Two-photon absorption of styryl-quinolinium, -pyridinium, and -barbituric acid derivatives, and intramolecular charge transfer

Article abstract of DOI:10.1139/v00-188

A series of new chromophores, styryl-parent end-capped with various donors, and with barbituric acid, methyl-pyridinium, and methyl-quinolinium as the acceptors, has been synthesized and characterized by element analysis or X-ray diffraction. Using the Z-scan system, their two-photon absorption (TPA) cross-section values (δ) have been determined under excitation with 10 Hz, and 1064 nm, 35 ps mode-locked Nd:YAG laser pulse in DMF with d_o = 0.05 M. The effective δ value is as high as 10.9 × 10^-48 cm⁴ s per photon for trans-4-(4′-N,N-diphenyl amino) styryl-N-methyl quinolinium iodide (DPASQI). The δ value increases from barbituric acid- to pyridinium- to quinolinium-derivatives apparently due to the increase in both the conjugated degree and planarity; however, when the acceptor is fixed, the δ value increases from dialkyl amino groups to diphenyl amino groups even though the latter is a weaker donor than the dialkyl amino groups. Theoretical calculations confirm that the increased distortion from planarity for the barbituric acid derivative makes its δ value decrease. The relatively large δ value for quinolinium- or pyridinium-derivatives originates from larger intramolecular charge transfer, which can be characterized by the difference of dipole moment (Δμ_ge) between the S₀ and S₁, and the transition dipole moment (M_ee′) between S₁ and S₂.

Full text of DOI:10.1139/v00-188

174
Two-photon absorption of styryl-quinolinium,
-pyridinium, and -barbituric acid derivatives, and
intramolecular charge transfer
XiaoMei Wang, Chun Wang, WenTao Yu, YuFang Zhou, Xian Zhao, Qi Fang, and
MinHua Jiang
Abstract: A series of new chromophores, styryl-parent end-capped with various donors, and with barbituric acid,
methyl-pyridinium, and methyl-quinolinium as the acceptors, have been synthesized and characterized by element anal-
ysis or X-ray diffraction. Using the Z-scan system, their two-photon absorption (TPA) cross-section values (δ) have
been determined under excitation with 10 Hz, and 1064 nm, 35 ps mode-locked Nd:YAG laser pulse in DMF with d0 = 0.05
M. The effective δ value is as high as 10.9 × 10–48 cm4 s per photon for trans-4-(4′-N,N-diphenyl amino) styryl-N-
methyl quinolinium iodide (DPASQI). The δ value increases from barbituric acid- to pyridinium- to quinolinium-
derivatives apparently due to the increase in both the conjugated degree and planarity; however, when the acceptor is
fixed, the δ value increases from dialkyl amino groups to diphenyl amino groups even though the latter is a weaker do-
nor than the dialkyl amino groups. Theoretical calculations confirm that the increased distortion from planarity for the
barbituric acid derivative makes its δ value decrease. The relatively large δ value for quinolinium- or pyridinium-
derivatives originates from larger intramolecular charge transfer, which can be characterized by the difference of dipole
moment (∆µge) between the S0 and S1, and the transition dipole moment (Mee′) between S1 and S2.
Key words: two-photon absorption, intramolecular charge transfer, styryl-quinolinium, styryl-pyridinium, styryl-
barbituric acid.
Résumé : On a synthétisé une série de nouveaux chromophores, un styryle de base recouvert de divers donneurs et des
accepteurs comme l’acide barbiturique et les ions méthylpyridinium et méthylquinoléinium et on les a caractérisés par
analyse élémentaire ou diffraction des rayons X. Faisant appel à un système à balayage Z et opérant dans le DMF avec
d0 0,05 M, une excitation de 10-Hz et une pulsation laser Nd:YAG de 35 ps en mode bloqué, on a déterminé les valeurs,
δ, des sections efficaces de leur absorption de deux photons. Dans le cas de l’iodure du trans-4-((4′-N,N-diphénylamino)
styryl-N-méthylquinoléinium (« DPASQI »), la valeur δ effective s’élève jusqu’à 10,9 × 10–48 cm4 s/photon. Par ail-
leurs, la valeur de δ augmente en passant de l’acide barbiturique, puis aux dérivés pyridinium et quinoléinium; il
semble que cette situation résulte d’une augmentation tant du degré de conjugaison que de la planarité. De plus,
lorsque l’accepteur est fixé, la valeur de δ augmente des groupes dialkylamino aux groupes diphénylamino même si ces
derniers sont des donneurs plus faibles que les groupes dialkylamino. Des calculs théoriques confirment que, dans le
cas de l’acide barbiturique, l’augmentation de la distorsion par rapport à la planarité provoque une diminution de la va-
leur de δ. La valeur relativement élevée de δ pour les dérivés quinolinéinium et pyridinium trouverait son origine dans
un important transfert de charge intramoléculaire qui peut être caractérisé par la différence du moment dipolaire
(∆µge) entre S0 et S1 et la transition du moment dipolaire (M ee′) entre S1 et S2.
Mots clés : absorption à deux photons, transfert de charge intramoléculaire, styryl-quinoléinium, styryl-pyridinium, sty-
ryl-acide barbiturique.
