A low-energy ion spectrometer with large field of view and wide energy range onboard a Chinese GEO satellite

1 Introduction

Space environment monitoring and weather forecast become extremely important for satellite safety. Charged particles in space are known to damage spacecraft and are regarded as a significant fraction of spacecraft anomalies (Baker 2000, Iucci et al. 2005, Quan et al. 2022). Galactic cosmic rays and solar energetic ions can penetrate the satellite shielding surfaces and generate single event effects. High-energy electrons (above 100 keV) can also penetrate the satellite shielding surfaces, resulting in internal charging and electrostatic discharges. Low-energy ions and electrons (below 100 keV) in the geospace environment possess relatively complex behaviors and variations driven by solar wind–magnetosphere interactions associated with the solar and geomagnetic activities, leading to satellite surface charging and subsequent discharging. They are also regarded as one of the major causes of satellite anomalies (Iucci et al. 2005, Ishii et al. 2021, Ferguson et al. 2015; and references therein).

Geosynchronous Earth orbit (GEO) satellites encounter plasma environments with vastly diverse characteristics. The balance of electric potential may be interrupted at the boundaries of those plasma environments (Koons et al. 2006). Surface charging or subsequent discharging is a serious challenge for satellites (Thomsen et al. 2013, Ganushkina et al. 2021, Spence et al. 1993, Nicolaou et al. 2020, Matéo-Vélez et al. 2018). GEO satellites usually encounter surface charging around the Spring and Autumn equinox points. The photoelectrons suddenly decrease in the Earth’s shadow and the satellites may be charged to a negative potential of several hundred volts, even several kilovolts. The hot plasma flux during geomagnetically active periods may also break the electric potential balance and charge the satellites. The in situ low-energy ion measurements may offer an alternative way to study surface charging (Sarno-Smith et al. 2016). On the one hand, the low-energy ion measurements could directly reflect the ion acceleration by the negative satellite potential. On the other hand, the low-energy ion measurements could partially signify the geomagnetic activity level. The top-hat hemispherical or toroidal electrostatic analyzer (ESA) (Carlson et al. 1982, Young et al. 1988) has been widely used to detect the space low-energy ions in most missions such as Wind (Lin et al. 1995), Cluster (Rème et al. 1997), FAST (Carlson et al. 2001), Venus Express (Barabash et al. 2007), Van Allen probes (Spence et al. 2013), GOES (Dichter et al. 2015), MAVEN (Halekas et al. 2015), and Tianwen-1 (Kong et al. 2020).

In this article, a miniaturized low-energy ion spectrometer (LEIS) payload onboard a Chinese GEO satellite is presented. This payload inherited our previous instrument for the SJ18 mission (Hu et al. 2019), employing a state-of-the-art cooperative design of a top-hat ESA and a pair of angular scanning deflectors to obtain a 360° × 90° large field of view (FOV). Further improvement of its capability is achieved in detecting the space ions with a wider energy range from 50 eV to 25 keV/q and higher resolutions. The LEIS is able to sample the energy flux distributions of ions in 3D space, measure the negative spacecraft surface charging, and monitor the space environment variations associated with magnetospheric storms or substorms.

2 Design of the LEIS payload

The recently launched Chinese navigation satellite is a three-axis stabilized spacecraft in GEO orbit, located at a geocentric distance of 6.6 Re in the geographical equatorial plane as illustrated in Figure 1(a). The typical top-hat ESA (Carlson et al. 1982, Young et al. 1988) has a 360° azimuthal FOV, but only a few degrees for the elevation angles. To obtain a larger FOV of the ESA on a three-axis stabilized spacecraft, a state-of-the-art design is developed by supplementing a pair of angular scanning deflectors into a typical top-hat ESA. The present LEIS payload is an update of our heritage instrument (Hu et al. 2019), persisting a large FOV coverage of 360° azimuthal angles and 90° elevation angles. Relative to the previous spectrometer, the present design increased the inner sphere radius and reduced the separation between the inner and outer spheres. Such key changes can enable us to achieve much higher energy and angular resolutions, as well as a larger analyzer constant that can make the detected energy range extend up to 25 keV/q even if keeping the same range of sphere bias voltage. In addition, the electronic and power supply systems have been further optimized and compacted to fulfill smaller power consumption, weight, and sizes. The present LEIS specifications are listed in Table 1.

