Real-time analysis of polymer properties and reaction kinetics allows scientists to control the outcome of a polymerization process. To illustrate this, a homopolymerization of acrylamide was produced using varying temperatures and semi-batch feeds of initiator and/or acrylamide monomer. This allows control of the rate of polymerization, absolute intrinsic viscosity and the weight average molecular weight of the polymer in real time.
Pilot reactor monitoring and control system
The ACOMP pilot reactor monitoring and control system (Advanced Polymer Monitoring Technologies, New Orleans, La.) has two standalone enclosures, a reactor control cabinet and an ACOMP cabinet (see Figure 1). The reactor control cabinet houses a 15-L circulating heater bath, two gas-feed controllers and two positive-displacement feed pumps, all of which facilitate accurate control of reactor temperature; precise feed of gas through the flowmeters; and independent supply of up to two liquid solutions simultaneously—monomer, initiator, chain transfer agent or other component. Controls for each component, including a human–machine interface (HMI), are housed in the reactor control cabinet. The reaction vessel is a 1.5-L stainless steel hemispherical bottom reactor that can be pressurized to 150 psi. A pump affixed to the bottom port of the reactor circulates the contents through a sampling loop.
Figure 1 – ACOMP pilot reactor monitoring and control system.The ACOMP cabinet houses the sample extraction and conditioning components and detector train for monitoring the polymer reaction. A UV/VIS spectrometer is used for determining monomer concentration; a differential pressure transducer measures intrinsic viscosity (IV); and a multi-angle, static light-scattering detector is used for assessing molecular weight (Mw).
Continuous sampling
Rather than taking discrete aliquots, the ACOMP system continuously extracts a small stream from the reactor sampling loop during the polymerization process. This is typically on the order of 0.10–1.0 cm3/min, depending on the conditions of the reaction, rate of polymerization and dilution necessary for accurate measurement. The polymer is then immediately diluted and homogenized with appropriate solvent. Dilution and cooling of the reactor contents quenches further polymerization, allowing for polymer characterization at the time of sample extraction. Once the sample has been conditioned for detector measurement, the stream continuously flows through the detectors to carry out measurement and characterization of the polymer properties (see Figure 2).
Figure 2 – ACOMP sampling process.Sample analysis
Sample stream analysis is similar to characterization of size exclusion chromatography (SEC) data, the difference being that ACOMP analysis takes place in real time without using chromatography to separate the sample based on size exclusion. The method allows for continuous, real-time characterization of the polymer’s absolute molecular weight and intrinsic viscosity. Since there is no delay between discrete samples, as is the case with SEC, a more complete representation of kinetics is achieved. This enables better measurement of the reaction kinetics and the ability to predict and control reaction kinetics in real time.
By utilizing the appropriate detectors for characterization, true polymer characteristics like absolute molecular weight, concentration of monomer, concentration of polymer and intrinsic viscosity can be determined during synthesis. UV absorption was used to monitor the concentration of acrylamide monomer from the beginning of the reaction, when no polymer was present, to the end of the reaction, after all monomer had been converted to polymer. By mass balance, understanding that all monomer in the reaction will convert to polymer, the concentration of polymer in the reactor could be calculated at every point in time based on direct measurement of the monomer. Information about polymer concentration, combined with absolute Rayleigh scattering intensity, allows determination of absolute Mw of the polymer product in the reactor. The IV of the polymer can be calculated once the polymer concentration value and voltage signal of the differential pressure transducer are known.1–3
Control of polymer characteristics during synthesis
To understand how to control a reaction for a parameter such as property or trajectory, the true Mw, IV, kinetic rates and trajectories of a polymerization process must be determined. The initiator, monomer feed, chain transfer agent, quenching agent rates and temperature can then be adjusted to produce a specific polymer. Using traditional methods, the user would refer to a model of the reaction or infrequent discrete data points, which do not always provide a clear picture of trends due to large gaps. The ACOMP method provides model-free guidance for control of these parameters and thus to achieve the desired final polymer product. Further integration of analysis and model predictive control algorithms will enable fully automatic, closed feedback control, thus removing the need for human active manual control.
Results
The isomorphic reaction pair shown in Figure 3 compares two homopolymerization acrylamide reactions with nearly identical kinetic profiles for polymer conversion. In the first reaction, the temperature was increased from 45 °C to 65 °C, shown by the dashed red line. The conversion, as monitored by the UV detector, is depicted by the solid red line; the solid blue line is the conversion trajectory of a reaction controlled by actively adding initiator at a fixed T = 45 °C (dashed blue line).
Figure 3 – Isomorphic reaction pair. An isothermal reaction with manually, active controlled addition of initiator (solid blue) allowed the guide trajectory (solid red) to be followed to within 5%.A second isomorphic reaction pair is shown in Figure 4. The first reaction, shown in red, is the same as that in Figure 3, i.e., temperature was increased to 65 °C at the fixed initiator charge. Shown in blue is the follow-up reaction, performed by adding initiator at T = 45 °C to follow the weight average molecular weight of the guide trajectory.
Figure 4 – Isomorphic reaction pair. Molecular weight of an isothermal reaction with manually, active controlled addition of initiator (blue curve) allowed the guide trajectory Mw (red curve) to be followed to within 10%.Figure 5 shows the intrinsic viscosity of a model semi-batch reaction (solid red) in which acrylamide monomer (Am) was flowed into the reactor from a concentrated reservoir. UV absorbance proportional to the amount of monomer in the reactor is represented by the dashed red line. The second reaction (blue) attempts to reproduce the model reaction IV by varying the monomer flow into the reactor as needed in order to drive the second reaction as close to the model reaction as possible. As can be seen, the intrinsic viscosities are within 5%.
Figure 5 – Isomorphic reaction pair. Intrinsic viscosity of an isothermal reaction with manually, active controlled addition of initiator (blue curve) allowed the guide trajectory IV (red curve) to be followed to within 5%.Conclusion
It is possible to monitor and control the reaction kinetics and properties of a polymer during synthesis. Future work will incorporate fully automated feedback control for monomer, initiator or other reactor feeds based on real-time ACOMP data analysis.
References
- Florenzano, F.H.; Strelitzki, R. et al. Absolute, online monitoring of polymerization reactions. Macromolecules 1998, 31, 7226–38.
- Kreft, T. and Reed, W.F. Predictive control of average composition and molecular weight distributions in semibatch free radical copolymerization reactions. Macromolecules 2009, 42(15), 5558–65.
- Reed, W.F. Automated continuous online monitoring of polymerization reactions (ACOMP) and related techniques. In Encyclopedia of Analytical Chemistry; pp 1–40; doi: 10.1002/9780470027318.a9288.
Michael F. Drenski is chief technology officer, and Wayne F. Reed is chief scientific officer, Advanced Polymer Monitoring Technologies, Inc., 1078 South Gayoso St., New Orleans, La. 70125, U.S.A.; tel.: 504-777-2805; e-mail: [email protected]; www.apmtinc.com. This work was supported by the U.S. Department of Energy’s Advanced Manufacturing Office.