To conserve energy and also for heat-sensitive substrates, low-temperature-curing powder coatings are receiving increasing attention in the coatings industry.1,2 The problems associated with low-temperature-curing powder coatings have been the subject of several patents and publications.3-9 Two major problems associated with low-temperature-curing powder coatings are:
    1. The development of haze on the surface due to an undesirable phenomenon called, “blooming.”

    2. Poor appearance and physical properties.

This presentation describes the development of a nonblooming polyester resin and recent work on the development of low-temperature-curing powder coatings with improved appearance and physical properties.

Development of a Nonblooming Polyester Resin

Basic understanding of the nature of the blooming effect has been provided by Martinazzo, et al.3 The authors of this publication identified the chemical nature of blooming by a variety of modern analytical techniques and found it to be a 22-member cyclic oligomer formed by the condensation of two molecules of terephthalic acid (TA) with two molecules of neopentyl glycol (NPG) as shown in Figure 1, with a crystalline melting temperature of about 275 deg C. Another important fact revealed in this article was that the high-performance gel permeation chromatography (HP-GPC) of the polyester resin showed a small peak at the tail end of the molecular weight distribution (MWD) plot with a retention time (Rt) similar to that for the blooming substance. This made it clear that what is called blooming is not a byproduct generated during the bake from the interaction between the resin and crosslinker or by thermal degradation, but is a substance already present in the resin itself.

Based on this information, we scanned the MWD plots of a large number of polyester resins, two of which are shown in Figure 2. Plot 1 depicts MORKOTE® 200, a Rohm and Haas TGIC polyester resin based predominantly on NPG/TA with some minor modifying monomers. In this plot, at about 36 minutes Rt, there is a relatively small but sharp peak corresponding to the low-molecular-weight cyclic oligomer responsible for blooming as identified by Martinazzo, et al.3 Plot 2 is for another similar polyester resin, MORKOTE 400, in which a portion of NPG is substituted with 2-butyl-2-ethyl-1,3-propane diol (BEPD). In this plot, the peak due to the cyclic oligomer is much smaller. This observation is the basis of our experimental work leading to the development of a nonblooming polyester resin.

Experimental

All polyesters were synthesized by a standard, two-step, melt fusion process. In the first step, a hydroxyl-terminated polyester is formed by reacting monomers in a resin kettle in the presence of esterification catalyst and color stabilizer under nitrogen. The temperature rises slowly to 240 deg C while the water of esterification is collected. When the water-collection rate becomes immeasurable, and the acid value of the resin (AN) falls below 5, the first stage of the reaction is complete, providing a hydroxyl terminal polyester. In the second stage, isophthalic acid (IPA) is added to the reaction mixture and heating continued. Finally, the heating is continued under reduced pressure of less than 150 mm Hg until a steady acid value is obtained. Application testing of various resin samples was done in a white test formulation, and the testing for blooming resistance was done in a high-gloss black formulation, both listed in Table 1. To test blooming resistance of a resin, high-gloss black-powder-coated steel panels were first cured for 10 minutes at 375 deg F (191 deg C). The cured panels were subjected to heat at 225 deg F (107 deg C) in an air-circulation oven for varying periods of time, and the change in their gloss values determined. For comparison, a commercial polyester resin, MORKOTE 300 and Grilesta V73-94, a nonblooming resin, were also tested. The tests were done in triplicate up to 16 hours for each polyester and time period.

Results and Discussions

After some initial unsuccessful attempts by modifying process conditions and by varying the ratio of TA/IPA to minimize blooming, our efforts were directed towards optimizing the glycol composition.

A series of polyesters were synthesized by varying the ratio of NPG/BEPD. The resin compositions are listed in Table 2, and their performance properties and blooming resistance test results in Table 3. The results clearly show that within four hours of testing, MORKOTE 300, based on NPG/TA/IPA, shows a drastic reduction in initial gloss due to blooming. All other resin samples, based on varying levels of BEPD substitution, show substantially less reduction in gloss. This is comparable in some cases to the commercial resin Grilesta V73-94. MWD plots of the two resins, MORKOTE 200 and a resin sample N with 16% BEPD substitution, are shown in Figure 3. A MWD plot for a BEPD substituted resin clearly shows that a small peak identified as due to the presence of cyclic oligomer has more or less disappeared. It is for this reason, we believe, that the polyester resins in which a portion of NPG is substituted with BEPD, show reduced tendency to bloom, depending upon the degree of substitution. An in-depth understanding of the structure property relationship in BEPD-based saturated polyester resins has been provided by Ahjopalo, et al.4 They studied the reactivity of various diols and diacids under different reaction conditions. They used molecular modeling and various analytical techniques such as GPC, Matrix-assisted Laser Desorption Ionization time of flight mass spectrometry (MALDI) and MALDI-MS to characterize linear and cyclic oligomers. They found the probability of the formation of macrocyclic structures decreases in the order BEPD-PA(Phthalic anhydride) >BEPD-AA(Adipic acid) >BEPD-IPA>BEPD-TA.

