Enhanced adhesion of electron beam resist by grafted monolayerpoly(methylmethacrylate-co-methacrylic acid) brushFrancesco Narda Viscomi, Ripon Kumar Dey, Roberto Caputo, and Bo Cui Citation: Journal of Vacuum Science & Technology B 33, 06FD06 (2015); doi: 10.1116/1.4935506 View online: http://dx.doi.org/10.1116/1.4935506 View Table of Contents: http://scitation.aip.org/content/avs/journal/jvstb/33/6?ver=pdfcov Published by the AVS: Science & Technology of Materials, Interfaces, and Processing Articles you may be interested in Active-matrix nanocrystalline Si electron emitter array with a function of electronic aberration correction formassively parallel electron beam direct-write lithography: Electron emission and pattern transfer characteristics J. Vac. Sci. Technol. B 31, 06F703 (2013); 10.1116/1.4827819 Comparison between ZEP and PMMA resists for nanoscale electron beam lithography experimentally and bynumerical modeling J. Vac. Sci. Technol. B 29, 06F306 (2011); 10.1116/1.3640794 Sub-10-nm half-pitch electron-beam lithography by using poly(methyl methacrylate) as a negative resist J. Vac. Sci. Technol. B 28, C6C58 (2010); 10.1116/1.3501353 Self-assembled monolayers of poly(ethylene glycol) siloxane as a resist for ultrahigh-resolution electron beamlithography on silicon oxide J. Vac. Sci. Technol. B 27, 2292 (2009); 10.1116/1.3212899 Technique for preparation of precise wafer cross sections and applications to electron beam lithography ofpoly(methylmethacrylate) resist J. Vac. Sci. Technol. B 20, 3085 (2002); 10.1116/1.1518020

Redistribution subject to AVS license or copyright; see http://scitation.aip.org/termsconditions. Download to IP: 129.97.40.81 On: Wed, 18 Nov 2015 22:43:11

Enhanced adhesion of electron beam resist by grafted monolayerpoly(methylmethacrylate-co-methacrylic acid) brush

Francesco Narda Viscomia)

Department of Electrical and Computer Engineering and Waterloo Institute for Nanotechnology (WIN),University of Waterloo, Waterloo, Ontario N2L 3G1, Canada and Department of Physics andCNR-NANOTEC, University of Calabria, 87036 Rende, Italy

Ripon Kumar DeyDepartment of Electrical and Computer Engineering and Waterloo Institute for Nanotechnology (WIN),University of Waterloo, Waterloo, Ontario N2L 3G1, Canada

Roberto Caputob)

Department of Physics and CNR-NANOTEC, University of Calabria, 87036 Rende, Italy and Laboratoryof Nanotechnology and Instrumentation in Optics (LNIO), ICD CNRS UMR 6281, University of Technologyof Troyes, CS 42060, 10004 Troyes, France

Bo Cuic)

Department of Electrical and Computer Engineering and Waterloo Institute for Nanotechnology (WIN),University of Waterloo, Waterloo, Ontario N2L 3G1, Canada

(Received 24 July 2015; accepted 29 October 2015; published 12 November 2015)

In electron beam lithography, poor resist adhesion to a substrate may lead to resist structure

detachment upon development. One popular method to promote resist adhesion is to modify the

substrate surface. In this study, the authors will show that a poly(methylmethacrylate-co-methacrylic

acid) [P(MMA-co-MAA)] monolayer “brush” can be grafted onto a silicon substrate using thermal

annealing that leads to chemical bonding of the P(MMA-co-MAA) copolymer to the hydroxyl

group-terminated substrate, followed by acetic acid wash to remove the bulk, unbonded copolymer.

The monolayer brush has a thickness of 12 nm. The authors will show that it can greatly improve the

adhesion of positive resist, the ZEP-520A, and negative resist polystyrene to bare silicon surfaces,

which led to high resolution patterning without resist detachment upon development. The

improvement was more dramatic when patterning dense sub-100 nm period grating structures. But

the improvement was negligible for an aluminum substrate, because, even without the brush layer,

resist adhesion to aluminum is found already to be strong enough to prevent resist structure peeling

off. The current simple and low cost method could be very useful when resist adhesion to the

substrate for a given developer is weak. VC 2015 American Vacuum Society.

