Crystal structures and luminescence properties of osmium complexes of cis-1,2-vinylenebis(diphenylarsine) and pyridyl ligands: Possible evidence for metal d, ligand d backbonding

Article abstract of DOI:10.1016/j.ica.2005.11.024

Divalent osmium complexes of the form [Os(N-N)2L-L](PF6-)2 where N-N was a polypyridyl, and L-L was either cis-1,2-bis(diphenylphosphino)ethene (dppene) or cis-1,2-vinylenebis(diphenylarsine) (dpaene) have been synthesized and characterized. X-ray structures were determined for three complexes and for the free dpaene molecule. It was observed that the P-C bond lengths, and C-P-C bond angles do not change significantly when complexed to osmium. It was observed the As-C bond lengths shorten by 2.3 pm and the C-As-C bond angles broaden by 5.6°when dpaene was complexed to osmium. These changes in the arsine structure may indicate a different method of backbonding between arsenic and osmium. It was found that the arsine complexes had absorption and emission that were to the red of analogous phosphine complexes. In violation of "energy gap law", the dpaene complexes were found to have higher quantum yields. This may be due to the way that the arsenic atoms bond to osmium.

Full text of DOI:10.1016/j.ica.2005.11.024

Inorganica Chimica Acta 359 (2006) 1093–1102
www.elsevier.com/locate/ica
Crystal structures and luminescence properties of osmium complexes of
cis-1,2-vinylenebis(diphenylarsine) and pyridyl ligands:
Possible evidence for metal d, ligand d backbonding
*
Brenden Carlson , Gregory D. Phelan, Jason B. Benedict, Werner Kaminsky, Larry Dalton
Chemistry Department, University of Washington, Box 351700, Seattle, WA 98195, United States
Received 1 July 2005; received in revised form 16 November 2005; accepted 23 November 2005
Available online 7 February 2006
Abstract
Divalent osmium complexes of the form ½OsðN–NÞ2L–LꢀðPF6ꢁÞ2 where N–N was a polypyridyl, and L–L was either cis-1,2-bis(diph-
enylphosphino)ethene (dppene) or cis-1,2-vinylenebis(diphenylarsine) (dpaene) have been synthesized and characterized. X-ray structures
were determined for three complexes and for the free dpaene molecule. It was observed that the P–C bond lengths, and C–P–C bond
angles do not change significantly when complexed to osmium. It was observed the As–C bond lengths shorten by 2.3 pm and the
C–As–C bond angles broaden by 5.6° when dpaene was complexed to osmium. These changes in the arsine structure may indicate a dif-
ferent method of backbonding between arsenic and osmium. It was found that the arsine complexes had absorption and emission that
were to the red of analogous phosphine complexes. In violation of ‘‘energy gap law’’, the dpaene complexes were found to have higher
quantum yields. This may be due to the way that the arsenic atoms bond to osmium.
Ó 2005 Elsevier B.V. All rights reserved.
Keywords: As ligands; Bond theory; Osmium complexes; Phosphane ligands; X-ray crystal structures
1. Introduction
been used as sensors, these include the porphyrin com-
plexes of palladium and platinum [8], and other complexes
utilizing Ir [9], or Ru [10]. For the purposes OLED and sen-
sor applications several osmium complexes have been
developed (Fig. 1). In this report we discuss the structures
and spectral properties of the complexes.
The bonding in arsine (phosphine, antimony) ligand
containing metal complexes may be thought of as having
two major components. There is r donation of the lone
pair contained in the s-orbital on the arsenic (phosphorus,
antimony) ligand to an empty d-orbital on the metal
(Fig. 2) [11]. For complexes in octahedral coordination,
as is the case for the osmium complexes being discussed
in this report, the donation of the ligand lone pair is to
the dz2 and the dx2ꢁy2 [11]. The other component is called
backbonding. This occurs by backdonation from a filled
d-orbital on the metal (dxy, dxz, dyz) to an empty orbital
on the phosphorus [11]. The question arises as to which
orbital on the arsenic is responsible for accepting the back
There is considerable interest in using metal complexes
in a variety of applications including light sources in
Organic Light Emitting Devices (OLEDs) [1]. Using metals
such as iridium [2], ruthenium [3], or osmium [4], it is pos-
sible to design complexes that can emit light ranging in
color from blue through red making them ideal candidates
for this technology [5]. The resulting luminophores could
be part of an OLED in optical displays.
Another possible use for metal complexes of the type
being reported could be for sensors. Luminescence of the
complexes may be used to detect various substances and
has been utilized in fields such as medicine [6], and beer
manufacturing [7]. Several phosphorescent materials have
*
Corresponding author. Tel.: +1 206 616 2974; fax: +1 206 685 8665.
E-mail address: wcarlson@u.washington.edu (B. Carlson).
0020-1693/$ - see front matter Ó 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.ica.2005.11.024

1094
B. Carlson et al. / Inorganica Chimica Acta 359 (2006) 1093–1102
structure of dppene has been previously reported [15]. In
R
X
R
the reports the C–P bond length was 181.5 pm to the
bridge, and 182.9 and 183.9 pm to the phenyl groups.
The average P–C bond length for dppene was found to
be 183.1(6). The average C–P–C bond angle was found to
be 102.0(5)°.
R2
H
H
Complex
X
As
P
R1
H
H
R
R
1
2
3
4
5
N
N
R
2+
As MeOPh H
Os
N
R
As CH
P
CH
CH
3
3
3
3
In our previous report, we obtained a structure of
dpaene from crystals that were grown from 1-butanol
[4a]. The structure had significant disorder, which greatly
increased the standard deviation of the bond lengths and
angles. This made meaningful comparisons between the
dpaene structure and the metal complexes derived from
dpaene difficult. Crystals of dpaene without disorder were
obtained from a dilute heptane solution giving a lower
standard deviation than previously reported (Fig. 3). The
crystallographic refinement parameters are tabulated in
Table 1. Bond angles and lengths are tabulated in Tables
2 and 3. The non-disordered structure closely agrees with
the major (88.9%) structure of the previous report; thus,
the structure grown from heptane was averaged with the
major structure grown from butanol. The average As–C
bond length was 196.3(11) pm. This result was slightly
shorter the expected bond length of 198 pm based upon
the covalent radii of arsenic (121 pm) and carbon (77 pm)
[16]. The average C–As–C bond angle was observed to be
97.3(12)°, considerably more acute than that of a tetrahe-
dral structure where the ideal bond angle would be
109.5°. It has been previously reported that a reluctance
exists for arsenic to hybridize, forming bonds with orthog-
onal p-orbitals [17]. The ‘‘lone pair’’ does not reside in an
orbital of ‘‘sp’’ character, rather an orbital that is of more
s character.
