Anal. Chem. 2006, 78, 3090-3096
High-Throughput Determination of Quantitative Structure-Property Relationships Using a Resonant Multisensor System: Solvent Resistance of Bisphenol A Polycarbonate Copolymers Radislav A. Potyrailo,*,† Patrick J. McCloskey,‡ Ronald J. Wroczynski,† and William G. Morris†
General Electric Company, Global Research Center, Niskayuna, New York 12309
Polymers are important materials for sensor, microfluidic, and other demanding applications. High-throughput screening methodology has been applied for the evaluation of the solvent resistance of a family of polycarbonate copolymers prepared from the reaction of bisphenol A (BPA), hydroquinone (HQ), and resorcinol (RS) in different solvents of practical importance, such as chloroform, tetrahydrofuran (THF), and methyl ethyl ketone (MEK). We employed a 24-channel acoustic-wave sensor system that provided previously unavailable capabilities for parallel evaluation of polymer solvent resistance. This high-throughput polymer evaluation approach assisted in construction of detailed solvent-resistance maps of polycarbonate copolymers and in determination of quantitative structure-property relationships. The best absolute solvent resistance of all studied copolymers was achieved in MEK, followed by chloroform and THF. A D-optimal mixture design was employed to explore the relationship between the copolymer compositions and their solvent resistance. The applied special cubic model for each solvent took into account the primary mixture terms such as BPA, HQ, and RS; binary interaction terms such as BPA-HQ, BPA-RS, and HQ-RS; and a ternary interaction term BPA-HQ-RS. A combination of the normal distribution of the model residuals and the very high values of adjusted R2 (0.97-0.99) demonstrated a good quality of the model. At a HQ concentration of 40 mol %, the solvent resistance was the highest for all tested solvents, and different concentrations of BPA (40 and 60 mol %) and RS (0 and 20 mol %) did not affect the solvent resistance. Without HQ, solvent resistance was decreasing with an increase of RS and decrease of BPA. Overall, with an increase of HQ concentration from 0 to 40 mol %, the solvent resistance of BPA-HQ-RS copolymers was improved by up to 3 times in THF, by 21 times in chloroform, and by 32 times in MEK. Rational design of functional materials for chemical and biological sensors, organic electronics, coatings, and other de* Corresponding author. E-mail:
[email protected]. † Materials Analysis and Chemical Sciences. ‡ Polymer and Chemical Technologies.
3090 Analytical Chemistry, Vol. 78, No. 9, May 1, 2006
manding applications is very attractive. To be quantitatively successful, rational material design requires detailed knowledge about how material properties relate to material function. This knowledge is often obtained from extensive experimental and simulation data;1-3 however, the increase of structural and functional complexity of materials limits the ability to rationally define the precise requirements that result in desired properties.4 Combinatorial and high-throughput experimentation methods in materials science provide new opportunities in research areas for which the multidimensional nature of the interactions between the function and the composition, preparation method, and enduse conditions of materials makes rational design difficult or impossible.5-9 Using these methods, rapid synthesis and performance screening of a relatively large number of compounds is performed on a short time scale with automation and robotic equipment that operates with multiple small-scale samples. A variety of recently reported parallel combinatorial polymerization reactors10-13 can be applied to accelerate polymer synthesis. However, analytical tools for rapid evaluation of combinatorially produced polymers are much less advanced and are highly desired.12 For example, conventional evaluations of solvent effects on engineering polymers require exposure of a relatively large amount of a sample to a solvent and rely on measurements of (1) Newnham, R. E. Cryst. Rev. 1988, 1, 253-280. (2) Ulmer II, C. W.; Smith, D. A.; Sumpter, B. G.; Noid, D. I. Comput. Theor. Polym. Sci. 1998, 8, 311-321. (3) Suman, M.; Freddi, M.; Massera, C.; Ugozzoli, F.; Dalcanale, E. J. Am. Chem. Soc. 2003, 125, 12068-12069. (4) Schultz, P. G. Appl. Catal., A 2003, 254, 3-4. (5) Jandeleit, B.; Schaefer, D. J.; Powers, T. S.; Turner, H. W.; Weinberg, W. H. Angew. Chem., Int. Ed. 1999, 38, 2494-2532. (6) Potyrailo, R. A., Amis, E. J., Eds. High Throughput Analysis: A Tool for Combinatorial Materials Science; Kluwer Academic/Plenum Publishers: New York, 2003. (7) Koinuma, H.; Takeuchi, I. Nat. Mater. 2004, 3, 429-438. (8) Potyrailo, R. A., Takeuchi, I., Eds. Special Feature on Combinatorial and High-Throughput Materials Research; Meas. Sci. Technol. 