---
title: ASPIICS Coronagraph on Proba-3
url: https://www.emergentmind.com/topics/aspiics-coronagraph
type: topic
---

# ASPIICS Coronagraph on Proba-3

ASPIICS, the **Association of Spacecraft for Polarimetric and Imaging Investigation of the Corona of the Sun**, is an externally occulted solar coronagraph aboard ESA’s Proba-3 formation flying mission. Its defining characteristic is the separation of the telescope and the external occulter onto two spacecraft that fly in precise formation, forming a giant coronagraph with a baseline of around 144 m. This configuration enables observations of the inner solar corona in eclipse-like conditions, close to the solar limb and with very low straylight, over a field of view extending to 3 $R_\odot$; regular observations reach down to 1.099 $R_\odot$, and early results report occasional access to 1.05 $R_\odot$ [2509.00253][2511.01679].

## 1. Mission configuration and scientific scope

ASPIICS was conceived as the primary science instrument of Proba-3, a dual-spacecraft mission in a highly elliptical Earth orbit. One spacecraft carries the telescope and the external occulter is mounted on the second spacecraft; during observations, the two spacecraft fly in a precise formation for 6 hours out of a 19.63 hour orbit, or up to 5.5 hours at a time in the first-results description. The very long distance between the external occulter and the telescope, around 144 m, represents an increase of two orders of magnitude compared to classical externally occulted solar coronagraphs [2509.00253][2511.01679].

The scientific rationale is the direct observation of the poorly covered coronal domain between EUV imagers and classical coronagraphs. The mission objectives described for ASPIICS include the structure and dynamics of the inner solar corona, the origin and acceleration of the slow solar wind, the onset and early evolution of coronal mass ejections, coronal heating, and shock formation in the low corona. This observational regime is explicitly framed as providing eclipse-like conditions in orbit rather than brief eclipse snapshots [2509.00253].

A central design result is that increasing the occulter–telescope distance reduces diffracted light for a given occultation geometry. In the instrument-design study, the relative intensity of diffracted light at the center of the entrance aperture is expressed as

$$
L_{\mathrm{A1} = \left\{ \pi^2 r_\odot \left[ 1-\left(\frac{r_\odot D}{R_\mathrm{EO}\right)^2 \right] \right\}^{-1} \frac{\lambda}{R_\mathrm{EO}
$$

and this scaling is used to motivate the large baseline and the resulting access to very low coronal heights [2509.00253].

## 2. Optical architecture, channels, and measurement modes

ASPIICS is described as a classic externally occulted Lyot coronagraph. The optical system comprises a large CFRP external occulter with a toroidal edge, a 50 mm entrance aperture, primary optics with focal length 734.6 mm, an internal occulter placed in the conjugate optical plane, a Lyot stop, a relay lens assembly, a filter wheel, and an APS detector with 2048 $\times$ 2048 pixels of 10 $\mu$m size corresponding to 2.817 arcsec per pixel. Formation-flying metrology is supported by a Shadow Position Sensor with eight photodiodes around the aperture and an Occulter Position Sensor Emitter using triple light emitters on the occulter spacecraft [2509.00253].

The spectral and polarimetric concept combines wideband white-light imaging, narrowband coronal-line imaging, and full polarimetry. The filter wheel contains six slots, including three wideband polarized channels, one unpolarized channel, and two narrowband channels. Detector operation is tile-based, with 32 $\times$ 32 tiles, and up to 3 exposures per acquisition are used to cover the coronal dynamic range [2509.00253].

| Filter | Peak / FWHM | Main contribution |
|---|---:|---|
| Wideband | 5510.6 Å / 295.5 Å | White-light continuum |
| Pol 0, 60, 120 | 5519.3 Å etc. / $\approx 296$ Å | Polarized continuum |
| Fe XIV | 5303.4 Å / 5.4 Å | Coronal green line (2 MK plasma) |
| He I D$_3$ | 5876.1 Å / 21.0 Å | Prominences (10,000 K–100,000 K) |

The nominal field of view is given as 1.099–3 $R_\odot$, with the instrument unvignetted above 1.174 $R_\odot$. Synoptic imaging is performed at 1 min cadence for the full field of view, while quarter-field modes allow cadences as short as 2 s. In first observations, the operational high-dynamic-range sequence is described as three rapidly acquired exposures, for example 0.1 s, 1 s, and 10 s, merged into HDR images calibrated in mean solar brightness [2509.00253][2511.01679].

## 3. He I D$_3$ coronagraphy and prominence diagnostics

A dedicated pre-launch study optimized the narrow-band filter for prominence and CME observations in the He I D$_3$ line. The analysis used a multi-level 1D non-LTE radiative transfer code for isothermal, isobaric prominence slabs at 8 kK, 30 kK, and 100 kK, with outputs including total energy in the He I D$_3$ line $E_{\rm D3}$, central intensity $I_0$, electron density $n_e$, and line-center optical thickness $\tau_0$. Four candidate filters centered at 5877.25 Å in vacuum were tested: a flat 20 Å top-hat and Gaussian profiles with FWHM of 5, 10, and 20 Å [1807.00155].

