micromachines

Article

Influence of Geometry and Surrounding Conditions

on Fluid Flow in Paper-Based Devices
Noosheen Walji and Brendan D. MacDonald *
Faculty of Engineering and Applied Science, University of Ontario Institute of Technology,
2000 Simcoe Street North, Oshawa, ON L1H 7K4, Canada
* Correspondence: brendan.macdonald@uoit.ca; Tel.: +1-905-721-8668 (ext. 5716); Fax: +1-905-721-3370
Academic Editors: Sergey S. Shevkoplyas and Nam-Trung Nguyen
Received: 31 January 2016; Accepted: 18 April 2016; Published: 25 April 2016

Abstract: Fluid flow behaviour in paper is of increasing interest due to the advantages and expanding
use of microfluidic paper-based analytical devices (known as PADs). Applications are expanding
from those which often have low sample fluid volumes, such as diagnostic testing, to those with
an abundance of sample fluid, such as water quality testing. The rapid development of enhanced
features in PADs, along with a need for increased sensitivity and specificity in the embedded
chemistry requires understanding the passively-driven fluid motion in paper to enable precise control
and consistency of the devices. It is particularly important to understand the influence of parameters
associated with larger fluid volumes and to quantify their impact. Here, we experimentally investigate
the impacts of several properties during imbibition in paper, including geometry (larger width and
length) and the surrounding conditions (humidity and temperature) using abundant fluid reservoirs.
Fluid flow velocity in paper was found to vary with temperature and width, but not with length of
the paper strip and humidity for the conditions we tested. We observed substantial post-wetting flow
for paper strips in contact with a large fluid reservoir.
Keywords: paper-based devices; paper microfluidics; imbibition; environmental monitoring

1. Introduction
Microfluidic paper-based analytical devices (PADs) are a versatile technology capable of
facilitating a variety of complex detection and diagnostic processes [13]. Microfluidics has traditionally
involved small fluid volumes (109 to 1018 L) [2] and PADs have also been used in applications
with small fluid volumes. Processing and analyzing small fluid volumes is critical in many fields,
particularly in diagnostics, due to the availability of low fluid volumes and convenience associated
with collecting small volumes of fluid, such as finger prick collection of blood samples. PADs have
been developed for a number of diagnostic analytical tests including infectious diseases [46], and
cancer detection and investigation [7]. PADs can also be used for testing where larger fluid volumes
are available as evidenced by the widespread application of PADs for pregnancy testing using urine
samples, and recently there has been a trend towards using PADs in applications where an abundance
of fluid is available, such as environmental testing, specifically for water quality testing. A number
of PADs have been developed for environmental testing including those testing for heavy metals
in water sources [811], heavy metals in industrial waste [12], bacterial contaminants [13], airborne
contaminants [14,15], and other biological and pollution targets [1620]. The availability of larger fluid
volumes makes it possible to process more fluid through the devices, for example in cases where greater
concentration may be desired; thus, the PADs can be designed with larger sizes and fed from large
reservoirs, which can influence the fluid flow behaviour during wicking. To design PADs capable of
exploiting larger fluid volumes we require an understanding of how the parameters associated with
larger fluid volumes, such as larger geometry and fluid reservoirs, influence the fluid flow behaviour.
Micromachines 2016, 7, 73; doi:10.3390/mi7050073

www.mdpi.com/journal/micromachines

Micromachines 2016, 7, 73

2 of 13

We address this need with a detailed experimental study to characterize the fluid flow behaviour in
paper with larger dimensions (beyond 5 mm) and fed from large reservoirs.
PADs are ideal for applications in low-resource settings due to their low cost, ease of use,
and passively-driven flow that allows for independence from auxiliary equipment [1,21]. A vital
consideration in the successful application of PADs on a global scale is the various climate zones they
are applied in. For example, one area of interest for microfluidic testing is diagnosing HIV and syphilis
in Rwanda, with expanding interest from healthcare workers in India and Tanzania [22]. Rwanda
has a tropical highland climate with temperatures ranging from 15 to 30 C and a humidity range
of 38% to 100% [23]. A paper-based hepatitis B detection test was encased in a pen format for safe
sample containment, with clinical application in Vietnam [4]. Vietnam experiences tropical climates
with temperatures between 20 and 35 C and humidity between 50% and 98% [23]. We are interested
in testing for contaminants, specifically arsenic levels, in drinking water sources in tube wells in
Bangladesh. The climate zone in Bangladesh is described as a tropical savannah, with temperatures
varying from 15 to 40 C, and humidity levels from 25% to 95% [23]. Fluid flowing in paper-based
devices can undergo evaporation during wicking and some PADs are encased in plastic to prevent
this evaporation [4,24,25]. This is particularly important for applications with low fluid volume so as
not to deplete the fluid and prevent the strip from drying out before the test is complete. For devices
where larger fluid volumes are available it is possible that the fluid flow behaviour will be influenced
by evaporation, and the devices may be more susceptible to the evaporation due to their larger sizes
and increased exposure to the surrounding conditions. In PADs where the consistency of results
in each test is necessary regardless of the point-of-interest location, a detailed understanding of the
influence of temperature and humidity on fluid flow behaviour during wicking is beneficial to the
design process.
The analytical capacity of PADs has been increasing rapidly due to the development of enhanced
features such as multi-dimensional device designs [17,2629], enrichment-based techniques [30],
filtering and multi-step reactions [31], dissolvable fluidic time delays [32], flow control mechanisms
such as pumps [33], and analyte concentration and transport after wetting [34]. These enhanced
features have enabled more complex chemistries to be incorporated into paper-based platforms,
which have expanded the contaminants that can be detected. Many of these contaminants occur
in the presence of other similar constituents, and have low detection limits, thus increasing the
need for sensitivity and specificity. Engineering highly sensitive and specific PADs requires precise
fluid flow control that enables pre-programming of the chemical reactions to yield predictable and
consistent signal readouts. The current demands for precise flow control have necessitated an in-depth
understanding of which parameters influence the flow behaviour and what their impact is; therefore,
it is particularly important to analyze how the parameters associated with larger fluid volumes and
environmental conditions influence the flow behaviour to ensure precise test results.
In this paper, we characterize fluid flow in paper-based devices using an experimental analysis to
determine which parameters influence imbibition and need to be considered when predicting flow
behaviour in PADs when there is an abundance of sample fluid volume available. Specifically, we
perform experiments to investigate the influence of surrounding temperature and humidity levels,
paper machine direction, strip length, and strip width on the wicking behaviour of fluid in a paper
strip when the source of the fluid is a large reservoir. This analysis of influential parameters can be
incorporated in design considerations for paper-based devices when larger fluid volumes are available
for a range of climate conditions.
2. Materials and Methods
2.1. Preparation of the Paper Strips
We used pure cellulose chromatography paper (Whatman Grade 1 CHR and 17 CHR, GE Healthcare,
Mississauga, ON, Canada) for the paper strips. Table 1 gives the properties of these papers, as provided

2. Materials and Methods

2.1. Preparation of the Paper Strips
Micromachines
2016,pure
7, 73
We used

13
cellulose chromatography paper (Whatman Grade 1 CHR and 17 CHR,3 ofGE
Healthcare, Mississauga, ON, Canada) for the paper strips. Table 1 gives the properties of these
papers, as provided by the supplier. Strip dimensions were printed onto the paper using an inkjet
by the supplier. Strip dimensions were printed onto the paper using an inkjet printer (HP Deskjet 2540,
printer (HP Deskjet 2540, Hewlett-Packard, Mississauga, ON, Canada), and the strips were cut using
Hewlett-Packard, Mississauga, ON, Canada), and the strips were cut using a craft paper cutter.
a craft paper cutter.

Table 1. Properties of Whatman chromatography paper.

