Patterns from Drying DropsKhellil Sefiane†

1School of Engineering, The University of Edinburgh,

King’s Buildings, Mayfield Road,

Edinburgh, EH9 3JL, United Kingdom.

†Email: ksefiane@ed.ac.uk

Abstract

The objective of this review is to investigate different deposition patterns from dried droplets

of a range of fluids: paints, polymers and biological fluids. This includes looking at

mechanisms controlling the patterns and how they can be manipulated for use in certain

applications such as medical diagnostics and nanotechnology.

This review introduces the fundamental properties of droplets during evaporation. These

include profile evolution (constant contact angle regime (CCAR) and constant radius regime

(CRR)) and the internal flow (Marangoni and Capillary flow (Deegan et al. [22])). The

understanding of these processes and the basic physics behind the phenomenon are crucial to

the understanding of the factors influencing the deposition patterns. It concludes with the

applications that each of these fluids can be used in and how the manipulation of the

deposition pattern is useful.

The most commonly seen pattern is the coffee-ring deposit [27] which can be seen frequently

in real life from tea/coffee stains and in water colour painting. This is caused by an outward

flow known as Capillary flow which carries suspended particles out to the edge of the wetted

area. Other patterns that were found were uniform, central deposits and concentric rings

which are caused by inward Marangoni flow. Complex biological fluids displayed an array of

different patterns which can be used to diagnose patients.

Content

1. Introduction............................................................................................................................

1.1 History of pure liquid drop evaporation.....................................................................

1.2 Profile Evolution........................................................................................................

1.3 Flow regimes inside a droplet....................................................................................

1.4 Complex evaporation situations.................................................................................

2. Patterns from Drying Drops.................................................................................................

2.1 Polymers.....................................................................................................................

2.2 Paints..........................................................................................................................

2.3 Biological fluid...........................................................................................................

2.4 Nano-particle suspensions..........................................................................................

3. Applications............................................................................................................................

3.1 Polymers.....................................................................................................................

3.2 Biomedicine...............................................................................................................

3.3 Nanotechnology..........................................................................................................

4. Conclusion...............................................................................................................................

References...................................................................................................................................

1. Introduction

Study of drop evaporation has been growing in number more rapidly since the 1980s. This

increase in activity is due to rising demand for its application in fields such as inkjet printing,

paints, polymers, nanotechnology and medical diagnostic techniques [1]. Two fields of

particular interest recently are nanotechnology, which has appeared in the past decade [2],

and diagnosis methodology, which appeared in the past two decades. Firstly appearing in the

‘Litos Test’ system from Russia [3] for the early diagnosis of urolithiasis. The mechanisms

behind basic drop evaporation, whereby a pure liquid drop evaporates into surrounding still

air, are fairly well understood. Recently improved understanding of the mechanisms behind

more complex drop evaporation has allowed fast improvements in these fields and will give

scope for future study.

1.1 History of pure liquid drop evaporation

The history of the study drop evaporation starts in 1877. When Maxwell published an article

entitled “Theory of the Wet Bulb Thermometer” in which he derived equations for basic drop

evaporation [2, 4-8]. Maxwell believed the drop evaporation process to be diffusion-

controlled [5] due to the difference in vapour concentration between the surface of the drop

and in the surrounding bulk air [9]. Further study in this area has proved his initial

hypothesis, although correct nonetheless incomplete, since it is now accepted that drop

evaporation arises by a combination of heat and mass transfer. Heat is transferred to the drop

from the surroundings by all three modes of heat transfer and mass is transferred from the

drop to the surroundings by convection and diffusion [2]. However, his rudimentary

equations were used as the basis for a lot of the study that followed. Further studies from

Maxwell time have also shown that the evaporation of drops is proportional to other factors

such as: vapour pressure, radius of the drop and surface tension.

Sreznevsky discovered that the evaporation of hemispherical drops from a flat plate was

proportional to the vapour pressure of the evaporating liquid [2]. Following this, Morse added

that evaporation rate was proportional to the radius of a spherical drop [4] and after analysing

his results, Langmuir decided that this was the case based on the relationship between

diffusion and conduction heat transfer [2, 4, 10].

The effect of convection mass transfer, in which, the evaporating vapour from the surface of

the drop is swept away by the surrounding air is known as Stefan flow. It was Fuchs that

introduced this concept into the Maxwell equations [5, 6]. Fuchs was a very influential man

in these early investigations. In 1959, his article entitled “Evaporation and droplet growth in

gaseous media” [11] was translated from Russian and became a crucial resource since it

analysed and criticised all major research work that had been carried out up until then [2].

Sazhin [6, 12], Erbil [2] and Tonini [5] explain the more recent progress made in the

mathematical modelling of drop evaporation, these go into more detail than this review will

cover. Each of these articles gives a thorough history of progresses made in drop evaporation

research starting from Maxwell time. Sazhin concentrates on drops of fuel whereas Tonini

uses general liquid drops and compares new models for unsteady state evaporation with the

more basic Maxwell, Stefan and Fuchs models. Erbil, in his review [2] explains the

evaporation of isolated sessile and spherical drops.

