Controlling the ion velocity in an ion sheath by applying an alternating current (AC) voltage to an electrode and/or a substrate is critical in plasma material processes. To externally control the velocity distribution of incident ions on a substrate, the application of tailored-waveform AC voltages instead of sinusoidal voltages has garnered interest in recent years. In this study, to investigate temporal changes in ion-velocity distributions, we developed a time-resolved laser-induced fluorescence spectroscopy (LIF) system using a continuous-wave diode laser as an excitation-laser source. A time-resolved LIF system entails the capture of temporally continuous and spectrally discrete LIF spectra during an AC voltage cycle. By measuring temporal changes in the LIF signal intensity at various excitation-laser wavelengths, the argon-ion velocity distribution near the electrode following the AC voltage can be characterized. The results of applying sinusoidal, triangular, and rectangular bias waveforms indicate that the LIF measurement scheme proposed herein can be used to investigate the dynamic behavior of ion-velocity distributions controlled by tailored-waveform AC voltages.

In plasma material processes such as plasma etching and plasma-enhanced chemical vapor deposition, ions accelerated toward the substrate in an ion sheath are critical in plasma–surface interactions, e.g., they provide activation energy for surface reactions.1,2 Accurate control of the flux and velocity distribution of ions on a substrate is necessary, particularly for fabricating nanometer-scale structures via plasma etching. In recent years, the application of various alternating current (AC)-voltage waveforms instead of sinusoidal waveforms to discharge electrodes and/or substrates has garnered attention. This tailored waveform approach aims to control plasma–surface interactions by adjusting ion-velocity distribution functions (IVDFs) on the substrate.3,4 The behaviors of ions near the electrode and substrate where the tailored-waveform AC voltages are applied has been investigated numerically.5–9 In addition, experimental studies pertaining to the tailored-waveform approach primarily focus on the analysis of time-averaged IVDFs.7,8,10,11 Previously, Wang and Wendt measured the temporal evolution of electric potential on a substrate.3 To optimize the tailored-waveform AC voltages for each application, one must understand the dynamic behaviors of the IVDF on a substrate during one AC-voltage cycle.

The methodologies for measuring IVDFs can be categorized into electrical, particle, and optical methods. Retarding-field energy analyzers (RFEAs) and energy-filtered mass spectrometers are electrical and particle measurement methods, respectively, which entail the counting of ions passing through direct current (DC)-biased grids.12–14 Laser-induced fluorescence spectroscopy (LIF) is an optical method that detects fluorescence light from target particles excited by a laser beam.15–18 The LIF method has been applied to the diagnostics of neutral particles and ions in plasmas such as hydrogen and impurity atoms in fusion plasmas,19–21 and radicals generated in flames and low-temperature plasmas.22–26 The LIF method can measure the velocity distribution of the target species, including ions, along the excitation-laser axis without perturbation of plasma and fluid parameters by scanning the excitation-laser wavelength. LIF measurement systems are generally categorized into two groups: one that uses a pulsed laser as the excitation-laser source, and one that uses continuous-wave (CW) lasers.

Because the IVDFs in bulk plasmas and ion-acceleration fields are crucial in various plasma technologies, the LIF method has been extensively used to obtain the IVDFs, and its capability has been confirmed based on comparison with measurement results yielded by RFEAs.27–29 To investigate the behavior of rare-gas ions in low-temperature plasmas in the steady state, the LIF method has been utilized to diagnose bulk plasmas,30–33 pre-sheaths,34–38 ion sheaths formed on floated,39,40 and DC-biased electrodes,41–45 and plumes of electric propulsion devices.46–52 

To understand the dynamic behaviors of ions in plasmas, time-resolved LIF systems have been developed and applied to diagnose spatiotemporal fluctuations in plasma thrusters,53–57 the temporal evolution of discharge and afterglow phases in pulsed plasmas,58,59 and ion sheaths formed on AC-biased electrodes.60–65 Time-resolved LIF measurements of ion sheaths were achieved by using pulsed dye lasers as the excitation-laser source and controlling the laser-pulse timing synchronized with the AC bias voltage applied to the electrode. To accurately evaluate the effects of tailored-waveform AC voltages on the IVDF, the temporal change in the IVDFs must be measured with sufficient data-acquisition timing in one cycle. However, achieving temporally continuous data acquisition using a pulsed-laser-based LIF method is challenging. The time-resolved LIF measurement using a CW diode laser was demonstrated to investigate plasma instability.66,67 They developed a digital demodulation technique using a field programmable gate array (FPGA)-based system66 and measured fluctuation frequencies of ion velocity at ∼0.5 MHz near a DC-biased electrode.67 

In this study, we develop a time-resolved LIF system using a CW diode laser as an excitation-laser source to analyze the dynamic behavior of IVDFs near AC-biased electrodes. Using a CW laser with a rapid fluorescence-signal detection system allows us to capture the entire temporal evolution of IVDFs with a time resolution that is sufficiently shorter than that of the bias voltage cycle. To detect LIF signals with a high time resolution, we modulate the laser intensity using an electro–optic modulator (EOM) with a modulation frequency of 20 MHz and detect the LIF signal intensity using an analog demodulation system including a phase-sensitive detector (PSD) with an output signal bandwidth of 2 MHz. A similar LIF system using an acousto–optic modulator and a PSD demonstrated potential for analyzing the temporal fluctuation of plasma plume in a Hall thruster at a time scale of 10 µs.54,55 In this study, we perform test measurements of bulk plasma generated under an electron cyclotron resonance (ECR) configuration to confirm the capability of our LIF system, such as its temporal resolution. Subsequently, we apply the time-resolved LIF system to understand the temporal evolution of IVDFs near the tailored-waveform AC-biased electrodes placed in an ECR plasma. The IVDFs measured with sinusoidal, triangular, and rectangular waveforms exhibit different temporal changes, thus demonstrating the high performance of our time-resolved LIF system.

