IAXO-D1 Micromegas Detector

• IAXO-D1 Micromegas detector is a low-background gaseous X-ray detector prototype developed for BabyIAXO, leveraging microbulk technology and radiopure materials from CAST.
• It features a time projection chamber with a 3 cm drift volume, a 4 µm aluminized Mylar window, and a 6×6 cm² microbulk readout integrated with AGET electronics for precise event reconstruction.
• The prototype achieves high spatial resolution (≈100 µm at 6 keV) and low intrinsic background, validating its performance for solar axion searches in a helioscope setup.

The IAXO-D1 Micromegas detector is a low-background gaseous X-ray detector prototype developed for BabyIAXO, the intermediate helioscope stage toward the International AXion Observatory. It is a microbulk Micromegas time projection chamber derived from the last generation of CAST Micromegas detectors and adapted to the BabyIAXO detection concept, in which solar-axion-induced X rays are focused by optics onto a small spot on the detector plane. Within that program, IAXO-D1 functions both as a detector prototype for BabyIAXO deployment and as an instrument for dedicated studies of spatial resolution, intrinsic background, radiopurity-oriented construction, shielding, and gas operation (Quintana et al., 9 Sep 2025, Ferrer-Ribas et al., 2023).

1. Lineage, naming, and experimental role

BabyIAXO is described as the intermediate experimental stage toward IAXO and as a fully fledged helioscope with discovery potential. Its architecture comprises a 10 m long dipole magnet with two 70 cm diameter bores, two detection lines, X-ray focusing optics, and low-background X-ray detectors, on a rotating platform enabling 12 hours per day of solar tracking (Ferrer-Ribas et al., 2023). In the broader IAXO conceptual design, ultra-low-background Micromegas X-ray detectors are the baseline focal-plane detectors, imaging focused signal spots of about $0.2\ \mathrm{cm^2}$ at the ends of magnet bores (Armengaud et al., 2014).

The designation “IAXO-D1” belongs to the BabyIAXO Micromegas prototype line. The same development program also includes IAXO-D0, an earlier prototype used for background-discrimination studies and surface-background reduction. In this sequence, D1 is presented as a new optimized BabyIAXO-relevant Micromegas prototype with improved radiopurity, optimized shielding, active veto integration, and new radiopure electronics (Ferrer-Ribas et al., 2023). A later status review characterizes IAXO-D1 more specifically as an underground, radiopure, shielded, low-background BabyIAXO/IAXO Micromegas prototype derived from the CAST microbulk detector concept and installed at the Laboratorio Subterráneo de Canfranc to validate intrinsic-background performance independently of surface cosmic rays (Altenmueller et al., 2024).

The detector’s physics role follows the standard helioscope chain. Solar axions convert into X rays in the magnetic field, the optics focus the emerging photons into a small focal spot, and the detector must record those few-keV events with high efficiency and very strong background rejection. Earlier CAST systems already combined low-background microbulk Micromegas with dedicated X-ray optics as a technological pathfinder for IAXO, but those papers did not use the name IAXO-D1; they are best regarded as direct precursors rather than the D1 prototype itself (Aznar et al., 2015).

2. Detector architecture and materials

IAXO-D1 is a small time projection chamber with a $3\,\mathrm{cm}$ conversion/drift volume. In the dedicated spatial-resolution campaign it was filled with Ar + $5\%$ isobutane at atmospheric pressure and operated in open flow at $5\,\mathrm{L/h}$. An alternative BabyIAXO mixture is under consideration at $500\,\mathrm{mbar}$: $50\%$ Xe, $48\%$ Ne, $2\%$ isobutane (Quintana et al., 9 Sep 2025).

X rays enter through a gas-tight $4\,\mu\mathrm{m}$ aluminized Mylar window, which also serves as the TPC cathode and is supported by a metallic strong-back. The chamber walls are $18\,\mathrm{mm}$-thick radiopure Cu-ETP copper, all gaskets are radiopure PTFE, and a Kapton field shaper is included to improve drift-field uniformity and reduce border effects; in the D1 design it is externally covered with $3\,\mathrm{cm}$0 PTFE to suppress copper fluorescence from the chamber body (Quintana et al., 9 Sep 2025). More generally, the BabyIAXO Micromegas construction line is described as microbulk Micromegas fabricated from radiopure copper and Kapton, reflecting the CAST-derived emphasis on radiopurity and low internal background (Altenmueller et al., 2024).

