Article · 2017-12-13 · An alternative approach is the block coplolymer (BCP) lithography which uses direct self-assembling of poly-mer patterns. Recently, it has attracted signiﬁcant

Delivered by Publishing Technology to: Sung Kyun Kwan UniversityIP: 115.145.196.100 On: Tue, 05 Jan 2016 05:00:30

Copyright: American Scientific Publishers

Copyright © 2015 American Scientific PublishersAll rights reservedPrinted in the United States of America

ArticleJournal of

Nanoscience and NanotechnologyVol. 15, 8093–8098, 2015

www.aspbs.com/jnn

Surface Modification of Block Copolymer ThroughSulfur Containing Plasma Treatment

Sang Wook Choi1� †, Jae Hee Shin1� †, Min Hwan Jeon2, Jeong Ho Mun3, Sang Ouk Kim3,Geun Young Yeom1�2�∗, and Kyong Nam Kim1�∗

1School of Advanced Materials Science and Engineering, Sungkyunkwan University (SKKU), Suwon,Gyeonggi-do 440-746, Republic of Korea

2SKKU Advanced Institute of Nanotechnology (SAINT), Sungkyunkwan University (SKKU), Suwon,Gyeonggi-do 440-746, Republic of Korea

3Department of Materials Science and Engineering, KAIST, Daejeon, 305-701, Republic of Korea

Some of the important issues of block copolymer (BCP) as an application to the potential low costnext generation lithography are thermal stability and deformation during pattern transfer process inaddition to defect density, line edge/width roughness, etc. In this study, sulfur containing plasmatreatment was used to modify the BCP and the effects of the plasma on the properties of plasmatreated BCP were investigated. The polystyrene hole pattern obtained from polystyrene polystyrene-block-poly(methyl methacrylate) (PS-b-PMMA) was initially degraded when the polystyrene hole wasannealed at 190 �C for 15 min. However, when the hole pattern was treated using sulfur containingplasmas using H2S or SF6 up to 2 min, possibly due to the sulfurization of the polystyrene holesurface, no change in the hole pattern was observed after the annealing even though there is a slightchange in hole shapes during the plasma treatment. The optimized plasma treated polystyrenepattern showed the superior characteristics as the mask layer by showing better thermal stability,higher chemical inertness, and higher etch selectivity during plasma etching.

Keywords: Block Copolymer (BCP), PS-b-PMMA, Plasma Treatment, H2S, SF6, Selectivity.

1. INTRODUCTIONAs the pattern geometry continues to shrink in the micro-electronic industry, traditional photolithography technol-ogy shows the limitation for sub-nano patterning.1–4

An alternative approach is the block coplolymer (BCP)lithography which uses direct self-assembling of poly-mer patterns. Recently, it has attracted significant atten-tion because of the characteristics of uniformly alignednanoscale feature arrays that can be obtained easily at lowcost and with a high throughput.1–8 Nevertheless, there arestill many issues coming from the material characteristicsof BCP that must be resolved to achieve the goal as thenext generation nanoscale lithography.

Generally, a BCP consists of two or more covalentlybonded polymeric monomer or block units, which is chem-ically different and attached to each other. Due to their

∗Authors to whom correspondence should be addressed.†These two authors contributed equally to this work.

differences in the chemistry, the blocks tend to separatelike oil and water; but due to their covalent linkage, thephase separation occurs over the length scales determinedby the length of the BCP molecules which ranges from afew nanometers to tens of nanometer. BCP is phase sep-arated into various features such as cylindrical, lamellar,and gyroid morphologies by the volume fraction of theconstituent blocks.3–6 Because this phase separation pro-cess depends on various parameters such as temperature,humidity etc., the fabricated patterns could be incomplete,and results in non-uniformity between patterns. By far,the most common BCP that researchers have investigatedas the BCP lithography is polystyrene-block-poly(methylmethacrylate) (PS-b-PMMA). The PMMA domains can beselectively removed, and the remaining polystyrene mate-rial serves as the mask for pattern transfer. Especially, thesoft mask material (polystyrene) remained from the phaseseparation tends to show drawbacks for highly selectivepattern transfer due to the relatively lower thermal stabil-ity and low etch selectivity.5�6�8 This results in a serious

J. Nanosci. Nanotechnol. 2015, Vol. 15, No. 10 1533-4880/2015/15/8093/006 doi:10.1166/jnn.2015.11286 8093



Surface Modification of Block Copolymer Through Sulfur Containing Plasma Treatment Choi et al.

