Access to this full-text is provided by The Royal Society.
Content available from Interface Focus
This content is subject to copyright.
royalsocietypublishing.org/journal/rsfs
Research
Cite this article: Kosmidis-Papadimitriou A,
Qi S, Squillace O, Rosik N, Bale M, Fryer PJ,
Zhang ZJ. 2021 Characteristics of respiratory
microdroplet nuclei on common substrates.
Interface Focus 12: 20210044.
https://doi.org/10.1098/rsfs.2021.0044
Received: 20 May 2021
Accepted: 2 November 2021
One contribution of 7 to a theme issue
‘Coronavirus and surfaces’.
Subject Areas:
environmental science, chemical physics,
biophysics
Keywords:
respiratory, microdroplet, nuclei, substrates,
transmission, virus
Author for correspondence:
Zhenyu J. Zhang
e-mail: z.j.zhang@bham.ac.uk
Characteristics of respiratory microdroplet
nuclei on common substrates
Alexandros Kosmidis-Papadimitriou
1
, Shaojun Qi
1
, Ophelie Squillace
1
,
Nicole Rosik
1
, Mark Bale
2
, Peter J. Fryer
1
and Zhenyu J. Zhang
1
1
School of Chemical Engineering, University of Birmingham, Edgbaston, Birmingham B15 2TT, UK
2
DoDxAct Ltd, Wells BA5 2LW, UK
AK-P, 0000-0002-1654-4630; OS, 0000-0003-3840-5065; PJF, 0000-0003-4767-7839;
ZJZ, 0000-0003-0243-2109
To evaluate the role of common substrates in the transmission of respiratory
viruses, in particular SARS-CoV-2, uniformly distributed microdroplets
(approx. 10 µm diameter) of artificial saliva were generated using an
advanced inkjet printing technology to replicate the aerosol droplets and
subsequently deposited on five substrates, including glass, polytetrafluor-
oethylene, stainless steel, acrylonitrile butadiene styrene and melamine.
The droplets were found to evaporate within a short timeframe (less than
3 s), which is consistent with previous reports concerning the drying kinetics
of picolitre droplets. Using fluorescence microscopy and atomic force
microscopy, we found that the surface deposited microdroplet nuclei present
two distinctive morphological features as the result of their drying mode,
which is controlled by both interfacial energy and surface roughness. Nano-
mechanical measurements confirm that the nuclei deposited on all substrates
possess similar surface adhesion (approx. 20 nN) and Young’s modulus
(approx. 4 MPa), supporting the proposed core–shell structure of the
nuclei. We suggest that appropriate antiviral surface strategies, e.g. functio-
nalization, chemical deposition, could be developed to modulate the
evaporation process of microdroplet nuclei and subsequently mitigate the
possible surface viability and transmissibility of respiratory virus.
1. Introduction
Transmission of respiratory viruses can take place in different modes, either
directly via contact between individuals, indirectly via commonly touched objects
or surfaces, or directly through the air in the form of large droplets or small aero-
sols [1]. Surface transmission, in particularly via surface fomite, was viewed as one
of the primary concerns since the initial stages of the pandemic caused by severe
acute respiratory syndrome coronavirus-2 (SARS-CoV-2) in 2020 [2]. One of the
early studies suggested that SARS-CoV-2 remains viable in aerosols for at least
3 h, and that SARS-CoV-2 is more stable on plastic and stainless steel (SS) than
on copper and cardboard [3], which highlighted the unique role of common sur-
faces in virus transmission during the pandemic. The impact of substrate on the
surface viability of virus was further demonstrated with a large range of Middle
East respiratory syndrome coronavirus (MERS-CoV) [4].
There have been compelling arguments that the transmission of SARS-CoV-2
after touching surfaces should be considered as relatively minimal [5], given that
it is improbable that an infected person coughs or sneezes on a surface (with suf-
ficient quantity of infectious virus), and someone else touches that surface shortly
after (within 1–2 h) [6]. This rationale is sound and sensible, on the assumption
that surface transmission takes place via a large quantity of respiratory fluid
and that the virus would be inactivated beyond the timeframe suggested. An
extensive list of evidence, including superspreading events, long-range trans-
mission, asymptomatic transmission, was given to support that the dominating
© 2021 The Authors. Published by the Royal Society under the terms of the Creative Commons Attribution
License http://creativecommons.org/licenses/by/4.0/, which permits unrestricted use, provided the original
author and source are credited.
transmission route of SARS-CoV-2 is airborne [7,8]. Although
the likelihood of surface transmission is no longer as signifi-
cant as perceived at the initial stage of the pandemic,
the possible presence of infectious virus on solid substrates,
in particular on high-touch environmental surfaces,
could have significant implications for both social and
healthcare practice.
There remain significant knowledge gaps in drawing con-
clusions on the possible role of surface in preserving and
transmitting SARS-CoV-2 due to the multitude of complexities
involved. This is highlighted in a systematic review by Onak-
poya and colleagues focusing on the role of fomite
transmission over 64 studies [9], which concluded that no evi-
dence is available to confirm viral infectivity or transmissibility
via fomites, but that none of the studies surveyed is sufficiently
methodologically robust to adequately address the question.
