Initial Results of Low Earth Orbit Space Radiation Dosimeter on Board the Next Generation Small Satellite-2

The PD of LEO-DOS employs a silicon photodiode particle sensor, which is situated between two 10 mm layers of A-150 TEP, as illustrated in Fig. 2. This configuration, with one layer above and one below the sensor, allows for the assessment of the effective dose to the human body in accordance with ICRP Publication 103 (Charles, 2008). The particle sensor for the PD is a n-type silicon photodiode sensor with an effective area of 20 × 20 mm and 650 ± 30 μm thickness, developed for the application to the cosmic ray energetic and mass for the international space station experiment for the International Space Station (ISS) in Korea. Its properties as a particle detector are investigated in reference (Hyun et al. 2015).

Fig. 2. Schematic view of a particle dosimeter of LEO-DOS. The silicon photodiode particle sensor is sandwiched between two layers of 10 mm A-150 tissue-equivalent plastic, with one layer on the upper side and one on the lower side. An A225 charge sensitive preamplifier and shaping amplifier is placed close to the silicon sensor to reduce stray noise. LEO-DOS, low Earth orbit space radiation dosimeter.

Download Original Figure

The absorbed dose for charged particles with a silicon detector is calculated by dividing the sum of the deposited energy in the silicon sensor by the mass of the silicon sensor. The calculation formula for the absorbed dose in silicon can be expressed as follows:

where N_i is the count of the i-th channel, ε_i is the energy of the i-th channel, and m is the mass of the silicon sensor.

D_Si is converted to D_H20 by converting LET in silicon (L_Si) to LET in water (L_H20) through the Benton equation (Benton et al. 2010) as follows:

where l̅ is a mean chord length of the PD of 919 μm obtained by dividing the thickness of the silicon sensor of PD of 650 μm by cos (45°) considering the isotropic incidence of the charged particles in space.

The biological effectiveness factor, also known as the quality factor (Q(L)) can be expressed as a function of LET (L) according to ICRP publication 60 as follows (ICRP 1991):

The dose equivalent can be expressed as the product of absorbed dose and mean quality factor (Q_mean) as shown in the following equation:

Fig. 3 presents a schematic view of the NS configuration. A 1.5” × 1.5” stilbene (C₁₄H₁₂) organic scintillator was adopted as a sensor of a fast NS. To mitigate the influence of charged particles in near-Earth space, a hollow plastic sensor (EJ-200, ELJEN Technology, Sweetwater, TX, USA) coated with EJ-510 reflective paint is installed around the stilbene sensor, serving as a VETO sensor to filter out these particles. Each scintillator sensor is optically coupled with an array of SiPMs (J-series 60035, SensL, now ON Semiconductor, Cork, Ireland). The stilbene sensor is coupled with a 4 × 4 SiPM array, while the VETO sensor is coupled with a 24-SiPM array.

Fig. 3. Schematic view of a neutron spectrometer which consist of stilbene and VETO sensor. The VETO sensor, designed with a hollow structure, employed an EJ-200 plastic detector coated with EJ-510 reflective paint.

Download Original Figure

The advantage of using stilbene for a fast neutron dosimeter is that it does not require an unfolding algorithm to obtain dosimetric parameters such as the absorbed dose (D) and equivalent dose (H) (Miller et al. 2020). Additionally, since the stilbene sensor has tissue-equivalent characteristics, the impact of neutrons in LEO on the human body can be evaluated through the stilbene absorbed dose.

The absorbed dose (D_stilbene) in stilbene is calculated by dividing the sum of the recoil proton energy in the stilbene crystal by the mass of the stilbene sensor. The calculation formula for the absorbed dose in stilbene in the energy range from 2 to 8 MeV neutrons can be expressed as follows:

where N_i is the count of the i-th channel, ε_p,i is the recoil proton energy of the i-th channel, and m is the mass of the stilbene crystal.

Since the stilbene scintillator in NS has a quenching effect, the measured electron equivalent energy (ε_ee) in stilbene was converted to proton energy (ε_p) using the light output function to obtain the recoil proton energy (Hansen et al. 2002).

The equivalent dose (H_stilbene) for neutrons in low Earth orbit, measured by the stilbene sensor, can be obtained by multiplying the stilbene absorbed dose (D_stilbene) by the radiation weighting factor (w_R) according to the energy of the neutron, as suggested in ICRP 60 (ICRP 1991). In ICRP 60, the radiation weighting factor is 10 for neutrons in the range of 2–20 MeV. The equivalent dose in the stilbene in the energy range from 2 to 8 MeV neutrons is as follows:

The circuit configuration of the LEO-DOS system comprises a PD, NS, and a Space Craft Interface (SCIF) electronics, as illustrated in Fig. 4. All components used in the LEO-DOS readout electronics are commercial off-the-shelf parts, screened to withstand a total ionizing dose level of 30 krad.