CH
X
N
R
R
Fig. 1. Structure of the osmium complexes.
P, As,Sb
d-orbital
R3X
Os
Filled Metal d-Orbital
backbond
bond
X = P, As, Sb
filled s orbital
donation
X = P, As, Sb
empty σ* orbital
R3X
Filled Metal d-Orbital
backbond
donation
Fig. 2. Sketch of possible bonding scenarios involving arsine (phosphine,
antimony) ligands.
donation from the metal since there are no apparent orbi-
tals to accept charge. This accepting orbital has been
described as being either a d-orbital [12] or an antibonding
r-orbital; current consensus is that the latter is more
appropriate [13]. Some crystallographic reports have
shown that the bond lengths between phosphorus/arsenic
and the R groups attached to it lengthen, which supports
the notion that a r* is responsible for backbonding with
the metal [13,14]. It has been generally accepted that the
phosphine/arsine ligands backbond strongly when based
upon strongly electron withdrawing groups such as fluo-
rine. With the AsF3 ligand it has been shown that bond
lengths increase 4 pm (due to r* backbonding) while bond
angles remain unchanged at 95° [14]. Ligands with more
electron donating groups such as methyl tend to be stron-
ger sigma donors and backbond more weakly. Our data
suggests that r* backbonding is not taking place and the
complexes being presented may be an example of metal
d–ligand d backbonding.
Structure of dppene and dpaene containing complexes.
X-ray structures for complexes 2 and 5 are illustrated in
Figs. 4 and 5. The crystallographic refinement parameters
are tabulated in Table 1. In our previous report we dis-
cussed the structure of several osmium-dppene containing
In this report we have synthesized several osmium com-
plexes and have obtained their crystal structures. The
results from the crystallography lead us to conclude that
for these complexes the arsenic back bonds through a sys-
tem that is primarily of d character.
2. Results and discussion
Fig. 3. Crystal structure of cis-1,2-vinylenebis(diphenylarsine) grown
from heptane with 50% probability spheres. Hydrogen removed for
clarity.
Structure of cis-1,2-bis(diphenylphosphino)ethene (dppe-
ne)and cis-1,2-vinylenebis(dipheylarsine) (dpaene). The

Table 1
Crystallographic data for the structures provided
Property
Dpaene
26H22As2
1
2
3
4
5
DMB
Empirical formula
C
2(C50H38As2N4Os), C53H44Cl6F12N4OsP4 C97H81As2Cl15N4O10OsS2 C60H58As2Cl4F12N4OsP2 C65H62F12N4OsP4 C51H47.5As2Cl2F12N4OsP2
0.69(C14H16),
3(C2H3N), 4(F6P)
Formula weight
Temperature (K)
484.28
130(2)
1432.47
130(2)
1491.70
130(2)
2398.57
130(2)
1606.88
130(2)
1441.27
130(2)
1417.31
130(2)
˚
Wavelength (A)
Crystal/color
0.71073
cut block/colorless
monoclinic, Pc (No. 7) triclinic, P1
0.71073
prism/orange
0.71073
cut block/red
triclinic, P1
0.71073
cut block/orange
triclinic, P1 (No. 2)
0.71073
cut block/orange
monoclinic, P21/c
0.71073
plate/red
0.71073
plate/orange
ꢀ
ꢀ
ꢀ
Crystal system,
space group
Unit cell dimensions
monoclinic, P21/c monoclinic, C2/c
˚
a (A)
12.5970(3)
5.61500(10)
17.0450(5)
90.00
117.1660(11)
90
12.0330(2)
13.6250(3)
19.5740(5)
99.0570(7)
97.3220(7)
112.9261(11)
11.9760(12)
13.1660(17)
20.256(3)
95.551(6)
99.932(5)
15.6190(7)
16.3000(4)
21.0090(9)
71.386(2)
75.8890(14)
84.734(2)
13.2030(2)
13.6150(2)
37.3350(5)
90
109.0581(5)
90
15.3740(5)
14.2610(6)
28.9730(11)
90
105.611(2)
90
48.5050(2)
11.4120(3)
20.1190(5)
90
103.0440(7)
90
˚
b (A)
˚
c (A)
a (°)
b (°)
c (°)
113.176(4)
3
˚
Volume (A )
1072.63(4)
1.499
4454/3739
2855.25(11)
1.666
12528/8605
2843.9(6)
1.742
14977/9393
4915.4(3)
1.621
18382/11398
6343.44(16)
1.683
26446/14579
6118.0(4)
1.565
20796/11404
10849.3(4)
1.735
33470/11604
Density (Mg mꢁ3
Reflections
)
collected/unique
Final R indices
[I > 2r(I)]
R1
wR2
0.0576
0.1505
0.0549
0.1318
0.0705
0.1291
0.0595
0.1514
0.0574
0.1006
0.0630
0.1420
0.0547
0.1273
R indices (all data)
R1
wR2
0.0732
0.1401
0.0950
0.1516
0.2037
0.1672
0.0747
0.1607
0.1553
0.1285
0.1369
0.1819
0.0924
0.1440

1096
B. Carlson et al. / Inorganica Chimica Acta 359 (2006) 1093–1102
Table 2
C6
C–As–C bond angles (°) for complexes for dpaene and osmium complexes
C7
C43
Compound C(Ph)–As(1)–C(br) C(Ph)–As(1)–C(br) C(Ph)–As(1)–C(Ph)
C41
dpaene
1
99.7(3)
103.7(3)
102.4(4)
102.3(3)
103.4(3)
103.7(5)
96.1(3)
102.7(3)
104.2(4)
103.4(3)
103.5(3)
103.6(5)
97.6(3)
103.2(3)
104.0(4)
100.5(3)
100.0(3)
101.3(4)
C44
C46
C47
C4
3
5
C3
P1
DMB
DPB [4a]
N3
Complex
C(Ph)–As(2)–C(br) C(Ph)–As(2)–C(br) C(Ph)–As(2)–C(Ph)
C1
dpaene
1
97.0(3)
106.5(3)
102.2(4)
102.2(3)
105.8(3)
104.8(5)
98.4(3)
102.2(3)
101.3(4)
102.1(3)
101.9(3)
102.6(5)
96.6(3)
101.8(3)
102.4(4)
104.6(3)
101.7(3)
102.4(5)
N4
Os1
C2
3
N1
5
N2
DMB
DPB [4a]
P2
C28
C20
C19
br, bridge; Ph, phenyl; DPB, [Os(4,40-diphenyl-2,20-bipyridine)2(dpaene)]2+
2(Tsꢁ).
