See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/307447138

Influence of surface forces and wall effects on

the minimum fluidization velocity of liquid-
solid micro-fluidized bed

Article May 2016

CITATIONS READS

0 11

3 authors, including:

Orlando Lopes do Nascimento Vladimir Zivkovic

Newcastle University Newcastle University
1 PUBLICATION 0 CITATIONS 31 PUBLICATIONS 131 CITATIONS

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Intensified by Design for the intensification of processes involving solids handling View
project

Hydrodynamics Study of Liquid-Solid Micro Circulating Fluidised Bed View project

All in-text references underlined in blue are linked to publications on ResearchGate, Available from: Orlando Lopes do Nascimento
letting you access and read them immediately. Retrieved on: 21 October 2016
 Influence of surface forces and wall effects on the minimum
fluidization velocity of liquid-solid micro-fluidized bed
Orlando L. do Nascimento, David A. Reay, Vladimir Zivkovic*

School of Chemical Engineering and Advanced Materials, Newcastle University, Merz

Court, Newcastle-upon-Tyne, NE1 7RU, UK
*Email: vladimir.zivkovic@ncl.ac.uk

ABSTRACT
Micro-fluidized beds represent a novel means of significantly enhancing mixing, mass

and heat transfer under the low Reynolds number flows that dominate in micro- devices

used in microfluidics and chemical micro-process technologies. This is one way of

implementing process intensification. Major differences of micro-fluidized beds from

their classical macro-scale counterparts are the critical importance of surface forces and

almost unavoidable wall effects due to their small bed size. Surface forces can become

dominant over gravity and hydrodynamics forces at the microscale and fluidization could

either be hindered or even prevented through the adhesion of particles to the walls of the

bed. We have used the acid-base theory of van Oss, Chaudhury and Good combined with

the Derjaguin approximation to estimate the wall adhesion forces for comparison with

hydrodynamics forces. Our new experiments show interesting fluidization behaviour at

the boundary of micro-flow as a result of interplay between the ratio of surface and

hydrodynamics forces and wall effects which both influence the minimum fluidization

velocity.

Keywords: Fluidization, Micro-Fluidized Bed, Microfluidics, Surface Forces, Wall

Effects, Process Intensification.

1
1. Introduction

Microfluidics [1-3] is the science and technology of processing of small volumes of fluids

in conduits having dimensions of the order of tens to hundreds of micrometres. This

research area holds promise in disparate fields ranging from automation of chemical

analysis [4-6] to medical diagnostics [7-9] through to process intensification [10-13]. This

promise is frustrated, however, because the heat and mass transport central to these and

other applications is dominated by the molecular diffusion that comes with the inevitable

laminar flow found in micron-sized conduits. Fluidized beds have long been used at the

macro-scale to enhance mixing and, thereby, heat and mass transport. Recent modelling

[14, 15] and experimental work [16-21] has demonstrated that liquid-solid micro-

fluidized beds are feasible, offering the potential to not only overcome diffusion-limited

fluid mixing, heat and mass transport in simple micron-sized channels, but also to provide

higher sensitivity and multi-modal detection in the diagnostic context by virtue of the

large surface area per unit volume that comes from use of micro-particles [22, 23].

In general, the main difference between micro- and macro-scale flows is the importance

of surface forces relative to volumetric forces such as gravity. Based on this criterion, the

widely asserted boundary between the two regimes is set at 1 mm in the general

microfluidics field [24, 25]. Not unsurprisingly, our recent work [19, 20, 26] confirmed

that surface forces play an important role in the micro-fluidized bed (FB) as well, for

example surface forces can prevent fluidization in a FB in some cases. We showed that

the acid-base model of van Oss, Chaudhury and Good combined with the Derjaguin

approximation [27] can successfully predict the propensity of micro-particles to adhere

to the walls of FBs using common liquid fluidizing media [20, 26]. Furthermore, we

used a comparison of surface and hydrodynamic driving forces to estimate the boundary
2
between micro- and macro-scale fluidization at 1 cm with stricter limits at 1 mm, the same

as for microfluidics [26].

