Self Crosslinking Polymeric Dispersants Used in Emulsion Polymerization

Acrylic dispersions have found widespread use in modern coatings technology. Their excellent durability makes them suitable for indoor and outdoor decorative paints, and they can be formulated into high-resistance coatings for industrial uses.

Several properties are important for high-performance coatings, including the following.

These properties can be controlled by several parameters in acrylic dispersion design, such as controlled particle morphology, polymer backbone compositions, polymer Tgs and the use of crosslinking. Additionally, the film formation process needs to be controlled, since good performance can only be achieved from waterborne dispersions after good film formation and entanglement of the polymer chains across the particle interfaces. Conventional acrylic dispersions contain surfactants used for stabilization of the polymer particles. Generally, these surfactants negatively affect the desired application properties, especially resistance properties and outdoor durability, which are very dependent on surfactants. Surfactants are sometimes referred to as "necessary evils" in acrylic dispersions.1

Figure 1 / Surfactant Migration

Surfactants in Emulsion Polymerization

Surfactants play a crucial role in emulsion polymerization. They are required for emulsification of the monomers, formation of micelles as polymerization loci and colloidal stabilization of the polymer particles. In addition, surfactants reduce the surface tension of the resulting dispersion, which is required for wetting of a substrate and for film formation. With surfactants, dispersions can be formulated into a coating because the surfactants will help avoid destabilization of the dispersion with the addition of formulation components. In fact, they may even be an aid to effective mixing of these components and the dispersion.

Surfactants are, in most cases, water soluble and mobile components in the film. They have a tendency to cluster together or migrate (see Figure 1), either to the film-air interface or the film-substrate interface.1 In Figure 2, surfactant exudation to the surface of an acrylic dispersion film is shown, where the exuded surfactant crystallized at the polymer air interface. The surfactants can in this way seriously affect the water sensitivity of the film, as well as the adhesion characteristics.

Surfactants that remain on the interfaces of the polymer particles will create hydrophilic channels through the film, which may cause undesirable effects such as water transport through the film to the substrate.

Several options have been studied to reduce the negative effects of surfactants while maintaining the positive aspects, such as the use of copolymerizable surfactants1-2 and polymeric surfactants. Copolymerizable surfactants will be chemically bound to the polymer particles and cannot migrate through the film or to the surface freely. As a result, films from acrylic dispersions prepared with copolymerizable surfactants should have better water resistance and adhesion properties.

There are, however, some issues associated with the use of copolymerizable surfactants. The reactive groups of these surfactants are often not of the same reactivity as the reactive groups of the monomers used, which may lead to poor incorporation or homopolymerization of the surfactants in the water phase. On the other hand, if incorporation is good the resulting dispersion can have high surface tension because there is no free surfactant left, which impairs wetting, leveling and formulation of the dispersion. To promote good copolymerization with the acrylic monomers, the reactive group of the surfactant should be in the hydrophobe of the surfactant, and such surfactants are hardly commercially available.

Figure 2 / AFM Image of Exudation of Surfactant from an Acrylic Dispersion Film Left is the height, right is the phase image

Alternatively, polymeric surfactants can be used.3 These are much less mobile than conventional surfactants and are usually very strongly absorbed to the particle surface because of their multiplicity of hydrophobes. Again, they will not easily form clusters or migrate to the surface of a film. These polymeric surfactants can be prepared by a variety of methods including bulk polymerization, suspension polymerization or emulsion polymerization.

It is also possible to prepare surfactant-free acrylic dispersions, which are solely stabilized by the anionic initiator (persulphate) end groups,4 but in general only low-solids systems can be prepared this way. The solids can be raised by adding ionic monomers, but generally the resulting systems have low tolerance toward the addition of formulating component.

Another route to obtaining surfactant-free dispersions is the preparation of self-dispersing polymers in organic solvents, which are subsequently dispersed into water.5 After dispersing, the organic solvents are distilled off to yield a secondary dispersion or artificial latex. Here the dispersing groups are usually acid groups that have been neutralized to salts by the addition of an amine. The ionized polymers have molecular weights of 10K-50K, are surface active and are known to be useful in the stabilization of hydrophobic polymers. In this case, core-shell particle morphologies are created where the ionized polymer forms the shell, and the hydrophobic polymer forms the core of the particle. The ionized polymer in this case functions as the stabilizer for the hydrophobic core.

