Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Calibration of RADMON Radiation Monitor Onboard Aalto-1 CubeSat

Calibration of RADMON Radiation Monitor Onboard Aalto-1 CubeSat RADMON is a small radiation monitor designed and assembled by students of University of Turku and University of Helsinki. It is own on-board Aalto- 1, a 3-unit CubeSat in low Earth orbit at about 500 km altitude. The detec- tor unit of the instrument consists of two detectors, a Si solid-state detector and a CsI(Tl) scintillator, and utilizes the DE-E technique to determine the total energy and species of each particle hitting the detector. We present the results of the on-ground and in- ight calibration campaigns of the in- strument, as well as the characterization of its response through extensive simulations within the Geant4 framework. The overall energy calibration Email addresses: philipp.oleynik@utu.fi (Philipp Oleynik ), rami.vainio@utu.fi (Rami Vainio ), arjupu@utu.fi (Arttu Punkkinen), dudnik@rian.kharkov.ua (Oleksiy Dudnik), jan.gieseler@utu.fi (Jan Gieseler), hannu-pekka.hedman@utu.fi (Hannu-Pekka Hedman), heli.hietala@utu.fi (Heli Hietala), edward.haeggstrom@helsinki.fi (Edward Hggstr om), petri.niemela@aalto.fi (Petri Niemel a), juhpe@utu.fi (Juhani Peltonen), jaan.praks@aalto.fi (Jaan Praks), rpunk@utu.fi (Risto Punkkinen), teansa@utu.fi (Tero S antti), eino.valtonen@utu.fi (Eino Valtonen) Accepted to Advances in Space Research November 19, 2019 arXiv:1911.07550v1 [astro-ph.IM] 18 Nov 2019 margin achieved is about 5%. The full instrument response to protons and electrons is presented and the issue of proton contamination of the electron channels is quanti ed and discussed. Keywords: Radiation belts, Electron precipitation, Solar energetic particles, CubeSats 1. Introduction RADMON (Peltonen et al., 2014; Kestil a et al., 2013) is a small radia- tion monitor on-board the 3-unit Aalto-1 CubeSat (Kestil a et al., 2013). The satellite was launched by a PSLV-C38 rocket from India on 23 June 2017 into the low Earth orbit with an inclination of 97 degrees and an average altitude of 505 km (Praks et al., 2018). Aalto-1 is Finland's rst national satellite mission and it is designed, assembled and operated by students at Aalto Uni- versity, Espoo. The RADMON experiment was designed and assembled by students, in the University of Turku and University of Helsinki. Here we will describe the RADMON detector unit and characterize the response of the detector to the energetic particle radiation it measures, including the calibra- tion of the instrument. First scienti c results of the RADMON experiment are presented by Gieseler et al. (2019). 2. RADMON Instrument RADMON (Peltonen et al., 2014) consist of four subsystems: the Detector Unit, the Analog Electronics Board, the Digital Electronics Board and the Power Supply Board stacked in a compact con guration with a volume of 0.4 units (Fig. 1), a mass of 360 g, and a power consumption of 1 W. The detector signals are ampli ed and continuously digitized at 10 MHz sampling rate on the Analog Electronics Board and then transmitted through the instrument bus connecting the boards to the Digital Electronics Board, where a Xilinx Virtex-4 LX15 eld programmable gate array (FPGA) handles the signal processing from the pulse detection and pulse height determination to the classi cation and counting of particle events in spectral channels of 2 Figure 1: RADMON radiation monitor on-board Aalto-1. The instrument consists of a stack of three printed circuit boards and a detector unit in a brass housing, and an aluminum frame on the top. A square aluminum entrance window in front covers the detector unit inside the brass container. the instrument. The details of the RADMON electronics are presented by Peltonen et al. (2014). 2.1. Instrument geometry The detector unit of the instrument consists of a 350-mm-thick silicon de- 2 3 tector with an active area of 2.12.1 mm and a 101010 mm CsI(Tl) scintillation detector acting as a calorimeter. The scintillator is coupled to a 1010 mm p-i-n photodoide for optical photon readout. The scintillator crystal cube has ve sides wrapped with a white PTFE lm which improves light collection on the photodiode. Signal processing circuits of the detec- tors are independent; they produce voltage pulses, which are then digitized by corresponding analog-to-digital converters (ADC) (Peltonen et al., 2014). The detectors are housed in a brass container with walls thick enough to stop protons below 50 MeV and electrons below 8 MeV. The brass container is xed to an aluminum frame, which also carries the electronics of the instru- ment. 3 incident particle Figure 2: A cross section of the detector unit. Topmost layer is an aluminum entrance window, a brass container is light-brown, the silicon detector is orange, its passive 3.53.5 mm area is gray, and its supporting structure is dark-green. The CsI(Tl) scintillator cube is light-green, below which there is the Hamamatsu S3590-08 silicon photodiode with a depletion layer of 300 mm. The frontal opening in the container collimates incoming particles to a solid angle around =5 and is covered by a 280 mm thick aluminum entrance window (Fig. 2). The entrance window is opaque for electrons with energies below 0.24 MeV and for protons with energies below 6.5 MeV. An incident particle rst passes the aluminum entrance window, depositing some energy there, which we account for during the calibration, then the particle deposits some energy in the silicon detector and the rest is likely to be deposited in the scintillator. For most protons with energies 10 { 50 MeV the scintillator acts as a calorimeter, so that the particle energy is high enough to penetrate the silicon detector, but low enough to be absorbed in the scintillator. Con- sidering such protons one can write an equation based on the dependency of the particle range, R [g cm ], in medium on energy: R (E ) = R (E ) + d  ; (1) Si 0 Si CsI Si Si where E is the incident particle energy when it enters the silicon detector, E is the energy deposited in CsI scintillator, d = 350 mm is the thickness CsI Si of the silicon detector, and  = 2:33 g cm is the silicon density. E , in this Si 0 case, is the sum of E and E , energy deposited in silicon. This equation CsI Si 4 could be used to determine a set of points on the [E E ] plane which Si CsI can be used both for particle discrimination and instrument calibration. Assuming that the energy of the particle is fully absorbed in the instru- ment and the range function is reversible, the equation (1) can be solved for energy deposited in silicon E vs incident particle energy E . The range{ Si 0 energy relation is commonly approximated by the Bragg-Kleeman rule as R (E) = A  E , where the incident particle energy is normalized to a Si 0 certain value which is 1 MeV in the present work. However, this approxima- tion di ers slightly from the measured range at lower energies (Berger et al., 1992). We have chosen an approximation which gives less than 2% error in the 2 { 200 MeV energy range, given by (Attix, 2008) R (E) = + : (2) Si 1 MeV The best t with the PSTAR (Berger et al., 1992) proton range data yields 3 2 3 2 = 1:76, = 12:5 10 g cm , and = 4:3 10 g cm . Since the energy{ range relation is reversible, one can obtain a solution of the equation (1) for Si 1= E d 0 Si Si E = E MeV : (3) Si 0 1 MeV Using the assumption E + E = E it is possible to describe the curve on Si CsI 0 the E E plane by the equation Si CsI 1= E d CsI Si Si E = + MeV E : (4) Si CsI 1 MeV The equation (4) de nes a so-called "banana" curve with parameters xed by the instrument geometry. It is independent on E because each incident energy value is represented on the curve by a dot, therefore a continuous energy spectrum of incoming protons produces a continuous curve, with a shape determined only by the detector geometry. The true incoming particle energy di ers from E by several MeV of energy absorbed in the aluminum entrance window. The shape of the curve is independent on E , but the value of the energy threshold is a ected by the energy absorbed in the entrance 5 window. A zero angle of incidence is assumed between the proton momen- tum direction and the normal to the silicon detector plane. For an angle of incidence  deviating from zero the e ective silicon thickness is greater by a factor of 1= cos . In space, protons originate from all directions and the brass collimator restricts incident directions to a  20 half-width cone for protons with energies below 50 MeV. This restriction applied to the factor of 1= cos  alters the e ective thickness of the silicon detector by  6% which results in a slight blurring of the curve provided by (4) towards higher E , Si since particles deposit more energy in the e ectively thicker silicon detector, see gure 3. 2.2. Particle counters A particle must hit both detectors to be registered. The detection logic rejects any single detector hits or events that have deposited energy below a threshold. This logic eliminates single hits coming from side-penetrating particles or bremsstrahlung X-rays from electrons scattered in the brass con- tainer or spacecraft structures. In the present simulations, we consider the detection of an incident particle and secondary ones it might produce as a separate event. There is a nite chance that a primary or secondary parti- cle creates the coincidence condition by hitting, e.g., the scintillator within a temporal coincidence window with another particle that has deposited energy in the silicon detector, or vice versa. Taking into account observed detection rates and the coincidence window interval we nd this e ect negligible. The discrimination of incoming particles to protons and electrons is ac- complished using the E{E technique (Goulding et al., 1964; Birks, 1964b; Shimoda et al., 1979) applied to digitized pulse heights of the detector sig- nals. A linear combination of these digitized pulse heights in both detectors is used as a measure of the total deposited energy E . Particle counters are accumulated for 15 seconds and then stored with a time stamp. After the current counter values are stored, all counters are reset to zero. The aperture of the instrument rotates with the satellite and scans over all pitch angles evenly since 15 seconds cover a few rotation periods. Protons are separated from electrons using their locus on the E E Si CsI 6 plane. There are two regions on the plane which follow the "banana" curve on the (Fig. 3) where a detected particle is counted as a proton. Protons are further classi ed into nine energy channels p1 { p9 by a sequence of thresholds for the E . Each channel is a counter which is incremented when conditions for the E are met. Energies deposited by electrons do not produce a distinct curve on the plane due to substantial scattering in both detectors, but they are con ned by a certain region with borders quite far from the proton curve. One can estimate the energy deposited by an electron as a sum of deposited energies in both detectors. The detected electrons are counted in ve electron channels e1 { e5 depending on the total deposited energy. Any particle detection which cannot be classi ed as a detection of an electron or a proton increments one of the two outlier counters, o1 or o2, below and above the proton track, respectively. For each channel and deposited energy it is possible to describe a de- tection probability density function p = p(E ; channel), which we fur- channel 0 ther denote as instrument response. The instrument response is obtained by Monte-Carlo simulations and is described in section 4.1. 