A second major issue in FBs is the high potential for the particle-to-bed diameter ratios

to be greater than 0.1, leading to significant influence of the bed walls on the packing of

the particles in the bed and subsequently fluidization behaviour [17, 19, 20]. The bed

voidage in the FB (micro-packed beds as the starting fluidization point) is indeed

substantially higher compared to macroscale beds, leading to a significant increase in the

minimum fluidization velocity [19, 20]. The bed expansion behaviour also varies with the

particle-to-bed ratio confirming strong wall effects as in the original Richardson-Zaki

correlation for viscous flow [28]. Our preliminary experiments indicate that the

Richardson-Zaki exponent, n, increases significantly in a linear manner with the particle-

to-bed diameter ratio only when the ratio exceeds 0.1 [20]. Subsequently, Tang et al. [21]

confirmed our findings if only for narrow particle size distributions, while the trend is

opposite for particles with wider distribution. Their study shows that wall effects are still

influencing minimum fluidization behaviour for very high bed to particle ratios of up to

30, i.e. particle-to-bed diameter ratios greater than 0.03 [21]. In gas-solid miniaturized

beds experiments, the influence of the wall effects on the minimum fluidization velocity

was observed for even lower particle-to-bed diameter ratios of 0.02 [29, 30].

We performed new experiments with glass micro-particles and water as a fluidizing

medium at a boundary of a micro-fluidization according to our previous mentioned study.

Specifically, we used micro-machined Perspex fluidized bed of 1 mm2 and 4 mm2 square

cross sections in this study. The structure of this paper is as follows. We will first outline

the experimental details, including the apparatus, the particulate and liquid materials, and

the experimental procedures used. Secondly, a brief theoretical background for adhesion

3
between a wall and particles immersed in a liquid is given. This is followed by a

presentation of the results obtained using this theory and their discussion.

Fig. 1: Schematic of (a) micro-fluidized bed and (b) experimental setup for
visualization.

2. Experimental details

2.1. Micro-fluidized bed

The micro-fluidized bed was made by milling square cross section millimetre channels

into a Perspex block fitted with a distributor. Two micro-bed units were used in

experiments with a cross section of 1 mm x 1mm (i.e. the hydraulic diameter D =1 mm)

and 2 mm x 2 mm (D = 2 mm) and height of 100 mm as shown in Fig. 1(a). The distributor

was 1.5 mm thick, porous polyethylene sheet with mean pore size of 21m (SPC

Technologies Ltd, UK). Fig. 1(b) gives schematics of the experimental set-up which

consisted of syringe pump (AL-4000, WPI Inc., US) to pump water as a fluidizing

medium and Euromex Nexius trinocular digital microscope fitted with a LED ring for

illumination of the bed. A USB digital camera (JB Microscopes Ltd, UK) connected to a

microscope was used for recording the micro-fluidization expansion behaviour. The

4
images were stored on a PC for offline analysis. The micro-fluidized bed was placed on

a laboratory jack for easy height adjustment.

2.2. Particulate and liquid materials

We considered two different groups of as-supplied particles: (1) four different sized soda

lime glass microspheres of density p = 2500 kg/m3 and average diameter = 26.5, 35,

58 and 115 m with a standard deviation of 1.5, 3, 5 and 9 m respectively (Cospheric

LLC, US); and (2) five different sized poly(methyl methacrylate) (PMMA) particles of

density p1200 kg/m3 and average diameter = 23.5, 35.0, 41.5, 58 and 115 m

with a standard deviation of 1.5, 3, 1.5, 5 and 9 m respectively (Cospheric LLC, US).

All fluidization experiments were undertaken using tap water as a fluidizing medium.

2.3. Experimental procedure

Flowrates were calibrated in the range of fluidization velocities with and without particles

inside the bed as in some cases a high pressure drop across the porous distributor caused

a considerable but linear discrepancy from syringe pump readings. The pressure drop

across the micro-fluidized bed was not measured in the present work. Consequently, the

minimum fluidization velocity was obtained from visual observation of expansion

behaviour. ImageJ [31] was used for offline image analysis to obtain the height of the

fluidized bed as a function of water flow rate. The minimum fluidization velocity, Umf,

was obtained by extrapolation of the relative fluidized bed height plots as a function of

liquid velocity to the packed bed height velocity in both fluidization and de-fluidization

operation, i.e. for increasing and decreasing flow rates experiments respectively [16, 19,

21]. A minimal of three cycles of fluidization and defluidization were performed to

5
determine the minimum fluidization velocity for the entire range of experimental

conditions. Experiments were performed at room temperature of 20 2C.

