PHYSICAL REVIEW E 89, 032507 (2014)
Role of space charges on light-induced effects in nematic liquid crystals doped by methyl red L. Lucchetti and F. Simoni Dipartimento di Scienze e Ingegneria della Materia, dell’Ambiente ed Urbanistica and CNISM, Universit`a Politecnica delle Marche, Ancona, Italy (Received 5 December 2013; published 24 March 2014) We show that both the extraordinarily large nonlinear response and the light-induced permanent reorientation in liquid crystals doped by the azo dye methyl red originates from the same phenomenon of modification of the charge density on the irradiated surface. The demonstration is done by applying ac voltage to the samples, showing that in this case no permanent anchoring is possible. The measurements confirm the role of photoisomerization that gives a transient contribution to the actual reorientation process only in the high dose regime. This result allows us to draw a picture for light-induced effects that might be applied to a large class of compounds. DOI: 10.1103/PhysRevE.89.032507
PACS number(s): 61.30.Gd, 61.30.Hn, 82.50.Hp, 78.15.+e
Light-induced effects in dye-doped liquid crystals have attracted interest in the past years for two main reasons. The first one is the research of a suitable photoalignment technique to be used in display technology and in optical memory devices [1]. The second originates from the discovery of the huge nonlinear response of these mixtures that might have interesting applications in the field of optical processing with extremely low power [2]. One of the most studied systems in this field is the mixture of the nematic liquid crystal pentylcyanobiphenil and the azo dye methyl red (5CB + MR) at a weight concentration typically between 0.1% and 1%. In this material the abovementioned phenomena are strongly affected by light-induced surface effects as reported by different authors [3–9]. It has been demonstrated that the competition between light-induced desorption and light-induced adsorption determines the final orientation of the liquid crystal on the surface irradiated with a wavelength in the absorption band of MR, allowing photoalignment perpendicular or parallel to the impinging light polarization, depending on the dose of the irradiation [4,10]. In summary the photoalignment process has been described in the following way: (i) a so-called “dark-adsorbed” layer of molecules is adsorbed on the boundary surface during sample preparation, before any irradiation of the sample; (ii) upon illumination, dark-adsorbed molecules in the trans state absorb light and may undergo desorption from the surface pushing the easy axis outward from the light polarization direction; (iii) upon illumination, molecules in the bulk absorb light going to the cis state and may undergo adsorption on the illuminated surface pushing the easy axis towards the light polarization direction; (iv) light absorption in the bulk, leading to photoisomerization of the molecules, gives rise to optical torque on the liquid crystals (LC) director pushing the easy axis outward from the light polarization direction; (v) the combination of the abovementioned phenomena, dependent on energy density and polarization of light, produces the final orientation of LC on the irradiated surface; i.e, for low energy density, light-induced desorption is dominant on adsorption leading to reorientation outward from the light polarization direction, while for high energy density the opposite occurs. On the other hand in the past we studied the nonlinear optical properties of this mixture, pointing out that it is possible to achieve a colossal response that can be easily controlled 1539-3755/2014/89(3)/032507(5)
by the electric field [2,11–13]. We have demonstrated that the huge nonlinear optical response is due to a light-induced reduction of the charge density on the irradiated surface, i.e., desorption of charges from the surface, that affects the actual anchoring conditions and, due to LC elasticity, the LC orientation in the bulk. This is the surface-induced nonlinear effect (SINE). This nonlinear optical behavior can be observed at very low doses; that is, the boundary between permanent light-induced anchoring fixing the easy axis and a reversible nonlinear response is given by the provided energy density. The basic question arising from the results of the mentioned investigations is the following: is there a link between the transient reorientation accounting for the extraordinarily large nonlinear response observed at low light intensity and the light-induced permanent reorientation observed at higher light doses? Are they parts of the same process or do they have a different origin? More than a decade ago we proposed a positive answer to this question through a detailed analysis of the diffraction efficiency of transient and permanent gratings recorded in these types of samples using different pump polarizations [5]; however a clear demonstration of the role of charges in the adsorption and desorption processes, whose balance determines the final easy axis on the irradiated surface, has not been given yet. At the same time no theoretical models explaining the changes of the easy axis on the irradiated surface have taken into account that charges are involved in the desorption or adsorption process. In this work we want to give a final answer to the above-proposed question. Let us suppose the nonlinear behavior is due to transient desorption of the charge complex from the irradiated surface (already demonstrated in Refs. [5–11]) while the memory effect is due to permanent adsorption of the same charge complex. In this case the two transient and permanent effects actually have the same origin, being different only in the light dose impinging on the sample. Actually there is a very simple way to verify this model: compare the effect of irradiation of this type of sample by applying an external voltage and compare the results obtained when V = 0 V, V > 0 V dc., and V > 0 V ac. In fact while the dc field favors the formation of a space charge at the sample surfaces, an ac field will prevent it. As a consequence, according to our model, we should observe all the reorientation processes at V = 0 V and possibly enhanced at V > 0 with a
032507-1
©2014 American Physical Society
L. LUCCHETTI AND F. SIMONI
PHYSICAL REVIEW E 89, 032507 (2014)
off
(a)
S (arb. units)
