Pulsed Laser Epitaxy

• Pulsed laser epitaxy is a technique that combines pulsed laser deposition with epitaxial control to achieve atomically sharp interfaces and high crystalline quality.
• The method precisely tunes parameters like laser fluence, substrate temperature, and oxygen partial pressure to control film stoichiometry and phase selection.
• Applications include quantum materials, complex oxides, and chalcogenides, offering tailored electronic, magnetic, and optical functionalities for device integration.

Pulsed laser deposition (PLD) is widely used to grow epitaxial thin films of quantum materials such as complex oxides, and the closely related term pulsed laser epitaxy (PLE) is used for PLD growth in which the deposited layer is epitaxial. In the systems reported across oxides, chalcogenides, fluorides, antiperovskites, semiconductors, and magnetic garnets, a high-power laser pulse ablates a target, generates a plasma plume, and deposits material onto a heated substrate; the resulting film structure and properties are controlled by substrate temperature, background gas or oxygen partial pressure, laser fluence or pulse energy, deposition rate, plume geometry, and the chemical and structural characteristics of the interface (Ulbrandt et al., 2019, Harris et al., 2020, Ok et al., 2020).

1. Process definition and control space

In practical use, pulsed laser epitaxial growth combines stoichiometric or near-stoichiometric transfer from a solid target with a strongly nonequilibrium deposition environment. Reported implementations include KrF excimer lasers at $248\,\mathrm{nm}$ or $246\,\mathrm{nm}$, ArF excimer lasers at $193\,\mathrm{nm}$, and Nd:YAG excitation at $532\,\mathrm{nm}$, $355\,\mathrm{nm}$, or $1064\,\mathrm{nm}$; substrates range from SrTiO$_3$, Al$_2$O$_3$, MgO, TiO$_2$, LiF, YSZ, KTP, Si with oxide buffers, and GGG, while ambient conditions span high vacuum to oxygen partial pressures from $246\,\mathrm{nm}$0 to $246\,\mathrm{nm}$1 depending on the materials system (Nichols et al., 2014, Hiramatsu et al., 2014, Keßler et al., 2024).

The same general method supports very different growth strategies. Some films are deposited directly at elevated temperature, such as Ba$246\,\mathrm{nm}$2IrO$246\,\mathrm{nm}$3 on SrTiO$246\,\mathrm{nm}$4 at $246\,\mathrm{nm}$5C in $246\,\mathrm{nm}$6 oxygen, KTiOAsO$246\,\mathrm{nm}$7 on KTP at $246\,\mathrm{nm}$8C in $246\,\mathrm{nm}$9 oxygen, and ScF$193\,\mathrm{nm}$0 on LiF at $193\,\mathrm{nm}$1C under $193\,\mathrm{nm}$2 vacuum (Nichols et al., 2014, Clavel et al., 28 Mar 2025, Jayakody et al., 2024). Others are deposited at room temperature and crystallized by in-situ annealing, as in Mn$193\,\mathrm{nm}$3Sn on Al$193\,\mathrm{nm}$4O$193\,\mathrm{nm}$5 or MgO (Gao et al., 2022). Large-area implementations add further constraints: for LaSrMnO$193\,\mathrm{nm}$6 on 4" Si wafers, the plume is smaller than the wafer, homogeneous growth is achieved by scanning the plume across the wafer and rotating the wafer under a large heater, and the full plume is collected on the wafer, resulting in a compositional varying flux of species on the wafer (Groenen et al., 2016).

This diversity suggests that pulsed laser epitaxial growth is best understood as a controllable process space rather than a single recipe. The common objective is atomically sharp interfaces, high crystalline quality, and a growth window in which particle flux, kinetic energy, oxidation state, and surface mobility are jointly compatible with single crystal epitaxy.

