Organoimido-Polyoxometalate Nonlinear Optical Chromophores: A Structural, Spectroscopic, and Computational Study

Article abstract of DOI:10.1021/acs.inorgchem.7b00708

Ten organoimido polyoxometalate (POM)-based chromophores have been synthesized and studied by hyper-Rayleigh scattering (HRS), Stark and Resonance Raman spectroscopies, and density functional theory (DFT) calculations. HRS β₀ values for chromop

Full text of DOI:10.1021/acs.inorgchem.7b00708

Article
pubs.acs.org/IC
Organoimido-Polyoxometalate Nonlinear Optical Chromophores: A
Structural, Spectroscopic, and Computational Study
Ahmed Al-Yasari,† Nick Van Steerteghem, Hayleigh Kearns, Hani El Moll, Karen Faulds,#
,∇
‡
#
†,∥
Joseph A. Wright,† Bruce S. Brunschwig, Koen Clays, and John Fielden*
§
‡
,†
†
School of Chemistry, University of East Anglia, Norwich, NR4 7TJ, United Kingdom
‡
Department of Chemistry, University of Leuven, Celestijnenlaan 200D, B-3001 Leuven, Belgium
§
Beckman Institute, California Institute of Technology, 1200 East California Blvd., MC 139-74, Pasadena, California 91125, United
States
#
Department of Pure and Applied Chemistry, University of Strathclyde, 99 George Street, Glasgow, G1 1RD, United Kingdom
∇College of Pharmacy, University of Kerbala, Kerbala, Iraq
*
S Supporting Information
ABSTRACT: Ten organoimido polyoxometalate (POM)-based
chromophores have been synthesized and studied by hyper-Rayleigh
scattering (HRS), Stark and Resonance Raman spectroscopies, and
density functional theory (DFT) calculations. HRS β values for
0
chromophores with resonance electron donors are significant (up to
30
+
1
39 × 10− esu, ∼5 times greater than that of the DAS cation), but
systems with no donor, or the −NO acceptor show no activity, in
some cases, despite large DFT-predicted β-values. In active systems
2
with short (phenyl) π-bridges, β values comfortably exceed that of the
0
purely organic structural analogue N,N-dimethyl-4-nitroaniline
3
/2
(
DMPNA), and intrinsic β-values, β /N (where N is the number
0
of bridge π-electrons) thus appear to break empirical performance
3/2
limits (β /N vs λmax) for planar organic systems. However, β values
0
0
obtained for extended systems with a diphenylacetylene bridge are
comparable to or lower than that of their nitro analogue, N,N-dimethyl-4-[(4-nitrophenyl)ethynyl]-aniline (DMNPEA).
Resonance Raman spectroscopy confirms the involvement of the POM in the electronic transitions, whether donor groups are
present or not, but Stark spectroscopy indicates that, in their absence, the transitions have little dipolar character (hence, NLO
inactive), consistent with DFT-calculated frontier orbitals, which extend over both POM and organic group. Stark and DFT also
suggest that β is enhanced in the short compounds because the extension of charge transfer (CT) onto the POM increases
changes in the excited-state dipole moment. With extended π-systems, this effect does not increase CT distances, relative to a
−
NO acceptor, so β values do not exceed that of DMNPEA. Overall, our results show that (i) the organoimido−POM unit is
2
0
an efficient acceptor for second-order NLO, but an ineffective donor; (ii) the nature of electronic transitions in arylimido-POMs
is strongly influenced by the substituents of the aryl group; and (iii) organoimido-POMs outperform organic acceptors with short
π-bridges, but lose their advantage with extended π-conjugation.
INTRODUCTION
components. For example, connection of lacunary Keggin
■
n−
n−
(
XW O ) or Dawson (X W O ) POMs to organic or
11 39 2 17 61
Polyoxometalates (POMs) are a class of anionic, molecular
metal oxide clusters, whose range of properties is commensu-
rate with their enormous variety of structural types. POMs can
be derivatized with organic groups, and recent years have seen
great advances in the synthetic chemistry of such POM
5
−7
metallo-organic chromophores
can create systems where
1
photoexcitation of the chromophore leads to multiple charge
6c,7
accumulation on the electron accepting POM, and reduction
7
of protons to dihydrogen. However, electronic isolation of the
“
hybrids”both in terms of initial derivatization of the POM,
POM from the organic subunits means that these properties are
an enhancement (due to enforced spatial proximity) of
behavior that occurs with unconnected POMs and chromo-
phores. Instead, we are interested in studying new optical and
and a range of organic transformations that can be used to post-
2
functionalize the resulting hybrid. These have allowed the
construction of POM-organic architectures of increasing
3
complexity, including polymers and POMs linked to catalytic
4
5,6
centers or light-harvesting chromophores. Such research is
motivated by the promise of emergent or synergic properties
resulting from the combination of the organic and inorganic
Received: March 22, 2017
©
XXXX American Chemical Society
A
DOI: 10.1021/acs.inorgchem.7b00708
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
photophysical behaviors that emerge when strong electronic
coupling occurs between POMs and appended organic groups.
Organoimido POM derivativestypically of the Lindqvist
using Snowy Range (SnRI) portable Raman spectrometers. Two
excitation sources were used: a diode laser with 532 nm excitation (50
mW of power) and a diode laser with 785 nm excitation (100 mW).
2
−
Acquisition times were 0.2 s at 532 nm and 1 s at 785 nm. The SnRI
family ([M O ] , where M = Mo, W)demonstrate this
6
19
−1
5
32 nm spectrometer has a wavenumber range from 200 cm to 3200
strong electronic coupling through new electronic transitions
−1
−1
cm , whereas the 785 nm spectrometer has a range from 200 cm to
8
and shifted POM reduction potentials. Their synthetic
−1
2
000 cm . The Raman measurements were normalized to a
−1
chemistry is well-developed, and many species can be accessed
cyclohexane standard, with respect to the peak at 801 cm . Spectra
were baseline-corrected using a multipoint linear fit and a level and
zero-leveling mode in Grams software (AI 7.0).
2c,9
through a DCC-mediated amine/POM coupling,
followed
by various organic post-functionalizations. However, the
photonic properties of these materials have so far been
Hyper-Rayleigh Scattering. General details of the hyper-Rayleigh
18
1
0
scattering (HRS) experiment have been discussed elsewhere, as have
been the experimental procedure and data analysis protocol used for
addressed only slightly, beyond calculations suggesting
second-order nonlinear optical coefficients (first hyperpolariz-
1
9
ability, β > 10−27 esu in some cases) that compete with high-
the fs measurements in this study. Measurements were carried out
−5
using dilute (ca. 10 M) filtered (Millipore, 0.45 μm) acetonitrile
11
performance organic materials, but surprisingly seem to be
based on POM-to-nitro CT. Given the relatively high energies
of organoimido-POM electronic transitions, and the challenge
of obtaining high-activity molecular nonlinear optical (NLO)
materials with adequate transparency (reabsorption of visible
light can cause lowered efficiency, overheating and instability),
this has encouraged us to explore the arylimido Lindqvist anion
as a platform for new high-activity, high-transparency second-
order NLO materials. Such materials are of potential value to
technologies in telecommunications and optical/electro-optical
solutions, so that self-absorption of the SHG signal was negligible,
verified by the linear relation between signal and concentration.
30
Crystal violet was used as an external reference (β
esu in methanol,
= 338 × 10−
xxx,800
1
9a
corrected for local field factors at optical
frequencies). All 800 nm measurements were performed using the
3+
8
00 nm fundamental of a regenerative mode-locked Ti :sapphire laser
(Spectra Physics, model Tsunami, 100 fs pulses, 2 W, 80 MHz). An
absence of demodulation for most compounds, i.e., constant values of
β versus frequency, showed that no multiphoton fluorescence
contributions to the HRS signals were present at 400 nm. This may
indicate (i) a lack of fluorescence, (ii) spectral filtering out of
fluorescence, or (iii) the fluorescence lifetime is too short for its
demodulation to be observed within the bandwidth of the instrument.
Where present, fluorescence has been corrected by demodulation
fitting to provide a fluorescence free value of β. The reported β values
are the averages taken from measurements at different amplitude
modulation frequencies. Measurements at 1064 nm were carried out in
acetonitrile, using a Spectra-Physics InSight DS+ laser (average power
of 1 W, sub-100 fs pulses, 80 MHz). In this setup, the collection optics
are coupled to a spectrograph (Bruker, Model 500is/sm) and an
EMCCD camera (Andor Solis model iXon Ultra 897). Correction for
multiphoton-induced fluorescence was done by subtracting the broad
MPF background signal from the narrow HRS peak (fwhm = ±9 nm).
The higher accuracy of this setup enables us to use the solvent as an
1
2
computing, but current approaches generally rely on
multidimensional chromophores that can be synthetically
complex, and/or involve use of precious metals (e.g., Ru) to
13
provide an octupolar core.
We recently found that a small family of arylimido
polyoxometalate derivatives have experimentally determined
(
by hyper-Rayleigh Scattering, HRS) static, resonance
−30
14
corrected β values of up to 133 × 10 esu. This compares
to 25 × 10
0
−30
+
esu for technologically valuable DAS under
nonresonant conditions, and in some cases the POM-based
chromophores break through empirical NLO performance
15
limits for planar organics, as defined by electron-number
adjusted β versus the absorption wavelength λmax. Therefore,
organoimido POMs are among a small group of dipolar
chromophores, comprising “tictoid” compounds with twisted π-
−
30
internal reference (acetonitrile, β
= 0.258 × 10
esu; βzzz,1064
=
HRS,1064
−30
20
0
.623 × 10 esu).
Stark Spectroscopy. Stark spectra were collected in butyronitrile
1
6
17
glasses at 77 K (estimated local field correction fint = 1.33). Apparatus
bridges and other systems with highly unusual donor sets,
21
and data collection procedure were as previously reported, but with a
that have been seen to exceed these limits, but are the only to
Xe arc lamp as the light source in place of a tungsten filament bulb.
Each spectrum was measured at least twice, and spectra were modeled
with a sum of two or three Gaussian curves that reproduce the band of
interest. The second, first, and zeroth derivatives of the Gaussian
do so with conventional planar π-bridges and donor groups
−
(
e.g, −NR , −OR, O ). As such, they may offer a low-cost,
2
convenient route to high-performance, high-transparency NLO
materials. Herein, we expand the series of organoimido-POM
chromophores, establishing structure−activity relationships,
and use Resonance Raman and Stark spectroscopies, and
density functional theory (DFT) calculations to elucidate the
nature of the charge-transfer transitions responsible for their
NLO properties. These show that the derivatives lack dipolar
character (and are hence NLO inactive) in the absence of
resonance electron donors. However, with suitable resonance
donors and short π-bridges, the imido-POM acceptors enhance
NLO activity versus organic analogues by extending charge
transfer onto the POM, increasing dipole moment changes.
22
curves were then used to fit the Stark spectra with Liptay’s equation.
The dipole moment change (Δμ = μ − μ , where μ and μ are the
12
e
g
e
g
respective excited- and ground-state dipole moments) was then
calculated from the coefficient of the second derivative component.
This assumes that the two-state model is applicable to the systems
investigated here, which have relatively high transition energies, and
may be complicated from mixing with other states close in energy.
Thus, values obtained for the compounds with the highest energy
absorption bands in this series are subject to additional uncertainty.
