Chapter 2

Methods of Fabrication

2.1 Patterning Techniques

The techniques of the formation of micro- and nano-patterns on surfaces with regard to chemical and biological substances is essential in sensor technology. These techniques have been summarized in a review by Hook, Voelcker, and Thissen (1). More detailed references can be found in this particular review. An overview of patterning techniques is given in Table 2.1.

Table 2.1 Patterning techniques (1)

Method	Advantage	Disadvantage
Contact and noncontact printing	Quick and simple pattern formation	Poor resolution, pattern dimensions limited
Electron beam lithography	Highly resolved patterning, no need for a mask	Slow processing time, clean environment required, expensive
Dip pen nanolithography	Highly resolved patterning, high energy particle beams not required.	Slow processing time
Photolithography	Highly resolved patterning, large surface areas	Rigorous experimental protocols.
Templating	Low cost, quick and highly resolved	inert environment, dimensions and geometries restricted
Imprint lithography	Low cost, high resolution	Limited number of molecules and substrates
Microcontact printing	Simple method	Limited number of molecules and substrates
Microfluidics	Experimentally simple
Micromoulding	Experimentally simple	Limited pattern geometries
Magnetic force based patterning	Directly patterns cells and biomolecules	Conducted in an inert environment

2.2 Coating Techniques

A comparative study using different coating techniques with adhesives has been published (2). The most popular techniques used are spin coating, spray coating, and drop coating.

2.2.1 Dip Coating

The dip coating process consists of the following basic steps (3):

2.2.2 Spin Coating

Spin coating is used in order to deposit uniform thin films onto flat surfaces (4). Spin coating is intensively used in photolithography, to deposit layers of photoresist in the range of 1 μm. Typically the process of spin coating lasts less then 1 min. Commercial devices are available. The major reasons for the popularity of this technique is the simplicity of operation and the uniformity and thinness of coated layers.

The coating is placed in dilute solution on the surface. The surface is rotated at high speed. In this way, the fluid is displaced due to the centrifugal force. The rotation is continued until the fluid spins off the edges of the surface and until the desired thickness of the film is established. During this process, the solvent may evaporate.

The higher the angular speed, the thinner the film will become. In addition, the thickness of the film will depend on the concentration of the solution and the nature of the solvent. Spin coating is widely used in microfabrication. Thin films with a thickness below 10 nm can be fabricated. The maximum allowable rotation speed of the support in spin coating process decreases from 3600 rpm for 200 mm wafers to 1600 rpm for 300 mm wafers. This is in order to avoid problems due to striation and laminar-turbulent transition over the wafers during the process.

Physical models have been established to investigate the effects of spin coating process parameters including the rotation speed, initial viscosity and initially dispensed volume on the final average film thickness (5–7) These models assume a uniform film thickness over the entire surface. Thus information on the average film thickness is given.

Another one-dimensional model has been developed to simulate the coating process from a dispensing to a drying phase in more detail by adopting moving meshes (8). In that study, the effects of various parameters of the spin coating process on the coated film profile are investigated. The model is used to optimize some parameters in a 450 mm wafer coating process to minimize the photoresist consumption.

Coating defects can occur, if convective flow has not completely ceased when this skin forms. Skin formation can be eliminated or delayed by partially saturating the overlying gas with a solvent or by using mixed solvents (6).

The deposited film may be removed from the surface, e.g., in order to fabricate membranes. In addition, composite membranes can be prepared by spin coating onto a micro-porous polymeric support. As a variant, precoating of the support is possible (9).

2.2.3 Spray Coating

In the spray coating technique, the coating is applied by spraying. The most common devices use a compressed gas, such as air in order to atomize and focus the particles to be coated.

2.2.4 Drop Coating

This coating technique is done by simply applying a drop of the solution on top of the sensor.

2.2.5 Electrospray

A shadow-masked electrospray method has been used in the fabrication of a carbon black/poly(vinyl pyrrolidone) composite sensor (10). This method of fabrication, is advantageous for the preparation of an active layer with optimum average thickness and porous microstructure compared to the common drop casting method. The thickness is proportional to the number of scans of electrospraying. The field assisted generation of small droplets results in the formation of rough and porous microstructures.

The microstructures at the surface were observed by scanning electron microscope (SEM) experiments at the center region of the films that have been deposited by both electrospray and drop casting methods. The SEM images clearly reveal that the electrospray method develop more highly porous substructures in comparison to those deposited by the drop casting method. Actually, a rough, porous morphology is advantageous to achieve a highly sensitive and fast responding sensor (10).