C29
C31
C22
complexes [18]. The result here closely resembles the results
in our previous report. The average Os–P bonds length was
229.3(8) pm, which was ꢂ13 pm shorter than what would
be expected for a single bond based upon the covalent radii
of osmium (133 pm) and phosphorous (110 pm) [16]. The
average P–C bond of the complexed dppene ligand was
182.8(15) pm. The C–P–C bond angles for the complex
were 102.8(14)°.
X-ray structures are given for complexes 1, 3, 4 and
DMB are illustrated in Figs. 6–9. The crystallographic
refinement parameters are tabulated in Table 1. The aver-
age C–As–C bond angle was observed to be 102.9(14)°
for the complexes (Table 2). The average lengths (Table
3) of the Os–N, Os–As, and As–C bonds were measured
to be 208.8(27) pm, 240.9(6), and 194.0(9) pm. The As–Os
bonds were ꢂ 13 pm shorter than what would be expected
for a single bond based upon the covalent radii of the
atoms: arsenic (121 pm), osmium (133 pm) [16].
C34
C24
Fig. 4. Crystal structure of complex 2, [osmium (1,10-phenanthroline)2
cis-1,2-bis(diphenylphosphino)ethene]2+, with 50% probability spheres.
Solvent, hydrogen atoms, and counter ions removed for clarity.
to undergo very little, if any, structural change. For the
dppene system, both the C–P bond lengths and C–P–C
bond angles were well within standard deviation of the free
ligand [15]. However, dpaene was observed to undergo sig-
nificant structural change when complexed to osmium. The
difference in behavior between dppene and dpaene may be
explained by the r and p contributions of the ligand–Os
bonds.
Os–ligand p bonds through the p system of the bridging
unit may be possible. For each of the complexes the C@C
bond length was measured to be: complexes 1 and 3
Nature of bonding of dppene and dpaene complexed to
osmium. The behavior of dppene and dpaene when com-
plexed to osmium were observed to be different from each
other. Dppene when complexed to osmium was observed
[129.3(12) pm], complex
5
[130.8(9) pm], DMB
[132.7(10) pm], and DPB [129.6(11) pm], all of which were
normal for a C@C bond, and similar to the C@C bond in
the free dpaene structure, which was measured to be
Table 3
˚
Bond lengths for free ligand, dpaene, and osmium complexes (A)
Complex
N1
N2
N3
N4
As1
As2
1
2.133(5)
2.088(7)
2.109(5)
2.079(6)
2.054(9)
2.091(5)
2.102(6)
2.088(5)
2.099(6)
2.094(8)
2.133(5)
2.098(6)
2.075(5)
2.092(5)
2.011(11)
2.086(5)
2.077(7)
2.100(5)
2.092(6)
2.057(9)
2.4088(6)
2.4122(7)
2.4130(7)
2.4000(7)
2.3997(15)
2.4056(7)
2.4108(8)
2.4129(7)
2.4068(7)
2.4200(15)
3
4
DMB
DPB [4a]
Arsenic
As(1)–C(br)
As(1)–C(Ph)
As(1)–C(Ph)
As(2)–C(br)
As(2)–C(Ph)
As2–C(Ph)
dpaene
1
1.946(8)
1.944(7)
1.952(8)
1.938(7)
1.936(8)
1.932(11)
1.951(7)
1.952(7)
1.939(7)
1.937(7)
1.953(7)
1.930(10)
1.979(7)
1.955(7)
1.932(7)
1.921(7)
1.943(7)
1.945(11)
1.953(7)
1.948(6)
1.939(7)
1.928(6)
1.945(7)
1.929(9)
1.959(8)
1.945(6)
1.935(7)
1.921(7)
1.941(7)
1.944(11)
1.969(7)
1.948(6)
1.935(8)
1.943(6)
1.942(7)
1.941(11)
3
4
DMB
DPB [4a]
br, bridge; Ph, phenyl, DPB = [Os(4,40-diphenyl-2,20-bipyridine)2 (dpaene)]2+ 2(Tsꢁ).

B. Carlson et al. / Inorganica Chimica Acta 359 (2006) 1093–1102
1097
Fig. 5. Crystal structure of complex 5, [osmium (3,4,7,8-tetramethyl-1,10-
phenanthroline)2 cis-1,2-bis(diphenylphosphino)ethene2+, with 50% prob-
ability spheres. Hydrogen atoms, solvent, and counter ions removed for
clarity.
Fig. 7. Crystal structure of complex 3, [osmium (4,7-bis(4-methoxyphe-
nyl)-1,10-phenanthroline)2 cis-1,2-vinylenebis(diphenylarsine)]2+
with
50% probability spheres. Solvent, hydrogen atoms, and counter ions
removed for clarity.
,
Fig. 8. Crystal structure of complex 4, [osmium (3,4,7,8-tetramethyl-1,10-
phenanthroline)2 cis-1,2-vinylenebis(diphenylarsine)]2+, with 50% proba-
bility spheres. Solvent, hydrogen atoms, and counter ions removed for
clarity.
