Discovering Optimal Materials for Wide Bandgap Semiconductors

In the pursuit of more powerful, efficient, and compact electronic devices, wide bandgap materials are emerging as a leading force. Their superiority over silicon in key aspects such as efficiency, higher junction temperatures, power density, thinner drift regions, and switching speeds positions them as the materials of choice for the future of power electronics.

As researchers dig deeper into overcoming challenges and refining manufacturing processes, wide bandgap materials are not just an alternative but an indicator of a new era in electronic engineering. Moreover, Wide bandgap semiconductors in ceramic packaging materials have shown significantly similar thermal coefficient of expansion to electrically insulating ceramics as shown in Figure 1 which enables them to be used for high temperature and power applications. However, wide bandgap materials are not the ideal choice for bipolar devices as the ratio of electron-to-hole mobility is lower than traditional silicon.

To compare different limits of semiconductor materials for varied power devices, the use of specific on-resistance R_Onsp is widely accepted as it provides an approximate figure of merit to each semiconductor material. This specific on-resistance expression is based on the theoretical maximum breakdown voltage for a sudden pn junction. Moreover, the breakdown voltage is derived from the bond between carrier multiplication and critical electric field which is represented in terms of bandgap energy of the material. Therefore, it is clearly seen that while keeping the expense and maturity of material processing for different semiconductor materials in mind, the ideal choice is GaN for unipolar devices whereas C is a better choice for bipolar devices.

Figure 1: A Comparative Study of the Coefficient of Thermal Expansion Between Several Wide Bandgap Semiconductor Materials and Package Ceramics

Exploring New Materials for Semiconductors with Emphasis on Bandgap Energy and Critical Electric Field

The performance of semiconductors is evaluated by certain electrical parameters like theoretical maximum junction temperature, critical field, and thermal conductivity which decides the overall effectiveness of the material. The current leading semiconductor materials in focus are Gap, SiC, and diamond, as these materials have thermal conductivity more than or close to traditional Si and GaN semiconductors. However, diamond is the pioneer as it has the largest bandgap and thermal conductivity with the highest electron mobility of any of the materials listed above. Although having such superior qualities, diamond falls back due to low CTE and less mature device fabrication technology which can ultimately cause a thermo-mechanical mismatch thereby decreasing power and efficiency.

On the other hand, SiC and GaN materials are more suited as package materials as they provide a much better thermo-mechanical relationship due to higher CTE when compared to diamond, moreover, the fabrication technology for these devices is much more technologically advanced and developed. In the current industry, manufacturers have started using GaN material for electro-optic and RF applications, and SiC is used for commercial power devices like Schottky devices. In general, it is desirable to enhance the thermal conductivity of both the associated package material and semiconductor material.

SiC displays a higher thermal conductivity when compared to other materials like GaN, however, this advantage is offset by GaN's larger bandgap and other electric properties, such as the theoretical maximum junction temperature of operation and the critical electric field. Consequently, it is currently concluded that GaN stands out as one of the most promising semiconductor materials for advanced power electronic devices mostly due to its close coefficient of thermal expansion match with insulating ceramics, a reasonable electron mobility value, and a thermal conductivity similar to that of Si.

The conventional on-resistance calculations, which rely on the critical electric field of the material, have been proven to be unreliable. Through recent calculations incorporating additional data, a more accurate assessment of on-resistance reveals an even greater advantage for wider bandgap materials. Consequently, this highlights the preference for Gallium Nitride over Silicon Carbide in terms of on-resistance. The revised relationship between the bandgap and critical electric field is shown in Equation 1, where ε is the permittivity of the semiconductor material, q is the electron charge, and E_G is the bandgap energy.

Equation 1: Revised Equation Defining the Relationship Between Critical Electric Field and Bandgap Energy

Using the known values of critical electric field for various materials, we can directly find the bandgap energy for the same materials, which is favourable for evaluating the performance of direct gap semiconductors as it is easy to obtain large values of critical field. On the other hand, the relationship between Specific On-Resistance and band gap for semiconductor materials is given in Equation 2 which provides similar results as Equation 1.

Equation 2: Equation Defining the Relationship Between Specific On-Resistance and Bandgap Energy

The current age industries are bound to find the optimal semiconductor material systems for future power electronic devices, with Gallium Nitride, Silicon Carbide, and Carbon coming out as prime contenders. The critical criteria of possessing an electron-to-hole mobility ratio below three and an electron mobility surpassing 1000 cm/V s tapers down the preference to a handful of materials, including diamond. The near-one mobility ratio of diamond makes it highly suitable for bipolar device designs, especially in high-temperature operating conditions. Its mixture of characteristics, comprising a close mobility ratio, high carrier mobilities, a substantial bandgap, and excellent thermal conductivity, positions diamond as an optimal choice for electronic devices across varying power levels and types.

However, acknowledging current fabrication technology and considerations of bandgap size and thermo-mechanical properties, Gallium Nitride emerges as the primary choice for the upcoming technological advancement. Silicon Carbide, being less desirable due to material quality and impurity doping issues, remains a worthy candidate. The analysis comparing various electrical parameters highlights the theoretical superiority of GaN over Si in optimizing electrical behavior for power electronic applications, overcoming the shortcomings of traditional Silicon.

Over the past decade, it has become clearer that the traditional relationship between energy bandgap and critical electric field, designed for a handful of semiconductor materials, lacks universal applicability. New relationships have been derived based on data from different semiconductors which include both direct-gap and indirect-gap materials, demanding attention to the need for more precise approaches in evaluating semiconductor materials for advanced electronic applications.