Received August 1, 2007

Revised October 16, 2007

Introduction

Chitosan is a linear chain polysaccharide made of glucosamine and N-acetylglucosamine, joined together by β-1,4-glycosidic bonds. The more highly acetylated and insoluble antecedent of chitosan, chitin, is a commonly occurring natural material that is a primary component of the shells and exoskeletons of crustaceans and many insects as well as the cell walls of some fungal species. Although no firmly established nomenclature boundary exists between chitin and chitosan, it has been proposed that the material may be considered chitosan when soluble in dilute acidic conditions, which occurs when deacetylation is at least 60%.¹ Both chitin and chitosan possess a variety of attractive properties, including biocompatibility, biodegradability, and low toxicity. Additionally, these materials are noted for their ability to bind and sequester metals. Sorption of metal ions can occur by chelation mechanisms in neutral solutions or by electrostatic and ion exchange mechanisms in acidic solutions.²

Because of its predominantly deacetylated structure, chitosan is rich in primary amine groups. When chitosan is introduced to acids, protonation of amine groups leads to periodic positive charges along the polymer backbone and results in water-soluble chitosan salts. The chemical functionality conferred by the abundance of amine groups also makes chitosan particularly suitable for chemical modification.³ A number of studies have been published on chitosans modified through their primary amine groups to incorporate thiol functionality.^4–8 Sulfur-rich chitosan derivatives have been found to be good sorbents for a variety of metal ions, including Hg(II), Cd(II), Co(II), Ni(II), Pb(II), and Cr(III).^{4, 9–11}

Mercury, in particular, is often cited for its interaction with sulfur groups. Key to the sulfur−metal interaction is the tendency for thiol groups to selectively form stable complexes with large, soft ions such as Hg(II).^{12, 13} An overarching goal of this thiolated chitosan study was to create a thin film surface that is selectively responsive to mercury.

We have synthesized a number of previously described thiolated chitosans, including covalently linked conjugates of chitosan with thioglycolic acid, 2-iminothiolane, and cysteine. These materials were processed into cross-linked hydrogel thin films with interference-derived coloration, analogous to the specular effects observed in many natural structures and organisms.^{14, 15} The wavelengths of light reflected from each sample change with film thickness and refractive index, giving rise to a platform for detecting analyte-induced changes in film geometry and/or optical properties. The films were exposed to 50 ppm solutions of a variety of salts to test their responses to metal ions and counterions of interest.

Experimental Section

Materials. Low-molecular-weight chitosan was purchased from Sigma-Aldrich (St. Louis, MO). The carbodiimide, N-(3-dimethylaminopropyl)-N′-ethylcarbodiimide hydrochloride (EDC), used for coupling reactions between amines and carboxylic acids, was supplied by Fluka BioChemika (Switzerland). Thiol-containing small molecules and all salts used for metal ion studies were purchased from Sigma-Aldrich. Chitosan conjugates were made with the thiol molecules L-cysteine, thioglycolic acid (TGA), 2-iminothiolane and (±)-α-lipoic acid. Metal ion studies were conducted using the compounds: chromium(VI) oxide, mercury(II) nitrate monohydrate, lead(II) nitrate, cadmium(II) chloride, arsenic(III) oxide, sodium nitrate, cobalt(II) acetate, disodium hydrogen phosphate, sodium chloride, sodium sulfate, lead(II) acetate, nickel(II) sulfate heptahydrate, sodium acetate, chromium(III) nitrate nonahydrate, and chromium(III) chloride. All chemicals were used as received. Silicon substrates were obtained as single-sided polished, mechanical grade (100) plane wafers (University Wafer, South Boston, MA).