Fig. 6. Crystal structure of complex 1, [osmium (1,10-phenanthroline)2
cis-1,2-vinylenebis(diphenylarsine)]2+, with 50% probability spheres. Sol-
vent, hydrogen atoms, and counter ions removed for clarity.
ability of the phosphorus was a function of the mixing of
P–R r* and the 3d-orbitals [20]. Increasing back donation
from the metal to the phosphors leads to longer P–R dis-
tances and smaller R–P–R angles, whereas the opposite
was true for increasing r contribution [13b]. For the Os–
dppene complexes the two effects cancel; thus, it is observed
that the C–P bond lengths and the C–P–C bond angles
remain unchanged form those of the free ligand [13b,19].
The 99% confidence interval for the 30 As–C bond lengths
of the complexes was 193.6–194.4 pm, and of the 30 C–As–
C bond angles 101.7–104.0°. The 99% confidence level for
the 12 bond lengths of the dpaene ligand was 195.5–
197.2 pm, and of the 12 bond angles 96.5–98.1°. Thus, to
133(1) pm. This is evidence for little participation, if any, of
the C@C bond in the complexation to osmium, and very
little strain being placed upon this bond by complexing
the dpaene ligand to the metal. The p system of the phenyls
may also participate in receiving electron density from the
filled d orbital of the metal, but due to the distance between
the phenyl groups and osmium (365—596 pm), and the
lack of proper orientation of the phenyl groups towards
the metal, this possibility is remote.
The contributions of r and p components to metal–
phosphorus bonds of PR3 complexes by measurement of
the P–R bond lengths and R–P–R bond angles has been
reported [19]. The basis for this was that the p-accepting

1098
B. Carlson et al. / Inorganica Chimica Acta 359 (2006) 1093–1102
As4d–Os5d backbonding analogous to what was observed
for complexes based upon triphenyl antimony.
Spectral properties. The photophysical properties of this
general class of osmium complexes has been previously
reported [22]. Complexes 1, 5 and DMB have not been
reported in the literature and have been characterized.
Emission properties of complexes 2 and 6 have been
reported elsewhere [23,24]; however, the crystal structure
is reported in this report and the absorbance and emission
properties were measured for direct comparison to the
complexes made with the dpaene ligand. The measured
emission lifetime and quantum yield of complexes 2 and
6 were in good agreement with the published values; how-
ever, the spectra were somewhat blue shifted. It has been
previously reported that dpaene complexes of osmium with
extended p systems were an exception to the energy gap
law, where complexes that had red shifted emission exhib-
ited greater quantum yields [4a]. A similar trend was
observed in these complexes without extended p systems.
The absorbance, and emission are tabulated in Table 4,
and illustrated in Figs. 10 and 11. Additional spectra
may be found in the supplementary materials. The orbital
transition that defines the MC state (dp–dr*) are of
Fig. 9. Crystal structure of complex DMB, [osmium (4,40-dimethyl-2,20-
bipyridine)2 cis-1,2-vinylenebis(diphenylarsine)]2+, with 50% probability
spheres. Hydrogen atoms and counter ions removed for clarity.
99% confidence, it may be stated that the As–C bond
lengths shorten, and the C–As–C bond angles broaden
when the ligand is complexed to osmium. In the same logic
with the phosphine complexes, this could be taken as evi-
dence for an increased r component in the r/p ratio for
the Os–As bond, but this is not in accordance with the
observed data. Both the P–Os and the As–Os bonds exhibit
a similar 13 pm shortening in the bond than would be
expected. Secondly, dpaene and dppene exert a similar
trans effect on the phenanthroline ligand; thus, both
dppene and dpaene show similar backbonding.
The proposal in this report involves significant mixing of
the As 4d-orbitals into the As–C r* overlapping with the
filled Os 5d -orbitals to explain the observed shortening
of the As–C bond length and increase C–As–C in the bond
angle when complexed to osmium. Similar trends of
decreasing C–M bonds and increases in C–M–C bond
angles were observed in triphenyl antimony ligands when
complexed to platinum [21]. The authors concluded that
the increase in bond angles in conjunction with a decrease
in the sigma bonds was evidence for Sb5d–Pt5d orbital
overlap leading to backbonding [21a]. Thus, the dpaene–
osmium complexes being reported may be an example of
L mol-1 cm-1
x 103
70
60
50
40
30
20
10
0
250
350
450
550
650
750
Wavelength (nm)
Fig. 10. Absorbance spectra of complex 1 (s), complex 4 (m), and
complex DMB (*).
Table 4
Spectral properties of the various complexes
Complex
LC (nm) (e)
1MLCT (nm) (e)
3MLCT (nm) (e)
Emissiona
s (ns)b
Uc
1
267 (52000)
267 (57000)
276 (68000)
277 (51000)
286 (47000)
384 (12000)
376 (13000)
377 (17000)
367 (13000)
385 (9300)
500 (2400)
480 (3000)
450 (5000)
441 (4000)
487 (2600)
606
595
588
572
624
1580
1840
2800
3400
400
0.19
0.18
0.27
0.24
0.09
2
4
5
DMB
a
Nanometers.
Emission lifetime.
Emission quantum yield.
b
c

B. Carlson et al. / Inorganica Chimica Acta 359 (2006) 1093–1102
1099
be 595 nm. Yet, both were observed to have similar
quantum yields. The same was observed for complexes
4 and 5. Complex 4 was observed to have emission at
588 nm and complex 5 at 572 nm. Despite the red shift
in emission, complex 4 was measured to have a greater
quantum yield than complex 5. The dppene analog of
DMB has been reported with a quantum yield of 0.03
[24], while the dpaene complex being reported here
ab. units
1.2
1
0.8
0.6
0.4
0.2
0
was observed to have
a quantum yield of 0.09.
Therefore it was observed that the dpaene complexes
being reported here violate the energy gap law similar
to what was observed in complexes with extended p sys-
tems. The violation of the energy gap law by the dpaene
complexes may be due the differences in bonding
between the dpaene and dppene ligands with the
osmium core.