The following equations and definitions are integral to the subsequent
discussion:
A two-state analysis of the ICT transitions gives
Δμab2 = Δμ122 + 4μ122
(1)
EXPERIMENTAL SECTION
■
where Δμ is the dipole moment change between the diabatic states,
ab
Synthesis, Characterization, and DFT Calculations. Details
pertaining to synthetic chemistry and standard chemical and physical
characterization (NMR spectroscopy, X-ray crystallography, etc.), and
DFT calculations can be found in the Supporting Information (SI).
Raman Spectroscopy. The POM derivatives were prepared as 5
mM solutions in acetonitrile. Samples were analyzed in glass cuvettes
using 600 μL of the solution. Raman measurements were performed
Δμ is the observed (adiabatic) dipole moment change, and μ is the
12
12
2
1
transition dipole. The value of μ12 can be determined from the
oscillator strength fos of the transition, using the relation
1
/2
⎛
⎜
⎝
fos
⎞
|μ | =
⎟
⎠
1
2
1.08 × 10−5E
max
(2)
B
DOI: 10.1021/acs.inorgchem.7b00708
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
where Emax is the energy of the ICT maximum (in wavenumbers) and
Structures have been elucidated by NMR and electrospray
mass spectrometry, supported by elemental analysis and, in
most cases, X-ray crystal structures. The organosilyl Keggin
anion (11) provides a counterpoint to the imido-Lindqvist
μ12 is given in eÅ. The latter is converted to Debye units upon
2
multiplying by 4.803. The degree of delocalization cb and electronic
coupling matrix element Hab for the diabatic states are given by
species, because it has a strong −NH donor but much weaker
Emax(μ12)
Δμab
2
|
H | =
ab
electronic communication across the C−Si−O−W bridge,
while compounds 12 and 13 provide the closest possible purely
organic analogues to the POM derivatives. The synthetic
approach to the organo-imido compounds was centered on the
N,N′-dicyclohexylcarbodiimide (DCC) mediated coupling of
anilines with [NBu ] [Mo O ], to produce arylimido Lindqvist
(3)
(4)
⎡
1/2⎤
2
⎛
⎝
⎞
cb2 = 1 1 − ⎜
Δμ
12
⎢
⎥
⎟
⎜
2
2 ⎟
⎢
⎥
2
Δμ12 + 4μ12
⎠
⎣
⎦
4
2
6
19
8
,9,14
If the hyperpolarizability tensor β has only nonzero elements along
the ICT direction, then this quantity is given by
derivatives.
Notably, we found that (i) anhydrous
0
dimethylsulfoxide (DMSO) was a better solvent for this
chemistry than the commonly used acetonitrile, and (ii) careful
control of temperature and a slight excess of hexamolybdate
helped to prevent the formation of undesired (and difficult to
separate) bis-imido products. Short chromophores 1−5 were
thus accessed by reacting the parent amine with hexamolybdate.
Of the extended systems, 6 and 8 were obtained by direct
reaction of hexamolybdate with the parent ligand, while 7, 9,
and 10 employed a Sonogashira coupling between 1 and the
appropriate alkyne.
2
β = 3
Δμ (μ )
1
2
12
2
0
(
Emax
)
(5)
The Gaussian fitting procedure used for the Stark spectra is
dependent on a variety of parameters as well as the choice of baseline.
Thus, while the precision of calculated β values is ca. 20%, given a
consistent fitting approach, the actual error of the fitted parameters
versus their true values is difficult to estimate, and likely larger. The
experimental errors for μ12 and Δμ are estimated to be ±20% when
the second derivative component dominates the fit, and significantly
larger when it does not. Errors for Δμ , H , and c are estimated as
30%, ±30%, and ±50%, respectively.
0
12
1
NMR Spectroscopy. H NMR spectroscopy is an effective
2
ab
ab
b
tool for primary characterization of (diamagnetic) arylimido-
hexamolybdates. Compared to the parent anilines, the chemical
shifts for the protons ortho- to the imido bonds are shifted
downfield by up to 1 ppm. This indicates the electron-
withdrawing effect that the POM cluster has on the attached
aryl group, and is illustrated for compound 4 in Figure 1.
±
RESULTS AND DISCUSSION
Chromophore Design and Synthesis. To investigate the
effects of donor/acceptor substituents and π-conjugation on the
NLO properties of arylimido-Lindqvist species, we synthesized
the family of anionic chromophores 1−10 (see Chart 1) as
tetrabutylammonium salts.
■
Chart 1. Polyoxometalate-Based Chromophores and Nitro
Analogues Investigated in This Study
1
Figure 1. H NMR spectrum of the aromatic region of 4, and its
amino precursor showing the downfield shift of the ortho- aromatic
resonances upon formation of the imido-Lindqvist group.
Chemical shifts for the protons of the 10 imido-Lindqvist
derivatives are presented in Table 1. Generally, and logically,
the chemical shifts of the protons H ortho to the imido group
a
shift upfield as the donor strength of the D/A substituent
increasesdecreasing from 7.49 ppm to 7.15 ppm and from
7
.62 ppm to 7.46 ppm in the short and long pyrrole/NMe
2
pairs 4/5 and 8/9. Other resonances mostly follow this pattern,
most notably H (ortho to the D/A group), although not
d
completely consistentlyperhaps due to variation in the
strength of conjugation and/or ring current effects between
the different species. In −NO derivatives 3 and 7, the highest δ
2
(
ca. 8.2 ppm) is assigned to H , consistent with the high δ
d
observed in organic nitro compounds, yet H shifts upfield,
a
compared to all but 5 (short −NMe ), suggesting that the
2
strong −NO acceptor may pull electron density from the
2
imido-Lindqvist unit to result in increased shielding of H . This
a
10
is consistent with DFT calculations (including our own, vide
infra) suggesting that, in nitro-substituted organoimido-
hexamolybdates, the overall direction of charge transfer is
C
DOI: 10.1021/acs.inorgchem.7b00708
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
1
a
Table 1. Selected H NMR Chemical Shifts for 1−10
Chemical Shift, δ (ppm)
Donor/Acceptor
Imido Ring
Ring
NMe , pyrrole, or
2
a
b
c
d
NPh2
b
b
b
b
b
1
2
3
4
5
6
7
8
9
0
7.75
7.49
7.34
7.49
7.15
7.78
7.61
7.62
7.46
7.46
6.99
7.19
8.22
7.32
6.63
7.55
7.26
7.52
7.20
7.20
7.20
3.07
6.31
6.33
7.31
7.75
7.55
7.26
7.37
7.23
8.24
7.25
6.63
6.72
c
7.23
2.98
1
a
Recorded at 300 or 500 MHz in CD CN; all δ are given in ppm with
3
b
c
respect to TMS. n = 0. Overlapped with pyrrole signals. Where
possible, assignments have been confirmed using two-dimensional
(
2D) NMR techniques.
Figure 2. Crystal structures of new chromophoric anions obtained in
this study, labeled with the appropriate compound number. [Legend:
C, gray; N, blue; O, red; Mo, green; and I, purple. H atoms are white
circles with arbitrary radii.] Disordered parts have been omitted for the
sake of clarity. Thermal ellipsoids are at the 30% probability level.
from the POM/imido group and to the nitro group, although it
does not necessarily indicate that the POM is an effective
donor, as proposed elsewhere.
10c,d
X-ray Crystal Structures. We have obtained high-quality
X-ray crystal structures for all of the organoimido compounds
in the study, with the exception of 3, which failed to produce
diffraction-quality crystals. Our structures of 4, 8, and 9 were
23
Table 2. Selected Bond Lengths in Compounds 1−5
14
previously published, those of the other anions, including 5,
are displayed in Figure 2. Full details are presented in the SI.
The Mo−O bond lengths of the imido-hexamolybdate
clusters show a similar pattern of lengthening and shortening
a
compound
r1
r2
r3
r4
2
−
8
1
2
2.089(4)
1.426(8)
1.411(5)
1.375(5)
1.375(6)
1.390(8)
1.397(5)
1.421(6)
1.388(6)
1.385(8)
1.382(5)
1.355(6)
1.383(6)
1.394(7)
1.396(6)
1.407(5)
versus [Mo O ] to that previously described (see the SI),
6
19
while the bond lengths of the organic ligands show trends
consistent with the donor−acceptor nature of the derivatives.
Moreover, while these structural phenomena do not necessarily
occur in solution (where temperature and local electric field are
different to the crystal), they are consistent with the solution
photophysical and electrochemical properties. In the “short”
chromophores 1−5, a pronounced contraction of the phenyl
b
4
5
a
Average of two chemically identical, crystallographically independent
C−C bonds. Average of two crystallographically independent anions.
b
functionalized compounds 6−8 show a coplanar arrangement
of the two phenyl rings; 45° and 86° twists are observed with
the strong NH and NMe donors. Since the barriers to
C−C bonds between the ortho- and meta- carbons (Table 2, r )
3
compared to the other phenyl C−C bonds (r /r ), is observed
2
4
2
2
on moving from −I and alkynyl derivatives 1 and 2, to the
rotation in unhindered phenylacetylenes are very low (ca. 1 kcal
−1
24
strong donor −NMe in 5. This is consistent with a significant
mol ), both twisted and coplanar arrangements are often
observed in their crystal structures. Even so, the observed twists
fit with NMR (vide supra), electrochemical, and Stark
spectroscopic measurements (vide infra), suggesting that
conjugation across the phenylacetylene unit is weaker with
the stronger donors. Such weakened conjugation would further
reduce the barrier to rotation and increase the probability of
isolating a twisted geometry in the crystal.
2
contribution from a quinoidal resonance form when a strong
donor group is present. Further evidence of this is seen in a
shortened donor-N to phenyl bond length in 5, compared to
the weak pyrrole donor system 4. However, there is no
significant change in the C−N and N−Mo imido bond
distances.
This shortening of the ortho-to-meta C−C bonds (r ) is far
3
less pronounced in the extended analogue (10), only
statistically significant on the donor ring, and absent in the
other extended systems. This implies, logically, that commu-
nication between organic donor and POM is weaker with the
longer bridge. Moreover, while the acceptor/weak donor
Electronic Spectroscopy and Electrochemistry. Con-
2
−
nection of arylimido groups to [Mo O ] produces a new
6
19
transition red-shifted from the O → Mo bands, and also not
present in the parent amines. Previously, we tentatively
assigned this low energy (LE) band as ligand-to-POM charge
D
DOI: 10.1021/acs.inorgchem.7b00708
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
1
4
transfer (LPCT). This was consistent with our data, and
previous descriptions as a ligand-to-metal CT transition red-
second harmonic. In addition, electron-number adjusted β
3/2
values, β /N
(where N is the number of π-conjugated
0
2c,8,9
shifted by conjugation of the imido and aryl π-systems.
electrons) have been calculated to facilitate comparison of the
intrinsic performance of the POM chromophores with organic
materials.
However, the possibility that other electronic transitions
contribute to the bands, for example, π−π* transitions lowered
in energy by interaction with the POM, or imido-to-aryl CT
transitions (particularly with acceptor groups) cannot be
discounted. Indeed, the LE band red-shifts upon replacement of
The data indicate that without a resonance electron donor,
POM-based chromophores are NLO-inactive, as scattered
second harmonic (SH) light failed to exceed the solvent
background at both wavelengths for iodo derivatives 1 and 6,
ethynyl derivative 2, and nitro derivatives 3 and 7. This
dramatically contrasts with computational studies predicting
significant β values for organoimido systems with no resonance
10
−
I or −CCH with the strong resonance acceptor −NO ,
2
showing that CT is not exclusively to the POM and mandating
use of the more general assignment “intra-hybrid charge
transfer” (IHCT). Other data (Stark spectroscopy, DFT,
Raman spectroscopy) also suggest multiple origins for the LE
absorptions (vide infra).
