Factors that Control the Wet-Property Development of Waterborne Inks.
The property development from a resin fortified film forming emulsion depends on the rate of acid-salt decomposition and annealing of particles in the ink film. The acid-salt decomposition in clear polymer films from a film forming emulsion is diffusion controlled, a slow process that depends on the free volume of the polymer or Tg. Water resistance properties can take hours at room temperature to develop, but develop much faster at elevated temperatures. More than 80 percent of the ammonium acrylate moieties need to decompose and leave the polymer film before acceptable water resistance is obtained.
The performance demands placed on inks used in the flexible packaging industry are extensive. Inks are expected to print well on a variety of flexible substrates having good quality print, excellent water and chemical resistance properties, adhesion, crinkle resistance and heat resistance.
Overall, the performance of waterborne polymer emulsions is deficient compared to solventborne polymers, which dominate the flexible packaging market. The performance of waterborne polymers in film applications can be better understood through inter-particle diffusion and the decomposition of the ammonium acrylate moieties (acid-salts) to acrylic acid moieties and ammonia. (1,2,3)
This paper focuses on decomposition of acid-salts in films prepared from a polymeric supported waterborne emulsion. The emulsion studied contains a soft and glassy phase. It has a minimum film formation temperature of + 12[degrees]C. A theory that accounts for the development of water resistance properties is proposed. The theory summarizes the relationships between acid salt decomposition and the development of water resistance properties.
Results and Discussion
Water Up-Take Studies
One aspect of this study was to document the rate that thin dry polymer films (2.5 microns), cast from a film forming emulsion, absorbed water. Two sets of films were cast over ZnSe ATR crystals and aged at 23[degrees]C for two hours and at 60[degrees]C for 16 hours, respectively. Films of a commercial polyamide resin designed for ink printing were also studied. Figure 1 shows a diagram of the ATR crystal and the polymer film.
[FIGURE 1 OMITTED]
Figure 2 shows the absorption of water as a function of time as water diffuses into the polymer film. After aging, pure water was placed over the film and the water up-take was measured by infrared spectroscopy (FT-IR). Each data point took approximately one second to collect. The top plot in Figure 2 shows the absorption of water into the waterborne film forming emulsion after aging a 2.5 microns film for two hours at 25[degrees]C. The film was saturated with water in less than two minutes. However, the solvent-based polyamide film had not reached water saturation after one day. Heat aging films of the waterborne emulsion drastically reduced the rate of water up-take. It was postulated from these water up-take studies that the rate of water-uptake in the emulsion films was controlled either by film formation or residual ammonium acrylate moieties within the films.
[FIGURE 2 OMITTED]
To determine the influence of residual ammonium acrylate moieties in emulsion films, their rates of decomposition were studied under a variety of conditions, and the amount of residual acid-salt in ink films was correlated to resistance properties. Prior to this study, the rate and mechanism of acid-salt decomposition in the continuous phase was unknown. Also, the link between the amount of acid-salt decomposition and water resistance was unknown. The results and discussion below show that if films prepared using waterborne technology are to compete with the water resistance properties of solvent-based polyamide resins, then it is necessary to decompose most of the ammonium acid-salts (> 80 percent) to obtain the desired properties. Scheme 1 shows the reaction for the decomposition of the ammonium acrylate salt.
Use of FT-IRATR for Mechanistic Studies
This study used infrared spectroscopy to monitor the level of ammonium acrylate salt in the film, where the acid-salt absorbs infrared radiation strongly at 1549 [cm.sup.-1]. A calibration0 plot of film thickness versus the absorbance of styrene at 700 [cm.sup.-1] shows that an accurate film thickness is measurable directly from the IR spectrum of film. The thickness for each film was measured by obtaining the fringe pattern of the film over ZnSe using a UV-visible spectrometer. The thickness was then calculated using the well known equation below:
Film thickness = N ([[lambda].sub.1] [[lambda].sub.2]) 2n([[lambda].sub.2] - [[lambda].sub.1])
where N is the number of "peaks" in the fringe pattern between wavelengths 1 and 2 and is the refractive index of the polymer. Figure 3 shows the relationship between the absorbance at 700 [cm.sup.-1] and film thickness; thus, using this plot the film thickness for any sample can be obtained from the IR spectrum of a polymeric film.
[FIGURE 3 OMITTED]
Figures 4a and 4b show the kinetic results for a 1.5 micron film. The left plot below shows the decomposition kinetics. The fraction of acid-salt moieties that are decomposed in the film (1-[A.sub.t]/[A.sub.o]) is plotted as a function of time where [A.sub.t] is the infrared absorption from the acid-salt moieties at time (t), and [A.sub.o] is the initial infrared absorbance of the acid-salt moieties.
