ECR Hydrogen Research Articles

Probe measurement of an ECR hydrogen plasma facing the C12A7 electride surface

Plasma parameters of an ECR discharge excited in a compact ion source with a replaceable plasma electrode (PE) were measured with a Langmuir probe. The PE was installed to investigate how an electride PE changes the plasma parameters from those with a Mo PE. The effect upon the plasma parameters in front of the PE due to the material will be elucidated by changing the probe position axially along the beam extraction direction. Analyses of the measured probe characteristics revealed that electride PE has a higher electron density (ne) and lower electron temperature (Te) than the Mo PE. When the PE was biased negatively, the ratio of the positive saturation current (Ii ) to the negative saturation current (Ie +Ih- ) for the electride PE was about 1.6 times of the ratio for the Mo PE.

Temperature peaking at beginning of breakdown in 2.45 GHz pulsed off-resonance electron cyclotron resonance ion source hydrogen plasma

An experimental study of temperature and density evolution during breakdown in off-resonance ECR hydrogen plasma is presented. Under square 2.45 GHz microwave excitation pulses with a frequency of 50 Hz and relative high microwave power, unexpected transient temperature peaks that reach 18 eV during 20 μs are reported at very beginning of plasma breakdown. Decays of such peaks reach final stable temperatures of 5 eV at flat top microwave excitation pulse. Evidence of interplay between incoming power and duty cycle giving different kind of plasma parameters evolutions engaged to microwave coupling times is observed. Under relative high power conditions where short microwave coupling times are recorded, high temperature peaks are measured. However, for lower incoming powers and longer coupling times, temperature evolves gradually to a higher final temperature without peaking. On the other hand, the early instant where temperature peaks are observed also suggest a possible connection with preglow processes during breakdown in ECRIS plasmas.

Surface Cleaning Effects of Silicon Substrates by ECR Hydrogen Plasma on Subsequent Homoepitaxial Growth

We have demonstrated the preparation of the almost defect-free homoepitaxial layer and the defective layer, respectively, with and without applying the in-situ cleaning of the silicon substrate surface using electron cyclotron resonance hydrogen plasma. Secondary ion mass spectroscopy indicated that the interfacial oxygen and carbon concentrations, respectively, decreased and increased with the in-situ cleaning. We have investigated the effect of process parameters such as microwave power, d.c bias, and cleaning time, on the epitaxial growth, by evaluating the cross-sectional transmission electron microscopy images of the subsequently deposited Si homoepitaxial film.

Low temperature growth of homoepitaxial film on Si substrate cleaned in-situ by ECR hydrogen plasma

Low temperature growth of homoepitaxial film on Si substrate cleaned in-situ by ECR hydrogen plasma

Plasma cleaning of carbon species for silicon homoepitaxial growth

As device dimensions are reduced into the sub-micron region in pursuit of higher integration density and better circuit performance, low temperature processing, including low temperature cleaning and low temperature epitaxial growth, is becoming important. In order to achieve the high-quality epitaxy with the low temperature processing, the technique of in-situ plasma cleaning has been developed [1, 2]. Although there have been some reports on the plasma cleaning of surface oxygen prior to epitaxial growth, there are rare reports on the systematic study on the removal of surface carbon. Carbon is much more tenacious and stable on the silicon surface than oxygen and temperatures over 1100 ◦C are required to remove carbon from the surface by thermal desorption process. Here, we have performed the low temperature in-situ cleaning in order to reduce the interfacial carbon concentration. Substrates were 4 inch, czochralski-grown, p-type (100) silicon with 0.5–20 -cm resistivity. The wafers were RCA cleaned and HF dipped for 20–30 s in 10:1 aqueous solutions and rinsed in DI (deionized) water and then dried by blowing nitrogen on them. All the processes were done inside the class 100 cleanroom and it took only 10 s to load the wafer into the load lock chamber of the CVD reactor after the wafer was blowdried. After the wafers were transferred and loaded onto the heater stage, the main chamber was pumped down and ultimately 1–2 × 10−8 Torr could be attained. In-situ predepositon wafer cleaning was done by using ECR hydrogen plasma. The ECR hydrogen plasma is able to deliver a higher density of low energy and light hydrogen ions to the wafer, resulting in highly efficient cleaning without substrate damage [2]. The ECR chamber was at the side of the CVD chamber. ECR was operated at the 2.45 GHz S-band microwave frequency. Depositions were done by flowing 10 sccm SiH4 without carrier gases, immediately after the plasma was extinguished. We have applied the in-situ cleaning condition with a microwave power of 300 W, DC bias of 10 V, pressure of 1 mTorr, and the cleaning temperature of 600 ◦C and subsequently have deposited the silicon epitaxial layer. Fig. 1 shows the XTEM image of silicon epilayer and the interface, revealing that the high-quality epitaxial layer is produced by applying the in-situ plasma cleaning. Fig. 2 shows the SIMS depth profile

