Tetrapalladium(II) bisazobenzene and azoazoxybenzene complexes: Syntheses, electronic structures, and optical properties

Article abstract of DOI:10.1021/om200479g

Three new tetranuclear, chloride-bridged Pd^II complexes have been prepared by using 1,4-bis[(E)-phenylazobenzenes (BPABs) and an azoazoxybenzene derivative. The latter was obtained unexpectedly by reacting N,N-dimethyl-4-nitrosoaniline with 2,3

Full text of DOI:10.1021/om200479g

ARTICLE
pubs.acs.org/Organometallics
Tetrapalladium(II) Bisazobenzene and Azoazoxybenzene Complexes:
Syntheses, Electronic Structures, and Optical Properties
Octavia A. Blackburn, Benjamin J. Coe,* and Madeleine Helliwell
School of Chemistry, University of Manchester, Oxford Road, Manchester M13 9PL, U.K.
S Supporting Information
b
ABSTRACT: Three new tetranuclear, chloride-bridged PdII
complexes have been prepared by using 1,4-bis[(E)-phenyla-
zo]benzenes (BPABs) and an azoazoxybenzene derivative. The
latter was obtained unexpectedly by reacting N,N-dimethyl-4-
nitrosoaniline with 2,3,5,6-tetramethylphenylenediamine in gla-
cial acetic acid. Characterization includes 1H NMR and UVꢀvis
spectroscopies, and single-crystal X-ray structures have been
obtained for two of the proligands and one complex. As
expected, the methyl groups on the central aryl rings cause
severe departures from planarity. The proligands show an
intense UV absorption (λmax ≈ 350 nm) due to π f π* intramolecular charge transfer, together with weaker n f π* bands in
the visible region. The complexes display an intense visible band (λmax = 449ꢀ475 nm) and multiple near-UV absorptions. While
the electronic absorption spectra of the BPAB complexes are quite similar, the azoazoxybenzene complex shows a doubly intense
visible band. Dichroic ratios (DRs) measured in two different liquid crystal (LC) hosts show that complexation is not beneficial, as
noted for dinuclear azobenzene complexes previously. The shape of the new complexes appears to give relatively poor alignment
within the LC. Time-dependent density functional theory reveals differing origins of the low-energy transitions for BPAB and
azoazoxybenzene complexes, those for the latter involving charge transfer within the NdN(O)C6H3NMe2 moieties. The origin of
the low-energy transition of the BPAB complex also differs from that in the related dinuclear complexes. The directions of the μ12
vectors calculated with respect to the long molecular axis generally correlate with the observed DR values, with better alignment for
the low-energy transitions that are relatively more dichroic.
’ INTRODUCTION
Palladacyclic complexes have relevance to many emerging appli-
cations involving for example catalysis,1 nonlinear optics,2 or liquid
crystalline materials.3 We described recently studies with dinuclear
cyclopalladated azobenzenes (e.g., 1ꢀ4, Figure 1),4 as part of our
work on the dichroic properties of metal complexes, a relatively
unexplored area.5 In this context, we have studied also NiII and
PdII complexes of linearly connected azobenzene ligands.6
Dichroic dyes show orientation-dependent absorption intensi-
ties and are of interest for potential use in guestꢀhost liquid
crystal displays (GH-LCDs). In a GH-LCD, an applied electric
field is used to manipulate the orientation of the molecules and
thus change the appearance of the display.7 This technology has
promise for applications such as electronic paper.8 The dichroic
ratio (DR) is the ratio of the absorbances when illuminated by
Figure 1. Dominant isomeric structures of the chloride-bridged
light polarized parallel (A||) and perpendicular (A^) to the liquid
dipalladium(II) complexes studied previously.4
crystal (LC) director. DR values large enough for commercial use
require the dye molecules to be well aligned in the LC matrix,
with the transition dipole moment (μ12) of the color-inducing
electronic transition(s) coinciding with the long molecular axis.
In this article, we describe the synthesis and properties of
tetranuclear PdII complexes that incorporate bisazobenzene or
azoazoxybenzene ligands to create a rectangular motif. While
dinuclear cyclopalladated azobenzenes are well-known, to our
knowledge cyclopalladation of bisazobenzenes or azoazoxyben-
zenes is unprecedented. In order to obtain the desired motif,
palladation of the central aryl rings is prevented by methyl
Received: June 3, 2011
Published: August 30, 2011
r
2011 American Chemical Society
4910
dx.doi.org/10.1021/om200479g Organometallics 2011, 30, 4910–4923
|

Organometallics
ARTICLE
acid (37%, 4 mL) cooled in an ice bath to below 5 ꢀC was added
dropwise a solution of NaNO2 (0.92 g, 13.3 mmol) in water (10 mL).
The solution was stirred at 0ꢀ5 ꢀC for 1 h. This orange diazonium salt
solution was then added slowly to a cooled solution of phenol (1.15 g,
12.2 mmol) and NaOH (1.12 g, 28.0 mmol) in water (20 mL),
maintained below 5 ꢀC for the duration of the addition. The resulting
suspension was stirred with cooling for 1 h, then allowed to warm to
room temperature and stirred for a further 1 h. The suspension was
filtered to remove a dark brown precipitate, and the filtrate was acidified
to pH 5 by using 1 M HCl. The resulting orange precipitate was filtered
off, and crude 5 obtained by silica gel column chromatography eluting
with 92:8 dichloromethane/ethyl acetate (yield 80 mg): δH (400 MHz,
CD3COCD3) 9.12 (2 H, br, OH), 7.86 (4 H, d, J = 8.9, B), 7.05
(4 H, d, J = 8.9, A), 2.12 (12 H, s, 4Me); m/z (ꢀES) 373 ([M ꢀ H]ꢀ).
This material was combined with 1-bromoheptane (84 mg, 0.469 mmol)
and K2CO3 (118 mg, 0.854 mmol) in DMF (10 mL), and the mixture
heated under reflux for 3 h. The resulting solution was poured into cold
water, and the precipitate was filtered off and recrystallized from ethanol
to give an orange solid (70 mg, 2%): δH (400 MHz, CDCl3) 7.92 (4 H, d,
J = 8.9, B), 7.03 (4 H, d, J = 8.9, A), 4.05 (4 H, t, J = 6.6, 2OCH2), 2.12
(12 H, s, 4Me), 1.83 (4 H, qnt, J = 6.5, 2CH2), 1.52ꢀ1.28 (16 H, 8CH2),
0.90 (6 H, t, J = 6.7, 2Me); m/z (+ES) 571 ([M + H]+). Anal. Calcd (%)
for C36H50N4O2: C, 75.8; H, 8.8; N, 9.8. Found: C, 75.6; H, 9.0; N, 9.8.
Diffraction-quality crystals were grown by layering hexane on top of a
dichloromethane solution.
Synthesis of 1-Amino-4-[4-(ethoxycarbonyl)-(E)-phenylazo]-
2,3,5,6-tetramethylbenzene (9). To a stirred solution of ethyl-4-
aminobenzoate (1.50 g, 9.08 mmol) in dichloromethane (30 mL) was
added a solution of oxone (11.20 g, 18.2 mmol) in water (120 mL). The
mixture was stirred under argon for 1 h, and the organic phase acquired a
green color. The layers were separated, and the aqueous phase washed
with dichloromethane (2 ꢁ 20 mL). The combined organic layers
were washed with 1 M HCl (2 ꢁ 50 mL), saturated aqueous NaHCO3
(2 ꢁ 50 mL), and water (2 ꢁ 50 mL), then dried over MgSO4 and
evaporated to give crude 8 as a pale yellow solid (1.27 g): δH (400 MHz,
CDCl3) 8.30 (2 H, d, J = 8.8, C6H4), 7.93 (2 H, d, J = 8.8, C6H4), 4.43
(2 H, q, J = 7.1, CH2), 1.43 (3 H, t, J = 7.1, Me); m/z (ꢀES) 180 ([M +
H]ꢀ). This material was used without further purification due to limited
stability; it was combined with TMPD (1.16 g, 7.06 mmol) in toluene
(50 mL) and glacial acetic acid (2 mL) and heated at reflux for 24 h. After
cooling to room temperature, the dark red solution was washed with water
(3 ꢁ 50 mL), dried over MgSO4, and evaporated to dryness. Purification
was achieved by silica gel column chromatography eluting with dichlor-
omethane to yield a dark red solid (465 mg, 16%): δH (400 MHz, CDCl3)
8.17 (2 H, d, J = 8.5, D), 7.87 (2 H, d, J = 8.5, C), 4.42 (2 H, q, J = 7.1,
CH2), 3.93 (2 H, s, NH2), 2.38 (6 H, s, 2Me), 2.17 (6 H, s, 2Me), 1.43
(3 H, t, J = 7.1, CO2CH2Me): m/z (+ES) 326 ([M + H]+), 348 ([M +
Na]+). Anal. Calcd (%) for C19H23N3O2: C, 70.1; H, 7.1; N, 12.9.
Found: C, 69.9; H, 7.1; N, 12.8.
Figure 2. The structures of (E)-azobenzene (a) and BPAB in its anti (b)
and syn (c) conformations, showing the approximate μ12 vectors (pink
dashed lines). The solid blue lines represent the long molecular axes.
substituents. Related tetranuclear complexes of diimine ligands
have been investigated previously for their liquid crystalline
properties.9,10
Azobenzenes are commonly used as dichroic dyes owing to
their long slender shape and the relatively close alignment of μ12
for the intense π f π* intramolecular charge-transfer (ICT)
transition with the long molecular axis. Given that the DRs of
bisazobenzenes exceed those of azobenzenes,11 we anticipated
that the new tetrapalladium complexes might outperform their
dinuclear counterparts studied previously. Furthermore, the
angle between the μ12 vector and the long molecular axis for a
1,4-bis[(E)-phenylazo]benzene (BPAB) in its anti conforma-
tion (Figure 2) has been calculated to be 4.38ꢀ,12 compared to
17ꢀ for (E)-azobenzene.13 In the tetrapalladated structures
described here, the ligands are forced to adopt a syn arrange-
ment, in which the μ12 vector coincides exactly with the long
molecular axis.
’ EXPERIMENTAL SECTION
Materials and Procedures. The complex PdCl2(NCPh)214 was
synthesized by published methods. All other chemicals were used as
supplied by Sigma Aldrich or Alfa Aesar, and the LCs used for dichroism
measurements were supplied by Merck and Synthon. Products were
dried at room temperature overnight in a vacuum desiccator (CaSO4) or
by direct attachment to a high-vacuum line for several hours prior to
characterization.
1
General Physical Measurements. H NMR spectra were re-
corded on Bruker UltraShield 500 or AV 400 spectrometers, and all
shifts are quoted with respect to TMS. The fine splitting of phenyl ring
AA0BB0 patterns is ignored, and the signals are reported as simple
“doublets”, with “J values” (in Hz) referring to the two most intense
peaks. Electrospray mass spectrometry was performed on a Micromass
Platform II, and MALDI mass spectra were recorded on a Shimadzu
Axima Confidence instrument with dichloromethane as the solvent. In
cases where clusters of peaks are observed due to the presence of various
isotopes, the most intense peak is always quoted. Elemental analyses
were performed by the Microanalytical Laboratory, University of
Manchester. UVꢀvis spectra were obtained by using a Shimadzu UV-
2401 PC spectrometer. Dichroism measurements were performed on
a custom-built Hewlett-Packard machine supplied by HP Research
(U.K.). Abbreviations: br = broad; s = singlet; d = doublet; dd = doublet
of doublets; t = triplet; q = quartet; qnt = quintet.
Synthesis of 1-[4-(Ethoxycarbonyl)-(E)-phenylazo]-4-[4-n-
heptoxy-(E)-phenylazo]-2,3,5,6-tetramethylbenzene (11H2).
9 (130 mg, 0.399 mmol) was dissolved in glacial acetic acid (15 mL)
and propionic acid (2.4 mL) and cooled to below 5 ꢀC in an ice bath.
Nitrosyl sulfuric acid (40 wt % solution in sulfuric acid, 140 mg,
0.441 mmol) was added slowly to the stirred dark red solution at 0ꢀ5 ꢀC,
and stirring was continued for 1 h. This diazonium salt solution was then
added slowly to a cooled solution of phenol (41 mg, 0.436 mmol) and
NaOH (48 mg, 1.20 mmol) in water (15 mL). The pH was maintained
above 7 by adding 2 M aqueous NaOH, and the temperature kept below
5 ꢀC during the addition. The red suspension was stirred for 1 h with
cooling, then allowed to warm to room temperature and stirred for a further
1 h. The solid was filtered off and purified by silica gel column chromatog-
raphy eluting with dichloromethane/ethyl acetate (99:1), followed
by recrystallization from ethanol to give 10 as a red solid (30 mg): δH
Synthesis of 1,4-Bis[4-n-heptoxy-(E)-phenylazo]-2,3,5,6-
tetramethylbenzene (6H2). To a stirred solution of 2,3,5,6-tetra-
methylphenylenediamine (TMPD, 1.00 g, 6.09 mmol) in hydrochloric
4911
dx.doi.org/10.1021/om200479g |Organometallics 2011, 30, 4910–4923

