February 15, 1998 / Vol. 23, No.

4 / OPTICS LETTERS 253

Optically induced space-charge fields, dc voltage,

and extraordinarily large
nonlinearity in dye-doped nematic liquid crystals
I. C. Khoo, S. Slussarenko,* B. D. Guenther, Min-Yi Shih, P. Chen, and W. V. Wood
Department of Electrical Engineering, Pennsylvania State University, University Park, Pennsylvania 16802

Received September 25, 1997

We have observed extraordinarily large optical nonlinearity in Methyl Red – doped nematic liquid-crystal film.
Grating diffraction can be generated with an optical intensity as low as 40 mWycm2 , and a refractive-index
change coefficient of more than 6 cm2 yW is obtained. The effect is attributed to formation of an optically
induced dc space-charge field and to the resulting reorientation of the highly birefringent nematic director
axis.  1998 Optical Society of America
OCIS codes: 010.1080, 190.0190, 160.5320, 160.3710, 100.0100.

Nematic liquid-crystal films containing photoexcitable torques, and cis–trans conformational processes1 – 3
dopants exhibit numerous interesting nonlinear optical as a contributing mechanism to the observed effect.
phenomena.1 – 6 In particular, studies1 – 3 have shown Furthermore, the dynamics of these reorientation
that the photoexcited dopant dye molecules can exert processes are also measured to be of the milliseconds-
intermolecular torques to realign the liquid-crystalline and-longer time scale that is typical of director axis
director axis. Other studies 4 – 6 have reported the ob- reorientation rather than of a thermal effect, as the
servation of optically induced dc space-charge f ields latter is characterized by time constants in the submil-
that, in combination with a bias dc field, can cre- lisecond regime for such thermal diffusion lengths.7
ate a director axis reorientation and produce a large With reference to Fig. 1, if the two pumps and the
refractive-index change. However, the necessity of us- probe’s polarizations are in the x direction, no diffrac-
ing dc bias voltages close to the threshold for dy- tion is observed. If the probe polarization is oriented
namic scattering is an undesirable feature, especially along the y axis, the diffraction is maximum. When
for practical applications. both pump beam polarizations are along the y di-
Here we report observation of an extraordinarily rection, self-diffraction from these beams is observed,
large purely optically induced refractive-index change whereas the x-polarized probe beam yields no diffrac-
in Methyl Red dye– doped nematic liquid-crystal film. tion. In this case the self-diffraction effect becomes
The liquid crystal used is pentylcyanobiphenyl (5CB) more pronounced if the sample is tilted around the
doped with traces of the laser dye Methyl Red (Aldrich x axis such that the incident beams make an oblique
Chemicals) at concentrations ranging from 0.5% to incident angle with respect to the normal to the cell
1%. Homeotropic alignment is achieved by treatment window. These polarization studies show that the ex-
of the indium tin oxide– coated cell windows with the traordinary ray’s refractive-index change is caused by
surfactant hexadecyl trimethyl ammonium bromide the reorientation of the director axis in the y z plane.
(HTAB). The absorption constant b of these f ilms
is 194 cm21 at the l ­ 488 nm argon-laser line. The
transmission is 85% and 60% for f ilm thicknesses
of 6 and 25 mm, respectively. A linearly polarized
argon laser is divided into two equal-power writing
beams; the beam diameter is 1.5 cm. These beams,
propagating in the y z plane, are normally incident
to the cell window and overlap at a small angle a,
as shown in Fig. 1. This configuration imparts a
sinusoidal optical intensity function I ­ 2I0 s1 1 cos qyd
along the y axis, where I0 is the intensity of one writing
beam. A linearly polarized 5-mW He–Ne laser is used
to probe the grating.
In general, easily visible grating diffractions are
generated with total input power as low as 100 mW
(intensity 2I0 of 40 mWycm2 ). These intensities
are ,2 orders of magnitude smaller than those em-
ployed in the research reported Refs. 1–6. This,
as well as the polarization dependence and the Fig. 1. Dependence of probe diffraction efficiency on total
observation of photovoltage and other dc–ac bias writing-beam intensity. Sample thickness, 6 mm; a ­
field dependence as described in what follows, 0.04 rad; probe beam angle b ­ 0.1 rad. Inset, schematic
has also ruled out thermal effects, intermolecular of the optical wave mixing geometry.

