Article pubs.acs.org/Macromolecules
Cylinder Orientation and Shear Alignment in Thin Films of Polystyrene−Poly(n‑hexyl methacrylate) Diblock Copolymers Raleigh L. Davis,† Paul M. Chaikin,‡ and Richard A. Register*,† †
Department of Chemical and Biological Engineering and Princeton Institute for the Science and Technology of Materials, Princeton University, Princeton, New Jersey 08544, United States ‡ Department of Physics, New York University, New York, New York 10003, United States S Supporting Information *
ABSTRACT: Thin films of cylinder-forming polystyrene−poly(n-hexyl methacrylate) diblock copolymers, PS−PHMA, are attractive as nanolithographic templates, an application which demands good control over the thin-film structureyet the quality of order achieved to date in PS−PHMA films has fallen short of this ideal. In the present work, a series of PS−PHMA diblocks of varying composition, all forming PS cylinders, are synthesized and their morphology studied as a function of film thickness in both nonsheared and shear-aligned films. In nonsheared films, the cylinder axis orientation relative to the surface switches from parallel to perpendicular as a function of film thickness; this oscillation is damped out as the fraction of the cylinder-forming block (PS) increases, away from the sphere−cylinder phase boundary. In aligned films, thicknesses which possess the highest coverage of parallel cylinders prior to shear show the highest quality of alignment postshear, as measured by the in-plane orientational order parameter of the cylinder axis. In well-aligned samples of optimal thickness, the quality of alignment is limited by isolated dislocations, whose density is highest at high PS contents, and by undulations in the cylinders’ trajectories, whose impact is most severe at low PS contents; consequently, polymers whose compositions lie in the middle of the cylinder-forming region exhibit the highest quality of alignment.
■
nanolithography. When aligned via shear, thin films of PS− PHMA have been used to create nanowire polarizing grids for deep ultraviolet light.29,30 A primary advantage of PS−PHMA for this application is that the two blocks etch at sufficiently different rates to permit an all-dry etching process during pattern transfer; this etch contrast can be further enhanced through heavy metal staining.30 A major challenge, however, has been the relatively poor quality of alignment achieved in PS−PHMA thin films compared with other cylinder-forming polymers.30,31 This is in part because the thin film morphology of PS−PHMA, both before and after shear, often consists of mixed patterns of in-plane cylinders and hexagonally packed dots, which reflect either cylinders which have transformed into discrete spherical microdomains32 or cylinders which have reoriented to be normal to the substrate.33 Because of PHMA’s34 low surface energy (as compared to PS35) and the generally attractive interactions between poly(alkyl methacrylates) and silica,36 when cast as a thin film onto a substrate with a silica surface, the PHMA matrix is expected to wet both interfaces, thus prompting in-plane cylinder formation at film thicknesses which correspond to approximately integer multiples of the bulk interplanar spacing. When at an incommensurate thickness, however, the cylinders would need to be strained from their equilibrium dimensions to maintain a
INTRODUCTION Currently there is enormous interest in the use of block copolymer thin films as patterning templates for production of materials with nanoscale features.1−5 This approach leverages block copolymers’ capacity to form a variety of periodic structures, such as alternating lamellae, hexagonally packed cylinders, or body-centered-cubic spheres, on size scales of 10− 100 nm.6,7 In bulk block copolymers, the size and type of features produced are primarily dependent on f, the block volume fraction, and χN, where χ is the Flory−Huggins interaction parameter and N is the degree of polymerization. When confined to a thin film, however, additional factors such as surface interactions and commensurability effects, caused by discrepancies between the preferred microdomain spacing and the film thickness, can heavily influence the morphology and the microdomain orientation.8−10 For most nanolithographic applications, the microdomains must have well-defined orientational or positional order. A variety of techniques have been developed to impart such order in both the in-plane and out-ofplane directions.11 One technique in particular, shear alignment, can align microdomains both in the bulk12−14 and in confined systems.15 In thin films, application of a shear stress at the film surface promotes in-plane orientational order in the direction of applied shear; this has previously been shown to align certain sphere-,16−19 cylinder-,15,19−27 and lamellaforming28 block copolymer thin films. Poly(styrene)−poly(n-hexyl methacrylate) block copolymers (PS−PHMA) forming PS cylinders are of particular interest for © 2014 American Chemical Society