Organometallics
ARTICLE
Table 1. Crystallographic Data and Refinement Details for the Proligands 6H2 and 15H2 and Complex 7
6H2
15H2
7
formula
C
36H50N4O2
C31H41N5O2
515.69
C73H98Cl6N8O4Pd4
1789.89
M
570.80
cryst syst
triclinic
triclinic
triclinic
space group
P1
P1
P1
a/Å
5.4082(9)
9.1179(15)
16.912(3)
98.067(3)
96.340(3)
102.416(3)
797.8(2)
1
5.6665(9)
9.3701(14)
27.180(4)
82.750(3)
87.528(3)
75.106(3)
1383.4(4)
2
10.730(3)
11.944(3)
16.333(4)
97.487(4)
94.532(4)
115.473(4)
1852.2(8)
1
b/Å
c/Å
R/deg
β/deg
γ/deg
U/Å3
Z
T/K
100(2)
100(2)
100(2)
μ/mmꢀ1
0.074
0.079
1.225
cryst size/mm
appearance
no. of reflns collected
no. of indep reflns (Rint
reflns with I > 2σ(I)
goodness-of-fit on F2
0.60 ꢁ 0.50 ꢁ 0.20
orange plate
4696
0.60 ꢁ 0.60 ꢁ 0.02
yellow plate
8641
0.30 ꢁ 0.25 ꢁ 0.10
yellow plate
14 280
)
3194 (0.0355)
2310
3978 (0.0843)
1891
7395 (0.0293)
5855
0.981
0.777
1.105
final R1, wR2 [I > 2σ(I)]
0.0457, 0.1045
0.0688, 0.1291
0.274, ꢀ0.285
0.0436, 0.0639
0.1197, 0.0796
0.166, ꢀ0.148
0.0485, 0.0991
0.0679, 0.1060
0.987, ꢀ0.877
(all data)
peak and hole/e Åꢀ3
(500 MHz, CDCl3) 8.23 (2 H, d, J = 8.7, D), 7.96 (2 H, d, J = 8.7, C), 7.89
(2H, d, J=8.8, B), 6.96 (2H, d, J=8.8, A), 5.53(1H, br, OH), 4.44 (2H, q,
J = 7.1, CH2), 2.19 (6 H, s, 2Me), 2.12 (6 H, s, 2Me), 1.44 (3 H, t, J = 7.1,
CO2CH2Me); m/z (+ES) 431 ([M + H]+); m/z (ꢀES) 429 ([M ꢀ H]ꢀ).
This compound (90 mg) was reacted with 1-bromoheptane (45 mg, 0.251
mmol) and K2CO3 (58 mg, 0.420 mmol) in DMF (10 mL) in a manner
identical to 6H2. The product was purified by recrystallization from hexane/
ethanol to give an orange-red solid (36 mg, 6%): δH (400 MHz, CDCl3)
8.23 (2 H, d, J = 8.8, D), 7.97ꢀ7.90 (4 H, B + C), 7.04 (2 H, d, J = 9.1, A),
4.43 (2 H, q, J = 7.1, CO2CH2), 4.06 (2 H, t, J = 6.6, OCH2), 2.20 (6 H, s,
2Me), 2.12 (6 H, s, 2Me), 1.84 (2 H, qnt, J = 6.7, CH2), 1.53ꢀ1.29 (11 H,
C4H8 + CO2CH2Me), 0.91 (3 H, t, J = 7.0, Me); m/z (+ES) 529 ([M +
H]+), 551 ([M + Na]+). Anal. Calcd (%) for C32H40N4O3: C, 72.7; H, 7.6;
N, 10.6. Found: C, 72.5; H, 7.5; N, 10.3.
Synthesis of 1-Amino-4-[4-(dimethylamino)-(E)-phenyl-
ONN-azoxy]-2,3,5,6-tetramethylbenzene (13). N,N-Dimethyl-
4-nitrosoaniline (1.00 g, 6.66 mmol) and TMPD (1.09 g, 6.64 mmol)
were combined in glacial acetic acid (100 mL) and stirred under a flow of
argon for 24 h. The dark green solution was diluted with water (100 mL),
and the product extracted with dichloromethane (3 ꢁ 50 mL). The
combined organic extracts were washed with water (2 ꢁ 100 mL), dried
over MgSO4, and evaporated to give a dark khaki solid. Purification was
achieved by recrystallization from ethanol/dichloromethane to yield a
gold-colored solid, and further crops were obtained by subsequent
recrystallizations of the filtrate (1.19 g, 57%): δH (400 MHz, CDCl3)
8.23 (2 H, d, J = 9.4, C), 6.68 (2 H, d, J = 9.4, D), 3.58 (2 H, br, NH2),
3.07 (6 H, s, NMe2), 2.11 (6 H, s, 2Me), 2.00 (6 H, s, 2Me); m/z (+ES)
313 ([M + H]+). Anal. Calcd (%) for C18H24N4O: C, 69.2; H, 7.7; N,
17.9. Found: C, 69.0; H, 7.8; N, 17.6.
(37%, 1.0 mL) and water (20 mL) gave a dark yellow-green solution.
Coupling with phenol (0.32 g, 3.40 mmol) in 1 M aqueous NaOH (7.8 mL)
produced a yellow-brown precipitate. Recrystallization from dichloro-
methane/ethanol gave 14 as a yellow solid (250 mg): δH (400 MHz,
CD3SOCD3) 8.14 (2 H, d, J = 9.3, C), 7.77 (2 H, d, J = 8.8, B), 6.96 (2 H, d,
J = 8.8, A), 6.81 (2 H, d, J = 9.4, D), 3.05 (6 H, s, NMe2), 2.04 (6 H, s, 2Me),
1.92 (6 H, s, 2Me); m/z (+ES) 418 ([M + H]+), 440 ([M + Na]+); m/z
(ꢀES) 416 ([M ꢀ H]ꢀ). This compound (200 mg) was reacted with
1-bromoheptane (107 mg, 0.597 mmol) and K2CO3 (137 mg, 0.991 mmol)
in DMF (15 mL) in a manner identical to 6H2. The crude product was
recrystallized from dichloromethane/ethanol to give an orange solid (175
mg, 13%): δH (400 MHz, CDCl3) 8.26 (2 H, d, J = 9.3, C), 7.91 (2 H, d, J =
9.0, B), 7.02 (2 H, d, J = 9.0, A), 6.71 (2 H, d, J = 9.4, D), 4.05 (2 H, t, J = 6.5,
OCH2), 3.09 (6 H, s, NMe2), 2.12 (6 H, s, 2Me), 2.03 (6 H, s, 2Me), 1.84
(2 H, qnt, J = 6.7, CH2), 1.53ꢀ1.25 (8 H, C4H8), 0.91 (3 H, t, J = 7.0, Me);
m/z (+ES) 517 ([M + H]+), 539 ([M + Na]+). Anal. Calcd (%) for
C31H41N5O2: C, 72.2; H, 8.0; N, 13.6. Found: C, 72.1; H, 8.3; N, 13.6.
Diffraction-quality crystals were grown by slow evaporation of a dichlor-
omethane solution.
Synthesis of Tetra(μ-chloro)bis{μ-[2,3,5,6-tetramethyl-1,4-
phenylenebis[(E)-azo-4-(n-heptoxy)-1,2-phenylene]]}tetra-
palladium, Pd4(μ-Cl)4(6)2 (7). 6H2 (30 mg, 0.053 mmol) and PdCl2
(NCPh)2 (45 mg, 0.117 mmol) were combined in ethanol (20 mL), and
the mixture was heated under reflux for 8 h. After cooling, the ochre
precipitate was filtered off and washed with a small amount of ethanol
(yield 35 mg, 77%): δH (400 MHz, CDCl3) 7.77 (4 H, d, J = 8.6, c), 6.95
(4 H, d, J = 2.5, a), 6.74 (4 H, dd, Jbc = 8.4, Jab = 2.5, b), 4.12 (8 H, t, J = 6.4,
4OCH2), 2.21 (24 H, s, 8Me), 1.85 (8 H, qnt, J = 6.7, 4CH2), 1.54ꢀ1.28
(32 H, 4C4H8), 0.91 (12 H, t, J = 7.2, 4Me); m/z (+MALDI) 1668 ([M ꢀ
Cl]+), 1729 ([M + Na]+). Anal. Calcd (%) for C72H96Cl4N8O4Pd4: C,
50.7; H, 5.7; N, 6.6. Found: C, 50.7; H, 5.7; N, 6.5. Diffraction-quality
crystals were grown by layering ethanol on top of a dichloromethane
solution.
Synthesis of 1-[4-(Dimethylamino)-(E)-phenyl-ONN-azoxy]-
4-(4-n-heptoxyphenylazo)-2,3,5,6-tetramethylbenzene (15H2).
Precursor 14 was prepared in a manner identical to 5. Addition of NaNO2
(0.26 g, 3.77 mmol) to 13 (1.00 g, 3.20 mmol) in concentrated HCl
4912
dx.doi.org/10.1021/om200479g |Organometallics 2011, 30, 4910–4923