3. Theoretical background

Theoretical calculations were detailed in a previous paper [26] so here a procedure is

briefly outlined. The free energy of interaction between two different solid surfaces

immersed in a liquid, G1w2, is calculated from Dupree equation using the acid-base

approach developed by van Oss, Chaudhury and Good [27]. If a free energy is less than

zero (i.e. G1w2 <0), adhesion between the two solid surfaces occurs in the presence of

the liquid phase, otherwise interactions are repulsive. However, the situation is dynamic

involving fluid flow so one need to compare the adhesion force to the hydrodynamic force

particles are experiencing. We used the Derjaguin approximation to obtain adhesion

forces where the particles and the bed wall are approximated by spheres and a flat plate

which leads to [27]

Fadh G1w2 d p (3)

Due to difficulties in calculating directly drag forces [26], hydraulic forces are obtained

through a simple force balance where the drag force, Fd, is equal to buoyant weight of a

particle, Fbw, as given by

1
Fd Fbw ( p f )( d 3p ) g (4)
6

4. Results and discussion

4.1. Free Energy Predictions for Experimental Systems

6
The surface tensions parameters of water and two solids (i.e. PMMA and glass)

considered in this experimental study are shown in Table 1 as taken from various sources.

The free energies of interaction, G1w2, for the two particles/wall material/liquid triplets

were evaluated using all combinations of these values. The free energies obtained from

the combinations for a triplet were then used to obtain the averages and standard

deviations shown in Table 2 along with the associated minima and maxima.

Table1: Surface tension components (in mJ/m2) for water and solids used in this study
taken from various literature sources.
Material LW AB + - Reference
Water 72.8 21.8 51.0 25.5 25.5 van Oss [27]
Glass 59.8 42.0 17.8 1.97 40.22 Freitas & Sharma [32]
64.37 42.3 22.07 2.9 42.0 Clint & Wicks [33]
51.7 33.7 18.0 1.3 62.2 van Oss [27]
PMMA 44.65 42 2.65 0.55 3.2 Della Volpe & Siboni [34]
44.58 41.2 3.38 0.38 7.5 Clint & Wicks [33]
40.6 40.6 0 0 12 van Oss [27]
39.21 36.68 2.53 0.16 10.02 Zdziennicka [35]

Table 2: Free energy of interaction (in mJ/m2) between glass or PMMA particles and
the sidewalls of the PMMA (Perspex) microfluidized bed in water. The mean,12 ,
standard deviation, (), minimum, () , and maximum, () , values are
obtained by using different sets of the surface tension parameters drawn from Table 1.
Surf. 1 Surf. 2 Liquid (G) () () Interaction
Glass PMMA Water 4.62 9.28 19.21 12.38 Attractive
PMMA PMMA Water -47.22 7.49 -62.76 -37.83 Attractive

In the first case of glass particles immersed in water, the majority of the parameter

combinations predict that glass microparticles will be attracted to a PMMA bed walls

when immersed in water; only when the glass parameters of van Oss [27] are used with

the PMMA parameters except that of Della Volpe & Siboni [34] do we get a positive

free energy (3 out of 12 combinations). The average free energy for the latter three

combinations suggest no propensity for adhesion (mean value of 8.00 mJ/m2 with
7
standard deviation of 3.80 mJ/m2), while the average of only adhesion predicting values

(9 out 12 combinations) is -8.86 mJ/m2 with a standard deviation of 5.76 mJ/m2. Overall

the average of all 12 combinations is -4.62 mJ/m2 with an obviously large standard

deviation of 9.28 mJ/m2 (Table 2), which is due to a large spread of calculated values as

can be seen from the reported minimal and maximal values. However, taking into account

all of these results suggest strongly that glass microparticles have a weak propensity to

adhere to a PMMA surfaces in the presence of water. Indeed, we noticed glass particle

sticking to the wall when the fluid flow is stopped and they even stick above fluidized

bed in our de-fluidization experiments as shown in Figure 2(a) and (b). We noticed that

some particles would roll-out off the walls and get back to the bulk of the fluidized bed

during the de-fluidization cycle of the experiments.

In the second case involving PMMA particles in the presence of water, all 16

combinations of the PMMA surface tension parameters were used rather than a simplified

formula for interaction between identical surfaces in liquid [27] as essentially bed and

particle materials are almost certainly of different characteristics, coming from different

sources. All the evaluated free energies are substantially less than zero, not unsurprisingly

given the hydrophobicity of PMMA. With an average free energy of interaction of 47.22

mJ/m2 and standard deviation of 7.49 mJ/m2 (Table 2), the theory certainly suggests a

strong propensity for PMMA particles adhesion to a PMMA bed walls in presence of

water as shown in Figure 2(c) and (d). The free energy interaction between identical

surfaces (4 out of 16 combinations) is slightly different with a mean of -46.65 mJ/m2 and

standard deviation of 9.97 mJ/m2. This suggests strong cohesion forces between PMMA

particles while immersed in water which is not present in glass particles case as the glass

8
surface is hydrophilic and consequently interactions are repulsive (mean free energy of

22.16 mJ/m2).

