Nanophase-Separated Amphiphilic Conetworks as Versatile Matrixes

Sebastian UlrichAmin SadeghpourRené M. RossiNico BrunsLuciano F. Boesel .... Pietro La Manna , Pellegrino Musto , Giancarlo Galli , Elisa Martinelli...
0 downloads 0 Views 663KB Size
Download PDF
Anal. Chem. 2006, 78, 6376-6383

Nanophase-Separated Amphiphilic Conetworks as Versatile Matrixes for Optical Chemical and Biochemical Sensors Michael Hanko,† Nico Bruns,‡ Sara Rentmeister,† Jo 1 rg C. Tiller,*,‡ and Ju 1 rgen Heinze*,†

Freiburg Materials Research Center (FMF), Institute of Physical Chemistry and Institute of Macromolecular Chemistry, Albert-Ludwigs-University Freiburg, Stefan-Meier-Strasse 21, 79104 Freiburg, Germany

As a novel class of sensor matrixes, nanophase-separated amphiphilic polymeric conetworks (APCNs) open a new dimension for optical chemical and biochemical sensing. These conetworks consist of a hydrophilic phaseswe used poly(2-hydroxyethyl acrylate), poly(2-(dimethylamino)ethyl acrylate), or polycationic poly(2-(trimethylammonium)ethyl acrylate)sand of a hydrophobic phasespoly(dimethylsiloxane). Sensors can be prepared by simple impregnation of the matrix. Due to nanophase separation, there is a spatial separation between areas in which the indicator reagents are well immobilized and areas that advantageously take care of the diffusive transport of the analyte, whereby these functionalities of the contrary phases can be exchanged. Thanks to the huge interface between the contrary phases, the accessibility of the indicator reagents is good, which makes it possible to design sensors with high sensitivity. To demonstrate the advantages of APCNs as matrixes, different prototypes of sensors were prepared, e.g., one to determine gaseous chlorine based on its reaction with immobilized o-tolidine and another to determine vaporous acids based on immobilized bromophenol blue dianions. As a breakthrough in biochemical sensing, we are also able to present an easily producible, optically transparent biochemical sensor to determine peroxides in nonpolar organic medias based on coimmobilized horseradish peroxidase and 2,2′azino-bis(3-ethylbenzothiazoline-6-sulfonate). Chemical and biochemical sensor technologies play an important role in areas where rapid detection, high sensitivity, and specificity are important.1 A large number hereof are optical (bio)chemical sensors (optodes2) based on immobilized indicator reagents. They are widely used because they are easy to produce and allow for cost-efficient and quick measurements.3-8 * To whom correspondence concerning the polymers should be addressed: (fax) (49) 761 203 4709; (e-mail) [email protected]. To whom correspondence concerning the sensors should be addressed: (fax) (49) 761 203 6237; (e-mail) [email protected]. † Institute of Physical Chemistry. ‡ Institute of Macromolecular Chemistry. (1) Weetall, H. H. Biosens. Bioelectron. 1999, 14, 237-242. (2) Luebbers, D. W.; Opitz, N. Z. Naturforsch. C 1975, 30, 532-533. (3) Wolfbeis, O. S. Anal. Chem. 2000, 72, 81R-89R. (4) Wolfbeis, O. S. Anal. Chem. 2002, 74, 2663-2678. (5) Wolfbeis, O. S. Anal. Chem. 2004, 76, 3269-3284.

6376 Analytical Chemistry, Vol. 78, No. 18, September 15, 2006

Generally, indicator reagents and enzymes are immobilized by covalent or noncovalent methods. Covalently, indicator reagents and enzymes can be immobilized by coupling to polymeric supports. Furthermore, enzymes can be immobilized by crosslinking with bifunctional reagents or through an avidin-biotin linkage.9-12 Noncovalent methods include adsorption onto transparent supports, doping or impregnation of the sensor matrixes, entrapment in cross-linked polymeric materials, or simple containment using membrane or film devices permeable only for the analytes. The most common methods are doping and impregnation of the sensor matrixes.13 In these cases, the indicator reagents and the enzymes are dissolved in the sensor matrixes and stabilized by molecular forces such as van der Waals forces, dipolar interactions, ionic interactions, hydrogen bonds, or a combination of these forces.7,8,14-17 The matrix itself has to be transparent, chemically inert, and permeable for the analytes and should sufficiently stabilize the indicator reagent. Most materials for optode matrixes are either hydrophilic or hydrophobic16 and can be classified in the following groups: sol-gels, porous glasses, hydrogels, and the usual polymers.3-5,13,18-25 In particular, poly(siloxanes), fluoropolymers, (6) Ralfs, M.; Heinze, J. Sens. Actuators, B 1997, 44, 257-261. (7) Alexy, M.; Voss, G.; Heinze, J. Anal. Bioanal. Chem. 2005, 382, 16281641. (8) Alexy, M.; Hanko, M.; Rentmeister, S.; Heinze, J. Sens. Actuators, B 2006, 114, 916-927. (9) Bickerstaff, G. F., Ed. Immobilization of Enzymes and Cells; Humana Press: Totowa, NJ, 1997. (10) Tiller, J.; Berlin, P.; Klemm, D. Biotechnol. Appl. Biochem. 1999, 30, 15562. (11) Kuswandi, B.; Andres, R.; Narayanaswamy, R. Analyst 2001, 126, 14691491. (12) Konry, T.; Novoa, A.; Cosnier, S.; Marks, R. S. Anal. Chem. 2003, 75, 26332639. (13) Wolfbeis, O. S.; Reisfeld, R.; Oehme, I. Struct. Bonding 1996, 85, 51-98. (14) Klimant, I.; Wolfbeis, O. S.; Leiner, M. J. P.; Karpf, H.; Kovacs, B. Eur. Pat. Appl. EP 93-890131, 1994. (15) Heinze, J.; Ralfs, J. Ger. Offen. DE 99-19957708, 2001. (16) Oehme, I.; Wolfbeis, O. S. Mikrochim. Acta 1997, 126, 177-192. (17) Stangelmayer, A.; Klimant, I.; Wolfbeis, O. S. Fresenius J. Anal. Chem. 1998, 362, 73-76. (18) Schooner, F.; Simard, R. E.; Pandian, S. J. Food Sci. 1991, 56, 1229-1232. (19) Lev, O. Analusis 1992, 20, 543-553. (20) Lev, O.; Kuyavskaya, B. I.; Sacharov, Y.; Rottman, C.; Kuselman, A.; Avnir, D.; Ottolenghi, M. Proc. SPIE-Int. Soc. Opt. Eng. 1993, 1716, 357-366. (21) Zemskii, V. I.; Kolesnikov, Y. L.; Novikov, A. F. J. Opt. Technol. (Translation of Opticheskii Z.) 1998, 65, 765-769. (22) Ohyama, T.; Maruo, Y. Y.; Tanaka, T.; Hayashi, T. Sens. Actuators, B 2000, 64, 142-146. 10.1021/ac060634+ CCC: $33.50

