Photo-darkening and photo-bleaching are well known phenomena in As-Se and Ge-Se chalcogenide glasses, respectively. Consequently, a systematic dependence of photo-induced optical changes in GexAs45-xSe55 glass series on x is expected between these two extremes. This prediction of photosensitivity on Ge/As ratio has been exploited to demonstrate the first intrinsically photo-stable chalcogenide glass at x~10, which would be suitable for fabricating photo-insensitive optical components for various applications.
©2008 Optical Society of America
Infrared (IR) transmitting chalcogenide glasses (ChG) are known for their photosensitivity to band-gap radiation, which produces several types of photo-induced changes in structure and properties [1,2]. Whereas photosensitivity is useful in applications such as optical writing, photolithography, etc. , a photo-stable glass, which is not easily affected by exposure to light, is needed for many applications in IR optics. Indeed, potential applications of ChG have been restricted due to the drift in their optical properties with exposure to light. Although it was noted that the photo-darkening of amorphous chalcogenide films is influenced by doping with Cu or Sn impurities, little attention has been devoted to optimize intrinsic ChG coomposition against such unwanted photo-induced changes [4,5]. Accordingly, in the present work we have investigated, under in situ band-gap radiation, the changes in the transmission spectra of GexAs45-xSe55 glasses. Note that the extreme members of this pseudo-binary ChG series exhibit opposite photo-responses viz. photo-darkening in binary As-Se (i.e. x=0) and photo-bleaching in some Ge-Se compositions (e.g. x=33) . As a result, the first intrinsic ChG composition that is not affected by band-gap laser irradiation is discovered.
GexAs45-xSe55 compositions form stable glasses as they are located in the center of Ge-As-Se glass forming region . Samples of bulk GexAs45-xSe55 with x=0, 10, 20 and 33 were prepared by the melt-quench method using 99.999% pure germanium, arsenic and selenium. The elements were vacuum sealed in quartz ampoules and gradually heated to 850 °C and rocked for 10 h. The melt was then quenched in ice water to form bulk glass. It was used as the source material for depositing thin film on microscope glass slide substrate by vacuum evaporation (Edwards Coating System E306A) in a vacuum of about 1×10-6 Torr. The deposition rate was ~10 Å/s, continuously measured using a quartz crystal rate/thickness monitor (Sigma Instruments SQM-160). The total thickness of the film was 1.0 µm, which is approximately the penetration depth of band-gap light for our samples. It is well known that such a low deposition rate produces film composition that is very close to the starting bulk materials .
Photoinduced optical changes in the samples were determined as a function of time by in situ optical spectroscopy, as described previously . The thin film samples were illuminated at room temperature within 5 mm diameter spot with a diode laser of wavelength, λ=660 nm and intensity=150 mW/cm2. The diameter of the relatively weak probing white light spot from the spectrometer was 2 mm. The two beams were directed such that they crossed each other at the sample with the diode laser spot completely overlapping the light spot from the spectrometer. The data before and during laser illumination were collected every 10 milliseconds. It should be noted that such a small exciting laser intensity and large beam spot on the sample would result in a relatively insignificant temperature increase during illumination . Therefore the observed changes are not due to temperature rise but mainly due to photo-effects. Our interest was primarily in the absorption at wavelengths near the band gap, where photoinduced effects are expected to be the largest. Therefore, simultaneously with the full spectrum, transmission signals were also recorded at six fixed wavelengths within the region of steepest increase in absorption, using different channels of the spectrometer, for example at 590, 600, 610, 620, 630 and 635 nm for the sample with x=10.
3. Results and discussion
Figure 1 shows a three-dimensional plot of transmitted intensity for GexAs45-xSe55 films after laser is turned on as a function of time and wavelength in the vicinity of absorption edge where interference fringes are absent (λ=550 to 630 nm). Figure 2 shows simpler, two-dimensional transmission spectra as a function of wavelength before and during illumination of the various samples, and as a function of time at indicated wavelengths. To assess the effect of probe light, spectra are recorded for the ‘dark condition’ i.e. without the pump laser. As seen in Fig. 2, no change in the spectra is observed until the pump laser is turned on, thus confirming that the probe light has no impact on the optical transmission. Note that there is no significant change in the transmission of all the samples for 10 seconds or so after the laser is turned on.