Organometallics
ARTICLE
Scheme 1. Syntheses of Bisazobenzene Proligands and Their Chloride-Bridged Tetrapalladium(II) Complexes, Showing the
Possible Structural Isomers and 1H NMR Labelsa
a Reagents and solvents: (i) NaNO2/HCl then NaOH/phenol; (ii) n-C7H15Br/K2CO3/DMF; (iii) PdCl2(NCPh)2 in ethanol under reflux.
Synthesis of Tetra(μ-chloro)bis{μ-[2,3,5,6-tetramethyl-1-[(E)-
azo-4-(n-heptoxy)-1,2-phenylene]-4-[(E)-azo-4-(ethoxycarbonyl)-
1,2-phenylene]phenylene]}tetrapalladium, Pd4(μ-Cl)4(11)2
(12). This compound was prepared in a manner identical to 7 by
using 11H2 (30 mg, 0.057 mmol) and PdCl2(NCPh)2 (44 mg, 0.115
mmol). Purification by recrystallization from dichloromethane/etha-
nol and washing with diethyl ether gave a yellow solid (32 mg, 69%).
1H NMR spectroscopy reveals a ca. 1:1 mixture of isomers with very
closely overlapped signals: δH (400 MHz, CDCl3) 8.15ꢀ8.12 (4 H, 2s,
d + d0), 8.00ꢀ7.94 (8 H, e + e0, f + f0), 7.80ꢀ7.76 (4 H, 2d, c + c0),
7.00ꢀ6.95 (4 H, 2d, a + a0), 6.77ꢀ6.73 (4 H, 2dd, b + b0), 4.49ꢀ4.41 (8
H, 4CO2CH2), 4.17ꢀ4.09 (8 H, 4OCH2), 2.23 (24 H, s, 8Me), 2.21
(24 H, s, 8Me), 1.85 (8 H, qnt, J = 7.1, 4CH2), 1.54ꢀ1.30 (44 H,
4C4H8 + 4CO2CH2Me), 0.92 (12 H, t, J = 6.8, 4Me); m/z (+MALDI)
1586 ([M ꢀ Cl]+), 1644 ([M + Na]+). Anal. Calcd (%) for
C64H76Cl4N8O6Pd4: C, 47.4; H, 4.7; N, 6.9. Found: C, 47.8; H, 4.7;
N, 6.7.
Synthesis of Tetra(μ-chloro)bis{μ-[2,3,5,6-tetramethyl-
1-[(E)-ONN-azoxy-4-(dimethylamino)-1,2-phenylene]-4-[(E)-azo-
4-(n-heptoxy)-1,2-phenylene]phenylene]}tetrapalladium,
Pd4(μ-Cl)4(15)2 (16). This compound was prepared in a manner
identical to 7 by using 15H2 (40 mg, 0.078 mmol) and PdCl2(NCPh)2
(68 mg, 0.177 mmol). The product was purified by filtration of a
dichloromethane solution through Celite, followed by evaporation of
the solvent to yield an orange solid (35 mg, 56%). 1H NMR spectros-
copy reveals a dominant isomer with ca. 20% of a minor isomer. Major
isomer: δH (400 MHz, CDCl3) 7.80 (2 H, d, J = 8.6, c), 7.36 (2 H, d, J =
9.1, f0), 6.96 (2 H, d, J = 2.5, a), 6.77ꢀ6.73 (4 H, b + d), 6.44 (2 H, dd, Jef =
10.0, Jed = 2.5, e), 4.10 (4 H, t, J = 6.6, 2OCH2), 3.16 (12 H, s, NMe2),
2.20 (12 H, s, 4Me), 2.17 (12 H, s, 4Me), 1.84 (4 H, qnt, J = 6.7, 2CH2),
1.52ꢀ1.29 (16 H, 2C4H8), 0.91 (6 H, t, J = 6.9, 2Me); m/z (+MALDI)
1560 ([M ꢀ Cl]+), 1596 ([M + H]+), 1619 ([M + H + Na]+), 1633
([M + K]+). Anal. Calcd (%) for C62H78Cl4N10O4Pd4: C, 46.7; H, 4.9;
N, 8.8. Found: C, 46.4; H, 5.3; N, 8.3.
4913
dx.doi.org/10.1021/om200479g |Organometallics 2011, 30, 4910–4923

