First hyperpolarizabilities of dipolar, bis-dipolar, and octupolar molecules

Article abstract of DOI:10.1016/j.cplett.2004.12.112

A series of dipolar (1), bis-dipolar (2), and octupolar molecules (3) containing 1, 2, and 6 dipolar units within a molecule has been synthesized and their hyperpolarizabilities were analyzed. The β_HRS increases in the order, 1 < 2 < 3. The 'monomeric' β_zzz increases by approximately twofold from 1 to 2, whereas β_zzz of 2 and β_xxx of 3 are similar. Noteworthy is the parallel increase in the hyperpolarizability tensor with the λ_max.

Full text of DOI:10.1016/j.cplett.2004.12.112

Chemical Physics Letters 403 (2005) 68–71
www.elsevier.com/locate/cplett
First hyperpolarizabilities of dipolar, bis-dipolar, and
octupolar molecules
a
Si Kyung Yang , Hyun Cheol Ahn , Seung-Joon Jeon , Inge Asselberghs ,
a
a
b
b
a,
*
Koen Clays , Andre Persoons ,b, Bong Rae Cho
*
´
a
Molecular Opto-Electronics Laboratory, Department of Chemistry and Center for Electro- and Photo-Responsive Molecules, Korea University,
1-Anamdong, Seoul 136-701, Republic of Korea
b
Center for Research on Molecular and Photonics, Laboratory of Chemical and Biological Dynamics, University of Leuven, Celestijnenlaan 200D,
B-3001, Leuven, Belgium
Received 12 November 2004; in final form 22 December 2004
Available online 11 January 2005
Abstract
A series of dipolar (1), bis-dipolar (2), and octupolar molecules (3) containing 1, 2, and 6 dipolar units within a molecule has been
synthesized and their hyperpolarizabilities were analyzed. The bHRS increases in the order, 1 < 2 < 3. The ꢀmonomericꢁ bzzz increases
by approximately twofold from 1 to 2, whereas bzzz of 2 and bxxx of 3 are similar. Noteworthy is the parallel increase in the hyper-
polarizability tensor with the kmax
.
ꢀ 2005 Published by Elsevier B.V.
1. Introduction
lar tensor components [3–6]. It has been suggested that,
unlike the dipolar and octupolar molecules [7–12], the
multi-dipolar molecules combining several charge
transfer units could provide larger hyperpolarizabili-
ties, without classic bathochromic shift of the charge
transfer bands. Recently, a series of binaphthyl and
carbazole derivatives have been synthesized and their
hyperpolarizability tensors were analyzed by using a
vector model [5,6]. The results show that the off-diago-
nal tensor components of the ꢀdimersꢁ increased signif-
icantly. However, no enhancement in the ꢀmonomericꢁ
bzzz has been observed. Moreover, there is no report
regarding the tensor analysis for molecules containing
more than two dipolar units.
In this work, we have synthesized a series of dipolar,
bis-dipolar, and octupolar molecules containing 1, 2,
and 6 dipolar units within a molecule and analyzed their
hyperpolarizability tensors (1–3) (Chart 1). We were
interested in learning the effects of successively increas-
ing the dipolar units within a molecule on their linear
and non-linear optical properties.
Synthesis of organic non-linear optical materials
with large first hyperpolarizability is of considerable
interest because of their potential applications in vari-
ous photonic technologies [1]. Most organic NLO
materials are elongated push–pull p-conjugated system
in which the donor and acceptor groups are bound to
the opposite terminals. The first hyperpolarizability of
such molecules is usually dominated by tensor compo-
nent along the charge transfer direction bzzz [2]. For fu-
ture application of organic NLO materials in all-optic
devices, it is important to know the values of indepen-
dent and non-zero components of the molecular hyper-
polarizability. This information is useful in assessing
how well the molecules will be arranged in the solid
state to take advantage of the largest possible molecu-
*
Corresponding author. Fax: +82 2 3290 3121.
E-mail addresses: andre.persoons@fys.kuleuven.ac.be
(A. Persoons), chobr@korea.ac.kr (B.R. Cho).