It is seen from Fig. 1 that after a latent period of no change the film with x=0 shows monotonous shift of the absorption edge to longer wavelengths with laser irradiation time. The films with x=22 and 33 reveal complex changes, that is, a slight red shift at the beginning, followed by a remarkable blue shift with irradiation time. It takes 11 hours for the x=22 sample before reaching saturation of blue shift or photo-bleaching; the blue shift continues in the x=33 sample even after illumination for 20 hours (Fig. 3), which indicates photo-bleaching rate is much slower than photo-darkening, as well as dependent on Ge concentration. The sequential photo-darkening and photo-bleaching behaviors show that the As-Se and Ge-Se parts of the glass network respond to pump laser rather independently of each other in these glasses. This inference is a new insight of the photoinduced phenomenon in chalcogenide glasses that have been believed to be structurally homogenous. By comparison, Fig. 2 shows that the photoinduced changes in the film with x=10 are almost absent with the transmission change within 8.6% for the whole wavelength region from visible to near infrared, which is comparable to that found in As2Se3:Sn . Although the data in Fig. 3 are shown for λ=620 nm, a similar behavior of photoinduced changes in absorption is observed at other wavelengths in the vicinity of absorption edge.
The mechanism of photo-effects in the present glass series can be understood in terms of the compositional heterogeneities created during film formation from the vapor that contains various atomic fragments such as found in As-S system [11, 12]. Samples without Ge (x=0) or with a low content of Ge are over-stoichiometric with regard to As and are chemically metastable. During the exposure, the excess metal clusters may react as expressed by the shift of the following equilibrium, mostly from the left to right, resulting in the shift of absorption edge towards longer wavelength (photo-darkening) [13-15]:
The photo-bleaching process of samples (x=22, 33) with lower As content can be explained as follows. During the evaporation of GexAs45-xSe55 glasses, the vapors of GeSe2 are thermally dissociated into GeSeu (u<2) fragments as proposed by Spandau and Klanberg [15-17]. Under laser irradiation, heteropolar Ge-Se bonds are formed from less stable Ge-Ge/Se-Se homopolar bonds according to the following reaction:
Thus the energy gap of the layers is increased (Eopt(GeSe2)>Eopt(Ge2Se3, or GeSe)), and the chalcogenide films are photo-bleached.
The change in absorption coefficient, Δα, at various fixed wavelengths is calculated from Δα=(-1/d)ln(T/T0), where d is the thickness of the films and T/T0 is the transmitted signal relative to that of the initial signal (i.e., the signal measured before diode laser is switched on) [18, 19]. Although there is a possible change in reflection due to the change in refractive index during illumination, its magnitude would be too small to affect the values of absorption coefficient significantly . Quite likely the illumination may also cause an increase in the thickness (d) of the films, which may influence the calculation of the absorption coefficient. However, for the present discussion, we assume d to remain constant as the likely changes in d by illumination are small, compared to the changes in absorption coefficient . In any case, an increase in d will result in a decrease in absorption coefficient, which is opposite to that observed in photo-darkening (Δα increases in photodarkening).
Figure 3 shows an example of the time evolution of Δα calculated from the changes in transmission at various wavelengths in the absorption edge region. During illumination, Δα of the film with x=0 increases continuously at all the wavelengths before saturating, that of x=22 or 33 increases slightly at the beginning of laser irradiation, then decreases markedly, while that of x=10 undergoes relatively negligible increase-decrease-increase with time.
As noted above, photo-darkening/bleaching response of chalcogenide glasses is closely related to Ge/As ratio. The variation of absorption coefficient Δα versus Ge content in Fig. 4 illustrates this fact explicitly. Note that photo-induced optical changes in GexAs45-xSe55 glass series vary from photo-darkening (x=0) to photo-bleaching (x=22, 33). Most interestingly, the glass with x=10 shows negligible photosensitivity and the smallest Δα even after long time illumination.
Among the large number of chalcogenide glasses reported in literature, x=10 is the first glass composition that has shown remarkable intrinsic photo-stability i.e. without adding any dopants such as Sn. It is expected to be an optimum material for infrared optics for use under laser irradiation conditions. Further experiments are in progress to verify the stability of this glass over much longer time period of months, even years.
We thank the National Science Foundation of China (NSFC50702021) for its financial support and NSF’s International Materials Institute for New Functionality in Glass (IMI-NFG) (DMR-0409588) for initiating and supporting our international collaboration.