2005, 16. (9) Potyrailo, R. A. Angew. Chem., Int. Ed. 2006, 45, 702-723. (10) Komon, Z. J. A.; Diamond, G. M.; Leclerc, M. K.; Murphy, V.; Okazaki, M.; Bazan, G. C. J. Am. Chem. Soc. 2002, 124, 15280-15285. (11) Boussie, T. R.; Diamond, G. M.; Goh, C.; Hall, K. A.; LaPointe, A. M.; Leclerc, M.; Lund, C.; Murphy, V.; Shoemaker, J. A. W.; Tracht, U.; Turner, H.; Zhang, J.; Uno, T.; Rosen, R. K.; Stevens, J. C. J. Am. Chem. Soc. 2003, 125, 4306-4317. (12) Hoogenboom, R.; Meier, M. A. R.; Schubert, U. S. Macromol. Rapid Commun. 2003, 24, 15-32. (13) Potyrailo, R. A.; Wroczynski, R. J.; Lemmon, J. P.; Flanagan, W. P.; Siclovan, O. P. J. Comb. Chem. 2003, 5, 8-17. 10.1021/ac0519662 CCC: $33.50
© 2006 American Chemical Society Published on Web 03/24/2006
solvent-induced cracks under stress.14-17 It is not possible to predict the solvent crazing resistance of a polymer across the whole spectrum of organic liquids;14 thus, environmental stress cracking resistance is required in applications for which engineering polymers are in contact with solvents because of the possibility of premature cracking and embrittlement of stressed or strained polymers.14-17 It has been reported that solubility of a variety of polymers in different solvents correlated15 or did not correlate16,17 with the environmental stress cracking resistance. Polymersolvent evaluations also rely on measurements of solvent uptake18-20 and weight loss.21 Since all of these determinations are manually performed, they cannot handle multiple samples to match the throughput of combinatorial polymerization reactor arrays. Our sensor-based high-throughput screening infrastructure22-24 is applied to better understand solvent resistance of polycarbonate copolymers for sensing and microfluidic applications. In sensing applications, formulated sensor materials are common for analysis of chemical and biological species in air and liquids.25-28 Often, a solvent which is used for the preparation of the sensor formulation can attack a plastic substrate of choice, forcing the use of either less attractive substrate materials or a complicated sensorassembly process. For example, ion-selective membranes were solvent-cast separately and attached to a poly(methyl methacrylate) (PMMA) substrate using an adhesive tape.25,26 Casting the membrane directly onto the substrate was impossible because the tetrahydrofuran solvent of the membrane formulation dissolved the PMMA surface and changed the membrane composition in an unpredictable manner.25 Thus, we employed solvent-resistant polycarbonate copolymers as supports for deposition of solventbased sensor formulations.29 Solvent-resistant polymers also attract interest for microfluidic applications as an alternative to glass and silicon19,30 Examples of solvent-resistant polymeric microfluidics (14) Kambour, R. P.; Gruner, C. L.; Romagosa, E. E. Macromolecules 1974, 7, 248-253. (15) Bernier, G. A.; Kambour, R. P. Macromolecules 1968, 1, 393-400. (16) Kambour, R. P.; Romagosa, E. E.; Gruner, C. L. Macromolecules 1972, 5, 335-340. (17) Li, X. Polym. Degrad. Stab. 2005, 90, 44-52. (18) Huang, J.-C.; Zhu, Z.-k.; Yin, J.; Qian, X.-f.; Sun, Y.-Y. Polymer 2001, 42, 873-877. (19) Rolland, J. P.; Van Dam, R. M.; Schorzman, D. A.; Quake, S. R.; DeSimone, J. M. J. Am. Chem. Soc. 2004, 126, 2322-2323. (20) Takeichi, T.; Ujiie, K.; Inoue, K. Polymer 2005, 46, 11225-11231. (21) Qi, Y.; Ding, J.; Day, M.; Jiang, J.; Callender, C. L. Chem. Mater. 2005, 17, 676-682. (22) Potyrailo, R. A.; Morris, W. G.; Wroczynski, R. J. In High Throughput Analysis: A Tool for Combinatorial Materials Science; Potyrailo, R. A., Amis, E. J., Eds.; Kluwer Academic/Plenum Publishers: New York, 2003, Chapter 11. (23) Potyrailo, R. A.; Morris, W. G.; Wroczynski, R. J. Rev. Sci. Instrum. 2004, 75, 2177-2186. (24) Potyrailo, R. A.; Morris, W. G.; Wroczynski, R. J.; McCloskey, P. J. J. Comb. Chem. 2004, 6, 869-873. (25) Johnson, R. D.; Badr, I. H. A.; Barrett, G.; Lai, S.; Lu, Y.; Madou, M. J.; Bachas, L. G. Anal. Chem. 2001, 73, 3940-3946. (26) Badr, I. H. A.; Johnson, R. D.; Madou, M. J.; Bachas, L. G. Anal. Chem. 2002, 74, 5569-5575. (27) Arshak, K.; Moore, E.; Cavanagh, L.; Harris, J.; McConigly, B.; Cunniffe, C.; Lyons, G.; Clifford, S. Composites, Part A 2004. (28) Cho, E. J.; Tao, Z.; Tang, Y.; Tehan, E. C.; Bright, F. V.; Hicks, W. L., Jr.; Gardella, J. A., Jr.; Hard, R. Appl. Spectrosc. 2002, 56, 1385-1389. (29) Potyrailo, R. A.; McCloskey, P. J.; Ramesh, N.; Surman, C. M. Sensor Devices Containing Copolymer Substrates for Analysis of Chemical and Biological Species in Water and Air; U.S. Patent Application 2005133697, 2005. (30) Harrison, C.; Cabral, J. T.; Stafford, C. M.; Karim, A.; Amis, E. J. J. Micromech. Microeng. 2004, 14, 153-158.