© 2006 American Chemical Society Published on Web 07/27/2006

Figure 1. Loading scheme of the nanophase-separated amphiphilic conetworks: Loading of the hydrophobic phase with a hydrophobic indicator reagent dissolved in a nonpolar solvent (upper way); loading of the hydrophilic phase with a hydrophilic indicator reagent, enzyme, or both dissolved in a polar solvent (lower way).

organic glassy polymers (e.g., polystyrene, poly(methyl methacrylate), poly(vinyl chloride)), and cellulose derivatives are widely used. They are also used as copolymers and with different amounts of plasticizers increasing the gas permeability.16,26-31 Given the large number of indicator reagents, their different chemical and physical properties, and the different fields of optode application, it is not surprising that up to now no class of matrix materials has been known that enables the versatile immobilization of indicator reagents and simultaneously fulfills the other properties required for optode matrixes, such as optical transparency and clarity, chemical and mechanical stability, and high analyte permeability.27 Here, we present a technology that provides a novel and versatile class of matrix materials that allows for the good immobilization of indicator reagents and of enzymes and simultaneously provides for a high permeability for the analytes. It is based on nonporous, nanostructured amphiphilic polymeric conetworks (APCNs)32 with a cocontinuous morphology of the phases and with an extremely large interface between the conetworks’ (23) Wolfbeis, O. S.; Oehme, I.; Papkovskaya, N.; Klimant, I. Biosens. Bioelectron. 2000, 15, 69-76. (24) Rickus J. L.; Dunn, B.; Zink, J. I. In Optical Biosensors: Present and Future; Ligler, F. S., Rowe Taitt, C. A., Eds.; Elsevier Science B.V.: Amsterdam; The Netherlands, 2002; pp 427-456. (25) Rayss, J. Proc. SPIE-Int. Soc. Opt. Eng. 2005, 5855, 17-19. (26) Morf, W. E.; Seiler, K.; Rusterholz, B.; Simon, W. Anal. Chem. 1990, 62, 738-742. (27) Baldini, F.; Bracci, S. In Polymer Sensors and Actuators; Osada, Y., De Rossi, D. E., Eds.; Springer-Verlag: Berlin; Germany, 2000; pp 91-107. (28) Tiller, J.; Berlin, P.; Klemm, D. J. Appl. Polym. Sci. 2000, 75, 904-915. (29) Nezel, T.; Fakler, A.; Zhylyak, G.; Mohr, G. J.; Spichiger-Keller, U. E. Sens. Actuators, B 2000, 70, 165-169. (30) Amao, Y. Microchim. Acta 2003, 143, 1-12. (31) Berlin, P.; Klemm, D.; Jung, A.; Liebegott, H.; Rieseler, R.; Tiller, J. Cellulose 2003, 10, 343-367. (32) Erdodi, G.; Kennedy, J. P. Prog. Polym. Sci. 2006, 31, 1-18.

hydrophilic and hydrophobic phases due to nanophase separation. Amphiphilic conetworks are a class of promising materials for applications such as contact lenses,33 pervaporation membranes,34 drug delivery systems,35 and catalyst supports.36-38 The APCNs are optically transparent; they can be produced easily and subsequently loaded with indicator reagents and, in addition, with enzymes by simple impregnation to prepare (bio)chemical sensors. Therefore, this method does not require additives or chemical modifications of the indicator reagents or of the enzymes. The principle of APCNs as matrixes for the immobilization of indicator reagents and enzymes is based on the peculiar swelling properties of these polymers. Because in polar liquids the hydrophilic phase is swollen, this phase can be separately loaded with hydrophilic indicator reagents and enzymes, if desired. The dissolved indicator reagents/enzymes diffuse from their solution into the swollen hydrophilic phase of the conetwork. Upon drying, the phase shrinks, entrapping the indicator reagent/enzyme in a hydrophilic and enzyme-friendly environment.37 In the same way, the hydrophobic phase can be loaded separately with hydrophobic indicator reagents from their solution in nonpolar organic solvents (Figure 1). Advantageously, one phase of the amphiphilic conetwork immobilizes the indicator reagent/enzyme and the contrary phase is responsible for the diffusion of the analyte to the internal (33) Nicolson, P. C.; Vogt, J. Biomaterials 2001, 22, 3273-3283. (34) Du Prez, F. E.; Goethals, E. J.; Schue, R.; Qariouh, H.; Schue, F. Polym. Int. 1998, 46, 117-125. (35) Keszler, B.; Kennedy, J. P.; Mackey, P. W. J. Controlled Release 1993, 25, 115-121. (36) Kralik, M.; Zecca, M.; Bianchin, P.; D’Archivio, A. A.; Galantini, L.; Corain, B. J. Mol. Catal. A 1998, 130, 85-93. (37) Bruns, N.; Tiller, J. C. Nano Lett. 2005, 5, 45-48. (38) Savin, G.; Bruns, N.; Thomann, Y.; Tiller, J. C. Macromolecules 2005, 38, 7536-7539.

