Objective The laser with stable output energy is critical for lithography. For the closed-loop control of the deep-ultraviolet excimer light source, the laser pulse parameters need to be measured on-line with high precision. Traditional excimer laser energy measurement devices are based on low repetition rate. In order to meet the pulse measurement requirements of high repetition rate excimer lasers, the on-line measurement system for pulse parameters of high repetition rate excimer laser is designed. The pulse energy acquisition is realized by using peak holding circuit and high speed ADC. The optimization method for peak holding circuit design is proposed by simulating the parasitic equivalent model of peak holding circuit.

Methods The measurement system for pulse parameters that is designed includes the light path of the former stage and the processing circuit of the latter stage. The measurement system for pulse parameters is placed at the output of the main optical path, and the laser output light passes through the beam splitter, with 95% of the energy output light being used for subsequent lithography work, and 5% of the output light being used for detection, which effectively reduces the loss of the output signal, as shown in Fig.4. The block diagram of the detection circuit is shown in Fig.5. The principle and transient analysis of the peak holding circuit are shown in Fig.6-8. The detection circuit includes a trans-impedance amplification circuit, a filter circuit, a peak holding circuit, and a switching circuit for synchronous triggering of the signal. The peak voltage is maintained and broadened by the peak holding circuit, then the peak value is sampled by a 16-bit 20 MHz high-speed and high-precision ADC, and the collected signal is processed by FPGA.

Results and Discussions The test performance of the measurement system for pulse parameters is carried out with an ArF excimer laser. The output of the measurement system for pulse parameters is square wave signal with wide pulse width, which is convenient for ADC acquisition. The response time of the measurement system for pulse parameters is about 150 ns. Besides, the repeatability experiment of the measurement system for pulse parameters yielded a relative error of 0.22% relative to the standard energy meter while the ArF excimer laser works at 6 kHz. And the max relative error is 0.56%. A comparison of the reference energy values of a standard energy meter and those measured by the measurement system for pulse parameters is shown in Fig.13 with the laser operated in constant energy mode. The relationship between the output voltage of the measurement system for pulse parameters and the energy of input signal is shown in Fig.14 with 500 pluses, and the curve is fitted by the least squares method to obtain the fitting relationship. According to the analysis, the correlation coefficient of the linearity is 99.83%.

Conclusions The system is designed for measuring the pulse parameters of a deep ultraviolet excimer light source for photolithography, and on-line accurate measurement of the pulse parameters of an excimer laser is achieved. The feasibility of the measurement scheme is verified by comparing the theoretical calculation with the measured results. Based on the equivalent model and response analysis of the hardware circuit, the reason and elimination method of the peak effect of the peak holding circuit are studied. The measurement system was tested on a 193 nm ArF excimer light source at a constant energy mode of 6 kHz, and realized the on-line extraction of repetition frequency and pulse width and the measurement of repetition, consistency and linearity of pulse energy, the results verify that the linearity of the pulse parameter measurement system is 99.83%, the relative error with the standard energy meter is not more than 0.56%, and the average error is 0.22%. It can meet the requirement of on-line measurement of pulse parameters of ArF excimer light source.