DAMPE: Dark Matter Particle Explorer

• DAMPE is a satellite-based high-energy observatory that measures cosmic rays, gamma rays, and potential dark matter signals using advanced calorimetry and tracking.
• It integrates four sub-detectors (PSD, STK, BGO, NUD) to achieve precise energy, charge, and directional measurements across a wide energy range.
• Extensive on-orbit calibration and robust background rejection—up to 9×10⁴ for protons—ensure reliable spectral analyses and deep astrophysical insights.

Searching arXiv for DAMPE mission and instrumentation papers to ground the article in current arXiv-indexed sources. The DArk Matter Particle Explorer (DAMPE) is a satellite-borne high-energy cosmic ray and gamma-ray observatory launched on December 17, 2015 into a 500 km Sun-synchronous orbit. It was designed for the indirect detection of dark matter, precision measurements of cosmic rays, and gamma-ray astronomy, with sensitivity to high energy electrons and gammas in the energy range 5 GeV to 10 TeV and to hundreds of TeV for nuclei. DAMPE combines a large geometric factor of $\sim 0.3\,\mathrm{m}^2\,\mathrm{sr}$, a deep calorimeter, precision tracking, charge measurement, and neutron detection; it has been operating smoothly in space for more than 8 years since launch on December 17, 2015 (Chang et al., 2017, Li et al., 2024).

1. Mission profile and scientific scope

DAMPE is one of the four scientific space science missions within the framework of the Strategic Pioneer Program on Space Science of the Chinese Academy of Sciences. It is a general purpose high-energy cosmic-ray and gamma-ray observatory whose main objectives are the study of galactic cosmic rays, the electron-positron energy spectrum, gamma-ray astronomy, and the search for dark matter signatures in their spectra (Chang et al., 2017, Alemanno, 2022).

From its earliest mission descriptions, DAMPE was framed as an instrument for precise and high-statistics measurements of electrons, protons, and heavy nuclei, for the study of high-energy gamma rays from astrophysical sources, and for the search for indirect signatures of dark matter through anomalies or features in cosmic electron/positron and photon spectra (Gargano, 2017). The energy region between $1$ and $100\,\mathrm{TeV}$ was identified as a regime to be explored with higher precision than previous experiments, while later mission summaries emphasized spectral measurements up to very high energies, including cosmic electrons/positrons and gamma rays up to tens of TeV, and protons and nuclei up to hundreds of TeV (Gargano, 2017, Silveri, 2022).

This mission scope places DAMPE at the interface between direct cosmic-ray detection in space and indirect ground-based measurements. A plausible implication is that DAMPE was conceived not only as a discovery instrument for spectral structures, but also as a calibration bridge across measurement techniques and energy domains.

2. Detector architecture

DAMPE consists of four principal sub-detectors: the Plastic Scintillator Detector (PSD), the Silicon-Tungsten Tracker-converter (STK), the BGO imaging calorimeter, and the Neutron Detector (NUD). Their functions are complementary: charge identification and gamma vetoing in the PSD, precision trajectory reconstruction and photon conversion in the STK, energy measurement and shower imaging in the BGO calorimeter, and additional lepton/hadron discrimination in the NUD (Chang et al., 2017, Yu et al., 2017, Tykhonov et al., 2017, Zhang et al., 2016).

Sub-detector	Salient properties	Primary functions
PSD	82 modules in two layers; active area larger than 82 cm $\times$ 82 cm; efficiency for $Z=1$ particles can reach 0.9999	Charge measurement from H to Fe; photons/electrons separation
STK	768 silicon micro-strip sensors in 6 double layers; 3 tungsten layers; total active area 6.6 m$^2$	Tracking, charge estimation, photon conversion and direction reconstruction
BGO calorimeter	14 layers, 308 BGO crystal bars; 32 radiation lengths deep; $1.6\,\Lambda_I$	Energy measurement, 3D shower imaging, electromagnetic/hadronic discrimination, trigger
NUD	4 boron-loaded plastic scintillators	Additional hadron/lepton separation via neutron activity