Analytical Chemistry, Vol. 78, No. 18, September 15, 2006

6377

interface between the phases, where the (enzyme-catalyzed) reaction with the indicator reagent occurs. In the case of sensors that are applied in liquids, the indicator reagent/enzyme has to be immobilized in the phase of the conetwork with the contrary polarity of the liquid sample. It is obvious that in liquids the permeability of the amphiphilic conetwork for the analyte is high, concerning the swelling of the (unloaded) phase. In the case of sensors that were designed for the application in gas phases, the accessibility of the immobilized indicator reagent depends on the specific (gas) permeability of the conetwork phases. The spatial separation between areas for the immobilization of the indicator reagent and areas for the diffusive transport of the analyte, combined with the huge interface between the contrary phases due to nanophase separation, causes great advantages. Especially for chemical gas sensors, the combination of a highly functionalized phase providing for a good and stable immobilization of the indicator reagent/enzyme and of a highly gas-permeable contrary phase (e.g., silicones) is very beneficial. This paper reports on prototypes of sensors that possess an increased sensitivity compared to sensors based on the usual matrix materials. We also present a sensor with a polycationic amphiphilic matrix that has good mechanical properties. To the best of our knowledge, for the first time, nanophase-separated amphiphilic conetworks have also enabled the design and easy preparation of optically clear and transparent biochemical sensors that can be used in nonpolar organic media. EXPERIMENTAL SECTION Methods. Attenuated total reflection Fourier transform infrared spectroscopy (ATR-FTIR) was carried out on a Bruker Vektor 22 spectrometer, equipped with a Golden Gate accessory (Specac). Tapping-mode atomic force microscopy (AFM) was carried out with a Nanoscope III scanning probe instrument (Digital Instruments) at ambient conditions in phase mode using NCL-W (tapping mode) cantilevers (Nanosensors). The measurements were performed on the surface of the samples and on cryofractures of the films. UV/VIS/SWIR absorption spectra (190-2700 nm) were received in transmission mode with an Omega 20 spectrophotometer (Bruins Instruments). Gas Preparation System. Gaseous chlorine was taken from a test gas cylinder (Messer Griesheim) and diluted with synthetic air (20.5% O2 in N2) from another gas cylinder. Digital mass flow controllers 5850S (Brooks) and F-201C-RAA-33-E (Bronkhorst HiTec) with an adjustable range of gas flow between 0 and 1000 mL min-1 were used to adjust the gas concentrations. The total gas flow was fixed at a constant flow of 1000 mL min-1 for all measurements. Gas flows were monitored by mass flow meters EL-FLOW (Bronkhorst Hi-Tec). All experiments were carried out under ambient conditions and at room temperature (23 ( 2 °C). To keep the relative humidity (RH) constant, an adjustable volume of synthetic air was saturated with water with help of a gas-washing bottle. The same method was used to adjust the concentration of acetic acid. The concentration was determined from the timedependent mass loss of acetic acid in the evaporator. Relative humidity and temperature were measured with a Sensmitter SHT 75 (Driesen+Kern). The sensors were exposed to the gases by mounting them inside a flow cell with a cross sectional area of 0.5 × 1.0 cm2 (Hellma) placed in the optical path of the spectrophotometer. The 6378

Analytical Chemistry, Vol. 78, No. 18, September 15, 2006

flow cell was replaced by a normal quartz cuvette (Hellma) for measurements in liquid media. Materials. Bifunctional poly(dimethylsiloxane) (PDMS) macromonomers (R,ω-methacryloxypropyl poly(dimethylsiloxane) (MA-PDMS-MA)) (Mn ) 5200 g mol-1) and 2-(trimethylsilyloxy)ethyl acrylate (TMSOEA) were synthesized as described previously.39,40 The other acrylates were obtained from Alfa Aesar (Karlsruhe, Germany) and distilled before use. All other reagents and solvents were p.a. grade and used without further treatment. Preparation of Surface-Attached APCN Films. For our experiments, we chose nanophase-separated amphiphilic polymeric conetworks prepared by radical copolymerization of differently substituted acrylic monomers and bifunctional MA-PDMSMA macromonomers. Two different types of APCNs were prepared: nonionic and cationic. The nonionic conetworks consisted either of poly(2-hydroxyethyl acrylate) (PHEA) and PDMS (referred to as PHEA-l-PDMS) in ratios of 25:75, 58:42 and 77:23 wt % or of 36 wt % poly(2-(dimethylamino)ethyl acrylate) (PDMAEA) and 64 wt % PDMS (PDMAEA-l-PDMS). The cationic conetwork consisted of poly(2-(trimethylammonium)ethyl acrylate) (PTMAEA) linked by PDMS (PTMAEA-lPDMS) and was prepared from PDMAEA-l-PDMS by quaternization with dimethyl sulfate. For the preparation of surface-attached APCN films, methacrylate-modified glass slides (15 × 7 × 1 mm3) were prepared as described earlier.39 An unmodified glass sheet (200 × 150 × 3 mm3) was coated with several parallel strips of an adhesive poly(propylene) (PP) tape (Tesa film) that served as spacers to adjust the thickness of the films. Afterward, 12 µL of the monomer mixture was dropped under an argon atmosphere between two PP spacers onto the unmodified glass sheet and a methacrylatemodified glass slide was immediately placed on top. The initial batches of the conetwork films were prepared in the same way, but without using PP spacers. For polymerization, the slides were irradiated six times for 60 s from a distance of 30 cm using a 500-W Nitraphot light bulb (Osram), each time interrupted for 15 s to avoid excessive heat. After removing the conetwork-coated slides from the bottom glass sheet, they were immersed in different solvents as described below to remove residual monomers, remnants of the photoinitiator, solvents, and protecting groups, if present. The reaction scheme for the preparation of the different APCN films is shown in Figure 2. PHEA-l-PDMS Films. The monomer mixture consists of the calculated volumes of TMSOEA and MA-PDMS-MA, in which 26 mmol L-1 photoinitiator Lucirin TPO (BASF) was dissolved. After bulk polymerization, the conetwork-coated slides were immersed three times in a 50 vol % aqueous solution of methanol for 1 day in order to remove the protecting trimethylsilyl groups. The completeness of this removal was controlled by ATR-FT-IR measurements. Then the slides were rinsed with methanol and dried at 40 °C/0.1 mbar overnight. PDMAEA-l-PDMS Films. The monomer solution consists of 33 vol % Decalin as solvent, 22 vol % 2-(dimethylamino)ethyl acrylate (DMAEA), 45 vol % MA-PDMS-MA, and 17.3 mmol L-1 (39) Bruns, N.; Scherble, J.; Hartmann, L.; Thomann, R.; Iva´n, B.; Mu ¨ lhaupt, R.; Tiller, J. C. Macromolecules 2005, 38, 2431-2438. (40) Scherble, J.; Iva´n, B.; Mu ¨lhaupt, R. Macromol. Chem. Phys. 2002, 203, 18661871.