Upon film formation of such secondary dispersions, the matrix of the film will consist of the ionized polymers and hydrophobic particles are dispersed in this matrix. Since the ionized polymers are usually low in molecular weight and contain fairly high levels of copolymerized and ionized acid, the matrix of the film will tend to be the weakest part of the system with regard to sensitivity to water and toughness of the film. The low molecular weight of the ionized polymers will aid the coalescence process and entanglement of the polymer chains to build up strength in the film.

Figure 3 / Film Formation of Acrylic Dispersions

Film Formation

Proper film formation6 is crucial to achieve a good performance of the final film. The film formation process can be divided in several stages7-8 (see Figure 3) where the entanglement step especially will determine the final properties. During the entanglement process, the interface between initially separate particles will gradually fade away. The strength of the film, both in mechanical and in resistance terms, will increase during the entanglement process.

Acrylic dispersions generally have very high molecular weights, much higher than conventional solventborne polymers, such as alkyds. This is beneficial for the overall film properties, but makes the entanglement step very slow. The buildup of properties in acrylic dispersion films may, therefore, take days or even longer. In some cases, optimal film formation is never achieved because the coalescents that keep the polymers mobile evaporate before entanglement is complete.

Especially high-Tg and high-molecular-weight acrylic dispersions can achieve very good application properties, but for good film formation, high levels of coalescents are required. With the ongoing drive toward more environmentally friendly coatings, the aim is to reduce this coalescent demand further with zero VOCs as the ultimate goal. In recent years, several options have been investigated and published aimed at combining the good properties of high-Tg polymer backbones with good film formation.

One method is the use of a very hydrophilic acid and, in some cases, low-molecular-weight polymer backbone. If a sufficient amount of acrylic or methacrylic acid is incorporated into the backbone, these dispersions can be either alkaline soluble or alkaline swellable.9-10 In both cases, the minimum film formation temperature (MFT) and the coalescent demand are significantly reduced. To maintain a workable balance between viscosity and solids content, low-molecular-weight polymers or oligomers are used. Since low-molecular-weight polymers are not very water- and chemical-resistant, and hydrophilic backbone polymers have poor water resistance, they are often combined with some form of crosslinking after or during film formation. Clearly the poor resistance properties are the most important drawback of this approach.

Applying Crosslinking to Improve Chemical Resistance

While good MFT-hardness balance can be achieved with hydrophilic oligomers, this approach has the disadvantage of inferior chemical resistance. This can be overcome by post crosslinking the low-molecular-weight polymer. Crosslinking can be effective in two ways: chemical reaction between polymer chains following interdiffusion and crosslinking at the interfaces of the particles. In the first case, a coherently formed film is fixed and the molecular weight of the chains present in this film is increased. In the second case, two particles are chemically bound by crosslinking during or just after film formation. Formation of a chemical link between particles has a similar effect on properties, such as entanglement between polymer chains, but can take place on a much shorter time scale. A variety of crosslinking reactions have been described in past years. There are so-called two-component crosslinking systems where one of the crosslinking components is added just before application of the dispersion. More desirable for this work are self-crosslinking (one-pot) systems where all reactive components are present and long-term storage stable. The crosslinking reaction can be triggered by the evaporation of water upon drying, a change of pH, or by curing at elevated temperature, where the crosslinking reaction is faster, or reactive groups are de-blocked. Examples of suitable crosslinking systems are the reaction of amines with epoxy functionality11 where either can be on the polymer backbone, the auto-oxidation of incorporated fatty acid groups,12-13 the self condensation of alkoxy-silane functionality,14 the self condensation of n-methylolacrylamide,15-16 metal-ion co-ordination with backbone functional groups such as acetoacetoxy15 groups or acid groups, and the reaction of acetoacetoxy groups with amines17-18 or acetoacetoxy groups with unsaturated groups in a Michael reaction.19

Results and Discussion

This article discusses the use of polymeric self-crosslinking dispersants for the preparation of surfactant-free acrylic dispersions20 with a core shell morphology.