2. Plume energetics, deposition rate, and interlayer kinetics

A central feature of pulsed laser epitaxy is the very high instantaneous deposition flux produced by each laser pulse. During homoepitaxy of SrTiO$193\,\mathrm{nm}$7, in-situ x-ray scattering showed that energetic PLD suppresses the lateral growth of two-dimensional islands, which suggests that energetic particles break up smaller islands, while fast interlayer transport occurs for both energetic and thermalized PLD and does not depend strongly on the kinetic energy of the incident particles (Ulbrandt et al., 2019). In later combined in-situ x-ray reflectivity and kinetic Monte Carlo work on SrTiO$193\,\mathrm{nm}$8, specular reflectivity monitored interlayer transport and diffuse scattering revealed the evolution of in-plane length scales; transient islands on top of stable two-dimensional islands formed immediately after the deposition pulse and then ripened via detachment and diffusion, and the detachment energy barrier played a dominant role in determining the recovery time constant (Ulbrandt et al., 23 Sep 2025).

The kinetic description used across these studies is explicitly thermally activated. Surface diffusion was written as

$193\,\mathrm{nm}$9

and hopping or detachment rates were written as

$532\,\mathrm{nm}$0

These expressions formalize a recurrent experimental result: deposition pulses create nonequilibrium surface populations, while the dwell time between pulses determines how far the system relaxes toward an epitaxial configuration (Groenen et al., 2016, Ulbrandt et al., 23 Sep 2025).

Deposition rate is itself a primary growth variable. For cobalt-doped BaFe$532\,\mathrm{nm}$1As$532\,\mathrm{nm}$2, the optimal deposition rate, which could be tuned by pulse energy, was independent of laser wavelength, and high-quality epitaxial growth was achieved near a deposition rate of approximately $532\,\mathrm{nm}$3Å/s regardless of whether the ablation source was ArF, KrF, or Nd:YAG (Hiramatsu et al., 2014). For Sr$532\,\mathrm{nm}$4IrO$532\,\mathrm{nm}$5, reduced PLD plume dimensions and slow deposition rates were identified as the key for stabilizing pure Sr$532\,\mathrm{nm}$6IrO$532\,\mathrm{nm}$7 phase thin films (Seo et al., 2016). These results argue against a purely wavelength-based view of PLD optimization and place greater weight on flux, pulse-by-pulse supersaturation, and the time available for reconstruction.

Direct plasma diagnostics further quantify the plume. During PLD of FeSe, Langmuir probe measurements gave electron temperatures from $532\,\mathrm{nm}$8 to $532\,\mathrm{nm}$9, plasma densities from $355\,\mathrm{nm}$0 to $355\,\mathrm{nm}$1, and two distinct plasma regimes separated by a threshold of about $355\,\mathrm{nm}$2 delivered to the target. The low-energy/low-density regime lacked the high-energy ion tail and produced a gentler arrival of species at the substrate; this regime was described as highly favorable for the epitaxial growth of monolayers and larger crystal grains, with reduced defect formation (Harris et al., 2020).

3. Strain, substrate chemistry, and buffer-layer control

Substrate selection in pulsed laser epitaxy is not limited to lattice mismatch minimization. It also includes symmetry, surface chemistry, thermal expansion compatibility, and, in many systems, the use of an intervening buffer layer that supplies the correct atomic template. A clear example is PdCrO$355\,\mathrm{nm}$3, for which direct growth of epitaxial films without impurity phases was not successful, despite examination of several substrates. The difficulty was attributed to both the chemical and structural dissimilarities between the substrates and the volatile nature of the PdO layer. This difficulty was overcome by growing CuCrO$355\,\mathrm{nm}$4 buffer layers before PdCrO$355\,\mathrm{nm}$5, and only monolayer thick buffer layer was sufficient to grow the correct PdCrO$355\,\mathrm{nm}$6 phase; the result indicated that the epitaxy of Pd-based delafossites is extremely sensitive to the chemistry and structure of the interface, necessitating near perfect substrate materials (Ok et al., 2020).