10
donor, and indeed finding high activity for nitro-function-
alized systemsthis activity being ascribed to POM (or imido)
to −NO charge transfer. Our HRS results, and resonance
2
The IHCT bands certainly show LPCT character, demon-
strated by the red-shift in λmax with increased donor strength
i.e., from 4 to 5 or from 8 to 10; see Figure 3 and Table 3).
Raman and Stark spectroscopy (vide infra), suggest that such
any such charge transfer must be very weak, because the
electron-withdrawing effect of the POM’s delocalized, vacant
Mo 4d orbitals is more significant than the weak electron-
donating ability of its oxo-groups. Unsurprisingly, the amino-
aryl silyl Keggin derivative (11), which shows little evidence of
donor−acceptor communication via electronic spectroscopy or
electrochemistry, is also NLO-inactive.
(
However, all five imido-Lindqvist species with resonance
electron donors (short chromophores 4 and 5, and extended
systems 8−10) show significant (large) dynamic β values at
zzz
both measured wavelengths, with the highest values of 813 ×
−30
−30
1
0
esu (800 nm) and 440 × 10
esu (1064 nm) being
obtained for 10. βzzz values are consistently lower at 1064 nm,
most likely because any resonance enhancement is effectively
eliminated, but the overall trend in the dynamic first
hyperpolarizabilities is similar at the two wavelengths, with
extended, dimethylamino donor 10 being the most active, and
short, pyrrole donor 4 being the least active. The small
differences (relative performance of 5 and 8) are likely due to
uncertainties in the 400 nm data, relating to resonance
contributions, so the 1064 nm data should be considered
more reliable. Notably, the trend in the 1064 nm data is
consistent with established structure activity relationships for
organic systems, whereby increased donor strength and
conjugation length lead to higher β.
Figure 3. Electronic absorption spectra of 4 (blue), 5 (dark blue), 8
red), and 10 (deep red) in acetonitrile at 298 K.
(
Red-shift also occurs with extended conjugation, except for the
NMe derivatives 5 and 10: remarkably, 5 has λ 3 nm to
−
2
max
the red of the 421 nm obtained for extended system 10, and
also shows a slight change in band shape. This is consistent
with X-ray crystallographic evidence for weakened conjugation
in extended chromophores with strong donors (vide supra),
and particularly strong donor−POM communication in 5.
Electrochemistry also indicates that donor−POM and accept-
or−POM communication is strong in the short chromophores,
but weakens in extended systems. Short chromophores 1−5
show a strong dependence in reduction potential upon the
group para- to the imido-N, following the expected order of
donor strength NO < I < CCH < Py < NMe across a range of
Static first hyperpolarizabilities, β , from the 1064 nm data
0
facilitate comparison with purely organic chromophores, and
reveal that the POM derivatives generally have high activities
given their relatively small π-systems, simple planar structures
(alkyne bridges are generally considered less efficient than
25
alkenes in 2nd order NLO ), moderate donor strengths and
high transition energies. All five active chromophores have β
values comfortably exceeding that of the technologically
2
2
0
>
140 mV. However, the reduction potentials of the five
+
−30
extended systems cover a range of only 27 mV, and show no
valuable DAS cation (25 × 10 esu, λmax = 470 nm), and
several more active stilbazolium chromophores, but with
significantly less-red-shifted absorption profiles. The most
pertinent comparison, however, is with nitro-acceptor-based
dependence on the organic moiety’s donor or acceptor
1
characteristics. Thus, H NMR, X-ray crystal structures,
electron spectroscopy, and electrochemistry all suggest that π-
conjugation is weakened in extended systems with strong
donors.
12 and 13 since, for both imido-POM and −NO the acceptor
2
is based on an N-substituent with a formal positive charge, that
Hyper-Rayleigh Scattering. Table 4 shows βzzz values
can delocalize charge away from the π-bridge. This shows a
obtained at 800 nm, and βzzz and β values obtained at 1064 nm
large enhancement in β for the POM when a short phenyl
0
0
for 1−13. The 1064 nm source was used to obtain the static,
spacer is used −5 is nearly three times more active than its nitro
analogue 12, but with the longer phenylacetlyene spacer the
performance of POM 10 is lower than (but within experimental
resonance-corrected β values, because λ
for all of the
0
max
chromophores is distant (in all cases, >100 nm shorter than the
32 nm second harmonic), and residual absorption is minimal
5
error of) its −NO analogue 13. Notably, 4, with a weak pyrrole
2
at this wavelength. At 800 nm, β0 would be severely
underestimated, because of the close proximity of λmax to the
donor, also has higher β values than 12, and both short POM
0
3/2
chromophores have very high β /N values (3.78 for 4, 5.90
0
E
DOI: 10.1021/acs.inorgchem.7b00708
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 3. Electronic Absorption Data and Reduction Potentials (E1/2) for 1−11 and [NBu ] [Mo O ] (TBAMo )
4 2
6
19
6
a
ε (× 10 M cm−1
3
−1
b
λmax (nm)
)
Emax (eV)
assignment
E1/2 (V)
ΔE, vs Ag/AgCl (mV)
TBAMo6
223
24.9
13.4
7.1
5.56
4.75
3.84
O → Mo
O → Mo
O → Mo
−0.315
−0.315
−0.315
69
69
69
261
23
3
1
242
37.0
32.1
27.0
5.12
4.59
3.49
O → Mo and π → π*
O → Mo and π → π*
IHCT
−0.476
−0.476
−0.476
63
63
63
2
3
70
55
2
3
264 (36.2)
4.69
3.46
O → Mo and π → π*
−0.499
−0.499
61
61
358 (27.0)
IHCT
216
38.3
26.3
20.3
30.2
5.74
4.88
4.32
3.35
O → Mo and π → π*
O→Mo and π→π*
O → Mo and π → π*
IHCT
−0.434
−0.434
−0.434
−0.434
71
71
71
71
254
287
371
4
5
6
7
223
39.6
28.2
32.3
5.56
4.41
3.34
O → Mo and π → π*
O → Mo and π → π*
IHCT
−0.500
−0.500
−0.500
63
63
63
2
3
81
71
221
37.5
30.6
32.0
5.61
4.80
2.92
O → Mo and π → π*
O → Mo and π → π*
IHCT
−0.575
−0.575
−0.575
71
71
71
2
4
58
24
236
31.4
31.8
40.0
5.26
4.15
3.25
O → Mo and π → π*
O → Mo and π → π*
IHCT
−0.504
−0.504
−0.504
69
69
69
2
3
99
81
241
36.0
29.9
49.6
5.14
4.60
3.18
O → Mo and π → π*
O → Mo and π → π*
IHCT
−0.486
−0.486
−0.486
64
64
64
2
3
69
89
8
9
290
86
40.1
43.8
4.28
3.21
O → Mo and π → π*
−0.496
−0.496
61
61
3
IHCT
281
06
39.2
35.8
4.41
3.05
O → Mo and π → π*
−0.476
−0.476
63
63
4
IHCT
1
1
0
1
248
36.2
44.5
41.2
5.02
4.24
2.94
O → Mo and π → π*
O → Mo and π → π*
IHCT
−0.498
−0.498
−0.498
61
61
61
2
4
92
21
263
64.0
4.72
4.72
O → W and π → π*
O → W and π → π*
−0.331
−0.850
69
71
2
63
64.0
a
−5
b
−3
Concentrations ca. 10 M in MeCN. Solutions ca. 10 M in analyte and 0.1 M in [NBu ][BF ] at a glassy carbon working electrode with a scan
rate of 100 mV s . Ferrocene internal reference E = 0.46 V, ΔE = 80 mV.
4
4
−1
1/2
p
for 5) for their modest λmax. This places them comfortably
beyond the apparent limit described by Kuzyk et al. for planar
organic chromophores (Figure 4), with reference compound 12
dipolar chromophores exceed the Figure 4 apparent limit as
comfortably as 4 and 5. These typically feature unusual, highly
twisted “TICT” π-systems that favor charge separation
(dramatically increasing the contribution of Δμ12 in the two-
state model). This, and observation of the highest intrinsic β in
the chromophores with the smallest π-systems (opposite to the
normal trend in purely organic systems such as 12 and 13),
indicate that the POM must be playing a significant role in the
NLO activity of the phenyl bridged “short” systems. These
results raise questions about the nature of the IHCT bands in
the HRS active/inactive compounds, and the reason for the
unusually high activity of the short POM derivatives versus
15
(
×
whose β value agrees well with the previously determined 29
10 esu ) close to the limit. Since the β obtained for 13 is
0
0
−30
26
significantly greater than that previously measured by EFISH
46 × 10− esu, the benchmark in our previous work ), under
30
14
(
our conditions this compound together with POMs 9 and 10
also appears to (just) exceed the apparent limit. While this
means comparisons with other studies must be made with
caution, and it must be emphasized that POM electrons are not
3
/2
included in N; the β /N values make it clear that the POM
0
has a large effect on short π-systems, and only a few other
F
DOI: 10.1021/acs.inorgchem.7b00708
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 4. HRS Data at 800 and 1064 nm for 1−13
βzzz,800 (× 10− esu)
30
a
βzzz,1064 (× 10−30 esu)a
β0,1064 (× 10−30 esu)b
β0/N3/2
compound
λmax (nm)
1
2
3
4
5
6
7
8
9
355
358
371
371
424
381
389
386
406
421
262
394
406
inactive
inactive
inactive
inactive
inactive
inactive
292 ± 34
302 ± 58
inactive
123 ± 10
283 ± 7
inactive
56 ± 5
87 ± 2
3.78
5.90
inactive
inactive
557 ± 56
716 ± 63
813 ± 144
inactive
143 ± 10
372 ± 22
440 ± 55
inactive
59 ± 5
133 ± 8
139 ± 17
1.12
2.53
2.65
1
1
1
1
0
1
2
3
not measured
not measured
84 ± 13
452 ± 68
33 ± 5
161 ± 24
2.17
3.08
a
βzzz assuming a single dominant tensor component, measured using 800 and 1064 nm fundamental laser beams. The quoted units (esu) can be
3
3
−2
20
b
22
converted into SI units (C m J ) by dividing by a factor of 2.693 × 10 . Nonresonant, static β estimated from β using the two-state model.
zzz
Figure 4. Electron-number-adjusted β values (β /N3/2) of NLO-active
0
organoimido-POMs, and a range of planar, dipolar organic
chromophores from the literature, plotted against their wavelengths
of maximum absorption (λmax).
comparable organic materials, that we address below in a
combined spectroscopic and computational approach.
Raman Spectroscopy. In Resonance Raman spectroscopy,
Raman modes of parts of the molecule or assembly associated
27
with an excited transition are enhanced in intensity. By using
one wavelength (785 nm) far from the absorption of any of the
POM derivatives, and one (532 nm) coinciding with the tail of
the absorption, it is therefore possible to test for involvement
(
but not the role) of the POM. We performed these
measurements on a subset of the POM derivatives: 1, 3, 5−
7
, and 10, plus [NBu ] [Mo O ] and nitro analogues 12 and
4 2 6 19
1
3.
