Printer Friendly

Novel water-dispersible glycidyl carbamate (GC) resins and waterborne amine-cured coatings.

Abstract Water-dispersible glycidyl carbamate (GC) functional resins were synthesized and crosslinked using a water-dispersible amine to form coatings. GC functional resins are synthesized by the reaction of an isocyanate functional compound with glycidol to yield a carbamate (urethane) linkage (-NHCO-) and reactive epoxy group. The combination of both functionalities in a single resin structure imparts excellent mechanical and chemical properties to the coatings. Previous studies on the development of GC coatings have focused on solvent-borne coating systems. In this study, GC resins were modified by incorporating nonionic hydrophilic groups to produce water-dispersible resins. To determine the influence of the content of hydrophilic groups on dispersion stability, aqueous dispersions were made from a series of hydrophilically modified GC resins and characterized for particle size and dispersion stability. The composition of a typical, dispersed GC resin particle was predicted using Monte Carlo simulations. Stable GC dispersions were used to prepare amine-cured coatings. The coatings were characterized for solvent resistance, water resistance, hardness, flexibility, adhesion, and surface morphology. It was observed that GC resins were able to be dispersed in water without using any surfactant and by minimal mixing force (hand mixing) and produced coating films with good properties when crosslinked with a compatible waterborne amine crosslinker.

Keywords Glycidyl carbamate resins, Polyurethane, Epoxy, Hydrophilic modification, Waterborne coatings, Monte Carlo

Introduction

Glycidyl carbamate (GC) resins are produced by the reaction of an isocyanate functional resin with glycidol. GC resins contain both urethane (-NHCO-) and epoxy functional groups in their structure. (1) Polyurethane and epoxy resins are widely used in numerous commercial applications, such as coatings, composites, high performance polymers, etc. Polyurethane coatings offer excellent toughness, adhesion, flexibility, and chemical resistance. Epoxy coatings offer excellent corrosion and solvent resistance, adhesion, and versatility in crosslinking (curing) mechanisms. (2), (3) The combination of urethane and epoxy functional groups in GC resins imparts an excellent set of properties to GC-based coatings. Most of the GC coatings developed in the past were solvent-based coatings and exhibited an excellent combination of mechanical and chemical properties. GC resins can be crosslinked using amines or by self-crosslinking to produce high performance coating systems. (1), (4-6) GC resins have also been used to produce hybrid sol-gel coatings with an excellent combination of properties. (7-9) GC resins studied thus far can show a very high viscosity due to intermolecular hydrogen bonding; however, a previous study showed that GC resins can be modified with alcohols to reduce their viscosity up to 90%, and the resulting impact on properties of the coatings was studied. (10)

Environmental concerns have led to stringent regulations and a demand for reduction in volatile organic compound (VOC) emission. Methods, such as water-borne coatings, UV curing, high solids, and powder-coating, techniques, are all viable approaches to reduce VOCs. (11-13)

Waterborne polymers are an important class of materials for coatings due to increasing environmental regulations. Water as a solvent for coatings has good potential as a carrier or the solvent in the development of low or zero VOC environmentally friendly coatings. While water may have certain limitations in coating applications due to its high heat capacity, high surface tension, and slow and humidity-dependent evaporation rate, research on waterborne coatings has gained significant attention. In general, resins used as binders in coating applications are hydrophobic and are difficult to disperse in water. Common techniques adopted to disperse resins in water are the use of surfactants, incorporation of hydrophilic groups in the resin structure, or the synthesis of latex by emulsion polymerization. The hydrophilic component in waterborne coatings is an essential component to enable dispersion but may cause multiple undesirable issues such as water sensitivity, low chemical resistance, phase separation, and poor appearance. Resins for waterborne coatings should be designed considering the ease of applicability and performance attributes. It is also highly desirable to incorporate an appropriate level of hydrophilic groups in the resin structures to be able to obtain adequate dispersion by minimal shear force (i.e., hand mixing). (14-23) Film formation and the final coating properties of waterborne coatings are dependent on good coalescence between dispersed resin particles during the film formation process. A coalescing solvent, a volatile plasticizer, is usually employed to improve coalescence of particles. However, it may contribute to the VOC of coatings. (24) Also, a current trend is to minimize or eliminate the use of surfactants (external emulsifiers) in waterborne coatings. Surfactants remain in coatings after film formation and, over a period of time, they can diffuse to the surface, cause phase separation, increase hydrophilicity, and decrease the water resistance of coatings. (21)

There have been significant efforts in the development of waterborne resin and crosslinker systems to meet environmental regulations and demanding coating applications. Widely used coating binders, such as polyurethanes, epoxies, polyesters, and alkyds, are rendered dispersible in water by the incorporation of nonionic hydrophilic groups (e.g., polyethers) or water-reducible ionic (anionic and cationic) groups in resin structures. (24) Bayer Inc. has developed water-dispers-ible polyisocyanates for use in waterborne 2K poly-urethane coatings. (17), (21), (23) Air Products and Chemicals Inc., and Resolution Performance Products (Hexion Specialty Chemicals, now known as Momentive) have developed waterborne epoxy resins and crosslinkers by incorporating hydrophilic nonionic or ionic groups in epoxy resins or amine crosslinkers. (14), (15), (25-29)

In preliminary experiments, Edwards developed a water-dispersible GC resin using Bayer's hydrophilic polyisocyanate. The hydrophilic GC resin could be dispersed in water with the help of a surfactant. The coatings crosslinked with a waterborne amine cross-linker had good properties. (5)

The aim of the research presented in this article was to synthesize water-dispersible GC resins by incorporating nonionic hydrophilic groups in the resin structure and to study the influence of the extent of hydrophilic groups on coating properties. The nonionic hydrophilic group incorporated in GC resin was methoxy poly(ethylene glycol) (mPEG). The hydrophilic group content in the resin was varied by incorporating mPEG having different chain lengths and by varying the amount in the GC resins. The influence of the degree of hydrophilic group modification on coating properties was studied.