Fig. 2. Optical micrograph showing adhesion of 35 m glass (a & b) and PMMA (c & d)
particles to the PMMA walls of 1 mm2 square micro-fluidized bed in water as the
fluidizing medium. Adhesion is shown above packed bed with no fluid flow (a & c) and
above fluidized bed during de-fluidization cycle of experiments at volumetric flow rate
of 160 l/hour (b & d).

105

104

103
F adh /F d

102

101

100
10-5 10-4 10-3 10-2

Particle diameter d p (m)

Fig. 3. Ratio of adhesion forces, Fadh, to drag force, Fd, for glass (filled circles) and
PMMA (empty circles) micro-particles as a function of particle size. Lines represent
theoretical prediction for any given particle size.

9
4.2. Estimated adhesion/drag force ratio as a function of particle diameter

Although the free energy of interaction identifies the nature of the forces (attractive or

repulsive), the ratio of forces at play is more useful when seeking to understand if

adhesion will in fact happen and influence the dynamics of the process. Fig. 3 shows the

ratio of estimated adhesion forces to drag force as a function of particle diameter for the

glass and PMMA particles inside a PMMA bed with water as the fluidizing medium; the

2 gradient of the lines in this figure arises from the adhesion force scaling directly with

particle diameter, Eq. 3, whilst the drag force scales with the cube of the diameter, Eq. 4.

In the cases of PMMA micro-particles fluidized by water in a PMMA micro-fluidized

bed, adhesion forces are some 3 to 5 orders of magnitude larger than the estimated drag

forces, indicating particle adhesion to the bed walls is highly likely to influence the

fluidization process. However, weaker adhesion forces and higher particle density in the

case of glass micro-particles gives adhesion forces comparable to drag forces, i.e. only 1

to 3 orders of magnitude larger. Specifically for the largest 115 m glass particles the

estimated ratio of adhesion force to drag force is only Fadh/Fd = 26.3. This is very close

to the boundary ratio value (order of 10) from our previous paper [26] for which we

hypothesised the surface forces can be neglected.

Indeed as hypothesized for a system with comparable adhesion and hydrodynamics forces

[26], we achieved smooth fluidization of the largest 115 m glass particles in PMMA

micro-fluidized bed using water as fluidizing medium as shown in Fig. 4(a). Perhaps

surprisingly, we were able to achieve smooth fluidization for each size of glass micro-

particles (only the smallest 26.5 m glass particles fluidization is shown in Fig. 4(b))

indicating that transition regime is more broad and/or more nuanced and complex than

anticipated in our previous paper [26]. Even more unexpected, we attained smooth

10
fluidization for all size of PMMA particles (e.g. Fig 4(c) and (d) for 115 m and 23.5 m

PMMA particles respectively) despite adhesion forces which are 3 to 5 orders of

magnitude larger than the drag forces. This is completely different to our previous study

with the system having a similar ratio of adhesion to surface forces where fluidization

was impossible due to the particles completely adhering to the walls of the micro-

fluidized bed - we will return to this point in sub-section 4.5. Although no complete de-

fluidization was observed, in all cases the onset of fluidization was postponed which

indicates that the adhesion forces between particles and the bed walls must be overcome

before the fluidization starts.

Fig. 4. Optical micrographs showing successful fluidization with: (a) 115 m glass, (b)
26.5 m glass, (c) 115 m PMMA and (d) 23.5 m PMMA microparticles in a Perspex
(PMMA) D = 1 mm square micro-fluidized bed at different water volumetric flow rates.

11
4.3. Determination of minimum fluidization velocity from expansion graphs

Typical plots of relative bed height as a function of superficial fluid velocity are given in

Figure 5. In all cases the bed expansion ratio shows a linear trend with fluidization

velocity similar to previous micro-fluidization experiments [19, 21]. Extrapolation to the

packed bed height was used to determine experimental fluidization velocity both for

increasing, Umf,u, and decreasing, Umf,d, flow rate curves as shown in Figure 5. In some

cases the extrapolation fitted line will intersect with the vertical zero expansion line

slightly below the experimental packed bed point. In those instances the last packed bed

point before bed expansion is reported as experimental Umf instead of intersection point.

This illustrates that the determined Umf are slightly underestimated using this method

compared to more reliable pressure drop measurements [17]. We used arithmetic mean

of Umf,u and Umf,d values with associated standard deviations as experimental Umf,exp in

further analysis. The theoretical minimum fluidization velocities, Umf,t was calculated

using the Ergun equation [36], with an estimated initial packed bed voidage of mf = 0.40

0.01 in line with previous experiments [21] and trend of rectangular packed bed voidage

with particle-to-bed ratios [37, 38]. The mean value of the theoretically calculated

minimum fluidization velocities are shown by arrows on the plots. The experimental

minimum fluidization velocities are proportionally greater for PMMA particles compared

to glass particles also for the smaller particle diameter compared to the bigger ones. This

strongly suggests that ratio of adhesion forces to the hydrodynamic forces is influencing

the incipient fluidization velocity.

