Eur. Phys. J.

E (2023) 46:113
https://doi.org/10.1140/epje/s10189-023-00374-8
THE EUROPEAN
PHYSICAL JOURNAL E
Regular Article - Soft Matter

Drying behaviour of nanofluid sessile droplets on

self-affine vis-à-vis corrugated nanorough surfaces
Deeksha Rani and Subhendu Sarkara

Surface Modification and Applications Laboratory (SMAL), Department of Physics, Indian Institute of Technology Ropar,
Nangal Road, Rupnagar, Punjab 140001, India

Received 1 September 2023 / Accepted 24 October 2023 / Published online 24 November 2023
© The Author(s), under exclusive licence to EDP Sciences, SIF and Springer-Verlag GmbH Germany,
part of Springer Nature 2023

Abstract In recent years, evaporative self-assembly of sessile droplets has gained considerable attention
owing to its wide applicability in many areas. While the phenomenon is well studied for smooth and isotrop-
ically rough (self-affine) surfaces, investigations comparing the outcomes on self-affine vis-à-vis corrugated
surfaces remains to be done. In this experimental work, we compare the wetting and evaporation dynam-
ics of nano-colloidal microlitre droplets on self-affine and corrugated nanorough surfaces having identical
roughnesses and interface properties. The coupled influence of particle size, concentration, and surface
structuring has been explored. Differences in wettability and evaporation dynamics are observed, which
are explained via the interaction between wetting fluid and anisotropic surface roughness. Our findings
exhibit different temporal behaviour of contact radius and angle in the evaporation process of the droplets.
Further, the corrugated surface exhibits anisotropic wettability with a monotonic change in droplet shape
as evaporation proceeds, finally giving rise to irregular dried patterns. The scaled rim width and crack
spacing of the particulate deposits are examined. Our results can inspire fabrication of surfaces that can
facilitate direction-dependent droplet motion for specific applications.

1 Introduction the surface [9,10]. The dried droplet patterns can be

influenced by the internal flows, solute properties, con-
The phenomenon of surface wetting and evaporation- tact line dynamics, atmospheric conditions, wettability
induced particle self-assembly has been an area of active of surface, and interparticle as well as particle–substrate
theoretical as well as experimental research since it was interactions.
first introduced by Young [1,2]. The topic gained atten- Especially for the case of droplet evaporation, sev-
tion due to its wide applicability in areas like print- eral studies to date focus on the investigation of
ing, lubrication, coating, waterproofing, microfluidics, the evaporation patterns, the evaporation modes [11],
forensics, biotechnology, nanochromatography, analyte particle–liquid–substrate interactions [12], hydropho-
detection, etc. [3–7]. The evaporation of a sessile droplet bicity (hydrophilicity) of the substrate [13], etc. An
on a surface satisfying Laplace’s equation [8] is primar- ideal solid surface is considered to be atomically
ily governed by the evaporation flux from the droplet smooth, and hence, its wetting and subsequent evap-
surface associated with pinning or depinning of the oration characteristics are decided by the interfacial
droplet. This effect has led to several interesting phe- free energies [14]. On the contrary, a real (or fabri-
nomena reported both by experimental and theoreti- cated) surface is topographically rough at the macro- or
cal groups during the last few years. In this context, micro-scale or chemically heterogeneous. The presence
the evaporation dynamics of microlitre-sized droplets of of such textured topography is one of the key factors
colloidal suspensions exhibit spatio-temporal behaviour that directly affect the interaction between the liquid
at the solid–liquid interface, wherein it can get pinned, and substrate [15]. In particular, when a pure liquid is
eventually producing a ring-shaped deposition pattern, put on a self-affine surface [16,17], the droplet usually
often referred to as the “coffee ring effect” (CRE) as a preserves its typical hemispherical shape thereby con-
result of particle migration towards the pinned three- fining itself at the CL leading to enhancement in evapo-
phase contact line (CL) owing to non-uniform evapora- ration rate [13,18]. This usually arises due to the liquid
tive fluxes [2]. The formation of deposits and flow fields interacting with the asperities of the rough surface and
may, however, be hindered by flow convection within driving it to expand laterally at the surface, resulting
the drop as a result of variations in surface tension along in an increase in liquid–vapour area and a decrease in
surface tension at the contact line. Studies suggest that
a
e-mail: sarkar@iitrpr.ac.in (corresponding author) surface structuring including shape, width and height

