Langmuir 2008, 24, 355-358
355
Patchy Particles by Glancing Angle Deposition Amar B. Pawar and Ilona Kretzschmar* Department of Chemical Engineering, The City College of New York, 140th Street & ConVent AVenue, New York, New York 10031 ReceiVed September 28, 2007. In Final Form: NoVember 12, 2007 The application of glancing angle deposition (GLAD) as a means to produce patchy particles is reported. Shadow effects are caused by neighboring particles within the particle monolayer. The patch geometry is determined by the angle of incidence of the vapor rays and the monolayer orientation. A mathematical model is used to study the patch geometry and to calculate the area of the patch. The smallest patch produced with GLAD is 3.7% of the particle surface.
Introduction Recently, much attention has been given to understanding the assembly of functionally anisotropic building blocks using a molecular simulation approach.1-5 The increased interest has led to the generation of many anisotropic building blocks, and a new unifying nomenclature has been proposed.6 Janus particles, particles with positively and negatively charged halves, present the simplest of the new building blocks with anisotropy dimension A (surface coverage).6 Vertical metal vapor deposition onto a monolayer of latex particles is a straightforward procedure leading to Janus particles.7,8 Experimentally, the clustering of charged Janus particles has been observed using fluorescence microscopy. In addition, Monte Carlo simulations were employed to verify the observed cluster shapes computationally.9 Janus particles present an important class of building blocks for directional assembly. However, in the context of assembling specific target structures, precise control over particle interactions is required. The two halves of a Janus particle are too large and allow only limited control over the assembly direction. Patchy particles (i.e., particles with more than one patch or patches that are less than 50% of the total particle surface) present the next generation of particles for assembly. In terms of the new nomenclature,6 this refers to moving along anisotropy dimension A (surface coverage) as well as adding anisotropy dimension E (branching). The patches serve as specific interaction sites,2,5 and the formation of various structures has been predicted in simulations.3 However, little attention has been given to the fabrication of patchy particles10-14 as compared to Janus particles.7-8,15-18 Zhang et al.3 have implemented nanosphere * Corresponding author. E-mail:
[email protected]. (1) Glotzer, S. C.; Horsch, M. A.; Iacovella, C. R.; Zhang, Z. L.; Chan, E. R.; Zhang, X. Curr. Opin. Colloid Interface Sci. 2005, 10, 287-295. (2) Zhang, Z.; Keys, A. S.; Chen, T.; Glotzer, S. C. Langmuir 2005, 21, 1154711551. (3) Zhang, Z.; Glotzer, S. C. Nano Lett. 2004, 4, 1407-1413. (4) Chen, T.; Lamm, M. H.; Glotzer, S. C. J. Chem. Phys. 2004, 121, 39193929. (5) Pawar, A. B.; Kretzschmar, I.; Aranovich, G.; Donohue, M. D. J. Phys. Chem. B 2007, 111, 2081-2089. (6) Glotzer, S. C.; Solomon, M. J. Nat. Mater. 2007, 6, 557-562. (7) Takei, H.; Shimizu, N. Langmuir 1997, 13, 1865-1868. (8) Perro, A.; Reculusa, S.; Ravaine, S.; Bourgeat-Lami, E. B.; Duguet, E. J. Mater. Chem. 2005, 15, 3745-3760. (9) Hong, L.; Cacciuto, A.; Luijten, E.; Granick, S. Nano Lett. 2006, 6, 25102514. (10) Zhang, G.; Wang, D.; Mo¨hwald, H. Chem. Mater. 2006, 18, 3985-3992. (11) Zhang, G.; Wang, D.; Mo¨hwald, H. Nano Lett. 2005, 5, 143-146. (12) Zhang, G.; Wang, D.; Mo¨hwald, H. Angew. Chem., Int. Ed. 2005, 44, 7767-7770. (13) Snyder, C. E.; Yake, A. M.; Feick, J. D.; Velegol, D. Langmuir 2005, 21, 4813-4815. (14) Yake, A. M.; Snyder, C. E.; Velegol, D. Langmuir 2007, 23, 9069-9075. (15) Hong, L.; Jiang, S.; Granick, S, Langmuir 2006, 22, 9495-9499.
