1. Introduction

At present,demand for mass storage has been increasing owing to increased information. Therefore,memory devices require improved performance to meet the needs of an information hungry society. For such reasons,the demand for non-volatile memory has increased rapidly^[1]. Phase change memory (PCM) is a new non-volatile memory that uses reversible phase transition of a chalcogenide resistor^{[2, 3]}. The idea of using an amorphous-to-crystalline phase transition for information storage was born way back in the 1960s by Ovshinsky who suggested a memory switch based upon changes in the properties of amorphous and crystalline phases of multi-component chalcogenides^[4]. In this technology the information storage relies on a current induced reversible phase transition. A current pulse focused onto the area corresponding to the bit size,is employed to heat the PCM material,which induces the change between the amorphous and crystalline phases^[5]. Since the two phases feature a different resistivity^[6],the electrical readout is performed with a low power probing electrical pulse. In this case the idea is to electrically induce the phase change in a PCM cell and associate the stored information to the corresponding high and low resistance values. Since both states are stable at room temperature^[7],no energy is required to keep data stored. This results in an inherently non-volatile memory technology. There are set and reset operations in PCM. A current pulse of low intensity and relatively long duration crystallizes the film (set operation),while a pulse of higher intensity but shorter duration is used for amorphization (reset operation). Readout is performed at a lower current that does not change the phase^[4]. This is shown schematically in Figure 1. A critical property of phase change material is the threshold switching^{[5, 8, 9, 10]}. Without this effect PCM would not be a feasible technology,because in the high resistance state,extremely high voltages would be required to deliver enough power to the cell to heat it above the crystallization temperature^[6]. However,when a voltage above a particular threshold ($V_{\rm t})$ is applied to a phase change material in the amorphous phase,the resulting large electrical fields greatly increase the electrical conductivity^[5]. With the previously resistive material now suddenly conducting,a large current flows which can then heat the material. The threshold switching effect serves to make possible the use of an applied voltage of a few volts,despite the high initial resistance of the device in the reset state^{[5, 6, 10]}.

Alloys on the pseudo ternary tie line in Ge--Sb--Te phase diagram as well as the eutectic of Sb--Te and Ge--Sb binary alloys have been the extensively studied and optimized material systems for PCM technology due to their excellent properties^{[1, 11, 12, 13, 14, 15, 16]} such as a reasonably high reflectivity contrast between the amorphous and crystalline states to ensure a successful READ process,high stability at room temperature to ensure longer archivability,first or rapid reversible switching between the two storage states. Despite their widespread application,these materials have shown potential drawbacks which preclude their application in electronic storage as PCM devices^[12]. These drawbacks include: low crystallization temperature preventing their application (e.g. data retention) at elevated temperatures,low SET (crystalline) resistance which increases the reset current of the PCM device. It has been reported that materials with high optical contrast such as GST also show a relatively large increase in mass density upon crystallization^[17]. Repeated cycling in PCM applications can lead to void formation caused by the large difference in mass density between the two phases causing subsequent cell failure^[7]. Elemental segregation is another failure mechanism that is observed for PCM cells with GST phase change material^[7]. Antimony-rich areas are formed on top of the bottom electrode^[3] which deteriorates data retention because antimony-rich alloys are known to have a reduced crystallization temperature and archival life stability^[18]. A number of studies have been carried out in order to acquire advanced materials exhibiting lower power consumption,rapid phase transition,cyclic reliability and higher data retention ability. Research has shown that films from Sn$_{2}$Se$_{3}$ posses a higher crystalline resistance as compared to those from GST alloys. Despite the enormous potential technological importance of this material,very little information on its material properties has been reported. Therefore in this research,the crystallization kinetics of Sn$_{40}$Se$_{60}$ alloys has been studied to establish their suitability for PCM application.

2. Experimental procedure

The constituent elements (5 N pure) were weighed according to their atomic weight percentages. The mixture was placed in a clean silica tube and heated in ambient argon for the elements to melt. The ampoule was then sealed. The heated melt was cooled to obtain the desired alloy. The alloy was then ground in a porcelain mortar to obtain fine particles of the alloy. Ordinary glass slides were used as substrates for preparing Sn$_{40}$Se$_{60}$ films. The films were prepared by thermal evaporation technique in a vacuum. The film sheet resistance was measured by employing a four-point probe setup with a square array following a procedure proposed by Van dar Pauw. Sheet resistance of the films was measured at various temperatures in order to determine the crystallization temperatures. The sheet resistance versus temperature measurements was performed at different heating rates and the corresponding crystallization temperatures ($T_{\rm c})$ were determined from the minimum of the derivative of sheet resistance. The heating rates were varied between 1 and 10 K/min. For each run,a fresh specimen of the film was used. The Kissinger plots were used in determining the activation energy.

