Hidden Cavity Formations by Nanocrystalline Self-Assembly on

Letter pubs.acs.org/Langmuir

Hidden Cavity Formations by Nanocrystalline Self-Assembly on Various Substrates with Different Hydrophobicities Iltai Kim† and Kenneth D. Kihm* Department of Mechanical, Aerospace, and Biomedical Engineering, University of Tennessee, Knoxville, Tennessee 37996-2210, United States S Supporting Information *

ABSTRACT: The effect of surface hydrophobicity is examined in the formation of hidden complex cavities during evaporation-induced nanocrystalline self-assembly taking place on three different substrates bearing different levels of hydrophobicity, namely, cover glass (CG), a gold thin film (Au), and a polystyrene dish (PS). It turns out that the DLVO theory, the relative thermal conductivities between the substrate and nanofluids, and the relationship between the evaporation and the radial outflow motions of nanoparticles comprehensively explain why the number of cavity cells is proportional to nanoparticle concentration and inversely proportional to surface hydrophobicity.

1. INTRODUCTION Evaporative self-assembly of nanofluids has attracted increasing attention over the past decade, ever since Deegan et al. systematically explained the mechanism of coffee stain ring formation.1 Nevertheless, the formation of hidden cavities inside the self-assembly remained largely unknown because of the fine scale and complexity associated with the three-phase transition. In the former study,2 the formation of the hidden cavities was observed to be due to a complex competition between Marangoni flow, capillary flow, and near-surface forces. In addition, several interactive forces compete with one another during the evaporative formation of hidden cavities. These include the forces driven by the surface temperature gradient, electrostatic forces between nanoparticles, van der Waals forces between nanoparticles, and the substrate surface as well as the internally driven thermophoretic forces that work to replenish the evaporated liquid.2−6 Therefore, it is anticipated that the formation of hidden cavities involves the multiple competitions between the evaporation of liquid and the radial outflow motion of the nanoparticles,8 the relative thermal conductivities between the substrate and nanofluid, and also the electrostatic and van der Waals forces between the nanoparticles and the substrate (i.e., the DLVO force7). Recently, Bhardwaj et al.7 showed the role of DLVO interactions in forming self-assembled patterns of colloidal particles that were substantially larger than nanoparticles. Shen et al.8 investigated the formation of coffee ring structure using the dimensionless time scales between the evaporation and the radial motion of nanoparticles. Reistenpart et al.9 showed that the internal Marangoni flow direction depends upon the relative thermal conductivity between the substrate and the evaporating liquid. Despite numerous studies of self-assembly pattern variations associated with surface hydrophobicity,7,10−17 © 2012 American Chemical Society

very few of them have focused on the formation of hidden cavities inside, and little is known about the effect of the surface hydrophobicity of the solid substrate and the nanofluid concentrations. The use of surface plasmon resonance (SPR) reflectance imaging has recently enabled the quantitative and dynamic characterization of hidden complex cavities formed by the evaporative self-assembly of nanofluids.2 The mechanism for hidden cavity formation is explained by stagnation flow balanced between outward radial flow and inward Marangoni flow based on experimental observation. For the case of a sessile drop, Barash et al.18 have shown that the convective flow that drives the assembly gives rise to hydrodynamic vortices, which aid the formation of hidden cavities. A study of 2D particle agglomeration suggested that this was the case.19 Furthermore, this study addressed the formation of multiple aggregation islands using externally controlled interparticle forces.