[http://dx.doi.org/10.1116/1.4935506]

I. INTRODUCTION

The performance of electron beam and other lithographies

is not only limited by “intrinsic” properties of the resist,

notably sensitivity and contrast, but also by “extrinsic”

factors, such as adhesion to the substrate. Currently, the

record densest line array pattern generated by electron beam

lithography (EBL), with an array periodicity down to 9 nm,

was obtained using hydrogen silsesquioxane (HSQ) resist,1

rather than using PMMA that has comparable or higher con-

trast. This is because the exposed HSQ effectively becomes

silicon dioxide that is more rigid than organic polymer resist

and it firmly adheres to the silicon substrate (with a natural

oxide layer atop) via Si–O–Si bond, whereas ultranarrow

PMMA lines would be deformed and detached from the sub-

strate due to capillary forces arising during the drying of the

rinsing liquid. Therefore, it is critical to modify the substrate

surface in order to promote the adhesion of the resist. For

instance, for photolithography, hexamethyldisilazane

(HMDS) is routinely applied to promote adhesion of photo-

resist to the substrate. For EBL using polystyrene (PS) resist,

or ZEP-520A resist with methyl ethyl ketone (MEK):methyl

isobutyl ketone (MIBK) developer, coating the substrate

with a thin layer of cross-linked polymer (here antireflection

coating) was found to greatly improve the resist adhesion.2,3

For EBL using HSQ resist on metallic or ceramic substrates,

a significant improvement in resist adhesion was achieved

by surface modification with (3-mercaptopropyl) trimeth-

oxysilane and poly(diallyldimethylammonium) chloride

(PDDA) for Au substrate, and with (3-aminopropyl) trie-

thoxysilane (APTES) for Mo substrate.4 Self assembled

monolayer (SAM) of APTES and PDDA can also improve

HSQ adhesion to Cr, Cu, and ITO substrate.4

The resist adhesion is characterized by the “work of

adhesion” W, defined as W¼ r13þ r23 � r12, where r13 is

the interfacial tension between the resist and the liquid

(developer), r23 is the interfacial tension between the sub-

strate and the liquid, and r12 is the interfacial tension

between the resist and the substrate. More details for the the-

oretical basics of resist adhesion can be found elsewhere.5,6

a)Present address: Department of Physics and Mathematics, University of

Hull, Hull HU6 7RX, United Kingdom.b)Electronic mail: roberto.caputo@fis.unical.itc)Electronic mail: bcui@uwaterloo.ca

06FD06-1 J. Vac. Sci. Technol. B 33(6), Nov/Dec 2015 2166-2746/2015/33(6)/06FD06/6/$30.00 VC 2015 American Vacuum Society 06FD06-1

Redistribution subject to AVS license or copyright; see http://scitation.aip.org/termsconditions. Download to IP: 129.97.40.81 On: Wed, 18 Nov 2015 22:43:11

It is considered that good resist adhesion will result if the

work of adhesion is greater than 5 mN/m.5 Apparently, the

adhesion can be greatly affected by the surface energy of the

liquid developer and the substrate, and W may even become

negative (resist autopeeling off in developer). For photoli-

thography, baking the substrate to drive off adsorbed water

is critical because otherwise the interfacial tension between

the aqueous developer and the substrate (i.e., the r23 term)

will be very low, leading to low work of adhesion. That is, a

hydrophilic substrate surface will promote the lateral pene-

tration of the aqueous developer into the resist/substrate

interface, weakening the adhesion or even peeling off the

resist structure by capillary force during liquid rinsing and

drying. By applying HMDS that is hydrophobic due to its

methyl group termination, the r23 term will increase and the

FIG. 1. (a) Chemical structure of P(MMA-co-MAA) (molar ratio MMA:MAA

1:0.016) and (b) grafting process of the copolymer onto –OH terminated sili-

con surface, resulting in a strong Si–O–C bond and release of water byproduct.

FIG. 3. (Color online) SEM images of grating structure exposed in ZEP resist and developed using MEK:MIBK for 1 min. (a) 100 nm period, with ZEP directly

coated on silicon, showing almost all the lines were detached; (b) 100 nm period, with ZEP coated on silicon treated with PMMA brush, showing well defined pattern;

(c) 175 nm period, coated on silicon; and (d) 175 nm period, coated on PMMA brush, both showing well defined pattern. Dose 0.117 nC/cm, exposed at 20 keV.

