Controlling particle dispersion in latex paints containing associative thickeners.

Abstract Latex paints contain several types of particles including polymer binder, primary pigment, extenders, and colorants. When the paints contain associative thickeners, control of particle dispersion can be very complicated due to the interaction of the particles with dispersants, surfactants, and the associative thickener itself. In particular, dispersion of the pigments can act independently of dispersion of the binder particles. The consequences of this situation are manifested in the physical properties of the paint and of the films it forms. This paper describes these interactions in terms of their colloid chemistry and shows the consequences of additive choices on the particle dispersion and also the optical properties of model paint films.

Keywords Dispersants, Thickeners, Latexes, Colloids, Latex. Dispersion, Flocculation, Hiding. Pigment optics, Titanium dioxide

Introduction

Background

The applications properties of latex paints have improved significantly with the advent of associative thickeners. (1) Figure 1 shows two reasons for the increase in use of associative thickeners: decreased low shear viscosity leading to better flow and leveling and increased high shear viscosity leading to higher film build and improved brush feel. An additional rheological benefit is reduced spatter due to increased extensional viscosity. Since associative thickeners are rheology modifiers, it is natural that most of the literature deals with this aspect of their behavior. The fourth benefit of associative thickeners is not rheology related, but deals with their ability to form uniform networks of latex and pigment particles leading to improved physical properties such as higher gloss and hiding. It is this aspect with which this paper deals in terms of the colloidal interactions of associative thickeners and pigments. The dispersion state of pigments has a profound effect on film optical properties and the correct choice of dispersant type and level for the particular type of associative thickener is critical.

The two most common types of associative thickener are Hydrophobically modified Ethoxylated Urethanes (HEUR) and Hydrophobically modified Alkali-Swellable Emulsion (HASE). The objectives of this paper are to clarify the dispersion/flocculation behavior of pigments thickened with associative thickeners, compare with nonassociative systems, and link this behavior to the film properties of gloss and hiding for both HEUR- and HASE-thickened systems.

Dispersion of particles in latex coatings formulations

In practical coatings formulations, nonassociative thickeners do not adsorb onto latex or pigment particles. This leads to the particle distribution pictured in Fig. 2a where the thickener solution is a separate phase that crowds the pigment and latex particles together. A much more uniform distribution can be achieved with associative thickeners with the proper choice of latex parameters, dispersant type, and surfactant. This is pictured in Fig. 2b and leads to improved rheological and applications properties. Associative thickeners are unique in that the dispersion state of the latex particles and each type of pigment particle (i.e., primary, extender, and colorant pigments) are independent, leading to the possibility of poor pigment dispersion/good latex dispersion (Fig. 2c) or good pigment dispersion/poor latex dispersion (Fig. 2d). In addition, the thickener associations can be minimized to the point where it acts as a nonassociative thickener. Figures 2a, 2c, and 2d all depict a phenomenon known as depletion flocculation. This is usually the cause of degraded properties and will be discussed in detail in the next section. In this paper the latex particles are all well dispersed and the pigment dispersion state is explored.

[FIGURE 1 OMITTED]

Another form of particle-particle interaction is bridging flocculation. As the name implies, particles are connected together in close proximity by a single thickness of thickener molecules. Bridging flocculation was discussed in detail in previous papers (2-5) and is not covered in detail in this paper due to the fact that it is fairly rare in fully formulated paints. This is because bridging usually occurs at additive levels much below those found in actual coatings formulations. Therefore, particle stability and especially depletion flocculation will be the focus of the next section.

[FIGURE 2 OMITTED]

Interparticle potential energy and flocculation

The particles in coatings formulations are subject to attractive and repulsive forces that lead to a metastable situation. Whether the particles are well dispersed or flocculated depends on their interparticle potential energy [V.sub.tot]. This can be represented in its simplest form by:

[V.sub.tot] = [V.sub.lelec] + [V.sub.vdw] + [V.sub.depl] (1)

where [V.sub.elec] is the repulsive energy keeping the particles apart by charge-charge repulsion, [V.sub.vdw] is the attractive energy all matter experiences due to Van der Waals forces, and [V.sub.depl] is the depletion energy (attractive) resulting from the osmotic force exerted by nonabsorbing (i.e., nonassociative) polymers on particles in suspension. (6,7) Previous papers in this series dealt with latex dispersion. It is worth noting that pigment particles are more difficult to disperse due to inherently higher Van der Waals attraction. This can make high gloss and hiding difficult to achieve in latex paints.

