Polymerization Reaction Monitoring for PSA Production Using an ATR-FTIR Probe

The Carl Dahlquist Best Paper Award was presented to Renata Jovanovic at the PSTC TECH XXVI Seminar in May 2003 in Washington, DC.

The utilization of Attenuated Total Reflectance – Fourier Transform Infrared (ATR-FTIR) spectroscopy for the characterization of pressure-sensitive products is well established. Some recent advances in sensor technology have opened new application areas of ATR-FTIR spectroscopy. One of them is the monitoring of chemical reactions, which are of interest for pressure-sensitive adhesive production. In this study, ATR-FTIR spectroscopy was used for monitoring polymerizations of acrylic monomers. Current results offer a basis for the development of ATR-FTIR spectroscopy as not only a monitoring tool, but also as a tool for process control.

Figure 1. Typical ATR-FTIR Spectra of BA Solution Polymerization in Toluene

_{(Published with permission from Ref. 5)}

The manufacture of pressure-sensitive adhesives includes synthesis and formulation. Two common synthesis methods for compounds with built-in PSA properties are polymerization in solution and emulsion.

₁When producing compounds with built-in PSA properties, the variation in the chemical and physical characteristics of the reaction components and/or the disturbance of process parameters can seriously affect the desired structure-property relationships of the final PSA. Modeling and/or monitoring of polymerization reactions are the means to ensure desired final product properties despite these disturbances. Models can be used effectively in these situations, but they are usually required to be robust and have an extensive parameter database. Reaction monitoring systems combined with feedback control strategies can be a solution to real-time process control.

Figure 2a. Decrease in the Absorbance for BA and VAc Characteristic Peaks with Time Indicates Monomer Consumption During the Reaction

_{(Published with permission from Ref. 5)}

Among several reaction monitoring techniques, in-line monitoring is preferred₂because it overcomes the major disadvantages of off- and on-line monitoring techniques. This includes considerable time lags, in one case, and the necessity of sampling loops, which are prone to polymer buildup or sample disturbance in the other. Different techniques are suitable for polymerization reaction monitoring (e.g., gas chromatography, calorimetry, ultrasound, spectroscopic techniques.

_3-4) Among the spectroscopic techniques, midrange infrared (MID IR) offers kinetic and structural information about reaction components without the application of complex chemometrics methods for analysis or expensive modifications to the reactor configuration. Recent advantages in sensing and signal transfer technologies have enabled the use of Attenuated Total Reflectance – Fourier Transform Infrared (ATR-FTIR) spectroscopy for remote, real-time reaction monitoring of chemical reactions. Curing, solution and emulsion polymerizations have been monitored using a range of homemade and commercial monitoring systems based on ATR-FTIR spectroscopy.

_3-4

In this study, a commercially available ATR-FTIR reaction monitoring system was used. It operates in the MID IR region (4000-650 cm_-1) and consists of electronic and optical modules. A purged path for the IR beam from source to detector and back was ensured using a set of mirrors and conduits ending with a remote sampling device. The sampling device consists of a stainless steel body and a six-reflection bi-layer diamond-composite ATR element. The external processor is used for data acquisition and manipulation. Details of the probe are discussed elsewhere.₅ The basis for monitoring the reaction is the determination of characteristic absorbances assigned to monomer consumption or polymer buildup. These can be used to calculate polymer composition and overall conversion.

Figure 2b. Decrease in the Absorbance for BA and VAc Characteristic Peaks with Time Indicates Monomer Consumption During the Reaction

_{(Published with permission from Ref. 5)}

In the work presented here, an ATR-FTIR probe was used to monitor solution and batch emulsion polymerization reactions of different monomers suitable for PSA production. The use of the ATR-FTIR probe started with off-line monitoring of homogeneous systems such as solution homo- and co-polymerizations₅and continued with the more demanding real-time monitoring of heterogeneous systems such as emulsion or miniemulsion polymerizations.₆The data obtained using ATR-FTIR spectroscopy was compared to those obtained using traditional off-line monitoring techniques such as gravimetry,₁H-NMR spectroscopy and gas chromatography. The results presented here were obtained for butyl acrylate and vinyl acetate homo- and co-polymerizations. Successful ATR-FTIR monitoring was also performed with other monomers of interest for PSA production such as styrene and methyl methacrylate.₇