For complex oxides on silicon, buffer-layer engineering can shift the accessible temperature window. In LaSrMnO$355\,\mathrm{nm}$7 growth on YSZ/CeO$355\,\mathrm{nm}$8/Si, the introduction of a thin SrRuO$355\,\mathrm{nm}$9 layer significantly enhanced the epitaxial quality and crystallinity of LaSrMnO$1064\,\mathrm{nm}$0 grown at reduced temperatures, and ferromagnetic behavior was observed for growth temperatures as low as $1064\,\mathrm{nm}$1C. The suggested mechanism was improved surface diffusion, ensuring sufficient intermixing of surface species for formation of the correct phase under large-area, full-plume collection conditions (Groenen et al., 2016).

Epitaxial strain can also stabilize phases that in bulk require extreme thermodynamic conditions. Ba$1064\,\mathrm{nm}$2IrO$1064\,\mathrm{nm}$3 bulk synthesis requires high pressure $1064\,\mathrm{nm}$4 GPa, but epitaxial Ba$1064\,\mathrm{nm}$5IrO$1064\,\mathrm{nm}$6 films were stabilized on SrTiO$1064\,\mathrm{nm}$7 by in-plane compressive strain. Using

$1064\,\mathrm{nm}$8

with $1064\,\mathrm{nm}$9GPa and $_3$0, the in-plane pressure was estimated as $_3$1GPa, satisfying the bulk synthesis requirement (Nichols et al., 2014). In KTiOAsO$_3$2/KTiOPO$_3$3, the chosen architecture exploited a low lattice parameter mismatch and a refractive-index contrast, yielding well-oriented epitaxy up to about $_3$4 after deposition and improved crystal state after oxygen annealing (Clavel et al., 28 Mar 2025).

A recurrent misconception is that lattice mismatch alone determines success. ScF$_3$5 did not form detectable crystalline films on SrTiO$_3$6 or MgO under a wide range of growth conditions, whereas LiF enabled successful fabrication of epitaxial ScF$_3$7 films because chemical compatibility and near-perfect lattice match were key factors (Jayakody et al., 2024). In PdCrO$_3$8, MgO also failed to nucleate the correct phase despite a relatively small mismatch, again underscoring that interface chemistry and structural registry can dominate over scalar mismatch values (Ok et al., 2020).

4. Stoichiometry management and phase selection

Because PLD is a nonequilibrium transfer process, stoichiometry management is frequently decisive for phase-pure epitaxy. The Ruddlesden-Popper iridates provide a canonical case. Conventional PLD methods often result in mixed phases of Sr$_3$9IrO$_2$0, Sr$_2$1Ir$_2$2O$_2$3, and SrIrO$_2$4, but reduced PLD plume dimensions and slow deposition rates were identified as the key for stabilizing pure Sr$_2$5IrO$_2$6 phase thin films, with real-time in-situ monitoring of optical spectra used for phase identification (Seo et al., 2016). A related study selectively grew high quality single-phase Sr$_2$7IrO$_2$8, Sr$_2$9Ir$_3$0O$_3$1, and SrIrO$_3$2 from a single Sr$_3$3Ir$_3$4O$_3$5 target by tuning background oxygen pressure and epitaxial strain, and showed that identical growth conditions on different substrates could stabilize different phases. The paper explicitly stated that this could not be explained by a simple kinetic model (Gutiérrez-Llorente et al., 2017).

Stoichiometry compensation by target engineering is especially important for volatile constituents. In AgCrSe$_3$6, an Ag-rich PLD target was used to compensate for Ag deficiency in thin films. This suppressed the nucleation of impurity phases and resulted in c-axis-oriented and single-phase AgCrSe$_3$7 on lattice-matched YSZ(111) (Tajima et al., 28 May 2025). In hybrid PLD of sulfides, tert-butyl disulfide was introduced as an external sulfur source to regulate the stoichiometry of the deposited films. By decoupling cation and anion delivery, the method precisely tuned film composition and yielded epitaxial or textured sulfide films with improved crystallinity and improved surface and interface roughness compared to the state-of-the-art (Surendran et al., 2023).