0146-9592/98/040253-03$10.00/0  1998 Optical Society of America

254 OPTICS LETTERS / Vol. 23, No. 4 / February 15, 1998

An estimate of the refractive-index change coeff i- the light is reversed. Figure 3 shows that the photo-
cient n2 can be obtained from Fig. 1 for probe diffrac- voltage rises rapidly as the input intensity is increased,
tion efficiency versus writing beam intensity for the and it reaches a saturated value of ,12 mV for intensi-
case when the pump beams are polarized in the x di- ties above 2 mWycm2 .
rection and the probe polarization is along the y axis. Similar photoinduced voltage pulses were observed
Because the condition lg 2 .. dl holds, the diffrac- previously10,11 in Methyl Red– doped p-methoxylben-
tion is in the Raman–Nath regime. The f irst-order zyladine-p0 -butyl analine films sandwiched between
diffraction eff iciency h is thus given by h ­ J1 2 (phase SnO2 -coated Pyrex glass plates and also in 5CB films.
shift) ,spDndyld2 . Using the values h ­ 0.05 at a There the photovoltaic effect was attributed to pho-
pump beam intensity of 1.3 mWycm2 from Fig. 1, we tocharge generation in the Methyl Red – doped liquid
obtain Dn ­ 7.45 3 1023 and a magnitude of n2 ­ crystal and subsequent diffusion and redistribution of
6 cm2 yW. Similar values of n2 are obtained for the the ions with different mobilities; exchanges of ions be-
25-mm-thick samples and also at other argon-laser tween the HTAB-coated indium tin oxide electrodes
lines within the absorption band of Methyl Red dye. and the photocharges producing doped nematic film
We attribute the director axis reorientation effect could also be involved.
to the optically induced space-charge f ields, similar Because of the dielectric and conductivity anisot-
to the effect reported in Refs. 4 – 6. One source of ropies and the spatial variation of the director axis
space-charge field is the photorefractivelike compo- orientation, calculating the space-charge distribution
nent that results from generation of photoions from and the resultant space-charge fields and director axis
excited dye molecules4,8 and subsequent migration and reorientation is a rather complex multidimensional
diffusion of these charged ions. The other two sources electrodynamics problem that is clearly outside the
arise from the dielectric and conductivity anisot-
ropies, the so-called Carr –Helfrich effect,6,9 which re-
quire a dc field in the z direction. As we show below,
we observe an internally generated dc voltage along z,
so these space-charge f ields could also come into play.
Previous theory and experiments6 have shown that
under the action of these space-charge fields the re-
orientation process is optimal if the grating spacing is
of the order of twice the f ilm thickness. This is again
confirmed in our experimental results (see Fig. 2).
It is instructive to compare the observed index
change coefficient of the Methyl Red– doped f ilm with
that of f ilm doped with Fullerene sC60 d, although we
have to bear in mind that the latter requires an ex-
ternally applied voltage. From Ref. 5, the n2 value for
a 25-mm-thick C60 -doped film is of the order of 2.3 3
1023 sDn , 1.4 3 1023 ; 2I0 , 0.6 Wycm2 d. The maxi-
mum solubility of C60 in 5CB corresponds to a concen- Fig. 2. Dependence of probe diffraction eff iciency on grat-
tration of 0.05% by weight, and the absorption constant ing spacing for the 6-mm-thick sample. Writing beam in-
b was measured to be ,5 cm21 , so n2 yb , 0.46 3 1023 . tensity, 3.5 mWycm2 .
On the other hand, the Methyl Red–doped film of
the same thickness gives n2 yb , 6y194 , 3.1 3 1022 .
As the exact nature of this newly observed effect re-
mains to be determined, at present we can only of-
fer the heuristic explanation that the much larger
nonlinearity of Methyl Red– doped f ilm is due simply
to the greater solubility of Methyl Red dye in 5CB
as well as to its higher photocharge production eff i-
ciency, which is ref lected in the following photovoltage
measurement.
Charge migration and diffusion from the grating
maximum to the minimum produce a spatially vary-
ing (grating function) space-charge f ield in the y di-
rection. Because of absorption, the optical intensity
varies monotonically along the z direction, and thus
a (unidirectional) dc space-charge field is also estab-
lished along z. The latter component is manifested
in the form of a dc voltage drop across the entrance
and exit indium tin oxide cell electrodes. The inset in Fig. 3. Observed photovoltage dependence on the optical
Fig. 3 shows an oscilloscope trace of the dc voltage de- illumination intensity in a 25-mm-thick nematic film. In-