Figure 3 shows the total interparticle potential energy curve for particles that are well dispersed (i.e., not flocculated). Note that as the particles approach one another, they just keep experiencing increased repulsive energy. Figure 4 shows the curves of all of the contributing energies for particles that are experiencing depletion flocculation. These all add up to the [V.sub.tot] curve. Note that the particles do not actually touch each other in the depletion state. Rather they exist about 10 nm apart in a shallow secondary energy minimum. This is not coagulation, which occurs when particles are in actual physical contact in the primary energy minimum (not shown in the figure). Since the particles are only weakly attracted to each other (compared to coagulation) shear forces can redisperse them whereupon they will reflocculate.

[FIGURE 3 OMITTED]

[FIGURE 4 OMITTED]

When no energy barrier to flocculation exists, as in Fig. 4, the rate of flocculation is governed by diffusion-limited aggregation kinetics (DLA) wherein the rate of flocculation is proportional to the number of particles per unit volume and inversely proportional to the viscosity of the continuous phase. When an energy barrier exists, as pictured in Fig. 5, the rate of flocculation is slower and it is governed by reaction-limited aggregation kinetics (RLA). For RLA the rate of flocculation is the DLA rate multiplied by the factor exp (-[V.sub.max]/kT) where [V.sub.max] is the height of the energy barrier as noted in Fig. 5. An energy barrier may be caused by an adsorbed species, for example. DLA usually produces larger, more irregularly shaped flocs than RLA. This can have consequences for film properties such as gloss and hiding in addition to rheological effects. A more detailed description of DLA and RLA has been presented previously for latexes. (8)

[FIGURE 5 OMITTED]

[FIGURE 6 OMITTED]

Dispersion diagrams

Earlier work in this series introduced the dispersion phase diagrams (DPD) for latexes (2,3) and pigments. (9) DPDs provide a convenient way to visualize the regions where good dispersion, bridging flocculation, and depletion flocculation of particles occur as a function of thickener concentration and dispersant or surfactant. Figure 6 is a generalized DPD with the different regions labeled. The goal of coatings formulation is to target the good dispersion region so that applications properties are optimized. As expected, the viscosity of a system will increase with increasing associative thickener concentration. At a given thickener concentration, the particles usually experience bridging flocculation at low dispersant (or surfactant) levels, then good dispersion, and finally depletion flocculation as the associative thickener is displaced from the particles and is converted into a nonassociative polymer. DPDs will be used later in this paper to depict the effects of using low to high carboxylate-containing dispersants on Ti[O.sub.2] dispersion in both HEUR and HASE systems.

[FIGURE 7 OMITTED]

Degree of dispersion

It is useful to have a guideline for assessing the extent of particle flocculation in a dispersion. For the purposes of this work, a scale of 1-5 was established wherein 5 is a good, well-dispersed system and 1 represents particle flocculation that results in very large floc structure. Figure 7 is a series of three micrographs showing examples of dispersion ratings 5, 3, and 1 of Ti[O.sub.2] dispersions. At equilibrium, the dispersion in the good region of a DPD is a "5" whereas it is usually a "1" in the bridging and depletion regions. When a depletion-flocculated dispersion of rating "1" is sheared, it momentarily becomes a "5," then transitions through "4," "3," "2," and finally to "1." How quickly this happens depends on whether the system is undergoing DLA or RLA kinetics of flocculation. Flocculation rate is usually slower near the flocculation phase boundary because some associative polymer is still adsorbed at that point before being totally displaced by dispersant or surfactant, for example.

Experimental

Materials

The following materials were used to determine the pigment phase behavior:

Model Associative Polymers -- HEUR-type polyoxy-ethylene backbone with terminal C12 hydrophobes (average molecular weight of 50,000) and HASE-type with MAA/EA copolymer backbone and pendant [C.sub.12][H.sub.25] hydrophobes (average molecular weight of 400,000).

Nonassociative Polymer -- HEC of comparable molecular volume to the HASE.

Pigments -- commercial grade interior Ti[O.sub.2] (alumina-rich surface) and commercial grade colorants phthalo blue, lampblack, and red iron oxide.

Dispersants -- high carboxylate dispersant (polyacrylic acid), a range of dispersants with varying amounts of carboxylate (i.e., acidic monomer), a hydrophobic dispersant (olefin/maleic acid copolymer), inorganic dispersant potassium tripolyphosphate (KTPP).

Surfactants -- anionic surfactant sodium dodecyl sulfate (SDS) and nonionic surfactant octylphenol ethoxylate 40 (OP (EO)[.sub.40]).

Pigment Solids -- 0.10 volume fraction for Ti[O.sub.2] and 0.05 volume fraction for colorants.

pH -- adjusted to 9.0-9.5 with N[H.sub.4]OH

Model Paints -- 20 PVC, 35% VS model paints containing interior Ti[O.sub.2], acrylic latex, typical additives, and thickener (HEUR, HASE, or HEC), drawn down and analyzed for contrast ratio, 20[degrees] gloss and 60[degrees] gloss. Paint viscosity was adjusted to 95-105 KU.