The results (Figure 5, as well as Figures S7−S9 in the SI)
show that, in all of the organoimido-POMs, vibrational modes
at ca. 990 cm− are enhanced with excitation at 532 nm, vs 785
nm (where they are almost undetectable). No such effect is
1
−1
observed in [NBu ] [Mo O ]. The enhanced 990 cm band
Figure 5. Raman spectra of [NBu ] [Mo O ] (top), 1 (middle), and
10 (bottom) excited at 532 and 785 nm. Concentration = 5 mM.
4
2
6
19
4
2
6
19
is absent from POM-free 12 and 13, and at similar energy to
2−
the symmetric a MoO mode of [Mo O ] (980−986
1
g
6
19
−
1
28
band with 532 cm−1 excitation implies involvement of the POM
and/or imido group in the IHCT transition, in all of the
investigated imido-POMs.
cm ), the IR-active MoNAr stretch of the imido
2
9
−1
derivatives (ca. 975 cm ), and the a mode of the reduced
1
3−
−1 28a
anion [Mo O ] (970 cm ). Thus, enhancement of this
6
19
G
DOI: 10.1021/acs.inorgchem.7b00708
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 5. ICT Absorption and Stark Spectroscopic Data for 1−10, and Organic Analogues 12 and 13
h
Amp
Fwhh
μ12
(D)
Δμ
Δμ
r12
rab
H
β
∑β
12
(D)
ab
(D)
ab
0
0
a
a
b
c
d
e
(Å) (Å) (× 10 cm−1)
f
g
3
cb2
i
−30
j
−30
k
λmax (nm)
357 (371)
340)
Emax (eV)
(norm) (eV)
fos
(× 10 esu) (× 10 esu)
l
m
1
2
3
3.47 (3.30)
(3.55)
0.37
0.63
0.36
0.52
0.18
0.50
3.8
6.1
0
0
0
0
m
(
l
m
361 (381)
353)
3.43 (3.25)
(3.50)
0.34
0.66
0.34
0.50
0.15
0.46
3.5
5.9
0
0
0
m
(
0
m
m
m
371 (410)
3.34 (3.03)
(3.21)
0.05
0.40
0.56
0.18
0.37
0.54
0.01
0.25
0.61
1.1
4.7
6.7
6.7
4.5
5.4
7.0
10.4
14.5
1.4
0.9
1.1
1.5
2.2
3.0
3.8
11.7
13.0
0.98
0.72
0.68
1
11
23
35
35
35
(
(
387)
356)
(3.49)
4
5
6
7
8
9
0
2
3
381 (409)
3.25 (3.03)
(3.24)
0.11
0.67
0.23
0.16
0.45
0.33
0.03
0.59
0.16
1.6
6.9
3.4
6.3
7.7
3.9
7.0
15.9
7.9
1.3
1.6
0.8
1.5
3.3
1.6
5.6
11.4
12.5
0.95
0.74
0.75
2
41
4
47
47
47
(
(
383)
347)
(3.58)
438 (448)
2.83 (2.77)
(2.86)
0.20
0.48
0.31
0.17
0.31
0.67
0.06
0.30
0.45
2.5
5.2
6.2
7.9
12.2
6.7
9.3
16.1
14.1
1.7
2.5
1.4
1.9
3.3
2.9
6.0
7.5
0.93
0.88
0.74
7
48
31
86
86
86
(
(
437)
396)
(3.12)
11.0
m
m
m
390 (422)
3.18 (2.94)
(3.12)
0.14
0.50
0.36
0.14
0.37
0.47
0.06
0.58
0.58
2.3
7.0
6.7
4.2
5.2
4.9
6.3
14.9
14.3
0.9
1.1
1.0
1.3
3.1
3.0
8.8
11.8
12.9
0.84
0.68
0.67
3
31
22
56
56
56
(
(
398)
365)
(3.40)
m
m
m
396 (426)
3.13 (2.91)
(3.08)
0.18
0.43
0.39
0.14
0.33
0.52
0.09
0.51
0.79
2.8
6.6
7.9
4.8
6.3
7.4
14.7
19.0
1.0
1.3
2.2
1.5
3.1
4.0
9.0
11.2
11.2
0.83
0.71
0.78
5
34
70
110
110
110
(
(
403)
373)
(3.32)
10.4
400 (432)
3.10 (2.87)
(3.05)
0.20
0.42
0.38
0.15
0.33
0.53
0.10
0.46
0.73
3.0
6.3
7.7
8.9
9.4
10.7
15.8
18.7
1.8
2.0
2.2
2.2
3.3
3.9
6.5
9.8
0.92
0.80
0.79
11
47
69
130
130
130
(
(
407)
378)
(3.28)
10.8
10.8
434 (468)
2.86 (2.65)
(2.81)
0.23
0.41
0.36
0.18
0.31
0.47
0.10
0.33
0.47
3.2
5.5
6.4
18.8
20.7
24.0
19.8
23.4
27.2
3.9
4.3
5.0
4.1
4.9
5.7
3.4
5.3
5.7
0.97
0.94
0.94
31
94
250
250
250
(
(
442)
413)
(3.01)
128
1
1
1
456 (482)
2.72 (2.57)
(2.72)
0.29
0.40
0.30
0.19
0.28
0.36
0.16
0.32
0.34
4.0
5.6
5.6
22.3
23.4
23.3
23.7
26.0
25.8
4.6
4.9
4.8
4.9
5.4
5.4
3.5
4.7
5.1
0.97
0.95
0.95
64
116
99
280
280
280
(
(
456)
427)
(2.90)
408 (422)
3.04 (2.94)
(3.07)
0.24
0.49
0.27
0.17
0.28
0.49
0.07
0.24
0.24
2.4
4.5
4.4
8.1
7.3
8.1
9.4
11.6
12.0
1.7
1.5
1.7
2.0
2.4
2.5
6.2
9.6
9.8
0.93
0.82
0.84
7
18
17
42
42
42
(
(
404)
377)
(3.29)
431 (461)
2.88 (2.69)
(2.86)
0.16
0.49
0.35
0.26
0.38
0.58
0.07
0.34
0.40
2.7
5.6
5.9
24.7
21.9
22.2
25.3
24.6
25.1
5.1
4.6
4.6
5.3
5.1
5.2
2.3
5.3
5.8
0.99
0.95
0.94
29
99
93
220
220
220
(
(
434)
401)
(3.09)
a
In butyronitrile at 77 K; observed absorption maxima with maxima for Gaussian fitting functions in brackets. Data in all subsequent columns relate
b
−9
2
to fitted curves. Obtained from (4.6 × 10 M cm ) ε × fw1/2, where εmax is the maximal molar extinction coefficient and fw is the full width at
max
1/2
−1
c
d
e
f
half height (Fwhh, given in cm ). Calculated using eq 2. Calculated from f Δμ using fint = 1.33. Calculated from eq 1. Delocalized electron-
int
12
g
h
i
transfer distance calculated from Δμ /e. Effective (localized) electron-transfer distance calculated from Δμ /e. Calculated from eq 3. Calculated
12
ab
j
k
l
from eq 4. Calculated from eq 5. Sum of the β vales from each individual Gaussian function, to 2 sf. Fitted using only contributions from first and
0
m
second derivatives; zeroeth derivative contribution set to zero. Inactive by HRS.
However, there are significant variations in the strength of
enhancement between the different organoimido compounds.
It is especially strong in 10 (Figure 5), most likely because this
compound has the most significant residual absorption at 532
nm. However, it is weak enough in the other two extended
compounds (6 and 7), by comparison with short compounds 1
and 3 whose λmax is further from 532 nm, to suggest that
involvement of the POM/imido in the transition is weaker (or
different) in extended systems that lack a resonance donor.
This is consistent with observations by other techniques. The
anomaly is 5, which, despite the second-largest 532 nm
absorption after 10, a short bridge, and clear evidence for very
strong aryl-POM interaction by all other techniques, shows a
−1
weak enhancement at 990 cm vs 1, 3, and 10. It may be that
the 990 cm− band is more associated with MoNAr than
MoO, and that the extended CT distance for 5 vs nitro-
1
H
DOI: 10.1021/acs.inorgchem.7b00708
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
analogue 12 (but not for 10 vs 13, vide infra) and its unusually
strong red-shift implies the acceptor for CT is more strongly
centered on the POM framework for 5 than for 10. As POMs
show only minimal structural changes upon reduction, this
could cause less vibrational excitation of MoNAr than an
acceptor spread more onto the aryl imido unit. This is
consistent with the DFT-calculated lowest energy transitions of
Δμ /Δμ and more significant Stark-derived β . These are
1
2
ab
0
consistent with the larger π-systems and a CT component away
from the POM and toward −NO . Nonzero Stark β values for
2
0
HRS inactive POMs may arise because Stark spectroscopy does
not take directional opposition between CT processes (e.g., aryl
→ POM, aryl → NO ) into account, and because the dipolar
2
character responsible for NLO activity is dampened by the
highly delocalized nature of both aryl and POM moieties.
5
and 10 in solution.
In 3 and 7, the −NO group provides an additional probe.
Comparing 7 with the organic −NO analogue 13, the much
2
2
Thus, it is significant that organic nitro compounds 12 and 13
lower Δμ (and β ) is consistent with the HRS inactivity of 7,
12
0
show very strong enhancement of the −NO band at ca. 855
and Resonance Raman measurements that suggest weakened
2
−1
cm upon moving from 785 nm to 532 nm, but this effect is
much weaker in 3 and 7 (other ArNO vibrations are too
CT to −NO with a connected POM.
2
30
Compared to the inactive systems, the five HRS-active POMs
(4, 5, 8−10), and the organic chromophores (12, 13), all show
more significant red-shifts of 10−35 nm in the PrCN glasses,
with the trend in magnitude of red-shift matching the trend in
β0. This resembles the behavior of some organic charge transfer
dyes, which have large thermochromic shifts associated with
large excited-state dipole moments, and much smaller
2
close to others observed in −NO -free systems to be
2
diagnostic). Since 7 has more residual absorption at 532 nm
than 12 (see Figure S10 in the SI), this suggests that the POM/
imido group suppresses CT to the NO group, and most likely
2
vice versa, leading to a weak dipole moment change and the
observed HRS inactivity.
21,22
32
Stark Spectroscopy. Stark spectroscopy
can measure
thermochromic shifts for less-polar systems. The fits (Figure
dipole moment changes and other parameterse.g., donor−
acceptor electronic coupling, delocalizationrelevant to CT
transitions and molecular NLO performance. Moreover, β0
calculated from Stark data, while yielding values that are often
significantly larger than those measured from real SH light (e.g.,
S12 in the SI) also show more dipolar character for the active
compounds, as second derivative contributions dominate at
least one (in most cases, all) of the three Gaussians associated
with the IHCT peak. This is reflected in Δμ /Δμ values that
1
2
ab
are generally higher than in HRS inactive systems. Stark derived
1
0,31
via HRS or EFISH),
generally agree in trend with
∑β (sum of all contributing Gaussians) values follow a very
0
nonresonant HRS and EFISH. Thus, Stark spectroscopy is a
valuable means to assess the relative second-order NLO
performance of a group of related chromophores in an
experiment that is not complicated by resonance enhancement
similar trend in magnitude to those from 1064 nm HRS, with
extended −NMe derivative 10 giving the highest value of 280
2
−
30
× 10 esu and short pyrrole 4 showing the lowest value, at 47
−
30
× 10
esu. The similarity of the trends between the two
13e
or reabsorption of SH light.
