J. Micro/Nanolith. MEMS MOEMS 6!2", 023005 !

Apr–Jun 2007"

Plasma cleaning of lithium off of collector optics

material for use in extreme ultraviolet
lithography applications
Martin J. Neumann Abstract. One of the critical issues within extreme ultraviolet lithography
Reece A. DeFrees is mirror lifetime and the degradation due to debris from the pinch. This
Huatan Qiu research investigated and showed the efficacy of using a helium second-
David N. Ruzic ary plasma and heat for removal of Li debris from collecting on the sur-
University of Illinois face of collector optics. A He helicon plasma, which minimizes self-
Department of Nuclear, Plasma, and biasing and sputtering, has good extreme ultraviolet !EUV" photon
Radiological Engineering wavelength transmission and preferential sputtering of lithium compared
103 South Goodwin to other collector optics material. Through the combined use of heating
Urbana, Illinois 61801 and a He secondary plasma, EUV collector sample surface roughness
E-mail: mneumann@uiuc.edu and surface composition was able to be maintained near as-received
status. The use of the He secondary plasma while the collector optics
sample is exposed to Li debris shows promise as an in situ cleaning
Oleh Khodykin process for collector optics and can extend the lifetime of collector
Alex Ershov optics. © 2007 Society of Photo-Optical Instrumentation Engineers.
Cymer Corporation #DOI: 10.1117/1.2750651$
17075 Thornmint Court
San Diego, California 92127 Subject terms: extreme ultraviolet; mirrors; optics; lithography; plasmas;
reflectivity.
Paper 06096R received Dec. 13, 2006; revised manuscript received Mar. 6, 2007;
Robert L. Bristol accepted for publication Mar. 14, 2007; published online Jun. 25, 2007. This paper
is a revision of a paper presented at the SPIE conference on Emerging Litho-
Intel Corporation graphic Technologies X, Feb. 2006, San Jose, Calif. The paper presented there
Components Research RA3-252 appears !unrefereed" in SPIE Proceedings Vol. 6151.
5200 NE Elam Young Parkway
Hillsboro, Oregon 97123

1 Introduction and Li are condensable metal vapors and can coat the mir-
ror collection optics, which will degrade the mirror’s reflec-
According to the International Technology Roadmap for
tivity. Output power, imaging capability, wafer throughput,
Semiconductors !ITRS", the 32-nm node will require next-
generation lithography for semiconductor productions with and overall cost of ownership are important factors in de-
a light wavelength of 13.5 nm, which is termed extreme termining which of the sources tool suppliers and chip
ultraviolet lithography !EUVL".1–3 Currently, there are sev- manufacturers will select, but one commonality with all
eral processes under development to produce EUV photons EUVL sources is that the collector optics must be robust
at this wavelength including laser produced plasma pinches and have a long lifetime.
!LPP", gas discharged plasma pinches !GDPP", and Although GDPP, LPP, and vacuum arc convert about 1
vacuum arcs.4 Some drawbacks of the various schemes in- to 4% of net deposited plasma energy into EUV photons,
clude a low collection efficiency, the degradation of the the remaining energy generates out-of-band radiation and
electrode system, and the limited lifetime of the collector produces highly energetic ions and neutrals in the dense hot
optics.4,5 plasma that move outward in all directions. A fraction of
The initial choice of a target material for the EUV pinch these ions strike electrode surfaces, injection nozzles, and
was xenon, as it yields a modest series of line transitions in the vacuum chamber producing low-energy sputtered at-
the 13.5± 0.2-nm range with a 1% conversion efficiency.4 oms, and another fraction will reflect and create a source of
Now that EUV source power requirements are approaching medium-energy gas atoms. Highly energetic ions will also
200 W, assuming a 10 mJ/ cm2 resist, xenon is becoming travel the same line of sight path as the desired photons into
increasingly untenable as this exposes the plasma-facing the collector optics.6 These sources of ion and low-energy
collector mirror optics to conditions that are extremely sputtered atoms are collectively labeled “debris,” and miti-
harsh and damaging. gation techniques are the subject of ongoing efforts to
As such, alternative target materials, such as Li and Sn, block this debris from reaching the collector optics. Various
have been identified to have a higher electrical conversion mitigation techniques include foiltraps and E and B repul-
efficiency and yield a broader band spectra at 13.5± 0.2 nm sions fields.6,7 Foiltraps are the industry standard technique
for Sn and a narrow line spectrum at 13.5 nm for Li and and are effective, but they still allow debris to reach the
both of these target materials have a higher EUV conver- collector optics. For high-volume manufacturing operation
sion efficiency, on the order of 1 to 4%.4 However, both Sn !100 wafers per hour", the collector optics must not lose
more than 10% of absolute EUV reflectivity over 1011
1537-1646/2007/$25.00 © 2007 SPIE pulses or 30 000 h.8

