Access Official Technical PDF

Share Summary:LinkedInEmail

High thermal conductivity can make a material attractive for removing heat from semiconductor devices, but it is not the only property that matters when integrating a heat spreader with a chip. Boron arsenide thermal expansion is also important: a mismatch between joined materials as temperature changes can contribute to thermal stress at an interface. Measurements of cubic boron arsenide (BAs) provide a clearer basis for evaluating this property alongside its heat-transport potential.

From 300 K to 773 K, the measured linear thermal expansion coefficient of BAs rises from (4.2 ± 0.4) × 10-6 K-1 to (6.7 ± 0.7) × 10-6 K-1. The room-temperature result is close to first-principles calculations that account for longer-range atomic interactions. Together with specific-heat data, it also supports a room-temperature Grüneisen parameter of 0.84 ± 0.09, consistent with the calculated value of 0.82.

Why thermal expansion matters in semiconductor assemblies

As electronic devices become more compact and transistor density increases, removing heat from localized hot spots remains a key engineering challenge. Poor heat dissipation can affect device performance and contribute to failure mechanisms. Materials with high thermal conductivity are therefore of interest as possible routes for spreading heat away from active regions.

However, a heat-spreading layer and a semiconductor device expand by different amounts as their temperature changes. If their coefficients of thermal expansion differ substantially, the resulting mismatch can create thermal stress in the assembly. This makes expansion compatibility relevant to materials selection, bonding and the reliability of device architectures—not just to the thermal conductivity of an individual material.

Diamond illustrates this trade-off. It has very high thermal conductivity, but its thermal expansion coefficient is much lower than those of common semiconductors such as silicon and gallium arsenide. The mismatch can contribute to stress at a bonded interface. BAs is of interest because its measured expansion falls closer to the values of these semiconductor materials.

Boron arsenide thermal expansion: measurements and results

The BAs crystals were examined using high-temperature X-ray diffraction (XRD). The lattice parameter—the dimension describing the cubic crystal’s repeating structure—was determined at temperatures between 300 K and 773 K. It increased monotonically across that range. The linear thermal expansion coefficient was calculated from how the lattice parameter changes with temperature.

At 300 K, the measured coefficient is (4.2 ± 0.4) × 10-6 K-1. It increases with temperature, reaching (6.7 ± 0.7) × 10-6 K-1 at 773 K. For a cubic material, the volumetric coefficient is approximately three times the linear coefficient; using the room-temperature measurement gives about 1.26 × 10-5 K-1.

The XRD patterns showed no phase change in the sample up to 773 K under the measurement conditions. This observation establishes the structural behavior of the tested material over the reported range; it should not be taken as a general qualification of a device or an operating limit for every BAs component.

What the calculations reveal about lattice anharmonicity

Thermal expansion is linked to lattice vibrations and their anharmonic behavior. In the study, first-principles calculations using interactions extending to fifth-nearest neighbours produced a room-temperature linear expansion coefficient of 4.0 × 10-6 K-1, close to the measured value. The calculated and measured results agree well over most of the temperature range, with a small difference at higher temperatures.

This comparison helps explain why earlier theoretical estimates differed. A previous calculation using shorter-range anharmonic force constants reported a lower room-temperature value of 3.04 × 10-6 K-1. The results indicate that including longer-range atomic interactions is important for accurately describing BAs lattice anharmonicity and predicting its thermal expansion.

The Grüneisen parameter provides another way to characterize the relationship between lattice vibrations and changes in material volume. Combining measured thermal expansion and specific heat with a separately measured bulk modulus gives a room-temperature value of 0.84 ± 0.09. The first-principles result is 0.82. Agreement between these values supports the consistency of the measurements and calculations; both are substantially below some earlier theoretical estimates.

Implications for material selection and integration

Compared with diamond and cubic boron nitride, BAs has a thermal expansion coefficient closer to those of silicon and III-V semiconductors. At room temperature, its coefficient is about 50% higher than silicon’s and about 40% lower than gallium arsenide’s. These comparisons do not establish that BAs is a drop-in replacement for another material, but they identify a potentially useful compatibility advantage for semiconductor thermal-management designs.

For engineers and system integrators evaluating heat-spreading materials, the findings underline the need to consider several properties together:

  • Thermal conductivity: BAs crystals have been reported with room-temperature values around 1000 W m-1 K-1, while impurities and defects can reduce thermal conductivity.
  • Expansion compatibility: BAs expands more like common semiconductors than diamond or cubic boron nitride, which can matter when assessing thermally induced stress.
  • Temperature-dependent behavior: The measured coefficient increases from 300 K to 773 K, so comparisons should reflect the temperatures relevant to an application.
  • Model accuracy: Reliable prediction of expansion requires accounting for sufficiently long-range atomic interactions.

Conclusion

Cubic boron arsenide combines promising thermal-transport potential with a thermal expansion coefficient that is more compatible with silicon and III-V semiconductors than those of diamond and cubic boron nitride. The close agreement between measured and calculated expansion—and between measured and calculated Grüneisen parameters—also strengthens the understanding of BAs lattice anharmonicity. For thermal-management decisions, the practical takeaway is to assess conductivity, expansion mismatch and temperature range together, while treating material compatibility as one part of a broader integration evaluation.

Access Official Technical PDF

About The Author