Figure 2. Reaction scheme for the preparation of the amphiphilic conetwork films: (a) PHEA-l-PDMS, R ) O-(CH2)2-OH; (b) PDMAEA-lPDMS, R ) O-(CH2)2-N(CH3)2; (c) PTMAEA-l-PDMS, R ) O-(CH2)2-N(CH3)3+‚1/2SO42-.

Lucirin TPO. After polymerization, the conetwork-coated slides were incubated in dichloromethane three times for 1 day in order to remove the Decalin. Then the slides were incubated in methanol for another day and dried at 40 °C/0.1 mbar overnight. PTMAEA-l-PDMS Films. For the quaternization of the PDMAEA-l-PDMS films, they were immersed in a 10 vol % solution of dimethyl sulfate in acetonitrile for 1 h at room temperature, rinsed with acetonitrile, and incubated in fresh acetonitrile for another hour. The solvent was replaced by water overnight, and then the slides were dried at 40 °C/0.1 mbar for 1 day. Loading of the APCN Films with Indicator Reagents. To prepare the prototypes of (bio)chemical sensors, the amphiphilic polymeric conetwork films were loaded with indicator reagents and, where necessary, with enzymes by impregnation. PHEA-l-PDMS/o-Tolidine Sensors. The hydrophilic phase of the PHEA-l-PDMS films (25:75 and 77:23) (prepared without using PP spacers) was loaded with o-tolidine by immersing them in a 3 mg mL-1 (14.1 mmol L-1) solution of o-tolidine in methanol for 30 min. The sensors were air-dried, residual o-tolidine was removed from the rear side of the glass carriers with a cotton bud, and then the colorless and optically clear sensors were dried for 60 min at 25 °C/0.1 mbar. PTMAEA-l-PDMS/BPB Sensors. The hydrophilic phase of the PTMAEA-l-PDMS films was loaded with bromophenol blue dianions (BPB) by an anion exchange procedure. For this, PTMAEAl-PDMS films containing sulfate as anions were immersed in a 107 µmol L-1 aqueous solution of bromophenol blue sodium salt (pH 7) for 60 min and washed three times with water. Then the deep blue and optically clear sensors were dried overnight at 40 °C/0.1 mbar. PHEA-l-PDMS (58:42)/HRP/ABTS Sensors. Optochemical biosensors for determination of peroxides in nonpolar organic solvents were prepared by immersing the PHEA-l-PDMS (58:42) films in a 0.1 mol L-1 phosphate buffer (pH 7.0) overnight. Then the conetwork films were transferred in 2 mL of a solution containing 1 mg mL-1 horseradish peroxidase (HRP) and 2,2′azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) diammonium salt (ABTS); 1.7 mmol L-1) in phosphate buffer and incubated

overnight. The colorless and optically clear biochemical sensors obtained were washed with phosphate buffer and dried under nitrogen at room temperature until constant weight. RESULTS AND DISCUSSION Properties of the Amphiphilic Conetwork Films. Covalently attached amphiphilic conetwork films of the same batch as well as of different batches were easy to reproduce. All types of prepared conetwork films were optically clear and transparent. The absorption spectra of the dry conetwork films in the shortwavelength infrared (SWIR) were suitable means of controlling the relative film thickness. Several distinctive absorption bands could be associated with the polyacrylates (2281, 2342, and 2481 nm) and with the PDMS (2292, 2371, 2468, and 2650 nm), as Figure 3 demonstrates. The absolute thicknesses of the films were measured on cryofractures of the conetwork films with an optical microscope (Olympus Optical Co.) and with help of a P-10 Long Scan Profiler (KLA-Tencor). The thickness of the PHEA-l-PDMS films of the initial batches, which were prepared without using PP spacers, was found to be 20 ( 5 µm. Films, which were prepared using PP spacers, had thicknesses of 60 ( 4 (PHEA-l-PDMS (58:42)) and 43 ( 3 µm (PDMAEA-l-PDMS). The deviation in film thickness is owing to a small amount of the monomer mixture/ monomer solution that crept between the PP spacer and the methacrylate-modified glass slide when the slide was placed on the monomeric drop. The nanoscopic morphology of similar amphiphilic conetwork films has already been described.39 Figure 4 shows AFM images of the prepared APCN films, that were obtained in the phase mode, visualizing the nanoscopic phase separation of these amphiphilic conetworks. In the images presented here, the hard polyacrylic phase appears brighter than the soft PDMS phase. The AFM measurements were performed on the surface of the conetwork films and on film cross sections made by fracturing the films at -196 °C. In all cases, the AFM images performed on the surface of the films showed for the most part the same morphology as the bulk morphology performed on the cross sections. The images Analytical Chemistry, Vol. 78, No. 18, September 15, 2006