2.3. Pulse height data Raw pulse height data is also available for calibration purposes and can be downlinked during calibration campaigns. The amount of data is sub- stantially higher during these observations, so it was necessary to keep the campaigns short for the sake of continuous observations in the normal particle counting mode, yet long enough to enable in- ight calibration in real space environment, which was impossible to reproduce on ground-based facilities available to the team. The available pulse height data covers di erent regions such as high- latitude belts where particles from higher L-shells can be observed, quiet equatorial regions, and the South Atlantic geomagnetic anomaly, where the proton contamination dominates over the electron ux in electron channels e2 { e5. See Gieseler et al. (2019) for a description of the measured radiation environment. 7 2.4. On-ground calibration The instrument was originally calibrated on the ground using an acceler- ator beam at Abo Akademi University. The beam energy was kept constant (17 MeV) during the calibration. The instrument was placed into a par- allel proton beam arriving straight to the aperture through exchangeable aluminum sheets of di erent thickness in order to control the incident proton energy for the instrument. The calibration setup was modeled within the Geant4 framework (Agostinelli et al., 2003; Allison et al., 2006; Allison et al., 2016) in order to estimate en- ergies deposited in both detectors of the instrument. Each detector was considered to have a linear response function, so its digital signal N ADC could be characterized by a pair of parameters N = a  E + b LSB ADC det (least signi cant bits), where a is the overall gain and b is an o set. These two parameters were calculated using the least squares method applied to measurements. The detector calibration results are presented in table 1. Table 1: Gains and o sets for the RADMON detectors obtained during the on-ground and in- ight calibration campaigns. Values without errors in the in- ight calibration were kept xed. a [LSB/MeV] b [LSB] a [LSB/MeV] b [LSB] Si Si CsI CsI on-ground: 140 6 3 15 18:3 0:5 9 5 in- ight: 140 3 14:6 0:3 0 First calibration data obtained in space right after the launch showed that the gain previously calculated for the CsI scintillation detector deviated from the present gain value of the detector by up to 20%. One possible reason for that is degradation of the optical contact between the scintillator crystal and the photodiode readout, since the change was detected right in the beginning of orbital measurements. This fact called for an in- ight calibration campaign to be performed using protons observed in low Earth orbit. The campaign consisted of two runs, the second one was preformed with an elevated energy detection threshold in order to keep electron counts low and reserve the telemetry for proton counts. The in- ight calibration 8 campaign data is analyzed in the present work. 3. Simulations 3.1. Proton track calibration As it was mentioned in section 2.1, protons with energies of 10 { 50 MeV form a distinct non-linear feature on the E E plane. The posi- Si CsI tion and the curvature of that proton track are de ned only by the detector dimensions and materials. The linearity of the readout electronics is well established during on-ground tests, making this track a perfect tool to cali- brate gains and o sets of the RADMON detectors in space. During several calibration intervals in orbit, protons and electrons with continuous spectra were observed and the pulse height data were delivered to ground. The data points were analyzed on the [ADC { ADC ] plane. (Si) (CsI) Since protons in space are not intrinsically collimated, there is a noticeable background above and below the proton "banana". We have processed the data using the HDBSCAN algorithm (McInnes et al., 2017), which detects clusters in noisy data. 568 data points were selected for calibration. In order to determine proper gains and o sets using these data one needs an analytical description of the observed curve in ADC units. The readout is linear, but there are non-linear processes in the scintillator and its readout photodiode. Scintillation processes in CsI(Tl) are thoroughly studied (Gwin and Mur- ray, 1963a,b). The scintillation has two components with di erent decay times (Benrachi et al., 1989), with their ratio depending on speci c energy loss dE=dx. Thus, it is particle species dependent. Moreover, the light out- put is quenched by recombination of electron-hole pairs in the scintillator medium (Birks e ect) (Birks, 1964a). The e ect is negligible for electrons, but plays an important role in proton detection. It can be described in the di erential form as dL S(dE=dx) = ; (5) dx 1 + kB(dE=dx) where L is the light output, E is the particle energy, S is a normalization constant, and kB is a measure of the Birks e ect in uence on the light 9 output. This equation must be integrated in order to derive L(E) function, which then is used to get a relation between the digitized pulse height and the particle energy as seen by the scintillator ADC (E). This function has Si scintillation detector gain and o sets as parameters to be calibrated. Horn et al. (1992) asserted that dE=dx  1=E for the sake of analytical integration of the equation 5 but, as was pointed by Avdeichikov et al. (2000), this approximation is substantially limited. We have approximated dE=dx for protons in the energy range relevant to the RADMON calibration as dE=dx E with = 0:678, which ts well the experimental data (Berger et al., 1992). Integration of the equation (5) gives the L(E ) function expressed CsI through a hyper-geometric function F 2 1 1 1 E CsI L(E )  E  1 F (1; ; + 1; ) : (6) CsI CsI 2 1 kB a Finally, we obtain 1 1 E CsI ADC = a  E  1 F (1; ; + 1; ) + b : (7) (CsI) CsI CsI 2 1 CsI kB a This function was used to t the analytical description of the proton track to the chosen experimental points. The kB constant was adopted from Avde- ichikov et al. (2000). For the silicon detector a linear function was used ADC = a  E + b ; (8) (Si) Si Si Si where gain a and o set b were initialized with the values measured on Si Si ground. The t was done by least-squares method; the shortest distance to the curve from each experimental point was taken as the error estimator. 3.2. Geant4 model A complete response function for each particle channel was obtained from simulations carried out within the Geant4 framework (Agostinelli et al., 2003; Allison et al., 2006; Allison et al., 2016). The whole Aalto-1 satellite with the RADMON instrument was modeled as a realistic 3D model described by a 10 GDML (Chytracek et al., 2006) script. All parts of the RADMON instrument were modeled exactly as they are in reality. The rest of the satellite was partially simpli ed by replacing ne details of sub-systems like the on-board computer and batteries with an aluminum foam lling the space inside the 3U CubeSat frame. The foam density was calculated from a previously measured mass of the spacecraft subsystems. The model was placed into a cubic Geant4 world with dimensions of 300300300 cm . A particle source used in the model was a sphere with a radius of 30 cm, which fully enclosed the Aalto-1 satellite model. Each par- ticle was rst placed into a uniformly random position on the sphere, with U (0; 1) and cos   U (0; 1). The particle momentum direction was cho- sen according to the uniform Lambertian angular distribution (Greenwood, 2002). During a simulation run, the particle energy remained constant. In order to cover the necessary energy range, we used a quasi-logarithmic grid of energies. The sensitive volumes in the model were the 2.12.1 mm silicon detector, the 101010 mm CsI(Tl) scintillation crystal, and the p-i-n photodiode, which was modeled as a 300 mm thick 1010 mm silicon detector in order to take direct particle energy deposits into account. The photodiode is normally lit by scintillation light, but in some cases, it can be hit directly by a particle. Such hits are indistinguishable from light pulses from the scintillator due to moderately high integration time, but their contribution to the response is quite small since the photodiode is installed at the back of the instrument aperture. Direct hits ionize the photodiode at proton energies from about 55 MeV and make no special features in the instrument response functions. The modeling software recorded deposited energies for each particle in each volume. We summed energy deposits from primary particles and sec- ondary ones, exactly the way the real instrument does. All physical inter- actions possible at the simulated energy range were enabled in the software. The detection was counted only if the energy deposit was above 70 keV in both detectors. The threshold of 70 keV imitates the noise oor of the instru- ment. Direct hits in the photodiode were recorded as an auxiliary component to be added to the scintillator output following Bird et al. (1994). All single 11 hits were discarded, since the instrument detection logic behaves the same way in hardware. We used gains and o sets obtained from the calibration to provide the in- strument responses by simulating the RADMON detection logic and particle channel classi ers. Each detection record collected from the Geant4 model was rst converted from energy units to the ADC units. Then the simulated detection was examined whether it must be classi ed as an electron or a proton, or it must be counted in one of the outlier channels. The data for simulated proton detections was utilized to quantify the proton contamina- tion of electron channels. Energy channel classi cation was performed by an algorithm implemented in RADMON based on analysis of the parameter E = (10  ADC + 48 (Si) ADC )=64 . The division in this formula is the standard integer division, (CsI) since the RADMON does all arithmetics in the integer domain. The linear combination, based on in- ight calibration, overestimates the gain of the scintillator but the e ect is taken into in the calculated energy response of each energy channel. 