Modification of Chitosan. Modification procedures for thiolation of chitosan were adapted from work published by Bernkop-Schnürch and colleagues.^{5, 8, 16} Briefly, thiol-containing acids were reacted with chitosan in dilute solution to covalently bond the acids with amine groups on the chitosan chain, resulting in amide formation. Compounds A, B, and D were reacted in the presence of a carbodiimide to facilitate the amide-forming condensation reaction. Compound C proceeded via a ring-opening mechanism with no carbodiimide present (Figure 1).

Figure 1. Chitosan is composed of subunits of glucosamine (x) and N-acetylated glucosamine (y), where x > y. The structure is readily modified by grafting small molecules to the polymer’s amine functionality, most often in a carbodiimide-mediated process. Derivatives used in this study include conjugates of chitosan and L-cysteine (compound A), thioglycolic acid (compounds B and B′), and 2-iminothiolane (compound C). Synthesis of a fourth derivative using alcohol-solvated lipoic acid resulted in partial acetylation at the amine and primary alcohol sites (compound D). Click to Enlarge

For each synthesis reaction, the pH was adjusted to 4.5–5.0. The solutions were stirred at room temperature overnight. Solutions were then transferred into 3500 molecular weight cutoff dialysis tubing (Pierce Biotechnology, Rockford, IL) for removal of unreacted reagents and salts. Solutions were dialyzed in five cycles over the course of a week in room temperature ultrapure water. Following dialysis, the products were recovered by lyophilization and stored at −20 °C.

Compound A. L-Cysteine (0.660 g) was dissolved in 400 mL of water. Activation of the carboxylic acid groups of the thiol molecules was achieved by adding EDC, at a concentration of 50 mM. A slurry of chitosan (1.0 g) and acetic acid (4 mL) was added and allowed to dissolve.

Compound B. Thioglycolic acid (0.38 mL) was dissolved in 400 mL of water, along with 50 mM EDC. Chitosan (1.0 g) and acetic acid (4 mL) were added to the solution to create compound B. A second formulation was prepared with 2.65 mL of thioglycolic acid to make compound B′.

Compound C. 2-Iminothiolane hydrochloride (0.406 g) was dissolved in 400 mL of water before addition of a 1.0 g/4 mL slurry of chitosan and acetic acid. The synthesis proceeded, without aid of a carbodiimide, via nucleophilic attack of the imino carbon by electrons of the chitosan amine group. This ring-opening reaction resulted in the formation of a carboxamidine linkage featuring a terminal thiol group.

Compound D. Because of the low solubility of lipoic acid in water, significant modification of the above procedure was necessary. Chitosan (1.0 g) and acetic acid (4.0 mL) were added to 50 mL of water and allowed to dissolve. Just prior to the introduction of lipoic acid, EDC (0.48 g) was dissolved. Lipoic acid (0.5 g) dissolved in a minimal amount of methanol was added to the chitosan solution. Methanol was added until the precipitated lipoic acid redissolved. The final solvent ratio was 50 mL of water to 75 mL of methanol.

Material Analysis. Elemental analysis of carbon, nitrogen, hydrogen, and sulfur content was performed by atomic absorption spectroscopy (Atlantic Microlab, Inc., Norcross, GA).

Colorimetric testing of sulfur concentrations was also performed using 5,5′-dithiobis(2-nitrobenzoic acid), commonly known as Ellman’s reagent. This procedure was adapted from previously published methods.^{6, 7, 17, 18} First, 10 mg samples of chitosan and each derivative were carefully weighed and placed in 1.5 mL microcentrifuge tubes. The samples were hydrated with 1 mL of phosphate buffered saline (PBS) at pH 8. To each tube was also added 0.5 mL of indicator solution, prepared by dissolving 40 mg of Ellman’s reagent in 40 mL of PBS. A series of L-cysteine solutions, ranging from 1 × 10⁻² to 1 ×10⁻⁹ M, were similarly prepared as calibration standards.

