🔭 astrophysics

Impact of segmented deformable mirrors on high-contrast testbeds for exoplanet imaging with future large space telescopes: contrast stability assessment on the HiCAT bench

This study assesses the impact of segmented deformable mirror misalignments on high-contrast exoplanet imaging by combining HiCAT testbed experiments and digital twin simulations to demonstrate that sub-nanometer aberrations significantly degrade contrast, thereby underscoring the critical need for precise cophasing control in future large space telescopes like the Habitable Worlds Observatory.

Original authors: Benjamin Buralli, Mamadou N'Diaye, Raphaël Pourcelot, Marcel Carbillet, Emiel H. Por, Iva Laginja, Ludovic Canas, Sarah Steiger, Peter Petrone, Meiji M. Nguyen, Bryony Nickson, Susan F. Redmond, Anany

Published 2026-04-29

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Benjamin Buralli, Mamadou N'Diaye, Raphaël Pourcelot, Marcel Carbillet, Emiel H. Por, Iva Laginja, Ludovic Canas, Sarah Steiger, Peter Petrone, Meiji M. Nguyen, Bryony Nickson, Susan F. Redmond, Ananya Sahoo, Laurent Pueyo, Marshall D. Perrin, Rémi Soummer

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 trying to take a photo of a tiny, faint firefly sitting right next to a blindingly bright spotlight. That is the challenge astronomers face when trying to image Earth-like planets orbiting other stars. The star's light is so overwhelming that it drowns out the planet, much like trying to see a candle flame next to a searchlight.

To solve this, scientists use special telescopes with "coronagraphs"—devices that act like a cosmic pair of sunglasses to block the star's glare. However, for these sunglasses to work perfectly, the telescope's mirror needs to be absolutely perfect and perfectly still.

The Problem: The Mirror is a Puzzle, Not a Sheet
Future space telescopes will need to be huge to see these distant planets. But you can't launch a single giant mirror into space; it won't fit in a rocket. So, engineers build them out of many smaller mirror pieces, like a puzzle, called a "segmented mirror."

The paper describes a test called HiCAT, which is a laboratory model of such a telescope. It uses a special mirror made of 37 tiny hexagonal segments (like a honeycomb). The goal is to see if these tiny pieces stay perfectly aligned. Even if they drift apart by a distance smaller than the width of a human hair (specifically, less than a nanometer), it creates ripples in the light that ruin the image, making the "firefly" (the planet) disappear again.

The Experiment: Watching the Mirror Breathe
The researchers wanted to know: Do these mirror segments wobble or drift over time, even when we tell them to stay still?

To find out, they set up a high-tech "stethoscope" for light called a Zernike Wavefront Sensor. Think of this sensor as a super-sensitive ruler that can measure the shape of the light beam with incredible precision. They placed this sensor right after the mirror to watch the segments for 30 minutes.

The Challenges: Air and Moving Parts
The experiment wasn't easy. The lab had two main problems:

Air Turbulence: Just like heat rising off a hot road, the air inside the lab moved around, creating tiny distortions in the light.
The Flipping Mirror: To get the light to the sensor, a mirror had to physically flip out of the way. When this motor moved, it got hot and wobbled slightly, messing up the alignment.

The team had to be clever. They used the surrounding mirror segments as a "reference group." Since the air turbulence affects all nearby segments equally, they could measure the wobble of the neighbors and subtract it from the main segment's measurement. It's like if you were trying to measure a single person's heartbeat, but the whole room was shaking; you'd measure the room's shake first and remove it to see the actual heartbeat.

The Findings: The Mirror is Surprisingly Steady
After cleaning up the data, the results were very encouraging:

Sub-Nanometer Stability: The mirror segments were incredibly stable. They drifted by only a few picometers per minute. To visualize this: a picometer is to a millimeter what a millimeter is to the distance from the Earth to the Moon. The mirror was essentially holding its breath perfectly.
Sensor Precision: The sensor they built could detect these tiny movements with a precision of less than 100 picometers.

The Impact: Why a Tiny Wobble Matters
The researchers then used a "digital twin" (a computer simulation) of the telescope to see what happens if these tiny, real-world movements actually occur in the final image.

The Result: Even with such tiny movements, the "darkness" of the space around the star (where the planet should be) got about 2.5 times brighter (less dark) than it would be if the mirror were perfectly frozen.
The Comparison: This simulated result matched very closely with the actual images taken in the lab. This proves that the tiny, natural drifts of the mirror segments are a major reason why the images aren't as perfect as the theoretical best.

The Bottom Line
This paper shows that while these segmented mirrors are incredibly stable, they aren't perfectly still. That tiny amount of movement is enough to degrade the quality of the image significantly.

The takeaway for future space missions (like the planned Habitable Worlds Observatory) is clear: We need extremely precise systems to constantly sense and correct these tiny movements in real-time. If we want to take a clear picture of an Earth-like planet, we can't just build a segmented mirror; we have to build a mirror that knows exactly how to hold its shape against the slightest drift.

1. Problem Statement

Future flagship space telescopes, such as the Habitable Worlds Observatory (HWO), aim to directly image Earth-like exoplanets around Sun-like stars. This requires achieving extreme starlight suppression (contrast levels of $10^{-10}$ ). To achieve the necessary collecting area, these telescopes will utilize segmented primary mirrors.