2.2.6 Rapid Expansion of Supercritical Solutions

Coatings with a high surface-to-volume ratio can be deposited using the rapid expansion of supercritical solutions (RESS) technique. This spray-on technique can be used with both glassy and viscoelastic polymers (11).

Actually, this particular phenomenon of RESS was already observed and documented in 1879. The precipitation of solids from supercritical fluids was described as (12):

When the solid is precipitated by suddenly reducing the pressure, it is crystalline, and may be brought down as a snow in the gas, or on the glass as a frost.

RESS advantageously uses the change in solubility that occurs during the sudden decompression of a supercritical solution that contains a nonvolatile solute. The decompression is achieved through an orifice or capillary nozzle (13). The rapid expansion results in a high supersaturation. This is followed by a homogeneous nucleation and growth of solute particles. In general, the particles have a narrow size distribution with a very small size (14).

The RESS technique is useful to comminute shock-sensitive solids, to produce intimate mixtures of amorphous materials, to form polymeric microspheres and deposit thin films. Critical properties of common RESS solvents are provided. The solvents include carbon dioxide, propane, n-pentane, propylene, ethanol, and water. In all cases the RESS process requires dissolving of at least one solid in the supercritical fluid (15).

The process of aerosol formation has been mathematically modelled (16). The particle size is very sensitive to the temperature at which the solute is dissolved in the supercritical fluid and to the temperature to which the saturated mixture is preheated prior to the expansion. A variety of morphologies of the coatings can be achieved, including uniform films and particles of different shapes and sizes as well as fibers.

RESS has been used for a wide range of polymer supercritical fluid systems. Also processing multicomponent systems in this way is possible. A representative apparatus for method has been presented including instrumental details, as well as SEM of coatings that can be produced (17–19). A RESS apparatus is schematically shown in Figure 2.1.

Specific surfactants can be used for the solubilization of the materials in compressed CO₂. The selection of a surfactant with a CO₂-phobic segment, which has affinity for the functional material to be precipitated is crucial to such a surfactant supported process (20).

The precipitated nanomaterial can be collected by a number of methods. For example, the precipitated nanomaterials may be injected into a suitable substrate sheet for immobilization, or the nanomaterials may be collected in a suitable liquid. Due to the surfactant coating of the nanomaterials during the depressurization process, the nanomaterials will be stable and not undergo significant agglomeration. Thereby, discrete nanoparticles can be obtained depending on the processing conditions (20).

2.3 Electrospinning

Electrospinning process is an efficient method for the fabrication of a nanofibrous sensing film (21–23).

During this process, a polymer solution is placed into a syringe and the solution is extruded from the needle tip at a constant rate by a syringe pump. Between the needle and the target an electrical potential is applied to induce an electrostatic force on the surface of the droplet of the polymer solution. This causes a stretching of the droplet. When the voltage is sufficiently high, a thin jet of the polymer solution is formed at the surface of the droplet and causes the formation of nanofibers. These fibers are then deposited on the target (24).

The fibers can be assembled as a three-dimensional structured fibrous membrane with a controllable pore structure and high specific surface area. The device is shown in Figure 2.2.

In conventional electrospinning techniques, the solvent for the polymer must be sufficiently volatile to be vaporized during electrospinning at atmospheric pressure. Using a pressurized formulation and a pressurized collection vessel, polymeric formulations can be electrospun — despite viscosities which are too high for successful electrospinning under ambient conditions (25). Moreover, nonvolatile organic solvents can be used.

For example, in a pressurized environment refractory polymers can be electrospun. These polymers are difficult to dissolve in solution or they do not melt even at high temperatures. Electrospun fibers produced at high pressures using CO₂ exhibit an internal cell-like morphology with a coherent skin that is ruptured along the fiber axis in contrast to the fibers electrospun in the absence of an supercritical fluid and at ambient conditions (25).

2.4 Molecular Imprinted Polymers

The use of molecularly imprinted polymers (MIPs) is a rather general approach to chemical recognition techniques. It allows the rational design and assembly of materials in a stable and reusable form (26).

The origins of MIPs trace back to Polyakov (27). It was exemplified by the polymerization of sodium silicate with ammonium carbonate. When the polymerization was carried out in the presence of benzene, the silica particles exhibited a higher uptake of benzene.