Organometallics
ARTICLE
Scheme 2. Synthesis of the Azoazoxybenzene Proligand
15H2 and Its Chloride-Bridged Tetrapalladium(II) Complex,
Showing the Possible Structural Isomers and 1H NMR Labelsa
asymmetric unit. The atom Cl2 is disordered over two sites, with
occupancies of 60(2):40(2), and the C6Me4 group is also disordered
over two sites, with occupancies of 55(1):45(1). Therefore, for the
disordered atoms Cl2 and Cl2A ISOR was used to reduce ellipsoid
elongation and thus prevent the adjacent ellipsoids from encroaching on
each other’s space. Also, the atoms of the C6Me4 unit (C1ꢀC10) were
refined isotropically because of the near coincidence of some of them.
Crystallographic data and refinement details are presented in Table 1.
Dichroic Ratio Measurements. DR values were measured by
using proprietary equipment and software courtesy of Hewlett-Packard
Ltd. LC cells were purchased from Linkam Scientific, and syringe filters
were purchased from Fisher Scientific UK Ltd. A dye (ca. 1 wt %) was
dissolved in the LC by stirring above the clearing temperature for ca. 16 h.
Slow cooling was followed by filtration through a PTFE syringe filter of
pore size 0.2 μm to remove any undissolved dye. Glass cells, with a
5 μm wide gap and rubbed polyimide surface to give antiparallel
alignment, were filled by capillary action above the clearing temperature
of the LC and allowed to cool. The absorption of the solution was
measured parallel and perpendicular to the direction of rubbing by using
polarized light and 90ꢀ rotation of the cell. The DR (A^/A ) was
calculated at the absorption maximum, except when this lies below the
low wavelength limit of the spectrometer (380 nm). A cell containing
only the LC was used as a reference, and the quoted DR values are
averages from at least two separate measurements. The error on each
value is approximated to be a maximum of (10%, as determined from
repeated measurements with a standard black dye (Merck).
Theoretical Studies. Structure optimization (B3LYP/Def2-SVP/
SDD)19ꢀ21 and time-dependent density functional theory (TD-DFT)
calculations were conducted on models of complexes 7 and 16 (denoted
70 and 160) by using the Gaussian 03 suite of programs.22 The first 80
excited states were calculated by using the B3LYP exchangeꢀcorrelation
functional and Def-2 SVP/SDD basis sets with inclusion of a COSMO23
solvent continuum model (dichloromethane). Wavelength ranges of
ca. 323ꢀ446 nm (70) and 329ꢀ468 nm (160) were covered. The
UVꢀvis absorption spectra were simulated by using the GaussSum
program.24
’ RESULTS AND DISCUSSION
Synthesis. Two tetrapalladium(II) complexes of bisazoben-
zene ligands (7 and 12) and one of an azoazoxybenzene ligand
(16) have been prepared (Schemes 1 and 2). The symmetric
dialkoxy-substituted proligand 6H2 was synthesized by a double
diazonium coupling of TMPD with phenoxide, followed by
alkylation with 1-bromo-n-heptane. This coupling reaction gave
a very low yield, which was not improved by increasing the
reaction time or by using the more powerful diazotizing agent
nitrosyl sulfuric acid.
a The reagents and reaction conditions (i)ꢀ(iii) are as in Scheme 1.
X-ray Crystallography. Crystal structures of the proligands 6H2
and 15H2 and of the complex 7 have been obtained. Data were collected
on a Bruker APEX CCD X-ray diffractometer by using Mo KR radiation
(λ = 0.71073 Å), and the data were processed by using the Bruker
SAINT15 and SADABS16 software packages. The structures were solved
by direct methods by using SHELXS-9717 and refined by full-matrix
least-squares on all Fo2 data by using SHELXL-97.18 With the exception
of 7, all non-hydrogen atoms were refined anisotropically, and hydrogen
atoms were included in idealized positions by using the riding model,
with thermal parameters of 1.2 times those of aromatic parent carbon
atoms and 1.5 times those of methyl parent carbons. For 6H2, there is
half a molecule in the asymmetric unit, with the other half generated by
inversion. For 15H2, the data were cut at 0.9 Å resolution because the
crystal diffracted rather weakly at high angle. The atoms C27ꢀC31 of
the C7H15 group are disordered over two semioccupied sites. For 7,
there is half a Pd complex and half a molecule of dichloromethane in the
The syntheses of proligands 11H2 and 15H2 both begin with
condensations of substituted nitrosoarenes with TMPD. Ethyl-4-
nitrosobenzoate (8) was prepared by oxidizing ethyl-4-amino-
benzoate with Oxone (2KHSO5 KHSO4 K2SO4), following an
3
3
adapted literature procedure.25 Condensing 8 with TMPD in
toluene with an acetic acid catalyst yielded the azobenzene (9) in
moderate yield (20%). Using instead acetic acid as the solvent led
to self-reaction of 8 to give an asymmetric product, possibly an
azoxybenzene. Diazonium coupling of 9 with phenoxide afforded
the bisazobenzene 10 in moderate yield (17%). Acetic and
propionic acids were used as solvents, with nitrosyl sulfuric acid
as the diazotizing agent, owing to the low aqueous solubility of 9.
Reacting commercial N,N-dimethyl-4-nitrosoaniline with
TMPD in acetic acid gave the unexpected azoxybenzene com-
pound 13. To our knowledge, the formation of an azoxybenzene
4914
dx.doi.org/10.1021/om200479g |Organometallics 2011, 30, 4910–4923

Organometallics
ARTICLE
Figure 3. Aromatic regions of the 1H NMR spectra of complexes 7 (top), 12 (middle), and 16 (bottom) recorded at 400 MHz in CDCl3 at 293 K. See
Schemes 1 and 2 for proton labels.
Table 2. UVꢀVis Absorption Data for the New Proligands and Their Tetranuclear PdII Complexes in Dichloromethanea
terminal substituents
proligand
λmax, nm (ε, 103 Mꢀ1 cmꢀ1)/Emax, eV
complex
λmax, nm (ε, 103 Mꢀ1 cmꢀ1)/Emax, eV
OC7H15-n
6H2
454 (4.2)/2.73
345 (32.1)/3.59
7
451 (26.0)/2.75
380 (49.5)/3.26
322 (43.8)/3.85
449 (18.9)/2.76
381 (43.5)/3.25
310 (50.5)/4.00
475 (48.8)/2.61
373 (33.8)/3.32
335 (35.4)/3.70
300 (35.4)/4.13
OC7H15-n/CO2Et
OC7H15-n/NMe2
11H2
15H2
467 (3.7)/2.65
350 (29.2)/3.54
12
16
367 (34.7)/3.38
320 (19.5)/3.87b
a Solutions ca. 2 ꢁ 10ꢀ6 to 1 ꢁ 10ꢀ5 M. The ε values are the averages from measurements made at four different concentrations (with ε showing no
significant concentration dependence). b Shoulder.
by single-step reaction of a nitrosoarene and an aniline has
not been documented previously. Attempts to synthesize a
symmetric bisazoxybenzene by increasing the proportion of
N,N-dimethyl-4-nitrosoaniline in the reaction gave only an
asymmetric product via self-reaction of the nitroso compound
(as above). Nitrosoarenes often form azoxybenzenes in acidic
conditions, probably via a radical mechanism.26
1H NMR Spectroscopy. The proligands 6H2, 11H2, and 15H2
all show the expected AA0BB0 signals, and the aromatic regions of
the spectra of the complexes are shown in Figure 3. Signals were
assigned with the aid of COSY studies where appropriate. The
highly symmetric complex 7 gives three aromatic signals
(Figure 3, top), a singlet due to the eight methyl groups, and
one set of heptoxy signals. The signals for protons aꢀc shift
upfield, indicating a relative shielding, on complexation of 6H2.
The largest shift (0.29 ppm) occurs for proton b, located para
with respect to the site of palladation.
The spectra of complexes 12 and 16 are more complicated
than that of 7 due to lower symmetry and the presence of cis and
trans isomers (Schemes 1 and 2), which give strongly overlapped
signals. While 12 comprises a ca. 1:1 isomeric ratio, one isomer
dominates for 16. In dinuclear chloride-bridged palladium com-
plexes of azobenzene ligands, the trans isomer is favored due to
the differing trans effects of the C- and N-based donor sites.
However, in the present complexes, the cis and trans isomers
have the same arrangement of donor sites about the metal (see
Crystallography below), so the distribution of isomers depends
on the substituents, and the presence of the azoxy moiety in 16.
The chloride-bridged tetrapalladium complexes 7, 12, and 16
were synthesized by reacting the bisazobenzene and azoazox-
ybenzene proligands with PdCl2(NCPh)2 in refluxing ethanol.
The products precipitated out in good yields (70ꢀ92%). For
asymmetric ligands, a mixture of cis and trans isomers (referring
to the relative orientations of ligands) was produced, as ascer-
1
tained by H NMR spectroscopy (see below). Attempts to
produce analogous acetate-bridged complexes, structurally simi-
lar to reported imine-based species,9,10 were unsuccessful. These
observations may be due to steric congestion of the methyl
groups as a result of the nonplanar nature of the bridging
acetate ligands. To our knowledge, 16 is the first example of a
mixed azobenzene/azoxybenzene chloride-bridged palladium(II)
complex.
4915
dx.doi.org/10.1021/om200479g |Organometallics 2011, 30, 4910–4923