The PSD is the topmost detector and serves two distinct roles. First, it measures the charge state of relativistic ions from $Z=1$ to $Z=26$; second, it functions as an anti-coincidence detector for gamma-ray analyses, because neutral particles do not produce a scintillation signal while charged particles do (Yu et al., 2017). The STK provides six independent measurements of both $x$ and $1$0 coordinates, and the insertion of tungsten sheets after the first three tracking layers gives about one radiation length of conversion material for $1$1 events (Tykhonov et al., 2017). The BGO calorimeter is the main sub-detector for electromagnetic energy measurement; in the flight configuration it is 14 layers deep and measures shower development with fine longitudinal and lateral segmentation (Zhang et al., 2016). The NUD, located at the bottom of the stack, records delayed neutron activity, which is more abundant in hadronic than in electromagnetic showers (Chang et al., 2017).

The combined system was engineered to provide precise energy, charge, and direction measurements with strong redundancy. In mission-level performance summaries, the STK angular resolution at normal incidence and $1$2 is given as $1$3, and the BGO calorimeter energy resolution for electrons is given as $1$4 at $1$5 (Silveri, 2022, Chang et al., 2017).

3. Calibration, alignment, and long-term stability

A defining feature of DAMPE is the extent of its calibration program. The on-orbit calibration procedures documented for the PSD, STK, BGO, and NUD use Galactic cosmic rays, bright known GeV gamma-ray sources, and charge injection into the front-end electronics. The same calibration framework also determines the boundary of the South Atlantic Anomaly (SAA), the live time, and the alignments of the detectors (Ambrosi et al., 2019).

For the BGO calorimeter, both ground and on-orbit calibration chains are explicit. Ground calibration includes pedestal, minimum ionizing particle (MIP) peak, dynode ratio, and attenuation length measurements; the position dependence of light yield along a bar is modeled through

$1$6

with the position-independent combined signal

$1$7

The raw calorimetric energy is then reconstructed as

$1$8

Beam-test studies reported better than $1$9 at $100\,\mathrm{TeV}$0 and $100\,\mathrm{TeV}$1 at $100\,\mathrm{TeV}$2 for raw electron energy reconstruction, with excellent agreement between data and simulation (Zhang et al., 2016).

The STK requires recurrent geometric calibration because its science reach depends directly on sub-strip position accuracy. Internal alignment parameters are determined on orbit with non-showering protons and helium nuclei, solving for per-sensor translations and rotations. After alignment, the position resolution in internal planes was reported as $100\,\mathrm{TeV}$3 for protons and $100\,\mathrm{TeV}$4 for helium at incidence angles $100\,\mathrm{TeV}$5, while time stability remained within 4\%; in operational summaries, re-alignment every 2 weeks kept the position resolution $100\,\mathrm{TeV}$6 for all layers and angles (Tykhonov et al., 2017, Tykhonov et al., 2018).

Long-term trigger stability has also been quantified. A new on-orbit threshold calibration method was developed to account explicitly for electronic noise, and the average increase rate of the trigger thresholds for the first four BGO layers was found to be about $100\,\mathrm{TeV}$7 per year. This threshold evolution produces variations of the high-energy trigger efficiency of cosmic-ray electrons by about $100\,\mathrm{TeV}$8 per year at $100\,\mathrm{TeV}$9 and less than about $\times$0 above $\times$1 (Li et al., 2024). This indicates that threshold drift is significant for low-energy studies but negligible for the high-energy regime central to most DAMPE spectral analyses.

4. Event reconstruction and background rejection

DAMPE’s scientific output depends on separating rare electromagnetic events from a much larger charged-particle background. The BGO calorimeter is central to this task because it provides both precise deposited energy and high-dimensional shower-shape information.