Later, Pauling presented a theory of the structure and the process of formation of antibodies. A material can be assembled as a protein complement, i.e., an antibody, by using the foreign intruder as a template (26,28). The history of molecular imprinting has been profoundly reviewed (29). The review by Alexander contains the remarkable number of 1495 references.

An MIP can be described as a plastic cast or mold of the molecule of interest, where recognition is based on shape, much like a lock and key. MIPs are made by adding the molecule of interest to a solution of binding molecules that can be chemically incorporated into a polymer. Such binders usually exhibit an affinity for the target as they form complexes.

The basics of molecular imprinting is shown in Figure 2.3. In the first step, a molecule with functional groups X suitable to form complexes with the pendant moieties Y of a vinyl monomer is allowed to form the complexes. Then, another vinyl monomer is added with the groups R₁. This monomer must have the capability to copolymerize with the complex forming vinyl monomer. Then the copolymerization reaction is started and the complexed molecule is embedded in the polymeric matrix. Finally, the molecule with functional groups X us removed from the matrix. In this way, gaps are formed with just the size and shape to absorb selectively related molecules.

2.4.1 Influence of Cross Linking Agents

In MIP the crosslinking agent is essential for the prearranged complex, which is formed between the functional monomers and the template (30). The degree of crosslinking influences the properties of the final polymer. Sensors have been formed from methacrylic acid (MA) and ethylene glycol dimethacrylate as crosslinking agent with niacinamide as template molecule.

The selectivity of these sensors increases with the amount of crosslinking agent. On the other hand at high amounts of crosslinking agent, the selectivity is not enhanced (30).

2.5 Sensor Arrays

A series of polymers, i.e., poly(aniline) (PANI), poly(pyrrole) (PPY), and poly(3-methylthiophene) has been synthesized and doped with various mineral acids, and their anions (31). The doped polymer films were deposited on a miniature system of gold raster electrodes. These were formed on a glass ceramic substrate using photolithography.

The sensor array was exposed to certain chemical compounds and the response in electrical conductivity was measured. The data obtained in this way was analyzed by the discriminant technique.

The analysis has been based on the calculation of the distance between vectors of all pairs of the analytes under investigation in a multidimensional Euclidian space. The number of dimensions represent the number of sensors in the array (32).

For each pair of analytes with numbers i and j the analyte separability factor S_i,j is defined as

(2.1)

and are the vectors of the mean responses of the analytes i and j, ζ is the sum of the standard deviation, and is the Euclidian norm, which is defined as the square root of the sum of the absolute squares of its elements.

The analysis of the discrimination ability of the sensor arrays revealed that a set of four to six sensors is sufficient for the reliable detection of up to nine analytes. The analytes used are shown in Table 2.2.

Table 2.2 Analytes under investigation (31)

Analyte	Vapor Pressure /[mm Hg] at 20°C
Amyl alcohol	2.2
Xylene	6
Isobutyl alcohol	9.2
Toluene	22.3
Isopropyl alcohol	33.1
Ethyl alcohol	42.8
Benzene	77.1
Chloroform	161.9
Acetone	182

A comparative analysis of the responses of various conducting organic polymer films to the vapors of organic solvents has been conducted (33). The effect of organic solvents on PANI films always led to an increase in electrical conductivity. In contrast, PPY and poly(3-methylthiophene) films show both an increase or a decrease of the electrical conductivity.

The form of the response curves suggest that the effect of organic solvent vapors can be traced back to the following processes (33):

• Absorption of the analyte vapor by the polymer and ion dipole interaction between polycation-radical fragments of polymer and anionic dopants and the molecules of organic solvent, and
• Change of the electron state of the polymer due to specific interaction, e.g., formation of hydrogen bonds, charge transfer complexes, redox interaction between the molecules of organic solvent and polymer chain.

The first process is governed by the porosity of the polymer film and the polarity of the organic solvent. The interaction between the charged fragments of the polymer and the dopants becomes weaker, which results in an increase of the mobility of the charge carriers in the polymer and effects an increase of the electrical conductivity.

The specific interaction results from the chemical nature of analyte and electron state of the polymer. A specific interaction may either increase or decrease the electrical conductivity of the polymer (33).

2.5.1 Conducting Polymer Compositions

A conductive chemical sensor is constructed from at least one electrode pair and a plurality of photopolymerized electrically conducting polymer compositions. These compositions are in contact with separate electrode pairs.