Experimental

Materials

Two polyisocyanatcs used were hexamethylene diiso-cyanate isocyanurate (Desmodur N 3600 with %NCO of 23) and a hydrophilic polyisocyanate (Bayhydur XP 7165 with %NCO of 18.3) provided by Bayer MaterialScience. The NCO equivalent weight of Desmodur N 3600 is 180 g/eq and that of Bayhydur XP 7165 is 245 g/eq. Glycidol was supplied by Dixie Chemical and was stored refrigerated to minimize the formation of impurities. Methoxy poly(ethylene glycol) (mPEG) oligomers of number average molecular weight 350, 550, and 750 g/mol, were obtained from Aldrich. K-KAT 6212, a zirconium chelate complex, provided by King Industries, was utilized to catalyze the isocyanate and hydroxyl reactions to form the GC resins. All the reagents were used as received without any further purification. Amine crosslinkers, Anquamine 731 and Anquamine 419, provided by Air Products, have amine hydrogen equivalent weights (g/H) of 200 and 284, respectively. Anquamine 731 is a water-based curing agent reportedly designed to emulsify and crosslink liquid epoxy resin without the use of any surfactants. Anquamine 419, waterborne curing agent, is a modified aliphatic amine used in waterborne epoxy dispersions. Surfactant, Triton GR-7M (anionic surfactant based on dioctyl sulfosuccinates), obtained from Dow Chemical, was used in selected coating formulations.

Synthesis of mPEG-modified GC resins

A 500-mL four-necked reaction vessel was used for the synthesis of the GC resins. The vessel was fitted with a condenser, nitrogen inlet, and Model 210 J-KEM temperature controller and mechanical stirrer. A water bath was used for heating and cooling the vessel. The synthesis was done in two steps. The first step involved the reaction of the isocyanate with mPEG, and the second step involved the reaction of glycidol with the remaining isocyanate groups. The stoichiometric equivalent amounts (% by mol) of mPEG based on isocyanate groups used to react with isocyanate were 5%, 10%, and 15%. The overall stoichiometric equivalent amount of isocyanate, mPEG, and glycidol based on-NCO and-OH groups during the synthesis was maintained at 1:1. The catalyst K-KAT 6212, in the form of solution in tertiary butyl acetate (1-2 wt%) was used for the reaction of isocyanate and hydroxyl groups.

The reaction vessel was charged with Desmodur N 3600 followed by the addition of the required amount of mPEG. The reaction mixture was stirred at 50-60 [degrees] C for about (45-60) min to ensure a homogeneous mixture. The catalyst (K-KAT 6212) amount added after the mixing time was 0.03 wt% (of the total reaction mass). The reaction was continued for (4-5) h at about (70-85) [degrees] C before the addition of glycidol. The amount of glycidol added was divided over (3-4) intervals over (4-5) h. Glycidol addition in the first interval was done at 45 [degrees] C. The subsequent addition of glycidol was done at the temperature of about (55-60) [degrees] C, and this temperature was maintained until the isocyanate peak (at 2271 c[m.sup.(-1)]) had disappeared in FTIR. A GC resin was also synthesized by reacting hydrophilic isocyanate, Bayhydur XP 7165, with glycidol. The reaction of Bayhydur XP 7165 with glycidol was performed using K-KAT 6212 as a catalyst.

Characterization

The FTIR measurements were performed using a Nicolet 8700 FTIR spectrometer from Thermo Scientific. Sample aliquots were taken and coated on a potassium bromide salt plate. Spectra acquisitions were based on 64 scans with a data spacing of 1.98 c[m.sup.(-1)]. The change in band absorption of isocyanate (2272 c[m.sup.(-1)]),-OH and -NH (3750-3000 c[m.sup.(-1)]), amide (1244 c[m.sup.(-1)], and epoxide (910 c[m.sup.(-1)]) bands were utilized to follow the reaction progress.

13[.sup.C] NMR was carried out using a JEOL-ECA (400 MHz) NMR spectrometer coupled with an auto-sampler accessory. The spectra were run at 24 [degrees] C with 1000 scans. All the spectra were collected by dissolving (50-70) mg samples in 0.7 mL CDCI3. The spectra were analyzed using Delta NMR processing and control software (Version 4.3.5).

Epoxy equivalent weight (EEW) of the resins was determined by titration with hydrogen bromide (HBr) according to ASTM D1652. A required amount of resin (0.8-1.0 g) was dissolved in 5-10 mL of chloroform and was titrated against a standardized HBr solution prepared in glacial acetic acid. The indicator used was a solution of crystal violet in glacial acetic acid. The end point of the titration was the appearance of a permanent yellow-green color.

The particle size was determined using dynamic light scattering (DLS) with a Nicomp 380 Submicron Particle Sizer (Particle Sizing Systems, Santa Barbara, CA). The experiments were carried out at room temperature. The samples for particle size analysis were prepared by dispersing the GC resins into water (30% solids) using a high-speed laboratory homogenizer (5000 rpm). The particle size was determined periodically for samples that did not show permanent agglomeration and settlement.

