highly ordered square arrays from a templated abc triblock terpolymer

Published: June 16, 2011

r 2011 American Chemical Society 2849 dx.doi.org/10.1021/nl201262f |Nano Lett. 2011, 11, 2849–2855

LETTER

pubs.acs.org/NanoLett

Highly Ordered Square Arrays from a Templated ABC TriblockTerpolymerJeong Gon Son,† Jessica Gwyther,§ Jae-Byum Chang,† Karl K. Berggren,‡ Ian Manners,§ andCaroline A. Ross*,†

†Department of Materials Science and Engineering and ‡Department of Electrical Engineering and Computer Science,Massachusetts Institute of Technology, Cambridge, Massachusetts 02139, United States§School of Chemistry, University of Bristol, Bristol BS8 ITS U.K.

bS Supporting Information

Self-assembly of block copolymer thin films enables theformation of highly regular sub-10 nm microdomain patterns

at low cost in a simple process and is currently of great interest fornanolithography and device fabrication.1�5 Many groups havestudied the self-assembly of diblock copolymer films, but thesematerials are limited to certain geometries such as arrays ofparallel lines or close-packed dots. To extend the available varietyof microdomain geometries, triblock terpolymer self-assemblyhas been explored,6�8 which can generate geometries such asring-shape patterns9 or square-symmetry patterns.10 In particular,the square-symmetry array is a key geometry that is consideredessential for future device fabrication and which may be useful inmaking high density magnetic patterned media11 and via arrays inintegrated circuits. Several strategies have been developed formaking square-symmetry arrays from block copolymer self-assem-bly, including the templated self-assembly of a diblock copolymeron a square chemically patterned substrate with the same periodi-city as the copolymer,12 body-centered cubic packing of sphericalmicrodomains in triblock terpolymer films of specific thickness,13

square-symmetry cylindrical microdomains from a blend of twohydrogen-bonded diblock copolymers,14,15 diblock copolymerblend films in small square wells,16 and the self-assembly of thinfilms of a triblock terpolymer with a bulk square symmetry.10

Despite these successes, the square-symmetry patterns do nothave good long-range order in the absence of templating because

microphase separation leads to the formation of regions ofmicrodomains with short-range order but random in-planeorientation. The extent of short-range order, that is, the average“grain size” of the pattern, is determined by the annealing processand by the surface properties of the substrate.17 Long-range orderhas been imposed on diblock copolymer thin films by usingchemical or topographical substrate patterns,3,4,18�24 for exam-ple the use of sparse arrays of topographical posts to patterndense arrays of spherical25 or cylindrical26 microdomains, drivenby the strong affinity of the template to one domain of thediblock copolymer. However, there has been very little work onthe templating of triblock terpolymer films, at least in partbecause it is less straightforward to select appropriate topogra-phical or chemical substrate patterns to direct the assembly of atriblock terpolymer. As an example, we showed previously thatthe orientation of the microdomains in a square-pattern triblockterpolymer film with respect to a shallow substrate trench couldbe controlled by using a brush layer.10

In this paper, we demonstrate first the formation of highlyordered square patterns from thin films of a triblock terpolymerby use of a substrate brush layer, and second, registration of the

Received: April 14, 2011Revised: May 23, 2011

ABSTRACT: Square-symmetry patterns are of interest in nano-lithography but are not easily obtained from self-assembly of adiblock copolymer. Instead, we demonstrate highly ordered 44 nmperiod square patterns formed in a thin film of polyisoprene-block-polystyrene-block-polyferrocenylsilane (PI-b-PS-b-PFS) triblockterpolymer blended with 15% PS homopolymer by controllingthe film thickness, solvent anneal conditions, the surface chemistryand topography of the substrates. The square patterns consist ofPFS pillars that remained after removal of the PI and PS with anoxygen plasma. On an unpatterned smooth substrate, the averagegrain size of the square pattern was increased dramatically to severalmicrometers by the use of brush layers and specific solvent anneal conditions. Templated self-assembly of well-ordered squarepatterns was demonstrated on substrates containing nanoscale topographical sidewalls and posts, written by electron beamlithography, in which the sidewalls and base of the substrate were independently chemically functionalized.