Figure 1

(a) Schematic diagram of the satellite flying along the GEO orbit around the Earth. The inset photograph is a flight model of the LEIS payload. (b) Structural and functional schematic diagram of the LEIS payload.

The LEIS payload mainly consists of two parts: a sensor head and an electronic box. In Figure 1(a), the insert photograph is its flight model. Figure 1(b) shows its structural and functional schematics. The sensor head is a rotationally symmetric structure including a pair of angular scanning deflectors (signed as def1 and def2), a top-hat ESA, and a position-sensitive detector. The top-hat ESA has an inner sphere electrode (radius R ₁ = 36 mm), an outer sphere electrode (radius R ₂ = 38 mm), and a small “top-hat” section (radius R ₃ = 40 mm). The position sensitive detector (PSD) is composed of a pair of annular microchannel plates (MCPs) and 16 discrete anodes. As the ESA works, the outer sphere, top hat, and collimator electrodes are grounded, while two deflectors and the inner sphere electrodes are applied to the required voltages. The elevation angles of incident ions from −45° to +45° are selected by scanning the deflector voltages (U _def1 and U _def2). The ion energies are analyzed by sweeping ESA bias voltages (i.e., U _inner). After the selection of deflector and ESA voltages, the incident angles and energies of ions are known. These ions can pass through the ESA and hit the PSD positioned at the ESA exit. The produced charge pulses are collected by the 16 discrete anodes, each anode corresponding to 22.5° azimuth angles (signed as 16 channels from Ch00 to Ch15).

The electronic box includes a digital processing unit (DPU), a high-voltage power supply unit (HVU), and a low-voltage power supply unit. Figure 2 presents the DPU schematics and photographs of the printed circuit board (PCB). Sixteen-channel analog signals that are output from 16 discrete anodes in the sensor head are fast amplified and transferred into digital signals by 16 independent low-noise charge-sensitive amplifier/discriminators (A121). The threshold of each A121 chip for the input signal can be adjusted remotely by changing the reference voltage on the basis of our specially designed self-test module. The radiation-hardened DPU is developed on the basis of an antifuse Field Programmable Gate Array for data acquisition, processing, and storage, as well as the control of HV (high voltage) power supplies, self-test, status monitor, and communication. Two types of memory devices of electrically erasable programmable read only memory (EFPROM) and static random access moemory (SRAM) are applied to store the parameters of different HV values and the measured data of the payload, respectively. The RS422 communication method is employed to receive instructions from the satellite and respond to the satellite. A user-defined serial protocol is used to upload scientific data to the satellite.

Figure 2

(a) Schematics of the DPU and (b) photograph of its PCB.

Figure 3 shows the HVU schematics and its PCB photograph, which involve three sweeping HV power supplies and a fixed HV supply. To achieve the sweeping and fixed HV outputs, a widely used cooperation of optocouplers with negative and positive HV modules is employed. Two bipolar sweeping HV supplies (−3 kV to +3 kV) are designed specifically and applied on a pair of entrance deflectors. One unipolar sweeping negative HV (0 to −3 kV) is placed on the inner sphere electrode, and an adjustable negative HV (0 to −3 kV) is ready for the MCP electrodes. The low voltages required by the DPU and HV power supplies are conveniently obtained through the conversion of a primary power supply from the satellite platform in a low voltage board.

Figure 3

(a) Schematics of the HVU and (b) photograph of its PCB.

Operation modes of the present payload have inherited our previous design (Hu et al. 2009). Only a brief description is presented here. In the self-test mode, fundamental functions and parameters of the instrument can easily be tested on the basis of the specially designed self-test circuit module. In the normal measurement, the voltages of U _inner (0 to −3 kV), U _def1, and U _def2 (−3 kV to +3 kV) are swept synchronously with an exponential change. Sixty-four basic energy bins with 8.5% of increment step by step are made for covering the ion energies from 50 eV to 25 keV/q, and 15 angular bins for the elevation angles from −45° to 45°. Each elevation angle takes 128 steps as a minor cycle including 64 upward and 64 downward sweeping steps. There are 16 minor cycles in which the last cycle is designed for recording the possible UV light response and background noise by setting all sweeping voltages to zero. It is noted that the sweeping steps of HV supplies and the stay time at each of steps can be rearranged via RS422 instructions.