Coating formulation preparation

Waterborne coating formulations were made in plastic cups by dispersing the required amount of resin in water (deionized water) using a wooden stirrer. Surfactant, Triton GR-7 M, was used in selected formulations to disperse the GC resins in the water. To the dispersed resins, the required amount of amine crosslinker was added, and the resulting formulations were kept at ambient for about 5 min to stabilize the formulations. After the stabilization time, the coating formulations containing crosslinker were mixed by hand using a wooden stirrer for about 5 min and kept at ambient for about (10-20) min (induction time) before making drawdowns. The films were drawndown at 8 mils (200 [mu]m) wet thickness on steel panels (smooth-finished Q panels, type QD 36, 0.5 x 76 x 152 mm) cleaned with p-xylene. The coatings were cured at ambient conditions for about 2 weeks before determining their water resistance, solvent resistance, and other coating properties. Dry film thickness of the coatings was between 65 and 70 [mu]m. The free films for tensile testing were obtained by carefully removing the cured coating films from the steel substrate using a razor blade.

For the salt spray experiments, the coatings were applied in three layers onto steel and treated aluminum panels. The steel panels were cleaned with p-xylene for degreasing before applying coatings on them. The aluminum panels used in this experiment were cleaned using MEK for degreasing and Brulin Cleaner (Formula 815MX) with abrasive pad. The aluminum panels were further treated with deoxidizer solution (35% butanol, 25% isopropanol, 18% orthophosphoric acid, and 22% volume deionized water), and Alodine 5700 (chromate-free conversion coating). The treated aluminum panels were kept at ambient overnight before applying coatings on them. The first coat was applied at 3 mils (75 [mu]m) on steel (QD) and aluminum (A12024-T0) substrates, and the coatings were kept at ambient for 2 days. The second coat was applied at 4 mils (100 [mu]m), and the coatings were kept at ambient for 2 days. The third and final coat was applied at 6 mils (150 [mu]m), and the coatings were cured at 80 [degrees] C for 1 h. The coatings were kept at ambient for 8-9 days before salt spray experiments. Dry film thickness of the coatings was between 85 and 95 [mu] m. The free films for DMA were obtained by carefully removing the cured coating films from the steel substrate using a razor blade.

Coating performance

Initially all of the coatings were characterized for water resistance, solvent resistance, flexibility, hardness, contact angle, and adhesion. The coatings which had good water and solvent resistance were selected for further characterization. The selected coatings were characterized to determine their tensile properties (elongation at break and Young's modulus), glass transition temperature, dynamic mechanical properties, thermal stability, barrier properties, and corrosion resistance.

Water resistance of the coatings was determined by water drop test and water double rubs test. The water drop test was carried out by placing a drop of water (approximate weight ~0.05 g) on the coating surface. The water drop was covered with a glass slide (2 cm x 2 cm) to avoid evaporation of water. The glass slide was removed from the coating surface after 1 h. The surface of the coating was examined for defects such as bubble formation, film delamination, or any permanent marks. Six replicates of each coating formulation were tested. The number of coatings that did not show any surface defects, bubble formation, delamination, and permanent marks was reported. Thus, a water resistance test result reported as 6 indicates that all the six replicates passed the test (no surface defects), and the water resistance was the best. MEK double rubs were carried out according to ASTM D 5402 to assess the solvent resistance and development of cure. A 26-oz (737 g) hammer with three layers of cheesecloth wrapped around the hammerhead was soaked in MEK. The hammer head was rewet with MEK after 30-50 double rubs. The number of double rubs was noted once the metal surface of the panel was visible due to removal of the coating layer during the test. The water double rub test was carried out in a similar way as that of the methyl ethyl ketone (MEK) double rub test by replacing MEK with deionized water. For the water double rub test, the number of double rubs was noted once the formation of bubble, film delamination, permanent mark, or appearance of metal surface due to the removal of coating was observed. To determine relative hydro-philicity/hydrophobicity of the coatings, water contact angle was measured using a First Ten Angstroms FTA 100 series instrument.

Konig pendulum hardness of the coatings was measured following ASTM D 4366. The hardness test results are reported in seconds (s). Reverse impact strength of the coatings was determined following ASTM D 2794 using a Gardener impact tester. The maximum drop height was 43 in (1.09 m), and the drop weight was 4 lb (1.814 kg). Crazing or loss of adhesion was noted, and inch-pounds (in-lbs) were reported at film finish failure. Samples that did not fail were noted as having an impact strength of >172 in-lbs (1.98 kg m). The conical mandrel test was also used according to ASTM D 522 for the determination of flexibility of the coatings. The results of the flexibility test were reported as the length of a crack (cm) formed on the coating during the test. Cross hatch adhesion of the coatings was evaluated using a Gardco cross hatch adhesion instrument following ASTM D 3359. Tensile testing of the coatings was performed using Instron 5542 instrument. The test specimens were prepared according to ASTM D 638. The test was carried out at 10 mm/min at ambient conditions. Elongation at break and Young's modulus of the coatings were recorded. An average of at least three samples is reported.

Differential scanning calorimetry (DSC)

A TA Instruments Q1000 differential scanning calorimeter (DSC) coupled with an auto sampler accessory was employed to determine the glass transition temperature ([T.sub.g]) of the coatings. DSC experiments were performed by placing a sample into the conventional aluminum pans. The samples were subjected to a heat-cool-heat cycle. The samples were heated to 200[degrees]C and then cooled to -75[degrees]C, and held there for 5 min. DSC thermograms were taken from -75 to 250[degrees]C at a heating rate of 10[degrees]C/min. Glass transition temperature was determined as the temperature of the inflection at the mid-point.

Thermogravimetric analysis (TGA)

TGA was performed using a TA Instruments Q500. Temperature was ramped from ambient to 800[degrees]C with a ramp rate of 10[degrees]C/min. A nitrogen atmosphere was used during the test. Weight retained was plotted as a function of temperature.