Figure 1 shows a schematic diagram of the ECR plasma source and the setup for the LIF measurement used in this study. An antenna-excited ECR plasma source (EPS-120, Novelion Systems) was installed on a Gaseous Electronics Conference radio-frequency reference cell.68,69 The ECR plasma was generated by a 2.45-GHz microwave at 200 W in argon (Ar) gas at 0.05 Pa. The plasma parameters, electron temperature (Te) and density (ne), were measured using a conventional Langmuir probe. The measurement results in the discharge condition were Te = 5 eV and ne = 2 × 1010 cm−3. A stainless-steel plate electrode connected to the AC/DC power supply (“bias plate” in Fig. 1) was placed 1 mm from the center of the discharge cell. The excitation-laser beam was propagated perpendicular to the bias plate and placed into a beam dump through a hole (diameter = 3.5 mm) in the bias plate. We set the x-axis as the axis along the excitation-laser path and the origin at the center of the discharge cell.

FIG. 1.

Schematic diagram of the ECR plasma source and optical arrangement for LIF measurement near the AC-biased electrode used in this study.

FIG. 1.

Schematic diagram of the ECR plasma source and optical arrangement for LIF measurement near the AC-biased electrode used in this study.

Close modal

The target particles for LIF diagnostics in this study were Ar ions in the metastable state 3d4F7/2 (Ar+m). Figure 2 partially shows an Ar+ Grotrian diagram related to our LIF measurement. The wavelength of the excitation laser was 668.61 nm (3d4F7/2 → 4p4D5/2), and we detected fluorescence light intensities at a wavelength of 442.60 nm (4p4D5/2 → 4s4P3/2).68 An excitation CW laser beam with a wavelength of 668.61 nm was emitted from an external-cavity diode laser (DL pro, TOPTICA), and its intensity was amplified by a tapered amplifier (BoosTA pro, TOPTICA). The excitation-laser wavelength was monitored using a laser wavelength meter (WS/6-200, HighFinesse) with an absolute accuracy of 200 MHz (0.0003 nm at 670 nm).

FIG. 2.

Partial Grotrian diagram of the argon ion (Ar+) related to the LIF measurement performed in this study.

FIG. 2.

Partial Grotrian diagram of the argon ion (Ar+) related to the LIF measurement performed in this study.

Close modal

To confirm the capability of ion-sheath diagnostics using the plasma and LIF setups, we first performed time-averaged LIF measurements under a conventional LIF arrangement, as shown in Fig. 3. The amplitude of the excitation laser was modulated using a mechanical chopper (SR540, 30-slot blade, Stanford Research Systems) with a modulation frequency of 1.6 kHz. A laser beam was introduced into the chamber through a lens with a focal length of 750 mm to maintain a beam diameter of 2–3 mm within the measurement region. The fluorescent light from the measurement spot was collected using a lens with a diameter of 100 mm and detected using a photomultiplier tube (PMTSS, Thorlabs) through an optical fiber and an optical bandpass filter. The size of the measurement spot on the excitation-laser axis was ∼1 mmϕ, as confirmed by inserting 455 nm light from a light emitting diode via the optical fiber. The lens and fiber connector were fixed to a one-dimensional movable stage to scan the measurement position along the x-axis. The center wavelength and full-width at half maximum of the optical bandpass filter were 442.7 and 1 nm, respectively. The LIF-signal intensity was measured using a lock-in amplifier with a 1.6-kHz reference signal from the chopper controller, and the temporal changes in the LIF-signal intensity were recorded using an oscilloscope (Wavesurfer 3024z, Teledyne LeCroy). In the time-averaged measurements, the time constant in the lock-in amplifier and the scan speed of the laser wavelength were 100 ms and 0.24 pm/s, respectively.

FIG. 3.

Schematic illustration of (a) optical and (b) electrical setups for time-averaged LIF measurement. PMT: photomultiplier tube.

FIG. 3.

Schematic illustration of (a) optical and (b) electrical setups for time-averaged LIF measurement. PMT: photomultiplier tube.

Close modal

Figure 4 shows the LIF spectra and corresponding IDVFs measured at position x = 0 mm under three different DC bias voltages: +10, −10, and −30 V. We define vi as the ion velocity toward the bias plate. The vi, when the LIF signal intensity reaches its maximum, increases when the DC bias voltage decreases. The LIF spectrum measured at −30 V has a lower peak height and a wider broadening width than those measured at the other two bias voltages. It can be speculated that this is because the measurement spot is not a point but has a certain volume. Because the potential slope in the ion sheath is steeper in the case of −30 V, the LIF system detected ion acceleration inside the measurement spot.

