🔬 materials science

Microscopic and macroscopic characterization: MBE-grown versus sputter-deposited Au/Co/Au thin films for CISS and MIPAC effect studies

This study demonstrates that while the microscopic chirality-induced spin selectivity (CISS) effect remains robust across both molecular beam epitaxy and sputter-deposited Au/Co/Au films, the macroscopic magnetization induced by proximity (MIPAC) effects are highly sensitive to substrate microstructure, manifesting significantly only in sputter-deposited samples.

Original authors: Lokesh Rasabathina, Thi Ngoc Ha Nguyen, Aleksandr Kazimir, Rico Ehrler, Julia Krone, Franziska Schölzel, Zihao Liu, Peter Heinig, Markus Gößler, Irene Coin, Christina Lamers, Georgeta Salvan, Lech Tom

Published 2026-06-30

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Lokesh Rasabathina, Thi Ngoc Ha Nguyen, Aleksandr Kazimir, Rico Ehrler, Julia Krone, Franziska Schölzel, Zihao Liu, Peter Heinig, Markus Gößler, Irene Coin, Christina Lamers, Georgeta Salvan, Lech Tomasz Baczewski, Christoph Tegenkamp, Olav Hellwig

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine you are trying to build a tiny, high-tech switch that can tell the difference between "left-handed" and "right-handed" molecules. This is the world of CISS (Chirality-Induced Spin Selectivity). Think of chiral molecules like spiral staircases or screws: some twist left, some twist right. The CISS effect is like a bouncer at a club who only lets people with a specific "spin" (a quantum property of electrons) enter if they are walking up a left-handed staircase, but blocks them on a right-handed one.

To test this, scientists need a very specific floor for these molecules to stand on: a thin sandwich of gold and cobalt. The big question this paper asks is: Does it matter how we build the floor?

The researchers built two versions of this "sandwich floor":

The "Master Chef" Version (MBE): Made using Molecular Beam Epitaxy. This is like growing a crystal in a lab with extreme precision, atom by atom. It's expensive and slow, but the result is a perfectly smooth, orderly surface.
The "Home Cook" Version (Sputter-Deposited): Made using magnetron sputtering. This is like spraying paint onto a wall. It's faster, cheaper, and easier to scale up for real-world devices, but the surface is a bit rougher and more chaotic.

Here is what they found when they tested both floors with special spiral-shaped peptides (tiny protein chains):

1. The "Under the Microscope" View (The Microscopic Scale)

When the scientists looked at the floors with a super-powerful microscope (STM), they saw some differences in the "big picture" structure.

The Master Chef floor had a neat, interconnected network of terraces, like a well-planned city grid.
The Home Cook floor looked more like a field of isolated, packed grains, like a sandy beach.

However, when they zoomed in really close to the actual surface where the molecules stand, both floors looked surprisingly similar. They were both smooth enough for the molecules to sit comfortably.

The Result: When they tested the "bouncer" effect (the CISS effect) on both floors, it worked exactly the same way. Whether the floor was the expensive, perfect one or the cheaper, rougher one, the molecules successfully filtered the electron spins with about 80% efficiency.

The Takeaway: The "bouncer" doesn't care if the floor is a perfect city grid or a sandy beach, as long as the immediate spot where the molecule stands is smooth. The microscopic magic works on both.

2. The "Big Picture" View (The Macroscopic Scale)

Next, they looked at the magnetic properties of the whole floor, not just the tiny spot where the molecule sits. This is called the MIPAC effect. Imagine the floor is a giant magnet that can be flipped. They wanted to see if putting the molecules on the floor made the magnet harder or easier to flip.

The Result: This is where the two floors acted differently.

The Master Chef floor (MBE): Putting the molecules on it changed nothing. The magnet flipped just as easily as before.
The Home Cook floor (Sputter): Putting the molecules on it made the magnet harder to flip. The "coercivity" (the force needed to flip the magnet) went up by about 11%, and the magnetic "domains" (tiny regions of magnetism) moved slower.

The Analogy: Think of the magnetic domains as cars driving on a highway.

On the Master Chef floor, the highway is a smooth, open road. The cars (magnetic domains) zoom along, and adding the molecules didn't put up any new roadblocks.
On the Home Cook floor, the highway is a bit bumpier with more "grain boundaries" (like potholes or construction zones). When the molecules landed, they acted like extra speed bumps or traffic cones, making it harder for the cars to move. The molecules got stuck in the rough spots of the cheaper floor, slowing the whole system down.

The Bottom Line

The paper concludes that:

The "Spin Filter" (CISS) is tough: It works just as well on cheap, rough, spray-deposited films as it does on expensive, perfect crystal films. If you want to build a spin-based sensor, you don't need the most expensive manufacturing method.
The "Magnetic Change" (MIPAC) is sensitive: If you are looking for changes in the overall magnetism of the material, the way the material is made matters a lot. The rougher, cheaper films react differently to the molecules than the perfect ones.

In short, the "magic" of the molecule works everywhere, but the "reaction" of the floor depends on how smooth or rough the floor was built.