3.3. Bowtie analysis The RADMON particle channels were selected and combined in order to get several integral channels which data could be directly converted to ux without complex procedures involving response curves. We have calculated geometric factors and threshold energies for the following channels using a Van Allen "bowtie" analysis described by e.g. Sorensen et al. (2005). During the simulations the incident particle energy was set by a quasi- logarithmic grid with 48 energies for each decade. Geometric factors were calculated for each energy bin using the expression N (channel; E ) d bin G(channel; E ) = A ; (9) bin r N (E ) s bin where N (channel; E ) is a number of particles detected in an instrument d bin particle channel, N (E )is a number of particles shot in a particular energy s bin bin, and A is an area of radiating sphere. All geometric factors for available 12 energy bins were combined to make a discrete function G(E) for each instru- ment channel. For a combination of channels a corresponding combination of geometric factors was calculated. These geometric factors were used for an integral bowtie analysis accord- ing to the formula f (E)G(E)dE G = ; (10) I 1 f (E)dE where f (E) is a modeled di erential ux given as a power-law with a range of indices and E is a threshold energy for the analysis. If a channel can be characterized as an integral one, there is a speci c value E of the energy E 0 t for which G is the same for a wide range of power-law indices of f (E) (see gure 5). A mean value and a con dence interval for G are calculated through statistical analysis of a distribution of G values obtained for di erent power- law indices, and E is evaluated as a middle of an energy band where standard deviation of G does not exceed three times its minimal value. A con dence interval for E is this energy band width. For a nice integral channel its con dence interval is practically the width of an energy bin where E lies. An integral ux is then de ned by F (E > E ) = ; (11) where R is a count rate in the channel for which G and E are de ned. I 0 4. Results 4.1. Response of the particle channels We have obtained curves for geometric factors as a function of incident particle energy for each channel (Figure 6 and 7). The electron channels e2 { e5 are integral channels. They have long tails stretching up to high energies, yet each channel shows a clear threshold en- ergy. The response curves allow calculation of the integral threshold energies and corresponding geometric factors for these channels. 13 We have converted proton channels from di erential ones to four integral channels and one di erential in high energy range, in order to obtain a set of well-de ned energy ranges to be quoted for the channels. The measurement channels p1..p4 have responses similar to boxcar functions in the nominal energy range, but they also have relatively strong high-energy side bands, which limits their use as di erential channels. At 50 MeV and above the brass collimator becomes transparent to protons, so they start to be regis- tered also in the rst proton channels. Protons of energies higher than 100 MeV penetrate the whole satellite. They deposit little energy both in the scintillator and the silicon detector. Thus, they miss the proton track and get into the outlier channel o1 or contaminate electron channels. 4.2. Angular sensitivity The angular sensitivity plot (Figure 8) shows how the RADMON aperture gradually expands when energy increases beyond 50 MeV. This widening of a sensitive aperture is also seen on the response curves. Protons with energies slightly above 100 MeV are much less likely capable to get to the proton channels of the instrument, whereas protons of higher energies could be detected virtually anywhere on the [ADC { ADC ] plane. (Si) (CsI) 4.3. Contamination issues During the mission high count rates in electron channel have been ob- served while the spacecraft passed the South Atlantic Anomaly. Here we characterize the proton contamination of the electron channels of the RAD- MON to avoid misinterpretation of the observations. The contamination geometric factors are presented in Figure 9. The highest contamination ex- ists in channels e4 and e5. These responses fully explain the counting rates in the electron channels e3{e5 observed inside South Atlantic Anomaly. 4.4. Channel e1 issue The electron channel e1 was initially planned to measure electron ux from about 1 MeV energies. However, RADMON scintillator channel picks up noise from the environment, which led to the necessity of increasing the 14 detection threshold of the channel to values higher than foreseen, in order to avoid contaminating the measurements. During the mission we have observed very few counts in e1 and identi ed the reason to be a very narrow energy region of remaining non-zero sensitivity in e1. In addition, this narrow band is highly sensitive to pulse detection eciency near the threshold and drift of parameters of the RADMON signal processing pipeline, whereas the rest of the channels are stable due to the larger detector pulse heights. Therefore, the channel e1 has to be discarded from the scienti c dataset. 4.5. E ective geometric factors and energies of the channels The bowtie analysis provided integral geometric factors for instrument electron channels e2 { e4 and composite proton channels. All channels are integral: the e2 { e4 channels have long tails on the high energy side, and the i1 { i4 channels are composite ones combined from the instrument proton channels p1 { p9 in such a way that they have step-like response curves. They are presented in table 2. Table 2: Bowtie cuto energies and geometric factors for integral particle channels. e2 { e4 are electron integral channels and i1 { i4 are proton integral channels. The con dence intervals are at a level of 95%. Channel Cuto energy [MeV] Geometric factor [cm sr] e2 1:51 0:1 0:0108 0:0005 e3 3:1 0:2 0:0160 0:0005 e4 6:0 0:7 0:0119 0:0008 i1=(p1 : : : p9) 10:4 0:3 0:0228 0:0004 i2=(p2 : : : p9) 18:5 0:7 0:0256 0:0009 i3=(p3 : : : p9) 23:7 1:8 0:0219 0:0011 i4=(p4 : : : p9) 29 4 0:0187 0:0014 We have a di erential proton channel for high energy protons, i5 = (p5 : : : p9), it has sensitivity from 40 to 80 MeV with a di erential geo- metric factor G E = 0:78 0:09 cm sr MeV at an energy of 42 5 MeV. i5 15 5. Summary and Conclusions A realistic 3D model of the Aalto-1 satellite with the RADMON radiation monitor was constructed in a Geant4 simulation framework. Structures of the satellite and the instrument were described by a GDML script. The virtual model was placed into omnidirectional monoenergetic ux of protons and electrons. The energies of simulated particles covers the RADMON sensitiv- ity range, also extending to higher energies in order to study contamination issues. We have calculated energy response curves for protons and electrons for all instrument channels in a wide energy range. These responses include ge- ometric factors for electron channels contaminated by high energy protons. We have constructed four integral proton channels from the instrument chan- nels p1 { p9 and one di erential channel sensitive for protons of 40 { 80 MeV. The e2 { e4 electron channels of the RADMON instrument are integral ones. The obtained results allow conversion of count rates in the individual channels of the RADMON instrument to isotropic ux measurements in low Earth orbit. The data description will be published in a separate paper Gieseler et al. (2019). Acknowledgements. This work was performed in the framework of the Finnish Centre of Excellence in Research of Sustainable Space (FORESAIL) funded by the Academy of Finland (grants 312357 and 312358). We also gratefully acknowledge the e orts of dozens of students in Aalto University, University of Turku and University of Helsinki for their work in the Aalto-1 satellite and RADMON projects. Aalto University MIDE is thanked for nancial support for building Aalto-1. Aalto University, University of Turku, RUAG, Space Systems Finland, and Nokia sponsored the launch of the satellite. Compu- tations necessary for the presented modeling were conducted on the Pleione cluster at the University of Turku. 16 Figure 3: A simulated "banana" curve for protons with energies from 8 to 100 MeV. Color of the dots in upper panel denotes the incident proton energy. The red curve is the one de ned by the equation (4) where E is shifted by 0.15 MeV to correct for a 2% error of CsI the analytical range approximation at lower energies. The lower panel of the plot shows a residual between the analytical curve and Geant4 simulations. 8 MeV is close to the detection threshold, so there are few particles on the leftmost part of the curve compared to the central part. Stripe pattern on the lower panel appears due to simulations are carried out for monoenergetic particles on a discrete energy grid. The sharp curve change at 50 MeV is caused by CsI starting to be transparent, so that the assumption for equation 1 is no longer true. Protons leave less and less energy in CsI at incident energies of 55 MeV and above. 17 Figure 4: Proton "banana" curve t using calibration data from orbit. The gain value obtained for the scintillation detector is 68:5  1:2 keV/LSB. Shaded polygons are the regions where incoming particles are classi ed as protons or electrons. There are two populations of particles in the electron region below. They originate from two calibration campaigns, one of which had higher detection threshold in order to deliver more data covering the proton track. 18 Figure 5: Bowtie analysis for the e3 instrument channel. E = 3:08 0:06 MeV. Curves show G for di erent power-law indices of f (E) in a range of [2 : : : 7]. Figure 6: RADMON response functions for electron channels e1 { e5. Electron channels from e2 to e4 are integral channels with a distinct threshold. The black curve above individual response curves shows the integral sensitivity to electrons. 