Experimental Methods

Solution Polymerizations. Six homo- and co-polymerizations of butyl acrylate (BA) and vinyl acetate (VAc) in toluene were performed. The concentration of toluene was varied from 50 to 80 wt. percent. N-dodecyl mercaptan was used as a chain transfer agent and 2,2’- azobisisobutyronitrile (AIBN) as an initiator. Standard procedures were followed for reagent purification.₅All reactions were performed in glass ampoules at 60 degrees C. For each measurement, the reaction in two ampoules was simultaneously quenched and the contents of one ampoule were analyzed by gravimetry and₁H-NMR (Proton Nuclear Magnetic Resonance) spectroscopy. The contents of the second ampoule were used for off-line determination of conversion and copolymer composition using the ATR-FTIR probe. The probe was inserted into a vial containing the reaction mixture. For all samples, air was collected as a background spectrum and automatically subtracted from the sample spectra. All spectra were collected at 64 scans and a resolution of 8 cm_-1.

Figure 3. BA Solution Homopolymerization Monitored Using Gravimetry and ATR-FTIR Spectroscopy

_{(Published with permission from Ref. 5)}

Emulsion Polymerizations. Eight BA/VAc emulsion copolymerizations were performed in two automated reactors of 1.2 and 5L. Two polymerization temperatures were used (60 and 80 degrees C) and solid contents were 30 wt. percent and 50 wt. percent. For two runs, sodium dodecyl sulfate was used as a stabilizer while poly(vinyl alcohol) was used in all other runs. Ammonium persulfate and NaHCO₃were used as initiator and buffer, respectively. N-dodecyl mercaptan was used as a chain transfer agent in all reactions. Standard procedures were followed for monomer purification.₅In this case, the ATR-FTIR probe was used in-line for real-time reaction monitoring. The background spectrum was collected at the appropriate temperature and stirring rate prior to the addition of monomers and initiator. Collection of the reaction spectra was started simultaneously with the injection of initiator. The best signal to noise ratio for the spectra collection was determined to be 256 scans at resolution of 4 cm_-1. To ensure continuous monitoring, the spectra were collected every 2 min.

Figure 4. VAc Solution Polymerization Monitored Using Gravimetry and ATR-FTIR Spectroscopy

_{(Published with permission from Ref. 5)}

Miniemulsion Polymerizations. Several miniemulsion homo- and co-polymerizations of BA and VAc were also monitored using ATR-FTIR spectroscopy. These reactions were performed in a 1.2L, automated reactor at 60 degrees C. The solids content was 30 percent. The same number of scans and resolution was used as for the emulsion polymerizations. The difference was in the collection of background spectra. Due to the nature of the miniemulsion polymerization, the reaction components were sonicated prior to charging to the reactor. Thus, the background spectrum in this case included monomer(s). The reaction spectra were also collected beginning with the injection of initiator.

Results and Discussion

Solution Polymerization Monitoring. In this part of the project, the main objective was the identification of characteristic absorbances and their utilization for copolymer composition and overall conversion monitoring. Given this objective and the nature of the experimental runs, off-line measurements were a good initial start for the development of a measurement method based on ATR-FTIR spectroscopy. The results obtained were compared to gravimetric measurements.

The basis for the application of ATR-FTIR spectroscopy in monomer conversion determination is Beer’s law. The absorbance of the component in the reaction mixture is directly proportional to its concentration. The absorbance can be measured as peak height, peak height ratio, peak area or peak area ratio. Thus, a simple expression for conversion is obtained (Equation 1):

Figure 5. BA/VAc Solution Polymerization in Toluene

_{(Published with permission from Ref. 5)}

The first step was the homopolymerization of BA and VAc in toluene, in order to identify the characteristic absorbance peaks for each monomer that exhibited a change with reaction time. Figure 1 shows a typical spectra of BA homopolymerization in toluene. For the purpose of clarity, only a fingerprint region of the spectra is shown. The absorbances at several wavenumbers (e.g., 810 cm_-1, 984 cm_-1, 1409 cm_-1, 1640cm_-1) showed potentially useful time trends for reaction monitoring. In all cases, the absorbances decreased with time, indicating a decrease in monomer concentration due to its consumption during the polymerization. Similar trends were found at 876 cm_-1, 949 cm_-1, 1293 cm_-1, and 1648 cm_-1when spectra of the VAc homopolymerization in toluene were investigated.