Fig. 4. Schematic diagram of LEO-DOS signal processing showing a PD readout, NS readout and SCIF part. PSD, pulse shaping discrimination; LEO-DOS, low Earth orbit space radiation dosimeter; PD, particle dosimeter; NS, neutron spectrometer; SCIF, Space Craft InterFace.

Download Original Figure

2.3.2 Neutron Spectrometer (NS) Signal Pulse Processing

A cylindrical 1.5” × 1.5” Stilbene (C₁₄H₁₂) scintillator was optically coupled with a 4 × 4 Silicon photomultipliers (SiPM) array. SiPMs are a solid-state alternative to PMTs for detecting faint scintillation light. Individual SiPMs consist of arrays containing thousands of avalanche photodiode pixels operating in Geiger mode. Although SiPM parts are relatively small (e.g., the SensL J series is available in 6 × 6 mm square formats), arrays of SiPMs with uniform output can cover larger areas.

The bias voltage 27.5 V for the SiPM was supplied from the C14156 (https://www.hamamatsu.com), bias power supply with built-in temperature compensation function. The outputs of the 16 individual SiPMs were passively summed and amplified through an OPA656N(https://www.ti.com) trans-impedance amplifier. The output signals were connected to a pulse shaping discrimination (PSD) circuit, enabling the separation of gamma signals and neutron signals. For the PSD function, the total charge integrator (T), short charge integrator (S), and long charge integrator (L) are implemented in the NS-Charge-Integrator circuits. To quantitatively distinguish neutron and gamma-ray interactions, we define a PSD parameter as the charge ratio:

$$\text{PSD ratio}=\frac{\left(\text{L}-\text{S}\right)}{\text{L}}$$

(8)

The PSD image construction is conducted in the NS-Processor circuit, with the PSD ratio plotted against the total energy, represented on the x-axis. In the PSD image shown in Fig. 5, two distinct bands emerge for scintillators that respond differently to neutrons and gammas. The upper band corresponds to the neutron response, while the lower band corresponds to the gamma response. Often, a figure of merit (FOM) is defined as the distance between the two bands in a particular energy range, divided by the sum of the full-widths at half-maximum (FWHM) of the PSD ratio bands. A high FOM is usually interpreted as a good indicator of how well a certain scintillator device can distinguish between neutrons and gammas. The FOM is then computed by dividing the difference between the centroids by the sum of their FWHM:

Fig. 5. The PSD Image of stilbene NS. The PSD ratio is plotted against the total energy which is represented on the x-axis as channel. The PSD image is stored in the internal SRAM of the NS-MPU, with a resolution of 128 × 128 pixels. The upper band corresponds to the neutron response, while the lower band corresponds to the gamma response. PSD, pulse shaping discrimination; NS, neutron spectrometer.

Download Original Figure

$$\text{FOM}=\frac{x_n-x_{\gamma }}{{\text{FWHM}}_n+{\text{FWHM}}_{\gamma }}$$

(9)

where X_n and X_γ are the peaks of the neutron and gamma-ray bands, respectively, and FWHM_n and FWHM_γ are the full widths at half maximum of the two distributions.

In the NS-AMP circuit, a trigger signal is generated using the level discriminator of the SiPM output signal. The trigger signal creates integration time windows of 10, 0.2, and 0.5 μs for total integration, short integration (S), and long integration (L), respectively. The output of three integrator circuits were digitized using the 12-bit ADC embedded in the MSP432P401R, which serves as the NS-MPU. The PSD image with 128 × 128 pixel was constructed and stored in the MPU’s SRAM. A good FOM of more than 4.2 was obtained with our circuit. The VETO scintillator was also optically coupled with a 24 SiPM array. In the VETO-AMP, the summed SiPM array output is shaped and amplified, and the VETO spectrum is then obtained in the NS-MPU. Meanwhile, before processing the PSD logic in the microprocessor, the presence or absence of a VETO signal is determined. If the VETO signal exceeds the set VETO threshold value, it is considered a charged particle; if it is lower, it is classified as a neutron signal. Due to the difficulty in experimentally determining the VETO threshold value for distinguishing between neutrons and charged particles, it was set to 0.5 MeV based on simulation results (see Section 3).