Therefore, we use Stark
techniques and the fact that the extended systems, which show
more (albeit still weak) absorption at the 532 nm HRS SH
wavelength and are hence more vulnerable to resonance effects,
perform more strongly by Stark, implies that the HRS-
spectroscopy as a complementary technique to understand
the IHCT transitions and confirm trends in β0.
Fitting the Stark spectra using derivatives of the observed
absorption spectra was generally unsatisfactory, as only the
compounds with the strongest donor groups yielded reasonable
fits. Thus, we fit the IHCT transitions using two or three
Gaussian peaks to fit the absorption spectra, and then used
derivatives of each of the Gaussians to fit the Stark spectra. In
some cases, large zeroth (and often first)-derivative compo-
nents were observed to contribute to the Stark spectrum. In
these cases, the fits with the zeroth-derivative component set to
zero were performed. The results are discussed below (see
Table 5).
determined β values are not enhanced by resonance. Trends
0
within the HRS-active species follow expectations for the
structure of the organic component: as donor strength and
conjugation length increase, Δμ /Δμ (averaged) increase,
1
2
ab
2
b
H (D/A electronic coupling) decreases and c moves further
ab
from the 0.5 associated with complete delocalization. Although
the differences between individual compounds are often within
Stark experimental errors, these findings are consistent with our
other experimental observations (vide supra).
Stark data also help explain the high HRS β of 4 and 5
0
First, the five HRS inactive imido-POMs (1−3, 6, and 7) all
versus organic analogue 12, and more similar behavior of 9 and
10 to that of 13. For extended system 10, Stark and HRS β0
values are effectively within experimental error of those of
nitro-analogue 13. Moreover, values of Δμ and Δμ differ by
10% or less, showing that, in 10, CT distances are not extended
by the POM. This fits with time-dependent density functional
theory (TD-DFT) calculated acceptor orbitals for 10 that are
distributed over the phenylimido ring as well as the POM (vide
infra), and the simple consideration that, proportionally, the
POM will have a smaller effect on CT distances in larger π-
show minimal (0 to 9 nm) red-shifts in λ in PrCN glasses
max
versus room-temperature electronic spectra in MeCN or PrCN.
Visual inspection of the fits (Figure S11 in the SI) shows that
most are dominated by first-derivative contributions relating to
polarizability. The phenyliodo and phenylethynyl derivatives 1
and 2 consequently have zero values of Δμ , β , etc., consistent
12
ab
1
2
0
with their lack of measurable HRS signal, and the qualitative
conclusion that their transitions are highly delocalized and, at
most, weakly dipolar in nature. This suggests that, in the
absence of resonance donors/acceptors or extended conjuga-
tion, the lower-energy electronic transitions of arylimido-
Lindqvist anions are best seen as π−π* transitions perturbed
and lowered in energy by participation from POM-oxo donor,
and Mo d-orbital acceptor levels. This is broadly consistent
with TD-DFT calculations (vide infra). Significant second
derivative (CT) components emerge with extension of
systems. In the short POM chromophore 5, however, ∑β is
0
dominated by the contribution of the middle Gaussian, which
has Δμ = 12.2 D, and a localized dipole moment change Δμ
1
2
ab
= 16.2 D (CT distance r = 3.3 Å). These Δμ and r values are
ab
ca. 40%−50% larger than for any of the Gaussians of the direct
organic analogue 12, which is approximately twice the
estimated experimental error. Combined with other exper-
imental observations (for example, the pronounced red shift)
and TD-DFT calculations, these increased Δμ and r values
conjugation (6) and/or introduction of the −NO resonance
2
acceptor group (3, 7), leading to small but appreciable values of
I
DOI: 10.1021/acs.inorgchem.7b00708
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
imply that extension of charge transfer onto the POM increases
based HOMO to the POM, consistent with the much higher
charge separation in 4 and 5 vs −NO analogues. Thus, the
Stark Δμ . However, for short −NO derivative 3 and all of the
2
12
2
unusually high intrinsic β of 4 and 5 results from this increased
extended systems (see Tables S6 and S9−S13, and Figures S16
and S19−S23 in the SI), TD-DFT finds that the strongest low-
energy transition is from the phenylimido ring, toward the D/A
ring, regardless of substituent. This agrees with previous
charge separation, that occurs without the fall in electronic
coupling associated with extended organic π-systems (H and
ab
2
cb for 5 and 12 are similar), or adding to the number of
3
/2
10
electrons in calculation of β /N
.
calculations, but contradicts experiment. However, higher-
0
DFT Calculations. TD-DFT calculations were performed
energy CT processes, both toward and away from the POM,
are also present, whose contribution varies with functionaliza-
tion of the organic group.
To test whether these discrepancies originate from
comparing gas-phase calculations with solution/glass-based
using the SAOP XC potential and TZ2P basis set, through the
3
3
Amsterdam Density Functional. These methods are a
development of those previously published for organoimido-
10
POM NLO properties (LB94/TZP), and, in principle, an
improvement due to the newer and more accurate XC
measurements, we also calculated β in an acetonitrile solvent
0
3
4
potential and larger basis set. Full details are given in the
force field. To our knowledge, such calculations have not been
performed on organoimido-POMs before. They are unopti-
mized, but reproduce the experimental trend in β for the HRS
active compounds (Table S14 in the SI) while eliminating
SI, including calculated electronic transitions and orbitals
(
Tables S4−S20, Figures S14−S27), and a comparison with
prior calculations on the most similar compounds in the
literature (see Chart S1 in the SI).
negative βzzz values for all but the −NO derivatives (which no
2
Our gas-phase calculations of β yield results similar to those
longer have the largest magnitude). Moreover, calculated
orbital-to-orbital transitions obtained on a subset (1, 3, 5, 10;
see Tables S16−S19, and Figures S24 to S27 in the SI) show
that CT to the POM becomes stronger and lower energy upon
solvation, producing a picture that is, in most cases, qualitatively
consistent with experiment. The effect is shown for 10 in
Figure 6, and can be visualized in a simple way for all
0
1
0c−e
in the literature,
and thus show trends at variance to those
seen by experiment. The largest values are obtained for
extended −NO derivative 7 (inactive by HRS) with the
2
negative sign for β
indicating that this arises from a net CT
zzz,0
away from the POM, and toward the nitro group as described
1
0c−e
elsewhere.
For other extended compounds, β tends to
0
decrease as donor strength increasesopposing the exper-
imental trend. (See Table 6.) Calculated lowest transition
Table 6. DFT-Calculated Lowest-Energy Transitions and β0
Values for 1−10 in the Gas Phase
a
b
c
d
ΔEGE
fos
S
βzzz,0
βvec,0
e
f
f
1
2
3
4
5
6
7
8
9
3.12
3.04
2.75
3.10
2.92
2.46
1.63
2.38
2.64
2.65
0.3125
0.3927
0.8070
0.3918
0.1956
1.5141
0.9374
1.4441
0.8468
1.8002
A
19.3
−7.8
−52.7
10.9
10.1
7.1
A
A
A
A
A
A
A
A
A
e
32.4
5.0
Figure 6. Strongest low energy orbital-to-orbital transitions in (a)
compound 10 in the gas phase and in solution and (b) compound 5 in
solution.
37.1
19.8
215.7
487.3
268
−355.7
−805.3
−442.2
−115.0
−143.9
e
e
f
f
compounds through the calculated ground state dipole
moments (Table S20). Notably, the calculated lowest energy
transitions of 5 and 10 also reveal that the acceptor orbital
spreads over the phenylimido ring of 10, but concentrates more
on the POM in 5, consistent with the increased Δμ vs nitro
70.9
90.4
10
a
b
c
Lowest energy transition. Symmetry. Dominant βzzz tensor
1
2
component along the molecular axis. Negative values indicate that
NLO effects originate from CT away from the POM. Orientationally
averaged β value. Weighted average of two or more closely spaced,
similar transitions. Sum of two or more closely spaced, similar
d
analogues observed for 5 but not 10 (Figure 6).
e
Even so, computed solution β values for the active systems
0
0
f
are approximately an order of magnitude larger than experi-
ment, and significant values are found for inactive ones. It thus
seems that in most cases CT to the POM is overestimated vs
opposing CT processes. Calculated electronic spectra obtained
for 1, 3, 5, and 10 confirm this through a trend in fos with
energy that is generally opposite to that measured by Stark, and
also show an underestimate of transition energies. This will
3
10d
transitions.
energies ΔE also run counter to the experimental trend. For
GE
short compounds 1−5, calculated β are a better match for
0
experiment (plausible alternative Stark fits of 1 and 2 produce
β0 of ca. 10 × 10− esu) but still the largest value is obtained
30
have a large effect, since in calculations β α 1/ΔE
.
GE
0
for the −NO derivative 5. Analysis of TD-DFT orbital-to-
Overall, experiment and computation reveal a complex mixture
of CT and π → π* processes whose strength, direction and
energy varies with donor/acceptor, π-bridge, and solvation.
Fully addressing this in silico will require the attention of
theoreticians and computational specialists, as the large
divergence from experiment makes the appropriateness of our
methods an open question. However, prior work has shown
reasonable agreement between similar methods to ours, several
2
orbital transitions provides some points of qualitative agree-
ment with values of Δμ, etc., measured by Stark, while
illustrating some of the disagreements with experiment. For
example, the lowest energy transitions of 1 and 2 are essentially
π → π* transitions of the phenylimido group perturbed by the
POM (Figures S14 and S15 in the Supporting Information),
consistent with the minimal dipolar character observed through
a low (or zero) Stark-determined Δμ . In 5 (Figure S18) all
1
0d
10e
other DFT functionals,
and Hartree−Fock calculations,
1
2
transitions predominantly involve CT from a phenylimido-
and we can identify two modifications that may improve results
J
DOI: 10.1021/acs.inorgchem.7b00708
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
with our approach. One is to include POM-NBu + ion pairing
4
ASSOCIATED CONTENT
Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.inorg-
chem.7b00708.
■
in the solvation model, as this likely affects the interaction with
acetonitrile. The other is to introduce conformational averaging
of properties as the DFT optimized structures of 6−8 and 10
all show coplanar phenyl rings, but the energy barrier to
*
S
−1
24
rotation should be only ca. 1 kcal mol . Indeed, a calculation
on 10 in its crystallized geometry (86° twist) lowers its solution
β0 10-fold. Future DFT computations on imido-POM β may
also be improved by larger basis sets with more diffuse
functions.
Full synthetic details and characterization including
crystallographic details and CIF files for 1, 2, 5, 6, 7,
and 10; additional Raman spectra; example Stark spectra
and fits; details of DFT calculations, including pictorial
representations of frontier orbitals and frontier orbital
energies (PDF)
SUMMARY AND CONCLUSION
■
Accession Codes
Our results show that imido-Lindqvist clusters are an unusually
efficient acceptor for use with short (phenyl) π-bridges in
second-order NLO materials. With organic resonance electron
donors, the resulting chromophores offer better transparency/
nonlinearity tradeoffs than similar purely organic systems and
exceed empirical performance limits that apply to the vast
majority of dipolar organic CT systems. However, extension of
the chromophores by addition of a phenylacetylene spacer,
while increasing absolute β-values, brings their performance
back into line with comparable organic systems. Stark
spectroscopy, supported by DFT calculations and Raman
measurements, suggests that the unusually high intrinsic second
order NLO activity of phenyl bridged systems arises because
charge transfer extends onto the POM, giving unusually large
dipole moment changes (Δμ) for the size of π-system. At the
same time, donor/acceptor electronic coupling remains much
higher than if Δμ were increased by extending the π-system.
