Microfluidic device on a nonwoven fabric: A potential biosensor for lactate detection

Abstract

In the present study, a novel, wearable textile based microfluidic device was developed that provides a non-invasive, rapid, semi-quantitative detection of the lactate level in simulated sweat solution. The potential application was envisioned to be a biosensor that can monitor an athlete’s physical status during exercise. A photolithography technique was used for the fabrication of hydrophilic micro channels and reservoirs surrounded by hydrophobic barriers made from SU-8 negative photoresist. The reservoirs were functionalized by co-immobilization of lactate oxidase (LOX) and horseradish peroxidase (POX) enzymes. LOX uses L-(+)-Lactic acid as substrate and produces H₂O₂ which is a POX substrate. Then, POX oxidases H₂O₂ in the presence of 2,2′-azino-bis(3-ethylbenzthiazoline-6-sulfonic acid) diammonium salt (ABTS) and results in color formation. The studies showed that excess amount of analyte presence resulted in analyte inhibition. It was also shown that analyte pH and temperature were effective on the color formation. For effective results, analyte pH and temperature should be ≥5℃ and 25–30℃, respectively. Lower pH and higher temperature values resulted in a decrease in the enzyme activity. The textile based biosensor system could make a semi-quantitative visual detection to differentiate between the normal (<5 mM) and high (≥5 mM) lactate level: while a high lactate level led to a denser purple color formation, normal levels led to a light purple formation and a green color started to be observed.

Keywords

biosensor lactate detection microfluidic device nonwoven simulated sweat photolithography

Quantitative measurements of metabolites, biomarkers and proteins are usually performed for accurate medical diagnosis. Detection technologies used must be lightweight, robust and easy-to-use in order to meet the requirements of point-of-care testing.¹ Therefore, laboratory tests were miniaturized and automated using lab-on-chip (LOC) devices which have been developed recently. LOC devices allow the construction of portable and disposable systems that use very little volume of analytes and reagents.²

Microfluidic technology applications are developing rapidly in the fields of biochemical and chemical analysis since they provide control and transmission of small quantities of liquid samples for rapid analysis.³ So far microfluidic devices for biological analysis were manufactured by molding or etching channels into glass, PDMS, silicone, plastics or other polymers.⁵ Nowadays, paper-based microfluidic diagnostic devices are also explored as suitable candidates for such applications.¹ Diagnostic devices made of patterned papers are known as microfluidic paper-based analytical devices (µPADs) combining the capabilities of traditional microfluidics with the simplicity of diagnostic strip tests.^5–8 They do not require any power to control the fluid flow thanks to their inherent capillarity. Fabrication of microfluidic paper-based analytical devices is achieved by patterning hydrophilic channels onto paper surfaces demarcated by hydrophobic barriers.⁴ Currently, development of paper-based devices has become an appealing technology and various methods have been developed for their fabrication such as photolithography,^4,7 polydimethylsiloxane (PDMS) plotting,⁸ inkjet printing,^9,10 cutting,¹¹ plasma etching,¹² wax printing^13–15 and wax screen-printing.¹⁶

Martinez et al. defined a method to fabricate well-defined, millimeter-sized channels placed on the hydrophilic paper surrounded by hydrophobic barriers. They used urine solution to detect glucose and protein simultaneously on a paper-based analytical device.⁶ Dungchai et al. utilized multiple indicators for a single analyte detection using µPAD to increase the visual distinction capacity of color change based on analyte concentrations. Semi-quantitative measurement of lactate, glucose and uric acid was conducted on a µPAD using serum and urine samples to control functionality of this device in real samples.¹⁷ Lu et al. reported a simple fabrication method of paper-based microfluidic devices using wax to comprise hydrophobic and hydrophilic regions on the paper for biological analysis. Glucose and bovine serum albumin (BSA) assays were carried out to demonstrate the applicability of patterned paper.¹⁸ Ellerbee et al. described a microfluidic paper-based analytical device including an optical colorimeter to determine the concentration of the analytes depending on the optical absorbance of color density.¹⁹ Klasner et al. reported a method for the fabrication of a paper-based microfluidic device made of a novel polymer blend that can be used for conducting the analysis of glucose, urinary ketones and salivary nitrite.²⁰

Apart from the paper-based analytical devices, fabric has been investigated as a suitable base material for fabrication of flexible and wearable microfluidic devices to perform bio-chemical analysis of body fluids.^21–26 In these studies, body fluids such as sweat, blood, etc., have been usually used for the monitoring of some chemical and physical parameters like pH, conductivity, sweat rate, etc., and detection has been performed visually via a signal or color change on the detection zones of the sensing devices.^27–30

Sweat is not only an easily accessible body fluid generated during physical activity but also a filtrate of blood plasma including various substances such as ammonia, sodium, potassium, chloride, calcium, bicarbonate, and also organic compounds, e.g. lactate and glucose.³¹ Content of sweat not only varies between individuals, but also varies according to the sweat rate, the degree of hydration, the type of exercise, and the state of heat acclimatization.²¹ Considering this fact, monitoring sweat content reveals rich and valuable information about the physiological condition of the individual.²⁹ Since collecting sweat for physical analysis involves some difficulties, a limited number of studies related with this issue have been conducted in the literature.²¹ In the Biotex EU-funded project, a wearable sensing device, integrating a fabric-based fluid handling platform for collection and analysis of sweat using optical and electrochemical methods, was described.^31,32 Coyle et al. designed a fabric-based fluid handling system comprising a polyester/lycra® blend wicking textile that could collect and analyze sweat effectively during exercise for real-time assessment. The proposed sensor was a wearable platform comprising of both a fluid-handling system and an optical component where colorimetric analysis was conducted through light emitted diodes placed in the textile.²² Curto et al. reported on the fabrication of a textile-based, wireless, real-time, sweat pH analysis system and the integration of miniaturized electronic components to the microfluidic platform.³³ Curto et al. presented a simple, flexible, disposable and wearable micro-fluidic platform for monitoring the pH of sweat occurring during physical activity through the help of iono-gels.²⁹ Schazmann et al. in their work proposed a sodium sensor belt containing ion selective electrodes which was developed for the real-time assessment of sweat sodium concentration having an application in the diagnosis of cystic fibrosis disease.³⁴

