rus
Issue #2/2025
A. I. Maldzigati, G. V. Fedotova, D. G. Samkanashvili
Channel Secondary Electron Multiplier with Extended Dynamic Range
Channel Secondary Electron Multiplier with Extended Dynamic Range
DOI: 10.22184/1993-7296.FRos.2025.19.2.148.153
The results of efforts on extending the dynamic range of a channel-type secondary-electron multiplier are reported. Design and process solutions are described, which allowed to achieve an increase of the output current value in linear mode. Preliminary electron scrubbing of the detector is proposed.
The results of efforts on extending the dynamic range of a channel-type secondary-electron multiplier are reported. Design and process solutions are described, which allowed to achieve an increase of the output current value in linear mode. Preliminary electron scrubbing of the detector is proposed.
Channel Secondary Electron Multiplier With Extended Dynamic Range
A. I. Maldzigati, G. V. Fedotova, D. G. Samkanashvili
VTС Baspik, Vladikavkaz, Russia
The results of efforts on extending the dynamic range of a channel-type secondary-electron multiplier are reported. Design and process solutions are described, which allowed to achieve an increase of the output current value in linear mode. Preliminary electron scrubbing of the detector is proposed.
Keywords: secondary electron multiplier, channel electron multiplier, dynamic range, mass-spectrometer, analog mode, pulse counting mode, electron scrubbing
Article received: 13.11.2024
Article accepted: 23.01.2025
Currently, channel-type secondary electron multipliers (SEMs) are widely used as detectors in mass spectrometry. The vast majority of channel SEMs are made of glass, although some are made of coated ceramic materials or are a combination of glass and ceramic. The advantages of these detectors are their compactness, low power consumption, long lifetime, low dark noise, no need for a voltage divider, the ability to operate at low vacuum levels and to withstand multiple cycles between vacuum and atmosphere. All these factors make it possible to create very competitive detectors for various applications on the basis of channel SEMs [1].
A critical feature for the detectors used in mass spectrometry is their dynamic range, which is a measure of the maximum count rate in pulse counting mode or the maximum linear output current in analog mode.
A classic representative of channel SEM is SEM‑6 (Fig. 1) from VTC Baspik, which has proved to be a simple and reliable detector of charged particles, ultraviolet and X-ray radiation. However, SEM‑6 was designed for operation in the pulse counting mode, which nowadays restricts the scope of its application, since most modern SEMs operate in mass spectrometers in analog mode, which requires a larger output current to extend the linearity range.
The output current in channel SEMs maintains linearity up to a value equal to about 10% of the conduction current value [1]. This constraint is associated with the fact that the conduction current supplies the electrons needed to compensate for the positive charge generated after the electron avalanches leave the output of the channeltron, supporting the secondary electron emission process. In turn, the conduction current depends on the channel resistance and increases with decreasing resistance.
For a standard SEM‑6, according to the specifications, the channel supply current can vary from 5 µA to 20 µA at Usupply = 4 kV, hence the linear output current can be as high as 2 µA.
To obtain high channel conduction current it is necessary to reduce the channel resistance, and this can be achieved as follows:
change the thermal hydrogen treatment mode for deeper glass reduction;
decrease channel aspect ratio by using a shorter channeltron.
This task was accomplished in the course of modernization by:
developing a technology for thermal hydrogen reduction of channeltrons to the resistance level of about 1 · 107 Ohm, which is an order of magnitude less than that of the standard SEM‑6;
optimizing aspect ratio, which makes it possible to have the channeltron resistance after thermal hydrogen reduction equal to R = (0.8–0.9) · 107 Ohm.
In addition, the need to interrupt the reduced layer on the outer surface of the channeltron has been established, since the current flowing on the outer surface is not involved in reducing the conductivity of the channel after the electron avalanche has passed, but causes heating and unstable operation. In practice, interruption of conductivity on the outer surface was implemented by mechanical removal of the reduced layer on the outer surface within 1–2 mm width with the help of an abrasive material – a “groove”.