123
113 Page 2 of 11 Eur. Phys. J. E (2023) 46 :113

of substrate nano-asperities can influence several pro- will lead to interesting anisotropic drying patterns on
cesses like droplet evaporation [19], interfacial particle– the substrate surface.
substrate forces [20], dried droplet morphology [21], etc. In this work, we systematically investigated in detail
Surface roughness can, however, be classified into the wetting and evaporation characteristics of droplets
isotropic (self-affine) as well as anisotropic ones. For of colloidal suspensions on self-affine vis-à-vis corru-
self-affine roughness geometries and chemical hetero- gated nano-patterned surfaces having identical root-
geneities, the sessile droplets exhibit an isotropic beha mean-square (rms) roughnesses. The latter helped in
viour in terms of its shape and CL uniformity [22,23]. distinguishing the effect of overall roughness from that
Experimenting with pure liquid sessile droplets, Chen of its intricate surface morphology. Both surfaces exhib-
et al. demonstrated that the droplet contact angle ited particular sets of evaporation modes depending
(CA) and the evaporation modes are influenced by on the surface morphology, with the periodic rough
μm sized and spaced asperities [24]. Garg and co- surface showing a monotonically increasing anisotropic
workers showed that superwetting behaviour can be wettability which can be explained in the light of pin-
achieved using similar structures of ZnO micropillars ning in the presence of corrugation crests. In addition,
having spatial periodicities of μm order [25]. Recently, droplet deformation was also observed on the periodic
work by Hou and co-workers suggested the role of the surface, which corroborated with the transitions taking
topographical parameters in affecting the surface ten- place in the different modes of evaporation for the sur-
sion at the CL thereby leading to Cassie–Wenzel tran- faces. Finally, the deposition patterns formed by the
sitions [26]. In contrast, topographically anisotropic dried droplets on both surfaces were studied in the
surfaces in the forms of parallel lines, grooves, wrin- light of the corresponding morphological characteris-
kles, pillars, etc., present a non-uniform shape and tics. Delaminated effects were examined in the dried
apparent contact angle at the CL thereby exhibit- deposits and explained in light of stress effects.
ing anisotropic wetting behaviour [27,28]. Of late,
this phenomenon has gathered considerable interest
owing to potential applications in microfluidics, direc-
tional flow control devices, air-cooling applications, 2 Experimental methods
heat exchangers, drainage enhancement, MEMS, etc.
[29]. The physical origins of the anisotropic wetting Samples of size 1 cm × 1 cm were cut from undoped sil-
phenomenon are attributed to the liquid CL encoun- icon wafers of orientation (100) with a diameter ∼ 2
tering a physical discontinuity [30] or chemical het- and resistivity ∼ 100 ohm-cm. The samples were con-
erogeneity present on a substrate [31]. On anisotropic secutively cleaned with acetone and isopropanol by
surfaces, droplets prefer to spread or roll freely along ultrasonication for 10 min, respectively. Rinsing with
parallel orientation of grooves, while being restrained de-ionized (DI) water (ρ ≈ 18.2MΩ@250 C, Total car-
in other orthogonal direction or vice-versa. Zhao and bon content (TOC) < 5 ppb and Bacterial count <
co-workers had observed weak anisotropic wetting phe- 0.1 CFU/mL) was followed thereafter using a Milli-
nomenon of water droplets on polymer surfaces exhibit- pore DirectQ water purification system. For wet etch-
ing shallow periodic grooves of 0.4 μm periodicity and ing, HNA solution (using HF, HNO3 , and CH3 COOH
50 nm depth [32]. Studies on different hierarchical sys- by volume ratio of 3:3:5) was prepared [34]. A homo-
tems imprinted on polystyrene (PS) and PMMA thin geneous solution was made using a magnetic stirrer for
films with 2 μm height and 250 nm pitch clearly indi- 10 min, and then, the sample was etched ultrasonically
cate a comparatively good anisotropic wettability [28]. for 2 min. Samples were washed with de-ionized water
A further study showed that wavy microscopic struc- and dried after etching. The etching time was optimized
tures fabricated on PDMS surfaces having a periodic- to achieve a certain surface roughness. A second set
ity of 41 μm with a root-mean-square (rms) roughness of samples was prepared by low-energy Ar+ ion beam
variation between 20 and 140 nm exhibited wettability irradiation at 500 eV incident beam energy and 13.3
changes from anisotropic to isotropic with the increase mA incident beam current using a Kaufman type 4 cm
in surface roughness [33]. broad beam ion source. Upon evacuation, a base pres-
Although the above studies explored the wetting sure of 4 × 10−7 mbar was achieved prior to bombard-
phenomenon, they lacked information on the follow- ment inside the vacuum chamber. The ion beam was
ing aspects. Firstly, the evaporation dynamics was not incident at an angle of 67◦ to the surface normal. Ion
explored in any of the above studies. Secondly, they irradiation was done for 5 mins which amounted to a
were all done using pure water on microstructured sur- total fluence of 2.5 × 1019 ions cm−2 [35,36]. Both the
faces. It is imminent, in this context, that a study above processes render the top surface amorphous. The
on nanostructured surfaces using nano-colloidal water resulting surfaces were investigated by field emission
droplets will yield interesting results. Further, droplets scanning electron microscope (FESEM) (JEOL JSM-
of colloidal suspensions on such surfaces could exhibit 7610F Plus, JEOL Ltd., Japan). The surfaces were
unconventional evaporation dynamics influenced by also investigated using atomic force microscopy (AFM)
particle-surface interactions at scales 103 orders of mag- which was operated in the tapping mode with 512 × 512
nitude different from what is normally experienced on lines of image resolution and 3 and 5 μm scan sizes. All
surfaces with μm sized structures. Consequentially, this samples were analysed utilizing AFM data to calculate
periodicity, roughness and other surface properties.