FIG. 2. (Color online) AFM image of grafted P(MMA-co-MAA) monolayer

brush exposed with a square array pattern and developed by MIBK:IPA.

The residual thickness at the exposed area should be very small since

exposed PMMA (contains very small amount of MAA) has much shorter

chain length than the unexposed one. Therefore, the pattern height of 12 nm

is approximately the monolayer thickness.

06FD06-2 Viscomi et al.: Enhanced adhesion of electron beam resist 06FD06-2

J. Vac. Sci. Technol. B, Vol. 33, No. 6, Nov/Dec 2015

Redistribution subject to AVS license or copyright; see http://scitation.aip.org/termsconditions. Download to IP: 129.97.40.81 On: Wed, 18 Nov 2015 22:43:11

r12 term will decrease (as photoresist consisting of novolac

binder is also rich with methyl group), which both contribute

to the increased work of adhesion.

Finally, secondary and Auger electrons generated by the

e-beam exposure at the substrate–resist interface were found

to greatly decrease positive resist adhesion because chain

scission caused by those electrons makes the resist at the

interface subject to dissolution by the developer. This effect

can be minimized by using low atomic number substrate

such as carbon that has a low yield of secondary electrons.7

Moreover, denser pattern leads to more exposure due to

proximity effect at the area not directly exposed by the pri-

mary beam; thus, the resist structure is more prone to peeling

off. This effect is not an issue for negative resist for which

electron beam exposure may lead to stronger adhesion,

which is the case for HSQ coated on silicon substrate.

In this paper, we modified the substrate surface by graft-

ing a monolayer of poly(methylmethacrylate-co-methacrylic

acid) [P(MMA-co-MAA)] brush that led to a remarkable

improvement of the resist performance for positive resist

ZEP-520A and negative resist polystyrene. Yet, the improve-

ment is negligible when the resist was coated on Al sublayer,

as the resist adhered well to Al even without the copolymer

brush layer. Our method has some advantages compared

with those cited above. The grafting is simpler than the

SAM process, and the copolymer is cheaper than the SAM

surfactant. It is also easy to work with, and the very thin

monolayer would not significantly affect the subsequent

pattern transfer process by direct etch or lift-off.

II. EXPERIMENT

The P(MMA-co-MAA) brush grafting took place in the

same way as forming the PMMA-r-polystyrene brush, which

is widely used to provide a neutral surface for self-assembly

of di-block copolymer PMMA-b-polystyrene.8,9 The silicon

surface after solvent cleaning was activated by oxygen

plasma to give –OH termination onto the surface, and then,

it was coated with a thick layer of P(MMA-co-MAA) (molar

ratio MMA:MAA 1:0.016, Mw 34 kg/mol, Mn 15 kg/mol,

Sigma-Aldrich, St. Louis, Missouri, USA). Here, MAA

contains the desirable –COOH group to promote the grafting

process. Next, the copolymer was annealed overnight at

160 �C to induce the chemical reaction between –OH on the

resist and –OH on the surface (Fig. 1), which led to the pro-

duction of water and chemical bonding between the resist

and substrate (–Si–O–C– for silicon substrate). The anneal-

ing temperature and time are not optimized, and the time

would be significantly shorter for higher temperatures; but

the temperature should be below 200 �C to avoid degradation

of the PMAA component.10

The bulk of copolymer that was not chemically bonded

to the substrate was washed away by dipping the sample in

acetic acid and then rinsing with de-ionized water. The result

is a surface covered with a copolymer monolayer. The name

brush is given for the single layer of pending molecules,

which remind the bristles of a brush. As the monolayer brush

is very thin, the properties of the resist coated above are not

significantly altered.

In this work, the adhesion of both positive and negative

tone resists has been studied. The ZEP-520A (Zeon

Chemicals) has been chosen as a positive resist. This resist

adheres well to silicon when using high resolution developers,

notably amyl acetate and xylene. But when using a stronger

developer, the fine resist structure was found detached after

development.3 As such, its problem of adhesion on silicon

surface makes it the perfect candidate to verify the efficacy of

our process. The ZEP resist was coated by spin-coating at

4000 rpm for 40 s and baked at 180 �C for 8 min to give a final

film thickness of 190 nm. After exposure, it was developed

for 1 min using MEK:MIBK ratio 60:40 and rinsed by isopro-

panol before drying. This developer offers much higher sensi-

tivity than the company suggested high resolution developers

(amyl acetate or xylene), but it attacks the resist more aggres-

sively because it is more polar and penetrates more easily into

the resist/substrate interface, which detaches the resist. To test

the adhesion property, we exposed a pattern consisting of gra-

tings with different pitches ranging from 60 to 200 nm. As a

comparison and to show the effects induced by the brush layer

on the resist adhesion, we also exposed identical pattern under

identical condition on regular samples (resist on silicon).

