Polydiacetylene Nanofiber Composites as a Colorimetric Sensor Responding To Escherichia coli and pH.

Polydiacetylenes (PDAs) are conjugative polymers that demonstrate color changes as a response to an external stimulus. In this study, 10,12-pentacosadiynoic acid (PCDA) was mixed with a supporting polymer including poly(ethylene oxide) (PEO) and polyurethane (PU), and the mixture solution was electrospun to construct fiber composites. The electrospun fibers were then photopolymerized using UV irradiation to produce PEO-PDA and PU-PDA nanofiber mats with a fiber diameter ranging from 130 nm to 2.5 μm. The morphologies of both PEO-PDA and PU-PDA nanofibers were dependent on electrospinning parameters such as the ratio of PCDA to PEO or PCDA to PU and the total polymer concentrations. Scanning electron microscopy images showed beaded fibers of PEO-PDA and PU-PDA at 2 and 18 w/v % concentrations, respectively. Smooth fibers were found when the solvent concentration was increased to 3.75 w/v % in PEO-PDA and 25 w/v % in PU-PDA fibers. Both PEO-PDA and PU-PDA nanofiber composites demonstrated excellent colorimetric responses to the presence of Escherichia coli ATCC25922 bacterial cells and the changes in pH as external stimuli. The nanofibers underwent a rapid colorimetric response when exposed directly to E. coli ATCC25922 grown on Luria-Bertani agar. The comparison between the PEO-PDA and PU-PDA suggested that the combination of PEO and PDA is favorable because it provides a sensitive response to the presence of E. coli. The results were compared with samples of a PDA polymer in the absence of a matrix polymer. The colorimetric response was similar when the PDA polymer and the PDA nanofiber composites were exposed to pH changes, and the color change was found to occur at pH 10 and enhanced at pH 11-13. The PDA-containing nanofiber composites showed stronger colorimetric responses than those of the PDA polymer only, suggesting their potential as biosensors and chemosensors.

Introduction

Bacterial infection of wounds, including burns, diabetic foot ulcers, and surgical-site infections, impact approximately 2 million people and cost more than $18 billion in direct medical cost annually in the United States.[1] For example, chronically infected diabetic foot ulcers are the most critical wound care problem worldwide, and 14–24% of these cases eventually suffer an amputation.[2] Wound dressings are typically used to protect the wound and surrounding tissue from contamination and to promote wound healing. Conventional wound dressings are generally not suitable for chronic and acute wounds, nor are they suitable for treatment or monitoring the infection status of chronic and acute wounds. Novel strategies are in great need for early detection of wound infection to prevent further complications and to enhance the healing process. One recently used strategy is to introduce biosensors that can monitor in situ the presence of bacteria in wounds and hence improve wound care and management. A biosensor is an analytical device that is made of an analyte in combination with a biological element such as an enzyme, antibody, or nucleic acid.[3,4] The biological element interacts with the analyte, resulting in a biological response that can be converted into electrical or electrochemical signals.[5] In particular, colorimetric biosensors have been attractive because of their ease of use, rapid response, high prevision, accuracy, and cost-effectiveness.[6] If colorimetric biosensors could be integrated into wound dressings, the wound dressings could potentially enable early detection of infections in wounds by providing in situ monitoring of the wound condition via colorimetric indication.[7] Biosensors can function in different transducers, including electrochemical, optical, electronic, piezoelectric, gravimetric, and pyroelectric.[8] In electrochemical biosensor development, conjugated polymers can be used to create the bio-transducer components in the sensor.[9] Polyaniline, polypyrrole, and polydiacetylene (PDA) are mostly used conjugative polymers to construct electrochemical biosensors because their electrochemical properties are usually associated with visible colorimetric changes.[10] PDAs are especially attractive because they exhibit a blue-to-red color transition visible to the naked eye when they are subjected to external stimuli such as changes in temperature,[7,11−13] pH,[14] and the presence of bacterial cells,[15] and aromatic compounds.[16] PDAs were first synthesized by Wegner in 1969, who identified the polymerization as a 1,4-addition reaction.[17] The potential use of PDAs in developing biosensors was first reported in 1993 when Bednarski et al. coated a glass slide with a PDA bilayer and developed a direct colorimetric detection method for sialic acid, a receptor-specific ligand for the influenza virus.[18] Since then, a range of sensing systems were developed using PDAs such as films,[19] crystals,[20] and fibers.[21] Our previous studies showed that PDAs can be used in nanofiber composites for sensor applications, demonstrating their potential use in wound dressings and medical textiles.[7] Sensor sensitivity is generally increased via an increase in the contact surface area of biosensors. The surface area of nanofibers can be exceptionally high because of small diameters and high aspect ratios such as the diameter-to-length ratio. Nanofiber composites exhibit excellent flexibility as well as high permeability, which meet ergonomic and physiological preferences in the application of medical textiles such as wound bandages and surgical dressings. Because of the high cost and low spinnability of PDAs, the PDAs have been successfully incorporated into fiber/nanofiber composites with other polymers, such as polymethyl methacrylate, polystyrene, tetraethyl orthosilicate, poly(ethylene oxide) (PEO), which serve as supporting components or matrix polymers in the composites. The PDA-containing nanofiber composites could be used to develop wound bandages and surgical dressings with biosensing capabilities.[10,22]

