RU200541U1 - Well type gas electron multiplier - Google Patents

Abstract

The utility model relates to devices for detecting charged particles and ionizing radiation and can be applied in many areas of experimental physics and applied research. A well-type gas electron multiplier (GEM) contains a first electrode made on a board of a one-sided foil dielectric with through holes; a resistive second electrode made of diamond-like carbon (DLC), deposited in the form of a continuous film on the printed grid of the additional board, on which the reading elements are also located; moreover, the boards are combined into one multilayer in such a way that the resistive layer is the bottom of the holes, and the reading elements made with a pitch of the holes are located opposite the holes. What is new is that a low-resistance impedance is introduced between the first and second electrodes. The low impedance impedance is made, for example, by a capacitor with a capacity of at least 1000 pF. With the thickness of the first board, for example, 500 μm, with the pitch and diameter of the GEM holes, for example, 500 μm and 200 μm, respectively, and the lattice pitch of 0.5 mm, the surface resistance of the APU resistive layer is selected, for example, 25 MΩ / square. The technical effect of the PM is to increase the response rate by suppressing the ion tail in the output signal, which shortens the pulses on the reading electrode and reduces the dead time of the recording channels, reduces the likelihood of pulse overlap and miscalculations at high loads. 2 wp f-ly, 6 dwg

Description

The utility model (UM) refers to devices for detecting charged particles and ionizing radiation and can be applied in many areas of experimental physics and applied research. The gas electron multiplier (GEM, abbreviation GEM - Gas Electron Multiplier) is a key element of gas-discharge detectors designed to register charged particles and other ionizing radiation. Signal multiplication occurs as a result of avalanche multiplication of electrons in the holes. The proposed well-type GEM can be used under extremely harsh radiation conditions in calorimeters, as well as in large-area muon chambers, displacing other microstructure detectors. In addition, on the basis of the proposed GEM, detectors of thermal and cold neutrons can be created.

A typical detector based on a GEM contains three units: a cathode with a cathode (working or drift) gap and an amplifying element - GEM with an anode (induction) gap. Primary electrons drift in the working gap to the holes of the GEM, in the holes are accelerated under the action of an electric field to an energy at which avalanche ionization is possible. The output receives a charge, amplified for its reliable registration. There are many methods and schemes for registering coordinates, for example, strips (metal strips) or pads (metal pads) are created on the reading electrode, the number of which gives the desired coordinates of the charge.

The GEM was first described in [1] (Sauli. F, GEM: a new concept for electron amplification in gas // Nuclear Instruments and Methods, A 386 (1997), p.531) and patent [2] (Sauli F, Radiation detector of very high performance // US006011265A). Structurally, this GEM is made on a polyimide film (Kapton) 50 microns thick, covered on both sides with copper foil, in which a lot of through holes with a diameter of 50-70 microns with a pitch of 140 microns and placed in a hexagonal geometry are made. Gas electron multipliers of this design are called "thin", they are made by photolithography and chemical etching of both metal and canton.

The advantage of the new detector is its high speed of response, up to 10 ⁸ Hz / cm ² . The disadvantage is the significant electrical capacitance between the plates of the capacitor formed by the electrodes. Since this capacitance is charged, the electrical energy accumulated on the capacitor can damage the device in case of a breakdown. With a detector size of 100 × 100 mm ² and a film thickness of 50 microns, the interelectrode capacitance is 8000 pF, the energy stored by such a capacitor is 1 mJ at a voltage of 500 V. In the event of breakdown (and breakdown is caused by strongly ionizing nuclei particles formed in the structure under the influence of space radiation), this energy is released in the spark channel in the hole in the form of heat, significant heating of the film, melting of the electrode material and failure of the GEM is possible.

