Observation of giant local photoinduced birefringence in Ge25As30S45 thin films

K. Palanjyan; R. Vallée; T. Galstian

doi:10.1364/OME.5.001122

1. Introduction

Chalcogenide glass (ChG) films are the subject of numerous investigations due to their potential applications in photonics based on their wide transmission range (from visible to IR) and high photosensitivity [1–5]. Integrated optics devices that use ChG thin films are particularly interesting for the fabrication of high-index-contrast planar waveguide-coupled microdisk resonators, planar waveguides integrated on a photonic chip, submicrometer-thick low-loss waveguides, photonic circuits, which are capable to process optical data streams entirely in the optical domain, etc [6–12].

In many such guided optical devices (particularly those manufactured by photoexposition or photo patterning) the local birefringence is an important parameter, which has to be carefully considered due to its role played in the polarization mode dispersion. Thus, spectral broadening was already studied in some photonic chips and attributed to a non-uniform birefringence [12]. The control of the polarization modes was described in similar devices by using the photoinduced birefringence (PIB) [13]. The PIB in ChG materials was studied by many research groups (see for instance [14, 15]). However, to the best of our knowledge, all previous reports are based on the use of ChG films with thicknesses L that are noticeably larger than the typical penetration depth l of the excitation light. The typical values of l are at the micrometer scale, which, by the way, is of the same order of magnitude as the ChG film thicknesses used in integrated optical systems. As a result, the obtained values of PIB in research laboratories are spatially averaged and do not represent the real values that would be generated in photo patterned thin integrated circuits employing ChG films.

The aim of the present work is to investigate the local value of the PIB. To that purpose we have chosen the photosensitive Ge-As-S glass family, known for its structural stability. Indeed, the incorporation of germanium Ge into arsenic sulfide As-S glass enhances the material properties by increasing its network connectivity, resulting thus in a three-dimensional structure and a large increase of glass transition temperature (Tg ~350°C as compared to ~180°C for As-S glasses).

Several studies of the above-mentioned photo-induced scalar phenomena in this Ge-As-S vitreous system have been reported in the literature [16, 17]. Figure 1 shows the transmission spectra of annealed thin (of 3µm thickness) films of Ge₂₅As₃₀S₄₅ before (black dotted curve) and after (red solid curve) irradiation with bandgap light. Depending upon the fact if the samples were as-prepared or annealed, the studies revealed both photo-induced bleaching and darkening, respectively. Moreover, a unique non-monotonous phenomenon, the coexistence of fast photodarkening and slow photobleaching was also observed in Ge-As-S glasses [18].

Fig. 1 Transmission spectra (in non-polarized light) of non-exposed (black, dotted curve) and exposed (red, solid curve) Ge₂₅As₃₀S₄₅ thin films of 3 µm thickness. Samples were irradiated at 514 nm for 30 min.

Download Full Size | PDF

In addition, some preliminary investigations of PIB and photoinduced dichroism (PID) phenomena have already been conducted in similar compounds [19, 20]. In a recent work, we have studied the spatially averaged PID in Ge₂₅As₃₀S₄₅ films and proposed a model of consecutive bond transitions to explain its dynamics [21]. Here, we show that the local value of PIB in Ge₂₅As₃₀S₄₅ thin films could be an order of magnitude higher than its previously reported average value [19].

2. Experimental set-up and procedure

Bulk glass samples of Ge₂₅As₃₀S₄₅ composition were first synthesized by the traditional melt-quenching method in silica ampoule sealed under vacuum. The thin films (with thickness over 3 µm) were prepared by e-beam evaporation technique, as described in details in a previous report [19]. All experiments were conducted on thin films, which were annealed at 350°C during 1 hour and slowly cooled down to room temperature in ambient atmosphere.

The experimental setup, used to study the PIB, is schematically represented in Fig. 2.The photosensitivity was studied by exciting the ChG film by s-polarized continuous wave (CW) Argon ion laser (operating at 514.5 nm) at normal incidence on the ChG film (from air-ChG film side, x < 0, Fig. 2). Samples were exposed with different intensities varying from 2 W/cm² to 15 W/cm². The power of this excitation (or pump) beam was controlled by filters (F₁). A Glan prism polarizer (P) was used to obtain linear polarization. A CW He–Ne laser beam (operating at 632.8 nm) at incidence angle γ was used as a probe to analyze the PIB.

