Safety, health and environmental issues continue to drive the trend toward lower VOC coatings.

An important feature of this crosslinking monomer compared to epoxy, aziridine or isocyanate curing systems is its relatively low toxicity, which makes it especially attractive for high-performance, environmentally friendly, nonhazardous coatings. It is for these reasons that AAEM-containing waterborne coatings have gained such high interest as evidenced by recent patents2-5 and literature reports.6,7

Polymers containing acetoacetoxy (AcAc) and other functionalities containing active hydrogen atoms, such as diketo and acetoacetamide, crosslink rapidly at room temperature in the presence of diamines. An alkali-soluble polymer with AcAc functionality dissolved in water at high pH might therefore be expected to gel upon the addition of a soluble primary diamine. The coatings described herein are based on the discovery that certain alkali-soluble acrylic polymers with high levels of AcAc functionality will not gel upon the addition of diamine but rather will form stable, translucent colloidal dispersions. Such polymers are useful in their own right and can further serve as crosslinkable support resins for emulsion polymerization.8

Ambient cure coatings based on these polymers can meet the requirements of long-term storage stability and rapid development of film properties for furniture and other applications. The compositions use non-polymeric diamines to effect the crosslinking reaction. The base polymers are conveniently made by a multi-stage emulsion polymerization process.


The crosslinking reaction between AcAc groups and diamine proceeds rapidly at room temperature (see Figure 2). This reaction is well known and has been widely investigated in high-performance two-pack solvent- and waterborne coatings systems.

It is known that in waterborne systems, such as those based on acrylic latex polymers, the AcAc moiety can exhibit hydrolytic stability problems, especially at elevated temperatures.9 The hydrolysis of AAEM followed by a decarboxylation reaction (see Figure 3) leads to the formation of a hydroxyethyl group and a beta-keto acid. The beta-keto acid in turn decomposes into acetone and carbon dioxide at a range of pHs and temperatures.10

In designing aqueous crosslinking polymers with AcAc functionality, it is important to minimize or eliminate this reaction. The glass-transition temperature (Tg) of a copolymer with an AcAc pendant moiety is considerably lower than the Tg with equivalent amounts of hydroxyethyl moieties. Therefore, significant hydrolysis will increase the MFT and impede the film formation process. It will also reduce the crosslink density in cured films and may negatively affect film properties.

Another problem that can affect these systems is pre-crosslinking. This can occur either in the wet state, which may give rise to gel formation, or during film formation just prior to particle coalescence. Both phenomena will negatively influence the film properties of the coating.11 However, storage-stable compositions can be obtained when acetoacetoxy functionality is combined in the right equivalent ratio with acid functionality, preferably in the same polymer molecule. Many alkali soluble or swellable acrylic polymers with carboxy and acetoacetoxy functionality exhibit exceptional stability in solution when mixed with diamines, making them especially useful for the design of waterborne self-crosslinking coatings.

The chemistry of an ammonia solution of the described AcAc-functional polymers involves an interaction of five different equilibrium represented in Figure 4.

The mechanism of drying and crosslinking of AcAc functional latices has been studied, and various publications on the subject have been issued.11,12 Those publications report the critical film formation conditions that allow crosslinking to take place before film formation proceeds. The novel proprietary systems discussed in this article do not have these critical film formation conditions because the polymer is in aqueous solution and the degree of pre-crosslinking is effectively reduced by way of the mechanisms mentioned (see Figure 4, A–E) as long as the pH is high. During the drying and curing process, the volatile amine(s) leave the film, shifting the equilibrium of reactions B and C toward the carboxylic acid and AcAc. The polymer forms a film, and crosslinking proceeds. A key reaction during film formation and curing is reaction E, in which ammonia is displaced by an amine group of the nonvolatile diamine.

Experimental - Preparation of Coatings Binders

Preparation of Acid-Functional Resins. The dispersions used to make these coatings are based on alkali-soluble resins that can be prepared by emulsion polymerization. A typical example was made using a monomer mixture of methyl methacrylate (MMA), butyl acrylate (BA), acetoacetoxyethyl methacrylate (AAEM) and methacrylic acid (MAA) in a ratio of 21:37:30:12 by weight. A chain transfer agent was used to obtain low molecular weight (Mw <30,000). The polymerization was conducted at 80ºC with a one-hour monomer feed using the formula shown in Table 1.