The core consists of normal high-molecular-weight hydrophobic acrylic polymer. One purpose of the shell is to stabilize the core against flocculation and coagulation. To do this, the shell is made up from high-Tg hydrophilic alkaline-soluble polymer chains that are modified with hydrophobic components to make them highly surface active (such polymers will be referred to as polymeric dispersants).

Figure 4 / Surface Activity of Various Polymers vs. Concentration

Polymeric Self-Crosslinking Surfactant

Figure 4 compares the surface tension of aqueous solutions of such polymeric dispersants to polymethacrylic acid and a commercially available hydrophilic alkaline soluble acrylic oligomer and a commercially available styrene-acrylic acid oligomer. As can be observed, at very low concentration the surface tension of the polymeric dispersant solution is quickly reduced and equilibrates. The final surface tension is approximately 43 mN/m, while for polymethacrylic acid this is >60 mN/m.

This is likely caused by the multiple hydrophobic groups that a polymeric dispersant contains, which causes the polymer chain to stretch along the surface of the air water interface. The figure shows that using a commercially available hydrophilic alkaline polymer is far less effective in reducing the surface tension.

To demonstrate that the self-crosslinking polymeric dispersants are effective stabilizers for the preparation of acrylic dispersions, a REACT-IR experiment was performed. The REACT-IR probe detects small particle size (monomer swollen) particles and water-dissolved monomers. Large monomer droplets are outside the detection range of the REACT-IR probe. Monomer is added to a solution of self-crosslinking polymeric dispersant in which the probe is immersed.

Figure 5 / React-IR Reaction Plot and Profile Showing the Excellent Emulsification Properties of the Self-Crosslinking Acrylic Polymeric Dispersant

Figure 5 shows that the IR peaks from the monomer appear and grow during a period of about 10 minutes after addition of the monomers, after which equilibrium is reached. (In the left plot the peaks at a wave number around 1200. This same peak is plotted in intensity vs. time on the right. Some baseline shift will occur due to temperature variations during the reaction.) The change in this peak indicates that monomer is being absorbed into the self-crosslinking polymeric dispersant aggregates until equilibrium is reached.

A visual check at this stage confirms that no large monomer droplets remain, and all monomer isemulsified. When initiator is added at this point conversion of the monomer is almost instantaneous, with a large temperature exotherm. This is especially clear in the intensity vs. time plot on the right. This suggests that the polymerization mechanism has strong resemblance to that of a mini emulsion polymerization.

Figure 6 / React-IR Reaction Plot of SLS and Monomer

The resulting dispersion is stable and generally has a very low particle size, which confirms that these self-crosslinking polymeric surfactants are efficient stabilizers.

For comparison, a similar REACT-IR experiment was done with sodium lauryl sulphate as the surfactant (see Figure 6), and also with a commercially available styrene-acrylic acid oligomer as the stabilizer (see Figure 7).

Figure 6 does not show a monomer peak, nor does any reaction take place after addition of the initiator. This can be explained by the stirring of the experiment, which is done by a simple flat-blade stirrer at low rpm. While this method of stirring was adequate for emulsification of all monomer when the self-crosslinking polymeric dispersant was used, it proves insufficient for proper emulsification of the monomer by SLS. Visually the reaction mixture does not become white or milky, but stays a bit gray with clear inhomogenity. With this kind of inhomogenity it is clear that no small monomer droplets will be created.

The emulsification is so poor that even when radicals are formed only the very small amount of monomer dissolved in the water phase will react, but due to insufficient transport of monomer the reaction does not proceed. In the experiment with the commercially available styrene-acrylic acid oligomer again no monomer peak is observed (see Figure 7), indicating that no small monomer swollen particles are formed.

Figure 7 / React-IR Reaction Plot and Profile of Styrene-Acrylic Acid Oligomer and Monomer

The dispersion appears to be more milky than the SLS experiment, indicating that some monomer emulsion droplets have been formed. Upon addition of the initiator, a monomer peak appears and starts to slowly decay as it is being converted into polymer, indicating a normal emulsion polymerization that is rate limited by transport of monomer. A polymer peak is formed at the same time the monomer peak appears, and this peak slowly gains in intensity, clearly indicating that monomer swollen polymer particles are present.

