Access Official Technical PDF

Share Summary:LinkedInEmail

Predicting how silicon responds to intense, rapid heating requires more than equilibrium values for its overall thermal properties. In semiconductor laser processing, energy can be deposited on short time and spatial scales, creating conditions where the electron and phonon subsystems may not remain in local thermodynamic equilibrium. Models of this behavior need to describe the properties of each subsystem. In particular, silicon electron thermal conductivity and heat capacity depend on temperature, carrier concentration, and the statistical state of the charge carriers.

A quantum-statistical treatment provides a way to estimate these electron properties across a broad range of conditions. It accounts for the changing population of electrons and holes, including the effects of carrier degeneracy and temperature-dependent bandgap narrowing. These factors become especially important as silicon is heated toward melting, when its electronic structure and carrier concentration can change substantially.

Why model silicon’s electron subsystem?

In a solid, heat is transported and stored by both the lattice and free charge carriers. Under ordinary equilibrium conditions, measured thermal conductivity and heat capacity generally describe the combined material response. Those total values are useful, but they do not by themselves separate the electron contribution from the phonon, or lattice, contribution.

That distinction matters in models of rapid laser heating. When energy is deposited faster than the subsystems can equilibrate, a single equilibrium temperature and a combined set of material properties may not adequately represent the evolving state. A two-subsystem description treats electrons and phonons as interacting parts of the material and requires thermophysical properties for each.

Experiments provide important reference data, but measurements are limited to particular conditions and can be difficult near melting. A theoretical model can help estimate electron properties across a wider temperature range. The study considered intrinsic silicon from 300 K to 2000 K and compared calculated properties with available experimental results.

Silicon electron thermal conductivity and carrier statistics

The carrier population is central to calculating electron thermal conductivity. In intrinsic silicon, thermal excitation produces electrons in the conduction band and holes in the valence band. Under equilibrium conditions, their concentrations are equal. As temperature rises, more carriers are generated, while electrons and holes also recombine.

At low temperatures or low carrier concentrations, the electron gas can be treated as nondegenerate. In this regime, the Maxwell-Boltzmann approximation describes the carrier distribution. With stronger heating, however, the carrier concentration increases and the Fermi level can approach or enter a band. The classical approximation then becomes unsuitable; the model must use Fermi-Dirac statistics, which describe the probability that an available energy state is occupied.

Carrier concentrations calculated with this approach depend on Fermi-Dirac integrals. Because most of these integrals cannot be evaluated analytically, the study uses analytical approximations for integrals of several orders. These provide a practical way to calculate carrier populations and related properties across a range of degeneracy without assuming that the gas is always classical.

How bandgap narrowing affects the model

Silicon’s bandgap is not constant under strong heating. It narrows as temperature increases, partly because of lattice expansion and electron-lattice interactions. At sufficiently high carrier concentrations, collective interactions among carriers also contribute. As the gap becomes smaller, thermal excitation can produce more free carriers, which in turn affects the electronic properties being calculated.

The model therefore incorporates a temperature- and carrier-density-dependent bandgap. This is important near phase-transition conditions: treating the bandgap as fixed can misrepresent carrier concentration and, as a result, the electron subsystem’s heat capacity and thermal conductivity. The calculations also determine the Fermi energy consistently with the intrinsic carrier balance.

Separating electron and lattice contributions

Thermal conductivity

Total thermal conductivity is represented as the sum of lattice and free-carrier contributions. The electron component is calculated using carrier concentrations, mobility, and Fermi-Dirac integrals. The lattice component is estimated separately so that the combined result can be compared with measurements of silicon’s total thermal conductivity.

In the solid-state temperature range examined, lattice conductivity is the larger contribution. The calculated electron share remains small before melting, reaching no more than 4.7% of the total in the reported results. Above melting, the available comparison is limited to experimental equilibrium thermal conductivity for molten silicon; at 2000 K, the reported electron contribution is 4% of that measured total. These comparisons provide context for the electron term without implying that equilibrium measurements fully describe rapid, nonequilibrium heating.

Heat capacity

Heat capacity describes the energy required to raise a system’s temperature. For silicon, the model divides total heat capacity into lattice and free-carrier terms. The electron contribution is obtained from the temperature derivative of the energy density of electron-hole pairs, including both their kinetic energy and the energy associated with the bandgap.

As temperature increases, the calculated electron heat capacity grows relative to the lattice component. It remains a small part of the total in the reported range, reaching 1.4% at the equilibrium melting temperature and 1.72% at 2000 K. The calculated total heat capacity agrees well with the experimental data in the reported ranges below 1000 K and from the melting point to 2000 K.

Practical implications for thermal modeling

For engineers and technical teams building models of silicon processing, the findings reinforce several practical points:

  • Use temperature-dependent carrier properties: carrier concentration and mobility cannot be treated as fixed across strong heating conditions.
  • Choose statistics appropriate to the regime: Maxwell-Boltzmann assumptions may fail when the carrier gas becomes degenerate.
  • Account for bandgap changes: narrowing links temperature and carrier density to the electronic response.
  • Keep subsystem contributions distinct: total equilibrium measurements do not directly identify electron and lattice behavior separately.

Overall, quantum-statistical calculations offer a framework for estimating silicon’s electron heat capacity and thermal conductivity when heating drives the material beyond simple classical assumptions. The calculated properties show acceptable quantitative agreement with the experimental comparisons reported in the study. For modeling rapid heating and phase-transition conditions, the key implication is to represent carrier statistics and bandgap narrowing explicitly, while keeping the electron and lattice contributions separate where the process model requires it.

Access Official Technical PDF

About The Author