Expand this Topic clickable element to expand a topic
Skip to content
Optica Publishing Group

Optical properties of as-deposited, annealed and laser-treated Ge2Sb2Te5 thin films

Open Access Open Access

Abstract

Knowledge of the correct optical properties of phase change materials (PCM) is important in the development of many devices. Here, we report on the differences in optical properties of as-deposited, annealed, laser-reamorphized and laser-recrystallized $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ thin films. While as-deposited and laser-reamorphized $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ show a similar spectral dependences of refractive indices and extinction coefficients with a small decrease in optical band gap from 0.63 to 0.59 eV, spectral dependences of annealed and laser-recrystallized $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ differ significantly from each other in the entire measured spectral region with values of optical band gap 0.51 and 0.45 eV, respectively. We discuss the main differences in the optical functions between both crystalline phases using X-ray diffraction studies. We further present that the laser-reamorphized phase transforms to the crystalline phase at about 12 °C lower temperature in contrast to the as-deposited amorphous phase.

© 2022 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement

1. Introduction

Today, phase change chalcogenide-based materials are successfully used commercially for data storage [1] due to their unique ability to quickly and reversibly switch between amorphous and crystalline phases. This unique class of materials is best characterized by the Te-based phase-change alloys along the $\mathrm{(GeTe)_{x}-(Sb_{2}Te_{3})_{1-x}}$ pseudo-binary tie-line [2,3], although $\mathrm{Sb_{2}S_{3}}$ and $\mathrm{Sb_{2}Se_{3}}$ [4] materials are taken into consideration as well for their ultra low optical loss in the telecommunication spectral band. The phase transition is associated with the large contrast in electrical and optical properties between both states, which has been recently also adopted in neuro-inspired computational devices [5,6] and nanophotonics [7,8] such as optically-driven displays [9], reconfigurable photonic switches [10], sensors [11], etc.

A certain understanding of the huge reversible change in physical properties during switching between amorphous and crystalline phases comes from their specific atomic bonding. Considering the Extended X-ray Absorption Fine Structure (EXAFS) studies performed on the $\mathrm{(GeTe)_{x}−(Sb_{2}Te_{3})_{1−x}}$ pseudo-binary tie-line [12] revealed the following picture. As-deposited amorphous $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ (fully satisfying the Mott (8-N) rule for amorphous semiconductors [13]) consists of Ge-Te and Sb-Te bonds including "wrong" Ge-Ge and Ge-Sb bonds with almost all Ge atoms tetrahedrally coordinated. On the other hand, crystalline $\mathrm{Ge_{2}Sb_{2}Te_{5}}$, present in real devices, can be found in a low temperature metastable distorted "rocksalt-like" cubic phase [2], where the anion sublattice is formed by the Te atoms and the cation sublattice is occupied by the Ge and Sb atoms along with 20% of structural vacancies. Locally, Ge and Sb atoms form distorted octahedral units with three shorter and three longer Ge(Sb)-Te bonds [12]. Interaction between back-lobes of the $p$-orbitals involved in covalent bonding leads to the creation of resonant (or metavalent) bonds [14,15], making participating atoms six-fold coordinated. It has been reported that the installation or disruption of these resonant bonds (metavalent bonds) in the crystalline phase has a decisive influence on the drastic optical and electrical contrast in $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ [16].

The reversible transformation between the crystalline and amorphous phases can be easily achieved by applying either a short intense laser or electrical current pulse that quickly locally transforms the originally crystalline phase to the amorphous state via a melt-quench process or through non-thermal electronic excitation [17,18]. Conversely, a long, less intense laser or electrical pulse or a heat converts the amorphous phase back into the crystalline phase.

Currently, the massive development in nanophotonics requires the knowledge of optical parameters of phase-change materials for the design and optimization of new optical components using various simulation packages. For this purpose, however, the optical constants adopted from measurements recorded on as-deposited and thermally crystallized thin films do not necessarily correspond to those occurring in real devices. For example, in terms of amorphous state, ab-inition simulations revealed and a subsequent experiment confirmed that a melt-quenched amorphous phase is an intrinsic mixture of the Ge atoms in tetrahedrally coordinated units with a significant fraction of defected octahedral configuration [1921]. A similar discrepancy was also reported for thermally crystallized $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ in the range of temperatures from 150 to $180 \,^{\circ }{\rm C}$, indicating an unexpected irreversible linear increase in the optical band gap from 0.48 to 0.505 eV [22]. Subsequent X-ray diffraction (XRD) studies have offered a possible explanation by a different composition of nuclei and developed crystals [23]. In addition, recent intensive investigations of laser-induced crystallization of $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ disclosed that this process is a very complex problem [2430], when the crystal growth and possible film damages depend on a selected wavelength, pulse duration, and a laser intensity. Another issue is related to the profile of a laser beam, which generates areas with different surface morphology and grain sizes of grown crystals. All the above facts undoubtedly affect the optical parameters of $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ phases.

