Lab on a Chip

Cite this: DOI: 10.1039/c0xx00000x

www.rsc.org/loc

Dynamic Article Links ►

TECHNICAL INNOVATION

This journal is © The Royal Society of Chemistry [year] [journal], [year], [vol], 00–00 | 1

Xurography Actuated Valving For Centrifugal Flow Control

David J. Kinahana,b

*, Philip L. Earlya,b

, Abhishek Vembadia,c

, Eoghan MacNamarad,

Niamh A. Kilcawleya,b

, Thomas Glennond, Dermot Diamond

d, Dermot Brabazon

c and Jens Ducrée*

a,b

Received (in XXX, XXX) Xth XXXXXXXXX 20XX, Accepted Xth XXXXXXXXX 20XX

DOI: 10.1039/b000000x 5

We introduce a novel instrument controlled valving scheme for centrifugal platforms which is based upon

xurography. In a first approach, which is akin to previously presented event-triggered flow control, the

valves are composed of a pneumatic chamber sealed by a dissolvable film (DF) and by a pierceable

membrane. Liquid is initially prevented from wetting the DF by the counter pressure of a trapped gas. Via

a channel, this pocket is pneumatically connected to a vent, sealed by the pierceable membrane, located 10

on the top surface of the disc. By scouring the top surface of the disc, along a pre-defined track by a

robotic knife-cutter, the trapped gas is released and so the liquid can wet and disintegrate the DF. In order

to automate assay protocols without the need to integrate DFs, we extend this xurography-based flow

control concept by selective venting of chambers subjected to pneumatic over-pressure or vacuum

suction. Unlike most instrument controlled flow-control mechanisms, in this approach to valve actuation 15

can occur during disc rotation. To demonstrate the potential of this flow control approach, we designed a

disc architecture to automate the liquid handling as the backbone of a biplex liver assay panel. We

demonstrate valve actuation during rotation, using the robotic arm, using this disc with visualisation via

dyed water. We then demonstrate the biplex liver assay, using calibration reagent, by stopping the disc

and manually piercing the membrane to actuate the same valves. 20

1. Introduction

Over the recent decades centrifugal microfluidic systems have

increasingly entered the arena of biomedical point-of-care

diagnostics 1-3. In these “Lab-on-a-Disc” (LoaD) systems, a typic-

ally disc-shaped cartridge is rotated at defined frequency protocol 25

by a robust and low-cost spindle motor. Its simple actuation

scheme renders these LoaD devices independent of external

pressure sources and their pneumatic connectors, thus ensuring

ease-of-use and minimum maintenance.

Centrifugal flow control is predominantly implemented by chan-30

ges in the spin rate4. For instance, in capillary burst valves, the

centrifugally induced hydrostatic pressure overcomes the counter-

acting capillary pressure imposed by a hydrophobic constriction.

Apart from the system-innate spindle motor, flow control can also

be achieved by additional, instrument-based modules, for exam-35

ple by laser ablating a sacrificial membrane5 or by heating a

phase-change material 6.

Recently, Kinahan et al. introduced event-triggered flow control7,

8 based on dissolvable-film (DF) membranes9. These valves are

composed of a dead-end pneumatic chamber sealed by two DFs. 40

Upon removal of the so-called control film (CF) through contact

with a secondary liquid, the pneumatic chamber is vented to let

the primary liquid escape by dissolving the so-called the load film

(LF). By cascading such centrifugo-pneumatic DF valves, ‘event-

triggered’ flow control has been implemented, e.g. for automating 45

an unprecedented number of individual liquid handling steps in a

single disc at an essentially constant spin rate. However, a major

drawback of this event-triggered valving is its poor flexibility of

timing which depends on the dissolution time of the DFs and time

for liquid to move about the disc. Particularly, these timing issues 50

can limit the implementation of biomedical assays requiring

extended incubation.

In order to allow discretionary timing, we introduce here an instr-

ument-controlled scheme which is effectively an actuated centri-

fugo-pneumatic DF valve. In this case the CF is replaced by a 55

vent, sealed by a pierceable adhesive tape, which is located on the

top surface of the disc (Fig. 1). A blade, adapted from a

commercial knife cutter, is mounted on a 2-axis robotic arm and

positioned above the disc (Fig. 2). During rotation, the blade is

lowered to perforate the adhesive tape and thus venting the 60

pneumatic chamber to atmosphere. Only films located along the

same orbit are actuated (quasi) in parallel, so the order of radial

positions where the knife cutter touches the disc defines the

sequence of valving.