In this study, supporting polymers of PEO or polyurethane (PU) were separately mixed with the PDA monomer, 10,12-pentacosadiynoic acid (PCDA), and the mixtures were used in an electrospinning apparatus to create nanofiber composites. The nanofibers were then photopolymerized via UV irradiation resulting in PEO–PDA and PU–PDA nanofiber composites. The biosensing properties of the PEO–PDA and PU–PDA nanofiber composites were evaluated for their potential application as wound bandages and surgical dressings. PEO is a linear, semi-crystalline, biocompatible, nontoxic polymer approved for internal use in food, cosmetics, and pharmaceutical and personal care products.[23] PEO is widely used because of its amphiphilic properties as well as its solubility in aqueous and organic solvents[24] and its stability in air.[25] On the other hand, PU is also frequently used in medical products because of its good barrier properties and oxygen permeability. PU has been traditionally proven to be a bio- and hemocompatible material.[26] The electrospun nanofibers of PEO–PDA and PU–PDA were exposed to Escherichia coli (ATCC25922) culture, and the colorimetric behaviors were monitored to evaluate their biosensing properties. It was found that both nanofiber composites demonstrated colorimetric responses (CRs) to E. coli; however, the colorimetric changes in the PEO–PDA nanofibers responded differently than those changes in PU–PDA nanofibers. The colorimetric properties were measured using a photospectrometer as a function of exposure time to E. coli. The conditions of the bacterial culture, such as pH in Luria–Bertani (LB) media have a great impact on the bacteria and their growth. Thus, colorimetric responses in the nanofiber composites were also monitored using the photospectrometer as a function of pH in a buffer solution without the presence of E. coli to study the correlation between biosensing behaviors and pH. The results showed no visible color changes in the PDA nanofibers at pH < 11. Because most of known bacteria live in pH from 6 to 7, the results suggested that no false signal in detecting bacteria would occur at physiological pH. Our results suggested a great potential of using the PDA-containing nanofiber composites in wound bandages and surgical dressings to detect the presence of bacterial infection and to continuously monitor wound health status.

Results and Discussion

Fiber Morphology

Figure A,B shows the scanning electron microscope (SEM) images of PU–PDA and PEO–PDA electrospun fibers obtained at different spinning conditions. Polymer concentration and mass ratios of matrix polymers to PCDA (the monomer of PDA) had a great impact on fiber morphology. The morphology of PU–PDA fibers was generally different from that of PEO–PDA fibers. Beads on fibers were present in the PEO–PDA fibers obtained at 2 w/v % concentrations and the PU–PDA fibers obtained at 18 w/v % concentrations when the mass ratio of matrix polymers to PCDA was 2:1. During electrospinning, PDA solutions with low viscosity experienced a low viscoelastic force resulting in a partial break up in the electrical jet. Because of the effect of surface tension, free solvent molecules in the solution accumulated into a spherical shape, causing a formation of beads. The number of beads was significantly reduced when the concentrations were higher at 25 w/v % for PU–PDA fibers and 3.75 w/v % for PEO–PDA fibers. The mass ratio between matrix polymers to PCDA had a similar effect on fiber morphology. Fewer beads were formed at the high mass ratio (6:1) for both PU–PDA and PEO–PDA fibers. The formation of beads was attributed to the insufficient polymer chain entanglement at low concentrations. In principle, the polymer chains are stretched by electrostatic forces during electrospinning, resulting in linear fibers only if sufficient polymer chain entanglements prevent breakage and discontinuity in the solidified fibers. A low concentration of the polymers results in insufficient chain entanglements that result in beaded fibers.[23,29]