3 Performance of the LEIS payload

3.1 Results of simulation and calibration on the ground

In the design of the payload, we take advantage of Simion software (http://simion.com) to study the ESA performance in detail. The methods used in the simulation have been described previously (Hu et al. 2009). A rectangular ion (proton) beam with uniform position distribution is launched and fully covers the ESA entrance. For all small position units of the ion beam, a set number of protons in a bunch with a Gaussian distribution of 30% energy spread are conically emitted along the same established angular direction with small angular spread in a 0.25° increment one by one.

First, we set U _def1 = U _def2 = 0 to study the case that the incident ions with elevation angle β = 0 fly into the ESA. Under a series of specific sphere bias voltages U _inner and variable ion energies from 40 eV to 30 keV, we record the protons that can arrive at the exit of the ESA through the selection of a spherical electric field and find the relation between the ion energy and sphere bias voltage. Figure 4(a) shows the ion energy response of the ESA at a series of U _inner, and the peak shapes are almost the same for different U _inner. From the energy response spectra in Figure 4(a), we can obtain the energy resolution ΔE/E of the ESA as 8.2%, as shown in Figure 4(b). Similarly, we also deduce that the analyzer constant K is 8.9 by fitting the ratio of the energy E/q of ions to ESA bias voltage U _inner as shown in Figure 4(c). Note that it is key to obtain the analyzer constant K because this parameter enables us to determine the energy of ions on the basis of sphere bias voltage. We also simulated the responses of ESA to the elevation and azimuth angles of ions, as well as the effective length of ion entrance positions. Figure 4(d) shows the angular distribution of ions at elevation angle β = 0, and the resolution of the elevation angle is determined to be 2.1°. Figure 4(e) presents the angular distribution of ions at azimuth angle α = 0, and the resolution of the azimuth angle is 5.4°. Compared with our previous spectrometer (Hu et al. 2019), the present energy and angular resolutions are much better. In addition, the present analyzer constant K is larger than the previous one, meaning much wider energy range can be achieved. These obvious improvements of performance are benefitted from the increase of the inner sphere radius and the reduction of the separation between the inner and outer spheres. Figure 4(f) illustrates the entrance position spectrum of the incident ions, and the effective length is estimated to be approximately 17 mm. Taking the entrance aperture ΔR = 2 mm, the active area A _e is approximately 34 mm². Therefore, another key parameter called the geometric factor (GF) can be estimated by the formula GF = A e × 〈 Δ β × Δ E / E 〉 × ∫ cos α d α (Young et al. 1988). The present GF of ESA for each 22.5° azimuth angular channel is obtained to be 3.9 × 10⁻⁴ cm² sr eV/eV. Considering the 50% MCP efficiency and the 81% transmittance of two molybdenum grids, the absolute GF for each anode channel is 1.6 × 10⁻⁴ cm² sr eV/eV. Note that the GF directly links the measured count rate to the incident ion energy flux.

Figure 4

Simulation results: (a) energy response spectrum, (b) energy resolution, (c) analyzer constant K, (d) elevation angle response, (e) azimuth angle response, and (f) effective length of the entrance aperture for the incident ions.

To investigate the response of ESA for different elevation angles β of incident ions, we apply voltages to entrance deflectors, satisfying U _def1 = −U _def2, i.e., the deflecting voltage U _def = 2U _def1 = −2U _def2. A series of angles are chosen from −45° to 45° as the elevation angles of incident ions, and the voltages of U _inner, U _def1, and U _def2 are swept synchronously. The elevation angle β of incoming ions is determined by the voltage U _def and the energy E/q of the ions. Figure 5(a) depicts the elevation angular spectrum with the change of U _def/(E/q), in which the normalized GFs for different β are extracted and plotted in Figure 5(c). Relative to that at β = 0° (3.9 × 10⁻⁴ cm² sr eV/eV), there is a small difference for different β values. The resolution of the elevation angles is therefore estimated to be less than 3°, as shown in Figure 5(d). From Figure 5(a), we can obtain the relation between the elevation angle and U _def/(E/q), which can be fitted well by a linear function β = 126.31 U d e f / ( E / q ) + 0.133 , as shown in Figure 5(b). This function plays an important role in determining the elevation angle of specific energy ions according to the sweeping voltage; that is, under the given voltages of entrance deflectors and the inner sphere, the energy and elevation angle of incident ions can be determined.