dc field. On the contrary we expect no effect when an ac field is applied. It is worth mentioning that similar measurements have already been performed on undoped 5CB twisted cells to study the Freedericksz transition with and without light illumination [14]. Experiments have been carried out on 5CB planar cells doped with a small amount of MR. The dye weight concentration used was 1%. One commercial substrate treated to give strong planar anchoring (from INSTEC) was used as one of the two substrates of the cells; the other was a bare ITO coated glass. The first substrate had ITO coating as well to make the cell conductive. Cell thickness was controlled by Mylar spacers and carefully measured by means of spectroscopic techniques. Thirteen-micrometer Mylar spacers were used and the effective cell thickness measured by the spectroscopic method ranged between 19 and 20 μm. Once assembled, cells were filled with the 5CB + MR mixture in the isotropic state and then slowly cooled down to room temperature. During cooling cells were kept under the action of a magnetic field with the axis parallel to the rubbing direction of the planar surface. This procedure resulted in a satisfactory uniform planar alignment of LC over the whole cell. A dc or ac voltage was applied orthogonally to the cell substrates. A pump beam from an Ar+ laser (λ = 465 nm) impinged on the cell at normal incidence from the side of the unrubbed surface focused on the sample by a 20-cm plano-convex lens. The pump polarization direction was at an angle of 45° with respect to the initial easy axis of the cell. In this way both light-induced desorption and light-induced adsorption of dye molecules are expected to occur [3,4]. A probe beam of a low power He-Ne laser impinged on the other side of the sample with polarization parallel to the rubbing direction of the boundary surface and focused by a 10-cm focal lens in the central spot of the pump beam. The probe light transmitted by the cell and by a polarizer crossed to its initial polarization was monitored by a photodetector connected to an oscilloscope and a PC, as a function of time and of the pump energy dose. The frequency used in the case of ac voltage application was 1 kHz. We initially fixed the light power at 18 μW, giving an estimated intensity of 0.35 W/cm2 , which according to our previous experiments [12] allows a nonlinear response of the system, not inducing any memory effect on a short time scale. We should stress here that we do not use any special coating (such as PVCN-F) to enhance the permanent anchoring phenomenon as has been used in other experiments in the past [3,4,10,15]; therefore the energy dose needed to induce such an effect is higher in our samples. In Fig. 1 we report the signal measured versus the irradiation time (that is, for an increasing light dose). The observed signal is due to rotation of probe light polarization consequent to liquid crystal director reorientation [16,17]. Curve (a) is obtained with no applied voltage, while Curve (b) is obtained with an ac voltage applied (V = 0.2 Vrms ) at 1 kHz. It is worth noting that the applied voltage is below the threshold for the electric Freedericksz transition (measured as V = 0.9 Vrms ), then it is not able to induce any additional director reorientation. At V = 0, initially we observe an increase of the signal followed by a decrease to zero and then a uniform increase up to a saturation value obtained for energy densities higher than 1.3 kJ/cm2 . This behavior corresponds to the one
(b)
0
200
400
600
800
1000 1200 1400 1600 1800
E (J/cm2)
FIG. 1. Transmitted probe signal versus impinging light dose. Curve (a) is obtained with no applied voltage, while curve (b) is obtained with an ac voltage applied (V = 0.2 Vrms ) at 1 kHz. The vertical dashed line indicates the switching off of the pump beam. Inset: Detail of the region between 0 and 200 J/cm2 .
reported already several years ago in similar samples where the irradiated surface included a coating by PVCN-F [17]. At V = 0.2 Vrms the signal is zero up to an energy density of E = 1.1 kJ/cm2 , and it increases in correspondence to a change of slope of the signal at V = 0 V. Up to this value of the light dose the behavior is the one expected by our model since in the absence of space charge density (ac field applied) we do not observe any director reorientation. What happens above this value of energy density becomes clear if one takes into account the correspondent rotation of the liquid crystal director as is discussed below. By switching off the pump beam (vertical dashed line in Fig. 1) we observe a slight initial increase of the signal for V = 0, signal that keeps otherwise constant in time stabilizing to a value correspondent to the one before the switching off of the pump beam. We also observe a decay to zero of the signal for Vac > 0 . As mentioned, by checking the polarization rotation of the beam transmitted by the sample it becomes easy to understand the origin of the signal reported in Fig. 1. In fact in our experimental configuration due to the adiabatic condition the exit probe beam polarization follows the director orientation at the irradiated surface; therefore by rotation of the exit polarizer one can find again the extinction condition and determine the amount and the sign of rotation, which is taken as positive if towards the pump beam polarization and negative if outwards from the pump beam polarization. After determining the sign, the amount of rotation can actually be easily obtained from the data of Fig. 1 just applying the Malus law [18]. In Fig. 2 we report the reorientation angle correspondent to the two conditions reported in Fig. 1 [curve (a) V = 0 V and curve (b) V = 0.2 Vrms ]. Curve (a) shows the well known behavior already reported [17] where an initial negative reorientation is followed by a positive reorientation that becomes permanent, saturating at a certain value. On the contrary when a small ac field is applied, as is already clear in Fig. 1, we do not get any reorientation until E = 1.1 kJ/cm2 ; then we observe
032507-2
ROLE OF SPACE CHARGES ON LIGHT-INDUCED . . .
PHYSICAL REVIEW E 89, 032507 (2014)
off
10
20
(a)
8
6 10
τ (s)
φ (deg)
15
4 5
(b)
2 0
0
200
400
600
800
0
1000 1200 1400 1600 1800
0
E (J/cm2)
20
40
60
E
80
100
120
(J/cm2)
FIG. 2. Reorientation angle correspondent to the signal reported in Fig. 1. Again the vertical dashed line indicates the switching off of the pump beam.
