European Journal of Scientific Research ISSN 1450-216X / 1450-202X Vol. 126 No 2 November, 2014, pp.142 - 151 http://www.europeanjournalofscientificresearch.com
Micromegas Detector: Conception and Development based on Neon gasat High Energy Physics using 55Fe X-ray Source HamidMounir Spectrometry Laboratory of Materials and Archaeomaterials (LASMAR) Faculty of Science, University Moulay Ismail, Meknes – Morocco E-mail:
[email protected] SeddikBri Electrical Engineering Departments, High School of Technology, ESTM University My Ismail - Meknes- Morocco Abstract Recent years have seen the development of many fields of gas detectors. The MICROMEGAS (Micro-Mesh Gas Structure) appeared as the very promising detectors. It is major families of position detectors in High Energy Physics, detects and localizes energy deposit by photon and charged particles over large area. In this work, MATLAB simulations of This Detector are represented. The aim of this work is to simulate the principle parameters of MICROMEGAS detector (Townsend coefficients and amplification gain). We choose Neon-isobutane (Ne-iC4H10) and Ne-dimethyl-ether (Ne-DME) gas mixture, using 55Fe radioactive 5.9keV X-rays Source. In the normal condition, (T=298 K, p=1atm). The simulated results are nearly consistent with experimental data, and provide important information in the MICROMEGAS, concept and operating. Our simulation predicts that further improvements are still possible for give a best spatial and temporal resolution for MICROMEGAS detector.
Keywords:
MICROMAEGAS, Gas mixture, X-ray Source, Avalanche phenomenon, MATLAB Programing, Amplification
1. Introduction MICROMEGAS [1] is a high gain gaseous detector, which can stand up alone without a need of an additional pre-amplification. It is a new gaseous detector initially developed for track-in high-rate, high-energy, physics experiments since 1990. It shows higher counting rate capacity up to 108 .mm2 .s1 , position-sensitive with spatial resolution better than 100 microns and good performance of radiation hardness [2]-[3], which has been developed since 1996 at SACLAY, France [4]. MICROMEGAS is a Parallel Plate Detector (PPD) with three electrodes, cathode, micromesh and anode and a narrow amplification space, typically 50-100 microns, between the micromesh and the anode. The detector gain depends directly on the distance between the micromesh and anode; therefore, controlling the geometry of the amplification region is crucial in order to achieve good energy resolution and excellent timing properties [5]. These results were confirmed by a similar structure having wider amplification gap and thicker metallic grid [6].
Micromegas Detector: Conception and Development based on Neon gasat High Energy Physics using 55Fe X-ray Source
143
2. MICROMEGAS Description and Operating Detailed descriptions of MICROMEGAS are given in [1]-[3] and [7]. A two-stage parallel plate avalanche chamber has a narrow amplification gap defined by the anode plane and a cathode plane made by Ni electroformed micromesh. Several 500 LPI, 5x5 cm2 and of micro-mesh pitch (pg) with a conversion gap of 3 mm, an amplification gap of 100 mm with a strip pitch of 317 mm, have been designed and fabricated. The parallelism between the micromesh grid and the anode is maintaining by spacers of 150 mm in diameter and placing every 2 mm. They are printed on a thin epoxy substrate by conventional lithography of a photo-resistive polyamide film. The thickness of the film defines the amplification gap. This cheap and simple process allows the construction of large detectors with excellent uniformity and energy resolution over the whole surface.Figure1 shows geometric and physic descriptions of µ-MEGAS detector. Figure 1: Processes description of MICROMEGAS detector
2.1. MCROMEGAS: Concept and Configuration The geometry of MICROMEGAS detector is then, reproduced by form of cuboids. The coordinate system used is showing in the diagram below, where: − ≤≤ݔ ଶ ଶ ೄೝ ೄೝ ൞ − (1) ≤≤ݕ ଶ ଶ −∆ௌ௨௦௧௧ ≤ ∆ ≤ ݖௗ௧ Where pg is a micro-mesh pitch, p stripisthe strip pitch, ∆ௗ௧ is the drift gap and ∆ௌ௨௦௧௧ is the substrate height, see figure 2 below
144
HamidMounir and SeddikBri
Figure 2: Coordinate system and geometric parameters (a), (b) And μ-mesh Prototype (c), (d) used in MICROMEGAS Detector
2.2. Electric Field Configuration The knowledge of the shape of the electric field lines close to the micromesh is a key issue for an optimaloperation of the detector and especially for an efficient transfer of electrons to the amplification gap. Theelectric field is homogeneous in both the conversion and the amplification gap. It exhibits a funnel like shapearound the openings of the micro-grid: field lines are highly compressed towards the middle of the openings, intoa small pathway equal to a few microns in diameter. The compression factor is directly proportional to the ratioof the electric fields between the two gaps.Figure 3 displays details of the field lines near the grid used in the present test. Figure 3: Profile of the electric field lines close to the µ-Mesh in MICROMEGAS detector
Micromegas Detector: Conception and Development based on Neon gasat High Energy Physics using 55Fe X-ray Source
145