Fig. 2 shows a curve illustrating channeltron heating up with and without a “groove” during operation. The curve shows that a channeltron with a high conduction current (about 300 μA) without a “groove” heats up much more than the same channeltron with a “groove”. Heating up to 76 °C has occurred without external heating, and in application systems there may still be heating from an external source, hence the device may heat up to unacceptable values.
Fig. 3 shows a plot of the conduction current vs. operating time for the same device before and after making a “groove” on the outer surface.
From the plot in Fig. 3, it can be seen that the conduction current increased to a value of more than 1000 µA on the channeltron without the “groove”. In addition to heating up and unstable operation, such high conduction currents increase the power consumption of the device, which in some cases is also critical.
After interrupting the reduced layer on the channeltron outer surface, the channel resistance increased to R = 3 · 107 Ohm and allowed us to obtain a channel conduction current of 100 µA at Usupply = 3 kV.
Fig. 4 shows the plateau characteristic for the prototype. The plateau characteristic which is the dependence of the count rate on the voltage across the channeltron allows to determine the operating point of the device, where the plateau is reached in the pulse counting mode. Additional voltage increases improve the gain, but the count rate remains essentially constant.
The curve in Fig. 4 shows that the pulse count rate reaches a plateau at Usupply = 1.8 kV and does not change until Usupply = 3 kV or more. The optimal operating point is about 50–100 volts behind the knee of the curve in Fig. 4. As the multiplier ages, the knee shifts to the right and the voltage must be increased.
The dependence of gain on output current enables the determination of the maximum linear output current in analog mode. Fig. 5 shows a plot of gain as a function of output current for the prototype at different values of the initial gain M: 6.3∙107 and 1∙107. It can be seen that the smaller the gain, the larger the linear mode output current.
Fig. 6 shows for comparison similar dependencies for foreign channeltrons such as the six-channel Magnum channeltron and a typical single-channel channeltron (Channeltron 4700). [1]. It can be seen that the performance of the developed SEMs at appropriate gain exceeds in linearity the performance of typical single-channel foreign analogs.
When the channeltron is initially put into operation, a rapid decrease in gain is observed due to desorption of gases adsorbed by the working surface under the action of electron bombardment. The hit molecules are ionized by the passing electron flux, and the resulting positive ions move in the opposite direction. Colliding with the channeltron walls, these ions release additional electrons generating spurious pulses due to which in the first period of operation the output signal has an increased level, which at first decreases and then stabilizes as the channeltron is cleaned from surface contaminants. Therefore, to obtain stable output readings related only to the level of the incoming signal, the channeltron must be treated before operation. This is accomplished by operating the channeltron at some specified voltage with an input signal for a short period of time to degas the device.
Performance behavior during electron scrubbing of the detector has been investigated and preconditioning before measurements is proposed.
Fig. 7 shows the time dependence of the gain of the developed ion detectors at different initial gain values and different input signal levels. The value of the transmitted charge is calculated from the level of the measured output current.
It can be seen that as the amount of output current (hence the transmitted charge) increases, the gain stabilization occurs in a shorter period of time. A typical mode of “conditioning” of the devices consists in applying a voltage to the detector corresponding to a gain of about M = 5∙107 with an input signal providing an output current of 2.5–3 µA for 30–40 minutes, which corresponds to a charge collection of about 0.007 C.
As a result of implementation of the developed design and process solutions on the basis of SEM‑6M secondary-electron multiplier a modification of SEM‑6M‑1 with extended dynamic range with typical parameters given in Tab. has been developed.
Practical testing of the application of the upgraded SEM‑6M‑1 in instruments has shown the possibility of its successful application for the needs of advanced mass spectrometry.
AUTHORS
Maldzigati Alan Ilyich, Senior Engineer, laboratory of “MCP – detectors”, VTС Baspik, Vladikavkaz, Russia.
ORCID 0009-0006-3267-9558
Fedotova Galina Vasilyevna, Head of the laboratory of “MCP – detectors”, VTС Baspik, Vladikavkaz, Russia.
David Gennadievich Samkanashvili, Director of Science and Innovation, VTС Baspik, Vladikavkaz, Russia.
