Search

OBJECTIVE: To investigate the clinical characteristics, risk factors and prognosis of ulcerative keratitis (UK) in patients with rheumatoid arthritis (RA). METHODS: Sixteen patients with UK (the UK group) were screened among 4 773 RA patients hospitalized at Peking University People's Hospital from January 2003 to May 2021, and 72 patients with RA without UK in the control group were selected by propensity score matching (PSM). The clinical features, laboratory data and medications between the two groups were collected and analyzed. RESULTS: The average age of the patients with RA was 60 years at the time of the complication of UK. UK appeared on average in the 16th year of RA, with the majority of UK occurring in the 7th year of RA and beyond (14 cases), however, a minority of UK was the first symptom of RA (1 case). UK associated with RA manifested as simple marginal UK in 8 cases, marginal UK with perforation in 5 cases, marginal UK with uveitis in 2 cases and central UK with perforation in 1 case. The number of swollen joints was significantly higher in the UK group than in the control group [6.0(2.5, 23.0) vs. 3.0(1.0, 9.8), Z=-2.047, P=0.041], and the proportion of secondary Sjögren syndrome (SS, 37.5% vs. 6.9%, χ(2)=11.175, P=0.004) and interstitial lung disease (ILD, 37.5% vs. 8.3%, χ(2)=9.456, P=0.008) were significantly higher in the RA patients with UK than the patients without UK. The uses of sulfasalazine (12.5% vs. 48.6%, χ(2)=7.006, P=0.008), leflunomide (31.3% vs. 63.9%, χ(2)=5.723, P=0.017) and gold salt(6.3% vs. 33.8%, χ(2)=4.841, P=0.032)were significantly lower in the UK group than in the control group. However, there was no statistically significant comparison between the two groups for the 28-joint disease activity score-erythrocyte sedimentation rate (5.3±1.8 vs. 5.1±1.6, t=0.309, P=0.761). Logistic analysis showed the number of swollen joints (OR=1.148), secondary SS (OR=79.118), complicated with ILD (OR=6.596) and the use of sulfasalazine (OR=0.037) were independent relevant factors (P < 0.05) for UK in RA patients. CONCLUSION: The number of swollen joints, secondary SS, complicated with ILD and the use of sulfasalazine are independent relevant factors for UK in patients with RA.

In pulse-modulated (PM) radio frequency (RF) atmospheric pressure glow discharges (APGDs), the alpha–gamma mode transition is analyzed and confirmed with $I$ – $V$ characteristic curve. The behaviors of alpha–gamma mode transition in PM RF APGDs are more complicated than that in continuous wave (CW) RF APGDs. In terms of amplitude of applied RF voltage just before a pulse falling edge ( $V_{\mathrm {appls}}$ ), the critical $V_{\mathrm {appls}}$ at the alpha–gamma mode transition in a PM RF APGD increases with the decreasing duty ratio and is higher than that of a CW RF APGD. The feature of the alpha–gamma mode transition in a PM RF APGD is attributed to the gas temperature rise depending on both the RF voltage and duty ratio. The alpha–gamma mode transition is verified to occur at the initial phase of a PM RF APGD and the onset is earlier at a higher transient RF voltage. As the mode transition occurs, a RF voltage spike is formed in the pulse waveform.

In a pulse-modulated radio frequency atmospheric pressure dielectric barrier discharge in helium, the influence of the pulse width on the transition from a patterned glow discharge to a homogeneous glow discharge is experimentally investigated. The experimental results show that the discharge mode can transit from a patterned glow to a homogeneous glow at a moderate pulse width ( 7 μ s) but only has a patterned glow at a short pulse width ( 4 μ s) or a homogeneous glow at a long pulse width ( 18 μ s). The characteristic regions of patterned and homogeneous glow discharges are also acquired. The results can be explained using the breakdown characteristics of the pulse-modulated radio frequency atmospheric pressure discharge.

The physical polarities of electromagnetic waves propagating in a magnetoplasma-filled cylindrical waveguide are studied by numerical calculations and simple analytical analyses. In the plane of the frequency-axial wave number. the obtained results show that dispersion curves of right-hand polarized (RHP) and left-hand polarized (LHP) electromagnetic waves propagating in an infinite magnetoplasma determine characteristic regions in which wave modes propagating in a magnetoplasma-filled cylindrical waveguide have different physical polarities. The frequency region slightly below and above ω p1 (ω r1 denotes the lower-branch frequency of the RHP electromagnetic wave propagating in an infinite magnetoplasma) is characterized by pure or physical dominant RHP modes. In the frequency region of ω l ω r2 can be thought of as a RHP area. The physical polarity of wave modes clarified in this paper is helpful to understand the interaction between charged particles and waves propagating in a magnetoplasma-filled cylindrical waveguide.