1.2 Profile Evolution

During drop evaporation, two accepted theories for profile evolution are constant radius

regime (also known as constant contact area regime) and constant contact angle regime [13,

14]. In the case for constant radius regime (CRR) the droplet is pinned to the surface and the

height of the drop falls as the fluid evaporates. In constant contact angle regime (CCAR) the

radius decreases but the height remains constant. In reality a combination of both of these

usually occurs since a drop usually follows a CRR until a critical contact angle is reached at

which point the remaining fluid follows a CCAR until the fluid is fully evaporated.

Manipulation of the type of flow inside the droplet can control the type of deposition pattern

left by the drop [13]. Later, this combination of regimes can be used to explain the patterns

left by dried biological fluid drops, for example, a thick outer ring of protein with crystallised

salt in the centre which is seen in experiments carried out by Yakhno et al. [15] and Meloy

Gorr et al. [16].

Guilizzoni et al. [17] investigated the shape of the water droplet before and during

evaporation looking at contact angle, height and contact area as parameters. Their experiment

investigated the effect of surface effusivity, finned surfaces and Weber number on the rate of

evaporation and the droplet shape. The results show that droplets with a higher weber number

(a measure of the relative importance of the fluid's inertia compared to its surface tension)

gave more promising results since they had a larger contact area on finned surfaces which is

as expected. Those with smaller Weber numbers did not have a larger contact area in the

finned surface. The further understanding of profile evolution in this case would have a large

impact in the applications of dropwise cooling since contact area is directly proportional to

heat transfer.

Panwar et al. [18] found that a water sessile drop on a glass or polycarbonate surface

followed the CRR. On the other hand, Xu et al. [19] investigated a complicated system in

which micro-posts were positioned at regular intervals and then coated in the highly

hydrophobic material, Teflon. A water droplet containing varying concentrations of

suspended gold micro-particles was then placed onto the surface and the profile regimes were

recorded. The results showed that the fluid exhibited CCAR for a short while as it passed a

post and CRR between posts until it finally exhibited a fixed mode. This experiment has huge

applications in the deposition of nanoparticles onto surfaces used in areas such as

nanotechnology.

1.3 Flow regimes inside a droplet

A wide area of study is the effects of different flows inside the droplet. Two important flow

regimes are Capillary flow (driven by continuity) and Marangoni flow (driven by surface

tension gradients). The manipulation of these flow regimes can lead to different patterns left

from the evaporation of drying drops [20]. For example, in inkjet printing, the coffee ring

effect caused by Capillary flow provides a sharper image, however in other applications such

as thin film coating the uniform deposition caused by Marangoni flow is more desirable [1].

In Capillary flow it is assumed that evaporation occurs at the base edge of the drop and the

fluid then flows radially outwards to replace the evaporated fluid. In this particular flow

regime, the droplet usually maintains a constant radius and the contact angle/height decrease.

The deposition of particles at the wetted contact line creates the commonly seen ‘coffee ring

effect’ or ring pattern [1, 21, 22].

Marangoni flow can be the opposite of Capillary flow in that it causes the fluid to circulate

inwards. Surface tension decreases as temperature increases. In a drop, the base is cooler than

the apex which creates a temperature gradient and in turn a surface tension gradient. The

stronger surface tension at the base of the droplet pulls the fluid from the apex downwards

which creates circular motions of fluid along the interface. A detailed account of Marangoni

flow will not be discussed in this review, however the reader is referred to some works that

explain this phenomenon in more detailed fashion [1, 23, 24].

The work by Hu et al. [20] described how the manipulation of each type of flow can lead to

different patterns from drying drops. They proposed that the deposition patterns could be

altered by changing the flow inside the droplet by manipulating its temperature profile. They

proposed two methods of controlling the temperature profile; radial heat transfer to its surface

or using resistive micro heaters. Their experimental results proved this theory and show that

the flow can in fact be influenced which can greatly improve applications that require precise

material deposition. An article by Ristenpart et al. [25] also shows that the manipulation of

Marangoni flow inside the droplet can affect the deposition pattern left from drying. Their

experiment proved that Marangoni flow can be influenced by the relative thermal

conductivities of both substrate and liquid and that the direction of flow changed once a

critical contact angle had been reached.

1.4 Complex evaporation situations

In more complex evaporation situations, there are numerous different factors that must be

considered which make it much more complicated to model. Modelling more complex

situations requires a connection between several mechanisms that act within the drop and

since the modelling of pure liquid systems is still being explored, the modelling of complex

systems is very difficult. It appears that many current models ignore fundamental

mechanisms inside the droplet to simplify the situation allowing it to be modelled. For

example, lubrication theory neglects the vapour dynamics and models evaporation purely on

the evaporation of a liquid film [3]. Some experiments use simplified fluids to model more

complex ones, as seen in the experiments carried out by Meloy Gorr et al. [16] who use a

solution of lysozyme and NaCl to represent a biological fluid.

Works by Sazhin [6, 12] explain very clearly the problems faced in advanced modelling of

droplet heating and evaporation very clearly. Sazhin et al. [6, 12] review current models of

droplet evaporation and then identify unsolved questions. The review identifies further areas

of study that will be vital for further improvements in the modelling of droplet evaporation. It

explains how the use of computational fluid dynamics (CFD) is difficult because some

models may be too simple and others too complex. Models based on the complete Navier-

Stokes equations are the most complicated and are rarely used since the methods required to

solve them are too cumbersome.