tected across a 6-mm-thick f ilm. The polarity of the set, typical oscilloscope trace of the detected photovoltage in
observed voltage is reversed if the incident direction of a 6-mm-thick film under an optical intensity of 1 mWycm2 .
 February 15, 1998 / Vol. 23, No. 4 / OPTICS LETTERS
Fig. 4. Diffraction efficiency versus applied dc voltage in
a 25-mm-thick film; writing beam intensity, 1.5 mWycm2 .
Fig. 5. Observed grating writing times as a function of the
total writing beam intensity in a 6-mm-thick nematic film.
The buildup time for the photovoltage behaves analogously.
scope of this Letter. It suff ices to note that, because of
this multidimensional distribution of the induced space
charges, the induced index grating is not well def ined
in terms of its shift with respect to the incident optical
intensity function, unlike the case of dye- or C60 -doped
films with a bias dc voltage.4,5 This limitation may
explain why we fail to observe a significant two-beam-
coupling effect in the present case. In other words, the
nonlinear index change appears to be a local effect.
We investigated the effect of applying external dc
fields across the cell windows, i.e, along z. Figure 4
shows the diffraction eff iciency as a function of applied
dc voltage. Positive bias up to a maximum value at
0.4 V enhances the effect only minimally (from 28%
to 30%). Application of higher bias voltage (in either
direction) tends to lower the diffraction efficiency of the
grating. At higher applied voltage s61 Vd, instability
sets in and destroys the sample/grating. Applying ac
voltages, we found that they too diminish the grating
diffraction; an applied ac voltage of ,10 V peak to peak
turns off the grating diffraction completely. These
observations can all be explained simply in terms of the
255
quenching action of the applied field on the molecular
reorientation. The magnitudes of the ac/dc voltages
needed to quench the molecular reorientation correlate
well with the ac/dc dielectric anisotropies.
The grating formation time is dictated mainly by the
dynamics of the photoinduced space charges/f ields; as
the optical intensity increases, the buildup time short-
ens (see Fig. 5). At higher intensities s.20 mWycm2 d
the buildup time can be shortened to ,50 ms. The
relaxation time is given7 by tr , gyKsp 2 d22 1 q2 d,
where g is the viscosity coefficient and K is the
elastic constant, g , 0.15 P, and K , 1026 dyn. For
d ­ 6 mm (and lg ­ 12 mm) and d ­ 25 mm (and
lg ­ 50 mm) we get tr ­ 50 ms and tr ­ 0.78 s, respec-
tively. These time constants are in agreement with
those observed experimentally. We did observe that
under high-intensity illumination the induced grating
contains a persistent component that lasts for tens of
minutes after the writing beam is turned off, similar to
those observed in Rhodamine 6G dye– doped f ilm5 and
Methyl Red–doped f ilm,3 for which changes in surface
realignment and anchoring conditions are significant.
In conclusion, we have observed an extraordinarily
large optically induced mechanism for refractive-index
change in Methyl Red –doped nematic film. The effect
is potentially useful for application in coherent wave
mixing and adaptive optics devices and in development
of new liquid-crystal light valves and spatial light
modulators.
This research is supported by the U.S. Air Force
Phillips Laboratory and the U.S. Army Research Of-
fice. S. Slussarenko is partially supported by a U.S.
Civilian Research and Development Foundation tra-
vel award.
*Permanent address, Institute of Physics, Ukrainian
Academy of Sciences, Prospect Nauki 46, Kiev-22,
252650, Ukraine.
References
1. I. Janossy and T. Kosa, Opt. Lett. 17, 1183 (1992).
2. W. M. Gibbons, P. J. Shannon, S. T. Sun, and B. J.
Swetlin, Nature (London) 351, 49 (1991).
3. A. G. Chen and D. J. Brady, Opt. Lett. 17, 1231 (1992);
see also F. Simoni, O. Francescangeli, Y. Reznikov, and
S. Slussarenko, Opt. Lett. 549 (1997).
4. E. V. Rudenko and A. V. Sukhov, JETP 78, 875 (1994).
5. I. C. Khoo, H. Li, and Y. Liang, Opt. Lett. 19, 1723
(1994); I. C. Khoo, Opt. Lett. 20, 2137 (1995); IEEE J.
Quantum Electron. 32, 525 (1996).
6. G. P. Wiederrecht, B. A. Yoon, and M. R. Wasielewski,
Science 270, 1794 (1995).
7. I. C. Khoo, R. G. Lindquist, R. R. Michael, R. J.
Mansf ield, and P. LoPresti, J. Appl. Phys. 69, 3853
(1991); see also I. C. Khoo, Liquid Crystals: Physical
Properties and Nonlinear Optical Phenomena (Wiley
Interscience, New York, 1994).
8. F. P. Schafer, ed., Dye Lasers (Springer-Verlag, Heidel-
berg, 1974).
9. W. Helfrich, J. Chem. Phys. 51, 4092 (1969).
10. S. Sato, Jpn. J. Appl. Phys. 20, 1989 (1981).
11. L. K. Vistin, P.-A. V. Kazlauskas, and S. Paeda, Sov.
Phys. Dokl. 29, 207 (1984).