Questions about the relevant optical parameters of the amorphous and crystalline phase thus remain open and, considering their paramount importance in designing new optical elements using phase change materials, urgently require further study to avoid misleading conclusions. In this paper, we show and discuss differences in the optical properties of as-deposited, annealed, laser-reamorphized and laser-recrystallized $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ thin films. In addition, by measuring the temperature-dependent sheet resistance, we also demonstrate a different crystallization behavior between the as-deposited and laser-reamorphized phases, which affects the thermal stability of a device.

2. Materials and methods

A $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ bulk sample was prepared by a direct synthesis of Ge, Sb and Te elements of 5N purity, which were sealed in a quartz ampule at a residual pressure of 10$^{-3}$ Pa. The synthesis was carried out in a rocking furnace at a temperature of $950 \,^{\circ }{\rm C}$ for 24h and the ampule was subsequently quenched in the cold water.

Thin films of $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ were deposited on fused silica substrates using a flash evaporation technique with a deposition rate of 0.3 nms$^{-1}$ and a background pressure of 2$\times$10$^{-4}$ Pa. The thickness of the deposited thin films was 25 nm. Half of samples were subsequently crystallized at $200 \,^{\circ }{\rm C}$ for one hour in an Argon ambient to prevent sample oxidation. Prior to the application of the laser beam, annealed and as-deposited thin films were coated with a polystyrene overlayer with a thickness of a few microns to protect the $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ thin film against air oxidation. The polystyrene layer was applied by blade coating from a toluene solution (1 g polystyrene per 4 ml toluene) and air dried for four hours. We would like to emphasize that the laser beam was applied through the fused silica substrate, which is transparent for the used ultraviolet nanosecond pulse laser. After complete drying (24 h), the polystyrene layer peeled off by itself, which was an advantage for subsequent studies of optical properties.

The composition of the as-deposited and annealed thin films of $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ was verified by the energy-dispersive x-ray analysis (EDAX).

A KrF excimer laser (Lambda Physik COMPex 102) with a wavelength of 248 nm and a pulse length of 20 ns was used for optical switching between crystalline and amorphous states. This type of laser was chosen to prevent or at least minimize the occurrence of inhomogeneities caused by the profile of the laser beam, in which the manufacturer guarantees uniform intensity over the entire surface of the beam. One pulse with a laser intensity of 81 mJ / area (7 mm $\times$ 25 mm) was applied to successfully obtain a laser-reamorphized spot. Subsequently, laser-recrystallization was performed using the laser pulse intensity of 57 mJ / area that covered the re-amorphous part.

Ellipsometric spectra were recorded using a VASE ellipsometer (Woollam Co. Ltd.) over a spectral range 0.7 - 6.5 eV. The angle of incidence was varied from $65\,^{\circ }$ to $75\,^{\circ }$ with a step of $5\,^{\circ }$. The fitting model consisted of a fused silica semi-infinite substrate and a $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ single layer. The surface roughness was considered by means of the Bruggeman effective medium approximation with a mixture of 50% voids and 50% $\mathrm{Ge_{2}Sb_{2}Te_{5}}$. The optical constants of amorphous and crystalline $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ films were parameterized by the Tauc-Lorentz model (both amorphous phases) with one (laser-crystallized) or two (annealed) additional Gaussian oscillators. The optical gap was determined from the Tauc plot.

X-ray diffraction measurements were carried out using a diffractometer (Empyrean Malvern Panalytical) operated in the grazing incidence geometry at the angle of 2 $\deg$ for as-deposited amorphous and laser-reamorphized $\mathrm{Ge_{2}Sb_{2}Te_{5}}$, while two-theta/omega mode was used for the annealed and laser-crystallized samples of $\mathrm{Ge_{2}Sb_{2}Te_{5}}$. The FullProf software package [31] was used to obtain accurate lattice parameters of annealed and laser-recrystallized $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ samples.

The sheet resistance was measured using a van der Pauw four probe setup equipped with a heating cell (THMSG 600). Measurements were performed over a temperature range from 20 to 220$\,^{\circ }{\rm C}$ with a heating rate of 2$\,^{\circ }{\rm C}$/ min. An Ar inert atmosphere was used to prevent sample oxidation.

3. Results and discussion

Initially, we tuned the pulse energy of the nanosecond KrF laser to find the threshold energy to successfully achieve the re-amorphous area on the crystalline background. By gradual decreasing the pulse energy from 130 to 50 mJ / area (7 mm $\times$ 25 mm), we found that 81 mJ/area (amorphization) and 57 mJ/area (crystallization) represent the best conditions for multiple reversible switching of the $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ thin film without apparent film damages (ablation). We note that our optimum laser fluencies are in good agreement with recently reported values [32,33]. At first glance, the obvious optical contrast (photos obtained in the transmission mode) between the as deposited amorphous and annealed (crystalline) phases can be seen in Fig. 1 (A,B), which is well reproduced by applying a single-pulse excitation to the annealed $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ film, resulting in the laser-reamorphized phase and another single laser pulse, with lower intensity, applied on the same area transforms the latter phase to the laser-recrystallized phase as shown in Fig. 1 (C,D). While the deposited and laser-reamorphized phases look similar, it was observed with the naked eye that the laser-recrystallized phase has a lighter color compared to the annealed phase. The detailed nature of this difference will be discussed below.

 figure: Fig. 1.