Figure 1

SEM images of PU–PDA (A) and PEO–PCDA (B) electrospun fibers. Fibers presented in the upper row are prepared with a ratio of 2:1 matrix polymers to PCDA, and those presented in the lower row are prepared with a ratio of 6:1 matrix polymers to PCDA. The images in the columns present fibers obtained at low and high polymer concentrations. Graphs (C) and (D) depict the average diameter of electrospun fibers for PU–PDA and PEO–PDA, respectively.

Matrix polymers used in electrospinning are primarily to enhance the spinnability of PDAs because it is difficult to electrospin PDAs alone owing to their impaired solubility and low viscosity. Therefore, the concentration of the matrix polymer (PU or PEO) was calculated based upon the mass ratio and total polymer concentration. Figure B,C shows the correlation between the fiber diameter and the concentrations of PU and PEO, respectively. The fiber diameters of the PDA fibers prepared with both PU and PEO varied with the change in mass ratios. The diameter of fiber samples of PU–PDA and PEO–PDA increased linearly with the increase of the matrix polymer concentration (PU–PEO), which is attributed to the higher viscosity and surface tension of the electrospinning solution.[30] In addition, beads on the string were formed at low concentrations because of the low viscosity of the electrospun solution. Coarse fibers were formed at the high mass ratio of matrix polymers (PU or PEO) to PCDA. On the other hand, the SEM images shown in Figure suggest that the surface roughness of the fibers increased with an increase in the fiber diameter. The surface roughness of PU–PDA and PEO–PDA fibers appeared slightly different. This might be due not only to the nature of different matrix polymers but also to the different solvents used. Surface morphology of PEO and PDA electrospun fibers was previously reported differently when different solvents were used including chloroform, methylene chloride, and dimethyl formamide.[31]

Chemical Analysis via ATR–FTIR

Colorimetric changes in conjugated polymers are usually due to conformational changes in the conjugated macromolecules such as PDAs. In this study, PDAs were mixed with PEO or PU, resulting in PEO–PDA and PU–PDA nanofiber composites. This leads to a question whether the PEO or PU mixed with PDA has an effect on conjugated macromolecules at a molecular level and hence influences electrochemical properties in the nanofiber composites. Therefore, chemical analysis was carried out via attenuated total reflectance–Fourier transform infrared (ATR–FTIR) for PDAs including PDAs embedded in PEO and PU fibers and PDA in the absence of matrix polymers. ATR–FTIR measurements were conducted on samples in the blue phase prior to chromatic changes, and the spectra are shown in Figure . Spectrum A corresponds to the PU–PDA fibers, where the following resonance features were interpreted for characterization; IR νmax (cm–1): 3324 (N–H stretch), 2919–2847 (C–H stretch), 1692 (C=O stretch), 1104 (C–O stretch), and 723 (C–H bend). Similar features were observed in spectrum B representing PEO–PDA fibers; IR νmax (cm–1): 2919–2847 (C–H stretch), 1691 (C=O stretch), 1097 (C–O stretch), and 723 (C–H bend). The resulting resonance features were similar to previously reported results for PDAs, and the spectra were consistent with the anticipated structures.[7,32] PDAs synthesized in the absence of a matrix polymer were also characterized by ATR–FTIR before the color transition and are shown in Figure . Spectrum C represents PDA in the absence of a matrix polymer with key vibrational bands originating at IR νmax (cm–1): 2918–2847 (C–H stretch), 1690 (C=O stretch), and 722 (C–H bend). Resonance features that indicate the presence of PDAs are seen on all spectra, where the most prominent stretch is due to the carbonyl. Substitution at the terminal hydroxyl groups of the dicarboxylic acid monomers is shown by the absence of resonance features that correspond to those groups. The results suggest that the characteristic macromolecular structures of PDA were not altered after the composite polymer was formed with the matrix polymer. Thus, the electrochemical and colorimetric properties are not significantly altered in the nanofiber composites.[33]

Figure 2

ATR–FTIR spectra of polymeric fibers and PDA polymers. The signals correspond to PU–PDA fibers (A), PEO–PDA fibers (B), and PDA (C).