Structure-related limitations of current paper-based microfluidics—i.e. low efficiency of sample delivery within the microfluidic device and the anisotropy in rates of liquid spreading in the x,y-plane versus the z direction due to two-dimensional orientation of fibers in the x,y-plane of a sheet of paper, have been reported.^7,35–37 Furthermore, the limited distribution of fluid samples in the z-direction in paper poses limitation for the creation of more complex 3-D microfluidic devices for multiple assay processes.³⁷ Due to the current limitations of paper-based microfluidic devices, it is pointed out that there is a need for further research on the use of alternative materials, fabrication techniques and detection methods for fabrication of new types of microfluidic systems.^36,37

In the current study, a novel, disposable and wearable biochemical analytical device was designed and fabricated by patterning micro channels and reservoirs using SU-8 photoresist polymer through a photolithography technique on an inherently absorbant bicomponent Evolon® nonwoven fabric. To the best of our knowledge, this is the first study demonstrating the utilization of an Evolon® nonwoven fabric for a biosensing application, i.e. lactate detection. The nonwoven fabric platform was found to be advantageous over paper since a more efficient liquid sample delivery, i.e. wicking rate, was achieved. Moreover, the three-dimensional, isotropic structure of nonwoven fabric led to a complete penetration of channel forming SU-8 polymer through the entire thickness of Evolon® creating a well-demarcated hydrophobic–hydrophilic contrast of the pattern.

The proposed device, enabling a non-invasive, rapid semi-quantitative analysis of lactate level in simulated sweat solution, may be potentially used for the observation of an athlete’s physical condition during physical performance.

Materials and method

Materials

A 100 g/m² Evolon® nonwoven fabric was used as the basis for the microfluidic device which was made from split polyester/nylon microfibers. Freudenberg’s spunbonded and hydroentangled Evolon® fabric was chosen for its highly absorbant structure to collect and deliver the simulated sweat solution to the detection zones via microchannels. SU-8 3050, an epoxy based negative photoresist polymer, was used to form defined areas of hydrophilic reservoirs (detection zones) and hydrophilic channels demarcated by hydrophobic barriers on the fabric surface. SU-8 polymer was purchased from MicroChem Corporation (Newton, MA). A developer (1-methoxy-2-propanol acetate solution) was also purchased from MicroChem Corporation (Newton, MA). Iso-propyl alcohol 2-propanol was supplied from Zag Chemical Industry Research, Development and Calibration Laboratory. Lactate Oxidase (L0638 Sigma) (LOX) and peroxidase (P8375 Sigma) (POX) enzymes were used to functionalize the reservoirs for lactate detection. L-(+)-Lactic acid (L1750 Sigma), the lactate source, and 2,2′-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) diammonium salt (A1888 Sigma) (ABTS) were used as analytes for LOX and POX, respectively. For comparison, a filter paper, a base material commonly used for current paper-based microfluidics, was used during vertical wicking tests.

Method

Fabrication of microfluidic device on nonwoven substrate

A microfluidic device comprising micro channels linked to the reservoirs was fabricated using a photolithography technique for analyte (lactate in simulated sweat) detection. A microfluidic system pattern including micro channels and reservoirs was designed using Tanner Tools L-Edit 13.0 drawing program (Figure 1(a)). As shown in the figure, for each analysis, a three-reservoir system interconnected with each other via microchannels (800 µm) was used. The middle reservoir was designed for the introduction of the analyte and two flanking reservoirs were used for detection and internal control. The detection reservoir contained the enzymes and ABTS as coloring agent. In order to make sure that the color formation could only be seen as a result of enzymatic reaction in lactate presence, two different controls were used. In the first one, both enzymes were present but ABTS was not added, in the other one, ABTS was added but not the enzymes. Three different reservoir diameters, 4, 6 and 8 mm, were tested. A round shape was selected to construct the reservoirs in order to avoid the different wetting characteristics which could arise from the presence of sharp corners. For fabricating the pattern on a nonwoven substrate, a transparent photomask comprising the pattern (Figure 1(b)) was used during photolithography. The detailed procedure concerning the optimization of fabrication process was described elsewhere³⁸, but a brief description of the process and new design developed for this biosensor application is summarized below.

Figure 1.

(a) Designed model of the microfluidic system, (b) photomask used in photolithography.

Nonwoven fabric was spin coated using SU-8 as a negative photoresist to create hydrophilic channels and hydrophilic reservoirs demarcated by hydrophobic barriers on the fabric surface. SU-8 polymer was readily coated on the fabric due to the fabric’s inherent absorbant structure.

