Micro-fluidic mixer and method of determining pathogen inactivation via antimicrobial solutions

Abstract

A sample of produce wash water containing an antimicrobial sanitizer fluid, and a reference pathogen fluid are both injected into a pathogen inactivation region of a micro-fluidic mixer. The produce wash water (i.e. sanitizer fluid/pathogen fluid mix) is directed through mixer elements in the pathogen inactivation region of the micro-fluidic mixer. In the sanitizer deactivation region, a sanitizer deactivation solution is added to the sanitizer fluid/pathogen fluid mix to produce a deactivated solution. The deactivated solution is evaluated for the presence of the pathogen and the characteristics of the sanitizer. In the preferred embodiment, the sanitizer comprises chlorine and the pathogen comprises E. coli bacteria.

Claims

1. A micro-fluidic sanitizer analysis system comprising: a pathogen inactivation channel comprising: (a) a first Y-injection mixer comprising a first and second inlets; and, (b) at least first and second S-shaped serpentine pathogen inactivation channels having chaotic mixer elements and being connected in series, a first end of the first S-shaped serpentine channel being connected to the first Y-injection mixer; and, a sanitizer deactivation channel operatively connected to the pathogen inactivation channel, the sanitizer deactivation channel comprising: (c) a second Y-injection mixer comprising third and fourth inlets, a second end of the pathogen inactivation serpentine channels comprising the third inlet: and, (d) at least third and fourth sanitizer deactivation S-shaped serpentine channels having chaotic mixer elements and being connected in series, a first end of the sanitizer deactivation S-shaped serpentine channels being connected to the second Y-injection mixer, and a second end of the sanitizer deactivation S-shaped serpentine channels comprising a sanitizer deactivation channel outlet; wherein the micro-fluidic system is configured so that after a produce wash sample is injected into the first or second inlet, the produce wash sample is mixed with a reference bacterial solution component in the first Y-injection mixer, and then directed to the third inlet, and then mixed with a sanitizer deactivating solution component in the second Y-injection mixer, and then the resulting deactivated sanitizer solution is directed out of the sanitizer deactivation channel outlet, the resulting deactivated sanitizer solution then being analyzed.

2. The micro-fluidic system of claim 1 wherein the system is structured so that: the first injection mixer, the first and second S-shaped serpentine pathogen inactivation channels, second Y-injection mixer, and third and fourth sanitizer deactivation S-shaped serpentine channels are all in fluid communication and are connected in series.

3. The micro-fluidic system of claim 1 wherein the system is structured so that the pathogen inactivation channel comprises at least one Dean's mixer.

4. The micro-fluidic system of claim 1 wherein the produce wash sample comprises a chlorine produce wash solution.

5. The micro-fluidic system of claim 1 wherein the reference bacterial solution component comprises E. coli.

6. The micro-fluidic system of claim 1 wherein the system comprises a continuous fluid channel, the pathogen inactivation channel having the same structure as the structure of the sanitizer deactivation channel.

7. The micro-fluidic system of claim 1 wherein the system is structured so that the pathogen inactivation channel comprises at least one Dean's mixer.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) This patent or application file contains at least one drawing executed in color. Copies of this patent or patent application publication with color drawing(s) will be provided by the Office upon request and payment of the necessary fee.

(2) FIG. 1 is an elevational schematic of the micro-fluidic mixer described herein.

(3) FIG. 2 is a (black and white) photograph of the micro-fluidic mixer described herein.

(4) FIG. 3 shows the chlorine concentration and bacterial suspension during a test period.

(5) FIG. 4 is a schematic view of the computational fluid dynamics modeled mixing of bacteria, chlorine, and neutralizing solutions inside the micro-fluidic mixer (described herein) at a flow rate equivalent to 0.75 contact time.

(6) FIG. 5 is a graph showing the correlation between volumetric flow rate and contact time.

(7) FIG. 6 is a graph showing the effects of chlorine concentration and contact time on log reductions of E. coli.

(8) FIG. 7 is a graphic comparison of experimental test results to three selected bacterial inactivation models.

(9) FIG. 8 is a contour graph of log reductions (1 to 5 log) of E. coli at different chlorine concentrations and contact times.

(10) FIG. 9 is a decimal reduction time (D-value) required at certain free chlorine concentrations to achieve 90% (1-log reduction) of E. coli in water.

DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS

(11) As generally shown in FIG. 1, the current invention is directed to a micro-fluidic mixer 10. The micro-fluidic mixer 10 comprises two mixing regions 20, 30. The first region generally comprises a bacterial pathogen inactivation region 20, and the second region comprises a dechlorination region 30.