For electron/proton separation in the $\times$2 to $\times$3 interval, a dedicated machine-learning analysis was developed using Boosted Decision Trees (BDT) in the TMVA toolkit. Candidate features include the energy fraction in each BGO layer, layer-by-layer lateral spread, global longitudinal variables, and fit parameters of the shower development. The per-layer lateral spread is defined as

$\times$4

and the longitudinal profile is fit with

$\times$5

with shower-maximum depth given by $\times$6 (Zhao et al., 2018).

Training is performed separately in each energy bin using Monte Carlo electron and proton samples passing pre-selection. The retained classifier uses 1000 trees with max depth 3, and the standard working point is $\times$7 electron selection efficiency. In the sample bin $\times$8–$\times$9, pre-selection cuts give a rejection factor of 3, hard suppression gives a factor of 9, and the BDT rejection at $Z=1$0 signal efficiency is $Z=1$1 per incident proton, corresponding to a rejection factor of $Z=1$2. The total background rejection is thus

$Z=1$3

and the reported result is better than $Z=1$4 proton rejection at $Z=1$5 electron efficiency in every $Z=1$6 bin from $Z=1$7–$Z=1$8 (Zhao et al., 2018).

For gamma-ray selection, DAMPE combines BGO shower discrimination with STK track reconstruction and PSD veto logic. Electromagnetic showers are required to be narrower and earlier than hadronic showers through cuts on variables such as the number of hit crystals in the upper BGO layers, per-layer energy fractions, summed lateral spread in the top layers, and concentration variables. After the BGO cuts, $Z=1$9 of hadrons are rejected, while electromagnetic showers from electrons and photons are retained with $^2$0 efficiency above a few GeV. Charged-particle rejection is then enforced in the PSD by requiring the summed deposit in the relevant PSD bars to remain below an energy-dependent threshold, which reduces electron contamination to $^2$1. Monte Carlo studies give a proton rejection factor of $^2$2, an electron rejection factor of $^2$3, effective acceptance of $^2$4 at $^2$5 and $^2$6 at $^2$7, and an expected charged-particle contamination of $^2$8 above $^2$9 (Xu et al., 2017).

5. Gamma-ray response, PSF calibration, and line-search methodology

In gamma-ray astronomy, DAMPE’s response modeling depends critically on the calibration of its Point Spread Function (PSF) and the exploitation of its high energy resolution. The PSF is parameterized with a double King function,

$1.6\,\Lambda_I$0

where

$1.6\,\Lambda_I$1

Simulations showed that a single King function is insufficient because the tail component materially affects the 68\% containment radius (Duan et al., 2024).

Flight calibration of the PSF uses Vela, Geminga, and Crab pulsars in the $1.6\,\Lambda_I$2–$1.6\,\Lambda_I$3 range, together with 51 bright, hard-spectrum AGNs in the $1.6\,\Lambda_I$4–$1.6\,\Lambda_I$5 range. In the unbinned likelihood fits, the tail parameters are fixed to the simulation values and the core width is allowed to vary. The resulting empirical correction rescales the simulated core width as

$1.6\,\Lambda_I$6

with best-fit parameters $1.6\,\Lambda_I$7 and $1.6\,\Lambda_I$8. After scaling, the reduced $1.6\,\Lambda_I$9 for fits to the Vela pulsar angular distribution decreases from 61 to 10, and the maximum observed fractional improvement in point source sensitivity is $Z=1$0 at $Z=1$1 (Duan et al., 2024).

The same instrumental properties make DAMPE suitable for spectral-line searches. In a five-year search for monochromatic and sharp $Z=1$2-ray structures, two dedicated data sets were defined: LineSearch, containing photons that pair-convert in the STK, and BgoOnly, which for the first time in the analysis of calorimeter-based gamma-ray observatories used photons converting directly in the BGO calorimeter. The combined acceptance is $Z=1$3 at $Z=1$4 and $Z=1$5 between $Z=1$6–$Z=1$7, while the energy resolution is better than $Z=1$8 above $Z=1$9 and reaches $Z=26$0 above $Z=26$1. The analysis used signal-to-noise optimized regions of interest for different dark-matter density profiles and an unbinned likelihood with a sliding energy window. The reported result is that no line signals or candidates are found between $Z=26$2 and $Z=26$3 in the Galaxy; nevertheless, constraints on $Z=26$4 and $Z=26$5 are competitive with previous Fermi-LAT results, and below $Z=26$6 the lower limits on the decay lifetime are stronger by a factor of a few (Alemanno et al., 2021).