KEYWORDS: Block copolymer, triblock terpolymer, self-assembly, templated self-assembly, square pattern, nanolithography

2850 dx.doi.org/10.1021/nl201262f |Nano Lett. 2011, 11, 2849–2855

Nano Letters LETTER

microdomains using topographical posts and square wells. Thiswork extends the ability to precisely register and control micro-domain locations to triblock terpolymers, facilitating their use innanolithography and augmenting the range of available patterngeometries.

The films were made from a polyisoprene-block-polystyrene-block-polyferrocenylsilane (PI-b-PS-b-PFS) triblock terpolymerblended with 15% PS homopolymer in which the bulk morphologyconsists of PI and PFS cylinders in an alternating square-symmetryarrangement within a PS matrix. We previously showed that thinfilms of this blend could produce square-symmetry patterns withlimited long-range order.10 We show here that highly orderedsquare patterns with average grain size on the scale of micro-meters can be achieved by controlling the surface chemistry ofthe substrate using a brush layer that differs chemically from allthree of the blocks in the triblock terpolymer. Furthermore,topographical templates in which the sidewalls and substrate aretreated independently with different brush layers were used toimpose long-range order and registration of the square patterns.

An A-b-B-b-C triblock terpolymer can form alternating squaresymmetry cylinders of A and C in a matrix of B (Figure 1a) whenB is the majority block and the Flory�Huggins interaction

parameter between A and C, χAC is higher than χAB andχBC.

6,7 PI-b-PS-b-PFS is a suitable material according to thesolubility parameters δ, which are 17.0 (MPa)1/2 for PI, 18.5(MPa)1/2 for PS and 18.7 (MPa)1/2 for PFS.27,28 The interactionparameter χAB is proportional to the square of the difference insolubility parameters between A and B.29 We used an 82 kg/molPI-b-PS-b-PFS triblock terpolymer with volume fractions of 25,65, and 10%, respectively, blended with 15 wt % of 27 kg/mol PShomopolymer that increases the range of film thicknesses overwhich perpendicular cylinders form.10 Additionally, because thePFS moiety contains iron and silicon, it etches slowly comparedto the PI and PS blocks when treated with an oxygen plasma,which simplifies pattern transfer to other materials.

Thin films of the triblock/homopolymer blend were formedon brush-coated Si wafers by spin-coating a toluene solution ofthe blend. The polymer film was solvent annealed at roomtemperature in chloroform vapor to obtain square-symmetrymicrodomain patterns. The solvent vapor pressure in the cham-ber was controlled using a leak valve, Figure 1a, which wasadjusted to vary the degree of swelling of the films, themobility ofthe block segments30 and the interaction parameters,31 and topromote self-assembly of the microdomains.32 Solvent annealingcan change the volume fractions of the domains as a result ofselective uptake of solvent molecules within each block, whichcan drive order�order transitions,24,30 and the effective χ para-meter (χeff) decreases linearly with the fraction of solventmolecules incorporated.29 The film thickness change caused byswelling of the films was monitored using in situ film thicknessmeasurement by spectral reflectometry.30 Chloroform vapor wasused to swell the films from 128 to 178% thickness change.Figure 1b�d shows SEM images after etching of 32 nm (thereference unswelled film thickness) thick PI-b-PS-b-PFS triblockterpolymer/15% PS homopolymer blended films swelled to (b)128, (c) 152, and (d) 178% of the unswelled film thickness. In thecase of the 128% swelled film, the final PFS morphology,observed after removing the PI and PS using an oxygen plasma,was poorly ordered indicating that this degree of swelling was notenough to provide an adequate mobility. The 152% swelled filmformed highly ordered square array patterns showing an opti-mum degree of mobility without lowering the interaction para-meter excessively, but the 178% swelled films again formedpoorly ordered microdomain patterns. In this case the poororder may be a result of a change in volume fraction of thedomains. Figure S1 of the Supporting Information shows thathomopolymers of the three blocks swell similarly to the triblockblend of chloroform at low swelling ratios, but at high swellingratios the PFS swells less than the other blocks, which maypromote a morphological transition. In addition, the high swell-ing ratio is expected to lower the interaction parameters (χeffPS/PI,χeffPS/PFS, χeffPI/PFS), which lowers the driving force for micro-phase separation.