Colorimetric Response To Bacteria

PEO–PDA and PU–PDA fiber composites were immersed in E. coli culture, and their colorimetric properties were monitored as a function of contact time. The fibers were in the blue phase before they were immersed, and after immersion, the fibers demonstrated color changes to red immediately. The fiber samples that were immersed in a control dish containing only LB agar prepared without E. coli remained blue, and no visible colorimetric change was observed. To confirm our hypothesis involving membrane-secreted compounds, additional controls were investigated by using the pellet or the supernatant. Our findings indicated that the fiber in contact with the pellet underwent a color transition from blue to red, whereas the fiber in contact with the supernatant did not fully show a red shift. The fiber mat that was exposed to the supernatant appeared purple. The change suggests that bacterially secreted proteins that were present in the media might be capable of initiating the color transition.[15]Figures S2 and S3 show photographs of colorimetric responses of the PU–PDA and PEO–PDA fibers every 30 min from t = 0 min to t = 3 h, respectively. The color changes were different in the fibers obtained at various electrospinning parameters.

The color transition from blue to red was rapid in the fibers with a high mass ratio (6:1). The slowest transition of color was observed in the fibers that were prepared at the lowest concentrations and the lowest mass ratio. It was found that the color of beaded fibers remained unchanged for a longer time than that of the fibers without a significant amount of beads. The beads on the fibers might disturb the interaction with bacterial cells and cause a delay in color transition. Continuous fibers without beads exhibited a fast and prominent color change. The color change became less drastic in the PEO–PDA fibers at 1.5 h and in the PU–PDA fibers at 2.5 h. The intensity of the red color in the PEO–PDA fibers reached its maximum after 3 h. A similar trend was found in the PU–PDA fibers. The color transition continued, and no significant color change was observed after 3 h. The PU–PDA fibers obtained at a mass ratio of 6:1 and 18 w/v % concentration exhibited an early and vivid color transition as compared to the rest of fibers. The color change became more distinct as the pH of the solution increased and the concentration of bacteria increased.

Representative reflectance spectra are shown in Figure for the PEO–PDA fiber mat that was obtained at a 2:1 mass ratio and 3.75 w/v % solvent concentration. Untreated polymerized blue fiber mats that were used as controls showed a slight increase in the reflectance value at 640 nm, but remained unchanged at 540 nm. After the fibers were exposed to E. coli for 30 min, they started to change color and exhibited a wide reflectance peak at 610 nm, which bypassed the original peak at 640 nm. The reflectance switch on the spectra was continuously developed when the interaction between the fibers and E. coli continued for 3 h.

Figure 3

Reflectance spectra of the PEO–PDA fiber mat at a 2:1 mass ratio and 3.75 w/v % solvent concentration over time after the exposure to E. coli. The solid line represents the blue PDA fibers (control), and the different dotted lines represent the fibers after exposure to E. coli for different time periods.

A comparison between the PEO–PDA and PU–PDA fibers is made to investigate the colorimetric transition in different fiber composites. Both the PEO–PDA and PU–PDA fibers obtained at a 6:1 mass ratio showed a pronounced reflectance switch. It was found that the low switch was shown by the fine fibers obtained at a 2:1 PEO–PDA ratio and 2 w/v % polymer concentration, whereas a high switch was shown by the coarse fiber obtained at a 6:1 PEO–PCDA ratio and 3.75 w/v % polymer concentration. Our earlier discussion on the fiber size indicated that an increase in the amount of PEO in electrospinning solution influenced the fiber size, resulting in an increase of the fiber diameter. The increase of the fiber diameter thus expanded the exposed surface area and enhanced the color transition from blue to red. Additionally, the fiber surface became rough with the increase of the fiber diameter. The high reflectance of the color switch from blue to red that was associated with the coarser fibers was probably due to the high reflectance surface area of the rough fibers.