Furthermore, we also demonstrate how similar flow control can 65

be achieved by using pneumatic overpressure or vacuum suction 10. While this technique works without incorporation of DFs, it

comes at the expense of limiting the maximum spin rate.

As a proof-of-concept, we demonstrate the automation of multi-

liquid, multi-step sample preparation and reagent handling 70

protocol towards a biplex liver assay panel. Critically for on-disc

blood processing, the valves can be externally actuated during

2 | Journal Name, [year], [vol], 00–00 This journal is © The Royal Society of Chemistry [year]

Figure 1: Operation of the xurography-enabled DF valve. (a) The valve consists of a pneumatic chamber with two outlets closed by a DF tab and PSA. (b) The gas pocket trapped between the meniscus of the incoming liquid and these outlets initially prevents the wetting of the DF. (c) To open the valve, a knife blade mounted on a robotic arm (cf. Fig. 2) scours the top of the disc, and thereby piercing the PSA. (d) As the gas pocket decompresses, the DF is

wetted and dissolves to clear the passage of liquid to the outlet. 5

rotation, thus ensuring efficient plasma extraction. This highly

controllable valving scheme, where liquid handling steps at any

location on the disc can be controlled through interaction with its

external surface, represents a viable path towards large-scale

integration (LSI) and automation of multiplex bio-assays. 10

2. Experimental

2.1 Manufacture of Robotic Arm

The robotic arm was composed of two linear precision rails

(Reliance Precision Ltd, UK). It was mounted in an XZ-axis

configuration on two pillar supports. The pillars were 15

manufactured from chemically bonded laser-cut sheets of

poly(methyl-methacrylate) (PMMA). Both axes were powered

using stepper actuators (Firgelli Technologies Inc.). The stepper

motor on the X-axis used an integrated potentiometer to provide

position feedback to custom program (LabVIEW, National In-20

struments) via a data acquisition card (National Instruments).

This stepper motor could also accurately position the knife cutter

assembly, which is mounted on the x-axis, to an accuracy of

0.5mm. Along with the Z-axis rail, the knife-cutter assembly was

composed of a backing plate and a stepper actuator which had a 25

stroke of 10 mm without positional feedback. Therefore, the

vertical actuator was connected to the blade holder via a spring-

damped connection. In its retracted state, the blade did not

contact the top surface of the disc. In the extended state, the blade

could scour the top surface of the disc. Thus, tolerances in disc 30

thickness and mounting were compensated for by the spring

damping. The knife blade was adapted from a commercial knife-

cutter (Graphtec, Japan).

2.2 Disc Manufacture

The discs are manufactured using multilayer polymer fabrication 35

methods using methods previously described 7, 11. The over-

pressure / vacuum-suction valves were manufacture using disc

which were assembled from five layers. Laser-cut PMMA sheets

were used for the upper layer, the reservoir layer and the base

layer. Vents and loading holes were created from voids in the 40

upper layer and reservoirs were created from voids in the base

layer. Connecting microchannels were created on the upper and

lower pressure sensitive adhesive (PSA) (Adhesives Research,

Limerick, Ireland) from voids created by a knife-cutter. The

stacking of the two PSA layers permitted microchannels cros-45

sings. The discs for demonstrating xurography actuated DF

valves were manufactured from four PMMA and four PSA layers

as previously described 7 in order to incorporate the DF tabs.

Finally the vent layer of the disc was covered with a transparent

film (Adhesives Research) to seal the disc. Loading holes and 50

appropriate vents were then pierced by a scalpel.

2.3 Xurography Actuated DF Valves

The disc implements a biplex liver assay test (see Section 2.4)

starting with plasma extraction. Next, 10-µl volumes of plasma

are metered and then transferred to the periphery of the disc 55

where they mix with the assay reagents alkaline phosphatase

(ALP), active direct bilirubin (DB) and passive DB. While the

DB assays involve simpler end-point measurements, the ALP re-

quires the addition of a stop reagent after a defined reaction

interval. Assay automation thus requires five sequentially actu-60

ated valves.

Figure 2: Robotic arm. (a) 3D rendering. (b) The robotic arm mounted on spin stand. The liver assay disc (cf. Fig. 3) is mounted on the experimental rig.