The visible-light contribution from prominences was modeled as Thomson scattering on prominence electrons:

$$
E_{\rm VL} = \sigma_T~W(h, \lambda_c)~I_{\rm tot}~n_{e}~D
$$

with

$$
I_{\rm tot} = \int I_0(\lambda) G(\lambda) d\lambda,
$$

and the total narrow-band signal was written as

$$
E_{\rm tot}({\rm nb}) = E_{\rm D3}({\rm nb}) + E_{\rm VL}({\rm nb}).
$$

Because concurrent broad-band visible-light images can be used, the D$_3$ and VL contributions can be separated in analysis [1807.00155].

The principal filter-selection result is that an optimal narrow-band filter should be flat or somewhere between flat and Gaussian with FWHM of 20 Å in order to detect fast moving prominence structures. The study explicitly examined line-of-sight velocities of 0, 100, and 300 km s$^{-1}$; 1 Å shift is given as approximately 50 km s$^{-1}$, so a 20 Å width covers up to 500–600 km s$^{-1}$. At 300 km s$^{-1}$, relative D$_3$ losses are reported as negligible for the 20 Å flat filter, approximately 20% for the 20 Å Gaussian, approximately 60% for the 10 Å Gaussian, and almost complete loss for the 5 Å Gaussian [1807.00155].

Thermally, the maximum emission in the He I D$_3$ line is at 30 kK and the minimal at 100 kK. The ratio $E_{\rm D3}/E_{\rm VL}$ is identified as a temperature diagnostic because both signals are optically thin and the ratio is independent of prominence thickness. Reported ranges are $0.1 < E_{\rm D3}/E_{\rm VL} < 10$ for hot structures at 100 kK, $1 < E_{\rm D3}/E_{\rm VL} < 100$ for cool structures at 8 kK, and $10 < E_{\rm D3}/E_{\rm VL} < 1000$ for warm structures at 30 kK; the paper’s summary also states that the ratio is up to 10 for hot prominence structures, up to 100 for cool structures and up to 1000 for warm structures [1807.00155].

The same study also addresses contamination by Na I D$_1$ and D$_2$. For prominences, contamination is typically below 5%, rarely up to 10–15%, and maximum observed up to 20% in rare cases; for a 20 Å filter, the sodium contribution remains below 13%, whereas a 30 Å filter can raise it to 17% for blueshifted structures. The result is an explicit trade-off: the proposed 20 Å filter favors prominence and CME science at the expense of comet detection in sodium lines [1807.00155].

## 4. Diffraction, misalignment sensitivity, and internal occulter strategy

Diffraction of solar disk light at the external occulter is identified as the major source of straylight in coronagraphs, and ASPIICS diffraction modeling was developed before launch and tested against flight data after launch. The propagation formalism used in the diffraction studies is Fourier-optical and Fresnel-based. In the 2026 observational analysis, the propagated wave amplitude is written as

$$
\Psi_z(x,y) = \frac{\exp \left( \frac{i\pi (x^2+y^2)}{\lambda z} \right)}{i\lambda z} \mathcal F
\left[ A(\xi,\eta) \Psi(\xi,\eta) \exp\left(\frac{i\pi (\xi^2+\eta^2)}{\lambda z}\right) \right],
$$

and for the circular external occulter the entrance-aperture field is expressed through a Hankel-transform formulation [2604.21559].

Before launch, the misalignment study extended an axi-symmetrical diffraction model to non-symmetrical cases and arbitrary misalignments. It compared Sun shift, external-occultor shift, coronagraph shift, coronagraph tilt, longitudinal misalignments, and smaller perturbations such as Lyot stop shift and lens tilts. The dominant result is unambiguous: the most important misalignment is the tilt of the telescope with respect to the line connecting the center of the external occulter and the entrance aperture. The study concludes that the internal occulter with $R = 1.1 R_\odot$ is large enough to compensate possible misalignments in ASPIICS and that apodizing the edge of the internal occulter leads to additional suppression of diffracted light [1801.04204].

The same work formulates the practical alignment prescription as a co-alignment problem between the diffraction fringe from the external occulter and the internal occulter. Its recommended orientation strategy is to point the coronagraph to the center of the external occulter. With a 25 arcsec tilt, the geometric image of the external occulter can extend beyond an undersized internal occulter; increasing $R_{IO}$ from 1.662 mm to 1.694 mm restores suppression in the modeled worst-case tilt, and a 0.1–0.2 mm linear-gradient apodization suppresses diffracted light more effectively than simple oversizing while reducing the vignetting penalty [1801.04204].

Post-launch observations substantiate the model. Diffraction becomes directly measurable during calibration maneuvers with large off-pointings of about 0.8$^\circ$ or moderate off-pointings of about 20–40 arcsec, where the bright diffraction ring and its double-peak profile become visible. Early observations are reported to fully confirm all the qualitative properties of diffracted light suggested by the model. After fine-tuning, including an adjusted distance parameter $l$, an updated solar limb-darkening function at $\lambda = 550$ nm, and an empirical multiplicative factor $k = 1.2$, the model achieves quantitative correspondence at the level of 30%–50%, depending on the configuration [2604.21559].

