CHARACTERIZATION OF MICROFLUIDIC COMPONENTS FOR

LOW-COST POINT-OF-CARE DEVICES S. Hugo

1* , K. Land

1 and H. Becker

2

1Council for Scientific and Industrial Research, SOUTH AFRICA and

2microfluidic ChipShop, GERMANY

ABSTRACT

This paper presents the characterization of microfluidic components for the realization of low-cost point-of-care

diagnostic devices, with focus on full blood count applications. We present a set-up to enable automated actuation of

device components utilizing parameters similar to those produced by manual actuation. Initial results show that simple

microfluidic components can be used to achieve repeatable and accurate results for sample and reagent introduction and

propulsion, as well as mixing and dispersion of sample and reagent, without the need for complex microfluidic

operations.

KEYWORDS: Simple microfluidic components, Blister pack, Point-of-care, Full blood count

INTRODUCTION

Increasing emphasis is being placed on low-cost point-of-care diagnostic systems, particularly in under-resourced

settings [1] to enable instant diagnosis and to improve healthcare. Although numerous and varied advances in the field of

microfluidics have enabled point-of-care systems to be realized [2], there is often a trade-off between cost and operational

integrity of the device. This work addresses these limitations by exploring the minimum requirements for microfluidic

component complexity to achieve results equivalent to those of accepted standard techniques for blood cell counting

applications and illustrates the repeatability and robustness of device components for 1) sample introduction into the

device through a plugging mechanism, 2) reagent introduction and propulsion via a blister pack, and 3) mixing and

dispersion of sample and reagent. The successful implementation of these components rely on the forces and speeds

applied to the components for actuation, and were investigated in this work towards the development of a final device.

FUNCTIONAL PRINCIPLE

The simple microfluidic components were integrated into a device that was manufactured using injection molding.

Figure 1 shows the microfluidic device components to prepare a sample of blood for visual analysis using a dilution

factor of 1:20.

Figure 1: (a) Microfluidic device showing sample inlet port (A), blister pack (B), mixing chamber (C), air actuation

port (D), and visualization chamber (E). (b) Schematic designs of the expected fluid flow resulting from actuation for

inlet port plugging (A), blister pack compression (B) with subsequent mixing (C), and air actuation for dispensing of

sample (D) into visualization chamber (E).

EXPERIMENTAL

An experimental set-up was implemented to allow for detailed characterization of each of the microfluidic device

components. Figure 2 shows the set-up with the parameters that can be set and measured. An external syringe pump is

used for air actuation. High speed cameras provide side and bottom views of the microfluidic device, enabling volumes,

flow rates, as well as positions and times of contact of the actuators with the device components to be determined.

For initial testing, the sample was a 10 µl blood simulant containing yeast cells (5 x1012

cells per liter), to represent an

average red blood cell concentration found in human blood. The blister pack mounted on the microfluidic device con-

tained 190 µl colored water to simulate the blood preparation reagent to be mixed with the blood simulant.

978-0-9798064-6-9/µTAS 2013/$20©13CBMS-0001 461 17th International Conference on MiniaturizedSystems for Chemistry and Life Sciences27-31 October 2013, Freiburg, Germany

Figure 2: Experimental set-up for device component characterization showing the equipment and corresponding

information that can be extracted.

RESULTS AND DISCUSSION

Manual actuation parameters were first extracted using the experimental set-up to determine a range of input parameters

for automated actuation of the devices. Over a series of 10 tests for manual plugging of the inlet port, compression of the

blister pack and air introduction, an average peak force of 13.82 N was recorded for plugging of the sample inlet port, with

minimum and maximum forces recorded as 8.83 N and 19.12 N, respectively. The average speed of actuation was

determined based on the rise time interval of the force from zero to a peak force required to fully plug the inlet. The distance

travelled in this time was assumed to be 4 mm, which is the distance from the inlet port opening to the inlet port base. The

average speed for manual inlet port plugging was calculated to be 1.49 mm/s, with minimum and maximum speeds

determined as 0.59 mm/s and 2.58 mm/s, respectively.