Contribution of the members
of the author’s team
The article is based on the work of all members of the author’s team.: Maldzigati A. I. – conducting experiments, processing and discussing the results; Fedotova G. V. – processing and discussing the results; Samkanashvili D. G. – organization of work.
Conflict of interest
The authors declare that there is no conflict of interest. All the authors participated in the writing of the manuscript in terms of the contribution of each of them to the work and agree with the full text of the manuscript.
A. I. Maldzigati, G. V. Fedotova, D. G. Samkanashvili
VTС Baspik, Vladikavkaz, Russia
The results of efforts on extending the dynamic range of a channel-type secondary-electron multiplier are reported. Design and process solutions are described, which allowed to achieve an increase of the output current value in linear mode. Preliminary electron scrubbing of the detector is proposed.
Keywords: secondary electron multiplier, channel electron multiplier, dynamic range, mass-spectrometer, analog mode, pulse counting mode, electron scrubbing
Article received: 13.11.2024
Article accepted: 23.01.2025
Currently, channel-type secondary electron multipliers (SEMs) are widely used as detectors in mass spectrometry. The vast majority of channel SEMs are made of glass, although some are made of coated ceramic materials or are a combination of glass and ceramic. The advantages of these detectors are their compactness, low power consumption, long lifetime, low dark noise, no need for a voltage divider, the ability to operate at low vacuum levels and to withstand multiple cycles between vacuum and atmosphere. All these factors make it possible to create very competitive detectors for various applications on the basis of channel SEMs [1].
A critical feature for the detectors used in mass spectrometry is their dynamic range, which is a measure of the maximum count rate in pulse counting mode or the maximum linear output current in analog mode.
A classic representative of channel SEM is SEM‑6 (Fig. 1) from VTC Baspik, which has proved to be a simple and reliable detector of charged particles, ultraviolet and X-ray radiation. However, SEM‑6 was designed for operation in the pulse counting mode, which nowadays restricts the scope of its application, since most modern SEMs operate in mass spectrometers in analog mode, which requires a larger output current to extend the linearity range.
The output current in channel SEMs maintains linearity up to a value equal to about 10% of the conduction current value [1]. This constraint is associated with the fact that the conduction current supplies the electrons needed to compensate for the positive charge generated after the electron avalanches leave the output of the channeltron, supporting the secondary electron emission process. In turn, the conduction current depends on the channel resistance and increases with decreasing resistance.
For a standard SEM‑6, according to the specifications, the channel supply current can vary from 5 µA to 20 µA at Usupply = 4 kV, hence the linear output current can be as high as 2 µA.
To obtain high channel conduction current it is necessary to reduce the channel resistance, and this can be achieved as follows:
change the thermal hydrogen treatment mode for deeper glass reduction;
decrease channel aspect ratio by using a shorter channeltron.
This task was accomplished in the course of modernization by:
developing a technology for thermal hydrogen reduction of channeltrons to the resistance level of about 1 · 107 Ohm, which is an order of magnitude less than that of the standard SEM‑6;
optimizing aspect ratio, which makes it possible to have the channeltron resistance after thermal hydrogen reduction equal to R = (0.8–0.9) · 107 Ohm.
In addition, the need to interrupt the reduced layer on the outer surface of the channeltron has been established, since the current flowing on the outer surface is not involved in reducing the conductivity of the channel after the electron avalanche has passed, but causes heating and unstable operation. In practice, interruption of conductivity on the outer surface was implemented by mechanical removal of the reduced layer on the outer surface within 1–2 mm width with the help of an abrasive material – a “groove”.
Fig. 2 shows a curve illustrating channeltron heating up with and without a “groove” during operation. The curve shows that a channeltron with a high conduction current (about 300 μA) without a “groove” heats up much more than the same channeltron with a “groove”. Heating up to 76 °C has occurred without external heating, and in application systems there may still be heating from an external source, hence the device may heat up to unacceptable values.
Fig. 3 shows a plot of the conduction current vs. operating time for the same device before and after making a “groove” on the outer surface.