Figure 5

Simulation results for different elevation angles: (a) Elevation angular spectrum as the function of U _def/(E/q), (b) the linear relationship between the elevation angle β and U _def/(E/q), (c) the normalized GF for different elevation angles, and (d) the elevation angular resolution.

The calibration experiment was carried out by employing our developed vacuum test facility, which was described previously (Hu et al. 2019). Due to the limitation of unavailable beam time of ion accelerator facility used previously, a sketchy Penning ionization ion gun was used. In the experiment, several specific energies were chosen to obtain the energy response spectra and the analyzer constant. The elevation angles and azimuth angles were varied by rotating the LEIS payload through the 3-axis turntable. The ion beam flux is measured using a faraday cup that is moved in and out of the beam at the location of the sensor aperture. For a series of specific ion angle and energy, the voltages of U _inner, U _def1, and U _def2 are swept synchronously and the counts of ions that hit the PSD are recorded. Consequently, there are many energy and angle response spectra obtained at different energies, 14 elevation angles, and 16 azimuth angles. Here, only a part of them is presented. Figure 6 shows the experimental results for 1 of 16 azimuth angle channels (Ch06) at different ion energies and elevation angles. The energy response spectrum is presented in Figure 6(b), from which the pear positions are obtained for the corresponding energies. The analyzer constant is therefore deduced and plotted in Figure 6(c). Through a linear function fit, the analyzer constant is determined to be K = 8.94 (i.e., the scope of the line), which is very close to the simulated value of 8.9. Because the range of ion energy is limited from 1,000 to 4,000 eV and the energy is not monochromatic, present calibration experiments were not enough to cover the whole energy range and to truly evaluate the energy resolution. Detailed simulations made up for the shortage of calibration experiments. Figure 6(c) depicts the elevation angular spectrum measured at a specific energy (E/q = 3,000 eV/e), from which the angular resolution can be estimated to be about 3° and it is consistent with the simulation. From the measured incident ion current and the counts from energy and angular response spectrum, the GF is estimated to be ∼1.0 × 10⁻⁴ cm² sr eV/eV. Furthermore, according to the elevation angles and the corresponding deflector voltages, the relationship between the elevation angle β and U _def/(E/q) is obtained and plotted in Figure 6(d). Fitting the data by a linear function, we obtained β = 115.78 U d e f / ( E / q ) + 2.07 that generally reproduces the simulation result. The small difference between experiment and simulation probably come from the uncertainty in experiments such as ion energy and angle values.

Figure 6

Calibration experimental results: (a) energy response spectrum, (b) analyzer constant K, (c) elevation angle response spectrum, and (d) the linear relationship between the elevation angle β and U _def/(E/q).

3.2 Preliminary results of on-orbit measurement

The LEIS payload fulfilled the degassing at a suitable period of time after the satellite launch. Subsequently, within the permitted time, inflight tests were performed including electronics self-checking, HV operation, data storage and transmission, communication with ground stations, etc. Furthermore, the adjustment and optimization of amplifier thresholds and MCP voltages have been made via RS422 instructions. The normal ion measurement started on September 24, 2021. In light of the work scheme introduced above in which the voltages of U _inner, U _def1, and U _def2 are applied by sweeping synchronously with an exponential variation, three-dimensional (3D) measurements on energetic ions in space with a large FOV of 360° × 90° and a wide energy range up to 25 keV/q have been achieved. As an example, Figure 7 presents the raw data from on-orbit measurement in the time of 0.2 s for the individual 584, 4,000, and 22,666 eV energy bins with 8.5% bandwidths. Since the density of space ions in GEO is extremely small, the statistic counts recorded in a very short time for one energy bin are not large but visible. Note that the higher counts observed in 22,666 eV energy bin are because this bin has a larger energy bandwidth. In addition, it is much easier to see from Figure 7(c) that the ions from certain elevation angles cannot be detected by several channels due to the shielding of the spacecraft itself.

Figure 7

Raw data from on-orbit measurement in the instant time of 0.2 s for (a) 584 eV energy bin with 8.5% bandwidth, (b) 4,000 eV energy bin with 8.5% bandwidth, and (c) 22,666 eV energy bin with 8.5% bandwidth.