Table 1. Properties of Whatman chromatography paper.
2) 2
Paper
TypeType
Basis
Weight
(g/m
Paper
Basis
Weight
(g/m
)
Whatman
1 CHR
Whatman
1 CHR
87 87
Whatman
17 CHR
325325
Whatman
17 CHR

Thickness
Thickness (mm)
(mm)
0.18
0.18
0.70
0.70

Porosity Flow
Flow
(mm/min)
Porosity
RateRate
(mm/min)
67.8%
4.334.33
67.8%
69.1%
69.1%
6.336.33

2.2. Experimental
2.2.
Experimental Apparatus
Apparatus
Experiments were
conducted in
in aa temperature
temperature and
and humidity
humidity controlled
controlled chamber
chamber measuring
measuring
Experiments
were conducted
51.5 cm
cm
41.5
41.5 cm
cm 41
41 cm,
cm, as
as shown
shown in
in Figure
Figure 1a.
1a. The
The temperature
temperature in
in the
the chamber
chamber was
was adjusted
adjusted
51.5
using aa 250
250 W
W heat
heat lamp,
lamp, measured
measured using
using aa digital
digital thermometer,
thermometer, and
and controlled
controlled using
using aa digital
digital
using
temperature
controller
(TC).
The
humidity
in
the
chamber
was
established
using
a
submerged
temperature controller (TC). The humidity in the chamber was established using a submerged heater
heater
in ancontainer
open container
of monitored
water, monitored
a hygrometer,
digital hygrometer,
and controlled
in
an open
of water,
using ausing
digital
and controlled
using ausing
digitala
digital
humidity
controller
(HC).
The
heater
and
submerged
heater
were
placed
on
the
right
side
of
humidity controller (HC). The heater and submerged heater were placed on the right side of the
the
chamber,
while
the
reservoir
and
sensors
were
placed
on
the
left
side
of
the
chamber,
chamber, while the reservoir and sensors were placed on the left side of the chamber, approximately
approximately
25 the
cm heaters.
away from
heaters.was
Theplaced
reservoir
placed
cm left
away
from
thechamber,
left wall
25
cm away from
Thethe
reservoir
3 cmwas
away
from3 the
wall
of the
of
the
chamber,
and
the
sensors
were
mounted
on
the
left
wall
of
the
chamber,
no
higher
than
10 cm
and the sensors were mounted on the left wall of the chamber, no higher than 10 cm above the reservoir
above
the strip,
reservoir
and that
paper
to ensure
the temperature
and humidity
readings
and
paper
to ensure
the strip,
temperature
and that
humidity
readings monitored
the reservoir
and
monitored
the
reservoir
and
paper
strip
conditions
as
accurately
as
possible.
There
was
no
active
paper strip conditions as accurately as possible. There was no active mixing of the air in the chamber
mixing
air in
the chamber
orderchanges
to avoid
after changes
in conditions
the
in
orderoftothe
avoid
convection,
andinafter
inconvection,
conditions and
the chamber
was allowed
to reach
chamber wasprior
allowed
to reach equilibrium
prior to conducting experiments.
equilibrium
to conducting
experiments.

Figure 1.
1. (a)
(a) Schematic
Schematic of
of experimental
experimental setup
using paper
paper
Figure
setup to
to examine
examine wicking
wicking behaviour
behaviour in
in PADs
PADs using
strips dipped
dipped in
in aa reservoir;
reservoir; (b)
(b) paper
paper strip
strip wicking
wicking from
from reservoir
reservoir during
during experimentation.
experimentation.
strips

The paper
The
paper strip
strip (Whatman,
(Whatman, GE
GEHealthcare,
Healthcare,Mississauga,
Mississauga,ON,
ON,Canada)
Canada)was
wasfolded
foldedand
andplaced
placedinina
10
cm
diameter
petri
dish
filled
with
a
12.5
mmol
aqueous
solution
of
Allura
red
food
colour
dye
a 10 cm diameter petri dish filled with a 12.5 mmol aqueous solution of Allura red food colour dye
(Sigma Aldrich,
Aldrich, Oakville,
Oakville,ON,
ON,Canada)
Canada)and
andtotokeep
keepthe
thestrip
strip
level
opposite
end
was
supported
(Sigma
level
thethe
opposite
end
was
supported
by
an inverted petri dish as shown in Figure 1b. The wicking process was recorded using a Nikon Digital
SLR camera (Nikon, Mississauga, ON, Canada) with an AF-S DX Micro-NIKKOR 40 mm f/2.8 G lens

Micromachines 2016, 7, 73
Micromachines 2016, 7, 73

4 of 13

4 of 13

by an Mississauga,
inverted petri ON,
dish Canada).
as shown A
in JEOL
Figure6400
1b. The
wicking
processmicroscope
was recorded
using(JEOL
a Nikon
(Nikon,
scanning
electron
(SEM)
USA,
Digital SLR camera (Nikon, Mississauga, ON, Canada) with an AF-S DX Micro-NIKKOR 40 mm
St. Hubert, QC, Canada) was used to take micrograph images.
f/2.8 G lens (Nikon, Mississauga, ON, Canada). A JEOL 6400 scanning electron microscope (SEM)
(JEOL
USA,
Hubert, Canada) was used to take micrograph images.
3. Results
andSt.Discussion

Experiments
were conducted to study the influence of the surrounding temperature and humidity
3. Results
and Discussion
levels, the paper machine direction, and the dimensions of the paper strips including strip length and
Experiments were conducted to study the influence of the surrounding temperature and
strip width on the wicking behaviour of fluid in a paper strip fed from a large fluid reservoir.
humidity levels, the paper machine direction, and the dimensions of the paper strips including strip
andofstrip
width on
the wicking behaviour of fluid in a paper strip fed from a large
3.1.length
Influence
Temperature
on Wicking
fluid reservoir.
Figure 2 shows the experimental results for fluid flow in paper strips with a 10 mm width and a
of Temperature
on Wicking
45 3.1.
mmInfluence
length, tested
in temperature
conditions ranging from 15 to 45 C at a fixed relative humidity
of 30%.Figure
After 2analyzing
the data from the experiments, it was clear that the temperature of the fluid
shows the experimental results for fluid flow in paper strips with a 10 mm width and a
in 45
themm
reservoir
influenced
the wicking
behaviour
rather
the air
temperature
(Ta ).
f ) was
length,(T
tested
inwhat
temperature
conditions
ranging
from 15 to
45 Cthan
at a fixed
relative
humidity
It can
be
observed
in
Figure
2
that
the
flow
profiles
follow
the
typical
Washburn
behaviour
and
of 30%. After analyzing the data from the experiments, it was clear that the temperature of the fluidthe
distance
travelled(T
byf) was
the fluid
is proportional
to behaviour
the squarerather
root of
time
in the reservoir
whatfront
influenced
the wicking
than
the[35]:
air temperature (Ta).
It can be observed in Figure 2 that the flow profiles
follow the typical Washburn behaviour and the
d
cos
distance travelled by the fluid front is proportional
to the square root of time [35]:
rc t
(1)
x
2
cos
(1)
=

2
where x is distance, is the surface tension, is the contact angle, is viscosity of the fluid, rc
where
is distance,
is the
surface
tension,
the also
contact
is that
viscosity
of thewicking
fluid, time
is
is the
average
pore radius,
and
t is the
time. Itiscan
be angle,
observed
the total

the
average
pore
radius,
and

is
the
time.
It
can
also
be
observed
that
the
total
wicking
time
decreased as temperature increased. In conditions of 15 C fluid and air temperature, a wicking time
as temperature
increased.
Infor
conditions
15 Cto
fluid
andaair
temperature,
a wicking
of decreased
approximately
11 min was
required
the fluidoffront
travel
distance
of 45 mm.
With time
a fluid
of
approximately
11
min
was
required
for
the
fluid
front
to
travel
a
distance
of
45
mm.
With
a fluid