Dynamic mechanical analysis (DMA)

A TA Instruments Q800 Dynamic Mechanical Analysis system was employed to determine the viscoelastic properties of the cured coating films. The dimensions of free films used were of 23-26 mm in length, 5 mm in width, and 0.09-0.1 mm in thickness. Poisson's ratio was assumed to be 0.4 for all of the coatings films. The experiments were carried out within a temperature range of -20 to 200[degrees]C with a temperature ramp rate of 5[degrees]C/min at a frequency of 1 Hz. The storage modulus values (E') in the rubbery plateau region (well above [T.sub.g]) are used to calculate the crosslink density of the coatings. Equation (1) was used to calculate crosslink density (ve) of the coatings (30), (31):

E' = 3veRT (1)

where E' = storage modulus (Pa); ve = crosslink density (mol/L); R = gas constant (8.3 J/K/mol); and T = temperature (K).

Atomic force microscope (AFM)

The atomic force microscope (AFM) used was a Dimension 3100[R] microscope with a Nanoscope IIIa controller (Digital Instruments, Inc., California). Experiments were performed using tapping mode in air at laboratory conditions using silicon probes with force constant 2.8 N/m and nominal resonant frequency 17 kHz. A sample area of 20 [mu]m x 20 [mu]m was scanned.

Salt spray test

In the salt spray test, the coatings were exposed to a continuous salt spray (5% NaCl in deionized water) at 35[degrees]C for 10 days. Images of the coatings were taken periodically by scanning.

Results and discussion

Synthesis of water-dispersible GC resins

The synthesis of the water-dispersible GC resins was accomplished by carrying out the reaction of the desired amount of mPEG with HDI polyisocyanate, followed by reaction of the remaining isocyanate functional groups with glycidol. The various amounts of molecular weight and mol% of mPEG used for reacting with isocyanate groups were 350, 550, and 750 and 5%, 10%, and 15%, respectively. Scheme 1 shows the reaction scheme for the synthesis of the water-dispersible GC resins.

A control GC resin was synthesized by reacting glycidol with a commercial hydrophilic isocyanate (Bayhydur XP 7165), described as an isocyanurate resin condensate derived from hexamethylene diisocy-anate containing a polyether chain (the exact molecular weight and mol% of polyether chain are a trade secret of Bayer MaterialScience). Table 1 shows the GC resins synthesized in this study.
Table 1: GC resins synthesized

Resin          mPEG molecular      Mol% of    EEW
                   weight (g/mol)     mPEG  (g/eq)

R1                            350        5     416
R2                            550        5     369
R3                            750        5     391
R4                            350       10     324
R5                            550       10     373
R6                            750       10     304
R7                            350       15     485
R8                            550       15     417
R9                            750       15     500
R10          Made from commercial              358
(control)  hydrophilic isocyanate


[ILLUSTRATION OMITTED]

[FIGURE 1 OMITTED]

Characterization of the structures of the synthesized GC resins was done using (13)C NMR. Figure 1 shows the NMR spectrum of resin R7. The peaks at 45 and 50 ppm for epoxy group and at 59, 70, and 72 ppm for mPEG molecule indicated that epoxy group and mPEG were incorporated into the structure of GC resins.

Since only a fraction of the initial isocyanate groups are reacted with mPEG, the synthesized resins consist of a mixture of hydrophilic molecules containing mPEG and hydrophobic molecules containing no mPEG. It should also be noted that while the poly-isocyanurate is illustrated as being a trifunctional molecule, the commercial material is composed of trimer (~70%), pentamer (<20%), heptamer, and higher molecular weight oligomers (<15%). A Monte Carlo simulation was performed to understand the possible distribution of hydrophilic and hydrophobic molecules in GC resins. A commercial software package, DryAdd-Pro+ v 4.33, was used for the simulation. The simulation software creates a distribution of the reactive sites and lets these sites react according to user-specified information and thereby simulates the individual random reaction events of a real-time reaction. For the simulation, the composition of the commercial polyisocyanurate used in this study was considered to be 70 wt% of trimer, 15 wt% of pentamer, and 15 wt% of heptamer. The water-dispersible resin selected for the simulation was R2, IGC-mPEG550-5%. The simulation was carried out by specifying the two steps during the synthesis. The first step was the reaction of mPEG with the fraction of isocyanate groups in the isocyanurate, and the second step was the reaction of glycidol with the remaining isocyanate groups in the isocyanurate. The simulation provided the number and weight distribution of the hydrophilic (containing mPEG) and hydrophobic (no mPEG) molecules in the final resin composition. The weight and number distribution of the hydrophilic and hydrophobic species in the resin having a significant amount is shown in Table 2. In the simulated composition of the resin, trimer species constituted the highest weight, and number fraction for both hydrophobic as well as hydrophilic molecules, which is logical since the initial polyisocyanate is primarily comprised of trimer. The largest amounts of the hydrophilic molecules were based on the reaction of trimer with one mPEG and two glycidol molecules (T-1mPEG-2GDL). Based on the simulation, the hydrophobic and hydrophilic molecules, which constituted to the major fraction of the resin composition, are shown in Fig. 2.
Table 2: Simulated weight and number distribution of hydrophobic
and hydrophilic molecules in the GC resin

Composition            Wt%   % by number

Hydrophobic molecules
  T-3GDL                 54.9         68.5
  P-4GDL                 10.5          8.4
  H-5GDL                  9.5          5.5
  Total                  74.9         82.4

Hydrophilic molecules
  T-1mPEG-2GDL           14.3         10.8
  T-2mPEG-1GDL            1.1          0.6
  P-1mPEG-3GDL            3.1          1.7
  H-1mPEG-4GDL            3.4          1.5
  Total                  21.9         14.6
T = trimer, P = pentamer, H = heptamer, GDL = glycidol