Operationally, the residual diffraction burden is small. In the majority of the field of view the diffracted light is two orders of magnitude below the coronal signal; in the outer part of the vignetting zone it rises to at most 10% of the coronal brightness; and in typical images with nominal pointing it contributes negligibly, at the approximately 1% level or less. This resolves a common concern about the large-baseline geometry: the principal challenge remains diffraction control, but the combination of formation flying, internal occultation, and model-based calibration keeps the diffracted component below the scientific signal in standard observing conditions [2604.21559].

## 5. Optical ghosts and algorithmic removal

A separate pre-launch analysis addressed optical ghost images formed inside the telescope. The optical layout implemented in Zemax included the entrance aperture, primary objective O1, internal occulter, field lens O2, Lyot stop, relay objective O3, filter wheel, detector glass with ND50% filter, and the 2048 $\times$ 2048 pixel CMOS detector. Sequential and non-sequential raytracing were used to identify the ghost-producing surfaces and to quantify their geometrical behavior and energetics [1812.03990].

The dominant ghost contributors are not the primary objective but backreflections involving the detector and nearby optics: detector glass, filter glass, and relay lenses, especially O3/L5 and to a lesser extent O3/L4. The primary objective produces a virtually negligible ghost because the large distance between the external occulter and the primary objective makes these rays highly divergent, leaving less than 0.03% relative to the dominant ghosts on the detector. On this basis, the study concludes that the use of the Lyot spot in ASPIICS is not necessary [1812.03990].

The correction formalism exploits linearity. If the recorded image is

$$
\mathbf{R} = \mathbf{I} + \mathbf{G}(\mathbf{I}),
$$

then one computes

$$
\mathbf{G}(\mathbf{R}) = \mathbf{G}(\mathbf{I}) + \mathbf{G}(\mathbf{G}(\mathbf{I})),
$$

and applies the single-step estimate

$$
\mathbf{C} = \mathbf{R} - \mathbf{G}(\mathbf{R}) \approx \mathbf{I}.
$$

The justification is that $\mathbf{G}(\mathbf{I}) \ll \mathbf{I}$ and $\mathbf{G}(\mathbf{G}(\mathbf{I})) \ll \mathbf{G}(\mathbf{I})$. The reported residual after one subtraction is below $10^{-5}$ of the coronal signal even in worst-case scenarios [1812.03990].

Ghost light is most relevant in the outer field of view, where bright inner-corona signal is redistributed outward while the true coronal signal is weak. With the ND50% detector glass present, contamination can reach up to about 3% of the signal in the outer field of view; without detector glass it can reach about 10%, although the correction algorithm remains effective. The detector glass reduces the total ghost signal by a factor of about 4. The study further states that, even before subtraction, ghosts are weaker than diffracted light and scattering is less significant still [1812.03990].

## 6. Observational performance and first scientific results

After launch on 5 December 2024, ASPIICS began producing coronal observations that match the mission’s intended low-straylight regime. First results report quasi-stationary structures such as coronal loops, streamers, quiescent prominences, and dynamic phenomena including erupting prominences, coronal mass ejections, jets, slow solar wind outflows, and coronal inflows. The reported observing characteristics are high spatial resolution of 5.6 arcsec, temporal resolution of 30 s, and continuous observing intervals up to 5.5 hours [2511.01679].

The most distinctive early result is the observation of weak, widespread, and persistent small-scale outflows and inflows between 1.3 and 3 $R_\odot$ at a high spatial and temporal resolution for the first time. Outflows described as mini-blobs show speeds of 100–400 km s$^{-1}$, while inflows, often dark, occur with deceleration from about 350 km s$^{-1}$ to about 100 km s$^{-1}$. The authors state that omnipresent and complex fine-scale dynamics is observed so clearly in this region of the corona for the first time, extending the observed scale range of the variable slow solar wind formation region [2511.01679].

The instrument’s multi-band design is already used for thermal and structural discrimination. Fe XIV images reveal hot 2 MK loops, He I images resolve prominence fine structure, and white-light data show leading edges and density morphology. The first-results study states that multi-passband imaging allows tracking prominences in He I D$_3$, hot CME plasma in Fe XIV, and CME leading edges in white light synchronously. For CME cores between 1.5 and 3 $R_\odot$, the kinematics are reported as observed for the first time in this gap region, with the core trajectory fitted by a linear plus power-law form [2511.01679].

Polarimetric capability is also part of the scientific return. The degree and direction of polarization are mapped with sufficient accuracy, with tangential polarization angle reported to $\pm 2^\circ$, and electron density profiles are inverted from polarized brightness using the van de Hulst method. These density results are stated to match classical K-corona models. A plausible implication is that the combination of precise formation flying, low diffraction straylight, and multi-exposure HDR acquisition makes ASPIICS not only a morphology instrument but also a quantitative coronal diagnostic platform across the previously underobserved 1.1–3 $R_\odot$ domain [2511.01679].

Source: https://www.emergentmind.com/topics/aspiics-coronagraph