Fig. 1. Typical top-view images of (A) as-deposited, (B) annealed at $200 \,^{\circ }{\rm C}$ for 1h, (C) laser-reamorphized and (D) laser-recrystallized 25 nm thin film of $\mathrm{Ge_{2}Sb_{2}Te_{5}}$. The scale is the same for all images. Photos obtained in the transmission mode.

Download Full Size | PDF

To characterize and compare the structural changes in different states of $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ after annealing or a laser exposure, we measured XRD patterns of as-deposited, annealed and laser-treated $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ thin films as demonstrated in Fig. 2. We verified that both the as-deposited and laser reamorphized $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ films were completely amorphous, as evidenced only by the broad features characteristic of amorphous materials without crystalline peaks as shown in Fig. 2 (A). The results also confirmed that we had chosen the right laser conditions for the reamorphization process. After annealing of as-deposited $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ at $200 \,^{\circ }{\rm C}$ for 1h, a typical metastable distorted "rock-salt-like" cubic phase was observed. Similarly, upon application of a single ultraviolet laser pulse, the laser-reamophized phase transformed back into the metastable cubic phase as demonstrated in Fig. 2 (B). However, a detailed analysis of both crystalline phases, shown in Fig. 2 (C) revealed that the Bragg reflection peaks of the laser-recrystallized $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ are more intense, narrower and maxima of Bragg reflection peaks are shifted towards a lower angle of 2 theta compared to the annealed sample, which clearly indicates that the two crystalline phases are not the same. An earlier study pointed to the fact that the onset of crystallization is associated with the formation of nuclei with a higher Sb content, resulting in a larger lattice parameter [23]. The authors claimed that amorphous $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ first transforms to $\mathrm{Ge_{1}Sb_{4}Te_{7}}$ and after annealing above $180 \,^{\circ }{\rm C}$ the rocksalt-like phase of $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ is formed. To discern lattice parameters of the annealed and laser-recrystallized phases, we evaluated XRD patterns using the FullProf software package [31]. All fits can be seen in Fig. 1(S) in the supplementary information. At the beginning, we considered that both crystalline phases are formed by a single metastable face-centered cubic phase. As expected, the laser-recrystallized $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ phase can be well fitted using this model providing a lattice constant of 6.00 Å, which is a typical value for Ge-Sb-Te alloys [2]. On the other hand, using the same approach, experimental XRD data for the annealed phase cannot simply be interpreted by a single metastable cubic phase, suggesting that the annealed phase is possibly a rhombohedral phase or consists of two cubic phases with very close lattice constants. Through thorough analysis, we found that the XRD pattern of the annealed phase is best reproduced by the second option, where one phase is similar to the cubic phase found in the laser-recrystallized phase and the other has a smaller lattice constant of 5.98 Å. It should be emphasized that the energy-dispersive x-ray analysis (EDAX) confirmed that the composition of the as-deposited and annealed layers nominally correspond to the $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ alloy. A possible explanation for this phenomenon can be found in a recent study using X-ray Absorption Near-edge Structure (XANES) [16], which revealed that crystallization proceeds in two steps. The rapid and dominant stage occurs at the crystallization temperature. The second step continues well above the crystallization temperature (up to T$_{\boldsymbol C}$ + $60 \,^{\circ }{\rm C}$) and is characterized by increasing the degree of ordering. Because laser-recrystallization takes place on a nanosecond time scale, we hypothesize that the laser-recrystallized phase can be attributed to a rapidly crystallized phase with some covalently bonded elements, while the annealed phase is completely converted to a crystalline phase. Moreover, a recent report on in-situ thermal crystallization of $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ using a high resolution transmission electron microscopy revealed the co-existence of the fcc phase with the so-called "a vacancy ordered cubic" phase at temperatures above $200 \,^{\circ }{\rm C}$ [34], which may confirm the use of two cubic phases for the fit of the annealed $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ phase. Other reports revealed Te [35] or a Te-rich [36] phase segregation at the grain boundary and tetrahedrally coordinated Ge atoms within the fcc phase [37] using a high resolution transmission electron microscopy. Interestingly, a clear evidence of differences in structural units between laser- and thermally- crystallized $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ was observed by Raman spectroscopy [27]. The authors found that the Raman vibration attributed to ${\mathrm {Sb_{m}Te_{3}} }$ structural units is suppressed in the laser-crystallized $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ phase, while ${\mathrm {GeTe_{4-n}Ge_{n}}}$ (n = 0, 1, 2) tetrahedral units play a decisive role in both crystalline phases. In addition, the Raman peak ascribed to ${\mathrm {Sb_{m}Te_{3}} }$ structural units is more intense for the laser-crystallized $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ phase irradiated by a low laser fluence and ${\mathrm {GeTe_{4-n}Ge_{n}}}$ (n = 0, 1, 2) tetrahedra dominate in the laser-crystallized structure when a high laser fluence is applied [32].

 figure: Fig. 2.

Fig. 2. XRD patterns of (A) as-deposited and laser-reamorphized, (B) annealed at $200 \,^{\circ }{\rm C}$ for 1h and laser-crystallized 25 nm thin film of $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ and (C) a detailed look at two dominant Bragg reflections along (200) and (220) the crystallographic planes for both crystalline phases of $\mathrm{Ge_{2}Sb_{2}Te_{5}}$.