123
Eur. Phys. J. E (2023) 46 :113 Page 3 of 11 113

For the sessile drop evaporation experiments, col- mated from the line profile measurements. Other sta-
loidal solutions of Ludox silica nanoparticles (Sigma- tistical data pertaining to AFM analyses are furnished
Aldrich) with average particle sizes of 12 nm (HS-30; in Table 1 of SI. The roughness evolution for the chemi-
30 wt% suspension in H2 O) and 22 nm (TMA; 34 wt% cally etched surface is primarily governed by the chem-
suspension in H2 O) were used. The colloidal concen- ical kinetics that gains control over the etching mecha-
trations employed for the studies were 0.2, 0.5 and nism [34]. The reaction encompasses an oxidation step
0.7 vol%. These were achieved by diluting the above followed by a dissolution of silicon oxide. For IIS on
nanoparticle solutions with appropriate amounts of DI the other hand, the evolved roughness and periodic-
water in order to make the concentrations the same. To ity are ascribed to the self-organization of the surface
ensure homogeneity, the colloidal suspension was son- atoms caused by an energetic ion beam as explained
icated for 20 mins prior to each experiment. A 1 μL in literature [35–38]. More importantly, the orienta-
sessile droplet was drop cast on each sample using an tion of the nano-corrugations formed on the surface is
inbuilt micropipette in the drop shape analyser (Kruss an outcome of irradiation-induced instabilities due to
DSA25E). Evaporation of sessile droplet was carried curvature-dependent sputtering and a surface-diffusive
out with measurement of base diameter, wetting con- smoothing mechanism which reduces the surface ten-
tact angle and volume. Advancing and receding con- sion of the irradiated surface.
tact angles were obtained during the onset of contact
line motion while injecting or withdrawing liquid using
a Theta Flex optical tensiometer (Biolin Scientific). 3.2 Wettability of surface
This involved the insertion of a stainless steel syringe
tip with 0.5 mm outer diameter into a 5 μL droplet Figure 2 shows representative optical images for ses-
and gradually changing the volume at a flow rate of sile droplets of colloidal suspensions on the isotropic
1 μL/min. All experiments were performed in a tem- (Fig. 2a) and anisotropic surfaces (Fig. 2b, c) in direc-
perature and humidity-controlled chamber at 25 ± 1◦ C tions parallel and perpendicular to the direction of cor-
and 40 ± 5% relative humidity. The resulting deposition rugations, respectively. The schematic (not to scale)
patterns were studied using a Nikon Eclipse LV100ND shows measurement geometry for parallel and perpen-
optical microscope with ∼ 5×to 50× magnification. All dicular directions with respect to the corrugation direc-
experiments were repeated 4–5 times to ensure repro- tion. IIS samples were marked on the back side indicat-
ducibility. ing the ion beam direction prior to irradiation. This
acted as a reference for the subsequent experiments of
AFM and droplet angle to ascertain the relative ori-
entation of the corrugations with respect to the mark.
3 Result For the WES, the contact angle (CA) was found to
be the same irrespective of the direction of measure-
3.1 Surface morphology ment, implying that it exhibits isotropic wettability
with a slightly hydrophobic nature evidenced by a con-
Figure 1 shows FESEM images of the nanostructured tact angle of about 96◦ as displayed in Fig. 2a. This is
surfaces. It is clear that the wet etched surface (WES) primarily due to strong pinning of the sessile droplet
shows an isotropic self-affine morphology where the on the isotropic nanorough surface which can be con-
surface heights scale by different amounts in different firmed by a higher contact angle hysteresis caused by
directions [16], whereas the ion irradiated surface (IIS) the asperities on WES (Table 2 in SI). Studies suggest
shows an anisotropic morphology with nanometer-order that the presence of surface asperities pin the CL con-
corrugated structures formed on its surface in a pref- siderably, leading to an increase in particle concentra-
erential direction. Additionally, Fig. S1 of Supporting tion reaching the random-packing limit at times much
Information (SI) shows 2 × 2 μm2 AFM 2D topograph- earlier than what happens for a flat surface [39]. The
ical images of the above surfaces. The 2D Fast Fourier closely spaced height fluctuations contribute towards
Transformation (FFT) insets corroborate the observed this strong pinning phenomenon. The drop takes a
periodicities since the WES morphology shows a circu- spherical-cap shape, and CA is found to be uniform
lar FFT pattern with no characteristic spots. The FFT along the CL as noticed in a previous study [40]. The
for the IIS surface on the other hand shows distinct corrugated surface, on the other hand, exhibits static
side-lobes, thereby indicating the presence of periodic contact angles measured in orthogonal directions, i.e.
structures with a characteristic wavelength. The sur- both parallel and perpendicular to the corrugations.
face uniformity is evident from both the FESEM and These angles are found to be slightly less than that
AFM images. AFM measurements reveal that the root- of the WES. The lowest CA and better wettability
mean-square (rms) roughness of the chemically etched were exhibited by IIS in the parallel orientation, while
surface is 3.5 ± 0.2 nm while that of the irradiated sur- the highest by WES. It is further observed that mea-
face is 3.6 ± 0.3 nm (± indicates the standard devia- surements in the perpendicular direction show larger
tion in measurements). These values have been aver- contact angles (75◦ ) than the parallel one (69◦ ). This
aged over 4–5 scans acquired at different places across clearly suggests formation of elliptical droplets as com-
the sample. In addition, the corrugations formed on the pared to the spherical ones for the case of WES. The
irradiated surface have a periodicity of 40 nm as esti- corresponding CA measurements have been taken care

123
113 Page 4 of 11 Eur. Phys. J. E (2023) 46 :113

Fig. 1 Field-emission scanning electron microscopy (FESEM) images showing surface topographies of a WES and b IIS
surfaces, respectively

Fig. 2 Top panel: Schematic (not to scale) showing top tration and 12 nm particle size showing apparent wetting
view of droplets on WES and IIS. The orthogonal view- contact angle on a WES and IIS in b parallel and c per-
ing directions for IIS are shown by red arrows with respect pendicular direction of corrugations. Yellow arrows indicate
to the corrugation periodicities. Bottom panel: Side view of the droplet contact line from where the measurements have
1 μL droplets of colloidal suspensions of 0.5 vol% concen- been made

of by elliptical fits using the Young–Laplace method by direction exhibits larger hysteresis as well as contact
considering the surface curvature of the droplet result- angle as compared to the parallel one. For compari-
ing from the equilibrium between surface tension and son, experiments were also carried out for non-colloidal
weight of the liquid [41]. The irradiated substrate, thus, water droplets to observe the pinning effect and sep-
shows hydrophilicity with anisotropic wettability lead- arate out the influence of colloidal particles on the
ing to a droplet deformation during evaporation. The wetting phenomenon (results given below). Anisotropic
anisotropicity was further confirmed by advancing and wettability has been observed in earlier studies as well
receding angle experiments, leading to the demonstra- [30,32]. Their results showed that the apparent CA
tion of contact angle hysteresis for the two cases (Table when observed in a direction parallel to the corruga-
2 of SI) indicating the pinning effect at CL caused by tions/grooves was higher than that for the perpendicu-
surface asperities. It is observed that the perpendicular lar direction. It is to be noted here that the corrugations

