Growth-controlled photochromism in yttrium oxyhydride thin films deposited by HiPIMS and pulsed-DC magnetron sputtering

This study demonstrates that while both HiPIMS and pulsed-DC magnetron sputtering can produce photochromic yttrium oxyhydride films, the pulsed-DC method yields superior photochromic contrast and a lower band gap due to distinct microstructural orientations and a lower oxygen-to-hydrogen ratio, highlighting the critical influence of growth conditions on performance.

The Gist

Imagine you have a window that can change its own color. When the sun is shining, it turns dark to keep your house cool. When the sun goes down, it turns clear again so you can see outside. This is the dream of "smart windows," and the material scientists are trying to perfect for this job is called Yttrium Oxyhydride (YHO).

This paper is a report card comparing two different methods of building these special windows: one called HiPIMS and the other pulsed-DCMS. Think of these as two different chefs trying to bake the exact same cake, but using different ovens and techniques.

Here is the breakdown of what they found, using some everyday analogies:

1. The Two "Chefs" (The Deposition Methods)

To make the window film, the scientists spray atoms of Yttrium (a metal) onto a glass surface in a vacuum chamber, then let it react with air to turn into the smart material.

• The Pulsed-DC Chef (The Traditional Baker): This method is like a steady, gentle rain. It sprays atoms mostly as neutral particles. It's a bit like a calm drizzle.
• The HiPIMS Chef (The High-Voltage Storm): This method uses intense, short bursts of electricity. It's like a lightning storm. It doesn't just spray atoms; it zaps them, turning a huge number of them into charged ions (electrically active particles).
  • The Discovery: The HiPIMS storm was so powerful that it ionized almost all the Yttrium atoms, whereas the traditional method mostly left them neutral.

2. The "Goldilocks" Pressure (The Critical Pressure)

To get a clear, working window, the scientists had to find the "just right" amount of air pressure in the chamber.

• Too little pressure: The film becomes too dense and tight, like a brick wall. Oxygen can't get in to do its job, and the window stays dark and useless.
• Too much pressure: The film becomes too loose and full of holes. It gets very clear, but it loses its ability to change color.
• The Sweet Spot: They found that the HiPIMS storm needed a much higher pressure (more "air" in the room) to work correctly than the traditional drizzle. It's as if the HiPIMS storm was so energetic it needed more space to settle down properly, while the traditional method worked fine in a tighter space.

3. The Results: Who Made the Better Window?

Even though both methods made clear windows, they performed very differently when it came to changing color.

• The Traditional Chef (Pulsed-DCMS) Wins:

  • Performance: These windows turned dark much more dramatically (about 34% darker) when the sun hit them.
  • Structure: Under a microscope, these films looked like a neatly organized army of soldiers all facing the same direction (a specific crystal orientation). This order helped them work better.
  • Recipe: They had just the right amount of oxygen and hydrogen mixed in.

• The HiPIMS Chef (The Storm) Struggles:

  • Performance: These windows barely changed color (only about 9% darker).
  • Structure: These films were a bit more chaotic, like a crowd of people milling about in random directions.
  • Recipe: Because the HiPIMS process took longer and happened at higher pressures, the film accidentally absorbed a bit too much oxygen and not enough hydrogen. It was like the cake got a little too much flour and not enough sugar.

4. The Trade-Off: Speed vs. Strength

There is one interesting twist. The windows that changed color the most (the traditional ones) took a long time to turn back clear (bleach) once the sun went down. The ones that barely changed color (HiPIMS) turned clear much faster.

• Analogy: Think of it like a heavy door. The traditional window is a heavy, solid door that swings wide open (great contrast) but takes a long time to close. The HiPIMS window is a light screen door that barely moves but snaps shut instantly.

The Big Takeaway

The scientists learned that how you build the material is just as important as what you build.

• Even though HiPIMS is a "fancy" technology that usually makes stronger, denser materials for other uses, it wasn't the best fit for this specific smart window material in its current setup.
• The traditional method produced a more ordered, "just-right" chemical mix that allowed the window to darken significantly.

In short: The "gentle rain" method (pulsed-DCMS) baked a better smart window than the "lightning storm" method (HiPIMS) in this experiment. However, the scientists believe that if they tweak the HiPIMS storm (like improving the vacuum or speeding up the process), it might eventually become the superior method. For now, the traditional method is the winner for making windows that really know how to change color.

Technical Summary

1. Problem Statement

Oxygen-containing yttrium hydride (YHO) is a promising material for smart window applications due to its ability to reversibly darken under sunlight (positive photochromism) without external power. However, optimizing YHO performance requires a precise balance of chemical composition, microstructure, and defect density. While reactive magnetron sputtering is the standard synthesis method, the specific influence of high-energy deposition regimes on YHO properties remains under-explored.

The study addresses the gap in understanding how High Power Impulse Magnetron Sputtering (HiPIMS), which generates highly ionized fluxes, compares to conventional pulsed-DC Magnetron Sputtering (pulsed-DCMS) in terms of:

• Film density and microstructure.
• Oxygen-to-hydrogen stoichiometry.
• Photochromic contrast and kinetics.
• The critical working pressure required to achieve transparent, photochromic films.

