Phase-change materials offer a promising route for the practical realisation of new forms of general-purpose and ‘brain-like’ computers. An experimental proof-of-principle of such remakable capabilities is presented that includes (i) the reliable execution by a phase-change ‘processor’ of the four basic arithmetic functions of addition, subtraction, multiplication and division, (ii) the demonstration of an ‘integrate and fire’ hardware neuron using a single phase-change cell and (iii) the expostion of synaptic-like functionality via the ‘memflector’, an optical analogue of the memristor.

A theoretical study of the write, read, and erase processes in electrical scanning probe storage on phase-change media is presented. Electrical, thermal, and phase-transformation mechanisms are considered to produce a physically realistic description of this new approach to ultrahigh-density data storage. Models developed are applied to the design of a suitable storage layer stack with the necessary electrical, thermal, and tribological properties to support recorded bits of nanometric scale. The detailed structure of nanoscale crystalline and amorphous bits is also predicted. For an optimized trilayer stack comprising Ge/sub 2/Sb/sub 2/Te/sub 5/ sandwiched by amorphous or diamond-like carbon layers, crystalline bits were roughly trapezoidal in shape while amorphous bits were semi-ellipsoidal. In both cases, the energy required to write individual bits was very low (of the order of a few hundred picoJoules). Amorphous marks could be directly overwritten (erased), but crystalline bits could not. Readout performance was investigated by calculating the readout current as the tip scanned over isolated bits and bit patterns of increasing density. The highest readout contrast was generated by isolated crystalline bits in an amorphous matrix, but the narrowest readout pulses arose from isolated amorphous marks in a crystalline background. To assess the ultimate density capability of electrical probe recording the role of write-induced intersymbol interference and the thermodynamic stability of nanoscale marks were also studied.

The phase transition between the amorphous and crystalline states of Ge₂Sb₂Te₅ has been studied by exposure of thin films to series of 60 femtosecond (fs) amplified laser pulses. The analysis of microscope images of marks of tens of microns in size provide an opportunity to examine the effect of a continuous range of optical fluence. For a fixed number of pulses, the dependence of the area of the crystalline mark upon the fluence is well described by simple algebraic results that provide strong evidence that thermal transport within the sample is one-dimensional (vertical). The crystalline mark area was thus defined by the incident fs laser beam profile rather than by lateral heat diffusion, with a sharp transition between the crystalline and amorphous materials as confirmed from line scans of the microscope images. A simplified, one-dimensional model that accounts for optical absorption, thermal transport and thermally activated crystallization provides values of the optical reflectivity and mark area that are in very good quantitative agreement with the experimental data, further justifying the one-dimensional heat flow assumption. Typically, for fluences below the damage threshold, the crystalline mark has annular shape, with the fluence at the centre of the irradiated mark being sufficient to induce melting. The fluence at the centre of the mark was correlated with the melt depth from the thermal model to correctly predict the observed melt fluence thresholds and to explain the closure and persistence of the annular crystalline marks as functions of laser fluence and pulse number. A solid elliptical mark may be obtained for smaller fluences. The analysis of marks made by amplified fs pulses present a new and effective means of observing the crystallization dynamics of phase-change material at elevated temperatures near the melting point, which provided estimates of the growth velocity in the range 7-9 m/s. Furthermore, finer control over the crystallization process in phase-change media can be obtained by controlling the number of pulses which, along with the laser fluence, can be tailored to any medium stack with relaxed restrictions on the thermal properties of the layers in the stack.

A systematic design of practicable media suitable for rewritable, ultrahigh density (>;1Tbit/sq.in.), high data rate (>;1Mbit/s/tip) scanning-probe phase-change memories is presented. The basic design requirements were met by a Si/TiN/Ge₂Sb₂Te₅ (GST)/diamond-like carbon structure, with properly tailored electrical and thermal conductivities. Various alternatives for providing rewritability were investigated. In the first case, amorphous marks were written into a crystalline starting phase and subsequently erased by recrystallization, as in other already established phase-change memory technologies. Results imply that this approach is also appropriate for probe-based memories. However, experimentally, the successful writing of amorphous bits using scanning electrical probes has not been widely reported. In light of this, a second approach has been studied, that of writing crystalline bits in an amorphous starting matrix, with subsequent erasure by reamorphization. With conventional phase-change materials, such as continuous films of GST, this approach invariably leads to the formation of a crystalline “halo” surrounding the erased (reamorphized) region, with severe adverse consequences on the achievable density. Suppression of the “halo” was achieved using patterned media or slow-growth phase-change media, with the latter seemingly more viable.

A mark-length write strategy for multiterabit per square inch scanned-probe memories is described that promises to increase the achievable user density by at least 50%, and potentially up to 100% or more, over conventional approaches. The viability of the write strategy has been demonstrated by experimental scanning probe write/read measurements on phase-change (GeSbTe) media. The advantages offered by adopting mark-length recording are likely to be equally applicable to other forms of scanned probe storage.