400
500
600
700
800
900
Wavelength (nm)
3. Conclusion
Fig. 11. Emission spectra of complex 1 (s), complex 4 (m), and complex
DMB (*).
We have determined the crystal structure for 6 osmium
complexes and the dpaene ligand. We found that when
dppene was complexed to osmium, very little structural
change was evident in the dppene ligand versus the free
ligand. When dpaene was complexed to osmium, bond
lengths were observed to 99% confidence become shorter,
and the bond angles broader versus the free dpaene ligand.
From these structural changes we determined that the
backbonding between As and Os took place through
dAs–dOs overlap.
The photophysical properties of the complexes were
measured of the complexes. It was found that the dpaene
complexes had absorption and emission that were to the
red of analogous dppene complexes. In violation of energy
gap law, the dpaene complexes were found to have higher
quantum yields. This may be due to the way that the
arsenic atoms bond to osmium. The photophysical proper-
ties of the complexes make them suitable as oxygen and/or
pressure sensor materials, in organic electronics, or in lumi-
nescent electrochemical cells.
orbitals of like symmetry and thus are Laporte forbidden.
This transition may be very weak, with e usually less than
100 L molꢁ1 cmꢁ1 and thus is ‘‘buried’’ under the much
stronger LC and MLCT transitions. The LC transition
for the complexes was observed as a sharp peak between
260 and 280 nm. The extinction coefficients were measured
to be between 47000 and 68000 L molꢁ1 cmꢁ1. Two
MLCT bands were observed for these complexes. The
stronger of the two corresponds to the singlet transition
and appears as a somewhat broad transition with a maxi-
mum just below 400 nm. The spin-forbidden triplet MLCT
absorption band appears at 500 nm and arises due to the
spin-orbit coupling constant of the heavy osmium center,
the strong back bonding, and metal-ligand orbital inter-
mixing. The singlet MLCT band for the complexes was
observed between 364 and 391 nm (e = 9000–
17000 L molꢁ1 cmꢁ1) and the triplet band was observed
between 441 and 500 nm (e = 2400–5000 L molꢁ1 cmꢁ1).
The complexes feature a smooth, unstructured expo-
nential Gaussian emission typical of MLCT. The emis-
sion lifetime was measured to range between 400 and
3400 ns. The complexes with the dpaene ligand were
observed to have faster decay of the excited state. This
may be due to the heavy atom effect of arsenic relative
to phosphorus. Emission of the complexes was measured
to between 588 and 624 nm. The emission quantum
yields of the complexes were found to range between
0.09 and 0.26. The emission of the dpaene ligands were
red shifted relative to the same complexes made with
dppene ligands. According to the ‘‘energy gap law’’
there should be a decrease in emission quantum yield
for the dpaene complexes due to the decrease in energy
between the excited and ground states [22]. This was not
observed. The dpaene complexes tended to have greater
quantum yields. The emission of complex 1 was mea-
sured to be 606 nm while complex 2 was measured to
4. Experimental
General procedure for synthesis of osmium complexes [23].
The osmium complexes were synthesized by reacting 1.000 g
(2.08 mmol) of (NH4)2OsCl6 (Alfa) with 2.05 equivalents of
polypyridyl (N–N) ligand in 25 mL of refluxing DMF
(Aldrich) under inert atmosphere for 3 h. The resulting solu-
tion was filtered, washed with DMF, cooled to 0 °C, and
then added dropwise to a water solution of sodium dithio-
nite (2.00 g in 400 mL) at 0 °C. The resulting purple precip-
itate of Os(N–N)2Cl2 was filtered and washed with
deionized water. Os(N–N)2Cl2 was reacted with 1.05 eq.
of cis-1,2vinylenebis(dipheylarsine) (dpaene, X-ray struc-
ture illustrated as Fig. 3) [4a] or cis-1,2-vinylenebis(diphe-
nylphosphine) (dppene) ligand in a refluxing mixture of
2,20-ethoxyethoxyethaonol (Aldrich) and glycerol (75:25
by volume) for 2 h under inert atmosphere. The complexes

1100
B. Carlson et al. / Inorganica Chimica Acta 359 (2006) 1093–1102
were precipitated by dropwise addition to a saturated water
solution of KPF6, filtered, and washed with water. Com-
plexes were then purified on silica using either toluene/ace-
tonitrile or acetonitrile/water/KPF6 solvent systems. The
structures of the resulting complexes are shown in Fig. 1,
and the crystal structures of complexes 2 and 6 are illus-
trated in Figs. 4 and 5, and complexes 1, 3, 4, and DMB
are illustrated in Figs. 6–9. Elemental analysis, mass spec-
trometry and yield data have been summarized in Table 5.
Complex 1: Anal. Calc. C, 45.33; H, 2.89; N, 4.23. Found:
C, 45.50; H, 2.79; N, 4.33%. Mass spectrometry (MALDI):
1029.925(0.747%), 1030.944(5.03%), 1031.966 (29.51%),
1032.987(57.86%), 1034.001(84.58%), 1035.018 (59.30%),
1036.03(100%), 137.044(68.07%), 1038.061 (19.18%),
1039.083(3.19%). Yield: 72%. Complex 2: Anal. Calc. C,
48.55; H, 3.10; N, 4.53. Found: C, 48.15; H, 2.99; N, 4.58%.
Mass spectrometry (MALDI): 943.18(1.66%), 944.16
(20.70%), 945.17(46.44), 946.17 (83.20%), 947.18(37.42%),
948.20(100%), 949.22(48.51%), 950.23(8.69%). Yield: 68%.