LEO-DOS aboard NEXTSAT-2 was launched on May 25, 2023, at 6:24 PM. The launch and early orbit phase (LEOP) for LEO-DOS was successfully conducted from June 8 to June 15, and normal observations have been ongoing since June 22. During the LEOP period, NEXTSat-2 maintained a stable altitude and temperature, ensuring the proper operation of LEO-DOS. Throughout the day, the temperature of the LEO-DOS increased from 15°C at initial power-on to 20°C after approximately 4 hours, and then stabilized between 20°C and 30°C, depending on the attitude of NEXTSat-2.

Fig. 6 illustrates the daily variation in the dose measured by LEO-DOS on June 25, 2023, during the LEOP period. The presented data, measured at 120-second intervals, are plotted against UT. The average altitude of the satellite was 556 ± 9.29 km over 15 orbits. It shows 9 passes through the SAA regions, 30 passes through the high-latitude GCR regions of both hemispheres, 30 passes through the geomagnetic equator. The SAA pass time was approximately 9 minutes, and the GCR region pass time was around 20 minutes.

Fig. 6. Daily variation in the dose measured by LEO-DOS on June 25, 2023, during the LEOP period. The figure shows, from top to bottom, the satellite altitude, LEO-DOS temperature, PD absorbed dose, and NS absorbed dose variations throughout the day. LEO-DOS, low Earth orbit space radiation dosimeter; LEOP, launch and early orbit phase; PD, particle dosimeter.

Download Original Figure

The SAA dose rates reached the highest value of 8,445 μGy h^–1 for PD and 2.8 μGy h^–1. The GCR dose rates varied from 0.6 to 15 μGy h^–1 for PD and from 0.002 to 0.04 μGy h^–1 for NS. The daily average GCR dose rates were 4.98 μGy h^–1 for PD, 0.02 μGy h^–1 for NS. This variation in dose according to the orbit showed reasonable agreement with the dose variation measured on the ISS (as shown in Fig. 4 of Dachev et al. 2015), confirming that the observations are consistent and normal.

The geographical distribution of the dose rate for charged particles and neutrons, measured by the LEO-DOS, is shown in the two panels of Fig. 7, where horizontal axis is the longitude, vertical axis is the latitude, and the dose rate intensity is color-coded. Measurements for the three-month period from June 2023 to September 2023 are plotted. In the case of PD, the maximal dose rates are recorded in the region of the SAA from the trapped protons. In the case of NS, most observations in the SAA region were vetoed due to the influence of strongly trapped charged particles, causing the SAA region to appear empty. Occasionally, when a SEP event occurred, traces of SEP were observed along the satellite’s trajectory in the ORB region.

Fig. 7. Dose map for 3 months by LEO-DOS. (a) Absorbed dose map for PD. (b) Absorbed dose map for NS. In the case of NS, the SAA region appears empty due to the VETO effect caused by strongly trapped charged particles. LEO-DOS, low Earth orbit space radiation dosimeter; PD, particle dosimeter; NS, neutron spectrometer.

Download Original Figure

We can now create a detailed global radiation map using LEO-DOS for radiation sources such as GCR, IRB, ORB, and SEP in LEO, including dosimetric parameters such as absorbed dose and equivalent dose for charged particles and neutrons, following the source selection process described in Section 4.3.

The typical radiation sources in the LEO are GCR coming from beyond the solar system, trapped protons in the SAA region of the IRB, trapped electrons from the ORB, and SEP. In this section, we treat the source selection method according to the data selection procedure established by Dachev (Dachev 2009; Dachev et al. 2015).

Fig. 8 characterizes the predominant radiation sources detected by LEO-DOS, using dose rate from flux and dose to flux (D/F) dependencies. The abscissa plots the measured flux in cm^–2 s^–1, while the ordinate shows the dose rate in μGy h^–1 and the D/F ratio in nGy cm^–2 particle^–1 from July to September 2023. The red experimental points along the diagonal represent dose rate values that are linearly dependent on the flux, while the black points lying almost horizontally indicate the D/F ratio. From this figure, when the dose to flux ratio (D/F) exceeds 1.12, it can be characterized as an IRB source, whereas when the D/F ratio is less than 1.12, it can be characterized as an ORB source (Dachev 2009). However, the GCR, IRB, and ORB sources in dose rate < 15 μGy h^–1 region are intermixed each other, making it difficult to distinguish between them. Regions with flux < 4 cm² s^–1 exhibit values of 15 μSv h^–1, and this value was used as the criterion to differentiate between GCR and ORB sources.

Fig. 8. Analysis of the main radiation sources for LEO-DOS relies on the dose rate derived from flux and the dose-to-flux (D/F) ratio dependencies observed from July to September 2023. LEO-DOS, low Earth orbit space radiation dosimeter.