Each polymer composition may include an organic polymer capable of specifically interacting with the analytes. Thus, the sensor comprises a plurality of different polymer compositions, each with a dedicated electrode pair, to generate a collection of signals that provide a fingerprint unique to a particular analyte (34).

The individual analytes are recognized with an artificial neural network, i.e., a pattern recognition technique. A commercially available artificial neural network program (MATLAB®) was used to train the network. In addition, the technique of chemometrics has been used for pattern recognition (35,36).

Most conjugated polymers (CPs) prepared electrochemically using pyrrole, methylpyrrole, thiophene, phenylene, and 5-carboxyindole are of limited use, because of the complexity of the preparation processes.

As an alternative, using the technique of photopolymerization has been successfully developed. The foremost advantage of the photopolymerization process is that the CPs are formed on any substrate that may be conductive or nonconductive. The photopolymerization process works in addition onto other monomers, such as aniline, methylpyrrole, and thiophene. Some blends are shown in Table 2.3.

Table 2.3 Blends of conducting polymer and organic polymer dissolved in THF (34)

Experiments with a variety of polymers suggest that the photopolymerization process is a flexible procedure that can produce a series of electrically CP compositions with different functionalities, depending on the formulation. Further, polymer- based sensors can have a wider combination of sensor functionality than metal oxide-based sensors.

The photopolymerization process allows the fabrication of three-dimensional elements. This has the advantage of yielding a high surface area for interaction between the sensor element and the analyte. In order to prepare a three-dimensional sensing element, the blended polymerized composite is fixed onto a cylindrical substrate. For building the sensor, the sensor polymer material is deposited in a controlled fashion onto the interdigitated electrode areas of the substrate chip.

2.5.2 Surface Imprinting

Surface imprinting is a sophisticated technique used to create artificial receptors on polymer thin films for various substances, such as proteins, plant viruses, and yeasts (37–40). A variety of slightly differing techniques have been developed that are subsequently discussed.

Uniform surface imprinted nanoparticles can be prepared by introducing vinyl groups onto the surface of silica beads. This method includes a modification of the surface by vinyl moieties, followed by the copolymerization of functional monomers by a reversible addition-fragmentation chain transfer (RAFT) precipitation polymerization. The resultant RAFT surface imprinted nano sized polymers are spherical- shaped monodisperse particles. The method of preparing surface imprinted polymers is claimed to be superior in comparison to conventional MIPs. The method was tested with atrazine in spiked corn and lettuce samples. Recoveries of up to 93% could be found using a one-step extraction (41).

For example, 3-methacryloxypropyltrimethoxysilane can be used for the surface modification of silica. Onto that surface MA can be grafted. Ethylene glycol diglycidyl ether or 2,2′-(ethylenebis(oxymethylene))bisoxirane is used for crosslinking. Cytisine has been used as an analyte (42).

Quercetin imprinted core shell silica microspheres can be prepared using a surface imprinting technique. The imprinted materials are prepared by a sol-gel technique using 3-aminopropyltriethoxysilane as the functional monomer and tetraethyl orthosilicate as the crosslinking agent (43).

Thiocyanate functionalized silica gel particles have been prepared using Cd²⁺ as a template. 3-Thiocyanatopropyltriethoxysilane is used as the functional monomer, and epichlorohydrin as the crosslinking agent. The materials are suitable for the selective removal of Cd²⁺ from an aqueous solution. The adsorption and desorption of the ions is fast. A substantial binding capacity at a pH of 4.2–8.6 is observed. Repeated use is possible (44). Adsorption capacities for Cd²⁺ using various imprinted sorbents are collected in Table 2.4. Methods of the analysis and removal of Cd²⁺ are of importance because of its high toxicity and thus because of environmental concerns (45).

Table 2.4 Adsorption capacities for cadmium ions (44)

Reagent	Imprinting Method	Capacity/[mg g−1]
3-Thiocyanatopropyltriethoxysilane	Surface	44.7
3-Mercaptopropyltrimethoxysilane	Surface	31.9
Aminoethylaminopropyltrimethoxysilane	Surface	22.4
Poly(ethyleneimine)	Surface	18.7
Diazoaminobenzene-vinylpyridine	Bulk	10.4
Bis(2-aminoethyl)but-2-enediamide	Bulk	32.6
Salicylic aldehyde and 4-vinylpyridine	Bulk	0.48
N-Methacrylolyl-cysteine methyl ester	Bulk	3.0
Vinylimidazole	Bulk	4.6