The finished emulsion was diluted to approximately 18% NV and neutralized with 30% ammonia to different calculated degrees of neutralization (DN), which resulted in clear or translucent liquids. The samples were labeled A–E.

A stoichiometric amount of hexamethylene diamine (1:1 ratio of -NH2 to AcAc) was added to part of each neutralized emulsion (Samples labeled AX–EX).

Preparation of Emulsion Polymers. The preparation of emulsion polymers using AcAc-functional resins as supports offers advantages of higher solids capability and a wider range of end-use properties. A particularly useful manner of making the polymers8 involves a two-stage process in which a partially neutralized resin is made in situ and used to support a second stage polymerization. By including a functional monomer such as AAEM in the first stage, the final addition of diamine results in a crosslinked colloidal dispersion. AAEM and/or other crosslinking monomers can be included in the second stage.

A typical two-stage polymerization of this type was conducted with MMA, BA, AAEM and MAA as monomers in the first stage, and S, BA, 2-ethylhexyl acrylate (2-EHA), and AAEM in the second stage. The calculated Fox Tg of stage 1 was 10ºC and stage 2 was 12ºC. A sample (Sample F) was drawn for testing after neutralization to a pH of 8.5 with ammonia. Sample F had an average particle size of 120 nm by QELS determination (Brookhaven BI-90 Analyzer). A diamine, Jeffamine™ EDR 148 was then added (1 equivalent per mole of AAEM) to a separate portion (Sample G) of the polymer. The characteristics of Sample G are shown in Table 2.

Preparation and Testing of Films

Acid Functional Resin Films. The above acid-functional resin samples, A–E and AX–EX, were drawn down with a #32 wire wound rod on aluminum panels, allowed to dry for 2 hours, and the films were tested for solvent (MEK) resistance and König hardness. The results are shown in Table 3.

The results show the effect of acetoacetoxy-diamine crosslinking on film hardness and solvent resistance. They suggest that the crosslinking reaction is relatively fast at room temperature and that the degree of pre- neutralization with ammonia has minimal effect on crosslinking effectiveness. Separate GC experiments showed that free hexamethylene diamine (HMDA) disappeared from such samples within about an hour after being added.

Additional studies have shown that the percent NV of these compositions must be kept within certain limits to avoid gel formation upon the addition of polyfunctional amines. Typically, solids in the range of 15–20% by weight can be achieved with good colloidal stability. The monomer selection, level of AAEM and acid, resin mw, and amine type are key factors that determine the maximum solids for stability. The HMDA-containing dispersions referenced in Table 3 were still colloidally stable after one year of storage at room temperature.

Emulsion Polymer Films. Films of the final emulsion polymers without and with the Jeffamine EDR-148 (Samples F and G, respectively) were cast on gloss cards and allowed to dry at room temperature for one week. They were then tested for acetone resistance (10 sec. spot test and double rubs) and 48 vol % denatured ethanol resistance (double rubs). The Jeffamine “EDR-148”-containing dispersion was stored for four weeks at 40ºC and was re-tested on a gloss card as before. The results, summarized in Table 4, give evidence of effective ambient curing and good storage stability.

Studies of the Crosslinking Reaction

13C NMR studies14 were conducted to help elucidate the nature of the crosslinking reaction. Polymer films were prepared from aqueous mixtures of emulsion polymer, ammonia and 5-amino-1-pentanol (5AP) as a model compound. These contained one equivalent of ammonia per equivalent of acid and one equivalent of 5AP per equivalent of acetoacetate. The molar ratio of acid to acetoacetate was approximately 1:1 in the polymer, although the acetoacetate was present in a slight excess. The mixture, at 20% solids, had a pH of 8.35 after initial neutralization with ammonia and a final pH of 10.3 after addition of 5AP. The films were dried for 26 hours at ambient conditions.

The NMR spectrum of the film revealed the complete disappearance of the acetoacetate resonance at d50 ppm and the appearance of two new resonances at d82 and d83 ppm, signaling the formation of the 5AP enamine and the ammonia enamine, respectively. The ratio of 5AP enamine to ammonia enamine was 2:1 (see Figure 5a). The polymer film was dried further at 60ºC for four days and was analyzed after 17 hours and 96 hours of heating. The ratio increased to 3:1 at 17 hours and 5.5:1 after 96 hours (see Figure 5b).