0009-2614/$ - see front matter ꢀ 2005 Published by Elsevier B.V.
doi:10.1016/j.cplett.2004.12.112

S.K. Yang et al. / Chemical Physics Letters 403 (2005) 68–71
69
Chart 1
NC
CN
A and B. To prepare E, A and C were reacted to obtain
E0, which was then deprotectd. Compound 3 was ob-
tained in 49% yield by the condensation between D
and E. The structures of 2 and 3 were unambiguously
NBu2
NBu2
NC
NEt2
1
3
Bu2N
NC
NBu2
N
N
1
characterized by H and 13C NMR, IR, and elemental
CN
analysis. The NMR spectra of 3 show four singlets
due to the phenyl protons and two doublets for the viny-
lic protons and fifteen 13C resonance peaks, indicating a
symmetrical structure (see Supplementary materials).
Moreover, the similar chemical shifts for the vinylic pro-
tons of 2 and 3 reveal that there is no ring current effect
in the latter, probably because the dipolar units are con-
nected at the meta positions to interrupt the conjugation
[15].
2
CN
CN
Et2N
NEt2
Bu2N
NBu2
Chart 1.
2. Experimental
2.1. Synthesis
Compounds 1–3 have been synthesized either by the
literature method [12] or by Scheme 1 (see Supplemen-
tary materials for details).
3.2. One-photon absorption and fluorescence spectra
The absorption and fluorescence spectra for 1–3 are
displayed in Fig. 1. The kmax increases from 393 to 433
nm by the change in the structure from 1 to 2, as expected
for the extended conjugation. Interestingly, kmax for 2
and 3 are almost the same, although the latter shows
three absorption bands. This is again due to the inter-
rupted conjugation. However, none of the fluorescence
spectra have multiple peaks, which indicates that the
emission occurs from the lowest excited states with the
largest oscillator strength. The peak positions of the
absorption and fluorescence spectra are summarized in
Table 1. In contrast to the absorption spectra, the fluo-
rescence spectra exhibit systematic bathochromic shift
with increasing conjugation length. All of the com-
pounds show large Stokes shifts ranging from 5080
cmꢀ1 for 1 to 6700 cmꢀ1 for 3. As expected, the fluores-
cence Stokes shift increases as the conjugation length in-
creases. This means that the energy gap between the
ground and the excited states decreases monotonically
in the order, 1 > 2 > 3 (Table 1). As the conjugation
2.2. Determination of first hyperpolarizabilty (b)
The femtosecond HRS measurements were per-
formed in THF at 1300 nm by using Disperse Red 1
(DR1, b = 54 · 10ꢀ30 esu in CHCl3 at 1300 nm) as the
external reference [13,14].
3. Results and discussion
3.1. Synthesis
Compound 1 was prepared as described [12]. Synthe-
sis of 2 and 3 is shown in Scheme 1. A was prepared in
85% yield by the NBS bromination of 1,3-dicyano-
4,6-dimethylbenzene followed by phosphorylation.
Horner–Wittig condensation between A and p-diethyla-
minobenzaldehyde afforded 2 in 68% yield. B was pre-
pared by the alkylation of m-phenylenediamine with
P(OBu)3 followed by formylation. Mono-protection of
B with neopentyl glycol afforded C in 46% yield. D
was prepared in 42% yield by the condensation between
1.0
0.8
0.6
0.4
0.2
0.0
NC
(EtO)2(O)P
CN
P(O)(OEt)2
Bu2N
2,
NBu2
CHO
Bu2N
(b)
NBu2
O
(a)
OHC
OHC
B
C
A
O
Bu2N
NBu2
CN
CN
(c)
(d)
A
3
+
:
B
1
D
NC
CN
P(O)(OEt)2
(e)
(EtO)2(O)P
3
CHO
NBu2
CHO
Bu2N
NBu2
A
1
+
:
C
2
NBu2
NC
CN
E
400
600
800
Scheme 1. Reagents and conditions: (a) p-Et2NC6H4CHO, LDA/
THF, 0 ꢁC, 12 h; (b) Me2C(CH2OH)2, p-TsOH, 80 ꢁC, 1 h; (c) LDA/
THF, 0 ꢁC, 12 h; (d) i: LDA/THF, 0 ꢁC,12 h; ii: AcOH, 65 ꢁC, 18 h; (e)
LDA/THF, 0 ꢁC, 12 h.
