🔬 materials science

The Surface Sensitivity of X-ray Second Harmonic Generation as a Function of Energy

This study utilizes analytical and computational methods to demonstrate that while X-ray second harmonic generation in diamond is highly surface-sensitive near the carbon K-edge, its sensitivity shifts to being overwhelmingly dominated by the bulk quadrupole response as photon energy increases, with specific crystal orientations significantly influencing this behavior.

Original authors: Daniel Schacher, Tod A. Pascal, Keith V. Lawler, Craig P. Schwartz

Published 2026-02-27

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Daniel Schacher, Tod A. Pascal, Keith V. Lawler, Craig P. Schwartz

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 take a photograph of a single layer of paint on a wall, but you are using a flashlight that is so powerful and strange it can see through the wall itself. That is essentially what this paper is about, but instead of paint and a flashlight, the scientists are studying diamond using X-rays to see its surface.

Here is the story of their discovery, broken down into simple concepts:

The Goal: Seeing the Surface vs. The Bulk

In the world of light, there is a special trick called Second Harmonic Generation (SHG). Think of it like a musical instrument: if you play a low note (the input light), this trick turns it into a high note (the output light) that is exactly double the pitch.

Usually, this trick only works on the surface of a material if the material is perfectly symmetrical inside (like a diamond). It's like trying to hear a whisper in a quiet room; if the room is noisy (symmetrical inside), you can't hear the whisper (the surface signal). But if the room is quiet (at the surface), the whisper is loud and clear.

The scientists wanted to know: How deep does our "X-ray flashlight" see? Does it only see the surface, or does it get so powerful that it sees the whole diamond block, drowning out the surface signal?

The Analogy: The "Whisper" vs. The "Roar"

Imagine the surface of the diamond is a person whispering a secret.

The Surface Signal (The Whisper): This is the information we want. It tells us about the very top layer of atoms.
The Bulk Signal (The Roar): This is the noise coming from deep inside the diamond. In a perfect diamond, this noise shouldn't exist, but at high energies, the X-rays make the electrons inside the diamond "scream" (a plasma-like response), creating a loud roar that drowns out the whisper.

The paper asks: At what energy level does the "Roar" become so loud that we can no longer hear the "Whisper"?

The Experiment: Tuning the Radio

The researchers used a supercomputer to simulate shining X-rays on diamond at different energy levels (from "soft" X-rays, which are lower energy, to "hard" X-rays, which are very high energy).

They looked at two different "angles" of the diamond surface, like looking at a brick wall from the side versus looking at it from the top:

The (001) Surface: A flat, orderly surface.
The (111) Surface: A more complex, jagged surface.

The Findings: The "Sweet Spot"

Here is what they discovered, using a simple energy scale:

Near the "Carbon K-Edge" (approx. 285 eV): This is the "Goldilocks" zone. The X-ray energy is just right to make the surface atoms very excited. Here, the Whisper is loud. The signal is almost entirely from the surface. It's like being in a library where everyone is silent except for the person right next to you.
- Analogy: You are standing right next to a speaker; you hear them perfectly, and the background noise is quiet.
Around 1,000 eV: The "Roar" starts to get louder. The signal from the deep inside of the diamond begins to mix with the surface signal. The surface is still visible, but it's getting harder to tell where the surface ends and the inside begins.
- Analogy: The background music in the library has turned up. You can still hear the whisper, but you have to strain a bit.
Above 3,000 eV (and definitely by 7,000 eV): The Roar wins. The X-rays are so energetic that they make the entire diamond block vibrate. The signal is now dominated by the "bulk" (the inside). The surface information is completely drowned out.
- Analogy: The library is now a rock concert. You cannot hear the whisper at all; you only hear the band playing inside the building.

The Twist: Orientation Matters

The scientists also found that the "shape" of the surface matters.

On the flat (001) surface, the transition from "hearing the whisper" to "hearing the roar" happens smoothly as you increase the energy.
On the complex (111) surface, the behavior is a bit wobbly and unpredictable, likely because the atoms are arranged in a more complicated pattern. It's like trying to hear a whisper in a room with weird echoes; sometimes the sound bounces back in confusing ways.

The Big Takeaway

This paper is a guidebook for scientists who want to use X-rays to study surfaces.

If you want to study the surface: You must use soft X-rays and tune them to be very close to the specific energy where carbon atoms love to absorb light (the resonance). If you go too high in energy, you lose the surface information.
If you want to study the inside: You can crank up the energy to the "hard X-ray" range, and you will get a clear picture of the bulk material.

In short: X-ray Second Harmonic Generation is a powerful tool for seeing surfaces, but only if you don't turn the volume up too high. Once the energy gets too high, the tool stops looking at the surface and starts looking at the whole block of diamond.

1. Problem Statement

Second Harmonic Generation (SHG) is a powerful nonlinear optical technique traditionally used to probe surfaces and interfaces in centrosymmetric media (like diamond or silicon). In the optical and infrared regimes, SHG is highly surface-sensitive because the bulk electric-dipole response vanishes due to inversion symmetry, leaving only the surface-induced dipole signal.

However, as photon energies increase into the X-ray regime (specifically soft and hard X-rays), this surface specificity becomes compromised.

The Challenge: At X-ray energies, higher-order multipole terms (specifically electric-quadrupole and magnetic-dipole contributions) become significant. Unlike the dipole term, these higher-order terms are allowed in the bulk of centrosymmetric crystals.
The Question: It is unclear at what specific photon energy the transition occurs where the SHG signal shifts from being dominated by surface-induced dipoles to being dominated by bulk multipole responses. Understanding this crossover is critical for determining the utility of X-ray SHG as a surface probe versus a bulk structural probe.