In the first step of the fabrication process, the back side of a 100 g/m² Evolon® nonwoven fabric taped on the silicon wafer, was cleaned with nitrogen gas (1--2 bar) and spin coated using SU-8 3050 negative photoresist polymer. Approximately, 3 ml of SU-8 3050 was poured on the fabric surface and spun at 3000 rpm for 30 s to disperse the photoresist polymer homogenously on the surface. Then, soft baking was performed at 95℃ for 25 min. After that, the fabric taped on the silicon wafer was kept on a cold surface for 1–2 min, then again for 1–2 min on a hot plate at 95℃. Similarly, in the second step, the front side of the fabric was coated with a single layer of SU-8 and baked at 95℃ for 25 min. It was kept on a cold surface for 1–2 min and then on a hot plate at 95℃ for 1–2 min.

Subsequently, the front side of the fabric coated with SU-8 was exposed to UV light for 8 s through a photomask. Similarly, the back side of the fabric was exposed to UV light for 8 s to obtain a complete hydrophobic, leak-proof surface. After this, fabric was kept on a hot plate at 65℃ for 1 min, which was followed by 5 min of post exposure bake at 95℃ to cross-link the exposed portions of the photoresist. Then, a wet etching process was performed during which the unpolymerized SU-8 was removed by submerging the fabric into a developer (1-methoxy-2-propanol acetate) solution for 1 h. This is followed by rinsing the fabric with isopropyl alcohol to clean the developer from the fabric surface. Later on, deionized water was used to clean the remaining alcohol on the fabric surface. After the wet etching process, nonwoven surface was hard baked at 65℃ for 2 h in a convection oven to further cross link the resist. The production steps are shown in Figure 2.

Figure 2.

Fabrication steps used in photolithography.

Schematic cross sectional view of reservoirs and micro channels obtained on the Evolon® nonwoven fabric and the actual patterned fabric are shown in Figure 3.

Figure 3.

(a) Cross sectional view of the fabric with reservoirs and micro channels; (b) pattern obtained on the nonwoven fabric surface (average diameter measured on eight different reservoirs is 6.03 ± 0.56 mm).

Testing and characterization of the microfluidic device

Leakproofing of the microfluidic device

Fluid absorption without leakage was a very critical property in the microfluidic device because it affected the reaction on the detection zones causing cross-reactivity. When analytes and reagents leaked along the reservoir boundaries of the device, reproducibility could not be achieved. Therefore, hydrophilic areas must be surrounded by hydrophobic barriers smoothly; in other words a good hydrophilic/hydrophobic contrast of the pattern is required. After the fabrication steps in photolithography, characterization of the obtained pattern on the fabric surface was carried out using a Zeiss EVO MA10 SEM (scanning electron microscope). SEM analysis was conducted to analyze the formation of hydrophilic channels and reservoirs and the penetration of photoresist material into the surface of textile material.

Vertical wicking

The capillary action wicking ability of Evolon® fabric was compared with that of filter paper. For this purpose, vertical wicking tests were performed according to BS 3424-18 standard method for determination of capillary action wicking, which also indicates the resistance of fabrics to capillary wicking.³⁹ For testing, four strip samples, each 150 mm × 50 mm, two with their length in the machine direction and two with their length in the cross direction, were cut. 5 mm of a strip sample was suspended in a colored solution which was prepared by dissolving a red ink (Monopol drawing ink) in deionized water. The colored solution was used for easier reading of the wicking height. Then the wicking height in millimetres to which the colored solution had risen above the bottom of the test sample was recorded every 5 min. Two measurements were performed in both the machine direction and the cross direction. The mean wicking height values for paper and Evolon® samples in both directions were calculated and plotted against time (min).

Fabric stiffness

The change in fabric hand such as rigidity resulted from SU-8 treatment was determined according to ASTM D4032 standard test method for stiffness of fabric by the circular bend procedure.⁴⁰ In this test, a plunger forces a flat, folded swatch of fabric, 102 × 102 mm, through a 1.50 in diameter orifice in a platform. The maximum force in N required to push the fabric through the orifice was recorded. The average of two individual sample readings was calculated. Fabric stiffness or resistance to bending was used as an indication of the wearability of the microfluidic device.

Bio functionalization of the microfluidic device

In order to functionalize the reservoirs, co-immobilization of lactate oxidase (LOX) and horseradish peroxidase (POX) was conducted. LOX uses L-(+)-Lactic acid as substrate and produces H₂O₂ which is then used as a substrate by POX. In order to indirectly follow the reaction, ABTS, a dye which is prone to oxidation by POX in the presence of H₂O₂, was added to the measurement chambers and color formation based on the reaction in Figure 4 was observed for detection.

Figure 4.

Enzymatic reactions yielding color formation on the textile surfaces.

Preparation of the enzyme and analyte solutions

LOX enzyme solution was prepared in a 10 mM phosphate buffer (pH 7.0, 37℃). Enzyme amount was set at 2 unit/ml. POX enzyme solution was prepared in a 0.1 mM PBS (phosphate buffered saline) buffer (pH 7.0, 37℃). The enzyme amount was set at 2000 unit/ml. The pH value of enzyme solution was set at 7, the value accepted as the best pH value for POX activity.⁴¹ The ABTS solution (9.1 mM, pH: 7.0⁴²) was prepared in a 0.1 mM PBS buffer solution. Different analyte solutions (1–200 mM) were prepared by using L-(+)-Lactic Acid at pH 6.5 in 0.1 mM PBS buffer.⁴³

Optimization studies for the analyte and ABTS

Firstly, optimization studies were performed in solution and a linear working range was determined accordingly. For this purpose, enzyme-ABTS solution was prepared using 450 µL ABTS, 20 µL LOX (0.04 units) and 30 µL POX (60 units). This solution (50 µL) was added into different tubes, and 1 µL of analyte solutions in different concentrations (1–200 mM) were added for analysis. Color densities depending on the analyte concentration were measured by using spectrophotometer (Biorad Model 3559 UV Microplate) at 404 nm after 10 min of analyte addition. Secondly, in order to determine the effect of ABTS concentration to color density, different ABTS solutions (0.1–50 mg ABTS/7.5 ml) were prepared and tested. Finally, different enzyme amounts were tested to obtain best color visualization.