6379

Figure 3. Normalized SWIR absorption spectra of a PHEA film, of MA-PDMS-MA (dissolved in CCl4), and of the prepared APCN films.

of PHEA-l-PDMS (25:75) and PDMAEA-l-PDMS indicated a cocontinuous spongelike morphology, in which interconnected PDMS domains were enclosed in the polyacrylic phase, which formed the walls of the sponge structure. The diameter of the PDMS domains was in the range from 12 to 38 nm; the polyacrylic walls had a thickness of 2-20 nm. To apply APCNs as sensor matrixes, an independent swelling of the polymer phases was beneficial. This assumed that both the hydrophilic polyacrylic walls and the hydrophobic PDMS domains were interconnected throughout the bulk and present on the surface of the films, as the AFM images of PHEA-l-PDMS (25:75) and PDMAEA-l-PDMS films demonstrate. Figure 4 also reveals the huge interface between the hydrophilic and the hydrophobic phases. The differences in the nanoscopic morphology between PDMAEA-l-PDMS films (Figure 4b) and the corresponding qua-

ternized PTMAEA-l-PDMS films (Figure 4c) were impressive. In contrast to the nonquaternized film, the AFM image of the PTMAEA-l-PDMS film indicated a morphology where spherical PDMS domains (12-54 nm in diameter) were embedded in a polyacrylic matrix. Moreover, some of the PDMS domains have coalesced, probably due to the volume increase of the polyacrylic phase during quaternization. It was obvious that the hard polyacrylic PTMAEA phase required much more space than in the nonquaternized state, owing to the high charge of the polyacrylic phase. Thus, the thickness of the films increased ∼20% during quaternization. Although in PTMAEA-l-PDMS films only the hydrophilic phase showed a continuous morphology, we used these films as sensor matrixes for the immobilization of anionic indicator reagents because of theirsconsidering the high charge of the polymersgood mechanical properties and optical transparency, to demonstrate the versatility of our concept. A possibility to get charged conetworks with a cocontinuous morphology consists of the copolymerization of a charged acrylate together with an indifferent acrylate as hydrophilic phase,41 but this was not accomplished in this work. The quaternization of the PDMAEA-l-PDMS films resulted in the formation of a new absorption band in the SWIR spectrum at 2250 nm (cf. Figure 3). The quaternizing ratio of the tertiary ammonium groups of the PDMAEA-l-PDMS films was determined by this absorption band to be 97%. To address the question of the anion exchange capacity of the quaternized PTMAEA-l-PDMS films, some of the films were immersed in a 50 µmol L-1 aqueous solution of potassium dichromate for 60 min, washed twice with water, and dried at room air. The chromate, absorbed by the polymeric film through anionic exchange, caused an absorption band with maximum at 372 nm. With the aid of the LambertBeer law, the chromate concentration inside the amphiphilic conetwork was calculated and compared with the initial chromate concentration of the loading solution. The result was a concentration increase of a factor of 273, indicating a high exchange capacity of the PTMAEA-l-PDMS films for chromate anions. Characteristics of the Sensors. Determining Gaseous Chlorine with PHEA-l-PDMS/o-Tolidine Sensors. o-Tolidine, immobilized in a matrix of the block copolymer poly(dimethylsiloxane)-b-poly(carbonate) (PDMS-b-PC), was used in previous studies as a sensitive indicator reagent for the design of disposable sensors determining Cl2.6 Because one of the main influences to the sensors’ sensitivity is the permeability of the matrix for the analyte,

Figure 4. AFM phase mode images of conetwork film surfaces: PHEA-l-PDMS (25:75) (a), PDMAEA-l-PDMS (b), and PTMAEA-l-PDMS (c). The polyacrylic phase appears bright; PDMS shows dark. 6380 Analytical Chemistry, Vol. 78, No. 18, September 15, 2006

Figure 5. Absorption spectra before and after exposure to 0.5 ppm Cl2 (0% RH, 23 °C, 1 L min-1, 45 min) (a), and initial absorption change at 655 nm of a PHEA-l-PDMS (25:75)/o-tolidine sensor (b); comparison with the response of a PHEA-l-PDMS (77:23)/o-tolidine sensor (c). The exposures were started after 10 s.

the sensitivity of sensors made of o-tolidine immobilized in amphiphilic conetworks toward gaseous chlorine was compared to sensors consisting of the highly gas-permeable PDMS-b-PC matrix. For this, PHEA-l-PDMS (25:75)/o-tolidine sensors were exposed to 0.5 ppm (v) Cl2 in synthetic air. The presence of Cl2 resulted in the formation of a colored radical cation of o-tolidine, which shows absorption bands with maximums found at 381, 655, and 912 nm. Figure 5a shows the absorption spectra before and after 45 min of exposure of such a sensor. The plot of the absorption change at 655 nm versus exposure time showed a linear dependency with a steep slope at the beginning of the exposure (Figure 5b). According to Ralfs,6 the rate of absorption change in the linear region after beginning of the exposure depends on the concentra(41) Tiller, J. C.; Sprich, C.; Hartmann, L. J. Controlled Release 2005, 103, 355367.