(c) Monochrome image acquired while the disc is spinning at 50 Hz. The track of knife cutter can be seen scratching the PSA layer to vent and open Valve 1. 65

This journal is © The Royal Society of Chemistry [2013] Lab Chip, 2013, [vol], 00–00 | 3

Figure 3: The Lab-on-a-Disc for a biplex liver assay panel measuring Direct Bilirubin (DB) and Alkaline Phosphatase (ALP). (a) The disc displays three se-parate structures which are operated in parallel. (b-e) Schematic of blood processing on the disc, (b) sample centrifugation, (c) opening of Valve 1 and metering of plasma, (d) opening of Valve 2, 3 and 4, and transfer of metered sample (10 µl) to mix with reagents. (e) After 3 minutes, the knife cutter opens Valve 5 to release the reagent stopping the ALP assay. Note that, at all times, the disc is subject to Euler-force based “shake-mode” mixing. (f) 5

Image of the disc just prior to opening of Valve 5. This disc is processing 100% Calibrant. (g) Comparison of calibration data acquired on benchtop (n = 3) and on-disc (n = 1).

The disc was tested on a custom spin stand12 on which the robotic

arm was mounted. During disc rotation, a servo-motor (FESTO,

Germany) periodically generates a trigger pulse. 10

This motor was controlled using a CANOpen interface via a

custom LabVIEW (National Instruments, UK) program. The

angular location of the trigger pulse is set using this software

interface, thus permitting visualization of an arbitrary azimuthal

position on the disc. 15

As the robotic arm was above the disc, conventional ‘top down’

visualisation resulted in a limited view of the disc. Thus, a mono-

chrome CCD camera (PCO, Germany), synchronised with the

motor and a strobed light source, was mounted at an angle to vi-

sualise both, the movement of liquid as well as the knife cutter. 20

The efficacy of the knife cutter was evaluated with (red) dyed

4 | Journal Name, [year], [vol], 00–00 This journal is © The Royal Society of Chemistry [year]

Figure 4: A Lab-on-a-Disc for the biplex liver assay using over-pressure / vacuum suction xurography valves. Chambers with overpressure are

represented by light green and negatively pressurised valves are represented by light red. This disc was tested using dye solutions. Each LUO was initiated by stopping the disc and manually venting the pneumatic chambers.

water. Initially, the disc (manufactured using DFs) is rotated at 5

60 Hz and the valves remained closed (the suction / over-pressure

disc remained sealed up to 20 Hz). Next, the spin rate was

reduced to 5 Hz when the knife cutter was engaged to scour the

top of the disc along the radius of Valve 1. The valve vented

within ~5 s of the blade touching top PSA. Subsequently, the 10

blade was retracted and the disc was accelerated up to 30 Hz to

drive the liquid into the metering structures. Next, the disc was

decelerated to 5 Hz and the blade was re-positioned to trigger

Valves 2, 3 and 4 in parallel about 5 s after touchdown of the

blade. The disc was again accelerated to 30 Hz to empty the 15

metering structures. This procedure was repeated to trigger Valve

5.

Actuation of the valves was also attempted at higher spin rates

(10 Hz initially), but excessive vibration in the robotic arm as a

consequence of the mechanical drag caused damage to the appa-20

ratus. Yet, the maximum spin rate of xurography actuation might

be elevated through improved mechanical design. Due to this,

implementation of the biplex liver assay on the disc was

implemented by stopping the disc between each Laboratory Unit

Operation4 (LUO) for piercing the upper layer of PSA with a 25

scalpel. This method was also used for the overpressure / suction

valves.

2.4 Biplex Liver Assay Panel

To demonstrate timing of flow control with the xurography-actu-

ated flow control scheme, we implemented a multi-step. 30

absorbance based biplex liver assay panel comprising of DB

(DB411, Randox, Ireland) and ALP (AP311, Randox, Ireland) as

adapted from that previously described by Nwankire et al. 13.

While the DB assay completes after 15 minutes, the ALP reaction

must be halted after 3 minutes by addition of ‘stop reagent’ 35

immediately followed by the absorbance measurement. Along

with processing a blood sample, the assay needs to be calibrated

with known concentrations of DB (0 and 100%) and ALP

(CAL2350, Randox, Ireland). The DB concentration is

determined from two assays, termed ‘active’ and ‘passive’ which 40

are read out on a commercial plate reader (TECAN Infinite® 200

PRO).