Complex 4: Anal. Calc. C, 48.47; H, 3.79; N, 3.90. Found:
C, 48.53; H, 3.75; N, 3.95%. Mass spectrometry (MALDI):
1141.201(1.61%), 1142.218(5.86%), 1143.245(37.91%),
1144.268(64.30%), 1145.281(91.69%), 1146.298(55.45%),
1147.312(100%), 1148.329(69.34%), 1149.347(24.74%),
1150.369(5.58%), 1151.385(1.18%). Yield: 54%. Complex 5:
Anal. Calc. C, 51.63; H, 4.03; N, 4.15. Found: C, 51.39; H,
4.00; N, 4.11%. Mass spectrometry (MALDI): 1053.26
(0.95%), 1054.28(4.11%), 1055.30(32.92%), 1056.32
(68.58%), 1057.35(95.51%), 1058.37(73.62%), 1059.39 (100
%), 1060.41(84.76%), 1061.44(32.93%), 1062.47 (7.27%).
Yield: 46%. DMB: Anal. Calc. C, 45.05; H, 3.48; N, 4.20.
Found: C, 45.38; H, 3.62; N, 4.06%. Mass spectrometry
(MALDI): 1038.127(1.48%), 1039.112(21.81%), 140.13
(51.25%), 1041.156(87.75%), 1042.181(63.81%), 1043.202
(100%),1044.225(82.24%),1045.25(29.52%),1046.289(5.56%).
Additional absorbance and emission spectra, lifetime
complex in methylene chloride. Suitable crystals that were
orange in color of complex 3 were grown from layering a
50/50 mixture of ethyl acetate/hexane on top of a chloro-
form solution of the complex. Diffusion of the hexane/ethyl
acetate layer into the chloroform layer grew the crystals of
3. Clear crystal sections were cut from the raw material and
were mounted in a random orientation on a glass fiber on a
˚
KAPPA CCD diffractometer using Mo Ka (k = 0.71073 A)
radiation. When removed from the mother liquor, the crys-
tals of the larger complexes like 3 rapidly lost solvent,
which would lead to decomposition of the crystal. The
problem of rapid solvent release from the crystals was
solved by preparing and mounting the crystals under oil
followed by rapidly cooling to the 130(2) K temperature
for data collection. The image plate distance was generally
30 mm; however, a distance of 42.3 mm was used for com-
plex 4 and 34.5 mm was used for complex 5. / and x scans
of 0–200° were used in steps of D/ = 2°, Dx = 2° for
dpaene, 1.8° for complex 1, 1° for complexes 2, 4 and 5,
1.5° for complex 3, and 1.3° for complex DMB. Exposure
times per 1° were 15s (dpaene), 20s (complex 1), 30s for
complexes 2 and DMB, 12s (complex 3), 15s (complex 4),
and 40s (complex 5). 2H Ranges are 4.92–60.02° (dpaene),
4.30–55.38° (complex 1), 4.16–52.68° (complex 2), 5.92–
60.26° (complex 3), 4.30–56.68° (complex 4), 4.08–50.40°
˚
(complex 5), resolution limits was at least 0.823 A in all
cases. Further crystallographic data are summarized in
Tables 1–3.
Structure solution and refinement. Cell constants and ori-
entation matrices from collected diffraction data were
obtained by least squares refinements from up to 16894
(dpaene), 16894 (complex 1), 113503 (complex 2), 58044
(complex 3), 151132 (complex 4), 35911 (complex 5), and
104063 (DMB) full and partial reflections. The structures
were solved by direct methods using SIR97 and DIRDIF,
provided by the refinement package MaXus [25]. Missing
atoms were found by difference-Fourier synthesis. The
non-hydrogen atoms were refined with anisotropic temper-
ature factors. Scattering factors are from Waasmaier and
Kirfel [26]. The structures were refined with SHELXL-97
and ORTEP plots were generated with ORTEP32 [27]. Table
1 summarizes the crystal data, experimental conditions,
and refinement data for these structures.
1
measurements, and H NMR are available in the Supple-
mentary Materials section.
X-ray structure determination of dpaene, complexes 1–5,
and DMB. Suitable colorless crystals of the dpaene ligand
were grown from the slow evaporation of heptane. Pris-
matic crystals suitable for diffraction of complexes 1, 5
and DMB were grown from slow evaporation of a solution
of the complex in 2 mL of acetonitile and 18 mL of toluene.
Crystals of complexes 2 (red) and 4 (orange) were grown by
the slow diffusion of ether into a layer of a solution of the
Comꢁplexes 1 and 2 are close to C2-symmetry. However,
the PF6 counter ions were observed in two different posi-
tions in the asymmetric unit, which reduce the overall
Table 5
Elemental analysis, mass spectrometry (m/z) and yield data
Compound
Calculated
C
Found (%)
C
Mass
Yield
H
N
H
N
Complex 1
Complex 2
Complex 4
Complex 5
DMB
45.33
48.55
48.47
51.63
45.05
2.89
3.10
3.79
4.03
3.48
4.23
4.53
3.90
4.15
4.20
45.50
48.15
48.53
51.39
45.38
2.79
2.99
3.75
4.00
3.62
4.33
4.58
3.95
4.11
4.06
1036.0
948.2
1147.3
1059.4
1043.2
72
68
54
46
85

B. Carlson et al. / Inorganica Chimica Acta 359 (2006) 1093–1102
1101
ꢀ
symmetry to P1. Complex 1 crystallized with two disor-
paper. These data can be obtained free of charge via
www.ccdc.cam.ac.uk/conts/retrieving.html (or from The
Director, CCDC, 12, Union Road, Cambridge CB2 1EZ,
UK (fax: +44 1223 336 033; e-mail: deposit@ccdc.cam.
ac.uk or http://www.ccdc.cam.ac.uk).
Complexes. Further details of the crystal structure
investigation(s) may be obtained from the Fachinforma-
tionszentrum Karlsruhe, 76344 Eggenstein-Leopoldshafen,
Germany (fax: (49) 7247 808 666; e-mail: crysdata@fiz-
karlsruhe.de) on quoting the depository numbers CSD-
415295, -415296, -415297, -415298, -415299, and -415300).
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.ica.