Fig. 2 Experimental setup used for the study of PIB: P-polarizer, M–mirror, λ/2- half-wave plate, S-sample; A-analyzer; F₁ and F₂-filters, d-diaphragm, D-detector.

Download Full Size | PDF

The angle γ between incident pump and probe beams was $\approx 10^{°}$ . We have used a half-wave plate (λ/2) to align the linear polarization of the probe beam at 45° with respect to the linear polarization of the pump beam (the diameters of the probe and pump beams at the surface of the ChG film were ∅_pr ≈0.3 mm and ∅_p ≈2 mm, respectively). The two laser beams were aligned to overlap on the sample (S) with the help of a dielectric mirror (M). After passing through the sample, the probe beam was filtered by an analyzer (A) that was oriented at 50° (see hereafter) with respect to the original polarization of the probe beam (defined by the λ/2 plate). The intensity of the probe beam was set as low as possible (≈3.5 mW/cm²) to prevent any photo induced change. An interferential filter (F₂) was used to transmit only the wavelengths above 600 nm (to reduce the noise from the pump beam). A diaphragm (d) was also used in front of the photo detector (D) to further reduce noise.

It is worth mentioning that we might use multiple optical arms or a rotating polarizer (instead of a fixed one) to obtain more information. However, the use of a simple traditional polarimetric setup was preferred here to compare easily the obtained PIB results with those previously reported.

3. Results and discussions

The transmission spectrum (recorded on a Varian Cary 50 spectrophotometer) of the unexposed Ge₂₅As₃₀S₄₅ thin film exhibits typical Fabry-Perot oscillations that become noticeable for $λ$ larger than 600 nm [19]. Thus, in our study of PIB, we have to consider not only the transmission, but also the reflection of the probe. In addition, the transmitted probe beam intensity is relatively low due to the high absorption of the material at 632.8 nm. For this reason, instead of using an analyzer that is strictly perpendicular with respect to the original probe’s polarization, we aligned them at 50° with respect to each other. This helped to increase the sensitivity of the set-up. Furthermore, we have normalized the transmission (after the analyzer) by the values of measured transmission without analyzer (see typical curves on Fig. 3). Finally, we have used the approximation of small photo induced dichroism (at the probe wavelength). As we can see from our calculations (described below), these operations allow the elimination of contributions (in the value of the PIB) from the Fabry-Perot oscillations as well as from the material absorption [19].

Fig. 3 Transmitted intensity of the probe beam (3 μm thick sample is used versus time for pump intensity of 8W/cm2. The points 1 and 1’ represent the established values of excitation and 2 and 2’ represent the established values of relaxation corresponding to the probe transmission without (drawn by triangles in the fig.) and with analyzer (drawn by squares in the fig.), respectively.

Download Full Size | PDF

In general, the intensity of the probe beam, after the sample (before the analyzer), may be calculated (as $I_{1} = \vec{J_{1}} \times \vec{J_{1}^{*}}$ ) by using the Jones vector $\vec{J_{1}}$ of the probe [22], which may be presented in the first approximation as:

\vec{J_{1}} = \frac{\sqrt{2}}{2} \hat{T} [\begin{matrix} e^{- α_{∥} + i φ_{∥}} \\ e^{- α_{⊥} + i φ_{⊥}} \end{matrix}]

where

α_{∥, ⊥}

and

φ_{∥, ⊥}

are the photo induced parallel and perpendicular components (with respect to the pump beam’s polarization) of absorption and of the phase delay, respectively, and

\hat{T}

is the ChG film’s “transmission” matrix related to Fabry-Perot reflection losses. In general, those losses are polarization dependent. However, given their interferential nature (involving reflections from front and back interfaces of the ChG film) and the strong absorption of the film, we shall make an approximation of polarization independence of those loses to simplify the main demonstration of this work (the high local birefringence). We can thus present those losses by a scalar coefficient T.

In the same way, the intensity of the probe beam, after the analyzer, may be expressed (as $I_{2} = \vec{J_{2}} \times \vec{J_{2}^{*}}$ ) by the corresponding Jones vector $\vec{J_{2}}$ as:

\vec{J_{2}} = \frac{\sqrt{2}}{2} T [\begin{matrix} \cos θ \times e^{- α_{∥} + i φ_{∥}} - \sin θ \times e^{- α_{⊥} + i φ_{⊥}} \\ 0 \end{matrix}]

where

θ

is the angle between the original probe polarization and the analyzer.