tion of Cl2 and on the amount of accessible o-tolidine. The steeper the slope in this region at a given gas concentration, the more sensitive is the applied sensor. A linear regression of the absorption values at 655 nm after beginning of the exposures analogous to Ralfssresulted in a rate of absorption change of 0.045 min-1. Compared to sensors using PDMS-b-PC as matrix material (0.038 min-1 at the wavelength of the absorption maximum in PDMS-b-PC of 665 nm), this was an increase of ∼20% in sensitivity. The comparison between the concentrations of o-tolidine, immobilized in the different polymeric matrixes, was even more impressive. In PDMS-b-PC films (thickness 0.58 µm), the concentration of o-tolidine was 247 mmol L-1. In PHEA-l-PDMS (25: 75) films (∼20 µm in thickness), the concentration of o-tolidine was found to be only 25 mmol L-1 (mean concentration relating to both phases), determined by UV/visible spectroscopy. The higher rate of absorption change even at a lower concentration of o-tolidine indicated the good accessibility of the indicator reagent even in deeper layers of the ACPN matrix. Storing the prepared sensors at room temperature under an argon atmosphere had no influence on their stability and reactivity during the examination period of two weeks. Interestingly, sensors with amphiphilic conetworks consisting of 77 wt % PHEA (PHEA-l-PDMS (77:23)) were much less sensitive (Figure 5c). This can be explained by the fact that in PHEA-lPDMS films with 77 wt % PHEA, the PDMS domains, which are responsible for the diffusive transport of the analyte, are not interconnected. As shown in previous investigations,39 the PDMS domains in these conetwork films are mostly isolated and spherical. This underlines the role of a continuous siliconphase morphology for the gas diffusion inside the films, as found for PHEA-l-PDMS (25:75) films. The interconnected PDMS phase with its high gas permeability 42 ensures a fast diffusion of gaseous molecules in the deeper layers of the polymeric films, where the reaction with the indicator reagent, which is immobilized in the contrary phase, takes place in the region of the internal interface between the two phases. Thus, the described amphiphilic conetworks consisting of PHEA-l-PDMS with a cocontinuous morphology of the phases combine the good gas permeability of silicones with the good accessibility of wellimmobilized hydrophilic indicator reagents. The resulting increase in the sensitivity of the sensor, due to the good accessibility of the deeper layers for the analyte, could enable the design of threedimensional sensors.43 Optical Determination of Acid Gases and Vapors with Reusable PTMAEA-l-PDMS/BPB Sensors. Immobilized bromophenol blue salts are widely used for producing sensors determining basic or acid vapors and gases.44,45 The indicator reagent has a yellow to blue visual transition interval in the pH range of 3.0-4.6 (in aqueous solutions) that corresponds to the deprotonation step from the monoanionic to the dianionic species (pK ) 3.8).46 For these bromophenol blue dianions, the quaternized PTMAEA-l(42) Pauly, S. In Polymer Handbook; Brandrup, J., Immergut, E. H., Grulke, E. A., Eds.; John Wiley and Sons: New York, 1989; pp 435-449. (43) Åsberg, P.; Ingana¨s, O. Biosens. Bioelectron. 2003, 19, 199-207. (44) Stangelmayer, A.; Klimant, I.; Wolfbeis, O. S. Fresenius J. Anal. Chem. 1998, 362, 73-76. (45) Zhu, C.; Hard, C.; Lin, C.; Gitsov, I. J. Polym. Sci., Part A: Polym. Chem. 2005, 43, 4017-4029. (46) Tamura, Z.; Terada, R.; Ohno, K.; Maeda, M. Anal. Sci. 1999, 15, 339342.

Analytical Chemistry, Vol. 78, No. 18, September 15, 2006

6381

Figure 6. Change in absorption at 430 nm of a PTMAEA-l-PDMS/ BPB sensor during exposure to different concentrations of acetic acid (65% RH, 23 °C, 1 L min-1, 240 min).

PDMS films promised a good immobilization matrix with a high loading capacity and good mechanical properties due to the embedded, elastic PDMS domains. To investigate this, sensors were prepared by exchanging the sulfate of PTMAEA-l-PDMS for bromophenol blue dianions as aforementioned. After drying, the resulting deep blue sensors were exposed to different amounts of acetic acid (AcOH) in humidified synthetic air (RH ) 65%). The presence of acetic acid caused the formation of a new absorption band at 430 nm and the decrease in the absorption band at 599 nm (initial absorption >2). The protonation of the immobilized indicator was reversible. Depending on the AcOH concentration, the response time was in the range of 3-30 min. The plot of the absorption change at 430 nm and at different concentrations of AcOH versus the exposure time is shown in Figure 6. Again, the substantial change in absorption is obvious. Some of the sensors were stored under room air for at least two months whereupon neither a leaching of the indicator reagent nor a change in reactivity was observed. Optical Biochemical Sensor for Determination of Peroxides in Nonpolar Organic Solvents. Normally, enzymes are applied in polar, mostly aqueous media ensuring a high enzyme activity and stability, but there is a strong interest in organic-phase enzymatic assay owing to its usefulness for the direct analysis of waterinsoluble samples.47,48 Different approaches to enhance solubility, activity, and stability of enzymes for biocatalysis in organic media have been published,49,50 like, by example, modifying enzymes with poly(ethylene glycol) 51-55 or the preparation of surfactant-enzyme (47) Klibanov, A. M. Trends Biochem. Sci. 1989, 14, 141-144. (48) Sto ¨cklein, W. F. M.; Scheller, F. W. In Frontiers in Biosensorics. I. Fundamental Aspects; Scheller, F. W., Ed.; Birkha¨user: Berlin, 1997; pp 8396. (49) Hayes, D. G. In Modern Protein Chemistry: Practical Aspects; Howard, G. C., Brown, W. E., Eds.; CRC Press LLC: Boca Raton, FL, 2002; pp 179225. (50) Castro, G. R.; Knubovets, T. Crit. Rev. Biotechnol. 2003, 23, 195-231. (51) Inada, Y.; Takahashi, K.; Yoshimoto, T.; Kodera, Y.; Matsushima, A.; Saito, Y. Trends Biotechnol. 1988, 6, 131-134. (52) Yang, L.; Murray, R. W. Anal. Chem. 1994, 66, 2710-2718. (53) Mabrouk, P. A. J. Am. Chem. Soc. 1995, 117, 2141-2146. (54) Matsushima, A.; Kodera, Y.; Hiroto, M.; Nishimura, H.; Inada, Y. J. Mol. Catal. B: Enzym. 1996, 2, 1-17. (55) Secundo, F.; Ottlina, G.; Carrea, G. In Methods in Biotechnology. Vol. 15: Enzymes in Nonaqueous Solvents: Methods and Protocols; Vulfson, E. N., Halling, P. J., Holland, H. L., Eds.; Humana Press Inc.: Totowa, NJ, 2001; pp 77-81.