[FIGURE 4 OMITTED]
In Figure 4a, a value of 1.0 indicates that all the acid-salt moieties have decomposed to acrylic acid moieties and ammonia. Figure 4b shows the Ln([A.sub.o]/[A.sub.t]) plotted as a function of time. The use of infrared spectroscopy to follow the decomposition of the acid-salt assumes that Beer's Law is valid during the decomposition and that the infrared absorbance of the acid salt is directly proportional to its concentration. This analytical technique has previously been used successfully to show that the decomposition kinetics of diethyl amine-based carboxylic acid-containing films follow first-order kinetics at elevated temperatures. (3)
Figure 4a shows that for a 1.5 micron film, the acid-salt decomposition is ~90 percent complete after four hours. For acid-salt decomposition that is kinetically controlled, plotting the data as Ln(A.sub.o/[A.sub.t]) versus time should give data that correlates to a straight line. (3) The first-order rate equation is
Equation 1 Ln[[A.sub.o]/[A.sub.t]] = kt
where k is the first-order rate constant and t is the time. The data points on Figure 4b would correlate to give a curved line, not a straight line; therefore, the decomposition of the acid-salt in film is not kinetically controlled at 21[degrees]C. Controlled experiments (to be discussed shortly) also show that the acid salt decomposition rate is a function of film thickness. If the acid-salt decomposition rate was kinetically controlled, the rate of decomposition would be independent of film thickness.
Diffusion control is an alternative mechanism for the data in Figures 4a/b. Equation 2 is the diffusion equation for ammonia diffusing to the surface of a film where [M.sub.t] is the mass of the ammonia at time t and [M.sub.[infinity]] is the mass of ammonia available, 1 is the thickness of the film, and D is the diffusion constant.
Equation 2 [M.sub.t]/[M.sub.[infinity]] = 1 - 8/[[pi].sup.2] [summation over (m=o)] 1/[(2m+1).sup.2] exp [-D[(2m+1).sup.2][[pi].sup.2]t/[l.sup.2]]
In this model the diffusion constant, D, or more accurately the diffusion coefficient, is assumed constant throughout the decomposition. To use Eq. 2, it is assumed that the amount of ammonia produced is equal to the amount of acid-salt decomposed in the film; thus, the rate of acid-salt decomposition is controlled by the ability of the ammonia to diffuse from the film.
By following the rate of acid-salt decomposition by transmission infrared spectroscopy, a diffusion model can be obtained. The data in Figures 4a and 5 are the same; however, Figure 5 includes a correlation plot where regression analysis is applied to Eq. 2. (4)
[FIGURE 5 OMITTED]
The diffusion coefficient from the regression analysis is 0.242 [[micro].sup.2] [hr.sup.-1] and a correlation coefficient of 0.9713. The fit is excellent when the fraction of salt decomposed is <0.60, but goes from good to poor as more and more acid-salt decomposes.
Determination of the Diffusion Coefficient, D, as a Function of Acrylic Acid Moieites Formed in the Film
In many systems the diffusion coefficient is found to depend either linearly or exponentially on concentration. (5) To model the system under study, average diffusion coefficients were calculated using Eq. 2 from the data collected at three thicknesses. At each decomposition point in a set of data (left side of Eq. 2), iterative substitution for D was done on the right side of Eq. 2 until it equated to the left side of Eq. 2. The diffusion coefficients obtained are shown in Figure 6.
[FIGURE 6 OMITTED]
For thick films (3.7 and 10.7 microns at 21[degrees]C) the decrease in diffusion coefficients correlates with the fraction of acrylic acid moieties formed in the film. The decrease in diffusion coefficients as the salt decomposes to acid suggests a major change in the environment or matrix of the film. The iteration procedure (Figure 7) is checked by substituting the diffusion coefficients obtained into Eq. 2. The trace line is not a correlation, but rather, shows the mathematical validity of the iteration procedure. The theoretical level of acrylic acid moieties in the glassy phase of the film (after all the acid-salt has decomposed) is 26 percent.