[Traduit par la Rédaction] Wang et al 182
Introduction
Organic conjugated chromophores with a large two-photon
absorption (TPA) cross-section emerge as very attractive ele-
ments for use in a number of optical applications such as op-
tical data storage, up-converted lasing, and optical power
limiting (1–10). Numerous organic compounds have been
investigated experimentally and theoretically in order to un-
derstand the structure–property relationship of materials with
large TPA cross-sections (11–17). Some general structural
parameters for increasing the molecular TPA cross-section
values (δ) have gradually been drawn in several laboratories
(11–15). For example: (i) increasing the donor and acceptor
strength; (ii) changing the character of the conjugated
bridge; (iii) increasing the planarity of the chromophores; and
(iv) extending the conjugated length. The results of theoreti-
Received June 19, 2000. Published on the NRC Research
Press Web site on March 14, 2001.
X.M. Wang, C. Wang, W.T. Yu, Y.F. Zhou, X. Zhao, Q.
Fang, and M.H. Jiang.1 State Key Laboratory of Crystal
Materials, Shandong University Jinan, 250100 China.
1Corresponding author (e-mail: mhjiang@sdu.edu.cn).
Can. J. Chem. 79: 174–182 (2001)
© 2001 NRC Canada
DOI: 10.1139/cjc-79-2-174

Wang et al
175
Table 1. Optical properties and calculated structural parameters for all chromophores.
∆µge
(debye)
Mee′
(debye)
β
δ (× 10–48
)
θd/o a
Chromophore
(cm GW–1)
cm4 s photon–1
HMASBA
PSPI
70
0.7
3.4
8.1
—
0.60
0.91
1.09
1.12
1.42
1.41
1.75
3.9
5.7
6.8
7.0
8.8
Planar
Planar
Planar
Planar
Planar
Planar
8.5
8.2
8.6
9.8
—
DEASPI
HEASPI
DPASPI
HMASQI
DPASQI
8.2
11.8
10.6
12.1
8.8
10.9
—
aThe dihedral angle between two rings of the backbone.
Table 2. Crystallographic data for DPASPI and HMASQI.
DPASPI
HMASQI
Formula
MW
C26H23IN2
490.36
C21H23IN2O
446.31
T (K)
293
293
System
Space group
Orthorhombic
Pbca
Monoclinic
P21/n
a, b, c (Å)
10.5638(14), 11.1909(19), 37.255(4)
10.0512(11), 16.2879(12), 12.3328(8)
α, β, γ (°)
V (Å3)
90
90, 102.307(6), 90
1972.6(3)
4404.2(11)
Z
8
4
D calcd. (g cm–1)
Absorption coefficient (mm–1)
F (000)
1.479
1.468
1968
1.503
1.634
896
Crystal size (mm)
θ range (deg)
Index range
Refl. collected
Independent refl.
Max and min transmission
Final R indices [I > 2σ(I)]
R indices
Largest diff. peak and hole (e Å–3)
GoF on F2
0.20 × 0.28 × 0.25
2.2–25.0
–12 ≤ h ≤ 1, –13 ≤ k ≤ 1, – 44 ≤ l ≤ 1
4906
3847 [R (int) = 0.064]
0.301 and 0.202
R1 = 0.0776, wR2 = 0.2372
R1 = 0.1409, wR2 = 0.2372
1.79 and –1.11
0.35 × 0.30 × 0.20
2.10–26.00
–1 ≤ h ≤ 12, –1 ≤ k ≤ 20, –15 ≤ l ≤ 15
4906
3882 [R (int) = 0.0230]
0.689 and 0.561
R1 = 0.0358, wR2 = 0.0824
R1 = 0.0558, wR2 = 0.0939
0.5809 and – 0.499
1.066
0.93
Data/restraint/parameters
3847 / 0 / 262
3882 / 0 / 226
cal studies further reveal that the TPA δ value is related to
the imaginary part of the third-order polarizability (16),
between the two excited states. The first two terms in eq. [2]
make a positive contribution to the TPA process (17).
It is evident that the TPA δ value is dependent not only on
the difference between the ground- and excited-state dipole
moment but also the transition dipole moment between two
excited states. It is rational to deduce that materials with
large ∆µge and Mee′ must have exhibited high δ values due to
stronger intramolecular charge transfer.
Based on this idea, we calculated structural parameters,
such as ∆µge and Mee′, for π-donor–acceptor compounds.
One of the better candidate groups for generating a larger intra-
molecular charge shift is the substituted styryl-pyridinium
cations. Other promising candidates are the substituted
styryl-quinolinium cations. These kinds of salts possess the
fused aromatic heterocyclic ring that may improve π-electron
delocalization and result in better TPA properties. In addi-
tion the structural parameters for the barbituric acid deriva-
tive have also been calculated to examine the influence
planarity and conjugated length upon TPA property. The cal-
culation results are shown in Table 1.
4π hω2
[1]
δ(ω) =
L4 Im γ (−ω ;ω ,ω , −ω )
n2 c2
where h is Planck’s constant, n is the index of refraction of
the medium, L is a local field factor (equal to 1 for vacuum),
and c or ω is the speed of light or the frequency of light.