FIG. 4.

Ion-velocity distribution functions (IVDFs) measured at x = 0 mm under three DC-bias voltages: +10, −10, and −30 V.

FIG. 4.

Ion-velocity distribution functions (IVDFs) measured at x = 0 mm under three DC-bias voltages: +10, −10, and −30 V.

Close modal
Figure 5 shows the spatial distributions of the average ion velocity toward the bias plate vi_ave measured with different DC bias voltages and without the bias plate. “Without bias plate” in this paper means a condition where we remove the bias plate and move the beam dump toward the chamber wall (to the right in Fig. 1). Here, vi_ave is calculated as follows:
vi_ave=vifvidvifvidvi,
(1)
where f(vi) is the IVDF measured using time-averaged LIF. As shown in Fig. 5, ions accelerated toward the bias plate. The acceleration depends on the amplitude of the DC bias voltage. Since the Bohm velocity calculated from the plasma parameters was 3.4 km/s and vi_ave exceeded the Bohm velocity when we applied a DC bias voltage lower than the floating potential Vf = −5 V, the LIF system captured the behaviors of ions inside the ion sheath. Based on the results, we performed time-resolved LIF measurements with an AC bias voltage under the same discharge conditions and set the measurement spot at 1 mm from the bias plate (x = 0 mm).
FIG. 5.

Spatial distribution of average ion velocity vi_ave measured with and without a bias plate. (a) Data obtained at all measurement positions, and (b) focused view near the bias plate. Vertical dotted lines show the position of the bias plate (x = 1 mm).

FIG. 5.

Spatial distribution of average ion velocity vi_ave measured with and without a bias plate. (a) Data obtained at all measurement positions, and (b) focused view near the bias plate. Vertical dotted lines show the position of the bias plate (x = 1 mm).

Close modal

Figure 6 shows a schematic diagram of the time-resolved LIF system demonstrated in this study. To measure temporal changes in the IVDF in one cycle of the AC bias voltage, the amplitude of the excitation laser was modulated using an EOM (EO-AM-R-20-C1, Thorlabs) with a modulation frequency of 20 MHz, and the LIF signal in the photomultiplier output was measured using an analog PSD module (SD20IF2-0S, R&K) with a 20 MHz reference signal from a function generator used for driving the EOM. Because the lifetime of excited Ar+ in the state 4p4D5/2, which is the upper state in our LIF scheme, was 7.5 ns,70 the laser modulation frequency at 20 MHz was sufficiently lower than the physical limit to detect the fluorescence signal. The expected temporal resolution of the time-resolved LIF system was ∼500 ns since the output bandwidth of the PSD was 2 MHz. The signal was recorded using an oscilloscope, and the averaging over 2000 times was triggered by a rectangular voltage synchronized with the AC bias voltage. Measurement uncertainties in the time-resolved LIF system in the ion velocity and time are caused by the noise in the signal, the volume of the measurement spot, and the output bandwidth of PSD, respectively.

FIG. 6.

Schematic illustration of (a) optical and (b) electrical setups for time-resolved LIF measurement. EOM: electro–optic modulator. PD: photodetector. PSD: phase-sensitive detector.

FIG. 6.

Schematic illustration of (a) optical and (b) electrical setups for time-resolved LIF measurement. EOM: electro–optic modulator. PD: photodetector. PSD: phase-sensitive detector.

Close modal

In an experiment introduced in Sec. III A, in addition to performing EOM modulation, we mechanically chopped the excitation laser. To monitor the mechanical chopping of the excitation laser, we inserted an amplified photodetector (PDA10A2, Thorlabs) with an output bandwidth of 150 MHz, as shown in Fig. 6(a).

To diagnose the dynamic behavior of IVDF by the time-resolved LIF system, we applied AC bias voltages at 10 kHz with a peak-to-peak amplitude Vpp of 40 V and an offset of −11.23 V to a bias plate (see Sec. III B). The offset was the sum of the floating potential (−5 V) and self-bias voltage calculated from the IV curve. We simulated conventional plasma processing through a blocking capacitor. The AC bias voltage waveforms were sinusoidal, triangular, and rectangular.

To evaluate the capability of the time-resolved LIF system, we measured the LIF signal with amplitude modulation of the excitation laser at two frequencies simultaneously, i.e., 20 MHz and 1.6 kHz. We generated the ECR plasma without a bias plate, i.e., bulk-plasma diagnostics. Figure 7 shows the measured temporal changes in the excitation-laser amplitude and detected LIF-signal intensity. The LIF signal intensity was modulated by the excitation-laser intensity using a mechanical chopper. Here, we defined the LIF-signal intensity as the difference between the ON and OFF phases of the excitation laser, as shown in Fig. 7.

FIG. 7.

Temporal changes in (a) excitation-laser amplitude and (b) LIF signal measured by a photodetector and time-resolved LIF system, respectively. LIF-signal intensity refers to the difference between the ON and OFF phases of the excitation laser herein.

FIG. 7.

Temporal changes in (a) excitation-laser amplitude and (b) LIF signal measured by a photodetector and time-resolved LIF system, respectively. LIF-signal intensity refers to the difference between the ON and OFF phases of the excitation laser herein.