It is easy to verify that we can exclude the influence of the Fabry-Perot losses in the calculation of the PIB if its polarization dependence is small (neglected) and if we use the following normalized form of probe intensity:

I_{n o r m} = \frac{{\hat{J}}_{2} \times {\hat{J}}_{2}^{*}}{{\hat{J}}_{1} \times {\hat{J}}_{1}^{*}} = \frac{\cos^{2} θ \times e^{- 2 α_{∥}} + \sin^{2} θ \times e^{- 2 α_{⊥}} - 2 \times \cos θ \times \sin θ \times e^{- (α_{∥} + α_{∥})} \times \cos (φ_{∥} - φ_{⊥})}{e^{- 2 α_{∥}} + e^{- 2 α_{⊥}}}

As the wavelength of He-Ne laser beam is relatively far from the ChG’s band gap, in the first approximation, the PID (at 632.8 nm) can be neglected also

(α_{∥} \approx α_{⊥})

to further simplify the analyses. This hypothesis is confirmed by the experimental results reported in Ref [21], which show that the photoinduced absorption (often called “photo darkening” or PD) in these materials may be rather significant, while the PID is very low at this probe wavelength

(\frac{α_{∥}}{α_{⊥}} \approx 1, 05)

. Thus, we can see that the normalized transmitted intensity value depends only upon the birefringence:

I_{n o r m} = 0.5 (1 - \sin 2 θ \times \cos Δ φ)

Δ φ = arc \cos (\frac{1 - 2 \times I_{n o r m}}{\sin 2 θ})

Where

Δ φ = φ_{∥} - φ_{⊥}

, which, in the case of a uniform PIB

Δ n

, is

Δ φ = \frac{2 π Δ n L}{λ}

.

The corresponding experimental study was performed on ChG thin films of 3 µm thickness with pump intensity of 8W/cm². Figure 3 shows the typical dynamics of excitation and partial relaxation process.

We can clearly observe (Fig. 3) that, with switching-on the pump beam, the probe beam’s transmission first sharply increases and then decreases very slowly. We observe a partial recovery of the transmitted light when the excitation beam is switched-off, leading to a significant remnant PIB.

Based on the Eq. (4)’), we can calculate the average (along the thickness of the film) birefringence $Δ n = \frac{λ Δ φ}{2 π L}$ by using the measured normalized intensity. This last one is obtained (from the experimental data, presented in Fig. 3) by dividing the established value of transmitted intensity recorded with analyzer (point 2) by the one recorded without analyzer (point 2’). Furthermore, these values of PIB can be calculated (using our experimental data) for various pump intensities and thus plotted as a function of pump intensity, as presented in Fig. 4.As one can see, the slope of the PIB decreases with increasing the excitation intensity and becomes negative above 8 W/cm². This behavior confirms our previous results, which have pointed out that the PIB is created by a limited number of pre-existing photosensitive units, but not newly photo created ones [19].

Fig. 4 Dependence of the established values of PIB (under CW excitation) upon the excitation intensity for the 3 μm Ge₂₅As₃₀S₄₅ thin film. The line is used to guide eyes only.

Download Full Size | PDF

It is worth to mention that no change of thickness after exposition was observed by DekTak profilometer study (within the error limit: ± 0.01nm), excluding the possibility of photo-induced expansion or depression, which could affect our calculations of the PIB values.

Our additional studies of the PD and PID also show the same linear character of excitation in the low excitation range of the pump intensities [21]. Indeed, we can see that, for the 3 μm Ge₂₅As₃₀S₄₅ thin film, the maximal average value of the PIB is Δn_av = 0.03 ± 0.003.

In fact, as mentioned in the introduction, we should also consider the exponential decrease of the pump intensity in the material. Thus, the relative phase delay (due to the local PiB) will also change in the material with an exponential law and for a given thickness (L = 3 μm) and initial intensity, the differential phase modulation of the probe beam will be:

Δ φ = \frac{2 π}{λ_{0}} \int_{0}^{L} Δ n (z) d z

where

Δ n (z)

is the value of the local anisotropy at z.

As we can see from Fig. 4 and Ref [19, 21], for the range of low pump intensities, both PID $(Δ α)$ and PIB $(Δ n)$ are linearly proportional to the input pump intensity:

Δ n = μ I (z)

where

μ

is the coefficient of proportionality that we shall use later as a fit parameter.