After peak identification, the next step was the determination of the best mathematical expression for the absorbance. Theoretically, regardless of the peak selection and the expression for absorbance, the results should be similar. However, in this case, the best fit between gravimetry and ATR-FTIR data was obtained when the absorbances at 810cm_-1for BA and at 1293 cm_-1for VAc (see Figure 2) were used. A peak height to two-point baseline was used in both cases. Thus, Equation 1 becomes (seeEquation 2):

Figures 3-4 show typical results for BA and VAc homopolymerizations in toluene. Similar results were obtained when different solvent concentrations were used. In all cases, ATR-FTIR spectroscopy showed good agreement with the standard gravimetric method. The estimated error for ATR-FTIR spectroscopic data was 2.5 wt. percent.₇

After the determination of conversion vs. time data of the solution homopolymerizations, ATR-FTIR spectroscopy was used for off-line monitoring of BA/VAc solution copolymerizations. In addition to low toluene absorbance, an additional challenge was to find the characteristic absorbances for each of the two monomers that would not overlap, in order to use them for the determination of individual monomer conversions. Figure 5 shows the results are shown for a BA/VAc solution copolymerization in toluene.

In the copolymerization case, the following expression was used to calculate overall conversion using ATR-FTIR spectroscopic data (see Equation 3):

Figure 6. Typical Real-Time Reaction Spectra of BA/VAc Emulsion Copolymerization Obtained Using an ATR-FTIR Probe

_{(Published with permission from Ref. 6)}

wherem_i/(m_i+m_j) is the weight fraction of monomer i in the reaction mixture and x_i is the individual monomer conversion determined from Equation 2. The data obtained using ATR-FTIR spectroscopy were compared to standard ₁H-NMR spectroscopy and gravimetric data. As in the homopolymerization cases, good agreement between the different techniques was obtained.

Emulsion Polymerization Monitoring. After the successful completion of off-line monitoring of homogenous reaction systems, the ATR-FTIR probe was used for the more complex, in-line monitoring of the heterogeneous emulsion polymerization system. Real-time monitoring of such a system is interesting not only because there are several phases present, but also because of the compartmentalization of monomer(s) among them, especially for highly water soluble monomers such as VAc. Typical reaction spectra collected in-line are shown in Figure 6. Real-time peak profiles, i.e., normalized absorbances vs. time (Figure 7) were used as input data in Equation 2 to determine the individual monomer conversions. Once the individual monomer conversions were obtained, Equation 3 was used to determine overall conversion.

Figure 7. Real-Time Peak Profiles
_{(Published with permission from Ref. 6)}

Several different copolymer compositions were investigated. In addition, two different stabilizers were used: one was an electrostatic stabilizer, sodium dodecyl sulfate (SDS) and the other was a steric stabilizer, poly(vinyl alcohol) (PVOH). Due to the presence of grafting reactions when PVOH was used, the obtained polymers were not soluble in the range of organic solvents acceptable for ₁H-NMR analysis. Thus, gas chromatography was employed as a standard technique for off-line monitoring of copolymer composition. A typical result using SDS as stabilizer is shown in Figure 8. Once again, ATR-FTIR spectroscopy showed good agreement compared to standard techniques. In this case, the heterogeneity of the system did not affect the results obtained using the probe. This was possibly due to several reasons. First, good mixing of the reaction mixture was obtained. Second, the probe was positioned just 2 mm above the impeller element ensuring that a representative sample was in contact with the probe. In addition, each spectrum was recorded from an average of 256 scans. Similar results were observed when PVOH was used for the same initial BA/VAc composition but at a higher solids content (50 percent). A typical conversion vs. time curve under such conditions is shown in Figure 9.

Figure 8. BA/VAc Emulsion Copolymerization (SDS Stabilizer) Monitored Using an ATR-FTIR Probe and Standard Techniques
_{(Published with permission from Ref. 6)}

In emulsion polymerization monitoring, the ATR-FTIR probe has shown great potential. While off-line monitoring of copolymer composition using ₁H-NMR spectroscopy was not possible due to the insolubility of the obtained polymers, in situ measurements using ATR-FTIR spectroscopy were successful. In addition, the probe was used to successfully monitor an induction period when it occurred (Figure 8). This enables the operator to estimate early during the reaction the amount of initiator to be added in the reactor for corrective purposes. Furthermore, in the case of a catastrophic coagulation, the spectra obtained showed the disturbance along the time axis much earlier than the separation of phases was observed in the samples taken from the reactor (Figure 10). Further investigation is needed in order to obtain a more quantitative representation of these phenomena but, nonetheless, this information can save time and financial resources that are required for reactor cleaning.