When the frequency (ω) is left out of consideration, the third-
order polarizability (γ) can be expressed as eq. [2] (17–18),
M 2ge
E 2ge
M 2ge
Ege
∆µ2
M 2ee
Ege′
ge
[2]
γxxxx = 24
+
−
∑
Ege
e′
where ∆µge is the difference of dipole moment between the
ground state (S0) and the first excited state (S1); Mge or Mee′ is
the transition dipole moment between S0 and S1 or between
the two lowest excited states, that is, S1 and S2, and Ege or Ege′
is the difference between ground and excited state energy or
© 2001 NRC Canada

176
Can. J. Chem. Vol. 79, 2001
Scheme 1.
methyl-pyridinium and further to methyl-quinolinium moi-
ety, and also by varying the donor strengths in the order of:
dialkylamino group > diphenylamino group. On the basis of
correlated quantum-chemical calculation and experimental
measurement results, we further demonstrate that large ∆µge
and Mee′ can result in large intramolecular charge transfers,
which have contributed to enhance the two-photon absorp-
tion cross-section.
ethyl-N-ethyl amino)benzaldehyde and 4-(N-hydroxyethyl-
N-methyl amino)benzaldehyde were synthesized according
to ref. 11.
4-Methyl-N-methylquilinonium iodide was synthesized by
the reaction of lepidine (4.3 g, 30 mmol) and methyl iodide
(4.4 g, 31 mmol) in ethanol (30 mL) at reflux temperature
for 2 h. After cooling, the solution was filtered and recry-
stallized in ethanol. Yellow needle crystals were obtained
with yield 80%.
Experimental
Synthesis of chromophores
Synthesis
Chromophores were synthesized by the Knoevenagel con-
densation reaction of the appropriate substituted benzalde-
hyde and the acceptor components such as 4-methyl-N-
methylpyridinium iodide and 4-methyl-N-methylquinolinium
iodide with a catalytic amount of piperidine. For example, a
mixture of equal molar 4-(N,N-diphenyl amino)benzalde-
hyde (3.5 g, 13 mmol) and 4-methyl-N-methylquinolinium
iodide (3.7 g, 13 mmol) were stirred in ethanol (20 mL)
at reflux temperature overnight in the presence of 4–5 drops
piperdine. After cooling, the crude product of DPASQI
The decomposition temperatures were measured on a
Perkin–Elmer TGS-2 thermogravimetric analyzer at a rate of
20°C min–1 under nitrogen atmosphere. Element analyses
were performed on a Perkin 2400. Mass spectra were re-
corded with an HP5989 apparatus. NMR spectra were deter-
mined on FX-90Q NMR spectrometer (TMS as standard).
Synthesis of donor and acceptor components
4-Pyrrolidinyl benzaldehyde, 4-(N,N-diethyl amino)benzalde-
hyde, 4-(N,N-diphenyl amino)benzaldehyde, 4-(N-hydroxy-
© 2001 NRC Canada

Wang et al
177
Table 3. Selected bond lengths (Å) and angles (°) for HMASQI.
N(1)—C(10)
N(1)—C(2)
N(1)—C(1)
N(2)—C(16)
N(2)—C(21)
N(2)—C(19)
O(1)—C(20)
C(2)—C(7)
C(17)—C(18)
C(2)-N(1)-C(10)
C(10)-N(1)-C(1)
C(2)-N(1)-C(1)
C(16)-N(2)-C(21)
C(16)-N(2)-C(19)
C(21)-N(2)-C(19)
N(1)-C(2)-C(3)
N(1)-C(2)-C(7)
C(3)-C(2)-C(7)
C(4)-C(3)-C(2)
C(18)-C(17)-(16)
C(17)-C(16)-(13)
1.331(5)
1.381(5)
1.475(5)
1.353(5)
1.464(6)
1.480(6)
1.465(6)
1.413(5)
1.379(6)
120.5(3)
118.9(4)
120.6(3)
121.6(4)
121.9(4)
115.6(4)
120.4(4)
119.2(3)
120.4(4)
120.4(4)
121.1(4)
122.6(4)
C(2)—C(3)
C(3)—C(4)
C(4)—C(5)
C(5)—C(6)
C(6)—C(7)
C(7)—C(8)
C(8)—C(9)
C(8)—C(11)
C(19)—C(20)
C(3)-C(4)-C(5)
C(6)-C(5)-C(4)
C(5)-C(6)-C(7)
C(2)-C(7)-C(6)
C(2)-C(7)-C(8)
C(6)-C(7)-C(8)
C(9)-C(8)-C(11)
C(9)-C(8)-C(7)
C(11)-C(8)-C(7)
C(10)-C(9)-C(8)
N(1)-C(10)-C(9)
N(1)-C(10)-C(9)
1.406(6)
1.355(6)
1.402(7)
1.367(6)
1.414(5)
1.448(5)
1.386(5)
1.436(5)
1.460(7)
120.4(4)
120.1(4)
121.4(4)
117.3(4)
120.3(3)
122.4(4)
122.9(4)
115.9(3)
121.3(3)
121.9(4)
122.3(4)
122.3(4)
C(9)—C(10)
C(11)—C(123)
C(12)—C(13)
C(13)—C(18)
C(13)—C(14)
C(14)—C(15)
C(15)—C(16)
C(16)—C(17)
1.372(6)
1.352(5)
1.442(5)
1.395(5)
1.403(5)
1.362(6)
1.408(6)
1.405(6)
C(19)-C(20)-O(1)
C(12)-C(11)-C(8)
C(11)-C(12)-C(13)
C(18)-C(13)-C(14)
C(18)-C(13)-C(12)
C(14)-C(13)- C(12)
C(15)-C(14)-C(13)
C(14)-C(15)-C(16)
N(2)-C(16)-C(17)
N(2)-C(16)-C(15)
C(17)-C(16)-C(15)
C(20)-C(19)-N(2)
105.4(4)
124.9(4)
126.4(4)
115.9(4)
123.6(3)
120.4(4)
122.0(4)
122.3(4)
120.5(4)
123.5(4)
116.0(4)
109.9(4)
trans-4-[4′-(N-Hydroxyethyl-N-ethyl) amino)styryl]-N-methyl
pyridinium iodide (HEASPI): Purified by column chroma-
tography on silica gel with ethanol as the eluent. Yield 85%
and Td = 286.7°C. Element analysis: Anal. calcd.: C 52.69,
H 5.69, N 6.83; found: C 52.43, H 5.21, N 6.45.