temperature of 35 C (air temperature of 45 C), wicking time was reduced by three minutes,
and a
temperature of 35 C (air temperature of 45 C), wicking time was reduced by three minutes, and a
total duration of approximately 8 min was required for the fluid front to travel 45 mm in the paper strip.
total duration of approximately 8 min was required for the fluid front to travel 45 mm in the paper
These results demonstrate that the time required for the fluid front to travel the length of the paper strip
strip. These results demonstrate that the time required for the fluid front to travel the length of the
decreases as temperature increases, and correspondingly speed increases. This increase in wicking
paper strip decreases as temperature increases, and correspondingly speed increases. This increase
speed can be attributed to the decrease in the viscosity of water as its temperature increases. To analyze
in wicking speed can be attributed to the decrease in the viscosity of water as its temperature
thisincreases.
justification,
Figure 3 shows a comparison of the experimental results for wicking distance at 2, 4,
To analyze this justification, Figure 3 shows a comparison of the experimental results for
and
6
min
versus
theatinverse
of viscosity
varying
fluid
temperatures.
Thisfluid
correlation
between
wicking distance
2, 4, androot
6 min
versus theatinverse
root
of viscosity
at varying
temperatures.
wicking
distance and
viscosity
follows
the Washburn
model
Equation
(1), wheremodel
the wicking
distance
This correlation
between
wicking
distance
and viscosity
follows
the Washburn
Equation
(1),
is inversely
proportional
to
the
root
of
viscosity.
The
corresponding
dependence
of
wicking
distance
where the wicking distance is inversely proportional to the root of viscosity. The corresponding
and
the viscosity
on temperature
in Figure
confirms
that the
fluidin
viscosity
dependence
of term
wicking
distance and shown
the viscosity
term3on
temperature
shown
Figure 3accounts
confirmsfor
thethat
observed
variations
temperature.
the fluid
viscosity with
accounts
for the observed variations with temperature.

Figure
Experimentalresults
resultsfor
forwicking
wicking in
in 11 CHR
CHR strips
inin
length
at at
Figure
2. 2.
Experimental
strips10
10mm
mmininwidth
widthand
and4545mm
mm
length

a) and fluid (Tf) temperature conditions varying from 15 to 45 C.

ambient
(T
ambient (Ta ) and fluid (Tf ) temperature conditions varying from 15 to 45 C.

Micromachines 2016, 7, 73
Micromachines 2016, 7, 73
Micromachines 2016, 7, 73

5 of 13
5 of 13

5 of 13

Figure 3. Comparison of the inverse root of viscosity (from the Washburn equation) to
experimental
data for of
wicking
distance
at viscosity
2,root
4, and
6viscosity
minthe
inWashburn
1(from
CHR, the
10equation)
mm
width
strips
at varying
Figure
3. Comparison
the
inverse
of (from
Washburn
equation)
to
Figure
3. Comparison
theofinverse
root of
to experimental
data
data at
for2,wicking
2, 4, 10
and
6 min
in 1strips
CHR,at10
mm width
strips atconditions.
varying
temperature
conditions.
forexperimental
wicking distance
4, and 6distance
min in 1atCHR,
mm
width
varying
temperature
temperature conditions.

3.2.
3.2. Influence
Influence of
of Humidity
Humidity on
on Wicking
Wicking
3.2. Influence of Humidity on Wicking
Wicking
in paper
paper strips
strips with
with the
the same
same dimensions
dimensions as
as the
the temperature
temperature tests
tests (10
(10 mm
mm width
width and
and
Wicking in
Wicking
in paper
strips
with the same
dimensions
as from
the temperature
tests
(10 mm
width and
45
mm
length)
were
tested
in
humidity
conditions
ranging
30%
to
85%
relative
humidity
(H),
at
45 mm length) were tested in humidity conditions ranging from 30% to 85% relative humidity (H),
45
mm
length)
were
tested
in
humidity
conditions
ranging
from
30%
to
85%
relative
humidity
(H),
at
a fixed
airair
temperature
ofof2020C,
results
are
shown
in Figure
4. Despite
the
C,and
at
temperature
andthe
theexperimental
experimentalresults
resultsare
areshown
shown
Figure
Despite
aa fixed
fixed air
temperature
of 20 C,
and
the
experimental
inin
Figure
4. 4.
Despite
thethe
variations
in
humidity,
the
wicking
time
for
the
fluid
to
fill
the
full
length
of
the
paper
strip
had
a
variations
ininhumidity,
fluid to
to fill
fill the
thefull
fulllength
lengthofofthe
thepaper
paper
strip
had
variations
humidity,the
thewicking
wickingtime
time for
for the
the fluid
strip
had
a a
consistent
value
of
approximately
7
min.
There
was
no
observable
increase
or
decrease
in
wicking
consistent
value
was no
no observable
observableincrease
increaseorordecrease
decrease
wicking
consistent
valueofofapproximately
approximately77min.
min. There
There was
in in
wicking
velocity
beyond
the
experimental
error
that
corresponded
to
increases
in
humidity
for
the
conditions
velocity
beyond
the
toincreases
increasesininhumidity
humidityforfor
the
conditions
velocity
beyond
theexperimental
experimentalerror
errorthat
that corresponded
corresponded to
the
conditions
of our
our experiment.
experiment. This
result
indicates
that
PADs
developed for
for applications
applications fed
from
large
fluid
of
This
result
indicates
that
PADs
developed
fed
from
large
fluid
of our experiment. This result indicates
PADs developed for applications fed from large fluid
reservoirs
and
settings
with
varying
humidity
levels
can
yield
consistent
flow
behaviour
regardless
reservoirs
and
settings
with
varying
reservoirs
and
settings
with
varyinghumidity
humiditylevels
levelscan
canyield
yieldconsistent
consistentflow
flow behaviour
behaviour regardless of
of
fluctuating
levels
of
humidity.
fluctuating
levels
of humidity.
of fluctuating
levels
of humidity.

Figure
4.
Experimentalresults
resultsfor
forwicking
wicking in
in 11 CHR
CHR strips
mm
in in
length,
forfor
Figure
4.
Experimental
strips10
10mm
mminin
inwidth
widthand
and4545
45
mm
length,
Figure
4. Experimental
results for
wicking in
1 CHR strips
10
mm
width
and
mm
in length,
for
humidity
(H)
conditions
varying
from
30%
to
85%.
humidity (H) conditions varying from 30% to 85%.
humidity (H) conditions varying from 30% to 85%.

3.3. Influence of Machine Direction on Wicking

3.3. Influence
Influence of
of Machine
Machine Direction
Direction on
3.3.
on Wicking
Wicking
During the paper production process the cellulose pulp is laid into sheets, and it has been
During the
process
the the
cellulose
pulppulp
is laidisinto
and it has
been has
observed
During
thepaper
paperproduction
production
process
cellulose
laidsheets,
into sheets,
been
observed
that
paper
fibres
tend to align
parallel
to
the direction
of the
machine
[36,37].and
Theitmachine
that
paperthat
fibres
tendfibres
to align
parallel
to the
direction
the machine
[36,37].
The[36,37].
machineThe
direction
of
observed
paper
tend
to align
parallel
to theofdirection
of the
machine
machine

Micromachines 2016,
73
Micromachines
2016, 7,
7, 73

of 13
13
66 of

direction of the chromatography paper was labelled by the manufacturer, so an investigation to

the chromatography paper was labelled by the manufacturer, so an investigation to determine any
determine any potential impacts of fibre arrangement could be conducted. Hypotheses for the
potential impacts of fibre arrangement could be conducted. Hypotheses for the specific fluid transport
specific fluid transport mechanism in paper include wicking through capillary-like structures
mechanism in paper include wicking through capillary-like structures formed by an alignment of
formed by an alignment of pores, or wicking along adjacent fibres [38]. It is expected that fluid
pores, or wicking along adjacent fibres [38]. It is expected that fluid flowing in the direction of the
flowing in the direction of the fibres will encounter fewer obstructions from the fibres and therefore
fibres will encounter fewer obstructions from the fibres and therefore travel through the paper strip in
travel through the paper strip in a shorter time.
a shorter time.
The machine direction is labelled in the SEM micrograph inset of Figure 5 for Whatman 17 CHR
The machine direction is labelled in the SEM micrograph inset of Figure 5 for Whatman 17 CHR
paper, and the tendency for fibre alignment parallel to machine direction can be observed. Figure 5
paper, and the tendency for fibre alignment parallel to machine direction can be observed. Figure 5
shows the results of a comparison for fluid flow in the machine direction and cross direction for
shows the results of a comparison for fluid flow in the machine direction and cross direction for
Whatman 17 CHR paper with a width (w) of 10 mm. It can be seen that fluid flow in the machine
Whatman 17 CHR paper with a width (w) of 10 mm. It can be seen that fluid flow in the machine
direction is faster than in the cross direction and this observation becomes more distinct as wicking
direction is faster than in the cross direction and this observation becomes more distinct as wicking
distance increases. In Figure 5, it can be seen that the fluid in the machine direction strip wicked to a
distance increases. In Figure 5, it can be seen that the fluid in the machine direction strip wicked to
length of 45 mm 30% faster than in the cross direction strip. For our experimental investigations we
a length of 45 mm 30% faster than in the cross direction strip. For our experimental investigations
chose to use the machine direction, since the chromatography paper is designed to operate in the
we chose to use the machine direction, since the chromatography paper is designed to operate in the
machine direction and the fluid flow is less inhibited.
machine direction and the fluid flow is less inhibited.

