pubs.acs.org/Langmuir © 2010 American Chemical Society
)
Cracking of Drying Latex Films: An ESEM Experiment Kalin I. Dragnevski,*,†, Alexander F. Routh,‡ Martin W. Murray,§ and Athene M. Donald† †
)
Sector of Biological & Soft Systems, Cavendish Laboratory, Department of Physics, University of Cambridge, 19 JJ Thomson Avenue, Cambridge CB3 0HE, U.K., ‡Department of Chemical Engineering and Biotechnology & BP Institute, University of Cambridge, Madingley Rise, Cambridge CB3 0EZ, U.K., and §Akzo Nobel, Wilton Applied Research Group, The Wilton Centre, Redcar TS10 4RF, U.K. Now at the Department of Engineering Science, University of Oxford, Parks Road, Oxford OX1 3PJ, UK Received November 30, 2009. Revised Manuscript Received January 20, 2010 In this study environmental scanning electron microscopy was used to observe the cracking of drying latex films below their glass-transition temperature. By controlling the relative humidity so that it decreases linearly with time, a critical level of humidity at which cracking occurs can be determined and this is measured as a function of film thickness. It was found that the cracking humidity decreases with increases in film thickness for thicknesses in the range of 30 to 100 μm and then remains almost unchanged. A scaling argument can be used to fit the data very well and indicates that cracking occurs as soon as the entire film is consolidated into close packing.
Introduction For many years polymer latices, used for the production of architectural coatings, commonly referred to as paints, have been the subject of extensive theoretical and experimental research. Latex, which is an example of a wet insulating material, can be defined as an aqueous colloidal suspension of spherical polymer particles. If water is allowed to evaporate from the system, then the latex will undergo a series of transformations resulting in the formation of a dry polymer film. Although opinion is divided as to the exact mechanisms that underpin this transformation, most workers agree that this process, known as film formation, can be divided into four main stages that can be described as follows:1-9 stage I - dispersed suspension of polymer particles; stage II concentrated suspension of particles in contact with each other, surrounded by solvent-filled interstices; stage III - continuous ordered array of particles deformed by van der Waals and capillary forces; and stage IV - a molecularly continuous film formed as a result of the diffusion of polymer chains across interfaces between particles. A schematic representation of the process is shown in Figure 1. However, depending on the drying conditions, the formulation, and the thickness of the coating, the films may display a number of cracks. Therefore it is important to understand the mechanism by which the films fail in order to achieve better *Corresponding author. E-mail:
[email protected]. (1) Brown, G. L. Formation of films from polymer dispersions. J. Polym. Sci. 1956, 22, 423. (2) Vanderhoff, J. W. Mechanism of film formation of latices. Chem. Process Eng. 1970, 51, 89. (3) Voyutskii, S. S.; Ustinova, Z. M. Role of autohesion during film formation of latex. J. Adhes. 1977, 9, 39. (4) Sheettz, D. P. Formation of films by drying of latex. J. Appl. Polym. Sci. 1965, 9, 3759-3773. (5) Boczar, E. M.; Dionne, B. C.; Fu, Z.; Kirk, A. B.; Lesko, P. M.; Koller, A. D. Spectroscopic studies of polymer interdiffusion during film formation. Macromolecules 1993, 26, 5772. (6) Winnik, M. A. Latex film formation. Curr. Opin. Colloid Interface Sci. 1997, 2, 192-199. (7) Keddie, J. L. Film formation of latex. Mater. Sci. Eng. 1997, R21. (8) Routh, A. F.; Russel, W. B. A process model for latex film formation: limiting regimes for individual driving forces. Langmuir 1999, 15, 7762-7773. (9) Steward, P. A.; Hearn, J.; Wilkinson, M. C. An overview of polymer latex film formation and properties. Adv. Colloid Interface Sci. 2000, 86, 195-267.