2. EXPERIMENTAL SECTION We now report our findings on the different formations of hidden complex cavities that result from varying the self-assembly parameters, such as the surface hydrophobicity and nanoparticle concentration. Our experiment was conducted with Al2O3 nanoparticles of 47 nm average diameter, which were sonically mixed with deionized water, without any additives. The laboratory conditions were controlled to remain at 40% humidity and 21 ± 0.5 °C. The microscopy imaging system allows simultaneous observations of the dorsal view, side view, and the natural fringe ventral view of an evaporating drop on a substrate (Figure 1). The natural fringes are constructed from the Received: April 16, 2012 Revised: May 27, 2012 Published: June 7, 2012 9195

dx.doi.org/10.1021/la301526f | Langmuir 2012, 28, 9195−9200

Langmuir

Letter

Figure 1. Experimental schematic illustration and dorsal view microscopy images of evaporative self-assembled patterns of Al2O3 nanofluids of three different initial volume concentrations on three different substrates (a cover glass (CG), a gold thin film (Au), and a polystyrene dish (PS)) with an increasing level of hydrophobicity. Cracks are identified by the bright lines and curves. interference of two reflected rays: one from the substrate surface and the other from the cavity ceiling surface. Thus, these two rays represent the optical path length differential, or the cavity heights. The aqueous contact angles of the three tested substrates were found to increase with their hydrophobicity levels, starting from a cover glass (CG, θc ≈ 52°), a gold thin film (Au, θc ≈ 76°), and finally a polystyrene surface (PS, θc ≈ 90°).

where Lm is the mean distance between the two particles, Dp is the nanoparticle diffusivity, D is the nanofluidic drop diameter, K is a characteristic parameter (L2/t) representing the droplet evaporation rate (K = (4Dν(C0 − C∞)/ρL), Dν is the molecular diffusion coefficient of the vapor in air, C0 and C∞ are the densities of the saturated vapor in close proximity above the liquid−air interface and the ambient vapor density, respectively, ρL is the liquid density, and θinitial and θreceding are the contact angles of the droplet at the beginning and while receding, respectively (θinitial ≫ θreceding for most cases). The physical implication of CR is that slow evaporation requires a longer time, so nanoparticles have sufficient time to migrate and form a pinned contact line (i.e., CR ≪ 1). However, when evaporation takes place too quickly, nanoparticles self-assemble prematurely, before reaching the contact line (i.e., CR ≫ 1). The calculated CR values are much less than unity for all of the tested conditions (Figure 2), which is consistent with the fact that most of the self-assembled patterns are well defined to result in the pronounced contact lines. In addition, CR decreases with increasing nanoparticle concentration, demonstrating that rings are more definite and broader at higher concentrations. However, CR is of the same order for the three different substrates, and similar self-assembly patterns resulted for the same nanoparticle concentrations. Note that the dorsal images in Figure 1 do not show the hidden cavities because of the opacity of the self-assembled cavity ceilings. Thus, the bright lines and curves in Figure 1

3. RESULTS AND DISCUSSIONS The dorsal view images (Figure 1) demonstrate consistency with the predicted relationship between the evaporation and the radial outflow motion of nanoparticles, given by a dimensionless number CR as8 τparticle CR = τevap (1) where τparticle relates the time for two adjacent nanoparticles to meet near the contact line τparticle =

Lm 2 2Dp

(2)

and τevap represents the evaporation time for nanoparticles to form the initial monolayer within the droplet τevap =

θinitial − θreceding ⎛ D ⎞2 π⎜ ⎟ ⎝2⎠ 4K

(3) 9196

dx.doi.org/10.1021/la301526f | Langmuir 2012, 28, 9195−9200

Langmuir

Letter

fringes are constructed from the cavity shape and size. Separate sets of bright/dark fringes correspond to individual cavity cells. Indeed, most of the cracks shown in Figure 1 are located between the neighboring cellular fringes identified in Figure 3. The weaker contrast of the images for the Au substrate is due to the notably reduced transparency of the gold thin film of approximately 50 nm thickness. The hidden cavities are preferably formed in the ring area, with sizes ranging from tens of micrometers to 1 mm, and the number of cavity cells tends to increase with increasing nanoparticle concentration but somewhat decreases with an increasing level of substrate hydrophobicity (Figure 4; please refer to the Supporting Information regarding the dynamic processes of cavity cell formation during the evaporative selfassembly occurring on different substrates). Cavities are generated more quickly for a nanofluid with higher nanoparticle concentrations because of the faster evaporation rates. It is also observed that cracks are formed before cavities, and these cracks mostly define the main boundaries between individual cavity cells. For the self-assembly of nanofluids at higher concentrations, a notably greater number of cracks (visible or invisible from the dorsal view) would develop because of the higher intrinsic stresses and their rates of growth within the ring structures. Therefore, it is conjectured that the increased number of cavity cells for nanofluids at higher nanoparticle concentrations can be attributed to the increased number of cracks. The number of cavity cells also demonstrates a positive correlation with increasing DLVO force,7 which represents the