Removal efficiency of organic contaminants on Si wafer by dry cleaning using UV/O 3 and ECR plasma

The removal efficiency of the organic contaminants existing on the surface of silicon wafers by a dry cleaning method using UV/O 3 and ECR plasma is discussed in this paper. After cleaning, Si wafers are characterized by attenuated total reflection-Fourier transform infrared spectroscopy (ATR-FTIR) and atomic force microscropy (AFM). In UV/O 3 cleaning, the exposure time to reach the detection limit for organic contaminants in ATR-FTIR is about 10 min and the RMS surface roughness reduces with increasing exposure time. In ECR hydrogen plasma cleaning, the RMS surface roughness reduces significantly with increasing the exposure time but the removal of organic contaminants from the silicon wafer is not very effective. In ECR oxygen plasma cleaning, the exposure times to reach both the detection limit and the lowest RMS roughness are 40 and 10 s, respectively, and the optimum exposure time is suggested to be 30–40 s, considering both the effects of cleaning and surface roughening. Therefore, dry cleaning using ECR oxygen plasma seems to be more effective than that using ECR hydrogen plasma or the UV/O 3 cleaning for the removal of organic contaminants. Also, the removal mechanisms of the organic contaminants in UV/O 3 and ECR plasma cleanings are discussed in detail.

Measurements of ECR silane plasma parameters

The plasma parameters of an ECR silane plasma were measured with both a heated Langmuir probe and an 8 mm microwave interferometer. It was found that the heated Langmuir probe is a powerful tool for measuring the plasma parameters. Many negative ions were also found to be produced in the plasma. An attempt of ion temperature measurements in an ECR hydrogen plasma was made with an ion-sensitive probe and negative ions were found to exist.

Improved stability of a-Si:H fabricated from SiH 2Cl 2 by ECR hydrogen plasma

Marked improvement was observed in a-Si:H prepared from SiH 2Cl 2 at the high growth rate of 15 Å/s by ECR-hydrogen plasma. The defect densities in the annealed state and the saturated state after light soaking are maintained at low levels, (1–5) × 10 15 cm −3 and (2–5) × 10 16 cm −3, for the films with a wide range of the structural parameters such as the optical gap (1.75–1.95 eV), hydrogen content (5–20 at.%), and content of SiH 2 bonds (2–10 at.%). In addition, no marked changes were observed in the behavior of light induced degradation as the Fermi level position of the films was varied in a range of 0.6 eV. This leads us to conclude that the defect precursors are efficiently eliminated from the Si-network during the film growth.

Electrically reversible depassivation/passivation mechanism in polycrystalline silicon

An electrically reversible depassivation/passivation phenomenon has been found for hydrogen passivated polysilicon. This reversible effect is seen in both ECR and rf hydrogen passivated n-channel thin films transistors (TFTs) but is not seen in the corresponding hydrogen passivated p-channel TFTs, nor is it seen in either n- or p-channel TFTs before hydrogenation. This phenomenon was observed while room temperature bias stressing n- and p-channel TFTs fabricated on the same solid phase crystallized polysilicon films on quartz substrates. A model involving hydrogen release or capture at defects and positively charged hydrogen motion in device electric field is proposed. This phenomenon has significant implications for polysilicon TFT design and operation.