Note: Concentrations of thickener and surfactant are expressed as wt% of the continuous phase and concentrations of dispersant are expressed as wt% of pigment.

Determination of particle dispersion

Aqueous mixtures of pigment, thickener, and dispersant were prepared in clear glass containers at 0.10 volume fraction pigment and allowed to equilibrate for at least 60 h before evaluation. Particle dispersion was assessed by both visual inspection and microscopy. In some samples, the bridging flocculation region could not be conclusively distinguished from coagulation resulting from very low dispersant levels. Therefore. the region was treated as a combined bridging flocculation/coagulation region. Depletion flocculation could be confirmed by the fact that dilution of the sample with water to below the critical flocculation concentration yielded a well-dispersed system. For DPDs, an average of 30-40 samples were prepared for each system to define the dispersion and flocculation regions with some precision. This information was used to create the dispersion diagrams. Rate of pigment flocculation was assessed microscopically by shearing the sample under a cover slip and observing the rate of floe formation and the size of the floes. The degree of flocculation was rated according to a scale of 1-5 described in a previous section (see Fig. 7). A rating of 1 would be expected for a nonassociative thickened dispersion and a rating of 5 would be expected for an associative thickened dispersion with correct choice and level of dispersant.

Results

Adsorption of dispersant on interior Ti[O.sub.2]

Good dispersants serve as coupling agents to induce adsorption of associative thickener onto pigment surfaces. Dispersants can vary from polycarboxylates (i.e., polyacids) that are very hydrophilic to copolymers of carboxylates and hydrophobic monomers that are considered hydrophobic dispersants. Generally speaking, polycarboxylates are not good coupling agents between pigments and associative thickeners. In this study the full range of dispersant composition was explored, so the first step was to measure the adsorption characteristics of low- and high-carboxylate dispersant on interior Ti[O.sub.2]. Figures 8a and 8b show the low-carboxylate adsorption isotherm and the high-carboxylate adsorption isotherm, respectively. The two most important characteristics are that they both have similar adsorption curves and that they both are close to saturation in the range of 0.5-0.75% dispersant based on pigment weight. This information will be used later to help explain adsorption behavior of the associative thickeners.

[FIGURE 8 OMITTED]

Nonassociative thickeners

As a baseline, the DPD was determined for interior grade Ti[O.sub.2] and high-carboxylate dispersant thickened with nonassociative thickener, specifically HEC. Figure 9 is the DPD. This is typical of nonassociative thickeners: depletion flocculated everywhere except at very low thickener concentration below the critical flocculation concentration. The diagram looks the same whether a high- or low-carboxylate dispersant is used. For nonassociative thickeners, the flocculation rate is fast and is governed by DLA kinetics. The only way to obtain a reasonably good dispersion of pigment in a nonassociative system is to get uniform hetero-flocculation of the pigment and latex, and this is difficult to achieve.

[FIGURE 9 OMITTED]

HEUR thickener/Ti[O.sub.2] dispersions

Background

The properties of coatings prepared with HEUR thickeners are known to be sensitive to the choice of dispersant and also the choice of Ti[O.sub.2] grade. (9-11) As mentioned previously, this work will concentrate on dispersant effects and utilize only interior grade Ti[O.sub.2] as the primary pigment. Other grades of Ti[O.sub.2] have been discussed in a previous paper in this series and in other publications. In order to achieve maximum benefit, the pigment must become part of the associative polymer network just as the latex does. To achieve good dispersion the dispersant should be bifunctional, having ionic functionality to interact with the pigment and hydrophobic functionality to interact with the associative thickener hydrophobes. The first study to be done was to characterize the adsorption of HEUR onto the pigment as a function dispersant.

[FIGURE 10 OMITTED]

[FIGURE 11 OMITTED]

Adsorption of HEUR thickener on Ti[O.sub.2]

As a starting point, the adsorption of HEUR onto Ti[O.sub.2] as a function of dispersant type and concentration was determined. Some work had already been done in this area in the literature. (12-14) Figure 10a shows how adsorption of HEUR is significantly enhanced as a low-carboxylate (i.e., hydrophobic) dispersant is added. After saturation by the dispersant, HEUR adsorption begins to decline due to excess dispersant in the continuous phase interacting with the HEUR molecules. (15) As one moves to the high-carboxylate dispersant (Fig. 10b), the adsorption levels of HEUR are much lower, but significant, and decline to near zero by the time the Ti[O.sub.2] is saturated with dispersant. High-carboxylate dispersant in the continuous phase does not interact directly with the HEUR in solution except for possibly an electrolyte effect.

