Open Access
Add to BibSonomyAbstract
The induced transient dichroism and the transient optical Kerr effect (OKE) response of azo-doped isotropic liquid crystals were studied in picosecond time scale. Photoisomerization occurs when the azo-dye molecules are resonantly excited. As a result, a significant enhancement of OKE signal was found in the dye doped liquid crystals with respect to the pure liquid crystals. The enhancement was attributed to the extra torque exerted on liquid crystals after anisotropic orientational distribution of trans and cis isomers were generated. A modified mean field theory model was applied to treat the interactions of trans and cis isomers with liquid crystals. A small-pump enhancement factor of 3.8 was obtained, which is comparable with that of the well-known anthraquinone dye doped LC system.
©2006 Optical Society of America
1. Introduction
In 1990, it was found that photo-excitation of small amount of doped anthraquinone dye molecules could drastically enhance the optical re-orientation ability of nematic liquid crystals (LC). Since then, the effect (Jánossy effect) has attracted much attention,[1–3] because it gives a classical and important example of modifying host properties by guest (defect) molecules, it also provides an efficient way to orient LC molecules much more easily by optical field, which potentially leads to applications such as optical alignment of LC cells. The mechanism of Jánossy effect was concluded to be that after resonant polarized excitation, the generated orientational anisotropy of anthraquinone in ground and excited state will create an additional torque (mean field) to orient LC molecules [5–8]. The mechanism also holds generically in host-guest systems such as dyes in isotropic LC or liquids.[9]
However, guest-host interaction in another important type of azo-dye doped liquid crystal system is not well-understood.[10] Since azo-dye represents a whole family of molecules that can response very effectively to light through trans-cis photo-isomerization, understanding the real mechanism of azo-dye interaction with liquid crystal is very important to both physicists and chemists. Although Jánossy gave a phenomenological explanation based on an assumption that the trans and cis isomers have opposite sign of the additional torque,[11–12] the physical connections of the explanation with the microscopic model which describes well the generic Jánossy effect is not clear, its extension to include the dynamic process of enhancement is not possible either. Experimental work to reveal the early re-orientation dynamics of such a system is obviously necessary to support a physical explanation.
In this paper, we studied the photo-induced transient dichroism and the optical Kerr effect (OKE) response of the azo-doped LC. It was found that the enhancement of optical reorientation ability of LC comes from anisotropic angular distribution of two isomers. A small-pump enhancement factor of 3.8 was obtained, which is comparable with that of the well-known anthraquinone dye doped LC system. Its dynamical process can still be well understood under the theoretical framework of mean field theory.
2. Transient dichroism
It is well known that trans conformation (more or less rod-like) is a more stable state for azo-dye. After being excited, the trans molecule with its molecular axis parallel to the pump light polarization direction will be preferentially excited and transformed with certain probability to cis conformation which adopts a V-like shape.
Generally, information about the rotational decay of dyes in the ground state can be obtained by pump-probe transient absorption spectroscopy. A linearly polarized pump light generates an anisotropic orientation distribution. An optically delayed weak light beam then probes the time-resolved change of the ground state population. The probe light intensity follows I(t)=I0exp[-α(t)d], where I0 is the intensity of input beam, d is the sample thickness, α(t) is the absorbance at the probe beam wavelength at time t after the pump. We have
in which α(0) and I(0) are the absorbance and probe intensity without pump. By varying the probe beam polarization, transient dichroism can be obtained from:
where the subscripts denote the polarization direction either parallel (//) or perpendicular (⊥) to the pump beam polarization. As the system is isotropic before the pump, we have α ⊥(0)=α //(0)=α 0, and I ⊥(0)=I //(0)=I 0 exp(-α 0 d). Therefore,
In the case of azo-dye, ground state population of both trans and cis isomers should be included after photo-excitation. They will be denoted by the subscript index i=t, c, respectively. Assuming their temporal orientational order parameters to be
the non-equilibrium absorbance can be expressed as [13]:
where Ni is the total number of dye molecules per unit volume in state i; Qi is the orientational order parameter (normalized to Ni); σI′ is the absorption cross section for the probe beam; and Ai is the coefficients that account for the possibility of having transition dipoles not parallel to the molecule axis (Ai=1 for the parallel case).