FIG. 3. Relaxation time vs impinging energy dose. The solid line is a guide for the eyes. The vertical dashed line indicates the boundary between transient and permanent light-induced reorientation.
an increasingly negative one, which decays to zero when the pump beam is switched off. In correspondence with the pump switching off the positive reorientation occurring at V = 0 V gets a slight increase, in perfect agreement with results reported by the group of Reznikov [15]. These observations point out the role of photoisomerization of the azo molecules. In fact Fig. 2 shows that the signal observed when the ac voltage is applied (above E = 1.1 kJ/cm2 ) is due to a reorientation outward from the pump polarization that is not permanent; therefore it is consequent to photoisomerization of bulk molecules, an effect that disappears when the pump is switched off. We should underline that it is the first time that the photoisomerization effect in dye-doped liquid crystals is observed not overlapped on other phenomena that are here quenched by the ac field. Our observation makes clear that this process becomes relevant only at high doses, while surface effects are the dominant ones at lower energy densities. Additional information on the overall process is obtained by looking at the lifetime of the induced reorientation; this helps in distinguishing between transient reorientation that leads to nonlinear optical response and permanent reorientation that leads to light-induced anchoring. Measurements were performed at V = 0 V, using a pump power of P = 18 μW, and switching off the pump beam at increasing values of the impinging energy density. Then the relaxation time was evaluated by an exponential fit of the relaxation curves. The relaxation time increases with the irradiation dose, starting from 1 s at E = 8 J/cm2 up to 10 s at E = 104 J/cm2 . For higher values of the impinging energy density, the signal does not relax to zero within 30 min from switching off the pump beam. Therefore we can speak of transient reorientation up to an energy density of 104 J/cm2 , while above this value reorientation approaches a condition that is stable in time. In Fig. 3 a dashed line located at about 110 J/cm2 shows the asymptotic behavior towards the permanent reorientation. The continuous line is a guide to the eye showing the diverging increase of the reorientation lifetime. Comparing
these results to data reported in Figs. 1 and 2, it becomes clear that we get a transient reorientation when driven by lightinduced desorption, while we get a long-lasting to permanent reorientation when driven by light-induced adsorption. The behavior of the photoinduced reorientation has also been tested on an external dc field. According to our interpretation, the application of a dc field might enhance the signal, provided it has the right sign. If the observed light-induced effects in MR doped 5CB are all due to the presence of the charge complex present in the mixture, a dc field of the right polarity should enhance the effects of adsorption and desorption of these complexes, whereas a field of the opposite polarity should prevent the effect from occurring. In this respect, measurements with dc fields of different polarities should also give information about the sign of the charge complexes. Figure 4 reports the probe signal transmitted by the cell between crossed polarizers, in the case of positive (a) and negative (b) polarity of the dc field applied to the irradiated surface. The field is again lower than the threshold for the electric Freedericksz transition (note that the measured threshold in the case of the dc field is higher than the one measured with the ac voltage). For positive polarity the measured signal is comparable with the one obtained at zero field reported in Fig. 1 [curve (a)], being slightly enhanced with respect to it. If the field polarity is reversed, the signal has a different behavior that is quite similar to that observed in Fig. 1 [curve (b)] for the ac field. The two curves are also compared to the signal detected for V = 0 V [curve (c)]. The data of Fig. 4 are in good agreement with the proposed model: the application of a positive dc field slightly enhances the overall effect of light-induced desorption and adsorption of dye molecules on the irradiated surface, while the application of a negative dc field quenches both these phenomena and leaves the photoisomerization unperturbed. From these results we have the additional information that negative charge complexes are responsible for the light-induced desorption and adsorption processes.
032507-3
L. LUCCHETTI AND F. SIMONI
PHYSICAL REVIEW E 89, 032507 (2014)
form, which tends to align the surface director parallel to the easy axis. In this way the total surface free energy density becomes 1 ε LD 1 (nˆ s · Es )2 − W (nˆ S · πˆ )2 , (2) FS = − 2 2 2
(a)
S (arb.units)
(c)
where ε = ε// − ε⊥ , LD is the Debye screening length, and ns is the surface director. This term must be added to the intrinsic surface energy having the usual Rapini-Papoular
with W being the anchoring strength and π the easy axis. The surface field Es is connected to the surface charge density and this leads to a dependence of the anchoring energy on it. The observed light-induced reorientation outward and toward the pump light polarization due to desorption and adsorption of charged complexes corresponds to transient and permanent modification of σ and leads directly to a change of the anchoring energy that can be transient or permanent depending on the energy dose impinging on the cell (i.e., depending on the amount of the modification of σ ). We want to summarize here the main results of the present report. First of all we have demonstrated by a very simple measurement that the light-induced processes widely investigated for more than a decade on nematic liquid crystals doped by methyl red have the same origin, involving a basic role played by charges on the irradiated surface. Namely, the only difference between the transient response leading to huge nonlinear response and light-induced permanent anchoring is the amount of light dose. Linked to this demonstration is the definite observation that only light-induced adsorption leads to a permanent director reorientation, while the desorption effect has a limited recovery time, being sufficiently small at low light intensity to allow a huge nonlinear optical response. Another important observation is that the process in this compound involves negative charges. This fact should be cleared by a more detailed investigation on the photoinduced charge generation in these types of materials. Finally we have confirmed the limited role of photoisomerization in these processes occurring only at high dose levels and we have isolated this effect from the other ones by quenching them via ac field application. We point out that the features of the light-induced effects occurring in the studied compound can be envisaged to be typical of the large class of donor-acceptor azobenzene compounds. Actually a systematic investigation of them to look for optimization of the discussed phenomena is still missing.