Langmuir 2010, 26(11), 7747–7751
control over this, in many cases, unwanted if natural process. Similar to film formation, opinion is divided as to the mechanism by which cracks form, with a number of mechanisms currently being proposed.10-15 It is suggested that the frequency of cracking and hence the average crack spacing come from balancing the elastic energy released during fracture with the energy required to create the new surface. Therefore, the resultant prediction is that the crack spacing scales with film thickness. However, it has also been shown16 that the crack spacing can be controlled by hydrodynamics. A number of authors have used cantilever techniques to study the stresses that cause cracking.17-19 It has been shown that for a drying film the meniscus of the air-water interface between latex particles gives a capillary pressure in the fluid that is below atmospheric. Therefore, atmospheric pressure, pushing on the film surface, puts the entire system into compression, and hence the exudation of solvent results in a reduction or even elimination of the capillary pressure until evaporation increases it again. Recently, Dufresne et al.20 showed that cracking films are in fact wet except at the position of the cracks. This demonstrated that the capillary pressure is responsible for the film failure. More (10) Jagla, E. A. Stable propagation of an ordered array of cracks during directional drying. Phys. Rev. E 2002, 65, 6147. (11) Bleuth, J. L. Cracking of thin bonded films in residual tension. Int. J. Solids Struct. 1992, 29, 1657-1675. (12) Bordia, R. K.; Jagota, A. Crack growth and damage in constrained sintering films. J. Am. Ceram. Soc. 1993, 76, 2475-2485. (13) Parker, A. P. Stability array of multiple edge cracks. Eng. Fract. Mech. 1999, 62, 577-591. (14) Thouless, M. D. Crack spacing in brittle films on elastic substrates. J. Am. Ceram. Soc. 1990, 73, 2144-2146. (15) Schulze, G. W.; Erdogan, F. Periodic cracking of elastic coatings. Int. J. Solids Struct. 1998, 35, 3615-3634. (16) Lee, P. W.; Routh, A. F. Why do drying films crack? Langmuir 2004, 20, 9885-9888. (17) Peterson, C.; Heldmann, C.; Johannsmann, D. Internal stresses during film formation of polymer lattices. Langmuir 1999, 15, 7745-7751. (18) Martinez, C. J.; Lewis, J. A. Shape evolution and stress development during latex-silica film formation. Langmuir 2002, 18, 4689-4698. (19) Tirumkudulu, M. S.; Russel, W. B. Role of capillary stresses in film formation. Langmuir 2004, 20, 2947-2961. (20) Dufresne, E. R.; Corwin, E. I.; Greenblatt, N. A.; Ashmore, J.; Wang, D. Y.; Dinsmore, A. D.; Cheng, J. X.; Xie, X. S.; Hutchinson, J. W.; Weitz, D. A. Flow and fracture in drying nanoparticle suspensions. Phys. Rev. Lett. 2003, 81, 4501.
Published on Web 02/05/2010
DOI: 10.1021/la904512q
7747
Article
Figure 1. Schematic diagram of the process of latex film formation.
recently, Routh et al.16 introduced a new scaling to describe the spacing between cracks in drying dispersions. The scaling relates to the distance that solvent can flow, to relieve capillary stresses, as a film fails. When used for its traditional applications (i.e., as a paint or adhesive), latex is applied in its wet state to a surface and allowed to dry under ambient conditions. Therefore, conventional electron microscopy, with its extreme drying and specimen preparation requirements is not a suitable characterization technique for the examination of latices in their natural wet state. However, environmental scanning electron microscopy (ESEM),21 which offers the possibility of imaging wet and insulating materials, has been successfully used in the study of a number of systems and dynamic processes including latices and film formation.22-26 Unlike conventional electron microscopy, where high vacuum is maintained throughout the instrument, ESEM is based on the use of a multiple aperture graduated vacuum system, which allows specimens to be imaged under water vapor or other auxiliary gases such as nitrogen and nitrous oxide.24 In this way the chamber can be held at pressures of up to 40 Torr (1 Torr = 133 Pa) while the gun and column remain at pressures of ∼7.5 10-7 Torr.24 Moreover, by using a correct pumpdown procedure27 and by controlling the temperature of the specimen, which in the ESEM is usually done by using a Peltier stage, dehydration can be inhibited and samples can be imaged in their natural state. Furthermore, by taking into consideration the saturated vapor pressure (SVP) curve for water as a function of temperature28 and by increasing (21) Danilatos, G. Review & outline of environmental SEM at present. J. Microsc. 