Download Full Size | PDF

Differences in the structure of amorphous and crystalline layers can be observed in their optical properties. To obtain optical constants from ellipsometric data, the band structure of semiconductor materials must be considered. In this work, we selected dispersion models, which are Kramers - Kronig consistent for describing the optical properties of all amorphous and crystalline $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ thin films with physically realistic significance. For this reason, the Tauc-Lorentz model was used to parameterize $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ electric permittivity and adapt the ellipsometric measurements of amorphous $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ and the Tauc-Lorentz model with one or two additional Gaussian oscillators was employed for crystalline phases [22]. The Tauc plot was used to determine the optical band gap, $\mathbf{E}_{g}^{Opt}$ for crystalline states.

Upon fitting the ellipsometry data, we obtained spectral dispersions of the refractive index, n and extinction coefficient, k and optical band gap, $\mathbf{E}_{g}^{Opt}$ of as-deposited amorphous, annealed, laser-reamorphized and laser-recrystallized $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ phases shown in Fig. 3. At first glance, a typical optical contrast between amorphous and crystalline $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ can be observed, which is in good agreement with trends reported in previous works [22,38]. However, a deeper look disclosed that optical functions between two studied amorphous samples and between two crystalline phases are different. In terms of amorphous state, the as-deposited and laser-reamorphized phases have a very similar spectral dependence of the optical constants, which confirms that the laser reamorphization of the $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ crystalline layer was successful. But, a detailed analysis revealed that n and k values of the laser-reamorphized phase in the entire measured region are slightly shifted towards optical constants for the crystalline phase. As reported, a chemical nature of laser-reamorphized (or melt-quenched) $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ was described as a mixture of the Ge atoms in octahedral (crystal-like) and tetrahedral configurations in comparable concentrations, in contrast to the as-deposited amorphous phase that is dominated by the Ge atoms in the tetrahedral configuration [21]. The fact that in the laser-reamorphized phase resonance bonding in the octahedral configurations is localized to within several interatomic distances, their contribution to the the overall optical functions are not significant, but it cannot simply be neglected. For example, the values of $\Delta$n and $\Delta$k at the telecommunication wavelength of 1550 nm are 0.1 and 0.08, respectively.

 figure: Fig. 3.

Fig. 3. Optical properties of as-deposited and laser-reamorphized and annealed at $200 \,^{\circ }{\rm C}$ for 1h and laser-crystallized 25 nm thin films of $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ . (A) Refractive index, n, (B) Extinction coefficient, k, and (C) Optical band gap, $\mathbf{E}_{g}^{Opt}$.

Download Full Size | PDF

The drastic change in optical functions of the crystalline phase stems from the formation of continuous resonant bonding network in the face-centered cubic phase [1416]. It is apparent from Fig. 3 (A) and (B) that the annealed and laser-recrystallized phases have very distinct spectral dependences of optical constants in the measured range of wavelengths from 300 to 2200 nm. More specifically, in the spectral range from 300 to 1600 (1200) nm, the values of n and k are significantly higher for the annealed phase, while larger values can be observed at longer wavelengths in the case of laser-recrystallized phase. As above mentioned, the intrinsic structural disorder caused by the existence of non-resonantly bonded atoms, which may concentrate at the grain boundary [36] or within the fcc phase [37] can decrease the values of optical functions of the laser-recrystallized phase in the spectral range from 300 to 1600 (1200) nm. Hypothetically, if the laser-crystallized phase is the Sb-rich phase [23] compared to nominal composition of $\mathrm{Ge_{2}Sb_{2}Te_{5}}$, Sb atoms in the fcc phase induce structural vacancies that weaken the continuous resonant network and thus it can decrease the values of optical constants in the above mentioned spectral range as reported in [38]. Interestingly, the larger content of the Sb atoms in the fcc phase increases light absorption at wavelengths longer than $\approx$1200 nm [38], which could be one of possible explanations for the larger k value in this spectral region observed for laser-recrystallized $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ compared to the annealed phase. However, this phenomenon is not fully understood and needs to be studied further.