The readout is a microbulk Micromegas with a two-dimensional X-Y strip pattern covering $3\,\mathrm{cm}$1, with 120 strips per axis at a pitch of $3\,\mathrm{cm}$2. The microbulk mesh has $3\,\mathrm{cm}$3-diameter holes. The detector is connected to the front-end electronics by a custom solder-less “face-to-face connector,” in which a flat Kapton cable with copper pads is mechanically compressed between screwed copper parts to contact the readout pads; this avoids soldering at the detector interface and is relevant to both reliability and radiopurity-oriented implementation (Quintana et al., 9 Sep 2025).

Element	IAXO-D1 value or feature	Context
Drift volume	$3\,\mathrm{cm}$4 deep	BabyIAXO spatial-resolution study
Entrance window	$3\,\mathrm{cm}$5 aluminized Mylar	Cathode, gas-tight, strong-back supported
Chamber body	$3\,\mathrm{cm}$6 Cu-ETP copper	Radiopure construction
Gaskets	Radiopure PTFE	Low-background design
Field shaper	Kapton, $3\,\mathrm{cm}$7 PTFE cover	Uniformity and fluorescence suppression
Readout area	$3\,\mathrm{cm}$8	Two-dimensional X-Y strips
Segmentation	120 strips/axis, $3\,\mathrm{cm}$9 pitch	High granularity
Mesh holes	$5\%$0 diameter	Microbulk mesh
Test gas	Ar + $5\%$1 isobutane	Atmospheric pressure, $5\%$2
Alternative gas	$5\%$3 Xe, $5\%$4 Ne, $5\%$5 isobutane	$5\%$6, under consideration

3. Electronics, calibration, and reconstruction

In laboratory characterization, different IAXO-D1 prototypes were assembled and operated with AGET electronics in an Argon–5% Isobutane mixture. A $5\%$7Fe source was used for characterization, and a representative event showed five triggered strips; the reconstructed spectrum resolved the $5\%$8Fe $5\%$9 line, the $5\,\mathrm{L/h}$0Fe $5\,\mathrm{L/h}$1 line, and the low-energy argon escape peak (Ferrer-Ribas et al., 2023). In the SOLEIL synchrotron campaign, the detector was read out with AGET electronics interfaced through a FEMINOS card, using a $5\,\mathrm{L/h}$2 flat Kapton cable. The strip capacitance, including the cable contribution, was measured to be about $5\,\mathrm{L/h}$3 (Quintana et al., 9 Sep 2025).

The spatial-resolution study used a self-triggered acquisition, and because of the high rate only hit channels were recorded rather than the full channel set. This differed from the intended BabyIAXO low-rate regime, where event rates are expected below $5\,\mathrm{L/h}$4, and limited detailed studies of diffusion and reconstruction (Quintana et al., 9 Sep 2025). Related IAXO/BabyIAXO developments emphasize the move from CAST AFTER-based electronics to AGET-type autotrigger strip readout, since strip-triggering can reduce threshold relative to mesh-triggering because the mesh capacitance is much higher than that of individual strips (Altenmueller et al., 2024).

Event reconstruction in the D1 spatial study was performed with the REST-for-Physics framework. The chain had three stages: raw signal analysis, detector-hits analysis, and track reconstruction. Raw signal analysis removed noise events and isolated pulses; detector-hits analysis converted channel signals into energy deposits with physical $5\,\mathrm{L/h}$5 coordinates and inferred $5\,\mathrm{L/h}$6 from charge-collection time using a Garfield++ drift velocity; track reconstruction connected hits into 3D tracks through shortest-path algorithms and produced topological observables (Quintana et al., 9 Sep 2025). For that specific analysis, event selection was intentionally simple: events were required to have at least one hit in both X and Y, and only single-track events were kept, reflecting the point-like nature of photoabsorbed X rays in the relevant energy range (Quintana et al., 9 Sep 2025).