The samples were gently agitated for 6 h at room temperature before being centrifuged. The absorbance of each supernatant was measured using a USB2000 miniature fiber optic spectrometer in transmission mode, with a DH-2000 deuterium and tungsten halogen light source (Ocean Optics, Dunedin, FL). To determine the sulfur concentrations in each sample, their measured absorbance intensities at 450 nm were compared with the sigmoidal dose–response curve generated from the L-cysteine calibration solutions.

Infrared spectra were obtained using a Nicolet 380 FTIR spectrometer (Thermo Scientific, Waltham, MA) using lyophilized polymer samples on a diamond ATR crystal.

¹H NMR analysis was performed with a 300 MHz instrument (Varian, Inc., Palo Alto, CA). NMR samples were prepared by hydrating 5 mg of polymer sample with 20 µL of acetic acid-d₄ before adding 1 mL of deuterium oxide. Spectra were obtained at room temperature with a spin speed of 20 Hz, a 31.5° pulse, and an acquisition time of 3.744 s.

Thin Film Preparation. The unmodified and modified chitosans were dissolved in 1% (w/v) acetic acid solution (0.4 g per 30 mL) to produce viscous syrups. To render films insoluble following deposition, 60 mg of the thermally activated and water-soluble cross-linking agent hexamethylene 1,6-di(aminocarboxysulfonate) (HDACS) was added to each solution.^19–21 The solutions were vacuum filtered through a coarse fritted glass funnel and degassed in a sonicator for 5 min to minimize bubbles. The solutions were then dispensed onto small fragments of cleaned silicon wafers. The cleaning preparation of the silicon substrates is described elsewhere.^{19, 22, 23} The solution-coated wafers were spun on a spin processor (Laurell Technologies, North Wales, PA) at speeds between 1500 and 3000 rpm. Samples were dried on a mildly warm hot plate under a stream of argon and were placed in a 60 °C oven overnight to allow HDACS cross-linking to take place.

Metal Ion Analysis. An M-2000U variable-angle spectroscopic ellipsometer (J. A. Woollam Co., Lincoln, NE) was used to measure the thicknesses and refractive indices of the thin films. Refractive index values were calculated at a wavelength of 589.3 nm using a three-term Cauchy equation. Film reflectance was measured, normal to the film surface, using the aforementioned fiber optic spectrometer in reflectance mode. Further details of the analysis procedures have been described previously.²³

Three replicate thin films comprised each sample set. Following initial measurements, the samples were submersed in aqueous solutions containing a dissolved salt of the metal species of interest. Each solution contained 50 ppm of the selected cation or counterion. After five minutes, the films were removed from the solutions and dried under argon on a warm hot plate. Thickness, refractive index, and reflectance measurements were once again taken to determine any change brought about by exposure to the ionic analytes.

Results and Discussion

Extent of Thiolation. The addition of small sulfur molecules to the chitosan backbone was evaluated by the percent sulfur found in atomic absorption. Elemental analysis results from unmodified chitosan (provided by Sigma-Aldrich) and modified chitosan derivatives are summarized in Table 1.

Atomic absorption analysis indicated measurable amounts of sulfur in compounds A, B, and C. Of interest is the lack of sulfur found in compound D (the lipoic acid conjugate). To confirm this result, a second sulfur-specific testing methodology was employed using Ellman’s reagent. The results of this study are presented in Figure 2. The relative trends in sulfur concentrations measured with Ellman’s method are in good agreement with the values arrived at by elemental analysis despite the fact that the sulfur concentrations were outside the linear Beer’s law region of the calibration curve. Compound D has a slight response for sulfur beyond the baseline set by unmodified chitosan; however, its level of error does not preclude the possibility of a false positive. The faint sulfurous odor detected on the lipoic sample suggests the possible presence of unbound thiols. The observation of this odor after dialysis suggests that a small amount of lipoic acid may be associated in a noncovalent manner with the biopolymer sample, perhaps through electrostatic bonding as reported by Bernkop-Schnürch et al.²⁴