The phase transition can be characterized by the change in $\mathbf{E}_{g}^{Opt}$ as shown in Fig. 3 (C). There is a significant decrease in $\mathbf{E}_{g}^{Opt}$ between as-deposited and annealed $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ from 0.63 eV to 0.51 eV, which is in good agreement with recently reported values [10,22,38]. Due to different deposition conditions and techniques, the $\mathbf{E}_{g}^{Opt}$ value of as-deposited amorphous $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ in the literature ranges from 0.63 to 0.8 eV [10,22,38], while the $\mathbf{E}_{g}^{Opt}$ value for the annealed fcc phase is $\approx$ 0.5 eV [10,22,38]. In the laser-reamorphized state, the Tauc-Lorentz model showed $\mathbf{E}_{g}^{Opt}$ of 0.59 eV, which is a lower value in comparison with $\mathbf{E}_{g}^{Opt}$ of the as-deposited phase, but consistently shifted towards $\mathbf{E}_{g}^{Opt}$ of the annealed crystalline state similar to the n and k spectral dependences and can be therefore attributed to the polyamorphic nature of the amorphous phases, as discussed above. Interestingly, we observed that $\mathbf{E}_{g}^{Opt}$ of laser-recrystallized $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ is unexpectedly 0.45 eV, which is a lower value that any reported values for crystalline $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ in the fcc phase. As demonstrated in [22], thermally crystallized $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ exhibits an irreversible linear increase in $\mathbf{E}_{g}^{Opt}$ from 0.48 to 0.505 eV in the range of temperatures from 150 to $180 \,^{\circ }{\rm C}$. From a thermodynamic point of view, the annealed and laser crystallized phases are formed at different times (hours vs. nanoseconds) and thus we suppose that $\mathbf{E}_{g}^{Opt}$ of laser-crystallized $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ could be close to the beginning of the observed unusual trend. Bearing in mind the temperature dependence of the optical band gap of as-deposited $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ [22] from RT to $220 \,^{\circ }{\rm C}$, we assume that the obtained $\mathbf{E}_{g}^{Opt}$ corresponding to laser-reamorphized and laser-recrystallized $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ are located close to the onset and just at the end of the phase transition. However, we should also take in mind that $\mathbf{E}_{g}^{Opt}$ decreases with increasing Sb content in the fcc phase from 0.51 ($\mathrm{Ge_{2}Sb_{2}Te_{5}}$) to 0.49 eV ($\mathrm{GeSb_{2}Te_{4}}$) [38] and this possibility should not be completely omitted. Optical constants at the telecommunication wavelength of 1550 nm for all $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ phases and corresponding $\mathbf{E}_{g}^{Opt}$ are summarized in Table 1.

Tables Icon

Table 1. Refractive index, n and Extinction coefficient, k of all $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ phases taken at the telecommunication wavelength $\lambda$= 1550 nm and Optical band gap, $\mathbf{E}_{g}^{Opt}$.

Finally, we would like to focus on amorphous phases. The influence of the initial state of as-deposited and laser-reamorphized $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ on the amorphous-to-crystal transition was thoroughly studied by means of electrical transport experiments, as demonstrated in Fig. 4. Of special interest is the observed significantly lower crystallization temperature of laser-reamorphized $\mathrm{Ge_{2}Sb_{2}Te_{5}}$, which is $128 \,^{\circ }{\rm C}$ compared to $140 \,^{\circ }{\rm C}$ for the as-deposited counterpart. Typically, the presence of a high concentration of crystal-like structural motifs [39], which can support the formation of crystal nuclei, or the use of a heterogeneous nucleation [4042] can decrease the crystallization temperature. The fact that the crystallization temperature of laser-reamorphized $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ is $12 \,^{\circ }{\rm C}$ lower in contrast to the as-deposited amorphous phase is unambiguous evidence of a difference in local structure of both phases. To the above mentioned nature of both amorphous phases, as-deposited amorphous $\mathrm{Ge_{2}Sb_{2}Te_{5}}$, prepared through a vapor phase, can be exclusively considered as a statistically ordered amorphous phase, whereas the laser-reamorphized $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ can be characterized as a chemically ordered amorphous phase, which arises from memory effects of the initial cubic phase, having a strong impact on the decreased value of crystallization temperature. It should be emphasized that a similar trend was observed in the pressure-induced amorphous $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ phase, which crystallizes to the initial metastable cubic phase at $\approx 117 \,^{\circ }{\rm C}$ under ambient pressure [43].

 figure: Fig. 4.

Fig. 4. The sheet resistance of the as-deposited and laser-reamorphized 25 nm thin film of $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ measured in the range of temperatures from RT to $220 \,^{\circ }{\rm C}$.

Download Full Size | PDF

4. Conclusion

In conclusion, we demonstrated the differences in optical properties of as-deposited, annealed, laser-reamorphized and laser-recrystallized $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ thin films. While as-deposited and laser-reamorphized $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ show a similar spectral dependences of refractive indices and extinction coefficients with a small decrease in optical band gap from 0.63 to 0.59 eV, spectral dependences of annealed and laser-recrystallized $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ differ significantly from each other with values of optical band gap 0.51 and 0.45 eV, respectively. Using X-ray diffraction studies, we attributed the lower band gap value of the laser-recrystallized $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ phase to the formation of a single crystalline metastable cubic phase with a lattice parameter of 6.00 Å compared to the annealed phase at $200 \,^{\circ }{\rm C}$ for 1h, resulting in broader asymmetric diffraction peaks, which can be determined by two cubic phases with a lattice constants of 5.98 and 6.00 Å. Compared to the previously reported data on the temperature dependence of the optical band gap of as-deposited $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ [22] from RT to $220 \,^{\circ }{\rm C}$, we found that the optical band gaps corresponding to laser-reamorphized and laser-recrystallized $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ are located close to the onset and just at the end of the phase transition. We further presented that the laser-reamorphized phase transforms to the crystalline phase at about $12 \,^{\circ }{\rm C}$ lower temperature in contrast to the as-deposited amorphous phase. The observed decrease in the crystallization temperature of the laser-reamorphized phase is thought to be a result of the retaining of structural order derived from the original cubic $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ phase, which is preserved in the laser-reamorphized phase.