Chiral caves have been constructed for the separation of enantiomers (46). The method uses the grafting of dimethylaminoethyl methacrylate onto the surfaces of small silica gel particles. Instead of silica particles, crosslinked poly(vinyl alcohol) can be used as base material for surface grafting (47). On such microspheres, acrylonitrile (AN) was graft polymerized. A cerium salt is used as a redox initiator. The AN moieties are then modified by hydroxylamine in an amidoximation reaction so that a poly(amidoxime) will be formed. This material is then imprinted with uric acid. When the adsorption of uric acid reaches saturation, glutaraldehyde is added as the crosslinking agent. The process of imprinting is shown schematically in Figure 2.4.

The selectivity coefficient of the non-imprinted microspheres for uric acid relative to guanine is only 1.3, thus essentially no selectivity is observed. After imprinting, the selectivity coefficient is enhanced to 14, which indicates an excellent recognition selectivity and binding affinity towards uric acid (47).

Surface imprinted polymer composites have been prepared on tetratitanate whiskers as the carrier material. For example, 4-vinylpyridine was used as the functional monomer and dibenzothiophene as template (48).

Erythrocytes have been used as templates for MIPs. Despite the flexibility of these cells, the imprints show a high selectivity for different blood groups. This was demonstrated by a sensor based on a quartz crystal microbalance. It has been shown that the imprinting process can create selective synthetic antibodies for erythrocyte antigens (49).

2.5.3 Molecular Imprinted Sensor Arrays

The sensor array format has proven to be an effective method of transforming sensors of modest selectivity into highly selective and discriminating sensors. The primary challenge in developing such sensor arrays is the arrangement of a sufficient number of elements that possess sufficient different binding affinities for the analytes of interest (50).

The use of MIPs for recognition of such sensor arrays is particularly advantageous as they can be rapidly and inexpensively prepared with tailored selectiveness. In addition, the array format helps to compensate a low selectivity and cross reactivity of a single MIP sensor (50).

The first example of a sensor array assembled from molecularly imprinted materials was by Mirsky and coworkers in 2003 (51). MIP sensor arrays have shown a broad utility, as even analytes that were not used as templates in the imprinting process can be effectively discriminated (50).

2.6 Ink Jet Fabrication

Imaging sensors can be fabricated by reproducibly printing an array of photopolymerizable sensing elements. For example, a pH sensitive indicator, has been deposited on the surface of an optical fiber image guide. The reproducibility of the microjet printing process is excellent for micrometer sized polymer sensors. pH sensors were evaluated in terms of their pH sensing ability, response time, and hysteresis. A fluorescence imaging system was used for the detection. It seems that the microjet technique has distinct advantages over other fabrication methods (52). An imprinted device is shown in Figure 2.5.

2.6.1 Inkjet Printed Chemical Sensor Array

It was shown that a chemical sensor array can be fabricated by ink jet printing (53). Poly(thiophene)s with a variety of side chains were used. The polymers are summarized in Table 2.5. A custom ink jet system is suitable to selectively deposit the polymers onto an array of transduction electrodes.

Table 2.5 Polymers used for ink jet printing (53)

Homopolymers
Poly(3-hexylthiophene)
Poly(3-dodecylthiophene)
Poly(3-methoxyethoxyethoxymethylthiophene)
Bromoester terminated poly(3-hexylthiophene)
Benzyl terminated Poly(3-hexylthiophene)
Copolymers	Molar Ratio
Poly(3-hexylthiophene-b-styrene)	65:35
Poly(3-hexylthiophene-b-methyl acrylate)	80:20
Poly(hexylthiophene-b-butyl acrylate)	82:18
Poly(3-dodecylthiophene-ran-3-methylthiophene)	50:50

The sensor arrays are prepared on a conductive silicon substrate with an insulating thin silica surface. Two masks are used for fabrication. A 5 nm layer of titanium is covered by a 50 nm layer of gold. Titanium is an adhesion improver for gold. The gold layer is the contact layer for the polymers. The work function of gold is close to that of the polymers, therefore an ohmic resistive contact is formed. The second mask is used to add a final coating of gold to make the sensor more robust. Details of the sensor geometry and the method of fabrication can be found in the original literature (53).