2005.11.024.
dered toluene molecules with the disorder introduced by
the center of inversion and an acetonitrile molecule, and
complex 2 crystallized with three methylene chloride mole-
cules per molecule 2. Complex 3 exhibited disorder at car-
bon 45, which was one of the methoxy groups. The larger
complexes (such as 3) co-crystallized with considerable
amount of solvent molecules, which required the aforemen-
tioned precautions during sample preparation and mea-
surement. There were 5 chloroform molecules crystallized
with one molecule 3. Two methylene chloride molecules
accompany one molecule 4. Complex 5 crystallized with
one molecule of toluene per molecule of 5, and one of
the phenyl groups was found to be disordered. This disor-
der is intrinsic. In general, disorder is never related to
reduced data quality: contrary to general believe, a crystal
needs not to show perfect order in all its atoms as long as
all its features obey the space group symmetry. DMB crys-
tallized solvent free from a solution of 2 mL acetonitrile
and 18 mL toluene. The molecular structures and crystallo-
graphic number schemes are illustrated in the ORTEP-32
drawings, Figs. 3–9.
References
[1] C.W. Tang, S.A. Van Slyke, C.H. Chen, Appl. Phys. Lett. 51 (1987)
21.
[2] (a) C. Adachi, M.A. Baldo, S.R. Forrest, M.E. Thompson, Appl.
Phys. Lett. 77 (2000) 2650;
(b) S. Lamansky, P. Djurovich, D. Murphy, F. Abdel-Razzaq, H.-E.
Lee, C. Adachi, P.E. Burrows, S.R. Forrest, M.E. Thompson, J. Am.
Chem. Soc. 123 (18) (2001) 4304;
(c) F. Chen, Y. Yang, M.E. Thompson, J. Kido, Appl. Phys. Lett. 80
(13) (2002) 2308.
All the Os complexes being reported show octahedral
coordination sphere around the central metal atom. The
phenanthroline ligands are arranged by pseudo-twofold
rotation such that each can pi–pi stack with one phenyl
group of the dpaene ligand.
[3] (a) E.S. Handy, A.J. Pal, M.F. Rubner, J. Am. Chem. Soc. 121 (14)
(1999) 3525;
(b) H. Rudmann, S. Shimada, M.F. Rubner, J. Am. Chem. Soc. 124
(17) (2002) 4918;
Absorbance and emission measurements. Absorbance was
measured on a Shimadzu UV-1601 spectrophotometer.
Emission was measured at 22 °C on a Perkin–Elmer
LS50B fluorescence spectrophotometer. The wavelength
sensitivity of the instrument was calibrated prior to mea-
surements using a standard 20 W tungsten lamp of known
output. Photoluminescence (PL) quantum yields to 10%
of the Os complexes (UOs) in de-oxygenated (by argon
purge) acetonitrile solutions were obtained using Ru(II)
tris(2,20-bipyridine) dihexafluorophosphate as the stan-
dard, which has a known quantum yield of 0.042, using
the following equation [28]:
(c) C.-Y. Liu, A.J. Bard, J. Am. Chem. Soc. 124 (16) (2002) 4190;
(d) H. Rudmann, M.F. Rubner, Appl. Phys. 90 (9) (2000) 4338.
[4] (a) B. Carlson, G.D. Phelan, W. Kaminsky, L. Dalton, X. Jiang, S.
Liu, A.K.-Y. Jen, J. Am. Chem. Soc. 124 (47) (2002) 14162;
(b) S. Bernhard, X. Gao, G.G. Malliaras, H.D. Abruna, Adv. Mater.
14 (6) (2002) 433;
(c) Y.G. Ma, H.Y. Zhang, J.C. Shen, C.M. Che, Synth. Met. 94
(1998) 245;
(d) X.Z. Jiang, A.K.-Y. Jen, B. Carlson, L.R. Dalton, Appl. Phys.
Lett. 80 (5) (2002) 713;
(e) Y.-L. Tung, P.-C. Wu, C.-S. Liu, Y. Chi, J.-K. Yu, Y.-H. Hu, P.-
T. Chou, S.-M. Peng, G.-H. Lee, Y. Tao, A.J. Carty, C.-F. Shu, F.-
I. Wu, Organometallics 23 (15) (2004) 3745.
[5] C. Adachi, M.A. Baldo, S.R. Forrest, M.E. Thompson, Appl. Phys.
Lett. 77 (2000) 2650.
[6] (a)V.I.Ogurtsov,D.B.Papkovsky,Sensor.Actutor.BB88(1)(2003)89;
(b) M.P. Xavier, D. Garcia-Fresnadillo, M.C. Moreno-Bondi, G.
Orellana, Anal. Chem. 70 (24) (1998) 5184;
abs Ru area Os
UOs
¼
ꢃ
ꢃ 0:042
ð1Þ
area Ru abs Os
Samples were excited through the LC state at 280 nm with
an absorption of 0.070 for the standard and samples, and
were repeated at an absorption of 0.050 for standard and
samples. Temperature for the measurements was
22 2 °C.
(c) J.M. Costa-Fernandez, M.E. Diaz-Garcia, A. Sanz-Medel, Anal.
Chim. Acta 360 (1–3) (1998) 17;
(d) H.N. McMurray, P. Douglas, C. Busa, M.S. Garley, J. Photochem.
Photobiol. A 80 (1–3) (1994) 283.
[7] F. Schael, S. Engelhard, O. Reich, Chem. Ingenieur Tech. 73 (4)
(2001) 381.
[8] (a) E. Puklin, B. Carlson, S. Gouin, C. Costin, E. Green, S.
Ponomarev, H. Tanji, M. Gouterman, J. Appl. Polym. Sci. 77 (13)
(2000) 2795;
Acknowledgement
This research was supported by the AFOSR through the
Smart Skins MURI center.
(b) H.-F. Ji, Y. Shen, J.P. Hubner, B.F. Carroll, R.H. Schmehl, J.A.
Simon, K.S. Schanze, Appl. Spectrosc. 54 (6) (2000) 856;
(c) S. Gouin, M. Gouterman, J. App. Polym. Sci. 77 (13) (2000) 2815;
(d) G.E. Khalil, C. Costin, J. Crafton, G. Jones, S. Grenoble, M.
Gouterman, J.B. Callis, L.R. Dalton, Sensor Actuat B B97 (1) (2004)
13.