However, segmented mirrors introduce unique challenges:

Thermo-mechanical drifts can cause nanometer-scale misalignments (piston, tip, and tilt) between segments.
These misalignments generate mid-order spatial frequency aberrations that degrade coronagraphic performance, potentially limiting contrast to $10^{-6}$ if uncorrected.
While primary mirror phasing systems exist, segmented Deformable Mirrors (DMs) in the optical train (or in laboratory testbeds) are used to correct residual errors and emulate segment drifts.
The Gap: There is a critical need to quantify the intrinsic temporal stability of segmented DMs. If the DM itself drifts or exhibits instability, it limits the achievable contrast in high-contrast testbeds and complicates the validation of wavefront control strategies for future missions.

2. Methodology

The study was conducted on the HiCAT (High-contrast Imager for Complex Aperture Telescopes) testbed at the Space Telescope Science Institute and Université Côte d'Azur.

Experimental Setup

Testbed: HiCAT uses an IrisAO segmented DM with 37 hexagonal segments (3 rings around a central segment).
Sensing: A Mid-Order Zernike Wavefront Sensor (MOWFS) was integrated to measure segment-level aberrations.
- The MOWFS uses a Zernike phase-contrast method with a focal plane mask ( $2.5 \lambda/D$ ) to convert phase aberrations into intensity variations.
- It specifically targets mid-spatial frequencies (segment misalignments) rather than low-order global tip/tilt.
Control Algorithm: An Interaction Matrix approach was used.
- Piston, Tip, and Tilt (PTT) modes were sequentially applied to each of the 37 segments with a 10 nm peak-to-valley (PTV) amplitude.
- A command matrix (pseudo-inverse of the interaction matrix) was derived to reconstruct wavefront errors from MOWFS intensity measurements.

Data Processing & Noise Mitigation

Raw data exhibited significant noise due to:

Air turbulence within the HiCAT enclosure.
Mechanical instability of the flip-mount mirror directing light to the MOWFS.

To address this, the authors implemented:

Synthetic Reference Signal: Using neighboring segments (which were kept flat) as a reference to estimate and subtract local turbulence contributions from the actuated segment's measurement.
Active Alignment: An algorithm to automatically realign the ZWFS mask to minimize global tip-tilt errors caused by the flip-mount drift.

Simulation (Digital Twin)

A digital twin of HiCAT (using the HCIPy and CATKit2 frameworks) was used to simulate coronagraphic images.
Procedure:
1. Generate a "perfect" dark hole with a contrast of $0.5 \times 10^{-8}$ using Electric Field Conjugation (EFC) in the simulator.
2. Inject synthetic temporal wavefront errors into the simulated DM. These errors were generated based on the statistical properties (Power Spectral Density) of the experimentally measured stability of the central segment.
3. Analyze the resulting contrast degradation in the dark zone.

3. Key Contributions

Quantification of DM Stability: First comprehensive measurement of the temporal stability of an IrisAO segmented DM in a high-contrast environment, isolating segment-level drifts.
Noise Reduction Technique: Development of a robust method to subtract turbulence and mechanical drift from segmented DM stability measurements using neighboring segments as references.
Linking Lab to Mission: Bridging the gap between laboratory component stability and system-level contrast performance by simulating the propagation of measured DM errors into coronagraphic contrast degradation.
MOWFS Characterization: Validated the MOWFS capability to measure segment errors with sub-nanometric precision (down to ~60 pm RMS for tip/tilt).

4. Key Results

Temporal Stability Measurements

Stability Rates: The temporal evolution of wavefront errors (WFE) on the central segment was measured as:
- Piston: 5.9 pm RMS/min
- Tip: 4.8 pm RMS/min
- Tilt: 4.1 pm RMS/min
Overall Statistics: Across all 37 segments, the average temporal standard deviation over 30 minutes was:
- Piston: $22 \pm 18$ pm RMS/min
- Tip: $7 \pm 2$ pm RMS/min
- Tilt: $7 \pm 5$ pm RMS/min
Observation: The outer ring segments showed higher noise (up to ~1.26 nm) due to lower signal-to-noise ratios and fewer reference neighbors, but the inner segments demonstrated excellent stability at the tens of picometers level.

Contrast Impact Analysis

Simulation Results: When injecting the measured stability variations into the HiCAT digital twin:
- The contrast degraded from a static flat position of $0.47 \times 10^{-8}$ to $1.2 \times 10^{-8}$ .
- This represents a degradation factor of ~2.5.
Comparison with Reality: The experimental contrast measured on the physical HiCAT bench (reported in a 2024 paper) was $2.5 \times 10^{-8}$ .
Conclusion: The DM stability alone explains a significant portion (factor of 2.5) of the gap between the theoretical best-case contrast and the experimental reality. The remaining discrepancy is attributed to other factors like non-linearities in the DM at low amplitudes, other active component instabilities, and residual static aberrations.

5. Significance and Implications

Validation of Control Strategies: The results confirm that segmented DMs can be stable enough (at the picometer level) to support high-contrast imaging, provided that cophasing sensing and control strategies are robust.
Mission Requirements: For the Habitable Worlds Observatory (HWO), this study highlights that segment-level stability is a critical specification. Even sub-nanometer drifts can degrade contrast by factors of 2–3, which is unacceptable for $10^{-10}$ contrast goals.
Testbed Design: Future testbeds must account for intrinsic DM drifts to avoid misinterpreting component instability as a failure of the coronagraph or wavefront control algorithm.
Future Work: The authors suggest that modeling non-linearities (quantization steps) in the DM response and investigating electrical correlations between actuator channels are necessary to fully close the gap between simulation and experimental results.

In summary, this paper demonstrates that while segmented DMs exhibit excellent intrinsic stability (tens of picometers), their temporal drift is a non-negligible factor in high-contrast imaging. Accurate characterization and modeling of these drifts are essential for designing the wavefront control systems required for future Earth-analog exoplanet missions.