, чем у тонких ГЭУ [3] (Breskin A., et al. А concise review on THGEM detectors // Nucl. Instrum. Meth. A 598 (2009), 107). Преимуществом толстых ГЭУ является простота проектирования и относительно низкая стоимость изготовления в промышленности печатных плат. Однако накопленная электрическая энергия на образованном электродами конденсаторе не меньше чем в тонких ГЭУ, т.к. с увеличением толщины уменьшается емкость, но увеличивается рабочее напряжение, а энергия пропорциональна квадрату напряжения. Накопленная на межэлектродной емкости электрическая энергия в случае пробоя может вывести прибор из строя.Known "thick" GEM (English abbreviation THGEM - thick GEM), made on a double-sided foil glass-fiber laminate using the technology of production of printed circuit boards with a dielectric thickness of 10 times

than in thin GEM [3] (Breskin A., et al. A concise review on THGEM detectors // Nucl. Instrum. Meth. A 598 (2009), 107). The advantage of thick power plants is the simplicity of design and the relatively low cost of manufacturing printed circuit boards in the industry. However, the accumulated electrical energy on the capacitor formed by the electrodes is not less than in thin GEMs, since with increasing thickness, the capacitance decreases, but the operating voltage increases, and the energy is proportional to the square of the voltage. The electric energy accumulated on the interelectrode capacitance in the event of a breakdown can damage the device.

Another disadvantage of a thick GEM is the presence of an ionic "tail" in the signal, which broadens the pulse to ~ 1 μs, leads to overlapping of pulses and to miscalculations at high loads; loss of efficiency.

, которая нанесена на отдельную плату из стеклотекстолита толщиной 0.1 мм. Резистивный слой нанесен сверху на медную печатную решетку в виде клетки с размером ячейки 1 см², которая используется, с одной стороны, для подачи напряжения на резистивный слой, с другой - для стекания заряда с него. Составляющие конструкцию платы объединены склеиванием в одну многослойную плату таким образом, что резистивный слой является дном отверстий, проводники решетки размещены между отверстиями, а считывающие годы - напротив отверстий ГЭУ.Well-known GEM (Well-type GEM was first proposed in: Bellazzini R., et al. The WELL detector. Nucl. Instr. And Meth. A423, 1999, P.125-134. The proposed detector with metal electrodes is unreliable, like GEM. ) type (eng. well), in which holes are not through, but blind - well type [4] (Arazi L., et al. Beam Studies of the Segmented Resistive WELL: a Potential Thin Sampling Element for Digital Hadron Calorimetry // arXiv : 1305.158, Nucl. Instr. And Methods A 732, 2013, p. 199). The cited GEM of a well type was made through drilling a board made of one-sided fiberglass with a thickness of 0.4 mm, where the foil is the first electrode of the GEM, the diameter of the holes is 0.5 mm, with rims of 0.1 mm around the holes with the placement of holes in a square geometry with a pitch of 0.96 mm. The second electrode, which serves as an anode, is a resistive electrode in the form of a thin film of a mixture of graphite and epoxy resin with linear surface resistance

, which is applied to a separate board made of fiberglass, 0.1 mm thick. The resistive layer is applied on top of a copper printed lattice in the form of a cell with a cell size of 1 cm ² , which is used, on the one hand, to supply voltage to the resistive layer, and on the other, to drain the charge from it. The boards constituting the structure are combined by gluing into one multilayer board in such a way that the resistive layer is the bottom of the holes, the lattice conductors are placed between the holes, and the reading years are opposite the GEM holes.

The disadvantage of the detector is the low speed of ~ 10 ⁵ Hz / cm2, due to the following factors: a rather long distance of charge drainage, 0.5 cm, and the presence of an ion "tail" in the signal, which is induced by ions moving in the hole, which also broadens the pulse, leads to superposition pulses at high loads.

The prototype of the claimed invention is a well-type GEM [5] (Kashchuk AP and other well-type gas electron multiplier (GEM). RF patent for utility model No. 198153). The device is constructed as follows: it contains the first electrode made on a board made of a one-sided foil-clad dielectric with through holes; a second diamond-like carbon (DLC) electrode deposited in the form of a continuous film on the printed grating of the additional board, on which the reading elements are also placed; moreover, the boards are combined into one multilayer board in such a way that the resistive layer is the bottom of the holes, the conductors of the printed lattice surround the group of GEM holes, and the reading elements made with a pitch of holes are located opposite the holes.