Funding

Ministerstvo Školství, Mládeže a Tělovýchovy (LM2018103).

Acknowledgments

The work was carried out under the auspices of CEMNAT at the University of Pardubice.

Disclosures

The authors declare no conflicts of interest.

Data availability

Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.

Supplemental document

See Supplement 1 for supporting content.

References

1. F. T. Hady, A. Foong, B. Veal, and D. Williams, “Platform storage performance with 3D XPoint technology,” Proc. IEEE 105(9), 1822–1833 (2017). [CrossRef]  

2. N. Yamada, E. Ohno, K. Nishiuchi, N. Akahira, and M. Takao, “Rapid phase-transitions of GeTe–Sb2Te3 pseudobinary amorphous thin-films for an optical disk memory,” J. Appl. Phys. 69(5), 2849–2856 (1991). [CrossRef]  

3. M. Wuttig and N. Yamada, “Phase-change materials for rewriteable data storage,” Nat. Mater. 6(11), 824–832 (2007). [CrossRef]  

4. M. Delaney, I. Zeimpekis, D. Lawson, D. W. Hewak, and O. L. Muskens, “A new family of ultralow loss reversible phase-change materials for photonic integrated circuits: Sb2S3 and Sb2Se3,” Adv. Funct. Mater. 30(36), 2002447 (2020). [CrossRef]  

5. S. Ambrogio, P. Narayanan, H. Tsai, R. M. Shelby, I. Boybat, C. di Nolfo, S. Sidler, M. Giordano, M. Bodini, N. C. P. Farinha, B. Killeen, C. Cheng, Y. Jaoudi, and G. W. Burr, “Equivalent-accuracy accelerated neural-network training using analogue memory,” Nature 558(7708), 60–67 (2018). [CrossRef]  

6. W. Zhang, R. Mazzarello, M. Wuttig, and E. Ma, “Designing crystallization in phase-change materials for universal memory and neuro-inspired computing,” Nat. Rev. Mater. 4(3), 150–168 (2019). [CrossRef]  

7. M. Wuttig, H. Bhaskaran, and T. Taubner, “Phase-change materials for non-volatile photonic applications,” Nat. Photonics 11(8), 465–476 (2017). [CrossRef]  

8. S. Abdollahramezani, O. Hemmatyar, H. Taghinejad, A. Krasnok, Y. Kiarashinejad, M. Zandehshahvar, A. Alu, and A. Adibi, “Tunable nanophotonics enabled by chalcogenide phase-change materials,” Nanophotonics 9(5), 1189–1241 (2020). [CrossRef]  

9. P. Hosseini, C. D. Wright, and H. Bhaskaran, “An optoelectronic framework enabled by low-dimensional phase-change films,” Nature 511(7508), 206–211 (2014). [CrossRef]  

10. T. Y. Teo, M. Krbal, J. Mistrik, J. Prikryl, L. Lu, and R. E. Simpson, “Comparison and analysis of phase change materials-based reconfigurable silicon photonic directional couplers,” Opt. Mater. Express 12(2), 606–621 (2022). [CrossRef]  

11. L. Shengyong, L. Yilin, and A. Xiaochuan, “Realization of a sensitivity multi-band tunable sensor through modulating the phase change of Ge2Sb2Te5,” Opt. Laser Technol. 144, 107428 (2021). [CrossRef]  

12. A. V. Kolobov, P. Fons, A. I. Frenkel, A. L. Ankudinov, J. Tominaga, and T. Uruga, “Understanding the phase-change mechanism of rewritable optical media,” Nat. Mater. 3(10), 703–708 (2004). [CrossRef]  

13. N. Mott, “Electrons in disordered structures,” Adv. Phys. 16(61), 49–144 (1967). [CrossRef]  

14. G. Lucovsky and R. M. White, “Effects of resonance bonding on the properties of crystalline and amorphous semiconductors,” Phys. Rev. B 8(2), 660–667 (1973). [CrossRef]  

15. K. Shportko, S. Kremers, M. Woda, D. Lencer, J. Robertson, and M. Wuttig, “Resonant bonding in crystalline phase-change materials,” Nat. Mater. 7(8), 653–658 (2008). [CrossRef]  

16. M. Krbal, A. V. Kolobov, P. Fons, J. Tominaga, S. R. Elliott, J. Hegedus, A. Giussani, K. Perumal, R. Calarco, T. Matsunaga, N. Yamada, K. Nitta, and T. Uruga, “Crystalline GeTe-based phase-change alloys: Disorder in order,” Phys. Rev. B 86(4), 045212 (2012). [CrossRef]  

17. K. V. Mitrofanov, P. Fons, K. Makino, R. Terashima, T. Shimada, A. V. Kolobov, J. Tominaga, V. Bragaglia, A. Giussani, R. Calarco, H. Riechert, T. Sato, T. Katayama, K. Ogawa, T. Togashi, M. Yabashi, S. Wall, D. Brewe, and M. Hase, “Sub-nanometre resolution of atomic motion during electronic excitation in phase-change materials,” Sci. Rep. 6(1), 20633 (2016). [CrossRef]  

