chem. 2012, 51, 7415−7424