6382 Analytical Chemistry, Vol. 78, No. 18, September 15, 2006

complexes.56-58 For the design of biochemical sensors operating in nonpolar organic solvents, especially for works on organic-phase enzyme electrodes, different techniques for enzyme immobilization have been carried out.9-12,48,59,60 In general, the massive entrapment technique, which allows for a high enzyme loading of the matrix, is preferable because the immobilized enzymes can be protected by numerous matrixes so that the biochemical sensors suffer less interference from external factors. For this, hydrophilic polymers such as hydrogels or sol-gels can be used as ideal matrixes,61-63 but in organic media, the sensor responses are generally rather sluggish due to the innately slow diffusive transport of the analytes in the hydrophilic matrix.64-66 Matrixes of sol-gels may be exclude nonpolar analytes altogether, resulting in poor analytical sensitivity. The key disadvantage of sol-gel-based optical (bio)chemical sensors is the relatively limited range of material compositions due to the need of optical transparency.63 The addition of hydrophobic particles to hydrogels improves the diffusive transport,67 but it comes along with a loss in optical clarity and without an interconnected structure of these additives. As shown above for the PHEA-l-PDMS/o-tolidine sensors, a cocontinuous morphology of the phases is of advantage for the diffusive transport of analytes into deeper layers of the matrix. Nanophase-separated amphiphilic conetworks improve this approach by optical clarity and transparency and by interconnected hydrophobic domains with a huge interface to the hydrophilic phase increasing the accessibility of the enzyme. Previous investigations have shown that enzymes can be easily immobilized in the APCN films by impregnation. The immobilization of enzymes in amphiphilic conetworks possesses the ability to stabilize and enhance the catalytic activity of enzymes in organic solvents.37,38,68 On the basis of these promising results, we have examined the possibility of coimmobilizing HRP and ABTS as chromophoric substrate in the prepared PHEA-l-PDMS (58:42) films and the suitability of these disposable PHEA-l-PDMS (58: 42)/HRP/ABTS biosensors for the determination of peroxides in nonpolar organic solvents. We chose HRP because this enzyme is currently used for determining hydrogen peroxide69-74 and also has been widely employed as a model enzyme in nonaqueous enzymology.57,75-78 (56) Kamiya, N.; Inoue, M.; Goto, M.; Nakamura, N.; Naruta, Y. Biotechnol. Prog. 2000, 16, 52-58. (57) Bindhu, L. V.; Emilia Abraham, T. Biochem. Eng. J. 2003, 15, 47-57. (58) Yang, L.; Dordick, J. S.; Garde, S. Biophys. J. 2004, 87, 812-821. (59) Avila, G. P.; Salvador, A.; de la Guardia, M. Analyst 1997, 122, 1543-1547. (60) Ada´nyi, N.; Va´radi, M. Eur. Food Res. Technol. 2003, 218, 99-104. (61) Dong, S.; Guo, Y. Anal. Chem. 1994, 66, 3895-3899. (62) Lee, K. Y.; Mooney, D. J. Chem. Rev. 2001, 101, 1869-1879. (63) Jin, W.; Brennan, J. D. Anal. Chim. Acta 2002, 461, 1-36. (64) Brink, L. E. S.; Tramper, J. Biotechnol. Bioeng. 1985, 27, 1258-1269. (65) Zaks, A.; Klibanov, A. M. J. Biol. Chem. 1988, 263, 8017-8021. (66) Veronese, F. M.; Mammucari, C.; Schiavon, F.; Schiavon, O.; Lora, S.; Secundo, F.; Chilin, A.; Guiotto, A. Farmaco 2001, 56, 541-547. (67) Wu, X. J.; Choi, M. M. F. Anal. Chem. 2004, 76, 4279-4285. (68) Bruns, N.; Tiller, J. C. Ger. Offen. DE 2003-10343794, 2005. (69) Wang, J.; Lin, Y.; Chen, L. Analyst 1993, 118, 277-280. (70) Chut, S. L.; Li, J.; Tan, S. N. Analyst 1997, 122, 1431-1434. (71) Navas Dı´az, A.; Ramos Peinado, M. C.; Torijas Minguez, M. C. Anal. Chim. Acta 1998, 363, 221-227. (72) Schulte-Ladbeck, R.; Kolla, P.; Karst, U. Analyst 2002, 127, 1152-1154. (73) Smith, K.; Silvernail, N. J.; Rodgers, K. R.; Elgren, T. E.; Castro, M.; Parker, R. M. J. Am. Chem. Soc. 2002, 124, 4247-4252. (74) Qian, L.; Yang, X. Talanta 2006, 68, 721-727. (75) Dordick, J. S.; Marletta, M. A.; Klibanov, A. M. Proc. Natl. Acad. Sci. U.S.A. 1986, 83, 6255-6257.

for at least 1 h. Storing the prepared biochemical sensors at 4 °C under dry conditions (argon atmosphere) had no influence on their reactivity for at least two weeks. These preliminary results show the excellent suitability of amphiphilic conetworks for easily producing optically clear and transparent biochemical sensors for very promising applications in nonpolar organic media.