[FIGURE 2 OMITTED]

Dispersion stability

Dispersions of the synthesized GC resins were made at 30% solids using a high-speed homogenizer. The dispersions were kept at ambient conditions and their dispersion stability after 1, 6, and 14 days was evaluated. Dispersion stability was determined by visual examination to determine if phase separation (settlement) occurred in the test samples, and the results are shown in Table 3. In general, dispersions of relatively low hydrophilic content resins showed phase separation (settlement), while dispersions of more highly hydrophilic resins were stable and showed no phase separation. Particle size analysis was performed on the dispersions that did not show any phase separation. Dispersion stability was found to be strongly dependent upon mPEG chain length and its amount in the GC resins. Resin R1, being the most hydrophobic resin as it contains the smallest mPEG chain length (Mn of 350) in the smallest amount (5 mol%) could not be dispersed in water using HSD and remained phase separated. Control GC resin, R10, showed similar behavior to that of R1; it did not go into water by HSD and remained phase separated. Resin R2 containing mPEG of molecular weight of 550 at 5 mol% showed better dispersibility compared to that of R1. Dispersion made with R2 showed formation of agglomerates at the bottom of the container. However, the agglomerates could be easily redispersed with hand mixing resulting into a good dispersion. Resin R4 showed some phase separation after 6 days but formed a good dispersion following hand mixing. On the fourteenth day, R4 had resulted in a viscous nonflowable dispersion that upon further dilution with water to obtain a flowable dispersion contained no agglomerates. Resin R7 dispersion was similar to that of R4 dispersion except that R7 formed large agglomerates after diluting with water. Dispersion of resin R5 did not phase separate on the sixth day. The dispersion of resin R5 was found to be slightly phase separated on the fourteenth day but could be redispersed well by hand mixing. Dispersions of resins R3, R6, R8, and R9 showed no phase separation after 14 days. Particle size analysis of these dispersions did not indicate the formation of large agglomerates over 14 days. Resins R3, R6, and R9 contained mPEG of 750 molecular weight at levels of 5, 10, and 15 mol%. Resin R8 contained mPEG of molecular weight 550 at 15 mol%. Thus, the chain length as well as the level of modification influenced the dispersion stability.
Table 3: Dispersion stability of GC resins

Resin                  1 day                  6 days

               Dispersion  Particle   Dispersion  Particle
               stability   size (nm)  stability   size (nm)

R1             C                   -  c                   -
R2             B                   -  B                   -
R3             A                  25  A                  11
R4             B                   -  B                   -
R5             A                  24  A                  31
R6             A                  15  A                  11
R7             B                   -  B                   -
R8             A                  15  A                   9
R9             A                   4  A                   9
R10 (control)  C                   -  C                   -

                      14 days

               Dispersion  Particle
               stability   size (nm)
R1             C                   -
R2             B                   -
R3             A                  23
R4             D                   -
R5             B                   -
R6             A                   4
R7             E                   _
R8             A                  13
R9             A                   3
R10 (control)  C                   -

A = no phase separation, B = partial phase separation redispersible
by hand, C = did not disperse in water by HSD and remained phase
separated, D = nonflowable viscous dispersion became flowable after
addition of water and hand mixing, no agglomeration, E = nonflowable
viscous dispersion, showed large agglomerates after addition of
water and hand mixing, "-" particle size experiments not performed
because dispersions showed phase separation


The Monte Carlo simulation performed to determine the distribution of the hydrophilic and hydrophobic molecules in hydrophilic GC resins showed that the major fraction of the hydrophilic molecules is trimer with one mPEG and two GC groups. Based on this information, a schematic representation of a typical, dispersed particle of GC resin in water is shown in Fig, 3. Figure 3 shows that the particle consists largely of hydrophobic GC molecules (containing no mPEG) stabilized by hydrophilic GC molecules with mPEG chains extending out into the aqueous medium. GC resins contain strong hydrogen bonding groups such as urethane (-NHCO-) and carbonyl (-CO) responsible for their high viscosity. The high viscosity of neat GC resins resulting from hydrogen bonding indicates a strong interaction among GC molecules which may make them less favorable for interaction with water molecules.

Waterborne coating formulations from the GC resins were made by mixing the components in water by hand. The amine crosslinker (Anquamine 731) was then added to the dispersion of GC resin in water made with hand mixing. GC resins (except Rl and R10) were dispersed in water without using any surfactant. A small amount of surfactant was used to aid in dispersing resins R1 and R10, because these resins could not be dispersed solely in water by hand without surfactant. Table 4 shows the coating formulations of the GC resins used in this study. Films were drawdown on steel panels and were cured under ambient laboratory conditions for about 2 weeks before evaluating their water resistance and other coating properties.
Table 4: Coating formulations of the GC resins

Coating          Resin       Resin  Water  Surfactant  Crosslinker
formulation                (wt%)  (wt%)    (wt%)        (E:A)

F1           R1               62   37.3         0.7          1.1
F2           R2               66     34           0          1.1
F3           R3               63     37           0          1.1
F4           R4               66     34           0          1.1
F5           R5               67     33           0          1.1
F6           R6               67     33           0          1.1
F7           R7               67     33           0          1.1
F8           R8               66     34           0          1.1
F9           R9               66     34           0          1.1
F10(control  R10(control)     63   36.3         0.7          1.1

The crosslinker used was Anquamine 731


[FIGURE 3 OMITTED]

[FIGURE 4 OMITTED]

[FIGURE 5 OMITTED]