Figure 5.
5. Experimental
Experimental observations
observations for
for wicking
wicking in
in aa 10
10 mm
mm wide
wide 17
17 CHR
CHR paper
paper strip,
strip, in
in the
the machine
machine
Figure
direction
(MD)
and
cross
machine
direction
(CMD).
direction (MD) and cross machine direction (CMD).

3.4. Wicking
Wicking in
in Paper
Paper Strips
Strips of
of Varying
Varying Lengths
3.4.
Lengths
Paper-based devices
used for
for analytical
analytical testing
testing in
in situations
situations where
where there
there is
is an
an abundance
abundance
Paper-based
devices can
can be
be used
of sample
fluid volume,
for example
in water
testing, and
and longer
longer paper
paper strip
strip lengths
lengths can
can be
be
of
sample fluid
volume, for
example in
water quality
quality testing,
exploited
for
a
number
of
functions
including
additional
reaction
steps
or
to
increase
the
concentration
exploited for a number of functions including additional reaction steps or to increase the
of an analyte by
flowing
more
fluid over
a reaction
zone.
To examine
theexamine
impacts the
of paper
strip
concentration
of an
analyte
by flowing
more
fluid over
a reaction
zone. To
impacts
of
dimensions
on fluid wicking,
of varying
lengths lengths
(L) were(L)tested
ascertain
if fluid
flow
paper
strip dimensions
on fluid strips
wicking,
strips of varying
were to
tested
to ascertain
if fluid
behaviour
in paper
is influenced
by the length
the strip.
Figure
experimental
results for
flow
behaviour
in paper
is influenced
by theoflength
of the
strip.6 shows
Figure the
6 shows
the experimental
fluid flow
paper
strips
of varying
and a length
fixed width
10 mm,
withofthe
at
results
forin
fluid
flow
in paper
stripslength
of varying
and aoffixed
width
10 fluid
mm, temperature
with the fluid
C), 25% humidity, and fed from a large fluid reservoir. Experimental
20 C (ambient
of 22 temperature
temperature
attemperature
20 C (ambient
of 22 C), 25% humidity, and fed from a large fluid
data demonstrates
that as
thedemonstrates
lengths of thethat
paper
increased
from
25 tostrips
65 mm,
there was
no
reservoir.
Experimental
data
as strips
the lengths
of the
paper
increased
from
subsequent
or no
decrease
in flow
velocityorbeyond
the
Asthe
such,
length is a
25
to 65 mm,increase
there was
subsequent
increase
decrease
inexperimental
flow velocityerror.
beyond
experimental
flexible
PADs,
andparameter
can be adjusted
to meet
larger
requirements
without
error.
Asdesign
such, parameter
length is a for
flexible
design
for PADs,
and
can size
be adjusted
to meet
larger
impacting
the wicking
behaviour
in a the
paper-based
device. in a paper-based device.
size
requirements
without
impacting
wicking behaviour

Micromachines 2016,
2016, 7,
73
Micromachines
7, 73

of 13
13
77 of

Figure 6.
6. Experimental
Experimentalresults
resultsfor
forwicking
wicking
1 CHR
strips
of lengths
(L) varying
65 mm,
Figure
inin
1 CHR
strips
of lengths
(L) varying
fromfrom
25 to25
65to
mm,
and
and
a
width
of
10
mm.
a width of 10 mm.

3.5. Wicking
Wicking in
in Paper
Paper Strips
Strips of
of Varying
Varying Widths
Widths
3.5.
In situations
situations where
where paper-based
paper-based tests
tests are
are required
required and
and there
there is
is an
an abundance
abundance of
of fluid
fluid volume,
volume,
In
wider
bebe
exploited
for for
a number
of functions
including
larger larger
test signal
wider paper
paperstrip
striplengths
lengthscan
can
exploited
a number
of functions
including
test output
signal
for easefor
of reading,
easier handling
of the test
or to
increase
amount
fluid collected
output
ease of reading,
easier handling
of by
thethe
testuser,
by the
user,
or to the
increase
theof
amount
of fluid
in a givenintime.
Impacts
of the paper
strip
width
investigated
for straight
paper strips
collected
a given
time. Impacts
of the
paper
stripwere
width
were investigated
for straight
paperwith
stripsa
constant
width and
fedand
from
a large
reservoir.
In order In
to order
examine
the influence
of the width
of
with
a constant
width
fed
from fluid
a large
fluid reservoir.
to examine
the influence
of the
the
paper
strip
on
fluid
flow
behaviour,
we
used
paper
strips
with
a
length
of
45
mm
and
a
range
of
width of the paper strip on fluid flow behaviour, we used paper strips with a length of 45 mm and a
widths
5 tofrom
40 mm
increments
of 5 mm. of 5 mm.
range offrom
widths
5 toin40
mm in increments
Paper-based devices
with
width
values
thatthat
fluctuate
along
the
Paper-based
deviceshave
havebeen
beendesigned
designedand
andanalyzed
analyzed
with
width
values
fluctuate
along
paper
strip
as
a
method
to
control
fluid
flow
by
altering
the
area
exposed
to
the
imbibition
front
and
the paper strip as a method to control fluid flow by altering the area exposed to the imbibition front
thus thus
slowing
downdown
the wicking
of the of
fluid
width/area)
or speeding
up the flowing
and
slowing
the wicking
the(with
fluidincreased
(with increased
width/area)
or speeding
up the
fluid (with
decreased
width/area)
[24,26]. [24,26].
It is expected
that no changes
will occur
in occur
the wicking
flowing
fluid
(with decreased
width/area)
It is expected
that no changes
will
in the
behaviour
of
the
fluid
when
the
width
is
varied
for
straight
cut
strips
with
a
constant
width
value,
wicking behaviour of the fluid when the width is varied for straight cut strips with a constant width
since the
area
continuous.
Previous
studies
have have
found
that that
different
stripstrip
widths
influence
the
value,
since
theisarea
is continuous.
Previous
studies
found
different
widths
influence
flow
in
straight
paper
strips
with
hydrophobic
barriers
resulting
in
varying
wicking
times
[39,40].
the flow in straight paper strips with hydrophobic barriers resulting in varying wicking
Songok
et al. investigated
channelsmall
widths
of 0.5,widths
1.0 andof1.5
and 1.5
found
variation
in
times
[39,40].
Songok et al.small
investigated
channel
0.5,mm,
1.0 and
mm,the
and
found the
wicking
to be speed
due toto
the
theshape
droplet
thatdroplet
was feeding
the feeding
channelthe
from
abovefrom
[39].
variationspeed
in wicking
beshape
due toofthe
of the
that was
channel
Hong
and
Kim
investigated
small
channel
widths
of
1,
2,
and
4
mm
and
concluded
that
the
effects
of
above [39]. Hong and Kim investigated small channel widths of 1, 2, and 4 mm and concluded that
hydrophobic
channel boundaries
significant
a widthwith
on the
orderon
of the
1 mm
[40].
slower
the
effects of hydrophobic
channelare
boundaries
arewith
significant
a width
order
ofThe
1 mm
[40].
flowslower
in channels
smallerwith
widths
was attributed
toattributed
the surfacetotension
forcestension
at the hydrophobic
The
flow with
in channels
smaller
widths was
the surface
forces at the
boundaries that
oppose the
and the
therefore
expect nothey
variation
stripsfor
with
cut
hydrophobic
boundaries
thatflow,
oppose
flow, they
and therefore
expectfornopaper
variation
paper
edges.with
Paper
strips
with
cut edges
larger
than
a fewlarger
millimeters
not been analyzed
to
strips
cut
edges.
Paper
strips and
withwidths
cut edges
and
widths
than ahave
few millimeters
have not
determine
the
influence
of
width
on
the
fluid
flow
behaviour
and
we
provide
our
observations
here
been analyzed to determine the influence of width on the fluid flow behaviour and we provide our
for straight cut
strip
from
to 40widths
mm. from 5 to 40 mm.
observations
here
forwidths
straight
cut 5strip
experimental
results
for fluid
flow in
PADs
of varying
using
Whatman
Figure 77shows
showsthethe
experimental
results
for fluid
flow
in PADs
of width
varying
width
using
1Whatman
CHR, which
has
a
thickness
of
0.18
mm.
The
total
wicking
time
for
paper
strips
of
45
mm
in
length
1 CHR, which has a thickness of 0.18 mm. The total wicking time for paper strips of 45 mm
was
observed
to decrease
the width
increased,
corresponding
to an increase
wickingofspeed
for
in length
was observed
to as
decrease
as the
width increased,
corresponding
to anofincrease
wicking
increasing
width. This
confirms
the width
fluid flow
thin strips,
speed
for increasing
width.
Thisthat
confirms
that influences
the width the
influences
thebehaviour
fluid flowfor
behaviour
for with
thin
diminishing
dependencydependency
as the stripas
width
increases.
The same The
set of
experiments
were repeated
strips,
with diminishing
the strip
width increases.
same
set of experiments
were
with
thicker
paper (Whatman
17 CHR with
a thickness
of 0.70 mm)
to give
repeated
withchromatography
thicker chromatography
paper (Whatman
17 CHR
with a thickness
of 0.70
mm)the
to
results
in
Figure
8.
For
the
thicker
paper,
no
variations
associated
with
width
were
observed
within
give the results in Figure 8. For the thicker paper, no variations associated with width were observed
the experimental
error. Iterror.
is therefore
crucial to
consider
the influence
of widthofonwidth
the flow
rates
for
within
the experimental
It is therefore
crucial
to consider
the influence
on the
flow