Optical micrographs of sensor arrays are shown in Figure 2.6. (a) Completed, wire bonded, test chip showing. 24 electrode patterns with ink-jetted polymers on 20 of these. The sensors in the first column are used for reference. The rest of the electrodes have one type polymer jetted on each column. From left to right, the polymers are poly(3-hexylthiophene)-b-polystyrene), poly(hexylthiophene)-b-poly(butyl acrylate), Poly(3-methoxyethoxyethoxymethylthiophene), poly(3-dodecylthiophene-ran-3-methylthiophene), poly(3-hexylthiophene)-b-poly(methyl acrylate). (b) Enlarged view of the gold spiral electrodes with no polymer. (c) Spiral electrodes with jetted poly(3-hexylthiophene) formed from 10 drops of 5 mg/mL polymer concentration dissolved in trichlorobenzene (53).

A series of volatile organic vapors have been tested as analytes. These were acetone, methylene chloride, toluene, and cyclohexane. Sensing is based on the fact that the conductivity response to the vapors is dependent on the chemical structure of the polymers. The signals were analyzed by principal component analysis. Then the results have been correlated to the chemical structures of the different polymers (53).

References

1. A. Hook, N. Voelcker, and H. Thissen, Patterned and switchable surfaces for biomolecular manipulation, Acta Biomaterialia, 5(7):2350–2370, September 2009.

2. D. Pestov, O. Guney-Altay, N. Levit, and G. Tepper, Improving the stability of surface acoustic wave (saw) chemical sensor coatings using photopolymerization, Sensors and Actuators B: Chemical, 126(2):557–561, October 2007.

3. Dip coating, [electronic:] http://en.wikipedia.org/wiki/Dip-coating, 2010. Wikipedia.

4. Spin coating, [electronic:] http://en.wikipedia.org/wiki/Spin_coating, 2010. Wikipedia.

5. D.E. Bornside, C.W. Macosko, and L.E. Scriven, Modeling of spin coating, Journal of imaging technology, 13(4):122–130, August 1987.

6. D.E. Bornside, C.W. Macosko, and L.E. Scriven, Spin coating: One-dimensional model, Journal of Applied Physics, 66(11):5185–5193, December 1989.

7. T. Ohara, Y. Matsumoto, and H. Ohashi, The film formation dynamics in spin coating, Physics of Fluids A, 1(12):1949–1959, December 1989.

8. J.-Y. Jung, Y.T. Kang, and J. Koo, Development of a new simulation model of spin coating process and its application to optimize the 450 mm wafer coating process, International Journal of Heat and Mass Transfer, 53(9–10):1712–1717, April 2010.

9. J. Le Roux and D. Paul, Preparation of composite membranes by a spin coating process, Journal of Membrane Science, 74(3):233–252, November 1992.

10. Y.S. Kim, Fabrication of carbon black-polymer composite sensors using a position-selective and thickness-controlled electrospray method, Sensors and Actuators B: Chemical, 147(1):137–144, May 2010.

11. N. Levit, D. Pestov, and G. Tepper, High surface area polymer coatings for SAW-based chemical sensor applications, Sensors and Actuators B: Chemical, 82(2–3):241–249, February 2002.

12. J.B. Hannay and J. Hogarth, On the solubility of solids in gases, Proc. R. Soc., 30:178–188, January 1879.

13. D.W. Matson, K.A. Norton, and R. Smith, Making powders and films from supercritical fluid solutions, Chemtech, 19(8):480–486, August 1989.

14. J.W. Tom and P.G. Debenedetti, Particle formation with supercritical fluids–a review, Journal of Aerosol Science, 22(5):555–584, 1991.

15. R.E. Sievers and U. Karst, Methods for fine particle formation, US Patent 5 639 441, assigned to Board of Regents of University of Colorado (Boulder, CO), June 17, 1997.

16. X. Kwauk and P.G. Debenedetti, Mathematical modeling of aerosol formation by rapid expansion of supercritical solutions in a converging nozzle, Journal of Aerosol Science, 24(4):445–469, June 1993.

17. R.D. Smith, Supercritical fluid molecular spray thin films and fine powders, US Patent 4734451, assigned to Battelle Memorial Institute (Richland, WA), March 29, 1988.

18. J.L. Fulton and G. Deverman, Electrostatic deposition of particles generated from rapid expansion of supercritical fluid solutions, WO Patent 03 101 624, assigned to Battelle Memorial Institute, December 11, 2003.