Close modal

By measuring the LIF-signal intensity while changing the excitation-laser wavelength, we obtained an LIF spectrum, as shown by the red points in Fig. 8. In addition, we plotted a measurement result of the time-averaged LIF under the same discharge conditions. The fluorescence spectra obtained by the time-resolved and time-averaged LIF systems showed similar peak wavelengths and broadening corresponding to vi_ave and the ion temperature, respectively. Hence, we conclude that the time-resolved LIF system developed in this study can appropriately measure IVDFs in the ECR plasma.

FIG. 8.

IVDFs measured using time-resolved (red points) and time-averaged (black line) LIF systems in bulk ECR plasma. The IVDFs are normalized by maximum values.

FIG. 8.

IVDFs measured using time-resolved (red points) and time-averaged (black line) LIF systems in bulk ECR plasma. The IVDFs are normalized by maximum values.

Close modal

Figure 9 shows the temporal changes in the LIF-signal intensity measured using the time-resolved LIF system with three different excitation-laser wavelengths plotted with the bias-voltage waveform (10-kHz sinusoidal waveform). The LIF-signal intensity depended on the excitation-laser wavelength, which corresponded to the ion velocity along the excitation-laser beam. To analyze the dynamic behaviors of the IVDF, we summarized the measurement results for all laser wavelengths, as shown in Fig. 10. The red region represents vi_ave, and the temporal change in vi_ave follows the bias-voltage waveform. The thickness of the horizontal lines is decided just for a clear view, not indicating velocity uncertainties. In contrast to a previous study that used a pulsed excitation laser,62 the time-resolved LIF measurement performed in this study outputs continuous temporal-change data of the IVDF with discrete velocity information.

FIG. 9.

(a) AC bias voltage waveform applied to bias plate and temporal changes in LIF-signal intensity measured by time-resolved LIF system with three different excitation-laser wavelengths, i.e., (b) 668.6010, (c) 668.6036, and (d) 668.6060 nm, which correspond to vi ∼ 5.7, 4.6, and 3.5 km/s, respectively.

FIG. 9.

(a) AC bias voltage waveform applied to bias plate and temporal changes in LIF-signal intensity measured by time-resolved LIF system with three different excitation-laser wavelengths, i.e., (b) 668.6010, (c) 668.6036, and (d) 668.6060 nm, which correspond to vi ∼ 5.7, 4.6, and 3.5 km/s, respectively.

Close modal
FIG. 10.

(a) AC bias voltage waveform applied to bias plate and (b) temporal change in IVDF measured at position x = 0 mm.

FIG. 10.

(a) AC bias voltage waveform applied to bias plate and (b) temporal change in IVDF measured at position x = 0 mm.

Close modal

To evaluate the performance of the time-resolved LIF system, we plotted the IVDFs for five phases of the sinusoidal bias voltage based on the data shown in Fig. 10. The instantaneous IVDFs plotted in Fig. 11 were compared with the IVDFs (LIF spectra) measured using the time-averaged LIF system by applying the corresponding DC bias voltages to the bias plate. The IVDFs measured using both the time-resolved and time-averaged LIF systems were normalized by the maximum LIF-signal intensity at a bias voltage of 0 V. Based on the results, the IVDFs measured by applying the AC and DC bias voltages exhibited similar vi_ave features at each voltage. This indicates that the instantaneous applied voltage determines the IVDF under discharge and AC bias conditions without a detectable delay. Moreover, this result suggests that the time-resolved LIF system is sufficiently swift in capturing temporal changes in the IVDF following the AC bias voltage at 10 kHz.

FIG. 11.

Instantaneous IVDFs at five timings measured by the time-resolved LIF system (red points) and time-averaged IVDFs under DC bias voltages corresponding to instantaneous applied voltages (black line). Timings shown in Fig. 10 and applied DC bias voltages were (a) 25 µs/+10 V, (b) 42 µs/0 V, (c) 50 µs/−10 V, (d) 58 µs/−20 V, and (e) 75 µs/−30 V.

FIG. 11.

Instantaneous IVDFs at five timings measured by the time-resolved LIF system (red points) and time-averaged IVDFs under DC bias voltages corresponding to instantaneous applied voltages (black line). Timings shown in Fig. 10 and applied DC bias voltages were (a) 25 µs/+10 V, (b) 42 µs/0 V, (c) 50 µs/−10 V, (d) 58 µs/−20 V, and (e) 75 µs/−30 V.

Close modal

For further evaluation, we compared the time-integrated IVDF for one cycle of a 10 kHz sinusoidal AC bias voltage calculated from the result of the time-resolved LIF measurement with the IVDF measured by the time-averaged LIF system applying a 10-kHz sinusoidal AC bias voltage. The two IVDFs shown in Fig. 12 were normalized to the maximum values for each measurement. The similarity of the IVDFs measured by the different systems under the same discharge conditions suggests that the time-resolved LIF system appropriately measures temporal changes in the IVDF near the AC-biased electrode.

FIG. 12.

Comparison between time-integrated IVDF calculated from one-cycle data of time-resolved LIF measurement (red points) and IVDF measured by a time-averaged LIF system (black line). The AC bias voltage was a 10-kHz sinusoidal waveform.

FIG. 12.