Organometallics
ARTICLE
For both 12 and 16, the signals for the protons para to the site of
palladation (e and e0) shift upfield by ca. 0.27 ppm on complexing
11H2 and 15H2, respectively. In 16, the signals for the protons
f and f0 shift upfield by 0.90 ppm when compared with 15H2. This
large shift may be attributable to the proximity of these protons to
the azoxy oxygen (formally Oꢀ) enforced by complexation.
Electronic Spectroscopy. The proligands 6H2, 11H2, and 15H2
and complexes 7, 12, and 16 have been characterized by UVꢀvis
spectroscopy in dichloromethane. The results are presented in
Table 2, and representative spectra are shown in Figure 4. Note
that 12 and 16 were studied as cis/trans isomeric mixtures.
6H2 and 11H2 display intense absorption bands with respec-
tive λmax values of 345 and 350 nm due to π f π* ICT transitions
(Figure 4a). In each case, a weaker band at lower energy of n f
π* origin is also apparent. Both bands shift bathochromically on
replacing one of the heptoxy groups with ꢀCO2Et, with a larger
relative change for the lower energy absorption. 15H2 gives an
intense ICT band with λmax = 367 nm, but in this case the n f π*
transition is not resolved. The red-shift of the ICT band on
moving from 6H2/11H2 to 15H2 is attributable to the strongly
electron-donating ꢀNMe2 substituent, while the diminished
intensity of the n f π* transition is due to participation of
n-electrons in the NꢀO bond.27 The high-energy shoulder on
the ICT band of 15H2 may be ascribed to a π f π* transition
where the donor orbital involves the nitrogens and oxygen of the
azoxy moiety.28
In acetonitrile solutions, the ICT absorption of BPAB is
shifted bathochromically relative to that of (E)-azobenzene
(λmax = 361 cf. 316 nm), with an almost doubled molar extinction
coefficient (ε = 53 cf. 29 ꢁ 103 Mꢀ1 cmꢀ1).29 Matsui et al. have
measured the UVꢀvis spectra of a large series of BPAB deriva-
tives, almost all of which have λmax > 400 nm in hexane.12,30
Therefore, the λmax and ε values for 6H2 and 11H2 are lower than
might be expected. However, their ε values are comparable with
those reported for several tetrafluoro-p-phenylene-centered dyes
in hexane.12 Clearly, the nonplanarity induced by the methyl
substituents, as observed in the solid state for 6H2 and 11H2 (see
below), attenuates π-conjugation, causing λmax and ε to decrease.
On formation of the tetrapalladated structures, at least three
bands are observed between 270 and 550 nm (Figure 4b). The
spectra of 7 and 12 are similar, but that of 16 is distinctly
different. This variation in optical absorption behavior is evident
also in the solid state, 7 and 12 appearing dull yellow, while 16 is
bright orange.
Complexes 7 and 12 both display three resolved absorption
maxima, with the two higher energy bands having similar ε values
and the lowest energy band being roughly half as intense. These
observations contrast with the behavior of related dipalladium
complexes 1ꢀ4 (Figure 1), for which all three bands have comp-
arable intensities.4 The intensities of the bands at ca. 450 nm for 7
and 12 aresimilar to those of 1ꢀ4 and diminishedwhencompared
with the ICT bands of their proligands. The latter change may be
attributed to decreased coplanarity of the aryl rings on com-
plexation. The intensities of the higher energy bands for 7 and 12
are about twice those of 1ꢀ4, correlating with the number of
Figure 4. UVꢀvis absorption spectra in dichloromethane recorded at
293 K: (a) the proligands 6H2 (red), 11H2 (green), and 15H2 (blue);
(b) the complexes 7 (red), 12 (green), and 16 (blue).
Figure 5. Representation of the molecular structure of 6H2 (50% probability ellipsoids).
Figure 6. Representation of the molecular structure of 15H2, with the disorder of the heptyl chain removed for clarity (50% probability ellipsoids).
4916
dx.doi.org/10.1021/om200479g |Organometallics 2011, 30, 4910–4923

Organometallics
ARTICLE
Figure 7. Representation of the molecular structure of 7, with the disorder of the inner Cl atoms/C6Me4 groups and the H atoms removed for clarity
(50% probability ellipsoids).
{Pd2Cl2}2+ units. The λmax values of 7 and 12 differ significantly
only for the highest energy band, which is blue-shifted in 12 when
compared with 7. Slight differences in ε are also observed, and a
high-energy shoulder appears on the 381 nm band of 12
(Figure 4b). These differences are attributable to lowered sym-
metry and the presence of isomers for 12.
When compared with the lowest energy bands of 7 and 12, that
of complex 16 is approximately twice as intense and red-shifted.
In contrast, the higher energy absorptions of 16 are less intense than
those of 7 and 12. Very little attention has been paid previously to
the UVꢀvis spectra of cyclopalladated azoxybenzenes, but one
report notes three bands (λmax = 230, 329, and 433 nm in
dichloromethane) for a symmetric dipalladium complex.27
Crystallography. Single-crystal X-ray structures have been ob-
tained for the proligands 6H2 and 15H2 and complex 7. Represen-
tations of the molecular structures are shown in Figures 5ꢀ7, and
selected bond distances and angles are given in Table 3.
Only two other structures containing BPAB units have been
reported,31,32 while structures of related 1,2/1,3-isomeric com-
pounds are more common. The structure of 6H2 shows an anti
conformation (Figure 5). The large dihedral angle of 72.71(8)ꢀ
between the central and outer aryl rings is attributable to the
steric effect of the methyl substituents. The structure of unsub-
stituted BPAB also shows an anti conformation, but deviates only
slightly from planarity (by ca. 3ꢀ),31 whereas the derivative
Disperse Orange 29 (DO29, 1-[4-hydroxy-(E)-phenylazo]-
4-[4-nitro-(E)-phenylazo]-2-methoxybenzene) adopts a syn
conformation with dihedral angles of 10.9ꢀ and 14.1ꢀ.32 The
NdN bond distances in 6H2 are slightly longer than those in
BPAB or DO29, perhaps owing to the methyl substituents. In
DO29, the NdN bond nearest to the ꢀOMe group is elongated
by ca. 0.02 Å when compared with the one on the other side of
the molecule.32
The structure of 15H2 (Figure 6) shows that the azoxy oxygen
atom is bonded to the nitrogen nearest to the ꢀNMe2 group, and
the terminal five carbons of the heptoxy chain are disordered over
two sites. To our knowledge, this is the first crystal structure of an
azoazoxybenzene. However, there are many reported structures
of azoxybenzenes, including their cyclopalladated complexes,33
and bisazoxybenzenes.34 Zaleski and co-workers have used X-ray
crystallography extensively in their studies on para-substituted
azoxybenzenes.35
Table 3. Selected Interatomic Distances (Å) and Angles
(deg) for the Proligands 6H2 and 15H2 and Complex 7
6H2
15H2
7
NdN
CꢀN
1.260(2)
1.249(3)
1.273(3)a
1.461(3)a
1.443(3)a
1.443(3)
1.437(3)
1.271(5)
1.274(5)
1.393(5)c
1.384(6)c
1.44(1)d
1.48(1)d
1.48(1)d
1.43(1)d
1.434(2)b
1.424(2)
NꢀO
1.265(2)
PdꢀC
1.948(4)
1.952(4)
1.993(4)
1.993(4)
2.321(1)
2.314(1)
2.399(3)e
2.507(6)e
2.475(4)e
2.408(4)e
111.3(4)c
111.4(4)c
111.1(5)d
119.7(6)d
111.1(5)d
120.6(6)d
PdꢀN
PdꢀCl
CꢀNꢀN
113.7(1)
113.5(1)
117.2(2)a
113.5(2)a
113.3(2)
113.6(2)
NꢀNꢀO
CꢀPdꢀN
126.2(2)
79.2(2)
79.3(2)
95.75(4)
89.7(1)e
88.7(2)e
PdꢀClꢀPd
a Involving the NNO fragment. b Central. c Within the chelate rings.
d Involving the (disordered) C6Me4 units. e Involving the inner (disordered)
Cl atoms (i.e., trans to C).
While the azo NdN bond distance in 15H2 is similar to that in
6H2, the azoxy NdN distance is longer by ca. 0.02 Å; this and the
NꢀO distance have values typical for azoxybenzenes. Also as for 6H2,
15H2 shows large dihedral angles between the terminal and central
aryl rings; 72.2(2)ꢀ and 74.7(2)ꢀ for the heptoxy- and ꢀNMe2-
substituted rings, respectively. The slight difference in these twists
4917
dx.doi.org/10.1021/om200479g |Organometallics 2011, 30, 4910–4923

