for alternating phase shift mask fabrication

Transcription

1 Practical phase control technique for alternating phase shift mask fabrication Miho Takahashi,*l Akihiro Miyake,*2 Hidetaka Saitou,*2 Hiroyuki Miyashita,*3 and Shiaki Murai*2 *1 Mask Engineering Development Center, Hitachi, Ltd., 20-1, Josuihon-cho 5chome, Kodaira-shi, Tokyol , Japan *2 Semiconductor Components Operations, Dai Nippon Printing Co.,Ltd., *3 Semiconductor Components Laboratory, Dai Nippon Printing Co.,Ltd., 2-1, Fukuoka 2chome, Kamifukuoka-.shi, Saitama , Japan ABSTRACT Techniques of fabricating altpsms of single trench structure by quartz etching have been developed. A two-step etching process was adopted to make the side-etching structure. Shifter areas were etched by dry-method in first step and by wet-method in second step. In the experiments, by a RIE type dry-etching machine, a hydrofluoric (HF) acid etching system, a phase measurement system (MPM-248) and 6025-mask plate were used. We optimized dryetching parameter, such as pressure, RF power and composition of gas mixture ratio. The smallest phase difference between the device-pattern and the monitor-pattern was obtained in low pressure with inactive gas. The dry etching uniformity was improved and range value was reduced from 2.7 degrees to 1.1 degrees by using type B of the plate holder. In wet etching process, two modes of movement and two concentration of buffered HF acid were compared from the viewpoint of etching uniformity. The uniformity was reduced to two third of the benchmark condition. For A1tPSMs production quality, phase mean deviation is within +/-2 degrees and uniformities are within 2.5 degrees. The technique described in this study can be applied to 150 nm node altpsms fabrication and shows an extensibility to 130 nm node. Key word: phase-shift mask, altpsms, side-etching structure, dry etching, hydrofluoric acid etching, phase mean deviation, uniformities 1. INTRODUCTION Mask error enhancement factor (MEF) value becomes larger for the geometry smaller than exposure wavelength. Alternating phase shift masks (altpsms) are effective for reducing MEF value. A phase shift mask proposed by Levenson et al. consists of a glass plate, chromium opaque patterns, and phase shifters'. The typical structures of altpsms are SOG (spin on glass) on Cr type2, SOG under Cr type3 and quartz (Qz) trench type4 (single trench and dual trench). Qz-trench type can be produced in an easier process than SOG type and keep off a phase uniformity error caused by difference of SOG thickness. We have been developing a technique of fabricating altpsms of single trench structure by quartz etching. A1tPSMs have been applied to 180 nm memory device productions with KrF lithography. According to the miniaturization of the pattern size of ULSI devices, phase mean deviation In Photomask and Next-Generation Lithography Mask Technology VII, Hiroaki Morimoto, 394 Editor, Proceedings of SPIE Vol (2000) X1001$15.OO

2 and phase uniformity are more and more important. ITRS roadmap in 1999 shows altpsms phase mean deviation is degree and its uniformity is +/-2 degree for l3onm techno1ogy5. The most important development subjects of altpsms fabrication are on accurate phase control and shifter-defect free process. This paper will discuss a new phase control technique, which includes improvement of etching uniformity and precise etch-depth control. 2. EXPERIMENTAL 2.1 Fabrication process A two-step etching process was adopted to make the side-etching structure as shown in Fig.1. Shifter areas were etched to about 60 degrees by a RIE type dry-etching machine in the first step. After resist removing, phase-angles of monitor pattern were measured by using the direct phase measurement system, MPM-248 (Lasertec Corporation). Followed by the second step, additional etching was done by a hydrofluoric (HF) acid etching system to form the side-etching structure. The etching time was so determined that the phase is adjusted to be 180 degrees, based on the phase value obtained in the first step. 2.2 Improvement of etching uniformity and precise etch-depth control Dry etching process To reduce the phase angle difference that was caused by micro-loading effect between the device-pattern and the monitor-pattern, we optimized dry-etching parameters, such as pressure, RF power, and composition of the gas mixture ratio. We investigated phase difference between monitor and device pattern, etching uniformity, etching-rate and selectivity for three conditions. The uniformity was confirmed with 25 points (120mm X 120mm area) on 6025 mask plate. Species ofgas were tetra fluorocarbon (CF4) for an active gas and an inactive gas. The effect of the plate holder shape on etching uniformity was evaluated HF etching process Two modes of movement in the dipping method (simple and complex movement) and two concentrations of buffered hydrofluoric (HF) acid were compared from the viewpoint of etching uniformity. The uniformity was confirmed with 49 points phase measurement (120mm X 120mm area) on 6025 mask plate. 3. RESULTS AND DISCUSSION 3.1 Characteristics for dry etching The experimental results of the dry etching are summarized in Table I. The results for high pressure, low pressure, and low pressure with an inactive gas are shown in No.1 - No.3, respectively. Quartz/resist selectivity is 0.4, 1.2, and 1.4 respectively, for No.2 and No.3 are sufficient to etch 60 degree. The smallest phase difference between the device-pattern and the monitor-pattern was obtained in condition No.3. The uniformities in the three conditions are almost the same, but they are not enough for our target specification. Fig.2 shows SEM micrographs in the three conditions. The trench patterns look almost the same. We chose 395