[1] M. M. Chrzanowski, J. Zielinski, M. Olifierczuk, J. Kedzierski, and E. Nowinowski-Kruszelnicki, J. Achiev. Mater. Manuf. Eng. 48, 7 (2011). [2] L. Lucchetti, M. Di Fabrizio, O. Francescangeli, and F. Simoni, Opt. Commun. 233, 417 (2004). [3] E. Ouskova, D. Fedorenko, Y. Reznikov, S. V. Shiyanovskii, L. Su, J. L. West, O. V. Kuksenok, O. Francescangeli, and F. Simoni, Phys. Rev. E 63, 021701 (2001). [4] E. Ouskova, Y. Reznikov, S. V. Shiyanovskii, L. Su, J. L. West, O. V. Kuksenok, O. Francescangeli, and F. Simoni, Phys. Rev. E 64, 051709 (2001). [5] L. Lucchetti, M. Di Fabrizio, O. Francescangeli, and F. Simoni, J. Nonlinear Opt. Phys. Mater. 11, 13 (2002).
[6] L. Lucchetti, D. E. Lucchetta, O. Francescangeli, and F. Simoni, Mol. Cryst. Liq. Cryst. 375, 641 (2002). [7] R. Ramos-Garcia, I. Lazo-Martinez, I. Guizar-Iturbide, A. Sanchez-Castillo, M. Boffety, and P. Ruck, Mol. Cryst. Liq. Cryst. 454, 179 (2006). [8] V. Boichuk, S. Kucheev, J. Parka, V. Reshetnyak, Y. Reznikov, I. Shiyanovskaya, K. D. Singer, and S. Slussarenko, J. Appl. Phys. 90, 5963 (2001). [9] J. Zhang, V. Ostroverkhov, K. D. Singer, V. Reshetnyak, and Y. Reznikov, Opt. Lett. 25, 414 (2000). [10] D. Fedorenko, E. Ouskova, V. Reshetnyak, and Y. Reznikov, Phys. Rev. E 73, 031701 (2006).
(b)
0
200
400
600
800
1000 1200 1400 1600 1800
E (J/cm2)
FIG. 4. Probe signal transmitted by the cell between the crossed polarizers, in the cases of positive (a) and negative (b) polarity of the dc field applied to the irradiated surface. The curves are also compared to the one obtained with no external field (c).
The presence of charged complexes in dye-doped LC samples has been directly observed by measurements of photocurrent [19] and photovoltage [20,21] by different authors. Moreover, the effects of charged objects on surface anchoring energy has also been studied several years ago [22,23]. From these works it is known that when a solid substrate is in contact with a nematic LC selective ion adsorption takes place. This creates an electric surface field in liquid crystal cells, which extends in the bulk over the Debye screening length [23]. The consequent surface field, Es , strongly affects the surface anchoring energy of the nematic director. This happens because the coupling of the surface field Es with the nematic director gives rise to an additional term to the surface free energy density of the form [23]: 1 ε LD FE = − (nˆ s · Es )2 , (1) 2 2
032507-4
ROLE OF SPACE CHARGES ON LIGHT-INDUCED . . .
PHYSICAL REVIEW E 89, 032507 (2014)
[11] L. Lucchetti, M. Gentili, and F. Simoni. Opt. Express 14, 2236 (2006). [12] L. Lucchetti, M. Gentili, F. Simoni, S. Pavliuchenko, S. Subota, and V. Reshetnyak, Phys. Rev. E 78, 061706 (2008). [13] L. Lucchetti, M. Gentili, and F. Simoni. IEEE J. Sel. Top. Quantum Electron. 12, 422 (2006). [14] J. Merlin, E. Chao, M. Winkler, and K. D. Singer, Opt. Exp. 13, 5024 (2005). [15] D. Fedorenko, K. Slyusarenko, E. Ouskova, V. Reshetnyak, K. R. Ha, R. Karapinar, and Y. Reznikov, Phys. Rev. E 77, 061705 (2008). [16] D. Voloshchenko, A. Khizhnyak, Y. Reznikov, and V. Reshetyak, Jpn. J. Appl. Phys. 34, 566 (1995).
[17] O. Francescangeli, S. Slussarenko, F. Simoni, D. Andrienko, V. Reshetnyak, and Y. Reznikov, Phys. Rev. Lett. 82, 1855 (1999). [18] F. A. Jenkins and H. E. White, Fundamentals of Optics (McGraw-Hill, New York, 1976), p. 1746. [19] E. V. Rudenko and A. V. Sukov, J. Electrochem. Plat. Technol. 78, 875 (1994). [20] S. Sato, Jpn. J. Appl. Phys. 20, 1989 (1981). [21] I. C. Khoo, S. Slussarenko, B. D. Guenther, M.-Y. Shih, P. Chen, and W. V. Wood, Opt. Lett. 23, 253 (1998). [22] A. V. Zakharov and R. Y. Dong, Phys. Rev. E 64, 042701 (2001). [23] G. Barbero and G. Durand, J. Appl. Phys. 67, 2678 (1990).