problem because, even though the BCP lithography candefine the nanoscale mask features, the device performanceis eventually decided by the quality of pattern transfer.6

To improve the selectivity during the pattern transfer,cryogenic methods using an extremely low temperatureduring the pattern transfer have been investigated.1�5�6�9–11

These methods showed relatively high selective etchingover the BCP mask, however, controlling the cryogenicprocess temperature during the pattern transfer process isnot an easy task. Another approach for highly selectivepattern transfer could be achieved through sulfonation ofthe soft mask which forms SO3H functional group. Theresearches on the sulfonated polystyrene ball have beenreported a long time ago and have shown the charac-teristics of the increased mechanical strength, decreasedswelling, high thermal stability etc.12–14 The current rep-resentative surface sulfurization methods can be dividedinto wet method and gaseous method. In the case of wetsulfonation, Turbak et al.15 have been investigated but thismethod showed the limitation such as difficulty in con-trolling sulfonation amount and repeatability. For the drysulfonation, Kucera et al.12�13 showed the possibility of sul-fonation of polystyrene ball using gaseous SO3. However,this method also showed drawbacks such as long time of afew days required for sulfonation and difficulty in control-ling the degree of sulfonation. In addition, the above meth-ods can not be applicable to the polystyrene for nanoscaleBCP lithography due to the long sulfonation time and pos-sible dimensional change of nanosize polystyrene featuresduring the sulfonation.In this study, a sulfur containing plasma treatment

method has been investigated for sulfurization of nanoscalepolystryrene features instead of sulfonation to investigatethe possibility of transferring the polystyrene patterns bet-ter through thermal processing, plasma processing, etc. byimproving thermal and chemical stability. For the plasmatreatment, sulfur-containing plasmas using H2S or SF6

were used, and the change of cylindrical hole pattern sizewas measured after a following thermal processing. Theoptimized plasma treatment showed better thermal stabil-ity, well-tailored hole pattern distribution, and improvedetch selectivity.

2. EXPERIMENTAL DETAILSAs the BCP pattern sample, a polystyrene hole patterncomposed of 40 nm diameter holes was used. To formthe 40 nm diameter polystyrene hole pattern, 140 kg/molpolystyrene and 65 kg/mol PMMA were dissolved in atoluene and the solution was spin-coated on the siliconwafer. After the spin coating, a thermal annealing was con-ducted at 230 �C for 40 h to accomplish 40 nm diame-ter self-assembled hole morphology of PS-b-PMMA. ThePMMA domains in the BCP film were selectively removedby O2 plasma reactive ion etching (RIE). The resultant

nano-sized polystyrene hole pattern was applied as a litho-graphic mask for pattern transfer. The details on the fabri-cation of the BCP hole pattern can be found elsewhere.16–18

Both the sufurization process and the pattern trans-fer process were conducted with an inductively coupledplasma (ICP) system equipped with a substrate biasingsystem (shown in Fig. 1). The silicon wafer with thepolystyrene hole pattern was inserted in the processingchamber using a loadlock system without breaking the vac-uum while the substrate was cooled at about 10 �C with achiller. For the sulfurization process, the plasma treatmentwas performed at 13.56 MHz 50 W of ICP source power,at 8 mTorr of operating pressure, and with 76 sccm ofH2S or SF6 flow rate. After the sulfurization, the stabilityof the polystytene hole pattern was estimated by thermalannealing at 190 �C for 15 min in a vacuum furnace. Forthe pattern transfer process, the Cl2 plasma was formedby applying 13.56 MHz 50 W of the ICP source powerat 8 mTorr of operating pressure and with 45 sccm of Cl2flow rate. For the pattern transfer,−150 V of DC bias volt-age was applied to the substrate by supplying an additional13.56 MHz rf power to the substrate.Field emission secondary electron microscope (FE-EM,

Hitachi S-4700) was used to estimate the change of thepolystyrene hole pattern before and after the processingand X-ray photon spectroscopy (XPS; ESCA2000, VGMicrotech Inc.,) was used to analyze the surface bind-ing state of the sulfurized polystyrene surface after plasmatreatment. The etched height after pattern transfer wasmeasured with a stylus profilometer (Tencor Alpha-step500).

3. RESULTS AND DISCUSSIONTo evaluate the thermal stability of the 40 nm diameterpolystyrene hole pattern, the fabricated polystyrene holepattern was thermally annealed and the degradation of thehole pattern was observed using SEM. Figure 2 shows theSEM images of (a) the reference pattern (40 nm diameterpolystyrene hole pattern) and the polystyrene hole patterns

Figure 1. Schematic diagram of the ICP system used in the experiment.