Wave Length (nm)
Fig. 1. Normalized one-photon absorption and emission spectra for 1
(——), 2 (------), 3 (ꢁ ꢁ ꢁꢁ ꢁ ꢁ) in THF, respectively.

70
S.K. Yang et al. / Chemical Physics Letters 403 (2005) 68–71
Table 1
Linear and non-linear optical properties of 1–3
fb
DmST
sd,e
bf
bzzz
bzzz,static
a
c
g
h
kmax
1
2
3
393 (2.18)
433 (6.01)
429 (7.87)
0.44
1.4
5080
5430
6700
–
6
1
2
2
15
32
2
3
9
16
1
2
2.4
2.6
16
23
2.0
35 3i
18 2j
a
b
c
d
e
f
kmax of the one-photon absorption spectra in nm. The numbers in the parenthesis are the molar extinction coefficients (10ꢀ4e).
Oscillator strength estimated by the relationship: f = 4.319 · 10ꢀ9 Æ e Æ m1/2
Stokes shift in cmꢀ1 (1/kmax ꢀ 1/kfl).
.
Fluorescence life time in ns.
Standard deviation, 0.2 ns.
First hyperpolarizability measured at 1300 nm.
The bzzz of monomeric dipolar subunit except otherwise noted.
bzzz at zero frequency except otherwise noted.
The bxxx of monomeric dipolar subunit.
bxxx at zero frequency.
g
h
i
j
length increases, the charge transfer character of the
excited electronic state would increase. This would pre-
dict that the solvation energy of the excited electronic
state becomes large and the fluorescence Stokes shift
will increase too. This pattern is clearly observed
experimentally.
Compound 3 is a cyclic octupole containing 6 dipolar
units. It emits multiphoton fluorescence at 650 nm with
the fluorescence lifetime of (2.6 0.2) ns. The experimen-
tally measured bHRS is (23 2) · 10ꢀ30 esu, which is the
largest in this series of compounds. Assuming octupolar
D3h symmetry, then only the bxxx = ꢀbxyy = ꢀbyxy
=
ꢀbyyx are the non-zero hyperpolarizability tensor ele-
ments (molecule in the XY plane, Z-axis perpendicular
to the molecule), and bxxx = (35 3) · 10ꢀ30 esu. This
symmetry has been experimentally verified by depolariza-
tion measurements, i.e., q = 1.41 0.13, which is very
similar to 1.5 expected for pure octupolar symmetry [7].
The bxxx,static,3 calculated by using the three level model
is (18 2) · 10ꢀ30 esu (Table 1) [17].
3.3. First hyperpolarizability (b) of 1–3
The femtosecond HRS measurements were performed
in THF at 1300 nm by using Disperse Red 1 (DR1,
b = 54 · 10ꢀ30 esu in CHCl3 at 1300 nm) as the external
reference [13]. Small local field correction factor for the
field difference between CHCl3 and THF at the optical
frequencies has been taken into account. The b value
for 1 was (6 1) · 10ꢀ30 esu, irrespective of symmetry
assumptions. Assuming only a single major charge-trans-
fer along the molecular Z-axis, there is also a single major
The first hyperpolarizability of a dipolar molecule
can be expressed by the two-level model (Eq. (1)), where
l01 is the transition moment between the ground and
first excited CT state, l0 and l1 are the dipole moments
in the ground and the first excited CT state, x01 is the
CT energy and x is the energy of the incident laser light,
respectively [16]. The equation is identical to that of
three-level model of molecular non-linearity for the
octupoles if (l1 ꢀ l0) is substituted by l12 which is the
transition moment connecting the two degenerate ex-
cited states [17]
hyperpolarizability
b
tensor
element,
zzz,1 = (15 2) · 10ꢀ30 esu. With CT band with maxi-
mum absorption band at 393 nm and for measurement
at 1300 nm, the resonance enhancement factor is 1.73
[16]. Therefore, bzzz,static,1 = (9 1) · 10ꢀ30 esu (Table 1).