Oxide growth imposes a different stoichiometric balance: under-oxidation, over-oxidation, and target evolution can all become limiting. For RuO$_3$8 and IrO$_3$9 on TiO$_2$0(110), the main challenges included a deteriorating material flux due to laser induced target metallization and the delicate balance of under- vs over-oxidation of the stubborn Ru and Ir metals. Optimal conditions were identified as $_2$1, $_2$2 for RuO$_2$3 and $_2$4, $_2$5 for IrO$_2$6, explicitly balancing oxidation against particle mobility during nucleation (Keßler et al., 2024).

These examples suggest that phase selectivity in pulsed laser epitaxial growth is usually an entangled problem of plume composition, surface kinetics, volatile-species compensation, and substrate-controlled thermodynamics, rather than a simple matter of nominal target stoichiometry.

5. Structural signatures, defects, and growth modes

The structural toolkit of pulsed laser epitaxy is correspondingly broad: $_2$7-2$_2$8 scans, rocking curves, reciprocal-space mapping, $_2$9-scans, pole figures, RHEED, AFM, SEM, XRR, HAADF-STEM, EDX, and in some cases in-situ optical spectroscopy or Langmuir probes. High crystalline quality is routinely demonstrated by narrow rocking curves. Representative values include $246\,\mathrm{nm}$00 for the BIO 006 reflection, comparable to the SrTiO$246\,\mathrm{nm}$01 substrate at $246\,\mathrm{nm}$02, $246\,\mathrm{nm}$03 for Mn$246\,\mathrm{nm}$04Sn(0001), $246\,\mathrm{nm}$05 for the KTA (800) reflection, $246\,\mathrm{nm}$06 for Bi$246\,\mathrm{nm}$07O$246\,\mathrm{nm}$08Se on SrTiO$246\,\mathrm{nm}$09, and $246\,\mathrm{nm}$10 for the ScF$246\,\mathrm{nm}$11(001) peak on LiF (Nichols et al., 2014, Gao et al., 2022, Clavel et al., 28 Mar 2025, Song et al., 2019, Jayakody et al., 2024).

Growth modes vary strongly with temperature, rate, and thickness. In Bi$246\,\mathrm{nm}$12O$246\,\mathrm{nm}$13Se on SrTiO$246\,\mathrm{nm}$14, a growth mode transition to three dimensional island from quasi-2D layer was observed as deposition time was prolonged, and Laue oscillations disappeared for thicker films as roughness and relaxation set in (Song et al., 2019). In RuO$246\,\mathrm{nm}$15(110)/TiO$246\,\mathrm{nm}$16(110), layer-by-layer growth persisted up to $246\,\mathrm{nm}$17 unit cells if growth proceeded at high deposition rates, whereas low deposition rates fostered initial 3D island growth and cluster formation; these islands eventually merged and growth continued in a step flow mode (Keßler et al., 2024). In ReS$246\,\mathrm{nm}$18, films grew with $246\,\mathrm{nm}$19 ReS$246\,\mathrm{nm}$20 perpendicular to sapphire or MoS$246\,\mathrm{nm}$21 template below $246\,\mathrm{nm}$22C, but became polycrystalline above $246\,\mathrm{nm}$23C, with the MoS$246\,\mathrm{nm}$24 template producing smoother and structurally superior films (Vishal et al., 2018).

Defect populations are equally system-specific. Very thin NiO(111) films on c-sapphire showed co-existence of $246\,\mathrm{nm}$25-rotated triangular domains, very low density of $246\,\mathrm{nm}$26 dislocations, screw dislocation density of $246\,\mathrm{nm}$27, and lowest edge dislocation density of $246\,\mathrm{nm}$28; continuous epitaxial films as thin as $246\,\mathrm{nm}$29 were obtained (Yadav et al., 2021). AgCrSe$246\,\mathrm{nm}$30 on YSZ(111) exhibited both twisted and polar domains (Tajima et al., 28 May 2025). ScF$246\,\mathrm{nm}$31 on LiF provided an unusual case in which principal-axis peaks were sharp but the $246\,\mathrm{nm}$32 reflection was extremely broad; the broadening was attributed not to classic epitaxial disorder but to disorder in the F$246\,\mathrm{nm}$33 octahedral rotations caused by the thermal-expansion mismatch of a negative-thermal-expansion film on a positive-thermal-expansion substrate (Jayakody et al., 2024).

