Modeling Coherent Nonlinear Microscopy of Axially Layered Anisotropic Materials Using FDTD

This paper expands a previously developed FDTD-based numerical pipeline to quantitatively model coherent nonlinear microscopy, specifically third-harmonic generation, in axially layered anisotropic materials obeying Kleinman Symmetry, thereby overcoming prior limitations regarding diagonal-only nonlinear susceptibilities and enabling better interpretation of images affected by sample linear heterogeneity.

The Gist

Imagine you are trying to take a super-clear photograph of a tiny, invisible world inside a cell. To do this, scientists use a special kind of camera called Coherent Nonlinear Microscopy. Instead of using a flash, this camera fires a laser beam so intense that it makes the tiny structures inside the cell "glow" with their own light.

However, there's a catch. The physics of how this light glows is incredibly complicated. It's like trying to predict how a complex machine will vibrate when you hit it, but the machine is made of different materials, has weird shapes, and the vibrations depend on the exact angle you hit it from.

For a long time, scientists had to use simplified math to understand these images. They had to pretend the materials were simple and uniform, which meant they couldn't explain what was happening in complex, real-life biological tissues (like the cornea of an eye or layers of skin).

The New Tool: A "Digital Time Machine"

The authors of this paper, Mohammad and Nicolas, have built a new digital simulation tool (based on a method called FDTD) that acts like a "time machine" for light. Instead of guessing, they simulate exactly how light waves bounce, twist, and interact with every single layer of a sample, second by second.

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

1. The Problem: The "Diagonal" Limitation

Previously, their simulation tool was like a toy car that could only drive straight. It could only handle materials where the light behaved in a simple, straight-line way (called "diagonal" properties).

• The Reality: Real biological materials, like collagen fibers in your skin or cornea, are like complex 3D mazes. Light doesn't just go straight; it bounces off walls, twists, and changes direction depending on the angle.
• The Limitation: Because their software had restrictions, they couldn't model these "mazes" accurately. They could only simulate simple, flat materials.

2. The Breakthrough: Unlocking the "Maze"

The team upgraded their software to handle anisotropic materials.

• The Analogy: Imagine you were previously only allowed to build with square Lego bricks. Now, they've given you specialized, angled bricks that can be stacked in layers to build complex, twisting towers.
• The Science: They expanded their code to handle "layers" of material stacked along the path of the light beam. They also taught the computer to understand Kleinman Symmetry, which is a fancy rule that says, "If you rotate the light or the material, the physics stays consistent." This allowed them to simulate how light interacts with complex, layered structures like the cornea of an eye.

3. What They Tested (The "Lab Rats")

To prove their new tool works, they ran three types of tests:

• Test A: The Simple Glass Slab (Isotropic)

  • The Setup: They simulated a block of glass surrounded by water.
  • The Result: They showed that when they used a circularly spinning laser (like a spinning top), the light didn't glow at all. When they used a straight laser, it glowed brightly. This matched the known laws of physics perfectly, proving their "engine" runs smoothly.

• Test B: The Collagen Fiber (Anisotropic)

  • The Setup: They simulated a layer of collagen (the protein that gives skin and eyes their structure). Collagen is like a bundle of straws all pointing in the same direction.
  • The Result: When they hit this bundle with light, the simulation showed the light glowing in specific patterns that matched real-world experiments. It proved their tool could handle the "twisting" nature of real biological tissue.

• Test C: The Double-Action (SHG + THG)

  • The Setup: Some materials do two things at once: they double the light's frequency (SHG) and triple it (THG).
  • The Result: They simulated a material that does both. Their tool successfully predicted how these two different "glows" would mix and interfere with each other, something previous simplified models couldn't do accurately.

Why This Matters

Think of this new simulation as a flight simulator for light.

• Before: Scientists could only simulate flying a plane in a straight line on a calm day.
• Now: They can simulate flying through a storm, over mountains, and through complex wind tunnels.

This is huge for biology and medicine because:

1. Better Diagnosis: It helps doctors and researchers understand exactly why certain tissues look the way they do under a microscope, leading to better diagnoses of diseases.
2. No More Guessing: Instead of guessing how light behaves in a messy, layered cell, they can now calculate it precisely.
3. Future Proof: Their method can be expanded to simulate even more complex light interactions, potentially helping design better lasers and medical imaging tools in the future.

In short: They took a rigid, limited tool and turned it into a flexible, powerful engine that can finally simulate the messy, beautiful, and complex reality of how light interacts with living tissue.

Technical Summary

Here is a detailed technical summary of the paper "Modeling Coherent Nonlinear Microscopy of Axially Layered Anisotropic Materials Using FDTD."

1. Problem Statement

Quantitative interpretation of coherent nonlinear microscopy images (such as Third-Harmonic Generation, THG) is hindered by complex phase-matching conditions, particularly when samples exhibit linear optical heterogeneity (e.g., refractive index variations).

• Limitations of Existing Models: Traditional semi-analytical frameworks (like Angular Spectrum Representation with Green's functions) assume homogeneous linear optical properties. They fail to account for micron-scale refractive index heterogeneities common in biological samples, which significantly impact image contrast in techniques like Coherent Anti-Stokes Raman Scattering (CARS) and THG.
• Limitations of Previous FDTD Approaches: The authors' previous Finite-Difference Time-Domain (FDTD) models could handle linear heterogeneity but were restricted to diagonal nonlinear susceptibility tensors. This limitation, imposed by software constraints, meant the models could only accurately simulate linear polarizations along specific axes (x or y) and could not handle general anisotropic materials or non-diagonal tensor elements required for Second-Harmonic Generation (SHG) and complex polarization states.