3. Results and discussion

The graph in Figure 2(a) shows the sheet resistance versus temperature for 200 nm Sn$_{40}$Se$_{60}$ thin film heated at a rate of 5 K/min within a temperature range of 25 C and 245 C. At room temperature,the sheet resistance of Sn$_{40}$Se$_{60}$ was found to be 195 M$\Omega $/$\square$. This value decreased to 1560 $\Omega $/$\square$ upon annealing and an abrupt decrease was observed between 150 and 175 C. This abrupt decrease is an indication that Sn$_{40}$Se$_{60}$ is a phase change material. The crystallization temperature obtained from the derivative of sheet resistance vs. temperature was 156.6 $\pm$ 0.3 C. Although this value is low,it is higher than the reported value of 149 C for the GST for similar heating rate^{[13, 14, 16, 19, 20, 21]}. This is desirable because it increases data retention at high temperatures. The change in sheet resistance during phase transition is more than five orders of magnitude which implies that the material is suitable for PCM application^[22].

Figure 2(b) shows the variation of sheet resistance versus temperature for Sn$_{40}$Se$_{60}$ thin films annealed at different heating rates. The results show a positive shift in the steep decline of sheet resistance as the heating rate is increased. For the heating rates of 1,2.5,5,7.5 and 10 K/min the crystallization temperatures obtained from the minimum of the derivatives of the sheet resistance are 126.7 $\pm$ 0.2 C,135.9 $\pm$ 0.2 C,156.6 $\pm$ 0.3 C,166.1 $\pm$ 0.4 C and 175.0 $\pm$ 0.3 C respectively. The Kissinger plot for Sn$_{40}$Se$_{60}$ alloy is shown in Figure 3. The obtained value of the slope was used to calculate the activation energy according to the Kissinger Equation (1)^[23].

$\ln (\beta T_{\rm C}^{-2})=-\frac{\Delta E}{K_{\rm B}}\left(\frac{1}{T_{\rm C}} \right)+\,{\rm constant}.$

(1)

The activation energy ($\Delta E$) was found to be 0.62 $\pm $ 0.07~eV.

This value compares well with what was obtained by another group^[22] on a similar combination where they reported an activation energy of 0.57 $\pm$ 0.09 eV. The obtained value is much higher than that reported for GST (0.39 eV)^{[16, 17, 18, 19, 20, 21, 24, 25]}. This is a desirable property since the activation energy affects resistance to thermally activated re-crystallization. It has been reported that the resistance to re-crystallization is proportional to the activation energy^[26]. Therefore,the high obtained value of activation energy for this alloy implies that PCM cells made from its samples will have higher resistance to recrystallization. The values of crystallization temperature of the alloys at all heating rates: 1,2.5,5,7.5 and 10 K/min was much higher than the PCM operating temperature which is 85 C^[7]. This is an important advantage for this alloy because it is essential to prevent self transition of recording materials between the two phases. Hence it can be expected that a PCM made from this alloy will remain stable in its amorphous and crystalline states at the operating temperature.

4. Conclusion

The crystallization kinetics of Sn$_{40}$Se$_{60}$ thin films has been successfully investigated using sheet resistance versus temperature measurements. From the results obtained,Sn$_{40}$Se$_{60}$ is a promising alloy for PCM applications because of its high sheet resistance in the amorphous state of 195 M$\Omega $/$\square$,high electrical contrast of five orders of magnitude,and high crystallization temperature of 156.6 $\pm$ 0.3 C. The crystallization kinetics results of Sn$_{40}$Se$_{60}$ thin film obtained in this research provide critical insight in the search for better PCM materials that exhibit superior properties better than the current levels.

Acknowledgements

The authors would like to acknowledge the assistance they received from the Physics Department of Kenyatta University in the form of facilitation to use laboratories at the University of Nairobi and Kenyatta University in acquiring necessary data for this work. We also acknowledge the Physics Laboratory technical staff for their technical support while carrying out the laboratory work.