Significant endeavors have been dedicated to diminishing dislocation densities and releasing stress in heteroepitaxial diamonds. In 2017, Schreck et al. fabricated an around 92 mm diameter, 1.6 mm thick freestanding heteroepitaxial diamond plate on Ir/YSZ/Si (001) (with YSZ = yttria stabilized zirconia), the largest single-crystal diamond ever reported^[14]. This achievement was facilitated by applying high-power conditions (915 MHz) in their microwave-plasma chemical vapor deposition (MPCVD) setup to ensure a more homogeneous growth environment. Besides, Kim et al. grew 1-inch high-quality self-supported diamonds with a thickness ranging from 500−600 μm after double-sided polishing on Ir (001)/sapphire (112¯0) using the microneedle method^[11]. Subsequently, they accomplished 2-inch high-quality freestanding monocrystalline diamonds on A-sapphire substrates with a misoriented angle of up to 7° in a step-flow growth mode^[12]. This letter presents a practical and dependable approach for growing sizable diamond single crystals. We introduce the utilization of a laser beam to create patterns on the surface of a 50 nm diamond layer grown on BEN-treated Ir/YSZ/Si (001) substrates. These laser-patterned templates effectively ease the stress within the diamond layer, thereby facilitating the growth of a robust 2-inch freestanding diamond. The resultant diamond remains intact, devoid of cracks, promising crystal quality and low dislocation density.

Fig. 1 illustrates schematically the fabrication process of the freestanding heteroepitaxial diamond grown on Ir/YSZ/Si (001) via laser-patterned templates. Firstly, a 100 nm thick YSZ epilayer and a 120 nm thick Ir epilayer were sequentially deposited on a 2-inch Si (001) single-crystal substrate using a PLD setup. Details of preparing the YSZ and Ir epilayers are described in our previous work^{[20, 21]}. Then, BEN treatment for diamond nucleation was performed on the prepared Ir/YSZ/Si (001) substrate in a specialized apparatus (DN2106) based on microwave plasma. The bias voltage was −300 V, and the duration was 40 min. Onto the BEN-treated Ir composite substrate, a thin diamond film of around 50 nm was grown. A 355-nm ultraviolet laser with a power of 500 mW was employed to create 500 μm × 500 μm square patterns with lateral faces along <110> direction. The width treated by the laser is 20 μm. The growth of a thick diamond layer was then carried out using the MPCVD reactor under optimized growth conditions. The diamond patterns underwent an epitaxial lateral overgrowth (ELO) process to achieve a coalescence, forming a closed and continuous diamond epilayer at the first 200 μm thickness. At the latter stage of the CVD growth, nitrogen was slightly added to the ambient gas to increase the growth rate. During the cooling process, the diamond/Ir layers automatically detached from the Si substrate due to the coefficient of thermal expansion (CTE) induced thermal strain. The growth of the thick diamond lasted for 80 h, resulting in a final thickness of around 400 μm.

In short, 2-inch free-standing diamonds were prepared by using heteroepitaxy on composite Ir/YSZ/Si (001) substrates. To release stress, patterned templates were fabricated using laser etching after the initial growth of 50-nm-diamond. Then, the subsequent growth was completed on a patterned template. The FWHM of the diamond (400) and (311) X-ray rocking curves were 313 and 359 arcsecs, respectively. The dislocation density determined by counting etching pits was approximately 2.2 × 10⁷ cm⁻². These results evidence that the laser-patterned method can effectively release stress during the growth of large-size diamonds, offering a simpler and more cost-effective alternative to the traditional photolithography-patterned scheme.

Fig. 3(a) shows a picture of the 2-inch freestanding diamond crystal with a thickness of 400 μm. The dim color is a result of nitrogen impurities. Owing to the thermal strain and the partial attachment of the diamond to the Ir interface, the diamond/Ir layers detached from the Si substrate during the cooling process. This diamond plate was obtained without cracks and fracturing. To quantitatively evaluate dislocation density and its distribution, we subjected the diamond surface to H₂/O₂ plasma etching within the MPCVD setup. Under the etching conditions involving a microwave power of 3 kW, a gas pressure of 150 Torr, and a substrate temperature of 850 °C, the process was executed for 20 min. The density of these dislocation features was determined to be approximately 2.2 × 10⁷ cm⁻², a value consistent with the lowest dislocation levels observed in heteroepitaxial diamonds.

As an ultra-wide bandgap semiconductor, diamond garners significant interest due to its exceptional physical properties^[1–3]. These superior characteristics make diamonds highly promising for applications in power electronics^[4], deep-ultraviolet detectors^[5], high-energy particle detectors^[6], and quantum devices based on color centers^[7]. A high-quality inch-sized monocrystalline diamond wafer is fundamental for massive semiconductor device research. Nevertheless, preparing such diamond wafers constitutes one of the major technological challenges in commercializing diamond materials and devices. Over several decades of development, the heteroepitaxial growth of diamonds on Ir composite substrates, combined with bias-enhanced nucleation (BEN) process, has emerged as the most promising method for producing inch-sized monocrystalline diamonds^[8–15]. However, achieving large-sized heteroepitaxial diamonds in practice remains challenging. The disparity in lattice parameters between hetero-substrates and diamonds leads to high threading dislocation densities, typically in the range of 10⁷−10⁹ cm⁻² for a few hundred micrometers thickness^[16]. Additionally, the complex strain control within the diamond–iridium composite system and the thermal mismatch due to differences in thermal expansion coefficients induce stress up to the GPa level in diamond thin films^{[17, 18]}. This stress often causes cracking of the diamond epitaxial layers, hindering the growth of thick diamond wafers^[19].

Fig. 2 shows the X-ray diffraction (XRD) results of the heteroepitaxial diamond grown on Ir/YSZ/Si (001) for 80 h. The resultant diamond is around 400 μm thick. The XRD θ−2θ scan shown in Fig. 2(a) exhibits four prominent diffraction peaks corresponding to Ir (200) (2θ = 47.3°), Si (400) (2θ = 69.2°), Ir (400) (2θ = 106.7°), and diamond (400) (2θ = 119.3°), verifying a pure (001)-oriented alignment and an excellent crystallinity. Mutual orientations among the diamond, Ir, YSZ, and Si substrate were verified by an in-plane φ scan of the {111} diffraction peaks of each layer at χ = 54.74°. As shown in Fig. 2(b), the φ scan results of diamond and iridium layers show an exact quadruple symmetry with φ angles occurring at φ = 0°, 90°, 180°, and 270°. No diffraction peaks of the YSZ layer and silicon substrate (not shown here) were detected, probably because they exceeded the penetration depth of X-rays. It was proved that the Ir/YSZ/Si substrate has a cube-on-cube crystallographic epitaxial relationship in our previous work. Therefore, this result demonstrates the same crystallographic epitaxial relationship in the diamond/Ir/YSZ/Si system. Furthermore, the X-ray pole figure of diamond {111} diffractions from the as-grown diamond crystal is shown in Fig. 2(c). The absence of additional peaks except for four {111} diffraction peaks evidences that no twinning is found in the diamond crystal. The X-ray rocking curves (XRCs) of diamond (400), (311), and (220) diffractions are displayed in Figs. 2(d)−2(f), respectively. The full width at half maximum (FWHM) of respective XRCs are 313.5, 359.1, and 1228.6 arcsecs, respectively, confirming the good epitaxial quality of the resultant diamond.