3 condition No.3 because of the smallest phase difference between two different pattern sizes, then the smallest micro-loading effect. There is nothing wrong with SEM micrograph of altpsms structure using condition No.3 dry etching (Fig.3). Table II shows the results on the dry etching uniformity for the two types of plate holders. Type A consists of Aluminum (Al) and type B is set up quartz plate on type A. Dry etching condition is No.3. The uniformity reduces from 2.7 degrees to 1.1 degrees by using type B holder. The holder plate setting up quartz has good performance for etching uniformity. 3.2 Characteristics of wet etching The results of the wet-etching uniformity are shown in Fig.4 and Table III. The etching uniformity of the benchmark condition, simple movement and HF acid concentration 1 (relative value), is 2.5 degrees. The other hand, we investigated the uniformity of changed movement method or lower HF acid concentration. That of each condition is 1.8, 1.2 degrees respectively, so the latter reduces to half of that of the benchmark condition. Since a complex movement is not suitable for automated production process, a simple movement with HF acid at a low concentration was adopted. Its disadvantage is only long etching time. 3.3 Trial production result Fig.5 shows the results of altpsms production quality about phase mean deviations and uniformities. The number ofthe sample plates is 12, and their mean deviations are within +1-2 degrees and uniformities are within 2.5 degrees. 4. SUMMARY The smallest micro-loading effect was obtained in low pressure with inactive gas. The dry etching uniformity reduced from 2.7 degrees to 1.1 degrees by using type B of the plate holder to set up quartz plate. In a wet etching, the uniformity was reduced from 2.5 degrees to 1.2 degrees by using lower concentration HF acid. For altpsms production quality, phase mean deviation is within +1-2 degrees and uniformities are within 2.5 degrees. We can supply defect free altpsms for 150 nm node. The results in this study indicate an extensibility of this technique to 130 nm. 5. ACKNOWLEDGMENT The authors would like to appreciate for technical advice to Mr.M.Kobayashi (of Hitachi) and also to appreciate to Mr.Y.Gotoh, Mr.N.Hasegawa and Mr.K.Suko (of Hitachi) for technical support. 6. REFERENCES 1) Levenson M.D., et al. "Improving Resolution in Photolithography with a Phase- Shifting Mask", IEEE, ED-29, 12, (1982) 396

4 2) H.Hoga, Y.Koizumi, F.Mizuno & H.Nakaune : Practical Method of Phase-Shifting Mask Fabrication", SPIE, vol.2254, 14 (1994) 3) Yoichi Takahashi, Hiroshi Fujita, Hisashi Moro-oka, Masa-aki Kurihira, Kazuo Suura, and Haitake Sano : "Primary Processes in E-Beam and Laser Lithographies for Phase-Shifting Mask Manufacturing", SPIE, vol. 1674, (1992) 4) J.Ushioda, Y.Seki, H.Tanabe & Y.Ogura : "Sub 0.1 jm ArF Excimer Laser Lithography with Alternating Phase-Shifting Masks", SPIE, vol.2512 (1995) 5) THE INTERNATIONAL TECHNOLOGY ROADMAP FOR SEMICONDUCTORS: LITHOGRAPHY, , (1999) 1stEtching (Dry Etch) F >- : Fig. 1 Two - step Etching process 2nd Etching (Wet Etch) Table I Experimental results of dry etching Condition No.1 No.2 [ No.3 Pressure High Low Low RFpower(W) A 2/3A A Gas CF4 CF4 CF4 w/inactive gas Etching-rate (nm/mm.) Selectivity Phase difference between Monitor and device pattern (degree) Etching uniformity (Range: degree)

5 (1).. (2)- jy 'Y I (3)( r" 1 :i (1) Condition No.1 (2) Condition No.2 (3) Condition No.3 Fig.2 SEM micrographs in three dry etching conditions (after dry etching) Fig.3 SEM micrograph of altpsms structure using condition No.3 Table II Dry etching uniformity for two substrate holders Dry etching condition No.3 No.3 [Substrate holder type Type A Type B FEtching uniformity (degree) 2.7 I.1 398

6 (1) ' ( (7 Ā B C D 8 F G (I) Benchmark condition (2) Complex movement (3) HF acid concentration 0.5 (Relative value) (2) (3) ] ABC0 E F G 6 2 fl i O U U106 10] A B CD E F 6 5 a O U ( Fig.4 Phase uniformity of each wet etching method Table ifi Experimental results of wet etching uniformity Movement during dipping Simple Complex Simple HF-acid Concentration 1 (Relative value) 0 S I Etching uniformity (degree)

7 ba 4 C 1.5 a) Sample No Sample No Fig.5 Trial production results of phase mean deviation and uniformity 400