12
Fig. 5. Relative bed height as a function of superficial fluid velocity, U, with increasing
flow rate (symbol) and decreasing flow rate (symbol) for (a) 115 m glass, (b) 26.5 m
glass, (c) 115 m PMMA and (d) 23.5 m PMMA microparticles in 1 mm2 micro-bed.
Errors are smaller than symbols.

4.4. Influence of Adhesion/Drag Force Ratio on the minimum fluidization velocity

We plotted ratios of experimentally determined fluidization velocity over the theoretical

calculated values using the Ergun equation [36] as a function of adhesion/drag force ratios

as shown in Fig. 6. In the case of glass micro-particles, Fig. 6 (a), the velocity ratio goes

below unity which is a consequence of underestimation of experimental minimum

fluidization velocity by the extrapolation of expansion lines, a technique applied here as

discussed in the previous sub-section. Yet, it seems that the velocity ratios level off at

around 0.8 at lower fluid velocities in the case of 4 mm2 (squares) square micro-fluidized

bed indicating probably a 20-30% underestimation in the determination of the Umf in our

13
experiments. Regardless of these systematic underestimations, the figure undoubtedly

shows that the increase in experimental minimum fluidization velocity is linearly scaling

with the force ratio both for glass and PMMA micro-particles fluidization. However, it is

noticeable that two distinct lines represent results in 1 mm2 (circles) and 4 mm2 (squares)

square micro-fluidized beds in both glass and PMMA particle fluidization experiments.

This clearly indicates that wall effects, i.e. the particle-to-bed diameter ratio, influence

the postponement of incipient fluidization in addition to the adhesion/drag force ratio.

Fig. 6. Ratio of experimental, Umf,exp, and theoretical, Umf,t, minimum fluidization

velocity as a function of adhesion/drag force ratio for (a) glass and (b) PMMA micro-
particles in D = 1 mm (circles) and D = 2 mm (squares) square micro-fluidized beds.

4.5. Influence of adhesion/drag force ratio and wall effects on the minimum

fluidization velocity

In order to take wall effects into account, we replotted the data from Fig. 6. as a function

of the simple product of two ratios: adhesion/drag force ratio and particle-to-bed diameter

ratio. In both particles cases, the data fall on one master line showing good linear trend

of increase of the minimum fluidization velocity as a function of the introduced ratios

product. The linear fitting to the glass micro-particles fluidization data gives a line of

14
slope 0.0510 0.0025 as shown in Fig. 7(a) (adjusted R2 = 0.9836) while for the PMMA

beads slope of linear fitting line is 0.0100 0.0006 (adjusted R2 = 0.9676), Fig. 7(b). We

speculate that the difference in the slope of the lines is due to particle cohesion which is

present for PMMA particles but not for glass beads immersed in water as reported in the

sub-section 4.1, but other factors like particle density cannot be completely excluded. In

general cohesion will increase the particle size due to agglomeration which will reduce

the importance of surface forces as the major factor and therefore reduce the slope of the

line, but more detailed investigation is needed to elucidate this further.

Fig. 7. Ratio of experimental, Umf,exp, and theoretical, Umf,t, minimum fluidization

velocity as a function of product of force and particle-to-bed diameter ratios for (a) glass
and (b) PMMA micro-particles in D= 1 mm (circles) and D = 2 mm (squares) square
micro-fluidized bed.
The discovered relationship can be used to predict the point of no-fluidization by

extrapolating experimental data to the particle terminal velocity, Ut, as shown in Fig. 8.

This complete de-fluidization happened in our previous study with 30.5 m glass and

PMMA beads inside a 400 x 175 m PDMS micro-channel [26]. In the case of PMMA

particles, Fig. 8(b), predicted a critical product ratio of the order of 5000 where the micro-

fluidization would be impossible which explains the no-fluidization observation from our

previous study at double of this critical ratio. However, extrapolation of the glass beads

micro-fluidization data points from this study predicts successful fluidization for the