Low-Temperature-Curing Powder Coatings

Low-temperature-curing powder coatings display poor surface appearance due to the lack of flow out of the melted powder film and poor physical properties due to inadequate cure. Also, in high-gloss coatings, the development of haze on the surface due to blooming becomes a serious problem if suitable nonblooming polyester resin is not used. To overcome these problems associated with low-temperature-curing powder coatings, we have developed semicrystalline carboxyl functional polyester resins, which form a co-reactable mixture with an amorphous nonblooming carboxyl polyester resin described earlier in a TGIC/polyester coating system. The use of semicystalline polyester resins in powder coatings has been a subject of recent publications.7, 10-13

Experimental

Semicrystalline polyester resins were synthesized by a two-step process described earlier. The properties of various resins are listed in Table 4. In DSC evaluation, each resin displayed a sharp melting peak (Tm2deg C), with a shoulder at a lower temperature (Tm1deg C). Raw materials used in preparing coating formulations are listed in Table 5.

Coating formulations were prepared as listed in Table 6. All ingredients were bag blended and extruded using a twin-screw extruder under the following conditions:

Feed Zone Temperature Cool

Mixing Zone Temperature 160 deg F

Screw Speed 400 rpm

Aluminum Oxide “C” was added in an amount of 0.2% to the extrudate prior to grinding. The extrudate was ground on a Brinkmann grinder and screened through a 200-mesh screen. The resulting powder was sprayed with a Nordson electrostatic spray gun on to a 3” x 5” clean cold rolled steel panel (Q-Panel) and cured at 300 deg F (149 deg C) for 10 minutes. The test results obtained with various powder coatings are listed in Table 6. Formulations 7 and 8 were cured at 2.5 - 3.5 mil coating thickness at 300 deg F for 15 minutes.

Results and Discussions

1. Samples 1 and 4 containing Resin A give excellent cure and mechanical properties, but sample 4 gives a much smoother coating, which is free from blooming.

2. Semicrystalline Resin B with higher “f” gives textured finish with excellent cure and mechanical properties.

3. When cured at 275 deg F for 15 minutes, samples 4, 5 and 6 all give excellent cure. Sample 6 (control - comparative example) does not develop full mechanical properties, whereas samples 4 and 5 under such low-temperature cure conditions develop very good mechanical properties and appearance.

4. A similar effect is observed in case of samples 7 and 8. In these samples, even at 45% pigment loading, sample 8 containing semicrystalline Resin A develops nearly full mechanical properties.

Conclusions

A nonblooming polyester resin has been developed in which a portion of NPG has been substituted with BEPD. GPC analysis of BEPD-substituted resins clearly show reduced levels of cyclic oligomers responsible for blooming. The probability of the formation of macrocyclic structures has been shown to decrease in the order of BEPD-PA>BEPD-AA>BEPD-IPA>BEPD-TA. By substituting a portion of amorphous carboxyl polyester resin with a semicrystalline resin, a coating with excellent appearance and physical properties can be obtained at temperatures as low as 275 deg F in 15 minutes.

For more information, contact Navin B. Shah, Rohm and Haas Powder Coatings, Research & Development Center, 3 Commerce Dr., Reading, PA 19607; phone 610/775.6690; or e-mail nshah@rohmhaas.com.

This paper was presented at the International Waterborne, High-Solids and Powder Coatings Symposium, February 26-28, 2003, New Orleans.

References

1 Rijkse K. Mod. Pt. & Ctgs. April 2001, 36-39.

2 Bailey, J. Ind. Pt. & Powder December 1997, 18-23.

3 ?Martinazzo, F. et al. 13th PRA International Conference November 15-17, 1993, Brussels, Hoechst Sara S.P.A

4 L. Ahjopalo et al. Polymer 41 2000, 8283-8290.

5 Morton International, Inc. U.S. Patent 5,880,223, March 1999.

6 ?McWhorter Technologies. U.S. Patent 5,637,654, June 1997.

7 ?Courtaulds Coatings (Holdings) Ltd. U.S. Patent 6,184,311 B1, February 2001.

8 ?Shell Oil Company. U.S. Patent 6,187,875 B1, February 2001.

9 ?Bergmans A. et al. Pt. & Powder Ctgs. October 2001, 17-19.

10 UCB, S.A. EP 1,067,159 A1, October 2001.

11 ?UCB, S.A. WO 99/32564, July 1999.

12 Rohm and Haas Company. U.S. Patent 6,309,751 B1, October 2001.

13 ?Johansson M. et al. J. Ctgs. Tech. Vol 72, No. 906 (2001), 49-54.

Rohm and Haas Company. U.S. Patent 6,294,610 B1, September 2001.