Download Original Figure

Due to the difficulty in categorizing radiation sources based on the D/F and flux relationship, the L-value was used to distinguish the radiation sources. The L-value, L-shell or McILwain L-parameter is a parameter used in space physics to describe the location of a magnetic field line in the Earth’s magnetosphere (Dachev et al. 2017). It corresponds to the radial distance from the Earth’s center to the point where the magnetic field line crosses the magnetic equator, measured in Earth radii, if the magnetospheric magnetic field is a dipole field.

Fig. 9 shows the distribution of the dose rates measured with the LEO-DOS instrument against McIlwain’s L values. Fig. 9(a) represents the SAA and GCR regions, generated during the period from August 17 to August 18, 2023. The SAA is an area where the IRB comes closer to the Earth’s surface due to a displacement of the magnetic dipole axes from the Earth’s center. The IRB is located at altitudes ranging from 0.2 to 2.0 Earth radii at the geomagnetic equator. Within the SAA region, electron energies can reach up to 10 MeV, while proton energies extend up to 700 MeV. The SAA region is characterized by L-values ranging from 1 to 2.5. These findings are in good agreement with the referenced literature (Dachev et al. 2017). In this region, the radiation dose rate of the LEO-DOS is measured at 8,774 μGy h^–1. The maximal particle fluxes at an L-value of approximately 1.24 are recorded as 1,706 particles cm^–2 s^–1. From LEO-DOS data, SAA events are characterized by dose rate > 15 μGy h^–1, L value < 2.5 and D/ F > 1.0 nGy cm^–2 particle^–1.

Fig. 9. Distribution of the dose rates measured with the LEO-DOS instrument against McIlwain’s L values. (a) SAA and GCR region: Galactic cosmic rays (GCR) are seen everywhere and at all times. SAA (L < 2.5) region created by high energy protons when the LEO-DOS crosses the region of the SAA. (b) ORB region is generated by sporadic high-energy electrons seen mainly at 3 < L < 7. (c) SEP region are generated by SEP event of Sun. LEO-DOS, low Earth orbit space radiation dosimeter; SAA, South Atlantic Anomaly; ORB, outer radiation belt.

Download Original Figure

GCR is charged particle that originated from sources beyond the solar system and consists of 98% proton and heavy ions (baryon component), 2% electrons. GCR are seen everywhere and at all times. In regions where L > 4, the energy deposited in the silicon sensor exhibits a spectrum similar to that of GCR (Dachev et al. 2017). The GCR plot in Fig. 9(a) was generated by selecting a period from October 17 to October 30, 2023, during which ORB events were not distinguished. From this plot, we adopt an upper GCR value of D < 15 μGy h^–1 as the selection criterion.

Fig. 9(b) illustrates the ORB region generated during the period from September 21 to September 24, 2023, when ORB events were characterized by very high ORB fluxes and dose rates. Sporadic high-energy electrons are seen mainly at 3 < L < 7 where the ORB is usually situated. It primarily contains electrons with energies extending up to a few MeV. Thus, they are characterized by dose rate D < 15 μGy h^–1, D/F < 1.12 nGy cm² particle^–1 and flux > 4 cm^–2 s^–1. These criteria are essential for understanding and mitigating radiation effects in space environments. The ORB poses challenges for spacecraft and satellite design due to its elevated radiation levels.

Fig. 9(c) depicts the dose variation caused by SEP events during period from October 5 to October 9, 2023. SEPs are primarily produced by solar flares and are associated with intense solar activity. SEPs cause a clear increase in the dose in L > 4 when an event occurs. The ion energies of SEPs typically range between a few MeV and larger per nucleon. They are characterized by dose rate > 15 μGy h^–1, L > 4 and D/F > 1.12 nGy cm² particle^–1. These criteria are crucial for assessing the impact of SEPs on spacecraft, satellites, and astronaut health during space missions. Understanding and mitigating the effects of SEPs is essential for safe space exploration.

Table 4 presents the statistics for radiation sources based on selected criteria from July to September 2023. These include the average dose rate (D_si) (μGy h^–1) and (D_H2O) (μGy h^–1), the average quality factors (Q_av), and the dose equivalent rates (H) (μSv h^–1) for PD, as well as the average dose rate (D) (μGy h^–1) and the equivalent dose rates (H) (μSv h^–1) for NS. Additionally, four SEP events during the same period are also presented.