Therefore the dichroism can be written as
As the trans isomer has much larger absorbance in the wavelength region that we are interested, Eq. (6) can be simplified as
As a result, the rotational decay constants τt and τc can be deduced from dichroism measurement.
The system we studied is azobenzene dye (dimethylamino-nitroazobenzene (DMANAB)) doped 4’-n-pentyl-4-cyanobiphenyl (5CB, from Merck) LC. The dye concentration was 0.05% by weight. The sample was placed in a 1mm fused quartz cell. The cell was kept at 47 °C (above the clear point of 5CB, which is 35.4 °C). Transient dichroisms at several probe wavelengths were measured. The pump wavelength was 532nm. The output of the picosecond optical parametric generators (OPG) was used as the probe beam. The same experimental configuration as described in Ref. [15] was used to separate the contribution of birefringence from that of dichroism.
Fig. 1 Time resolved transient absorbance observed at 520nm. Rotational decay lifetimes 1815±148ps and 227±47ps were obtained by bi-exponential fitting (solid line).
The transient absorption probed at 520nm was shown in Fig. 1. Two time constants were obtained by fitting the decay curve, they are 1815±149 ps and 227±47 ps. Table 1 summarizes the time constants and their relative weights (A1/A2) measured at different wavelengths. As it is known that the absorption band of the cis isomer locates in the shorter wavelength region, we concluded that the short time constant is associated with the cis isomer. On the other hand, the change of τi over wavelength is not conclusive due to the relatively poor S/N at shorter wavelength. Another check of assignment of τt and τc was made by measuring the change of A1/A2 when pump light intensity increases. With higher pump intensity, presumably more population of cis isomer is expected, thus A1/A2 will increase, though with an obvious saturation tendency. This was confirmed in experiment and shown in Fig. 2. In Fig. 2, experimental data were also fitted by a saturation curve, a saturation intensity of 15.2 mJ/cm2 can be deduced.
Table 1. The results of the transient dichromism measurement at several wavelengths.
Fig. 2. Relation of A1/A2 to pump intensity. The solid line is a saturation fitting which yields a saturation pump intensity of 15.2 mJ/cm2.
3. Optical Kerr effect in azo-doped LC
Treating trans and cis isomers as the “ground” and “excited” state molecules respectively as in a generic Janossy effect analysis, the dynamics of the trans and cis moleculars’ order parameters can be written as the following after some modification of Eq. (2) in reference [9]:
where τt and τc are the decay times of Qt and Qc due to rotational diffusion, respectively, αt and αc are light absorption probability per unit time, and ϕtc, ϕct are quantum efficiencies of the trans-cis and cis-trans transitions, respectively. Although the two isomers transform to each other through an excited state, and experimental evidence showed that the excited state had a lifetime of a few picoseconds [14], the influence of excited state was ignored in this work. Furthermore, we assumed that after a trans-cis-trans transformation cycle, the possibility that a re-generated trans isomer keeps its original orientation is low (i.e. Ptc, Pct<<1, this assumption was proved to be reasonable in a separate experiment in a nanosecond time scale [16]). Then Eq. (8) can be re-written as:
If the light pulse duration is much shorter than τi, Qi(t) takes the form in Eq. (4). Based on the mean-field model, and assumed that trans and cis isomers can be treated as different doping species, the ordering parameter of liquid crystal Qh follows [6]:
Contributions to the order parameter of liquid crystals are from optical torque (∝I0(t)), trans (∝Qt(t)) and cis (∝Qc(t)) orientational anisotropy. In Eq. (10), A, B, ηi are constants which are determined by the interaction strengths between dye and liquid crystals.[6,7] It should be noted that conventionally dynamical response of isotropic LC can be reproduced by three time constants in different time scale [9]. In Eq. (10), two longer time constants (τh1, τh2) are included, as the third time constant (~ps) which represents the electronic response of individual LC molecules only has effects during the existence of the pump pulse and thus can be ignored.