J. Micro/Nanolith. MEMS MOEMS 023005-1 Apr–Jun 2007/Vol. 6!2"

Downloaded From: http://nanolithography.spiedigitallibrary.org/ on 12/25/2012 Terms of Use: http://spiedl.org/terms

 Neumann et al.: Plasma cleaning of lithium off of collector optics material…

However, even as effective as these mitigation tech-

niques are, there are limitations, and some debris will still
reach the collector optics. The use of condensable fuels,
such as Sn and Li, are especially difficult because of the
vapor deposition and coating on materials. Lithium is espe-
cially difficult to deal with because of its chemical reactiv-
ity and compound formation. However, Li is advantageous
to use over Sn because it is lightweight and has a low
vaporization point, allowing it to vaporize off of the optical
surface during operation when the optics are kept at an
elevated temperature. Also, Li forms lower energy debris so
it causes less damage to the collection optics compared to
the heavier and higher energy debris formed with the use of
Sn. Thus, the use of Li as a potential fuel in an EUVL
source has attracted considerable interest.
Prevention and removal of Li debris from the collector Fig. 1 Transmission coefficients for 6-eV photons in the EUV spec-
optics is an important area of study so as to develop long trum through various gases at 78 mTorr through 100 cm !see Ref.
lasting collector optics and operating regimes in addition to 14".
expanding the knowledge base about Li transport and inter-
action. This investigation was designed to study a second-
ary plasma interaction with EUV collector optics for use in bombarding flux of sub-kilo electron volt Li ions for a com-
subsequent removal of the Li debris from the optics sur- mercially viable high-volume manufacturing component.
face. The fourth degradation mechanism is due to elevated
temperatures of more than 200°C. Such products may also
2 EUV Collector Optics Degradation Mechanism diffuse to the substrate, in principle, increasing substrate
Normal-incident EUV mirror optics, typically used for LPP roughness if the substrate acts as a diffusion barrier. En-
sources, consist of a smooth Si substrate that is covered hanced thermal interdiffusion of the high and low index
with alternating layers of high and low absorbance thin materials, such as Si and Mo, within the mirror structure,
films, such as Mo and Si, each with a layer thickness of can occur.12 This necessitates the development of mirror
#3.5 nm.4 These multilayer mirror layers work as Bragg optics that maintain their stability at elevated temperatures.
reflectors adding weak reflections from many surfaces in There is also a need for effective diffusion barriers with
phase. For high-volume manufacturing operation !on the low EUV absorption that can resist the diffusion of debris
order of 100 wafers per hour", the collector optics must not within the mirror matrix.
lose more than 10% of absolute EUV reflectivity over Various leading mitigation techniques to limit the
about 1 year of operation, which is on the order of 1011 amount of debris that reach the collector include the use of
pulses for an average repetition rate of 5 kHz.8 Therefore, mass-limited droplet targets, tape targets, ambient gas
erosion of the bilayers and surface scattering of the incident buffer, electrostatic repulsion fields, and permanent and
photons must be kept to a minimum. pulsed magnetic fields.13 Foiltraps, in which a set of thin
EUV mirror optics are subjected to degradation through plates are placed parallel to the light rays to catch the
four main pathways.9 The first is buildup of low-energy Brownian motion of debris in an ambient buffer gas, have
neutral debris on the surface of the mirror optics. If this proven particularly effective. However, none of these tech-
buildup is substantial such that the thickness of the buildup niques are yet able to reach a mirror lifetime of 1011 shots,
is on the order of a few wavelengths of 13.5 nm, this will necessitating the development of in situ mirror cleaning
disallow EUV reflection due to scattering and can act to techniques.
scatter the incident EUV photons away from the desired Investigation into such techniques has been somewhat
focal point. This effect only comes into play if there is a limited. One idea is to use oxygen, as high energy oxygen
large substantial amount of deposition on the surface of the ions can mechanically break the molecular bonds of the
mirror and is not significant if the debris buildup is on a surface molecules and remove some species from the sur-
small order of magnitude in comparison to 13.5 nm. This face. Atomic oxygen present in the plasma also readily re-
buildup of debris on the surface along with implantation of acts with the debris on the surface of the mirror optics and
debris leads to a second pathway of mirror optics destruc- can form more volatile compounds that can be more easily
tion coming from diffusion of debris material into the op- evaporated or removed. However, oxygen species are effec-
tics material that acts to change the mirror composition, tive predominately on hydrocarbons and not on condens-
blurring the defined layer boundaries, and changing the in- able metal vapor debris. Oxygen species are also very re-
dices of refraction, which leads to a decrease in EUV re- active with the mirror optics itself. In the case in which the
flection to the intermediate focal point. debris load is so severe that cleaning must happen while the
A third pathway of mirror degradation comes from the source is on, oxygen has the disadvantage of being highly
sputtering of the optics by higher energy incident ions from absorptive of EUV photons in comparison to other gases, as
the plasma pinch. Debris generation can be limited to a shown in Fig. 1, which shows transmission coefficients of
certain extent, but this production cannot be eliminated in various gases from 13 to 14 nm, gas pressure of 78 mTorr,
its entirety.5,10,11 Optical mirror surfaces for Li-based EUV and gas energy of 6 eV.14 A gas energy of 6 eV and
systems must be engineered to withstand a continuous 78 mTorr were chosen to illustrate this point because this is