In many cases, the fluid can be a suspension. This creates problems since factors such as

particle size, electrostatic interaction between different components in the fluid on the

substrate and each other have an effect on the patterns left from evaporation. A distinction

must be made between a suspension and a phase change system. In the evaporation of a

suspension such as coffee, the liquid evaporates leaving the solid deposition, the pattern of

which can be altered by changing the composition of the suspension [3]. In phase change

systems, such as salt or protein solutions, the residue changes phase to a gel or a solid as the

fluid evaporates. This is an important step in the biomedical applications since blood follows

this regime [15, 26]. Varying the concentration of different components within the fluid can

have a large effect on the behaviour of the drop.

Another area of investigation is the effect of the hydrophobicity of the substrate which has

been reported in numerous studies, including the one by Xu et al. [19]. As too is the effect of

contact line pinning and its dynamics. An everyday example of contact line pinning is when

small raindrops on a window occasionally stick to the surface which appears to contradict

gravity [27]. This situation occurs due to roughness on the glass surface which the water can

grip to [28]. As explained previously, the effects of Marangoni and Capillary flow also have

to be included in the modelling attempt.

2. Patterns from Drying Drops

There are a number of different patterns that can be left from the drying of a fluid drop. The

pattern depends on a number of factors, some of which have been highlighted previously.

Other factors that need to be considered are the effect of atmospheric temperature, substrate

temperature and surface roughness and pattern e.g. Xu et al. [19]. This section will describe

deposition patterns left from similar types of fluid such as polymers, paints, suspensions and

biological fluids and discuss the factors that affect the pattern left.

2.1 Polymers

Polymers exist in colloid suspensions. A lot of previous study has only involved Newtonian

suspensions, however, a polymer suspension is a Non-Newtonian fluid and the mechanisms

involved in its drying are poorly understood. This is an important area to gain understanding

in since polymers are regularly added to fluids used in applications such as inkjet printing to

manipulate the patterns left by the dried drops [1]. Further applications are explained later on.

It is accepted that the pattern left by a dried drop of colloid suspension is most commonly the

prominent coffee-ring effect [29]. Deegan [27] explains that the only conditions required to

form the ring are contact line pinning and evaporation. These are two conditions that are met

in most droplets making the ring a very common phenomenon. As explained previously, this

effect is caused by Capillary flow which sweeps the suspended particles outwards to the edge

of the wetted contact area which can then dry instantaneously to help pin the drop [1]. This is

known as self-pinning where the dried solute particles help to keep the radius of the drop

constant.

In an experiment carried out by Kajiya et al. [29], the change in polymer concentration inside

a drop of solution was recorded using fluorescent microscopy. This allowed the internal

transport processes to be seen clearly and is crucial to improving the understanding of fluid

dynamics inside the drop. In this experiment they used a solution of fluorescent polystyrene

and anisole. The results from the experiment showed that early in the drying process the

concentration of polystyrene increased at the contact line. After this the concentration in the

central region remained fairly constant throughout drying, until later stages. Capillary flow

was responsible for this early movement towards the outer edge and explains why the dried

central region can be very thin. The experiment also proved that the rate of Capillary flow is

proportional to evaporation rate. Therefore, the thickness of the coffee ring can be controlled

by changing the surrounding or substrate temperature. The ability to record the movement of

particles in the fluid opens up avenues to improve the understanding of the mechanisms

taking part in droplet evaporation.

Controlling the central region contained within the outer ring is more complicated. Research

by Jung-Hoon Kim et al. [13] looked at the deposit patterns left by evaporating drops of

water and polymer solutions. The polymer used was Poly(3,4-ethylenedioxythiophene)-

poly(styrenesulfonate) known as PEDOT-PSS. The deposition patterns were affected by

changing the substrate temperature which controlled the direction of internal flow during the

later stages of evaporation. They found that on a cooled substrate (temperature lower than the

droplet temperature), the Marangoni flow inwards was stronger than the outward Capillary

flow. This led to a larger concentration of deposits in the central region. On the room

temperature or heated surface, the Capillary flow was stronger creating the prominent coffee

ring pattern. This coffee ring pattern is concurrent with Kajiya et al. [29], who also saw this

pattern at room temperature. Using the conclusion from Kajiya and co-workers experiment,

Capillary flow is proportional to evaporation rate which is faster at higher substrate

temperatures, explaining the stronger capillary forces at higher temperatures. The ability to

control the deposition pattern has large applications in ink-jet printing, later explained. Figure

1 shows a summary of the cross sectional deposition patterns left by a dried drop at varying

substrate temperatures. The simple diagram below clearly shows how the central region has a

thicker deposits layer at lower temperatures and that the coffee ring effect appears at higher

temperatures.

Figure 1: Cross sectional interpretations of deposits left from dried drops of a solution of

water and PEDOT-PSS at different substrate temperatures

An experiment carried out by Yongjoon et al. [1] investigated the effect of adding different

sized particles into polymer solutions. They added particles of Polystyrene (1 μm and 6 μm)

and hollow glass beads (9-13 μm) separately into three different solutions: pure water,

polyethylene oxide (PEO) and xantham gum (XG). PEO is a flexible polymer and XG is semi

flexible. The results of this experiment highlight the effect of: Newtonian versus Non-

Newtonian fluids, polymer flexibility, polymer elasticity and suspended particle diameter on

the patterns left from the dried drops. Their results showed that suspensions containing

smaller particles, i.e. the 1 μm Polystyrene particles, produced the most noticeable coffee

stain pattern upon drying. Larger particles suspended in the water and in the PEO solution

were not deposited on the edge and formed a more uniform pattern in the middle. Larger

particles suspended in the XG solution did form the coffee stain pattern upon drying.