- наносекунды, существенно снижается средний ток и мощность разряда.In the case of an electrical breakdown in a GEM with a resistive electrode made of diamond-like carbon, a "local" capacitance is discharged, in the limit it is the capacitance of the nearest vicinity of the hole. This capacitance can be fractions of a pF, respectively, the time constant of the chain of discharge of this capacitance to the plasma in the hole

- nanoseconds, the average current and discharge power are significantly reduced.

The device is reliable due to the AAC, but the speed of operation is limited by two factors: the time for electrons to drain from the resistive surface onto the conducting lattice and the ion tail in the output signal. The charge drain time can be shortened by choosing the linear resistance of the resistive layer and the lattice pitch, then the ion tail will be the dominant factor limiting the response speed.

The technical effect of the PM is to increase the response rate by suppressing the ion tail in the output signal, which shortens the pulses on the reading electrode and reduces the dead time of the recording channels, reduces the likelihood of pulse overlap and miscalculations at high loads.

The technical effect is achieved by the fact that in the well-known GEM of the well-type, containing the first electrode made on a board made of a one-sided foil-clad dielectric with through holes, a resistive second electrode made of diamond-like carbon (DLC), deposited in the form of a continuous film on the printed lattice of an additional board, on which reading elements are also placed; moreover, the boards are combined into one multilayer board in such a way that the resistive layer is the bottom of the holes, the conductors of the printed lattice surround the group of holes of the GEM, and the reading elements made with a pitch of the holes are located opposite the holes, new is that a low-resistance impedance is introduced between the first and second electrodes, made, for example, a capacitor with a capacity of 1000 pF.

The low impedance impedance is made, for example, by a capacitor with a capacity of at least 1000 pF.

With the thickness of the first board, for example, 500 μm, with the pitch and diameter of the GEM holes, for example, 500 μm and 200 μm, respectively, and the lattice pitch of 0.5 mm, the surface resistance of the APU resistive layer is selected, for example, 25 MΩ / square.

At the same time, the design parameters are optimized: the surface resistance of the resistive layer and the lattice pitch for a given pitch and hole diameter are chosen such that the loss of the signal amplitude at the reading electrode is minimal, and this can be obtained by preventing the charge from draining too quickly from the resistive surface onto the lattice. and with a not too "dense" lattice, i.e. with a small lattice pitch.

The proposed power plant of the well type can be made on a one-sided foil glass fiber or canton with a thickness of, for example, 500 microns with holes with a diameter of 200 microns, located with the same pitch along the X and Y axes, for example, 500 microns.

The proposed technical solution is presented in Fig. 1 and FIG. 2, where the numbers indicate: 1 - the first electrode made on a board made of a one-sided foil-clad dielectric (The material can be fiberglass, polyimide film (kapton), lavsan, fluoroplastic, polyethylene, etc.) with through holes - 2; a second electrode 3 with a resistive layer deposited on top of the conductive printed grid of the additional board; 4 - the substrate of the conductive printed grating 5; 6 - strips of the X reading electrode; 7 - insulator, for example, in the form of a prepreg (used in the technology of manufacturing printed circuit boards), 8 - strips Y of the read electrode, while strips 6 and 8 are orthogonally oriented relative to each other; insulator 9 serving as a substrate for Y strips.

FIG. 1 shows a sectional view of the device of the proposed power plant of the well type, which shows the placement of the printed conductive grid, on which the APU is applied, relative to the holes of the power plant.

FIG. 2 shows a sectional view of the device of the proposed power plant of the well type, where the placement of the reading strips X and Y relative to the openings of the power plant is shown in other layers of the multilayer structure.

FIG. 3 shows various options for surrounding the GEM holes with lattice conductors with different lattice spacing.