J. Micro/Nanolith. MEMS MOEMS 023005-2 Apr–Jun 2007/Vol. 6!2"

Downloaded From: http://nanolithography.spiedigitallibrary.org/ on 12/25/2012 Terms of Use: http://spiedl.org/terms

 Neumann et al.: Plasma cleaning of lithium off of collector optics material…

Fig. 2 Theoretical sputtering coefficients for He+ ions normal inci-

dence on Li, Ru, and Mo !see Refs. 15 and 16".

the range in which experimental investigation was under-

taken and commercial operation around the collector optics
could be undertaken, but the relative comparison of trans-
mission coefficients for the various gases holds true at a Fig. 3 Schematic of SCOPE facility.
wide range of gas pressures and energies for this photon
wavelength range.14
Therefore, cleaning through oxidation is unlikely to be erating regimes that will aid in the advancements of EUV
effective. This work concentrates on the use of He as a mirror lifetime optics. SCOPE is composed of a unique
plasma species for secondary plasma source for cleaning. sample holder, magnetron, and a secondary plasma source.
Figure 1 shows He is a good choice from the standpoint of A 3-in. magnetron with a Li target and a He plasma is
EUV photon transmission through He gas. In addition, He used to sputter Li off of the target and deposit a Li film on
has a better theoretical preferential sputtering yield for Li mirror optics samples that are #6 cm away. For the plasma
versus the mirror capping layer of Ru and Mo at lower cleaning portion of SCOPE, a helicon plasma source is em-
energies, as shown in Fig. 2.15,16 This indicates that He ions ployed at 13.56 MHz from 0 to 3 kW. The antenna em-
can be directed to preferentially sputter Li debris from the ployed in SCOPE is an m = 0 helicon plasma stabilized with
surface of the mirror while minimizing damage to the mir- external magnetic fields. Helicon plasmas have been shown
ror itself. to work well as secondary plasma sources while avoiding
the production of debris through self-sputtering.17 This
3 Experiment Facility setup is illustrated in Fig. 3.
A new facility, the Surface Cleaning of EUV Optics by
Plasma Exposure !SCOPE" at the University of Illinois, has 4 Experimental Investigation
been constructed for the purpose of studying plasma clean- Experiments with the use of a He secondary plasma were
ing of optics materials for use in EUV applications. The performed to study the mitigation and removal of Li debris
SCOPE facility is a multifunctional device that is capable from EUV mirror optics, as laid out in Table 1.
of creating Li debris conditions incident upon EUV optics Atomic force microscopy !AFM" and scanning electron
materials so as to develop a model of Li transport and op- microscopy !SEM" surveys of the experimental samples are

Table 1 Experimental matrix and resulting RMS roughness.

He Background Bias Voltage Magnetron Helicon Sample RMS

Pressure on Sample Plasma Plasma Temperature Roughness

Sample 0 Control sample 2.94 nm

Sample 1 78 mT 0 Yes No 50°C 56.9 nm

Sample 2 78 mT −100 Yes No 50°C 30.5 nm

Sample 3 78 mT −100 Yes Yes 50°C 27.4 nm

Sample 4 78 mT 0 Yes No 400°C 26.4 nm

Sample 5 78 mT −100 Yes Yes 400°C 1.03 nm