123
Eur. Phys. J. E (2023) 46 :113 Page 5 of 11 113

accounted for their cases were in the μm range, while wetting behaviour for the droplet of colloidal suspen-
for the present study they are in the range of nanome- sion, with a maximum contact angle difference of 6◦
ters. Considering the shorter wavelength (40 nm) of the between parallel and perpendicular directions. For all
corrugations for the present case in comparison with the rough surfaces, it is observed that the larger parti-
the larger ones (∼ μm) as found in the earlier stud- cle size shows a higher CA for lower concentrations. The
ies, the pinning of the droplet is effectively dominated contact angles increase with increasing particle con-
by the crests of the undulations when the droplet is centration and size which indicates that the dispersed
observed in the parallel direction. A deciding factor here nanoparticles alter the force balance near the CL. This
is the number of crests over which the droplet rests. For is envisaged to be an outcome of particle–surface and
the present case, the number is very high as compared particle–particle interactions via Derjaguin–Landau–
to earlier studies since the wavelength is in the range Verwey–Overbeek (DLVO) forces [20,42], as discussed
of nanometers. A lower apparent CA (for the corru- in SI. These interactions become more effective as evap-
gated surface in the parallel direction) than the WES oration proceeds since the surface-to-volume ratio of
can be explained by the fact that the pinning by the particles increases, eventually leading to more particle-
well-separated corrugation crests is slightly lower than surface interactions. However, droplets having smaller
the densely placed height fluctuations for the isotropic particle sizes are also effective in raising CA for identical
rough surface. Hence, the squeezing is less for the par- droplet volumes but for higher concentrations. Radiom
allel configuration [40]. In the perpendicular direction, et al. reported similar findings in their work for parti-
the droplet is stretched along the corrugations. How- cle size 260 nm and concentrations ranging from 0.05 to
ever, the few defects along the corrugations act as pin- 2 vol% [43]. However, in their case, the substrate was
ning centres for the droplet. The three-phase CL thus considered to be smooth and hydrophilic.
resembles closer to a pristine surface (p-Si), except for Further, an estimation of total free energy change
the presence of a few defects at certain places. This (ΔG) based on a thermodynamic model [32] reveals dif-
leads to an initial CA of 75◦ which is quite close to that ferent values for the orthogonal directions. This model
of a pristine Si (73◦ ) surface. The above considerations separates out the free energy contributions due to sur-
explain the differing behaviour observed in the present face roughness and the shape of the sessile drop. Thus,
work with respect to the earlier ones. Finally, in order to it can differentiate between isotropic and anisotropic
check the surface chemical homogeneity/heterogeneity, wetting on a rough surface through CL characteris-
X-ray photoelectron spectroscopy (XPS) studies were tics along the two orthogonal directions of a corrugated
conducted on the surfaces. These measurements (Fig. surface. According to the model, the total free energy
S2 in SI) ascertained that the surface chemistries are change (ΔG) for a sinusoidal corrugated surface is given
almost identical for the surfaces. by
Figure 3 depicts the wetting contact angles on
nanostructured p-Si surfaces for various colloidal con-  
centrations of the droplet. For comparison, data for x
ΔG = −L γla cosθ + L γla (H0 − z )2 + x2 − H0
pure water (0% conc.) have also been furnished. It is cosα
observed that CA on the WES varies from 95◦ to 102◦ . (1)
For the IIS, the contact angles exhibit an increasing
value for both orientations in the range of 67◦ to 77◦ . Here, L is the length of the CL, γla and θ have their
In addition, the corrugated surface shows an anisotropic usual meanings, x is the horizontal projection of the
distance that the CL has moved across, α is the angle
defined by the corrugation (ripple) amplitude (d) and
wavelength (λ) (tan α = 2d/λ), H0 is the initial length
of the liquid front and z is the height of the three-
phase contact point from z = 0. The free energy of
adhesion is found to be negative for both the surfaces,
indicating the aggregation and adhesion of particles on
them. For IIS, ΔG/L is in the range −114 μJ/m and
−81 μJ/m in the parallel and perpendicular directions,
respectively, for smaller particle and lower volume frac-
tion as shown in Fig. 4. Thus, the parallel direction
indicates the least surface energy among the two con-
figurations. While in case of WES, the value of ΔG/L
is found to be −19 μJ/m which is larger than IIS. In
contrast, Zhao et al. demonstrated that the free energy
of a water droplet on micron-sized (both in terms of
depth and wavelength) grooves was greater in the par-
allel than in the perpendicular direction [32]. In their
Fig. 3 Measurement of the wetting contact angles of a case, the droplet CL length was of the order of μm
micro-liter sessile droplet on WES, IIS and pristine Si (p-Si) while for the present case it is of the order of mm. This
surface resulted in the second term on the rhs of Eq. 1 to be

