Dr. Stuart Croll
Research
Weathering Durability
- predictive modelling of coating property deterioration
- Monte Carlo simulation
- nanoscale changes in coating properties including molecular relaxations
- corrosion protection
- latex and pigment particle assembly
- drying behavior
- Internal Stresses in Films
- effect on adhesive and cohesive properties
- during weathering
- in novel crosslinking chemistries
- application of coatings' and materials science to the preservation and restoration of modern art
- How external events, polymer science, pigment development and analytical instrumentation lead to the current state of paint technology.
Weathering Durability
One of the most difficult questions to answer in any technology is that of its service life. This is particularly true of complex systems that comprise many parts or ingredients.
Chemical change during weathering is usually provoked by ultraviolet radiation in conjunction with moisture, thermal effects and pollution. The effect of ultraviolet radiation during the lifetime of a material depends on the spectral distribution of the radiation, its intensity, how much is absorbed by the material, and the quantum yield of the degradation process. A complete understanding of durability would include how those chemical changes translate into changes in end use properties and thus a prediction of the service life. We are attempting two approaches (more if we happen across them).
One can see below, in an example, how the absorbance in the UV-visible region increases as the exposure period extends; this is called “yellowing” when it becomes visually obvious.
In one approach, such data can be used to calculate the effective dosage if the quantum efficiencies of the various degradation processes are known in sufficient detail (we are working on it). Any acceleration in the absorbed dosage may allow us to identify an effective lifetime [S. G. Croll and A. D. Skaja, “Spectroscopic Adsorption And Effective Dosage in Accelerated Weathering of a Polyester-Urethane Coating,” J. Mats. Sci. 37(22), pp. 4889 - 4900, (2002)]. In addition, we are trying to link changes in mechanical properties with these spectroscopic trends.
Spectroscopic identification of chemical changes during degradation is very valuable, but it is difficult to deduce without other evidence what the consequences are for the physical properties that determine coating usefulness, e.g. whether they mean an increase in crosslinking, or chain scission and thus the consequences for the mechanical properties or appearance.
Another approach is to assume that degradation takes place in random, repeated events and model the process by Monte Carlo techniques. Our goal is to use chemical, spectroscopic information to learn about the individual events, then accumulate them into kinetics of coating degradation using computational experiments. Computer results are consistent with experimental data [ B. Hinderliter and S. Croll , “Monte Carlo approach to estimating the photodegradation of polymer coatings,” J. Coatings Tech. Research, Vol. 2(6), pp. 483 – 491, 2005 ]:
We find that taking the surface profile that is developed in the simulation and calculating the reflectance of that surface, as it changes with exposure, models real life experiments on gloss loss due to weathering. Similarly, we have had some success modeling the loss in mechanical properties with exposure and changes in surface wetting.
And if we look hard, sometimes we see things that we did not expect:
An early stage of degradation processes: nanoscale blisters formed on an Aircraft Urethane coating: before and after 5 weeks exposure in QUV® seen by Atomic Force Microscope tapping mode by Dr. Xiaofan Yang [X. F. Yang, C. Vang, D. E. Tallman, G. P. Bierwagen, S. G. Croll, S. Rohlik, “Weathering degradation of a polyurethane coating,” Polymer Degradation and Stability, Vol. 74, 341 (2001)].
Colloidal Particles in Paint
Understanding how a paint formulation translates into comparative numbers of particles, how the spacing between particles compares to their size and what controls their stabilization mechanisms improves efficient formulation design. The application of Derjaguin, Landau, Verwey and Overbeek (DLVO) theory of the electrostatic stabilization of colloids to paint formulations shows that composition and structural details of the particles need to be input in order for the theory to properly model stability. It is necessary to include the extent of the surface treatment on the pigment particles, the ionic strength of the continuous phase and the thickness of any adsorbed layer of dispersant polymer as well as knowing the zeta-potential of the particles under the prevailing conditions [S. G. Croll, “DLVO Theory Applied To TiO2 Pigments And Other Materials In Latex Paints,” Prog. Org. Coatings, 44, pp. 131 - 146 (2002)]. The spacing between particles in a random dispersion can be estimated from their particle sizes, their concentrations and the characteristic packing arrangement assigned to the dispersion.