Therefore, changing the polymer solution can have a large effect on the pattern and so can the

size of particle suspended within it. Figure 2 shows a table of photos of the final pattern from

the fully dried drop of each solution containing each size of particle. This figure has been

created from the figures contained within [1] to give a summary of the different patterns. The

original article shows a timeline of pictures for the evaporation of each drop.

Figure 2: Summary of final patterns left from drops of three solvents (water, PEO and XG)

with three different sized suspended particles as labelled on the diagram. Modified from [1]

A possible explanation for the difference in patterns seen between the two polymer solutions,

PEO and XG, is the difference in their viscosity. XG is a shear thinning fluid, so during low

shear drying, the viscosity remains high since it has a very high zero shear viscosity. The zero

shear viscosity is the viscosity exerted by a fluid while it is stationary. Capillary flow pushes

large particles towards the rim at the beginning of the drying process that are then trapped.

Therefore larger particles in higher viscosity fluids exhibit the coffee ring effect after drying.

This is backed up by the fact that solutions with lower concentration of XG do not exhibit

such dominant coffee stain outer ring patterns [1]. The collection of particles within the ring

can be explained by the attraction of the positive glass substrate and the negative suspended

particles.

Work by Uno et al. [30] investigated the effect of hydrophobic and hydrophilic substrates in

the pattern left by drying drops of polymer latex solution. As expected, on the hydrophilic

surface, the drop followed CRR while drying and produced a circular coffee-ring stain upon

drying. Because the particles are ‘attracted’ to a hydrophilic surface, they adsorbed to it

which creates the ring pattern. On the hydrophobic surface the drop maintained a more

spherical shape which followed CCAR during evaporation. The particles did not adsorb onto

the surface initially, however as the drop continued to evaporate the concentration of particles

increased which caused the particles to clump together and form aggregates. These

aggregates then adsorbed onto the surface to leave a random pattern of ‘spots’ upon drying as

the aggregates were deposited. Figure 3 is a clear illustration depicting the situation explained

here.

Figure 3: Explanation of pattern formation of latex polymer solution of hydrophilic and

hydrophobic substrates

2.2 Paints

Paint is used in small scale applications, such as artwork, and in larger scale applications such

as decorating or protecting metals against corrosion such as in bridges. Artists experiment

with the different patterns left from the drying of paint drops in their work. This is most

commonly used in watercolour paintings where the coffee ring pattern can be seen clearly.

The addition of solvents and other ingredients to paint can create a uniform drying pattern

which is preferential to form an even coating. In these cases, the addition of the ingredients

lessens or prevents the coffee ring effect that is usually seen in suspensions.

The different ingredients each have their own influence on the finish. For example, some are

made from a mixture of water and a thicker solvent. Upon drying, the water evaporates

quickly leaving the pigment carrying particles in the solvent which is highly viscous stopping

the particles from moving outwards. This creates a uniform drying pattern [21].

In an experiment by Abbasian et al. [31], the Marangoni flow during the evaporation of three

solvents that are commonly added to paints was investigated. These three solvents were

xylene, MEK and MiBK. It was found that the patterns left by the pure solvents and solutions

of different solvent ratios had different levels of unevenness. This experiment investigates

film drying, rather than droplets and this tends to be the case in most of the articles on the

patterns left by drying paint.

With new advances in the manipulation of capillary and Marongoni flow during drop

evaporation [20, 23], there could be new avenues of research opening up that investigate the

effect of different solvents on the patterns left from paint droplets. This is an important

application in spray painting.

2.3 Biological fluid

Biological fluids are very complicated since they contain a variety of different proteins and

electrolytes which all interact with each other and affect the mechanisms inside a droplet as it

dries. This makes the understanding of pattern formation very difficult, however, research has

been more concentrated on this area for the past two decades and repetitions of different

patterns are now being found [3, 16].

Yakhno et al. [15, 26, 32] have carried out a lot of investigative work into the patterns left

from biological fluids. In one article, they discussed the results from patterns left from dried

drops of four different fluids: pure water, 0.9 wt% NaCl solution, bovine serum albumin

(BSA) solution and a solution composing of 7 wt% BSA and 0.9 wt% NaCl. Although BSA

is a serum albumin derived from cows, it has a lot of applications in biomedicine including

Enzyme-linked immunosorbent assay (ELISA). This test uses antibodies and colour change

to detect a substance. The patterns left from the dried drop of each sample are shown in figure

4 below.

Figure 4: Patterns left from dried drops of; (a) water, (b) 0.9 wt% NaCl solution, (c) BSA

solution and (d) 7 wt% BSA and 0.9 wt% NaCl solution. Modified from [15]

It is clear from figure 4 that the NaCl solution (b) dries to leave a distinct outer ring of

crystals with some clusters randomly forming inside the ring. The BSA solution (c) dries to

leave a smooth, thick outer ring. It can be assumed that these patterns are formed by the NaCl

or BSA since there is no pattern left from the drop of pure water (a). When NaCl and BSA

are both in solution with water (d), the pattern left by the NaCl is very different since it

crystallises uniformly in the centre of the thick outer ring of deposited BSA, the pattern of

which remains unchanged from the pattern formed by the BSA solution.