The thickness of the films also affects the morphology andorientation of the microdomains. At 32 nm thickness, squaresymmetry perpendicular cylinder arrays were formed, as shownin Figure 1c. However, when 37 nm thickness films were swelledby 152%, the films formed discrete terraced structures containingboth in-plane and perpendicular cylinders (Figure 1e). Thickerfilms, such as 60 nm, showed predominantly in-plane cylinders.Such an orientation change with film thickness is well establishedfor diblock copolymer thin films, both theoretically33 andexperimentally,34 and is related to the commensurability betweenthe microdomain period and the film thickness.

Figure 1. (a) Schematics of the perpendicular orientation of alternatingcylinders with square symmetry in an ABC triblock terpolymer filmusing solvent annealing. (b�d) SEM images of oxidized PFS micro-domains from a 32 nm thick film of PI-b-PS-b-PFS triblock terpolymerblended with 15 wt % homopolymer PS after solvent annealing inchloroform vapor, after (b) 128%, (c) 152%, and (d) 178% swellingduring solvent annealing. Oxygen plasma etching removes the PI and PSdomains revealing oxidized PFS cylinders. Only the 152% swelled filmformed highly ordered square arrays with period 44 nm. (e) Thickerfilms, such as 37 nm, showed in-plane cylinders when the films wereswelled by 152%.


Nano Letters LETTER

We now describe the effect of surface chemistry on the averagegrain size of the square-array pattern. Ordering of diblockcopolymer domain patterns is known to be significantly affectedby the surface properties of the substrate,17 and we previouslyfound large differences in the correlation length of cylindricalpolystyrene-b-polydimethylsiloxane (PS-b-PDMS) block copo-lymer thin films deposited on pristine Si wafers, on PS brush-coated and on PDMS brush-coated substrates.35 Most studies ofbrush effects use a brush consisting of one of the blocks of thecopolymer, but it is also possible to use a different brush chemistry.In this work, we select brushes of PEO (polyethylene oxide), PB(polybutadiene), and P2VP (poly(2-vinylpyridine)) in additionto PS and PFS brushes. Brushes were formed by graftinghydroxyl-terminated short-chain homopolymers onto oxidizedSi wafers, then rinsing away the ungrafted chains.

Figure 2 shows SEM images of PI-b-PS-b-PFS/PS blend filmson various substrates after oxygen plasma treatment, and theaverage grain size obtained from the images. Because theoptimum film thicknesses to form perpendicular cylinder squarearrays can differ according to the wetting behavior of the film onthe substrate, we selected the optimum thickness as just belowthe thickness at which in-plane cylinders start to form. This was32 nm, except for the PFS brush sample where the optimum filmthickness was 39 nm. On an untreated prime Si wafer (seeFigure 2a), a relatively small grain size was observed because thehydroxyl-terminated polar native oxide surface strongly inter-acted with the polymer chains and lowered the mobility. In thecase of both PS and PFS brushes, the films also showed very smallgrain size (Figure 2b,c). However, on brush substrates that differfrom the three blocks in the triblock terpolymer, such as PB,P2VP, and PEO, (Figure 2d�f), the films formed surprisinglyhighly ordered PFS square arrays. Figure 2g shows the averagegrain sizes on each substrate. While the average grain size on theSi wafer, PS brush and PFS brush sample were 319, 214, and165 nm, respectively, the PB brush, P2VP brush, and PEO brushsubstrate, showed much larger grain sizes, 848 nm, 1.22, and 1.75μm, respectively.