The transient re-orientation of azo-doped isotopic liquid crystals under optical pump was measured by using OKE measurements. The OKE signal is proportional to the square of order parameter [9]:
Experimentally, the fundamental (1064 nm) output of a picosecond Nd:YAG laser was used as the probe. Its frequency-doubled light (532 nm) acted as the pump. The probe beam was substantially attenuated and went through a motorized optical delay line; it then passed through a pinhole with a 1 mm diameter. The pump and the probe beams were both linearly polarized, and the polarization of the probe beam was rotated 45° with respect to that of the pump. After a lens with a focal length of 15 cm, the two beams were focused tightly on the same spot of the sample. The diameter of the pump beam is about 0.5 mm and its overlapping size with the probe beam is about 120 µm. The pulse width is about 28 ps. After passing a polarizer that was cross-polarized with the probe polarization, the transmitted light was detected by a photodiode connected with gated integrators.
Fig. 3. The time resolved square root of OKE signal of the azo-doped LC and pure LC. Theoretical fitting was done by using Eq. (11).
Figure 3 shows a transient OKE curve measured at a pump intensity of 0.88 GW/cm2 (26 mJ/cm2). It apparently consists of a broad bump after the pump and a slowly decaying tail. The intensity of the tail is significantly larger than that of the pure LC, which implies that by exciting azo-dye molecules, large enhancement of nonlinearity occurs. By using the rotational diffusion lifetime of trans and cis isomers obtained in the previous section and taking τh1=14.3 ns, τh2=0.5 ns obtained by fitting the OKE curve of pure LC, the transient OKE curve can be fitted in terms of Eq. (10). The fitted curve agrees with the experimental result well. The first bump peaked at 300 ps results from the mediate time response of LC, while the large tail from the long-time rotational diffusion of liquid crystals. In addition, trans and cis isomers have different contributions to the LC reorientation, because their decay times (i.e. interaction time with LC) are very different, and the interacting strength could also be different.
The enhancement of the steady state orientation and its dependence on the intensity of the pump light were obtained by measuring OKE signal of both pure LC and the azo-dye doped LC after a delay of 1.7 ns. This time delay insures that the transient dynamics of the perturbed dye molecules dies away. The results were shown in Fig. 4. The enhancement goes down inversely with the increase of pump power. This can be understood as the following: the steady state population of the cis isomer will gradually saturate with the increase of the pump energy. The enhancement factor can be simply written as:
in which ηc is the small-pump enhancement factor, Is is the saturation intensity. Fitting the experimental points in terms of Eq.(12) yields ηc=3.8±0.4 and Is=14.8±2.9mJ/cm2. The enhancement factor is in the same level comparing with those in other systems like anthraquinone dye doped LC (ηc=3.2±0.3). [9] A separate check of the Is was carried out by fitting data in Fig. 2 with a saturation curve which yields a Is=15.2 mJ/cm2.
Fig. 4. The enhancement factor (solid circles) vs. pump intensity. Solid line is a fitted line in terms of Eq. (12). The pump intensity dependent OKE signals from pure LC (open squares) and azo-dye-doped LC (open triangles) were also plotted.
4. Summary
In summary, we studied the photo-induced transient dichroism in picosecond time scale and obtained the rotational diffusion lifetimes of trans and cis isomers. Time-resolved optical Kerr effect measurements were carried out to reveal the early dynamics of the re-orientation process of azo-dye doped liquid crystals. A small-pump orientation enhancement factor of 3.8 was obtained, which is comparable with that of the well-known anthraquinone dye doped LC system. The orientation enhancement was concluded to come from photo-isomerization generated angular anisotropy of the two isomers. Mean field theory was found to be still applicable for treating such a system.
Acknowledgments
This work was supported by National Natural Science Foundation of China (#10474015, 60478005, 50532030) and Shanghai Commission of Science and Technology (#06JC14010).