2. Methodology

The authors employed a dual approach combining analytical derivations and advanced computational simulations to map the surface-to-bulk transition in diamond.

A. Analytical Framework

Perturbation Theory: The study utilized the formalism of Freund and Levine, decomposing the electron-field interaction into $p \cdot A$ $p \cdot A$ (momentum-vector potential) and $A^2$ $A^{2}$ (vector potential squared) terms.
- The $B$ term (from $p \cdot A$ ) dominates in the long-wavelength limit and vanishes in centrosymmetric media under the dipole approximation (surface-sensitive).
- The $U$ term (from $A^2$ ) arises from charge density contributions and remains non-zero in the bulk (bulk-sensitive).
Ratio Analysis: The authors derived the ratio $|B/U|$ to estimate the crossover point where bulk plasma-like nonlinearities ( $U$ ) overwhelm surface dipole responses ( $B$ ).
Coupling Depth: They also referenced the work of Mizrahi and Sipe to calculate the "coupling depth" ( $\xi$ ), which defines the effective depth contributing to bulk SHG based on phase-mismatch and wavelength.

B. Computational Approach

Method: Velocity-Gauge Real-Time Time-Dependent Density Functional Theory (VG-RT-TDDFT). This method is crucial because it naturally includes the full multipole expansion (dipole, quadrupole, etc.) without relying on the dipole approximation.
Software: Customized version of the SIESTA code.
System Model: Diamond slabs modeled as (001)-terminated and (111)-terminated surfaces (16 layers total, with vacuum separation).
Layer Control Strategy: To isolate depth-dependent contributions, the authors manipulated the pseudopotentials. By treating the C 1s core states as either "valence" (active) or "core" (frozen) for specific layers, they could selectively "turn on" or "turn off" the SHG response of specific atomic layers.
- Surface Sensitivity Test: Comparing signals from the top 2 layers vs. the top 8 layers.
- Symmetry Test: Calculating tensor components that are symmetry-forbidden in the dipole limit (e.g., $\chi^{(2)}_{XZZ}$ for a (001) surface) to detect bulk quadrupole contributions.
Parameters: Simulations covered photon energies from 282 eV (near the C K-edge) up to 7 keV, with intensities ranging from $10^{15}$ to $10^{22}$ W/cm².

3. Key Contributions

Quantitative Crossover Mapping: The study provides a precise energy-dependent map of when X-ray SHG transitions from surface-dominated to bulk-dominated behavior in diamond.
Validation of Multipole Effects: It demonstrates that the $A^2$ term (bulk quadrupole response) becomes the dominant driver of SHG at high X-ray energies, validating theoretical predictions made decades ago by Freund and Levine.
Crystal Orientation Dependence: The paper highlights that surface sensitivity is not uniform; it depends significantly on crystal orientation (comparing (001) vs. (111) terminations), with (111) showing more complex, non-monotonic behavior due to symmetry and bond distance variations.
Methodological Advancement: It successfully applies VG-RT-TDDFT with controlled pseudopotential layering to deconvolve surface and bulk contributions in a way that standard perturbation theory (which often neglects higher-order bulk terms) cannot.

4. Key Results

Analytical Prediction: The ratio $|B/U|$ reaches unity (indicating equal surface and bulk contributions) at approximately 3400 eV. The coupling depth analysis suggests the transition begins earlier, in the range of 1000–3500 eV.
Near the C K-edge (~285 eV):
- SHG is highly surface-sensitive.
- For (001) surfaces, the signal is overwhelmingly dominated by the top two atomic layers (>95% contribution).
- This confirms X-ray SHG is an effective tool for probing buried interfaces and surface structures when tuned to resonant edges.
Intermediate Energies (1000 eV):
- The signal becomes largely bulk-sensitive.
- The contribution from deeper layers (3–8) rises rapidly, and the surface-to-bulk ratio drops significantly.
- Symmetry-forbidden tensor components (indicative of bulk quadrupole response) begin to emerge above zero.
High Energies (>3000 eV to 7000 eV):
- SHG is overwhelmingly bulk-dominated.
- By 3000 eV, the bulk response is roughly 10 times stronger than the surface contribution.
- By 7000 eV, the signal is effectively a bulk probe, with each layer contributing equally, consistent with a plasma-like response of nearly free electrons.
Orientation Effects: The (111)-terminated surface showed lower surface sensitivity near the K-edge compared to (001) and exhibited non-monotonic behavior, likely due to ionization effects and additional anisotropic tensor components allowed by $3m$ symmetry.

5. Significance

This work resolves a critical ambiguity in the field of nonlinear X-ray spectroscopy. It establishes that:

X-ray SHG is a tunable probe: Researchers can choose their photon energy to target specific depths.
- Soft X-rays (near resonance): Ideal for surface science, interface chemistry, and symmetry breaking at surfaces.
- Hard X-rays: Ideal for probing bulk structural parameters (bond lengths, interplanar spacings) via plasma-like nonlinearities.
Interpretation Guidelines: The paper provides a "rule of thumb" for experimentalists: to maintain surface sensitivity in diamond, measurements must remain close to the absorption edge (below ~1000 eV). Above this threshold, bulk effects dominate, and surface signals are likely swamped.
Theoretical Validation: It bridges the gap between early theoretical predictions (Freund/Levine) and modern experimental capabilities (XFELs), confirming that the transition from dipole-dominated surface physics to quadrupole-dominated bulk physics occurs in the soft-to-hard X-ray crossover region.

In conclusion, the paper demonstrates that while X-ray SHG loses its traditional surface specificity as energy increases, it remains a powerful, energy-tunable tool for investigating materials, provided the photon energy is selected based on the desired depth of interrogation.