123
113 Page 6 of 11 Eur. Phys. J. E (2023) 46 :113

15 and 8–10 min, respectively, for all combination sets

of nanoparticle sizes and concentrations.
During differential evaporation on IIS, the droplet
radius and angle decline with time and essentially
approach zero, indicating the switching from CCR
mode to CCA mode of evaporation. This transition is
determined by the CCR/CCA ratio which is calculated
with respect to the normalized time (T /T0 ). For all
concentrations, the CCR/CCA ratio was found to be
0.6 for parallel configuration and 0.5 for the perpen-
dicular direction of the corrugations as evaluated from
the data of Fig. 5b, c, respectively. The time span of
CCA modes is affected by the periodicity of corruga-
tions, which contributes to the receding of the droplet
CL more in a perpendicular direction. This receding is
Fig. 4 Total free energy change (ΔG) as a function of col-
driven by the adhesion between particle–substrate at
loidal concentration with different particle sizes for WES
CL and further influenced by the surface periodicity,
and IIS
roughness, and interfacial wetting states [44]. In gen-
eral, the transition between the modes of evaporation
is determined by the nature of the defects, with surfaces
positive with respect to the first one in the study by
having dense height fluctuations acting as hydrophobic
Zhao. In the present study, this turns out to be the
surfaces and surfaces with comparatively largely spaced
opposite, i.e. first term is greater than the second one.
protrusions acting as hydrophilic ones as found in our
experiments and discussed above (Sect. 3.2).
3.3 Evaporation characteristics Figure 6 depicts the variation in contact angle and
base radius (normalized) of the droplet of colloidal sus-
The evaporation dynamics of the colloidal sessile droplet pension on WES and IIS in parallel and perpendicu-
on WES and IIS for the particle size of 12 nm with lar directions for the larger particle size (22 nm). The
respect to change in contact angle and normalized overall distribution of modes of evaporation on both
droplet radius (R/R0 ) is shown in Fig. 5. For compar- surfaces with orientations was reported to be similar to
ison, data for pure water (0 vol%) are also plotted for the smaller particle size. For the determination of the
each case. Here, R0 is the initial radius and R is the CCR/CCA ratio, the parallel direction shows a ratio of
radius at a particular instant of time. All plots have 0.4 and the perpendicular direction that of 0.2 as calcu-
been made as a function of total normalized evaporation lated from Fig. 6b, c, respectively. With an increase in
period (T /T0 ). Here, T0 is the total time of evaporation the particle size, the receding time of the CL at CCA
and T is a particular instant of time during the evap- mode increases with a reduction in CCR mode dura-
oration process. The evaporation of the sessile droplet tion, as particles are loosely deposited next to each
on a substrate takes place in three stages, i.e. CCR other at the CL during evaporation. However, smaller
(constant contact radius) mode, CCA (constant contact size particles easily get pinned at the CL and can get
angle) mode and MM (Mixed) mode [24,39]. On WES, closely packed during evaporation with comparatively
the droplet predominantly evaporates in the CCR mode larger CCR mode. Therefore, particle size decides the
with a steady reduction in CA almost till the end where adhesion forces acting between particles and substrate
both angle and radius decrease during the short-lived as well as among particles inside the fluid during evap-
MM mode of evaporation as shown in Fig. 5a. Isotropic oration [42]. The particle deposition near the CL is the
roughness at the nanoscale contributes to intense pin- result of the competition between these forces exerted
ning of droplet at CL where most of the evaporation on each of the particles. During evaporation, as the
occurs in CCR mode thereby exhibiting a strong coffee- inward surface tension force overcomes all interfacial
ring effect. This is primarily due to the availability of forces, the depinning of the CL occurs faster for larger
numerous isotropically distributed height fluctuations particle sizes [45]. Thus, the coupled effect of topo-
in the form of pinning sites [39]. For IIS substrates, graphical periodicity, associated pinning sites and par-
droplet evaporation was studied in parallel and per- ticle size dominate the depinning of droplet boundary
pendicular directions of the corrugations as depicted in the perpendicular orientation.
in Fig. 5b, c, respectively. For both orientations, the The hydrophilic IIS shows a prominent signature of
droplet evaporates in CCR mode first, then in CCA anisotropic wettability for the sessile droplet which is
mode, and lastly in MM. The existence of 3 differ- driven by the periodicity of the asperities and can be
ent modes of evaporation for the case of IIS is simi- experimentally observed during evaporation. In order to
lar to that of a pristine Si surface reported earlier [39]. visualize the droplet deformation during evaporation,
However, the CCR mode observed for corrugated IIS the normalized droplet radii in the parallel and per-
is a short-lived one as compared to that of WES. The pendicular directions were plotted as a function of nor-
total evaporation times were observed to be different malized evaporation time as shown in Fig. 7. It is clear
for the surfaces. For WES and IIS, the times were 12– that, initially, the wetting area of the drop is almost cir-

123
Eur. Phys. J. E (2023) 46 :113 Page 7 of 11 113

Fig. 5 Variation in contact angle and radius with normalized time during evaporation of 12 nm particle size colloidal
droplet on a WES and IIS in b parallel and c perpendicular view direction. Data for pure water droplets have been added
for reference

Fig. 6 Variation in contact angle and radius with normalized time during evaporation of 22 nm particle size colloidal
droplet on a WES and IIS in b parallel and c perpendicular view direction. Data for pure water droplets have been added
for reference