8094 J. Nanosci. Nanotechnol. 15, 8093–8098, 2015



Choi et al. Surface Modification of Block Copolymer Through Sulfur Containing Plasma Treatment

Figure 2. SEM images of (a) the reference pattern (40 nm diame-ter polystyrene hole pattern) and the polystyrene hole patterns after theannealing at 190 �C in a vacuum furnace for (b) 5 min and (c) 15 min.

after the annealing at 190 �C in a vacuum furnace for (b)5 min and (c) 15 min. As shown in Figure 2(b), after theannealing for 5 min, some of polystryrene holes were col-lapsed and the rest of the holes showed the decrease inhole diameter. After the annealing for 15 min, as shownin Figure 2(c), all hole patterns were collapsed and irreg-ular curve patterns were generated. The collapse of thepolystytene holes is believed to be related to the reflowof the polystytene surface molecules during the annealingto decrease the surface energy. From these results, it canbe found out that the nano-size polystyrene hole patternapplied as the mask for the next generation BCP lithogra-phy has a very low thermal stability, therefore, it can showvarious problems during the following pattern transfer pro-cessing such as plasma processing.

To prevent the polystyrene pattern deformation and col-lapse during the following pattern transfer processing, thepolystyrene hole pattern was exposed to a sulfur contain-ing plasma for two minutes generated with 50 W of rfpower while flowing 76 sccm of H2S or SF6 at 8 mTorr

and was followed by annealing at 190 �C for 15 min. Theleft pictures of Figures 3(a)–(c) show the SEM imagesof polystyrene hole patterns before the plasma treatment(same as Fig. 2(a)), after the H2S plasma treatment, andafter the SF6 plasma treatment, respectively. As shown inthe left picture of Figure 3(a), the reference polystyrenehole pattern showed the hole size of 35∼39 nm beforethe plasma treatment. The irregular polystyrene hole dis-tribution is believed to be related to the imperfect phaseseparation between polystyrene and PMMA during theself-assembly process. As shown in Figure 3(b), after theH2S plasma treatment, the hole size was slightly deformedand the size was decreased to 21∼31 nm possibly due tothe hydrogen in the H2S which changes the surface mor-phology of polystyrene. However, when the polystyrenehole pattern was treated with the SF6 plasma, the holesize was increased slightly to 39∼40 nm by the etching ofpolystyrene surface with fluorine radicals, and the thick-ness of hole pattern was also decreased from 60 nm to53 nm after the SF6 plasma treatment (not shown).The right pictures of Figures 3(a)–(c) show the SEM

images of polystyrene hole patterns after the annealing ofthe reference pattern, the H2S plasma treated pattern, andthe SF6 plasma treated pattern on the left side at 190 �Cfor 15 min, respectively. As shown in the right picturesof Figure 3(b), after the annealing, even though the refer-ence polystyrene hole patterns were collapsed completelyas shown in the right picture of Figure 3(a) (the same pic-ture as Fig. 2(c)), only a slight change in the hole size wasobserved for the H2S plasma treated pattern. However, inthe case of the SF6 plasma treated pattern, as shown inFigure 3(c), no change in the hole size was observed afterthe annealing.During the annealing, not only the hole size but also the

thickness of the polystyrene hole pattern can be changed.Figure 4 shows the reduction percentage of polystyrenehole size and thickness after the annealing at 190 �C for15 min for the reference, the H2S plasma treated pat-tern, and the SF6 plasma treated pattern. In the case ofthe reference polystyrene hole pattern, not only the holepattern was completely collapsed as shown in Figures 2and 3(a), but also the mask thickness was significantlydecreased by 60% (from 60 to 25 nm). On the contrary,in the case of the H2S plasma treated polystyrene pattern,even though the hole size was decreased by about 25%,and the mask thickness was not changed after the anneal-ing. In the case of the SF6 plasma treated polystyrene pat-tern, both the hole size and the hole thickness were notchanged after the annealing even though the hole size wasslightly increased (from 35∼39 to 39∼40 nm) and the holethickness was slightly decreased (60 to 53 nm) during theplasma treatment. Therefore, the polystyrene hole patternexhibited significantly better thermal property after the sul-fur containing plasma treatments. Especially, compared tothe H2S plasma treatment, after the SF6 plasma treatment,

J. Nanosci. Nanotechnol. 15, 8093–8098, 2015 8095




Figure 3. SEM images of BCP patterns (a) without plasma treatment (same pictures as Figs. 2(b) and (c)), (b) after H2S plasma treatment for 2 min,(c) after SF6 plasma treatment for 2 min. Left pictures are before annealing and right pictures are after annealing at 190 �C for 15 min.