The crystallisation of salts within a protein outer ring is further documented by Meloy Gorr et

al. [16] who investigated the effect of lysozyme and NaCl concentrations on the patterns left

from the dried drops of the solutions. Lysozyme is a protein commonly found in tear drops

and saliva of humans. The combination of lysozyme and NaCl offers a very simplified

version of a biological fluid. In pure lysozyme solutions, a thick smooth ring of deposits was

left at the outside of the wetted area as described in the two previous experiments discussed.

As NaCl concentration increased, rough crystalline structures appeared in the centre of the

outer protein ring in dendritic patterns. However, in this experiment, unlike the one carried

out by Yakhno et al., at higher salt concentrations a clear ring formed inside the outer

lysozyme ring with the crystallised salt forming within that. This second ring contained larger

groups of lysozyme molecules.

Although this experiment is carried out using a very simplified version of a biological fluid,

the patterns produced are very similar to those formed by more complex fluids. Therefore,

investigating these simpler fluids might help understand the mechanisms behind the patterns

formed in more complex fluids and further enhance their uses in biomedical applications later

discussed.

Meloy Gorr et al. suggest that the general pattern of salts crystallising inside a protein ring is

due to the separation of the two fluids early on in the evaporation process [16]. They also

noted that initially the drop was pinned and followed a constant radius regime during drying.

Once the outer ring had formed, the drop then changer regime to a constant contact angle

regime until the drop had fully evaporated [16]. This phenomenon was only seen in solutions

with NaCl concentrations of lower than 0.5 wt%. The pattern seen in both of these

experiments can therefore be explained by coupling the suggestion of an early fluid split with

the presence of the two different regimes. If the protein separates from the fluid early, it

would experience Capillary flow, seen in constant radius regime, and would be deposited on

the edge of the initial contact area creating the thick outer ring. As evaporation continues, the

salt that is still in solution begins to crystallise but by this point the drop now experiences

constant contact angle regime so a uniform layer of crystals would be deposited inside the

outer protein ring.

Investigative work has been carried out on women who have just given birth, comparing

those who experienced normal, premature and premature to the point of threatened abortion

childbirth. Yakhno et al. [26, 32] have carried out some of this research. The patterns left

from drops of plasma of women who have just undergone childbirth are similar however a

difference was seen in the thickness of and size of crystals present in the ‘transition zone’ of

the dried drop. Those who experienced premature and extremely premature childbirth (b) had

a wider circle containing larger crystals compared to those who experienced normal

childbirth (a), the circle of crystals being talked about here are indicated by the black arrows

on the photos in figure 5 below.

Figure 5: Patterns left from drying drops of plasma from post childbirth women who

experienced normal childbirth (a) and premature child birth (b). Modified from [26]

Although it is not a biological fluid, a very similar phenomenon where the salt crystallises

inside a ring of potato starch was found by Choudhury et al. [33]. In this situation, the salt

crystallised into dendritic fractal formations with a thick outer ring of pure starch surrounding

it (the ring of starch is denoted by arrow A, near the bottom edge of figure 6).

Figure 6: Picture of a dried drop of starch and NaCl gel showing enlarged photos of certain

areas of the deposited pattern. Taken from [33]

2.4 Nano-particle suspensions

Many experiments have been carried out using gold particle suspensions [34, 35]. One

experiment by Budhadipta et al. [34] found a method for dispersing gold nanorods onto a

single walled carbon nanotube (CNT) macrostructure. Their method used the drying of drops

of nanorod suspensions on a single walled carbon nanotube substrate in ambient conditions.

They found that the factor controlling the pattern of GNRs on the substrate was the

anisotropic interaction between the nanorod and the single walled carbon nanotube substrate

which means that gold nanorods aligned with the direction of the single walled carbon

nanotube microfibers. The experiment was also carried out using triangle and polygonal

shaped gold particles and the same result was found. The spontaneous alignment with the

single walled carbon nanotube substrate is due to a combination of forces acting on the

nanoparticles. As the drop continues to evaporate, the volume of liquid present in the drop

falls until the nanoparticles are sitting in a very thin film of fluid. The capillary force then

promotes the particles to lie horizontally. Van der Waals forces promote the particles to

attach to the substrate with maximum contact area, in the case of nanorods this is horizontally

in the micro channels on the single walled carbon nanotube. Figure 7 shows the pattern of

gold nanorods on the single walled carbon nanotube substrate and clearly shows the

alignment in the micro channels. To achieve higher concentrations of gold nanorods on the

surface, the drop evaporation process was repeated on the same area.

Figure 7: SEM image of: (a) single walled carbon nanotube fibre [scale bar: 40 μm] and

(b,c) spontaneous gold nanorods alignment on single walled carbon nanotube fibre [scale

bar: 0.2 μm]. Gold nanorods deposited from dried drops of dilute gold nanorods solution.

From [34]

Darwich et al. [35] used the natural evaporation of nanodroplets to produce nanoring patterns

of gold particles. It can be seen that the gold particles form a ring spontaneously upon drying

or they can be introduced into a system which allows the diameter and placement of the

particles to be controlled. The ring of gold particles that forms around the edge of the wetted

area is the same as the coffee-ring effect that is seen in many other fluids explained

previously in section 2. The nanodroplets form quickly and reliably on the hydrophobic

substrate. The advantage of this technique over others which use nanosized moulds and

templates is that the rings can be made much quicker since the nanodroplet formation is fast.