Figure 2. Dependence of C R on the nanoparticle volume concentration for the three tested substrates with different hydrophobicities.

identify cracks that developed after dry-out. The formation of hidden cavities is visible only in the ventral view of natural fringe mapping, as seen in Figure 3 where the bright/dark

Figure 3. Ventral view microscopy images of hidden complex cavities of Al2O3 nanofluids of three different initial volume concentrations on three different substrates with increasing hydrophobicity: a cover glass (CG), a gold thin film (Au), and a polystyrene dish (PS). Families of bright/dark fringes are evolved in cellular shapes representing individual cavity cells. Indeed, most of the cracks shown in Figure 1 are located between the neighboring cellular fringes shown. The weaker contrast of the images for the Au substrate is due to the notably reduced transparency of the gold thin film of approximately 50 nm thickness. 9197

dx.doi.org/10.1021/la301526f | Langmuir 2012, 28, 9195−9200

Langmuir

Letter

Figure 4. Number of hidden cavities as a function of nanoparticle concentration for the three tested substrates.

combined attractions of the electrostatic and van der Waals forces between the nanoparticles and the substrate Figure 5. Dependence of FDLVO on the nanoparticle volume concentration for the three tested substrates.

FDLVO = Fel + Fvdw Fel = −

128πd pγpγsnkBTκ −1

2 ⎛ z ⎞ = a exp⎜ − −1 ⎟ ⎝ κ ⎠

FvdW =

⎛ z ⎞ exp⎜ − −1 ⎟ ⎝ κ ⎠

α 1 Ad p3 2 rtd 2 12 z (z + d p)

forces for the case of the cover glass substrate induce more pronounced self-pinning with ring patterns. Furthermore, FDLVO increases almost linearly with increasing nanoparticle concentration, showing that the reduced interparticular distance dramatically increases the electrostatic attractions and thus broader and more definite rings are formed at a higher concentration of nanoparticles. For a given substrate, the larger DLVO force at a higher concentration seems to create more hidden cavities. However, the increased DLVO forces with reduced hydrophobicity, from PS to CG, do not always show more cavity cells. The number of cavity cells of the Au film is the lowest with its hydrophobicity being in the middle. Thus, the DLVO force alone does not completely explain the dependence of the cavity numbers on the surface hydrophobicity. We believe that the additional consideration of the relative thermal conductivities between the substrate and nanofluid, kR = kS/kL, can supplement the physical explanations for the number of hidden cavities formed. Figure 6 shows that kR for the Au substrate is 3 orders of magnitude higher than that for the CG substrate and 4 orders of magnitude higher than that for the PS substrate. The thermal conductivity of the substrate, kS, is given as 1.05, 318, and 0.08 W/mK for the CG-, Au-, and PS substrates, respectively,33 and the thermal conductivity of nanofluids, kL, is given as 0.58, 0.61, and 0.66 W/mK for the 0.1, 1.25, and 10% volume concentrations, respectively.34,35 Consequently, the highly conductive Au substrate (kR ≫ 1) maintains a relatively uniform surface temperature whereas the air/liquid interface temperature is lowered because of the latent heat loss with evaporation. Thus, the thermocapillary−phoretic Marangoni flow is driven from the warmer contact line to the cooler top region of the droplet (inset photograph in Figure 6). Therefore, the radially outward flows along the substrate surface are induced to replenish the Marangoni flow movement. In contrast, for the poorly conducting PS substrate (kR < 1), the intensive evaporation near the contact line lowers the local