19 Figure 7: RADMON response functions for combined integral proton channels and a di erential one, which is also a combined channel p5..p9. Figure 8: RADMON angular sensitivity for protons in proton channels. 20 Figure 9: High energy proton contamination of electron channels e2 { e5. 21 References Agostinelli, S., Allison, J., Amako, K., Apostolakis, J., Araujo, H., Arce, P., Asai, M., Axen, D., Banerjee, S., Barrand, G., Behner, F., Bellagamba, L., Boudreau, J., Broglia, L., Brunengo, A., Burkhardt, H., Chauvie, S., Chuma, J., Chytracek, R., Cooperman, G., Cosmo, G., Degtyarenko, P., Dell'Acqua, A., Depaola, G., Dietrich, D., Enami, R., Feliciello, A., Fer- guson, C., Fesefeldt, H., Folger, G., Foppiano, F., Forti, A., Garelli, S., Giani, S., Giannitrapani, R., Gibin, D., Cadenas, J.G., Gonz alez, I., Abril, G.G., Greeniaus, G., Greiner, W., Grichine, V., Grossheim, A., Guatelli, S., Gumplinger, P., Hamatsu, R., Hashimoto, K., Hasui, H., Heikkinen, A., Howard, A., Ivanchenko, V., Johnson, A., Jones, F., Kallenbach, J., Kanaya, N., Kawabata, M., Kawabata, Y., Kawaguti, M., Kelner, S., Kent, P., Kimura, A., Kodama, T., Kokoulin, R., Kossov, M., Kurashige, H., Lamanna, E., Lamp en, T., Lara, V., Lefebure, V., Lei, F., Liendl, M., Lockman, W., Longo, F., Magni, S., Maire, M., Medernach, E., Minami- moto, K., de Freitas, P.M., Morita, Y., Murakami, K., Nagamatu, M., Nartallo, R., Nieminen, P., Nishimura, T., Ohtsubo, K., Okamura, M., O'Neale, S., Oohata, Y., Paech, K., Perl, J., Pfei er, A., Pia, M., Ranjard, F., Rybin, A., Sadilov, S., Salvo, E.D., Santin, G., Sasaki, T., Savvas, N., Sawada, Y., Scherer, S., Sei, S., Sirotenko, V., Smith, D., Starkov, N., Stoecker, H., Sulkimo, J., Takahata, M., Tanaka, S., Tcherniaev, E., Tehrani, E.S., Tropeano, M., Truscott, P., Uno, H., Urban, L., Urban, P., Verderi, M., Walkden, A., Wander, W., Weber, H., Wellisch, J., Wenaus, T., Williams, D., Wright, D., Yamada, T., Yoshida, H., Zschiesche, D., 2003. Geant4 { a simulation toolkit. Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment 506, 250 { 303. URL: http://www.sciencedirect. com/science/article/pii/S0168900203013688, doi:https://doi.org/ 10.1016/S0168-9002(03)01368-8. Allison, J., Amako, K., Apostolakis, J., Araujo, H., Arce Dubois, P., Asai, M., Barrand, G., Capra, R., Chauvie, S., Chytracek, R., Cirrone, G.A.P., Cooperman, G., Cosmo, G., Cuttone, G., Daquino, G.G., Donszelmann, 22 M., Dressel, M., Folger, G., Foppiano, F., Generowicz, J., Grichine, V., Guatelli, S., Gumplinger, P., Heikkinen, A., Hrivnacova, I., Howard, A., Incerti, S., Ivanchenko, V., Johnson, T., Jones, F., Koi, T., Kokoulin, R., Kossov, M., Kurashige, H., Lara, V., Larsson, S., Lei, F., Link, O., Longo, F., Maire, M., Mantero, A., Mascialino, B., McLaren, I., Mendez Lorenzo, P., Minamimoto, K., Murakami, K., Nieminen, P., Pandola, L., Parlati, S., Peralta, L., Perl, J., Pfei er, A., Pia, M.G., Ribon, A., Rodrigues, P., Russo, G., Sadilov, S., Santin, G., Sasaki, T., Smith, D., Starkov, N., Tanaka, S., Tcherniaev, E., Tome, B., Trindade, A., Truscott, P., Urban, L., Verderi, M., Walkden, A., Wellisch, J.P., Williams, D.C., Wright, D., Yoshida, H., 2006. Geant4 developments and applications. IEEE Transac- tions on Nuclear Science 53, 270{278. doi:10.1109/TNS.2006.869826. Allison, J., Amako, K., Apostolakis, J., Arce, P., Asai, M., Aso, T., Bagli, E., Bagulya, A., Banerjee, S., Barrand, G., Beck, B., Bogdanov, A., Brandt, D., Brown, J., Burkhardt, H., Canal, P., Cano-Ott, D., Chauvie, S., Cho, K., Cirrone, G., Cooperman, G., Cort es-Giraldo, M., Cosmo, G., Cuttone, G., Depaola, G., Desorgher, L., Dong, X., Dotti, A., Elvira, V., Folger, G., Francis, Z., Galoyan, A., Garnier, L., Gayer, M., Genser, K., Grichine, V., Guatelli, S., Gu eye, P., Gumplinger, P., Howard, A., H rivn a cov a, I., Hwang, S., Incerti, S., Ivanchenko, A., Ivanchenko, V., Jones, F., Jun, S., Kaitaniemi, P., Karakatsanis, N., Karamitros, M., Kelsey, M., Kimura, A., Koi, T., Kurashige, H., Lechner, A., Lee, S., Longo, F., Maire, M., Man- cusi, D., Mantero, A., Mendoza, E., Morgan, B., Murakami, K., Nikitina, T., Pandola, L., Paprocki, P., Perl, J., Petrovi c, I., Pia, M., Pokorski, W., Quesada, J., Raine, M., Reis, M., Ribon, A., Fira, A.R., Romano, F., Russo, G., Santin, G., Sasaki, T., Sawkey, D., Shin, J., Strakovsky, I., Taborda, A., Tanaka, S., Tom e, B., Toshito, T., Tran, H., Truscott, P., Urban, L., Uzhinsky, V., Verbeke, J., Verderi, M., Wendt, B., Wen- zel, H., Wright, D., Wright, D., Yamashita, T., Yarba, J., Yoshida, H., 2016. Recent developments in Geant4. Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment 835, 186 { 225. URL: http://www.sciencedirect. 23 com/science/article/pii/S0168900216306957, doi:https://doi.org/ 10.1016/j.nima.2016.06.125. Attix, F., 2008. Introduction to Radiological Physics and Radiation Dosime- try. Wiley-VCH Verlag GmbH & Co. Avdeichikov, V., Fomichev, A.S., Jakobsson, B., Rodin, A.M., Ter-Akopian, G.M., 2000. Range energy relation, range straggling and response function of CsI(Tl), BGO and GSO(Ce) scintillators for light ions 439, 158{166. Benrachi, F., Chambon, B., Cheynis, B., Drain, D., Pastor, C., Seghier, D., Zaid, K., Giorni, A., Heuer, D., Ll eres, A., Morand, C., Stassi, P., Viano, J., 1989. Investigation of the performance of CsI(Tl) for charged particle identi cation by pulse-shape analysis. Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment 281, 137 { 142. URL: http://www.sciencedirect. com/science/article/pii/0168900289912254, doi:https://doi.org/ 10.1016/0168-9002(89)91225-4. Berger, M.J., Coursey, J.S., Zucker, M.A., Chang, J., 1992. Stopping-Power & Range Tables for Electrons, Protons, and Helium Ions Share. NIST Stan- dard Reference Database 124. Technical Report NISTIR 4999. National Institute of Standards and Technology (NIST), Physics Laboratory (now Physical Measurement Laboratory (PML)). Bird, A., Carter, T., He, Z., Dean, A., Ramsden, D., 1994. Pulse shape analysis of signals from a CsI(T1)/photodiode detector. Nucl. Instruments Methods Phys. Res. Sect. A Accel. Spectrometers, Detect. Assoc. Equip. 353, 46{49. URL: https://www.sciencedirect.com/science/article/ pii/0168900294915997, doi:10.1016/0168-9002(94)91599-7. Birks, J.B., 1964a. The Scintillation Process in Alkali Halide Crystals. IEEE Trans. Nucl. Sci. 11, 4{11. doi:10.1109/TNS.1964.4323396. Birks, J.B., 1964b. The theory and practice of scintillation counting. Perga- mon, Oxford. 24 Chytracek, R., Mccormick, J., Pokorski, W., Santin, G., 2006. Geom- etry description markup language for physics simulation and analysis applications. IEEE Transactions on Nuclear Science 53, 2892{2896. doi:10.1109/TNS.2006.881062. Gieseler, J., Oleynik, P., Hietala, H., Vainio, R., Hedman, H.P., Peltonen, J., Punkkinen, A., Punkkinen, R., S antti, T., Hggstr om, E., Praks, J., Niemel a, P., Riwanto, B., Jovanovic, N., Mughal, M.R., 2019. Radiation monitor radmon aboard aalto-1 cubesat: First results. Submitted to Ad- vances in Space Research . Goulding, F., Landis, D., Cerny, J., Pehl, R., 1964. A new particle identi er technique for Z = 1 and Z = 2 particles in the energy range > 10 MeV. Nucl. Instruments Methods 31, 1{12. URL: https://www.sciencedirect.com/ science/article/pii/0029554X64903131, doi:10.1016/0029-554X(64) 90313-1. Greenwood, J., 2002. The correct and incorrect generation of a cosine distribution of scattered particles for Monte-Carlo modelling of vacuum systems. Vacuum 67, 217 { 222. URL: http://www.sciencedirect. com/science/article/pii/S0042207X02001732, doi:https://doi.org/ 10.1016/S0042-207X(02)00173-2. Gwin, R., Murray, R.B., 1963a. Scintillation process in CsI(Tl). 1. Com- parison with activator saturation model. Physical Review 131, 501{508. doi:{10.1103/PhysRev.131.501}. Gwin, R., Murray, R.B., 1963b. Scintillation process in CsI(Tl). 2. Emission spectra and possible role of self-trapped holes. Physical Review 131, 508{ 511. doi:{10.1103/PhysRev.131.508}. Horn, D., Ball, G., Galindo-Uribarri, A., Hagberg, E., Walker, R., Laforest, R., Pouliot, J., 1992. The mass dependence of CsI(Tl) scintillation re- sponse to heavy ions. Nucl. Instruments Methods Phys. Res. Sect. A Accel. Spectrometers, Detect. Assoc. Equip. 320, 273{276. URL: https://www. 25 sciencedirect.com/science/article/pii/0168900292907853, doi:10. 1016/0168-9002(92)90785-3. Kestil a, A., Tikka, T., Peitso, P., Rantanen, J., N asil a, A., Nordling, K., Saari, H., Vainio, R., Janhunen, P., Praks, J., Hallikainen, M., 2013. Aalto-1 nanosatellite { technical description and mission objectives. Geosci. Instrumentation, Methods Data Syst. 2, 121{130. doi:10.5194/ gi-2-121-2013. McInnes, L., Healy, J., Astels, S., 2017. hdbscan: Hierarchical density based clustering. The Journal of Open Source Software 2, 205{206. URL: https: //doi.org/10.21105/joss.00205, doi:10.21105/joss.00205. Peltonen, J., Hedman, H.P., Ilmanen, A., Lindroos, M., M a att anen, M., Pesonen, J., Punkkinen, R., Punkkinen, A., Vainio, R., Valtonen, E., S antti, T., Pentik ainen, J., Hggstr om, E., 2014. Electronics for the RAD- MON instrument on the Aalto-1 student satellite, in: 10th Eur. Work. Mi- croelectron. Educ. EWME 2014, pp. 161{166. doi:10.1109/EWME.2014. Praks, J., Niemel a, P., N asil a, A., Kestil a, A., Jovanovic, N., Riwanto, B., Tikka, T., Leppinen, H., Vainio, R., Janhunen, P., 2018. Miniature spectral imager in-orbit demonstration results from aalto-1 nanosatellite mission, in: Proceedings of 2018 IEEE International Geoscience and Remote Sens- ing Symposium, pp. 1986{1989. doi:10.1109/IGARSS.2018.8517658. Shimoda, T., Ishihara, M., Nagatani, K., Nomura, T., 1979. Simple DE{E particle identi cation with a wide dynamic range. Nucl. Instruments Meth- ods 165, 261{264. URL: https://www.sciencedirect.com/science/ article/pii/0029554X79902805, doi:10.1016/0029-554X(79)90280-5. Sorensen, T.C., Armstrong, T.P., Pavanasam, A.G., Taherion, S., 2005. Galileo energetic particle detector observations of the spatial distributions and energy spectra of >1 and >11 MeV electrons in the 10{40 RJ region of the jovian magnetosphere. 26 Icarus 178, 395 { 405. URL: http://www.sciencedirect.com/ science/article/pii/S0019103505002484, doi:https://doi.org/10. 1016/j.icarus.2005.07.006. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Astrophysics arXiv (Cornell University)

Loading next page...