The electrons liberated in the conversion gap by the ionizing radiation follow these lines and are focused into the multiplication gap where amplification process takes place. The ratio between the Electric Field in the amplification gap and that in the conversion gap must be set at large values (>5) to permit a full electron transmission, and to reduce a part of ion cloud, produced in the avalanche, to escape into the conversion gap. 2.3. Study of Electron Avalanche Amplification When electrons are subjected to an electric field more than ten kilovolts per centimeter, they acquire enough energy to ionize the atoms to gas; new electrons are created and will in turn be accelerated. A process chain snaps and gives rise to an avalanche. It is necessary to add the noble gas mixture forming the basis of a certain proportion of poly atomic gases known as quencher. The effects of this gas acting on the electron transport, it will also play a role in the development of the avalanche. It also will absorb electromagnetic radiation produced during the avalanche. Numerous levels of excitations, rotations and vibrations will enable it to absorb photons and thus limit the development of the avalanche. This stabilizes the operation of the detector and limits the development of avalanche process leading to a divergent and a hamstring problem. On the other hand, the energy radiated during the collection of ions on the cathode which could cause a new signal will be absorbed by these molecules. The avalanche phenomenon is characterized by the inverse mean free path for ionization is usually noted α or (Tw). This coefficient is called the first Townsend coefficient (amplification factor). It represents a number of electron-ion pairs produced per unit length of drift. In general α increase with the electric field if this field exceeds a certain threshold. An approximation of the first Townsend coefficient as a function of field strength E is given by several equations proposed to describe this ratio since 1941[8] - [9]. ZASTAWNY [10] suggests, to this simple model: ߙ = ݔ݁ܣቀ− ܤቁ (2) ܧ Where p is the atmospheric pressure, the coefficients A and B depend on the gas mixture used; these parameters must be adjusted for each gas mixture, considering the cross section of gases used [11] and [12]. Note that these coefficients as a function of electric field and pressure. The gain of a detector can be noted for a length of amplification d as ௗ
ݔ݁ = ܩቆන ߙ(ݔ݀ )ݔቇ
(3)
In this case of approximation of electric field uniform amplification in the amplification region, α is constant and simply written as )݀ߙ(ݔ݁ = ܩ (4)
3. Simulation Model The analytical study of our system leads us to extract some equations that lead us to model our detector described above. The simulation model of the physical processes occurring within MICROMEGAS is based on the MATLAB Program that is the tool of simulation, and more effective for our work. In the general case, there are three types of parameters affecting the detector: the parameters related to the chamber (geometry), the gas parameters (diffusion, gain, mixing coefficients), the parameters of the trace (angle, energy, signal, type of particles). The figure 4 shows the principle operating and the simulator used for MICROMEGAS detector modeling. 3.1. Parameters Configuration and Characterization of MICROMEGAS Table 1 below, shows some input output parameters used in the MICROMEGAS detector modeling.
146
HamidMounir and SeddikBri
Input output parameters of micromegas detector Gas Mixture
Radiation- Source Fe X-Rays (5, 9 keV)
Output parameters Drift Velocity (vd) Longitudinal Diffusion (DL) Transverse Diffusion (DT) Townsend Coefficient(Twi) Amplification Gain (Gi) Spatial Resolution Temporal Resolution
55
Input parameters Amplification Field E (kV/cm) Magnetic Field B (T) Gas Mixture Mixing coefficients (Ai, Bi, i=1 to 5) Temperature T (normal condition) Pressure p (atmospheric) Drift Field E (V/cm)
Neon- isobutane (Ne-iC4H10) Neon-dimethylEther (Ne-DME
Table 1:
3.2. Simulation of the First Townsend Coefficients From equation (2), and table 2 shown below [11], we can estimate the first amplification coefficient (Twi, i=1 to 5) for different proportions of each gas mixture; using X-rays (5, 9keV) as a radiation source, the same program remains what we need to change the values of the coefficients Ai and Bi[11][13]. These parameters are adjusted by experience [14]. Table 2:
Mixing coefficients for different gas mixtures of an amplification gap of 100 µm Gas MixtureProportion (%)A (cm-1)B (kV cm-1) 6 3400 11 3700 20 4400 30 5600 37 6900 6 3200 11 3300 30 3800 50 5800
Ne-isobutane
Ne-DME
48.90 56.80 74.00 97.80 118.80 46.00 50.70 65.30 99.80
3.2.1. Simulation of the First Townsend Coefficients based on Ne-isobutane The amplification coefficient of calculations results Twi(first Townsend coefficient, (i = 1 to 5)) for different proportions of Neon-isobutane are shown in figure 5. Figure 5: First Townsend Coefficient for Neon-isobutane 7000
Townsend Coeff [1/cm]
6000 5000 4000 3000 Ne-isobutane(6%) Ne-isobutane(11%)
2000
Ne-isobutane(20%) 1000
Ne-isobutane(30%) Ne-isobutane(37%)
0
025
100
200
300
400
500
600
700
800
900
1 000
E[kV/cm]
From figure 5, it seems that a proportion of Ne-isobutane given, the Townsend coefficient increases with increasing electric field and reaches a saturation value range as the field which is more important.