Previous studies proposed that the mechanism responsible for the color change might be attributed to the release of endotoxins, such as lipopolysaccharides, from the Gram-negative bacterial strain. As membrane-active compounds are secreted by proliferating bacteria, the PDA undergoes conformational transitions by perturbations that disrupt hydrogen bonding of the head-groups and favor binding of positively charged ions.[15] Further studies that explore the interaction between bacterial cells and PDAs are needed to investigate the driving force of the color change as a response to E. coli.

The colorimetric response of PEO–PDA and PU–PDA fibers is presented in Figure . The % CR between the blue and red reflectance correlates to the intensity of the color transition. Figure A shows that at the same time integral, the PEO–PDA fibers obtained at the PEO–PCDA mass ratio of 2:1 and the concentrations of 2 w/v % demonstrated lowest % CRs (2.6–3.6), which represent the lowest color transition. Figure B shows that at the same time integral, the PU–PDA fibers obtained at the PU–PCDA mass ratio of 2:1 and the concentrations of 18% also presented lowest % CRs (1.7–2.9), showing the lowest color transition. The increases in both mass ratio and concentration for both fibers were able to enhance the colorimetric response of the PDA-containing composites. It is in agreement with the previous conclusion that the increase in the concentration of the matrix polymer (PEO) increases the surface area of the fiber that is in contact with the bacterial cells and enhances colorimetric response behaviors. In addition, it is important to notice that by comparison of Figure A,B, the % CRs of the PEO–PDA composites were higher than those of the PU–PDA composites at the similar spinning conditions, suggesting faster color transition occurring in the PU–PDA composites. For example, the maximum % CRs by the PEO–PDA and PU–PDA are 15.2 and 10.6, respectively. This may be due to the interaction between the two components in the composites, which include the sensing component of PDA and the matrix polymer. The results show that the combination of PEO and PDA give rise to a favorable and sensitive response in the presence of E. coli. The mechanism of sensitive response by the PEO–PDA composites needs further investigation.

Figure 4

The colorimetric response values of PEO–PDA and PU–PDA fibers after direct contact with E. coli for 3 h.

It can be noticed that both PU–PDA (6:1 mass ratio and 18%) and PEO–PDA (2:1 mass ratio and 3.75%) fibers demonstrated a high % CR. These values were also associated with a relatively large variation shown in Figure , suggesting that strong colorimetric responses might not necessarily indicate high sensitivity and stability of color changes in these fibers.

PEO–PDA and PU–PDA nanofiber composites and PDA in the absence of the matrix polymer were separately dispersed in buffer solutions where the pH varied in the range of 0–14. The suspensions were measured using ultraviolet–visible (UV–vis), and the absorbance was used to calculate colorimetric responses. The absorbance spectra of the samples immersed in the buffer solution with pH 0–9 were nearly overlapped, suggesting no color change in this range of pH. When the pH increased to 10, 11, 12, and 13, the color of the PEO–PDA and PU–PDA nanofiber suspensions changed from blue to purple and finally to red. The color shift was significant at pH 11–13. The changes were likely due to the high concentration of the hydroxide ion at pH 11–13, which had a significant impact on the chemical environment and hydrogen bonding of the PDAs.[18,34] Previous work suggested that pH changes in PDA can alter the hydrogen bonding of the carboxylic acid head groups, inducing a conformational change in the PDA backbone that leads to a color transition of the polymer.[18,35] In this case, changes in the chemical environment cause a reflectance shift because of shortening of the p-conjugated bonds in the backbone of the polymer.[36] In the blue phase, π electrons within the polymer backbone are delocalized.[37] As the hydrogen bonding is disrupted by external stimuli, the π electrons become distorted and the polymer reaches the red phase.[38]

The intensity of the color transition was clear at pH 11 when the PDA demonstrated a color of purple, given that the color transition was reversible and was not fully completed at pH 11. When the pH was further increased to 12, the PDA nanofibers in the solution became red. The color change was captured by the increase of the colorimetric response calculated by the absorbance at 640 nm (blue phase) and 540 nm (red phase). The colorimetric responses were plotted against the pH in the buffer solutions, and the plots are shown in Figure . It was found that the color transition from blue to red was promoted at pH 12 and 13. Then, the colorimetric responses slightly dropped at the extremely alkaline condition (pH = 14).

