AIMS, at the School of Physics, Suranaree University of Technology, combines artificial intelligence with first-principles modeling to discover, understand and predict new materials.
AIMS is built on computational condensed matter physics: density functional theory (DFT) calculations of real materials, checked against X-ray absorption, infrared, photoemission and other experiments. Machine learning sits on top of that foundation. It makes the physics faster and reaches problems that physics alone cannot.
Electronic structure, band gap engineering and defect formation energies with VASP and Quantum ESPRESSO.
XAS, XPS and IR spectra that fingerprint local structure and validate experimental data.
Materials discovery, phase identification and image analysis for medical and agricultural problems.
Collaborators: NANOTEC (NSTDA), Thai Synchrotron National Lab, Tohoku University, KAIST, UC Berkeley, Case Western Reserve University and the Institute for Basic Science. We are looking for partners in experimental physics, chemistry, materials engineering and medical research.
Two tracks, one method: physics-based models for understanding, machine learning for speed.
The foundation of the group. First-principles calculations of defects, interfaces, surfaces and layered materials, always tied back to a measurement.
Where is the hydrogen, manganese or calcium atom, and what is it doing? We compute vibrational modes and X-ray absorption spectra for candidate defect sites and match them to infrared and XANES measurements in SrTiO3, BaTiO3 and CdTe.
Strain, intermixing and the two-dimensional electron gas at LaAlO3/SrTiO3, and what core-level binding energy shifts in XPS reveal about the species sitting on a surface.
Band structures of HfSe2 and other layered materials, tuned by strain and alkali-metal dosing, compared directly with angle-resolved photoemission. MXenes and negative electronic compressibility.
Hydrogen storage, lithium-sulfur and lithium-ion batteries, supercapacitors and photocatalysts: how defects, doping and strain change ion diffusion, adsorption and charge transport.
Electronic, optical and polarization properties of (Ba,Sr,Ca,Bi,Na)TiO3 with VASP, strain effects for piezo-photocatalysis, and elastic properties of materials under pressure.
Figures are from our published papers and conference talks. Click a figure to open it full size.
Machine learning built on top of that physics: faster spectra, smarter screening, and real-world data problems from medicine to energy markets.
Neural networks that predict XANES spectra from atomic structure instantly, inverse models that recover structure from XAS with calibrated uncertainty, and automated phase identification from XRD patterns.
Predicting the effect of cationic dopants on hydroxyapatite from large databases, then checking the best candidates against DFT elastic tensors. Neural networks for the mechanical properties of bio-cement.
Diabetic retinopathy classification from fundus images, optical coherence tomography with deep learning for chick gender determination, and real-time pupil tracking for retinal imaging.
Walk-forward benchmarks of gradient boosting and deep models for electricity prices on the Japan Electric Power Exchange, and neural networks for black hole mass in X-ray reverberating AGNs.
Our DFT and machine-learning work runs on high-performance computing (HPC) clusters. Students learn to use them well, a skill that carries into academic research and into industry careers in computational science and data science.
Several kinds of support are available to students who join the group. Availability, amounts and deadlines change each year, so contact us before you apply.
Support from the Research Unit (RU) on Advance Nano at Suranaree University of Technology, available to students working with the group.
A scholarship for graduate students (master's or PhD) who do their thesis with a SUT faculty member holding research funding from an external source.
Student support paid from Fundamental Fund research projects run in the group.
Paid research positions for graduates who already hold a master's or PhD degree.
Funding from industry projects and other sources, for people who want to work on applied problems with partners.
Not sure which one fits your situation?
Ask about fundingP. Sriboriboon, I. Fongkaew
N. Pandech, L. Ngamwongwan, S. Jungthawan, W. Sailuam, I. Fongkaew. 7th Sino-Thai Symposium on High Energy Physics, Astrophysics, Materials Science and Beyond
L. Liu. Stage 1 report, School of Physics, SUT
Source: Google Scholar profiles of Ittipon Fongkaew and Sirichok Jungthawan, merged and de-duplicated by title, November 2026.
Sirichok Jungthawan
Ittipon Fongkaew
Prof. Dr. Sukit LimpijumnongDistinguished Professor
Prof. Dr. Yoshiyuki KawazoeDistinguished Professor
Prof. Young-Hoon KiemDistinguished Professor
Keerati Keeratikarn
Panithan Sriboriboon
Xingji Tang
Longxun Liu
Ratsadakorn Yimsabai ManeewongWe welcome undergraduate, master's and PhD students curious about physics, programming and machine learning.
Get in touch