123
113 Page 8 of 11 Eur. Phys. J. E (2023) 46 :113

Fig. 7 Temporal deformation profiles of droplets on IIS surface during evaporation from both directional view (i.e. parallel
and perpendicular) with corresponding concentrations and particle sizes. Color bar on the right indicates normalized time
of evaporation (T /T0 )

cular as observed, although perturbed by the presence 3.4 Deposition pattern

of corrugations. At this point, the radii for both direc-
tions remain in CCR mode. However, as observed from The role of evaporation dynamics on the final distribu-
Fig. 7, after a normalized time of T /T0 = 0.27 for the tion of silica nanoparticles was investigated using opti-
12 nm particles and T /T0 = 0.15 for the 22 nm particles cal microscopy. The final deposition patterns left when
the radius (R⊥ ) in the perpendicular direction changes the solvent gets entirely evaporated are collectively
to CCA mode while that in the parallel direction (R ) dependent on the interaction between the colloidal par-
still remains in the earlier mode, thereby resulting in a ticles and substrate. Figure 8 shows the observed micro-
shape change of the droplet from circular to elliptical scopic images of the deposits on WES and IIS. Fig-
one. Figure S3 of SI shows a representative plot of such ure 8a–c and d–f refers to WES for different particle
a temporal radii evolution for 0.2 vol% concentration sizes and concentrations, while Fig. 8g–i and j–l refers
and a particle size of 12 nm. The typical evolution of to those for IIS. The patterns deposited on WES clearly
the radii during the evaporation process shows a mono- exhibit the widest coffee-ring due to intense pinning
tonic decrease until the complete disappearance of the along the droplet CL whereas irregular patterns are
droplet. However, it is evident that the residual drop formed on IIS. The particle concentration in the droplet
becomes more oblate as evaporation proceeds, suggest- has a considerable impact on the final deposition pat-
ing that slippage of the contact line along the grooves terns. Particles travelling towards CL during evapora-
takes place more readily than across the corrugation tion contribute to ring thickness as particle concentra-
peaks, with particle size and concentration having an tion and its size, both increases. In the deposit patterns,
impact on the magnitude of the effect. Occurrence of two separate regions can be defined, the exterior region
anisotropic shapes guided by underneath corrugated referring to the area outside the deposits that resem-
or other geometrical topographies have been observed ble ring-like patterns and the interior region referring
but only for non-colloidal droplets earlier [22,46]. Such to the area inside the deposits in the annular region.
topographies give rise to stick–slip motion, which even- Figure 9a plots the scaled rim width (with respect
tually decides the shape of the sessile droplet during to the dried droplet radius) with the colloidal concen-
the evaporation event. The presence of nanoparticles tration. As a droplet dries while shrinking in size and
and nano-roughened surfaces, as reported in our study, the particles get deposited at the rim, the rim width
exhibits a monotonic change of droplet shape till the increases while the dried droplet radius decreases in
end of the evaporation cycle. size. Hence, a higher value on the y-axis of the plot
refers to deposition as well as shrinkage of a droplet. On

123
Eur. Phys. J. E (2023) 46 :113 Page 9 of 11 113

Fig. 8 Optical images of final deposition pattern of droplets of colloidal suspensions with 12 nm and 22 nm particle size
on a–f WES and g–l IIS for various particle concentrations

WES, particles are easily transported to the edge where cles at a concentration of 0.2 vol%, which increases to
they remain pinned, thereby leading to a relatively large 59.4 ± 7.6 μm as the concentration of particles of the
number of particles being located at the ring after evap- droplet increases to 0.5 vol%. Similar trend is observed
oration. With the enhancement of the coffee-ring effect for the higher particle size as well, which goes from
due to increasing size of the particle, the cross-sectional 80.1 ± 5.5 to 117.6 ± 10.6 μm with all concentrations.
area of the ring increases. Thus, the value of rim width On the other hand, after an initial period of slippage
to the radius of deposition remains low between 0.1 to and repinning of CL during evaporation, it becomes evi-
0.2 because the radius of all deposits remains almost dent that the patterns deposited on IIS exhibit charac-
same while the thickness of ring increases. For the teristics similar to those commonly associated with the
case of IIS, the deposits become smaller than in WES coffee ring effect. However, this substrate also exhibits
patterns due to depinning of CL, but the rim width crack spacing with large fluctuations due to irregular-
rises concurrently with particle concentration, which is ity in the shape of the deposit. The deposits obtained
also evident from Fig. 8. As a result, the scaled rim by higher concentration drying droplets delaminate
width is significantly enhanced from 0.25 to 0.6 with from the solid substrate due to large stress generated
an increase in particle concentration. Simultaneously between the layers of deposits. The deposit patterns
for both cases, this ratio degraded with the increment for 0.7 vol% concentration small particle droplets show
in particle dimension as rim width is directly propor- delamination phenomena for both substrates, although
tional to the size of particle [2]. the effect is more pronounced for IIS. In general, delam-
The accumulation of nanoparticles occurs with evap- ination occurs when a nanoparticle-laden particulate
oration of the droplet of colloidal suspension, resulting layer gets detached from the substrate [48], as indi-
in a generation of particulate film over the solid sub- cated in Fig. 8a, g. Thus, it shows that there is no
strates. During the last phases, cracks are formed owing dependency on periodicity in case of delamination. The
to development of capillary stress inside the films. For delamination behaviour of particulates on rough sub-
both cases, with increasing particle size the total num- strates reduces, with no trace of any separation of film
ber of cracks decreases while the distance between adja- composed of higher-sized nanoparticles.
cent cracks increases as evident from Fig. 9b. This is
a direct outcome of the capillary stress (σ) and par-
ticle size (Rp ) having an inverse relationship, given
by σ = 2γ/Rp , where γ is the surface tension [47]. 4 Conclusion
Therefore, the stress for small particles becomes sig-
nificantly larger leading to an enhancement in crack To summarize, we experimentally study and compare
density. The separation between cracks was evaluated the evaporation characteristics of nano-colloidal ses-
as 34.8 ± 5.7 μm on WES for the case of 12 nm parti- sile droplets on self-affine vis-à-vis corrugated (period-