A disadvantage seems to be that it is slightly more difficult to control the size of the rings

which using a mould ensures. Figure 8 shows the nanoring deposition pattern of gold

particles.

Figure 8: AFM images of gold nanoring structures formed by the drying of nanodroplets of

gold suspension (a) Frame size 1 μm and (b) 0.3 μm

3. Applications

3.1 Polymers

The work explained previously, by Uno et al. [30], started due to the demand of materials to

coat building exteriors in, that would exhibit ‘self-cleaning’ properties. These materials

would therefore keep maintenance costs to a minimum since stains would be more easily

removed. The experiment carried out in their work aimed to give a better understanding of

why the striped stain patterns that form on building exteriors appear more commonly on

hydrophobic surfaces rather than hydrophilic. To do this they created a polymer latex solution

from styrene and p-styrenesulfonate to signify the rain drop since they believed the pattern

formation to be caused by the different pollutants present in the rain drop. The results from

this experiment proved that the aggregates adhered more firmly to a hydrophilic surface than

the hydrophobic and means that study into a suitable material should begin with hydrophobic

substances.

A second application for polymer solutions is in ink-jet printing. The manipulation of the

pattern of solute left on the substrate is important in this application [29]. The commonly seen

coffee ring pattern affects the thickness of the deposited polymer film created during ink jet

printing and also enhances the sharpness of the image.

3.2 Biomedicine

The composition of biological fluids are altered by disease and diet [3], and as earlier stated,

the patterns left from drying drops are altered by the composition of the fluid drop. This

means that the patterns left from the drying of biological fluids could be used as a cost

effective and fast means of diagnosis of certain diseases. Different biological fluids have been

investigated such as blood, serum (blood plasma with the blood clotting components

removed) and tear drops in various studies [15, 16, 26, 32, 36, 37].

The development of this diagnosis technique began with the ‘Litos’ test system which was

established in Russia. The use of this system arose after it was realised that urine salts of

patients suffering from urolithiasis, or more commonly known as kidney stones, would

crystallise in the biological fluid and therefore leave a distinct pattern [3, 38]. Since the

discovery of this phenomenon, it has attracted others to study the possibility of this diagnosis

technique. However, the technique relies on comparing a drop pattern from an infected

patient to that of a healthy patient (control). This requires the collection of numerous samples

for each disease/condition before a good comparison can be made. This explains why it has

taken a couple of decades for big improvements in diagnostics, since repeatable and distinct

patterns are now being seen in the residue left by drops of similar biological fluid.

The problem with using the deposition patterns as a means of diagnosis is the subjective

nature of the work since there are no exact patterns to look for, it is very much a judgment

made by eye. An advantage of this technique is that certain diseases, such as cancer and

kidney disease, that are difficult to diagnose early might be detected sooner since they can

cause high levels of certain proteins in the blood [3]. This change in protein production and

concentration can be spotted early using drop patterns since it will have an effect on the

pattern left from drying.

Work by Yakhno et al. [32] has proven that this attractive idea of diagnosis from deposition

patterns of blood drops is realistic. It is not only cheap and fast, but it is not invasive for the

patient and the procedure can be used by less qualified personnel. The experiment carried out

by Yakhno et al. looked at blood samples from different patients with one of eight conditions:

(a) control, (b) breast cancer, (c) lung cancer, (d) paraproteinemia, (e) in-time delivery, (f)

premature delivery, (g) threatened abortion and (h) hepatitis. The patterns left by samples

from three patients for each condition are shown in figure 9. This figure was modified from

an image in article [32] in which patterns from five patients were shown for each condition.

It can be seen from figure 9 that the patterns left by serum drops with different conditions

vary massively making them easy to differentiate between, however the patterns also differ

slightly from patient to patient with the same condition. This makes it difficult to diagnose

patients with certainty and makes the process very subjective.

Figure 9: Patterns left from dried drops of serum from blood samples of individuals with the

7 listed conditions and one control. Modified from [32]

The investigations of human tear fluid might give indications of ocular disease. In an article

by Filik [37] the importance of the concentrations of different proteins in the tear fluid are

described, they are responsible for keeping the cornea lubricated to allow blinking and

fighting infection etc. A fluctuation in protein levels could give an indication of ocular

disease. This article only describes the use of Raman Spectroscopy, but a similar diagnosis

could be concluded by comparing the patterns from dried drops of tear fluid since the levels

of protein will alter the patterns.

3.3 Nanotechnology

The advent of nanotechnology in the past decade has enticed a large number of people to

research this area. This can be seen by the large increase in the number of publications, there

were 720 papers published in 2010 and 2011 alone [39]. Nanotechnology can utilise the

patterns from drying drops to make new materials, such as gold nanorings or to deposit

materials onto micro-structures, for example gold nanorods spontaneously placed onto single

walled carbon nanotube surface [34]. Patterns formed by drying drops laden either with

nanoparticles or fullerenes have revealed complex formations. The full understanding of the

mechanisms behind these patterns and their exploitation for various technological purposes is

still an open field, Figure 10.

Figure 10 : (a) Patterns from drying Fullerenes, Y.Chen et al., Appl. Phys. Lett. 102, 041911 (2013)]. (b) Patterns formed from Al2O3-H2O nanofluid droplets at various temperature and concentrations [3].