FIG. 4 and FIG. 5 shows the equivalent circuits of the proposed well-type GEM with a resistive electrode in the form of a thin film of APU, and this layer is specially expanded to show the linear capacities of the RC structure on the printed grating (C _m ) -second electrode and the reading electrode (C _x ). The difference between the schemes FIG. 4 and FIG. 5 in the supply of the operating voltage to the GEM: either - V is supplied to the first electrode, indicated by 1, or + V to the second electrode, indicated by 5. The voltage is supplied from a voltage source through a resistor Rv, while one of the electrodes is grounded, thus a low-resistance impedance is introduced between the first and second electrodes, which is made, for example, a capacitor with a sufficiently large capacity of 1000 pF. With such a capacitance, the impedance X _c = 1 / ωC is 10-100 Ohm at frequencies in the range of 1-10 MHz. The figures show that when a low-resistance impedance is introduced between the first and second electrodes, the current source acts on a parallel chain of two capacities, one of which, C _m , is dynamic, varying within large limits as the charge Q spreads along the resistive surface.

FIG. 6 shows the change in the electron charge density ρ (t, r) on the surface of the resistive electrode at a fixed time constant of the resistive layer RC = 4 μs. This is a function of two variables - time t, counted from the moment the point charge Q appears (see Fig. 4 and Fig. 5) on the resistive surface, and the length of the drainage radius of the charge r, measured from the center of the hole.

сигнал наводится на стрипах. Более того, на порядок короче фронт сигнала. Две основные компоненты заряда - электронная и ионная разделяются полем и при своем движении в отверстии наводят импульсы на всех электродах. Электронный заряд Q "садится" на резистивный слой 3 - дно колодца, а ионный дрейфует в обратном направлении к электроду 1. Растекающийся вдоль поверхности резистивного слоя 3 заряд электронов Q дополнительно наводит импульсы отрицательной полярности на считывающих стрипах 6 -координата X и 8 - координата Y. Импульс отрицательной полярности относительно уровня +V2 наводится на решетке 5, а на электроде 1 - импульс положительной полярности относительно уровня V=0. В соответствии с законом Кирхгофа для токов, циркулирующих в контурах, суммарный сигнал от наведенных зарядов на электродах равняется нулю.The proposed device of a well-type power plant works as follows. Suppose that zero potential V1 = 0 is applied to the first electrode 1, and potential + V2 is applied to the second electrode 3. Let V2 be such a potential that an electric field arises in holes 2, which is large enough to accelerate electrons to an energy that causes secondary ionization of the working gas atoms with the creation of new electron-ion pairs. In this case, when one electron enters the hole, an avalanche process of electron multiplication occurs in the hole. With an electron free path of several microns and several acts of ionization of gas atoms with the creation of an electron-ion pair, a sufficiently large secondary charge can be formed at the depth of the well, which is determined by the gas amplification. The gas gain in a well-type power plant is the ratio of the secondary to the primary charge. We emphasize that there is no induction gap in the well GEM and there is no secondary charge loss in the induction gap, which is inherent in GEM and THGEM, respectively, by 2-3 times

the signal is induced on the strips. Moreover, the signal front is an order of magnitude shorter. The two main components of the charge - electronic and ionic - are separated by the field and, when moving in the hole, induce pulses on all electrodes. The electronic charge Q "sits" on the resistive layer 3 - the bottom of the well, and the ionic one drifts in the opposite direction to the electrode 1. The charge of electrons Q spreading along the surface of the resistive layer 3 additionally induces pulses of negative polarity on the readout strips 6 - the X coordinate and 8 - the Y coordinate A pulse of negative polarity relative to the + V2 level is induced on grating 5, and on electrode 1 - a pulse of positive polarity relative to level V = 0. In accordance with Kirchhoff's law for currents circulating in the circuits, the total signal from the induced charges on the electrodes is zero.

According to the solution of the telegraph equation shown in FIG. 6, after 50 ns, the charge drain radius exceeds the pitch of the GEM holes at RC = 4 μs, while the charge density Q, shown at t = 0 by the point in FIG. 4 and FIG. 5. In this case, the loss of charge Q is possible due to the flow of a part onto the lattice. This leads to a loss of signal at the readout electrode located behind the grating.