After the optimization studies in solution, optimization of the same parameters was performed for the textile surfaces. For this, LOX, POX and ABTS (1.65 mg) were sequentially dispensed on the reservoirs in different ratios (LOX: POX, 0.001–0.10 (units/reservoir): 1–10 (units/reservoir)). When the solutions were completely absorbed by the textile, analyte solutions were directly added on to the surfaces and color formation was monitored.

In the final part, studies were repeated on the final design of the biosensor system, which consisted of three connected reservoirs, one for detection (DR), one for control (CR) and a last one in the middle, for the introduction of the analyte solution. In this case, the LOX/ POX ratio was increased by 10× to obtain a more efficient color formation in the detection reservoirs.

Effect of temperature on the enzyme activity

In order to determine the effect of temperature on color formation, different temperatures (25–40℃) were tested. For this, analyte solutions were kept in an oven (Nuve EN400) at least for 30 min until they reached the specified temperature values. Then, they were dispensed onto the analyte reservoirs on textiles and kept in an oven set at the same temperature. Photos were taken at different time intervals to monitor the color formation (camera: Canon EOS 50D, Japan).

Effect of pH on the enzyme activity

To study the effect of pH on color formation, analyte solutions with different pH (pH: 4, 5 and 6) values corresponding the pH range of sweat were prepared. Analytes with different pH values were dispensed on the detection reservoirs at room temperature and photos were taken at different times to examine the color formation.

Shelf-life studies

Shelf-life of the enzyme immobilized textile surfaces was evaluated by storing enzyme immobilized textiles at + 4℃. Measurements were made after different storage durations (1–25 days) and color formation on reservoirs was photographed.

Results and discussions

Testing and characterization of the microfluidic device

Leakproofing of the microfluidic device

Fluid absorption both on the reservoirs and channels and their leakproofing were observed using a red-colored solution obtained by dissolving a red ink (Monopol drawing ink) in deionized water. In total, 30 µL of this colored solution was dispended on the central reservoirs using a micropipette. Then, in each line, the fluid transmission via micro channels to the other reservoirs located on the left- and right-hand side of the triple reservoir system was monitored. No liquid leakage was observed around the reservoirs and the channels and both the fluid flow and the color distribution seemed to be homogeneous (Figure 5).

Figure 5.

Top view of the channels with a width of 800 µm linked to the reservoirs with a diameter of 6 mm and liquid absorption on the reservoirs and channels during monitoring of system’s hydrophilicity and leakproofing.

Surface characterization of the proposed device was then conducted on the SEM (Zeiss EVO MA10 SEM) device to monitor the penetration of photoresist material into the surface of the nonwoven. Surface coating must be very smooth and hydrophobic barriers should surround the hydrophilic zones perfectly for a better performance. Effective detection of the color change after the introduction of the analyte depends on the interaction between the enzyme and simulated sweat. At this stage, leakproofing plays an important role and it must be ensured for successful and reproducible results. In SEM images, it was observed that fibers on the fabric surface were covered completely by the SU-8 negative photoresist. The top view of the reservoirs fabricated on the fabric surface showing boundaries of hydrophobic/hydrophilic regions are shown in Figure 6(a). The full penetration of the channel forming SU-8 polymer through the plane of the Evolon® fabric was also achieved, which contributed to the well-defined pattern in the fabric (Figure 6(b)). This is required to avoid the escaping of the aqueous fluid through the channels and spreading through the device.⁷ This requirement was pointed out as a constraint for patterning paper because liquids tend to spread in the x,y-plane of paper rather than in the z-direction due to the orientation of intertwined fibers in x,y-plane of a sheet of paper, leading to blurring of the patterns.⁷ In case of Evolon®, the hydroentaglement process renders a more isotropic, three-dimensional structure which allows the penetration of liquid both in the plane and through the plane of fabric.

Figure 6.

SEM image showing (a) the top view of hydrophilic (fibers) and hydrophobic zones on the fabric surface after the etching process; (b) the cross sectional view of hydrophilic and hydrophobic zones of the fabricated Evolon® fabric.

Vertical wicking

For the Evolon® fabric sample, the colored solution reached almost 80 mm in height within the first 5 min and a maximum sample height of 140 mm within 20 min in both machine and cross directions (Figure 7). On the other hand, for paper, within the first 5 min the solution reached 43 mm in height in the machine direction and 32 mm in height in the cross direction. For paper, wicking of the liquid stopped at a height of 118 mm in the machine direction and 100 mm in the cross direction which took almost 45 min (Figure 8). The wicking test results showed that the Evolon® fabric has a considerably higher wicking rate compared to filter paper to wick the fluid by capillary action. The high surface-to-volume ratio of microfibers of the Evolon® fabric contributed to its high wicking ability.

Figure 7.

Comparison of wicking rates of Evolon® fabric and paper in machine direction.

Figure 8.

Comparison of wicking rates of Evolon® fabric and paper in cross direction.