For compound 2, multiphoton fluorescence was ob-
served at 650 nm, in agreement with emission spectrum.
The fluorescence lifetime is estimated at (2.4 0.2) ns.
The value of b measured for 2 is (16 2) · 10ꢀ30, irre-
spective of symmetry assumptions. Assuming C2v sym-
metry with angle between two halves 120ꢁ, then the
relation between the only two non-zero tensor compo-
1
ꢀh2
l201ðl1 ꢀ l0Þ
x401
bzzz
¼
ꢂ
ꢂ
:
ð1Þ
x201
ðx021 ꢀ 4x2Þðx201 ꢀ x2Þ
Table 1 shows that b value increases monotonically with
the conjugation length in the order, 1 < 2 < 3. When the
chromophore structure is changed from 1 to 2, kmax in-
creases by 40 nm and the oscillator strength, which is
proportional to the square of the transition moment
ðl201Þ, increases by approximately threefold. Hence, the
nearly threefold increase in b from 1 to 2 is primarily
due to the smaller CT energy and larger transition mo-
ment. On the other hand, kmax remains nearly the same
and the oscillator strength increases significantly by the
change from 2 to 3. Although it is not possible to com-
nents (bzzz
b
b
,
b
zxx = bxzx = bxxz
zzz = (tanh/2)2 = 3 [6]. This assumption also results in
zzz = 8 · 10ꢀ30 esu, bzxx = bxzx = bxxz = 24 · 10ꢀ30 esu
)
becomes, bzxx/
for the molecular tensor components for the ꢀdimericꢁ
molecule 2. In addition, the bzzz for the monomeric part
is calculated to be 32 · 10ꢀ30 esu from the relationships
b
zzz = 2cos3 (h/2)bzzz and bzxx = bxzx = bxxz = 2 cos(h/
2)sin2(h/2)bzzz [6]. The value of bzzz,static,2 calculated as
above is (16 1) · 10ꢀ30 esu (Table 1).

S.K. Yang et al. / Chemical Physics Letters 403 (2005) 68–71
71
pare (l1 ꢀ l0) of 2 and l12 of 3 with existing data, the
Acknowledgments
significant increase in the b from 2 to 3 can be attributed
to the larger transition moment (Eq. (1)). Note that sig-
nificant enhancement of b is realized without bathochro-
mic shift by combining 6 dipolar units within a
octupolar molecule.
This work was supported by NRL-MOST and CRM-
KOSEF. S.K.Y. and H.C.A. were supported by BK21
program.
Comparison of the hyperpolarizability tensors reveals
that the ꢀmonomericꢁ bzzz,static,2 = 16 · 10ꢀ30 esu for a
single ꢀhalf contributionꢁ of 2 is approximately twice
the ꢀexpectedꢁ value from 1 (Table 1). In sharp contrast,
when two 2-methoxy-6-nitronaphthalene ꢀmonomersꢁ
were combined into a binaphthyl unit without sharing
the conjugation path, the ꢀmonomericꢁ bzzz derived from
the addition model was found to be identical to the ꢀdi-
mericꢁ molecule [5]. Similarly, the value of ꢀmonomericꢁ
bzzz for 3,6-dinitro-9-ethylcarbazole (DNEK) was smal-
ler than that of 3-dinitro-9-ethylcarbazole (NEK) [6]. It
should be noted that the kmax of the monomers and di-
mers are identical in both series of compounds [5,6]. On
the other hand, the kmax shows a large bathochromic
shift from 1 to 2 (vide supra). Hence, the large increase
in the bzzz values for the ꢀmonomericꢁ part observed in 2
can readily be attributed to the more efficient intramo-
lecular charge-transfer (ICT). Consistently, the ꢀmono-
mericꢁ bzzz for 2 and bxxx for 3 are similar,probably
because of the similar extent of ICT. This result suggests
an intersting possibility that multi-polar molecules with
large ꢀmonomericꢁ bzzz could be designed by increasing
the charge transfer character.