Figure 2. 5,5′-Dithiobis(2-nitrobenzoic acid) (Ellman’s reagent) was used to detect sulfur in the modified polymers. Each sample was washed in pH 8 PBS solution containing Ellman’s reagent. The positive response to sulfur was quantified by absorbance at 450 nm and correlated to molar sulfur concentrations of an L-cysteine calibration curve (shown inset). The response of unmodified chitosan at 4.5 × 10−5 M is assigned as the baseline response. Click to Enlarge

FTIR Analysis. Infrared spectra of chitosan and the four chitosan derivatives are shown in Figure 3. Evidence of conversion of the chitosan amine group is provided by the decrease of the amine NH bend signal at 1591 cm⁻¹ in each of the modified chitosans. Coupled with the loss of amine signal is evidence of amide formation in the modified chitosans, with notable amide I band signals between 1625 and 1675 cm⁻¹ and/or amide II signals between 1520 and 1560 cm⁻¹. Predictably, the CH bend around 1376 cm⁻¹ and the COC stretch near 1149 cm⁻¹ are essentially unchanged between unmodified chitosan and the chitosan derivatives. The spectra of compound D features signals at 1203 and 1185 cm⁻¹, which are attributed to the C(O)O stretch and OCC asymmetrical stretch associated with the ethanoate group.

Figure 3. The transmittance IR spectra of chitosan and chitosan derivatives are shown for comparison. All spectra possess similarly positioned CH bend and COC stretch signals. There is evidence of a loss of amine signal near 1591 cm−1 and strengthened amide I/II bands in the modified chitosans, suggesting that addition is occurring at the chitosan amine site. The spectra of compound D (N-acetyl chitosan ethanoate) features a C(O)C stretch near 1203 cm−1 and the OCC asymmetrical stretch at 1185 cm−1. Click to Enlarge

NMR Analysis. The percent degree of deacetylation was measured from ¹H NMR spectra using the method described by Hirai et al.:²⁵

An enlargement of the pertinent spectral region is shown in Figure 4. The position 2 singlet peak is located at 1.87 δ, and the positions 3, 4, 5, 6, and 6′ hydrogen atoms are observed at 3.71, 3.59, 3.58, and 3.53 δ, respectively. A broadened and distorted triplet, characteristic of strongly coupled methyl hydrogen absorption in the acetyl group, is observed centered at 1.87 δ. Using Hirai’s technique, the degree of deacetylation of the unmodified chitosan used in this study was measured to be 91.88%, similar to the value of 90.85% reported by the supplier.

Figure 4. In 1H spectra of chitosan, the N-acetyl methyl hydrogen atoms in unmodified chitosan exist as a broadened triplet upfield from other protons of interest at 1.87 δ. The relative intensity of the acetyl −CH3 signal was compared with signals for hydrogen at positions 2, 3, 4, 5, 6, and 6′ to determine the degree of deacetylation (91.88%). In the spectra of compound D, the relatively intensity of the N-acetyl signals is increased, representing a 76.47% degree of deacetylation. Also apparent is the emergence of ethanoate signal, centered at 2.04. Click to Enlarge

The structure of compound D is greatly elucidated by NMR and FTIR spectra, which suggest the addition of acetyl groups at both the amine and primary alcohol sites of the chitosan chain. The reacetylation of the amine is evidenced by the heightened intensity of the methyl hydrogen triplet centered at 1.87 δ, while the abutting signal corresponding to the acetyl addition at the primary alcohol position is observed upfield at 2.04 δ. Integration of these −CH₃ signals in compound D suggests that addition at the alcohol site is preferred 3-fold over the acetylation of the amine. The degree of N-deacetylation in compound D is 76.47%.

