Carboxyl functional polyesters have been prepared from 1,3-propanediol (PDO), neopentyl glycol (NPG), and a 76/24 terephthalic/isophthalic acid (TPA/IPA) combination. These polyesters were formulated with triglycidyl isocyanurate (TGIC) resins for evaluation as powder coatings. Coatings based on polyesters with PDO provided superior impact resistance and flexibility over those that contained no PDO. Overall, properties including hardness, adhesion, and chemical resistances are retained over a broad range of PDO concentrations. In addition, the viscosity of the polyester decreased as the level of PDO increased.

Synthesis of Carboxyl
Functional Polyesters

Carboxyl functional polyesters were synthesized in a 1-liter, round-bottom flask under a nitrogen purge using a two-stage process. In stage 1, raw materials including TPA, IPA, NPG and PDO (see Table 1) were charged to the reactor and the mixture was heated at 170-230?C to form a hydroxyl functional prepolymer. In stage 2, IPA was added to cap the hydroxyl groups and the esterification was continued to an acid value of 30 to 45. The total reaction time was approximately 10-15 hours. Dibutyltin oxide (0.4%) was used as a catalyst, and xylene and water were added to facilitate water removal during the reaction. Polyester compositions with molar substitutions of PDO for NPG from 0% to 100% are listed in Table 1.

Preparation of Powder Coatings

Polyester powder coatings prepared from PDO-derived polyesters were formulated with triglycidyl isocyanurate resin by way of equal equivalents of carboxyl/epoxy groups. PT-810 (TGIC) from Ciba was used as a crosslinking agent for the polyesters. Choline chloride (0.18%, Actiron CT-6 from Synthron Inc.) was used as a catalyst. A flow-control agent (Modaflow Powder III from Monsanto) and degassing agent benzoin (Uraflow-B from GCA Chemical Corp.) were also incorporated into the coatings. Pigmented powder coatings based on R-960 TiO2 (DuPont) at a pigment/binder ratio of 0.7/1 by weight were also evaluated. The final powder coating compositions are listed in Table 2.

All the materials were initially premixed in a high-speed mixer for 2 minutes to ensure homogeneous mixing, and the solids were then fragmented into small particles. The resulting intimate mixture was then continuously processed through a twin-screw extruder to produce a uniform viscous melt. The extrusion temperature was maintained at 80?C in both zones 1 and 2 at a processing speed of 120 rpm. Molten extrudates passed through a pair of water-cooled squeeze rolls afforded a friable product. The products were then pulverized using a hammer mill with liquid nitrogen fed slowly into the grinding chamber. Classification on an Alpine sieve afforded powders with particle sizes of 105 microns or less.

The final powders were electrostatically sprayed onto grounded cold-rolled steel panels (Q Panel S-36), and coating properties were evaluated after curing at

180degC for 18 minutes.

Results and Discussion

Characteristics of Polyester Resins: The PDO-based carboxyl functional polyester resins afforded properties similar to those based on NPG (see Table 3). Glass-transition temperatures (Tg) of the polyesters decreased with increasing PDO contents (see Figure 1). For example, the Tg values ranged from 64degC for the NPG polyester to 52degC for the 50% PDO polyester. Glass-transition temperatures for the carboxyl functional polyesters were 4-6? higher than the corresponding hydroxyl functional polyesters. This characteristic of the carboxyl polyesters is expected to improve the storage stability of the corresponding powder coatings. Tgs for the polyesters were reported for the second heating cycle by way of differential scanning calorimetry (DSC) at a scanning rate of 10?C/minute. The second heating cycle involved heating the samples to a melt and then quenching the resin prior to a second heating when the Tg was determined.

No crystallization or melting peaks were noted on the DSC curves of polyesters derived from NPG or NPG/PDO mixtures, indicating amorphous polyesters (see Figure 2). However, the polyester from pure PDO was a semi-crystalline polymer with crystallization and melting temperatures of approximately 109degC and 188degC, respectively. Therefore, partially replacing NPG with PDO up to 50 molar % provided amorphous polyesters suitable for coating applications. It should be noted that the carboxyl polyesters were similar to hydroxyl polyesters with respect to amorphous characteristics.

Reactivity of Polyesters

The reactivity of carboxyl functional polyester powders with TGIC resin crosslinking agent was studied by way of DSC at a scanning rate of 10?C/minute. DSC curves shown in Figure 3 and testing results listed in Table 4 indicate the onset of cure, exothermal peak and enthalpy are very close for all of the PDO-derived and NPG control coatings. Thus, polyesters based on NPG and PDO/NPG mixtures had similar reactivities with the TGIC crosslinking agent.

Processability

After premixing, the intimate powder mixture was continuously processed through a twin-screw extruder to produce a uniform viscous melt. The extrusion temperature was maintained at 80 degC in both zones 1 and 2 at 120 rpm. All of the powder mixtures from NPG and NPG/PDO polyesters were easily processed through the extruder (see Table 5). It was observed that the torque reduced with increasing the PDO content in the polyesters, indicating the reduction in the polyester viscosity. The powder based on 100%, 1,3-propanediol cannot be processed under the conditions due to its crystallinity and high melting point.