1991, 162, 391-402. (22) Keddie, J. L.; Meredith, P.; Jones, R. A. L.; Donald, A. M. Film formation of acrylic latices with varying concentrations of non-film forming latex particles. Langmuir 1996, 12, 3793-3801. (23) Meredith, P.; Donald, A. M. Study of wet polymer systems in an environmental SEM: some imaging considerations. J. Microscopy 1996, 181, 23-35. (24) Keddie, J. L.; Meredith, P.; Jones, R. A. L.; Donald, A. M. Kinetics of film formation in acrylic latices studied with multiple-angle-of-incidence ellipsometry and environmental SEM. Macromolecules 1995, 28, 2673-2682. (25) Donald, A. M.; He, C.; Royall, P.; Sferrazza, M.; Stelmashenko, N. A.; Thiel, B. L. Applications of environmental SEM to colloidal aggregation and film formation. Colloids Surf., A 2000, 174, 37-53. (26) Bogner, A.; Thollet, G.; Basset, D.; Jouneau, P.-H.; Gauthier, C. Wet STEM: A new development in environmental SEM for imaging nano-objects included in a liquid phase. Ultramicroscopy 2005, 104, 290-301. (27) Cameron, R. E.; Donald, A. M. Minimising sample evaporation in the environmental scanning electron microscope. J. Microsc. 1994, 173, 227. (28) Royall, P. The Behaviour of Silica in Matt Water-Based Lacquers. Ph.D. Thesis, University of Cambridge, Cambridge, U.K., 2000.
7748 DOI: 10.1021/la904512q
Dragnevski et al.
Figure 2. ESEM image of a propagating crack during an in situ drying experiment. Individual latex particles can also be seen within the structure of the film (RH = 70%).
the temperature of the specimen or reducing the chamber pressure, it is possible to control the evaporation conditions within the specimen chamber, which allows for the examination of processes such as film formation and cracking in a controlled way.
Materials and Methods An aqueous latex composition, supplied by ICI Plc, based on copolymers of methyl methacrylate (MMA) and 2-ethylhexylacrylate (2-EHA) was used in this study. The latex was initially about 55 wt % polymer with an average particle size of 300 nm, which was determined by using a Coulter LS230 light-scattering apparatus. The glass-transition temperature (Tg) of the latex was determined by differential scanning calorimetry (DSC), carried out on dry specimens, using a Perkin-Elmer Pyris 1 instrument. The measured temperature was 279.8 K. The minimum filmformation temperature (MFFT) of the latex was measured by using an MFFT-Bar and was found to be 278 K. Microstructural analysis was carried out on an FEI XL-30 environmental scanning electron microscope equipped with a Peltier stage. Wet samples from the latex formulation were cast onto standard (E)SEM stubs (10 mm diameter) by uniformly spreading the dispersion with a glass rod. These were then placed onto the cooling stage in the microscope chamber at a temperature of ca. 274 K. An evaporation-inhibiting pump-down sequence was then performed, with the ambient air progressively replaced by water vapor. Once the purging cycle was completed, water vapor pressures and working distances of 3.5-4.5 Torr and 9.5-11.5 mm were set, which provided suitable imaging environments. Imaging of the specimens was carried out at an accelerating voltage of 10 kV. Previous studies 28 have shown that the use of moderate beam voltages in combination with fairly high pressures results in minimal beam damage, which proved to be the case in this study. To determine the humidity at cracking accurately, the specimens were maintained wet after the initial pump-down sequence. Then the water content of the samples was slowly reduced by varying the temperature and/or pressure of the chamber. The relative humidity in the chamber was set to reduce linearly with time (e.g., H = 100 C2t, where H is the relative humidity, t is time, and C2 is a constant of around 1 s-1). This allowed an accurate measurement of the humidity at which the first cracks appeared spontaneously in the field of view. After that the samples were allowed to form films fully, and the thickness of the films was measured.
Experimental Results An example of a typical observation of crack formation during the drying experiments is shown in Figure 2. It can be Langmuir 2010, 26(11), 7747–7751
Dragnevski et al.