 
/lp/arxiv-cornell-university/calibration-of-radmon-radiation-monitor-onboard-aalto-1-cubesat-MAckgRzpEV
ISSN
0273-1177
eISSN
ARCH-3330
DOI
10.1016/j.asr.2019.11.020
Publisher site
See Article on Publisher Site

Abstract

RADMON is a small radiation monitor designed and assembled by students of University of Turku and University of Helsinki. It is own on-board Aalto- 1, a 3-unit CubeSat in low Earth orbit at about 500 km altitude. The detec- tor unit of the instrument consists of two detectors, a Si solid-state detector and a CsI(Tl) scintillator, and utilizes the DE-E technique to determine the total energy and species of each particle hitting the detector. We present the results of the on-ground and in- ight calibration campaigns of the in- strument, as well as the characterization of its response through extensive simulations within the Geant4 framework. The overall energy calibration Email addresses: philipp.oleynik@utu.fi (Philipp Oleynik ), rami.vainio@utu.fi (Rami Vainio ), arjupu@utu.fi (Arttu Punkkinen), dudnik@rian.kharkov.ua (Oleksiy Dudnik), jan.gieseler@utu.fi (Jan Gieseler), hannu-pekka.hedman@utu.fi (Hannu-Pekka Hedman), heli.hietala@utu.fi (Heli Hietala), edward.haeggstrom@helsinki.fi (Edward Hggstr om), petri.niemela@aalto.fi (Petri Niemel a), juhpe@utu.fi (Juhani Peltonen), jaan.praks@aalto.fi (Jaan Praks), rpunk@utu.fi (Risto Punkkinen), teansa@utu.fi (Tero S antti), eino.valtonen@utu.fi (Eino Valtonen) Accepted to Advances in Space Research November 19, 2019 arXiv:1911.07550v1 [astro-ph.IM] 18 Nov 2019 margin achieved is about 5%. The full instrument response to protons and electrons is presented and the issue of proton contamination of the electron channels is quanti ed and discussed. Keywords: Radiation belts, Electron precipitation, Solar energetic particles, CubeSats 1. Introduction RADMON (Peltonen et al., 2014; Kestil a et al., 2013) is a small radia- tion monitor on-board the 3-unit Aalto-1 CubeSat (Kestil a et al., 2013). The satellite was launched by a PSLV-C38 rocket from India on 23 June 2017 into the low Earth orbit with an inclination of 97 degrees and an average altitude of 505 km (Praks et al., 2018). Aalto-1 is Finland's rst national satellite mission and it is designed, assembled and operated by students at Aalto Uni- versity, Espoo. The RADMON experiment was designed and assembled by students, in the University of Turku and University of Helsinki. Here we will describe the RADMON detector unit and characterize the response of the detector to the energetic particle radiation it measures, including the calibra- tion of the instrument. First scienti c results of the RADMON experiment are presented by Gieseler et al. (2019). 2. RADMON Instrument RADMON (Peltonen et al., 2014) consist of four subsystems: the Detector Unit, the Analog Electronics Board, the Digital Electronics Board and the Power Supply Board stacked in a compact con guration with a volume of 0.4 units (Fig. 1), a mass of 360 g, and a power consumption of 1 W. The detector signals are ampli ed and continuously digitized at 10 MHz sampling rate on the Analog Electronics Board and then transmitted through the instrument bus connecting the boards to the Digital Electronics Board, where a Xilinx Virtex-4 LX15 eld programmable gate array (FPGA) handles the signal processing from the pulse detection and pulse height determination to the classi cation and counting of particle events in spectral channels of 2 Figure 1: RADMON radiation monitor on-board Aalto-1. The instrument consists of a stack of three printed circuit boards and a detector unit in a brass housing, and an aluminum frame on the top. A square aluminum entrance window in front covers the detector unit inside the brass container. the instrument. The details of the RADMON electronics are presented by Peltonen et al. (2014). 2.1. Instrument geometry The detector unit of the instrument consists of a 350-mm-thick silicon de- 2 3 tector with an active area of 2.12.1 mm and a 101010 mm CsI(Tl) scintillation detector acting as a calorimeter. The scintillator is coupled to a 1010 mm p-i-n photodoide for optical photon readout. The scintillator crystal cube has ve sides wrapped with a white PTFE lm which improves light collection on the photodiode. Signal processing circuits of the detec- tors are independent; they produce voltage pulses, which are then digitized by corresponding analog-to-digital converters (ADC) (Peltonen et al., 2014). The detectors are housed in a brass container with walls thick enough to stop protons below 50 MeV and electrons below 8 MeV. The brass container is xed to an aluminum frame, which also carries the electronics of the instru- ment. 3 incident particle Figure 2: A cross section of the detector unit. Topmost layer is an aluminum entrance window, a brass container is light-brown, the silicon detector is orange, its passive 3.53.5 mm area is gray, and its supporting structure is dark-green. The CsI(Tl) scintillator cube is light-green, below which there is the Hamamatsu S3590-08 silicon photodiode with a depletion layer of 300 mm. The frontal opening in the container collimates incoming particles to a solid angle around =5 and is covered by a 280 mm thick aluminum entrance window (Fig. 2). The entrance window is opaque for electrons with energies below 0.24 MeV and for protons with energies below 6.5 MeV. An incident particle rst passes the aluminum entrance window, depositing some energy there, which we account for during the calibration, then the particle deposits some energy in the silicon detector and the rest is likely to be deposited in the scintillator. For most protons with energies 10 { 50 MeV the scintillator acts as a calorimeter, so that the particle energy is high enough to penetrate the silicon detector, but low enough to be absorbed in the scintillator. Con- sidering such protons one can write an equation based on the dependency of the particle range, R [g cm ], in medium on energy: R (E ) = R (E ) + d  ; (1) Si 0 Si CsI Si Si where E is the incident particle energy when it enters the silicon detector, E is the energy deposited in CsI scintillator, d = 350 mm is the thickness CsI Si of the silicon detector, and  = 2:33 g cm is the silicon density. E , in this Si 0 case, is the sum of E and E , energy deposited in silicon. This equation CsI Si 4 could be used to determine a set of points on the [E E ] plane which Si CsI can be used both for particle discrimination and instrument calibration. Assuming that the energy of the particle is fully absorbed in the instru- ment and the range function is reversible, the equation (1) can be solved for energy deposited in silicon E vs incident particle energy E . The range{ Si 0 energy relation is commonly approximated by the Bragg-Kleeman rule as R (E) = A  E , where the incident particle energy is normalized to a Si 0 certain value which is 1 MeV in the present work. However, this approxima- tion di ers slightly from the measured range at lower energies (Berger et al., 1992). We have chosen an approximation which gives less than 2% error in the 2 { 200 MeV energy range, given by (Attix, 2008) R (E) = + : (2) Si 1 MeV The best t with the PSTAR (Berger et al., 1992) proton range data yields 3 2 3 2 = 1:76, = 12:5 10 g cm , and = 4:3 10 g cm . Since the energy{ range relation is reversible, one can obtain a solution of the equation (1) for Si 1= E d 0 Si Si E = E MeV : (3) Si 0 1 MeV Using the assumption E + E = E it is possible to describe the curve on Si CsI 0 the E E plane by the equation Si CsI 1= E d CsI Si Si E = + MeV E : (4) Si CsI 1 MeV The equation (4) de nes a so-called "banana" curve with parameters xed by the instrument geometry. It is independent on E because each incident energy value is represented on the curve by a dot, therefore a continuous energy spectrum of incoming protons produces a continuous curve, with a shape determined only by the detector geometry. The true incoming particle energy di ers from E by several MeV of energy absorbed in the aluminum entrance window. The shape of the curve is independent on E , but the value of the energy threshold is a ected by the energy absorbed in the entrance 5 window. A zero angle of incidence is assumed between the proton momen- tum direction and the normal to the silicon detector plane. For an angle of incidence  deviating from zero the e ective silicon thickness is greater by a factor of 1= cos . In space, protons originate from all directions and the brass collimator restricts incident directions to a  20 half-width cone for protons with energies below 50 MeV. This restriction applied to the factor of 1= cos  alters the e ective thickness of the silicon detector by  6% which results in a slight blurring of the curve provided by (4) towards higher E , Si since particles deposit more energy in the e ectively thicker silicon detector, see gure 3. 2.2. Particle counters A particle must hit both detectors to be registered. The detection logic rejects any single detector hits or events that have deposited energy below a threshold. This logic eliminates single hits coming from side-penetrating particles or bremsstrahlung X-rays from electrons scattered in the brass con- tainer or spacecraft structures. In the present simulations, we consider the detection of an incident particle and secondary ones it might produce as a separate event. There is a nite chance that a primary or secondary parti- cle creates the coincidence condition by hitting, e.g., the scintillator within a temporal coincidence window with another particle that has deposited energy in the silicon detector, or vice versa. Taking into account observed detection rates and the coincidence window interval we nd this e ect negligible. The discrimination of incoming particles to protons and electrons is ac- complished using the E{E technique (Goulding et al., 1964; Birks, 1964b; Shimoda et al., 1979) applied to digitized pulse heights of the detector sig- nals. A linear combination of these digitized pulse heights in both detectors is used as a measure of the total deposited energy E . Particle counters are accumulated for 15 seconds and then stored with a time stamp. After the current counter values are stored, all counters are reset to zero. The aperture of the instrument rotates with the satellite and scans over all pitch angles evenly since 15 seconds cover a few rotation periods. Protons are separated from electrons using their locus on the E E Si CsI 6 plane. There are two regions on the plane which follow the "banana" curve on the (Fig. 3) where a detected particle is counted as a proton. Protons are further classi ed into nine energy channels p1 { p9 by a sequence of thresholds for the E . Each channel is a counter which is incremented when conditions for the E are met. Energies deposited by electrons do not produce a distinct curve on the plane due to substantial scattering in both detectors, but they are con ned by a certain region with borders quite far from the proton curve. One can estimate the energy deposited by an electron as a sum of deposited energies in both detectors. The detected electrons are counted in ve electron channels e1 { e5 depending on the total deposited energy. Any particle detection which cannot be classi ed as a detection of an electron or a proton increments one of the two outlier counters, o1 or o2, below and above the proton track, respectively. For each channel and deposited energy it is possible to describe a de- tection probability density function p = p(E ; channel), which we fur- channel 0 ther denote as instrument response. The instrument response is obtained by Monte-Carlo simulations and is described in section 4.1. 