Comparison between time-integrated IVDF calculated from one-cycle data of time-resolved LIF measurement (red points) and IVDF measured by a time-averaged LIF system (black line). The AC bias voltage was a 10-kHz sinusoidal waveform.

Close modal

To ensure the potential of the time-resolved LIF system, we measured the temporal changes in the IVDF by applying tailored bias-voltage waveforms. First, we tested triangular waveforms with rise times of 25% (triangular 25%) and 75% (triangular 75%) for one cycle. Figures 13 and 14 show the temporal changes in the IVDF measured based on the triangular 25% and 75% bias-voltage waveforms, respectively. As shown, vi_ave followed the triangular bias-voltage waveforms, and the times at which vi_ave reached its maximum were at negative-peak timings of the bias voltage. Figure 15 shows a comparison between the time-integrated IVDF calculated from the results of the time-resolved LIF measurements and the IVDF measured using the time-averaged LIF system. The shapes of the IVDF measured by the two LIF systems were similar to each other, as in the case of the sinusoidal AC bias waveform.

FIG. 13.

(a) Waveform of a triangular 25% AC bias voltage applied to the bias plate, and (b) temporal change in IVDF measured at x = 0 mm.

FIG. 13.

(a) Waveform of a triangular 25% AC bias voltage applied to the bias plate, and (b) temporal change in IVDF measured at x = 0 mm.

Close modal
FIG. 14.

(a) Waveform of a triangular 75% AC bias voltage applied to the bias plate, and (b) temporal change in IVDF measured at x = 0 mm.

FIG. 14.

(a) Waveform of a triangular 75% AC bias voltage applied to the bias plate, and (b) temporal change in IVDF measured at x = 0 mm.

Close modal
FIG. 15.

Comparison between time-integrated IVDF calculated from the result of time-resolved LIF measurement (red points) and IVDF measured by a time-averaged LIF system (black line) under AC bias voltages of triangular (a) 25% and (b) 75%.

FIG. 15.

Comparison between time-integrated IVDF calculated from the result of time-resolved LIF measurement (red points) and IVDF measured by a time-averaged LIF system (black line) under AC bias voltages of triangular (a) 25% and (b) 75%.

Close modal

Figure 16 shows the temporal change in the IVDF measured by applying a rectangular bias voltage with a frequency of 10 kHz and a duty ratio of 50%. Low and high ion velocity phases were observed following the high and low phases of the bias voltage, respectively. The transition time from high to low ion velocity was longer than that from low to high ion velocity. This suggests that the timescales required to reduce and expand the ion sheath were different under the discharge conditions. Based on previous studies, the reduction of sheath thickness due to an increase in the bias voltage is governed by the ambipolar diffusion of electrons and ions from the bulk plasma.71 The time to expand the sheath by decreasing the bias voltage is determined by the ion-transit time, which is the duration required by ions to travel from the sheath edge to the electrode.72 The difference in the timescale of ambipolar diffusion and the ion-transit time should be considered when investigating ion-sheath dynamics. Figure 17 shows a comparison of the time-integrated IVDF measured by the time-resolved LIF system and the IVDF measured by the time-averaged LIF system for a rectangular waveform. Two peaks were indicated at low and high ion velocities, and the broadening of the IVDF peak at a higher velocity was larger than that at a lower velocity. These features were similarly confirmed by the instantaneous IVDFs shown in Fig. 16.

FIG. 16.

(a) Waveform of a rectangular AC bias voltage applied to the bias plate, and (b) temporal change in IVDF measured at x = 0 mm.

FIG. 16.

(a) Waveform of a rectangular AC bias voltage applied to the bias plate, and (b) temporal change in IVDF measured at x = 0 mm.

Close modal
FIG. 17.

Comparison between time-integrated IVDF calculated from the result of time-resolved LIF measurement (red points) and IVDF measured by a time-averaged LIF system (black line) under a 10-kHz rectangular AC bias-voltage waveform.

FIG. 17.

Comparison between time-integrated IVDF calculated from the result of time-resolved LIF measurement (red points) and IVDF measured by a time-averaged LIF system (black line) under a 10-kHz rectangular AC bias-voltage waveform.

Close modal

The results presented in this section indicate that the time-resolved LIF system developed in this study can capture the dynamic behavior of the IVDF near the AC-biased electrode, including the case where tailored bias-voltage waveforms were used. In the sinusoidal and triangular bias-voltage cases, the time-resolved LIF showed its ability to confirm that the IVDF followed the AC bias voltage well. In addition, we found that the minimum timescales required to increase and decrease vi_ave differed from the results of the time-resolved LIF measurement by applying the rectangular bias-voltage waveform.