Organometallics
ARTICLE
Table 4. Dichroic Ratios Measured in LC Hosts K15 and ZLI-2293 for UVꢀVis Bands at g380 nm
DR
DR
LC
K15
terminal substituent
proligand
380 nm
complex
380 nm
460ꢀ480 nm
OC7H15-n
6H2
5.3
6.3
5.2
8.1
5.5
7.2
7
2.6
2.7
2.4
2.4
2.6
2.4
4.9
5.6
3.6
4.5
4.8
3.6
ZLI-2293
K15
OC7H15-n/CO2Et
OC7H15-n/NMe2
11H2
15H2
12
16
ZLI-2293
K15
ZLI-2293
Table 5. Selected Interatomic Distances (Å) and Angles
(deg) for the Optimized (B3LYP/Def2-SVP/SDD) Structures
of 70 and 160
70
160
NdN
CꢀN
1.26
1.31a
1.26
1.42a,b
1.38b
1.43a
1.44
1.38b
1.44
NꢀO
PdꢀC
PdꢀN
1.23
1.98
2.04
1.98
2.04
2.05a
Figure 8. Electronic absorption profiles measured parallel (blue) and
perpendicular (red) to the LC director for complex 7 in K15.
PdꢀCl
2.40
2.41
2.49c
113.4b
116.5
2.48c
114.6a,b
113.3b
116.7a
116.6
123.7
80.2a
78.9
may be due to the presence of the azoxy oxygen atom. The NNO
moiety is essentially coplanar with the ꢀNMe2-substituted ring, as is
commonly observed in azoxybenzenes, possibly due to interaction of
the oxygen with the adjacent ring.35b The ꢀNMe2 group is also
almost perfectly coplanar with the adjacent ring. While the CꢀN(O)
distance is comparable with previously reported data, the other azoxy
CꢀN distance (1.443(3) Å) is slightly longer than corresponding
literature values,35 possibly due to the methyl groups. Deformation
about the azoxy moiety, which includes an NꢀNꢀO angle of
126.2(2)ꢀ, is typically noted for azoxybenzene structures, regardless
of the torsion angle of the aromatic rings.35a,d
The structure of 7 (Figure 7) contains an inversion center and
crystallizes with one molecule of dichloromethane per PdII
complex. The core of the molecule, comprising four metallocyc-
lic units, is essentially planar. The inner chlorine atoms are
disordered over two positions with unequal occupancy, forming
a Cl2AꢀCl1ꢀCl2 angle of ca. 12.0ꢀ. The central C6Me4 rings are
also disordered over two sites unequally and form large dihedral
angles with the two outer rings (ca. 74.3ꢀ88.0ꢀ). The latter
angles are increased significantly with respect to the proligand
6H2 due to the additional steric influence of the bridging Cl
atoms. The twisting observed in 7 is comparable to, although
generally larger than, that observed in the chloride-bridged
dipalladium complex of 2,6-dimethyl-(E)-azobenzene.36
CꢀNꢀN
NꢀNꢀO
CꢀPdꢀN
78.9
PdꢀClꢀPd
95.8
91.7c
96.2
92.1c
a Involving the NNO fragment. b Within the chelate ring. c For the inner
Cl atoms.
The PdꢀCl distances are longer when trans to C than N (Table 3),
due to the greater structural trans effect of the carbanionic
ligand.37,38 The inner PdꢀClꢀPd bond angles are therefore smaller
than those to the outer Cl atoms. The PdꢀC/N distances and the
CꢀPdꢀN angles are within the ranges expected for cyclopalladated
complexes.36,38 The PdꢀC distances (1.948(4)/1.952(4) Å) are
shorter than expected for single bonds, suggesting some multiple-
bond character.9 The intramolecular Pd Pd separation in 7 is
3 3 3
typical for chloride-bridged cyclopalladated complexes (3.437(1) Å)
and is naturally shorter than corresponding distances in bromide/
iodide-bridged tetrapalladated structures (3.662 and 3.888 Å,
respectively).9,10
Dichroic Ratios. DR values of the proligands and complexes
measured in the LCs K15 (4-pentyl-40-cyanobiphenyl) and ZLI-
229339 are shown in Table 4. Representative UVꢀvis spectra for
complex 7 in K15 are shown in Figure 8. For the complexes, DRs for
the two lowest energy bands were recorded. In most cases, K15 gives
The azo bonds in 7 are all locked in an E-conformation and
forced into a mutually syn arrangement. While the NdN distances
do not change significantly on complexation, the CꢀN distances
within the chelate rings shorten by ca. 0.04 Å with respect to 6H2.
The bridging chlorides are forced to sit trans to two nitrogens or
two carbanionic ligands, whereas in dinuclear chloride-bridged
species this can be avoided with a trans arrangement of ligands.
4918
dx.doi.org/10.1021/om200479g |Organometallics 2011, 30, 4910–4923

Organometallics
ARTICLE
Table 6. Data from TD-DFT Calculations Based on B3LYP/
Def2-SVP/SDD Geometries of Model Complexes 70 and 160
complex
ΔE (eV)
fos
major contributions
μ12 (D)
70
2.79
0.04
HOMOꢀ1 f LUMO
HOMO f LUMO+1
2.02
2.80
2.85
0.06
0.08
HOMOꢀ6 f LUMO
HOMOꢀ2 f LUMO
HOMOꢀ9 f LUMO+2
HOMOꢀ6 f LUMO
HOMOꢀ1 f LUMO+3
HOMOꢀ15 f LUMO+3
HOMOꢀ14 f LUMO
HOMOꢀ10 f LUMO+1
HOMOꢀ2 f LUMO
HOMOꢀ10 f LUMO+4
HOMOꢀ7 f LUMO
HOMOꢀ7 f LUMO
HOMOꢀ2 f LUMO+3
HOMO f LUMO+1
2.44
2.64
3.19
0.11
3.07
3.20
3.25
0.15
0.12
3.47
3.06
3.37
0.56
HOMOꢀ9 f LUMO+2
HOMOꢀ4 f LUMO+1
HOMOꢀ2 f LUMO
HOMOꢀ4 f LUMO+2
HOMOꢀ13 f LUMO+1
HOMOꢀ12 f LUMO
HOMOꢀ7 f LUMO+3
HOMOꢀ13 f LUMO+2
HOMOꢀ12 f LUMO+3
HOMO f LUMO
6.63
3.39
3.39
3.58
0.33
0.21
0.10
5.08
4.04
2.75
Figure 9. Respective TD-DFT-calculated (blue) and experimental
(green) UVꢀvis spectra of (a) 70 and 7; (b) 160 and 16. The ε-axes
refer to the experimental data only, and the vertical axes of the calculated
data are scaled to match the main experimental absorptions. The fos-axes
refer to the individual calculated transitions (red).
3.61
2.68
2.80
0.06
0.02
0.81
2.13
1.30
8.73
160
HOMO f LUMO+1
lower DR values than ZLI-2293 since the lower clearing tempera-
ture of the former corresponds with a lower order parameter (S) at
room temperature, and S generally correlates with DR.40
HOMOꢀ1 f LUMO+2
HOMOꢀ1 f LUMO+3
HOMO f LUMO+2
Compared with the azobenzenes we have studied previously,4
the new bisazobenzene and azoazoxybenzene proligands do not
show significantly larger DR values. Although it is established
that increasing the number of azo groups enhances dichroism,
adding lateral aryl substituents produces an opposite effect as the
breadth of the dye increases.11 The lower-than-expected DR
values obtained for 6H2, 11H2, and 15H2 are probably attribu-
table to their twisted conformations caused by the methyl
substituents.
Generally, the data for complexes 7, 12, and 16 are reminiscent
of our studies with 1ꢀ4 (Figure 1), with relatively larger DR
values at longer wavelengths.4 In both LCs, the DR at 380 nm
decreases distinctly on complexing 6H2, 11H2, and 15H2.
Notably, the DR values of 7, 12, and 16 are lower than those
observed for 1ꢀ4 at 380 nm, especially in ZLI-2293. However,
when measured at 460ꢀ480 nm, both tetra- and dipalladium
complexes give similar DR values. Even at these wavelengths, 7,
12, and 16 are still less dichroic than their proligands (although
some of the differences may not be significant).
HOMO f LUMO+3
2.82
3.16
0.02
0.44
HOMOꢀ10 f LUMO
HOMOꢀ9 f LUMO
HOMOꢀ6 f LUMO+1
HOMOꢀ2 f LUMO+1
HOMOꢀ11 f LUMO+5
HOMOꢀ11 f LUMO+6
HOMOꢀ10 f LUMO+4
HOMOꢀ8 f LUMO+1
HOMOꢀ6 f LUMO+1
HOMOꢀ11 f LUMO+3
HOMOꢀ7 f LUMO+2
HOMOꢀ6 f LUMO+3
HOMOꢀ8 f LUMO+1
HOMOꢀ5 f LUMO
HOMOꢀ16 f LUMO+1
HOMOꢀ14 f LUMO
1.27
6.06
3.22
3.23
0.10
0.08
2.90
2.61
3.36
3.64
0.56
0.16
6.64
3.61
Our results imply that μ12 alignment with the molecular axis is
most effective for the lower energy transitions of 7, 12, and 16.
Decreasing DR on complexation suggests less efficient ordering
within the LC matrix and/or skewing of μ12 away from the
molecular axis. It seems likely that the lack of planarity and large
size of the complexes may disrupt their alignment within the LC.
At 380 nm, all three complexes show very similar DRs in both
LCs, while at 460ꢀ480 nm, 12 has a DR lower than those of 7
and 16 in K15. In contrast, in ZLI-2293 at 460ꢀ480 nm, the DR
decreases steadily on moving along the series 7 f 12 f 16.
These observations indicate that only the shape of the central core is
important and that the μ12 directions are similar for analogous
4919
dx.doi.org/10.1021/om200479g |Organometallics 2011, 30, 4910–4923

