Skip to content

Three-Dimensional Temperature Field Model for Multi-Pulse Nanosecond Laser Ablation of Polycrystalline Diamond

FieldValue
JournalMachines
Published2026-06-01
AuthorsZiguang Wang, Hanping Zhang, Shelong Du, Yu Zhang, Xiaoguang Guo
DOI10.3390/machines14060626

A three-dimensional temperature model tracked measured ablation depth in multipulse nanosecond laser processing of polycrystalline diamond. Its relative error was below 20%. More scans, higher pulse energy, and higher pulse frequency increased thermal accumulation and ablation depth. Faster scanning reduced both.

The model combines a Gaussian spatial intensity distribution with a Gaussian temporal pulse. Lambert-Beer attenuation controls energy deposition with depth. Successive pulses are superposed to represent accumulation during scanning. Three-dimensional transient heat conduction then governs heat diffusion through the PCD.

The calculation uses MATLAB V2021, a structured grid, and explicit finite differences. Boundaries far from the laser region are isothermal, while the other outer surfaces are adiabatic. The authors also derive analytical integrals for single-pulse and multipulse irradiation under a semi-infinite assumption.

PCD is represented as homogeneous, continuous, and isotropic, with constant thermal properties. The initial temperature is 300 K. Density is 3.5 g/cm³, specific heat is 0.51 J per g·K, and thermal conductivity is 1800 W per m·K. The absorption efficiency is 0.25, and the absorption coefficient is 0.03 cm⁻¹.

The ablation depth is the depth where the centerline temperature reaches 4098 K. Graphitization is assumed to begin at 1073 K.

The experiments used 10 mm × 10 mm × 0.3 mm PCD specimens with an initial roughness of 0.362 µm. The surfaces were ultrasonically cleaned with anhydrous ethanol.

A 1064 nm fiber nanosecond laser was focused perpendicular to the surface. Pulse width was 100 ns, the focal spot radius was 18.78 µm, and the focusing lens had a 160 mm focal length. Scanning electron microscopy characterized the ablated surface morphology. A Bruker Alicona three-dimensional optical profilometer measured the surface profiles and ablation depths.

Each experiment group varied one process variable using the control variable method. Other conditions were held constant within each series.

Varied parameterTested levelsMeasured and modeled depth response
Number of scans5, 10, 15, 20, 25Depth increased with more scans
Scanning speed25, 50, 75, 100, 125 mm/sDepth decreased as speed increased
Single-pulse energy0.25, 0.3, 0.35, 0.4, 0.45 mJDepth increased with energy
Pulse frequency50, 60, 70, 80, 90 kHzDepth increased with frequency

The model and profilometry data followed the same direction in all four parameter series. Figure 5 used a ±20% error band around each simulated depth. Every measured depth fell within that band, and the relative error was below 20%.

At 0.30 mJ to 0.45 mJ, measured groove depth increased, lateral width expanded, and groove edges became steeper. The model associated this morphology with stronger surface and depth temperature gradients.

The literature review contrasts this work with earlier studies of morphology, graphitization, cooling, ultrafast ablation, and two-dimensional temperature fields. It cites femtosecond lasers as offering greater control of surface quality than nanosecond lasers. The present model addresses three-dimensional accumulation as successive nanosecond pulses arrive before preceding heat fully dissipates. The study does not experimentally compare laser types or their surface quality.

The surface temperature peaks at the spot center and decays radially in an approximately axisymmetric pattern. As single-pulse energy rises from 0.30 mJ to 0.45 mJ, the modeled center temperature rises, the high-temperature region expands, and isotherms extend outward. The modeled thermal footprint also reaches farther into the PCD.

Lateral gradient curves are antisymmetric around the center, with the largest changes near the central region. The absolute gradient increases with scans, energy, and frequency. Faster scanning reduces the gradient and makes the surface distribution more uniform.

Under the Gaussian source, a larger absolute lateral gradient means faster surface temperature decay. It also indicates a more concentrated hot region and a steeper removal boundary.

The depth-direction field peaks near the irradiated surface, then decays into the PCD. Its gradient reaches its most negative value near the surface and approaches zero with depth. More scans or pulse frequency shift the profile in the negative direction. Higher energy increases the gradient magnitude, while faster scanning reduces it.

The lateral gradient controls the removal boundary, while the depth-direction gradient primarily governs ablation depth. These relations connect the simulated thermal field to the measured groove morphology.

The experiments used a fixed 100 ns pulse width, so the pulse-width study was numerical. The modeled widths were 10 ns, 50 ns, 100 ns, and 200 ns. The analysis held the number of scans at 15, scanning speed at 75 mm/s, single-pulse energy at 0.35 mJ, and pulse frequency at 70 kHz.