To investigate the differences between the self-assembly of thefilms on different brush substrates, the top and bottom morpho-logies of the films were examined using atomic force microscopy

(AFM) without any etching. In each case, the top surface of thefilms was featureless (Figure 3a) suggesting that preferentialwetting of one component of the triblock terpolymer occurred atthe top surface of the films regardless of substrate. This prefer-ential wetting originates from surface energy minimization of thefilms. The surface energies of PI and PS are 32 and 40 mN/m,respectively, and the surface energy of PFS is known to be slightlyhigher than that of PS,36 so we expect the PI domains to beselectively exposed at the top surface of the films. However, at thebottom interface the morphology was highly dependent on thebrush layer. To observe the bottom surface, 50 nm SiO2 layerdeposited Si wafer was used as the substrate. The films wereimmersed in 5 wt % HF aqueous solution to dissolve the SiO2

Figure 2. SEM images and grain sizes of square-array patterns from a PI-b-PS-b-PFS blended with 15% PS thin film on (a) Si wafer, (b) PS brush,(c) PFS brush, (d) PB brush, (e) P2VP brush, (f) PEO brush-coated substrates. (g) Average grain sizes of the square patterns were measured based on low-magnification SEM images. The grain size of the square pattern is increased dramatically to several micrometers by the use of P2VP or PEO brush layers.

Figure 3. (a) AFM top surface image of thin films of PI-b-PS-b-PFSblended with 15% PS. A featureless surface was observed at the topsurface of the films. (b�d) AFM bottom interface images of the films on(b) PS brush, (c) PFS brush, and (d) PEO brush-coated substrates.Highly ordered square hole patterns were observed at the bottominterface of PEO and P2VP brush samples indicating the PFS cylinderscontact the lower interface, while no features were observed at thebottom interface of films on PS brush or PFS brush substrates.


Nano Letters LETTER

sacrificial layer and release the polymer film, which was picked upon another substrate to expose the bottom interface of the film.The films on PS or PFS brush-coated substrates did not showordered patterns at the bottom interface (Figure 3b,c), which wasattributed to selective wetting of the brush-coated substrate bythe PS and PFS blocks, respectively. However, on the PEO orP2VP brush-coated substrate, well-ordered square arrays of holepatterns were observed (Figure 3d). We believe that the Si-containing PFS domains selectively reacted with the HF solution,enabling their locations to be observed.37 The square symmetryhole arrays revealed that PFS domains directly contact thesubstrate.

Schematics of the film cross sections are shown in the insets ofFigure 3b�d. The films on PS or PFS coated brush substratewere assumed to form a PI layer at the top surface and fullywetted PS or PFS layers at the bottom interface, respectively.These surface layers transformed the alternating cylinder mor-phology into a spherical morphology at this thickness condition,which degraded the tendency for square packing that existsbetween the parallel cylinders in bulk. However, when thetriblock terpolymer films were annealed on the PEO or P2VPbrushed substrate, the PFS and PI cylinders contacted thebottom surface and oriented perpendicularly to the film to formthe square symmetry pattern.

ThePEO(δ∼20.2 (MPa)1/2) andP2VP(δ∼20.4 (MPa)1/2)28

brushes are relatively hydrophilic compared to all three ofthe triblock terpolymer blocks. However, solvent annealingaffects the magnitude of the interfacial energy between thebrushes and the three blocks of the triblock terpolymer. Inter-facial energy γ is proportional to the square root of theFlory�Huggins interaction parameter χ and therefore to thedifference in solubility parameters.29 Solvent annealing lowersthe effective χ parameter (χeff) linearly as more molecules areincorporated at the interface.29 Brushes made from one of thecomponents (PI, PS, or PFS) promote preferential wetting withzero interfacial energy regardless of the presence of solventmolecules, but for the PEO or P2VP brushes, incorporation ofmore solvent molecules lowers χeff between any of the domainsand the brush, which makes the brush closer to being a neutralsurface (i.e., γeff PI-PEO, γeff PS-PEO, and γeff PFS-PEO become moresimilar). This promotes a perpendicular orientation of thecylindrical triblock terpolymer domains on a PEO or P2VPbrush substrate with good long-range order. Several groups havealso reported that solvent annealing promotes perpendicularorientation of block copolymer microdomains in thin films evenon energetically non-neutral substrates.32 Thus, long-range orderof the triblock terpolymer films was successfully accomplished byoptimization of solvent annealing, film thickness and surfacechemistry of the substrate.