include systems for organic-phase synthesis,31 polymer synthesis,32 studies of polymeric and colloidal formulations,30 high-throughput measurements of immiscible fluids,32 and microfluidic combinatorial polymer research.33 Polycarbonates, their copolymers, and blends are used as materials for fabrication of microfluidic components34-38 and gas sensor materials.39,40 Einhorn demonstrated reactions of hydroquinone or resorcinol with phosgene to prepare polycarbonates.41 Hydroquinone polycarbonates were studied by Kricheldorf and Lubbers.42 Several binary and ternary copolycarbonates were prepared using monomers such as hydroquinone, biphenol, and substituted hydroquinones. Solvent resistance of hydroquinone polycarbonates has been reported.43 Brunelle reviewed in detail blends and copolymers of polycarbonate with dramatically improved solvent resistance.44 It was pointed out that polymerization of the hydroquinone and bisphenol A (BPA) cocyclics via anionically initiated, ring-opening polymerization leads to high molecular weight semicrystalline polymers with the capability of incorporating hydroquinone into the polycarbonate.44 These polycarbonates showed dramatically increased solvent resistance. In this study, we evaluate solvent resistance of several polycarbonate copolymers prepared from the reaction of hydroquinone (HQ), resorcinol (RS), and BPA. In sensor and microfluidic applications, our need is to have good solvent resistance of the polymers to prevent degradation of the substrate surface upon deposition of a sensor formulation and to prevent contamination of the solvent-containing sensor formulations or contamination of organic liquid reactions in microfluidic channels. Unfortunately, no comprehensive quantitative reference solubility data of unstressed copolymers is available to date.14-17,44 Our highthroughput polymer evaluation approach permitted the construction of detailed solvent-resistance maps, the development of quantitative structure-property relationships for BPA-HQ-RS copolymers, and provided new knowledge for the further development of the polymeric sensor and microfluidic components. CONCEPT OF COMBINATORIAL SCREENING OF COPOLYMER-SOLVENT INTERACTIONS In this study, we evaluate the solvent resistance of polycarbonate copolymers prepared from the reaction of bisphenol A, hydroquinone, and resorcinol. Synthesis of these copolymers has (31) Cygan, Z. T.; Cabral, J. T.; Beers, K. L.; Amis, E. J. Langmuir 2005, 21, 3629-3634. (32) Cabral, J. T.; Hudson, S. D.; Wu, T.; Beers, K. L.; Douglas, J. F.; Karim, A.; Amis, E. J. Polym. Mater. 2004, 90, 337-338. (33) Amis, E. J. Nat. Mater. 2004, 3, 83-85. (34) Martin, P. M.; Matson, D. W.; Bennett, W. D.; Lin, Y.; Hammerstrom, D. J. J. Vac. Sci. Technol. 1999, A 17, 2264-2269. (35) Becker, H.; Locascio, L. E. Talanta 2002, 56, 267-287. (36) Soper, S. A.; Henry, A. C.; Vaidya, B.; Galloway, M.; Wabuyele, M.; McCarley, R. L. Anal. Chim. Acta 2002, 470, 87-99. (37) Erickson, D.; Li, D. Anal. Chim. Acta 2004, 507, 11-26. (38) Madou, M. J. Fundamentals of Microfabrication. The Science of Miniaturization; CRC Press: Boca Raton, FL, 2002. (39) Freud, M. S.; Lewis, N. S. Proc. Natl. Acad. Sci. U.S.A. 1995, 92, 26522656. (40) Sivavec, T. M.; Potyrailo, R. A. Polymer Coatings for Chemical Sensors; U.S. Patent 6,357,278 B1, 2002. (41) Einhorn, A. Lieb. Ann. Chem. 1898, 300, 135-155. (42) Kricheldorf, H. R.; Lu ¨ bbers, D. Makromol. Chem. Rapid Commun. 1989, 10, 383-386. (43) Schnell, H. Angew. Chem. 1956, 68, 633-640. (44) Brunelle, D. J. Trends Polym. Sci. 1995, 3, 154-158.
