Skip to content

p -Type shallow doping and quality improvement of diamond via pulsed laser irradiation

FieldValue
JournalJournal of Physics D Applied Physics
Published2026-06-30
AuthorsAdedayo Alaga, Saif Al Arafin Taqy, Mahfuz Ahmed Azmain, Brian Samuels, Tajwar Al Mamun
DOI10.1088/1361-6463/ae845a

Pulsed laser irradiation produced p-type polycrystalline diamond with a hole mobility of 7.09 cm² V⁻¹ s⁻¹ on a commercial 300 µm PCD wafer. The treated material carried +1.166 × 10¹⁹ cm⁻³ holes and had a sheet resistivity of 603.7 Ω sq⁻¹. Its Raman quality factor was 100%, while its trans-polyacetylene defect peak fell 43.08% from the laser-damaged state.

Boric acid served as the boron source. The authors immersed each PCD sample in 10 ml of a 4% solution before irradiation with a 248 nm KrF excimer laser. The laser delivered 25 ns pulses at room temperature and ambient pressure.

The first raster exposure used 0.34 J cm⁻² at 1 Hz for 500 pulses. The recovery exposure used 0.17 J cm⁻² at 1 Hz for 4000 pulses.

The authors attribute doping to rapid photothermal decomposition, localized microplasma formation, and boron movement along grain boundaries. They interpret the second exposure as strain relief and defect recovery without renewed graphitization.

Each photon carried about 5.0 eV, below the reported 5.97 to 8.3 eV B-O bond range. The authors propose cumulative photothermal decomposition and multiphoton optical breakdown during the pulse train.

Photolithography defined a 5 × 5 mm² growth window on n-type silicon (100). PDDAC and detonated nanodiamond seeds were applied within the patterned region. Patterning aimed to improve growth uniformity and reduce stress from the mismatch between diamond and silicon thermal expansion.

Hot-filament CVD used nine tungsten filaments near 2200°C and a 700°C substrate. Growth lasted 8 h at 2.77 kPa. The gas flow was 57 sccm methane, 2000 sccm hydrogen, and 3 sccm oxygen. The resulting lab-grown PCD film was 3.0 to 3.5 µm thick.

The commercial comparison was a 300 µm MSE PCD wafer polished to Rₐ below 1 nm. After irradiation, a 3:1 volumetric mixture of sulfuric acid and nitric acid cleaned the samples for 20 min at 70°C. Temperatures above 75°C caused partial delamination at the film and silicon interface.

Contacts used 20 nm Ti under 100 nm Au and annealed at 470°C for 60 s in argon. The authors state that this anneal activated the dopant.

Raman spectra used a 532 nm laser over 1000 to 1800 cm⁻¹. Measurements came from the sample center and four quadrants. Gaussian peak fitting was normalized to the diamond peak at 1332 cm⁻¹.

The lab-grown quality factor fell from 94.01% to 93.70% after the high-energy exposure. The 4000-pulse recovery raised it to 94.16 ± 0.11%, above the starting value.

Across five measured locations, the diamond Raman full width at half maximum narrowed after recovery. It changed from 9.59 ± 1.28 cm⁻¹ to 6.06 ± 0.81 cm⁻¹.

For the commercial PCD, the width rose from 5.75 cm⁻¹ to 5.92 cm⁻¹ during high-energy exposure. It fell to 5.63 cm⁻¹ after recovery.

Its trans-polyacetylene area increased from 1467.86 a.u. to 2169.82 a.u. after high-energy exposure. Recovery lowered it to 1235.18 a.u., 43.08% below the disordered state.

AFM indicated that lab-grown roughness rose from 1.92 nm to 10.53 nm after irradiation. The commercial sample finished near 10 nm despite its initial Rₐ below 1 nm. The authors associate the Raman narrowing with stress relaxation and reduced defect-related carbon.

XPS used Al Kα radiation at 1486.6 eV with argon sputter profiling. The lab C1s spectrum placed C-C at 285.05 eV, C-B at 283.90 eV, and C-O at 287.2 eV. The commercial C-B peak was 283.87 eV after acid cleaning.

The authors use the retained bonding signal as evidence against removable surface residue. ToF-SIMS used Bi3+ analysis at 30 keV and Cs+ sputtering at 2 keV.

The boron signal was strongest at the surface and decreased with depth, while carbon remained nearly constant. Boron reached about 15 to 20 nm in the lab film and 20 to 25 nm in the commercial PCD. The authors link the deeper commercial profile to uniform optical coupling and heating on its smoother surface.

Current-voltage measurements used linear regression, and Hall measurements quantified carriers. Active transport thickness was assigned from the ToF-SIMS boron profiles.