better thermal stability of polystyrene hole pattern couldbe obtained.For the optimization of the sulfurization process for

the polystyrene hole pattern using the SF6 plasma, thehole size and thickness of the polystyrene pattern treatedby the SF6 plasma were measured as a function of theplasma treatment time after annealed at 190 �C for 15 minand the results are shown in Figure 5 with SEM imagesof the patterns after the annealing. Other plasma treat-ment conditions were the same as those in Figure 3(c).As shown in the figure, initially with the increase of SF6

plasma treatment time, the hole size was increased andthe hole thickness was decreased by the etching of the

hole surface in addition to sulfurization. However, whenthe SF6 plasma treatment time was longer than 2 min, thechange in the hole size and hole thickness was saturatedindicating sufficient sulfurization without etching of thepolystyrene surface. In addition, as shown in SEM images,the crack patterns between the polystyrene holes appearedto be decreased by the SF6 plasma treatment until the pat-tern was treated to 2∼2.5 min. However, when the pat-tern was treated for 3 min, the pattern deformation wasobserved possibly due to the damage by the SF6 plasma.The chemical binding states of the polystyrene hole

pattern surface after the SF6 plasma treatment for 2 minwere investigated using XPS to analyze the reason for the




Choi et al. Surface Modification of Block Copolymer Through Sulfur Containing Plasma Treatment

Figure 4. Reduction percentage of polystyrene hole size and thicknessafter the annealing at 190 �C for 15 min for the reference, the H2S plasmatreated pattern, and the SF6 plasma treated pattern.

improved property of the SF6 plasma treated polystyrenesurface. The XPS narrow scan data of S 2p on the SF6

plasma treated polystyrene surface is shown in Figure 6.As shown in Figure 6, before the SF6 plasma treat-ment, no peak related to S was observed. However, afterthe plasma treatment, the broad chemical bonding peaks(164 eV∼169.2 eV) possibly related to S such as SOx,SFx , etc. were observed. Also, with increasing the plasmatreatment time from 30 sec to 2 min, the peak intensi-ties related to S were increased. Therefore, it is believedthat the improvement of the polystyrene property after theSF6 plasma treatment is related to the sulfurization ofthe polystyrene surface by forming sulfur compounds onthe surface during the SF6 plasma treatment. (other exper-iment showed that F does not directly affect the improve-ment of polystyrene properties).

Figure 5. Hole size and hole thickness of the polystyrene patterntreated by the SF6 plasma measured as a function of the plasma treatmenttime after annealed at 190 �C for 15 min.

Figure 6. XPS narrow scan data of S 2p on the SF6 plasma treatedpolystyrene surface measured as a function of SF6 plasma treatment time.

As mentioned above, the device performance obtainedby the BCP lithography depends on the quality of patterntransfer through the following processing such as plasmaprocessing. Therefore, in addition to the thermal annealingtest, the chemical stability of the polystyrene hole patternbefore and after the SF6 plasma treatment was investi-gated by etching the silicon using the 40 mm diameterpolystyrene hole patterns as the etch mask. For the etchingexperiment, a chlorine plasma was generated by apply-ing 50 W of rf power to the ICP source and by applying−150 V of DC bias voltage to the substrate. The operatingpressure was maintained at 8 mTorr by flowing 45 sccmof Cl2 gas. The substrate temperature was kept at roomtemperature. Figure 7 shows the etch selectivities of sili-con over 40 nm diameter polystyrene hole patterns treatedand untreated by the SF6 plasma for 2 min before the

Figure 7. Etch selectivities of silicon over 40 nm diameter polystyrenehole patterns treated and untreated by the SF6 plasma for 2 min beforethe etching.

J. Nanosci. Nanotechnol. 15, 8093–8098, 2015 8097




etching. As shown in the figure, for the untreated (ref-erence) polystyrene hole pattern, the etch selectivity wasabout 2 due to the severe mask etching and collapsingduring the chlorine plasma etching. However, in the caseof the polystyrene pattern treated by the SF6 plasma, thepolystyrene mask was maintained without degradation dur-ing chlorine plasma etching. Also, the etch selectivity wasdramatically increased to 34 by decreasing the polystyrenemask etching significantly. Therefore, by the SF6 plasmatreatment, not only the thermal strength but also chemi-cal inertness was improved, and which are some of therequired properties of mask materials for the next genera-tion BCP lithography.

