Charged particle diagnostic system for lithium ultracold plasma experiments

. Charged particle detection systems are widely used in modern atomic physics experiments. In this paper we describe an experimental setup for ultracold plasma (UCP) diagnostics based on the basis of a microchannel plate-based detector (MCP). Two different readout techniques for the MCP output are described in detail. Expansion of the low-density UCP and the frequency-dependent response of the UCP to an external electric field are measured. Preliminary results are obtained.


Introduction
For several decades already, micro-channel plates (MCPs) have been established as a reliable tool for charged particle detection. Because of their unique properties such as high gain, low noise, and fast response time, they have been extensively utilized in various areas, including nuclear physics, atomic and molecular physics, and materials science [1][2][3]. An MCP consists of an array of micro-channels with diameters ranging from a few micrometers to dozens of micrometers, which makes it possible to detect charged particles with high sensitivity and resolution. The chevron stack of multiple MCPs provides high gain and amplification of secondary electrons.
There are several types of the MCP utilized for different applications. An MCP with a phosphorus screen provides high spatial resolution for various charged particle imaging applications [4,5]. By using a phosphorus screen on the output surface of the MCP, electron amplification is converted into visible light that can be detected by means of a camera or by means of another imaging device. Another way for achieving spatial resolution is an MCP which utilizes multiple anode arrays. This type of MCP has multiple anodes or output channels on its output surface, with each of them connected to a separate FPGA-based readout system [6]. By analyzing the signals from each anode, it is possible to determine the precise position of the incoming electrons or ions, which makes it possible to map the spatial distribution of a particular signal.

Fig.1.
Sketch of an experimental setup. The UCP is produced by means of photoionization of atoms within a magneto-optical trap (MOT). Then the electrons from the UCP are directed along the x-axis via an electric field from the extraction electrodes E1 and E2 towards the microchannel plate detector (MCP). The MOT region is shielded from the MCP by grounded wire mesh grid. In order to provide compensation for stray electric fields and in order to apply external high-frequency electric fields, eight electrodes surround the MOT region, which electrodes are indicated in the figure by numbers 1 to 8. The MCP connection wires designated as In, Out, and A are the MCP input, the MCP output, and the anode, respectively.
There are several readout techniques available for MCP detectors, including those which utilize digital storage oscilloscopes [7][8][9], pulse counting systems that are utilized for photon counting experiments, and FPGA-based systems that are particularly beneficial for positionsensitive multi-anode MCP detectors [6]. The above-said techniques make it possible to obtain precise time resolution of the signals measured. Integrating amplifiers with high values of time constants can be used for measuring all the charges accumulated on the anode. This technique is useful in case where high time resolution is not required.
Ultra-cold plasma (UCP) can be studied using various experimental techniques [10][11][12], including the use of an MCP for electron detection and imaging. MCPs are highly sensitive detectors that can detect and amplify signals from individual electrons or ions and, are frequently used for precision spectroscopy of Rydberg states [13,14].
In this paper, we describe a charged particle diagnostic system for experiments involving ultracold lithium plasma. In our ultra-high vacuum system, a chevron MCP detector assembly is mounted along with multiple electrodes, which makes it possible for us to direct charged particles and to apply external electric fields. A detailed description of two distinct signalreadout techniques for the MCP detector is presented. Single pulse events are counted by using a FPGA-based device. By utilizing the said method, we demonstrate the expansion rate measurement for a low-density UCP. For continuous measurements of the electron flux, the integrating amplifier (IA) is used. By means of this technique, the total charge accumulated on the anode is integrated over time. The response of an ultracold plasma to an external radiofrequency electric field is measured using an integrating amplifier.

Experimental setup
The key part of our experimental setup is the conventional magneto-optical trap (MOT) for lithium atoms. We utilize two distinct diode lasers that are equipped with tapered amplifiers. Our MOT is formed by three pairs of counter-propagating bichromatic laser beams, with opposite circular polarization. The frequency of the cooling laser is locked onto the lowfrequency hyperfine component 2 1/2 ( = 2) − 2 3/2 , while the frequency of the repumping laser is locked onto the high-frequency hyperfine component 2 1/2 ( = 1) − 2 3/2 . We confine approximately 0 ≈ 10 8 cold lithium atoms with the peak density of 0 ≈ 6 × 10 10 cm -3 , and the temperature of the atomic cloud is approximately 0.9 mK [15]. In order to carry out photoionization of cold atoms in the MOT, a continuous-wave laser beam with a wavelength of 350 nm is utilized [16]. The 350 nm laser beam intensity is controlled by an acousto-optical modulator (AOM).