. Аналогично можно оценить погонную емкость C_x. Что касается C_m, то это - динамическая емкость, равная нулю до появления заряда Q (см. Фиг. 4 и Фиг. 5) на резистивной поверхности и увеличивается во времени по мере растекания заряда электронов по резистивной поверхности.Due to the RC structure of the resistive layer 3, the capacitance C ₀ is the "local" capacitance of the vicinity of one hole: C ₀ = C / N, where C is the total capacitance between electrodes 1 and 3, N is the number of holes. For the above values, C ₀ ≈0.01 pF, and regardless of the area of the GEM, while the total capacitance C can be thousands of pF (depending on the size of the GEM). The surface resistance of the APU with a thickness of 0.1 μm is 25

... Similarly, one can estimate the linear capacity C _x . As for C _m , this is a dynamic capacitance equal to zero before the appearance of charge Q (see Fig. 4 and Fig. 5) on the resistive surface and increases with time as the electron charge spreads over the resistive surface.

Let us show under what conditions the suppression of the ionic component in the output signal I _x occurs, which increases the response speed. FIG. 4, the role of the low-resistance impedance is played by the capacitor C ₁ . In this diagram, electrode 1 is grounded and the second electrode, labeled 5, is connected to a voltage source through a high-resistance resistor Rv. At relatively high frequencies (eg RC >> 300 kHz), the resistive layer can be represented by "local" capacitances C _x and C _m . In this case, C _m is the dynamic capacity. If C ₂ >> C _m , then the equivalent capacity is less than the smaller value C _m C ₂ / (C _m + C ₂ ) ≈C _m . Measurements show that already at C ₂ = 1000 pF, the ion tail is suppressed to a level of ~ 1% relative to the amplitude of the main electronic component.

The same is observed and when applying the voltage -V1 to the first electrode when the second electrode is connected to the ground conductor (V2 = 0), and performs the role of low-resistance impedance of the coupling capacitor C ₁ (see. FIG. 5). The circuits shown in FIG. 4 and FIG. 5 are equivalent.

When a low impedance is applied between the first and second electrodes, the low frequencies of the signal (ion tail) are branched off into the grating circuit, and the high frequencies of the fast electronic component of the signal are branched into the circuit of the read electrode. The division of the charge occurs quite effectively due to the fact that the linear capacity C _{m is} dynamic - it changes from zero to a significant value as the charge spreads over the resistive electrode. The maximum value of this capacitance is determined by the size of the grating and the thickness of the resistive layer film:

where ∈ _r is the dielectric constant of the resistive material (APM) whose thickness is d, and the area of the electrodes A, determined by the size of the lattice cell, is the pitch and thickness of the printed conductors. Numerically, we obtain C _m = 350 pF at a resistive layer thickness d = 100 nm and a lattice cell area A = 1 mm ² .

Let us show that the speed of the GEM also limits the time it takes for the electron charge to drain off the resistive surface before its recombination on the lattice. The time of complete drainage of the charge determines the radius r = s / 2 and the speed ν, the latter is determined (W. Riegler. Electric fields, weighting fields, signals and charge diffusion in detectors including resistive materials // 2016 JINST 11 P11002.), As ν = 1 / (2∈ ₀ ⋅ R):

where ∈ ₀ = 8.85 ⋅ 10 ^-12 F / m is the electrical constant, R is the surface resistance of the resistive layer, s is the pitch of the GEM holes.

и для s=0.5 мм получим Т=110 не, что соответствует быстродействию порядка 10⁷ Гц/см², много выше, чем в прототипе. Поверхностное сопротивление пленки АПУ не зависит от размеров квадрата и определяется удельным сопротивлением и толщиной пленки. Так, измерения показывают, что АПУ толщиной 120 нм обеспечивает R=25

. При уменьшении R в 10 раз до 2.5

(толщина АПУ 180 нм - зависимость нелинейная) соответственно ускоряется стекание заряда и в 10 раз увеличивается быстродействие ~10⁸ Гц/см², что, как известно, является пределом для GEM с металлическими электродами [1].With R = 25

and for s = 0.5 mm we obtain T = 110 ns, which corresponds to a speed of the order of 10 ⁷ Hz / cm ² , much higher than in the prototype. The surface resistance of the APU film does not depend on the size of the square and is determined by the resistivity and film thickness. Thus, measurements show that AAP 120 nm thick provides R = 25

... With a decrease in R 10 times to 2.5

(the thickness of the APU is 180 nm - the dependence is nonlinear), respectively, the drainage of the charge is accelerated and the response speed increases by a factor of 10 ~ 10 ⁸ Hz / cm ² , which, as you know, is the limit for GEM with metal electrodes [1].