(12) As shown in FIG. 1, the pathogen inactivation region 20 is designed to mix a sample of produce wash water with a reference bacterial source. The combined wash water and bacterial solution leaving the pathogen inactivation region 20 is directed into the dechlorination region 30. In dechlorination region 20, a chlorine inactivating solution (i.e. a dechlorinating solution) is added to the fluid mix to inactivate the chlorine in the solution. From the dechlorination region 30, the effluent flows out an outlet 40. The effluent leaving the outlet 40 is examined to determine the quantity of surviving bacteria cells in the effluent.

(13) Specifically, as shown in FIG. 1, the pathogen inactivation region 20 comprises a first inlet 21 and a second inlet 22. A bacterial reference sample solution is injected into the first inlet 21. In the preferred embodiment, the bacterial sample solution comprises Escherichia coli (i.e. E. coli) O157:H7 (RM4406, ATCC 700728), but other types of bacteria or contaminant should be considered within the scope of the invention. A sample of produce wash water (or other studied solutions) is injected in the second inlet 22. The dashed-line arrows shown in FIG. 1 indicate the direction of fluid flow.

(14) The bacterial fluid and the wash water converge at a Y injection-mixer 24. The fluid is then directed though a Dean's vortex-mixer 26 configuration to thoroughly mix the wash water and bacterial fluids. The Dean's vortex mixer 26 includes chaotic mixer elements 28 that extend into the mixer's 28 main channels and create turbulence that further mixes the fluids.

(15) As the fluid leaves pathogen inactivation region 20 and enters the dechlorination region 30, the fluid flows into a second y-injection mixer 31. In the (second) y-injection mixer 31 the fluid is mixed with a dechlorination fluid entering from a dechlorination fluid inlet 32. The dechlorination fluid is mixed with the bacteria/wash water mix in a (second) Deans vortex 34 which also includes a (second) set of chaotic mixer elements 36. The (now) dechlorinated fluid flows out the outlet 40 where any surviving bacteria are analyzed.

(16) FIG. 2 shows a picture of the micro-fluidic mixer 10, as fabricated. The device was fabricated and assembled using modified microfabrication methods that are well known in the art. The master mold was fabricated on a silicon wafer with SU-8 microchannels via contact photolithography using a mask aligner according to the supplier's instruction. Polydimethylsiloxane (PDMS) replicas with patterned microchannels were prototyped via soft lithography by curing the PDMS mixture (base: curing agents=10:1) on the mast mold at 125 C. for 20 minutes. Cured replicas were further assembled into the micro-fluidic mixer on a clean silicon wafer via O.sub.2 plasma treatment (March Jupiter III, Westlake, Ohio, USA).

(17) FIG. 2 shows the bacteria solution port 21, wash water solution port 22, and the dechlorination solution 32 port, as well as the effluent outlet port 40. The inlet ports 21, 22, 32, and the outlet port 40 were created using Nanoports (IDEX Health and Science, Oak Harbor, Wash., USA) to provide seamless connection with an external syringe pump, which allows adjustable flow rate.

(18) Upon successful assembly of the device 10, the micro-fluidic mixer 10 was initially evaluated for its effectiveness and accuracy in providing a constant, stable input of solutions through the microchannel. The stability of bacterial concentration was tested by pumping the bacterial suspension (via bacteria inlet 21, FIGS. 1 and 2) and sterile water (via inlet wash water inlet 22 and dechlorination fluid inlet 32, FIGS. 1 and 2) at different flow rates. The effluent was collected and the bacteria were enumerated using modified Most Probable Number (MPN) procedures as described in the following section. The chlorine stability was tested similarly by pumping chlorine solution (via wash water inlet 22) with water, and the effluent was collected to quantify chlorine concentration using the DPD colorimetric method. In the follow-up chlorine stability test and dechlorinating efficiency studies, both chlorine (via wash water inlet 22) and dechlorinating solutions (via dechlorination inlet 32) were pumped into the micro-fluidic mixer with water. The effluent was collected to detect chlorine residual using the DPD method.

(19) The mixing efficiency and mixing patterns were evaluated by modeling scalar mixing (mixing efficiency) using the Computational Fluid Dynamics software (Autodesk Simulation CFD, 2014). The chlorine diffusion coefficient in water used in this modeling was 1.3810.sup.5 cm.sup.2/s. The solution has a low Reynolds number in this device, and the liquid was determined to be incompressible and the flow was characterized as laminar in the scalar mixing model. The boundary conditions (i.e. inlet volume flow rate, and outlet pressure) were adjusted to reflect the actual experimental value.