Fig.2.
The pressure evolution within our UHV system after the recent bakeout process. The observed peaks correspond to the activation of the titanium sublimation pump. The shaded region labeled as "LOW PRESSURE" represents the current pressure level, which cannot be measured by our vacuum gauge.
In order to detect charged particles, in our ultra-high vacuum (UHV) system we mount a chevron assembly of microchannel plates DV186A manufactured by Baspik. This MCP has a gain exceeding 10 7 at a power voltage of 2.2 kV. Fig. 1 shows schematically the MCP together with various intravacuum electrodes. In order to direct the charged particles towards the MCP, two extraction electrodes are utilized. Additional eight electrodes are mounted for the stray electric field compensation. In order to shield the MOT region from the highvoltage (HV) supplied to the MCP, a grounded stainless wire mesh is utilized. The electrodes are connected to the vacuum feedthrough flanges by means of Kapton insulated wires, which are shielded by a grounded stainless-steel shield. The electrodes which are used for extraction and which are labeled as E1 and E2 in Fig. 1, are positioned at the distance of 39 mm from the MOT region. In order to generate the extraction field, we utilize a two-channel, highvoltage amplifier A800DI, manufactured by FLC Electronics. This amplifier has a low output impedance of less than 0.1 , a high slew rate of 500 V/ s and is capable of producing an output voltage of up to ±400 V.
In our experimental setup it is crucial to achieve the UHV conditions. For some experiments, we utilize an optical dipole trap (ODT) in order to confine cold lithium atoms [17]. The lifetime of the cold atoms in the ODT is strongly dependent on the residual gas pressure. Therefore, we carefully choose all materials, wiring, and flanges in order to ensure compatibility with our ultra-high vacuum conditions. After mounting our detection system, the vacuum chamber is pumped by means of scroll and turbomolecular pumps until it achieves the pressure of 10 −6 Torr. The entire vacuum system is heated and baked for 10 days at 200 ∘ for outgassing. Fig. 2 shows the vacuum pressure after the bakeout process. Once the bakeout process is completed, the vacuum is maintained by means of an ion pump and a titanium sublimation pump (TSP). In order to achieve UHV conditions, we conduct three TSP runs. The peaks in Fig. 2 correspond to the activation of the titanium sublimation pump. The value of pressure in the vacuum system is beyond the measurable range of our vacuum gauge (in Fig. 2, this range of values is denoted "low pressure"). We estimate the approximate value of the vacuum pressure by measuring the lifetime of lithium atoms in the MOT [18,19]. , 2 = 200 , = 1 , = 1 nF, "MCP In" is the input voltage of the first MCP, "MCP Out" is the voltage applied to the output of the second MCP, and "Signal" indicates the output signal of the detector. We utilize two output readout configurations: (b) Digital pulse counting scheme, "Amp" is the low noise amplifier ZFL-1000LN+; (c) Integrating amplifier circuit. The DG645 digital delay generator is utilized for controlling the acousto-optical modulator ("AOM") and for triggering the FPGA-based counter and the oscilloscope ("Scope").
A schematic diagram of the MCP chevron and readout circuits can be seen in Fig. 3. The MCP chevron is connected to a high-voltage power supply SRS PS350 via a voltage divider circuit [ Fig. 3(a)]. The output of the MCP is isolated from HV power supply by means of a decoupling capacitor. Fig. 3(b) shows the scheme for digital pulse counting. The output of the MCP is connected to a low-noise amplifier ZFL-1000LN+, manufactured by Mini Circuit. The amplified signal is monitored by means of an oscilloscope having a bandwidth of 500 MHz (DPO7054 manufactured by Tektronix). The oscilloscope operates in the Fast Frame Acquisition mode, enabling us to capture over 250,000 waveforms per second. The oscilloscope is triggered by means of a pulse signal from the MCP, at a trigger level chosen on the falling edge. Each triggering event forms a digital TTL signal in the Auxiliary trigger output. This signal is collected and counted by means of using the digital FPGA counter NI PCIe-7820R. The digital delay/pulse generator SRS DG645 is utilized in order to control the intensity of the photoionizing laser, the delay between the plasma excitation pulse and the charged particles extraction field, and to start acquisition of the FPGA counter with the desired delay.
In Fig. 3(c), the electrical diagram of the integrating amplifier (IA) is shown, which is utilized for integrating over time the signal obtained from MCP [20]. For the integrator, the time constant = can be regulated with a variable resistor (1 ) and a capacitance (3 − 60 pF). The integrated signal is then passed through a low-noise operational amplifier ADA4817. In order to prevent any signal level saturation, the amplification level in the negative feedback loop can be adjusted by means of a variable resistor . This resistor is grounded using a capacitor 2 = 10 F. Further adjustment is carried out by utilizing a variable capacitance 1 , which makes fine adjustment in the range of 3 − 60 F possible. The amplified signal is captured and averaged by the oscilloscope. The averaged waveform recorded is sent to the computer via the Ethernet interface.