The reaction mixture still has a gray texture, indicating that a lot of poorly emulsified monomer isstill present. Reaction is slow and about half of the monomer is converted in 30 minutes. Another addition of initiator restarts the reaction with the same slow conversion rate. Apparently the styrene-acrylic acid oligomer is a bit more effective as a stabilizer than sodium lauryl sulphate and does give some emulsification with this poor stirring, but no very small emulsified droplets are formed, and the reaction is very slow and with low conversion.

Figure 8 / TEM Image of a Dispersion Based on a Self-Crosslinking Polymeric Surfactant

From these comparative experiments it must be concluded that the self-crosslinking polymeric dispersant has a remarkable emulsification power, when compared to conventional surfactant or commercially available and commonly used surface active oligomers. Even at inefficient stirring sub-micron monomer emulsion droplets are created that react at very high rates when initiator is added.

To verify that a core shell structure is obtained, the dispersion was analyzed with TEM. The self-crosslinking polymeric dispersant was modified so it could be selectively stained vs. the core polymer. Figure 8 shows that the stained self-crosslinking polymeric dispersant (the dark colored material) is present in the particle shell and in the water phase.

Figure 9 / Film Formation of Core Shell Particle

The Effect on Film Formation

This hard hydrophilic self-crosslinking polymeric dispersant will have good film formation since it is an alkaline-soluble polymer. Entanglement, however, is not applicable in the self-crosslinking polymeric dispersants, since their molecular weights are just below the entanglement Mw.

Consequently, a very fast mixing of the low-Mw hard hydrophilic polymeric dispersant will occur. This is then followed by a crosslinking reaction leading to a crosslinked continuous hard phase containing discontinues soft domains. The formation of crosslinks at the particle interface will give an even better buildup of film strength as polymer chain entanglement would, and generally at a much shorter time scale. By focusing crosslinking at the particle interface it is used most effectively. Many of the crosslinking techniques are expensive, and should be used efficiently. Precrosslinking should be avoided as much as possible, since this will negatively affect entanglement during film formation.

Alkaline-soluble systems on their own generally lack sufficient resistance properties to be useful. To overcome this problem, we have used the polymeric dispersants as stabilizers in a subsequent emulsion polymerization. This way, core shell particles are created that have the polymeric dispersant as the shell, and a high-molecular-weight resistant polymer in the core (see Figure 9).

Table 1

Relatively small quantities of polymeric stabilizer are required for effective stabilization of the core. By introducing the same self-crosslinking mechanism in the core polymer, performance of the system can be upgraded even further. Now not only the polymeric dispersant crosslinks with itself to form a crosslinked matrix in the resulting film, but this matrix is also grafted onto the resistant high-Mw cores of the particles. This effect is demonstrated in Table 1.

As can be observed, crosslinking of the core polymer alone has no effect on the hardness MFT balance, compared to no crosslinking at all. If just the polymeric dispersant is self-crosslinking the hardness MFT balance is improved. But the best result is achieved if both the polymer and the polymeric dispersant crosslink.

Figure 10 / Film Morphology

In effect, a core shell polymer with a film forming hard shell polymer and a high Mw, resistant core is created, where the hard component will form the matrix of the film and will then crosslink with itself and the core polymer (see Figure 10). Both the core and the shell will contribute to the overall hardness, toughness, and resistance of the final film. No surfactant will be used and hence no surfactant exudation will take place (see Figure 11) and no hydrophilic channels to the film can be created.

Figure 11 / AFM Image of a Film Cast from a Dispersion Containing Polymeric Dispersant Left is height, right is phase image

The right photo in Figure 11 is a phase image of a film cast from a dispersion containing a self-crosslinking polymeric dispersant, which was recorded in AFM tapping mode. The light areas represent hard material and the soft areas represent soft material. It can be observed that the surface of the film is mainly hard material, which confirms that the surface of the film is covered with the polymeric crosslinked dispersant. This is beneficial for hardness and anti-blocking. In the left photo, the height differences are depicted, and it can be seen that the surface it quite smooth, and individual particles can no longer be observed, which confirms that film formation was efficient.