Article
Figure 3. ESEM image sequence of the drying/rewetting experiment. Image (a) shows the wet film, which upon drying cracks (b). Rewetting of the film, induced by decreasing its temperature, causes the cracks to heal (c). Further redrying (d), in this instance, induced by increasing temperature, results in crack reformation. It is important to mention that because of the fact that the images were taken at low magnification it was not possible to observe individual latex particles.
Figure 4. ESEM images at (a) the crack tip and (b) the rear of a crack. seen that the crack does not propagate in an entirely straight line and probably follows a path of defects in the particle packing. Figure 3 demonstrates that upon rewetting a crack pattern will heal and completely disappear. However, upon further drying the same crack pattern is formed. This indicates that any defects in the particle packing are retained in the dry-rewet sequence. Partial coalescence prior to cracking would also mean that the cracks would appear in the same place after subsequent rewetting cycles. It is interesting to note that the crack width is larger in the rewetted case (Figure 2d) when compared to the original crack (Figure 2b). Figure 4 shows an ESEM image taken at the crack tip (Figure 4a) and at the rear of the same crack (Figure 4b). The crack tip image indicates that the film ruptures at the film surface and the crack then propagates toward the substrate as the crack moves forward. Figure 4b at the rear of the crack shows smaller, almost parallel cracks running close to the main crack. Similar Langmuir 2010, 26(11), 7747–7751
ESEM observations were made by Chiu et al.29 in drying granular ceramic films. They also suggested that the intermediate (packed) region is saturated with water. The results from the measurements of the critical cracking humidity as a function of the latex film thickness are presented in graphical form in Figure 5. The relative humidity was reduced from 100% at a constant rate, and the point at which cracks formed was recorded. Hence, for 50-μm-thick films the first cracks formed at a relative humidity of 75%. From the above results, it is clearly seen that the humidity at which cracking occurs decreases for film thicknesses of up to 100 μm and then becomes constant. Here it is important to note that it is believed that the humidity data for thicknesses (29) Chiu, R. C.; Garino, T. J.; Cima, M. J. Drying of granular ceramic films: 1, effects of processing variables on cracking behaviour. J. Am. Ceram. Soc. 1993, 76, 2258.
DOI: 10.1021/la904512q
7749
Article
Dragnevski et al.
Scaling Argument Consider a film of thickness H and volume fraction φ as shown in Figure 7. The evaporation rate E_ determines the change in film height according to dH ¼ -E_ dt
ð1Þ
The evaporation rate is mass-transfer-controlled and is given by E_ = k(p* - p¥), where k is a mass-transfer coefficient, p* is the saturated vapor pressure of water under the experimental conditions, and p¥ is the vapor pressure of water in the atmosphere. If the total pressure of the humidity chamber is P, then the humidity H is given by Figure 5. Critical cracking humidity as a function of film thickness.
H ¼
18P¥ 29ðP -P¥ Þ
ð2Þ
The relative humidity H is the humidity divided by the humidity at saturation, hence P¥ ðP -PÞ 100 H¼ P ðP -P¥ Þ
ð3Þ
If we assume that P . P* . p¥ then the relative humidity reduces to P¥ H ¼ 100 P
ð4Þ
Hence the linear reduction in the relative humidity with time, Hh=100 - C2t, relates to a linear reduction in the water partial pressure with time. Substitution into the evaporation rate expression and using eq 1 results in Figure 6. Surface microstructure of a drying latex showing cracks around the edges of the specimen.
dH p ¼ - k p ð100 -C2 tÞ dt 100
ð5Þ
where C2 is the rate at which the relative humidity is reduced. Solving eq 5 subject to the initial height being H0 gives H ¼ H0 -
kpC2 2 t 200
ð6Þ
An overall mass balance of Hφ = H0φ0 results in the following expression for the volume fraction in the film Figure 7. Film of height H and volume fraction φ evaporating at
_ rate E.
above 100 μm is artificial because cracking was observed only around the edges of the specimens (as shown in Figure 6). This is most probably due to small variations in the thickness of the sample and/or small fluctuations of the conditions under which the sample is being examined. It has also been suggested30 that this may be due to vertical nonuniformities in the volume fraction; the Peclet number becomes larger than 1. During this study it was not possible to carry out measurements on films