PS has been chosen as negative resist to study the effect

of P(MMA-co-MAA) brush monolayer on its adhesion to

FIG. 4. Very low magnification view of exposed polystyrene pattern, to

show all the grating patterns having different periods and exposed at expo-

nentially increasing doses from left to right. (a) Resist coated on bare silicon

wafer, showing only a few gratings with large periods and high exposure

doses survived without detachment. (b) Resist coated on P(MMA-co-MAA)

brush that was grafted to the silicon substrate.

06FD06-3 Viscomi et al.: Enhanced adhesion of electron beam resist 06FD06-3

JVST B - Nanotechnology and Microelectronics: Materials, Processing, Measurement, and Phenomena

Redistribution subject to AVS license or copyright; see http://scitation.aip.org/termsconditions. Download to IP: 129.97.40.81 On: Wed, 18 Nov 2015 22:43:11

substrate. We have chosen PS because it is a very versatile

and a popular negative resist, offering ultrahigh resolution

for low molecular weight,2 and ultrahigh sensitivity for high

molecular weight.11 As well, it can be thermally devel-

oped,12 coated by thermal evaporation to allow nanofabrica-

tion on irregular surfaces such as AFM cantilever and

optical fiber,13 or doped with metal Cr to greatly enhance its

etching resistance.14 Yet its adhesion to the silicon substrate

is weak, and a cross-linked polymer sublayer has been previ-

ously employed to promote its adhesion.2 Here, we used PS

with a molecular weight of 30 kg/mol dissolved in anisole

with a concentration of 2 wt./vol. %. After spin-coating, the

sample was baked at 160 �C for 8 min on a hotplate to give a

final film thickness of 80 nm.

The electron beam lithography was carried out using a

Leo 1530 scanning electron microscope equipped with a

Raith beam blanker and nanometer pattern generation sys-

tem. The acceleration voltage used for all the experiments

was 20 kV with a beam current of about 160 pA. Polystyrene

and ZEP were exposed with a line dose range of 0.4–2.0 and

0.04–0.20 nC/cm, respectively.

III. RESULTS AND DISCUSSION

Since PMMA (here contains very small amount of MAA)

itself is an electron beam resist, the monolayer thickness can

be obtained by AFM measurement of the pattern defined by

electron beam lithography. As shown in Fig. 2, the mono-

layer thickness was found to be about 12 nm.

We first studied the positive resist ZEP, and Fig. 3 shows

gratings written with a line dose of 0.117 nC/cm at 20 keV

electron energy. For 100 nm grating periodicity [Figs. 3(a)

and 3(b)], almost all lines were detached from the sample

surface without the brush layer, whereas they were firmly

attached when the surface had been treated with the copoly-

mer brush. For larger periodicity such as 175 nm [Figs. 3(c)

and 3(d)], both gratings survived the development process

without detachment. Here, the peeling off for the case of

100 nm period without the brush adhesion layer is caused by

the strong developer MEK:MIBK we used for attaining a

high sensitivity. This more polar developer would penetrate

more easily in-between the resist and substrate, weakening

the adhesion and leading to resist detachment due to capil-

lary force during liquid rinsing and drying. Besides consider-

ation of surface and interface energies at the development

step, the brush also promotes adhesion because the ZEP

polymer chains would get “clipped” (interlocked) with

the copolymer chain that would hinder the penetration of

developer into the interface. Finally, though PMMA (here

contains very small amount of MAA), which is also a posi-

tive tone resist, would also be “weakened” by exposure due

to proximity effect, this effect should be small compared to

ZEP directly coated onto silicon since PMMA is much less

FIG. 5. (Color online) SEM images of grating structure exposed in polystyrene resist and developed using THF for 1 min (a) 60 nm period, PS directly coated

on silicon, showing all the lines were detached; (b) 60 nm period, PS coated on silicon treated with P(MMA-co-MAA) brush, showing well defined pattern; (c)

200 nm period, PS directly coated on silicon; and (d) PS coated on silicon treated with P(MMA-co-MAA) brush.