Thus, the POM acceptor can increase Δμ, while sacrificing less
transition strength than organic modifications used to achieve
the same goal. In longer chromophores, the POM produces no
gains over a nitro acceptor. This is most likely because acceptor
orbitals spread onto the phenylimido ring, and proportionally
the POM acceptor must have less effect on CT distances when
the π-system is larger.
CCDC 1553805−1553810 contain the supplementary crystal-
lographic data for this paper. These data can be obtained free of
charge via www.ccdc.cam.ac.uk/data_request/cif, or by E-
mailing data_request@ccdc.cam.ac.uk, or by contacting The
Cambridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, U.K.; fax: +44 1223 336033.
AUTHOR INFORMATION
Corresponding Author
*E-mail: John.Fielden@uea.ac.uk.
■
ORCID
Joseph A. Wright: 0000-0001-9603-1001
John Fielden: 0000-0001-5963-7792
Present Address
∥
University of Hail, Faculty of Science, Department of
Chemistry, P.O. Box 2440, Hail, Saudi Arabia.
Notes
The authors declare no competing financial interest.
In addition to the Supporting Information and deposited CIF
files, data can be accessed by contacting the corresponding
author.
ACKNOWLEDGMENTS
■
We thank the UK EPSRC National Crystallography Service in
Southampton for X-ray data for 5, 7, and 10, the UK EPSRC
National Mass Spectrometry Facility (Swansea) for MS, and
the UK National Chemical Computing Service for access to
ADF on Slater. This work was supported by the Iraqi
Government (HCED scholarship to A.A.Y.), Royal Society of
Chemistry, EPSRC (EP/M00452X/1), EU FP7 (Marie Curie
IOF POMHYDCAT Contract No. 254339 to J.F.) and Fonds
voor Wetenschappelijk Onderzoek-Vlaanderen (FWO-V, Ph.D.
fellowship for N.V.S.). K.F. and H.K. would like to thank the
BBSRC (No. BB/M018652/1) for funding. B.S.B. acknowl-
edges support from the Beckman Institute of the California
Institute of Technology. Stark data were collected and analyzed
at the Molecular Materials Resource Center of the Beckman
Institute of the California Institute of Technology.
Compounds with no resonance donor, or with a resonance
acceptor, show no NLO activity, despite similar electronic
transition energies and extinction coefficients to active
compounds. With no resonance donor or acceptor, both TD-
DFT results and the small to zero Δμ measured by Stark
suggest that the electronic transitions are best described as π →
π* transitions perturbed by POM O 2p donor, and POM Mo
4d acceptor orbitals, and hence lack sufficient CT character to
be effective in quadratic NLO. With the −NO resonance
2
acceptor, CT character measured by Stark increases, but not
enough to explain the large DFT-computed β values. The root
0
of this computational inaccuracy appears to be (i) poor
estimation of transition energies, and (ii) inaccurate estimation
of the relative importance of transitions with directionally
opposed dipole moment changes. While (ii) is improved by
solvent correction, producing a similar trend to experimental
data, (i) worsens producing inflated β values. In future, this may
be addressed by larger basis sets with more diffuse functions,
improved solvent corrections, and rotational averaging for
phenylacetylene bridges.
Now that the fundamentals underpinning the performance of
imido-POM chromophores have been understood, future
efforts will be directed at optimizing performance by tailoring
the organic donor systems, and developing bulk NLO materials
suitable for application in devices.
REFERENCES
■
(1) (a) Polyoxometalates: From Platonic Solids to Anti-Retroviral
Activity; Pope, M. T., Mu
Dordrecht, The Netherlands, 1994. (b) Long, D.-L.; Burkholder, E.;
Cronin, L. Polyoxometalate clusters, nanostructures and materials:
From self-assembly to designer materials and devices. Chem. Soc. Rev.
̈
ller, A., Eds.; Kluwer Academic Publishers:
2
007, 36, 105. (c) Fielden, J.; Cronin, L. Coordination Clusters. In
Encyclopedia of Supramolecular Chemistry; Atwood, J. L., Steed, J. W.,
Eds.; Taylor & Francis: London, 2007. (d) Hill, C. L. Polyox-
ometalatesMulticomponent Molecular Vehicles To Probe Funda-
mental Issues and Practical Problems. Chem. Rev. 1998, 98, 1.
K
DOI: 10.1021/acs.inorgchem.7b00708
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(
e) Long, D.-L.; Cronin, L. Pushing the frontiers in polyoxometalate
and metal oxide cluster science. Dalton. Trans. 2012, 41, 9815.
2) (a) Dolbecq, A.; Dumas, E.; Mayer, C. R.; Mialane, P. Hybrid
Y.; Barnes, C. L.; Peng, Z. Hybrid Molecular Materials Based on
Covalently Linked Inorganic Polyoxometalates and Organic Con-
jugated Systems. Angew. Chem., Int. Ed. 2001, 40, 2290.
(
Organic−Inorganic Polyoxometalate Compounds: From Structural
Diversity to Applications. Chem. Rev. 2010, 110, 6009. (b) Proust, A.;
Matt, B.; Villanneau, R.; Guillemot, G.; Gouzerh, P.; Izzet, G.
Functionalization and post-functionalization: a step towards poly-
oxometalate-based materials. Chem. Soc. Rev. 2012, 41, 7605.
(9) (a) Wei, Y.; Xu, B.; Barnes, C. L.; Peng, Z. An Efficient and
Convenient Reaction Protocol to Organoimido Derivatives of
Polyoxometalates. J. Am. Chem. Soc. 2001, 123, 4083. (b) Xu, L.;
Lu, M.; Xu, B.; Wei, Y.; Peng, Z.; Powell, D. R. Towards Main-Chain-
Polyoxometalate-Containing Hybrid Polymers: A Highly Efficient
Approach to Bifunctionalized Organoimido Derivatives of Hexamo-
lybdates. Angew. Chem., Int. Ed. 2002, 41, 4129. (c) Lv, C.; Khan, R. N.
N.; Zhang, J.; Hu, J.; Hao, J.; Wei, Y. Bifunctionalization of
Polyoxometalates with Two Different Organoimido Ligands.
Chem.Eur. J. 2013, 19, 1174.
(
c) Zhang, J.; Xiao, F.; Hao, J.; Wei, Y. The chemistry of organoimido
derivatives of polyoxometalates. Dalton Trans. 2012, 41, 3599.
d) Proust, A.; Thouvenot, R.; Gouzerh, P. Functionalization of
(
polyoxometalates: Towards advanced applications in catalysis and
materials science. Chem. Commun. 2008, 1837.
(
3) (a) Judeinstein, P. Synthesis and properties of polyoxometalates
(
10) (a) Yan, L.; Yang, G.; Guan, W.; Su, Z.; Wang, R. Density
based inorganic−organic polymers. Chem. Mater. 1992, 4, 4. (b) Xu,
B.; Lu, M.; Kang, J.; Wang, D.; Brown, J.; Peng, Z. Synthesis and
Optical Properties of Conjugated Polymers Containing Polyoxometa-
late Clusters as Side-Chain Pendants. Chem. Mater. 2005, 17, 2841.
Functional Theory Study on the First Hyperpolarizabilities of
Organoimido Derivatives of Hexamolybdates. J. Phys. Chem. B 2005,
1
09, 22332. (b) Yang, G.; Guan, W.; Yan, L.; Su, Z.; Xu, L.; Wang, E.
B. Theoretical Study on the Electronic Spectrum and the Origin of
Remarkably Large Third-Order Nonlinear Optical Properties of
Organoimide Derivatives of Hexamolybdates. J. Phys. Chem. B 2006,
(
c) Lu, M.; Xie, B.; Kang, J.; Chen, F.-C.; Yang, Y.; Peng, Z. Synthesis
of Main-Chain Polyoxometalate-Containing Hybrid Polymers and
Their Applications in Photovoltaic Cells. Chem. Mater. 2005, 17, 402.
1
10, 23092. (c) Janjua, M. R. S. A.; Amin, M.; Ali, M.; Bashir, B.; Khan,
(
d) Chakraborty, S.; Keightley, A.; Dusevich, V.; Wang, Y.; Peng, Z.
M. U.; Iqbal, M. A.; Guan, W.; Yan, L.; Su, Z.−M. A DFT Study on
The Two-Dimensional Second-Order Nonlinear Optical (NLO)
Response of Terpyridine-Substituted Hexamolybdates: Physical In-
sight on 2D Inorganic−Organic Hybrid Functional Materials. Eur. J.
Inorg. Chem. 2012, 2012, 705. (d) Janjua, M. R. S. A. Quantum
Mechanical Design of Efficient Second-Order Nonlinear Optical
Materials Based on Heteroaromatic Imido-Substituted Hexamolyb-
dates: First Theoretical Framework of POM-Based Heterocyclic
Aromatic Rings. Inorg. Chem. 2012, 51, 11306. (e) Janjua, M. R. S.
A.; Khan, M. U.; Bashir, B.; Iqbal, M. A.; Song, Y.; Naqvi, S. A. R.;
Khan, Z. A. Effect of π-conjugation spacer on the first hyper-
polarizabilities of polymeric chain containing polyoxometalate cluster
as a side-chain pendant: A DFT study. Comput. Theor. Chem. 2012,
Synthesis and Optical Properties of a Rod−Coil Diblock Copolymer
with Polyoxometalate Clusters Covalently Attached to the Coil Block.
Chem. Mater. 2010, 22, 3995.
(
4) (a) Bar-Nahum, I.; Cohen, H.; Neumann, R. Organometallic−
Polyoxometalate Hybrid Compounds: Metallosalen Compounds
Modified by Keggin Type Polyoxometalates. Inorg. Chem. 2003, 42,
3
677. (b) Araghi, M.; Mirkhani, V.; Moghadam, M.; Tangestaninejad,
S.; Mohammdpoor-Baltork, I. Synthesis and characterization of a new
porphyrin−polyoxometalate hybrid material and investigation of its
catalytic activity. Dalton Trans. 2012, 41, 3087.
(
5) (a) Bonchio, M.; Carraro, M.; Scorrano, G.; Bagno, A.
Photooxidation in Water by New Hybrid Molecular Photocatalysts
Integrating an Organic Sensitizer with a Polyoxometalate Core. Adv.
Synth. Catal. 2004, 346, 648. (b) Allain, C.; Favette, S.; Chamoreau, L.-
M.; Vaissermann, J.; Ruhlmann, L.; Hasenknopf, B. Hybrid Organic−
Inorganic Porphyrin−Polyoxometalate Complexes. Eur. J. Inorg. Chem.
9
(
94, 34.
11) (a) Marder, S. R.; Perry, J. W.; Schaefer, W. P. Synthesis of
Organic Salts with Large Second-Order Optical Nonlinearities. Science
989, 245, 626. (b) Coe, B. J.; Harris, J. A.; Asselberghs, I.; Clays, K.;
1
2
(
008, 2008, 3433.