From the plot in Fig. 3, it can be seen that the conduction current increased to a value of more than 1000 µA on the channeltron without the “groove”. In addition to heating up and unstable operation, such high conduction currents increase the power consumption of the device, which in some cases is also critical.
After interrupting the reduced layer on the channeltron outer surface, the channel resistance increased to R = 3 · 107 Ohm and allowed us to obtain a channel conduction current of 100 µA at Usupply = 3 kV.
Fig. 4 shows the plateau characteristic for the prototype. The plateau characteristic which is the dependence of the count rate on the voltage across the channeltron allows to determine the operating point of the device, where the plateau is reached in the pulse counting mode. Additional voltage increases improve the gain, but the count rate remains essentially constant.
The curve in Fig. 4 shows that the pulse count rate reaches a plateau at Usupply = 1.8 kV and does not change until Usupply = 3 kV or more. The optimal operating point is about 50–100 volts behind the knee of the curve in Fig. 4. As the multiplier ages, the knee shifts to the right and the voltage must be increased.
The dependence of gain on output current enables the determination of the maximum linear output current in analog mode. Fig. 5 shows a plot of gain as a function of output current for the prototype at different values of the initial gain M: 6.3∙107 and 1∙107. It can be seen that the smaller the gain, the larger the linear mode output current.
Fig. 6 shows for comparison similar dependencies for foreign channeltrons such as the six-channel Magnum channeltron and a typical single-channel channeltron (Channeltron 4700). [1]. It can be seen that the performance of the developed SEMs at appropriate gain exceeds in linearity the performance of typical single-channel foreign analogs.
When the channeltron is initially put into operation, a rapid decrease in gain is observed due to desorption of gases adsorbed by the working surface under the action of electron bombardment. The hit molecules are ionized by the passing electron flux, and the resulting positive ions move in the opposite direction. Colliding with the channeltron walls, these ions release additional electrons generating spurious pulses due to which in the first period of operation the output signal has an increased level, which at first decreases and then stabilizes as the channeltron is cleaned from surface contaminants. Therefore, to obtain stable output readings related only to the level of the incoming signal, the channeltron must be treated before operation. This is accomplished by operating the channeltron at some specified voltage with an input signal for a short period of time to degas the device.
Performance behavior during electron scrubbing of the detector has been investigated and preconditioning before measurements is proposed.
Fig. 7 shows the time dependence of the gain of the developed ion detectors at different initial gain values and different input signal levels. The value of the transmitted charge is calculated from the level of the measured output current.
It can be seen that as the amount of output current (hence the transmitted charge) increases, the gain stabilization occurs in a shorter period of time. A typical mode of “conditioning” of the devices consists in applying a voltage to the detector corresponding to a gain of about M = 5∙107 with an input signal providing an output current of 2.5–3 µA for 30–40 minutes, which corresponds to a charge collection of about 0.007 C.
As a result of implementation of the developed design and process solutions on the basis of SEM‑6M secondary-electron multiplier a modification of SEM‑6M‑1 with extended dynamic range with typical parameters given in Tab. has been developed.
Practical testing of the application of the upgraded SEM‑6M‑1 in instruments has shown the possibility of its successful application for the needs of advanced mass spectrometry.
AUTHORS
Maldzigati Alan Ilyich, Senior Engineer, laboratory of “MCP – detectors”, VTС Baspik, Vladikavkaz, Russia.
ORCID 0009-0006-3267-9558
Fedotova Galina Vasilyevna, Head of the laboratory of “MCP – detectors”, VTС Baspik, Vladikavkaz, Russia.
David Gennadievich Samkanashvili, Director of Science and Innovation, VTС Baspik, Vladikavkaz, Russia.
Contribution of the members
of the author’s team
The article is based on the work of all members of the author’s team.: Maldzigati A. I. – conducting experiments, processing and discussing the results; Fedotova G. V. – processing and discussing the results; Samkanashvili D. G. – organization of work.
Conflict of interest
The authors declare that there is no conflict of interest. All the authors participated in the writing of the manuscript in terms of the contribution of each of them to the work and agree with the full text of the manuscript.
Readers feedback