Figure 7. Change in the absorption spectrum of a PHEA-l-PDMS (58:42)/HRP/ABTS biosensor under influence of a 10 mmol L-1 solution of tBuOOH in n-heptane (22.5 °C, 10 min) (a). Absorption changes at 420 nm of PHEA-l-PDMS (58:42)/HRP/ABTS sensors in a 10 mmol L-1 solution of tBuOOH in n-heptane (black rhombi), in pure n-heptane (gray rhombi); of a sensor prepared without ABTS (gray dots) and of a sensor prepared without HRP (gray triangles), both in a 10 mmol L-1 solution of tBuOOH in n-heptane (b).

The prepared PHEA-l-PDMS (58:42)/HRP/ABTS biosensors were mounted in a cuvette of the spectrophotometer without previous preswelling of the hydrophilic phase in an aqueous buffer solution. After adding 900 µL of a 10 mmol L-1 solution of tertbutyl hydroperoxide (tBuOOH) in n-heptane, the absorption spectra in the range from 375 to 1050 nm were obtained periodically. A rapid, initially linear increase in absorption at 420 nm was observed, indicating the formation of ABTS•+ radical cations79 (Figure 7a). Without tBuOOH, no reaction occurred. Sensors without immobilized HRP also showed no reaction, whereas at sensors without immobilized ABTS, the slow degradation of the enzyme by the peroxide could be observed (the absorption maximum of HRP was at 404 nm) (Figure 7b). These results indicated clearly that the observed absorption change was definitely the result of the HRP-catalyzed oxidation of ABTS by tBuOOH, indicating a well-working optical biosensor in n-heptane even in the dry state. No leaching of the reagents was observed during the immersion of the sensors in n-heptane (76) Gorman, L. A. S.; Dordick, J. S. Biotechnol. Bioeng. 1992, 39, 392-397. (77) Ryu, K.; Dordick, J. S. Biochemistry 1992, 31, 2588-2598. (78) Huang, Q.; Al-Azzam, W.; Griebenow, K.; Schweitzer-Stenner, R. Biophys. J. 2003, 84, 3285-3298. (79) Solı´s-Oba, M.; Ugalde-Saldı´var, V. M.; Gonza´lez, I.; Viniegra-Gonza´lez, G. J. Electroanal. Chem. 2005, 579, 59-66. (80) Hanko, M.; Bruns, N.; Tiller, J. C.; Heinze, J. Unpublished results.

CONCLUSION With this feasibility study we have, for the first time to our knowledge, demonstrated that the class of nanophase-separated amphiphilic conetworks can serve as versatile and advanced matrixes for optical chemical and biochemical sensors. Both the matrixes and the sensors are easy to prepare and the manufacturing can be scaled up. Amphiphilic conetworks can be prepared as films with a wide range of properties, such as ionic/nonionic, different compositions, or variable ratios of hydrophilic/hydrophobic phases, thereby enabling the versatile immobilization of indicator reagents. The APCNs can be loaded with well-known indicator reagents by simple impregnation without any need for chemical modification. Thus, in addition to the sensors reported here, preliminary results of sensors consisting of APCN matrixes loaded with β-carotene, diphenylamine, N,N′-diphenyl-1,4-phenylenediamine, iodide, or different pH indicators testify to the versatile suitability of APCN matrixes.80 Additionally, APCNs can even be loaded with enzymes to design biochemical sensors. The possibility of versatile immobilization of enzymes was shown in previous investigations, e.g., immobilization of peroxidases, lipases, cytochrome c, and R-chymotrypsin.80 Due to the cocontinuous morphology of the phases, the immobilized indicator reagents are accessible in all layers of the film, even if the polymeric film is up to 60 µm thick; hence, they could serve as three-dimensional sensors with increased capacity and sensitivity. In addition, the properties of the sensors can be influenced by different controllable parameters, such as conetwork composition, ratio of hydrophilic and hydrophobic phases, film thickness, and loading quantity. The innovative use of APCNs as matrixes can simplify the development of new (optical) sensors and the improvement of existing sensors. In addition, these materials open up new fields of application, e.g., in biosensing, as the pioneering results of a dry enzyme-based biochemical sensor for the application in nonpolar organic media demonstrate. Altogether, the nanophaseseparated amphiphilic conetworks presented here are very promising materials for a new class of versatile and advantageous sensor matrixes. Future work is focused on the development of new sensor systems for application in gaseous and liquid media and of further enzyme-based biosensors using APCN matrixes. ACKNOWLEDGMENT This work was supported by the European Union (MASTER project), by the Deutsche Forschungsgemeinschaft (SFB 428 and Emmy-Noether-Programm), and by the Fonds der Chemischen Industrie. NOTE ADDED AFTER ASAP PUBLICATION This paper was posted on July 27, 2006. Footnote 32 was later revised. The correct version was posted on August 1, 2006. Received for review April 6, 2006. Accepted June 23, 2006. AC060634+ Analytical Chemistry, Vol. 78, No. 18, September 15, 2006

6383