Appendix A. Supplementary materials
Supplementary data associated with this article can
be found, in the online version, at doi:10.1016/
j.cplett.2004.12.112.
References
[1] L.R. Dalton, W.H. Steier, B.H. Robinson, C. Zhang, A. Ren, S.
Garner, A. Chen, T. Londergan, L. Irwin, B. Carlson, L. Fifield,
G. Phelan, C. Kincaid, J. Amend, A. Jen, J. Mater. Chem. 9
(1999) 1905.
[2] L.T. Cheng, W. Tam, S.R. Marder, A.E. Stiegman, G. Rikken,
C.W. Spangler, J. Phys. Chem. 95 (1991) 10643.
[3] J.L. Oudar, J. Zyss, Phys. Rev. A 26 (1982) 2016.
[4] J. Wolff, D. La¨ngle, D. Hilllenbrand, R. Wortmann, R. Matsch-
iner, C. Glania, P. Kra¨mer, Adv. Mater. 9 (1997) 138.
[5] E. Henderickx, C. Boutton, K. Clays, A. Persoons, S. Van Es, T.
Biemans, B. Meijer, Chem. Phys. Lett. 270 (1997) 241.
[6] C. Boutton, K. Clays, A. Persoons, T. Wada, H. Sasabe, Chem.
Phys. Lett. 286 (1998) 241.
[7] S. Brasselet, J. Zyss, J. Opt. Soc. Am. B 15 (1998) 257.
[8] J.J. Wolff, R. Wortman, Adv. Phys. Org. Chem. 32 (1999) 121.
[9] J.J. Wolff, F. Siegler, R. Matschiner, R. Wortman, Angew. Chem.
Int. Ed. 39 (2000) 1436.
[10] C. Lambert, W. Gaschler, G. No¨ll, M. Weber, E. Scha¨mlzlin, C.
Bra¨uchle, K. Meerholz, J. Chem. Soc., Perkin Trans. 2 (2001) 964.
[11] B.R. Cho, S.B. Park, S.J. Lee, K.H. Son, S.H. Lee, M.-J. Lee, J.
Yoo, Y.K. Lee, G.J. Lee, T.I. Kang, M. Cho, S.-J. Jeon, J. Am.
Chem. Soc. 123 (2001) 6421.
4. Conclusion
In conclusion, we have synthesized a series of dipolar,
bis-dipolar, and octupolar molecules containing 1, 2,
and 6 dipolar units. The bHRS increases in the order,
1 < 2 < 3. Interestingly, the bHRS increases from 2 to 3
with a minimal red-shift in the absorption spectrum.
On the other hand, the ꢀmonomericꢁ bzzz (or bxxx) in-
creases by approximately twofold as the molecular
structure changes from 1 to 2, but remains nearly the
same for a further change to 3. Noteworthy is the paral-
lel increase between bzzz (or bxxx) and ICT.
[12] M.J. Lee, M. Piao, S.H. Lee, M.-J. Jeong, K.M. Kang, S.-J.
Heon, B.R. Cho, J. Mater. Chem. 13 (2003) 1030.
[13] G. Olbrechts, R. Strobbe, K. Clays, A. Persoons, Rev. Sci.
Instrum. 69 (1998) 2233.
[14] G. Zhang, S. Li, Y. Jiang, J. Phys. Chem. A 107 (2003) 5573.
[15] K. Mullen, H. Unterberg, W. Huber, O. Wennerstro¨m, U.
¨
Norinder, D. Tanner, B. Thulin, J. Am. Chem. Soc. 106 (1984)
7514.
[16] J.L. Oudar, D.S. Chemla, J. Chem. Phys. 66 (1977) 2664.
[17] M. Joffre, D. Yaron, R.J. Silbey, J. Zyss, J. Chem. Phys. 97 (1992)
5607.