At constant pulse energy, a shorter width raises instantaneous power density, center temperature, and near-surface concentration. A longer width allows more diffusion during the pulse and produces a more gradual field. The model predicted greater ablation depth as pulse width decreased.

The authors state that electron-phonon nonequilibrium, nonlinear absorption, and non-thermal ablation may matter at picosecond or femtosecond scales. They recommend a two-temperature model or molecular dynamics model for that regime.

The model omits melt splashing, plasma shielding, latent heat of phase change, recoil pressure, and material ejection. It also neglects convective and radiative heat exchange and holds density, specific heat, and thermal conductivity constant.

Neglecting plasma shielding can overestimate laser energy coupled into the PCD. Neglecting latent heat can overestimate local temperature rise. Omitting recoil and ejection limits predictions of detailed ablation morphology.

The authors therefore classify the calculation as a heat-conduction-based thermal response model. They position it for comparing relative parameter effects and temperature-gradient evolution.

Validation covered ablation depth within the tested parameter ranges. The experiments did not measure device-level heat dissipation, subsurface damage, or post-ablation graphitization.

The model links number of scans, scanning speed, pulse energy, and pulse frequency to thermal gradients and ablation depth. Reducing scans, pulse energy, or pulse frequency lowers thermal accumulation and both gradient magnitudes. Increasing scanning speed lowers heat input per unit area and reduces ablation depth.

Lateral gradients indicate the expected groove width and edge steepness. Depth-direction gradients indicate the expected removal depth. Process plans can therefore trade removal depth against thermal concentration using these indicators.

The paper motivates PCD machining for heat spreader manufacture, but the experiment did not qualify a heat spreader device. The model supports relative parameter selection within the tested conditions. Absolute temperatures and detailed groove morphology should remain tied to the tested material and conditions.

The process needs PCD plates with a defined thickness for nanosecond-laser microgroove ablation. The diamond materials catalog offers MPCVD-grown polycrystalline diamond plates and wafers up to 125 mm, with 0.1 to 500 µm material thickness, which covers the paper’s 0.3 mm specimen. 6CCVD’s engineering support can address custom fabrication and process requirements for this laser machining application.

View Original Abstract

Polycrystalline diamond has great potential for power-device heat dissipation and precision manufacturing owing to its exceptional hardness, excellent thermal conductivity, and superior wear resistance. However, the challenges of material removal and controlling thermal damage hinder high-quality machining. In this study, a three-dimensional transient temperature field model is developed for multi-pulse nanosecond laser ablation of polycrystalline diamond. The model incorporates the Gaussian spatial distribution of laser energy, Lambert-Beer depth-dependent absorption, multi-pulse energy superposition, and three-dimensional heat conduction. The heat conduction equation is numerically solved using MATLAB, and lateral and longitudinal temperature gradients are introduced to characterize thermal accumulation and material removal behavior. The model is validated by comparing the predicted ablation depths with experimental measurements, which show consistent variation trends. The results indicate that increasing the number of scans, single-pulse energy, and pulse frequency enhances thermal accumulation, expands the microgroove width, and increases the ablation depth, whereas increasing the scanning speed weakens thermal accumulation and reduces the ablation depth. In addition, a shorter pulse width increases the instantaneous power density and strengthens near-surface thermal concentration. This study provides theoretical guidance for controlling the heat-affected region and optimizing process parameters in precision laser machining of diamond.

  1. 2026 - Fully diamond-based embedded manifold microchannel heat sink: Achieving ultra-high heat flux cooling [Crossref]
  2. 2025 - Simultaneous Enhancement of Wear Resistance and Impact Toughness in Polycrystalline Diamond Compact via Gradient Structure Design [Crossref]
  3. 2025 - Direct Integration of Polycrystalline Diamond with 3C-SiC for Enhanced Thermal Management in GaN HEMTs: Impact of Grain Structure and Interface Engineering [Crossref]
  4. 2025 - Integration of top-side low-temperature diamond on AlGaN/GaN RF HEMT for device-level cooling [Crossref]
  5. 2023 - Preparation of 3-Inch Diamond Film on Silicon Substrate for Thermal Management [Crossref]
  6. 2022 - Wafer-scale polishing of polycrystalline MPACVD-diamond [Crossref]
  7. 2025 - Atomic-level flat polishing of polycrystalline diamond by combining plasma modification and chemical mechanical polishing [Crossref]
  8. 2024 - Machining damage in the fabrication of polycrystalline diamond micro end mill with laser and grinding combined processing method [Crossref]
  9. 2023 - The polishing methods for large area CVD diamond wafer [Crossref]
  10. 2024 - Energy beam-based direct and assisted polishing techniques for diamond: A review [Crossref]