Diamond sampleAssigned active depthSheet resistivityHole concentrationHole mobility
Lab-grown HFCVD PCD, 3.0 to 3.5 µm15 nm2.36 × 10⁴ Ω sq⁻¹+4.46 × 10¹⁹ cm⁻³0.594 cm² V⁻¹ s⁻¹
Commercial MSE PCD, 300 µm25 nm603.7 Ω sq⁻¹+1.166 × 10¹⁹ cm⁻³7.09 cm² V⁻¹ s⁻¹

The lab-grown resistance fell from 5.91 × 10⁷ Ω over ±10 V to 5.14 × 10⁴ Ω over ±6 V. Using assigned active thicknesses, the lab film had an estimated bulk resistivity of 3.54 × 10⁻² Ω cm. The commercial PCD estimate was 1.51 × 10⁻³ Ω cm. These bulk values depend on the SIMS depth assignment.

The commercial sample had lower carrier concentration but much higher mobility than the rough lab film. The authors relate its performance to fewer grain boundaries and fewer carrier-scattering centers. The paper reports three-orders-of-magnitude resistance reduction for the lab sample and six for the commercial sample.

XRD found no B₄C diffraction peaks near 35° and 37°. The lab-grown diamond (111) peak shifted from 44.13° to 44.04°. The commercial sample shifted from 43.94° to 43.71°. The authors interpret these lower angles as lattice expansion after boron substitution.

Temperature-dependent transport gave an activation energy of about 0.192 eV. For comparison, isolated substitutional boron in lightly doped diamond has an activation energy of 0.37 eV. The authors identify 0.192 eV as the expected value for heavily doped diamond.

Together, XPS, Hall measurements, and thermally activated transport support substitutional p-type doping.

Energy densities above 0.34 J cm⁻² with 500 pulses caused chipping and localized ablation. Densities below 0.17 J cm⁻² with 4000 pulses gave weak or undetectable doping.

Beyond 4000 pulses at 0.17 J cm⁻², XPS detected substrate peaks. The authors associate these signals with film thinning or partial delamination. The measured boron depth confines the demonstrated conduction to a 15 to 25 nm surface region.

Material selection must balance carrier concentration against grain-boundary scattering. The polished commercial sample delivered much higher mobility, although its carrier concentration was lower.

The paper positions laser doping as a localized post-growth alternative to ion implantation and dopant introduction during growth. Adopting this process requires oxidative acid cleaning and the 470°C activation anneal. The paper reports no completed device, long-term aging, or thermal stability data.

For p-type device research, 6CCVD offers MPCVD-grown single-crystal and polycrystalline diamond, including boron-doped diamond, with Ti and Au among its metallization options. Those options cover the paper’s 20 nm Ti and 100 nm Au contacts, although the lab-grown film used HFCVD rather than MPCVD. The diamond materials catalog presents the material family, and engineering support can review the 248 nm laser process and 15 to 25 nm boron-depth requirements.

View Original Abstract

Abstract Diamond, an ultrawide bandgap semiconductor with exceptional thermal, optical, and electrical properties, has attracted significant interest for next-generation electronic applications. However, its integration into micro and optoelectronic devices remains limited by doping challenges. Herein, pulsed laser irradiation (PLI) is employed as an approach for controlled boron doping in diamond. Boron incorporation is achieved via immersion in a boric acid precursor, followed by PLI at 0.34 J cm −2 to induce p-type conductivity in a patterned 5 × 5 mm 2 lab-grown HFCVD polycrystalline diamond (PCD). To mitigate laser-induced defects and suppress graphitization, a subsequent low-energy irradiation at 0.17 J cm −2 is applied. After an initial 500-pulse high-energy treatment, the quality factor (Q.F.) decreases from 94.01% to 93.70%, but recovers to 94.16% following 4000 pulses of low-energy exposure, yielding a net improvement of ∼0.5%. The treated sample exhibits a carrier concentration of +4.46 × 10 19 cm −3 and a hole mobility of 0.594 cm 2 V −1 s −1 . When applied to a commercial MSE PCD sample ( R a < 1 nm), this treatment maintains a Q.F. of 100% and reduces the transpolyacetylene (TPA) peak intensity by 43.08%. Electrical measurements reveal a carrier concentration of 1.166 × 10 19 cm −3 and a hole mobility of 7.09 cm 2 V −1 s −1 . These results demonstrate a practical route for achieving reliable p-type doping and improved structural quality in diamond for advanced electronic applications.

  1. 2023 - A review of thermal properties of CVD diamond films [Crossref]
  2. 2016 - Thermal conductivity of silicon carbide composites with highly oriented graphene nanoplatelets [Crossref]
  3. 2017 - Near-infrared refractive index of synthetic single crystal and polycrystalline diamonds at high temperatures [Crossref]
  4. 2021 - Diamond for electronics: materials, processing and devices [Crossref]
  5. 2025 - A perspective of doping in diamond: from nanoelectronics to quantum applications [Crossref]
  6. 2010 - INSPEC (Information service), and institution of electrical engineers
  7. 1997 - Ion-implantation in diamond and diamond films: doping, damage effects and their applications [Crossref]
  8. 2021 - Graphitization of synthetic diamond crystals: a morphological study [Crossref]
  9. 2021 - Review of SI-based GESN CVD growth and optoelectronic applications [Crossref]
  10. 2009 - CVD‐diamond detectors for experiments with hadrons, nuclei, and atoms