2. Methodology

The researchers synthesized YHO thin films using a two-step process on soda-lime glass and silicon substrates:

1. Deposition: Reactive sputtering of a Yttrium (Y) target in an Ar/H₂ atmosphere to form $\beta$-YH₂ films.
   • Techniques Compared: HiPIMS (140 Hz, 50 µs pulses) vs. Pulsed-DCMS (80 kHz, 2.5 µs reverse time).
   • Variables: Sputtering pressure was systematically varied (0.45–0.80 Pa for DCMS; 1.00–1.20 Pa for HiPIMS) while maintaining a constant average power (200 W).
2. Oxidation: Films were exposed to atmospheric air to convert $\beta$-YH₂ into the photochromic YHO phase.
3. Characterization:
   • Plasma Diagnostics: Optical Emission Spectroscopy (OES) to analyze ionization states and discharge waveforms.
   • Composition: Time-of-flight Energy Elastic Recoil Detection Analysis (ToF-E ERDA) for atomic density and stoichiometry (Y:H:O ratios).
   • Structure: X-ray Diffraction (XRD) for crystallite size, lattice parameters, and preferred orientation; Scanning Electron Microscopy (SEM) for morphology.
   • Optical/Photochromic: Spectrophotometry for transmittance ($T_{sol}$, $T_{lum}$) and band gap ($E_g$); UV-A illumination for photochromic contrast ($\Delta T_r$) and bleaching kinetics.

3. Key Contributions

• Discharge Physics: Demonstrated that HiPIMS discharges are dominated by highly ionized Yttrium ($Y^+$) due to self-sputter recycling, whereas pulsed-DCMS is dominated by neutral species and $Ar^+$.
• Critical Pressure Shift: Identified that the critical working pressure ($P_c$) required to form transparent films is significantly higher for HiPIMS ($\approx$ 1.0 Pa) than for pulsed-DCMS ($\approx$ 0.5 Pa).
• Microstructure-Property Decoupling: Showed that despite similar optical transmittance, the two techniques produce films with distinct crystallographic textures and stoichiometries, leading to vastly different photochromic performances.
• Growth Mechanism Insight: Linked the lower photochromic performance of HiPIMS films to non-optimal growth conditions (lower deposition rates, higher residual gas exposure) leading to partial oxidation during deposition.

4. Key Results

A. Plasma and Deposition Dynamics

• HiPIMS: Characterized by strong $Y^+$ emission and high ionization fractions. The process involves significant self-sputter recycling, resulting in a lower deposition rate (requiring 3.3–4.5× longer deposition times for similar thickness).
• Pulsed-DCMS: Dominated by $Ar^+$ and neutral species, with lower ionization and higher deposition rates.

B. Structural and Compositional Differences

• Critical Pressure ($P_c$):
  • Pulsed-DCMS: $P_c \approx 0.5$ Pa. Films below this are opaque; above, they become transparent.
  • HiPIMS: $P_c \approx 1.0$ Pa. Films require higher pressure to achieve sufficient porosity for oxygen diffusion during post-oxidation.
• Texture:
  • Pulsed-DCMS: Exhibits a pronounced <100> out-of-plane preferred orientation (kinetically limited growth).
  • HiPIMS: Exhibits a random polycrystalline orientation (closer to thermodynamic equilibrium).
• Density & Stoichiometry (at $P_c$):
  • Pulsed-DCMS (pDC0.55): Higher density (~3.90 g/cm³), lower O/H ratio ($YH_{2.40}O_{0.83}$), and lower optical band gap (2.70 eV).
  • HiPIMS (HiP1.05): Lower density (~3.65 g/cm³), higher O/H ratio ($YH_{2.38}O_{1.15}$), and higher optical band gap (2.94 eV).

C. Photochromic Performance

• Contrast ($\Delta T_r$):
  • Pulsed-DCMS: Significantly superior, achieving 34% contrast (at 0.55 Pa).
  • HiPIMS: Poor performance, maxing at 9% contrast (at 1.05 Pa).
• Kinetics:
  • Films with higher contrast (DCMS) exhibit slower bleaching (recovery time $\approx$ 180 min).
  • Films with lower contrast (HiPIMS) bleach faster (recovery time $\approx$ 45 min).
• Correlation: The superior performance of DCMS films is attributed to the lower oxygen-to-hydrogen ratio and the specific <100> texture, which may facilitate more favorable oxidation pathways or defect structures for photochromism.

5. Significance and Conclusion

The study establishes that thin-film growth conditions and microstructure are as critical as chemical composition in governing the photochromic performance of YHO.

• Optimization Insight: While HiPIMS is known for producing dense, high-quality coatings in other applications, it was found suboptimal for YHO photochromism under the tested conditions. The high ionization and self-sputtering led to partial oxidation during the hydride deposition step and a random texture that hindered the photochromic effect.
• Practical Implication: For smart window applications requiring high contrast, pulsed-DCMS remains the superior technique, provided the pressure is tuned just above the critical threshold ($P_c \approx 0.5$ Pa) to balance transparency and photochromic activity.
• Future Directions: The authors suggest that HiPIMS could potentially be optimized for YHO by improving base vacuum conditions, increasing deposition rates, and decoupling ion energy from ion flux to prevent premature oxidation and tailor the microstructure.

In summary, this work provides a critical benchmark for selecting sputtering technologies for photochromic oxides, highlighting that "denser is not always better" and that specific microstructural textures (like the <100> orientation in DCMS) are vital for maximizing optical switching capabilities.