19. H.R. Satvati and M.N. Lotfollahi, Effects of extraction temperature, extraction pressure and nozzle diameter on micronization of cholesterol by RESS process, Powder Technology, 210(2):109–114, June 2011.

20. G.C. Irvin, R. Jagannathan, S. Jagannathan, S. Sunderrajan, D.D. Tuschel, W.C. Lenhart, and D.J. Nelson, Surfactant assisted nanomaterial generation process, US Patent 7 276 184, assigned to Eastman Kodak Company (Rochester, NY), October 2, 2007.

21. B. Ding, J. Kim, Y. Miyazaki, and S. Shiratori, Electrospun nanofibrous membranes coated quartz crystal microbalance as gas sensor for NH₃ detection, Sensors and Actuators, B: Chemical, 101(3):373–380, 2004.

22. B. Ding, M. Wang, J. Yu, and G. Sun, Gas sensors based on electrospun nanofibers, Sensors, 9(3):1609–1624, 2009.

23. O. Landau, A. Rothschild, and E. Zussman, Processing-microstructure-properties correlation of ultrasensitive gas sensors produced by electrospinning, Chemistry of Materials, 21(1):9–11, 2009.

24. S. Liu, H. Sun, R. Nagarajan, J. Kumar, Z. Gu, J. Cho, and P. Kurup, Dynamic chemical vapor sensing with nanofibrous film based surface acoustic wave sensors, Sensors and Actuators A: Physical, 167(1):8–13, May 2011.

25. M. McHugh, Z. Shen, D. Gee, G. Karles, J. Nepomuceno, and G. Huvard, Method of producing fibers by electrospinning at high pressures, US Patent 7 935 298, assigned to Virginia Commonwealth University (Richmond, VA), May 3, 2011.

26. G.M. Murray and B.M. Arnold, Molecularly imprinted polymeric sensor for the detection of explosives, US Patent 6 872 786, assigned to The Johns Hopkins University (Baltimore, MD), March 29, 2005.

27. M.V. Polyakov and Z.F. Khim, Adsorption properties of silica gel and its structure, Zh. Fiz. Khim., Ser. B, 2(6):799–805, 1931.

28. L. Pauling, A theory of the structure and process of formation of antibodies, Journal of the American Chemical Society, 62(10):2643–2657, 1940.

29. C. Alexander, H.S. Andersson, L.I. Andersson, R.J. Ansell, N. Kirsch, I.A. Nicholls, J. O’Mahony, and M.J. Whitcombe, Molecular imprinting science and technology: A survey of the literature for the years up to and including 2003, Journal of Molecular Recognition, 19(2):106–180, March/April 2006.

30. Z. Zhang, H. Li, H. Liao, L. Nie, and S. Yao, Influence of cross-linkers’ amount on the performance of the piezoelectric sensor modified with molecularly imprinted polymers, Sensors and Actuators B: Chemical, 105 (2):176–182, March 2005.

31. A.L. Kukla, A.S. Pavluchenko, Y.M. Shirshov, N.V. Konoshchuk, and O.Y. Posudievsky, Application of sensor arrays based on thin films of conducting polymers for chemical recognition of volatile organic solvents, Sensors and Actuators B: Chemical, 135(2):541–551, January 2009.

32. A.S. Pavluchenko, Z.I. Kazantseva, I.A. Koshets, and Y.M. Shirshov, Application of linear classifiers for analysis of the sensor arrays descriptiveness for detection of the volatile compounds molecules, Sensor Electronics and Microsystem Technologies, 2:69–80, 2005.

33. O. Posudievsky, N. Konoschuk, A. Kukla, A. Pavluchenko, Y. Shirshov, and V. Pokhodenko, Comparative analysis of sensor responses of thin conducting polymer films to organic solvent vapors, Sensors and Actuators B: Chemical, 151(2):351–359, January 2011.

34. A. Gonzalez-Martin, J. Kim, H. Jabs, A. Solar, L.A. Rutherford, Jr., and D. Westerheim, Chemical sensors utilizing conducting polymer compositions, US Patent 6 994 777, assigned to Lynntech, Inc. (College Station, TX), February 7, 2006.

35. K.R. Beebe, R.J. Pell, and M.B. Seasholtz, Chemometrics: A Practical Guide, Wiley New York, New York, 1998.

36. R.G. Brereton, Chemometrics for Pattern Recognition, John Wiley & Sons, New York, 2009.

37. H. Shi, W.B. Tsai, M. Garrison, S. Ferrari, B.D. Ratner, et al., Template-imprinted nanostructered surfaces for protein recognition, Nature, 398 (6728):593–596, 1999.