Appendix A. Supplementary data
Crystal structure information: Dpaene. CCDC-268200
contain(s) the supplementary crystallographic data for this
[9] (a) E. Vander-Donckt, B. Camerman, F. Hendrick, R. Herne, R.
Vandeloise, Bull. Soc. Chim. Belg. 103 (1994) 207;

1102
B. Carlson et al. / Inorganica Chimica Acta 359 (2006) 1093–1102
[14] H.B. Kraatz, H. Jacobsen, T. Ziegler, P.M. Boorman, Organomet-
(b) Y. Amao, Y. Ishikawa, I. Okura, Anal. Chim. Acta 445 (2001)
177;
allics 12 (1993) 76.
(c) R. Gao, D.G. Ho, B. Hernandez, M. Selke, D. Murphy, P.I.
Djurovich, M.E. Thompson, J. Am. Chem. Soc. 124 (2002) 14828;
(d) B. Carlson, G. Khalil, M. Gouterman, L. Dalton, Polym. Prepr.
43 (2002) 590;
(e) M.C. DeRosa, D.J. Hodgson, G.D. Enright, B. Dawson, C.E.B.
Evans, R.J. Crutchley, J. Am. Chem. Soc. 126 (24) (2004) 7619.
[15] (a) S.J. Berners-Price, L.A. Colquhoun, P.C. Healy, K.A. Byriel,
J.V. Hanna, J. Chem. Soc., Dalton Trans. 23 (1992) 3357;
(b) P.C. Jones, M.C. Gimeno, Zeit. Fur Krist. 209 (1994) 688.
[16] G. Wulfsberg, in: Principles of Descriptive Inorganic Chemistry,
University Science Books, Mill Valley, CA, 1991, p. 447.
[17] N.C. Norman, Chemistry of Arsenic, Antimony, and Bismuth,
Blackie Academic, London, 1998.
[18] B. Carlson, G.D. Phelan, J.B. Benedict, W. Kaminsky, L. Dalton,
Inorg. Chim. Acta 357 (2004) 3967.
[19] B.J. Dunne, R.B. Morris, A.G. Orpen, J. Chem. Soc., Dalton Trans.
(1991) 653.
[10] (a) F. Jiang, R. Xu, D. Wang, X. Dong, G. Li, Gongneng Cailiao 31
(1) (2000) 72;
(b) K.S. Schanze, B.F. Carroll, S. Korotkevitch, M. Morris, AIAA J.
35 (2) (1997) 306;
(c) F.-Z. Jiang, R. Xu, D.-Y. Wang, X.-D. Dong, G.-C. Li, D.-B.L.-
J. Zhu, J. Mater. Res. 17 (6) (2002) 1312;
(d) H. Sakaue, J.W. Gregory, J.P. Sullivian, AIAA J. 40 (6) (2002)
1094;
[20] D.S. Marynick, J. Am. Chem. Soc. 106 (14) (1984) 4064.
[21] (a) O.F. Wendt, L.I. Elding, J. Chem. Soc., Dalton Trans. 24 (1997)
4725;
(e) R.D. Bowman, K.A. Kneas, J.N. Demas, A. Periasamy, J.
Microsc. 211 (2) (2003) 112;
(b) O.F. Wendt, A. Scodinu, L.I. Elding, Inorg. Chim. Acta 277
(1998) 237;
(f) F.J. Mingoarranz, M.C. Moreno-Bondi, D. Garcia-Fresnadillo,
C. de Dios, G. Orellana, Mikrochim. Acta 121 (1–4) (1995) 107.
[11] J.E. Huheey, E.A. Keiter, R.L. Keiter, Inorganic Chemistry: Princi-
ples of Structure and Reactivity, 4th ed., Harper Collins, New York,
1993, p. 489.
[12] F.A. Cotton, G. Wilkinson, C.A. Murillo, M. Bochmann, Advance
Inorganic Chemistry, 6thed., Wiley-Interscience, New York, 1999, p. 642.
[13] (a) A.G. Orpen, N.G. Connelly, J. Chem. Soc., Chem. Commun. 19
(1985) 1310;
(c) F.W. Heinemann, M. Gerisch, K. Schenzel, D. Steinborn, Zeit.
fuer Krist. 211 (6) (1996) 388.
[22] E.M. Kober, J.V. Caspar, R.S. Lumpkin, T.J. Meyer, Phys. Chem.
90 (1986) 3722.
[23] E. Kober, J. Caspar, P. Sullivan, T. Meyer, Inorg. Chem. 27 (1988)
4587.
[24] S.R. Johnson, T.D. Westmoreland, J.V. Caspar, K.R. Barqawi, T.J.
Meyer, Inorg. Chem. 27 (18) (1988) 3195.
[25] S. Mackay, C. Edwards, A. Henderson, C. Gilmore, N. Stewart, K.
Shankland, A. Donald, MaXus, University of Glasgow, Scotland,
1997.
[26] D. Waasmaier, A. Kirfel, Acta Crystallogr. Sect. A 51 (1995) 416.
[27] (a) L.J. Farrugia, Appl. Cryst. 30 (1997) 565;
(b) G.M. Sheldrick, SHELXL97, University of Go¨ttingen, Germany,
1997.
(b) A.G. Orpen, N.G. Connelly, Organometallics 9 (4) (1990) 1206;
(c) G. Pacchioni, P.S. Bagus, Inorg. Chem. 31 (21) (1992) 4391;
(d) A. Albinati, W.T. Klooster, T.F. Koetzle, J.B. Fortin, J.S. Ricci,
J.F. Eckert, P. Tina, A.J. Lough, R.H. Morris, A.P. Golombek,
Inorg. Chim. Acta 259 (1–2) (1997) 351;
(e) Y. Ohki, H. Suzuki, Angew. Chem., Inter. Ed. 41 (16) (2002) 2994;
(f) R.J. Morris, G.S. Girolami, Inorg. Chem. 29 (21) (1990) 4167.
[28] L. Wallace, D.P. Rillema, Inorg. Chem. 32 (1993) 3836.