(trans-4-[(4 ′-N,N-diphenyl amino) styryl]-N-methyl quino-
linium iodide) was filtered then purified by column chroma-
tography on silica gel using acetone–ethanol (1:1) as the
eluent. Needle deep-rose crystals were obtained after evapo-
ration with a yield of 55% and the decomposition tempera-
ture (Td) = 254.87°C. Element analysis: Anal. calcd.: C
66.67, H 4.66, N 5.18; found: C 66.03, H 4.59, N 5.12.
trans-4-[(4 ′-N-Hydroxyethyl-N-methyl amino) styryl]-N ′-
methyl quinolinium iodide (HMASQI) was synthesized by
the same method. Purified by column chromatography on
silica gel using acetonitrile–water (1:1) as the eluent. Strip-
like purple crystals were obtained after evaporation. Yield
60% and Td = 278.11°C. Element analysis: Anal. calcd.: C
56.51, H 5.19, N 6.28; found: C 56.48, H 5.13, N 6.38. (See
Table 3 for bond lengths and angles).
5-(4-N-Hydroxyethyl-N-methylamino tyryl) barbituric acid
(HMASBA): A mixture of equal molar 4-(N-hydroxyethyl-N-
methyl amino) benzaldehyde (3.58 g, 20 mmol) and barbitu-
ric acid (2.56 g, 20 mmol) was refluxed in the presence of
Et3N (3.2 g, 32 mmol) in water–ethanol (15:15 mL) until a
clear solution was obtained. The reaction was kept at reflux
for an additional 2 h, then the solvent was removed by evap-
oration. After 2–3 days the resulting precipitate was washed
several times with water and acetone, then dried. Red
microcrystals with a yield of 80% and Td = 245.49°C were
obtained. MS m/z: 289 (M+), 258, 120. 1H NMR δ: 10.73 (d,
J = 11.46 Hz, 2H), 8.24 (d, J = 8.6 Hz, 2H), 7.98 (s, 1H),
6.68 (q, J = 8.6 Hz, 2H), 4.71 (1H, s), 3.53 (4H, s), 3.07
(3H, s). Element analysis: Anal. calcd.: C 58.12, H 5.23, N
14.53; found: C 57.64, H 4.85, N 14.38.
Using the same method mentioned above, we synthesized
the chromophores of pyridinium derivatives.
trans-4-[4 ′-(N,N-Diphenylamino) styryl]-N-methyl pyridin-
ium iodide (DPASPI): The crude product was purified by
column chromatography on silica gel using ethane–chloro-
form (1:1) as the eluent. Yield 82% and Td = 260.2°C. Ele-
ment analysis: Anal. calcd.: C 63.68, H 4.73, N 5.71; found:
C 63.63, H 4.85, N 5.67. See Table 4 for bond lengths and
angles.
Crystal structure determination
Crystal data and other information on structure determina-
tion are given in Tables 2–4. X-ray diffraction data were
made on a Bruker P4 four-cycle diffractometer and graphite-
monochromated Mo Kα radiation was used (λ = 0.71073 Å).
The structures were solved by direct method and refined by
full-matrix least-squares on F2. The refinement was on F2 of
reflections with F0 4σ (F0) to minimize Σw (*F0* – *Fc*)2.
Anisotropic displacement parameters were refined for all
nonhydrogen atoms, and isotropic hydrogen atoms were
constrained to ride on their parent carbon atoms with fixed
bond lengths and idealized bond angles.
trans-4-(4 ′-Pyrrolidinyl styryl)-N-methyl pyridinium iodide
(PSPI): The crude product was purified by column chroma-
tography on silica gel with ethanol as the eluent. Yield 82%
and Td = 260.6°C. Element analysis: Anal. calcd.: C 55.11,
H 5.35, N 7.14; found: C 55.0, H 4.84, N 7.59.
trans-4-[4 ′-(N,N-Diethylamino) styryl]-N-methyl pyridinium
iodide (DEASPI): Purified by column chromatography on
silica gel with ethanol as the eluent. Yield 84% and Td =
264.7°C. Element analysis: Anal. calcd.: C 54.84, H 5.88, N
7.10; found: C 54.34, H 5.46, N 6.98.