06FD06-4 Viscomi et al.: Enhanced adhesion of electron beam resist 06FD06-4

J. Vac. Sci. Technol. B, Vol. 33, No. 6, Nov/Dec 2015

Redistribution subject to AVS license or copyright; see http://scitation.aip.org/termsconditions. Download to IP: 129.97.40.81 On: Wed, 18 Nov 2015 22:43:11

sensitive than ZEP resist that implies a less easy chain

scission.

Besides P(MMA-co-MAA) monolayer brush as shown

here, HMDS is also sometimes employed with an aim to pro-

mote adhesion of ZEP resist to silicon substrate.15 However,

to our knowledge, systematic study to evaluate the effect of

HMDS on ZEP resist adhesion was not reported. To this end,

we applied HMDS using a dedicated HMDS prime oven

(YES oven, Yield Engineering Systems, Inc.), and compared

the lithography results with or without HMDS. For both

cases, no resist peeling off was found when using the high

resolution developer amyl acetate. In fact, HMDS is not sug-

gested by the resist manufacturer as a necessary step before

coating ZEP. When using high sensitivity (yet low resolu-

tion) MEK:MIBK developer, the degree of resist peeling off

was found to be similar for the two cases, indicating HMDS

does not help promote adhesion of ZEP to silicon. This is

probably because (unlike aqueous developer for photoresist)

the solvent developer here can wet the HMDS layer and

penetrate into the resist–substrate interface; and HMDS is a

relatively small molecule that would give a rather smooth

surface, whereas P(MMA-co-MAA) has a long chain that

gives a brushlike surface to entangle with the copolymer

chain in the ZEP resist.

The experiment has been repeated with a negative tone

resist as well. For this purpose, we chose 30 kg/mol PS dis-

solved in anisole with 2 wt./vol. % concentration for spin-

coating to give an 80 nm-thick film, and used tetrahydrofuran

(THF) as developer and isopropanol as the rinsing liquid. As

clearly seen in Fig. 4 for the low magnification view of all

the exposed grating patterns, with brush, most grating

patterns remained undetached; whereas severe detachment is

evident when the polystyrene resist was coated directly on

a bare silicon wafer treated with oxygen plasma before

spin-coating. High magnification SEM images (Fig. 5)

shows that, with PS coated on the brush layer, a periodicity

down to 60 nm (30 nm half pitch) was well defined; whereas

the same grating structure was completely detached when PS

is coated directly onto a silicon wafer. For larger period,

such as 200 nm, both gratings were well defined without

resist detachment [Figs. 5(c) and 5(d)]. With its lower sur-

face energy and being less polar than PMMA, PS was found

to adhere poorly to silicon; and previously a cross-linked

polymer under-layer have been coated to promote the adhe-

sion in order to obtain an ultradense grating pattern of 20 nm

periodicity using a very low molecular weight PS2. Such an

under-layer would make the subsequent pattern transfer

more challenging since it is much thicker than the P(MMA-

co-MAA) brush used here.

It would also be interesting to see whether the same brush

layer will equally help the resist adhesion to nonsilicon sur-

face. To this end, we coated 20 nm of aluminum on silicon

wafer, followed by grafting the copolymer brush atop. As

can be seen from Fig. 6 for ZEP resist with 100 nm grating

period, the brush made no difference: the ZEP resist’s adhe-

sion to Al is already strong enough, and thus, the brush did

not give any further improvement. The difference is also

negligible for other grating periods (not shown). The reason

for such a good adhesion can be attributed to the roughness

of the Al film, which offers more anchoring points than an

atomically smooth crystalline silicon surface. Or, we can say

that a thin Al film will also function as an adhesion layer like

the P(MMA-co-MAA) brush, though it is less desirable since

vacuum deposition is a more costly process than P(MMA-

co-MAA) brush grafting and the Al layer will complicate the

subsequent patterns transfer process.

IV. CONCLUSIONS

We showed that a PMMA monolayer brush can be

grafted onto silicon substrate using thermal annealing that

leads to chemical bonding of P(MMA-co-MAA) to the

hydroxyl group-terminated substrate, followed by acetic

FIG. 6. (Color online) SEM images of 100 nm period grating structure

exposed in ZEP resist. (a) ZEP directly coated on aluminum surface and (b)

ZEP coated on aluminum film treated with P(MMA-co-MAA) brush. Both

gratings were very well defined.

