Research focus: Spacecraft radiation transport · Radiation shielding · VLEO/NEO environments · Monte Carlo modelling · Radiation effects in electronics

This repository presents a computational radiation-shielding experiment investigating how 1 mm, 3 mm and 5 mm structural shielding modifies the radiation environment experienced by an internal silicon sensor.

The study couples SPENVIS radiation-environment models with Geant4/GRAS Forward Monte Carlo (FMC) transport simulations and evaluates the resulting radiation field using multiple damage and exposure metrics:

TID · LET · TNID · TNID-V · Particle Fluence

The central finding is that shielding performance is not universally monotonic: increasing structural thickness can reduce some radiation components while increasing others through particle transport, fragmentation and secondary-particle production.

A conventional first-order shielding assumption is:

More shielding → less radiation at the protected component.

For spacecraft systems exposed to complex particle environments, this assumption is incomplete.

Increasing material thickness changes not only the amount of primary radiation reaching an internal component, but also the particle population through nuclear interactions, electromagnetic interactions and secondary-particle production.

This experiment therefore asks:

Does increasing structural shielding thickness always reduce the radiation exposure and damage experienced by an internal spacecraft sensor?

Three controlled shielding configurations were investigated:

The sensor location, spacecraft geometry and simulation methodology were kept consistent across the campaign.

The strongest TID attenuation occurred for the trapped-electron environments.

For the three trapped-electron segments, increasing shielding from 1 mm → 5 mm reduced TID by:

The dominant TID environment in the simulated campaign was trapped-electron Segment 3, where the 5 mm configuration reduced the deposited dose by approximately 70%.

::contentReference[oaicite:1]{index=1}

Detailed data: analysis/tables/02_TID_attenuation.csv

Figure: analysis/figures/02_TID_attenuation.png

The most important systems-level result is that increasing shielding thickness does not produce a universal reduction across all radiation metrics.

For example, GCR hydrogen showed:

• TID: +60.6%
• LET: −11.2%
• TNID: +39.5%
• TNID-V: +72.8%
• Fluence: +67.9%

from 1 mm → 5 mm.

Similarly, several heavy-ion GCR environments showed substantial increases in displacement-damage metrics.

This demonstrates why evaluating shielding effectiveness using TID alone can be misleading.

TID generally decreased strongly for trapped electrons and solar protons, but several GCR components showed non-monotonic behaviour.

The 1 mm → 5 mm response included:

• Trapped electron Segment 1: −80.1%
• Trapped electron Segment 2: −79.6%
• Trapped electron Segment 3: −69.6%
• Solar protons: −28.9%
• GCR carbon: −13.1%

However:

• GCR hydrogen: +60.6%
• GCR helium: +12.6%
• GCR oxygen: +7.1%
• GCR iron: +2.0%

Thus, additional shielding does not automatically translate into lower deposited dose for every particle population.

::contentReference[oaicite:2]{index=2}

LET also exhibited a particle-dependent shielding response.

Representative 1 mm → 5 mm changes include:

The response therefore cannot be represented by a single shielding attenuation factor.

Figure: analysis/figures/03_LET_response.png

Data: analysis/tables/03_LET_response.csv

TNID showed some of the clearest evidence that increasing structural thickness can alter the radiation environment in ways that are not captured by simple attenuation.

For several GCR species, TNID increased substantially between 1 mm and 5 mm:

The GCR silicon case showed the largest increase:

65.5 → 583.8 MeV/g

corresponding to approximately +792% from 1 mm to 5 mm.

Several of these increases are substantially larger than the propagated Monte Carlo uncertainty; for example, the GCR-Si 1 mm → 5 mm TNID change corresponds to approximately 6.5σ.

At the same time, trapped-electron environments showed strong attenuation:

• Segment 1: −76.4%
• Segment 2: −74.9%
• Segment 3: −66.3%

This contrast is one of the central results of the experiment.

The vacancy-related non-ionizing damage metric, TNID-V, showed the same fundamental behaviour.

Representative GCR changes from 1 mm → 5 mm:

For GCR oxygen:

2,544 → 5,079 MeV/g

corresponding to approximately +99.6%.

Conversely, trapped-electron TNID-V decreased strongly:

• Segment 1: −83.9%
• Segment 2: −82.9%
• Segment 3: −75.3%

These results reinforce the conclusion that shielding effectiveness is radiation-mechanism dependent rather than a simple function of material thickness.

Data: analysis/tables/04_TNID_TNIDV_response.csv

Figures:

• analysis/figures/04_TNID_response.png
• analysis/figures/04_TNID_V_response.png

Particle fluence provides another important indication that shielding changes the transported particle population rather than simply removing particles.

For example:

The increase in GCR fluence for several species is consistent with the possibility of secondary-particle generation and transport through the shielding structure.

Therefore:

Shielding should be treated as a particle-transport problem, not simply as an attenuation problem.

Figure: analysis/figures/05_total_fluence_response.png

Data: analysis/tables/05_fluence_response.csv

The computational workflow combines environment modelling, particle-source generation, Monte Carlo transport and engineering-level post-processing.

 SPACE ENVIRONMENT
 │
 ▼
 ┌───────────┐
 │ SPENVIS │
 └─────┬─────┘
 │
 Environment spectra
 │
 ▼
 Source MAC files
 │
 ▼
 ┌──────────────────────┐
 │ Geant4 General │
 │ Particle Source │
 └──────────┬───────────┘
 │
 ▼
 ┌──────────────────────┐
 │ Geant4 / GRAS │
 │ Forward Monte Carlo │
 │ Particle Transport │
 └──────────┬───────────┘
 │
 ▼
 Radiation interaction
 with spacecraft
 │
 ▼
 ┌──────────────────────┐
 │ Internal Si Sensor │
 └──────────┬───────────┘
 │
 ▼
 ┌──────────────────────────────────┐
 │ Radiation Response Metrics │
 │ │
 │ TID · LET · TNID · TNID-V │
 │ Particle Fluence │
 └────────────────┬─────────────────┘
 │
 ▼
 Python Post-Processing
 │
 ▼
 Engineering Interpretation