These results show that primary amine enamines are the major products of AcAc-containing polymer films, which indicate that crosslinking occurs when diamines are present in the system. This crosslinking can take place in the wet state and during film formation. However, these studies also indicate the presence of ammonia enamine in the film. The ammonia can be displaced by the primary amine over time, which increases the crosslink density. To summarize, NMR suggests that crosslinking occurs in the wet state and during film formation, but continues to advance as the films age at room temperature or under heating conditions. This helps to explain the observation that films prepared from these emulsion polymers exhibit high gloss and clarity but also have superior chemical resistance properties.

Tensile Property Measurements were used to further study the crosslinking reaction. A latex was prepared according to the general method described. It contained 10% AAEM based on total monomer weight and had a Tg (Fox) of –13ºC. The low Tg was chosen to permit the casting of films without the need for coalescing solvents. Two sets of films — one containing no diamine and the other with a stoichiometric amount of HMDA — were made by depositing latex polymer solutions in Teflon moulds and drying at ambient temperature for 48 hours.

Stress–strain measurements were performed on an “Instron 1011” with a 100 N load cell at a cross head speed of 12 mm/min. Four specimens with approximate dimensions of 2 x 16 x 60 mm were used for each sample. The computed values of Young’s modulus and the crosslink density (see Table 5) showed only modest increases due to the addition of the crosslinker. However, more significant effects of crosslinking were discernible in the values of the tensile yield strength and the maximum observed elongation given in Table 5. Tensile yield strength is defined as the stress at which the sample displays an increase in strain at constant stress.

Crosslink density was estimated from the Young’s modulus using the equation14 from the theory of rubber elasticity, Vc = E/3RT, for T>>Tg, where Vc is the crosslink density and E is Young’s modulus.

Performance Comparisons - Ambient Cure Systems

Optimization of this technology resulted in several experimental polymers having a desirable balance of properties for waterborne furniture coatings. It was found that developing the best performance with these polymers requires the proper selection of coalescents, thickeners and other additives. A combination of 5.7 phr butylglycol (EB), 2.8 phr propylene glycol phenyl ether (PPh) and 5.7 phr dipropylene glycol n-butyl ether (DPnB) gave an especially good balance of properties. Lowering the hydrophobic solvents in some cases left the films susceptible to cracking under varying temperature and humidity conditions. Butylglycol is used to decrease dry time and help incorporate the hydrophobic solvents.

Using this particular solvent blend, one of the experimental polymers, X-249, was compared with several other commercial one-package, self-crosslinking polymers (A, B and C) for chemical resistance, wet heat resistance, block resistance, and hardness development. Recommended starting point formulations from the suppliers were used to evaluate the remaining polymers as shown in Table 6.

Additional water was added to Formulas P and BF to reduce the spray viscosity to a range of 20–35 seconds, as shown in Table 7.

Formula CF showed poor response to associative thickeners. Rather than adding high amounts of thickener that might influence properties, no thickener was added to the formula.


The results of these evaluations are summarized in Tables 8–10. See the sidebar for details of the testing procedures.


One-pack ambient cure compositions have been developed that meet the requirements of long-term storage stability and rapid development of film properties for wood furniture and other applications. The acrylic binders in these coatings are based on acetoacetoxy chemistry and are conveniently made by a multi-stage emulsion polymerization process. Reactive diamines are used for crosslinking, which NMR analysis suggests takes place partly in the wet state but continues for some time after film formation. Stress-strain testing shows a marked increase in tensile yield strength in cured vs. uncured films. This data together with hardness, block and solvent resistance comparisons show the positive effects of crosslinking on coatings performance. The compositions appear to have advantages over latex systems, which rely on interparticle crosslinking for property development. This is related to the presence in the aqueous phase of the reactive AcAc groups for faster and more effective reaction with the diamine crosslinkers.


The authors thank Rein Janmaat for preparing the emulsions, and Glenn C. Calhoun, Dave L. Trumbo, and D. Sunil Jayasuriya for their contributions in the instrumental analyses.