Table 2

Film Formation and Properties of Model Dispersions

In the system under investigation there are several design parameters that can be adjusted to optimize application properties.

Ratio between shell material and core material
Tg of both phases
Hydrophilicity of the core
Acid value of the shell phase

The effects of these parameters on film formation characteristics, surface hardness and blocking properties were investigated. The Tg of the high-molecular-weight core and the ratio between high Tg shell and the core obviously affected surface hardness (see Table 2).

Table 3

From the first two entries, the Tg of the core has a huge effect on surface hardness and MFT. Both these effects are, however, reduced on increase of the shell:core ratio to 1:1. It is apparent that for a system comprising equal amounts of core and shell material, the continuous hard phase will determine the hardness, whereas the Tg of the core has no influence on this property. The advantage of a low-Tg core lies in a reduction of the MFT and hence reduced cosolvent demand for good film formation. The acid value did not have a significant effect on surface hardness (see Table 3), which can be expected since the effect of the difference in acid concentration on theoretical Tg was compensated for by other monomers. Since the acid concentration will influence the degree at which water can act as a plasticizer, it is expected to affect the MFT of these emulsions.

Table 4

While the surface hardness values seem to be independent of acid value, MFT was indeed affected by the acid value. Especially at high shell:core ratio (entries 7 and 8) MFT was significantly reduced with increasing acid value while maintaining good surface hardness. This can partly be explained with increasing water plastization and partly by taking into account that at higher shell:core ratio smaller particles will be formed. It is well known that smaller particles tend to give lower MFTs than larger particles with the same composition. While particle sizes for the polymer emulsions with a shell:core ratio of 1:4 were found to be in the range of 250-400 nm, at a shell:core ratio of 1:1 the particle size were between 40 and 70 nm.

Entry 7 yielded a very good balance between MFT and surface hardness, 15 deg C and 121, respectively. Reducing the acid value of the shell led to a marginal decrease in surface hardness while the MFT increased with 20 deg C.

These results show that shell:core ratio, Tg of the high-molecular-weight core, and acid value of the shell can be useful tools to control the balance between film formation on the one hand and film properties on the other.

For most applications, anti-blocking is a required property. Normally, an easy way of introducing anti-blocking is to increase the Tg of the film, which will cause poor film formation properties. For our model emulsions with a shell:core ratio of 1:1 good film formation could be combined with excellent blocking properties. In Table 4 blocking properties of model emulsions are shown as a function of chemical composition of the core. For both core compositions very good blocking and early blocking results were obtained. Interestingly, these good blocking properties were obtained for films containing 50% by weight of very low Tg material. Table 4 shows that at these low core Tgs good surface hardness properties were found and acceptable MFT's. Especially the polymer having a core containing methyl methacrylate gave a very interesting set of properties, with an MFT of only 15 deg C, high surface hardness and excellent blocking properties.

Table 5

Application in High-Performance Coatings

Variation of the design parameters can have a big impact on the application properties. When the design is optimized for application in industrial wood coatings the following properties can be achieved (see Table 5). Because of the absence of surfactant the system will be low foaming. The dispersion has a very low particle size, which results in excellent film formation and high transparency. An additional benefit of the low particle size is the almost translucent optical appearance of the binder, which is comparable to traditional solventborne coatings. Also wood wetting, which is an important property for industrial wood coatings, is excellent. This is caused by the good wetting potential of the self-crosslinking polymeric dispersant and the small particle size of the latex, which makes penetration into wood pores efficient.

By modifying some of the design parameters, a surfactant-free self-crosslinking acrylic dispersion for joinery can be created. In this application, flexibility is of importance combined with good anti-blocking properties (see Table 6). For achieving good outdoor durability the coating will have to be flexible enough to compensate for the changing wood dimensions, which are weather dependent. The coatings need to be fast drying and should be applicable by a variety of application techniques, such as spraying and dipping. Again these dispersions have a very low particle size and the desired translucent appearance.

Table 6-7

These coatings can be formulated and pigmented, and will still retain all properties. When formulated at 18% PVC the flexibility is still >80%, which is sufficient to achieve good outdoor durability.