Figure 5

The colorimetric response of 100% PDA polymers, PEO–PDA, and PU–PDA nanofiber composites as a function of pH when immersed in buffer solutions.

Similar results were obtained using 100% PDAs in the absence of matrix polymers as the PDA fibers. No color change was seen in pH values ranging from 0 to 9. The colorimetric response began at pH 11. At high pH values (12–14), the PDA polymer followed a pattern of the colorimetric response similar to the PDA nanofiber composites.

In comparison between the PDA polymer and the PDA nanofiber composites, it is important to notice that the colorimetric responses of the PDA nanofiber composites are higher than those of the PDA polymer at pH 11. The colorimetric response is due to the self-assembled PDA macromolecular chains, also called “synthetic vesicles”.[33] Electrospinning significantly enhanced the formation of self-assemblies of the PDA macromolecules and then improved the organization of synthetic vesicles within the nanofibers. Therefore, the colorimetric response was enhanced in the PDA nanofiber composites. In addition, the high surface area of the electrospun nanofibers was able to increase the colorimetric response as well. The results suggested that the PDA nanofiber composites are good candidates for biosensor applications.

Conclusion

Two polymers, PEO and PU, were separately mixed with DA monomers, and the mixtures were used to create PEO–PDA and PU–PDA nanofiber composites using an electrospinning method. Smooth and uniform nanofibers with diameters ranging from 130 nm to 2.5 μm were obtained at high concentrations and mass ratios. The variation in the nanofiber morphology was attributed to the mixture solutions and spinning conditions in electrospinning. Although the PDAs are known to exhibit colorimetric properties because of external stimuli, to the best of our knowledge, we are the first to report that the PDA nanofiber composites were able to respond to the presence of E. coli and the change of pH in buffer solutions. The colorimetric response to E. coli was more pronounced for 6:1 PEO–PDA at 3.75 w/v % concentration than for 2:1 PEO–PDA at 2 w/v % concentration. The % CR to E. coli was more intense in 6:1 PU–PDA at 25 w/v % concentration than for 6:1 PU–PDA at 18 w/v % concentration. The results suggest that the increase in the fiber diameter and the surface area enhanced the colorimetric sensitivity to E. coli. Both the PEO–PDA and PU–PDA nanofiber composites demonstrated a sensitive colorimetric response because of the presence of E. coli. The comparison results show that the combination of PEO and PDA is favorable in a sensitive response to E. coli. ATR–FTIR analysis confirmed that the colorimetric response was not caused by a variation of the functional groups within the polymer. Such results are in accordance with previous reports that suggest that the color change was induced by a conformational change in the molecular configuration of PDA when PDA interacted with E. coli. The colorimetric responses of the PDA only, PEO–PDA, and PU–PDA nanofiber composites were also investigated in buffer solutions with pH 0–14. Similar responses were found that the colorimetric response did not occur until pH 10, and then increased at pH 11, 12, and 13, followed by a slight drop at pH 14. Our results demonstrate that the PDA-containing nanofiber composites can be used in developing sensitive, flexible, and lightweight biosensors that are easy to use, durable, and do not require a power supply. The potential applications include lab-on-chip sensors, wearable sensors, and medical sensing textiles such as wound dressing and bandages.

Experimental Section

10,12-PCDA (98%) was purchased from GFS Organics (Columbus, OH); the PCDA monomer was used to synthesize PDAs in electrospinning. PEO (Mw = 300 000 g/mol) was purchased from Sigma-Aldrich (St. Louis, MO). PU Tecoflex SG-80A was kindly donated by Lubrizol Corporation (Brecksville, OH). PEO and PU were used separately as matrix polymers in nanofiber composites. LB and chloroform (≥99.8%) were purchased from Sigma-Aldrich (St. Louis, MO). Tetrahydrofuran (THF, 99%), N,N-dimethylformamide (DMF, 99.8%, Extra Dry, AcroSeal), and potassium hydroxide (KOH) were purchased from Fisher Scientific (Waltham, MA). LB broth/agar was used as the bacterial growth media. Chloroform, THF, and DMF were solvents for the preparation of electrospinning solutions. Hydrochloric acid (HCl) was purchased from EMD chemicals (Gibbstown, NJ, USA). HCl and KOH were used to analyze the polymers at various pH values. E. coli (Migula) castellani, and chalmers ATCC25922 were purchased from ATCC (Manassas, VA). Reagents were used as received without further purification.