6. Functional properties and technological directions

The relevance of pulsed laser epitaxial growth is ultimately established by the properties of the resulting films. In Ba$246\,\mathrm{nm}$34IrO$246\,\mathrm{nm}$35, the films were insulating but close to the metal-insulator transition boundary, with significantly smaller transport and optical gap energies than Sr$246\,\mathrm{nm}$36IrO$246\,\mathrm{nm}$37, together with enhanced electronic bandwidth and electronic-correlation energy (Nichols et al., 2014). In Mn$246\,\mathrm{nm}$38Sn, highly oriented epitaxial films showed a large anomalous Hall effect up to $246\,\mathrm{nm}$39 and longitudinal magneto-optical Kerr effect with $246\,\mathrm{nm}$40 at $246\,\mathrm{nm}$41 and $246\,\mathrm{nm}$42, comparable to Mn$246\,\mathrm{nm}$43Sn single crystals (Gao et al., 2022). In metallic PdCrO$246\,\mathrm{nm}$44, the antiferromagnetic transition at $246\,\mathrm{nm}$45 persisted down to $246\,\mathrm{nm}$46 thickness (Ok et al., 2020).

In semiconductor and chalcogenide systems, PLD epitaxy supports optoelectronic functionality while exposing interfacial limitations. Epitaxial Bi$246\,\mathrm{nm}$47O$246\,\mathrm{nm}$48Se on SrTiO$246\,\mathrm{nm}$49 reached a maximum room-temperature electron mobility of $246\,\mathrm{nm}$50 in a $246\,\mathrm{nm}$51-thick film, and the thickness-dependent mobility provided evidence that interface-scattering is likely to be the limiting factor for the relatively low electron mobility at low temperature (Song et al., 2019). AgCrSe$246\,\mathrm{nm}$52 films showed an absorption edge at around $246\,\mathrm{nm}$53 and a magnetic transition temperature at $246\,\mathrm{nm}$54, consistent with the reported values of direct bandgap and Néel temperature of bulk AgCrSe$246\,\mathrm{nm}$55 (Tajima et al., 28 May 2025). Hybrid PLD extended the accessible epitaxial space for vapor-pressure-mismatched sulfides, including BaS, SrS, CaS, ZnS, TiS$246\,\mathrm{nm}$56, BaZrS$246\,\mathrm{nm}$57, and BaTiS$246\,\mathrm{nm}$58, while maintaining smooth surfaces and interfaces (Surendran et al., 2023).

For integrated photonics and quantum materials, the method already reaches device-relevant heterostructures. Epitaxial KTiOAsO$246\,\mathrm{nm}$59 on KTiOPO$246\,\mathrm{nm}$60 formed a planar-waveguide geometry because KTA has a higher refractive index than KTP, and the Type-II phase-matching condition was written as

$246\,\mathrm{nm}$61

The deposited layer was about $246\,\mathrm{nm}$62, whereas about $246\,\mathrm{nm}$63 was targeted for second-harmonic generation at telecom wavelengths (Clavel et al., 28 Mar 2025). In Er-substituted yttrium iron garnet on GGG(111), single-phase, fully coherent growth with atomically sharp interfaces was maintained across $246\,\mathrm{nm}$64, and films with low Er content ($246\,\mathrm{nm}$65) retained nearly isotropic magnetization and exhibited a damping parameter only slightly higher than that of undoped YIG (Flajsman et al., 9 Nov 2025).

Taken together, these reports show that pulsed laser epitaxial growth is not defined by a single materials class or a single mechanism. Its distinctive capability is the coupling of pulsed nonequilibrium flux with epitaxial templating, so that deposition rate, plume energetics, interlayer transport, interface chemistry, strain, and stoichiometric compensation can be tuned to stabilize phases, orientations, and functionalities that are difficult to access by bulk synthesis or by continuous-flux thin-film methods.