The Evolon® fabric was found to be advantageous over paper in collecting and delivering the sample fluid to the detection zone, which is necessary for the rapid detection of analytes. The wicking rate of the material used for the microfluidic diagnostic device is very critical in the sense that the sample fluid should wick along the channels on the material as quickly as possible to minimize the sample evaporation during transport, which may lead to ineffective sample delivery within the microfluidic device.^35,36 Also if the distance from the introduction point of the analyte sample to the detection zone is considerable, then during solution spreading and wicking, the local analyte concentration may decrease limiting the sensitivity of the device.³⁷ The results also indicated that Evolon® may be preferred especially when the microfluidic pattern requires the sample to be introduced onto the material at a considerable distance away from the detection zone.

Fabric stiffness

Fabric stiffness was measured to determine if SU-8 treated Evolon® fabric could be used as a wearable device. The force value related to Evolon® fabric stiffness was measured as 85.05 ± 19.1 N for the SU-8 treated and 3.80 ± 0.50 N for the untreated sample. For comparison, the stiffness value was reported as 6.2 ± 1.5 N for a 153 g/m² shirting fabric according to the same test standard. The considerably low stiffness value obtained for the untreated Evolon® fabric may be attributed to the soft hand provided by the microfibrous fabric structure. SU-8 treatment evidently resulted in an increase in the bending resistance of fabric or stiffness which is inevitable considering the absorbent structure of Evolon® fabric provided by the high surface-to-volume ratio of microfibers. The measured stiffness of the SU-8 treated Evolon® fabric was comparable with the approximate stiffness value of 70 ± 16.0 N measured by the same test procedure for a standard denim fabric with a basis weight of 492 g/m².⁴⁰ It was reported that coating treatment on a fabric results in stiffening by penetration of coating resin into the fabric structure.⁴⁴ Therefore, it may be anticipated that a coating treatment would cause an increase in the stiffness value of aforementioned denim fabric, bringing its stiffness value closer to that of SU-8 treated Evolon® fabric. This indicated that despite the increase in the fabric stiffness after SU-8 treatment, the device was still wearable.

Biological activity tests

Optimization studies for the analyte and ABTS

Measurements made with the spectrophotometer showed that excess amount of analyte presence in tubes resulted in substrate (analyte) inhibition (Figure 9). The color density increased linearly in the presence of 1–20 mM analyte amount. It was constant between 20–40 mM and decreased right after 40 mM because of the inhibition. Therefore, it was found that measurements should be done in between 1–20 mM. Lactate concentration in the sweat ranges within these concentrations,⁴⁵ so that the detection could be performed without any problem in this range.

Figure 9.

Detection of enzymatic activity at different analyte concentrations.

By keeping the analyte concentration constant (20 mM), ABTS concentrations were varied for ABTS optimization studies. The most dense color density was obtained for 5 mg ABTS/7.5 ml and a decrease started right after that point. Change in the color density was also checked by using the three best ABTS concentrations (0.5–5–10 mg/7.5 ml; i.e. 0.067–0.67–1.33 mg/mL) for different analyte concentrations. Color density obtained, depending on the lactate level, was nearly 2× higher using 5 mg/7.5 ml.

After the optimization studies in solution, optimized parameters were used on the textile surfaces. Three different reservoir diameters were tested for their efficiency in analyte detection. The selection criteria were based on the design of a system which would allow the visualization of the color change by naked eyes with minimum enzyme usage. The results showed that 4 mm reservoirs were not suitable for the differentiation of the color change with naked eyes and 8 mm reservoirs require more enzyme and coloring agent usage, therefore 6 mm diameter reservoirs were chosen to be used for further studies. However, it was shown that color intensity was differently observed in solution and on textile fabric. Although different color densities depending on the lactate level was obtained in the solution (Figure 10(a)), no color formation was observed on the fabric surfaces (Figure 10(b)) when ABTS and enzymes were introduced together on the surface. A hollow green ring (Figure 8(c)) instead of a complete coloration was formed when each of them were introduced to the surface sequentially (Figure 10(c)). This was ascribed to the introduction method of the enzyme, ABTS and analyte to the same point of the reservoirs (center). A liquid flow occurred from the center towards the periphery and it was deduced that this fluid flow might have carried the molecules off the center. An enzyme ratio of LOX : POX, 0.006 : 6 (units/reservoir) was determined as the optimum ratio in order to obtain a high color density in the reservoirs. Two different controls were tested as described in under the section “Fabrication of microfluidic device on nonwoven substrate”, above, and no color formation was detected on the control groups, indicating the specificity of the color formation on lactate presence.

Figure 10.

Color formation as a result of analyte (lactate) introduction when enzymes and ABTS were; (a) mixed in solution, (b) mixed and dispensed onto textile surface and (c) separately dispensed onto the textile surface. (K = control (ABTS only); analyte is in mM.).

Color formation on the fabricated reservoirs on the fabric depending on the lactate concentration is shown in Figure 11. When a wider concentration range was scanned, a decrease in the color density was observed in concentrations above 40 mM because of analyte inhibition (Figure 11(a)). However, since the lactate level in sweat can have a maximum value at around 10–13 mM,⁴⁵ measurements made at high lactate levels of over this value were not physiologically relevant for a biosensor system that determines the lactate level in sweat. When a narrower concentration range of lactate level was scanned between 1–8 mM, two different color scales were obtained: color density was low in 1–3 mM and high in 4–8 mM lactate (Figure 11(b)).

Figure 11.