Micromegas Detector: Conception and Development based on Neon gasat High Energy Physics using 55Fe X-ray Source
147
3.2.2. Simulation of the First Townsend Coefficients based on Ne-Dimethyl-Ether The results simulation of first Townsend coefficients for different proportion based on gas mixture NeDME, are shown in figure 6. Figure 6: First Townsend Coefficient for Neon-Dimethyl-Ether 7000
Townsend Coeff [1/cm]
6 000 5 000 4 000 3 000 Ne-DME(6%)
2 000
Ne-DME(11%) Ne-DME(30%)
1 000 0
Ne-DME(50%) 025
100
200
300
400
500
600
700
800
900
1 000
E[kV/cm]
The figure 6 shows that the first Townsend coefficient Twi (i = 1 to 4) increases with the growth of the electric field and that this coefficient increases with the increase of proportion of NeDimethyl-Ether. 3.2.3 Comparison of Simulation Results of the First Townsend Coefficients based on (NeDME)/Ne-isobutane) with (Argon-isobutane /Argon-DME) The results of these parameters are compared to those found based gas mixture Ar-isobutane /Ar-DME in [15] - [11]. That is why we must submit all results for four gas mixtures on the same curve after we will determine the remarkable points for different proportions, clarifying the best gas mixture for more operation of the detector MICROMEGAS (temporal and spatial resolutions). Figure 7 shows the result obtained by this simulation. Figure 7: Comparison of First Townsend Coefficient for Ne-isobutane/Ar-isobutane and Ne-DME/Ar-DME
148
HamidMounir and SeddikBri 7000
6000 dTw3(isob)
T o w n s e n d C o e ff [1 /c m ]
5000
dTw1(isob)
4000
dTw3(DME)
dTw2(isob)
3000
dTw1(DME) dTw2(DME)
Ar-Isobutane( 6%) Ne-Isobutane( 6%) Ar-DME( 5%) Ne-DME( 6%) Ar-Isobutane( 10%) Ar-Isobutane( 11%) Ar-DME( 10%) Ne-DME( 11%) Ar-Isobutane( 30%) Ne-Isobutane( 30%) Ar-DME( 30%) Ne-DME( 30%)
2000
1000
0 0
100
200
300
400
500
600
700
800
900
1 000
E[kV/cm]
The Townsend coefficient simulated for different gas mixtures as a function of the induction field are shown in Figure 7, where we observe a good agreement for larger fields (above25 kV/cm). All gas mixtures show roughly the some asymptotic behavior for their Townsend coefficient: this coefficient reaches a plateau for sufficiently high induction electric field. A good choice of the gas mixtures guarantees an approximately constant Townsend coefficient for a wide range of the electric field, which is important for obtaining good position resolution and stable operation. Some of the parameterswhichare seton the curveare definedby the following equations: dT୵భ (isob) = T୵ (Ar − isobutane(6%)) − T୵ (Ne − isobutane(6%)) ۓ dT୵భ (DME) = T୵ (Ar − DME(6%)) − T୵ (Ne − DME(6%)) ۖ ۖ dT୵మ (isob) = T୵ (Ar − isobutane(10%)) − T୵ (Ne − isobutane(11%)) (5) (Ar (Ne dT (DME) = T − DME(10%)) − T − DME(11%)) ୵మ ୵ ୵ ۔ ۖ ۖ dT୵య (isob) = T୵ (Ar − isobutane(30%)) − T୵ (Ne − isobutane(30%)) dT୵య (DME) = T୵ (Ar − DME(30%)) − T୵ (Ne − DME(30%)) ە The table below present the parameters (cm-1) of equation (5), are calculated by MATLAB Program. Table 3:
Parameters of comparison between different gas mixtures calculated by MATLAB Program
E(kV/cm) 25 35 45 55 65 75 85 95 100 150 200 250 300 350 400
dTw1(isob) -98.10 -57.20 19.80 105.60 189.00 265.80 335.10 397.20 425.70 640.20 772.60 861.50 924.90 972.50 1009.40
dTw1(DME) 86.72 75.72 51.52 25.07 0.03 -22.56 -42.61 -60.32 -68.38 -127.74 -163.43 -187.02 -203.71 -216.12 -225.71
dTw2(isob) -92.66 -93.20 -58.84 -10.14 42.18 93.31 141.27 185.40 206.00 366.37 469.32 539.73 590.62 629.04 659.03
dTw2(DME) 90.59 96.63 86.16 70.10 52.94 36.44 21.20 7.35 0.95 -47.80 -78.22 -98.69 -113.34 -124.31 -132.83