032507-5
Role of space charges on light-induced effects in nematic liquid crystals doped by methyl red L. Lucchetti and F. Simoni Dipartimento di Scienze e Ingegneria della Materia, dell’Ambiente ed Urbanistica and CNISM, Universit`a Politecnica delle Marche, Ancona, Italy (Received 5 December 2013; published 24 March 2014) We show that both the extraordinarily large nonlinear response and the light-induced permanent reorientation in liquid crystals doped by the azo dye methyl red originates from the same phenomenon of modification of the charge density on the irradiated surface. The demonstration is done by applying ac voltage to the samples, showing that in this case no permanent anchoring is possible. The measurements confirm the role of photoisomerization that gives a transient contribution to the actual reorientation process only in the high dose regime. This result allows us to draw a picture for light-induced effects that might be applied to a large class of compounds. DOI: 10.1103/PhysRevE.89.032507
PACS number(s): 61.30.Gd, 61.30.Hn, 82.50.Hp, 78.15.+e
Light-induced effects in dye-doped liquid crystals have attracted interest in the past years for two main reasons. The first one is the research of a suitable photoalignment technique to be used in display technology and in optical memory devices [1]. The second originates from the discovery of the huge nonlinear response of these mixtures that might have interesting applications in the field of optical processing with extremely low power [2]. One of the most studied systems in this field is the mixture of the nematic liquid crystal pentylcyanobiphenil and the azo dye methyl red (5CB + MR) at a weight concentration typically between 0.1% and 1%. In this material the abovementioned phenomena are strongly affected by light-induced surface effects as reported by different authors [3–9]. It has been demonstrated that the competition between light-induced desorption and light-induced adsorption determines the final orientation of the liquid crystal on the surface irradiated with a wavelength in the absorption band of MR, allowing photoalignment perpendicular or parallel to the impinging light polarization, depending on the dose of the irradiation [4,10]. In summary the photoalignment process has been described in the following way: (i) a so-called “dark-adsorbed” layer of molecules is adsorbed on the boundary surface during sample preparation, before any irradiation of the sample; (ii) upon illumination, dark-adsorbed molecules in the trans state absorb light and may undergo desorption from the surface pushing the easy axis outward from the light polarization direction; (iii) upon illumination, molecules in the bulk absorb light going to the cis state and may undergo adsorption on the illuminated surface pushing the easy axis towards the light polarization direction; (iv) light absorption in the bulk, leading to photoisomerization of the molecules, gives rise to optical torque on the liquid crystals (LC) director pushing the easy axis outward from the light polarization direction; (v) the combination of the abovementioned phenomena, dependent on energy density and polarization of light, produces the final orientation of LC on the irradiated surface; i.e, for low energy density, light-induced desorption is dominant on adsorption leading to reorientation outward from the light polarization direction, while for high energy density the opposite occurs. On the other hand in the past we studied the nonlinear optical properties of this mixture, pointing out that it is possible to achieve a colossal response that can be easily controlled 1539-3755/2014/89(3)/032507(5)
by the electric field [2,11–13]. We have demonstrated that the huge nonlinear optical response is due to a light-induced reduction of the charge density on the irradiated surface, i.e., desorption of charges from the surface, that affects the actual anchoring conditions and, due to LC elasticity, the LC orientation in the bulk. This is the surface-induced nonlinear effect (SINE). This nonlinear optical behavior can be observed at very low doses; that is, the boundary between permanent light-induced anchoring fixing the easy axis and a reversible nonlinear response is given by the provided energy density. The basic question arising from the results of the mentioned investigations is the following: is there a link between the transient reorientation accounting for the extraordinarily large nonlinear response observed at low light intensity and the light-induced permanent reorientation observed at higher light doses? Are they parts of the same process or do they have a different origin? More than a decade ago we proposed a positive answer to this question through a detailed analysis of the diffraction efficiency of transient and permanent gratings recorded in these types of samples using different pump polarizations [5]; however a clear demonstration of the role of charges in the adsorption and desorption processes, whose balance determines the final easy axis on the irradiated surface, has not been given yet. At the same time no theoretical models explaining the changes of the easy axis on the irradiated surface have taken into account that charges are involved in the desorption or adsorption process. In this work we want to give a final answer to the above-proposed question. Let us suppose the nonlinear behavior is due to transient desorption of the charge complex from the irradiated surface (already demonstrated in Refs. [5–11]) while the memory effect is due to permanent adsorption of the same charge complex. In this case the two transient and permanent effects actually have the same origin, being different only in the light dose impinging on the sample. Actually there is a very simple way to verify this model: compare the effect of irradiation of this type of sample by applying an external voltage and compare the results obtained when V = 0 V, V > 0 V dc., and V > 0 V ac. In fact while the dc field favors the formation of a space charge at the sample surfaces, an ac field will prevent it. As a consequence, according to our model, we should observe all the reorientation processes at V = 0 V and possibly enhanced at V > 0 with a
032507-1
©2014 American Physical Society
L. LUCCHETTI AND F. SIMONI
PHYSICAL REVIEW E 89, 032507 (2014)
off
(a)
S (arb. units)