123
113 Page 10 of 11 Eur. Phys. J. E (2023) 46 :113

Fig. 9 a Ratio of rim width with the radius of deposition and b Crack spacing with colloidal concentration plotted for
deposition pattern observed on the rough surfaces for different particle sizes

icity ∼ 40 nm), yet identically rough substrates hav- process. Finally, delamination was found to occur for
ing an rms roughness of ∼ 3.5 nm. For an ideally smaller particles and larger concentrations on both sur-
smooth solid surface, the isotropic wetting behaviour faces, but was more noticeable for the periodic one. Our
is decided by the interfacial free energies [14] whereas findings impart insight into drying of sessile droplets
for a rough substrate, its interaction with the liquid of colloidal suspensions on nanorough anisotropic sur-
gains dominance [15]. Thus, the study of evaporation faces compared to isotropic ones. The results obtained
dynamics on identical rough surfaces with large differ- can serve as useful guidelines for designing other func-
ence in (an)isotropicity becomes imperative. The sur- tional surfaces for applications in areas of condensa-
faces, WES and IIS, are fabricated using chemical and tion [49], preferential fluid mass transfer [50], forced
ion-beam routes, respectively. The wettability, evapo- droplet motion [51], etc. Thus, our insights can cater to
ration kinetics and deposition patterns were investi- fabrication of surfaces that manipulate droplets, par-
gated in conjunction with textured surfaces and par- ticularly by utilizing micro/nano channels during the
ticles of varying size and concentration. The study drying process thereby advancing microfluidic areas of
unveils the enhancement in hydrophobicity on WES, application.
whereas IIS exhibits anisotropic wettability in paral-
lel and perpendicular directions to the surface period- Supplementary information The online version con-
icities. This finding differs from previous studies con- tains supplementary material available at https://doi.org/
ducted on anisotropic micro-structured surfaces, which 10.1140/epje/s10189-023-00374-8.
primarily focused on non-colloidal droplets [28,30,32].
The present results have been explained in the light of Acknowledgements The authors thank DST-SERB India
pinning by the corrugation crests for the anisotropic (Grant No. CRG/2018/001258) for partial financial support
surface. related to this work. We also thank the Central Research
The droplet was found to evaporate in CCR-MM Facility, IIT Ropar for AFM, FESEM and XPS measure-
modes on WES, whereas CCR-CCA-MM modes were ments. Deeksha Rani also would like to thank the Univer-
seen on IIS, with CCA being the predominant dry- sity Grants Commission (UGC), Government of India, for
financial support through a research fellowship.
ing mode. Evaporation studies carried out in orthog-
onal directions reveal that the ratio of the CCR to Data availability The datasets generated during and/or
CCA mode evaporation regimes was greater in parallel analysed during the current study are available from the
directions to the corrugations in IIS. With an increase corresponding author on reasonable request.
in particle size, this ratio reduces due to depinning
which occurs more rapidly at the CL. The presence
of nanoparticles and nano-roughened periodic surfaces
exhibited a monotonic change of droplet shape till the References
end of the evaporation cycle. Thus, IIS is responsible
for forming distorted ring arrangements, while WES 1. T. Young, Philos. Trans. R. Soc. Lond. 95, 65–87 (1805)
leads to the generation of coffee-ring patterns. The 2. R.D. Deegan, Phys. Rev. E - Stat. Phys. Plasmas Fluids
study of dried droplet pattern shows that the rim width Related Interdiscip. Topics 61, 475–485 (2000)
and crack spacing get more pronounced with particle 3. P. Calvert, Chem. Mater. 13, 3299–3305 (2001)
concentration and size for both surfaces. The forma- 4. T.-S.S. Wong, T.-H.H. Chen, X. Shen, C.-M.M. Ho,
tion of cracks and their spacing were discussed as a Anal. Chem. 83, 1871–1873 (2011)
consequence of capillary stress during the evaporation 5. W. Han, B. Li, Z. Lin, ACS Nano 7, 6079–6085 (2013)