The dispersions based on self-crosslinking polymeric dispersants can also readily be used as binders for high-performance inks. The alkaline-soluble polymeric dispersant provides good reversibility to the system, a key property for use in inks. When the ink is dried further and crosslinking will take place the final ink will be highly resistant even to ammonia, where for conventional inks this is commonly an issue. The core shell morphology provides the combination of good block resistance combined with very high flexibility, which makes these systems also very useful for application on flexible substrates such as PE films. An added advantage is that these dispersions have excellent adhesion properties to treated PE film, even after prolonged water exposure. The self-crosslinking polymeric dispersant provides very good wetting properties to various substrates, and makes high pigment loadings possible to give high-strength inks. High-solids systems (55-60% solids) can be prepared with this technology that have the benefit of very fast drying. Table 7 shows some of the properties that can be achieved in an ink binder based on this technology.

While in this article the focus has been on high-Tg shells and low-Tg cores, the use of low-Tg shells combined with high Tg cores can also lead to very interesting systems for a variety of end uses.

Conclusion

Self-crosslinking polymeric dispersants are efficient stabilizers for use in emulsions polymerization, giving good wetting properties and paint formulation stability to the resulting polymer dispersion. Contrary to conventional surfactants, they do not foam or migrate through the final film - two very beneficial features. When high-Tg self-crosslinking polymeric dispersants are used for the preparation of a core/shell polymer where the shell contains the self-crosslinking polymeric dispersant stabilizing a hydrophobic high-Mw core polymer, a very good MFT/hardness balance can be achieved. Here the self-crosslinking polymeric dispersant brings high performance to the coating, forming the crosslinked matrix where a conventional surfactant would only degrade the properties of the coating. Because of their superb film forming properties combined with crosslinking features the resulting core/shell dispersions require little or no coalescent, and yet give excellent film formation, hardness, anti-blocking and resistances against chemicals, making such dispersion very useful for high performance coating and ink applications.21

Acknowledgement

The authors would like to thank Danny Visser, Harriet van der Sande, Dorina van Haeringen, Anton Peters Marc Roelands and Guru Satguru for their contributions to this work.

This article won an Outstanding Paper Award at the 2002 International Waterborne, High Solids, and Powder Coatings Symposium in New Orleans. Symposium sponsored by The University of Southern Mississippi Department of Polymer Science.

For more information on dispersants, contact NeoResins, Sluisweg 12, PO Box 123 5140AC, Waalwijk, The Netherlands; or visit www.NeoResins.com.

References

1 A Schultz et al. The use of non-migrating surfactants in emulsion polymerization; Proc. of the 3rd biennial North American res. conf. on the science and techn. Of emulsion pol / pol coll. (1999).

2 Holmberg, K. Prog. Org. Coat., 20 (1992) 325-337.

3 Bouvy, A. Eur. Coat. J. 11/96 822-826.

4 Juang, M.S.; Krieger, I.M.; J. Pol. Sci.: Pol. Chem. Ed. Vol. 14, 2089-2107 (1976).

5 Schlarb, B.; Gyopar, M.; Rau, S. Haremza; Prog. Org. Coat., 26 (1995) 207-215.

6 Peters, A.; Overbeek, G.; Buckmann, A.; Padget, J.; Annable, T. Prog. Org. Coat. 29 (1996) 183-194.

7 Winnik, M.; Wang, Y.; Haley, F. J. Coat. Techn., Vol. 64, No. 811, August 1992.

8 Daniels, E.S.; Klein, A.; Prog. Org. Coat., 19 (1991) 359-378.

9 Jin, Z.Z.; Zhu, Y. J. Coat. Techn., vol. 60, No. 757, February 1988.

10 Padget, J.C. J. Coat. Techn., Vol. 66, No. 839, December-1994.

11 Geurts, J.M.; v Es., J.J.G.S; German, A.L. Prog. Org. Coat. 29 (1996) 107-115.

12 Chen, F.B.; Bufkin, B.G. J. Appl Pol. Sc.,Vol. 30,4551-4570 (1985).

13 Pulin, P.; Raval, D.A.; Mannari, V.M. J. Coat. Techn. Vol. 70, No. 883, August 1998.