dc field. On the contrary we expect no effect when an ac field is applied. It is worth mentioning that similar measurements have already been performed on undoped 5CB twisted cells to study the Freedericksz transition with and without light illumination [14]. Experiments have been carried out on 5CB planar cells doped with a small amount of MR. The dye weight concentration used was 1%. One commercial substrate treated to give strong planar anchoring (from INSTEC) was used as one of the two substrates of the cells; the other was a bare ITO coated glass. The first substrate had ITO coating as well to make the cell conductive. Cell thickness was controlled by Mylar spacers and carefully measured by means of spectroscopic techniques. Thirteen-micrometer Mylar spacers were used and the effective cell thickness measured by the spectroscopic method ranged between 19 and 20 μm. Once assembled, cells were filled with the 5CB + MR mixture in the isotropic state and then slowly cooled down to room temperature. During cooling cells were kept under the action of a magnetic field with the axis parallel to the rubbing direction of the planar surface. This procedure resulted in a satisfactory uniform planar alignment of LC over the whole cell. A dc or ac voltage was applied orthogonally to the cell substrates. A pump beam from an Ar+ laser (λ = 465 nm) impinged on the cell at normal incidence from the side of the unrubbed surface focused on the sample by a 20-cm plano-convex lens. The pump polarization direction was at an angle of 45° with respect to the initial easy axis of the cell. In this way both light-induced desorption and light-induced adsorption of dye molecules are expected to occur [3,4]. A probe beam of a low power He-Ne laser impinged on the other side of the sample with polarization parallel to the rubbing direction of the boundary surface and focused by a 10-cm focal lens in the central spot of the pump beam. The probe light transmitted by the cell and by a polarizer crossed to its initial polarization was monitored by a photodetector connected to an oscilloscope and a PC, as a function of time and of the pump energy dose. The frequency used in the case of ac voltage application was 1 kHz. We initially fixed the light power at 18 μW, giving an estimated intensity of 0.35 W/cm2 , which according to our previous experiments [12] allows a nonlinear response of the system, not inducing any memory effect on a short time scale. We should stress here that we do not use any special coating (such as PVCN-F) to enhance the permanent anchoring phenomenon as has been used in other experiments in the past [3,4,10,15]; therefore the energy dose needed to induce such an effect is higher in our samples. In Fig. 1 we report the signal measured versus the irradiation time (that is, for an increasing light dose). The observed signal is due to rotation of probe light polarization consequent to liquid crystal director reorientation [16,17]. Curve (a) is obtained with no applied voltage, while Curve (b) is obtained with an ac voltage applied (V = 0.2 Vrms ) at 1 kHz. It is worth noting that the applied voltage is below the threshold for the electric Freedericksz transition (measured as V = 0.9 Vrms ), then it is not able to induce any additional director reorientation. At V = 0, initially we observe an increase of the signal followed by a decrease to zero and then a uniform increase up to a saturation value obtained for energy densities higher than 1.3 kJ/cm2 . This behavior corresponds to the one
(b)
0
200
400
600
800
1000 1200 1400 1600 1800
E (J/cm2)
FIG. 1. Transmitted probe signal versus impinging light dose. Curve (a) is obtained with no applied voltage, while curve (b) is obtained with an ac voltage applied (V = 0.2 Vrms ) at 1 kHz. The vertical dashed line indicates the switching off of the pump beam. Inset: Detail of the region between 0 and 200 J/cm2 .
reported already several years ago in similar samples where the irradiated surface included a coating by PVCN-F [17]. At V = 0.2 Vrms the signal is zero up to an energy density of E = 1.1 kJ/cm2 , and it increases in correspondence to a change of slope of the signal at V = 0 V. Up to this value of the light dose the behavior is the one expected by our model since in the absence of space charge density (ac field applied) we do not observe any director reorientation. What happens above this value of energy density becomes clear if one takes into account the correspondent rotation of the liquid crystal director as is discussed below. By switching off the pump beam (vertical dashed line in Fig. 1) we observe a slight initial increase of the signal for V = 0, signal that keeps otherwise constant in time stabilizing to a value correspondent to the one before the switching off of the pump beam. We also observe a decay to zero of the signal for Vac > 0 . As mentioned, by checking the polarization rotation of the beam transmitted by the sample it becomes easy to understand the origin of the signal reported in Fig. 1. In fact in our experimental configuration due to the adiabatic condition the exit probe beam polarization follows the director orientation at the irradiated surface; therefore by rotation of the exit polarizer one can find again the extinction condition and determine the amount and the sign of rotation, which is taken as positive if towards the pump beam polarization and negative if outwards from the pump beam polarization. After determining the sign, the amount of rotation can actually be easily obtained from the data of Fig. 1 just applying the Malus law [18]. In Fig. 2 we report the reorientation angle correspondent to the two conditions reported in Fig. 1 [curve (a) V = 0 V and curve (b) V = 0.2 Vrms ]. Curve (a) shows the well known behavior already reported [17] where an initial negative reorientation is followed by a positive reorientation that becomes permanent, saturating at a certain value. On the contrary when a small ac field is applied, as is already clear in Fig. 1, we do not get any reorientation until E = 1.1 kJ/cm2 ; then we observe
032507-2
ROLE OF SPACE CHARGES ON LIGHT-INDUCED . . .
PHYSICAL REVIEW E 89, 032507 (2014)
off
10
20
(a)
8
6 10
τ (s)
φ (deg)
15
4 5
(b)
2 0
0
200
400
600
800
0
1000 1200 1400 1600 1800
0
E (J/cm2)
20
40
60
E
80
100
120
(J/cm2)
FIG. 2. Reorientation angle correspondent to the signal reported in Fig. 1. Again the vertical dashed line indicates the switching off of the pump beam.