Dispersion phase diagrams of HEUR-thickened Ti[O.sub.2] systems

The DPDs for Ti[O.sub.2] in HEUR were determined for the low-carboxylate dispersant (Fig. 11a) and the high-carboxylate dispersant (Fig. 11b). As expected from the adsorption data, the low-carboxylate dispersant yielded a large region of good dispersion, finally reaching depletion flocculation when there was enough excess dispersant in the continuous phase to disrupt the associative network. For the high-carboxylate dispersant, the initial HEUR adsorption level was apparently not high enough to give a good dispersion, so depletion flocculation dominates the diagram. HEUR adsorption was considerably lower than that for the low-carboxylate dispersant (see Fig. 10).

Optical properties of paints containing HEUR, Ti[O.sub.2], and dispersant

In order to confirm that the colloidal dispersion model actually translates into observed paint properties, model paints were prepared containing the same components used in the adsorption and dispersion studies. In the first set of paints, the dispersant level was kept at 1% of pigment solids as the % of carboxylate in the dispersant was varied. Both gloss and hiding were measured on the paint films. Figure 12 shows the data for 60[degrees] and 20[degrees] gloss and Fig. 13 is the corresponding data for contrast ratio. The gradually decreasing values with increasing carboxylate agree well with the colloidal dispersion concepts. In addition, if one considers that an optimized HEUR paint yielded 60[degrees] gloss = 83, 20[degrees] gloss = 43, and C.R. = 94.0, and that the comparable HEC-thickened paint yielded 60[degrees] gloss = 51, 20[degrees] gloss = 7, and C.R. = 91.1, then this very closely brackets the results of the dispersant series from good dispersion to depletion flocculated.

[FIGURE 12 OMITTED]

The next experiment was to determine the effect of high-carboxylate dispersant concentration on the optical properties of the paints. The dispersant concentration was varied from 0.1 to 1% based on the pigment weight. The surprising result is shown in Fig. 14 for gloss and Fig. 15 for C.R. At low levels of dispersant both gloss and hiding approached optimum levels even though the dispersion results indicated a flocculated pigment. As the dispersant level increased to saturation level (~0.5%) the optical properties degraded to that of the fully flocculated system. Perhaps the low level of HEUR adsorption was having a positive effect on the optical properties? Clearly a more detailed analysis of the dispersion properties of these systems was in order, especially as to the flocculation kinetics.

[FIGURE 13 OMITTED]

[FIGURE 14 OMITTED]

[FIGURE 15 OMITTED]

[FIGURE 16 OMITTED]

Flocculation rate and dispersion diagrams: HEUR

All of the DPDs discussed up to this point were equilibrium results, so it was decided to redo Diagram 11b, but assessing the pigment dispersion 30 s after shear had been applied. Remember that shear induces a temporary state of good dispersion from which particles then flocculate. As mentioned in the Methods section, these studies had to be done exclusively on the microscope. Figure 16 is the high-carboxylate DPD with dispersion ratings added at 0.1, 0.3, 0.5, 0.7, and 1% dispersant. These results clearly show that the "deeper" one goes into the depletion region, the worse (i.e., faster) the flocculation. The increased viscosity with increasing HEUR is also yielding an effect of slowing flocculation. The paints contained approximately 1-1.5% HEUR in the continuous phase. In the case where a pigment is flocculating in a paint film, the viscosity as it dries, and the open time will most likely have an effect on the degree of flocculation in the final film. The depletion flocculation rate appears to be governed by RLA kinetics at low dispersant levels where some HEUR is still adsorbed and by DLA when all of the HEUR is desorbed.

Effect of mixed dispersants on dispersion

One last dispersant effect was studied in HEUR systems before going on to surfactant effects. This was the effect of mixing small amounts of an inorganic hydrophilic dispersant into a well-dispersed Ti[O.sub.2] in HEUR. Potassium tripolyphosphate (KTPP) was added at 0.25% to Ti[O.sub.2] in 1% HEUR at various low-carboxylate dispersant levels. Table 1 lists the results of the study. Without KTPP, the expected excellent pigment dispersion was obtained at all dispersant levels. In all cases where KTPP was present the dispersion was very poor, irregardless of the low-carboxylate dispersant level. Clearly, one must be careful in formulating HEUR coatings to avoid mixing even small amounts of highly charged hydrophilic dispersants with "good" dispersants.

