3752
Langmuir 2008, 24, 3752-3758
Shape Changes during the Drying of Droplets of Suspensions Y. Zhang, S. Yang, L. Chen, and J. R. G. Evans* Queen Mary, UniVersity of London, Mile End Road, London E1 4NS, United Kingdom ReceiVed NoVember 7, 2007. In Final Form: January 11, 2008 During drying of droplets of suspensions, several flow regimes contribute to the radial flow of powder to the periphery to leave a pile-up of powder at the rim. It is shown that the shape of the droplet residues can be controlled both by restricting evaporation and by combining high and low boiling point solvents which modify particle flows and produce a range of droplet residues varying from a concave “doughnut” shape, sometimes with a central hole, to a convex dome shape. Addition of formamide to aqueous suspensions is shown to affect powder deposition by setting up a Marangoni flow rather than by reducing evaporation at the periphery. The results find direct application in thick-film combinatorial printing of ceramics to form small disks by droplet drying.
Introduction Thick-film combinatorial methods of ceramic discovery make use of ink-jet printing and related techniques to produce libraries of ceramic samples with different compositions.1 Ceramic powders, dispersed in suspension and partially stabilized against sedimentation with the help of dispersants, can be mixed, deposited as drops, and fired to produce libraries for subsequent high-throughput testing of properties on a wide range of compositions. The so-called “coffee ring” phenomenon2,3 hinders the production of disklike samples because powders accumulate at the edge of droplets instead of distributing uniformly. The sample becomes a ring or a well. Indeed, the phenomenon can be used to make ceramic well plates.4 For discrete library samples this hinders property measurement, X-ray diffraction, and spectroscopy. Samples with flat surfaces or convex domes that can be polished flat are preferable. Four types of flow influence the final drying pattern of a concentrated suspension in the form of a sessile drop.5 First is the lateral flow needed to compensate for evaporative losses from the droplet periphery,2,3 which is considered to be the main cause of the coffee ring effect. The second is the Marangoni flow,5-9 which affects the final patterns by inducing circulatory flows in the drying droplet. The third is due to the gravitational force,5 which induces sedimentation. Sedimentation is minimized in ceramic ink formulation by the use of dispersants but, except for very fine powders, is never completely extinguished. Finally, the Brownian motion produces particle collisions both in suspension and with the peripheral deposit of particles and influences particle diffusion. In this work, we modify the effects of the gravitational force during drying of small (5-30 µL) concentrated droplets and then introduce a drying control agent * Corresponding author. Present address: Department of Chemistry, University College London, 20 Gordon Street, London WC1H 0AJ. (1) Wang, J.; Evans, J. R. G. J. Comb. Chem. 2005, 7, 665-672. (2) Deegan, R. D.; Bakajin, O.; Dupont, T. F.; Huber, G.; Nagel, S. R.; Witten, T. A. Nature 1997, 389, 827-829. (3) Deegan, R. D.; Bakajin, O.; Dupont, T. F.; Huber, G.; Nagel, S. R.; Witten, T. A. Phys. ReV. E 2000, 62, 756-765. (4) Zhang, Y.; Chen, L. F.; Yang, S. F.; Evans, J. R. G. J. Am. Ceram. Soc. 2006, 89, 3858-3860. (5) Yiantsios, S. G.; Higgins, B. G. Phys. Fluids 2006, 18, No. 082103. (6) Hu, H.; Larson, R. G. J. Phys. Chem. B 2002, 106, 1334-1344. (7) Hu, H.; Larson, R. G. Langmuir 2005, 21, 3972-3980. (8) Hu, H.; Larson, R. G. J. Phys. Chem. B 2006, 110, 7090-7094. (9) Wang, H. T.; Wang, Z. B.; Huang, L. M.; Mitra, A.; Yan, Y. S. Langmuir 2001, 17, 2572-2574.
which has been used for controlling patterns on polymer films and in ink-jet printing.10-12 Experimental Details The experiment contains two sections: (1) placing small aqueous suspension droplets in a high-moisture environment, allowing sedimentation with very little evaporation, (2) introducing a drying control agent to the water and observing droplet residue changes. Five powders (Table 1) were used for preparing water-based inks. The dispersant used to stabilize all the inks was Darvan 821A (R.T. Vanderbilt Co., Inc., Norwalk, CT), which is an ammonium salt of poly(methacrylic acid) in aqueous solution. The dispersant dose was 2 wt % active dispersant based on ceramic powder. The inks were treated using an ultrasonic probe (model U200S, IKA, Staufen, Germany) at a duty cycle of 0.5 Hz and amplitude of 0.8 for 10 min. The inks are numbered, and their compositional details are listed in Table 2. In the first section, 10 µL droplets of the TiO2, Fe2O3, SrCO3, and Co3O4 inks (ink nos. 1-4) were deposited on a silicone release paper (grade SPT50/11, Cotek Papers Ltd., Glos, U.K.) using a digital transfer pipet (range 5-50 µL, Brand GMBH, Wertheim, Main, Germany). The silicone release paper is hydrophobic and provides a contact angle of 120° with singly distilled water. The ceramic inks, which tend to have lower surface energies, produce contact angles which are slightly greater than 90° initially. The droplets were dried in a closed vessel of 100 × 70 × 35 mm containing 5 mL of distilled water, giving RH ) 98 ( 2%, measured using a humidity sensor (HIH-3610, Honeywell Co., Illinois) at 20 °C. After 1 h, the vessel was uncapped and the droplets were dried quickly in an oven set to 80 °C. In parallel, droplets of the ink were dried in air at ambient conditions (approximately 40% RH, 20 °C) to provide a comparison. In the second section, a drying control agent, formamide (Merck kGaA, Darmstadt, Germany), which is a high boiling point solvent (Tb ) 210 °C) was used in a fixed ratio of 75 wt % water and 25 wt % formamide. TiO2 powder was dispersed in either water or the solvent mixture at solid loadings of 5, 10, and 20 vol % to form ink nos. 1 and 5-10 (Table 2). Sedimentation tests were conducted to explore the effects on the ink stability of introducing formamide. Droplets (20 µL) of the inks were deposited on silicone release paper and dried in air at ambient conditions (about 40% RH, 20 °C). The drying rate of the droplets was determined by weighing the droplets during drying using a microbalance (2100 mg ( 0.1 µg, model SE2, Sartorius Co., Surrey, U.K.). After drying, the shapes of the residues of the droplets were recorded and compared. (10) Park, J.; Moon, J. Langmuir 2006, 22, 3506-3513. (11) de Gans, B. J.; Schubert, U. S. Langmuir 2004, 20, 7789-7793. (12) Karabasheva, S.; Baluschev, S.; Graf, K. Appl. Phys. Lett. 2006, 89.
10.1021/la703484w CCC: $40.75 © 2008 American Chemical Society Published on Web 03/11/2008
Shape Changes during Drying of Suspension Droplets
Langmuir, Vol. 24, No. 8, 2008 3753
Table 1. Details of Oxide Powders powder
supplier
grade
TiO2 Fe2O3 SrCO3
Alfa Aesar, Lancashire, U.K. Alfa Aesar, Lancashire, U.K. Sigma-Aldrich Gmbh, Schnelldorf, Germany PI-KEM, Shropshire, U.K. Ex Condea Vista, Arizona
A14N51
Co3O4 Al2O3
1 2 3 4 5 6 7 8 9 10 11 a
density/ kg m-3
mean particle size/µm
472018
99.997 98.5 >99.9
4230 5240 3700
0.08