The key spatial metric was empirical. The spatial resolution was taken as the standard deviation $5\,\mathrm{L/h}$7 of the event mean-position distribution, extracted from a Gaussian fit. An alternative fit model, a step function convolved with a Gaussian, was tested to account for the square beam profile and produced no significant difference in the extracted $5\,\mathrm{L/h}$8 (Quintana et al., 9 Sep 2025). This suggests that, in the reported configuration, the measured widths were dominated by detector-plus-beam response rather than fit-model choice.

4. Low-background strategy and operating environments

The low-background program around IAXO-D1 combines radiopure detector construction, passive shielding, active vetoing, underground operation, and topology-based rejection. BabyIAXO requires a low-background X-ray detector with high efficiency in the 1–10 keV energy range and stringent background rejection capabilities, and the detector line is explicitly aimed at background levels of the order of $5\,\mathrm{L/h}$9 in the standard helioscope normalization (Quintana et al., 9 Sep 2025, Ferrer-Ribas et al., 2023).

The final BabyIAXO implementation is intended to surround the detector with a $500\,\mathrm{mbar}$0 passive lead shield and an active muon veto with nearly $500\,\mathrm{mbar}$1 coverage and target efficiency of $500\,\mathrm{mbar}$2, using $500\,\mathrm{mbar}$3-thick plastic scintillators about $500\,\mathrm{mbar}$4 wide and $500\,\mathrm{mbar}$5 long. A neutron tagger is also under development because cosmic-ray-induced neutrons may mimic X-ray signatures through nuclear recoils in the gas (Quintana et al., 9 Sep 2025). Earlier background-discrimination work with IAXO-D0 already implemented a $500\,\mathrm{mbar}$6 lead castle and a $500\,\mathrm{mbar}$7 triple-layer plastic-scintillator veto with cadmium between layers, specifically to tag cosmogenic-neutron background (Altenmüller et al., 2024).

On the surface, IAXO-D0 reached a final background of $500\,\mathrm{mbar}$8 in the 2–7 keV ROI and central 9 mm spot over 52.1 days, after Micromegas topology cuts, a muon-veto coincidence cut, and advanced veto cuts targeting delayed neutron-like signatures. The same work states that this is the lowest background level achieved at surface level for this detector class (Altenmüller et al., 2024). A later overview reports the same surface program as the IAXO-D0 stage, with $500\,\mathrm{mbar}$9 after the central-region and discrimination analysis and $50\%$0 after the additional neutron-tagging handle, with $50\%$1 efficiency for calibration events (Altenmueller et al., 2024).

IAXO-D1 serves the underground complement to that surface program. Preliminary results at LSC with passive shielding of $50\%$2 lead and $50\%$3–$50\%$4 internal copper gave $50\%$5 with a Xe-Ne mixture and $50\%$6 with Ar in the $50\%$7 range. The better Ar performance led to the suspicion of radon contamination in the Xe-Ne gas mixture (Altenmueller et al., 2024). In this respect, D1 is the intrinsic-background reference of the BabyIAXO Micromegas line.

5. Spatial-resolution measurements

The first dedicated spatial-resolution characterization of IAXO-D1 was carried out at the SOLEIL Metrology beamline with monochromatic low-energy X rays. The detector was scanned under various beam energies, positions, and drift field configurations, with beam energies of 5, 6, 7, 8.5, and 10 keV; drift field from $50\%$8 to $50\%$9; mesh voltage fixed at $48\%$0; and beam sizes between about $48\%$1 and $48\%$2 (Quintana et al., 9 Sep 2025).

At that operating point, the detector yielded a gain of about $48\%$3 and an energy resolution of about $48\%$4 at 6 keV. A practical complication was elevated electronic noise at SOLEIL, which required a threshold of about 120 ADC counts above baseline, compared with a typical laboratory threshold of 40 ADC. In the same campaign, the field shaper was left disconnected because connecting it introduced substantial DAQ noise; this omission became central to the interpretation of off-center and edge results (Quintana et al., 9 Sep 2025).