Color formation when lactate level was between (a) 1–100 mM and (b) 1–8 mM. (K = control (ABTS only).)

In the final design system which consisted of three connected reservoirs, a new color, purple, appeared with the increased analyte amount (Figure 12(a)). Formation of purple color could be attributed to the increase of LOX/POX ratio by a factor of 10. LOX enzyme (one unit = 1.0 µmole of L-lactate to pyruvate and H₂O₂ per min at pH 6.5 at 37℃) being an enzyme with lower specific activity compared to POX (one unit = 1.0 mg purpurogallin from pyrogallol in 20 s at pH 6.0 at 20℃), could be considered as the rate limiting enzyme in the coupled enzymatic reaction. When the LOX amount was increased, the H₂O₂ formation rate was increased and oxidation of higher amounts of ABTS in a short period of time contributed to the purple color formation. When H₂O₂ was directly added to the surfaces, purple color could be observed even in low concentrations (data not shown) and this also suggested the effect of H₂O₂ formation rate on color formation. ABTS was known to have two oxidation states corresponding to the oxidation of ABTS to its cation radical ABTS^•+ and subsequently to its dication ABTS²⁺. Color formation was found to be different depending on both concentration of oxidized ABTS and affinity of the reaction products (obtained from the ABTS enzymatic oxidation) for different textile or polymers and green, blue and purple colors have been reported.^46,47 In our case, different behaviors were obtained based on the enzyme ratio and analyte amount introduced to the system. It could be seen that purple color started to appear when the lactate level reached 5 mM and became dominant until 10 mM. Right after 10 mM, a decrease in purple color density was also observed, probably because of analyte inhibition. Denser green color formation on surfaces was obtained when the lactate level was <5 mM (Figure 12(b)). The experiments were repeated at different temperature and pH values as described in the following sections (n > 10), and it was seen that purple color formation could only be observed as a dominant color beginning from 5 mM lactate concentration. Only in two of the experiments, unexpected purple color formation was observed, one in the control reservoir and one in the 3 mM lactate applied reservoir. In 9 out of 10 cases, however, the color change from green to purple occurred at a specific concentration, namely after 5 mM lactate, so it could be said that acceptable repeatability was obtained in the distinction of normal and high level of lactate in sweat. Therefore, it can be concluded that when the lactate level was <5 mM green color formed, but between 5–10 mM purple color formation started to be observed and become dominant. In the control reservoir of the textiles (right-hand side, CR) no color formation was observed.

Figure 12.

The change in color formation based on lactate concentration. (a) Purple color become dominant when the lactate level was between 5–10 mM and (b) green color was dominant when the lactate level was <5 mM. (MR: Measurement reservoir (LOX/POX/ABTS); CR: control reservoir (ABTS only); I: Reservoir for the introduction of analyte solution.)

Effect of temperature on the enzyme activity

In order to determine the effect of temperature on the enzyme activity and color formation, analysis with 10 mM analyte solution was performed at different temperatures. Results showed that color formation was more intense when the temperature was in the range of 25–35℃ (Figure 13). Temperatures over 30℃ seems to result in a decrease in color formation. This decrease could be associated with the rapid evaporation of the solution from the open surface of the measurement chambers used for optimization studies. The final design would include a protective barrier on the top which would allow for more effective measurement.

Figure 13.

Effect of the temperature on the color visualization. (MR: Measurement reservoir (LOX/POX/ABTS); CR: control reservoir (ABTS only); I: Reservoir for the introduction of analyte solution.)

Effect of pH on the enzyme activity

Effect of pH on the enzyme activity was also determined using 10 mM analyte solution with different pH values. It was found that enzyme activity decreased with decreased pH values, and below pH 5.0 no purple color formation was seen. The optimum pH value was found to be 6.5 at 25℃ (Figure 14). It is known that the pH value of sweat could drop to 5.0,⁴⁸ therefore it could be said that lactate level in sweat could be determined using this biosensor system even after the pH drops to 5.0 during the exercise (Figure 14).

Figure 14.

Effect of the analyte pH on color formation. Purple color formation on the surfaces 10 min after the analyte addition. (MR: Measurement reservoir (LOX/POX/ABTS); CR: control reservoir (ABTS only); I: Reservoir for the introduction of analyte solution.)

Shelf-life studies

Finally, shelf-life of the enzyme immobilized fabric surfaces was studied. It was found that color formation on the fabric surfaces was less intense, but more homogeneous after storing the enzyme immobilized surfaces at +4℃ and they kept their activity for at least 25 days. This change in color formation probably resulted from a more homogeneous distribution of the enzymes on the reservoirs and more stable bonding of enzymes to the fabric. Due to the reversible nature of adsorption, the enzymes could be mobilized as a result of the fluid movement during the analyte introduction process. During the storage, however, enzymes might diffuse through the fibers and form more stable interactions so they were not affected as much by the analyte introduction as before and a more homogeneous color formation occured.

Conclusion

In this work a textile based microfluidic device, comprising micro channels linked to the reservoirs, was successfully fabricated using a photolithography technique. This wearable and disposable biochemical analytical device was designed to detect lactate in sweat with a potential application of monitoring an athlete’s physical status during exercise. Highly absorbent nonwoven fabric provided the capillary penetration of liquid through the fabricated micro channels to the detection zones without any extra power system. Hydrophilic zones comprising the micro channels and reservoirs were surrounded by hydrophobic barriers made from SU-8 negative photoresist polymer. Hydrophilicity and leakproofing of the channels and reservoirs were achieved. Since an effective detection of color change after the introduction of the analyte depends on the proper interaction between enzyme and analyte, leakproofing was found to be important for the functionality of the system and reproducibility.