Metal Ion Analysis. Films of unmodified chitosan exhibited minimal response to ultrapure water, with a change in thickness of only −1.3 ± 1.2 nm and a change in refractive index of −0.003 ± 0.003. The excellent water stability of these unmodified chitosan films is attributed to the high efficacy of the HDACS cross-linking agent. The resistance to change in water is improved from similar samples described in previously published work, in which HDACS cross-linked chitosan films decreased in thickness by −6.3 ± 0.7 nm when exposed to water.²² In this earlier study, the chitosan films had been coated from a solution containing 2.5% acetic acid, rather than the 1% acid solution used here. We contend that the more acidic environment of the earlier study was responsible for lowering the number of unprotonated amine groups available for reaction with HDACS, leading to a reduction in cross-link density and a heightened propensity for partial film dissolution.

Interactions between chitosan and metal ions are often ascribed to chelation mechanisms between the metals and chitosan amine groups, which possess free electron pairs that play a key role in polymer–metal interactions.² Chemical manipulations of the polymer through the amine group are likely to bring about changes in metal reactivity. In this case, the grafting of thiol-containing small molecules to chitosan amines has increased the potential for sulfur−metal interactions.

The physical and optical responses of thin films to several ionic solutions are shown in Figure 5. Measurements were made for cross-linked films of chitosan and modified chitosan derivatives and compared with data from previously published work involving Ca²⁺ cross-linked thin films of the biopolymer alginate.²³

Figure 5. Changes in thickness and refractive index (at 589.3 nm) were measured for films of unmodified and modified cross-linked chitosans after immersion in 50 ppm ionic solutions and compared with previously obtained results from thin films of Ca2+ cross-linked alginate. Click to Enlarge

It is important to note the measurements obtained of each film type after exposure to ultrapure water. Neither alginate nor unmodified chitosan exhibited statistically significant changes in thickness or index as a response to water. In contrast, three of the modified chitosan films decreased in thickness after water exposure (by −13.0 ± 0.4 nm, −6.5 ± 0.6 nm and −11.1 ± 0.1 nm for compounds A, B, and C, respectively). Compound D underwent minimal thickness change (−1.5 ± 0.5 nm) and is thus most like unmodified chitosan in its water response.

The issue of availability of free amines for cross-linking provides one explanation for the heightened sensitivity to water observed in thiolated chitosan samples. The added thiol molecules blocked amine groups that would be potential reaction sites for the HDACS cross-linking agent. By hindering cross-linking, the films were left more susceptible to partial dissolution in water. It is likely that this blocking effect could be overcome by adjusting the amount of cross-linker to account for the predicted addition of thiol groups, or by employing alternative cross-linking strategies through the chitosan alcohol group.

Steric hindrance provides another explanation for the reduction in cross-linking efficiency among the modified chitosans. The addition of grafted molecules to the polymer backbone may limit chain mobility and rotation, thus preventing nearby unblocked amine groups from assuming positions favorable for cross-linking. Compounds A and C, which feature larger side groups, are more prone to dissolution than compound B, with its less bulky thioglycolic acid side group. The cross-linking efficacy in compound D, evidenced by its response to water, is analogous to that of unmodified chitosan, suggesting that acetyl and ethanoate functionalities do not present significant steric obstacles for cross-linking.

Film responses to the three tested chromium solutions illustrate the potential for thin film selectivity for particular chemical speciation. In our previous work, it was found that Ca²⁺ alginate films exposed to trivalent chromium nitrate and chromium acetate solutions exhibited positive changes in thickness (7.2 ± 0.4 and 14.7 ± 1.0 nm) but underwent a dramatic decrease in thickness when exposed to hexavalent chromium oxide (−19.2 ± 0.8 nm).²³ Contrastingly, films of unmodified chitosan showed significant thickness increases in response to all three chromium solutions (9.4 ± 0.7, 12.6 ± 2.2, and 17.3 ± 0.8 nm, respectively). The responses of the modified chitosans are markedly different than the response of unmodified chitosan, even after accounting for the loss of thickness caused by water.

