LPP-EUV Light Source Development for High Volume Manufacturing Lithography

By Hakaru Mizoguchi, Junichi Fujimoto and Takashi Saitou

EUV Source System

We have developed a prototype of the first HVM EUV light source having 100 kHz 20 kW CO2 laser, 20 mm in diameter droplet generator, and magnetic field debris mitigation (Fig. 1).

Fig. 1: GL200E HVM EUV Source Device
Fig. 1: GL200E HVM EUV Source Device

The major specifications of the first generation HVM EUV light source are shown in Table 1.

Table 1: Specification of GL200E
Table 1: Specification of GL200E

High CE Experiment

We have investigated EUV plasma generation scheme by our small experimental tool which is operated at the repetition rate of 10 Hz (maximum). Fig. 2(A) shows the experimental setup for the basic investigation of EUV light generation and Sn debris mitigation. This tool is capable of simulating conditions of EUV light generation identical to those in ETS and GL200E, such as pulse duration and pulse energy of CO2 laser and pre-pulse laser, Sn droplet size, and magnetic field environment except for the repetition rate. The tool’s compactness makes it easier to measure and optimize various plasma generation parameters and results. The small experimental tool consists of various sub systems, such as a short-pulsed high-energy CO2 laser, a pre-pulse laser, a Sn droplet generator and a EUV vacuum vessel with a solenoid magnet. The droplet generator can supply a droplet with a diameter of around 20 mm. The system operates at a repetition rate of 10 Hz at maximum. The vacuum vessel is evacuated by a turbo molecular pump and a dry pump.

Fig. 2(A): 10 Hz EUV Source Device
Fig. 2(A): 10 Hz EUV Source Device

We have investigated the CE as a function of the droplet diameter with/without pre-pulse laser conditions. Fig. 2(B) shows the results, which indicate that high CE can be obtained even with the small droplet size. The pre-pulse laser condition is a key parameter for obtaining higher CE. The CE reached 3.3 percent with the 20 mm droplet by optimizing the pre-pulse laser conditions.

Fig. 2(B) Droplet Diameter vs. CE
Fig. 2(B): Droplet Diameter vs. CE

Currently we have demonstrated the new technical concept; High CE of 4.7 percent with 20 mm droplets and shorter pulse duration pre-pulse laser (10 ps).

Fig. 3: CE vs. CO2 Laser Pulse Energy by Pre-Pulse Laser Duration
Fig. 3: CE vs. CO2 Laser Pulse Energy by Pre-Pulse Laser Duration
Fig. 4: CO2 Laser Power Requirement at High Efficiency
Fig. 4: CO2 Laser Power Requirement at High Efficiency

The meaning of this Ce = 4.7 percent is shown in Fig. 4. In case of 250 W EUV power, we need only 21 kW CO2 laser. Also in case of 500 W EUV power, we need 40 kW CO2 laser. This technology reduces the CO2 laser power dramatically.

Magnetic Mitigation

When a Sn droplet target is irradiated with pre-pulse laser and/or CO2 laser beams, the Sn droplet is spread over in the vessel as plasma and several states of Sn. The Sn is classified generally into fragments, neutral atoms and ions. During spreading process Sn plasma emits EUV light. Residues of the plasma after emitting EUV light are eventually scattered inside the vessel. To prevent the collector mirror from being contaminated, Sn debris needs to be trapped before being deposited on the collector mirror. This Sn-excitation scheme is shown in Fig. 6. To enhance EUV energy and to maximize Sn debris mitigation, Sn ions should be maximized in these laser irradiation processes. We believe that the shape of Sn target is crucial. To realize it, the double laser irradiation process is utilized in our system. The scheme of collecting process is shown in Fig. 5.

Fig. 5: Debris Mitigation Schematic (2)
Fig. 5: Debris Mitigation Schematic (2)
Fig. 6: Neutral Sn Atom Measurement with LIF
Fig. 6: Neutral Sn Atom Measurement with LIF

Fig. 7 shows the result of the LIF measurements. When there is no main CO2 laser irradiation, the neutral atoms and the fragments are observed. With the CO2 laser irradiation, however, no neutral atoms and fragments are observed in these pictures. Fig. 8 shows the detailed data on the LIF signals. By comparing the signals with/without the CO2 laser irradiation, 93 percent of the Sn atoms are ionized and 7 percent of the Sn atoms remain as neutral atoms.

Fig. 7: LIF Measurement Result (1)
Fig. 7: LIF Measurement Result (1)
Fig. 8: LIF Measurement Result (2)
Fig. 8: LIF Measurement Result (2)
Fig. 9: Summary of Particle Observation
Fig. 9: Summary of Particle Observation

Experimentally measured magnetic field debris mitigation more than 98 percent of Sn ions collected by a magnetic field.

Latest Status of GL200E Construction

Based on the engineering data of the ETS and the small experimental device, now we are developing our first generation HVM light source, GL200E, in our facilities. Our GL200E system outlook as shown in Fig. 10 shows the first GL200E EUV light source system construction. The laser system is assembled in our clean room. The next step in the construction is the demonstration of power and short/long-term performance stability and debris free operation.

The output power is important for an HVM system. We have obtained plotted data by now. Currently, we have achieved 7 W EUV clean power at I/F with 90 kHz operation with 30 percent duty cycle (Fig. 11). Our final target 250 W is a challenge to enhance CO2 laser power from 13 kW to 20 kW and CE from 3 percent to 5 percent under Sn debris free operation.

Fig. 10: Picture of GL200E Prototype Construction
Fig. 10: Picture of GL200E Prototype Construction
Fig. 11: Picture of Proto EUV Source Operation
Fig. 11: Picture of Proto EUV Source Operation

Conclusion

We have investigated the EUV plasma generation scheme by small experimental tool which is facilitated less than 10 Hz operation. We have proposed double laser pulse irradiation method to generate LPP plasma efficiently. At this moment we have found the operation condition for obtaining CE of 4.7 percent by using 10 ps pre-pulse laser. Also, we have obtained 93 percent Sn ionization rate when the droplets are irradiated with double laser pulses under proper CO2 laser pulse conditions.

This work was partly supported by the New Energy and Industrial Technology Development Organization (NEDO), JAPAN.

Hakaru Mizoguchi, Junichi Fujimoto and Takashi Saitou are with Gigaphoton, Inc.

About the Author
Steven Glover is a proud member of the LIA staff. When he is not at work he is actively involved in several charitable efforts.
Posted in Laser Editorials Tagged