2.3. Pulse height data Raw pulse height data is also available for calibration purposes and can be downlinked during calibration campaigns. The amount of data is sub- stantially higher during these observations, so it was necessary to keep the campaigns short for the sake of continuous observations in the normal particle counting mode, yet long enough to enable in- ight calibration in real space environment, which was impossible to reproduce on ground-based facilities available to the team. The available pulse height data covers di erent regions such as high- latitude belts where particles from higher L-shells can be observed, quiet equatorial regions, and the South Atlantic geomagnetic anomaly, where the proton contamination dominates over the electron ux in electron channels e2 { e5. See Gieseler et al. (2019) for a description of the measured radiation environment. 7 2.4. On-ground calibration The instrument was originally calibrated on the ground using an acceler- ator beam at Abo Akademi University. The beam energy was kept constant (17 MeV) during the calibration. The instrument was placed into a par- allel proton beam arriving straight to the aperture through exchangeable aluminum sheets of di erent thickness in order to control the incident proton energy for the instrument. The calibration setup was modeled within the Geant4 framework (Agostinelli et al., 2003; Allison et al., 2006; Allison et al., 2016) in order to estimate en- ergies deposited in both detectors of the instrument. Each detector was considered to have a linear response function, so its digital signal N ADC could be characterized by a pair of parameters N = a  E + b LSB ADC det (least signi cant bits), where a is the overall gain and b is an o set. These two parameters were calculated using the least squares method applied to measurements. The detector calibration results are presented in table 1. Table 1: Gains and o sets for the RADMON detectors obtained during the on-ground and in- ight calibration campaigns. Values without errors in the in- ight calibration were kept xed. a [LSB/MeV] b [LSB] a [LSB/MeV] b [LSB] Si Si CsI CsI on-ground: 140 6 3 15 18:3 0:5 9 5 in- ight: 140 3 14:6 0:3 0 First calibration data obtained in space right after the launch showed that the gain previously calculated for the CsI scintillation detector deviated from the present gain value of the detector by up to 20%. One possible reason for that is degradation of the optical contact between the scintillator crystal and the photodiode readout, since the change was detected right in the beginning of orbital measurements. This fact called for an in- ight calibration campaign to be performed using protons observed in low Earth orbit. The campaign consisted of two runs, the second one was preformed with an elevated energy detection threshold in order to keep electron counts low and reserve the telemetry for proton counts. The in- ight calibration 8 campaign data is analyzed in the present work. 3. Simulations 3.1. Proton track calibration As it was mentioned in section 2.1, protons with energies of 10 { 50 MeV form a distinct non-linear feature on the E E plane. The posi- Si CsI tion and the curvature of that proton track are de ned only by the detector dimensions and materials. The linearity of the readout electronics is well established during on-ground tests, making this track a perfect tool to cali- brate gains and o sets of the RADMON detectors in space. During several calibration intervals in orbit, protons and electrons with continuous spectra were observed and the pulse height data were delivered to ground. The data points were analyzed on the [ADC { ADC ] plane. (Si) (CsI) Since protons in space are not intrinsically collimated, there is a noticeable background above and below the proton "banana". We have processed the data using the HDBSCAN algorithm (McInnes et al., 2017), which detects clusters in noisy data. 568 data points were selected for calibration. In order to determine proper gains and o sets using these data one needs an analytical description of the observed curve in ADC units. The readout is linear, but there are non-linear processes in the scintillator and its readout photodiode. Scintillation processes in CsI(Tl) are thoroughly studied (Gwin and Mur- ray, 1963a,b). The scintillation has two components with di erent decay times (Benrachi et al., 1989), with their ratio depending on speci c energy loss dE=dx. Thus, it is particle species dependent. Moreover, the light out- put is quenched by recombination of electron-hole pairs in the scintillator medium (Birks e ect) (Birks, 1964a). The e ect is negligible for electrons, but plays an important role in proton detection. It can be described in the di erential form as dL S(dE=dx) = ; (5) dx 1 + kB(dE=dx) where L is the light output, E is the particle energy, S is a normalization constant, and kB is a measure of the Birks e ect in uence on the light 9 output. This equation must be integrated in order to derive L(E) function, which then is used to get a relation between the digitized pulse height and the particle energy as seen by the scintillator ADC (E). This function has Si scintillation detector gain and o sets as parameters to be calibrated. Horn et al. (1992) asserted that dE=dx  1=E for the sake of analytical integration of the equation 5 but, as was pointed by Avdeichikov et al. (2000), this approximation is substantially limited. We have approximated dE=dx for protons in the energy range relevant to the RADMON calibration as dE=dx E with = 0:678, which ts well the experimental data (Berger et al., 1992). Integration of the equation (5) gives the L(E ) function expressed CsI through a hyper-geometric function F 2 1 1 1 E CsI L(E )  E  1 F (1; ; + 1; ) : (6) CsI CsI 2 1 kB a Finally, we obtain 1 1 E CsI ADC = a  E  1 F (1; ; + 1; ) + b : (7) (CsI) CsI CsI 2 1 CsI kB a This function was used to t the analytical description of the proton track to the chosen experimental points. The kB constant was adopted from Avde- ichikov et al. (2000). For the silicon detector a linear function was used ADC = a  E + b ; (8) (Si) Si Si Si where gain a and o set b were initialized with the values measured on Si Si ground. The t was done by least-squares method; the shortest distance to the curve from each experimental point was taken as the error estimator. 3.2. Geant4 model A complete response function for each particle channel was obtained from simulations carried out within the Geant4 framework (Agostinelli et al., 2003; Allison et al., 2006; Allison et al., 2016). The whole Aalto-1 satellite with the RADMON instrument was modeled as a realistic 3D model described by a 10 GDML (Chytracek et al., 2006) script. All parts of the RADMON instrument were modeled exactly as they are in reality. The rest of the satellite was partially simpli ed by replacing ne details of sub-systems like the on-board computer and batteries with an aluminum foam lling the space inside the 3U CubeSat frame. The foam density was calculated from a previously measured mass of the spacecraft subsystems. The model was placed into a cubic Geant4 world with dimensions of 300300300 cm . A particle source used in the model was a sphere with a radius of 30 cm, which fully enclosed the Aalto-1 satellite model. Each par- ticle was rst placed into a uniformly random position on the sphere, with U (0; 1) and cos   U (0; 1). The particle momentum direction was cho- sen according to the uniform Lambertian angular distribution (Greenwood, 2002). During a simulation run, the particle energy remained constant. In order to cover the necessary energy range, we used a quasi-logarithmic grid of energies. The sensitive volumes in the model were the 2.12.1 mm silicon detector, the 101010 mm CsI(Tl) scintillation crystal, and the p-i-n photodiode, which was modeled as a 300 mm thick 1010 mm silicon detector in order to take direct particle energy deposits into account. The photodiode is normally lit by scintillation light, but in some cases, it can be hit directly by a particle. Such hits are indistinguishable from light pulses from the scintillator due to moderately high integration time, but their contribution to the response is quite small since the photodiode is installed at the back of the instrument aperture. Direct hits ionize the photodiode at proton energies from about 55 MeV and make no special features in the instrument response functions. The modeling software recorded deposited energies for each particle in each volume. We summed energy deposits from primary particles and sec- ondary ones, exactly the way the real instrument does. All physical inter- actions possible at the simulated energy range were enabled in the software. The detection was counted only if the energy deposit was above 70 keV in both detectors. The threshold of 70 keV imitates the noise oor of the instru- ment. Direct hits in the photodiode were recorded as an auxiliary component to be added to the scintillator output following Bird et al. (1994). All single 11 hits were discarded, since the instrument detection logic behaves the same way in hardware. We used gains and o sets obtained from the calibration to provide the in- strument responses by simulating the RADMON detection logic and particle channel classi ers. Each detection record collected from the Geant4 model was rst converted from energy units to the ADC units. Then the simulated detection was examined whether it must be classi ed as an electron or a proton, or it must be counted in one of the outlier channels. The data for simulated proton detections was utilized to quantify the proton contamina- tion of electron channels. Energy channel classi cation was performed by an algorithm implemented in RADMON based on analysis of the parameter E = (10  ADC + 48 (Si) ADC )=64 . The division in this formula is the standard integer division, (CsI) since the RADMON does all arithmetics in the integer domain. The linear combination, based on in- ight calibration, overestimates the gain of the scintillator but the e ect is taken into in the calculated energy response of each energy channel. 