(4)

where dp is the nanoparticle diameter (47 nm), γp is the surface potential function of the Al2O3 nanoparticles with φs = 30 mV,20,21 and γs is the surface potential function of the substrates with φs = −65, −42, and −35 mV for the glass, Au, and PS substrates, respectively.21−23 n is the number of counterions far away, kB is the Bolztmann constant, T is the ambient temperature (294 K), κ−1 is the Debye length (∼1 μm for pure water24), A is the Hamaker constant, αrtd is the retardation constant (αrtd ≈ 0.1),24 and z is the nanoparticle distance from the substrate surface. The Hamaker constant, A, between Al2O3 nanopaticles and the substrate in the water medium is calculated to be 1.7, 8, and 1.5 for the CG, Au, and PS substrates (AAu > A CG > APS), respectively.25−32 Because both electrostatic and van der Waals forces are positive, FDLVO is always attractive between the nanoparticles and the substrates. The electrostatic force is a function of the surface potential (γs or φs) of a given substrate and its magnitude changes with varying hydrophobicity as well as with the corresponding surface potential. However, the van der Waals component is a primary function of the nanoparticle size dp. Assuming z ≈ κ−1, the calculations show that FDLVO increases with increasing concentration but decreases with increasing hydrophobicity (Figure 5). The relatively lower DLVO forces for the PS substrate indicate that there is not enough of a bonding force to attract nanoparticles to the PS substrate strongly. This is consistent with the observation that self-assembly on the PS substrate results in relatively weak pinning and substantial shrinkage of the drop as evaporation progresses (refer to Figure 3). In contrast, the stronger DLVO 9198

dx.doi.org/10.1021/la301526f | Langmuir 2012, 28, 9195−9200

Langmuir

■

Letter

ASSOCIATED CONTENT

S Supporting Information *

Videos of the dynamic processes of cavity cell formation during the evaporative self-assembly occurring on different substrates. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Present Address †

Center for Integrated Nanotechnologies, Sandia National Laboratory. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This research is partially supported by the University of Tennessee Basic Research Development Program (R011373164) and partially by the WCU (World Class University) Program at Seoul National University through the Korea Research Foundation, funded by the Ministry of Education, Science and Technology (R31-2008-000-10083-0).

Figure 6. Relative thermal conductivities of the substrate and liquid kR = kS/k for the three tested substrates with different nanoparticle volume concentrations.

■

REFERENCES

(1) Deegan, R. D.; Bakajin, O.; Dupont, T. F.; Huber, G.; Nagel, S. R.; Witten, T. A. Capillary flow as the cause of ring stains from dried liquid drops. Nature 1997, 389, 827−829. (2) Kim, I.; Kihm, K. D. Unveiling hidden complex cavities formed during nanocrystalline self-assembly. Langmuir 2009, 25, 1881−1884. (3) Xu, X. F.; Luo, J. B. Marangoni flow in an evaporating water droplet. Appl. Phys. Lett. 2007, 91, 3. (4) Min, Y. J.; Akbulut, M.; Kristiansen, K.; Golan, Y.; Israelachvili, J. The role of interparticle and external forces in nanoparticle assembly. Nat. Mater. 2008, 7, 527−538. (5) Ghasemi, H.; Ward, C. A. Energy transport by thermocapillary convection during sessile-water-droplet evaporation. Phys. Rev. Lett. 2010, 105, 4. (6) Nel, A. E.; Madler, L.; Velegol, D.; Xia, T.; Hoek, E. M. V.; Somasundaran, P.; Klaessig, F.; Castranova, V.; Thompson, M. Understanding biophysicochemical interactions at the nano-bio interface. Nat. Mater. 2009, 8, 543−557. (7) Bhardwaj, R.; Fang, X.; Somasundaran, P.; Attinger, D. Selfassembly of colloidal particles from evaporating droplets: role of DLVO interactions and proposition of a phase diagram. Langmuir 2010, 26, 7833−7842. (8) Shen, X. Y.; Ho, C. M.; Wong, T. S. Minimal size of coffee ring structure. J. Phys. Chem. B 2011, 114, 5269−5274. (9) Ristenpart, W. D.; Kim, P. G.; Domingues, C.; Wan, J.; Stone, H. A. Influence of substrate conductivity on circulation reversal in evaporating drops. Phys. Rev. Lett. 2007, 99, 4. (10) Mori, Y. Effect of surface hydrophobicity on interaction between particle and flat plate at final stage of wet coating process. Colloids Surf., A 2007, 311, 61−66. (11) Uno, K.; Hayashi, K.; Hayashi, T.; Ito, K.; Kitano, H. Particle adsorption in evaporating droplets of polymer latex dispersions on hydrophilic and hydrophobic surfaces. Colloid Polym. Sci. 1998, 276, 810−815. (12) Andreeva, L. V.; Koshkin, A. V.; Letiedev-Stepanov, P. V.; Petrov, A. N.; Alfimov, M. V. Driving forces of the solute selforganization in an evaporating liquid microdroplet. Colloids Surf., A 2007, 300, 300−306. (13) Sommer, A. P.; Rozlosnik, N. Formation of crystalline ring patterns on extremely hydrophobic supersmooth substrates: Extension of ring formation paradigms. Cryst. Growth Des. 2005, 5, 551−557.