[FIGURE 7 OMITTED]
To examine the dependence of D on the theoretical acid level in the film at a given time, the films at a thickness of 3.7 and 10.7 microns were correlated to Eq. 3:
Equation 3 D = [D.sub.o] [1-[alpha][1-[A.sub.t]/[A.sub.o]]]
where [D.sub.o] in Eq. 3 is the intercept from the plots in Figure 6, [A.sub.t] is the infrared absorbance of the acid-salt at time t, [A.sub.o] is the initial absorbance of the acid-salt at time zero, and [alpha] is a correlation parameter. Assuming Beer's Law holds under the condition of the experiment, 1-[A.sub.t]/[A.sub.o] is the theoretical level of acid that should be present in the film at a given time. The values for the 3.7 and 10.7 micron films obtained from regression analysis of the data in Figure 6 are 1.07 [+ or -] 0.01 and 1.18 [+ or -] 0.02, respectively. The Do values for the 3.7 and 10.7 micron films are 1.12 [+ or -] 0.02 and 1.30 [+ or -] 0.03, respectively.
Use of Eq. 2 and Eq. 3
Figure 8 shows the data for a film thickness of 3.7 microns and regression analysis using Eq. 2. Again the fit is poor. To improve the fit, Eq. 3 is substituted into Eq. 2 to obtain Eq. 4. The [D.sub.o] is obtained from the intercept from the plot in Figure 6. The data is fitted using Eq. 4 where [alpha] is a given value of 1.0.
[FIGURE 8 OMITTED]
Using a value of 1.0 eliminates the effect of the correlation coefficient and shows the true dependency of the diffusion coefficient on the acid formed in the film.
Equation 4. 1 - [A.sub.t]/[A.sub.[infinity]] = 1 - 8/[[pi].sup.2] [5.summation over (m=o)] 1/[(2m+1).sup.2] exp [-[D.sub.o][1-[alpha] [1-[A.sub.t]/[A.sub.o]]][(2m+1).sup.2][[pi].sup.2]t/[l.sup.2]]
One would expect the data collected for the films with a thickness of 3.7 and 10.7 microns to give identical linear correlations in Figure 6, but they do not. It is unclear why this discrepancy exists, but it may be due to the experimental error in obtaining the thickness of the films. Nevertheless, the correlations are reasonably close, and both correlations in Figure 6 show the dependency of the diffusion coefficient on amount of acid-salt present in the film.
It is postulated that this dependency is related to the free volume of the film. As the free volume of the film decreases because of acid-salt decomposition, the Tg of the film increases, and the diffusion coefficient decreases. This phenomenon deserves additional experimental work. Since the diffusion coefficients are a not linear function of the amount of acid formed in the film at a given time for the film with a thickness of 1.5 microns, it does not follow Eq. 3 or Eq. 4. As shown in Figure 5, the data for the 1.5 micron film appears somewhat linear above a theoretical acid level of 0.7 but flattens out or becomes constant below a theoretical value of 0.7. It is postulated that below 0.7, kinetic control begins to play a major role in the decomposition of the acid-salt; thus, as the thickness of these films decreases, a kinetic control mechanism begins to compete with the diffusion control mechanism.
This is illustrated with the equations in Figure 9. A variation of Eq. 2 gives the half-life equation for diffusion control. (6) It assumes that [t.sub.1/2] is sufficiently large so that all terms other than the first term in the series can be neglected.
[FIGURE 9 OMITTED]
Figure 10 shows a comparison of the data collected for a film thickness of 1.5 microns to the calculated acid level (and ammonia loss) if the mechanism operation is only diffusion control. Eq. 4 and the diffusion coefficient data from Figure 6 for a film thickness of 3.7 microns was used to calculate the theoretical acid level. This plot assumes that the diffusion coefficient is an inherent intrinsic property of the film.
[FIGURE 10 OMITTED]
The data in Figure 10 shows that for a given time, the measured values are less than the values one would expect if the ammonia loss mechanism from the film was only diffusion control. A calculated [t.sub.1/2] using Eq. 2 gives a value of 0.18. This value is less than the measured value of 0.32 hr. Although diffusion control contributes to the amount of carboxylic acid formed in the film, the rate of acid formation or ammonia loss is reduced due to kinetic control effects (Figure 9). At two hours or when 70 percent of the theoretical level of acid has formed, the diffusion coefficients control the observable effects and decomposition is essentially diffusion controlled.
Based on the decomposition results for the film with a thickness of 1.5 microns, it is expected that films thinner than 1.5 microns would have more of a kinetic component and less of a diffusion control component.
Additional Proof for Diffusion Control
The argument for diffusion control relies on the theory that the continuous phase is glassy with a Tg above the temperature at which the decomposition of the acid-salts observed. As shown by Eq. 2, the loss of ammonia by the decomposition of the acid-salts in the continuous phase depends on the thickness of the film. Figure 11 supports the decomposition mechanism for clear films of the film forming emulsion for a film thickness of 3.7 and 10.7 microns. The ammonium carboxylate salts in thicker films decompose slower than those in a thinner film. For a film thickness of 1.5 (Figure 7), 3.7 and 10.7 microns, the half-lives for the decomposition of the ammonium acrylate salts in each film thickness are 0.32, 1.4, and 12.5 hours, respectively. Ninety percent ammonium acrylate salt decomposition requires more than 15 half-lives.