Effect of surfactants on the dispersion

Surfactants may be added to coatings formulations overtly or as components of additives such as colorants. There is evidence in the literature that surfactants may interact with both the associative polymer and the pigment, in addition to the latex binder. (1) Two typical types of surfactant include anionics such as sodium dodecylsulfate (SDS) and nonionics such as alkylphenol ethoxylates (e.g., Octylphenol (EO)[.sub.40]) of different EO content and alkyl length. A study was done to evaluate dispersion of Ti[O.sub.2] in 1% HEUR solution containing 1% low-carboxylate dispersant and either SDS or OP (EO)[.sub.40]. The study was repeated using 1% high-carboxylate dispersant. The results are summarized in Table 2. The SDS severely degraded the low-carboxylate dispersions and had no effect on the already poor dispersions of the high-carboxylate samples. The nonionic surfactant had no effect on the already good dispersions of the low-carboxylate dispersant samples and caused gradual improvement in the initially poor dispersions of the high-carboxylate case. SDS is very efficient at destroying the associative thickener network and inducing depletion flocculation. By contrast, the ethoxylated nonionics are very compatible with the HEUR structure and do not cause flocculation although they do of course affect viscosity. Both SDS and ethoxylated nonionics have induced similar effects in latex systems thickened with HEUR. (2)

HASE thickener/Ti[O.sub.2] dispersions

Background

HASE associative thickeners differ structurally from HEUR in that they are polyelectrolytes and of much higher molecular weight and lower hydrophobe density. HASE thickeners derive more of their thickening action from molecular volume than typical HEUR thickeners. HASE thickeners are much less studied than HEUR thickeners in the literature both from a rheology standpoint and especially in terms of particle dispersion, so it is worth exploring their interactions to determine how good dispersions and coatings properties are achieved.

Dispersion phase diagrams of HASE-thickened Ti[O.sub.2] systems

To get a complete picture of dispersion phenomena in associative thickener systems, studies of dispersion and paint optical properties analogous to those for the HEUR systems were undertaken for HASE systems to compare and contrast the two thickeners. Figure 17a is the DPD for interior Ti[O.sub.2] with low-carboxylate dispersant and Fig. 17b is for the high-carboxylate case. Note that, unlike the HEUR system, there is no ridging floc/coagulation region. Since HASE molecules are polyelectrolytes, they can act like very high-molecular weight dispersants. This is why the high-carboxylate dispersant DPD has a significant region of good dispersion also. Initially the HASE disperses the pigment, but eventually the noncoupling dispersant displaces it, leading to depletion flocculation.

Optical properties of paints containing HASE, Ti[O.sub.2], and dispersant

Analogous paints to the dispersion diagrams were prepared and the optical properties of the resulting films assessed for a % carboxylate ladder of the dispersant. The gloss results and the C.R. results are in Figs. 18 and 19, respectively. Only a slight decrease in gloss was observed as % carboxylate increased and there was no statistical difference in C.R. across the series. Values were uniformly high. For comparison, an optimized HASE paint had 60[degrees] gloss = 82, 20[degrees] gloss = 41, and C.R. = 93.8. These results suggested a flocculation rate effect similar to what was observed in the HEUR case. When paints were made with increasing concentration of high-carboxylate dispersant, very little change in gloss or C.R. was observed across the series. This is shown in Figs. 20 and 21, respectively. This makes some sense considering that the good dispersion region of the HASE systems is considerably larger than for those of the HEUR system, so flocculated samples are not as far "into" the flocculation region as in the HEUR case. Also, there is less effect on viscosity in the HASE systems compared to the HEUR systems. Just to confirm that flocculation rate was a significant factor, the DPD was redetermined with floe kinetics just as was done for the comparable HEUR case.

Flocculation rate and dispersion diagrams: HASE

A flocculation rate study was run in the same manner as that for the HEUR case. Figure 22 shows the results. The paints in the optical study had a HASE concentration of about 1.5% based on continuous phase. The good dispersion ratings even up to 1% high-carboxylate dispersant confirm that flocculation rate is an important factor in the good optical properties across a range of dispersant composition and concentration. Another point needs to be made here. Although low-carboxylate dispersants produce very good dispersions, they can lead to excessive structure (i.e., "livering") upon aging. Therefore high-carboxylate dispersants are often the ones of choice for paints thickened with HASE as long as the dispersant level is kept below 1%. The good news is that the results reported here suggest that there is little sacrifice in performance when using high-carboxylate dispersants, and this seems to be a consequence of slower flocculation rates. This more extreme behavior distinguishes the HASE from the HEUR thickeners.

[FIGURE 17 OMITTED]

[FIGURE 18 OMITTED]

[FIGURE 19 OMITTED]

[FIGURE 20 OMITTED]

[FIGURE 21 OMITTED]

[FIGURE 22 OMITTED]

Effect of surfactants on dispersion

Surfactants in the coatings formulation may also be an issue for HASE-thickened paints. Therefore, a study was done to evaluate dispersion of Ti[O.sub.2] in 2% HASE solution containing 1% low-carboxylate dispersant and either SDS or OP (EO)[.sub.40]. The study was repeated using 1% high-carboxylate dispersant. The results are summarized in Table 3. Unlike the HEUR systems, the SDS had very little effect on the low-carboxylate dispersions, which were very good, and also had no effect on the already poor dispersions of the high-carboxylate samples. The nonionic surfactant had no effect on the already good dispersions of the low-carboxylate dispersant samples and, unlike the HEUR case, had no effect on the poor dispersions of the high-carboxylate case. This behavior reflects the differences in interaction of the surfactants with HASE compared to HEUR thickeners.