Organometallics
ARTICLE
Figure 10. TD-DFT-derived pictures of orbitals involved in transitions associated with the low-energy absorption band of 70 (isosurface value 0.03 au).
Figure 11. TD-DFT-derived pictures of HOMOs involved in higher energy transitions of 70 (isosurface value 0.03 au).
bands, when measuring at 380 nm. However, replacing two of the
heptoxy chains with ꢀCO2Et or ꢀNMe2 substituents does affect
the dichroic behavior at longer wavelengths.
Theoretical Studies. In order to rationalize the measured
UVꢀvis and DR data, TD-DFT calculations were carried out
on the methoxy analogues (denoted 70 and 160) of complexes 7
and 16 by using Gaussian 03.22 For 160, only the trans isomer is
considered.
angles between the central and outer rings of 85.47ꢀ and 82.42ꢀ.
The optimized geometry of 160 resembles that of 70. The relative
extension of the azoxy compared with the azo NdN distance,
observed crystallographically for 15H2, also occurs in 160. A
11.82ꢀ departure from planarity and dihedral angles between the
central and outer rings of 87.36ꢀ and 75.61ꢀ are predicted for 160.
The UVꢀvis spectra for 70 and 160 simulated by using
GaussSum24 are shown in Figure 9, with the experimental
spectra of 7 and 16. Predicted transition energies are presented in
Table 6 and MOs involved in major transitions are depicted in
Figures 10ꢀ13. The lowest energy bands are blue-shifted with
respect to the experimental spectra, especially for 160. Multiple
transitions contribute to this band for 70, while in 160 one transition
dominates. 160 also shows fewer significant transitions at higher
Selected optimized bond distances and angles are given in
Table 5. Comparison of the crystallographic data for 7 with
the optimized geometry of 70 shows reasonable agreement, the
largest discrepancies being +0.09 Å (PdꢀCl) and ca. +3ꢀ
(PdꢀClꢀPd). However, the optimized structure is more dis-
torted with a bend about the Cl Cl axis of 12.41ꢀ and dihedral
3 3 3
4920
dx.doi.org/10.1021/om200479g |Organometallics 2011, 30, 4910–4923

Organometallics
ARTICLE
Figure 12. TD-DFT-derived pictures of orbitals associated with the transition at ΔE = 2.80 eV for 160 (isosurface value 0.03 au).
Figure 13. TD-DFT-derived pictures of selected orbitals associated with the higher energy transitions of 160 (isosurface value 0.03 au).
4921
dx.doi.org/10.1021/om200479g |Organometallics 2011, 30, 4910–4923

Organometallics
ARTICLE
Figure 14. Calculated μ12 vectors (scaled to show direction) of 70 for transitions at (a) low energy and (b) high energy.
energies when compared with 70. In both cases, the predicted
absorption profiles do not match well with those observed, although
the presence of isomers of 16 is ignored in the calculations. For 70,
the discrepancy between the relative intensities of low- and high-
energy bands is large, as also noted in our previous studies with
dinuclear complexes.4 However, the oscillator strengths fos of the
predicted transitions correlate with the observed doubly intense
low-energy band for 16 when compared with 7. Three transitions
with fos < 0.1 form this band in 70, while the major low-energy
transition in 160 has fos = 0.81.
dimethyl-4-nitrosoaniline and TMPD in acidic conditions
yielded an unexpected azoxybenzene precursor. 1H NMR spectra
reveal that mixtures of cis and trans isomers are formed with
asymmetric ligands. Single-crystal X-ray structures of two proli-
gands and one complex show that the methyl groups on the
central aryl rings cause severe departures from planarity, espe-
cially in the complex, where the chloride bridges provide addi-
tional steric hindrance. UVꢀvis spectra are also influenced by the
structural distortions, decreased conjugation resulting in lower-
than-expected π f π* transition intensities. While the BPAB
complexes give quite similar spectra, the azoazoxybenzene com-
plex displays a doubly intense low-energy transition. The results of
DR measurements are reminiscent of those for related dinuclear
azobenzene complexes, with complexation proving unbeneficial.
Relatively poor alignment within the LC may result from the bulky
shape of the complexes. The DR values of the low-energy bands
are larger than those determined at higher energies. TD-DFT
calculations reveal that the origins of the low-energy transitions
differ for BPAB and azoazoxybenzene complexes, the latter
involving charge transfer from the C6H3NMe2 groups, explaining
the observed spectral differences. The origin of the low-energy
transition of the BPAB complex is also quite different from that in
the related dinuclear complexes. The directions of the μ12 vectors
with respect to the long molecular axes generally correlate with the
observed DR values, since alignment is best for the low-energy
transitions.
For the low-energy band of 70, the HOMOs involve the BPAB
ligands, metals, and chloride bridges, while the LUMOs are
primarily different phase combinations of BPAB π* orbitals
(Figure 10). These results contrast with those for related dinuclear
species, for which this band involves almost exclusively azoben-
zene-based transitions.4 The break in conjugation caused by the
twisted central rings influences the nature of the contributing MOs
in 70. The orbitals involved in the higher energy transitions show
distributions similar to those mentioned above, with the HOMOs
showing contributions from all over the complex (Figure 11).
The orbital origins of the lowest energy band of 160 contrast
with those of 70. The dominant transition at ΔE = 2.80 eV
involves the orbitals shown in Figure 12 where the degenerate
pairs HOMO/HOMOꢀ1 and LUMO+2/LUMO+3 are located
on the cyclopalladated NdN(O)C6H3NMe2 units. This transi-
tion has charge-transfer character (from amino donor to azoxy
acceptor) and accounts for the differences observed in the low-
energy region of the electronic spectra of 7/12 and 16. Two
transitions are dominant at higher energy; selected orbitals
involved are shown in Figure 13.
The μ12 directions have been approximated for the major
calculated transitions. For 70, the μ12 vectors of the two dominant
low-energy transitions align well with the long molecular axis,
while the lowest energy transition (ΔE = 2.79 eV) shows a
moderate degree of alignment. (Figure 14a). At higher energies,
some transitions retain good alignment, while others show
perpendicular μ12 vectors (Figure 14b). The lowest energy
transition with substantial intensity (ΔE = 2.80 eV) of 160 also
shows good axial alignment of μ12, with poorer alignment at
higher energies (see Supporting Information, Figure S1). These
results correlate broadly with the observed relatively higher DR
values for the low-energy bands.
’ ASSOCIATED CONTENT
S
Supporting Information. Crystallographic information in
b
CIF format; Cartesian coordinates of theoretically optimized geo-
metries for 70 and 160; calculated μ12 vectors of 160. This material is
available free of charge via the Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
*E-mail: b.coe@manchester.ac.uk.
’ ACKNOWLEDGMENT
We thank DyStar U.K. Ltd and the University of Manchester
for support. We are grateful to Dr. Michael G. Hutchings
(DyStar) for initiating this project and also to Dr. Joseph J. W.
McDouall (Manchester) for advice concerning TD-DFT calcu-
lations. Dr. Robin G. Pritchard (Manchester) is also gratefully
acknowledged for rerefining two of the X-ray crystal structures.
Hewlett-Packard Ltd are thanked for the use of the dichrometer
and generous donations of LCs.
’ CONCLUSION
New BPAB derivatives and an azoazoxybenzene have been
cyclopalladated to yield new chloride-bridged tetranuclear
palladium(II) complexes with a rectangular motif, using methyl
groups to block unwanted palladation. Reaction between N,N-
4922
dx.doi.org/10.1021/om200479g |Organometallics 2011, 30, 4910–4923