4. CONCLUSIONThe effect of sulfur-containing plasma treatment on theproperties of 40 nm diameter polystyrene hole patternobtained by PS-b-PMMA block copolymer was investi-gated. For the polystyrene hole pattern, the pattern waseasily degraded and collapsed during the thermal anneal-ing at 190 �C and during the plasma etching using chlorineat room temperature. However, when the polystyrene holepattern was treated by a SF6 plasma, the pattern was notchanged noticeably during the thermal annealing. In addi-tion, the pattern was not degraded and showed significantlylower etch rates during the chlorine reactive ion etching.The improved properties by the SF6 plasma treatment wererelated to sufurization not fluorination of the polystyrenesurface, because a H2S plasma treatment instead of the SF6

plasma treatment showed similar effects even though thepattern was slightly degraded during the plasma treatment.Therefore, through the sulfur containing plasma treatment,not only the thermal strength but also chemical inertnesswhich is required for the next generation BCP lithographycould be achieved.

Acknowledgments: This work was supported by theIndustrial Strategic Technology Development program

(10041926, Development of high-density plasma technolo-gies for thin-film deposition of nanoscale semiconductorsand flexible display processing) funded by the Ministryof Knowledge Economy (MKE, Korea) and was also sup-ported by the MOTIE (Ministry of Trade, Industry andEnergy (10049065)) and KSRC (Korea SemiconductorResearch Consortium) support program for developmentof future semiconductor devices.

References and Notes1. Z. Liu, X. Gu, J. Hwu, S. Sassolini, and D. L. Olynick, Nanotech-

nology 25, 285301 (2014).2. F. Yamashita, E. Nishimura, K. Yatsuda K, H. Mochiki, and

J. Bannister, Proc. SPIE Advanced Lithography, International Soci-ety for Optics and Photonics, 83280T (2012).

3. X. Gu, Z. Liu, I. Gunkel, S. Chourou, S. W. Hong, D. L. Olynick,and T. P. Russell, Adv. Mater. 24, 5688 (2012).

4. J. Bang, U. Jeong, D. Y. Ryu, T. P. Russell, and C. J. Hawker, Adv.Mater. 21, 4769 (2009).

5. X. Gu, I. Gunkel, and T. P. Russell, Philos. Trans. R. Soc., A 371(2013).

6. Z. Liu, Y. Wu, B. Harteneck, and D. Olynick, Nanotechnology24, 015305 (2013).

7. S. Rasappa, D. Borah, C. C. Faulkner, T. Lutz, M. T. Shaw, J. D.Holmes, and M. A. Morris, Nanotechnology 24, 065503 (2013).

8. S. Chang, V. P. Chuang, S. T. Boles, C. A. Ross, and C. V.Thompson, Adv. Funct. Mater. 19, 2495 (2009).

9. R. Dussart, T. Tillocher, P. Lefaucheux, and M. Boufnichel, J. Phys.D 47, 123001 (2014).

10. J. W. Bartha, J. Greschner, M. Puech, and P. Maquin, Microelectron.Eng. 27, 453 (1995).

11. Y. Wu, D. Olynick, A. Goodyear, C. Peroz, S. Dhuey, X. Liang, andS. Cabrini, Microelectron. Eng. 88, 2785 (2011).

12. F. Kucera and J. Jancar, Polym. Eng. Sci. 49, 1839 (2009).13. F. Kucera and J. Jancar, Polym. Eng. Sci. 38, 783 (1998).14. G. Y. Kim, R. Salovey, and J. Aklonis, Polym. Bull. 23, 543 (1990).15. A. F. Turbak, Ind. Eng. Chem., Prod. Res. Dev. 1, 275 (1962).16. S. H. Park, D. O. Shin, B. H. Kim, D. K. Yoon, K. S. Kim, S. Y.

Lee, S. H. Oh, S. W. Choi, S. C. Jeon, and S. O. Kim, Soft Matter6, 120 (2010).

17. D. H. Lee, D. O. Shin, W. J. Lee, and S. O. Kim, Adv. Mater.20, 2480 (2008).

18. S. J. Jeong, G. Xia, B. H. Kim, D. O. Shin, S. H. Kwon, S. W. Kang,and S. O. Kim, Adv. Mater. 20, 1898 (2008).

Received: 16 August 2014. Accepted: 10 March 2015.


Documents

Article · 2017-12-13 · An alternative approach is the block coplolymer (BCP) lithography which uses direct self-assembling of poly-mer patterns. Recently, it has attracted signiﬁcant