(a) (b)

3. Conclusions

The most common and well understood deposition pattern from a drying drop is the coffee-

ring effect. It is well documented in most papers regarding this topic [1, 20, 21, 25, 29, 40,

41]. Deegan et al. [27] explained that the only conditions required to form the ring are contact

line pinning and evaporation. These are two conditions that are met in most droplets making

the ring such a recurrent phenomenon. As explained previously, this effect is caused by

Capillary flow which sweeps the suspended particles outwards to the edge of the wetted

contact area which can then dry instantaneously to help pin the drop [1]. This is known as

self-pinning where the dried solute particles help to keep the radius of the drop constant.

The suppression or prevention of Capillary flow can be achieved by changing different

factors including; substrate temperature, suspended particle size and solvent type. By

changing the strength of Marangoni and Capillary flow inside the drop the deposition pattern

left can be manipulated. Other types of patterns seen from the drying of drops are; uniform,

central deposits and inner rings. Jung-Hoon Kim et al. [13] concentrated on the manipulation

of the central region of deposits in their experiment and found that a cooler substrate

temperature decreases the Capillary flow and increases the inward Marangoni flow. An

experiment by Yongjoon et al. [1] showed that larger suspended particles do not show the

coffee ring effect as predominantly as smaller particles and tended to be deposited in the

central region.

Biological fluids are incredibly complex, containing numerous different materials that

interact with each other to create very complicated patterns and make them very difficult to

model. Disease and diet can change biological fluid composition which changes the patterns

left from dried drops of infected patients, allowing for its use in diagnosing patients. A good

example of the different types of pattern seen from infected patients is given in figure 9. The

use of patterns from drying drops in diagnostics can start to be more widely used now that a

large database of repeatable patterns has been collected following the increase in research

over the past two decades [3, 26, 36].

The understanding of mechanisms behind patterns left from complex fluids is still lacking.

The experiment carried out by Kajiya et al. [29] can open up avenues to help improve the

existing knowledge of flow regimes inside an evaporating drop through their use of

fluorescent microscopy. It would be nice to see the patterns from drying drops used for the

early diagnosis of patients since it is a cheap and minimally invasive method. The patterns

left from drops of biological fluid of similar condition should be regularly recorded and with

the increase in repeatable patterns it can be hoped that this technique will be used more in the

future and with more confidence.

References

1. Choi, Y., J. Han, and C. Kim, 22. Pattern formation in drying of particle-laden sessile drops of polymer solutions on solid substrates. Korean Journal of Chemical Engineering, 2011. 28(11): p. 2130-2136.

2. Erbil, H.Y., 4. Evaporation of pure liquid sessile and spherical suspended drops: A review. Advances in Colloid and Interface Science, 2012. 170: p. 67-86.

3. Sefiane, K., 2. On the Formation of Regular Patterns from Drying Droplets and Their Potential Use for Bio-Medical Applications. JOURNAL OF BIONIC ENGINEERING, 2010. 7: p. S82-S93.

4. Picknett, R.G. and R. Bexon, 18. The evaporation of sessile or pendant drops in still air. Journal of Colloid and Interface Science, 1977. 61(2): p. 336-350.

5. Tonini, S. and G.E. Cossali, 19. An analytical model of liquid drop evaporation in gaseous environment. International Journal of Thermal Sciences, 2012. 57(0): p. 45-53.

6. Sazhin, S.S., 20. Advanced models of fuel droplet heating and evaporation. Progress in Energy and Combustion Science, 2006. 32(2): p. 162-214.

7. Holyst, R., Evaporation processes: 100 years of misconceptions. TCE: The Chemical Engineer, 2008(806): p. 36-37.

8. Fenwick, E. and J.C. Maxwell, A bibliography of James Clerk Maxwell. Part 1, Books by Maxwell / Edward Fenwick. 2009: Edinburgh : Edward Fenwick, 2009.

9. Crowe, C.T., M. Sommerfeld, and Y. Tsuji, Multiphase flows with droplets and particles / Clayton Crowe, Martin Sommerfeld, Yutaka Tsuji. 1998: Boca Raton, Fla. ; London : CRC Press, c1998.

10. Meric, R.A. and H.Y. Erbil, 15. Evaporation of sessile drops on solid surfaces: Pseudospherical cap geometry. Langmuir, 1998. 14(7): p. 1915-1920.

11. Fuchs, N.A., Evaporation and Droplet Growth in Gaseous Media, in Evaporation and Droplet Growth in Gaseous Media B2 - Evaporation and Droplet Growth in Gaseous Media. 1959, Pergamon Press: London.

12. Sazhin, S.S., et al., 8. New approaches to numerical modelling of droplet transient heating and evaporation. International Journal of Heat and Mass Transfer. 48: p. 4215-4228.

13. Kim, J.H., S.B. Park, and W.C. Zin, 38. Polymer Transports Inside Evaporating Water Droplets at Various Substrate Temperatures. Journal of Physical Chemistry C, 2011. 115(31): p. 15375-15383.

14. Erbil, H.Y., G. McHale, and M.I. Newton, 16. Drop evaporation on solid surfaces: Constant contact angle mode. Langmuir, 2002. 18(7): p. 2636-2641.