FIG. 3. Relaxation time vs impinging energy dose. The solid line is a guide for the eyes. The vertical dashed line indicates the boundary between transient and permanent light-induced reorientation.
an increasingly negative one, which decays to zero when the pump beam is switched off. In correspondence with the pump switching off the positive reorientation occurring at V = 0 V gets a slight increase, in perfect agreement with results reported by the group of Reznikov [15]. These observations point out the role of photoisomerization of the azo molecules. In fact Fig. 2 shows that the signal observed when the ac voltage is applied (above E = 1.1 kJ/cm2 ) is due to a reorientation outward from the pump polarization that is not permanent; therefore it is consequent to photoisomerization of bulk molecules, an effect that disappears when the pump is switched off. We should underline that it is the first time that the photoisomerization effect in dye-doped liquid crystals is observed not overlapped on other phenomena that are here quenched by the ac field. Our observation makes clear that this process becomes relevant only at high doses, while surface effects are the dominant ones at lower energy densities. Additional information on the overall process is obtained by looking at the lifetime of the induced reorientation; this helps in distinguishing between transient reorientation that leads to nonlinear optical response and permanent reorientation that leads to light-induced anchoring. Measurements were performed at V = 0 V, using a pump power of P = 18 μW, and switching off the pump beam at increasing values of the impinging energy density. Then the relaxation time was evaluated by an exponential fit of the relaxation curves. The relaxation time increases with the irradiation dose, starting from 1 s at E = 8 J/cm2 up to 10 s at E = 104 J/cm2 . For higher values of the impinging energy density, the signal does not relax to zero within 30 min from switching off the pump beam. Therefore we can speak of transient reorientation up to an energy density of 104 J/cm2 , while above this value reorientation approaches a condition that is stable in time. In Fig. 3 a dashed line located at about 110 J/cm2 shows the asymptotic behavior towards the permanent reorientation. The continuous line is a guide to the eye showing the diverging increase of the reorientation lifetime. Comparing
these results to data reported in Figs. 1 and 2, it becomes clear that we get a transient reorientation when driven by lightinduced desorption, while we get a long-lasting to permanent reorientation when driven by light-induced adsorption. The behavior of the photoinduced reorientation has also been tested on an external dc field. According to our interpretation, the application of a dc field might enhance the signal, provided it has the right sign. If the observed light-induced effects in MR doped 5CB are all due to the presence of the charge complex present in the mixture, a dc field of the right polarity should enhance the effects of adsorption and desorption of these complexes, whereas a field of the opposite polarity should prevent the effect from occurring. In this respect, measurements with dc fields of different polarities should also give information about the sign of the charge complexes. Figure 4 reports the probe signal transmitted by the cell between crossed polarizers, in the case of positive (a) and negative (b) polarity of the dc field applied to the irradiated surface. The field is again lower than the threshold for the electric Freedericksz transition (note that the measured threshold in the case of the dc field is higher than the one measured with the ac voltage). For positive polarity the measured signal is comparable with the one obtained at zero field reported in Fig. 1 [curve (a)], being slightly enhanced with respect to it. If the field polarity is reversed, the signal has a different behavior that is quite similar to that observed in Fig. 1 [curve (b)] for the ac field. The two curves are also compared to the signal detected for V = 0 V [curve (c)]. The data of Fig. 4 are in good agreement with the proposed model: the application of a positive dc field slightly enhances the overall effect of light-induced desorption and adsorption of dye molecules on the irradiated surface, while the application of a negative dc field quenches both these phenomena and leaves the photoisomerization unperturbed. From these results we have the additional information that negative charge complexes are responsible for the light-induced desorption and adsorption processes.
032507-3
L. LUCCHETTI AND F. SIMONI
PHYSICAL REVIEW E 89, 032507 (2014)
form, which tends to align the surface director parallel to the easy axis. In this way the total surface free energy density becomes 1 ε LD 1 (nˆ s · Es )2 − W (nˆ S · πˆ )2 , (2) FS = − 2 2 2
(a)
S (arb.units)
(c)
where ε = ε// − ε⊥ , LD is the Debye screening length, and ns is the surface director. This term must be added to the intrinsic surface energy having the usual Rapini-Papoular
with W being the anchoring strength and π the easy axis. The surface field Es is connected to the surface charge density and this leads to a dependence of the anchoring energy on it. The observed light-induced reorientation outward and toward the pump light polarization due to desorption and adsorption of charged complexes corresponds to transient and permanent modification of σ and leads directly to a change of the anchoring energy that can be transient or permanent depending on the energy dose impinging on the cell (i.e., depending on the amount of the modification of σ ). We want to summarize here the main results of the present report. First of all we have demonstrated by a very simple measurement that the light-induced processes widely investigated for more than a decade on nematic liquid crystals doped by methyl red have the same origin, involving a basic role played by charges on the irradiated surface. Namely, the only difference between the transient response leading to huge nonlinear response and light-induced permanent anchoring is the amount of light dose. Linked to this demonstration is the definite observation that only light-induced adsorption leads to a permanent director reorientation, while the desorption effect has a limited recovery time, being sufficiently small at low light intensity to allow a huge nonlinear optical response. Another important observation is that the process in this compound involves negative charges. This fact should be cleared by a more detailed investigation on the photoinduced charge generation in these types of materials. Finally we have confirmed the limited role of photoisomerization in these processes occurring only at high dose levels and we have isolated this effect from the other ones by quenching them via ac field application. We point out that the features of the light-induced effects occurring in the studied compound can be envisaged to be typical of the large class of donor-acceptor azobenzene compounds. Actually a systematic investigation of them to look for optimization of the discussed phenomena is still missing.