(20) The contact time between bacteria and chlorine solution was defined as the time required to pass through the pathogen deactivation region 20 (FIG. 1). The contact time was precisely controlled by adjusting the flow rate of the syringe pump that supplied the fluid to the micro-fluidic mixer 10. The mixing efficiency (scalar mixing) of the device and contact time distribution also were determined quantitatively using the trace function in the CFD software.

(21) Pathogen Inactivation Study

(22) A series of chlorine solutions with free chlorine concentrations of 0.5, 1, 5, 10, 20, 50 mg/L were freshly prepared using sodium hypochlorite. The free chlorine concentration before and after tests were verified using chlorine photometer (CP-15, HF Scientific Inc., Ft. Myers, Fla., USA). To determine the kinetics of chlorine on pathogen inactivation, chlorine solutions were pumped into the micro-fluidic mixer, together with the bacterial suspension of E. coli O157:H7 through a separate inlet. This was followed by injection of dechlorinating solution at the entry to Region B to stop the chlorine-pathogen reaction. To consider a dilution effect inside the micro-fluidic mixer (pathogen deactivation region 20, FIG. 1), the chlorine concentration in the start solution was double the effective concentration in the microchannel. All concentrations discussed in the study are effective concentrations after considering the dilution factors.

(23) The dechlorinating solution containing 0.1% sodium pyruvate and dechlorinating reagent (containing 110% sodium thiosulfate and 9099% sodium sulfate) was pumped via the dechlorination inlet 32 (FIG. 1) to stop the inactivation process. The contact time was controlled precisely by adjusting the solution flow rate and was determined quantitatively by CFD modeling. Pathogen survival was quantified using a modified Most Probable Number method as detailed in the inventors' previous publications.

(24) All experiments were repeated at least three times, and data was analyzed using SAS software (Version 9.3). E. coli O157:H7 populations were subjected to log transformation before statistical analysis. Data was analyzed as a two-way analysis of variance (ANOVA) with treatment and contact time as the main factors. Assumptions of normality and variance homogeneity were checked and the variance grouping technique was used to correct for variance heterogeneity. When effects were statistically significant, means comparisons were done using Tukey's range test with adjusted p-values to maintain experiment-wise error of 0.05.

(25) Results

(26) The effectiveness of the micro-fluidic mixer in maintaining stable concentrations of chlorine and bacterial suspensions during pumping was validated. The stability of the solution delivery first was tested by separately pumping chlorine (via wash water inlet 22, FIG. 1) with water, and then bacterial suspension (bacteria inlet 21, FIG. 1) with water, and collecting the effluent of each independent test for quantification. As shown in FIG. 3, both chlorine concentration (5 mg/L) and bacterial suspension (10.sup.7 MPN/mL) were stable for the entire 20 minute test period during continuous pumping at a flow rate of 0.16 mL/minute, or the equivalent of a 0.75 second contact time. The deviation of chlorine concentration was no more than 3%. E. coli O157:H7 concentrations were less than 2.5% (FIG. 3) from the respective theoretical values.

(27) The mixing efficiency of the pathogen inactivation region 20 and the dechlorination region 30 (FIG. 1) was also validated in separated experiments. The total pathogen inactivation (negative control) was validated by mixing chlorine (50 mg/L) and pathogen solution (10.sup.8 MPN/mL) inside the micro-fluidic mixer without the addition of dechlorinating solution, resulting in no bacterial survival detected from the collected effluent. The neutralizing power of dechlorinating reagent was demonstrated when pumping both chlorine and dechlorinating solutions into the micro-fluidic mixer, which resulted in no chlorine residue being detected in the collected effluent. These results are consistent with other reported micro-fluidic mixer studies with different solutions and flow rates.

(28) The mixing efficiency and mixing pattern were studied using CFD modeling of scalar mixing. During the modeling process, all solutions were considered as Newtonian fluids with non-slip boundary conditions. The solution has a low Reynolds number and laminar flow characteristics, due to the small scale. Achieving efficient and rapid mixing is one of the challenging tasks in micro-fluidic mixers, because mixing under laminar flow conditions depends on slow diffusion of chemicals and bacterial cells.

(29) Therefore, the study incorporated multiple passive micro-fluidic mixer designs that can be completed by simple modification of channel geometries into the following configurations: Y injection, serpentine channels, and channels with patterned grooves. These modifications and designs enhance folding, stretching and breaking of the laminar flows, which improves the mixing efficiency during flow. FIG. 4 shows the modeled scalar mixing of chlorine, bacteria, and dechlorinating solutions in our micro-fluidic mixer.