Received: June 19, 2014 Revised: July 23, 2014 Published: August 1, 2014 5277
dx.doi.org/10.1021/ma5012705 | Macromolecules 2014, 47, 5277−5285
Macromolecules
Article
Table 1. Summary of PS−PHMAs Used in This Study diblock Mn [kg/mol] PS−PHMA
wPS
PS Mn [kg/mol]
PS Đ
method 1
method 2
diblock Đ
dbulk [nm]
morphology
−15 −21 −26 −30 −35 −40
0.15 0.21 0.26 0.30 0.35 0.40
18.2 24.2 31.2 32.6 41.5 46.2
1.15 1.05 1.15 1.13 1.10 1.08
120 113 121 109 119 116
113 99 104 111 136 124
1.08 1.10 1.08 1.06 1.09 1.10
35.2 32.2 38.3 39.8 47.0 51.5
spheres cylinders cylinders cylinders cylinders perforated lamellae
described elsewhere;42 the other five PS−PHMA diblocks were synthesized using the same chemistry in different apparatus as described below. Tetrahydrofuran (THF) and methanol were purchased from Fisher Scientific. Both were degassed, and the THF was dried over the purple sodium−benzophenone complex for several days prior to use. S, HMA, and diphenylethylene (DPE) were purchased from Sigma-Aldrich. S and DPE were vacuum transferred from dibutylmagnesium after several freeze−pump−thaw cycles, while HMA was titrated with trioctylaluminum, turning the monomer a green-yellow color,43 prior to vacuum transfer. Trialkylaluminums are known to initiate free-radical polymerization of alkyl methacrylates in the presence of oxygen, so the HMA was degassed via freeze−pump− thaw cycles prior to titration.44 These three reagents were all distilled into evacuated flasks and either used immediately for polymerization or stored in a glovebox freezer at −20 °C. sec-Butyllithium and anhydrous LiCl were purchased from Sigma-Aldrich and used as received. The reactions were performed at ∼10 wt % solids inside an MBraun UNILab glovebox with N2 atmosphere (O2 and H2O concentrations 0.99. Error bars represent ±1 standard deviation of F from three or more images taken at the same film thickness.
the influence of polymer composition on in- vs out-of-plane orientation. With increased PS fraction, the peak centered around t/d ∼ 1 begins to broaden while simultaneously the valley at t/d ∼ 1.25 starts to vanish. Furthermore, only the diblock with the lowest wPS possesses a local valley between 2 < t/d < 3. For wPS = 0.30 and above, the switching has completely vanished, yielding films whose orientation is nearly independent of film thickness for t/d ≳ 1. This result indicates that as the polymer’s composition is shifted more firmly into the cylinderforming regions of the phase diagram, it becomes more energetically unfavorable for the cylindrical microdomains to reorient out-of-plane; this is likely due to the increased energy required to form the high-microdomain-curvature “capping
Figure 4. TM-AFM phase images of shear-aligned PS−PHMA-21 at thicknesses of (a) 20, (b) 30, (c) 44, and (d) 120 nm. Arrows indicate shear direction in each panel. Scale bar = 500 nm. (e) Plot of orientational order parameter, ψ6, as a function of normalized film thickness, t/d, for PS− PHMA-21 (red squares connected by red lines to guide the eye). Error bars represent ±1 standard deviation from three or more images taken at the same film thickness. For comparison, the F vs t/d data (black line) for this polymer (from Figure 2) are also shown. 5281
dx.doi.org/10.1021/ma5012705 | Macromolecules 2014, 47, 5277−5285
Macromolecules
Article
Figure 5. Plots of orientational order parameter, ψ6, and fractional coverage of lines, F, as a function of normalized film thickness, t/d, for PS−PHMA (a) -26, (b) -30, and (c) -35. Error bars represent ±1 standard deviation from three or more images taken at the same film thickness. Representative images for nonsheared and sheared films at t/d ∼ 1: (d) nonsheared and (e) sheared PS−PHMA-26, (f) nonsheared and (g) sheared PS−PHMA-30, (h) nonsheared and (i) sheared PS−PHMA-35. Scale bar = 500 nm; arrows indicate direction of applied shear in each panel.