J. Micro/Nanolith. MEMS MOEMS 023005-3 Apr–Jun 2007/Vol. 6!2"

Downloaded From: http://nanolithography.spiedigitallibrary.org/ on 12/25/2012 Terms of Use: http://spiedl.org/terms

 Neumann et al.: Plasma cleaning of lithium off of collector optics material…

Fig. 4 !a" Sample 0, !b" sample 1, !c" sample 2, !d" sample 3, !e" sample 4, and !f" sample 5.

shown in Fig. 4. When looking at the as-received mirror 1 ± 0.5 nm/ min as shown in the resulting profilometry re-
samples in Fig. 4!a", it should be noted that the SCOPE sult in Fig. 6. Figure 4!c" shows the AFM and SEM results
facility is not in a clean room; as can be seen in Fig. 4!a", for sample 2. Again, notice the roughness and the clumping
there was a certain amount of contamination or existing nature of the debris left on the surface of the mirror optics.
debris on unexposed mirror optics samples. This was con- This mirror would also not be reflective.
sidered the control for the remainder of the experiments. Sample 3 repeats the conditions of sample 2, but with
For the first experiment, sample 1 was kept at 50°C and the added He+ flux from the secondary helicon plasma
0 V with the magnetron on. At this temperature, the vapor source. Again, the magnetron was operated under the same
pressure and hence evaporation of Li from the surface is conditions. From Langmuir probe analysis, the new ne,
negligible.18 Because of the location of the sample with with both plasma sources in operation, was 3 ! 1012 cm−3, a
respect to the magnetron, the He plasma used to sputter Li factor of more than 100 times the density from the magne-
from the magnetron overlaps at the sample. Hence, with tron plasma alone, and Te is 6 eV. The debris film is less
just the magnetron plasma, this can be thought of as a sec- than the resolution of the profilometer, but an AFM survey,
ondary plasma source on the mirror optics that would not in Fig. 4!d", clearly shows that there is still debris present
normally be present in an EUVL source. From Langmuir on the surface of the mirror optic.
probe measurements at the sample surface with just the
magnetron plasma on, the electron density, ne, was deter-
mined to be 1 ! 1010 cm−3 and Te of 4.7 eV. However,
sputtering of the Li debris from the mirror sample can also
be deemed as negligible because at 0-V bias, this is in the
electron saturation region and no ion current is being drawn
to the sample. From a profilometry measurement of the
debris thickness on the sample, as shown in Fig. 5, the
deposition rate on the mirror optics was 3.33± .5 nm/ min.
Figure 4!b" shows the AFM survey and SEM survey of
sample 1 and show the Li film on the surface of the mirror
optics to be very rough and to vary widely across the
sample. In this condition, the EUV reflectivity would be
minimal.
Sample 2 was run under the same conditions but this
time a −100-V bias was applied so as to draw He+ ions in
from the magnetron He plasma and sputter Li from the Fig. 5 Profile of sample 1, run for 60 min at 78 mTorr with the mag-
surface of the mirror optics. This lowered the rate to netron He plasma and the mirror optics at 50°C and 0 V.