3.3. Bowtie analysis The RADMON particle channels were selected and combined in order to get several integral channels which data could be directly converted to ux without complex procedures involving response curves. We have calculated geometric factors and threshold energies for the following channels using a Van Allen "bowtie" analysis described by e.g. Sorensen et al. (2005). During the simulations the incident particle energy was set by a quasi- logarithmic grid with 48 energies for each decade. Geometric factors were calculated for each energy bin using the expression N (channel; E ) d bin G(channel; E ) = A ; (9) bin r N (E ) s bin where N (channel; E ) is a number of particles detected in an instrument d bin particle channel, N (E )is a number of particles shot in a particular energy s bin bin, and A is an area of radiating sphere. All geometric factors for available 12 energy bins were combined to make a discrete function G(E) for each instru- ment channel. For a combination of channels a corresponding combination of geometric factors was calculated. These geometric factors were used for an integral bowtie analysis accord- ing to the formula f (E)G(E)dE G = ; (10) I 1 f (E)dE where f (E) is a modeled di erential ux given as a power-law with a range of indices and E is a threshold energy for the analysis. If a channel can be characterized as an integral one, there is a speci c value E of the energy E 0 t for which G is the same for a wide range of power-law indices of f (E) (see gure 5). A mean value and a con dence interval for G are calculated through statistical analysis of a distribution of G values obtained for di erent power- law indices, and E is evaluated as a middle of an energy band where standard deviation of G does not exceed three times its minimal value. A con dence interval for E is this energy band width. For a nice integral channel its con dence interval is practically the width of an energy bin where E lies. An integral ux is then de ned by F (E > E ) = ; (11) where R is a count rate in the channel for which G and E are de ned. I 0 4. Results 4.1. Response of the particle channels We have obtained curves for geometric factors as a function of incident particle energy for each channel (Figure 6 and 7). The electron channels e2 { e5 are integral channels. They have long tails stretching up to high energies, yet each channel shows a clear threshold en- ergy. The response curves allow calculation of the integral threshold energies and corresponding geometric factors for these channels. 13 We have converted proton channels from di erential ones to four integral channels and one di erential in high energy range, in order to obtain a set of well-de ned energy ranges to be quoted for the channels. The measurement channels p1..p4 have responses similar to boxcar functions in the nominal energy range, but they also have relatively strong high-energy side bands, which limits their use as di erential channels. At 50 MeV and above the brass collimator becomes transparent to protons, so they start to be regis- tered also in the rst proton channels. Protons of energies higher than 100 MeV penetrate the whole satellite. They deposit little energy both in the scintillator and the silicon detector. Thus, they miss the proton track and get into the outlier channel o1 or contaminate electron channels. 4.2. Angular sensitivity The angular sensitivity plot (Figure 8) shows how the RADMON aperture gradually expands when energy increases beyond 50 MeV. This widening of a sensitive aperture is also seen on the response curves. Protons with energies slightly above 100 MeV are much less likely capable to get to the proton channels of the instrument, whereas protons of higher energies could be detected virtually anywhere on the [ADC { ADC ] plane. (Si) (CsI) 4.3. Contamination issues During the mission high count rates in electron channel have been ob- served while the spacecraft passed the South Atlantic Anomaly. Here we characterize the proton contamination of the electron channels of the RAD- MON to avoid misinterpretation of the observations. The contamination geometric factors are presented in Figure 9. The highest contamination ex- ists in channels e4 and e5. These responses fully explain the counting rates in the electron channels e3{e5 observed inside South Atlantic Anomaly. 4.4. Channel e1 issue The electron channel e1 was initially planned to measure electron ux from about 1 MeV energies. However, RADMON scintillator channel picks up noise from the environment, which led to the necessity of increasing the 14 detection threshold of the channel to values higher than foreseen, in order to avoid contaminating the measurements. During the mission we have observed very few counts in e1 and identi ed the reason to be a very narrow energy region of remaining non-zero sensitivity in e1. In addition, this narrow band is highly sensitive to pulse detection eciency near the threshold and drift of parameters of the RADMON signal processing pipeline, whereas the rest of the channels are stable due to the larger detector pulse heights. Therefore, the channel e1 has to be discarded from the scienti c dataset. 4.5. E ective geometric factors and energies of the channels The bowtie analysis provided integral geometric factors for instrument electron channels e2 { e4 and composite proton channels. All channels are integral: the e2 { e4 channels have long tails on the high energy side, and the i1 { i4 channels are composite ones combined from the instrument proton channels p1 { p9 in such a way that they have step-like response curves. They are presented in table 2. Table 2: Bowtie cuto energies and geometric factors for integral particle channels. e2 { e4 are electron integral channels and i1 { i4 are proton integral channels. The con dence intervals are at a level of 95%. Channel Cuto energy [MeV] Geometric factor [cm sr] e2 1:51 0:1 0:0108 0:0005 e3 3:1 0:2 0:0160 0:0005 e4 6:0 0:7 0:0119 0:0008 i1=(p1 : : : p9) 10:4 0:3 0:0228 0:0004 i2=(p2 : : : p9) 18:5 0:7 0:0256 0:0009 i3=(p3 : : : p9) 23:7 1:8 0:0219 0:0011 i4=(p4 : : : p9) 29 4 0:0187 0:0014 We have a di erential proton channel for high energy protons, i5 = (p5 : : : p9), it has sensitivity from 40 to 80 MeV with a di erential geo- metric factor G E = 0:78 0:09 cm sr MeV at an energy of 42 5 MeV. i5 15 5. Summary and Conclusions A realistic 3D model of the Aalto-1 satellite with the RADMON radiation monitor was constructed in a Geant4 simulation framework. Structures of the satellite and the instrument were described by a GDML script. The virtual model was placed into omnidirectional monoenergetic ux of protons and electrons. The energies of simulated particles covers the RADMON sensitiv- ity range, also extending to higher energies in order to study contamination issues. We have calculated energy response curves for protons and electrons for all instrument channels in a wide energy range. These responses include ge- ometric factors for electron channels contaminated by high energy protons. We have constructed four integral proton channels from the instrument chan- nels p1 { p9 and one di erential channel sensitive for protons of 40 { 80 MeV. The e2 { e4 electron channels of the RADMON instrument are integral ones. The obtained results allow conversion of count rates in the individual channels of the RADMON instrument to isotropic ux measurements in low Earth orbit. The data description will be published in a separate paper Gieseler et al. (2019). Acknowledgements. This work was performed in the framework of the Finnish Centre of Excellence in Research of Sustainable Space (FORESAIL) funded by the Academy of Finland (grants 312357 and 312358). We also gratefully acknowledge the e orts of dozens of students in Aalto University, University of Turku and University of Helsinki for their work in the Aalto-1 satellite and RADMON projects. Aalto University MIDE is thanked for nancial support for building Aalto-1. Aalto University, University of Turku, RUAG, Space Systems Finland, and Nokia sponsored the launch of the satellite. Compu- tations necessary for the presented modeling were conducted on the Pleione cluster at the University of Turku. 16 Figure 3: A simulated "banana" curve for protons with energies from 8 to 100 MeV. Color of the dots in upper panel denotes the incident proton energy. The red curve is the one de ned by the equation (4) where E is shifted by 0.15 MeV to correct for a 2% error of CsI the analytical range approximation at lower energies. The lower panel of the plot shows a residual between the analytical curve and Geant4 simulations. 8 MeV is close to the detection threshold, so there are few particles on the leftmost part of the curve compared to the central part. Stripe pattern on the lower panel appears due to simulations are carried out for monoenergetic particles on a discrete energy grid. The sharp curve change at 50 MeV is caused by CsI starting to be transparent, so that the assumption for equation 1 is no longer true. Protons leave less and less energy in CsI at incident energies of 55 MeV and above. 17 Figure 4: Proton "banana" curve t using calibration data from orbit. The gain value obtained for the scintillation detector is 68:5  1:2 keV/LSB. Shaded polygons are the regions where incoming particles are classi ed as protons or electrons. There are two populations of particles in the electron region below. They originate from two calibration campaigns, one of which had higher detection threshold in order to deliver more data covering the proton track. 18 Figure 5: Bowtie analysis for the e3 instrument channel. E = 3:08 0:06 MeV. Curves show G for di erent power-law indices of f (E) in a range of [2 : : : 7]. Figure 6: RADMON response functions for electron channels e1 { e5. Electron channels from e2 to e4 are integral channels with a distinct threshold. The black curve above individual response curves shows the integral sensitivity to electrons. 