References and links
1. K. Shimakawa, A. Kolobov, and S.R. Elliott, “Photoinduced effects and metastability in amorphous semiconductors and insulators,” Adv. Phys. 44, 475–588 (1995). [CrossRef]
2. K. Tanaka, “Photoinduced structural change in chalcogenide glass,” Rev. Solid. St. Sci. 4, 641 (1990).
3. C. Meneghini and A. Villeneuve, “As2S3 photosensitivity by two-photon absorption: holographic gratings and self-written channel waveguides,” J. Opt. Soc. Am. B 15, 2946–2950 (1998). [CrossRef]
5. P. Boolchand, D. G. Georgiev, and M. S. Iovu, “Molecular structure and quenching of photodarkening in As2Se3: Snx amorphous,” Chalc. Lett. 2, 27–34 (2005).
6. A. Zakery and M. Hatami, “Nonlinear optical properties of pulsed-laser-deposited GeAsSe films and simulation of a nonlinear directional coupler switch,” J. Opt. Soc. Am. B 22, 591–597 (2005). [CrossRef]
7. R.L. Myuller, G.M. Orlova, V.N. Timofeeva, and G. I. Ternova, Solid State Chemistry, Ed. Z. U. Borisva, (Consultants Bureau, New York1966), p. 232.
8. A. Kovalskiy, A. C. Miller, H. Jain, and M. Mitkova, “In-situ measurements of x-ray induced silver diffusion into Ge30Se70 thin film,” J. Am. Ceram. Soc. 91, 760–765 (2008). [CrossRef]
9. K. Antoine, H. Jain, and M. Vlcek, “Optical spectroscopy of a-As2Se3 under in situ laser irradiation,” J. Noncryst. Solids. 352, 595–600 (2006). [CrossRef]
10. C. R. Schardt, P. Lucas, A. Doraiswamy, P. Jivaganont, and J. H. Simmons, “Raman temperature measurement during photostructural changes in GexSe1-x glass,” J. Non-Cryst. Solids. 351, 1653–1657 (2005). [CrossRef]
11. M. Frumar, Z. Polak, Z. Cemosek, B. Frumarova, and T. Wanger, “Photoinduced effects in amorphous chalcogenides,” Chem. Papers 57, 310–327 (1997).
12. S.A. Solin and G.V. Papatheodorou, “Irreversible thermostructural transformations in amorphous As2S3 films: A light-scattering study,” Phys. Rev. B 15, 2084–2090 (1977). [CrossRef]
13. G. Chen, H. Jain, M. Vlcek, S. Khalid, J. Li, D. A. Drabold, and S.R. Elliott, “Observation of light polarization-dependent structural changes in chalcogenide glasses,” Appl. Phys. Lett. 82, 706–708 (2003). [CrossRef]
14. G. Chen, H. Jain, M. Vlcek, and A. Ganjoo, “Photoinduced volume change in arsenic chalcogenides by band-gap light,” Phys. Rev. B 74 (2006) 174203. [CrossRef]
15. M. Frumar, B. Frumar, T. Wagner, and P. Němec, “Photo-Induced Phenomena in Amorphous and Glassy chalcogenides,” in Photo-Induced Metastability in Amorphous Semiconductors, A. V. Kolobov, eds, (Wiley-VCH, 2003), pp.27–28.
16. R. Ston, M. Vlcek, and H. Jain, “Structure and photoinduced changes in bulk and films of As-Ge-S system,” J. Non-cryst. Solids 326&327, 220–225 (2003). [CrossRef]
17. A. Ganjoo, G. Chen, and H. Jain, “Photoinduced changes in the local structure of a-GeSe2 by in-situ EXAFS,” Phys. Chem. Glasses: Eur. J. Glass Sci. Technol. B 47, 177–181 (2006).
18. A. Ganjoo, K. Shimakawa, H. Kamiya, E. A. Davis, and J. Singh, “Percolative growth of photodarkening in amorphous As2S3 films,” Phys. Rev. B 62, R14601–R14604 (2000). [CrossRef]
19. A. Ganjoo and H. Jain, “Millisecond kinetics of photoinduced changes in the optical parameters of a-As2S3 films,” Phys. Rev. B 74, 024201 (2006). [CrossRef]
20. H. Hisakuni and Tanaka, “Giant photoexpansion in As2S3 glass,” Appl. Phys. Lett. 65, 2925–2927 (1994). [CrossRef]