[1] M. M. Chrzanowski, J. Zielinski, M. Olifierczuk, J. Kedzierski, and E. Nowinowski-Kruszelnicki, J. Achiev. Mater. Manuf. Eng. 48, 7 (2011). [2] L. Lucchetti, M. Di Fabrizio, O. Francescangeli, and F. Simoni, Opt. Commun. 233, 417 (2004). [3] E. Ouskova, D. Fedorenko, Y. Reznikov, S. V. Shiyanovskii, L. Su, J. L. West, O. V. Kuksenok, O. Francescangeli, and F. Simoni, Phys. Rev. E 63, 021701 (2001). [4] E. Ouskova, Y. Reznikov, S. V. Shiyanovskii, L. Su, J. L. West, O. V. Kuksenok, O. Francescangeli, and F. Simoni, Phys. Rev. E 64, 051709 (2001). [5] L. Lucchetti, M. Di Fabrizio, O. Francescangeli, and F. Simoni, J. Nonlinear Opt. Phys. Mater. 11, 13 (2002).
[6] L. Lucchetti, D. E. Lucchetta, O. Francescangeli, and F. Simoni, Mol. Cryst. Liq. Cryst. 375, 641 (2002). [7] R. Ramos-Garcia, I. Lazo-Martinez, I. Guizar-Iturbide, A. Sanchez-Castillo, M. Boffety, and P. Ruck, Mol. Cryst. Liq. Cryst. 454, 179 (2006). [8] V. Boichuk, S. Kucheev, J. Parka, V. Reshetnyak, Y. Reznikov, I. Shiyanovskaya, K. D. Singer, and S. Slussarenko, J. Appl. Phys. 90, 5963 (2001). [9] J. Zhang, V. Ostroverkhov, K. D. Singer, V. Reshetnyak, and Y. Reznikov, Opt. Lett. 25, 414 (2000). [10] D. Fedorenko, E. Ouskova, V. Reshetnyak, and Y. Reznikov, Phys. Rev. E 73, 031701 (2006).
(b)
0
200
400
600
800
1000 1200 1400 1600 1800
E (J/cm2)
FIG. 4. Probe signal transmitted by the cell between the crossed polarizers, in the cases of positive (a) and negative (b) polarity of the dc field applied to the irradiated surface. The curves are also compared to the one obtained with no external field (c).
The presence of charged complexes in dye-doped LC samples has been directly observed by measurements of photocurrent [19] and photovoltage [20,21] by different authors. Moreover, the effects of charged objects on surface anchoring energy has also been studied several years ago [22,23]. From these works it is known that when a solid substrate is in contact with a nematic LC selective ion adsorption takes place. This creates an electric surface field in liquid crystal cells, which extends in the bulk over the Debye screening length [23]. The consequent surface field, Es , strongly affects the surface anchoring energy of the nematic director. This happens because the coupling of the surface field Es with the nematic director gives rise to an additional term to the surface free energy density of the form [23]: 1 ε LD FE = − (nˆ s · Es )2 , (1) 2 2
032507-4
ROLE OF SPACE CHARGES ON LIGHT-INDUCED . . .
PHYSICAL REVIEW E 89, 032507 (2014)
[11] L. Lucchetti, M. Gentili, and F. Simoni. Opt. Express 14, 2236 (2006). [12] L. Lucchetti, M. Gentili, F. Simoni, S. Pavliuchenko, S. Subota, and V. Reshetnyak, Phys. Rev. E 78, 061706 (2008). [13] L. Lucchetti, M. Gentili, and F. Simoni. IEEE J. Sel. Top. Quantum Electron. 12, 422 (2006). [14] J. Merlin, E. Chao, M. Winkler, and K. D. Singer, Opt. Exp. 13, 5024 (2005). [15] D. Fedorenko, K. Slyusarenko, E. Ouskova, V. Reshetnyak, K. R. Ha, R. Karapinar, and Y. Reznikov, Phys. Rev. E 77, 061705 (2008). [16] D. Voloshchenko, A. Khizhnyak, Y. Reznikov, and V. Reshetyak, Jpn. J. Appl. Phys. 34, 566 (1995).
[17] O. Francescangeli, S. Slussarenko, F. Simoni, D. Andrienko, V. Reshetnyak, and Y. Reznikov, Phys. Rev. Lett. 82, 1855 (1999). [18] F. A. Jenkins and H. E. White, Fundamentals of Optics (McGraw-Hill, New York, 1976), p. 1746. [19] E. V. Rudenko and A. V. Sukov, J. Electrochem. Plat. Technol. 78, 875 (1994). [20] S. Sato, Jpn. J. Appl. Phys. 20, 1989 (1981). [21] I. C. Khoo, S. Slussarenko, B. D. Guenther, M.-Y. Shih, P. Chen, and W. V. Wood, Opt. Lett. 23, 253 (1998). [22] A. V. Zakharov and R. Y. Dong, Phys. Rev. E 64, 042701 (2001). [23] G. Barbero and G. Durand, J. Appl. Phys. 67, 2678 (1990).
032507-5