Thus a disc with nine reservoirs is required to array 0% calibrant,

100% calibrant and the blood sample against active DB, passive

DB and ALP. This is achieved by placing three identical 45

structures on a single disc, each of which can prove one sample

against active DB, passive DB and ALP (Fig. 3). Absorbance

data from a disc, used to process 0% calibrant and 100% calibrant

(Fig. 3h).

50

2.5 Xurography Actuated Overpressure / Suction Valves

In an effort to eliminate the integration of the DF, we developed

an alternative method where the liquid is initially retained in the

reservoirs by trapped air. Rather than by the DF, the liquid is

initially either retained by an upstream underpressure or a down-55

stream overpressure. Based on the same xurography-based me-

chanism, pressures are equilibrated with atmosphere by piercing

the top membrane. Figure 4 demonstrates the potential for auto-

mating the same biplex liver assay panel (Section 2.4) with this

alternative approach, thus permitting simpler disc manufacture; 60

on the downside, this valving scheme limits the maximum spin

rate of the valves as, at a sufficiently high spin-rate, the suction

back-pressure can be overcome 10. This can therefore limit the

usefulness of these valves in applications, such as leukocyte

extraction from blood 14, where high centrifugal forces are 65

advantageous.

Similarly, for the over pressure valve, the ‘burst frequency’ of the

unvented chamber, where the liquid will overcome the over

pressure and enter the collection chamber, is dependent on ratio

of the volume of an intermediate chamber (ΔV) to the combined 70

volume of the intermediate chamber and the collection chamber

(V). This ratio similarly determines the ‘design’ burst frequency

of centrifugo-pneumatic DF valves 9 and the ‘failure’ frequency

of event-triggered DF valves 7 . In the non DF case, the

intermediate chamber must be relatively large to reflect the size 75

of the collection chamber. However, the presence of these large

intermediate chambers effectively increases the dead volume of

liquid trapped within the valves.

3. Conclusions & Outlook

This novel, instrument-actuated flow-control mechanism renders 80

the timing of assay protocols independent from both, the spin rate

of the disc and the dissolution time of the DF to permit freely

This journal is © The Royal Society of Chemistry [2013] Lab Chip, 2013, [vol], 00–00 | 5

Figure 5: Concept of the proposed next-generation, xurography driven platform. This human-powered and thus fully autonomous system will employ a hand-crank to power the spindle. Pushing down spring-loaded cutter arms will initiate each LUO. The proposed system might also be 5

equipped with a mechanical tachometer to monitor the spin rate. Alternatively a mechanical brake system might be incorporated to throttle the (maximum) speed. Particularly applicable for non-quantitative assays delivering ‘Yes-No’ answers which can be read out without instrumentation, this fully mechanically operated platform will 10

be well-suitable for deployment to extremely resource-poor settings.

programmable timing of automated assay protocols on the Lab-

on-a-Disc platform, even “on the fly”. These enhanced capabili-

ties of the novel valving scheme are demonstrated by processing

a complex, multi-parameter liver assay including a calibrant as 15

well as stopping reagents. In this case, the valves were actuated

by stopping the disc and manually piercing the sealing

membrane; this approach was chosen due to the limited

mechanical reliability of the robotic arm. However, it is clear that

with more robust design this valving technology has great 20

potential for implementation of automated assays.

It should be noted that, during this study, care was taken during

disc design to place the actuation membranes on different radii

from loading and venting holes. This prevents possible damage to

the cutting blade during rotation. Similarly, it is important to 25

ensure that scouring the upper surface of the disc does not impact

on optical clarity should on disc measurement be required.

However, by connecting the actuation membranes to the valves

by long pneumatic channels, these valves provide great flexibility

to locate the actuation membranes (mostly) arbitrary radial 30

locations. One potential way to limit these issues might be to

locate the pierceable membranes on the underside of the disc.

This xurography-based flow control scheme has potential to be

particularly powerful for applications in low-resource settings15, 16

including diagnostics associated with emerging issues in 35

developing countries17. While our work shown here uses either a

2-axis robotic arm to actuate the valves, or we stop the disc and

manually pierce the membrane, in the future we propose to

simplify the platform where a number of low-cost, spring loaded

knife-blades are mounted at different disc radii. These blades 40

could be actuated by low-cost solenoid actuators. Mounting one

actuator / blade per LUO would reduce the cost and complexity

of this platform. In this approach the disc only requires to be

rotated using a low-cost servo-motor without position or speed

control. Indeed, for making the system fully autonomous, the 45

system could even operate in a power-free manual mode by

replacing the motor with a hand-crank while actuating the valves

through applying finger pressure to (appropriately shielded)

spring-loaded blades (Fig. 5). Such an approach will also enable

greater miniaturisation which can make the system more 50

appealing for point-of-use applications.