[FIGURE 11 OMITTED]
Effects of Relative Humidity
One concern during the diffusion studies was the effect of relative humidity on the decomposition of the acid-salts in the film. Films were aged under laboratory conditions at a relative humidity of about 50 percent, then placed in the sampling compartment of the FT-IR spectrometer. The relative humidity in the sampling compartment drifted from the humidity in the laboratory to ~ 0 percent over about 10 minutes. For kinetic studies done under laboratory conditions, samples were allowed to air dry for three minutes, placed in the sample compartment for three minutes, analyzed (one minute), then removed back to the laboratory bench (or constant temperature humidity room) for aging until the next analysis. The question was, did the low humidity of the sample compartment skew the kinetic data? To answer this question, a 6-micron wet-film was air-dried over ZnSe under laboratory conditions for three minutes, then placed into the sample compartment for four hours. The temperature for this comparison was 22[degrees]C. When the decomposition results for each conditioning technique was compared, no significant differences were observed.
Correlation of Kinetic Work and Applications Testing
Films of the film forming emulsion (slightly pigmented at 3 percent) were aged over freshly corona treated polyethylene at room temperature. Figure 12 shows the correlation between the amount of ammonium salt left in the film and tape adhesion. The data show (Figure 12) that for films aged at 21[degrees]C, when 20 percent of the ammonium acrylate salt is left in the film (dry film thickness, 1.5 microns), then optimal wet-tape adhesion occurs.
[FIGURE 12 OMITTED]
Applications Testing at 60[degrees]C
In applications testing, films are commonly heated to 60[degrees]C for one minute in a forced-air oven. It was of interest to examine the correlation between the chemistry and tape adhesion properties at this temperature. Figure 13 shows the correlation plot for the film forming emulsion films aged at 60[degrees]C. More than 90 percent of the ammonium acrylate salt must decompose (less than 10 percent of the acid salt remains) to obtain good tape adhesion. Interestingly, when ~20 percent of the ammonium acrylate salt is left, excellent water resistance is obtained. It should be noted that the wet-crinkle resistance was very poor at acid-salt decompositions above 95 percent.
[FIGURE 13 OMITTED]
The results above give the obvious conclusions that wet adhesion properties improve as the ammonium salt within the film decompose, and there is no correlation between the amount of ammonium salt that has decomposed and wet water crinkle resistance. Also, it should be noted that typical testing conditions (1 minute at 60[degrees]C) is the location on the plots above where the greatest change in chemistry and property development occurs. The cure temperature used for the plot in Figure 13 is still below the Tg of the continuous phase of the film. As a result, the decomposition of the acid-salts is still diffusion control; however, the diffusion coefficients are larger at this temperature than at room temperature. The message is that though the process is still diffusion control, the acid-salts decompose faster at 60[degrees]C than at 21[degrees]C.
[FIGURE 13 OMITTED]
Properties/Film Thickness Relationship
One of the questions the authors have asked themselves is, "what is the relation between the film thickness, acid-salt decomposition rate, and wet adhesion properties?" Since the rate of acid-salt decomposition slows as the thickness of the film increases (rate 1/thickness2), will the decrease in rate result in a slower development of application properties, or will the increase in film thickness overcome this effect?
To answer these questions, films were prepared at a dry film thickness of 5.2 microns, and the ammonium decomposition rates and wet adhesion properties determined. The results are shown in Figure 14. The films were cured at room temperature (21[degrees]C). The conclusion from Figure 14 is that as the dry film thickness of the film forming emulsion films increase beyond 2.5 microns, the rate of acid-salt decomposition will decrease as well as the rate of wet-tape adhesion development.
[FIGURE 14 OMITTED]
Developing water resistance properties in waterborne inks is related to the release of ammonia in the film forming emulsion. The mechanism of ammonia release from a film is diffusion controlled and the majority of ammonia must be released from the film before water resistance properties are fully developed.
A kinetic mechanism for thin films < 1.5 [micro] can be considered using differential equations that can be derived to describe the concentration of salt, free acid and ammonia in the film:
(1) [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII.]