The central quantitative result was a spatial resolution of approximately $48\%$5 at 6 keV. More precisely, the text states that the minimum is “slightly under $48\%$6” at 6 keV, while at 10 keV the resolution worsens to about $48\%$7. In the conclusion, the central performance over 5–10 keV is summarized as “100–300” $48\%$8 for operation at drift fields of at least $48\%$9 (Quintana et al., 9 Sep 2025). This comfortably exceeds the BabyIAXO requirement of a spatial resolution on the order of, or better than, $2\%$0.

The energy dependence is interpreted as a track-size effect: at higher X-ray energies, the photoelectron track becomes longer and spreads the deposited charge over a larger region, broadening the strip cluster and degrading the event centroid precision (Quintana et al., 9 Sep 2025). The drift-field scan showed a clear optimum near $2\%$1 for all tested energies. Below that field, resolution deteriorated quickly; above it, degradation was more gradual. The paper attributes the low-field data–simulation mismatch to gas impurities such as $2\%$2 or $2\%$3, which were not included in the simulation, and the higher-field deterioration to increased transverse diffusion (Quintana et al., 9 Sep 2025).

Position scans exposed field non-uniformities. At 6 keV and $2\%$4, the resolution along X reached a minimum of about $2\%$5 at the center, while along Y the minimum occurred around $2\%$6. The paper attributes the Y asymmetry to the high-voltage connections located on one side of the readout and oriented perpendicular to the Y axis, which distort the drift field. More generally, moving the beam only a few millimeters away from the center degrades the resolution by a factor of 3–4; even so, the worst measured resolution remains well below the $2\%$7 BabyIAXO requirement (Quintana et al., 9 Sep 2025). This indicates that the intrinsic 2D imaging capability is strong, while field uniformity and low-noise integration of the field shaper are decisive engineering details.

6. Relation to CAST heritage and BabyIAXO deployment

IAXO-D1 is explicitly based on the last generation of CAST Micromegas detectors, but it is embedded in the BabyIAXO low-background concept rather than being a simple carry-over of CAST hardware (Quintana et al., 9 Sep 2025). CAST established the microbulk Micromegas line as a rare-event X-ray technology with radiopure copper-and-Kapton construction, high granularity, topology-based rejection, and background levels that were progressively reduced below $2\%$8 at surface and to the $2\%$9 scale underground (Iguaz et al., 2015). The BabyIAXO/IAXO program inherits that platform and re-optimizes it for optics-coupled focal-plane operation, stringent shielding, and improved electronics (Ferrer-Ribas et al., 2023).

A parallel strand of development came from the CAST sunrise IAXO pathfinder detector, an optics-coupled microbulk Micromegas line installed at CAST and later operated in an extended run with a Xe-based gas mixture. In that pathfinder detector, the active conversion/drift volume had a 3 cm drift distance, a $4\,\mu\mathrm{m}$0 aluminized Mylar X-ray window, and a $4\,\mu\mathrm{m}$1 readout with $4\,\mu\mathrm{m}$2 channels at 0.5 mm pitch, all behind 10–15 cm lead shielding and a plastic-scintillator muon veto. The xenon-based mixture $4\,\mu\mathrm{m}$3 at 1.05 bar was used as an IAXO-relevant pathfinder exercise because it removed the argon fluorescence feature near 3 keV and provided higher X-ray detection efficiency (Collaboration et al., 2024). This pathfinder is not IAXO-D1 in nomenclature, but it is part of the same detector lineage.

Within BabyIAXO, the most important implication of the D1 measurements is that the detector already fulfills and surpasses the spatial-performance specification, while the underground results indicate that the intrinsic-background scale required by IAXO is credible in a CAST-derived microbulk platform (Quintana et al., 9 Sep 2025, Altenmueller et al., 2024). The studies also identify concrete optimization priorities: integrate the field shaper without introducing electronic noise, maintain laboratory-like thresholds rather than the elevated synchrotron threshold, preserve gas purity to avoid low-field collection losses, and complete the combined passive-shielding, active-veto, and radiopure-electronics program foreseen for BabyIAXO (Quintana et al., 9 Sep 2025). A plausible implication is that IAXO-D1 occupies a transitional position between CAST pathfinder operation and the mature BabyIAXO focal-plane detector architecture.