temperature below that of the rest of the drop surface. This results in a reversed Marangoni flow direction toward the cooler contact line region. It also reverses the internal flow direction to become radially inward along the substrate. The reason that the gold (Au) substrate leads to the smallest number of cavity cells formed is believed to be because the extremely large kR of gold drives strong radially outward flows toward the contact line, which may decrease the formation of cavity cells by overriding the DLVO force that tends to promote the formation of cavity cells. For the case of the polystyrene substrate, despite its lowest DLVO force, the radially inward driven flow along the substrate surface (because of kR < 1) is believed to promote cavity formation to a certain degree.

4. CONCLUSIONS On the basis of our experimental observations, we have discovered that the hidden complex cavity formation depends on self-assembly parameters such as the surface hydrophobicity and nanoparticle concentration. The natural fringe mapping technique shows that the number of cavity cells increases with increasing concentration but decreases with increasing hydrophobicity. We have attempted to explain these findings more comprehensively with integrative theoretical approaches, including DLVO theory, relative thermal conductivities between the substrate and nanofluids, and the relationship between the evaporation and the radial outflow motions of nanoparticles. Nevertheless, the exact details of cavity formation have yet to be further elucidated. For example, the physical mechanisms that precisely determine the number and size of the cavities are still unclear. We note the need for future research to delineate the complex underlying physics of hidden cavity formation further. This will include an extension of the scope of experiments toward more finely controlled surface hydrophobicities and a consideration of the effective distance of surface forces and cavity dimensions. 9199