for which all-dot patterns were present in the absence of shear, once sheared there is a partial dot-to-line transition (see Figure 4a) in which it appears that the dots are aligning in the direction of applied shear and in some cases elongating and connecting to form continuous cylindrical microdomains. This behavior is consistent with previous experimental52 and simulation53 results for similar block copolymer thin film systems. For 1 < t/d < 3, the local maxima in ψ6 correspond to well-aligned films containing integer numbers of layers of cylinders. In this thickness range, films of noninteger thicknesses show either a large number of dots or a very high defect density in all-line patterns. At t/d > 3 all images show well-aligned in-plane cylinders whose alignment quality is approximately independent of film thickness. Analogous studies were performed on the other cylinderforming PS−PHMA diblocks. PS−PHMA-26 displayed a qualitatively different relationship between ψ6 and t/d; unlike PS−PHMA-21, which possessed a strong oscillation in both F and ψ6, PS−PHMA-26 showed almost no dependence of ψ6 on t/d > 1. All thicknesses which showed a significant fraction of dots preshear show no dots postshear. This is again consistent with the notion that as wPS is increased, away from the sphere− cylinder phase boundary, then out-of-plane orientation (with its associated capping layers) becomes correspondingly less stable, and the cylinders more readily lie in-plane. The two highest-wPS polymers, PS−PHMA-30 and -35, show almost no dependence of ψ6 on thickness above t/d ∼ 0.8. This is perhaps unsurprising given that these polymers possessed nearly complete coverage of in-plane cylinders (F ∼ 1) preshear. To further compare the alignment behavior of the four polymers, Figure 6 shows alignment quality as a function of composition for both monolayer and four-layer films. For monolayer films, there is a monotonic increase in ψ6 with increased wPS; however, for the thicker films, there is no obvious trend with composition. This suggests that the compositional effects on alignment quality are magnified in the monolayer films desired for nanopatterning. Limitations on Alignment Quality. To investigate the maximum orientational order achievable by shear alignment in these systems, the four shear-aligned polymers were imaged at
Figure 6. Alignment quality vs wPS for films with t/d = 1 (red squares) and t/d = 4 (blue diamonds). Each point represents the average value from 10 AFM micrographs; error bars represent ±1 standard deviation.
integer values of t/d. At these thicknesses, all of the polymers displayed a high degree of alignment with complete coverage of in-plane cylinders (no dots) and relatively few isolated dislocations, which were the only lattice defects observed; a high-magnification AFM image of a typical dislocation is shown in Figure 7b. Because exclusively all-line patterns were present, we characterized the alignment in each image via ψ2, rather than ψ6, to allow for quantitative comparison with previous work on other cylinder-forming block copolymers.21 For such wellaligned samples, the measured quality of order is primarily limited by two factors: undulations in the cylinders’ trajectories and the frequency of dislocations. To determine to what degree these dislocations diminish the order in films of each polymer, the density of dislocations for each image was computed and normalized by the square of the cylinder periodicity (pAFM). These results are plotted in Figure 7a, where each point reflects a single AFM micrograph. Compared with previous work on cylinder-forming PS−PEP,21 the normalized dislocation densities for well-aligned PS−PHMA diblocks are relatively 5282
dx.doi.org/10.1021/ma5012705 | Macromolecules 2014, 47, 5277−5285
Macromolecules
Article
Figure 7. (a) ψ2 order parameter vs normalized dislocation density for monolayers (open symbols) and multilayers (closed symbols, up to 4 layers) of PS−PHMA-21 (red squares), -26 (orange diamonds), -30 (green triangles), and -35 (blue circles). Red line shows best-fit to all PS−PHMA-21 data (mono- and multilayer), while the black line shows the analogous best fit of the combined data for PS−PHMA-26, -30, and -35. (b) Highmagnification AFM of a typical dislocation (PS−PHMA-21, t = 30 nm). Representative micrographs of aligned monolayer films of (c) PS−PHMA21, (d) -26, (e) -30, and (f) -35 show the relative strength of undulations present in well-aligned films. Scale bars indicate 200 nm.