Figure 6 summarizes the changes in thickness and refractive index for each type of film. The results have been normalized to account for the response of each film type to ultrapure water. Solutions containing Hg(II), Cr(III), and Cr(VI) elicited significant thickness responses in each of the chitosan systems. The Hg(II) and Cr(VI) also caused significant refractive index shifts in all samples. Compound B, the cysteine conjugate, underwent significant thickness changes in responses to all solutions except sodium acetate and disodium hydrogen phosphate. Both disodium hydrogen phosphate and sodium sulfate solutions caused significant thickness responses in four of the five chitosan systems, suggesting that the phosphate and sulfate counterions may have noteworthy interactions with the films.

Figure 6. This matrix summarizes the thickness and refractive index responses of the chitosan and chitosan derivative films after exposure to 50 ppm ion solutions. A filled block indicates a statistically significant change in film thickness, and an X indicates a statistically significant change in refractive index. All results have been normalized against the response of each type of film to pure water. Click to Enlarge

The mean initial thicknesses for the 48 thin films used in each film set were 131.5 ± 2.5 nm (chitosan), 95.9 ± 0.6 nm (compound A), 84.9 ± 0.8 nm (compound B), 96.8 ± 0.9 (compound B′), 84.0 ± 1.7 nm (compound C), and 88.4 ± 0.4 nm (compound D). The reason for the reduced thicknesses of the modified chitosans stems from the reduced viscosities of their solutions prior to deposition on the silicon substrates. The modification processes shifted the solubility ranges of the modified chitosans closer to neutral pH. However, all films were coated from 1% acetic acid solutions, resulting in variability in solution viscosity. The mean initial refractive indices for each sample set were 1.513 (chitosan), 1.546 (compound A), 1.563 (compound B), 1.540 (compound B′), 1.523 (compound C), and 1.528 (compound D). There was minimal scatter in refractive index measurements, leading to standard error values below the significant digit threshold of the reported indices.

The results presented in this study have been evaluated under the assumptions that mean film thicknesses are sufficiently small and thickness distributions of the various film sets are sufficiently narrow to avoid any appreciable thickness-dependence of results. With substantially thicker films, approaching the dimensions of bulk materials, one might expect graduated physical/optical responses depending upon the diffusion of analytes into the hydrogel and the accessibility of functional groups. However, our biopolymer films in this thickness regime (typically <200 nm) have not exhibited appreciably different responses to analytes. We investigated the effects of initial thickness in a previous study with cross-linked alginate films, and found only minimal response differences among films with initial thickness ranging from 67 to 191 nm.²³

Film measurements were taken in an air-conditioned space, with a general temperature range of 21–23° and humidity levels typically between 45 and 55%. No formal study was conducted to evaluate the performance of ambient environmental conditions on film response. However, in instances where air-conditioning went offline and films were exposed to humid summer air, there was a noted increase in film thickness and decrease in refractive index, as would be expected of samples with a higher water content. We have found that films made on different days in the same conditions have highly reproducible results, and we have been able to reproduce experimental outcomes with films made months or years after those used in initial tests. However, to ensure that film measurements were taken with the most consistent ambient conditions, metal ion studies for each sample set were performed in single-day sessions, during which times environmental conditions were monitored for change.

Varying Sulfur Content. As previously noted, two preparations of the thioglycolic chitosan conjugate were made with differing amounts of thioglycolic acid. The more highly thiolated product, compound B′, contained more than twice the sulfur content of the less thiolated product, compound B.

The results of the thickness, refractive index, and reflectance measurements of the compounds B and B′ films were examined to determine whether statistically significant changes were brought about by exposure to ionic analytes. The effect of varying thiol content on film reflectance is shown in Figure 7. A table comparing thickness and refractive index measurements of compounds B and B′ films is contained in the Supporting Information.