lithography to pattern microsphere surfaces with multiple patches in highly symmetric arrangements. In this letter, we report the application of glancing angle deposition (GLAD),10,19-21 in which the sample is tilted, to the production of patchy particles using particles within the same monolayer as a shadowing mask. We show that the GLAD technique allows the generation of single patch areas as small as 3.7% of the total particle surface compared to a patch size of 1.3% that is achievable with Velegol’s particle lithography method.13 The effect of the angle of incidence and the monolayer orientation on the geometry of the patch resulting on the particle is determined. Computationally, the patch shape and size are verified using a simple mathematical model. Experimental Details Convective assembly22-24 in an acrylic cell with a circular bore is used to produce 2D arrays of sulfate latex polystyrene (PS) particles through the radial growth of a 2D close-packed monolayer from the center of the cell outward. Within such a 2D array of particles, close-packed domains of colloids with hexagonal symmetry having different monolayer orientations are observed. The average domain size observed is ∼15 000 µm2 (∼3000 particles of 2.4 µm diameter) within a 1-cm-diameter monolayer area. Gold vapor deposition on the 2D close-packed colloidal monolayer is performed inside a benchtop vacuum metal evaporation system (Cressington 308 R, Ted Pella, Inc.) at a pressure of 10-6 mbar. The angle of incidence of Au vapor flow, θ, measured from the substrate (90° being vertical to sample) is adjusted by tilting the sample stage in the vacuum chamber. The effect of the angle of incidence on the patch geometry is studied by varying θ from 30 to 10 and 2°. Following evaporation, the samples are imaged using a scanning electron microscope (EVO40 Zeiss). The patchy particles can be recovered from the silicon wafer substrate in the solution (e.g., water) by sonicating the wafer in water for a few minutes. Currently, we are able to produce 1.7 × 107 patchy particles corresponding to a 1-cm-diameter circular monolayer, but this is not a limit of the method. The details of the experiments and materials used are given in Supporting Information. (16) Cui, J. Q.; Kretzschmar, I. Langmuir 2006, 22, 8281-8284. (17) Paunov, V. N.; Cayre, O. AdV. Mater. 2004, 16, 788-791. (18) Cayre, O.; Paunov, V. N.; Velev, O. D. J. Mater. Chem. 2003, 13, 24452450. (19) Robbie, K.; Sit, J. C.; Bret, M. J. J. Vac. Sci. Technol., B 1998, 16, 1115-1122. (20) Karabacak, T.; Singh, J. P.; Zhao, Y.-P.; Wang, G.-C.; Lu, T.-M. Phys. ReV. B 2003, 68, 125408. (21) Zhao, Y.-P.; Ye, D.-X.; Wang, G.-C.; Lu, T.-M. Nano Lett. 2002, 2, 351-354. (22) Silvera-Batista, C. A.; Kretzschmar, I. Junior Scientist Conference ’06 Vienna University of Technology, Vienna, Austria, 2006. (23) Dimitrov, A. S.; Nagayama, K. Langmuir 1996, 12, 1303-1311. (24) Denkov, N. D.; Velev, O. D.; Kralchevsky, P. A.; Ivanov, I. B.; Yoshimura, H.; Nagayama, K. Langmuir 1992, 8, 3183-3190.
10.1021/la703005z CCC: $40.75 © 2008 American Chemical Society Published on Web 12/13/2007
356 Langmuir, Vol. 24, No. 2, 2008
Letters
Figure 1. Schematic of metal evaporation geometry and monolayer orientation with respect to the incident vapor beam. (A) x-z crosssectional view indicating angle of incidence θ. x-y cross section indicating monolayer orientation (B) at R ) 0° and (C) at R ) 30°. The patch geometry obtained by experiments is verified with that obtained mathematically using a simple geometrical model where the equations of the spheres (representing particles) and the inclined lines (representing incident rays) are solved simultaneously. The details of the geometrical model are described in Supporting Information. Figure 1 represents a schematic view of angled Au vapor deposition on a particle monolayer. The angle θ corresponds to the angle of incidence of the Au vapor beam measured from the substrate side (Figure 1.A). The monolayer orientation is identified by the angle R. For our studies, the reference monolayer orientation is chosen as depicted in Figure 1.B with R ) 0°. A different monolayer orientation can be found through clockwise rotation of the monolayer by R (Figure 1.C, R ) 30°) whereas the source position is fixed on the right side of the monolayer. It should be noted that because of the hexagonal symmetry the monolayer orientation repeats at R ) 60° (i.e., R ) 0 and 60° represent the same monolayer orientations as shown in Figure 1.B but with the particles having changed their positions within the monolayer).