2. Methodology

The authors developed an expanded FDTD numerical pipeline using Ansys Lumerical FDTD to simulate nonlinear microscopy in anisotropic, axially layered materials.

• Core Algorithm:
  • The simulation solves discretized Maxwell's equations in the time domain on a 3D grid.
  • Nonlinear polarization ($\vec{P}$) is calculated explicitly at every time step based on the electric field ($\vec{E}$) and material-specific nonlinear tensors ($\chi^{(2)}$ and $\chi^{(3)}$).
  • Kleinman Symmetry: To reduce computational load, the authors assumed Kleinman symmetry, allowing the reduction of the full tensor components into a manageable set of coefficients.
    • SHG ($\chi^{(2)}$): Reduced to 6 independent components.
    • THG ($\chi^{(3)}$): Reduced to 10 independent components.
• Material Implementation:
  • Custom material plugins were created to handle:
    1. Second-order nonlinear materials.
    2. Third-order nonlinear materials.
    3. Materials with both second- and third-order properties.
    4. Isotropic third-order materials (simplified input).
  • The code calculates nonlinear polarization by multiplying electric field vectors stored in memory, assuming the mesh points for $E_x, E_y, E_z$ are ordered identically.
• Geometric Constraints:
  • Due to software limitations regarding mesh optimization for complex geometries, the model is currently restricted to axially layered structures (layers stacked along the optical axis).
  • A pre-simulation check function (provided by Synopsys) is used to verify that the mesh ordering condition is met.
• Simulation Setup:
  • Simulations were run on CPUs (GPU implementation introduced excessive noise).
  • Linear dispersion was neglected (constant refractive indices) to isolate nonlinear effects.
  • Tightly focused Gaussian beams (NA = 0.95) were used to simulate high-resolution microscopy conditions.

3. Key Contributions

1. Extension to Anisotropic Tensors: The primary contribution is the successful implementation of non-diagonal nonlinear susceptibility tensors within the FDTD framework, enabling the modeling of anisotropic materials (e.g., collagen fibers).
2. Unified Framework for Degenerate and Non-Degenerate Processes: The model supports both degenerate processes (THG, SHG) and non-degenerate processes (Sum-Frequency Generation - SFG, Four-Wave Mixing - FWM, Third-Order Sum-Frequency Generation - TSFG).
3. Validation on Complex Geometries: The authors validated the model against well-described geometries, including isotropic interfaces and anisotropic uniaxial layers mimicking the corneal stroma.
4. Handling Mixed Nonlinearities: The framework can simultaneously model materials possessing both $\chi^{(2)}$ and $\chi^{(3)}$ properties, allowing for the study of interplay between cascaded second-order processes and direct third-order processes.

4. Results

The model was validated through four distinct simulation scenarios:

• Isotropic THG (Validation):

  • Simulated a glass-water interface.
  • Result: Correctly reproduced the known physical phenomenon where THG signal vanishes under circular polarization due to the cancellation of field components, while remaining strong under linear polarization.
  • Non-Degenerate THG: Successfully simulated FWM and TSFG signals from two orthogonal input beams, correctly predicting the polarization states of the generated frequencies (e.g., $2\omega_2 + \omega_1$parallel to$3\omega_1$).

• Anisotropic SHG (Collagen/Cornea):

  • Modeled a 2µm slab of uniaxial material (mimicking corneal stroma) with a specific $\chi^{(2)}$ tensor.
  • Result: The model reproduced the expected polarization dependence of SHG intensity as a function of the incident linear polarization angle. It correctly captured the generation of a $y$-polarized SHG component from an $x$-polarized input, a signature of non-diagonal tensor elements.

• Non-Degenerate Anisotropic SFG:

  • Simulated two beams at different wavelengths (1000 nm and 1200 nm) with orthogonal polarizations.
  • Result: Detected Sum-Frequency Generation (SFG) at 550 nm and correctly identified the polarization dependence of the SFG signal relative to the fundamental beams.

• Combined SHG and THG in Anisotropic Media:

  • Modeled a material with both $\chi^{(2)}$ and $\chi^{(3)}$ (collagen-like).
  • Result: The simulation successfully separated and analyzed the SHG (600 nm) and THG (400 nm) signals. It demonstrated that while SHG polarization dependence remained standard, the THG signal exhibited a complex, non-trivial polarization dependence resulting from the interference between direct third-order processes and cascaded second-order processes.

5. Significance and Future Perspectives

• Quantitative Imaging: This work provides a robust numerical tool to interpret coherent nonlinear microscopy images in biologically relevant, heterogeneous samples where linear refractive index variations cannot be ignored.
• Biological Applications: The ability to model anisotropic structures like collagen fibers (SHG) and lipid-water interfaces (THG) simultaneously allows for better understanding of tissue microstructure, orientation, and composition without labels.
• Generalizability: While currently limited to axially layered geometries due to meshing constraints, the authors posit that optimizing the meshing algorithm will make this approach applicable to arbitrary 3D geometries.
• Open Source: The authors are releasing the code and compilation instructions, facilitating the adoption of these advanced FDTD plugins by the broader microscopy community.

In summary, this paper bridges a critical gap between theoretical nonlinear optics and practical biological imaging by extending FDTD simulations to handle the full complexity of anisotropic, layered, and heterogeneous nonlinear materials.