18. X. Wang, X. Li, N. Chen, J. Bang, R. Nelson, C. Ertural, R. Dronskowski, H. Sun, and S. Zhang, “Time-dependent density-functional theory molecular-dynamics study on amorphization of Sc–Sb − Te alloy under optical excitation,” npj Comput. Mater 6(1), 31 (2020). [CrossRef]  

19. S. Caravati, M. Bernasconi, T. D. Kuhne, M. Krack, and M. Parrinello, “Coexistence of tetrahedral- and octahedral-like sites in amorphous phase change materials,” Appl. Phys. Lett. 91(17), 171906 (2007). [CrossRef]  

20. J. Hegedus and S. R. Elliott, “Microscopic origin of the fast crystallization ability of Ge-Sb-Te phase-change memory materials,” Nat. Mater. 7(5), 399–405 (2008). [CrossRef]  

21. M. Krbal, A. V. Kolobov, P. Fons, J. Tominaga, S. R. Elliott, J. Hegedus, and T. Uruga, “Intrinsic complexity of the melt-quenched amorphous Ge2Sb2Te5 memory alloy,” Phys. Rev. B 83(5), 054203 (2011). [CrossRef]  

22. J. Orava, T. Wagner, J. Sik, J. Prikryl, M. Frumar, and L. Benes, “Optical properties and phase change transition in Ge2Sb2Te5 flash evaporated thin films studied by temperature dependent spectroscopic ellipsometry,” J. Appl. Phys. 104(4), 043523 (2008). [CrossRef]  

23. E. Morales-Sanchez, B. Lain, E. Prokhorov, M. Hernadez-Landaverde, G. Trapaga, and J. Gonzalez-Hernandez, “Crystallization process in Ge2Sb2Te5 amorphous films,” Vacuum 84(7), 877–881 (2010). [CrossRef]  

24. S. H. Moller, E. H. Eriksen, P. L. Tonning, P. B. Jensen, J. Chevallier, and P. Balling, “Femtosecond-laser-induced modifications of Ge2Sb2Te5 thin films: Permanent optical change without amorphization,” Appl. Surf. Sci. 476, 221–231 (2019). [CrossRef]  

25. W. Li, F. Liu, J. Guo, Y. Zhang, Y. Wang, F. Liu, N. Sun, and W. Xiao, “Crystallization characteristics and local grain abnormal growth of amorphous Ge2Sb2Te5 films induced by a gaussian picosecond laser,” Opt. Laser Technol. 111, 585–591 (2019). [CrossRef]  

26. S. M. S. Privitera, I. Lopez Garcia, C. Bongiorno, V. Sousa, M. C. Cyrille, G. Navarro, C. Sabbione, E. Carria, and E. Rimini, “Crystallization properties of melt-quenched Ge-rich GeSbTe thin films for phase change memory applications,” J. Appl. Phys. 128(15), 155105 (2020). [CrossRef]  

27. J. Du, Z. Mu, L. Li, and J. Li, “A raman study on nanosecond-laser-induced multi-level switching of Ge2Sb2Te5 thin films,” Opt. Laser Technol. 144, 107393 (2021). [CrossRef]  

28. T. Kunkel, Y. Vorobyov, M. Smayev, P. Lazarenko, V. Veretennikov, V. Sigaev, and S. Kozyukhin, “Experimental observation of two-stage crystallization of Ge2Sb2Te5 amorphous thin films under the influence of a pulsed laser,” J. Alloys Compd. 851, 156924 (2021). [CrossRef]  

29. A. V. Kiselev, V. V. Ionin, A. A. Burtsev, N. N. Eliseev, V. A. Mikhalevsky, N. A. Arkharova, D. N. Khmelenin, and A. A. Lotin, “Dynamics of reversible optical properties switching of Ge2Sb2Te5 thin films at laser-induced phase transitions,” Opt. Laser Technol. 147, 107701 (2022). [CrossRef]  

30. T. Kunkel, Y. Vorobyov, M. Smayev, P. Lazarenko, A. Romashkin, and S. Kozyukhin, “Crystallization of Ge2Sb2Te5 thin film induced by a single femtosecond laser pulse: Experimental and theoretical study,” Mater. Sci. Semicond. Process. 139, 106350 (2022). [CrossRef]  

31. J. Rodriguez-Carvajal, “Recent advances in magnetic structure determination by neutron powder diffraction,” Phys. B 192(1-2), 55–69 (1993). [CrossRef]  

32. W. Zhou, F. Liu, N. Bai, Y. Wan, X. Lin, and J. Chen, “Crystallization of amorphous Ge2Sb2Te5 films induced by an ultraviolet laser,” Appl. Surf. Sci. 285, 97–101 (2013). [CrossRef]  

33. W. Q. Li and F. R. Liu, “Study on the crystallization mechanism of amorphous Ge2Sb2Te5 thin films induced by a short single pulsed laser,” J. Laser Appl. 30(3), 032601 (2018). [CrossRef]  