123
Eur. Phys. J. E (2023) 46 :113 Page 11 of 11 113

6. J.T. Wen, C.-M. Ho, P.B. Lillehoj, Langmuir 29, 8440– 33. S.G. Lee, H.S. Lim, D.Y. Lee, D. Kwak, K. Cho, Adv.
8446 (2013) Funct. Mater. 23, 547–553 (2013)
7. W. Zhou, A. Hu, S. Bai, Y. Ma, Q. Su, Nanoscale Res. 34. H. Robbins, B. Schwartz, J. Jpn. Soc. Precis. Eng. 51,
Lett. 9, 1–9 (2014) 1013–1018 (1960)
8. H. Hu, R.G. Larson, J. Phys. Chem. B 106, 1334–1344 35. J. Muñoz-Garcı́a, L. Vázquez, M. Castro, R. Gago,
(2002) A. Redondo-Cubero, A. Moreno- Barrado, R. Cuerno,
9. H. Hu, R.G. Larson, J. Phys. Chem. B 110, 7090–7094 Mater. Sci. Eng. R. Rep. 86, 1–44 (2014)
(2006) 36. Rakhi, S. Sarkar, Phys. Rev. B 106, 245420 (2022)
10. R. Gimenez, G.J. Soler-Illia, C.L.A. Berli, M.G. Bellino, 37. R.M. Bradley, J.M.E.J. Harper, Vac. Sci. Technol. A 6,
ACS Nano 14, 2702–2708 (2020) 2390–2395 (1988)
11. H.P. Jansen, K. Sotthewes, J. van Swigchem, H.J.W. 38. W.L. Chan, E. Chason, J. Appl. Phys. 101, 1 (2007)
Zandvliet, E.S. Kooij, Phys. Rev. E 88, 13008 (2013) 39. T. Pham, S. Kumar, Langmuir 33, 10061–10076 (2017)
12. X. Dai, N. Sun, S.O. Nielsen, B.B. Stogin, J. Wang, S. 40. Y. Chen, B. He, J. Lee, N.A. Patankar, J. Colloid Inter-
Yang, T.S. Wong, Sci. Adv. 4 (2018) face Sci. 281, 458–464 (2005)
13. D. Zang, S. Tarafdar, Y.Y. Tarasevich, M.D. Choud- 41. Scientific, K. ADVANCE-software @ www.kruss-
hury, T. Dutta, Phys. Rep. 804, 1–56 (2019) scientific.com. https://www.kruss-scientific.com/en/
14. H.-J. Butt, G. Karlheinz, M. Lappl, Contact Angle Phe- products-services/advance-software#
nomena and Wetting; pp. 118–144 (2003) 42. C.Q. LaMarche, S. Leadley, P. Liu, K.M. Kellogg, C.M.
15. D. Kim, N.M. Pugno, S. Ryu, Sci. Rep. 6, 37813 (2016) Hrenya, Chem. Eng. Sci. 158, 140–153 (2017)
16. A.-L. Barabási, H. E. Stanley, Fractal Concepts in Sur- 43. M. Radiom, C. Yang, W.K. Chan, Nanoscale Res. Lett.
face Growth. Cambridge University Press (1995) 8, 1–9 (2013)
17. P.K. Dhillon, S. Sarkar, Appl. Surf. Sci. 284, 569–574 44. W. Xu, C.-H. Choi, Phys. Rev. Lett. 109, 24504 (2012)
(2013) 45. B.M. Weon, J.H. Je, Phys. Rev. E - Stat. Nonlinear Soft
18. M. Hosseini, A. Rodriguez, W.A. Ducker, J. Colloid Matter Phys. Physical Review E - Statistical, Nonlinear,
Interface Sci. 633, 132–141 (2023) and Soft Matter Physics 82, 1–4 (2010)
19. W. Xu, C.H. Choi, J. Heat Transf. 134, 051022 (2012) 46. R. Raj, S. Adera, R. Enright, E.N. Wang, Nat. Com-
20. D. Lohani, S. Sarkar, Langmuir 34, 12751–12758 (2018) mun. 5, 4975 (2014)
21. C. Seyfert, E.J. Berenschot, N.R. Tas, A. Susarrey-Arce, 47. Y.O. Popov, Phys. Rev. E - Stat. Nonlinear Soft Matter
A. Marin, Soft Matter 17, 506–515 (2021) Phys. Physical Review E - Statistical, Nonlinear, and
22. G.G. Wells, É. Ruiz-Gutiérrez, Y. L. Lirzin, A. Nourry, Soft Matter Physics 71, 1–17 (2005)
B.V. Orme, M. Pradas, R. Ledesma-Aguilar, Nat. Com- 48. F. Giorgiutti-Dauphiné, L. Pauchard, Soft Matter 11,
mun. 9, 1–7 (2018) 1397–1402 (2015)
23. J. Zhang, F. Müller-Plathe, F. Leroy, Langmuir 31, 49. K.-C. Park, P. Kim, A. Grinthal, N. He, D. Fox, J.C.
7544–7552 (2015) Weaver, J. Aizenberg, Nature 531, 78–82 (2016)
24. X. Chen, R. Ma, J. Li, C. Hao, W. Guo, B.L. Luk, S.C. 50. M. Pradas, N. Savva, J.B. Benziger, I.G. Kevrekidis, S.
Li, S. Yao, Z. Wang, Phys. Rev. Lett. 109, 1–6 (2012) Kalliadasis, Langmuir 32, 4736–4745 (2016)
25. V. Garg, L. Qiao, P. Sarwate, C. Luo, Langmuir 30, 51. D. Herde, U. Thiele, S. Herminghaus, M. Brinkmann,
14469–14475 (2014) EPL (Europhysics Letters) 100, 16002 (2012)
26. B. Hou, C. Wu, X. Li, J. Huang, M. Chen, Appl. Surf.
Sci. 542, 148611 (2021) Springer Nature or its licensor (e.g. a society or other part-
27. D. Xia, L.M. Johnson, G.P. Lõpez, Adv. Mater. 24, ner) holds exclusive rights to this article under a publish-
1287–1302 (2012) ing agreement with the author(s) or other rightsholder(s);
28. F. Zhang, H.Y. Low, Langmuir 23, 7793–7798 (2007) author self-archiving of the accepted manuscript version of
29. A.D. Sommers, A.M. Jacobi, J. Micromech. Microeng. this article is solely governed by the terms of such publishing
16, 1571–1578, Intro application oriented (2006) agreement and applicable law.
30. D. Xia, S.R. Brueck, Nano Lett. 8, 2819–2824 (2008)
31. O. Bliznyuk, E. Vereshchagina, E.S. Kooij, B. Poelsema,
Phys. Rev. E - Stat. Nonlinear Soft Matter Phys. 79
(2009)
32. Y. Zhao, Q. Lu, M. Li, X. Li, Langmuir 23, 6212–6217
(2007)

123