For manual blister pack compression, an average force of 60.79 N was recorded for complete compression of the blister,

with minimum and maximum forces recorded as 60.12 N and 61.04 N. The average speed of actuation for blister pack

compression was calculated as 0.79 mm/s with minimum and maximum speeds determined as 0.45 mm/s and 1.29 mm/s,

respectively. The speed was calculated in the same way as the inlet port, but over 5 mm, which is the distance from the tip of

the blister to the surface of the device.

There is much greater variability in the inlet port plugging forces and speeds than for manual blister pack compression.

This is as a result of the fine positioning and alignment of the plug required for smooth insertion into the inlet port on the

device. Misalignment and angular offsets result in a greater force being required to insert the inlet plug.

For automated actuation testing, input parameters for the experimental set-up were based on the manual actuation values

determined. For the inlet plug actuator, a speed of 1.5 mm/s was set with a position that resulted in an approximate force of

13 N for complete inlet port plugging. For the blister pack actuator, the speed was set to 0.75 mm/s with an absolute position

that resulted in an approximate force of 60 N. Flow rates for manual air actuation were approximated from high speed

videos, and ranged between 500 µl/min and 1.8 mL/min. An air flow rate of 500 µl/min was set for automated actuation.

Figure 3 shows the results of automated actuation sequences for 5 tests, showing forces and fluidic functioning for inlet

port plugging (a) and blister pack compression (b).

Even with automated actuation, variations in inlet port plugging are apparent as a result of alignment, a critical factor

that is being addressed. The blister pack actuation results are more consistent, as alignment does not play as great a role.

The blister pack load cell resolution is more coarse than that of the inlet plug load cell as it has a larger load. Image pro-

cessing was used to determine the total volume dispensed for each test. According to design, the ideal case would be to

have a total volume of 200 µl (10 µl sample, 190µl reagent). Current tests show that an average volume of 190 µl is ob-

tained with a standard deviation of 5 %. Ideally this variation would need to be reduced to 0.5 % and would also be de-

pendent on the number of cells counted. Considerations as to whether or not active metering is required are thus neces-

sary and will be determined after more extensive testing of the device components.

Figure 4 illustrates the visual analysis of the mixed blood simulant and reagent dispensed via air actuation at 500

µl/min into a 20 µl visualization area with a depth of 100 µm. A homogeneous distribution of cells was observed

throughout the viewing area.

462

Figure 3: Results of automated actuation for inlet port plugging (a) and blister pack release (b), showing force over time

and corresponding fluidic operation schematics and photos for times indicated as (1) to (4) for test 2. (a) Progression of

sample being dispensed as the plug is inserted, and (b) sample movement into the mixing area as a result of blister pack

compression (2) and release (3), enabling the sample and reagent to be mixed.

Figure 4: Viewing chamber with corresponding microscope image (400X) of the central area.

CONCLUSION

The initial results show repeatable yet simple fluidic operations from sample introduction to sample analysis for blood

cell counting applications, utilizing an experimental set-up to allow for a range of parameters to be investigated. This

work provides a foundation for accurate and low-cost point-of-care blood count diagnostic systems such as a full blood

count to be realized, with particular relevance and benefit to rural clinics and hospitals, where the optimal balance be-

tween function and cost is crucial, and as a result of this work, achievable.

ACKNOWLEDGEMENTS

The authors gratefully acknowledge the CSIR for funding for this project, and the CSIR project team for their tech-

nical contributions. Devices were manufactured by microfluidic ChipShop.

REFERENCES

[1] C.D. Chin, V. Linder and S. Sia, Lab-on-a-chip devices for global health: Past studies and future opportunities, Lab

Chip, vol. 7, no. 41, pp 41-57, 2007.

[2] W.G. Lee, Y.-G. Kim, B.G. Chung, U. Demirci and A. Khademhosseini, Nano/Microfluidics for diagnosis of infec-

tious diseases in developing countries, Adv. Drug. Deliver. Rev., vol. 62, pp 449-457, 2007.

CONTACT

*S. Hugo, tel: +27-12-8413101; SHugo@csir.co.za

463