To investigate the dynamic behaviors of ions in the ion sheath formed in front of an AC-biased electrode, we developed a time-resolved LIF system using a CW diode laser as the excitation-laser source. The LIF system afforded temporally continuous signal detection with a time resolution of less than 1 µs. We confirmed the feasibility of the time-resolved LIF system for measuring temporal changes in the IVDF in plasmas through the diagnosis of a bulk ECR plasma generated in low-pressure Ar gas. To demonstrate the performance of the time-resolved LIF system in diagnosing ion dynamics in advanced plasma processes using tailored bias-voltage waveforms, we measured the temporal changes in the IVDF near the AC-biased electrode by applying sinusoidal, triangular, and rectangular waveforms. Based on a comparison of the results between time-resolved and conventional time-averaged LIF measurements under the application of a sinusoidal bias voltage, the time-resolved LIF system appropriately outputs the temporal changes in the IVDF following the bias-voltage waveform. This was similarly confirmed by the measurement results when two different waveforms of the triangular bias voltage were applied. When a rectangular bias voltage was applied, we observed a difference in the time duration between the increase and decrease in the ion velocity following the stepwise bias-voltage variation. This confirmed the results of previous fundamental studies pertaining to ion-sheath dynamics. Based on these findings, we conclude that the time-resolved LIF system developed in this study can contribute positively to plasma science and technology by revealing the dynamic interactions between charged particles and AC-biased electrodes.

This study was partly supported by a Grant-in-Aid for Young Scientists (Start-up) Grant No. 18H05849 and a Grant-in-Aid for Scientific Research (B) Grant No. 19H01886 from the Japan Society for the Promotion of Science. The authors thank Mr. S. Maeno at Novelion Systems Co., Ltd. and Tokyo Electron Ltd. for their support.

The authors have no conflicts to disclose.

Ryosuke Takahashi: Investigation (lead); Methodology (equal); Writing – original draft (equal). Seiya Kito: Investigation (supporting); Methodology (equal). Koji Eriguchi: Project administration (equal); Writing – review & editing (equal). Keiichiro Urabe: Conceptualization (lead); Methodology (equal); Project administration (equal); Writing – original draft (equal); Writing – review & editing (equal).

The data that support the findings of this study are available within the article.