We now show how these well-ordered square patterns can betemplated using topographically guiding wall and post patternsto enable pattern registration. Topographical ridge patterns witha rectangular layout were prepared by electron-beam patterningof an negative-tone inorganic hydrogen silsesquioxane (HSQ)resist (Figure 4a). The wall height was 32 nm, similar to the filmthickness, and the wall width was 20 nm. Preferential wetting ofthe sidewalls by one block is desirable in order to promoteordering,3,35 while nonpreferential wetting is desirable on thehorizontal surface of the substrate to improve the ordering asdiscussed above. These considerations suggest that a heteroge-neous surface is required38 with different brush layers on thevertical and horizontal surfaces. To accomplish this, a PFS brush

(molecular weight ∼5 kg/mol) was first grafted over the entiresubstrate, then treated with a short (∼2 s) oxygen reactive ionetch (RIE) that anisotropically etches the PFS brush on thehorizontal surfaces. Then the substrate was treated with ahydroxyl-terminated PEO brush that coated the horizontalsurfaces without displacing the PFS brush on the sidewalls.

The triblock terpolymer/PS homopolymer blend films formedhighly ordered and aligned square symmetry dot arrays in micro-meter-size rectangular cages, shown in Figure 4b,c. Rectangularcages of 500 nm � 500 nm, 2 μm � 3 μm and 1.5 μm � 4 μmeach formed a single-grain pattern. The orientation of the basevector of the PFS microdomain array parallel to the wall and thespacing between the PFS domains and thewall are consistent with abrush layer of PFS being present at the wall, so the first row of

Figure 4. Highly ordered square arrays formed from a triblock terpo-lymer film in an HSQ template. (a) Schematic of a square array ofmicrodomain triblock terpolymer films within a topographical templatewith preferential PFS brush coated sidewalls and nonpreferential PEObrush coated horizontal surfaces. (b,c) SEM images of highly alignedsquare arrays of oxidized PFSmicrodomains in (b) 1.5 and 2 μm� 3 μmand (c) 0.5 μm, 1 μm, and 1.5 μm � 4 μm templates.


Nano Letters LETTER

cylindrical microdomains that forms adjacent to the wall consist ofPI.10 The height of the HSQ sidewalls also affected the ordering ofthe square patterns, as can be seen in Figure S2 in the SupportingInformation. When the height exceeded the polymer film thick-ness, for example 50 nm, the film formed a meniscus near thesidewalls and in-plane cylinders were observed near the sidewalls.Sidewalls that were thinner than the film thickness, for example,25 nm, led to distortions in the square-symmetry arrays.

Sparse arrays of posts were also effective in templating thesquare arrays, as illustrated in Figure 5a. Arrays of electron-beampatterned HSQ posts were made with 88 nm � 132 nm spac-ing, commensurate with the triblock terpolymer pattern period,L0 = 44 nm. Arrays were also made containing paired posts with44 nm spacing in which the double-post motif was repeated on a176 nm � 132 nm period lattice. The height of the posts was25 nm, slightly below the film thickness, and the diameter wasapproximately 10 nm. The post sidewalls were coated with a PFS

brush and the horizontal surface of the substrate with a PEObrush. The PFS-coated posts acted as surrogates for perpendi-cular cylindrical PFS microdomains. Figure 5b shows square dotarrays formed on a commensurate post lattice with period ratioLpost,x/L0 = 2.0 and Lpost,y/L0 = 3.0. The basis vectors of the postlattice and that of triblock terpolymer microdomain lattice areaccurately aligned. Even sparser double-post arrays with Lpost,x/L0 = 3.0 and Lpost,y/L0 = 4.0 were also effective at templating thearray. We found earlier for post-templating of close-packedsphere arrays from a diblock copolymer that, as Lpost increases,the energy difference between different orientations of themicrodomain lattice is reduced and multiple microdomain latticeorientations form on different regions of the post lattice. In thepresent case of the triblock terpolymer, however, this tendency issuppressed by the use of the double-post motif, which promotes aparallel alignment of the post lattice and the microdomain lattice.This square pattern post templating is the first demonstration ofthe use of a topographical pattern to precisely register thelocation and orientation of the microdomains of a triblockterpolymer film.