dx.doi.org/10.1021/la301526f | Langmuir 2012, 28, 9195−9200

Langmuir

Letter

(14) Shin, D. H.; Lee, S. H.; Jung, J. Y.; Yoo, J. Y. Evaporating characteristics of sessile droplet on hydrophobic and hydrophilic surfaces. Microelectron. Eng. 2009, 86, 1350−1353. (15) Grant, L. M.; Tiberg, F.; Ducker, W. A. Nanometer-scale organization of ethylene oxide surfactants on graphite, hydrophilic silica, and hydrophobic silica. J. Phys. Chem. B 1998, 102, 4288−4294. (16) Grant, L. M.; Ducker, W. A. Effect of substrate hydrophobicity on surface-aggregate geometry: zwitterionic and nonionic surfactants. J. Phys. Chem. B 1997, 101, 5337−5345. (17) Kokkoli, E.; Zukoski, C. F. Surface pattern recognition by a colloidal particle. Langmuir 2001, 17, 369−376. (18) Barash, L. Y.; Bigioni, T. P.; Vinokur, V. M.; Shchur, L. N., Evaporation and fluid dynamics of a sessile drop of capillary size. Phys. Rev. E 2009, 79, 046301. (19) Miao, Y. H.; Liu, Y. H.; Hu, L. Y.; Helseth, L. E. Colloidal clustering of protein-coated microspheres in evaporating droplets. Soft Matter 2012, 8, 2267−2273. (20) Sieger, H.; Winterer, M.; Muhlenweg, H.; Michael, G.; Hahn, H. Controlling surface composition and zeta potential of chemical vapor synthesized alumina-silica nanoparticles. Chem. Vapor Depos. 2004, 10, 71−76. (21) Lee, J. H.; Hwang, K. S.; Jang, S. P.; Lee, B. H.; Kim, J. H.; Choi, S.; Choi, C. J. Effective viscosities and thermal conductivities of aqueous nanofluids containing of Al2O3 low volume concentrations nanoparticles. Int. J. Heat Mass Transfer 2008, 51, 2651−2656. (22) Lee, S. K.; Han, M. S.; Asokan, S.; Tung, C. H. Effective gene silencing by multilayered siRNA-coated gold nanoparticles. Small 2010, 7, 364−370. (23) Lackie, A. M. Effect of substratum wettability and charge on adhesion invitro and encapsulation invivo by insect hemocytes. J. Cell Sci. 1983, 63, 181−190. (24) Tien, C. Granular Filtration of Aerosols and Hydrosols; Butterworths: Boston, 1989. (25) Israelachvili, J. N. Intermolecular and Surface Forces; Academic Press: London, 1992. (26) Petosa, A. R.; Jaisi, D. P.; Quevedo, I. R.; Elimelech, M.; Tufenkji, N. Aggregation and deposition of engineered nanomaterials in aquatic environments: role of physicochemical interactions. Environ. Sci. Technol. 44, 6532−6549. (27) Gauthier, M.; Regnier, S.; Rougeot, P.; Chaillet, N. Analysis of forces for micromanipulations in dry and liquid media. J. Micromechatronics 2006, 3, 389−413. (28) Medout-Marere, V. A simple experimental way of measuring the Hamaker constant A(11) of divided solids by immersion calorimetry in apolar liquids. J. Colloid Interface Sci. 2000, 228, 434−437. (29) Zhao, Y.; Ng, H. T.; Hanson, E.; Dong, J. N.; Corti, D. S.; Franses, E. I. Computation of nonretarded London dispersion coefficients and Hamaker constants of copper phthalocyanine. J. Chem. Theory Comput. 2010, 6, 491−498. (30) Jaiswal, R. P.; Kumar, G.; Kilroy, C. M.; Beaudoin, S. P. Modeling and validation of the van der Waals force during the adhesion of nanoscale objects to rough surfaces: a detailed description. Langmuir 2009, 25, 10612−10623. (31) Ackler, H. D.; French, R. H.; Chiang, Y. M. Comparisons of Hamaker constants for ceramic systems with intervening vacuum or water: From force laws and physical properties. J. Colloid Interface Sci. 1996, 179, 460−469. (32) Butt, H.-J. g.; Cappella, B.; Kappl, M. Force measurements with the atomic force microscope: technique, interpretation and applications. Surf. Sci. Rep. 2005, 59, 1−152. (33) Bergman, T. L.; Lavine, A. S.; Incropera, F. P.; DeWitt, D. P. Fundamentals of Heat and Mass Transfer; Wiley: Hoboken, NJ, 2011. (34) Kleinstreuer, C.; Feng, Y. Experimental and theoretical studies of nanofluid thermal conductivity enhancement: a review. Nanoscale Res. Lett. 2011, 6, 229. (35) Mintsa, H. A.; Roy, G.; Nguyen, C. T.; Doucet, D. New temperature dependent thermal conductivity data for water-based nanofluids. Int. J. Therm. Sci. 2009, 48, 363−371.

9200

dx.doi.org/10.1021/la301526f | Langmuir 2012, 28, 9195−9200