(2) d[C.sub.A]/[d.sub.x] - [k.sub1] [C.sub.A] + [k.sub.2] [C.sub.B] [C.sub.C]
(2) d[C.sub.B]/[d.sub.x] - [k.sub1] [C.sub.A] + [k.sub.2] [C.sub.B] [C.sub.C]
(3) d[C.sub.C]/[d.sub.x] - [k.sub1] [C.sub.A] + [k.sub.2] [C.sub.B] [C.sub.C] - [k.sub.3] [C.sub.C]
[C.sub.A] concentration of carboxylic salt
[C.sub.B] concentration of carboxylic acid in the film
[C.sub.C] concentration of ammonia in the film
These equations were solved simultaneously as a function of time to study the time evolution of the different components using Athena Visual Workbench. The three rate constants were estimated using a nonlinear least squares estimation and are [k.sub.1] = 3.67 liters/hr., [k.sub.2] = 4.77 liter/mole/hr., [k.sub.3] = 1.04 liters/hr. These parameters were used to compare the salt compositions predicted by the model to the experimental data. It can be seen that the model presented by equations (2)-(4) are capable of describing the experimental data without the necessity of calculating different diffusion coefficients of ammonia from the film (Figure 6).
The model was used to simulate the concentration of each of the main components in the system, [COO- N[H.sub.4] +], [[COOH].sub.film] and N[H.sub.3 film], using the parameters above, and is shown in the figure below. This figure shows the expected concentration of carboxylic salt decreasing in time. However, once sufficient ammonia is built up in the film, the rate of decrease of the salt slows due to the equilibrium reaction. It is at this point that the deviation from 1st order kinetics occurs. The carboxylic acid continually increases with time, mirroring the concentration of the salt.
However, the ammonia in the film goes through a maximum concentration. This can be explained by the fact that it is produced by the decomposition and consumed by diffusion out of the film and into the atmosphere. The different rates of these competing reactions account for the maximum in the concentration observed. However, this analysis indicates that diffusion is still the rate determining step in the reaction series since the rate constant for the evaporation is approximately four times lower than the decomposition term.
(1.) Taylor, J.W. and Winnik, M.A., "Functional Latex and Thermoset Latex Films," Journal of Coatings Technology Research, Vol. 1, No. 3, 1-28 (2004).
(2.) Taylor, J.W. and Klots, T., "Applied Approach to Film Formation: The Glass Temperature Evolution of Plasticized Latex Films," European Coatings Journal, 6, 38 (2002).
(3.) Taylor, J.W., Collins, M.J., and Bassett, D.R.,"Study on the Chemistry of Polyguanidines and Precursors for Polycarbodiimides in Powder Coatings," Journal of Coatings Technology, Vol. 67, No. 846, 43-52 (1995).
(4.) Crank, J., "The Mathematics of Diffusion," 2nd Ed., Crarendon Press 1975, p. 238.
(5.) Crank, J., ibid, p. 241.
(6.) Neogi, P., in Transport Phenomena in Polymer Membranes, Neogi, P. (Ed.), Diffusion in Polymers, Marcel Dekker, Inc: NY, NY, pp. 173-209 (1996).
Dr. Tim Klots has been an applications research scientist with Johnson Polymer for the last five years. In this position, he uses physical characterization techniques to evaluate new materials and offer insight on performance and property relationships. He received his Ph.D. in 1988 at the University of Illinois, Urbana-Champaign and bachelor's degree from the University of the South in Sewanee, IL in 1983.
Andre van Meer is the applications team leader for Johnson Polymer's European research and development group. He has been with Johnson Polymer for 18 years out of his 20 years of working in this industry.
As a research engineer, Dave Schatz is also involved in product development. He has spent his entire 20-year career with Johnson Polymer focused on product development. He has a bachelor's degree in chemical engineering from the University of Wisconsin-Madison.
Dr. Jim Taylor received his bachelor's degree in chemistry from the University of the South and his Ph.D. in polymer/organic chemistry from the University of Tennessee. With 15 years of industry experience, Dr. Taylor joined Johnson Polymer in 1997 and began developing waterborne latex technology for coatings and inks. One aspect of his work has been aimed at understanding film formation for waterborne acrylic polymers in a variety of applications.
Dr. Dave Campbell is technical manager for Johnson Polymer's functional coatings and active packaging group. In this position, he is responsible for managing the technical activities of new product development. Dr. Campbell has been with Johnson Polymer for 20 years and has a Ph.D. in chemical engineering from ETH-Zurich, Switzerland, master's in engineering and bachelor's degrees from McMaster University, Hamilton Ontario, Canada.
|Printer friendly Cite/link Email Feedback|
|Author:||Klots, Timothy D.; van Meer, Andre W.; Schatz, David D.; Taylor, James W.; Campbell, J. David|
|Date:||Mar 1, 2005|
|Previous Article:||Primers for UV ink jet ink printing on papers.|
|Next Article:||2005 Calendar.|