19 Figure 7: RADMON response functions for combined integral proton channels and a di erential one, which is also a combined channel p5..p9. Figure 8: RADMON angular sensitivity for protons in proton channels. 20 Figure 9: High energy proton contamination of electron channels e2 { e5. 21 References Agostinelli, S., Allison, J., Amako, K., Apostolakis, J., Araujo, H., Arce, P., Asai, M., Axen, D., Banerjee, S., Barrand, G., Behner, F., Bellagamba, L., Boudreau, J., Broglia, L., Brunengo, A., Burkhardt, H., Chauvie, S., Chuma, J., Chytracek, R., Cooperman, G., Cosmo, G., Degtyarenko, P., Dell'Acqua, A., Depaola, G., Dietrich, D., Enami, R., Feliciello, A., Fer- guson, C., Fesefeldt, H., Folger, G., Foppiano, F., Forti, A., Garelli, S., Giani, S., Giannitrapani, R., Gibin, D., Cadenas, J.G., Gonz alez, I., Abril, G.G., Greeniaus, G., Greiner, W., Grichine, V., Grossheim, A., Guatelli, S., Gumplinger, P., Hamatsu, R., Hashimoto, K., Hasui, H., Heikkinen, A., Howard, A., Ivanchenko, V., Johnson, A., Jones, F., Kallenbach, J., Kanaya, N., Kawabata, M., Kawabata, Y., Kawaguti, M., Kelner, S., Kent, P., Kimura, A., Kodama, T., Kokoulin, R., Kossov, M., Kurashige, H., Lamanna, E., Lamp en, T., Lara, V., Lefebure, V., Lei, F., Liendl, M., Lockman, W., Longo, F., Magni, S., Maire, M., Medernach, E., Minami- moto, K., de Freitas, P.M., Morita, Y., Murakami, K., Nagamatu, M., Nartallo, R., Nieminen, P., Nishimura, T., Ohtsubo, K., Okamura, M., O'Neale, S., Oohata, Y., Paech, K., Perl, J., Pfei er, A., Pia, M., Ranjard, F., Rybin, A., Sadilov, S., Salvo, E.D., Santin, G., Sasaki, T., Savvas, N., Sawada, Y., Scherer, S., Sei, S., Sirotenko, V., Smith, D., Starkov, N., Stoecker, H., Sulkimo, J., Takahata, M., Tanaka, S., Tcherniaev, E., Tehrani, E.S., Tropeano, M., Truscott, P., Uno, H., Urban, L., Urban, P., Verderi, M., Walkden, A., Wander, W., Weber, H., Wellisch, J., Wenaus, T., Williams, D., Wright, D., Yamada, T., Yoshida, H., Zschiesche, D., 2003. Geant4 { a simulation toolkit. Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment 506, 250 { 303. URL: http://www.sciencedirect. com/science/article/pii/S0168900203013688, doi:https://doi.org/ 10.1016/S0168-9002(03)01368-8. Allison, J., Amako, K., Apostolakis, J., Araujo, H., Arce Dubois, P., Asai, M., Barrand, G., Capra, R., Chauvie, S., Chytracek, R., Cirrone, G.A.P., Cooperman, G., Cosmo, G., Cuttone, G., Daquino, G.G., Donszelmann, 22 M., Dressel, M., Folger, G., Foppiano, F., Generowicz, J., Grichine, V., Guatelli, S., Gumplinger, P., Heikkinen, A., Hrivnacova, I., Howard, A., Incerti, S., Ivanchenko, V., Johnson, T., Jones, F., Koi, T., Kokoulin, R., Kossov, M., Kurashige, H., Lara, V., Larsson, S., Lei, F., Link, O., Longo, F., Maire, M., Mantero, A., Mascialino, B., McLaren, I., Mendez Lorenzo, P., Minamimoto, K., Murakami, K., Nieminen, P., Pandola, L., Parlati, S., Peralta, L., Perl, J., Pfei er, A., Pia, M.G., Ribon, A., Rodrigues, P., Russo, G., Sadilov, S., Santin, G., Sasaki, T., Smith, D., Starkov, N., Tanaka, S., Tcherniaev, E., Tome, B., Trindade, A., Truscott, P., Urban, L., Verderi, M., Walkden, A., Wellisch, J.P., Williams, D.C., Wright, D., Yoshida, H., 2006. Geant4 developments and applications. IEEE Transac- tions on Nuclear Science 53, 270{278. doi:10.1109/TNS.2006.869826. Allison, J., Amako, K., Apostolakis, J., Arce, P., Asai, M., Aso, T., Bagli, E., Bagulya, A., Banerjee, S., Barrand, G., Beck, B., Bogdanov, A., Brandt, D., Brown, J., Burkhardt, H., Canal, P., Cano-Ott, D., Chauvie, S., Cho, K., Cirrone, G., Cooperman, G., Cort es-Giraldo, M., Cosmo, G., Cuttone, G., Depaola, G., Desorgher, L., Dong, X., Dotti, A., Elvira, V., Folger, G., Francis, Z., Galoyan, A., Garnier, L., Gayer, M., Genser, K., Grichine, V., Guatelli, S., Gu eye, P., Gumplinger, P., Howard, A., H rivn a cov a, I., Hwang, S., Incerti, S., Ivanchenko, A., Ivanchenko, V., Jones, F., Jun, S., Kaitaniemi, P., Karakatsanis, N., Karamitros, M., Kelsey, M., Kimura, A., Koi, T., Kurashige, H., Lechner, A., Lee, S., Longo, F., Maire, M., Man- cusi, D., Mantero, A., Mendoza, E., Morgan, B., Murakami, K., Nikitina, T., Pandola, L., Paprocki, P., Perl, J., Petrovi c, I., Pia, M., Pokorski, W., Quesada, J., Raine, M., Reis, M., Ribon, A., Fira, A.R., Romano, F., Russo, G., Santin, G., Sasaki, T., Sawkey, D., Shin, J., Strakovsky, I., Taborda, A., Tanaka, S., Tom e, B., Toshito, T., Tran, H., Truscott, P., Urban, L., Uzhinsky, V., Verbeke, J., Verderi, M., Wendt, B., Wen- zel, H., Wright, D., Wright, D., Yamashita, T., Yarba, J., Yoshida, H., 2016. Recent developments in Geant4. Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment 835, 186 { 225. URL: http://www.sciencedirect. 23 com/science/article/pii/S0168900216306957, doi:https://doi.org/ 10.1016/j.nima.2016.06.125. Attix, F., 2008. Introduction to Radiological Physics and Radiation Dosime- try. Wiley-VCH Verlag GmbH & Co. Avdeichikov, V., Fomichev, A.S., Jakobsson, B., Rodin, A.M., Ter-Akopian, G.M., 2000. Range energy relation, range straggling and response function of CsI(Tl), BGO and GSO(Ce) scintillators for light ions 439, 158{166. Benrachi, F., Chambon, B., Cheynis, B., Drain, D., Pastor, C., Seghier, D., Zaid, K., Giorni, A., Heuer, D., Ll eres, A., Morand, C., Stassi, P., Viano, J., 1989. Investigation of the performance of CsI(Tl) for charged particle identi cation by pulse-shape analysis. Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment 281, 137 { 142. URL: http://www.sciencedirect. com/science/article/pii/0168900289912254, doi:https://doi.org/ 10.1016/0168-9002(89)91225-4. Berger, M.J., Coursey, J.S., Zucker, M.A., Chang, J., 1992. Stopping-Power & Range Tables for Electrons, Protons, and Helium Ions Share. NIST Stan- dard Reference Database 124. Technical Report NISTIR 4999. National Institute of Standards and Technology (NIST), Physics Laboratory (now Physical Measurement Laboratory (PML)). Bird, A., Carter, T., He, Z., Dean, A., Ramsden, D., 1994. Pulse shape analysis of signals from a CsI(T1)/photodiode detector. Nucl. Instruments Methods Phys. Res. Sect. A Accel. Spectrometers, Detect. Assoc. Equip. 353, 46{49. URL: https://www.sciencedirect.com/science/article/ pii/0168900294915997, doi:10.1016/0168-9002(94)91599-7. Birks, J.B., 1964a. The Scintillation Process in Alkali Halide Crystals. IEEE Trans. Nucl. Sci. 11, 4{11. doi:10.1109/TNS.1964.4323396. Birks, J.B., 1964b. The theory and practice of scintillation counting. Perga- mon, Oxford. 24 Chytracek, R., Mccormick, J., Pokorski, W., Santin, G., 2006. Geom- etry description markup language for physics simulation and analysis applications. IEEE Transactions on Nuclear Science 53, 2892{2896. doi:10.1109/TNS.2006.881062. Gieseler, J., Oleynik, P., Hietala, H., Vainio, R., Hedman, H.P., Peltonen, J., Punkkinen, A., Punkkinen, R., S antti, T., Hggstr om, E., Praks, J., Niemel a, P., Riwanto, B., Jovanovic, N., Mughal, M.R., 2019. Radiation monitor radmon aboard aalto-1 cubesat: First results. Submitted to Ad- vances in Space Research . Goulding, F., Landis, D., Cerny, J., Pehl, R., 1964. A new particle identi er technique for Z = 1 and Z = 2 particles in the energy range > 10 MeV. Nucl. Instruments Methods 31, 1{12. URL: https://www.sciencedirect.com/ science/article/pii/0029554X64903131, doi:10.1016/0029-554X(64) 90313-1. Greenwood, J., 2002. The correct and incorrect generation of a cosine distribution of scattered particles for Monte-Carlo modelling of vacuum systems. Vacuum 67, 217 { 222. URL: http://www.sciencedirect. com/science/article/pii/S0042207X02001732, doi:https://doi.org/ 10.1016/S0042-207X(02)00173-2. Gwin, R., Murray, R.B., 1963a. Scintillation process in CsI(Tl). 1. Com- parison with activator saturation model. Physical Review 131, 501{508. doi:{10.1103/PhysRev.131.501}. Gwin, R., Murray, R.B., 1963b. Scintillation process in CsI(Tl). 2. Emission spectra and possible role of self-trapped holes. Physical Review 131, 508{ 511. doi:{10.1103/PhysRev.131.508}. Horn, D., Ball, G., Galindo-Uribarri, A., Hagberg, E., Walker, R., Laforest, R., Pouliot, J., 1992. The mass dependence of CsI(Tl) scintillation re- sponse to heavy ions. Nucl. Instruments Methods Phys. Res. Sect. A Accel. Spectrometers, Detect. Assoc. Equip. 320, 273{276. URL: https://www. 25 sciencedirect.com/science/article/pii/0168900292907853, doi:10. 1016/0168-9002(92)90785-3. Kestil a, A., Tikka, T., Peitso, P., Rantanen, J., N asil a, A., Nordling, K., Saari, H., Vainio, R., Janhunen, P., Praks, J., Hallikainen, M., 2013. Aalto-1 nanosatellite { technical description and mission objectives. Geosci. Instrumentation, Methods Data Syst. 2, 121{130. doi:10.5194/ gi-2-121-2013. McInnes, L., Healy, J., Astels, S., 2017. hdbscan: Hierarchical density based clustering. The Journal of Open Source Software 2, 205{206. URL: https: //doi.org/10.21105/joss.00205, doi:10.21105/joss.00205. Peltonen, J., Hedman, H.P., Ilmanen, A., Lindroos, M., M a att anen, M., Pesonen, J., Punkkinen, R., Punkkinen, A., Vainio, R., Valtonen, E., S antti, T., Pentik ainen, J., Hggstr om, E., 2014. Electronics for the RAD- MON instrument on the Aalto-1 student satellite, in: 10th Eur. Work. Mi- croelectron. Educ. EWME 2014, pp. 161{166. doi:10.1109/EWME.2014. Praks, J., Niemel a, P., N asil a, A., Kestil a, A., Jovanovic, N., Riwanto, B., Tikka, T., Leppinen, H., Vainio, R., Janhunen, P., 2018. Miniature spectral imager in-orbit demonstration results from aalto-1 nanosatellite mission, in: Proceedings of 2018 IEEE International Geoscience and Remote Sens- ing Symposium, pp. 1986{1989. doi:10.1109/IGARSS.2018.8517658. Shimoda, T., Ishihara, M., Nagatani, K., Nomura, T., 1979. Simple DE{E particle identi cation with a wide dynamic range. Nucl. Instruments Meth- ods 165, 261{264. URL: https://www.sciencedirect.com/science/ article/pii/0029554X79902805, doi:10.1016/0029-554X(79)90280-5. Sorensen, T.C., Armstrong, T.P., Pavanasam, A.G., Taherion, S., 2005. Galileo energetic particle detector observations of the spatial distributions and energy spectra of >1 and >11 MeV electrons in the 10{40 RJ region of the jovian magnetosphere. 26 Icarus 178, 395 { 405. URL: http://www.sciencedirect.com/ science/article/pii/S0019103505002484, doi:https://doi.org/10. 1016/j.icarus.2005.07.006.

Journal

AstrophysicsarXiv (Cornell University)

Published: Nov 18, 2019

There are no references for this article.