Figure 9. Conversion vs. Time for a BA/VAc Emulsion Copolymerization (PVOH Stabilizer) Using In-Line and Off-Line Monitoring Techniques
_{(Published with permission from Ref. 6)}

On the other hand, the probe was found to be extremely sensitive to temperature variations.₅ The temperature control had to be within ± 0.2 degrees C. Large deviations from the set point followed by a long stabilization period negatively affected the results while small deviations and a short stabilization period showed disagreement. After the set temperature was established the ATR-FTIR data showed good agreement with the standard techniques. In addition, at high concentrations of VAc in the presence of PVOH, deviations beyond experimental error were observed. Further investigation is needed to resolve this issue. After successfully monitoring several systems over a range of concentrations and solids contents_5-7, the speculation is that the presence of PVOH or its subtraction in the background might have caused this discrepancy. At lower VAc concentrations this method was appropriate, while at higher VAc concentrations, when more grafting of VAc occurs, the subtraction of PVOH as a background might not be appropriate.

Figure 10. Early Indication of Catastrophic Coagulation
_{(Published with permission from Ref. 6)}

Miniemulsion Polymerization Monitoring. Monitoring of miniemulsion polymerizations of BA/VAc and other monomers is currently in progress in our lab. Due to the fact that the preparation of miniemulsions includes homogenization of the reaction mixture prior to charging it to the reactor, the collected background contains monomer in addition to other components. In this case, the identification of characteristic peaks is more complex and possibly requires the application of chemometrics methods such as principal component analysis for data extraction.

Conclusion
For the production of PSA with the desired built-in performance characteristics, real-time process monitoring and control are essential process tools, especially for systems where compositional drift can occur or where variation in physical and chemical properties of reactants is expected. Even though there are different techniques for reaction monitoring, ATR-FTIR spectroscopy has advantages over other monitoring techniques. It has a potential for real-time monitoring of kinetic and structural changes of the reaction components with no requirements for sampling loops or complex changes to reactor design.

In this study, it has been shown that an ATR-FTIR probe can be successfully used for off-line and in-line monitoring of homogeneous and heterogeneous polymerization systems that are commonly employed in PSA synthesis. Used in the study for off-line monitoring of BA/VAc solution polymerizations, it has accurately monitored conversion and copolymer composition. When used off-line, no disadvantages were observed. The time lag was minimal, no calibration was required and the technique was accurate compared to standard techniques. When similar principles were applied for in-line monitoring of BA/VAc emulsion polymerizations, in most cases the probe was able to correctly monitor copolymer composition and conversion compared to traditional techniques. When used in real-time, the probe was also able to follow the occurrence of an induction period or to offer an early indication of catastrophic coagulation. On the other hand, strict temperature control was found to be a major requirement for the successful application of ATR-FTIR spectroscopy. The extent and the duration of the disturbance were major factors affecting the performance of the probe. In addition, some problems were experienced at high concentrations of water-soluble monomer combined with PVOH presence and high solids content. At this point, one speculation is that the methods developed for other feed compositions might not be applicable when grafting reactions are likely. The performance evaluation of the probe for real-time monitoring of miniemulsion polymerizations is still in progress.

Acknowledgements

The authors wish to acknowledge the National Engineering and Science Research Council of Canada and the Canada Foundation for Innovation for the financial support of this project.

References:

_{1 Benedek, I.; Heymans, L.J. Pressure-Sensitive Adhesives Technology, Marcel Dekker Inc.: New York, 1997
2 Dallin, P. Proc. Control Qual. 1997, 9(4), 167-172.

3 Hergeth, W.D. In Polymeric Dispersions: Principles and Applications. Asua, J.M. Eds.; Kluwer Academic Publishers: Dordrecht, 1997; 267-288.

4 Kammona, O.; Chatzi, E.G.; Kiparissides, C. J. Macromol. Sci.—Rev. Macromol. Chem. Phys. 1999, C39(1), 57-134.

5 Jovanovic, R.; Dube. M.A. J. Appl. Polym. Sci. 2001, 82(12), 2958-2977.

6 Jovanovic, R.; Dube. M.A. Polym. React. Eng. J. (in press).

7 Hua, H.; Dube. M..A. J. Polym. Sci.: Polym. Chem. 2001, 39, 1860-1876.}