dTw3(isob) -52.30 -107.60 -134.80 -130.70 -0.1038 -062.50 -13.30 39.90 67.00 321.90 521.30 671.80 787.40 878.20 951.20
dTw3(DME) -14.10 -4.97 14.03 36.48 59.13 80.59 100.35 118.33 126.68 190.83 231.55 259.25 279.23 294.28 306.02
Micromegas Detector: Conception and Development based on Neon gasat High Energy Physics using 55Fe X-ray Source 450 500 550 600 650 700 750 800 850 900 950 1000
1038.80 1062, 90 1082.90 1099.80 1114.20 1126.70 1137.70 1147.30 1155.90 1163.50 1170.40 1176.60
-233.34 -239.55 -244.71 -249.05 -252.76 -255.97 -258.77 -261.24 -263.43 -265.38 -267.14 -268.73
683.07 702.77 719.20 733.11 745.04 755.38 764.43 772.41 779.51 785.86 791.57 796.74
149 -139.64 -145.19 -149.81 -153.72 -157.06 -159.95 -162.48 -164.71 -166.69 -168.46 -170.05 -171.48
1011.00 1060.90 1103.00 1139.10 1170.40 1197.70 1221.70 1243.10 1262.10 1279.30 1294.80 1308.90
315.42 323.12 329.53 334.97 339.62 343.66 347.19 350.30 353.07 355.54 357.77 359.79
We find thatfor higher fields (above to 100 kV/cm);this model does notcorrectlyreproduce thedata[13]. 3.3. Simulation of the Amplification Gain From equations (2) and (4), we get the gain as a function of electric field E. For this, we consider the distance of the amplification gap of d=100 μm, the field E varies between 30 and 100 kV/cm, A and B are defined in table 2, these parameters depend on gas mixtures. We can optimize the amplification gain for different gas mixtures; it remains the same program that we must change the values of the coefficients Ai and Bi. 3.3.1. Simulation of the Amplification Gain based on Neon-isobutane Figure 8 present the simulated gain (GS) that is on the left, against that is on the right which is the experimental gain (GM) [11] - [15]. Figure 8: Simulated (GS) and Experimental (GM) gain based on Ne-isobutane for different proportions G a in
500 000
400 000
300 000
200 000
G1Ne-isobutane(6%) G1Ne-isobutane(11%) G1Ne-isobutane(20%) G1Ne-isobutane(30%) G1Ne-isobutane(37%)
(GM)
(GS)
100 000
0 30
35
40
45
50
55
60
65 E [kV/cm]
We note that the gain decreases with increasing amplification field by increasing the proportion of Ne-isobutane in addition; we note that the simulated gain is compatible with the measured gain. Although Neon gas is much lighter than Argon [15], we also used it as the basis of mixtures. The number of electrons produced in Neon average is three times lower than in Argon. This disadvantage can be reduced through the use of a greater proportion of isobutane. This gas has some advantages, including lower diffusion coefficient and the Townsend coefficient most important. So, hopefully we achieve greater gains and beam tests in a better spatial resolution. The maximum size of the avalanche in this case, considering one X-ray photon produced 169 electrons, is 9.107 electrons, which is close to
150
HamidMounir and SeddikBri
the limits of size. We still see a decrease in the maximum gain and higher operating voltages when we increase the proportion of quencher. 3.3.2. Simulation of the Amplification Gain based on Neon-Dimethyl-Ether (Ne-DME) The results obtained for Neon-Dimethyl-Ether are shown in figure 9 below. Figure 9: Simulated (GS) and Experimental (GM) gain based on Ne-DME for different proportions G a in
350000 300000 250000
G1 Ne-DME(6%) G2 Ne-DME(11%) G3 Ne-DME(30%) G4 Ne-DME(50%)
200000
(GS)
(GM)
150000 100000 50000 0 30
35
40
45
50
55
60 65 E [kV/cm]
We see from this figure that the gain decreases by increasing proportions of the quencher, this gain is closer to themeasuredgain. The DME is a quencher to reduce further dissemination. The gain of the detector with this mixture is very satisfying.