Extender pigment dispersions

Extender pigments were not strictly part of the work presented here, but it is worth mentioning that they follow the same dispersion rules as the other pigments discussed in this paper. One way to determine if a pigment is hydrophilic or hydrophobic is to attempt to disperse it in 1% HEUR solution without dispersant. A good dispersion indicates a hydrophobic pigment surface or at least one that adsorbs EO polymers. By this method, talc and anhydrous (calcined) kaolin produce dispersions with a 5 rating, and hydrous kaolin and calcium carbonate produce l's. Considering their surface structures, this makes sense. Dispersant choices follow the same guidelines as the primary pigments. Since colorant pigments contain other additives such as surfactants, they are more complicated and will be discussed next.

Colorant dispersions

Background

Colorants present a formulation challenge because they often contain unidentified dispersants and surfactants that can drastically reduce the viscosity of HEUR-thickened paints. Some studies of HEUR systems can be found in the literature, but they tend to concentrate on rheological effects (16,17) and tint strength, (18) but not direct dispersion effects. Also, studies of HASE systems are not well represented in the literature. Therefore three representative colorants were chosen to study the dispersion characteristics in 1% HEUR solution and 2% HASE solution. The colorants were red iron oxide, phthalo blue, and lampblack, representing a range of surface characteristics. Iron oxide is considered hydrophilic and phthalo blue is hydrophobic. (19) Lampblack has a more complex surface structure and high surface area, so its behavior is also more complex. (18) Surfactant effects were probed using the same surfactants as in the Ti[O.sub.2] work: SDS and OP (EO)[.sub.40].

Colorant dispersion in HEUR systems

Dispersion samples were prepared based on 0.05 volume fraction colorant in 1% HEUR solution and surfactant concentrations were based on the continuous phase. Results are listed in Table 4. All three colorants had an excellent 5 rating "as is" (i.e., without added surfactant), indicating significant adsorption interaction with the HEUR. The added SDS had no effect on the phthalo blue dispersion rating, but it gradually degraded the lampblack dispersion and had a sudden negative effect on the red iron oxide at higher SDS concentrations. The crystal surface structure of phthalo blue appears to be such that it can accommodate negatively charged surfactants. There is evidence in the literature for negative effects of SDS on colorant performance in HEUR coatings. (18) By contrast, the nonionic surfactant had no effect on the dispersions, yielding all 5's for all three colorants at all surfactant concentrations. This result is similar to the Ti[O.sub.2] discussed earlier in this paper and to latex results. (2) Nonionics can provide some steric stabilization and EO surfactants are quite compatible with HEURs. Clearly excess anionic surfactants should be avoided in HEUR-based formulations if possible.

Colorant dispersion in HASE systems

Dispersion samples were prepared based on 0.05 volume fraction colorant in 2% HASE solution and surfactant concentrations were based on the continuous phase. Results are listed in Table 5. Both red iron oxide and phthalo blue had excellent dispersions in HASE alone, whereas the lampblack was very poor. These results seem reasonable based on the surface composition of the colorants and the results from the HEUR study. SDS quickly degraded the red iron oxide dispersions and gradually degraded the phthalo blue, suggesting that too much charge was present in these systems as the SDS level increased. The lampblack dispersion remained very poor as expected for a hydrophobic pigment. The nonionic surfactant degraded the red iron oxide dispersion at all concentrations, whereas the phthalo blue remained a 5 at all nonionic surfactant levels. The nonionic surfactant improved the lampblack to the 2-3 range, but this is still a much worse dispersion than those of the phthalo blue, for example. These results suggest that HASE systems with colorant are more sensitive to added surfactant than HEUR systems. Therefore, care must be taken when formulating HASE coatings with colorants.

Discussion

Earlier work (2,3) established the dispersion behavior of latexes with respect to associative thickener and surfactant type and concentration. The current work extends this to pigments, where dispersants play a somewhat analogous role to the surfactants in latex systems. The rules are the same, no matter what the composition of particles in latex paints: additives that enhance the adsorption of associative thickeners on the particle surface lead to good dispersions and superior film properties, whereas additives that displace the associative thickeners eventually lead to depletion flocculation and consequently degraded film properties. Latex and pigment particles can act independently in terms of dispersion/flocculation, so it is necessary to pay attention to both the dispersant and surfactant content of latex paints thickened with associative thickeners. This is one of the ways that associative systems differ from nonassociative ones.