06FD06-5 Viscomi et al.: Enhanced adhesion of electron beam resist 06FD06-5

JVST B - Nanotechnology and Microelectronics: Materials, Processing, Measurement, and Phenomena

Redistribution subject to AVS license or copyright; see http://scitation.aip.org/termsconditions. Download to IP: 129.97.40.81 On: Wed, 18 Nov 2015 22:43:11

acid wash to remove the bulk, unbonded copolymer. The

monolayer brush has a thickness of 12 nm. We showed that

it can greatly improve the adhesion of positive resist ZEP-

520A and negative resist polystyrene to bare silicon surfa-

ces, which led to high resolution patterning without resist

detachment upon development. The improvement was more

dramatic when patterning dense sub-100 nm period grating

structures. But the improvement was negligible for alumi-

num substrate, because, even without the brush layer, resist

adhesion to aluminum is found already strong enough to

prevent resist structure peeling off. This simple and low-

cost method would be very useful when resist adhesion to

the substrate for a given developer is weak. Finally, since

P(MMA-co-MAA) is an electron beam resist and the mono-

layer brush can be grafted on any surface by spin or dip

coating, the process can be employed to fabricate nanostruc-

tures using electron beam lithography on irregular surfaces

such as an AFM cantilever, and this study will be published

elsewhere.

ACKNOWLEDGMENTS

F.N.V. acknowledges the financial support from

“Mobility of Students” (MOST) Program held by University

of Calabria. F.N.V. and R.C. would like to acknowledge the

contribution of the COST Action IC1208.

1J. K. W. Yang, B. Cord, H. Duan, K. K. Berggren, J. Klingfus, S. W. Nam,

K. B. Kim, and M. J. Rooks, J. Vac. Sci. Technol. B 27, 2622 (2009).2S. Ma, C. Con, M. Yavuz, and B. Cui, Nanoscale Res. Lett. 6, 446 (2011).3B. Shokouhi, J. Zhang, and B. Cui, Micro Nano Lett. 6, 992 (2011).4Z. Zhang, H. Duan, Y. Wu, W. Zhou, C. Liu, Y. Tang, and H. Li,

Microelectron. Eng. 128, 59 (2014).5J. Bauer, G. Drescher, and M. Illig, J. Vac. Sci. Technol. B 14, 2485 (1996).6J. Bauer, G. Drescher, H. Silz, H. Frankenfeld, and M. Illig, Proc. SPIE

3069, 640 (1997).7A. Schmidt, A. Clifton, W. Ehrfeld, G. Feiertag, H. Lehr, and M. Schmidt,

Microelectron. Eng. 30, 215 (1996).8P. Mansky, Y. Liu, E. Huang, T. P. Russell, and C. Hawker, Science 275,

1458 (1997).9D. Borah, S. Rasappa, R. Senthamaraikannan, M. T. Shaw, J. D. Holmes,

and M. A. Morris, J. Colloid Interface Sci. 393, 192 (2013).10B. Ho, Y. Lee, and W. Chin, J. Polym. Sci., Part A: Polym. Chem. 30,

2389 (1992).11C. Con, R. Dey, M. Ferguson, J. Zhang, R. Mansour, M. Yavuz, and B.

Cui, Microelectron. Eng. 98, 254 (2012).12C. Con, A. S. Abbas, M. Yavuz, and B. Cui, Adv. Nano Res. 1, 105 (2013).13J. Zhang, C. Con, and B. Cui, ACS Nano 8, 3483 (2014).14C. Con, J. Zhang, and B. Cui, Nanotechnology 25, 175301 (2014).15A. Stein, C. Jacobsen, K. Evans-Lutterodt, D. M. Tennant, G. Bogart, F.

Klemens, L. E. Ocola, B. J. Choi, and S. V. Sreenivasan, J. Vac. Sci.

Technol. B 21, 214 (2003).

06FD06-6 Viscomi et al.: Enhanced adhesion of electron beam resist 06FD06-6

J. Vac. Sci. Technol. B, Vol. 33, No. 6, Nov/Dec 2015

Redistribution subject to AVS license or copyright; see http://scitation.aip.org/termsconditions. Download to IP: 129.97.40.81 On: Wed, 18 Nov 2015 22:43:11