Table 2. Slopes and y-Intercepts from Fits of ψ2 vs Normalized Dislocation Densitya monolayer polymer PS−PHMA-21 PS−PHMA-26 PS−PHMA-30 PS−PHMA-35 PS−PHMA-26, -30, -35 simulations21 a
multilayer y-intercept
slope −3.4 −1.1 −1.1 −1.1 −1.2 −1.3
± ± ± ± ± ±
1.0 0.2 0.2 0.3 0.2 0.1
0.9947 0.9993 0.9992 0.9986 0.9990
± ± ± ± ±
0.0020 0.0004 0.0007 0.0020 0.0006
y-intercept
slope −4.1 −0.9 −0.8 −1.0 −0.9
± ± ± ± ±
3.0 0.2 0.3 0.4 0.2
0.9946 0.9987 0.9986 0.9995 0.9988
± ± ± ± ±
0.0010 0.0005 0.0008 0.0019 0.0005
All ± indicate one standard deviation.
high (10−2) were observed in PS−PHMA-35 films. It is interesting that the two polymers at the edges of the cylinder-forming composition window show the poorest quality of alignment (ψ2), but for different reasons: PS−PHMA-21 because of pronounced undulations (despite a relatively low defect density) and PS−PHMA-35 because of a relatively high defect density (despite minimal undulations). Thus, both of these disturbances must be considered to maximize the quality of order in the film, which is obtained for compositions in the middle of the cylinder-forming region.
polymers whose compositions lie in the middle of the cylinderforming region.
■
ASSOCIATED CONTENT
S Supporting Information *
Plot of dbulk against Mn determined by methods 1 and 2; AFM images of PS−PHMA-15 and -40; details concerning image filtering, calculation of the orientational order parameter, and the dislocation-finding algorithm; a representative larger-area AFM image containing three dislocations; and AFM images showing the lack of effect of quiescent annealing on the cylinder undulations. This material is available free of charge via the Internet at http://pubs.acs.org.
■
AUTHOR INFORMATION
Corresponding Author
*Tel +1 609 258 4691; fax +1 609 258 0211; e-mail register@ princeton.edu (R.A.R.). Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS The authors gratefully acknowledge Brian Michal for the synthesis of PS−PHMA-21, Aditya Koppikar for dn/dc measurements on PS and PHMA solutions, and Arash Nikoubashman and Athanassios Panagiotopoulos for helpful discussions. This work was generously supported by the National Science Foundation, MRSEC program, through the Princeton Center for Complex Materials (DMR-0819860).
■
REFERENCES
(1) Gu, X.; Gunkel, I.; Russell, T. P. Philos. Trans. R. Soc., A 2013, 370, 20120306. (2) Kim, J. K.; Yang, S. Y.; Lee, Y.; Kim, Y. Prog. Polym. Sci. 2010, 35, 1325−1349. (3) Kim, H.-C.; Park, S.-M.; Hinsberg, W. D. Chem. Rev. 2010, 110, 146−177. (4) Wang, J.-Y.; Chen, W.; Russell, T. P. Patterning with Block Copolymers. In Unconventional Nanopatterning Techniques and Applications; Rogers, J. A., Lee, H. H., Eds.; John Wiley & Sons: Hoboken, NJ, 2009; pp 233−289. (5) Darling, S. B. Prog. Polym. Sci. 2007, 32, 1152−1204. (6) Bates, F. S. Science 1991, 22, 898−905. (7) Hamley, I. W. The Physics of Block Copolymers; Oxford University Press: Oxford, 1998. (8) Hamley, I. W. Prog. Polym. Sci. 2009, 34, 1161−1210. (9) Russell, T. P.; Coulon, G.; Deline, V. R.; Miller, D. C. Macromolecules 1989, 22, 4600−4606. (10) Luo, M.; Seppala, J. E.; Albert, J. N. L.; Lewis, R. L., III; Mahadevapuram, N.; Stein, G. E.; Epps, T. H., III Macromolecules 2013, 46, 1803−1811. (11) Marencic, A. P.; Register, R. A. Annu. Rev. Chem. Biomol. Eng. 2010, 1, 277−297. (12) Keller, A.; Pedemonte, E.; Willmouth, F. M. Nature 1970, 225, 538. (13) Hadziioannou, G.; Picot, C.; Skoulios, A.; Ionescu, M. L.; Mathis, A.; Duplessix, R.; Gallot, Y.; Lingelser, J. P. Macromolecules 1982, 15, 263−267. (14) Morrison, F. A.; Mays, J. W.; Muthukumar, M.; Nakatani, A. I.; Han, C. C. Macromolecules 1993, 26, 5271−5273. (15) Angelescu, D. E.; Waller, J. H.; Adamson, D. H.; Deshpande, P.; Cho, S. Y.; Register, R. A.; Chaikin, P. M. Adv. Mater. 2004, 16, 1736− 1740. (16) Angelescu, D. E.; Waller, J. H.; Register, R. A.; Chaikin, P. M. Adv. Mater. 2005, 17, 1878−1881.
■
CONCLUSIONS The thin-film structure of a series of PS−PHMA diblock copolymers, with compositions spanning the cylinder-forming region of the phase diagram, was investigated. For incommensurate film thicknesses, these polymers do not terrace, but diblocks of lower PS weight fraction (wPS) showed continuous height variations with an amplitude corresponding to roughly half a cylinder layer spacing (“pseudo-terracing”). The PS− PHMA diblock with the lowest PS content showed the most dramatic effect of film thickness on morphology: nonsheared films of incommensurate thickness formed perpendicular cylinders rather than the usual parallel arrangement and sheared films of incommensurate thickness showed poorly ordered, highly defective patterns. However, for diblocks with higher weight fractions of PS, these effects were increasingly mitigated, such that the polymers with highest PS content showed essentially the same film structure by AFM regardless of film thickness. Alignment quality after shear was highly correlated with the fractional coverage by in-plane cylinders prior to shearing: both were lower for incommensurate film thicknesses, especially at low wPS. Alignment quality in diblock films is limited by two factors: the presence of isolated dislocations in the microdomain lattice and undulations in the cylinder trajectories. In PS−PHMA diblock thin films forming PS cylinders, the dislocation density increases with PS content, while the undulation magnitude decreases with PS content, such that the highest quality of alignment is obtained for 5284
dx.doi.org/10.1021/ma5012705 | Macromolecules 2014, 47, 5277−5285
Macromolecules
Article
(17) Wu, M. W.; Register, R. A.; Chaikin, P. M. Phys. Rev. E 2006, 74, 040801R. (18) Marencic, A. P.; Wu, M. W.; Register, R. A.; Chaikin, P. M. Macromolecules 2007, 40, 7299−7305. (19) Marencic, A. P.; Adamson, D. H.; Chaikin, P. M.; Register, R. A. Phys. Rev. E 2010, 81, 011503. (20) Pelletier, V.; Adamson, D. H.; Register, R. A.; Chaikin, P. M. Appl. Phys. Lett. 2007, 90, 163105. (21) Marencic, A. P.; Chaikin, P. M.; Register, R. A. Phys. Rev. E 2012, 86, 021507. (22) Kim, S. Y.; Gwyther, J.; Manners, I.; Chaikin, P. M.; Register, R. A. Adv. Mater. 2014, 26, 791−795. (23) Kwon, H. K.; Lopez, V. E.; Davis, R. L.; Kim, S. Y.; Burns, A. B.; Register, R. A. Polymer 2014, 55, 2059−2067. (24) Singh, G.; Yager, K. G.; Berry, B.; Kim, H. C.; Karim, A. ACS Nano 2012, 6, 10335−10342. (25) Ye, C. H.; Singh, G.; Wadley, M. L.; Karim, A.; Cavicchi, K. A.; Vogt, B. D. Macromolecules 2013, 46, 8608−8615. (26) Qiang, Z.; Zhang, L. H.; Stein, G. E.; Cavicchi, K. A.; Vogt, B. D. Macromolecules 2014, 47, 1109−1116. (27) Jeong, J. W.; Hur, Y. H.; Kim, H.-J.; Kim, J. M.; Park, W. I.; Kim, M. J.; Kim, B.; Jung, Y. S. ACS Nano 2013, 7, 6747−6757. (28) Pujari, S.; Keaton, M. A.; Chaikin, P. M.; Register, R. A. Soft Matter 2012, 8, 5358−5363. (29) Pelletier, V.; Asakawa, K.; Wu, M. W.; Adamson, D. H.; Register, R. A.; Chaikin, P. M. Appl. Phys. Lett. 2006, 88, 211114. (30) Papalia, J. M.; Adamson, D. H.; Chaikin, P. A.; Register, R. A. J. Appl. Phys. 2010, 107, 084305. (31) Marencic, A. P. Well-Ordered Block Copolymer Thin Films Using Shear-Alignment Techniques. Ph.D. Thesis, Princeton University. 2011. (32) Niihara, K.-I.; Sugimori, H.; Matsuwaki, U.; Hitaro, F.; Morita, H.; Doi, M.; Masunaga, H.; Sasaki, S.; Jinnai, H. Macromolecules 2008, 41, 9318−9325. (33) Nikoubashman, A.; Register, R. A.; Panagiotopoulos, A. Z. Macromolecules 2013, 46, 6651−6658. (34) Nielsen, B. V.; Nevell, T. G.; Barbu, E.; Smith, J. R.; Rees, G. D.; Tsibouklis, J. Biomed. Mater. 2011, 6, 015003. (35) Wu, S. J. Phys. Chem. 1970, 74, 632−638. (36) Keddie, J. L.; Jones, R. A. L.; Cory, R. A. Faraday Discuss. 1994, 98, 219−230. (37) Coulon, G.; Ausserre, D.; Russell, T. P. J. Phys. (Paris) 1990, 51, 777−786. (38) Coulon, G.; Collin, B.; Ausserre, D.; Chatenay, D.; Russell, T. P. J. Phys. (Paris) 1990, 51, 2801−2811. (39) Harrison, C.; Park, M.; Chaikin, P. M.; Register, R. A.; Adamson, D. H.; Yao, N. Macromolecules 1998, 31, 2185−2189. (40) Knoll, A.; Horvat, A.; Lyakhova, K. S.; Krausch, G.; Sevink, G. J. A.; Zvelindovsky, A. V.; Magerle, R. Phys. Rev. Lett. 2002, 89, 035501. (41) Zhang, X.; Berry, B. C.; Yager, K. G.; Kim, S.; Jones, R. L.; Satija, S.; Pickel, D. L.; Douglas, J. F.; Karim, A. ACS Nano 2008, 2, 2331− 2341. (42) Nikoubashman, A.; Davis, R. L.; Michal, B. T.; Chaikin, P. M.; Register, R. A.; Panagiotopoulos, A. Z. ACS Nano 2014, Submitted. (43) Allen, P. E. M.; Bateup, B. O.; Casey, B. A. J. Org. Chem. 1971, 29, 185−193. (44) Allen, R. D.; Long, T. E.; McGrath, J. E. Polym. Bull. 1986, 15, 127−134. (45) Varshney, S. K.; Hautekeer, J. P.; Fayt, R.; Jérôme, R.; Teyssié, P. Macromolecules 1990, 23, 2618−2622. (46) Wiles, D. M.; Bywater, S. Trans. Faraday Soc. 1965, 61, 150− 158. (47) Bushuk, W.; Benoît, H. Can. J. Chem. 1958, 36, 1616−1626. (48) Register, R. A.; Bell, T. R. J. Polym. Sci., Part B: Polym. Phys. 1992, 30, 569−575. (49) Stafford, C. M.; Roskov, K. E.; Epps, T. H., III; Fasolka, M. J. Rev. Sci. Instrum. 2006, 77, 023908. (50) Davis, R. L.; Jayaraman, S.; Chaikin, P. M.; Register, R. A. Langmuir 2014, 30, 5637−5644.
(51) Schneider, C. A.; Rasvand, W. S.; Eliceiri, K. W. Nat. Methods 2012, 9, 671−675. (52) Hong, Y.-R.; Adamson, D. H.; Chaikin, P. M.; Register, R. A. Soft Matter 2009, 5, 1687−1691. (53) Chremos, A.; Chaikin, P. M.; Register, R. A.; Panagiotopoulos, A. Z. Macromolecules 2012, 45, 4406−4415.
5285
dx.doi.org/10.1021/ma5012705 | Macromolecules 2014, 47, 5277−5285