Micromachines 2016, 7, 73
Micromachines
73
Micromachines 2016,
2016, 7,
7, 73

8 of 13
88 of
of 13
13

rates for thin paper strips with larger widths (greater than 5 mm) when larger PADs are being
rates for thin
paper strips
larger widths
(greater
than
5 mm) when larger PADs are being
designed
in situations
with with
an abundance
of sample
fluid
volume.
thin paper strips with larger widths (greater than 5 mm) when larger PADs are being designed in
designed in situations with an abundance of sample fluid volume.
situations with an abundance of sample fluid volume.

Figure
Figure 7.
7. Experimental
Experimental results
results for
for wicking
wicking in
in 11 CHR
CHR strips
strips of
of widths
widths (w)
(w) varying
varying from
from 55 to
to 40
40 mm,
mm, and
and
aaFigure
length
of
45
mm.
7.
Experimental
results
for
wicking
in
1
CHR
strips
of
widths
(w)
varying
from
5
to
40
mm,
and
length of 45 mm.
a length of 45 mm.

Figure 8.
8. Experimental
Experimentalresults
resultsfor
forwicking
wicking
CHR
strips
of widths
varying
5 tomm,
30 mm,
Figure
inin
1717
CHR
strips
of widths
(w) (w)
varying
fromfrom
5 to 30
and
and
a
length
of
45
mm.
aFigure
length8.ofExperimental
45 mm.
results for wicking in 17 CHR strips of widths (w) varying from 5 to 30 mm, and
a length of 45 mm.

Bohm et
et al.
al. [41]
[41] also
also observed
observed slower
slower flow
flow in
in channels
channels with
with smaller
smaller widths
widths for
for paper
paper strips
strips with
with
Bohm
Bohm et barriers
al. [41] also
flow
in
channels
withthe
smaller
widthson
forwidth
paperdiminished
strips with
hydrophobic
and
widths
11 to
found
dependence
hydrophobic
barriers
andobserved
widths of
ofslower
to 55 mm,
mm, and
and
found that
that
the
dependence
on
width
diminished
hydrophobic
and widths
1 to 5 mm,
and
foundnumber
that theof
on width
diminished
above
44 mm.
This
was
attributed
to
pores
were
terminated
at
above
mm. barriers
This behaviour
behaviour
was of
attributed
to the
the large
large
number
ofdependence
pores that
that were
terminated
at
above
4 mm.walls
This(dead-end
behaviourpores)
was
attributed
tohydrophobic
the
large number
ofwas
pores
that were
terminated
at
the
channel
walls
(dead-end
pores)
wherethe
the
hydrophobic
barrier
was
located,
interrupting
the
channel
where
barrier
located,
thusthus
interrupting
the
(dead-end
pores)
where
thetime
hydrophobic
thus
interrupting
the
flow
andwalls
resulting
in slower
fluid
flow
(longer
time
tobarrier
fill
a was
given
volume
of observations
space). Our
flowchannel
and resulting
in
slower
fluid
flow
(longer
to fill
a given
volume
oflocated,
space).
Our
the flow
andwere
resulting
in slower
fluid
flow
time
todependence
fill a given
of
Our
observations
with
edges
and
we found
the
tovolume
extend
widths
were
for strips
withfor
cutstrips
edges
and cut
we
found
the(longer
dependence
to
extend
beyond
widthsbeyond
of 4space).
mm,
up
to
observations
for
strips
with
cut
edges and
wemechanism
found the
dependence
toaextend
widths
of
4mm,
mm,however
up towere
40the
mm,
however
the
dead-end
pore
also provides
plausible
explanation
40
dead-end
pore
mechanism
also
provides
a plausible
explanation
forbeyond
the
behaviour
of 4observed.
mm,behaviour
up toA40video
mm,
however
the
pore mechanism
also provides
plausible explanation
for
the
we is
observed.
A
video
is included
ininformation
the supplementary
information
showing
we
included
indead-end
the supplementary
showingaside-by-side
tests
on a
for the behaviour
we
observed.
A avideo
is
included
in the
supplementary
showing
side-by-side
tests
a benchtop
with
a 10
mm
strip
beside
ademonstrates
35
mm strip. the
Theinformation
videodependence
demonstrates
benchtop
with
a 10on
mm
strip beside
35
mm
strip.
The
video
width
by
side-by-side
tests
on awicking
benchtop
withfor
a 10
mm
strip
beside
a under
35 mm
strip.
videowidths
demonstrates
the
widththe
dependence
by showing
the
varying
wicking
speed
for
two The
different
showing
varying
speed
two
different
widths
identical
conditions.
The under
video
the width
dependence
showing
the varying
wicking speed
for
two that
different
widths
identical
conditions.
Theby
video
alsodead-end
provides
substantiation
ofsince
the dead-end
pore
explanation
since
also
provides
substantiation
of the
pore
explanation
it shows
the
fluid
flowunder
at
the
identical conditions. The video also provides substantiation of the dead-end pore explanation since