Technical Summary: Microscopic and Macroscopic Characterization of MBE-grown versus Sputter-deposited Au/Co/Au Thin Films for CISS and MIPAC Studies

Problem Statement
The chirality-induced spin selectivity (CISS) effect enables spin-dependent transport at chiral molecule/metal interfaces and is a focal point for spintronic applications when combined with ferromagnetic thin films. Additionally, the adsorption of chiral molecules can induce macroscopic magnetic changes in the substrate, known as the magnetization induced by proximity of adsorbed chiral molecules (MIPAC) effect. While CISS has been extensively studied in molecular beam epitaxy (MBE)-grown systems, which offer atomically smooth surfaces and sharp interfaces, the influence of substrate microstructure on both CISS and MIPAC remains unclear. Most device-oriented studies utilize magnetron sputtering, a technique that allows for versatile layer structures but results in different interface roughness, grain structures, and surface morphologies compared to epitaxial growth. This study addresses the gap in understanding whether microscopic CISS signatures and macroscopic MIPAC responses are affected differently by these structural variations.

Methodology
The authors compared two types of Au/Co/Au ferromagnetic thin film stacks:

MBE-grown (Epitaxial): Fabricated on sapphire (Al2O3) substrates with a Pt/Au/Co/Au stack (1.2 nm Co).
Sputter-deposited (Polycrystalline): Fabricated on thermally oxidized Si substrates with a Ta/Pt/Au/Co/Au stack (0.9 nm Co).

Both samples were exposed to L-chiral $\alpha$ -helical alanine-rich peptides (L-polyalanine). The study employed a multi-scale characterization approach:

Structural Analysis: X-ray reflectivity (XRR) and X-ray diffraction (XRD) (out-of-plane, in-plane, rocking curves, and pole figures) were used to assess interface quality, crystallinity, and texture. Atomic force microscopy (AFM) and scanning tunneling microscopy (STM) provided real-space topography and local surface smoothness.
Chemical Verification: X-ray photoelectron spectroscopy (XPS) confirmed peptide adsorption and the formation of Au-S bonds.
Microscopic Magnetic Response: Scanning tunneling spectroscopy (STS) and magnetization-dependent tip-sample distance ( $\Delta z$ ) measurements were performed to detect local CISS effects (spin-polarized tunneling currents) after peptide adsorption.
Macroscopic Magnetic Response: Magneto-optical Kerr effect (MOKE) hysteresis loops and domain wall motion microscopy were used to measure changes in coercivity and domain wall velocity.

Key Results

Structural Differences: XRR and XRD confirmed that MBE-grown samples possess sharper interfaces, minimal intermixing, and highly uniform out-of-plane crystallite alignment (rocking curve FWHM of 0.14°). In contrast, sputter-deposited samples exhibited broader interfaces, partial intermixing, and a fiber texture with a broader distribution of crystallite orientations (rocking curve FWHM of 3.15°).
Surface Morphology: Despite significant differences in reciprocal space (crystallinity) and large-scale morphology (MBE showed interconnected terraces; sputtered showed isolated grains), both sample types displayed locally smooth Au(1 1 1) surface regions at the atomic scale (roughness $\approx \pm 0.4-0.5$ nm), suitable for peptide adsorption.
Microscopic CISS Response: STS and $\Delta z$ measurements revealed that both sample types exhibited a strong, magnetization-direction-dependent tunneling current after peptide exposure. The estimated spin polarization was 80% for MBE samples and 77% for sputter-deposited samples. This indicates that the microscopic CISS effect is robust and largely independent of the substrate's microstructure.
Macroscopic MIPAC Response: In contrast to the microscopic findings, macroscopic magnetic measurements showed a divergence. The sputter-deposited samples exhibited a slight but reproducible enhancement in coercivity (from 131 Oe to 145 Oe, an 11% increase) and a reduction in domain wall velocity (~15%) after peptide exposure. The MBE-grown samples showed no such coercivity enhancement.

Significance and Claims
The paper concludes that the observation of CISS and MIPAC effects is not uniform across different fabrication methods. The authors claim that:

Robustness of Microscopic CISS: The microscopic signature of the CISS effect (spin-selective transport) is robust against substrate microstructure variations, provided the local surface is smooth enough for molecular adsorption.
Sensitivity of Macroscopic MIPAC: Macroscopic magnetic responses (MIPAC-type effects), such as changes in coercivity and domain wall dynamics, are highly sensitive to the substrate's microstructure. The authors suggest that the isolated grain structure and grain boundaries in sputter-deposited films make them more susceptible to domain wall pinning induced by peptide adsorption, whereas the interconnected terrace network of MBE films is more immune to such pinning.
Caveats on Chirality: The authors modestly note that while a symmetric coercivity increase was observed in sputter-deposited samples, it remains to be fully clarified whether this is a true chirality-dependent MIPAC effect. They point out that a genuine chirality-sensitive interaction should ideally result in an asymmetric hysteresis loop (exchange-bias-like), whereas the observed symmetric increase suggests a non-chiral interaction or a pinning mechanism that requires further investigation with non-chiral reference molecules.

This study highlights the necessity of distinguishing between local electronic/spin transport phenomena and global magnetic property changes when designing chiral molecule-based spintronic devices, particularly when transitioning from fundamental epitaxial studies to application-oriented sputtered systems.

1. The "Under the Microscope" View (The Microscopic Scale)

2. The "Big Picture" View (The Macroscopic Scale)

The Bottom Line

Technical Summary: Microscopic and Macroscopic Characterization of MBE-grown versus Sputter-deposited Au/Co/Au Thin Films for CISS and MIPAC Studies

More like this