The average absorbed dose rate (D_si) of GCR measured with LEO-DOS PD during no quiet conditions was 4.72 μGy h^–1, which is higher than the average absorbed dose of GCR measured through Liulin onboard the ISS of 2.78 μGy h^–1 (Dachev et al. 2017). This difference is likely due to the inclination of the ISS, where Liulin is mounted which orbits the Earth with an inclination of 51.6°, while the NEXTSat-2, where the LEO-DOS is mounted, orbits the Earth with an inclination of 97.8°. Due to the higher inclination of the NEXTSat-2 orbit, the LEO-DOS passes through the polar region with a higher L-value, resulting in higher radiation exposure.

The average absorbed dose rates of PD for SEP and IRB sources show a more significant difference from Liulin’s measurement value than that of GCR. The reason for this difference is that the low energy protons dominant in SEP and IRB sources can be effectively shielded by the 10 mm TEP surrounding the PD, while the GCR has very high energy, most of which easily passes the 10 mm TEP. Due to this difference, a significant difference in the dose rates occurs for sources with a high proportion of low energy protons in PD with a 10 mm TEP shield and Liulin with a thin aluminum shield. In the case of the ORB region, most of the electrons below 3 MeV, which make up the majority of the ORB environment, are shielded by the 10 mm TEP of the PD, resulting in a huge difference in dose rates from Liulin’s measurement results.

Radiation protection is crucial for understanding how harmful space radiation doses are to the human body during space travel, especially in LEO. Using the tissue equivalent sensor, LEO-DOS measured the absorbed dose (D), quality factor (Q), and dose equivalent (H) of space radiation for charged particles and neutrons.

The cumulative dose over 30 days for charged particles was approximately 0.164 Sv, significantly lower than the maximum recommended safety level of 1.5 Sv for skin in LEO by the National Council on Radiation Protection (NCRP 2000). These results are consistent with those observed by Dachev et al. (2017) in the ISS radiation environment. For neutrons, the observed value was approximately 1/100th of that for charged particles, indicating a much safer level.

Fig. 10 illustrates the variations in SEP events from August 5 to August 12, 2023. The dose measured by LEO-DOS PD, plotted alongside the solar proton fluxes recorded by GOES-16 in geosynchronous orbit, shows the solar proton fluxes in the upper panel and the dose variation at 120-second intervals in the lower panel. The strong responses that regularly appear in the daily dose rate are associated with peaks from SAA encounters, as shown in Fig. 6. Excluding the SAA region, the dose variations measured by LEO-DOS show good agreement with the >50 MeV proton energy data from GOES-16. This is because the LEO-DOS PD has a self-shielding effect for proton energies below 40 MeV, as shown in Table 1.

Fig. 10. The variations in SEP events from August 5 to August 12, 2023. The upper panel displays the solar proton fluxes measured by GOES-16, while the lower panel presents the dose measured by LEO-DOS PD at 120 s intervals. The recurrent spikes in dose variation, associated to the SAA region as illustrated in Fig. 6. LEO-DOS, low Earth orbit space radiation dosimeter; PD, particle dosimeter; SAA, South Atlantic Anomaly.

Download Original Figure

In the case of NS, the measured absorbed dose (D) in the GCR region was 0.03 μGy h^–1. This value was higher than the simulated value of 0.013 μGy h^–1. In the MCNP simulation in this study, the absorbed dose and equivalent dose of NS were calculated using the neutron spectrum without shielding at 380 km altitude and 51.6° inclination obtained through simulation by Sato et al. (2006). The lower dose values in the MCNP simulation are due to the spectrum used in the simulation corresponding to a lower altitude and inclination compared to the LEO-DOS mission. Due to the high inclination of the LEO-DOS mission, NS passes through a polar region with low vertical cutoff rigidities, which can cause the generation of more albedo neutrons due to the higher GCR intensity (Dachev et al. 2023). Considering these points, it was concluded that the measured values showed good agreement with the values predicted by the simulation.

Radiation protection is crucial for understanding how harmful space radiation doses are to the human body during space travel, especially in LEO. Using the tissue equivalent sensor, LEO-DOS measured the absorbed dose (D), quality factor (Q), and dose equivalent (H) of space radiation.

Based on three months of observations with LEO-DOS, the measured biological effectiveness Q values for different radiation sources were 4.36 ± 0.27 for GCR and 1.73 ± 0.25 for SAA. Considering the detector geometry and the satellite’s inclination of 97.8°, these values are reasonably similar to those obtained by Dachev et al. (2017), which were 3.5 for GCR and 1.3 for SAA. Meanwhile, for the PD of LEO-DOS using a 10 mm TEP, the Q value for SEP was 2.82 ± 0.65, which is similar to the value of 2.5 measured on the ISS using a 40 mm TEP (Semkova et al. 2014).