On Bashing Servers: ShengBTE, A Boltzmann Transport Equation Solver.

[NNadir]

The paper to which I'll refer in this post is this one: Wu Li, Jesús Carrete, Nebil A. Katcho, Natalio Mingo, ShengBTE: A solver of the Boltzmann transport equation for phonons, Computer Physics Communications, Volume 185, Issue 6, 2014, Pages 1747-1758,

The latest public Bête noire is a sudden concern about the cost (and reliability) of electricity being driven by the power demands of servers usually in our end of the political spectrum with a sense of grave negatives, a supposition of that the whole affair is driven by dark rapacious billionaires.

I note, because of the rather dubious proposition that electric cars are "green" - they're more often not so than are so - few people wonder about replacing the world's gasoline demand with demand for electricity to power cars and the cost (and reliability) of electricity for this use.

It's, um, more than a little schizoid to compare these two states of affairs.

There are zero technologies than cannot be utilized for both good and bad purposes; the fault lies not with the technology itself, but with the ethical choice of how the technology is utilized. DU runs on servers, and so does Google Scholar, to which I appeal whenever a scientific (and sometimes even other academic) question pops into my mind.

As my survival is more and more precarious as I approach the human limits of aging, I'm trying to squeeze sometime into writing down some of the most meaningful energy ideas that have occurred to me - for better or worse - to leave to my son. Included in these is a general approach to utilizing depleted and "once through" uranium to generate plutonium, in a "breed and burn" system, something that will involve the isolation and use of thorium currently dumped along with lanthanide mine tailings.

I have long been familiar with the preparation and properties of uranium nitride, but as I was thinking about what to write, I realized that my files contain very little about thorium nitride.

The advantage of nitride fuels is their high thermal conductivity and high melting points which makes them ideal nuclear fuels, and a sort of neutral issue is that nitrides generate radioactive ¹⁴C and tritium, both of which produce angst among antinukes, but both of which are potentially very useful given their low neutron capture cross sections.

Working to correct this deficiency in my understanding of thorium nitride, I downloaded, among a multitude of others, this paper: Barbara Szpunar, Jayangani I. Ranasinghe, Linu Malakkal, Jerzy A. Szpunar, First principles investigation of thermal properties of thorium mononitride, Journal of Alloys and Compounds, Volume 879, 2021, 160467.

Excerpts from this thorium nitride paper:

Urania fuel, which is used in conventional nuclear reactors, is not suitable for some designs of new generation reactors (e.g., SuperCritical Water Reactor) due to its low thermal conductivity [1]. In the context of finding a sustainable development solution to the use of non-renewable energy sources, innovative research towards enhanced accident-tolerant nuclear fuel (EATF) that can withstand the loss of coolant for a long time is gaining momentum. EATF materials must have higher thermal conductivities (κ ) to prevent meltdown [2]. High-density metallic compounds, uranium silicide (U₃Si₂), uranium and thorium nitrides (UN, ThN) [3] have been proposed as alternative EATFs [2] for implementation as lower enrichment fuel.

In our previous papers [4], [5] we have investigated UN, which has the same cubic structure (Fmm symmetry) as ThN, discussed here. In contrast however, ThN is non-magnetic, but also metallic; therefore, its thermal conductivity does not deteriorate like the lattice-governed thermal conductivity in insulators (e.g. urania [6]). This is due to the increasing presence of electronic carriers with mobility as temperature rises. Since both electronic conductivity and electronic contribution to thermal conductivity are related to electron mobility, they can be derived from each other via the Wiedemann-Franz proportionality law (WFL), which is very useful in determining the contribution from phonons and electrons to the measured thermal conductivity...

A little further on they write:

Unlike urania, fewer such studies have been done on these alternative fuels such as ThN. Here we used the generalized gradient approximation (GGA) of the Perdew, Burke, and Ernzerhof functional (PBE) [8] as implemented in Quantum ESPRESSO (QE) code [9] and a software package, ShengBTE [10], to evaluate the phonon contribution to heat capacity and thermal conductivity.

I added the bold. Reference 10 is the paper referenced at the beginning of this post.

From that paper, reference 10, we see the following written about approaching an ab initio (first principles) approach to modeling the thermal conductivity of materials:

The lattice thermal conductivity, denoted as (kappa sub l) is a key property in countless applications. Important technologies that demand specific materials with tailored thermal conductivities include thermoelectricity [1], heat management [2] and development of non-volatile memory based on phase-change materials [3]. Knowing how thermal conductivity changes under extreme conditions is also essential for understanding the behavior of the Earth’s mantle [4]. Hence it is extremely desirable to have workflows available that yield predictive, parameter-free estimates of using only basic information about the chemical structure of the crystal. Unfortunately, until recently the development of such approaches has been hindered both by methodological difficulties and by CPU time constraints, and to date no widely available software package exists to tackle this problem...

...The phonon-contributed part of the total thermal conductivity is the lattice thermal conductivity. Phonon properties including frequencies, velocities and scattering rates are largely determined by interatomic force constants (IFCs). One important approach to study phonon transport in solids is the Boltzmann transport equation (BTE) [5]. However, solving this equation is far from trivial. Although it was originally formulated by Peierls in 1929, even as late as in 1960 looking for a direct solution to the BTE was regarded as a hopeless endeavor [5]. Instead, many solutions to the BTE conventionally rely on the relaxation time approximation (RTA) along with the Debye approximation, neglecting the true phonon dispersions, and several parameters are introduced to treat different scattering mechanisms. To improve on this, Callaway proposed a model [6] that treats the quasimomentum-conserving normal processes and the non-quasimomentum-conserving Umklapp processes on a different footing. Very recently, an improvement upon Callaway’s model has been proposed by Allen [7]. All these models involve parameters that are fitted to experimental data, and thus lack predictive power. In 1995 a practically feasible, iterative numerical method was proposed [8], [9], [10] to solve the BTE accurately. Early attempts started with a parameterized semiempirical interatomic potential, but such an approach suffers from problems of low accuracy and lack of transferability, since an appropriate potential for each compound must be developed...

A little further on:

In this paper we present a software package, ShengBTE,3 able to solve the Boltzmann transport equation for phonons starting from a set of IFCs obtained ab initio. The program can compute converged sets of phonon scattering rates and use them to obtain and many related quantities. ShengBTE harnesses the symmetries of the system to make these calculations more efficient and is able to deal with isotropic as well as anisotropic crystals. Moreover, it also implements an approximation developed by some of us to efficiently and accurately predict the thermal conductivity of nanowires under a diffusive boundary assumption [30].

(IFC = Interatomic Force Constants)

I have added the bold in all cases.

There is a considerable technical discussion in the full paper and a test run on indium arsenide, a conductive crystal. (Recently boron nitride crystals have been developed that have higher thermal conductivity than diamond, displacing it as the all time highest thermally conductive material.)

The point is that this program is one that can be utilized to save the world, and the use of electricity to run computers in this case is a positive use, as opposed to uses like spreading the insanity on "Truth Social" from the all caps orange pedophile in the White House, obviously a negative use.

We can have clean electricity, generated using nuclear energy, for which thorium nitride has interesting and possibly extremely valuable properties.

The ShengBTE program represents a tiny fraction of what computational power can do for us in a positive sense, (DU is another) and there are thousands more cases, and whining about the demands of electricity for servers can thus be as counterproductive as productive.

Have a nice weekend.