Therefore, the output probe intensity, corresponding to $Δ φ$ , will be expressed as:

I_{o u t} = I_{0} \sin^{2} (\frac{Δ φ}{2}) = I_{0} \sin^{2} (\frac{\frac{2 π}{λ_{0}} \int_{0}^{L} μ I (z) d z}{2})

In our experiments, we have measured the output probe’s intensity dependence (measured after the analyzer) upon the input pump intensity values, as presented in Fig. 5.Furthermore, we have adjusted the μ parameter (in the Eq. (6)) so that the calculated value of the normalized output probe intensity (z = 3 µm) becomes equal to the measured one (see Fig. 5).

Fig. 5 Experimentally measured dependence of the established output probe intensity upon the input pump intensity for the 3 μm Ge₂₅As₃₀S₄₅ thin film. The line is used to guide eyes only.

Download Full Size | PDF

The optimal values of the parameter μ were found for all pump intensities. Then, we have calculated the local value of $Δ n$ (Eq. (5) and Eq. (7)) by using the obtained values of μ.

Figure 6 shows the obtained dependences of the average value of the PIB and its local maximum values calculated at the “input” front of the ChG film) for different pump intensities (incident from the same “input” side).

Fig. 6 Average (solid curve) and local maximum (at the input front of the ChG film, dotted curve) values of the established PIB as a function of pump intensity for the 3 μm Ge₂₅As₃₀S₄₅ thin film. Solid and dashed lines are used to guide eyes only.

Download Full Size | PDF

As we can see from Fig. 6, the obtained local value of the PIB is almost one order of magnitude higher (maximum value achieved here is Δn_loc > 0.112) than the spatially averaged value that is usually reported in earlier studies.

4. Summary and conclusion

By simple polarimetric experiments and theoretical modeling (with an approximation of weak PID and weak polarization dependence of Fabry-Perot reflections), we have shown that the local value of the PIB in Ge₂₅As₃₀S₄₅ chalcogenide thin films may be an order of magnitude higher than its average value, usually reported in the literature. This should be taken into account during the design of new photo patterned integrated optics devices (channel waveguides, gratings, etc.) based on ChG thin films.

Acknowledgments

We acknowledge the financial support of Canadian Institute for Photonic Innovations (CIPI), Fonds Québécois de la Recherche sur la Nature et les Technologies (FRQNT) and Natural Sciences and Engineering Research Council of Canada (NSERC). We also thank Lens vector/TLCL Optical Research Inc. for their valuable advices.

References and links

1. C. H. Kwak, J. T. Kim, and S. S. Lee, “Nonlinear optical-image processing in amorophous As₂S₃ thin films,” Appl. Opt. 28, 737–739 (1989). [CrossRef] [PubMed]

2. H. M. Kim, J. W. Jeong, C. H. Kwak, and S. S. Lee, “Binary phase spatial modulation using photoinduced anisotropy in amorphous As₂S₃ thin film,” Appl. Opt. 34(26), 6008–6011 (1995). [CrossRef] [PubMed]

3. K. Tanaka, A. Saitoh, and N. Terakado, “Giant photo-expansion in chalcogenide glass,” J. Optoelectron. Adv. Mater. 8, 2058–2065 (2006).

4. T. V. Galstyan, J. F. Viens, A. Villeneuve, K. Richardson, and M. A. Duguay, “Photoinduced self-developing relief gratings in thin film chalcogenide As₂S₃ glasses,” J. Lightwave Technol. 15(8), 1343–1347 (1997). [CrossRef]

5. K. Tanaka and K. Ishida, “Photoinduced anisotropic structure in chalcogenide glasses,” J. Non-Cryst. Solids 227, 673–676 (1998). [CrossRef]

6. S. Spälter, H. Y. Hwang, J. Zimmermann, G. Lenz, T. Katsufuji, S. W. Cheong, and R. E. Slusher, “Strong self-phase modulation in planar chalcogenide glass waveguides,” Opt. Lett. 27(5), 363–365 (2002). [CrossRef] [PubMed]

7. D. Y. Choi, S. Madden, D. A. Bulla, R. Wang, A. Rode, and B. Luther-Davies, “Submicrometer-Thick Low-Loss As₂S₃ Planar Waveguides for Nonlinear Optical Devices,” IEEE Photon. Technol. Lett. 22(7), 495–497 (2010). [CrossRef]