15. Yakhno, T.A., et al., 3. Drops of biological fluids drying on a hard substrate: Variation of the morphology, weight, temperature, and mechanical properties. Technical Physics, 2010. 55(7): p. 929-935.

16. Heather Meloy, G., et al., 5. Salt-induced pattern formation in evaporating droplets of lysozyme solutions. Colloids and Surfaces B: Biointerfaces, 2012.

17. Guilizzoni, M. and G. Sotgia, 14. Experimental analysis on the shape and evaporation of water drops on high effusivity, microfinned surfaces. Experimental Thermal and Fluid Science, 2010. 34(1): p. 93-103.

18. Panwar, A.K., S.K. Barthwal, and S. Ray, Effect of evaporation on the contact angle of a sessile drop on solid substrates. Journal of Adhesion Science and Technology, 2003. 17(10): p. 1321-1329.

19. Xu, W. and C.-H. Choi, Experimental Studies on Evaporation Kinetics and Wetting Dynamics of Nanofluid Droplets on Superhydrophobic Surfaces of Micro-post Patterns. Journal of Adhesion Science & Technology, 2011. 25(12): p. 1305-1321.

20. Hu, H. and R.G. Larson, 32. Marangoni effect reverses coffee-ring depositions. Journal of Physical Chemistry B, 2006. 110(14): p. 7090-7094.

21. Yunker, P.J., 6. Coffee-rings and glasses: colloids out of equilibrium. 2012.22. Deegan, R.D. and O. Bakajin, 45. Capillary flow as the cause for ring stains from dried liquid

drops. Nature, 1997. 389(6653): p. 827.23. Hu, H. and R.G. Larson, 21. Analysis of the effects of Marangoni stresses on the microflow in

an evaporating sessile droplet. Langmuir, 2005. 21(9): p. 3972-3980.24. Girard, F., M. Antoni, and K. Sefiane, 25. On the effect of Marangoni flow on evaporation

rates of heated water drops. Langmuir, 2008. 24(17): p. 9207-9210.25. Ristenpart, W.D., et al., 33. Influence of substrate conductivity on circulation reversal in

evaporating drops. Physical Review Letters, 2007. 99(23).26. Yakhno, T.A., et al., 26. On the Existence of Regular Structures in Liquid Human Blood Serum

(Plasma) and Phase Transitions in the Course of Its Drying. Technical Physics, 2003. 48(4): p. 399.

27. Deegan, R.D., 23. Pattern formation in drying drops. Physical Review E, 2000. 61(1): p. 475-485.

28. Dettre, R.H. and R.E. Johnson, Contact Angle, Wettability, and Adhesion, ACS Advances in Chemistry, in Contact Angle, Wettability, and Adhesion, ACS Advances in Chemistry B2 - Contact Angle, Wettability, and Adhesion, ACS Advances in Chemistry. 1964, American Chemical Society: Washington. p. 136.

29. Kajiya*, T., D. Kaneko, and M. Doi, 35. Dynamical Visualization of “Coffee Stain Phenomenon” in Droplets of Polymer Solution via Fluorescent Microscopy. Langmuir, 2008. 24(21): p. 12369-12374.

30. Uno, K., et al., 24. Particle adsorption in evaporating droplets of polymer latex dispersions on hydrophilic and hydrophobic surfaces. Colloid and Polymer Science, 1998. 276(9): p. 810-815.

31. Abbasian, A., et al., 42. Study on different planforms of paint’s solvents and the effect of surfactants (on them). Progress in Organic Coatings, 2004. 49(3): p. 229-235.

32. Yakhno, T.A., et al., 34. The informative-capacity phenomenon of drying drops. Ieee Engineering in Medicine and Biology Magazine, 2005. 24(2): p. 96-104.

33. Choudhury, M.D., et al., 1. Pattern Formation in a Drying Drop of Starch Gel with Sodium Chloride. 2010.

34. Dan, B., et al., 43. Templating of Self-Alignment Patterns of Anisotropic Gold Nanoparticles on Ordered SWNT Macrostructures. Acs Applied Materials & Interfaces, 2011. 3(9): p. 3718-3724.

35. Darwich, S., et al., 44. Nanobubble and nanodroplet template growth of particle nanorings versus nanoholes in drying nanofluids and polymer films. Nanoscale, 2011. 3(3): p. 1211-1217.

36. Brutin, D., et al., 11. Pattern formation in drying drops of blood. JOURNAL OF FLUID MECHANICS, 2011. 667: p. 85-95.

37. Filik, J. and N. Stone, 29. Analysis of human tear fluid by Raman spectroscopy. Analytica Chimica Acta, 2008. 616(2): p. 177-184.

38. Christy, J.R.E., K. Sefiane, and E. Munro, 17. A Study of the Velocity Field during Evaporation of Sessile Water and Water/Ethanol Drops. Journal of Bionic Engineering, 2010. 7(4): p. 321-328.

39. Resource, E., 4. Advances in colloid and interface science [electronic resource]. 1967: Amsterdam ; New York : Elsevier.

40. Berteloot, G., et al., 27. Evaporation of a sessile droplet: Inside the coffee stain. Journal of Colloid and Interface Science, 2012. 370(1): p. 155-161.

41. Choi, Y., J. Han, and C. Kim, Pattern formation in drying of particle-laden sessile drops of polymer solutions on solid substrates. Korean Journal of Chemical Engineering, 2011. 28(11): p. 2130-2136.