Storage Stability

The glass-transition temperature of polyester resins for powder coatings should be high enough to achieve good storage stability. The Tg of polyester/TGIC powder precursors was measured by way of DSC and listed in Table 6. As expected, powder coatings formulated with up to 50% PDO-derived carboxyl polyesters had very good storage stability, since they had Tg values over 50degC.

Storage stability tests were performed by placing powders in a capped jar at 40degC for 10 days. Powders were subsequently examined for free-flowing properties (lumps not easily broken) each day for 10 days. Those with free-flowing properties after 10 days passed the test.

Inclined Plate Flow

The inclined plate flow properties of powder coatings were measured according to the PCI standard method. It is a useful indicator of the degree of flow occurring during the curing of powder-coated parts. The inclined plate flow is related to the melt viscosity of the base resin and is influenced by the reactivity of crosslinking agent and the polyester resins. It was noted that the plate flow increased with increasing PDO concentrations in the PDO/NPG mixtures at both 175?degC and 190degC (see Table 7). Therefore, incorporation of PDO improved the flow properties of the powder coatings. The powders also had higher plate flows at 175degC than at 190degC, because of higher reaction rates at higher temperatures.

Gel Time Reactivity

Gel time reactivity is the time required for a powder to advance to a gelled state through a liquid phase at a defined temperature. The test was performed by rubbing the powder coating with the tip of a wooden applicator stick over a hot plate until a solid gel was produced. Gel times for polyester powder coatings were determined at 180degC according to the PCI standard method. As shown in Table 7, all the coatings provided similar gel times, which is in accord with DSC studies.

Impact Resistance

Front and reverse impact resistance were tested according to the ASTM D-2794 standard method. The results in Table 8 indicated that PDO significantly improved the flexibility of both polyester/TGIC clear and pigmented powder coatings. For instance, the impact resistance increased from 60/30 in-lbs for the coating based on pure NPG polyester to 130/130 in-lbs for 30% PDO and to 160/160 in-lbs for 50% PDO-derived polyesters at the film thickness around 3.3 mils. Moreover, the impact resistance was strongly dependent on film thickness with thicker films having less flexibility (see Figure 4). These results were comparable to the results obtained with hydroxyl functional polyester and carboxyl polyester/epoxy hybrid systems.

Flexibility - Conical Mandrel Bend and T-Bend Test

Conical mandrel bend tests were performed by bending the coated panels on a conical mandrel tester (Gardner Laboratory Inc., 1/8" diameter) over a period of 3 seconds. The testing results are listed in Table 9. All clear and pigmented coating panels passed the test, e.g., no cracking.

Flexibility measured by way of the T-bend test revealed that incorporation of 15% PDO provided a 1T coating whereas pure NPG resulted in a 2T value. Coatings containing 30% and 50% PDO provided very good flexibility with 0T values. In the case of the pigmented coatings, pure NPG resulted in a 3T value compared to a 2T when 15% PDO and 0T when 30% PDO was incorporated. Therefore, these results provide further data on the contribution of PDO to improved flexibility for the powder coatings.

Gloss

The values for 20deg and 60deg gloss for the polyester/TGIC clear and pigmented powder coatings are shown in Figure 5. Incorporation of PDO gave similar gloss values in both clear and pigmented coatings compared to pure NPG coatings.

Hardness, Adhesion, and MEK Double-Rub Resistance

All coatings evaluated had excellent adhesion to cold-rolled steel substrates (see Table 10). They passed the crosshatch tape adhesion test in accordance with ASTM D-3359-92 with a value of 5B. Replacing NPG with PDO had little effect on the final pencil hardness. In the case of MEK double rub resistance, PDO-derived polyesters showed similar values when compared to pure NPG for both clear and pigmented coatings. Therefore, coatings based on PDO/NPG mixture combined good film hardness, impact flexibility with high gloss, and excellent adhesion with no change in solvent resistance.

Chemical and Stain Resistance

Coatings exposed to 10% HCl, 10% NaOH, gasoline, and mustard for 24 hours had excellent acid resistance compared to the control. Gasoline and 10% NaOH had a very slight effect on both PDO and NPG polyester coatings after a 24-hour exposure. All coatings exhibited very good stain resistance to mustard. PDO-derived coatings had no effect on the stain resistance (see Table 11). The data is presented in the form of ratings with 10 representing no effect and 1 indicating the most severe deterioration.

Conclusion

Carboxyl functional polyesters have been synthesized from mixtures of 1,3-propanediol (PDO) and neopentyl glycol (NPG). Increasing molar concentrations of PDO, e.g., 0, 15, 30, 50, and 100%, gradually reduced the glass-transition temperature of the polyesters with all resins up to 50% PDO having Tgs greater than 50degC.

Powder coating formulations based on these polyesters and TGIC crosslinking agent were easily processed in an extruder. The viscosity of the polyester decreased as the level of PDO increased, hence improved flowability resulted. Polyesters containing up to 50% PDO in the polyol mixture had good storage stability due to their higher Tg. Coatings properties indicated that replacing NPG with PDO significantly improved both impact resistance and flexibility. Other properties including hardness, adhesion, and chemical resistances were retained over a range of PDO concentrations.