The task has an optimum. So, if the charge spreads too quickly over the resistive surface, then part of the charge is lost, leaving not to the reading electrode, but to the grating - the shielding effect (for example, on a metal surface, the charge spreads instantly, and the reading electrode is completely shielded).

, выполненная с шагом 0.5 мм, не приводит к электромагнитному экранированию стрипов, расположенных напротив отверстий, поскольку заряд, севший на дно отверстия, имеет наибольшую плотность в центре отверстия и растекается до границ решетки не мгновенно, см. фиг. 6. Как видно, имеется промежуток времени ~5 нс, достаточный для индукции заряда Q на стрипы считывающих электродов 6 и 8. При этом пока C_m=0 (заряд еще не сел на резистивный слой), импеданс контура между вторым и первым электродами высокий (почти обрыв), а когда заряд растечется по поверхности, то емкость C_m резко увеличится, и импеданс контура станет низким. Низкие частоты, соответствующие относительно длинному ионному хвосту в сигнале I₁(t), ответвляются в динамическую емкость C_m на решетку. Таким образом, выходной сигнал I_x(t) существенно укорачивается.As the experiment shows, a lattice with AAP with surface resistance R = 25

, made with a pitch of 0.5 mm, does not lead to electromagnetic shielding of the strips located opposite the holes, since the charge that has landed at the bottom of the hole has the highest density in the center of the hole and does not spread out to the lattice boundaries instantly, see Fig. 6. As you can see, there is a time interval of ~ 5 ns, sufficient for the induction of the charge Q on the strips of the readout electrodes 6 and 8. While C _m = 0 (the charge has not yet settled on the resistive layer), the impedance of the circuit between the second and first electrodes is high (almost a break), and when the charge spreads over the surface, the capacitance C _m will increase sharply, and the loop impedance will become low. The low frequencies corresponding to the relatively long ion tail in the signal I ₁ (t) are branched into the dynamic capacitance C _m per grating. Thus, the output signal I _x (t) is significantly shortened.

Literature

1. Sauli. F, GEM: a new concept for electron amplification in gas // Nucl. Instr. and Meth., A 386 (1997), P. 531.

2. Sauli F, Radiation detector of very high performance // US006011265 A.

3. Breskin A., et al. A concise review on THGEM detectors // Nucl. Instrum. and Meth. A 598 (2009). P. 107.

4. Arazi L., et al. Beam Studies of the Segmented Resistive WELL: a Potential Thin Sampling Element for Digital Hadron Calorimetry // arXiv: 1305.1585vl, Nucl. Instr. and Methods A 732 (2013) 199-202.

5. Kashchuk A.P. and other well-type gas electron multiplier. Utility model patent No. 198153.

Claims (3)

1. Gas electron multiplier (GEM) well-type, containing the first electrode, made on a board of a one-sided foil dielectric with through holes; a resistive second electrode made of diamond-like carbon (DLC), deposited in the form of a continuous film on the printed grid of the additional board, on which the reading elements are also located; moreover, the boards are combined into one multilayer board in such a way that the resistive layer is the bottom of the holes, and the reading elements made with a pitch of the holes are located opposite the holes, characterized in that a low-resistance impedance is introduced between the first and second electrodes. 2. The device according to claim 1, characterized in that the low-impedance is made, for example, by a capacitor with a capacity of at least 1000 pF. 3. The device according to claim 1, characterized in that with the thickness of the first board, for example, 500 microns, with the pitch and diameter of the GEM holes, for example, 500 microns and 200 microns, respectively, and the grating pitch is 0.5 mm, the surface resistance of the resistive layer of the APU selectable, for example, 25 MOhm / square.

Images