Micromachines 2016, 7, 73
Micromachines 2016, 7, 73

9 of 13
9 of 13

it shows that the fluid flow at the middle of the paper strip progressed quicker than the flow at the
edges, indicating that the fluid near the cut edges was hindered. Since the edges are hindering the
middle of the paper strip progressed quicker than the flow at the edges, indicating that the fluid near
flow, wider strips would experience less hindrance than narrower strips and thus a quicker flow,
the cut edges was hindered. Since the edges are hindering the flow, wider strips would experience
which provides corroboration for the dead-end pore explanation.
less hindrance than narrower strips and thus a quicker flow, which provides corroboration for the
dead-end
pore explanation.
3.6.
Post-Wetting
Flow
3.6. Post-Wetting
Flow
Our most unexpected
observation was fluid flow in the paper strip after wetting was complete.
The paper
strips
remained observation
in contact with
reservoir
the liquid
frontwetting
had travelled
the full
Our most
unexpected
wasthe
fluid
flow in after
the paper
strip after
was complete.
length
of
the
strip.
Though
no
additional
fluid
transport
was
expected
due
to
capillary
forces,
The paper strips remained in contact with the reservoir after the liquid front had travelled thesince
full
wetting
was
complete,
further
flow,
or
post-wetting
flow,
was
observed
in
the
paper
strip.
length of the strip. Though no additional fluid transport was expected due to capillary forces, since
Post-wetting flow caused an increase in the amount of dye in the paper strip, and was detected
wetting was complete, further flow, or post-wetting flow, was observed in the paper strip. Post-wetting
visually as a darkening in the colour of the paper strip. ImageJ software was used to analyze images
flow caused an increase in the amount of dye in the paper strip, and was detected visually as a
of post-wetting flow to quantify our observations and rule out other causes such as evaporation or
darkening in the colour of the paper strip. ImageJ software was used to analyze images of post-wetting
diffusion.
flow to quantify our observations and rule out other causes such as evaporation or diffusion.
Figure 9 compares the colour intensity of a paper strip at the moment of complete wetting to the
Figure 9 compares the colour intensity of a paper strip at the moment of complete wetting to the
colour intensity 17 min after complete wetting. Colour intensity is measured by representing each
colour intensity 17 min after complete wetting. Colour intensity is measured by representing each pixel
pixel in the image with a numerical value between 0 and 255, where 0 is the value assigned to the
in the image with a numerical value between 0 and 255, where 0 is the value assigned to the darkest
darkest pixels and 255 is assigned to the brightest red pixels. The colour profile shows a shift to the
pixels and 255 is assigned to the brightest red pixels. The colour profile shows a shift to the left (lower
left (lower pixel values) that reflects the darkening in colour due to the increase in dye within the
pixel values) that reflects the darkening in colour due to the increase in dye within the paper strip.
paper strip. The darkening can also be observed in the paper strips in the inset of Figure 10. The
The darkening can also be observed in the paper strips in the inset of Figure 10. The overall darkening
overall darkening of the whole paper strip confirms that the amount of dye increased due to
of the whole paper strip confirms that the amount of dye increased due to post-wetting flow and not
post-wetting flow and not due to diffusion, since diffusion would result in a redistribution of the dye
due to diffusion, since diffusion would result in a redistribution of the dye rather than an overall
rather than an overall increase. To provide further verification of whether the colour change was due
increase. To provide further verification of whether the colour change was due to increased flow from
to increased flow from the reservoir or the diffusion of dye particles, a fully wetted strip was
the reservoir or the diffusion of dye particles, a fully wetted strip was removed from the reservoir
removed from the reservoir immediately after the fluid front reached the end of the strip and the
immediately after the fluid front reached the end of the strip and the colour was compared to a strip
colour was compared to a strip that remained in the reservoir. A slight redistribution of the dye was
that remained in the reservoir. A slight redistribution of the dye was detected as the removed strip
detected as the removed strip evened in colour, however, there was no distinct darkening in the
evened in colour, however, there was no distinct darkening in the colour of this strip. Additionally, the
colour of this strip. Additionally, the mass of both strips was measured 18 min after wetting. Both
mass of both strips was measured 18 min after wetting. Both strips were kept in the same location
strips were kept in the same location to ensure that both would undergo the same amount of
to ensure that both would undergo the same amount of evaporation and rule out evaporation as a
evaporation and rule out evaporation as a cause of the change in fluid volume within the strip. The
cause of the change in fluid volume within the strip. The mass of the paper strip that remained in the
mass of the paper strip that remained in the reservoir was 36% higher than the paper strip that was
reservoir was 36% higher than the paper strip that was removed from the reservoir, thus confirming
removed from the reservoir, thus confirming the presence of post-wetting flow.
the presence of post-wetting flow.

Figure 9.
9. Comparison
Comparison of
of colour
colour profiles
profiles of
of the
the red
red tone
tone in
in 11 CHR
CHR paper
paper strips
strips (a)
(a) upon
upon wetting
wetting for
for the
the
Figure
full length
length of
full
of the
the strip,
strip, and
and (b)
(b) 17
17 min
min after
after wetting
wetting to
to detect
detect flow
flow after
after wetting.
wetting.

Micromachines 2016, 7, 73
Micromachines 2016, 7, 73

Micromachines 2016, 7, 73

10 of 13

10 of 13

10 of 13

Figure
CHR paper
paperstrip
stripafter
afterwetting.
wetting.
Figure10.
10.Colour
Colourintensity
intensity of
of a
a 11 CHR

To determine the post-wetting flow behaviour as a function of time we analyzed the colour
To determine the post-wetting
flow
behaviour
as paper
a function
ofwetting.
time we analyzed the colour
Figure 10. Colour
intensity
of a 1 CHR
strip after
intensity of the paper strip after wetting was complete. The maximum value for red pixels is the
intensity of the paper strip after wetting was complete. The maximum value for red pixels is
numerical
value assigned
by the software
to the brightest
red pixel
in the image.
This
value was
To determine
the post-wetting
behaviour
a functionred
of time
the colour
the numerical
value assigned
by the flow
software
to theasbrightest
pixelwe
in analyzed
the image.
This value
measured
every
minute
after
saturation,
where
a
lower
value
represents
a
darker
red
intensity of every
the paper
strip after
after saturation,
wetting was where
complete.
The maximum
value for a
red
pixels red
iscolour
the
was measured
minute
a lower
value represents
darker
colour
corresponding
to an
increase
concentration,
and these
Figure
The
numerical value
assigned
byin
thedye
software
to the brightest
red values
pixel inare
theplotted
image. in
This
value10.
was
corresponding
to an increase
in dye
concentration,
and these
values are plotted
in Figure
10. The
data
data
demonstrates
decreasing
relationshipwhere
between
time value
and brightness
red colour
in the
measured
every aminute
after saturation,
a lower
representsofa the
darker
red colour
demonstrates
a
decreasing
relationship
between
time
and
brightness
of
the
red
colour
in
the
paper
strip
paper
strip that to
levels
off after
approximately
15and
min.
This
dataare
indicates
there10.exists
corresponding
an increase
in dye
concentration,
these
values
plotted that
in Figure
The a
that
levels
offquantity
after approximately
min.
Thispaper
data strip
indicates
that
there existsthe
a maximum
quantity of
maximum
ofa water
that 15
will
fill the
during
data demonstrates
decreasing
relationship
between
time
and post-wetting
brightness of flow.
red colour in the
waterpaper
that
will
theat
paper
strip
during
post-wetting
flow.
A close
look
the fibre
structure
of paper,15asmin.
seenThis
in the
micrograph
in Figure
stripfill
that
levels
off
after
approximately
dataSEM
indicates
that there
exists a11,
A
close
look
at
the
fibre
structure
of
paper,
as
seen
in
the
SEM
micrograph
in
Figure
11,
that
maximum
quantity pores
of water
will fill
during
post-wetting
flow.to be highly shows
shows
that interfibre
arethat
created
by the
the paper
spacesstrip
between
fibres,
which tend
variable
interfibre
pores
are
created
by
the
spaces
between
fibres,
which
tend
to
be
highly
variable
in
size
[38,42].
A close Alook
at by
theRoberts
fibre structure
of paper,
seen
infilling
the SEM
micrograph
in primary
Figure 11,
in size [38,42].
study
et al. observed
thatasthe
bulk
of pores
is not the
flow
shows
pores
are
by the
spaces
between
fibres,
which
tend
to be flow
highly
variable
A study
by that
Roberts
et al.rather
observed
that
bulk
filling
of pores
is not
the
primary
mechanism
mechanism
ininterfibre
paper,
it iscreated
flow the
along
the
channels
caused
by
fibre
overlap,
in other
words, in
in rather
size [38,42].
Afilm
study
by Roberts
et al. observed
the
bulk
filling
ofinpores
is words,
not
primary
flow
capillary
driven
flow.
Thechannels
cellulose
fibresthat
also
contain
pores
within
thethefibre,
known
as
paper,
it is flow
along
the
caused
by
fibre
overlap,
other
capillary
driven
mechanism
in
paper,
rather
it
is
flow
along
the
channels
caused
by
fibre
overlap,
in
other
words,
intrafibre
pores
[37,38].
Filling
of
these
intrafibre
pores
leads
to
fibre
swelling,
while
the
capillary
film flow. The cellulose fibres also contain pores within the fibre, known as intrafibre pores [37,38].
capillary
drivenbetween
film flow.
Thecan
fibresdeformation
also contain[37,43,44].
within
thethese
fibre,fibre
known
as
forces
a liquid
fibres
While
swelling
Filling
ofofthese
intrafibre
pores
leadscellulose
tocause
fibrefibre
swelling,
while thepores
capillary
forces
of a liquid
between
intrafibre
pores
[37,38].
Filling
of
these
intrafibre
pores
leads
to
fibre
swelling,
while
the
capillary
effects
not fibre
immediately
apparent
during While
imbibition,
provide
an explanation
for the
fibres
canare
cause
deformation
[37,43,44].
thesethey
fibrecould
swelling
effects
are not immediately
forces of a liquid between fibres can cause fibre deformation [37,43,44]. While these fibre swelling
presence
of
post-wetting
flow.
Movement
of
the
fibres
due
to
stretching
or
relaxation
caused
apparent
during imbibition, they could provide an explanation for the presence of post-wetting by
flow.
effects are not immediately apparent during imbibition, they could provide an explanation for the
elastic
forces
could
also
account
for fluidor
flow
within the
paperby
after
wetting.
Movement of the
fibres
due
to stretching
relaxation
caused
elastic
forces could also account for
presence of post-wetting flow. Movement of the fibres due to stretching or relaxation caused by