4. Summary and Concluding Remarks The detector operation is the result ofa detailed analysis of the physical phenomena.The calculation of the magnitudes of amplification (gain and Town send coefficient), depending on the thickness of the space, allowed amplification to determine the thickness of the space of the optimum gain, minimizing theeffect of slightgeometrical variationsof the micro-grid.An amplificationgap of100micronsis optimal ifit is desired to minimize the influence of the defects of planarity of the grid relative to the tracks. Simulations of μ-mega focused on the study of gas mixtures. To conduct this study we used a55 Fe source emitting photons of 5.9keV.For agap width of 100micronsand amicro-grid prototype 500LPI5x5cm2, we have shown that the coefficient is reduced by increasing Town send proportions which causes the quencher to the low diffusion in the drift space. Further more, the maximum gain of the neon-based mixtures is higher compared to argon mixtures that are found in the other reference, and exceeds 105, it can be as high as5.105 fora neon mixture of6% is obutane.We also noted in this case an increase of the maximum charge in the avalanche. It is of the order of 108 electrons (Raetherlimit). In order to optimize the spatial resolution of the detector for the localization of the particles, we have searched for gas mixtures with low diffusion; leading to a lower Town send coefficient. This led us to increase the proportion of the quencherinne on-based mixtures, which results in a lower maximum gain. In conclusion, the good agreement between experimental measurements and simulation suggests that our simulation model is all the more satisfying. However, if there are disagreements, the differences remain well below10%. We are tempted to interpret these observed differences, as it is in any case of systematic uncertainties inthe simulation, andphysical phenomenaoccurringwithin the detect or were not taken into account in our simulation program: as an example, the propagation of photonsin the avalanche contributing to the extension of the lateral size of the avalanche, and the Pening effect, the discharge phenomenon, ..., these phenomena are taken into account as a perspective in future work.
Micromegas Detector: Conception and Development based on Neon gasat High Energy Physics using 55Fe X-ray Source
151
In brief, through the various comparisons between real data and simulated data, we achieved a better understanding of Micro-mega chambers by improving its performance.
References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15]
Y. Giomataris, Ph. Rebourgeard, J.P. Robert and G. Charpak, (1996), Nucl. Instr. Meth. A376, PP: 29-35. Sauli F. (2002) Nucl. Instrum. Methods. A477, PP: 1-7. G. Charpak et al. (1998) Nucl. Instrum. Methods, A412, PP: 47-60. Y. Giomataris et al. (1996) Nucl. Instrum. Methods. A376, PP: 29-35. Y. Giomataris. (1998) Development and prospects of the new Gaseous detector MICROMEGAS, Nucl. Instrum. Meth. A 419, P 239. A. Sarvestani et al. (1997) Nucl. Instr. Meth. A410, P 238. G. Barrouch et al.(1999) Nucl. Instr. Meth. A423, PP: 32-48. S. A. Korff.(1955) Electrons and Nuclear Counters, Van Nostrand, Princeton. D. Mazed and M. Baaliouamer. (1999) a semi-microscopic derivation of gas gain formula for proportional counters. Nucl. Instr. and Meth. In Phys. Res. A 437, PP: 381-392. A. Zastawny. (1997) Standardization of gas amplification description in proportional Counters. Nucl. Instr.and Meth. In Phys. Res. A 385, PP: 239-242. M. PUIL Geal. (2000) the development of MICROMEGAS, a new detector gaseous positron micro-grid. PhD Thesis, University of Caen. Irene Peres.(19900) Kinetic modeling of electron and ion in the cylindrical proportional Counters. PhD. Thesis, University of PAUL SABATIER, TOULOUSE. A. Sharma et F. Sauli, (1993) Nuclear Instruments and Methods A334, PP: 420-424, Nuclear Instruments and Methods in Physics Research, A572, PP: 999–1003, 2007. H. Mounir, S. Bri, M. Haddad. (2013) MICROMEGAS PERFORMANCE BASED IN ARGON-ISOBUTANE AND ARGON-DIMETHYL-ETHER, ESJ, VOL.4, December.