Analytical Chemistry, Vol. 78, No. 9, May 1, 2006
3091
Figure 1. Composition space of BPA-HQ-RS copolymers evaluated using high-throughput sensor-based system. Red data points signify materials that were employed to build quantitative structureproperty relationships. Numbers are mole percent of monomers. Table 1. Composition of Copolymers monomer (mol %) sample no.
bisphenol A
hydroquinone
resorcinol
1 2 3 4 5 6 7
100 80 80 60 70 60 40
0 0 20 40 20 20 40
0 20 0 0 10 20 20 Figure 3. Acoustic-wave TSM sensor array system for mapping of solvent resistance of polycarbonate copolymers: (a) sensor resonators (8-mm diameter) arranged as a 6 × 4 array, compatible with available 24-well plates and (b) measurement schematic.
Figure 2. Concept of determination of polymer solvent resistance using TSM acoustic-wave sensors: (a) periodic exposure and withdrawal of a sensor resonator into a polymer solution and (b) determination of the dissolution rate of polymers in solvents from repetitive measurements of deposited polymers during the experiment.
been described elsewhere.44 The composition space of these BPA-HQ-RS copolymers is presented Figure 1. The amounts of monomers employed for copolymer synthesis are presented in Table 1. Pure HQ and RS polymers have poor mechanical properties44 and are of little practical interest. Thus, only a composition range shown in Figure 1 was studied. Highthroughput evaluation of the solvent resistance of BPA copolymers was performed in different solvents of practical importance, such as chloroform, tetrahydrofuran (THF), and methyl ethyl ketone (MEK). Measurements were performed using our acoustic-wave sensor array system designed for high-throughput materials characterization and operated in the thickness shear mode (TSM).22-24 TSM sensors are widely used for materials characterization.45 Our screening concept is depicted in Figure 2. For simplicity, operation of only one of the sensors from the sensor array is (45) Thompson, M.; Stone, D. C. Surface-Launched Acoustic Wave Sensors: Chemical Sensing and Thin-Film Characterization; Wiley: New York, 1997.
3092
Analytical Chemistry, Vol. 78, No. 9, May 1, 2006
illustrated. An initially clean sensor crystal is exposed to a solvent containing a small amount of polymer, followed by sensor withdrawal from the solution (Figure 2a). Quantification of the residual dissolved polymer deposited onto the sensor is performed by the measurements of the mass increase of the crystal, which is proportional to the amount of polymer film deposited onto the sensor from a polymer solution. Depending on the solvent resistance of the polymers, the dissolution rate of polymers in solvents will be different as determined from multiple measurements of deposited polymers during the experiment (Figure 2b). The measurements of frequency changes are performed when the sensors are periodically withdrawn from the solvents and the solvents are evaporated. In this way, the measured signal change is indicative of the amount of the deposited material from the solution given by45
∆fF ) -2f02(mF/A)(µQFQ)-1/2
(1)
where f0 is the fundamental resonant frequency of an unloaded device; µQ is the shear modulus of the piezoelectric AT-cut quartz substrate, 2.947 × 1011 g cm-1 s-2; FQ is the substrate density, 2.648 g cm-3; mF is the total mass of the coating deposited to both faces of the crystal; and A is the active surface area of one face of the crystal. EXPERIMENTAL SECTION Polymers and Solvents. Polycarbonate copolymers were prepared at GE Global Research as detailed earlier.44 Typical
molecular weights of all copolymers were in the range of 22 00033 000 g/mol with polycarbonate standards used for calibration. These slight differences in molecular weights did not produce detectable changes in solvent resistance for the same copolymer compositions as determined by traditional solvent-resistance measurements.17 Prepared copolymers were tested for solvent resistance in chloroform, THF, and MEK solvents (Mallinckrodt Baker Inc., Phillipsburg, NJ, J. T. Baker brand, reagent grade). High-Throughput Polymer-Solvent Screening. Measurements were performed using a 24-channel acoustic-wave sensor array system shown in Figure 3. Design details of the sensor system were reported earlier.22-24 For this application, sensor resonators were 8-mm-diameter AT-cut quartz crystals (International Crystal Manufactures, Oklahoma City, OK) polished with an optical finish of