For HEUR thickeners, hydrophilic dispersants (e.g., those with high-carboxylate content) tend to prevent adsorption of the thickener onto the pigment. As the pigment particle is covered with dispersant, the HEUR interaction decreases. The rate of depletion flocculation appears to be dependent on dispersant coverage, with higher coverage leading to less thickener adsorption and faster flocculation. In a drying film, slower flocculating pigment can still produce good physical properties such as high gloss and hiding because the good dispersion is "frozen in" as the paint dries and viscosity increases. HASE thickeners, because of their polyelectrolyte structure, tend to adsorb onto pigments more strongly than HEUR thickeners. This dual interaction makes HASE thickeners less sensitive to choice of dispersant. That said, flocculation rate is still an important phenomenon for HASE systems containing very hydrophilic dispersants. In HASE systems, hydrophobic dispersants can sometimes create too much structure and cause rheological problems. The results presented here strongly suggest that dispersant level should be optimized for both HASE and HEUR paints. Surface composition of pigments is a very important parameter and it is well to be aware of the different surfaces among the many grades of Ti[O.sub.2] and types of extender pigments and colorants.

Colorants present a special challenge because they have proprietary additives such as surfactants and dispersants that can have a significant impact on associative thickener interaction. However, colorants generally behave as expected based on their hydrophilic or hydrophobic surface composition. Anionic surfactants appear to create more problems in associative interactions involving pigments than nonionic surfactants. This is probably because they add ionic interactions into the mix, which can be very important for many pigment types.

The work presented here demonstrates the power of analyzing associative thickener systems in terms of colloidal interactions. There is a direct correlation between particle dispersion and both paint and film physical properties. Since associative thickener systems are complex, colloidal knowledge can help the formulator make better choices for additives such as dispersants and surfactants.

Conclusions

The following conclusions can be drawn based on the structure of the pigment phase diagrams and the paint optical properties of HEUR- and HASE-thickened systems generated in the work presented here:

1. Dispersion phase diagrams are useful for understanding the complex interactions of pigments with dispersants, associative thickeners, and surfactants.

2. HASE thickeners are more tolerant of dispersants with high-carboxylate content than HEUR thickeners.

3. Pigment flocculation and flocculation rate are important parameters for determining film properties such as gloss and hiding.

4. Dispersion properties of pigments are well correlated with film optical properties.

5. Anionic surfactants are more likely to have a negative effect on Ti[O.sub.2] and colorant dispersion than nonionic dispersants.

6. It is more difficult to achieve good colorant dispersion in HASE systems than in HEUR systems.

7. Close attention must be paid to dispersant and surfactant types and levels if optimum film properties are to be achieved.

8. Simplified systems (e.g., thickener/pigment/dispersant) can provide insight into the behavior of the actual coatings formulations.

Acknowledgment The author would like to thank the Rohm and Haas Co. for support and for permission to publish this work.

References

1. Glass, JE, "A Perspective on the History of and Current Research in Surfactant-Modified, Water-Soluble Polymers." J. Coat. Technol., 73 (913) 79 (2001)

2. Kostansek, E, "Using Dispersion/Flocculation Phase Diagrams to Visualize Interactions of Associative Polymers, Latexes and Surfactants." J. Coat. Technol., 75 (940) 27 (2003)

3. Kostansek, E, "Associative Polymer/Latex Dispersion Phase Diagrams II: HASE Thickeners." J. Coat. Technol. Res., 2 (6) 1 (2005)

4. Sperry, PR, Thibeault, JT, Kostansek, EC, "Flocculation and Rheological Characteristics of Mixtures of Latexes and Water-Soluble Polymeric Thickeners." Adv. Org. Coat. Sci. Technol., Series 9 (1) (1987)

5. Thibeault, JT, Sperry, PR, Schaller, EJ. "Effect of Surfactants and Cosolvents on the Behavior of Associative Thickeners in Latex Systems," In: Glass, JE (ed.) Water Soluble Polymers: Beauty with Performance, Advances in Chemistry Series 213, Chapter 20, American Chemical Society, Washington, D.C. (1986)

6. Sperry, PR, "A Simple Quantitative Model for the Volume Restriction Flocculation of Latex by Water-Soluble Polymers." J. Colloid Interface Sci., 87 375 (1982); Sperry, PR, "Morphology and Mechanism in Latex Flocculated by Volume Restriction," J. Colloid Interface Sci., 99 97 (1984)

7. Myers, D, Surfaces, Interfaces, and Colloids. (2nd ed.). Wiley-VCH, New York (1999)

8. Kostansek, E, "Controlled Coagulation of Emulsion Polymers." J. Coat. Technol. Res., 1 (1) 41 (2004)

9. Kostansek, E, "Associative Polymer/Particle Dispersion Phase Diagrams III: Pigments." J. Coat. Technol. Res., 3 (3) 165 (2006)

10. Lundberg, DJ, Glass, JE, "Pigment Stabilization Through Mixed Associative Thickener Interactions." J. Coat, Technol., 64 (807) 53 (1992)