The rf self-bias on the substrate in a rf inductively coupled plasma is controlled by varying the impedance of an external tuning inductor and capacitor network inserted between the substrate and the ground. In term of variations of tuned substrate self-bias (Vtsb) with tuning capacitance (Ct), three characteristic regions, namely continuous, instable, and bistable regions are found, corresponding to low, moderate, and high discharge power at a constant gas pressure, respectively. A novel self-oscillation observed in the instable region occurs only within a limited domain of Ct. With the increment in Ct, the self-oscillation frequency increases while its amplitude decreases. Near the boundary of the instable and bistable regions, the self-oscillations shift downward and upward at lower and higher Ct, respectively. The possible mechanism for the self-oscillation is discussed.

The influences of the pulse off-time on the breakdown voltage of the first pulse and the stable pulse discharge (having repeatedly undergone a process of ignition, maintenance, and extinction) are experimentally investigated in a pulse-modulated radio-frequency atmospheric pressure argon discharge. The experimental results show that the first pulse discharge breakdown voltage decreases, but the stable pulse discharge breakdown voltage increases with increasing the pulse off-time. In a large region of the pulse off-time, the luminescence property of the initial breakdown stage is studied using a high speed camera. The captured images at different pulse off-times demonstrate that the gas breakdown exhibits five key characteristics: single-point random breakdown, multi-point random breakdown, stable uniform breakdown, stable glow mixed with pattern breakdown, and stable nonuniform pattern breakdown. The physical reasons for these results are discussed.

The overshoot and undershoot of the applied voltage on the electrodes, the discharge current, and radio frequency (RF) power were observed at the initial phase of pulse-modulated (PM) RF atmospheric pressure discharges, but factors influencing the overshoot and undershoot have not been fully elucidated. In this paper, the experimental studies were performed to seek the reasons for the overshoot and undershoot. The experimental results show that the overshoot and undershoot are associated with the pulse frequency, the rise time of pulse signal, and the series capacitor Cs in the inversely L-shaped matching network. In the case of a high RF power discharge, these overshoot and undershoot become serious when shortening the rise time of a pulse signal (5 ns) or operating at a moderate pulse frequency (500 Hz or 1 kHz).

Summary form only given. Pulse-modulated RF atmospheric pressure discharges were investigated in recent years to reduce the thermal accumulation of the discharge and extend the operation region of the stable apha glow mode. Different pulse-voltage and current waveforms were acquired in previous experiments, but no attention was paid to the interpretation. We investigated this issue by the aid of positive and negative feedback effects deducted by varying the series capacitor in the inversely L-shaped matching network employed in our pulse-modulated RF atmospheric pressure discharge. The evolution of the RF pulse is found to be influence by the igniting voltage and the feedback effect, as a result, a variety of waveforms present by igniting the discharge in the rising edge or the plateau, and the positive or negative feedback region. The spikes or the bird-heads of the pulsed waveforms appear in the negative feedback region. The pulse undershot in final stage of the spike could be substantial that the RF discharge extinguished within a time interval and subsequently re-ignited. The positive feedback region is relatively narrower, and the pulsed RF discharge originally ignited in the positive feedback region can develop into the negative feedback region.

A tunable center-tapped transformer is proposed to increase the output of a rf magnetic probe and improve the signal-to-noise ratio. The tuning is implemented by a variable capacitor connected parallel with the primary winding of the tunable center-tapped transformer. Undesirable common-to-differential conversion is reduced by installing a compensating capacitor. In addition, a planar Faraday shield is installed between the windings of the transformer to further suppress the electrostatic coupling. It is found that tuning the variable capacitor can result in a resonance in the output voltage of the rf magnetic probe. The largest output voltage, achieved with the tunable magnetic probe under the optimal condition, is higher than that with a conventional one by an order of magnitude. Effects of the compensating capacitance on the common-mode output voltage are studied and discussed. Influences of parameters such as cable length, the coupling coefficient, and the step-up ratio of the transformer on the output voltage are also presented. Analytical derivations and numerical calculations based on the equivalent circuit are performed to elucidate the characteristics of the differential mode.

In a low-pressure radio-frequency (13.56 MHz), inductively coupled argon plasma generated by a normal cylindrical rf coil, electric field, current density, and absorbed power density is calculated from magnetic field measured with a phase-resolved magnetic probe. The anomalous skin effect (ASE) for the cylindrical rf coil is compared to those previously reported for the planar and re-entrant cylindrical rf coils. Physical reasons for our observed characteristics of ASE are presented. With the increasing discharge power, the size and the number of negative and positive power absorption regions evolve into several distinct patterns. For the low discharge power (at 156.9 W), there is one area of positive and one area of negative power absorption in the radial direction. For the medium discharge power (279 W–683.5 W), there are two areas of negative and two areas of positive power absorption. For the even higher discharge power (above 803.5 W), the number of areas is the same as that of the medium discharge p...