1.
M. A.
Lieberman
and
A. J.
Lichtenberg
, in
Principles of Plasma Discharges and Materials Processing
, 2nd ed. (
Wiley
,
Hoboken
,
2005
), p.
434
.
2.
G. S.
Oehrlein
and
S.
Hamaguchi
,
Plasma Sources Sci. Technol.
27
(
2
),
023001
(
2018
).
3.
S.-B.
Wang
and
A. E.
Wendt
,
J. Appl. Phys.
88
(
2
),
643
646
(
2000
).
4.
T.
Lafleur
,
Plasma Sources Sci. Technol.
25
(
1
),
013001
(
2016
).
5.
Z.-L.
Dai
and
Y.-N.
Wang
,
Surf. Coat. Technol.
165
(
3
),
224
231
(
2003
).
6.
Z.
Donkó
,
J.
Schulze
,
B. G.
Heil
, and
U.
Czarnetzki
,
J. Phys. D: Appl. Phys.
42
(
2
),
025205
(
2009
).
7.
J.-C.
Wang
,
P.
Tian
,
J.
Kenney
,
S.
Rauf
,
I.
Korolov
, and
J.
Schulze
,
Plasma Sources Sci. Technol.
30
(
7
),
075031
(
2021
).
8.
P.
Hartmann
,
I.
Korolov
,
J.
Escandón-López
,
W.
van Gennip
,
K.
Buskes
, and
J.
Schulze
,
Plasma Sources Sci. Technol.
31
(
5
),
055017
(
2022
).
9.
P.
Hartmann
,
I.
Korolov
,
J.
Escandón-López
,
W.
van Gennip
,
K.
Buskes
, and
J.
Schulze
,
J. Phys. D: Appl. Phys.
56
(
5
),
055202
(
2023
).
10.
E. V.
Barnat
and
T.-M.
Lu
,
J. Appl. Phys.
92
(
6
),
2984
2989
(
2002
).
11.
F. L.
Buzzi
,
Y.-H.
Ting
, and
A. E.
Wendt
,
Plasma Sources Sci. Technol.
18
(
2
),
025009
(
2009
).
12.
M.
Zeuner
,
H.
Neumann
, and
J.
Meichsner
,
J. Appl. Phys.
81
(
7
),
2985
2994
(
1997
).
13.
D.
Gahan
,
B.
Dolinaj
, and
M. B.
Hopkins
,
Rev. Sci. Instrum.
79
(
3
),
033502
(
2008
).
14.
J.
Benedikt
,
H.
Kersten
, and
A.
Piel
,
Plasma Sources Sci. Technol.
30
(
3
),
033001
(
2021
).
15.
R. A.
Stern
and
J. A.
Johnson
III
,
Phys. Rev. Lett.
34
(
25
),
1548
1551
(
1975
).
16.
R. W.
Dreyfus
,
J. M.
Jasinski
,
R. E.
Walkup
, and
G. S.
Selwyn
,
Pure Appl. Chem.
57
(
9
),
1265
1276
(
1985
).
17.
I. P.
Herman
,
Optical Diagnostics for Thin Film Processing
(
Academic Press
,
San Diego
,
1996
), p.
215
.
18.
R. K.
Hanson
,
R. M.
Spearrin
, and
C. S.
Goldenstein
,
Spectroscopy and Optical Diagnostics for Gases
(
Springer
,
Switzerland
,
2016
), p.
177
.
19.
C. H.
Muller
III
and
K. H.
Burrell
,
Phys. Rev. Lett.
47
(
5
),
330
333
(
1981
).
20.
E.
Hintz
,
Phys. Scr.
1982
(
T2B
),
454
458
.
21.
K.
Muraoka
and
M.
Maeda
,
Plasma Phys. Controlled Fusion
35
(
6
),
633
656
(
1993
).
22.
K.
Tachibana
,
T.
Mukai
, and
H.
Harima
,
Jpn. J. Appl. Phys.
30
(
7A
),
L1208
L1211
(
1991
).
23.
J. W.
Daily
,
Prog. Energy Combust. Sci.
23
(
2
),
133
199
(
1997
).
24.
C.
Suzuki
,
K.
Sasaki
, and
K.
Kadota
,
J. Appl. Phys.
82
(
11
),
5321
5326
(
1997
).
25.
J.
Amorim
,
G.
Baravian
, and
J.
Jolly
,
J. Phys. D: Appl. Phys.
33
(
9
),
R51
R65
(
2000
).
26.
A. K.
Patnaik
,
I.
Adamovich
,
J. R.
Gord
, and
S.
Roy
,
Plasma Sources Sci. Technol.
26
(
10
),
103001
(
2017
).
27.
Z.
Harvey
,
S. C.
Thakur
,
A.
Hansen
,
R.
Hardin
,
W. S.
Przybysz
, and
E. E.
Scime
,
Rev. Sci. Instrum.
79
(
10
),
10F314
(
2008
).
28.
N.
Gulbrandsen
,
Å.
Fredriksen
,
J.
Carr
, Jr.
, and
E.
Scime
,
Phys. Plasmas
22
(
3
),
033505
(
2015
).
29.
A.
Bennet
,
C.
Charles
, and
R.
Boswell
,
Phys. Plasmas
25
(
2
),
023516
(
2018
).
30.
G.
King
,
F. C.
Sze
,
P.
Mak
,
T. A.
Grotjohn
, and
J.
Asmussen
,
J. Vac. Sci. Technol., A
10
(
4
),
1265
1269
(
1992
).
31.
S.
Jun
,
H. Y.
Chang
, and
R.
McWilliams
,
Phys. Plasmas
13
(
5
),
052512
(
2006
).
32.
T.
Bieber
,
S.
Bardin
,
L.
de Poucques
,
F.
Brochard
,
R.
Hugon
,
J.-L.
Vasseur
, and
J.
Bougdira
,
Plasma Sources Sci. Technol.
20
(
1
),
015023
(
2011
).
33.
D.
Jiang
,
C.-S.
Yip
,
C.-Y.
Jin
,
W.
Zhang
,
L.
Wang
, and
G.-S.
Xu
,
Phys. Plasmas
29
(
6
),
063504
(
2022
).
34.
S. L.
Gulick
,
B. L.
Stansfield
,
Z.
Abou-Assaleh
,
C.
Boucher
,
J. P.
Matte
,
T. W.
Jhonston
, and
R.
Marchand
,
J. Nucl. Mater.
176-177
,
1059
1063
(
1990
).
35.
N.
Sadeghi
,
M.
van de Grift
,
D.
Vender
,
G. M. W.
Kroesen
, and
F. J.
de Hoog
,
Appl. Phys. Lett.
70
(
7
),
835
837
(
1997
).
36.
L.
Oksuz
,
M. A.
Khedr
, and
N.
Hershkowitz
,
Phys. Plasmas
8
(
5
),
1729
1733
(
2001
).
37.
G. D.
Severn
,
X.
Wang
,
E.
Ko
, and
N.