J. Micro/Nanolith. MEMS MOEMS 023005-4 Apr–Jun 2007/Vol. 6!2"

Downloaded From: http://nanolithography.spiedigitallibrary.org/ on 12/25/2012 Terms of Use: http://spiedl.org/terms

 Neumann et al.: Plasma cleaning of lithium off of collector optics material…

−100 V, the sample temperature was elevated to 400°C on

the surface, and there was a secondary He plasma from the
helicon source. The resultant Li debris film thickness was
less than the resolution of the profilometry and indicates
that there was not a Li film, but rather localized regions of
Li deposition. This is further confirmed by the AFM and
SEM surveys of this sample, as shown in Fig. 4!f". From
these results, it is clearly shown that the sample resembles
the as-received sample in Fig. 4!a". This clearly shows the
Li debris is both prevented from and effectively removed
from the mirror sample. Hence, reflectivity should be main-
tained. The buildup and incidence of Li debris on the
sample surface was minimal such that diffusion into the
mirror matrix was minimized. The important result shown
Fig. 6 Profile of sample 2, run for 60 min at 78 mTorr, with the here is that through the combined use of heating and a
magnetron He plasma and the sample at 50°C and −100 V. secondary He plasma source, mirror optics can be main-
tained in an as-received state. This process, combined with
other industry standard mitigation techniques, can provide a
This illustrates that although there is still debris on the way to further increase the protection and extend the life-
surface of the mirror optics, for everything else being held time of the collector optics with minimal loss of EUV pho-
constant, the increase in He+ ion flux to the mirror surface ton collection at the intermediate focus. The use of a sec-
acts to enhance removal of debris. The important illustra- ondary plasma and heating could be a viable in situ
tion here, though, is that with the addition of the added He+ collector mirror optics cleaning and debris prevention tech-
ion flux, there is significant enhanced sputtering of debris nique.
from the surface of the mirror optics. However, the coating
is still too thick to allow effective EUV reflectivity.
The next mirror sample, sample 4, was kept at the same
operating parameters of the magnetrons before, but the 5 Conclusion
sample temperature was elevated to 400°C on the surface,
and there was no secondary He plasma from the helicon The addition of heating and a secondary plasma act to-
source. The elevated temperature was chosen to increase gether to keep the EUV optics sample near as-received state
the Li evaporation rate; in addition, the multilayer coating with regards to surface roughness and debris buildup on the
of these samples were specifically engineered to withstand surface. This combination of heating and sputtering from
this elevated temperature.19 The profilometry measurement the secondary plasma cannot be thought to be indepen-
is shown in Fig. 7. From this profilometry measurement, dently exclusive of each other, though. Evaporative loss of
the net deposition rate is 0.33± 0.2 nm/ min. This can be Li has been shown to be enhanced through ion
difficult to discern because of the dust and contamination sputtering.20–22 However, when the collisions near the sur-
on the surface, but there is a sharp line from where the face are not of a binary nature or when the target atoms are
coverslip was at #3400 "m in Fig. 7. Figure 4!e" is an no longer stationary, the resultant collision involves a vol-
AFM and SEM survey of sample 4 showing that there is ume with a high number of moving atoms near the surface.
debris on the optics surface, but much reduced. The RMS The result is a temperature dependence of sputtering rate.22
roughness would still preclude effective focused EUV re- In this same vein, ion irradiation of the mirror optics
flectivity, though. sample can increase the evaporative loss that would not
Lastly, sample 5 was processed with the magnetron He normally be expected. To be liberated from the surface of
plasma at the same conditions, sample bias voltage of the mirror optics sample, the Li atoms must overcome the
binding energy holding it in place. This can be accom-
plished through the transfer of energy thermally or physi-
cally. That is, Li at an elevated temperature can be sput-
tered at a higher rate because the threshold for escape
energy from sputtering alone has been lowered due to the
additional energy supplied by thermal transfer. Li evapora-
tion can occur at a higher rate because of the contribution
of the kinetic energy from the He+ ion, which contributes to
overcoming the binding energy of the Li atom.21,23
These results show that the combination of heating EUV
optics and exposure to a relatively low-density secondary
He plasma that can minimize or even reverse the buildup of
surface roughness and debris while maintaining minimal
EUV photon absorption during exposure to EUV Li debris-
like conditions can maintain the surface roughness and
minimize debris buildup on the optics surface as that of an
Fig. 7 Profile of sample 4, run for 60 min at 78 mTorr with the mag- as-received state and provides a potentially viable in situ
netron He plasma and the mirror optics at 400°C and −100 V. mechanism for extending the lifetime of EUV optics.