Lactate analysis on textile based microfluidic device showed that enzyme behavior was different in the solution and on the textile surfaces. Therefore, two different optimization studies were conducted for each case. However, both optimization studies revealed that an excess amount of analyte presence resulted in analyte inhibition. Moreover, it was shown that analyte, pH and temperature also affected color formation. For effective results, pH and temperature should be ≥5℃ and 25–30℃, respectively. Lower pH and higher temperature values resulted in a decrease in enzyme activity. Although the pH measurement range is suitable for sweat analysis, the temperature range of the biosensor should be extended to higher values (37–40℃) to determine the lactate level in sweat. The textile based biosensor system could make a semi-quantitative visual detection to differentiate between the normal (<5 mM) and high (≥5 mM) lactate level: while high lactate level led to a denser purple color formation, normal levels led to a light purple formation and a green color started to be observed.

The wicking test results showed that Evolon® fabric has considerably higher wicking rate compared to paper. Moreover, complete penetration of SU-8 polymer through the thickness of Evolon® fabric rendered well-defined hydrophobic barriers of photoresist. The three-dimensional orientation of entangled fibers in Evolon® fabric caused isotropy in both lateral and vertical fluid flow leading to well-defined hydrophilic–hydrophobic contrast of the pattern and efficient sample delivery within the device. Besides, this nonwoven platform may be further explored for the fabrication of 3D microfluidic devices requiring controllable vertical fluid flow for multiple assay processes.

Finally, this study showed that nonwoven fabrics, produced cost-effectively and at high volumes, may be a choice of material for disposable analytical devices to carry out rapid biochemical detection of various analytes.

Footnotes

Funding

This work was supported by the Scientific and Technological Research Council of Turkey (TUBITAK) (grant number 111M483).

References

Zhang

Zha

. Fabrication of paper-based microfluidic device using printed circuit technology. AIP Adv 2012; 2: 022171(1–6)–022171(1–6).

Songjaroen

Dungchai

Chailapakul

. Novel, simple and low-cost alternative method for fabrication of paper-based microfluidics by wax dipping. Talanta 2011; 85: 2587–2593.

Tian

Shen

. Progress in patterned paper sizing for fabrication of paper-based microfluidic sensors. Cellulose 2010; 17: 649–659.

Martinez

Phillips

Whitesides

. Diagnostics for the developing world: Microfluidic paper-based analytical devices. Anal Chem 2010; 82: 3–10.

Martinez

Phillips

Carrilho

. Simple telemedicine for developing regions: Camera phones and paper-based microfluidic devices for real-time, off-site diagnosis. Anal Chem 2008; 80: 3699–3707.

Martinez

Phillips

Butte

. Patterned paper as a platform for inexpensive, low-volume, portable bioassays. Chemie International Edition 2007; 46: 1318–1320.

Martinez

Phillips

Wiley

. FLASH: A rapid method for prototyping paper-based microfluidic devices. Lab on Chip 2008; 8: 2146–2150.

Bruzewicz

Reches

Whitesides

. Low-cost printing of poly(dimethylsiloxane) barriers to define microchannels in paper. Anal Chem 2008; 80: 3387–3392.

Abe

Suzuki

Citterio

. Inkjet-printed microfluidic multianalyte chemical sensing paper. Anal Chem 2008; 80: 6928–6934.

10.

Tian

Garnier

. Fabrication of paper-based microfluidic sensors by printing. Colloid Surface B 2010; 76(2): 564–570.

11.

Fenton

Mascarenas

López

. Multiplex lateral-flow test strips fabricated by two-dimensional shaping. Appl ACS Mat Int 2009; 1(1): 124–129.

12.

Tian

Nguyen

. Paper-based microfluidic devices by plasma treatment. Anal Chem 2008; 80: 9131–9134.

13.

Shi

Jiang

. Rapid prototyping of paper-based microfluidics with wax for low-cost, portable bioassay. Electrophoresis 2009; 30: 1497–1500.

14.

Shi

Qin

. Fabrication and characterization of paper-based microfluidics prepared in nitrocellulose membrane by wax printing. Anal Chem 2009; 82: 329–335.

15.

Carrilho

Martinez

Whitesides

. Understanding wax printing: A simple micropatterning process for paper-based microfluidics. Anal Chem 2009; 81: 7091–7095.

16.

Dungchai

Chailapakul

Henry

. A low-cost, simple, and rapid fabrication method for paper-based microfluidics using wax screen-printing. Analyst 2011; 136: 77–82.

17.

Dungchai

Chailapakul

Henry

. Use of multiple colorimetric indicators for paper-based microfluidic devices. Anal Chim Acta 2010; 674: 227–233.

18.

Shi

Jiang

. Rapid prototyping of paper-based microfluidics with wax for low-cost, portable bioassay. Electrophoresis 2009; 30: 1497–1500.

19.

Ellerbee

Phillips

Siegel

. Quantifying colorimetric assays in paper-based microfluidic devices by measuring the transmission of light through paper. Anal Chem 2009; 81: 8447–8452.

20.

Klasner

Price

Hoeman

. Paper-based microfluidic devices for analysis of clinically relevant analytes present in urine and saliva. Anal Bioanal Chemistry 2010; 397: 1821–1829.

21.

Benito

F-Lopez

Coyle

Byrne

. Sweat-on-a-Chip: Analysing sweat in real time with disposable micro-devices. In: Proceedings of Sensors 2010 IEEE 1–4 November 2010, pp. 160–163Waikoloa, HI.