Linear optical property
The λmax values for all chromophores in DMF with d0 =
1.00 × 10–5 M have been measured with a Hitachi U-3500
© 2001 NRC Canada

178
Can. J. Chem. Vol. 79, 2001
Table 4. Selected bond lengths (Å) and angles (°) for DPASPI.
C(1)—N(1)
C(2)—N(1)
C(6)—N(1)
N(2)—C(12)
N(2)—C(15)
N(2)—C(21)
C(2)—C(3)
C(3)—C(4)
C(4)—C(5)
1.481(16)
1.355(13)
1.320(15)
1.424(13)
1.416(13)
1.416(13)
1.359(16)
1.353(17)
1.412(19)
1.524(17)
1.388(17)
117.6(9)
123.4(9)
118.8(8)
119.9(8)
116.8(8)
122.6(8)
121.6(8)
C(7)—C(8)
C(8)—C(9)
C(9)—C(14)
C(9)—C(10)
C(10)—C(11)
C(11)—C(12)
C(12)—C(13)
C(13)—C(14)
1.277(17)
1.427(16)
1.372(16)
1.431(16)
1.359(16)
1.381(15)
1.393(15)
1.375(16)
1.358(15)
1.425(15)
1.377(16)
126.2(10)
118.2(10)
115.6(10)
120.9(11)
122.0(10)
123.5(9)
C(17)—C(18)
1.360(17)
1.379(16)
1.362(16)
1.393(15)
1.387(60)
1.372(16)
1.404(19)
1.341(18)
1.359(16)
C(18)—C(19)
C(19)—C(20)
C(21)—C(22)
C(21)—C(26)
C(22)—C(23)
C(23)—C(24)
C(24)—C(25)
C(25)—C(26)
C(15)—C(16)
C(15)—C(20)
C(16)—C(17)
C(4)—C(7)
C(5)—C(6)
C(1)-N(1)-C(2)
C(1)-N(1)-C(6)
C(2)-N(1)-C(6)
C(12)-N(2)-C(15)
C(12)-N(2)-C(21)
C(15)-N(2)-C(21)
N(1)-C(2)-C(3)
C(2)-C(3)-C(4)
C(3)-C(4)-C(5)
C(3)-C(4)-C(7)
C(5)-C(4)-C(7)
C(4)-C(5)-C(6)
N(1)-C(6)-C(5)
C(4)-C(7)-C(8)
C(8)-C(9)-C(10)
C(8)-C(9)-C(14)
C(10)-C(9)-C(14)
C(9)-C(10)-C(11)
C(10)-C(11)-C(12)
N(2)-C(12)-C(11)
N(2)-C(12)-C(13)
C(11)-C(12)-C(13)
C(12)-C(13)-C(14)
C(9)-C(14)-C(13)
N(2)-C(15)-C(16)
N(2)-C(15)-C(20)
C(16)-C(15)-C(20)
C(7)-C(8)-C(9)
C(17)-C(18)-(19)
C(18)-C(19)-(20)
C(15)-C(20)-(19)
N(2)-C(21)-(22)
N(2)-C(21)-(26)
C(22)-C(21)-(26)
C(21)-C(22)-(23)
C(22)-C(23)-(24)
C(23)-C(24)-(25)
C(24)-C(25)-(26)
C(21)-C(26)-(25)
C(16)-C(17)-(18)
C(15)-C(16)-(17)
118.7(11)
122.8(11)
117.5(10)
121.1(10)
121.3(9)
117.5(10)
121.9(11)
117.8(11)
120.9(12)
120.8(11)
121.1(10)
120.4(10)
121.3(10)
118.5(9)
121.9(11)
116.3(11)
118.7(11)
125.0(11)
119.8(11)
121.6(11)
124.5(12)
118.1(10)
119.6(10)
123.7(11)
121.7(9)
119.0(9)
119.2(10)
123.7(12)
Thus, once an open-aperture Z-Scan (S = 1) is performed,
the TPA coefficient β can be unambiguously deduced and if
the concentration d0 (in units of mol L–1) of the solution is
known, the molecular TPA cross section δ (in units of cm4 s
per photon) can be determined by using the following rela-
tionship,
Linear optical property
The λmax values for all chromophores in DMF with d0 =
1.00 × 10–5 M have been measured with a Hitachi U-3500 re-
cording spectrophotometer from quartz curettes of 1 cm path.
Measurement of the two-photon absorption cross
section (δ value)
The Z-scan system with an open aperture (S = 1), which is
insensitive to the nonlinear refraction, can be used to mea-
sure the nonlinear absorption cross-section. Such Z-scan
traces with open apertures are expected to be symmetrical
with respect to the focus (Z = 0), where they have a mini-
mum transmittance (e.g., multiphoton absorption) or a maxi-
mum transmittance (e.g., saturation of absorption). In fact,
the nonlinear absorption coefficient (β) that is due to TPA
can be easily calculated from the transmittance curves.