J. Micro/Nanolith. MEMS MOEMS 023005-5 Apr–Jun 2007/Vol. 6!2"

Downloaded From: http://nanolithography.spiedigitallibrary.org/ on 12/25/2012 Terms of Use: http://spiedl.org/terms

 Neumann et al.: Plasma cleaning of lithium off of collector optics material…

Acknowledgments 11. U. Stamm, G. Schriever, and J. Kleinschmidt, “High-power GDPP

Z-pinch EUV source technology,” Chapter 14 in EUV Sources for
This work was supported by Cymer Corporation, Oleh Lithography, V. Bakshi, Ed., pp. 413–452, SPIE Press, Bellingham,
Khodykin, program manager. Surface characterizations of Wash. !2006".
12. C. L. Rettig, O. V. Khodykin, J. R. Hoffman, et al., “Protection of
sample materials were carried out in the Center for Mi- collector optics in an LPP based EUV source,” Emerging Litho-
croanalysis of Materials, University of Illinois, which is graphic Technologies IX, Proc. SPIE 5751, 910–918 !2005".
13. M. Richardson, C.-S. Koay, K. Takenoshita, C. Keyser, and M. Al-
partially supported by the United States Department of En- Rabban, “High conversion efficiency mass-limited Sn-based laser
ergy under Grant No. DEFG02-91-ER45439. plasma source for extreme ultraviolet lithography,” J. Vac. Sci. Tech-
nol. B 22!2", 785–790 !2004".
14. B. L. Henke, E. M. Gullikson, and J. C. Davis, “X-ray interactions:
Photoabsorption, scattering, transmission, and reflection at E
References = 50– 30,000 eV, Z = 1 – 92,” At. Data Nucl. Data Tables 54!2", 181–
342 !1993".
1. “International Technology Roadmap for Semiconductors 2005 edi- 15. J. P. Biersack and L. Haggmark, “A Monte Carlo program for the
tion: Lithography,” Report. International Technology Roadmap for transport of energetic ions in amorphous targets,” Nucl. Instrum.
Semiconductors !2005". Methods 174, 257 !1980".
2. “International Technology Roadmap for Semiconductors: 2005 edi- 16. J. P. Biersack, L. Haggmark, and U. Littmark, The Stopping and
tion executive summary,” Report. International Technology Roadmap Range of Ions in Solids, Pergamon Press, New York !1985"; SRIM
for Semiconductors !2005". software !2003".
3. V. Banine and R. Moors, “Plasma sources for EUV lithography ex- 17. D. B. Hayden, D. R. Juliano, M. N. Neumann, M. C. Allain, and D.
posure tools,” J. Phys. D 37!23", 3207–3212 !2004". N. Ruzic, “Helicon plasma source for ionized physical vapor deposi-
4. J. Jonkers, “High power extreme ultra-violet !EUV" light sources for tion,” Surf. Coat. Technol. 120–121, 401–404 !1999".