(30) The modeled result indicated that mixing can be achieved very rapidly and efficiently in the inventors' design. Therefore, the micro-fluidic mixer provides the means to conduct studies on pathogen inactivation kinetics in time periods as brief as 0.1 second. The CFD modeling also indicated that increased flow rates which result in decreased contact time, could increase mixing efficiency.

(31) The micro-fluidic mixer was also optimized for the adjustability of flow rate and contact time. By changing flow rate at different inlets, the time needed for fluid elements to pass through the mixing region can be varied. The relationship between the initial flow rate required and the specific contact time distribution can also be determined and predicted by the trace function in the CFD modeling software. FIG. 5 shows the correlation between inlet flow rate and contact time distribution. The micro-fluidic mixer had a channel dimension of 1 mm (width) by 0.1 mm (height), resulting in a cross-sectional area (A) of 0.1 mm.sup.2. The travel distance (D) of the chaotic mixer was 10 mm. The contact time (t) can be adjusted by changing the flow rate (r), and is calculated as:
t=DA/r(1)

(32) Ideally, every fluid element entering mixing the pathogen deactivation region 20 (FIG. 1) should spend the same amount of time before it enters the dechlorination region 30 (FIG. 1). However, with the folding, stretching, and breaking of the laminar flow condition, different fluid elements took different amounts of time to pass through the mixing patterns. In reality, not all fluid elements spend the same amount of time passing through the mixing regions in the micro-fluidic mixer. However, the variation did not affect the precision of contact time control with the coefficient of variation (C.sub.V)<0.05. The calculated flow rate was adopted in the following pathogen inactivation study to control the contact time from 0.1 to 1.5 seconds.

(33) The effect of chlorine concentration and exposure time on E. coli O157:H7 inactivation was tested using the micro-fluidic mixer after validation tests were completed. As shown in FIG. 6, the inactivation of E. coli O157:H7 was significantly affected by free chlorine concentration (P<0.0001), exposure time (P<0.0001), and their interactions (P<0.0001). The chlorine concentration and contact time exhibited a negative relationship, i.e., more concentrated chlorine solution required less contact time to achieve a 5-log reduction. With an E. coli O157:H7 suspension solution containing 110.sup.8 CFU/mL, 1.0 mg/L free chlorine solutions achieved 0.14 log reduction in 0.25 second, and 4.97 log reduction in 1.00 second. Increasing chlorine concentration significantly reduced exposure time required to achieve the same log reduction. With a free chlorine concentration of 10.0 mg/L, a 3.87 log reduction was achieved in 0.1 second, and a 5.77 log reduction in 0.25 second.

(34) The determination of minimum effective free chlorine concentration required to prevent pathogen survival and cross-contamination during produce wash is a major challenge facing the produce industry and regulatory agencies. The US industry-wide hazard analysis and critical control points (HACCP) program has set 1 mg/L free chlorine as the critical control limit (CCP) previously. Recent studies have shown that this chlorine concentration is insufficient in preventing pathogen cross-contamination.

(35) However, the effective minimum free chlorine concentration to prevent cross-contamination has not yet been determined and validated. During water-mediated cross-contamination, pathogens are first dislodged from the surface of the contaminated produce, survive in wash water, and then transfer to the originally uncontaminated produce as they move along with wash water. Therefore, the effective minimum free chlorine concentration required to prevent pathogen cross-contamination must be a concentration that can inactivate pathogens instantaneously. The inventors determined that a free chlorine concentration of 1 mg/L will achieve a 5-log reduction in 1.00 second, 5 mg/L in 0.50 second, and 10 and 20 mg/L in 0.25 second. Chlorine concentration at 0.5 mg/L and 50 mg/L were included in the study, but the accurate determination of exposure time required to achieve a 5-log reduction requires additional modification of the micro-fluidic mixer to accommodate such low and high chlorine concentrations.

(36) The results obtained with 1, 5, 10, and 20 mg/L in this study provide a reasonably good explanation for the observation that cross-contamination occurs at 1 mg/L, but not at 10 mg/L, as 1 mg/L requires more than 1.2 seconds to achieve a 5-log reduction in pathogen while 10 mg/L kills pathogens much faster (0.2 second for a 5-log reduction). The results were analyzed with the contour graph (FIG. 7), with the predicted log reduction (1-5 log) at different chlorine concentrations and contact times.

(37) Kinetic Models

(38) Kinetic modeling was also applied to simplify the complicated disinfection phenomena of produce wash system. Three models that are commonly used for studying bacterial disinfection kinetics with chemical-based disinfectant, including the Chick-Watson Model, the Hom Model, and the Selleck Model.