15
product ratio of around 200 as calculated for glass particles in our previous study. The

estimated value of the increase of the minimum fluidization velocity is approximately ten

times over the theoretical fluidization velocity but still approximately four times lower

than the particle terminal velocity. Indeed, in our previous study using an aqueous

solution of surfactant we managed to overcome adhesion forces but above the terminal

velocity so no-fluidization was achieved as particles were washed away [26]. We think

that the elasticity of PDMS material is probably the cause of this discrepancy. The area

of contact between particles and bed walls will be higher for PDMS rubber-like material

compared to rigid PMMA plastic used in this study. The contact area for glass/PDMS

system is approximetely 1.5 to 3 times bigger than for glass/PMMA wall contact as

estimated by the Hertz formula [39] with the elastic modulus and the Poissons ratio

values for the materials from various sources [40-43]. Consequently, the adhesion forces

are higher which explains partly the discrepancy between experiments in PMMA bed

with our previous PDMS bed measurement point. Furthermore, the difference in surface

roughness characteristics between micro-machined PMMA beds and PDMS channel

fabricated by the standard soft lithography technique [44] also can significantly influence

the adhesion dynamics [45]. The microfluidic PDMS channels probably have only small

surface coverage of small-scale nanometre surface asperities [46] while PMMA beds will

have in addition large-scale micrometre asperities as measured in similar system [21].

Because of particle-asperities interactions which are highly likely for rough PMMA beds,

the adhesion forces will be smaller in comparison with pure particle-plate interaction [45]

as used in our calculations. Consequently, smooth PDMS channels with low coverage of

small-scale asperities will have significantly bigger adhesion forces due to the increased

probability of having pure particle-plate contact [45], but detailed statistical information

16
on the surface roughness would be needed to correctly estimate adhesion forces between

rough surfaces using numerical simulations [45].

Fig. 8. Extrapolation of experimental data for glass (a) and PMMA (b) particles
fluidization from Fig. 7 to the ratio of the particle terminal velocity, Ut, over the
theoretical minimum fluidization velocity, Umf,t,. Experimental points for 30.5 m
particles from a previous study given as diamond points on Ut/Umf,t dashed-line while
lines are linear fitting to the experimental points of this study.

4.6. Hysteresis in micro-fluidization expansion behaviour

The expansion plots, Fig. 5, shows obvious and quite large hysteresis effects in expansion

lines for both glass and PMMA particles, probably due to considerable pressure overshoot

for fluidization cycle experiments as already measured in the micro-fluidized bed [17].

This may also be an indication of jamming transition as demonstrated by the careful

experimentation of Goldman and Swinney [47]. We quantify the degree of hysteresis as

the ratio of Umf,u and Umf,d for each size of particles. The velocity ratios were the order of

1.2-1.3 and 1.15-1.25 for PMMA and glass particles fluidization indicating larger

hysteresis for more adhesive PMMA beads. The plot of glass particles fluidization results

shows good trend of increase of hysteresis with the product of force and particle-to-bed

diameter ratios as shown in Fig. 9(b). On the other hand, this trend is not present in results

17
of PMMA particles fluidization with no obvious trend with either force ratio or particle-

to-bed diameter ratio alone (not shown here).

Fig. 9. The ratio of the experimental minimum fluidization velocity for increasing,
Umf,u, and decreasing, Umf,d, flow rate experiments versus product of force and particle-
to-bed diameter ratios for (a) glass and (b) PMMA micro-particles.

5. Conclusions

We studied fluidization of glass and PMMA micro-particles in 1 mm2 and 4 mm2 square

micro-fluidized beds made of Perspex (PMMA). As predicted by the acid-base theory of

van Oss, Chaudhury and Good there was noticeable adhesion to the bed walls for both

particle materials. The Derjaguin approximation was used to estimate adhesion forces

between micro-particles and bed walls for comparison with drag forces approximated by

the buoyant weight of the particles. The observed incipient fluidization in the micro-

fluidized bed was postponed in comparison with theoretical predictions based on

macroscopic experiments. This was influenced both by adhesion strength and wall

effects. The increase in the minimum fluidization velocity scales linearly with the simple

product of adhesion/drag force and particle-to-bed diameter ratios. Particle cohesion and

bed walls/particle materials properties like elasticity influence the scaling relationship but

further investigation is needed on this point. The hysteresis is proportional to the product

18
of force and particle-to-bed diameter ratios in case of glass particles but not for PMMA

particles results- further investigation is needed to elucidate this further.

ACKNOWLEDGEMENTS
OLdN acknowledges the EPSRC DTP award EP/M506382/1 for supporting his PhD

studies in the School of Chemical Engineering & Advanced Materials of Newcastle

University.