Water and solvent resistance

Table 5 shows the water and solvent resistance of the coatings. Water resistance of the waterborne GC coatings was determined by water drop and water double rubs tests. Water resistance of GC coatings was found to depend on the molecular weight and mol% of mPEG used in the resin. Coatings made from formulations F1, F2, and F4 exhibited excellent water resistance. The formulations F1, F2, and F4 were made from resins Rl, R2, and R4, respectively. Resins R1 and R2 contained 5 mol% mPEG having molecular weights of 350 and 550, respectively. Resin R4 contained 10 mol% mPEG having molecular weight of 350. The water resistance of the coating by water drop test made from the control GC resin, R10 (formulation F10) was inferior to that of the coatings made from the resins R1, R2, and R4 (formulations F1, F2, and F4). Addition of a small amount of surfactant (0.7 wt%) in coating formulation F1 to disperse GC resin Rl, made with low extent of hydrophilic modification (mPEG molecular weight of 350 at 5 mol%), did not appear to affect the water resistance of its coating.
Table 5: Water and solvent resistance of the coatings
Coating Water drop test Water double MEK double
              (6 = Best)  rubs  Rubs
F1                     6  >400  >400
F2                     6  >400  >400
F3                     3  >400  >400
F4                     5  >400  >400
F5                     2  >400  >400
F6                     0  >400  >400
F7                     2  >400  >400
F8                     0   325  >400
F9                     0   245  >400
F10(control)           4  >400  >400


Coatings became more hydrophilic with an increase in mPEG molecular weight and amount which resulted in a decrease of their water resistance. Coatings made from formulations F8 and F9 showed poor water resistance compared to that of the other coatings. The formulations F8 and F9 were made from the resins R8 and R9 and contained the large extent of hydrophilic part, 15 mol% mPEG having molecular weights of 550 and 750, respectively.

All of the GC coatings made in this study exhibited excellent solvent resistance with MEK double rubs values reaching above 400. The modification of GC resins by mPEG carried out in this experiment did not affect the solvent resistance of the coatings.

Water contact angle

The water contact angle of the GC coatings was determined to understand the relative hydrophilicity/ hydrophobicity of the coatings surfaces. Results of the contact angle measurements correlated well with the results of the water resistance of the coatings. Figure 4 shows a plot of water contact angle of the coatings. Increase in molecular weight and mol% of mPEG in GC resins decreased the water contact angle indicating the increase in hydrophilicity of the coatings. Water resistance of the coatings shown in Table 4 followed a similar trend. Increase in hydrophilicity of the coatings resulted in a decrease of water resistance of the coatings. Water resistance evaluated through the water drop test, and water double rubs of the coatings made from resins R8 and R9 (formulations F8 and F9, respectively) was lower compared with that of the other coatings. Coatings made from resins R8 and R9 showed relatively lower water contact angle (18[degrees] and 17[degrees], respectively) compared with that of the coatings made from the other resins. Resins R8 and R9 contained a large amount of the hydrophilic portion, with 550 and 750 molecular weight mPEG respectively, at 15 mol%. The results of water resistance and contact angle experiments suggest that the chain length as well as the amount of mPEG in GC resins influenced the relative hydrophilicity of the coatings.

Other coating properties

Initially, waterborne GC coatings had been prepared using a different crosslinker, Anquamine 419. However, the coatings crosslinked with Anquamine 419 had poor solvent and water double rubs <10), and a hazy or opaque appearance. On the other hand, waterborne GC coating crosslinked with Anquamine 731 cross-linker had good solvent and water double rubs (>245), and transparent and glossy appearance. AFM analysis on Anquamine 419 and Anquamine 731 crosslinked coatings was performed to understand the surface roughness of the coatings. Figure 5 shows the AFM images of the several of the GC resins crosslinked with the two different crosslinkers. Figure 5b also shows the reflection image over GC coating crosslinked with Anquamine 731. The AFM images in Fig. 5a show valleys, hills, and holes on the GC resins crosslinked with Anquamine 419. The images in Fig. 5a indicated incompatibility between the GC resin and the cross-linker, and poor coalescence between the resin and crosslinker particles. The AFM images in Fig. 5b show no valleys, hills, and holes on the surface of the coatings crosslinked with Anquamine 731. The images in Fig. 5b indicated good compatibility and coalescence between resin and crosslinker particles. Figures 5a and 5b also show the rms roughness of the coatings and indicate that the GC coatings based on Anquamine 731 were smoother compared to that of the GC coatings based on Anquamine 419.

Table 6 shows the performance of the GC coatings crosslinked using Anquamine 731. All of the GC coatings showed no cracks in conical mandrel bend test indicating excellent flexibility. Reverse impact test showed excellent resistance of the coatings to rapid deformation. The impact resistance value of most of the coatings reached the maximum (172 in-lb (1.98 kg m)) of the instrument.
Table 6: Performance of waterborne GC coatings

Coating      Conical      Reverse      Konig     Cross
             mandrel      impact     pendulum    hatch
                          (in-lb)              adhesion
             (0 cm =       (kgm)     hardness   (5B =
              Best)                    (s)     Best)

F1                 0  >172(>1.98)          44        5B
F2                 0   128 (1.47)          97        4B
F3                 0    112(1.29)          90        OB
F4                 0  >172(>1.97)          60        1B
F5                 0  >172(>1.97)          59        1B
F6                 0  >172(>1.97)          30        4B
F7                 0  >172(>1.97)          23        5B
F8                 0  >172(>1.97)          19        5B
F9                 0  >172(>1.97)          24        5B
F10(control        0  >172(>1.97)          39        5B


[FIGURE 6 OMITTED]

Further analysis of the coating properties

For further analysis of coating properties, coatings were selected from the results presented in Table 5, which show good water and solvent resistance determined by water drop test, water double rubs test, and MEK double rubs test. Thus, coatings F1, F2, F4, and F10 (control) based on resins R1, R2, R4, and R10 (control), respectively, were selected for further analysis.