Results and Discussion Vapor deposition at an angle of incidence θ creates patches on the particles depending on θ and the monolayer orientation in the domains, R. Figure 2 shows 2D arrays of 2.4 µm Aupatterned sulfated PS particles at different angles of incidence, θ ) 30, 10, and 2°, and varying monolayer orientation, R, between 0 and 46°. As θ is decreased from 90 to 2°, the shadow effect caused by the neighboring particles within the close-packed monolayer takes effect. The shadow effect determines the size and the shape of the Au patch created on each particle. As the angle θ becomes smaller, the shadow effect becomes stronger, and smaller patches are formed. As shown in Figure 2, the Au patch size diminishes considerably as θ varies from 30 to 10° and further decreases as θ is reduced to 2°. Scanning electron microscopy (SEM) images depict the top view (x-y view with respect to Figure 1) of the PS particle monolayers patterned with gold. The geometry of each patch obtained with the mathematical model is shown below each of the experimental SEM images with the corresponding angles. Experimental patch shapes and mathematically obtained patch shapes show very good agreement. The insets in the bottom panels of Figure 2 show the corresponding images of a single particle and its patch in a side view (y-z view). The images of particles modified with θ larger than 30°
Figure 2. Comparison of experimental (top) and calculated (bottom) gold patches on 2.4 µm sulfated polystyrene particles as a function of the angle of incidence (θ) and monolayer orientation (R). The images show the top view (x-y view) of the monolayer. (A) Patches obtained at an angle of incidence of θ ) 30° and monolayer orientations of R ) 0, 15, 30, and 46° (from left to right). (B) Patches obtained at an angle of incidence of θ ) 10° and monolayer orientations of R ) 0, 18, 29, and 40° (from left to right). (C) Patches obtained at an angle of incidence θ ) 2° and monolayer orientations of R ) 0, 14, 28, and 38° (from left to right). Bottom panel insets: y-z view (i.e., view from source side) of a single patchy particle in the monolayer obtained with the mathematical model. Scale bars in experimental images correspond to 2 µm.
are not reported here because a significant shadow effect is certainly present but not noticeable in the top-view images. The patch shape is a result of the arrangement of neighboring particles and the self-shadowing of the particle. The particle side facing the incidentAu rays is partially shadowed by neighboring particles from the next few rows of particles. The other side of the particle (i.e., the side away from the source) is partially shadowed because of the tangency of the incident rays to the particle itself. The boundary of the patch on the side of the particle facing away from the source (lagging boundary) is thus decided only by the angle of incidence θ, whereas the boundary on the side facing the incident rays (leading boundary) is effected by θ as well as the monolayer orientation R. As illustrated in Figure 2, the effect of monolayer orientation on the geometry of the patch is more pronounced at lower θ (10 and 2°) than at θ ) 30°. The pictures shown as insets indicate that as θ increases from 2 to 30° patches with sharp corners are observed on the particles (θ ) 30°). The sharp corners disappear as θ increases further (θ > 30°, data not shown) because of a diminished shadow effect.
Letters
Langmuir, Vol. 24, No. 2, 2008 357
Figure 4. Patch area in % surface area of the particle plotted as a function of the angle of incidence θ. Inset: Patch area in % surface area of the particle as a function of monolayer orientation (R).
Figure 3. Images of a single patchy particle in a monolayer demonstrating the shadow effect of neighboring particles at an angle of incidence of θ ) 2° and monolayer orientations of (A) R ) 0°, (B) R ) 15°, (C) R ) 30°, and (D) R ) 45° obtained by the mathematical model. The arrow in the bottom right corner indicates the direction of the incident rays. The top panel shows the top view of the monolayer, and the bottom panel depicts the patchy particle as seen from the source position at θ ) 2°. The dark-gray shaded spheres indicate the spheres causing the shadow effect.
Figure 3 shows the images of a single patchy particle within the monolayer demonstrating the shadow effect of neighboring particles at θ ) 2° and R ) 0, 15, 30, and 45° obtained using the mathematical model. The top panel of Figure 3 shows the top view of the monolayer with a single representative patchy particle. The patchy particle is indicated by the particle with the colored patch. The particles with dark gray shading are the particles that cause the shadow effect on the representative patchy particle. The source position is fixed on the right side of the monolayer and the incident ray direction is indicated by the arrow (lower right corner). The bottom panel shows the colored patch on the patchy particle along with the particles from the monolayer causing the shadow effect from the source position at θ ) 2°. When the monolayer orientation is such that R ) 0 and 30°, the patch boundary on the side of the particle facing the source (i.e., the leading boundary) is flattened, which leads to rectangular patches with a curved surface. The flattening of the leading boundaries is due to the fact that these boundaries are caused by particles that are aligned exactly with the particle that they are shadowing (i.e., the exactly aligned particle is particle 8 in Figure 3A (R ) 0°) and particle 5 in Figure 3C (R ) 30°)). At R ) 30°, particles 5, 9, and 13 in the monolayer are aligned (Figure 3C top panel), thus leading to the flat leading boundaries
for all angles of incidence. At the monolayer orientation of R ) 30°, the shadow effect of particles 2 and 4 is less prominent at lower angles of incidence (e.g., θ ) 2°) but with increasing θ the shadow effect of these particles becomes prominent and can be noted in Figure 2A,B with R ) 30 and R ) 29°, respectively. Patchy particles obtained with a monolayer orientation of R ) 0° do not always exhibit a flat leading boundary as seen in Figure 2C for θ ) 2°. As the angle of incidence, θ, increases, the shadow effect of particle 8 decreases and eventually plays no role in shadowing. The decreased shadow effect of particle 8 leads to patches with a sharp corner in the leading boundary (inset of Figure 2A with R ) 0°), which is caused by the combined shadow effect of particles 4 and 5. The leading boundary in the other monolayer orientations (Figure 3B,D) is not flat because of the fact that these boundaries are created by a combination of shadow effects from more than a single particle (e.g., the leading boundary in the case of R ) 15° is a combined shadow effect of particles 5, 8, and 12 within the monolayer). The patch boundary on the side of particles pointing away from the source (i.e., the lagging boundary) is caused by the tangency of the incident Au rays to the particle itself. As illustrated in Figure 3, the location of the lagging boundary is independent of the monolayer orientation R and is always flat. Figure 3A,C shows that the patches are symmetric about the x axis passing through the particle center for monolayer orientations of R ) 0 and 30°. Such patch symmetry is not observed for any other monolayer orientations. Looking at the particles from the source position, the patch geometries are mirror images of each other around R ) 30° (i.e., the patch geometry for R ) 15 and 45° are mirror images of each other). Figure 4 shows the plot of the percentage of the Au-patterned sphere area as a function of the angle of incidence, θ, at different monolayer orientations, R. The area of the patches is calculated from the patch geometries obtained using the mathematical model. We expect the patch areas to be within (0.5% of the indicated values in Figure 4. The small error in the area calculations is expected because the patch geometries are obtained on the basis of a grid of incident vapor rays hitting the particles with a ray spacing of 0.0005 units in x and y (Supporting Information). As already noted in Figure 3, the Au patch area behaves symmetrically around R ) 30° (i.e., the area and geometry of the patches are equal at R ) 20 and 40°, 10 and 50°, and 0 and 60° for any θ). The plot shows that all the monolayer orientations follow a similar
358 Langmuir, Vol. 24, No. 2, 2008
trend of increase in patch area with increase in θ. The increase in patch area flattens out at higher θ. The inset shows a plot of the percentage of the patch area as a function of monolayer orientation, R, at various θ. At lower angles of incidence (e.g., θ ) 2°), the change in the Au patch area is significant with the change in the monolayer orientation. The patch area varies from 3.7 to 5.7% as the monolayer orientation is varied between 0 and 30°. At higher θ, the change in the patch area with the monolayer orientation is negligible ( 30°) lead to patchy particles with similar patch sizes. Within a multidomain monolayer of particles, the grain boundaries
Letters
between different domains lead to a divergence in local patch geometries but show less variation in patch size. Defects within the monolayer such as vacancies or larger and smaller than average particles cause much variation with respect to the patch size. The formation of such defects can be easily avoided using particle solutions with very narrow particle size distributions. In summary, we are able to produce particles with patches as small as 3.7 ( 0.5% of the total particle surface area employing a simple glancing angle deposition technique. We find that the geometry of the patch is dependent on the angle of incidence of the vapor, θ, and the monolayer orientation, R. With the current setup, we are able to produce tiny patches on the particles with the source at incident angles as low as 2°. The mathematical model used to calculate the patch area on the modified particles shows that the patch area varies as a function of the angle of incidence, θ, in a similar fashion for all monolayer orientations, R. The patch area plot shows a significant dependence of the patch area on the monolayer orientation, R, at lower angles of incidence (θ below 30°). Our study demonstrates the potential of the GLAD technique toward producing patchy particles with geometrically different patches. The unique noncircular patches with the sharp corners obtained can be used to provide optical antenna effects to the particles through electromagnetic field enhancements.25 Currently, we are pursuing the generation of patchy particles modified with more than one material using GLAD. Changing the source position with respect to the monolayer orientation and the source material facilitates the formation of patchy particles with multiple patches of different materials. Acknowledgment. This work was supported by the National Science Foundation under a Career Award (no. CBET-06-44789) co-funded by the Division of Chemistry and the Office of Multidisciplinary Activities. I.K. thanks CCNY and CUNY for startup funds. Supporting Information Available: Experimental details and details of the geometrical model. This material is available free of charge via the Internet at http://pubs.acs.org. LA703005Z (25) Crozier, K. B.; Sundaramurthy, A.; Kino, G. S.; Quate, C. F. J. Appl. Phys. 2003, 94, 4632-4642.