Relative Number of particles in a paint formulation normalized with respect to the extender.
| Material | Amount, dry lb/100 gallons | Particle Diameter, nm | Material Density, kg/m3 | Relative number of particles |
| TiO2 pigment | 225 | 319 | 4030 | 18300 |
| Silica Extender | 160 | 7500 | 2610 | 1 |
| Calcined Clay | 50 | 1500 | 2630 | 39 |
| Latex | 180 | 300 | 1190 | 17700 |
For every one of the larger extender particles there are nearly 20000 TiO2 or latex particles. The exact result of this type of calculation depends on the formulation and detailed particle size, but it is clear that the dominant species are the latex and titanium dioxide particles.
In a paint at 34.5% volume solids concentration, with the ratio of ingredients in the first table above, for a geometry corresponding to loose random packing, 0.59, each particle is something like 76 nm apart. For a geometry that corresponds to dense random packing the spacing increases to 88 nm (a more efficient packing arrangement permits the same number of particles to be further apart). Particle separations are much less than the particle sizes. Even in a well dispersed and stabilized paint, particles tend to maintain their spatial relationships because motion involving several particles is necessary before any could change its local relationship with others. It is easy to see how the motion of particles becomes restricted by their neighbors, leading to increased viscosity.
Film Formation
If there was a long range repulsion between the particles, as there would be if the ionic strength of the continuous aqueous phase were very low, then the position of one particle would be influenced by the repulsion from its neighbors. For many years, conventional colloid stability studies were carried out in continuous phases that were very pure which allowed the stabilizing potential to extend comparatively far from the particle surface. In this situation one can appreciate how, as the paint dried, the particles would avoid each other and ensure that their concentration was maintained evenly throughout the suspension as it increased during drying. This would naturally lead to a gradual first phase of latex film drying wherein the concentration of latex particles increases but remains homogenous in the film.
If particles do not repel others until they are very close because high ionic strength has collapsed the Debye layer thickness, then they would pack together only when they are very close to their neighbors. Thus later ideas of latex drying [S. G. Croll, "Heat and Mass Transfer in Latex Paints during Drying," J. Coatings Tech., Vol. 59, No. 751, 81 (1987)] that involve a more sudden particle interaction during drying and a moving particle consolidation front with the formation of a packed layer of latex are more appropriate where there is a high ionic strength as in commercial aqueous paint formulations. Thus two views on latex paint drying result from a different appreciation of the ionic strength typical in paint.
Internal Stresses in Films
Stresses occur in coatings or other polymer composite systems by virtue of the densification that occurs during the crosslinking reactions or loss of volatiles due to cure; they may also occur due to differential thermal expansion between coating and substrate during the baking/stoving process. Dr. Croll spent his time at the National Research Council, Canada investigating these stresses and their origins. After he moved on, Dr. Dan Perera of the Coatings Research Institute, CoRI, in Belgium took up the interest and produced many useful and interesting articles. Now that Dr. Croll has returned to an ivory glasshouse at NDSU, he is restarting his activities in this area, particularly in the impact the stresses may have in weathering durability, or in the properties of novel crosslinking systems.
One can either measure the internal stress, usually by monitoring the bending of a coated substrate [S. G. Croll, J. Oil and Col. Chem. Assoc., Vol. 63(7), 271-275, (1980)] or measure the equivalent shrinkage when the coating is released from its substrate [ S. G. Croll, J. App. Polym. Sci., Vol. 23, No. 3, 847-858 (1979) ]. The stresses arise when a coating solidifies and can no longer flow to accommodate the loss in volume. In thermoplastic coatings, the solidification can be identified with the glass transition of the polymer plasticized by the evaporating solvent; in thermoset systems solidification occurs due crosslinking. Of course many systems will combine these effects. In thermoplastic systems, the stress or strain is independent of the thickness, but crosslinking coatings often have internal stresses that increase tremendously with coating thickness [S. G. Croll, J. of Coatings Technology, Vol. 53(672), pp. 85 - 92 (1981)].
Internal stresses or strains may be comparable to the cohesive strength of the coating and threaten the usefulness of a coating system. The data here shows internal strain where the upper branch represents shrinkage in films that spontaneously detached from the substrate and the lower branch was from films that were too thin and remained adhering [S. G. Croll, Polymer, Vol. 20, pp. 1423 - 1430, (1979)].
The lines here were generated from knowing how the glass transition changed
with pigment content and residual solvent content:
fs = solvent volume fraction at solidification, identified from the variation
in Tg with solvent content
fr = solvent volume fraction remaining in the “dry” films
fp = pigment volume fraction in the dry film
Adhesion failure due to internal stresses may be quantitatively related to the mechanical properties of the film and its adhesive strength and might be used as a non-intrusive method of measuring adhesion [S. G. Croll, J. Coatings Technology, Vol. 52(665), pp. 35 - 43 (1980)].