Based on the raw data measured over a relatively long time, the energy and flux distributions of ions in GEO space can be obtained according to the instrumental parameters and the applied voltages. The energy of ions is determined by the product of the analyzer constant (K = 8.9) and the ESA bias voltage. The differential energy flux of ions is obtained by the quotient of the recorded counts and the GF (GF = 1.6 × 10⁻⁴ cm² sr eV/eV). In the data processing, to obtain the true ion data, we deducted the background noise and possible UV contamination, which can be derived by the counts recorded at the last time of each cycle period in which the voltages of U _inner, U _def1, and U _def2 are all set to zero.

Figure 8 presents the preliminary results recorded on-orbit by different azimuth angle channels (Ch00, Ch04, Ch08, and Ch12) in a day from 2021-09-30 to 2021-10-01. This universal time (UT) corresponds to one magnetic local time (MLT) 12:00 and the next adjacent MLT 12:00. One can see from Figure 8(a1–a4) that the measured initial data are contaminated by the background noises and solar UV light response. Figure 8(b1–b4) shows the data from the background noises and solar UV light response obtained via the specially designed way. At the first glance, the bright pillars are easily found, which represent the UV response. The satellite position varies continually in a day; that is, each azimuth angle channel faces the Sun at different times. The on-orbit measurement results revealed in this case that the strong UV response for different channels emerged at different times. Figure 8(c1–c4) presents the true ion data obtained by using the initial data to subtract the background noises and UV response. For all channels, the recorded data display the identical remarkable enhancements of ion fluxes at the specific energy region (e.g., below 1 keV or above 10 keV) and a certain time of a day (e.g., MLT 00:00), which suggests the occurrence of surface charging or substorm events.

Figure 8

Preliminary results recorded on-orbit by different azimuth angle channels in a day from 2021-09-30 to 2021-10-01. The panels in the left column are the initial data. The panels in the middle column are the UV response and background noise. The panels in the right column are the true ion data obtained by using the initial data to subtract UV response and background noise. (a1, b1 and c1) for Ch00, (a2, b2 and c2) for Ch04, (a3, b3 and c3) for Ch08, and (a4, b4 and c4) for Ch12.

Figure 9 presents the ion energy and flux results measured by a representative channel (Ch05) at UT time from 2021-09-24/04:29 to 2021-10-03/04:29. As the figure shows, noticeable flux enhancements of ions with energies above 2 keV are observed clearly, like flying ribbons, indicating the signature of substorm occurrences. The local bright lines in the low-energy region of hundreds of eV are the results of spacecraft surface charging. Figure 9(b) shows the variation of ion fluxes recorded in 1 day from 2021-09-30/04:29 to 2021-10-01/04:29. One can easily see that the significant flux changes occurred at midnight and dawnside, revealing the intensive plasma injection signatures associated with the possible storm/substorm or surface charging events. When the spacecraft surface has a negative potential, the incoming ions can accelerate and then enter the ESA. As a result, the recorded energy achieves the increasing change relative to the initial energy. In other words the initial ions with energies smaller than the amount of surface potential will be directly speeded up and recorded in the same energy bins, resulting in significantly enhanced counts in such energy bins. As shown in Figure 9(b), an ion accumulation appeared in ∼1 keV energy bins near MLT 00:00 and the other one occurred in ∼500 eV energy bins near MLT 04:00, suggesting two possible surface charging signatures. The charging sustained for approximately 2 h. Therefore, the charging potential on the spacecraft surface can be deduced on the basis of the sudden enhancement of ion fluxes. In the present two cases, the electric potential on the satellite surface is determined to be about −500 V, even up to −2 kV. The further observation on magnetic storm/substorm or surface charging and their evolutions over much longer times will be presented in an article to be appeared (Shan et al. 2022).

Figure 9

Differential energy flux of ions measured by Ch05 during (a) 2021-09-24 to 2021-10-03 and (b) 2021-09-30 to 2021-10-01.

4 Summary

In this article, we report an LEIS developed for a Chinese navigation satellite in geosynchronous orbit. An advanced design combining a top-hat ESA with a pair of angular scanning deflectors enables us to achieve in situ measurement of low-energy ions in 3D space with a large FOV of 360° × 90° and a wide energy range from 50 eV to 25 keV per charge. The performance of the LEIS payload has been evaluated in detail by a series of simulations, calibration experiments, and environmental tests on the ground, as well as the measurements in flight along the satellite orbit. The preliminary results from on-orbit observations demonstrate that the capability of the present LEIS payload can well meet the mission requirements for sampling the low-energy ion distributions in 3D space, measuring the negative satellite potential, and monitoring the magnetospheric storm or substorm activities.