Acknowledgements

This work was supported by the European Union under Grant No,

FP7-KBBE-2013-7-613908-DECATHLON and Science

Foundation Ireland under Grant Nos. 10/CE/B1821, 12/IA/1567 55

and 12/RC/2289-INSIGHT.

Footnotes

* To whom correspondance should be addressed: david.kinahan@dcu.ie;

jens.ducree@dcu.ie a Biomedical Diagnostics Institute, Dublin City University, Glasnevin, 60

Dublin 9, Ireland b School of Physical Sciences, Dublin City University, Glasnevin, Dublin

9, Ireland c School of Mechanical and Manufacturing Engineering, Dublin City

University, Glasnevin, Dublin 9, Ireland 65

dThe Insight Centre for Data Analytics, National Centre for Sensor

Research, Dublin City University, Glasnevin, Dublin 9, Ireland

References

1. M. Madou, J. Zoval, G. Jia, H. Kido, J. Kim and N. Kim,

Annu. Rev. Biomed. Eng., 2006, 8, 601-628. 70

2. J. Ducrée, S. Haeberle, S. Lutz, S. Pausch, F. V. Stetten and R.

Zengerle, Journal of Micromechanics and Microengineering,

2007, 17, S103-S115. 3. R. Gorkin, J. Park, J. Siegrist, M. Amasia, B. S. Lee, J.-M.

Park, J. Kim, H. Kim, M. Madou and Y.-K. Cho, Lab on a 75

Chip, 2010, 10, 1758-1773. 4. O. Strohmeier, M. Keller, F. Schwemmer, S. Zehnle, D. Mark,

F. von Stetten, R. Zengerle and N. Paust, Chemical Society

Reviews, 2015, 44, 6187-6229. 5. J. L. Garcia-Cordero, D. Kurzbuch, F. Benito-Lopez, D. 80

Diamond, L. P. Lee and A. J. Ricco, Lab on a Chip, 2010, 10, 2680-2687.

6. Y.-k. Cho, J.-g. Lee, J.-m. Park, B.-s. Lee, Y. Lee and C. Ko,

Lab on a Chip, 2007, 7, 565-565. 7. D. J. Kinahan, S. M. Kearney, N. Dimov, M. T. Glynn and J. 85

Ducrée, Lab on a Chip, 2014, 14, 2249-2258.

8. D. J. Kinahan, S. M. Kearney, O. P. Faneuil, M. T. Glynn, N. Dimov and J. Ducrée, RSC Advances, 2015, 5, 1818-1826.

9. R. Gorkin III, C. E. Nwankire, J. Gaughran, X. Zhang, G. G.

Donohoe, M. Rook, R. O'Kennedy and J. Ducrée, Lab on a 90

Chip, 2012, 12, 2894-2902.

10. W. Al-Faqheri, F. Ibrahim, T. H. G. Thio, J. Moebius, K.

Joseph, H. Arof and M. Madou, PloS one, 2013, 8, e58523. 11. D. A. Bartholomeusz, R. W. Boutté and J. D. Andrade,

Microelectromechanical Systems, Journal of, 2005, 14, 1364-95

1374. 12. M. Grumann, T. Brenner, C. Beer, R. Zengerle and J. Ducrée,

Review of Scientific Instruments, 2005, 76, 1-6.

13. C. E. Nwankire, M. Czugala, R. Burger, K. J. Fraser, T. Glennon, B. E. Onwuliri, I. E. Nduaguibe, D. Diamond and J. 100

Ducrée, Biosensors and Bioelectronics, 2014, 56, 352-358.

14. D. J. Kinahan, S. M. Kearney, N. A. Kilcawley, P. L. Early, M. T. Glynn and J. Ducrée, PloS one, 2016, 11, e0155545.

15. M. T. Glynn, D. J. Kinahan and J. Ducrée, Lab on a Chip,

2014, 14, 2844-2851. 105

16. M. T. Glynn, D. J. Kinahan and J. Ducrée, Lab on a Chip,

2013, 13, 2731-2748.

17. B. M. Popkin, L. S. Adair and S. W. Ng, Nutrition reviews, 2012, 70, 3-21.

110