In summary, highly ordered square patterns with 44 nm periodand average grain size on the micrometer scale have beendemonstrated in films of a PI-b-PS-b-PFS triblock terpolymerblended with 15% PS homopolymer by the use of a substratebrush chemistry that differs from the three blocks of the triblockterpolymer, combined with solvent annealing to produce aspecific swelling ratio of the film. Moreover, topographicaltemplates consisting of rectangular walls grafted with a PFSbrush could produce “single grain” square symmetry patternsinside the templates, while post arrays were able to accuratelyalign and register the arrays. These highly ordered square arrayscan be useful as an etch mask for other functional materials,facilitating the fabrication of devices based on a Cartesian layout,such as integrated circuits.Methods. Substrate and Template Preparation. To graft the

brushes onto the oxidized Si substrates, hydroxyl-terminatedhomopolymers, including PS-OH (Mw ∼ 3 kg/mol, PolymerSource), PB-OH (Mw ∼ 5 kg/mol, Polymer Source), PFS-OH(Mw ∼ 5 kg/mol, synthesized) solutions in toluene, P2VP-OH(Mw ∼ 6 kg/mol, Polymer Source) solution in DMF, PEO-OH(Mw ∼ 5 kg/mol, Polymer Source) solution in chloroform werespin-coated on the substrate and annealed at 170 �C overnight invacuum. The substrates were then rinsed using the same spin-coating solvents to remove physically adsorbed polymer mol-ecules. The templates were fabricated using electron beampatterning of hydrogen silsesquioxane (HSQ), a negative-toneelectron beam resist. HSQ films (FOx 2% solids from DowCorning) were spin-coated on silicon substrates. Single-pixeldots and line patterns were exposed in a Raith 150 electron-beamlithography tool at 30 kV acceleration voltage. The samples weredeveloped in a 0.25 M NaOH/0.7 M NaCl in distilled water toremove unexposed resist and to reveal the topographical nanos-tructures. To obtain preferential wetting on the sidewalls andnonpreferential wetting on the horizontal surfaces of substrate,the patterned substrates were spin-coated with hydroxyl-termi-nated PFS brush and thermally treated at 170 �C overnight invacuum. A short (∼2 s) oxygen reactive ion etching (RIE)treatment was performed to anisotropically etch the brush layer.Then the hydroxyl-terminated PEO brush was spin-coated andthermally treated at 170 �C in vacuum.Self-Assembly of Triblock Terpolymer. We synthesized poly-

isoprene-b-polystyrene-b-polyferrocenylsilane (PI-b-PS-b-PFS)

Figure 5. (a) Schematic and (b,c) SEM images of highly ordered squarearrays formed on a sparse 2D lattice of HSQ (b) 88 nm� 132 nm singleposts and (c) 44 nm spacing double posts (brighter dots). The sub-strate was functionalized with a PEO brush layer and the post sidewallswere coated with PFS brush layer.


Nano Letters LETTER

with volume fractions of 25, 65, and 10%, respectively. The totalmolecular weight of this triblock terpolymer was approximately82 kg/mol. The details of PI-b-PS-b-PFS triblock terpolymersynthesis are given in a previous article.10 The 1 wt % of the triblockterpolymer was blended with 15 wt % of PS homopolymer(Mn∼ 27 kg/mol, PDI∼ 1.05, Polymer Source Inc.) solution intoluene. This blended solution was spin-coated on various brushcoated Si substrates with thickness of approximately 32 nm.Solvent annealing with controlled vapor pressure was executedusing 1 mL of chloroform in a glass chamber with a leak valveuntil the solvent was fully evaporated. For in situ film thicknessmeasurement, a Filmetrics F20-UV (Filmetrics Inc.) instrumentwas used. Oxygen reactive ion etching was carried out to removethe PI and PS microdomains leaving square-symmetry arrays ofoxidized PFS microdomains on the substrate.Characterization. To observe the morphology of the films at

the film�substrate interface, a Si wafer with 50 nm thick SiO2

layer was used as the substrate. The films were partially immersedin 5 wt % HF aqueous solution to dissolve the SiO2 sacrificiallayer and release the polymer film, which was picked up onanother substrate to expose the bottom interface of the film. Themorphology of the patterns was observed by field emission-scanning electronmicroscope (FE-SEM, Zeiss/LeoGemini 982)operated at 5 kV and AFM (Digital Instrument, Nanoscope IIIA)in tapping mode. The samples for FE-SEM were coated with athin Au�Pd alloy film in order to avoid charging effects. Averagegrain sizes of square patterns were calculated as the square root ofthe average grain area, calculated by image analysis from sets of 5SEM images (4 μm � 3 μm).