REFERENCES
[1] T.M. Squires, S.R. Quake, Microfluidics: Fluid physics at the nanoliter scale, Reviews
of modern physics, 77 (2005) 977.
[2] H.A. Stone, A.D. Stroock, A. Ajdari, Engineering flows in small devices:
Microfluidics towards lab-on-a-chip, Annu. Rev. Fluid Mech., 36 (2004) 381-411.
[3] G.M. Whitesides, The origins and the future of microfluidics, Nature, 442 (2006) 368-
373.
[4] P.S. Dittrich, K. Tachikawa, A. Manz, Micro Total Analysis Systems. Latest
Advancements and Trends, Anal. Chem., 78 (2006) 3887-3908.
[5] A. Manz, N. Graber, H.M. Widmer, Miniaturized total chemical analysis systems: A
novel concept for chemical sensing, Sensors and Actuators B, 1 (1990) 244-248.
[6] J. West, M. Becker, S. Tombrink, A. Manz, Micro Total Analysis Systems: Latest
Achievements, Anal. Chem., 80 (2008) 4403-4419.
[7] P. Abgrall, A.-M. Gu, Lab-on-chip technologies: making a microfluidic network and
coupling it into a complete microsystems: a review, Journal of Micromechanics and
Microengineering, 17 (2007) R15.
[8] S. Haeberle, R. Zengerle, Microfluidic platforms for lab-on-a-chip applications, Lab
on a Chip, 7 (2007) 1094-1110.
[9] J. Melin, S.R. Quake, Microfluidic Large-Scale Integration: The Evolution of Design
Rules for Biological Automation, Annu. Rev. Bioph. Biom., 36 (2007) 213-231.
[10] K.F. Jensen, Microreaction engineering -- is small better?, Chem. Eng. Sci., 56
(2001) 293-303.
[11] J.-C. Charpentier, Four main objectives for the future of chemical and process
engineering mainly concerned by the science and technologies of new materials
production, Chemical engineering journal, 107 (2005) 3-17.
[12] S.J. Haswell, Chemical technology: All together now, Nature, 441 (2006) 705-705.
[13] M.P. Dudukovic, Frontiers in Reactor Engineering, Science, 325 (2009) 698-701.
[14] J.J. Derksen, Scalar mixing by granular particles, AIChE J., 54 (2008) 1741-1747.

19
[15] J.J. Derksen, Scalar mixing with fixed and fluidized particles in micro-reactors,
Chem. Eng. Res. Des., 87 (2009) 550-556.
[16] B. Potic, S.R.A. Kersten, M. Ye, M.A. van der Hoef, J.A.M. Kuipers, W.P.M. van
Swaaij, Fluidization with hot compressed water in micro-reactors, Chemical Engineering
Science, 60 (2005) 5982-5990.
[17] E. Doroodchi, Z. Peng, M. Sathe, E. Abbasi-Shavazi, G.M. Evans, Fluidisation and
packed bed behaviour in capillary tubes, Powder Technol., 223 (2012) 131-136.
[18] E. Doroodchi, M. Sathe, G. Evans, B. Moghtaderi, Liquidliquid mixing using
micro-fluidised beds, Chem. Eng. Res. Des., 91 (2013) 2235-2242.
[19] V. Zivkovic, M.J. Biggs, Z.T. Alwahabi, Experimental study of a liquid fluidization
in a microfluidic channel, AIChE Journal, 59 (2013) 361-364.
[20] V. Zivkovic, M.N. Kashani, M.J. Biggs, Experimental and theoretical study of a
micro-fluidized bed, AIP Conference Proceedings, 1542 (2013) 93-96.
[21] C. Tang, M. Liu, Y. Li, Experimental investigation of hydrodynamics of liquidsolid
mini-fluidized beds, Particuology.
[22] S. Derveaux, B. Stubbe, K. Braeckmans, C. Roelant, K. Sato, J. Demeester, S. De
Smedt, Synergism between particle-based multiplexing and microfluidics technologies
may bring diagnostics closer to the patient, Analytical and Bioanalytical Chemistry, 391
(2008) 2453-2467.
[23] C.T. Lim, Y. Zhang, Bead-based microfluidic immunoassays: The next generation,
Biosensors and Bioelectronics, 22 (2007) 1197-1204.
[24] R.L. Hartman, K.F. Jensen, Microchemical systems for continuous-flow synthesis,
Lab Chip, 9 (2009) 2495-2507.
[25] A. Gunther, K.F. Jensen, Multiphase microfluidics: from flow characteristics to
chemical and materials synthesis, Lab Chip, 6 (2006) 1487-1503.
[26] V. Zivkovic, M.J. Biggs, On importance of surface forces in a microfluidic fluidized
bed, Chem. Eng. Sci., 126 (2015) 143-149.
[27] C.J. van Oss, Chapter Two The Apolar and Polar Properties of Liquid Water and
Other Condensed-Phase Materials, in: J.v.O. Carel (Ed.) Interface Science and
Technology, Elsevier2008, pp. 13-30.
[28] J.F. Richardson, W.N. Zaki, Sedimentation and fluidisation: Part I, Chem. Eng. Res.
Des., 75 (1997) S82-S99.
[29] X. Liu, G. Xu, S. Gao, Micro fluidized beds: Wall effect and operability, Chem. Eng.
J., 137 (2008) 302-307.
[30] A. Rao, J.S. Curtis, B.C. Hancock, C. Wassgren, The effect of column diameter and
bed height on minimum fluidization velocity, AIChE J., 56 (2010) 2304-2311.
[31] C.A. Schneider, W.S. Rasband, K.W. Eliceiri, NIH Image to ImageJ: 25 years of
image analysis, Nat Meth, 9 (2012) 671-675.
[32] A.M. Freitas, M.M. Sharma, Effect of surface hydrophobicity on the hydrodynamic
detachment of particles from surfaces, Langmuir, 15 (1999) 2466-2476.