A New Electrically Reversible Depassivation/Passivation Mechanism in Polycrystalline Silicon

AbstractAn electrically reversible depassivation/passivation phenomenon, recently found for hydrogen passivated polysilicon [1] is further explored in this report. This reversible effect is seen in both ECR and RE hydrogen passivated n-channel thin film transistors (TFTs) but is not seen in the corresponding hydrogen passivated pchannel TFrs, nor is it seen in either n- or p-channel TETs before hydrogenation. This phenomenon has been observed when room temperature bias stressing TFTs fabricated on solid phase or laser crystallized polysilicon films on quartz substrates [1]. A model involving hydrogen release or capture at defects, positively charged hydrogen motion in device electric fields, and subsequent hydrogen capture at other defects has been proposed. This phenomenon has significant implications for polycrystalline silicon TFT’ design and operation. By extension, it also offers significant insight into the stability problems of hydrogenated amorphous silicon.

Fabrication of Highly Stable a-Si:H from Chlorinated Precursors

ABSTRACTHigh quality a-Si:H was fabricated from SiH2Cl2 by ECR hydrogen plasma at high growth rate (>15 A/s). After sticking the precursors, SiHnClm (n+m≤3), stable Si-network was formed by chemical reactions enhanced on the heated substrate due to strong interaction between Cl and H bonded with Si. The defect densities of 3×1015 and 3×1016 cm3 were obtained for the films in the as-grown and the saturated states after light soaking, respectively. In these stable films, a surprisingly large amount of voids (2–10 vol%) was found by the simulation of the spectra of pseudo-dielectric constant measured by spectral ellipsometry.

Stable a-Si:H fabricated from halogenous silane by ECR hydrogen plasma

A-Si:H stable against light soaking was fabricated from SiCl 2H 2 fed into an activated hydrogen flow generated by an ECR plasma. Both hydrogen content (CH, at.%) and optical gap (Eg, eV) were reduced simultaneously with increasing substrate temperature (Ts, °C), which implies that structural relaxation is enhanced at the growing surface due to a strong interaction of hydrogen and chlorine. Meanwhile, hydrogen in the Si network was reduced to 6 at.% at a Ts of 220 °C or lower by hydrogen treatment without reducing the optical gap. These films showed some attractive features, viz., low electric conductivity of < 10 −13 (S/cm) at room temperature and low defect density, (3–5)× 10 15 (cm −3) in the annealed state and (3–5)× 10 16 (cm −3) in the saturated state after prolonged light illumination.

Real-time Monitoring and Control of Silicon Epitaxy Using Emission Fourier Transform Infrared Spectroscopy

AbstractReal-time epi-film thickness is measured by an Emission Fourier Transform Infrared Spectrometer (E/FT-IR). The E/FT-IR takes advantage of the heated wafer as the source of IR radiation. In our experiments, wafers were cleaned using in-situ ECR hydrogen plasma followed by film growth. The cleaning and deposition processes were monitored in real-time using the E/FT-IR technique. We have demonstrated the application of E/FT-IR for observing real-time growth rates and incubation times. Based on these real-time observations, the predeposition plasma cleaning process and the deposition process can be effectively monitored and controlled in real-time. Application of E/FT-IR in optimizing the predeposition hydrogen plasma cleaning process was demonstrated.

Electron Cyclotron Resonance Hydrogenation of Poly-Si Thin Film Transistors on SiO2/Si Substrates

ABSTRACTWe report the results of an ECR hydrogenation time-behavior study on polycrystalline silicon thin film transistors. The device characteristics, such as mobility, subthreshold slope, threshold voltage, off current and on/off current ratio, were measured as a function of the passivation time. It was found that there is an optimal passivation time for this particular TFT structure, and overexposure to ECR hydrogen plasma may cause the degradation of some device parameters.

Simulation Experiment on Removal of Carbon Film from Observation Window of Carbonized Fusion Machine

Although metal and oxygen impurities in nuclear fusion machines can be reduced by first-wall carbonization, this procedure contaminates the surfaces of observation windows and other parts. This paper describes the removal of carbon films from a quartz sample using a locally produced ECR hydrogen plasma as a method of overcoming this problem.