́
6) (a) Odobel, F.; Severac, M.; Pellegrin, Y.; Blart, E.; Fosse, C.;
Olbrechts, G.; Persoons, A.; Hupp, J. T.; Johnson, R. C.; Coles, S. J.;
Hursthouse, M. B.; Nakatani, K. Quadratic Nonlinear Optical
Properties of N-Aryl Stilbazolium Dyes. Adv. Funct. Mater. 2002, 12,
Cannizzo, C.; Mayer, C. R.; Elliott, K. J.; Harriman, A. Coupled
Sensitizer−Catalyst Dyads: Electron-Transfer Reactions in a Per-
ylene−Polyoxometalate Conjugate. Chem.Eur. J. 2009, 15, 3130.
1
10. (c) Coe, B. J.; Harris, J. A.; Asselberghs, I.; Wostyn, K.; Clays, K.;
Persoons, A.; Brunschwig, B. S.; Coles, S. J.; Gelbrich, T.; Light, M. E.;
Hursthouse, M. B.; Nakatani, K. Quadratic Optical Nonlinearities of
N-Methyl and N-Aryl Pyridinium Salts. Adv. Funct. Mater. 2003, 13,
(
b) Harriman, A.; Elliott, K. J.; Alamiry, M. A. H.; Le Pleux, L.;
́
Severac, M.; Pellegrin, Y.; Blart, E.; Fosse, C.; Cannizzo, C.; Meyer, C.
R.; Odobel, F. Intramolecular Electron Transfer Reactions Observed
for Dawson-Type Polyoxometalates Covalently Linked to Porphyrin
Residues. J. Phys. Chem. C 2009, 113, 5834. (c) Elliott, K. J.; Harriman,
A.; Le Pleux, L.; Pellegrin, Y.; Blart, E.; Mayer, C. R.; Odobel, F. A
porphyrin−polyoxometallate bio-inspired mimic for artificial photo-
synthesis. Phys. Chem. Chem. Phys. 2009, 11, 8767. (d) Matt, B.;
Coudret, C.; Viala, C.; Jouvenot, D.; Loiseau, F.; Izzet, G.; Proust, A.
Elaboration of Covalently Linked Polyoxometalates with Ruthenium
and Pyrene Chromophores and Characteriation of Their Photo-
physical Properties. Inorg. Chem. 2011, 50, 7761. (e) Matt, B.; Xiang,
X.; Kaledin, A. L.; Han, N.; Moussa, J.; Amouri, H.; Alves, S.; Hill, C.
L.; Lian, T.; Musaev, D. G.; Izzet, G.; Proust, A. Long lived charge
separation in iridium(III)-photosensitized polyoxometalates: Syn-
thesis, photophysical and computational studies of organometallic−
redox tunable oxide assemblies. Chem. Sci. 2013, 4, 1737.
3
47. (d) Coe, B. J.; Fielden, J.; Foxon, S. P.; Helliwell, M.; Asselberghs,
I.; Clays, K.; De Mey, K.; Brunschwig, B. S. Syntheses and Properties
of Two-Dimensional, Dicationic Nonlinear Optical Chromophores
Based on Pyrazinyl Cores. J. Org. Chem. 2010, 75, 8550. (e) Jang, S.-
H.; Luo, J.; Tucker, N. M.; Leclercq, A.; Zojer, E.; Haller, M. A.; Kim,
T.-D.; Kang, J.-W.; Firestone, K.; Bale, D.; Lao, D.; Benedict, J. B.;
́
Cohen, D.; Kaminsky, W.; Kahr, B.; Bredas, J.-L.; Reid, P.; Dalton, L.
R.; Jen, A. K.-Y. Pyrroline Chromophores for Electro-Optics. Chem.
Mater. 2006, 18, 2982. (f) Coe, B. J.; Fielden, J.; Foxon, S. P.; Harris, J.
A.; Helliwell, M.; Brunschwig, B. S.; Asselberghs, I.; Clays, K.; Garín, J.;
Orduna, J. Diquat Derivatives: Highly Active, Two-Dimensional
Nonlinear Optical Chromophores with Potential Redox Switchability.
J. Am. Chem. Soc. 2010, 132, 10498. (g) Coe, B. J.; Fielden, J.; Foxon,
S. P.; Helliwell, M.; Brunschwig, B. S.; Asselberghs, I.; Clays, K.;
Olesiak, J.; Matczyszyn, K.; Samoc, M. Quadratic and Cubic Nonlinear
Optical Properties of Salts of Diquat-Based Chromophores with
Diphenylamino Substituents. J. Phys. Chem. A 2010, 114, 12028.
(h) Cheng, L. T.; Tam, W.; Stevenson, S. H.; Meredith, G. R.; Rikken,
G.; Marder, S. R. Experimental investigations of organic molecular
nonlinear optical polarizabilities. 1. Methods and results on benzene
and stilbene derivatives. J. Phys. Chem. 1991, 95, 10631. (i) Cheng, L.
T.; Tam, W.; Marder, S. R.; Stiegman, A. E.; Rikken, G.; Spangler, C.
W. Experimental investigations of organic molecular nonlinear optical
(
7) Matt, B.; Fize, J.; Moussa, J.; Amouri, H.; Pereira, A.; Artero, V.;
Izzet, G.; Proust, A. Charge photo-accumulation and photocatalytic
hydrogen evolution under visible light at an iridium(III)-photo-
sensitized polyoxotungstate. Energy Environ. Sci. 2013, 6, 1504.
(
8) (a) Strong, J. B.; Yap, G. P. A.; Ostrander, R.; Liable-Sands, L. M.;
Rheingold, A. L.; Thouvenot, R.; Gouzerh, P.; Maatta, E. A. A New
Class of Functionalized Polyoxometalates: Synthetic, Structural,
Spectroscopic, and Electrochemical Studies of Organoimido Deriva-
2
−
tives of [Mo O ] . J. Am. Chem. Soc. 2000, 122, 639. (b) Xu, B.; Wei,
6
19
L
DOI: 10.1021/acs.inorgchem.7b00708
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
polarizabilities. 2. A study of conjugation dependences. J. Phys. Chem.
́
Okada, S.; Oikawa, H.; Nakanishi, H.; Vogel, H.; Beljonne, D.; Bredas,
1
(
991, 95, 10643.
12) (a) Nonlinear Optics of Organic Molecules and Polymers; Nalwa,
J.-L. Fourier analysis of the femtosecond hyper-Rayleigh scattering
signal from ionic fluorescent hemicyanine dyes. J. Opt. Soc. Am. B
2000, 17, 256. (d) Franz, E.; Harper, E. C.; Coe, B. J.; Zahradnik, P.;
Clays, K.; Asselberghs, I. Benzathiazoliums and pyridiniums for
second-order nonlinear optics. Proc. SPIE 2008, 6999, 699923−1−
699923−11.
H. S., Miyata, S., Eds.; CRC Press: Boca Raton, FL, 1997. (b) Marder,
S. R. Organic nonlinear optical materials: Where we have been and
where we are going. Chem. Commun. 2006, 131. (c) Kuzyk, M. G.
Using fundamental principles to understand and optimize nonlinear-
optical materials. J. Mater. Chem. 2009, 19, 7444.
(20) Campo, J.; Desmet, F.; Wenseleers, W.; Goovaerts, E. Highly
sensitive setup for tunable wavelength hyper-Rayleigh scattering with
parallel detection and calibration data for various solvents. Opt. Express
2009, 17, 4587.
(21) (a) Shin, Y. K.; Brunschwig, B. S.; Creutz, C.; Sutin, N.
Electroabsorption Spectroscopy of Charge-Transfer States of Tran-
sition-Metal Complexes. 2. Metal-to-Ligand and Ligand-to-Metal
Charge-Transfer Excited States of Pentaammineruthenium Complexes.
J. Phys. Chem. 1996, 100, 8157. (b) Coe, B. J.; Harris, J. A.;
Brunschwig, B. S. Electroabsorption Spectroscopic Studies of Dipolar
Ruthenium(II) Complexes Possessing Large Quadratic Nonlinear
Optical Responses. J. Phys. Chem. A 2002, 106, 897.
(22) (a) Liptay, W. In Excited States, Vol. 1; Lim, E. C., Ed.;
Academic Press: New York, 1974; pp 129−229. (b) Bublitz, G. U.;
Boxer, S. G. Stark Spectroscopy: Applications in Chemistry, Biology,
and Materials Science. Annu. Rev. Phys. Chem. 1997, 48, 213.
(c) Vance, F. W.; Williams, R. D.; Hupp, J. T. Electroabsorption
spectroscopy of molecular inorganic compounds. Int. Rev. Phys. Chem.
1998, 17, 307. (d) Brunschwig, B. S.; Creutz, C.; Sutin, N.
Electroabsorption spectroscopy of charge transfer states of transition
metal complexes. Coord. Chem. Rev. 1998, 177, 61.
(
13) (a) Dhenaut, C.; Ledoux, I.; Samuel, I. D. W.; Zyss, J.;
Bourgault, M.; Le Bozec, H. Chiral metal complexes with large
octupolar optical nonlinearities. Nature 1995, 374, 339. (b) McDonagh,
A. M.; Humphrey, M. G.; Samoc, M.; Luther-Davies, B.; Houbrechts,
S.; Wada, T.; Sasabe, H.; Persoons, A. Organometallic Complexes for
1
Nonlinear Optics. 16. Second and Third Order Optical Nonlinearities
of Octopolar Alkynylruthenium Complexes. J. Am. Chem. Soc. 1999,
1
21, 1405. (c) Di Bella, S. Second-order nonlinear optical properties of
transition metal complexes. Chem. Soc. Rev. 2001, 30, 355. (d) Coe, B.
J.; Fielden, J.; Foxon, S. P.; Brunschwig, B. S.; Asselberghs, I.; Clays,
K.; Samoc, A.; Samoc, M. Combining Very Large Quadratic and Cubic
Nonlinear Optical Responses in Extended, Tris-Chelate Metal-
lochromophores with Six π-Conjugated Pyridinium Substituents. J.
Am. Chem. Soc. 2010, 132, 3496. (e) Maury, O.; Le Bozec, H.
Molecular Engineering of Octupolar NLO Molecules and Materials
Based on Bipyridyl Metal Complexes. Acc. Chem. Res. 2005, 38, 691.
(
f) Le Bozec, H.; Renouard, T. Dipolar and Non-Dipolar Pyridine and
Bipyridine Metal Complexes for Nonlinear Optics. Eur. J. Inorg. Chem.
000, 2000, 229. (g) Alcaraz, G.; Euzenat, L.; Mongin; Katan, C.;
2
Ledoux, I.; Zyss, J.; Blanchard-Desce, M.; Vaultier, M. Improved
transparency−nonlinearity trade-off with boroxine-based octupolar
molecules. Chem. Commun. 2003, 2766. (h) Kang, H.; Zhu, P.; Yang,
Y.; Facchetti, A.; Marks, T. J. Self-Assembled Electrooptic Thin Films
with Remarkably Blue-Shifted Optical Absorption Based on an X-
Shaped Chromophore. J. Am. Chem. Soc. 2004, 126, 15974.
(23) Xiao, Z.; Zhu, Y.; Wei, Y.; Wang, Y. Synthesis and characteristic
of a new arylimido derivative of hexamolybdate with remote strong
electro-donating group (Bu N) [Mo O NC H N(CH )2−p]. Inorg.
4
2
6
18
6
4
3
Chem. Commun. 2006, 9, 400.