In a pulse-modulated (PM) radio-frequency (RF) capacitively coupled plasma, the DFT (Discrete Fourier Transform)-calculated RF power and the corresponding phase shift between voltage and current measured with calibrated voltage and current probes present oscillations in the pulse rising and falling edges. The oscillating phase shift between voltage and current obtained in the falling edge is outside the expected value for a resistive-capacitive RF discharge. Numerical simulation and analytical analysis are made to interpret these abnormal characteristics and seek an approach to obtaining the reliable resistive (active) RF power. The oscillation is proved to be originated from the oscillating non-zero reactive RF power of the capacitor(s) in the load. At the time instant when the reactive RF power within an integer RF period is not zero, the reactive RF power is mistakenly regarded as the active RF power in the DFT analysis, as a result, the corresponding phase is thus incorrect and even outside the expected value for a resistive-capacitive load. The resistive RF power and the phase can be only correctly calculated at the time instant when the reactive RF power is zero. For a series (or parallel) RC (resistor-capacitor) load and a combined RC load with the dominated series (or parallel) RC impedance, the time instant of the zero reactive RF power is calculated with one of the two proposed empirical formulae. In practice, the DFT-calculated resistive RF power is obtained according to the following procedures: (1) applying DFT to the measured RF voltage and current signals to obtain the power and time instants for minimal phase shifts between voltage and current; (2) selecting the empirical formula to calculate time instants of the zero reactive RF power; (3) getting resistive powers at time instants of the zero reactive RF power. In real PM RF capacitively coupled plasmas, the empirical formula for the series RC load is selected to calculate the resistive RF power. The accuracy of DFT-calculated resistive RF power is proved to be related to two kinds of errors. The first is the error of the time instant of the zero reactive RF power calculated using the empirical formula. This error is relatively lower when the requirement that the dominated parallel or series RC impedance is met and is almost independent of the impedance phase angle of a combined RC load. The second is the error of the DFT-calculated resistive RF power compared with the corresponding time integral RF power at the real zero reactive RF power. This error is independent of the load type or the load impedance but varies with the slope of PM RF voltage amplitude vs. time. The two kinds of errors both increase in the pulse rising and falling edges.

In pulsed RF atmospheric pressure glow discharges, the gas breakdown judged by the rapid drop in the amplitude of the pulsed RF voltage is no longer universally true. The steep increment of the plasma-absorbed RF power is proposed to determine the gas breakdown. The averaged plasma-absorbed RF power over a pulse period is used to evaluate effects of the preceding pulsed RF discharge on the breakdown voltage of the following one, finding that the breakdown voltage decreases with the increment in the averaged plasma-absorbed RF power under constant pulse duty ratio. Effects of the pulse off-time on the breakdown voltage and the breakdown delay time are also studied. The obtained dependence of the breakdown voltage on the pulse off-time is indicative of the transitional plasma diffusion processes in the afterglow. The breakdown voltage varies rapidly as the plasma diffuses fast in the region of moderate pulse off-time. The contribution of nitrogen atom recombination at the alumina surface is demonstrated in the prolonged memory effect on the breakdown delay time vs. the pulse off-time and experimentally validated by introducing a trace amount of nitrogen into argon at short and long pulse off-times.

In this paper, the resonant rf magnetic probe is upgraded by replacing the rotary capacitor in the old version with the series-connected coaxial cable. The numerical calculation and the measurement with the prototype probe show that the rf magnetic probe can achieve resonance at a middle length of the series-connected coaxial cable. The good electrical symmetry of the new rf magnetic probe is ensured by both the identity of series-connected coaxial cables and the new structure of the primary winding. Practical measurements conduced on an rf inductively coupled plasma source demonstrate that performances of the new rf magnetic probe are good.

Pulse-modulated RF atmospheric pressure glow discharges (APDGs) were investigated in recent years to reduce the thermal accumulation and extend the operation region of the stable alpha glow mode. Different pulse-modulated voltage and current waveforms were acquired in previous experiments, but no attention was paid to the interpretation. We investigated this issue and associated phenomenon via positive and negative feedback effects derived from varying the series capacitor in the inversely L-shaped matching network used in our pulse-modulated RF APGD source. The evolutions of pulse-modulated RF waveforms were found to be associated with the feedback region and the pulsed plasma absorbed RF power. In the positive feedback region, pulse-modulated RF APGDs are relatively stable. In the negative feedback region, wide spikes as well as undershoots occur in RF voltage and current waveforms and the plasma absorbed RF power. In case of a high RF power discharge with a low modulation frequency, the pulse-modulated RF APGD is extinguished and re-ignited due to the enhanced undershoot during the initial pulse phase. The pulse-modulated RF APGD can transit from positive to negative feedback region in a range of series capacitance. Experimental results are discussed by the aid of equivalent circuit, negative and positive feedback effects.