Organometallics
ARTICLE
’ REFERENCES
Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian
03, Revision C.02; Gaussian, Inc.: Wallingford, CT, 2004.
(23) Eckert, F.; Klamt, A. AIChE J. 2002, 48, 369.
(24) O’Boyle, N. M.; Tenderholt, A. L.; Langner, K. M. J. Comput.
Chem. 2008, 29, 839.
(1) (a) Herrmann, W. A.; B€ohm, V. P. W.; Reisinger, C.-P.
J. Organomet. Chem. 1999, 576, 23. (b) Dupont, J.; Pfeffer, M.; Spencer,
J. Eur. J. Inorg. Chem. 2001, 1917. (c) Beletskaya, I. P.; Cheprakov, A. V.
J. Organomet. Chem. 2004, 689, 4055. (d) Dupont, J.; Consorti, C. S.;
Spencer, J. Chem. Rev. 2005, 105, 2527.
(2) (a) Buey, J.; Coco, S.; Díez, L.; Espinet, P.; Martín-Alvarez, J. M.;
Miguel, J. A. Organometallics 1998, 17, 1750. (b) Aiello, I.; Dattilo, D.;
Ghedini, M.; Golemme, A. J. Am. Chem. Soc. 2001, 123, 5598. (c) Aiello,
I.; Dattilo, D.; Ghedini, M.; Bruno, A.; Termine, R.; Golemme, A. Adv.
Mater. 2002, 14, 1233. (d) Termine, R.; Talarico, M.; Aiello, I.; Dattilo,
D.; Pucci, D.; Ghedini, M.; Golemme, A. Opto-Electron. Rev. 2005, 13,
287. (e) Espinet, P.; Miguel, J. A.; Martín-Alvarez, J. M.; Villacampa, B.
J. Organomet. Chem. 2010, 695, 437.
(3) Ghedini, M.; Aiello, I.; Crispini, A.; Golemme, A.; La Deda, M.;
Pucci, D. Coord. Chem. Rev. 2006, 250, 1373. (b) Buey, J.; Espinet, P.
J. Organomet. Chem. 1996, 507, 137.
(4) Blackburn, O. A.; Coe, B. J. Organometallics 2011, 30, 2212.
(5) (a) Marshall, K. L.; Painter, G.; Lotito, K.; Noto, A. G.; Chang, P.
Mol. Cryst. Liq. Cryst. 2006, 454, 47. (b) Bruce, D. W.; Dunmur, D. A.;
Hunt, S. E.; Maitlis, P. M.; Orr, R. J. Mater. Chem. 1991, 1, 857.
(6) Blackburn, O. A.; Coe, B. J.; Fielden, J.; Helliwell, M.; McDouall,
J. J. W.; Hutchings, M. G. Inorg. Chem. 2010, 49, 9136.
(25) Priewisch, B.; R€uck-Braun, K. J. Org. Chem. 2005, 70, 2350.
(26) Cardellini, L.; Carloni, P.; Damiani, E.; Greci, L.; Stipa, P.;
Rizzoli, C.; Sgarabotto, P. J. Chem. Soc., Perkin Trans. 2 1994, 1589.
(27) Ghedini, M.; Pucci, D.; Crispini, A; Aiello, I.; Barigelletti, F.;
Gessi, A.; Francescangeli, O. Appl. Organomet. Chem. 1999, 13, 565.
(28) (a) Tosi, C. Spectrochim. Acta 1966, 22, 1701. (b) Gore, P. H.;
Wheeler, O. H. J. Am. Chem. Soc. 1956, 78, 2160.
(29) Cisnetti, F.; Ballardini, R.; Credi, A.; Gandolfi, M. T.; Masiero,
S.; Negri, F.; Pieraccini, S.; Spada, G. P. Chem.ꢀEur. J. 2004, 10, 2011.
(30) Matsui, M.; Nakagawa, H.; Joglekar, B.; Shibata, K.; Muramat-
su, H.; Abe, Y.; Kaneko, M. Liq. Cryst. 1996, 21, 669.
(31) Gilardi, R. D.; Karle, I. L. Acta Crystallogr., Sect. B 1972, 28,
1635.
(32) Freeman,H.S.;McIntosh,S.A.;Singh,P.Dyes Pigments 1997,35,11.
(33) Ghedini, M.;Morrone, S.;DeMunno, G.;Crispini, A.J. Organomet.
Chem. 1991, 415, 281.
(34) Wakahara, A.; Fujiwara, T.; Ikeda, M.; Tamura, Y.; Tomita, K.-I.
Acta Crystallogr. Sect. C 1983, 39, 99.
(35) (a) Ejsmont, K.; Domaꢀnski, A.; Kyzioz, J. B.; Zaleski, J. Acta
Crystallogr. Sect. C 2000, 56, 697. (b) Domaꢀnski, A.; Ejsmont, K.; Kyzioz,
J. B.; Zaleski, J. Acta Crystallogr. Sect. C 2001, 57, 467. (c) Ejsmont, K.;
Domaꢀnski, A.; Kyzioz, J. B.; Zaleski, J. Acta Crystallogr. Sect. C 2004,
60, o368. (d) Ejsmont, K.; Domaꢀnski, A.; Kyzioz, J. B.; Zaleski, J. J. Mol.
Struct. 2005, 753, 92.
(7) (a) Heilmeier, G. H.; Zanoni, L. A. Appl. Phys. Lett. 1968, 13, 91.
(b) White, D. L.; Taylor, G. N. J. Appl. Phys. 1974, 45, 4718.
(8) (a) Harris, S. Nat. Photonics 2010, 4, 748. (b) Sobel, A. Nat.
Mater. 2003, 2, 643. (c) Graham-Rowe, D. Nat. Photonics 2007, 1, 248.
(9) Vila, J. M.; Gayoso, M.; Pereira, M. T.; Rodriguez, M. C.;
Ortigueira, J. M.; Thornton-Pett, M. J. Organomet. Chem. 1992, 426, 267.
(10) Praefcke, K.; Diele, S.; Pickardt, J.; G€undo^gan, B.; N€utz, U.;
Singer, D. Liq. Cryst. 1995, 18, 857.
(36) Armentano, S.; Crispini, A.; De Munno, G.; Ghedini, M.; Neve,
F. Acta Crystallogr. Sect. C 1991, 47, 966.
(37) Barr, N.; Dyke, S. F.; Smith, G.; Kennard, C. H. L.; McKee, V.
J. Organomet. Chem. 1985, 288, 109.
(38) Armentano, S.; Crispini, A.; De Munno, G.; Ghedini, M.; Neve,
F. Acta Crystallogr. Sect. C 1991, 47, 2545.
(39) The precise composition of ZLI-2293 is unknown to us.
However, 1H NMR spectroscopy obtained in CDCl3 reveals it to
contain primarily 4,40-disubstituted cyclohexylbenzene derivatives with
long alkyl chains.
(40) (a) Ichinose, H.; Sato, H.; Naemura, S. Mol. Cryst. Liq. Cryst.
2004, 409, 401. (b) Cognard, J.; Phan, T. H. Mol. Cryst. Liq. Cryst. 1981,
68, 207.
(11) (a) Bahadur, B.; Sarna, R. K.; Bhide, V. G. Mol. Cryst. Liq. Cryst.
1981, 75, 121. (b) Bloom, A.; Hung, P. L. K. Mol. Cryst. Liq. Cryst. 1977,
40, 213. (c) Constant, J.; Raynes, E. P.; Shanks, I. A.; Coates, D.; Gray,
G. W.; McDonnell, D. G. J. Phys. D: Appl. Phys. 1978, 11, 479.
(12) Matsui, M.; Tanaka, N.; Andoh, N.; Funabiki, K.; Shibata, K.;
Muramatsu, H.; Ishigure, Y.; Kohyama, E.; Abe, Y.; Kaneko, M. Chem.
Mater. 1998, 10, 1921.
(13) Robin, M. B.; Simpson, W. T. J. Chem. Phys. 1962, 36, 580.
(14) Fan, G.-Z.; Li, T.; Li, G.-X. Appl. Organomet. Chem. 2006, 20,
656.
(15) SAINT (Version 6.45); Bruker AXS Inc.: Madison, WI, USA,
2003.
(16) SADABS (Version 2.10); Bruker AXS Inc.: Madison, WI, USA,
2003.
(17) Sheldrick, G. M. Acta Crystallogr., Sect. A 1990, 46, 467.
(18) Sheldrick, G. M. SHELXS 97, Programs for Crystal Structure Analysis
(Release 97-2); University of G€ottingen: G€ottingen, Germany, 1997.
(19) Becke, A. D. J. Chem. Phys. 1993, 98, 5648.
(20) Weigend, F.; Ahlrichs, R. Phys. Chem. Chem. Phys. 2005, 7,
3297.
(21) Andrae, D.; H€aussermann, U.; Dolg, M.; Stoll, H.; Preuss, H.
Theor. Chim. Acta 1990, 77, 123.
(22) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, V. G.; Montgomery, J. A., Vreven, Jr. T.;
Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.;
Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson,
G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.;
Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.;
Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Adamo, C.;
Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.;
Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.;
Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich,
S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.;
Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.;
Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.;
Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham,
M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.;
4923
dx.doi.org/10.1021/om200479g |Organometallics 2011, 30, 4910–4923