Results and discussion
A typical output signal of an MCP is shown in Fig. 4. In Figs. 4(a) and (b), we present direct output waveforms and waveforms after amplification. After the high-frequency amplifier, a second peak appears because of an impedance mismatch between the MCP output and the amplifier input. A high-frequency amplifier can also efficiently isolate the measured signal from the pulse field that is applied to the extraction electrodes (labeled as E1 and E2 in Fig. 1). In Fig. 4(c), the single pulse is shown for several time constants after its passing through an integrating amplifier.
The above-said digital and analog measurement techniques make it possible to measure various properties of our UCP. For some cases, it is necessary to obtain high temporal resolution. As shown in Fig. 5, expansion of the low-density UCP can be observed after the excitation beam is turned off. These data are obtained by means of digital techniques [ Fig. 3(b)]. The initial temperature of electrons is determined from the detuning value of the ionization laser above the ionization threshold, and for this measurement it is equal to = 0.5 K.
In some cases, it is more convenient to measure the time-averaged charge from the anode of the MCP detector. Fig. 6 shows the frequency response of the UCP to an external oscillating electric field. The RF field is applied to the top four electrodes, labeled 1 to 4 in Fig. 1. The measurements are performed by means of an integrating amplifier with a high value of the time constant. The solid line is the frequency response of the steady-state UCP [21]. This response is measured 300 s after the excitation beam is turned on. The dashed line shows the integrated frequency response of the expanding UCP, measured by means of integration of the signal over the first 50 s of the plasma expansion.  Fig. 6). This central frequency is associated with an electron density higher than that for expansion of the UCP (the dashed curve in Fig. 6). Additionally, the increased collision frequency results in the response width of approximately 8 MHz. A similar density-dependent response to the RF field is also observed in [22]. The data acquisition systems on the basis of oscilloscopes are often utilized for experiments that involve counting of photons over a time period [7][8][9]. However, the time taken to transfer the data to the storage device is a bottleneck of this technique. In [7], 265 measurements with a duration of 100 s were recorded over a period of approximately 2.7 hours. A trigger output option is a popular feature in many modern, low-cost oscilloscopes. Together with a FPGA-based counter, this will be a good solution for laboratory and educational applications. Furthermore, the oscilloscope trigger works as a discriminator for the received signal and makes it possible to adjust the discriminator level by means of selection of the triggering level.

Conclusion
In conclusion, we describe a charged particle diagnostic system for experiments involving ultracold lithium plasma. In the ultra-high vacuum system, a chevron MCP detector assembly along with multiple electrodes is mounted, which makes it possible to direct charged particles and to apply external electric fields. A detailed description of two distinct signal-readout techniques from the MCP detector is presented. Single-pulse events are counted by means of using a FPGA-based device and an oscilloscope, which works as a discriminator circuit. By means of utilizing this method, we demonstrate the expansion rate measurement for a lowdensity UCP. For continuous measurements of the electron flux, the integrating amplifier (IA) is utilized. The total charge accumulated on the anode is integrated over time. Response of the ultracold plasma to an external radio-frequency electric field is measured.