fluid elastic
flow within
the paper
after wetting.
forces could
also account
for fluid flow within the paper after wetting.

Figure 11. Scanning electron micrographs of cellulose fibre networks in (a) 1 CHR at 100
Figure
11. Scanning
electron micrographs
of cellulose
fibre networks
in (a)at1100
CHRmagnification,
at 100
Figure
11. Scanning
of CHR
cellulose
fibreand
networks
in (a)at1 10,000.
CHR
magnification,
(b) electron
1 CHR atmicrographs
10,000, (c) 17
at 100,
(d) 17 CHR
magnification, (b) 1 CHR at 10,000, (c) 17 CHR at 100, and (d) 17 CHR at 10,000.
(b) 1 CHR at 10,000, (c) 17 CHR at 100, and (d) 17 CHR at 10,000.

Micromachines 2016, 7, 73

11 of 13

For design of precise PADs when using a large fluid reservoir, it will be important to consider
the post-wetting effects in relation to the intensity of colorimetric signals by controlling the time of
submersion and elapsed time prior to test readout by the user. This phenomenon is of particular
interest for paper-based microfluidic applications where high sample volumes are readily available.
4. Conclusions
In this paper, we investigated the impact of surrounding conditions such as temperature and
humidity, paper machine direction, strip length, and strip width on fluid flow behaviour in paper strips
fed from a large fluid reservoir. Using experimental data, we determined that length and humidity
do not have a significant impact on flow behaviour in paper strips for the conditions we tested.
We noticed quicker wicking for flow parallel to the machine direction of the paper compared to the
cross (perpendicular) direction. Our experimental results indicated that the wicking time decreases as
temperature increases in proportion to decreases in the inverse root of the fluid viscosity. Additionally,
we observed that wicking time also decreases with increasing width for thin (1 CHR) paper strips,
a dependency which diminishes as width increases. Width was not found to influence the wicking
behaviour for the thicker paper strips (17 CHR). We also observed post-wetting flow when the paper
strips remained in contact with a fluid reservoir, which should be considered during design of PADs
used in conjunction with large sample volumes, for example, in environmental testing applications.
Supplementary Materials: The following are available online at http://www.mdpi.com/2072-666X/7/5/73/s1,
Video S1: Side-by-side tests on a benchtop with a 10 mm strip beside a 35 mm strip.
Acknowledgments: The authors gratefully acknowledge funding support from Grand Challenges Canada
through Stars in Global Health (Grant #: 0494-01-10).
Author Contributions: Noosheen Walji designed and performed the experiments, analyzed the data, and wrote
the paper. Brendan D. MacDonald supervised the work and edited the manuscript.
Conflicts of Interest: The authors declare no conflict of interest.

References
1.
2.
3.
4.
5.

6.

7.

8.

9.
10.

Martinez, A.W.; Phillips, S.T.; Whitesides, G.M.; Carrilho, E. Diagnostics for the developing world:
Microfluidic paper-based analytical devices. Anal. Chem. 2010, 82, 310. [CrossRef] [PubMed]
Whitesides, G.M. The origins and the future of microfluidics. Nature 2006, 442, 368373. [CrossRef] [PubMed]
Yetisen, A.K.; Akram, M.S.; Lowe, C.R. Paper-based microfluidic point-of-care diagnostic devices. Lab Chip
2013, 13, 22102251. [CrossRef] [PubMed]
Gong, M.M.; MacDonald, B.D.; Nguyen, T.V.; Van Nguyen, K.; Sinton, D. Lab-in-a-pen: A diagnostics format
familiar to patients for low-resource settings. Lab Chip 2014, 14, 957963. [CrossRef] [PubMed]
Veigas, B.; Jacob, J.M.; Costa, M.N.; Santos, D.S.; Viveiros, M.; Incio, J.; Martins, R.; Barquinha, P.;
Fortunato, E.; Baptista, P.V. Gold on paperpaper platform for Au-nanoprobe TB detection. Lab Chip
2012, 12, 48024808. [CrossRef] [PubMed]
Yen, C.W.; de Puig, H.; Tam, J.O.; Gmez-Mrquez, J.; Bosch, I.; Hamad-Schifferli, K.; Gehrke, L. Multicolored
silver nanoparticles for multiplexed disease diagnostics: Distinguishing dengue, yellow fever, and Ebola
viruses. Lab Chip 2015, 15, 16381641. [CrossRef] [PubMed]
Su, M.; Ge, L.; Kong, Q.; Zheng, X.; Ge, S.; Li, N.; Yu, J.; Yan, M. Cyto-sensing in electrochemical lab-on-paper
cyto-device for in-situ evaluation of multi-glycan expressions on cancer cells. Biosens. Bioelectron. 2015, 63,
232239. [CrossRef] [PubMed]
Apilux, A.; Dungchai, W.; Siangproh, W.; Praphairaksit, N.; Henry, C.S.; Chailapakul, O. Lab-on-paper with
dual electrochemical/colorimetric detection for simultaneous determination of gold and iron. Anal. Chem.
2010, 82, 17271732. [CrossRef] [PubMed]
Apilux, A.; Siangproh, W.; Praphairaksit, N.; Chailapakul, O. Simple and rapid colorimetric detection of Hg
(II) by a paper-based device using silver nanoplates. Talanta 2012, 97, 388394. [CrossRef] [PubMed]
Nath, P.; Arun, K.; Chanda, N. A paper based microfluidic device for the detection of arsenic using a gold
nanosensor. RSC Adv. 2014, 4, 5955859561. [CrossRef]

Micromachines 2016, 7, 73

11.
12.
13.
14.
15.

16.
17.

18.
19.

20.
21.
22.

23.
24.

25.

26.
27.
28.
29.
30.
31.
32.

33.