11. Tarng, M-R, et al., "Associative Thickeners in the Land of Commercial Reality: Coating Formulations," In: Glass, JE (ed.) Hydrophilic Polymers: Performance with Environmental Acceptability, Advances in Chemistry Series 248, Chapter 24. American Chemical Society, Washington, D.C. (1996)

12. Svanholm, T, Kronberg, B, Molenaar, F, "Adsorption Studies of Associative Interactions between Thickener and Pigment Particles." Prog. Org. Coat., 30 167 (1997)

13. Glass, JE, "Adsorption of HEUR Thickeners on Latex and Titanium Dioxide Disperse Phases." Adv. Colloid Interface Sci., 79 123 (1999)

14. Melville, I et al., "Pigment Thickener Interactions in Emulsion Paints." Polym. Paint Colour J., 177 (4187) 174 (1987)

15. Johnson, EA, "Interactions Between Rheology-Modifying and Pigment-Dispersing Agents." Farbe & Lack, 100 (9) 759 (1994)

16. Bergh, JS, Lundberg, DJ, Glass, JE, "Rheology of Associative Thickener Pigment and Pigmented Commercial Latex Dispersions." Prog. Org. Coat., 17 155 (1989)

17. Mahli, DM, Wegner, JM, Glass, JE, Phillips, DG, "Waterborne Latex Coatings of Color: I." J. Coat. Technol. Res., 2 (8) 627 (2005)

18. Mahli, DM, Wegner, JM, Glass, JE, Phillips, DG, "Waterborne Latex Coatings of Color: II." J. Coat. Technol. Res., 2 (8) 635 (2005)

19. Reiman, H et al., "Particles in Networks ("Partikel in Netzwerken")." Farbe & Lack, 108 91 (2002)

[c] FSCT and OCCA 2007

Presented at the 2006 FutureCoat! conference, sponsored by the Federation of Societies for Coatings Technology, in New Orleans, LA, on November 1-3, 2006.

E. Kostansek ([mailing address])

Rohm and Haas Company, P.O. Box 904, Spring House, PA 19477-0904, USA

e-mail: [e-mail]

Table 1: Effect of hydrophilic inorganic dispersant on 0.10 volume
fraction interior Ti[O.sub.2] dispersions in 1% HEUR solution

% Low-carboxylate dispersant  % KTPP  Dispersion rating

0.25                          0       5
0.5                           0       5
1.0                           0       5
0                             0.25    2
0.25                          0.25    1
0.5                           0.25    1
1.0                           0.25    1

Table 2: Effect of surfactants on 0.10 volume fraction interior
Ti[O.sub.2] dispersions in 1% HEUR solution

% Low-       % High-
carboxylate  carboxylate  %     % OP           Dispersion
dispersant   dispersant   SDS   (EO)[.sub.40]  rating

1            0            0     0              5
1            0            0.25  0              1
1            0            0.50  0              1
1            0            0     0.5            5
1            0            0     1              5
1            0            0     1.5            5
0            1            0     0              2
0            1            0.25  0              1
0            1            0.5   0              1
0            1            0     0.5            3
0            1            0     1              4
0            1            0     1.5            5

Table 3: Effect of surfactants on 0.10 volume fraction interior
Ti[O.sub.2] dispersions in 2% HASE solution

% Low-       % High-
carboxylate  carboxylate  %     % OP           Dispersion
dispersant   dispersant   SDS   (EO)[.sub.40]  rating

1            0            0     0              4
1            0            0.25  0              5
1            0            0.50  0              5
1            0            0     0.5            4
1            0            0     1              4
1            0            0     1.5            5
0            1            0     0              1
0            1            0.25  0              2
0            1            0.5   0              1
0            1            0     0.5            1
0            1            0     1              1
0            1            0     1.5            1

Table 4: Effect of surfactants on 0.05 volume fraction colorants in 1%
HEUR solution

                     Red iron oxide  Phthalo blue  Lampblack
%     % OP           dispersion      dispersion    dispersion
SDS   (EO)[.sub.40]  rating          rating        rating

0     0              5               5             5
0.25  0              5               5             5
0.5   0              5               5             3
0.75  0              1               5             2
1.0   0              1               5             1
0     0.5            5               5             5
0     1.0            5               5             5
0     1.5            5               5             5

Table 5: Effect of surfactants on 0.05 volume fraction colorants in 2%
HASE solution

                     Red iron oxide  Phthalo blue  Lampblack
%     % OP           dispersion      dispersion    dispersion
SDS   (EO)[.sub.40]  rating          rating        rating

0     0              5               5             1
0.25  0              1               4             2
0.5   0              1               3             1
0     0.5            1               5             2
0     1.0            1               5             3
0     1.5            1               5             2