’ASSOCIATED CONTENT

bS Supporting Information. Degree of swelling of homo-polymers compared to triblock blend films during solventannealing with controlled vapor pressure and additional SEMimages showing effects of template heights on the long-rangeorder of square arrays in self-assembled triblock terpolymer films.This material is available free of charge via the Internet at http://pubs.acs.org.

’AUTHOR INFORMATION

Corresponding Author*E-mail: [email protected].

’ACKNOWLEDGMENT

We gratefully acknowledge financial support from the Na-tional Science Foundation, the Semiconductor Research Cor-poration, the UCLA FENACenter, the Office of Naval Research,and the EPSRC. The Research Laboratory of Electronics Scan-ning-Electron-Beam Lithography Facility provided facilities forthis work.

’REFERENCES

(1) Lazzari, M.; Lopez-Quintela, M. A. Adv. Mater. 2003, 15 (19),1583–1594.(2) Park, M.; Harrison, C.; Chaikin, P. M.; Register, R. A.; Adamson,

D. H. Science 1997, 276 (5317), 1401–1404.(3) Segalman, R. A.; Yokoyama, H.; Kramer, E. J. Adv. Mater. 2001,

13 (15), 1152–þ.

(4) Cheng, J. Y.; Ross, C. A.; Smith, H. I.; Thomas, E. L. Adv. Mater.2006, 18 (19), 2505–2521.

(5) Son, J. G.; Bulliard, X.; Kang, H. M.; Nealey, P. F.; Char, K. Adv.Mater. 2008, 20 (19), 3643–þ.

(6) Stadler, R.; Auschra, C.; Beckmann, J.; Krappe, U.; Voight-Martin, I.; Leibler, L. Macromolecules 1995, 28 (9), 3080–3097.

(7) Zheng, W.; Wang, Z.-G. Macromolecules 1995, 28 (21), 7215–7223.

(8) Ludwigs, S.; Boker, A.; Voronov, A.; Rehse, N.; Magerle, R.;Krausch, G. Nat. Mater. 2003, 2 (11), 744–747.

(9) Chuang, V. P.; Ross, C. A.; Gwyther, J.; Manners, I. Adv. Mater.2009, 21 (37), 3789–3793.

(10) Chuang, V. P.; Gwyther, J.;Mickiewicz, R. A.;Manners, I.; Ross,C. A. Nano Lett. 2009, 9 (12), 4364–4369.

(11) Ruiz, R.; Dobisz, E.; Albrecht, T. R. ACS Nano 2010, 5 (1),79–84.

(12) Park, S. M.; Craig, G. S.W.; La, Y. H.; Solak, H. H.; Nealey, P. F.Macromolecules 2007, 40 (14), 5084–5094.

(13) Tang, C. B.; Bang, J.; Stein, G. E.; Fredrickson, G. H.; Hawker,C. J.; Kramer, E. J.; Sprung, M.; Wang, J.Macromolecules 2008, 41 (12),4328–4339.

(14) Tang, C. B.; Lennon, E. M.; Fredrickson, G. H.; Kramer, E. J.;Hawker, C. J. Science 2008, 322 (5900), 429–432.

(15) Tang, C. B.; Hur, S. M.; Stahl, B. C.; Sivanandan, K.; Dimitriou,M.; Pressly, E.; Fredrickson, G. H.; Kramer, E. J.; Hawker, C. J.Macromolecules 2010, 43 (6), 2880–2889.

(16) Hur, S. M.; Garcia-Cervera, C. J.; Kramer, E. J.; Fredrickson,G. H. Macromolecules 2009, 42 (15), 5861–5872.

(17) Harrison, C.; Chaikin, P. M.; Huse, D. A.; Register, R. A.;Adamson, D. H.; Daniel, A.; Huang, E.; Mansky, P.; Russell, T. P.;Hawker, C. J.; Egolf, D. A.; Melnikov, I. V.; Bodenschatz, E. Macro-molecules 2000, 33 (3), 857–865.