20
[33] J.H. Clint, A.C. Wicks, Adhesion under water: surface energy considerations,
International journal of adhesion and adhesives, 21 (2001) 267-273.
[34] C. Della Volpe, S. Siboni, Troubleshooting of surface free energy acid-base theory
applied to solid surfaces: The case of Good, van Oss and Chaudhury theory, in: K.L.
Mittal, H.R. Andersson (Eds.) Acid-base interactions: relevance to adhesion science and
technology, VSP, Utrecht, The Netherlands, 2000, pp. 55-91.
[35] A. Zdziennicka, The wettability of polytetrafluoroethylene and
polymethylmethacrylate with regard to interface behaviour of Triton X-165 and short
chain alcohol mixtures: I. Critical surface tension of wetting and adhesion work, Colloids
and Surfaces A: Physicochemical and Engineering Aspects, 367 (2010) 108-114.
[36] S. Ergun, Fluid flow through packed columns, Chem. Eng. Prog., 48 (1952) 89-94.
[37] M. Navvab Kashani, V. Zivkovic, H. Elekaei, M.J. Biggs, A new method for
reconstruction of the structure of micro-packed beds of spherical particles from desktop
X-ray microtomography images. Part A. Initial structure generation and porosity
determination, Chem. Eng. Sci., 146 (2016) 337-345.
[38] M. Navvab Kashani, V. Zivkovic, H. Elekaei, L.F. Herrera, K. Affleck, M.J. Biggs,
A new method for reconstruction of the structure of micro-packed beds of spherical
particles from desktop X-ray microtomography images. Part B. Structure refinement and
analysis, Chem. Eng. Sci., (2016) provisionaly accepted on 25/11/16.
[39] A.D. Zimon, Adhesion of dust and powder, Springer Science & Business
Media2012.
[40] J.M. Gere, Mechanics of Materials 5th, Brooks Cole, 2001.
[41] C. Ishiyama, Y. Higo, Effects of humidity on Young's modulus in poly(methyl
methacrylate), Journal of Polymer Science Part B: Polymer Physics, 40 (2002) 460-465.
[42] I.D. Johnston, D.K. McCluskey, C.K.L. Tan, M.C. Tracey, Mechanical
characterization of bulk Sylgard 184 for microfluidics and microengineering, Journal of
Micromechanics and Microengineering, 24 (2014) 035017.
[43] S.M. Tarsuslugil, R.M. OHara, N.J. Dunne, F.J. Buchanan, J.F. Orr, D.C. Barton,
R.K. Wilcox, Experimental and Computational Approach Investigating Burst Fracture
Augmentation Using PMMA and Calcium Phosphate Cements, Annals of biomedical
engineering, 42 (2014) 751-762.
[44] G.M. Whitesides, E. Ostuni, S. Takayama, X. Jiang, D.E. Ingber, Soft lithography
in biology and biochemistry, Annu. Rev. Biomed. Eng., 3 (2001) 335-373.
[45] C. Henry, J.-P. Minier, G. Lefevre, Numerical study on the adhesion and
reentrainment of nondeformable particles on surfaces: the role of surface roughness and
electrostatic forces, Langmuir, 28 (2011) 438-452.
[46] R. Jaeger, J. Ren, Y. Xie, S. Sundararajan, M.G. Olsen, B. Ganapathysubramanian,
Nanoscale surface roughness affects low Reynolds number flow: Experiments and
modeling, Applied Physics Letters, 101 (2012) 184102.
[47] D.I. Goldman, H.L. Swinney, Signatures of glass formation in a fluidized bed of hard
spheres, Physical review letters, 96 (2006) 145702.

21