(39) Watson developed an empirical logarithmic equation to relate the inactivation constant (k) to disinfectant concentration (C) and reaction time (t):

(40) $\begin{array}{cc}\log \left(\frac{N}{N_0}\right)=-{\mathrm{kC}}^{n}t& \left(2\right)\end{array}$
where N=number of pathogen cells per unit volume,

(41) N.sub.0=number of pathogen cells initially at time zero,

(42) k=strain- and condition-specific inactivation constant,

(43) C=free chlorine concentration,

(44) n=coefficient of dilution,

(45) t=reaction (contact) time.

(46) From the experimental data in FIG. 6, k and n were determined as:

(47) $\begin{array}{cc}\log \left(\frac{N}{N_0}\right)=-3.48C^{0.24}t& \left(3\right)\end{array}$

(48) The Hom model is a generalized empirical equation of the Chick-Watson model considering chlorine disinfection as a pseudo first-order reaction:

(49) $\begin{array}{cc}\log \left(\frac{N}{N_0}\right)=-{\mathrm{kC}}^n{t}^m& \left(4\right)\end{array}$

(50) where m is a reaction rate constant and other factors are as described above for equation 2. The constants were modeled using experimental data as:

(51) $\begin{array}{cc}\log \left(\frac{N}{N_0}\right)=-3.18C^{0.28}{t}^{0.18}& \left(5\right)\end{array}$

(52) The Selleck model was originally developed to predict chlorine inactivation of bacteria in wastewater. The model was empirical and can be adjusted to different sanitization systems:

(53) $\begin{array}{cc}\log \left(\frac{N}{N_0}\right)=-n\log \left(1+\frac{\mathrm{Ct}}{k}\right)& \left(6\right)\end{array}$

(54) The experimental data also was applied to compute values of empirical coefficient k and n, which yields:

(55) $\begin{array}{cc}\log \left(\frac{N}{N_0}\right)=-0.678\log \left(1+\frac{\mathrm{Ct}}{0.25}\right)& \left(7\right)\end{array}$

(56) The pathogen inactivation kinetics of the three selected models were compared statistically with the original experimental data. FIG. 8 shows the effect of chlorine concentration at a specific contact time (0.25 second). Both the Watson (P=0.860) and Hom (P=0.841) models reflected the pathogen inactivation scenarios with short contact time. The Selleck model was used to predict pathogen survival in chlorine over a substantially longer contact time (e.g. 3 hours) than was used in the studies reported here. Thus it is not surprising that the model didn't fit the results from this short time-course kinetic study (P<0.0001). The variation and difference between experimental data and modeled results could be explained by the heterogeneity of bacterial cells in a population relative to their inherent resistance to chlorine, which was distributed in a spatially- or time-dependent manner.

(57) In a chlorine-based disinfection process, the decimal reduction time (D-value) is defined as the time required at a specified free chlorine concentration to kill 90% (one log reduction) of the organisms being studied. The D-value here was calculated based on the empirical models developed in this study, namely the Watson model (equation 3) and the Hom model (equation 5). FIG. 9 shows the relationship between chlorine concentration and D-value. For the widely used 1 mg/L free chlorine concentration in HACCP programs, the D-value is 28.06 milliseconds. By increasing chlorine concentration to 5, 10, and 20 mg/mL, the D-value reduces to 5.72, 2.87, and 1.43 milliseconds respectively.

(58) Conclusions

(59) A micro-fluidic mixer useful for assessing pathogen inactivation kinetics at less than 1 second was designed, fabricated, and validated. This device also was used to determine the time and dose dependent response of pathogen inactivation via free chlorine. Test results indicate that 1) E. coli O157:H7 inactivation is significantly affected by free chlorine concentration (P<0.0001), contact time (P<0.0001), and their interactions (P<0.0001); 2) A 5-log reduction of E. coli O157:H7 requires exposing E. coli O157:H7 cells to a solution containing 1 mg/L free chlorine for at least 1.2 seconds, or a solution containing 10 mg/L free chlorine for 0.2 second. These findings provide critical information regarding the determination of minimum free chlorine concentration required to prevent pathogen survival and cross-contamination during fresh produce wash operations.

(60) For the foregoing reasons, it is clear that the method and apparatus described herein provides an innovative micro-fluidic mixer for determining whether sufficient free chlorine is present in a wash solution to inactivate a target pathogen. The disclosed method and apparatus may be modified and customized as required by a specific operation or application, and the individual components may be modified and defined, as required, to achieve the desired result.

(61) Although all of the materials of construction are not described, they may include a variety of compositions consistent with the function described herein. Such variations are not to be regarded as a departure from the spirit and scope of this disclosure, and all such modifications as would be obvious to one skilled in the art are intended to be included within the scope of the following claims.