38. F.L. Dickert, O. Hayden, R. Bindeus, K.-J. Mann, D. Blaas, and E. Waigmann, Bioimprinted QCM sensors for virus detection screening of plant sap, Analytical and Bioanalytical Chemistry, 378(8):1929–1934, April 2004.

39. O. Hayden and F.L. Dickert, Selective microorganism detection with cell surface imprinted polymers, Advanced Materials, 13(19):1480–1483, 2001.

40. H.-F. Wang, Y. He, T.-R. Ji, and X.-P. Yan, Surface molecular imprinting on Mn-doped ZnS quantum dots for room-temperature phosphorescence optosensing of pentachlorophenol in water, Analytical Chemistry, 81(4):1615–1621, February 2009.

41. S. Xu, J. Li, and L. Chen, Molecularly imprinted core-shell nanoparticles for determination of trace atrazine by reversible addition fragmentation chain transfer surface imprinting, Journal of Materials Chemistry, 21(12):4346, 2011.

42. B. Gao, Q. Niu, and R. Du, Preparation and recognition performance of cytisine alkaloid-imprinted material prepared using novel surface molecular imprinting technique, Journal of Separation Science, pp. NANA, March 2010.

43. P. Yang, W.D. Hou, H.D. Qiu, X. Liu, and S.X. Jiang, Preparation of quercetin imprinted core–shell organosilicate microspheres using surface imprinting technique, Chinese Chemical Letters, 23(5):615–618, May 2012.

44. Z.-C. Li, H.-T. Fan, Y. Zhang, M.-X. Chen, Z.-Y. Yu, X.-Q. Cao, and T. Sun, Cd(II)-imprinted polymer sorbents prepared by combination of surface imprinting technique with hydrothermal assisted sol–gel process for selective removal of cadmium(II) from aqueous solution, Chemical Engineering Journal, 171(2):703–710, July 2011.

45. S.E. Manahan, Environmental Chemistry, Lewis, Boca Raton, 1994.

46. B. Gao, Y. Chen, and J. Men, Constructing chiral caves and efficiently separating enantiomers of glutamic acid with novel surface-imprinting technique, Journal of Chromatography A, 1218(32):5441–5448, August 2011.

47. B. Gao, S. Liu, and Y. Li, Preparation and recognition performance of uric acid-imprinted material prepared with novel surface imprinting technique, Journal of Chromatography A, 1217(15):2226–2236, April 2010.

48. W.-Z. Xu, W. Zhou, L.-H. Bian, W.-H. Huang, and X.-Y Wu, Preparation of molecularly imprinted polymer by surface imprinting technique and its performance for adsorption of dibenzothiophene, Journal of Separation Science, 34(14):1746–1753, July 2011.

49. O. Hayden, K.-J. Mann, S. Krassnig, and F.L. Dickert, Biomimetic ABO blood-group typing, Angewandte Chemie International Edition, 45(16): 2626–2629, April 2006.

50. K.D. Shimizu and C.J. Stephenson, Molecularly imprinted polymer sensor arrays, Current Opinion in Chemical Biology, 14(6):743–750, December 2010.

51. T. Hirsch, H. Kettenberger, O.S. Wolfbeis, and V.M. Mirsky, A simple strategy for preparation of sensor arrays: molecularly structured monolayers as recognition elementselectronic supplementary information (ESI) available: experimental details and electrochemical measuring conditions. see http://www.rsc.org/suppdata/cc/b2/b210554c/, Chemical Communications, (3):432–433, January 2003.

52. J.C. Carter, R.M. Alvis, S.B. Brown, K.C. Langry, T.S. Wilson, M.T. McBride, M. Myrick, W.R. Cox, M.E. Grove, and B.W. Colston, Fabricating optical fiber imaging sensors using inkjet printing technology: A ph sensor proof-of-concept, Biosensors and Bioelectronics, 21 (7):1359–1364, January 2006.

53. B. Li, S. Santhanam, L. Schultz, M. Jeffries-EL, M.C. Iovu, G. Sauvé, J. Cooper, R. Zhang, J.C. Revelli, A.G. Kusne, and et al., Inkjet printed chemical sensor array based on polythiophene conductive polymers, Sensors and Actuators B: Chemical, 123(2):651–660, May 2007.