future lithography,” Plasma Sources Sci. Technol. 15!2", S8–S16 18. Veeco, Instruments Inc., “Vapor Pressure Curves for Elements,”
!2006". Woodbury, New York !2006"; http://www.veeco.com/library/
5. U. Stamm and K. Gabel, “Technology for LPP sources,” Chapter 19 resources_view_sub.php?id#30.
in EUV Sources for Lithography, V. Bakshi, Ed., pp. 413–452, SPIE 19. O. V. Khodykin, personal communication on sample procurement,
Press, Bellingham, Wash. !2006". Cymer Corp. !2005".
6. E. Vargas-Lopez, B. E. Jurczyk, M. A. Jaworski, M. J. Neumann, and 20. J. P. Allain, A. Hassanein, T. Burtseva, A. Yacout, Z. Insepov, S. Taj,
D. N. Ruzic, “Origins of debris and mitigation through a secondary and B. J. Rice, “Radiation-induced synergistic effects of athermal and
RF plasma system for discharge-produced EUV sources,” Microelec- thermal mechanisms on erosion and surface evolution of advanced
tron. Eng. 77!2", 95–102 !2005". electrode and condenser optics materials,” Emerging Lithographic
7. B. E. Jurczyk, E. Vargas-Lopez, M. Neumann, and D. N. Ruzic, Technologies VIII, Proc. SPIE 5374, 112–121 !2004".
“Illinois debris-mitigation EUV applications laboratory,” Microelec- 21. R. P. Doerner, M. J. Baldwin, S. I. Krasheninnikov, and D. G. White,
tron. Eng. 77!2", 103–109 !2005". “Behavior of high temperature liquid surfaces in contact with
8. J. P. Allain, A. Hassanien, M. Nieto, et al., “Erosion and degradation plasma,” J. Nucl. Mater. 313–316, 383–387 !2003".
of EUV lithography collector mirrors under particle bombardment,” 22. J. P. Allain, D. N. Ruzic, D. A. Alman, and M. D. Coventru, “A
Emerging Lithographic Technologies IX, Proc. SPIE 5751, 1110– model for ion-bombardment induced erosion enhancement with target
1117 !2005". temperature in liquid lithium,” Nucl. Instrum. Methods Phys. Res. B
9. D. N. Ruzic, “Origin of debris in EUV sources and its mitigation,” 239, 347–355 !2005".
Chapter 36 in EUV Sources for Lithography, V. Bakshi, Ed., pp. 23. R. P. Doerner and S. I. Krasheninnikov, “Particle-induced erosion of
957–994, SPIE Press, Bellingham, Wash. !2006". materials at elevated temperature,” J. Appl. Phys. 95!8", 4471–4475
10. I. V. Fomenkov et al., “Dense plasma focus source,” Chapter 12 in !2004".
EUV Sources for Lithography, V. Bakshi, Ed., pp. 373–393, SPIE
Press, Bellingham, Wash. !2006". Biographies and photographs of the authors not available.

J. Micro/Nanolith. MEMS MOEMS 023005-6 Apr–Jun 2007/Vol. 6!2"

Downloaded From: http://nanolithography.spiedigitallibrary.org/ on 12/25/2012 Terms of Use: http://spiedl.org/terms