PDA Synthesis

The diacetylene monomer, PCDA, was used in the polymerization of PDAs. PDA polymerized from PCDA was synthesized following our previously reported procedure.[7] Briefly, 2.56 g of PCDA (6.83 mmol) was dissolved in diethyl ether (35 mL) and filtered to remove any contaminant. Millipore water (18.2 MΩ cm) was added to yield a 1.07 w/v % suspension, which was sonicated at 65 °C for 30 min. The suspension was allowed to cool to room temperature, and then stored at 4 °C overnight. The suspension was transferred to a crystallizing dish with a magnetic stir bar and irradiated with UV light (254 nm) for 8 min.[27] After the photo-polymerization, the dark blue suspension was transferred to a round bottom flask protected from light to remove the solvent under vacuum. The solid PDA was then stored at 4 °C and characterized by ATR–FTIR and UV–vis.

Preparation of Electrospinning Solution

For the preparation of PDA-containing nanofiber composites, PCDA was mixed with PEO or PU to prepare electrospinning solutions by varying concentrations (w/v %) and mass ratios of PEO to PCDA and PU to PCDA. The mixture solution of PEO and PCDA was prepared by adding the calculated amount of PEO and PCDA in chloroform, followed by stirring overnight at room temperature on a hotplate stirrer at 600 rpm until a homogeneous solution in light pink was obtained. In mixing PU and PCDA, a binary solvent of 1:1 THF and DMF was used to prepare uniform solutions. It was attributed to the effect of solvent properties on the viscosity and surface tension of the solution. DMF is a dipolar aprotic solvent and has a high dielectric constant (36.7 at 25 °C) and dipole moment (3.8 D) as compared to THF, which has a low dielectric constant (7.6 at 25 °C) and dipole moment (1.7 D).[28] A 1:1 by volume mixture of these two solvents achieved a balance of solution viscosity and conductivity, which was favorable for the formation of PU–PDA nanofibers via electrospinning. The mixture solution of PU and PCDA was prepared by dissolving the required amount of PU and PCDA in THF and stirring the mixture overnight at room temperature. The mixture was then added to the same volume of DMF and continuously stirred until a uniform solution in light pink was obtained. The synthetic reaction for PU–PDA is depicted in Figure S1.

Electrospinning and PDA Polymerization

A customized vertical electrospinning apparatus was used to prepare nanofiber composites. The apparatus primarily consisted of a Gamma High Voltage Research ES50P power supply, a plastic syringe, a stainless steel needle, a Harvard PHD 2000 syringe pump, and an aluminum plate type collector. A mixture solution of PEO and PCDA or PU and PCDA was injected at 0.2 mL/h and 15 kV, and the resulting nanofibers were collected at a collection distance of 25 cm. The time for electrospinning a solution was 4 h to obtain a thick and colorless fiber mat. The as-spun fibers were stored in the dark overnight to avoid any changes due to light before the photopolymerization of PCDAs. A UV light (Spectroline, LONGLIFE filter, New York, USA) at 254 nm was used to photopolymerize PCDAs for 3 min, resulting in PDAs embedded in PEO or PU nanofibers and creating PEO–PDA or PU–PDA nanofiber composites. After UV irradiation, the color of the nanofiber composites became blue within 30 s and then became deep blue within 3 min.

Scanning Electron Microscopy

The size and morphology of electrospun nanofibers were studied using an SEM (JEOL, JSM 6500F, Tokyo, Japan). The fiber mats were kept overnight under vacuum to evaporate any residual solvent or moisture, followed by sputter-coating with 10 nm of gold to improve the conductivity of the fiber samples.

FTIR Spectroscopy

ATR–FTIR spectra of PCDA, PEO–PDA nanofibers, and PU–PDA nanofibers were collected on a Nicolet 6700 FTIR spectrometer (Thermo Electron Corp., Madison, WI) fitted with a Smart iTR ATR sampling accessory and a ZnSe crystal plate.