For a temporally Gaussian pulse, the normalized energy
transmittance can be given as (19),
[5]
hνβ = δN0 =
hνδNAd0 × 10–3
where hν is the energy of the incident photon. NA is Avoga-
dro’s constant.
We have accomplished TPA cross-section measurement
for all chromophores with the sample thickness of 2 mm
and a relatively high concentration of d0 = 0.05 M, using a
focussed 35 ps, 1064 nm mode-locked Nd:YAG laser.
Using the above equations, we can get the TPA coefficient (
β) and TPA δ value, which are presented in Table 1. The fi-
nal measured results of all samples are given in Table 1
with an experimental uncertainty of ±15%. The typical
open aperture Z-scan curves are shown in Fig. 1, which de-
picts the open-aperture trace at peak irradiance I0 = 3.64
GW cm–2.
+ ∞
2
1
[3]
T(z, S = 1) =
ln (1 + q0)(z,0)e− τ dτ
∫
− ∞
π q0(z,0)
where q0 (z,t) = β I0 Leff /(1 + z2/z 02). When there is no linear
absorption, Leff = L. L is the sample thickness, z0 and ± z are
the position of focus point and the range of scanning, re-
spectively, and I0 is the peak irradiance.
For | q0 | 1, the transmittance (T) can be expressed in terms
of the peak irradiance in a summation form more suitable for
numerical evaluation,
Calculation of the structures
Charge density distribution in both ground and excited
states, the difference between the ground and excited state
dipole moment (∆µge), the transition dipole moment between
two lowest states (M ee′), and the dihedral angle between two
rings of the backbone (θd/0 ) were all calculated with the
PM3 and ZINDO semiempirical methods.
∞
[−q0(z,0)]m
(m +1)3/ 2
βI0L /(1 + z2 /z 20)
∑
(m +1)3/ 2
m
[4]
T(z, S = 1) =
=
∑
m=0
© 2001 NRC Canada

Wang et al
179
Fig. 1. Normalized Z-scan transmittance of: (a) DPASQI and
(b) HMASQI, measured using ps pulses at λ = 1064 nm with
I0 = 3.64 GW cm–2.
Fig. 3. Packing diagram of HMASQI.
Fig. 4. Molecular structure of DPASPI.
molecule is overlapped with the acceptor-end of the other
molecule along the a-axis. The anions (I–) fill in the space
among the cations.
Fig. 2. Molecular structure of HMASQI.
The structures of DPASPI are shown in Figs. 4 and 5.
This crystal belongs to the orthorhombic crystal system of
centrosymmetric space group Pbca with eight molecules in
the unit cell. The two phenyl rings of the backbone are al-
most in the identical plane with the dihedral angle of 3.6°,
while the other two phenyl rings on the donor-end deviate
from this plane. From the packing diagram of DPASPI
(shown in Fig. 5), one can see that the donor groups overlap,
or stagger, each other while the acceptor-ends align face to
face with the definite degree. Along the c-axis, the donor-
and acceptor-groups present chain-like channels among the
space filling in the anions.
Two-photon absorption and intramolecular charge
transfer
monoclinic crystal system of the centrosymmetric space group
P21/n with the unit cell parameters of a = 10.0512(11) Å,
b = 16.2879(12) Å, c = 12.3328(8) Å, β = 102.307(6)°. Bond
lengths of C8—11, C11—12, and C12—13 are 1.436(5),
1.352(5), and 1.442(5) Å, respectively owing to strong π-
electron conjugation between the substituted phenyl ring and
quinolinium ring. Bond angles of pC6-7-8, pC2-7-8, pN1-
C2-C3, and pN1-C2-C7 are 122.4(4)°, 120.3(3)°, 120.4(4)o,
and 119.2(3)°, respectively, which shows that the
quinolinium ring possesses excellent planarity. The packing
diagram shows that HMASQI molecules stack to form a
layer-like structure along the b-axis. The donor-end of one
Linear absorption for all chromophores is shown in Fig. 6.
The influences from the quartz liquid cell and the solvent
DMF have been subtracted. One can see that maximum ab-
sorption bands are 463 nm for the barbituric acid derivative
and 458 ~ 482 nm for the pyridinium derivatives. Upon re-
placing the acceptor group with a larger conjugated frag-
ment (N-methyl-quinolinium) one can observe
a
bathochromic shift of the long-wave absorption band to 514
~ 554 nm. This suggests that the electronic interactions be-
tween the donor and acceptor groups induce the spectral
shift to lower energy in the order of: quilidinium derivatives >
pyridinium derivatives > barbituric acid derivatives.
© 2001 NRC Canada

180
Can. J. Chem. Vol. 79, 2001
Fig. 5. Packing diagram of DPASPI.
Fig. 6. Vis-IR absorption spectra of all chromophores in DMF
with d0= 1.00 × 10–5 M.
It must be noted that there is no linear absorption for all
chromophores in the region from ~700 to ~1300 nm. So
pumped by a 1064 nm laser pulse, the linear absorption can
be excluded because two-photon excitation via virtual inter-
mediate state can proceed, consistent with the fact that the
nonlinear absorption occurs at a frequency in the region of
one-photon transparency. It should be stressed that valid
structure–property relationships can only be ascertained for
those molecules that have their two-photon absorption peak
at approximately the same wavelength. If the two-photon ab-
sorption peak shifts greatly from the wavelength that is away
from the experimental wavelength of 1064 nm, then the
comparison of absolute cross-sections is misleading. From
Fig. 6, one can see that the maximum absorption wave-
lengths of all chromophores do not deviate too much. It is
for this reason that the structure–property relationships de-
duced from the experimental data in this work are valid at
1064 nm.