6. Scientific results, interpretation, and controversies

DAMPE’s best known early result is its measurement of the all-electron spectrum from $Z=26$7 to $Z=26$8, which revealed a spectral softening around $Z=26$9 and a tentative peak around $x$0 (Yuan et al., 2018). The softening has been treated in the literature as robust, whereas the $x$1 feature has consistently been described as tentative. In review-level discussions of the DAMPE electron data, the softening is attributed either to the maximum acceleration limits of electrons by astrophysical sources or to the breakdown of the assumption of a continuous source distribution at TeV energies, while the tentative peak would require a local source with a quasi-monochromatic injection spectrum (Yuan et al., 2017).

A recurring misconception is that the tentative $x$2 feature constitutes a detection of dark matter. The published theoretical analyses do not support that interpretation in such a direct form. In the dark-matter scenario, the favored explanation requires annihilation of $x$3 particles into $x$4 in a nearby clump or overdensity region, with an inferred clump mass of about $x$5–$x$6 or a density enhancement of $x$7–$x$8 times the canonical local density, which was described as relatively extreme compared with numerical simulations; a moderate enhancement of the annihilation cross section via, e.g., the Sommerfeld mechanism or non-thermal production is then required (Yuan et al., 2017). Follow-up phenomenological studies concluded that Fermi-LAT limits on dark matter annihilation are unable to probe these models in all cases, that Coma cluster radio emission can probe a substantial portion of the parameter space, and that SKA is projected to fully probe the studied models, while KM3NET may be the only upcoming high-energy instrument capable of ruling out the presence of such local overdensities in some scenarios (Beck et al., 2018, Beck et al., 2019).

For hadronic cosmic rays, DAMPE has reported multiple spectral structures. The proton spectrum was measured from $x$9 to $1$00, showing spectral hardening at $1$01 and spectral softening at $1$02 ($1$03 significance). The helium spectrum extends from $1$04 to $1$05, with a softening at $1$06 ($1$07 significance). DAMPE has also resolved nuclear charge peaks up to Ni, and has presented preliminary B/C information relevant to propagation studies (Silveri, 2022). These results are routinely interpreted as evidence that broken power-law descriptions are more appropriate than single power laws for the high-energy proton and helium spectra.

In gamma-ray astronomy, DAMPE has accumulated more than 220,000 photons detected above $1$08 and produced a catalog of 222 gamma-ray sources, most of them consistent with power-law spectra and cross-matched with the Fermi 4FGL catalog (Alemanno, 2022). With 102 months of data, it has also reported the detection of the Fermi bubbles at a significance of $1$09 and a GeV excess in the direction of Galactic center at $1$10 confidence, with spectra and morphology consistent with those observed by Fermi-LAT. The Galactic-center excess component was reported to be compatible with a dark-matter annihilation interpretation with a mass of $1$11 and a velocity-averaged cross section of $1$12 for the $1$13 channel, although unresolved astrophysical populations remain a plausible alternative (Alemanno et al., 29 Dec 2025).

Taken together, these measurements define DAMPE’s role in high-energy astroparticle physics: a precision calorimetric mission with strong particle-identification capability, whose scientific output ranges from charge-resolved nuclei spectra and all-electron measurements to high-energy gamma-ray source studies, dark-matter line searches, and diffuse gamma-ray analyses. The mission’s most durable significance lies not in any single anomaly, but in the combination of deep calorimetry, high energy resolution, stable long-term calibration, and broad energy reach, which has enabled multiple independent tests of cosmic-ray and gamma-ray phenomenology across the GeV–TeV domain (Silveri, 2022, Alemanno, 2022).