Table 7 shows for the above coatings the values of the elongation at break and Young's modulus determined by tensile test, [T.sub.g] determined by DSC, and crosslink density determined from DMA. The coatings showed low elongation at break. Young's modulus, [T.sub.g], and crosslink density of the coatings decreased as the extent of mPEG increased in the respective GC resins.
Table 7: Tensile properties of the screened coatings

Coatings      Elongation    Young's       [T.sub.g]       Crosslink
                  at      modulus(MPa)  ([degrees]C)  density(mol/L)
              break(mm)                   from DSC

F1                     6          3135            37            1.22
F2                     4          2931            35            0.95
F4                    11           978            27            0.61
F10(control)          15           819            34            0.94


Figure 6 shows the (a) storage modulus and (b) tan S curves for Fl, F2, F4, and F10 coatings. Crosslink density of the coatings was determined from the values of storage modulus well above [T.sub.g]. A decrease in the tan [DELTA] peak height indicates an increase in crosslink density. (32) Fl, F2, and F10 coatings had tan [DELTA] peak at similar heights and also had crosslink density values very close to each other. Coating F4 had the lowest crosslink density determined from its storage modulus curve and also had the highest tan[DELTA]peak. An increase in the amount of mPEG in the resins decreased the reactive epoxy group content in the coatings and decreased the crosslink density.

Thermal stability

TGA was performed on the coatings to determine their thermal stability. Results of TGA are shown in Fig. 7. In general, the coatings showed good thermal stability. The onset temperature for the degradation of the coatings was around 225[degrees]C.

[FIGURE 7 OMITTED]

[FIGURE 8 OMITTED]

[FIGURE 8 OMITTED]

Salt spray test

Salt spray tests were performed on the screened coatings on steel and aluminum substrates. The coatings on steel substrate were delaminated within 24 h of salt spray and the coatings on steel substrates were removed from the salt spray chamber. The coatings on aluminum substrate did not show delamination, blister, or creep over 10 days of salt spray. However, the panels became reddish brown dark over 240 h of salt spray. Figure 8 shows the pictures of the coatings after 10 days of salt spray test. It should be noted that the aluminum panels were treated with Alodine 5700 before applying the coatings on them. Unlike for the coatings on steel substrate, the coatings on aluminum substrate did not delaminate up to 240 h in salt spray testing. This could be due to surface treatment of the aluminum panels which improved the adhesion of the coatings.

Conclusions

Glycidyl carbamate resins can be made water dispersible by incorporating nonionic hydrophilic groups such as mPEG into the resin structures. Chain length and level of incorporation of mPEG in GC resins greatly influence their water dispersibility. All of the resins except R1 and R10 were able to be dispersed in water without organic cosolvents or surfactants. The coatings crosslinked with Anquamine 731 were transparent and had smooth surfaces. Coatings F1, F2, F4, and F10 based on GC resins R1, R2, R4, and R10, respectively, showed good water and solvent resistance in water drop, water double rubs, and MEK double rubs tests. The coatings had good impact strength, flexibility, adhesion, hardness, Young's modulus, and low elongation at break.

The salt spray tests indicated that substrate treatment had an influence on adhesion of the coatings. The coatings on steel substrates were delaminated in a day in a salt spray chamber while the coatings on treated aluminum substrate remained intact, and no blister or delamination was observed. The aluminum substrate became reddish brown over the period of exposure.

Acknowledgment This article is based on a research study supported by the United States Air Force under Contract No. FA9550-09-C-0150.

References

(1.) Edwards, PA, Striemer, G, Webster, DC, "Novel Polyure-thane Technology Through Glycidyl Carbamate Chemistry." J. Coat. Technol Res., 2 (7) 517-527 (2005)

(2.) Chattopadhyay, DK, Raju, KVSN, "Structural Engineering of Polyurethane Coatings for High Performance Applications." Prog. Polym. Sci, 32 (3) 352-18 (2007)

(3.) Tarlas, HD, "Amine Cured Epoxy Resin Coatings for Resistance to Atmospheric Corrosion." Mater. Prot., 9 (5) 37-42 (1970)

(4.) Edwards, PA, Striemer, G, Webster, DC, "Synthesis, Characterization and Self-Crosslinking of Glycidyl Carbamate Functional Resins." Prog Org Coat., 57 (2) 128-139 (2006)

(5.) Edwards, PA, "Glycidyl Carbamate Resins to Achieve Polyurethane Properties and Epoxide Reactivity." Ph.D. Dissertation, North Dakota State University, Fargo, 2004

(6.) Edwards, PA, Erickson, J, Webster, DC, "Synthesis and Self-crosslinking of Glycidyl Carbamate Functional Oligomers." Polym. Prep. (Am. Chem. Soc. Div. Polym. Chem.), 44 (1) 54-55 (2003)

(7.) Chattopadhyay, DK, Webster, DC, "Hybrid Coatings from Novel Silane-Modified Glycidyl Carbamate Resins and Amine Crosslinkers." Prog. Org. Coat., 66 (1) 73-85 (2009)

(8.) Chattopadhyay, DK, Zakula, AD, Webster, DC, "Organic-Inorganic Hybrid Coatings Prepared from Glycidyl Carbamate Resin, 3-Aminopropyl Trimethoxy Silane and Tetraethoxy-orthosilicate." Prog. Org. Coat, 64 (2-3) 128-137 (2009)