Two-photon absorption cross-sections measured in DMF
for the newly synthesized chromophores are summarized in
Table 1. From Table 1, one can see that when the acceptor
is fixed to the N-methyl pyridinium moiety, while the donor
strength decreases, the δ values increase from 5.7 × 10–48 cm4 s
per photon for PSPI, to 7.0 × 10–48 cm4 s per photon for
HEASPI, to 8.8 × 10–48 cm4 s per photon for DPASPI. This
suggests that the δ value is not always proportional to the do-
nor strength. Similar results can be observed in quinolinium
moiety. When acceptor is fixed to N-methyl quinolinium
moiety, with the decrease in donor strength, the two-photon
absorption cross-section values change from 8.8 × 10–48 cm4 s
per photon for HMASQI to 10.9 × 10–48 cm4 s per photon
for DPASQI.
Replacing barbituric acid by a pyridinium or a quinolinium
fragment, greatly decreases the dihedral angle between two
aromatic rings so as to be negligible. Consequently, a rela-
tively large value for the cross-section was obtained.
In a comparison of these newly synthesized compounds
with some other chromophores recently reported (13, 20),
we found that the δ value of 10.9 × 10–48 cm4 s per photon
for DPASQI is remarkable considering the limited conjuga-
tion length. For example, the TPA δ value for DPASQI is
nearly same as that of 6-(4-pyridyl)-N,N-diphenyl-2-
naphthylamine (δ = 12.9 × 10–48 cm4 s per photon) (13), but
almost twice that of the nitro-containing fluorescent deriva-
tive (δ = 6.5 × 10–48 cm4 s per photon) (20). The difference
of their TPA cross-section resulting from the variation of
both the acceptor group and the character of the conjugated
bridge shows that the quinolinium ring is a good acceptor
group and that the TPA δ value can be significantly enhanced
if a better planar moiety (as a conjugated bridge) is inserted
in a donor-quinolinium ring pair. Since the quinolinium ring
possesses better planarity, as confirmed by the crystallo-
graphic data of HEASQI mentioned above, this characteris-
tic has to contribute to the intramolecular charge transfer
that can be measured by such structural parameters as ∆µge
and Mee′ or the change of charge distribution on each atom
between the ground and the excited state.
The structural parameters calculated by the PM3 and
ZINDO methods are presented in Table 1. From this table,
one can conclude that the chromophores with large ∆µge and
M ee′ values possess relatively high δ value. For example,
M ee′ values in pyridinium derivatives change in the order of
DPASP+ > HEASP+ ~ PSP+ ~ DEASP+, while ∆µge values
change in the order of DPASP+ > HEASP+ ~ DEASP+ >>
PSP+. The measured δ values, with the order of DPASPI >
HEASPI ~ DEASPI > PSPI, confirmed that the experimental
results are in agreement with the calculation.
It is clear that with the acceptor structural factor changing
from barbituric acid to pyridinium and further to
quinolinium, δ values are gradually enhanced, in the range of
3.8 × 10–48 ~ 10.9 × 10–48 cm4 s per photon. The barbituric
acid derivative exhibits a relatively small cross-section,
which can be explained as the molecular planarity seems to
be critical in enhancing TPA cross-section values. Calcula-
tion shows that the dihedral angle between the benzene ring
and the barbituric acid (70° for HMASBA) distorts the pla-
narity and reduces the conjugated length, which is a disad-
vantage with respect to intramolecular charge transfer.
Charge density (e) distribution for part of the chromophore
cations in the ground state (shown in Fig. 7(a)), and in the
© 2001 NRC Canada

Wang et al
181
Fig. 7. Histogram of the π-charge distribution (e) in the:
(a) ground state and (b) in the excited state, for part of the
chromophores.
Conclusions
A series of new chromophores, styryl-parent end-capped
with various substituted amino groups as donors, and barbi-
turic acid, methyl-pyridinium, and methyl-quinolinium as
the acceptors, has been synthesized and characterized by ele-
ment analysis or X-ray diffraction.
We confirm both theoretically and experimentally that the
TPA cross-section is strongly influenced by intramolecular
charge transfer, which is characterized by the difference be-
tween the ground and excited state dipole moment (∆µge)
and the transition dipole moment between the excited states
(Mee′) or the change of charge distribution on each atom be-
tween ground and excited state. Based on above point of
view, the structure–property relationships of two-photon ab-
sorption (TPA) property can be interpreted reasonably,
which can provide some useful information for molecular
design with large TPA cross-sections.
The barbituric acid derivative possesses a relatively small
TPA cross-section value due to its being heavily distorted
from planarity and short conjugated length, which hinders
intramolecular charge transfer. In contrast to this,
quinolinium derivatives exhibit relatively large TPA cross-
section due to its excellent planarity.
Both the N,N-diphenylamino group and the quinolinium
ring are effective groups for enhancing TPA cross-sections,
although the N,N-diphenylamino group is not a strong donor
in comparison with the N,N-diakylamino group. Quantum
chemical calculation shows that when the N,N-
diphenylamino group and the quinolinium ring are matched,
a relatively large charge transfer is produced (∆µge = 12.1
debye), which confirms that TPA cross-section is not propor-
tional to electron-donating strength, but dominated by
intramolecular charge transfer.
excited state (shown in Fig. 7(b)), are plotted in histogram
form. One can see that in the ground state there is no alter-
nation of charge density distribution. For example, the
charge densities over D (donor) and atom C1 are both positive,
and the charge densities over A (acceptor) and atom C9 are
both positive too. Once in the excited state, the great charge
redistribution occurs, accompanying with the significant
charge positive–negative alternation, where the positive
charge increases on the donor-end (D), and in the meantime,
decreases on the acceptor-end (A). For example, one can see
that in the ground state, the charge distribution of the
Acknowledgments
This work was supported by a grant for State program of
China. The authors are grateful to Prof. Sun Zhenrong and
Prof. Ding Liangen in the Department of Physics, Normal
University of East China for their help to measure the two-
photon absorption cross-sections for all chromophores.
References
DPASP+ cation is in such a way: QD = +0.21 e and QA
+0.83 e; while in the excited state, QD = +0.93 e and QA
=
=
1. B.H. Cumpston, S.P. Ananthavel, and S. Barlow. Nature (Lon-
don), 3, 1041 (1999).
2. D.A. Parthenopoulos and P.M. Rentzepis. J. Appl. Phys. 68,
5814 (1990).
3. D.A. Parthenopoulos and P.M. Rentzepis. Science (Washing-
ton, D.C.), 245, 843 (1989).
4. J.E. Ehrlich, X.L. Wu, and I.Y.S. Lee. Opt. Lett. 22, 1843
(1997).
5. G.S. He, R. Gvishi, P.N. Prasad, and B.A. Reinhardt. Opt.
Commun. 117, 133 (1995).
6. G.S. He, R. Signorini, and P.N. Prasad. Appl. Opt. 37, 5720
(1998).
7. J.D. Bhawalkar, G.S. He, and P.N. Prasad. Rep. Prog. Phys.
59, 1052 (1996).
8. G.S. He, L. Yuan, and Y.P. Cui, M. Li, and P.N. Prasad. J.
Appl. Phys. 81, 2529 (1997).
+021 e, indicating a large charge redistribution on the two-
terminals occurs. It is interesting to note that such a
π-charge transfer degree is consistent with the ∆µge and Mee′
values and also in accordance with the enhancement of
δ value. So DPASPI exhibits relatively large δ values among
pyridinium derivatives because of its larger ∆µge (11.8
debye) and Mee′ (9.8 debye) values. HMASBA exhibits
smaller δ value because of its small change of dipole mo-
ment (0.8 debye). Its charge distribution is also small:
Qami = +0.04 e, Qbar = –0.27 e in the ground state, and Qami
= +0.25 e, Qbar = –0.28 e in the excited state, which shows
that only a weak charge transfer between the ground and the
excited state has occurred.
© 2001 NRC Canada

182
9. G.S. He, K.S. Kim, L.X. Yuan, N. Cheng, and P.N. Prasad.
Can. J. Chem. Vol. 79, 2001
15. C.W. Spangler. J. Mater. Chem. 9, 2013 (1999).
Appl. Phys. Lett. 71, 1619 (1997).
10. G.S. He, J.D. Bhawalkar, C.F, Zhao, C.K. Park, and P.N.
Prasad. Appl. Phys. Lett. 68, 3549 (1996).
16. M. Albota, D. Beljonne, J.L. Bras, J.E. Ehrlich, J.Y. Fu, A.A.
Heikal, and S.E. Hess. Science (Washington, D.C.), 281, 11
(1998).
11. C.F. Zhao, G.S. He, J.D. Bhawalkar, C.K. Park, and P.N.
Prasad. Chem. Mater. 7, 1979 (1995).
17. F. Meyers, S.R. Marder, B.M. Pierce, and J.L. Brédas. J. Am.
Chem. Soc. 116, 10 703 (1994).
12. C.F. Zhao, C.K. Park, and P.N. Prasad. Chem. Mater. 7, 1237
(1995).
13. B.A. Reinhardt, L.L. Brott, S.J. Clarson, A.G. Dillard, and J.C.
Bhatt. Chem. Mater. 10, 1863 (1998).
14. G.S. He, L.X. Yuan, N. Cheng, and P.N. Prasad. J. Opt. Soc.
Am. B, 14, 1079 (1997).
18. T. Kogej, D. Beljonne, F. Meyyers, J.W. Perry, S.R. Marder,
and J.L. Brédas. Chem. Phys. Lett. 298, 1 (1998).
19. S.B. Mansoor, A.A. Said, T.H. Wei, D.J. Hagan, and E.W.V.
Stryland. IEEE J. Quantum Electron. 26, 760 (1990).
20. K.D. Belfield, D.J. Hagan, E.W.V. Stryland, K.J. Schafer, and
R.A. Negres. Org. Lett. 1, 10 (1999).
© 2001 NRC Canada