(24) (a) Lin, V. S.; Therien, M. J. The Role of Porphyrin-to-
Porphyrin Linkage Topology in the Extensive Modulation of the
Absorptive and Emissive Properties of a Series of Ethynyl- and
Butadiynyl-Bridged Bis- and Tris(porphinato)zinc Chromophores.
Chem.Eur. J. 1995, 1, 645. (b) LeCours, S. M.; DiMagno, S. G.;
Therien, M. J. Exceptional Electronic Modulation of Porphyrins
through meso-Arylethynyl Groups. Electronic Spectroscopy, Elec-
tronic Structure, and Electrochemistry of [5,15-Bis[(aryl)ethynyl]-
10,20-diphenylporphinato]zinc(II) Complexes. X-ray Crystal Struc-
tures of [5,15-Bis[(4′-fluorophenyl)ethynyl]-10,20-
diphenylporphinato]zinc(II) and 5,15-Bis[(4′-methoxyphenyl)-
ethynyl]-10,20-diphenylporphyrin. J. Am. Chem. Soc. 1996, 118,
11854. (c) Kamat, N. P.; Liao, Z.; Moses, L. E.; Rawson, J.;
Therien, M. J.; Dmochowski, I. J.; Hammer, D. A. Sensing membrane
stress with near IR-emissive porphyrins. Proc. Natl. Acad. Sci. U. S. A.
2011, 108, 13984.
(25) (a) Barlow, S.; Marder, S. R. Electronic and optical properties of
conjugated group 8 metallocene derivatives. Chem. Commun. 2000,
1555. (b) Kim, H. M.; Cho, B. R. Second-order nonlinear optical
properties of octupolar molecules structure−property relationship. J.
Mater. Chem. 2009, 19, 7402. (c) Hennrich, G.; Asselberghs, I.; Clays,
K.; Persoons, A. Tuning Octopolar NLO Chromophores: Synthesis
and Spectroscopic Characterization of Persubstituted 1,3,5-Tris-
(ethynylphenyl)benzenes. J. Org. Chem. 2004, 69, 5077. (d) Cheng,
L. T.; Tam, W.; Marder, S. R.; Stiegman, A. E.; Rikken, G.; Spangler,
C. W. Experimental investigations of organic molecular nonlinear
optical polarizabilities. 2. A study of conjugation dependences. J. Phys.
Chem. 1991, 95, 10643. (e) Lee, J. Y.; Suh, S. B.; Kim, K. S. Polyenes
vs polyynes: Efficient π-frame for nonlinear optical pathways. J. Chem.
Phys. 2000, 112, 344. (f) Champagne, B.; Kirtman, B. Vibrational
versus electronic first hyperpolarizabilities of π-conjugated organic
molecules: an ab initio Hartree−Fock investigation upon the effects of
the nature of the linker. Chem. Phys. 1999, 245, 213.
(
14) Al-Yasari, A.; Van Steerteghem, N.; El Moll, H.; Clays, K.;
Fielden, J. Donor−acceptor organo-imido polyoxometalates: high
transparency, high activity redox-active NLO chromophores. Dalton
Trans. 2016, 45, 2818.
(
́
15) Tripathy, K.; Perez Moreno, J.; Kuzyk, M. G.; Coe, B. J.; Clays,
K.; Kelley, A. M. Why hyperpolarizabilities fall short of the
fundamental quantum limits. J. Chem. Phys. 2004, 121, 7932.
(
16) (a) Kang, H.; Facchetti, A.; Jiang, H.; Cariati, E.; Righetto, S.;
Ugo, R.; Zuccaccia, C.; Macchioni, A.; Stern, C. L.; Liu, Z.; Ho, S. T.;
Brown, E. C.; Ratner, M. A.; Marks, T. J. Ultralarge Hyperpolarizability
Twisted π-Electron System Electro-Optic Chromophores: Synthesis,
Solid-State and Solution-Phase Structural Characteristics, Electronic
Structures, Linear and Nonlinear Optical Properties, and Computa-
tional Studies. J. Am. Chem. Soc. 2007, 129, 3267. (b) Shi, Y.;
Frattarelli, D.; Watanabe, N.; Facchetti, A.; Cariati, E.; Righetto, S.;
Tordin, E.; Zuccaccia, C.; Macchioni, A.; Wegener, S. L.; Stern, C. L.;
Ratner, M. A.; Marks, T. J. Ultra-High-Response, Multiply Twisted
Electro-optic Chromophores: Influence of π-System Elongation and
Interplanar Torsion on Hyperpolarizability. J. Am. Chem. Soc. 2015,
1
(
37, 12521.
17) Beverina, L.; Sanguineti, A.; Battagliarin, G.; Ruffo, R.; Roberto,
D.; Righetto, S.; Soave, R.; Lo Presti, L.; Ugo, R.; Pagani, G. A. UV
absorbing zwitterionic pyridinium-tetrazolate: exceptional transpar-
ency/optical nonlinearity trade-off. Chem. Commun. 2011, 47, 292.
(
18) (a) Clays, K.; Persoons, A. Hyper-Rayleigh scattering in
solution. Phys. Rev. Lett. 1991, 66, 2980. (b) Clays, K.; Persoons, A.
Hyper-Rayleigh scattering in solution. Rev. Sci. Instrum. 1992, 63, 3285.
(
c) Hendrickx, E.; Clays, K.; Persoons, A. Hyper-Rayleigh Scattering in
Isotropic Solution. Acc. Chem. Res. 1998, 31, 675.
19) (a) Olbrechts, G.; Strobbe, R.; Clays, K.; Persoons, A. High-
(
frequency demodulation of multi-photon fluorescence in hyper-
Rayleigh scattering. Rev. Sci. Instrum. 1998, 69, 2233. (b) Olbrechts,
G.; Wostyn, K.; Clays, K.; Persoons, A. High-frequency demodulation
of multiphoton fluorescence in long-wavelength hyper-Rayleigh
scattering. Opt. Lett. 1999, 24, 403. (c) Clays, K.; Wostyn, K.;
Olbrechts, G.; Persoons, A.; Watanabe, A.; Nogi, K.; Duan, X.-M.;
(26) Flipse, M. C.; De Jonge, R.; Woudenberg, R. H.; Marsman, A.
W.; Van Walree, C. A.; Jenneskens, L. W. The determination of first
hyperpolarizabilities β using hyper-Rayleigh scattering: A caveat. Chem.
Phys. Lett. 1995, 245, 297.
M
DOI: 10.1021/acs.inorgchem.7b00708
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(
27) (a) Walsh, J. T.; Mallon, C. T.; Bond, A. M.; Keyes, T. E.;
Forster, R. J. Enhanced photocurrent production from thin films of
Ru(II) metallopolymer/Dawson polyoxotungstate adducts under
visible irradiation. Chem. Commun. 2012, 48, 3593. (b) Walsh, J. J.;
Long, D.-L.; Cronin, L.; Bond, A. M.; Forster, R. J.; Keyes, T. E.
2
+
Electronic and photophysical properties of adducts of [Ru(bpy)3]
4
−
and Dawson-type sulfite polyoxomolybdates α/β-[Mo O (SO ) ] .
18
54
3 2
Dalton Trans. 2011, 40, 2038. (c) Keyes, T. E.; Gicquel, E.; Guerin, L.;
Forster, R. J.; Hultgren, V. M.; Bond, A. M.; Wedd, A. G.
Photophysical and Novel Charge-Transfer Properties of Adducts
II
3
2+
4−
between [Ru (bpy) ] and [S Mo O ] . Inorg. Chem. 2003, 42,
2
18 62
7
897. (d) Walsh, J. J.; Long, D.-L.; Cronin, L.; Bond, A. M.; Forster, R.
J.; Keyes, T. E. Electronic and photophysical properties of adducts of
+
[
[
Ru(bpy)3]2 and Dawson-type sulfite polyoxomolybdates α/β-
4
−
Mo O (SO ) ] . Dalton Trans. 2011, 40, 2038. (e) Zhu, J.;
18
54
3 2
Zeng, Q.; O’Carroll, S.; Bond, A. M.; Keyes, T. E.; Forster, R. J.
Photocurrent generation from thin films of ruthenium metal-
lopolymer:polyoxometalate adducts using visible excitation. Electro-
chem. Commun. 2011, 13, 899.
(
28) (a) Bridgeman, A. J.; Cavigliasso, G. Density functional study of
the vibrational frequencies of Lindqvist polyanions. Chem. Phys. 2002,
79, 143. (b) Buckley, R. I.; Clark, R. J. H. Structural and electronic
properties of some polymolybdates reducible to molybdenum blues.
Coord. Chem. Rev. 1985, 65, 167. (c) Mattes, R.; Bierbusse, H.; Fuchs,
2
̈
J. Schwingungsspektren und Kraftkonstanten von Polyanionen mit
M O -Gruppen. Z. Anorg. Allg. Chem. 1971, 385, 230. (d) Rocchiccio-
6
19
li-Deltcheff, C.; Thouvenot, R.; Fouassier, M. Vibrational inves-
2
−
tigations of polyoxometallates. 1. Valence force field of Mo O
6
19
based on total isotopic substitution (oxygen-18, molybdenum-92,
molybdenum-100). Inorg. Chem. 1982, 21, 30.
(
29) She, S.; Bian, S.; Hao, J.; Zhang, J.; Zhang, J.; Wei, Y. Aliphatic
Organoimido Derivatives of Polyoxometalates Containing a Bioactive
Ligand. Chem.Eur. J. 2014, 20, 16987.
(
30) Osipov, V. G.; Shlyapochnikov, V. A.; Ponizovtsev, E. F.
Vibrational spectra of aromatic nitro compounds. J. Appl. Spectrosc.
968, 8, 598.
31) Coe, B. J.; Harris, J. A.; Clays, K.; Persoons, A.; Wostyn, A.;
1
(
Brunschwig, B. S. A comparison of the pentaammine(pyridyl)-
ruthenium(II) and 4-(dimethylamino)phenyl groups as electron
donors for quadratic non-linear optics. Chem. Commun. 2001, 1548.
(
32) Suppan, P.; Tsiamis, C. J. Temperature effects in solvatochromic
shifts. J. Chem. Soc., Faraday Trans. 2 1981, 77, 1553.
33) (a) te Velde, G.; Bickelhaupt, F. M.; Baerends, E. J.; Fonseca
(
Guerra, C.; van Gisbergen, S. J. A.; Snijders, J. G.; Ziegler, T.
Chemistry with ADF. J. Comput. Chem. 2001, 22, 931. (b) Fonseca
Guerra, C.; Snijders, J. G.; Te Velde, G.; Baerrends, E. Towards an
order-N DFT method. J. Theor. Chem. Acc. 1998, 99, 391.
(
c) ADF2014.10, SCM, Theoretical Chemistry, Vrije Universiteit,
Amsterdam, The Netherlands, http://www.scm.com.
34) (a) Gritsenko, O. V.; Schipper, P. R. T.; Baerends, E. J.
(
Approximation of the exchange-correlation Kohn-Sham potential with
a statistical average of different orbital model potentials. Chem. Phys.
Lett. 1999, 302, 199. (b) Schipper, P. R. T.; Gritsenko, O. V.; van
Gisbergen, S. J. A.; Baerends, E. J. Molecular calculations of excitation
energies and (hyper)polarizabilities with a statistical average of orbital
model exchange-correlation potentials. J. Chem. Phys. 2000, 112, 1344.
N
DOI: 10.1021/acs.inorgchem.7b00708
Inorg. Chem. XXXX, XXX, XXX−XXX