(9.) Chattopadhyay, DK, Muehlberg, AJ, Webster, DC, "Organic-Inorganic Hybrid Coatings Prepared from Glycidyl Carbamate Resins and Amino-Functional Silanes." Prog. Org. Coat., 63 (4) 405-15 (2008)

(10.) Harkal, UD. Muehlberg, AJ, Li, J, Garrett, JT, Webster, DC, "The Influence of Structural Modification and Composition of Glycidyl Carbamate Resins on Their Viscosity and Coating Performance." J. Coat. Technol Res., 7 (5) 531-546 (2010)

(11.) de Meijer, M, "Review on the Durability of Exterior Wood Coatings with Reduced VOC-Content." Prog. Org. Coat., 43(4)217-225(2001)

(12.) Geurink, PJA, Scherer, T, Buter, R, Steenbergen, A, Henderiks, H, "A Complete New Design for Waterborne 2-Pack PUR Coatings with Robust Application Properties." Prog. Org. Coat., 55 (2) 119-127 (2006)

(13.) Tanabe, H, Ohsugi, H, "A New Resin System for Super High Solids Coating." Prog. Org. Coat., 32 (1^) 197-203 (1997)

(14.) Elmore, JD, Kincaid, DS, Komar, PC, Nielsen, JE, "Water-borne Epoxy Protective Coatings for Metal." J. Coat. TechnoL, 74 (931) 63-72 (2002)

(15.) Galgoci, EC Komar, PC, Elmore, JD, "High Performance Waterborne Coatings Based on Dispersions of a Solid Epoxy Resin and an Amine-Functional Curing Agent." J. Coat. TechnoL, 71 (891) 45-52 (1999)

(16.) Marrion, AR, The Chemistry and Physics of Coatings, 2nd ed. The Royal Society of Chemistry, Cambridge. 2004

(17.) Melchiors, M, Sonntag, M, Kobusch. C Jurgens, E, "Recent Developments in Aqueous Two-Component Polyurethane (2K-PUR) Coatings." Prog. Org. Coat., 40 (1-4) 99-109 (2000)

(18.) Mundstock, H, Pires, R, Richter, F, Schmitz, J, "New Low Viscous Polyisocyanates for VOC Compliant Systems." Macromoi Symp., 187 281-292 (2002)

(19.) Overbeek, A, "Polymer Heterogeneity in Waterborne Coatings." J. Coat. TechnoL Res., 7 (1) 1-21 (2010)

(20.) Schoff, CK, "Organic Coatings: The Paradoxical Materials." Prog. Org. Coat., 52 (1) 21-27 (2005)

(21.) Shaffer, M, Stewart, R, Allen, K, Wylie, A, Lockhart, A, "Two Component Polyurethane Coatings: High Performance Crosslinkers Meet the Needs of Demanding Applications." JCT CoatingsTecK 6 (1), 50-55 (2009)

(22.) Taylor, JW, Winnik, MA, "Functional Latex and Thermoset Latex Films." J. Coat. Technol Res., 1 (3) 163-190 (2004)

(23.) Wicks. ZW, Wicks. DA, Rosthauser. JW. "Two Package Waterborne Urethane Systems." Prog. Org. Coat., 44 (2) 161-183 (2002)

(24.) Wicks, ZW, Jr. Jones, FN, Pappas, SP, Wicks, DA. Organic Coatings: Science and Technology, 3rd ed. John Wiley and Sons Inc., New Jersey, 2007

(25.) Back, GE, Wang, PC. Corley, LS. Elmore, KD. "Water Dispersible Epoxy Resins." US 6.956,086 B2, 2005

(26.) Corley, LS, Kincaid, DS, Young, "GC Amine-Terminated Polyamide in Oil-in-Water Emulsion." US 5.962.269, 1999

(27.) Corley, LS, Kincaid, DS. Young, GC, "Aqueous Dispersion of Polyamide-Amine Derived from Aminoalkylpiperazine Epoxy Resin." US 5,998,508, 1999

(28.) Matthias, L, Cook, M, Klippstein, A, "Method of Preparation of a Water Based Epoxy Curing Agent." US 7,615,584 B2, 2009

(29.) Stark. CJ. Elmore, JD, Back, GE, Wang, PC, Galgoci, EC, "Aqueous Dispersion of Epoxy Resins." US 6,221,934, 2001

(30.) Hill, LW, "Determination of Crosslink Density in Thermoset Coatings." Polym. Mater. Sci. Eng., 77 387-388 (1997)

(31.) Skaja, A, Fernando, D, Croll. S. "Mechanical Property Changes and Degradation During Accelerated Weathering of Polyester-Urethane Coatings." J. Coat. Technol. Res., 3(1)41-51 (2006)

(32.) Wold, CR, Soucek, MD, "Viscoelastic and Thermal Properties of Linseed Oil-Based Ceramer Coatings." Macromol. Chem. Phys., 201 (3) 382-392 (2000)

U. D. Harkal, A. J. Muehlberg, P. A. Edwards,

D. C. Webster

Department of Coatings and Polymeric Materials, North Dakota State University, PO Box 6050, Dept 2760, Fargo, ND, USA

e-mail: dean.webster@ndsu.edu

DOI 10.1007/s11998-011-9356-8
COPYRIGHT 2011 American Coatings Association, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2011 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Harkal, Umesh D.; Muehlberg, Andrew J.; Edwards, Peter A.; Webster, Dean C.
Publication:JCT Research
Article Type:Report
Date:Nov 1, 2011
Words:7267
Previous Article:Effect of [TiO.sub.2] surface treatment on the mechanical properties of cured epoxy resin.
Next Article:Nano-crystalline pulsed laser deposition hydroxyapatite thin films on Ti substrate for biomedical application.
Topics:

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters