§1 — Why laser heating in a DAC?
Diamond anvil cells routinely generate extreme static pressures, but many conditions of geophysical and materials interest require simultaneously high P and high T. Resistive heating of the entire cell body is capped near 1200–1500 K by gasket weakening, diamond softening, and anvil oxidation. Laser heating bypasses these constraints by focusing a high-power beam directly onto the sample through the diamond anvils, reaching 1200–7000 K at any pressure up to multi-megabar. It has become the standard path to deep-Earth conditions, melting curves, and synthesis of high-pressure phases [Aprilis et al. 2020, Crystals 10:459].

§2 — Laser types and their absorption physics
Two laser families are used in LH-DAC, chosen based on the optical properties of the sample [Campbell 2008, Rev. Sci. Instrum. 79:015108]:
• Near-IR: Nd:YAG / Nd:YLF / Yb:fiber (1064 nm) — the most common choice. Metals, iron oxides, and most silicates absorb strongly at 1 μm; the diamond anvils are transparent. Typical powers: 20–200 W CW. Excellent TEM&sub0;&sub0; beam quality. Used in double-sided geometry at all major beamlines: ESRF ID27, APS GSECARS 13-ID-D, SPring-8 BL10XU. Steep radial and axial temperature gradients — spectroscopic sampling of colder zones causes measured T to underestimate peak T by 5–15% [Benedetti & Loubeyre 2004, High Press. Res. 24:423].
• Mid-IR: CO₂ laser (10.6 μm) — the alternative for samples transparent at 1064 nm: MgO, silicates, noble gases, molecular solids. Diamond absorbs at 2000–2500 cm⁻¹ (two-phonon band) but is transparent at 943 cm⁻¹ (10.6 μm). Used in single-sided geometry through a ZnSe or ZnS window. More uniform heating in transparent materials gives lower axial gradients [Trots et al. 2013].

§3 — Power density and beam waist
The critical parameter is the beam waist w₀ (1/e² radius at the sample), typically 5–30 μm. Peak intensity at the sample [Campbell 2008]:
I₀ (W/μm²) = 2 × P (W) / (π × w₀² (μm²)) Typical conditions: 20–100 W, w₀ = 10–20 μm → I₀ ≈ 0.07–0.6 W/μm² for opaque metals. Radial temperature gradients are steep: ΔT ≈ 200–500 K/μm from the hotspot centre. Laser power must be stable to ±1–2% rms to avoid thermal runaway above 3000 K.

§4 — Double-sided heating
Single-sided heating creates a severe axial gradient — the laser face reaches target T while the opposite face may be 1000–2000 K cooler, depending on sample thickness and thermal conductivity. Double-sided heating uses two counter-propagating beams of equal power aligned collinearly through both anvils, reducing axial gradients by ~50% [Heinz & Jeanloz 1987, Geophys. Monogr. 39:113; Campbell 2008]. It is now the standard geometry at all synchrotron LH-DAC beamlines. Key requirements:
• Lateral alignment of the two beams: <5 μm offset
• Equal power split between upstream and downstream beam
• Independent temperature readout from both sides

§5 — Temperature measurement: spectroradiometry and Planck’s law
Above ~1200 K the heated sample emits measurable thermal radiation. Temperature is extracted by fitting the emission spectrum (600–900 nm) to the grey-body Planck function [Heinz & Jeanloz 1987; Benedetti & Loubeyre 2004]:
I(λ,T) = ε(λ) · C₁ · λ⁻⁵ · [exp(C₂/λT) − 1]⁻¹ C₁ = 3.742×10⁻¹⁶ W·m², C₂ = 1.4388×10⁻² m·K (CODATA 2018). The grey-body assumption (ε independent of λ) introduces systematic errors: wavelength-dependent emissivity causes fitted T to diverge from true T by up to 10–15% [Benedetti & Loubeyre 2004]. Typical measurement accuracy: ±50–200 K at 2000–5000 K. The spectroradiometric system must be calibrated with a certified standard lamp before each session. The 4-colour multispectral imaging technique of Campbell (2008) allows 2D temperature mapping across the hotspot, greatly reducing the gradient-related bias [Campbell 2008].

§6 — Pressure gauges in LH-DAC
Ruby fluorescence broadens and weakens above ~600 K and cannot be used during laser heating. Alternatives, in decreasing order of difficulty:
• Sm²⁺:YAG (688.29 nm): dλ/dT < 1×10⁻³ nm/K — nearly insensitive to temperature [Trots et al. 2013, J. Geophys. Res.]. Place the bead >50 μm from the hotspot (where T < 600 K). The best optical gauge for LH-DAC.
• Ruby before and after heating (room T): standard practice when Sm:YAG is unavailable; assumes no irreversible pressure change on quench.
• XRD of a co-loaded standard (Au, Pt, NaCl): most accurate, continuous in-situ; requires synchrotron access [Dewaele et al. 2008, Phys. Rev. B 78:104102].
• Diamond Raman edge: continuous in-situ; apply thermal correction dω/dT = −0.022 cm⁻¹/K.

§7 — Practical checklist
1. Load a Sm:YAG bead at >50 μm from the sample — outside the hotspot
2. Calibrate the spectroradiometer with a standard lamp before the session
3. Record P (ruby or Sm:YAG) before heating and after quench to room T
4. Align both beams with <5 μm lateral offset for double-sided geometry
5. Increase power slowly — watch for first glow onset (T > 1200 K)
6. Monitor power stability ±1–2% rms to avoid thermal runaway
7. After quench: verify for phase change by XRD or Raman spectroscopy

§8 — References
Heinz D.L. & Jeanloz R. (1987) — Temperature measurement in LH-DAC. Geophys. Monogr. 39:113
Campbell A.J. (2008) — Multispectral imaging radiometry in LH-DAC. Rev. Sci. Instrum. 79:015108
Benedetti L.R. & Loubeyre P. (2004) — Temperature gradients and emissivity in LH-DAC. High Press. Res. 24:423
Trots D.M. et al. (2013) — Sm:YAG primary fluorescence pressure scale. J. Geophys. Res. 118:5805
Dewaele A. et al. (2008) — Compression curves of transition metals with He medium. Phys. Rev. B 78:104102
Aprilis G. et al. (2020) — Practical review of LH-DAC. Crystals 10:459
Fedotenko T. et al. (2019) — LH-DAC system for synchrotron. Rev. Sci. Instrum. 90:104501