12 of 13

Shi, J.; Tang, F.; Xing, H.; Zheng, H.; Bi, L.; Wang, W. Electrochemical detection of Pb and Cd in paper-based
microfluidic devices. J. Braz. Chem. Soc. 2012, 23, 11241130. [CrossRef]
Nurak, T.; Praphairaksit, N.; Chailapakul, O. Fabrication of paper-based devices by lacquer spraying method
for the determination of nickel (II) ion in waste water. Talanta 2013, 114, 291296. [CrossRef] [PubMed]
Park, T.S.; Li, W.; McCracken, K.E.; Yoon, J.-Y. Smartphone quantifies Salmonella from paper microfluidics.
Lab Chip 2013, 13, 48324840. [CrossRef] [PubMed]
Cate, D.; Noblitt, S.D.; Volckens, J.; Henry, C. Multiplexed Paper Analytical Device for Measuring Airborne
Metal Particulates with Distance-Based Detection. Lab Chip 2015, 13, 23972404. [CrossRef] [PubMed]
Rattanarat, P.; Dungchai, W.; Cate, D.M.; Siangproh, W.; Volckens, J.; Chailapakul, O.; Henry, C.S.
A microfluidic paper-based analytical device for rapid quantification of particulate chromium.
Anal. Chim. Acta 2013, 800, 5055. [CrossRef] [PubMed]
Jayawardane, B.M.; McKelvie, I.D.; Kolev, S.D. A paper-based device for measurement of reactive phosphate
in water. Talanta 2012, 100, 454460. [CrossRef] [PubMed]
Wang, P.; Ge, L.; Yan, M.; Song, X.; Ge, S.; Yu, J. Paper-based three-dimensional electrochemical
immunodevice based on multi-walled carbon nanotubes functionalized paper for sensitive point-of-care
testing. Biosens. Bioelectron. 2012, 32, 238243. [CrossRef] [PubMed]
Yu, J.; Ge, L.; Huang, J.; Wang, S.; Ge, S. Microfluidic paper-based chemiluminescence biosensor for
simultaneous determination of glucose and uric acid. Lab Chip 2011, 11, 12861291. [CrossRef] [PubMed]
Zakir Hossain, S.M.; Luckham, R.E.; Smith, A.M.; Lebert, J.M.; Davies, L.M.; Pelton, R.H.; Filipe, C.D.M.;
Brennan, J.D. Development of a bioactive paper sensor for detection of neurotoxins using piezoelectric inkjet
printing of sol-gel-derived bioinks. Anal. Chem. 2009, 81, 54745483. [CrossRef] [PubMed]
Gallibu, C.; Gallibu, C.; Avoundjian, A.; Gomez, F.A. Easily Fabricated Microfluidic Devices Using Permanent
Marker Inks for Enzyme Assays. Micromachines 2016, 7, 1821. [CrossRef]
Hu, J.; Wang, S.; Wang, L.; Li, F.; Pingguan-Murphy, B.; Lu, T.J.; Xu, F. Advances in paper-based point-of-care
diagnostics. Biosens. Bioelectron. 2014, 54, 585597. [CrossRef] [PubMed]
Chin, C.D.; Laksanasopin, T.; Cheung, Y.K.; Steinmiller, D.; Linder, V.; Parsa, H.; Wang, J.; Moore, H.;
Rouse, R.; Umviligihozo, G.; et al. Microfluidics-based diagnostics of infectious diseases in the developing
world. Nat. Med. 2011, 17, 10151019. [CrossRef] [PubMed]
Cedar Lake Ventures Inc., WeatherSpark. Available online: https://weatherspark.com/ (accessed on
17 January 2016).
Mendez, S.; Fenton, E.M.; Gallegos, G.R.; Petsev, D.N.; Sibbett, S.S.; Stone, H.A.; Yi, Z.; Lopez, G.P. Imbibition
in Porous Membranes of Complex Shape: Quasi-Stationary Flow in Thin Rectangular Segments. Langmuir
2010, 26, 13801385. [CrossRef] [PubMed]
Pollock, N.R.; Mcgray, S.; Colby, D.J.; Noubary, F.; Nguyen, H.; Anh, T.; Khormaee, S.; Jain, S.; Hawkins, K.;
Kumar, S.; et al. Field Evaluation of a Prototype Paper-Based Point-of- Care Fingerstick Transaminase Test.
PLoS ONE 2013, 8, 110. [CrossRef] [PubMed]
Fu, E.; Lutz, B.; Kauffman, P.; Yager, P. Controlled reagent transport in disposable 2D paper networks.
Lab Chip 2010, 10, 918920. [CrossRef] [PubMed]
Martinez, A.W.; Phillips, S.T.; Whitesides, G.M. Three-dimensional microfluidic devices fabricated in layered
paper and tape. Proc. Natl. Acad. Sci. USA 2008, 105, 1960619611. [CrossRef] [PubMed]
Martinez, A.W.; Phillips, S.T.; Nie, Z.; Cheng, C.-M.; Carrilho, E.; Wiley, B.J.; Whitesides, G.M. Programmable
diagnostic devices made from paper and tape. Lab Chip 2010, 10, 24992504. [CrossRef] [PubMed]
Phillips, S.T.; Thom, N.K. Three-Dimensional, Paper-Based Microfluidic Devices Containing Internal Timers
for Running Time-Based Diagnostic Assays. Methods Mol. Biol. 2013, 949, 323.
Feng, L.; Li, X.; Li, H.; Yang, W.; Chen, L.; Guan, Y. Enhancement of sensitivity of paper-based sensor array
for the identification of heavy-metal ions. Anal. Chim. Acta 2013, 780, 7480. [CrossRef] [PubMed]
Li, X.; Tian, J.; Nguyen, T.; Shen, W. Paper-Based Microfluidic Devices by Plasma Treatment. Anal. Chem.
2008, 80, 91319134. [CrossRef] [PubMed]
Lutz, B.; Liang, T.; Fu, E.; Ramachandran, S.; Kauffman, P.; Yager, P. Dissolvable fluidic time delays
for programming multi-step assays in instrument-free paper diagnostics. Lab Chip 2013, 13, 28402847.
[CrossRef] [PubMed]
Zimmermann, M.; Schmid, H.; Delamarche, E. Capillary pumps for autonomous capillary systems. Lab Chip
2007, 7, 119125. [CrossRef] [PubMed]

Micromachines 2016, 7, 73

34.
35.
36.
37.
38.
39.

40.
41.

42.
43.
44.

13 of 13

Gong, M.M.; Zhang, P.; MacDonald, B.D.; Sinton, D. Nanoporous membranes enable concentration and
transport in fully wet paper-based assays. Anal. Chem. 2014, 86, 80908097. [CrossRef] [PubMed]
Washburn, E.W. The dynamics of capillary flow. Phys. Rev. 1921, 17, 273283. [CrossRef]
Cassidy, H.G. Investigation of Paper Chromatography. Anal. Chem. 1952, 24, 14151421. [CrossRef]
Rasi, M. Permeability Properties of Paper. Ph.D. Thesis, University of Jyvaskyla, Jyvaskyla, Finland, 2013.
Roberts, R.J.; Senden, T.J.; Knackstedt, M.A.; Lyne, M.B. Spreading of aqueous liquids in unsized papers is
by film flow. J. Pulp Pap. Sci. 2003, 29, 123131.
Songok, J.; Tuominen, M.; Teisala, H.; Haapanen, J.; Makela, J.; Kuusipalo, J.; Toivakka, M. Paper-Based
Microfluidics: Fabrication Technique and Dynamics of Capillary-Driven Surface Flow. ACS Appl.
Mater. Interfaces 2014, 6, 2006020066. [CrossRef] [PubMed]
Hong, S.; Kim, W. Dynamics of water imbibition through paper channels with wax boundaries.
Microfluid. Nanofluid. 2015, 19, 845853. [CrossRef]
Bhm, A.; Carstens, F.; Trieb, C.; Schabel, S.; Biesalski, M. Engineering microfluidic papers: Effect of fiber
source and paper sheet properties on capillary-driven fluid flow. Microfluid. Nanofluid. 2014, 16, 789799.
[CrossRef]
Dullien, F. Porous Media: Fluid Transport and Pore Structure; Academic Press: San Diesgo, CA, USA, 1992.
Duprat, C.; Protiere, S.; Beebe, A.; Stone, H.A. Wetting of flexible fibre arrays. Nature 2012, 7386, 510513.
[CrossRef] [PubMed]
Rost, M.; Laurson, L.; Dub, M.; Alava, M. Fluctuations in fluid invasion into disordered media.
Phys. Rev. Lett. 2007, 98, 14. [CrossRef] [PubMed]
2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC-BY) license (http://creativecommons.org/licenses/by/4.0/).