References and links
1. I. Jánossy, A.D. Lloyd, and B.S. Wherrett, “Anomalous optical Freedericksz transition in an absorbing liquid-crystal,” Mol. Cryst. Liq. Cryst. 179, 1–12 (1990).
2. D. Paparo, L. Marrucci, G. Abbate, E. Santamato, M. Kreuzer, P. Lehnert, and T. Vogeler, “Molecular-field-enhanced optical Kerr effect in absorbing liquids,” Phys. Rev. Lett. 78, 38–41 (1997). [CrossRef]
3. R. Muenster, M. Jarasch, X. Zhuang, and Y. R. Shen, “Dye-induced enhancement of optical nonlinearity in liquids and liquid crystals,” Phys. Rev. Lett. 78, 42–45 (1997). [CrossRef]
4. I. Jánossy and T. Kósa, “Influence of anthraquinone dyes on optical reorientation of nematic liquid crystals,” Opt. Lett. 17, 1183–1185 (1992). [CrossRef] [PubMed]
5. I. Jánossy, “Molecular interpretation of the absorption-induced optical reorientation of nematic liquid crystals,” Phys. Rev. E 49, 2957–2963 (1994). [CrossRef]
6. L. Marrucci, D. Paparo, G. Abbate, E. Santamato, M. Kreuzer, P. Lehnert, and T. Vogeler, “Enhanced optical nonlinearity by photoinduced molecular orientation in absorbing liquids,” Phys. Rev. A 58, 4926–4936 (1998). [CrossRef]
7. M. Kreuzer, L. Marrucci, and D. Paparo, “Light-induced modification of kinetic molecular properties: enhancement of optical Kerr effect in absorbing liquids, photoinduced torque and molecular motors in dye-doped nematics,” J. Nonlinear Opt. Phys. Mater. 9, 157–182 (2000).
8. M. Kreuzer, F. Hanisch, R. Eidenschink, D. Paparo, and L. Marrucci, “Large deuterium isotope effect in the optical nonlinearity of dye-doped liquid crystals,” Phys. Rev. Lett. 88, 013902 (2002). [CrossRef] [PubMed]
9. T. V. Truong, L. Xu, and Y. R. Shen, “Early dynamics of guest-host interaction in dye-doped liquid crystalline materials,” Phys. Rev. Lett. 90, 193902 (2003). [CrossRef] [PubMed]
10. M. I. Barnik, A. S. Zolotko, V. G. Rumyantseva, and D. B. Terskov, “Photoinduced director reorientation in aza-dye-doped nematic liquid-crystal,” Kristallografiya 40, 746–750 (1995).
11. I. Jánossy and L. Szabados, “Optical reorientation of nematic liquid crystals in the presence of photoisomerization,” Phys. Rev. E 58, 4598–4604 (1998). [CrossRef]
12. M. Becchi, I. Janossy, D. S. Shankar Rao, and D. Statman, “Anomalous intensity dependence of optical reorientation in azo-dye-doped nematic liquid crystals,” Phys. Rev. E 69, 051707 (2004). [CrossRef]
13. M. Kreuzer, E. Benkler, D. Paparo, G. Casillo, and L. Marrucci, “Molecular reorientation by photoinduced modulation of rotational mobility,” Phys. Rev. E 68, 011701 (2003). [CrossRef]
14. B. Schmidt, C. Sobotta, S. Malkmus, S. Laimgruber, M. Braun, W. Zinth, and P. Gilch, “Femtosecond fluorescence and absorption dynamics of an azobenzene with a strong push-pull substitution,” J. Phys. Chem. A 108, 4399–4404 (2004). [CrossRef]
15. A. Zeug, I. Ruckman, and B. Roder, “Picosecond transient dichroism and birefringence spectroscopy on pheophorbide-a molecules in solution,” J. Opt. B 3, S251–S258 (2001). [CrossRef]
16. Liyong Ji, Pei Yang, Xiu Liu, Liying Liu, and Lei Xu, unpublished results.