(18) Kim, S. O.; Solak, H. H.; Stoykovich, M. P.; Ferrier, N. J.; dePablo, J. J.; Nealey, P. F. Nature 2003, 424 (6947), 411–414.

(19) Stoykovich, M. P.; Muller, M.; Kim, S. O.; Solak, H. H.;Edwards, E. W.; de Pablo, J. J.; Nealey, P. F. Science 2005, 308 (5727),1442–1446.

(20) Ruiz, R.; Kang, H. M.; Detcheverry, F. A.; Dobisz, E.; Kercher,D. S.; Albrecht, T. R.; de Pablo, J. J.; Nealey, P. F. Science 2008, 321(5891), 936–939.

(21) Jeong, S. J.; Kim, J. E.;Moon, H. S.; Kim, B. H.; Kim, S.M.; Kim,J. B.; Kim, S. O. Nano Lett. 2009, 9 (6), 2300–2305.

(22) Park, S.; Lee, D. H.; Xu, J.; Kim, B.; Hong, S. W.; Jeong, U.; Xu,T.; Russell, T. P. Science 2009, 323 (5917), 1030–1033.

(23) Jung, Y. S.; Chang, J. B.; Verploegen, E.; Berggren, K. K.; Ross,C. A. Nano Lett. 2010, 10 (3), 1000–1005.

(24) Son, J. G.; Hannon, A. F.; Gotrik, K. W.; Alexander-Katz, A.;Ross, C. A. Adv. Mater. 2011, 23 (5), 634–639.

(25) Bita, I.; Yang, J. K. W.; Jung, Y. S.; Ross, C. A.; Thomas, E. L.;Berggren, K. K. Science 2008, 321 (5891), 939–943.

(26) Yang, J. K. W.; Jung, Y. S.; Chang, J. B.; Mickiewicz, R. A.;Alexander-Katz, A.; Ross, C. A.; Berggren, K. K.Nat. Nanotechnol. 2010,5 (4), 256–260.

(27) Kulbaba, K.; MacLachlan, M. J.; Evans, C. E. B.; Manners, I.Macromol. Chem. Phys. 2001, 202 (9), 1768–1775.

(28) Polymer Handbook, 4th ed.; Brandrup, J. I., Immergut, E. H.,Grulke, E. A., Grulke, A., Eds.; Wiley: New York, 1999.

(29) Helfand, E.; Tagami, Y. J. Chem. Phys. 1972, 56 (7), 3592–&.(30) Paik, M. Y.; Bosworth, J. K.; Smilges, D. M.; Schwartz, E. L.;

Andre, X.; Ober, C. K. Macromolecules 2010, 43 (9), 4253–4260.(31) Jung, Y. S.; Ross, C. A. Adv. Mater. 2009, 21 (24), 2540–.(32) Kim, S. H.; Misner, M. J.; Xu, T.; Kimura, M.; Russell, T. P. Adv.

Mater. 2004, 16 (3), 226–.(33) Fasolka, M. J.; Banerjee, P.; Mayes, A. M.; Pickett, G.; Balazs,

A. C. Macromolecules 2000, 33 (15), 5702–5712.(34) Morkved, T. L.; Jaeger, H. M. Europhys. Lett. 1997, 40 (6),

643–648.(35) Jung, Y. S.; Ross, C. A. Nano Lett. 2007, 7 (7), 2046–2050.


Nano Letters LETTER

(36) Lammertink, R. G. H.; Hempenius, M. A.; Vancso, G. J.; Shin,K.; Rafailovich, M. H.; Sokolov, J. Macromolecules 2001, 34 (7),2394–2394.(37) Son, J. G.; Bae, W. K.; Kang, H. M.; Nealey, P. F.; Char, K. ACS

Nano 2009, 3 (12), 3927–3934.(38) Park, S. M.; Stoykovich, M. P.; Ruiz, R.; Zhang, Y.; Black, C. T.;

Nealey, P. E. Adv. Mater. 2007, 19 (4), 607–611.

highly ordered square arrays from a templated abc triblock terpolymer

Documents