34. Y. Zheng, Y. Cheng, R. Huang, R. Qi, F. Rao, K. Ding, W. Yin, S. Song, W. Liu, Z. Song, and S. Feng, “Surface energy driven cubic-to-hexagonal grain growth of Ge2Sb2Te5 thin film,” Sci. Rep. 7(1), 5915 (2017). [CrossRef]  

35. L. Krusin-Elbaum, C. Cabral, K. N. Chen, M. Copel, D. W. Abraham, K. B. Reuter, S. M. Rossnagel, J. Bruley, and V. R. Deline, “Evidence for segregation of Te in Ge2Sb2Te5 films: effect on the phase-change stress,” Appl. Phys. Lett. 90(14), 141902 (2007). [CrossRef]  

36. G. Lucovsky and J. C. Phillips, “Reversible chemical phase separation in on-state of art ReWritable (RW) Ge2Sb2Te5 optical phase change memories,” J. Non-Cryst. Solids 354(19-25), 2753–2756 (2008). [CrossRef]  

37. X. Q. Liu, X. B. Li, L. Zhang, Y. Q. Cheng, Z. G. Yan, M. Xu, X. D. Han, S. B. Zhang, Z. Zhang, and E. Ma, “New structural picture of the Ge2Sb2Te5 phase-change alloy,” Phys. Rev. Lett. 106(2), 025501 (2011). [CrossRef]  

38. J.-W. Park, S. H. Eom, H. Lee, J. L. F. Da Silva, Y.-S. Kang, T.-Y. Lee, and Y. H. Khang, “Optical properties of pseudobinary GeTe, Ge2Sb2Te5, GeSb2Te4, GeSb4Te7, and Sb2Te3 from ellipsometry and density functional theory,” Phys. Rev. B 80(11), 115209 (2009). [CrossRef]  

39. D. Loke, T. H. Lee, W. J. Wang, L. P. Shi, R. Zhao, Y. C. Yeo, T. C. Chong, and S. R. Elliott, “Breaking the speed limits of phase-change memory,” Science 336(6088), 1566–1569 (2012). [CrossRef]  

40. R. E. Simpson, P. Fons, A. V. Kolobov, M. Krbal, and J. Tominaga, “Enhanced crystallization of GeTe from an Sb2Te3template,” Appl. Phys. Lett. 100(2), 021911 (2012). [CrossRef]  

41. P. Noe, C. Sabbione, N. Bernier, N. Castellani, F. Fillot, and F. Hippert, “Impact of interfaces on scenario of crystallization of phase change materials,” Acta Mater. 110, 142–148 (2016). [CrossRef]  

42. G. Liu, L. Wu, S. Zhang, W. Liu, S. Lv, M. Zhu, S. Song, and Z. Song, “Heterogeneous nucleus-induced crystallization for high-speed phase change memory applications,” Appl. Phys. Lett. 115(13), 133505 (2019). [CrossRef]  

43. M. Krbal, A. Kolobov, M. Hanfland, and P. Fons, “Crystallization of Ge2Sb2Te5 under high hydrostatic pressures: Differences in nanoscale atomic ordering in as-deposited and pressure-induced amorphous phases,” J. Alloys Compd. 874, 159980 (2021). [CrossRef]  

Supplementary Material (1)

NameDescription
Supplement 1       Supplementary

Data availability

Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.

Cited By

Optica participates in Crossref's Cited-By Linking service. Citing articles from Optica Publishing Group journals and other participating publishers are listed here.

Alert me when this article is cited.


Figures (4)

Fig. 1.
Fig. 1. Typical top-view images of (A) as-deposited, (B) annealed at $200 \,^{\circ }{\rm C}$ for 1h, (C) laser-reamorphized and (D) laser-recrystallized 25 nm thin film of $\mathrm{Ge_{2}Sb_{2}Te_{5}}$. The scale is the same for all images. Photos obtained in the transmission mode.
Fig. 2.
Fig. 2. XRD patterns of (A) as-deposited and laser-reamorphized, (B) annealed at $200 \,^{\circ }{\rm C}$ for 1h and laser-crystallized 25 nm thin film of $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ and (C) a detailed look at two dominant Bragg reflections along (200) and (220) the crystallographic planes for both crystalline phases of $\mathrm{Ge_{2}Sb_{2}Te_{5}}$.
Fig. 3.
Fig. 3. Optical properties of as-deposited and laser-reamorphized and annealed at $200 \,^{\circ }{\rm C}$ for 1h and laser-crystallized 25 nm thin films of $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ . (A) Refractive index, n, (B) Extinction coefficient, k, and (C) Optical band gap, $\mathbf{E}_{g}^{Opt}$.
Fig. 4.
Fig. 4. The sheet resistance of the as-deposited and laser-reamorphized 25 nm thin film of $\mathrm{Ge_{2}Sb_{2}Te_{5}}$ measured in the range of temperatures from RT to $220 \,^{\circ }{\rm C}$.

Tables (1)

Tables Icon

Table 1. Refractive index, n and Extinction coefficient, k of all G e 2 S b 2 T e 5 phases taken at the telecommunication wavelength λ = 1550 nm and Optical band gap, E g O p t .

Select as filters


Select Topics Cancel
© Copyright 2024 | Optica Publishing Group. All rights reserved, including rights for text and data mining and training of artificial technologies or similar technologies.