Hershkowitz
,
Phys. Rev. Lett.
90
(
14
),
145001
(
2003
).
38.
B.
Jacobs
,
W.
Gekelman
,
P.
Pribyl
,
M.
Barnes
, and
M.
Kilgore
,
Appl. Phys. Lett.
91
(
16
),
161505
(
2007
).
39.
N.
Sadeghi
,
T.
Nakano
,
D. J.
Trevor
, and
R. A.
Gottscho
,
J. Appl. Phys.
70
(
5
),
2552
2569
(
1991
).
40.
N.
Claire
,
G.
Bachet
,
U.
Stroth
, and
F.
Doveil
,
Phys. Plasmas
13
(
6
),
062103
(
2006
).
41.
M. J.
Goeckner
,
J.
Goree
, and
T. E.
Sheridan
,
Phys. Fluids B
4
(
6
),
1663
1670
(
1992
).
42.
M.
Watanabe
,
K.
Takiyama
, and
T.
Oda
,
Jpn. J. Appl. Phys.
39
(
2A
),
L116
L118
(
2000
).
43.
D.
Lee
,
G.
Severn
,
L.
Oksuz
, and
N.
Hershkowitz
,
J. Phys. D: Appl. Phys.
39
(
24
),
5230
5235
(
2006
).
44.
R.
Hood
,
B.
Scheiner
,
S. D.
Baalrud
,
M. M.
Hopkins
,
E. V.
Barnat
,
B. T.
Yee
,
R. L.
Merlino
, and
F.
Skiff
,
Phys. Plasmas
23
(
11
),
113503
(
2016
).
45.
H.
Lee
,
N.-K.
Kim
,
M.-G.
Lee
,
J.-W.
Kwon
,
S. H.
Son
,
N.
Bae
,
T.
Park
,
S.
Park
, and
G.-H.
Kim
,
Plasma Sources Sci. Technol.
31
(
8
),
084006
(
2022
).
46.
W. A.
Hargus
, Jr.
and
C. S.
Charles
,
J. Propul. Power
24
(
1
),
127
133
(
2008
).
47.
R.
Spektor
,
K. D.
Diamant
,
E. J.
Beiting
,
Y.
Raitses
, and
N. J.
Fisch
,
Phys. Plasmas
17
(
9
),
093502
(
2010
).
48.
S.
Mazouffre
,
Plasma Sources Sci. Technol.
22
(
1
),
013001
(
2013
).
49.
N.
Teshigahara
,
S.
Shinohara
,
Y.
Yamagata
,
D.
Kuwahara
, and
M.
Watanabe
,
Plasma Fusion Res.
9
,
3406055
(
2014
).
50.
R.
Tsukizaki
,
Y.
Yamamoto
,
D.
Koda
,
Y.
Yusuke
,
K.
Nishiyama
, and
H.
Kuninaka
,
Plasma Sources Sci. Technol.
27
(
1
),
015013
(
2018
).
51.
A. E.
Vinci
,
S.
Mazouffre
,
V.
Gómez
,
P.
Fajardo
, and
J.
Navarro-Cavallé
,
Plasma Sources Sci. Technol.
31
(
9
),
095007
(
2022
).
52.
T.
Morishita
,
R.
Tsukizaki
,
K.
Nishiyama
, and
H.
Kuninaka
,
J. Appl. Phys.
131
(
1
),
013301
(
2022
).
53.
S.
Mazouffre
,
D.
Gawron
, and
N.
Sadeghi
,
Phys. Plasmas
16
(
4
),
043504
(
2009
).
54.
C. J.
Durot
,
A. D.
Gallimore
, and
T. B.
Smith
,
Rev. Sci. Instrum.
85
(
1
),
013508
(
2014
).
55.
C. J.
Durot
, Ph.D. thesis,
University of Michigan
,
2016
.
56.
A.
Diallo
,
S.
Keller
,
Y.
Shi
,
Y.
Raitses
, and
S.
Mazouffre
,
Rev. Sci. Instrum.
86
(
3
),
033506
(
2015
).
57.
C. V.
Young
,
A. L.
Fabris
,
N. A.
MacDonald-Tenenbaum
,
W. A.
Hargus
, and
M. A.
Cappelli
,
Plasma Sources Sci. Technol.
27
(
9
),
094004
(
2018
).
58.
B.
Pelissier
and
N.
Sadeghi
,
Rev. Sci. Instrum.
67
(
10
),
3405
3410
(
1996
).
59.
C.
Biloiu
,
X.
Sun
,
E.
Choueiri
,
F.
Doss
,
E.
Scime
,
J.
Heard
,
R.
Spektor
, and
D.
Ventura
,
Plasma Sources Sci. Technol.
14
(
4
),
766
776
(
2005
).
60.
R. A.
Gottscho
,
R. H.
Burton
,
D. L.
Flamm
,
V. M.
Donnelly
, and
G. P.
Davis
,
J. Appl. Phys.
55
(
7
),
2707
2714
(
1984
).
61.
M. J.
Goeckner
,
S. M.
Malik
,
J. R.
Conrad
, and
R. A.
Breun
,
Phys. Plasmas
1
(
4
),
1064
1074
(
1994
).
62.
B.
Jacobs
,
W.
Gekelman
,
P.
Pribyl
, and
M.
Barnes
,
Phys. Rev. Lett.
105
(
7
),
075001
(
2010
).
63.
B.
Jacobs
,
W.
Gekelman
,
P.
Pribyl
, and
M.
Barnes
,
Phys. Plasmas
18
(
5
),
053503
(
2011
).
64.
N. B.
Moore
,
W.
Gekelman
,
P.
Pribyl
,
Y.
Zhang
, and
M. J.
Kushner
,
Phys. Plasmas
20
(
8
),
083506
(
2013
).
65.
N. B.
Moore
,
W.
Gekelman
, and
P.
Pribyl
,
J. Vac. Sci. Technol., A
34
(
2
),
021303
(
2016
).
66.
S. W.
Mattingly
and
F.
Skiff
,
Rev. Sci. Instrum.
89
(
4
),
043508
(
2018
).
67.
R.
Hood
,
S. D.
Baalrud
,
R. L.
Merlino
, and
F.
Skiff
,
Phys. Plasmas
27
(
5
),
053509
(
2020
).
68.
P. J.
Hargis
et al,
Rev. Sci. Instrum.
65
(
1
),
140
154
(
1994
).
69.
J. K.
Olthoff
and
K. E.
Greenberg
,
J. Res. Natl. Inst. Stand. Technol.
100
(
4
),
327
339
(
1995
).
70.
A.
Kramida
,
Yu.
Ralchenko
,
J.
Reader
, and
NISR ASD Team
, NIST atomic spectra database, version 5.10,
2022
, https://www.nist.gov/pml/atomic-spectra-database.
71.
S.
Mändl
,
R.
Günzel
, and
W.
Möller
,
J. Phys. D: Appl. Phys.
31
(
9
),
1109
1117
(
1998
).
72.
M. A.
Lieberman
,
J. Appl. Phys.
66
(
7
),
2926
2929
(
1989
).