22.

Coyle

Lau

Moyna

O'Gorman

Diamond

. Biotex-biosensing textiles for personalised healthcare management. IEEE T Inf Technol B 2010; 14(2): 364–370.

23.

Catrysse

Puers

Hertleer

. Towards the integration of textile sensors in a wireless monitoring suit. Sensor Actuat A—Phys 2004; 114: 302–311.

24.

Pandian

Mohanavelu

Safeer

. Smart vest: Wearable multi-parameter remote physiological monitoring system. Med Eng Phys 2008; 30: 466–477.

25.

Axisa

Schmitt

Gehin

. Flexible technologies and smart clothing for citizen medicine, home healthcare, and disease prevention. IEEE T Inf Technol B 2005; 9(3): 325–336.

26.

Coyle

Lau

K-T

. Fabric-based fluid handling platform with integrated analytical capability. In: Proceedings of the 29th annual international conference of the IEEE EMBS, Lyon, France, 23–26 August 2007, pp. 6451–6451.

27.

Diamond

Coyle

Scarmagnani

. Wireless sensor networks and chemo-/biosensing. Chemical Reviews 2008; 108: 652–679.

28.

Morris

Coyle

. Biosensing textile based patch with integrated optical detection system for sweat monitoring. Sensor Actuat B—Chem 2009; 139: 231–236.

29.

Curto

Fay

Coyle

. Real-time sweat pH monitoring based on a wearable chemical barcode micro-fluidic platform incorporating ionic liquids. Sensor Actuat B—Chem 2012; 171–172: 1327–1334.

30.

Coyle

Benito-Lopez

Radu

. Fibers and fabrics for chemical and biological sensing. Res J Text Apparel 2012; 14(4): 64–72.

31.

Coyle

Lau

. Design of a wearable sensing platform for sweat analysis. In: Proceedings of the pHealth2007, Chalkidiki, Greece, 20–22 June 2007 Available at: http://www4.dcu.ie/chemistry/asg/coyleshi/files/pHealth2007_Coyle.pdf.

32.

Morris

Schazmann

. Wearable technology for the real time analysis of sweat during exercise. In: Proceedings of Isabel: 2008 first international symposium on applied sciences in biomedical and commmunication technologies Aalborg, Denmark. Oct 25-28, 2008, pp. 109–110.

33.

Curto

Coyle

Byrne

. Concept and development of an autonomous wearable micro-fluidic platform for real time Ph sweat analysis. Sensor Actuat B—Chem 2012; 175: 263–270.

34.

Schazmann

Morris

Slater

. A wearable electrochemical sensor for the real-time measurement of sweat sodium concentration. Anal Meth 2010; 2: 342–348.

35.

Martinez

. Microfluidic paper-based analytical devices: From POCKET to paper-based ELISA. Bioanalysis 2011; 3(23): 2589–2592.

36.

Ballerini

Shen

. A perspective on paper-based microfluidics: Current status and future trends. Biomicrofluidics 2012; 6: 011301-1–011301-13.

37.

Liana

Raguse

Gooding

. Recent advances in paper-based sensors. Sensors 2012; 12: 11505–11526.

38.

Baysal

Kök

Trabzon

. Microfluidic nonwoven-based device as a potential biosensor for sweat analysis. Appl Mech Mat 2014; 490–491: 274–279.

39.

BS 3424-18–1986. Testing coated fabrics—Part 18: Method 21A: Determination of capillary action wicking.

40.

ASTM D4032. Standard test method for stiffness of fabric by the circular bend procedure.

41.

Sigma-Aldrich. Peroxidase Enzymes http://www.sigmaaldrich.com/life-science/metabolomics/enzyme-explorer/analytical-enzymes/peroxidase-enzymes.html. Accessed: 18 March 2014.

42.

Sigma-Aldrich. 2,2’-AZINO-BIS(3-ETHYLBENZTHIAZOLINE-6-SULFONIC ACID), DIAMMONIUM SALT http://www.sigmaaldrich.com/content/dam/sigma-aldrich/docs/Sigma/Product_Information_Sheet/a1888pis.pdf. Accessed: 18 March 2014.

43.

Sigma-Aldrich. Enzymatic assay of lactate oxidase. http://www.sigmaaldrich.com/etc/medialib/docs/Sigma/General_Information/2/lactate_oxidase.Par.0001.File.tmp/lactate_oxidase.pdf.

44.

Fung

. Coated and Laminated Textiles, Cambridge: Woodhead Publishing Limited, 2002.

45.

Akbaş

Pelvan

Ateş

. 100 m koşusu sonrası sporcuların kan laktat seviyeleri. Uluslararası Hakemli Akademik Sosyal Bilimler Dergisi 2011; 1: 101–107.

46.

Munteanu

F-D

Basto

Gubitz

. Staining of wool using the reaction products of ABTS oxidation by laccase: Synergetic effects of ultrasound and cyclic voltammetry. Ultrason Sonochem 2007; 14: 363–367.

47.

Cloete

Adriaanse

Swart

. Facile immobilization of enzymes on electrospun poly(styrene-alt-maleic anhydride) nanofibres. Polym Chem 2011; 2: 1479–1481.

48.

Lopez

Coyle

Byrne

. Simple barcode system based on inonogels for real time pH-sweat monitoring. In: Proceedings of the international conference on body sensor networks, Singapore. 7-9 June, 2010, pp. 291–296.