Figure 7. Thin films made from two formulations of the thioglycolic chitosan conjugate, compounds B and B′, exhibited changes in reflected color after exposure to metal ion solutions. The films exhibit similar reflectance changes after immersion in pure water but have markedly different responses to ionic solutions. Hg(II), Cr(III) (chloride and nitrate counterions), and Cr(VI) all caused red-shifting of film reflectance in compound B. However, only Hg(II) led to red-shifting for films made from compound B′, which had a higher concentration of thiol pendants. Click to Enlarge

Both compounds B and B′ have remarkably similar reflectance responses to water. The more highly thiolated B′ conjugate is slightly more blue-shifted than its less thiolated B counterpart after water exposure. However, the maximal reflected wavelengths after water exposure differ by only 4.1 ± 2.0 nm, a statistically real but small variation. There is no statistical difference between water-induced thickness changes in B and B′, suggesting that the extra thiol content in B′ does not correlate to a appreciably less water-stable film. On the other hand, the higher thiol concentration significantly affects film sensitivity to ionic analytes. Generally, the additional sulfur groups are tied to decreased film thickness response and a shift of maximal light reflectance to a lower wavelength, except in the case of mercury nitrate solution.

As shown in Figure 7, four of the analytes induced film red-shifting in compound B: Hg(NO₃)₂ and CrO₃, with significant red-shifting, and CrCl₃ and Cr(NO₃)₃ with relatively modest red-shifting. With the increased sulfur concentration of compound B′, the red-shifting response is solely confined to the Hg solution, thereby allowing explicit differentiation of Hg(NO₃)₂ solution from all others tested. For each of the chromium analytes, increased thiol concentration led to a blue-shifted reflectance response. The increased red-shifting of the B′ film after Hg(II) exposure is rooted in positive increases in both film thickness and refractive index. The preferential binding of Hg with thiolated chitosans in bulk form has been previously reported.^{26, 27} We contend that the change in metal ion selectivity resulted from changing the ratio of available primary amine groups and amide-bonded thioglycolic pendants.

Conclusions

Changes in physical and optical properties of thin films, caused by interactions with metal analyte species, can cause pronounced changes in the electromagnetic waves reflected by the films. When these changes take place in the visible regime, reflectance responses may be perceivable to the unaided eye and can serve as a strategy for optical-based sensing. This work demonstrates the potential for differentiating between various metals in aqueous solution using a variety of thin films made of chitosan and thiol chitosan conjugates. Each film type possessed a unique pattern of responses to the series of analyte solutions. By examining the response of several film types to a single solution, one can find a pattern of changes in reflection, thickness and refractive index to identify the analyte species present.

The usefulness of such a detection system is greatly enhanced by narrowing the response specificity of a single film type. By increasing the sulfur content in the thioglycolic chitosan conjugate, a variant polymer was made whose films selectively red-shifted in response to Hg(II) without the red-shifting response to Cr(III) and Cr(VI) found in the original formulation. These interaction differences are likely rooted in the ratio variation of amine and thiol functionalities along the polymer chain. Films made of this highly thiolated chitosan, compound B′, will be further investigated for their detection limit in measuring mercury salt solutions.

The difficulties of developing a biopolymer thin film sensor for real-world applications are rooted in the numerous complicating factors found in different water samples, including the presence of organic materials and ligands as well as variations in pH and temperatures. Perhaps the most predominant factor is the simultaneous presence of numerous metal ions and counterions, which may participate in complex synergistic and conflicting mechanisms as they compete for binding sites in the polymer thin film structure. However, metal ion selectivity in multi-ion systems has been achieved, primarily through chelation mechanisms.¹² Therefore, by controlling the functional groups on our thin films, the selectivity will be enhanced.

Supporting Information Available

Tables summarizing all changes in ellipsometer measurements of thickness and refractive index, with statistical treatments for compounds B and B′, are available. Also included are reflectance profiles, before and after analyte exposure. This material is available free of charge via the Internet at http://pubs.acs.org.

* Corresponding author. E-mail:cschauer@cbis.ece.drexel.edu.