8. J. Van Erps, F. Luan, M. D. Pelusi, T. Iredale, S. Madden, D. Y. Choi, D. A. Bulla, B. Luther-Davies, H. Thienpont, and B. J. Eggleton, “High-Resolution Optical Sampling of 640-Gb/s Data Using Four-Wave Mixing in Dispersion-Engineered Highly Nonlinear As₂S₃ Planar Waveguides,” J. Lightwave Technol. 28(2), 209–215 (2010). [CrossRef]

9. J. J. Hu, N. Carlie, N. N. Feng, L. Petit, A. Agarwal, K. Richardson, and L. Kimerling, “Planar waveguide-coupled, high-index-contrast, high-Q resonators in chalcogenide glass for sensing,” Opt. Lett. 33(21), 2500–2502 (2008). [CrossRef] [PubMed]

10. A. Ganjoo, H. Jain, C. Yu, R. Song, J. V. Ryan, J. Irudayaraj, Y. J. Ding, and C. G. Pantano, “Planar chalcogenide glass waveguides for IR evanescent wave sensors,” J. Non-Cryst. Solids 352(6-7), 584–588 (2006). [CrossRef]

11. M. L. Anne, J. Keirsse, V. Nazabal, K. Hyodo, S. Inoue, C. Boussard-Pledel, H. Lhermite, J. Charrier, K. Yanakata, O. Loreal, J. Le Person, F. Colas, C. Compère, and B. Bureau, “Chalcogenide Glass Optical Waveguides for Infrared Biosensing,” Sensors (Basel) 9(9), 7398–7411 (2009). [CrossRef] [PubMed]

12. B. J. Eggleton, C. G. Poulton, and R. Pant, “Inducing and harnessing stimulated Brillouin scattering in photonic integrated circuits,” Adv. Opt. Photonics 5(4), 536–587 (2013). [CrossRef]

13. A. Yesayan and T. Galstyan, “Mode control in integrated optical devices via photoinduced birefringence,” IEEE J. Quantum Electron. 35(8), 1142–1145 (1999). [CrossRef]

14. G. J. Adriaenssens, V. K. Tikhomirov, and S. R. Elliott, “Mechanism and kinetics of photoinduced anisotropy in chalcogenide glasses,” J. Non-Cryst. Solids 227, 688–693 (1998).

15. K. Tanaka, K. Ishida, and N. Yoshida, “Mechanism of photoinduced anisotropy in chalcogenide glasses,” Phys. Rev. B Condens. Matter 54(13), 9190–9195 (1996). [CrossRef] [PubMed]

16. D. Arsova, E. Skordeva, V. Pamukchieva, and E. Vateva, “Photoinduced changes in Ge-As-S thin films with various network rigidities,” J. Optoelectron. Adv. Mater. 7, 1259–1264 (2005).

17. M. Klebanov, V. Lyubin, D. Arsova, E. Vateva, and V. Pamukchieva, “Photoinduced anisotropy in photobleached Ge-As-S films,” Phys. B 301(3-4), 399–404 (2001). [CrossRef]

18. P. Khan, A. R. Barik, E. M. Vinod, K. S. Sangunni, H. Jain, and K. V. Adarsh, “Coexistence of fast photodarkening and slow photobleaching in Ge₁₉As₂₁Se₆₀ thin films,” Opt. Express 20(11), 12416–12421 (2012). [CrossRef] [PubMed]

19. K. Palanjyan, S. H. Messaddeq, Y. Messaddeq, R. Vallee, E. Knystautas, and T. Galstian, “Study of photoinduced birefringence vs As content in thin GeAsS films,” Opt. Mater. Express 3, 671–683 (2013).

20. M. Klebanov, V. Lyubin, D. Arsova, E. Vateva, and V. Pamukchieva, “Photoinduced anisotropy in photobleached Ge-As-S films,” Phys. B 301(3-4), 399–404 (2001). [CrossRef]

21. K. Palanjyan, R. Vallee, and T. Galstian, “Photoinduced bond conversions in GeAsS thin films,” J. Non-Cryst. Solids 410, 65–73 (2015). [CrossRef]

22. R. Clark Jones, “A new calculus for the treatment of optical systems V. A more general information, and description of another calculus,” J. Opt. Soc. Am. 37(2), 107 (1947). [CrossRef]

Observation of giant local photoinduced birefringence in Ge₂₅As₃₀S₄₅ thin films

Abstract

1. Introduction

2. Experimental set-up and procedure

3. Results and discussions

4. Summary and conclusion

Acknowledgments

References and links

Cited By

Figures (6)

Equations (8)

Optical Materials Express