Colorimetric Response To pH

The colorimetric response of PDAs to pH was investigated. The tested PDAs included 100% PDA powders (PPCDA synthesized in this study) and the PDA composites (PEO–PDA nanofibers obtained at a 2:1 mass ratio and 3.75% and PU–PDA nanofibers obtained at a 2:1 mass ratio and 25%). Buffer solutions were prepared by mixing HCl and KOH, resulting in a range of pH values from 0 to 14. The PDA polymer and the PEO–PDA and PU–PDA nanofiber composites were individually dispersed with 1 w/v % in the buffer solutions. All suspension solutions were stirred for 1 h to improve uniformity of the solutions. To obtain dry, solid materials, the solvent was removed using a lyophilizer, and the samples were analyzed using a Nicolet Evolution 300 UV–vis spectrophotometer in a diffuse reflectance mode. The colorimetric response (% CR) was calculated using eq as below.[14]where the initial percent blue (PB0) is the PB before the color change, and PB is the final PB after exposure to external stimuli. The PB value was calculated using eq where Ablue is the absorbance at 648 nm and Ared is the absorbance at 537 nm.

Bacterial Culture

E. coli (ATCC25922) was streaked on LB agar and was grown to saturation at 37 °C for 24 h. One colony was inoculated into LB broth medium and incubated at 37 °C with shaking at 200 rpm using MaxQ Shaker for 18 h. An aliquot (100 μL) of the overnight bacterial culture was spread on LB agar and incubated at 37 °C for 36 h. The E. coli culture was used to test the colorimetric responses of the obtained nanofiber composites.

Colorimetric Response To Bacteria

The colorimetric transition of the nanofiber composites upon direct exposure to E. coli was evaluated using a spectrophotometer (HunterLab ColorQuest XE) as a function of exposure time (0.5, 1.0, 1.5, 2.0, 2.5, and 3.0 h). Electrospun fiber mats were cut into 6.5 cm2 mats. Each 6.5 cm2 fiber mat was then placed flat in a clear Petri dish where E. coli was cultured, and the dish was closed and sealed tightly. This setup allowed bacteria to be in direct contact with the E. coli in the Petri dish. Because the fiber mat was flat on the bottom of the Petri dish and the Petri dish was clear, it was possible to conduct spectrophotometer measurements with the appropriate baseline. The method avoided cross-contamination and allowed the measurements to be taken in real time, without removing the fiber mat from the dish. Spectrophotometric measurements were then taken by placing the Petri dishes on the spectrophotometer measurement outlet while the fiber mats were still in the Petri dish. The colorimetric reflectance of the fiber mat was measured every 30 min for 3 h. The outside of the Petri dishes was wiped and cleaned with ethanol before each spectrophotometric measurement. For these measurements, three fiber samples obtained at each electrospinning condition were used in the spectrophotometer. At each time interval, three spectrophotometric measurements were conducted for the colorimetric analysis in each fiber mat. An average of the colorimetric reflectance of each fiber mat was calculated and used in the analysis of the colorimetric response. The experimental procedure was also followed for the controls, which included fiber samples in LB agar only, the supernatant, or the pellet. The LB agar was pristine and not used for previous bacterial growth. The supernatant and the pellet were obtained after the solution was centrifuged and filtered.

A reflectance value was obtained for all fiber mat samples using the ColorQuest spectrophotometer. Therefore, it was necessary to convert reflectance to absorbance, so that the colorimetric response could be calculated using eqs and 2. The absorbance of the blue and red phases can be calculated based on the reflectance values using eq as below.

The resulted absorbance value was used to calculate the colorimetric response to measure the efficiency of the color change of PDA in the fibers responding to E. coli. The reflectance of the blue phase in the spectrophotometer spectrum was measured at 640 nm, and the red reflectance was measured at 540 nm. These reflectance values were first converted into absorbance and then used to calculate the colorimetric response that represents the relative change from blue to red color of PDA before and after exposure to the bacterial cells.

In summary, Figure shows the experimental design for preparing electrospinning solutions, specified concentrations, mass ratios, and tests performed of PEO–PDA and PU–PDA fibers.

Figure 6

Experimental design chart of PDA electrospun fibers.