A similar reflection questioning the unlikelihood of indirect
transmission through contaminated surfaces has been reported
recently [10]. A core element that underpins the inconsistent
viewpoints on the transmission pathways of SARS-CoV-2 is
the physico-chemical properties of the microdroplets. Other
than the discrepancy over the definition of aerosols, droplets,
particles and droplet nuclei perceived by researchers from
different disciplines [11], diameters of the exhaled droplets
can cover a broad range, from 0.1 to 1000 µm [12,13], for
which the fluid mechanics and evaporation kinetics vary sub-
stantially. Literature suggests that sneezing may generate
droplets between 0.5 and 12 µm in diameter whereas breath-
ing and speaking may result in droplets with at an averaged
diameter of 1 µm [14–17].
It is therefore crucial to further understand the character-
istics of aerosol droplets once they are deposited on common
surfaces. Evaporation kinetics of microlitre and picolitre dro-
plets of pure liquid, e.g. water or organic solvents, have been
extensively studied in the past [18,19]. However, the complex
compositional nature of saliva, consisting of salts, proteins
and lipids could influence not only the evaporation process
of the microdroplet, but also the location of the viruses in
the final nuclei, and consequently their viability and trans-
missibility on solid substrates [20,21]. Previous studies in
which model respiratory liquids were used reported that
the generated microdroplets would crystallize on the sub-
strate as the result of their evaporation process [22,23]. It is
unclear whether the surface crystallization, as part of the
drying process, could inactivate the viruses because they
are enveloped in the crystals, or the crystals actually preserve
the viruses and prolong their viability upon rehydration. Fur-
thermore, characteristics of the microdroplet nuclei could
have a direct impact on the effectiveness of sample collection
protocols in that the dissolution kinetics of the crystalline
phase is much slower than the amorphous phase [24]. Finally,
adhesion of microdroplet nuclei to the underlying substrate
could be critical to contact transmission. Literature suggests
that the molecules contained in human saliva, such as sali-
vary agglutinin [25], salivary proteins MUC5B, cysteine-rich
glycoprotein 340 [26] and secretory leucocyte protease inhibi-
tor [27], could introduce inhibitory and antiviral effects.
However, artificial saliva that consisted of mucin and inor-
ganic electrolytes was used in the present work due to
health and safety restrictions.
In the present work, we investigated systematically the
morphological and nanomechanical properties of micro-
droplet nuclei and their formation kinetics as a function of
the surface in contact. Using an advanced inkjet printing
set-up, we were able to generate picolitre droplets and sub-
sequently deposit them onto five inanimate materials that
are commonly found in daily life, replicating the surface
transmission route of viruses. We found that surface charac-
teristics, including both surface energy and roughness,
could influence the evaporation process of microdroplets,
the structure of the resulting nuclei, and consequently the
viability and transmissibility of virus.
2. Methodology
2.1. Materials
Phosphate-buffered saline, mucin extracted from porcine stomach
(M1778), magnesium chloride, calcium chloride, ammonium
chloride and Dulbecco’s modified Eagle’smedium(DMEM)were
purchased from Sigma-Aldrich (Somerset, UK).
2.2. Artificial saliva
Artificial saliva, an aqueous mixture of salts, nutrients and
mucin, was prepared to replicate human saliva [28]. The addition
of inorganic ions in the mixture offers osmotic balance and
buffering especially during the injection process [28–30]. The arti-
ficial saliva composition used in the present work is shown in
table 1 [17,31–33].
2.3. Picolitre droplet generation
A customized Jetxpert Print Station (Imagexpert, NH) was used
to generate arrays of picolitre droplets of the artificial saliva that
were deposited onto various solid substrates. The print station
uses a conveyor belt instead of a single linear stage to improve
processing efficiency. The print head (GH2220, Ricoh, UK) was
thoroughly cleaned with distilled water, followed by the pre-
pared artificial saliva, prior to the droplet generation. To
prevent any potential clogging of the print head by the drying
saliva solution, an automated printing cycle that ejects 200 dro-
plets every 5 s was programmed to keep the nozzles wet and
prevent the accumulation of mucin.
To generate the desired droplet size, the print head driving
signals were optimized in terms of voltage amplitude and
pulse timing. The Jetxpert print station was equipped with a
strobing camera and the relevant software to measure the droplet
size. Once droplets with a diameter of approximately 10 µm were
produced consistently, the print head was placed over the con-
veyor stage to generate arrays of droplets that were
approximately 175 µm apart by printing with a resolution of
300 × 300 dpi. A standard USB camera (Hayear 1136) was
mounted above the printing stage to record videos (30 fps) of
the printed artificial saliva droplets and to monitor the averaged
drying time on each substrate. Five solid substrates of distinctive
Table 1. Composition of artificial saliva per litre of aqueous solution in
deionized water.
compound quantity compound quantity
MgCl
2
·7H
2
O 0.04 g Na
2
HPO
4
0.42 g
CaCl
2
·H
2
O 0.13 g (NH
2
)
2
CO 0.12 g
NH
4
Cl 0.11 g mucin 1.00 g
KCl 1.04 g KH
2
PO
4
0.21 g
DMEM 1 ml NaCl 0.88 g
royalsocietypublishing.org/journal/rsfs Interface Focus 12: 20210044
2
characteristics were used to represent the common surfaces: bor-
osilicate glass slides to represent glass and ceramics,
polytetrafluoroethylene (PTFE) to represent non-stick coatings,
SS that is commonly used for door handles and hand rails, acry-
lonitrile butadiene styrene (ABS) that has a wide application in
household and consumer goods, and melamine that is used for
dinnerware and laminate flooring.
2.4. Fluorescence microscopy
Fluorescence microscopy was carried out to identify the residues
of the deposited microdroplets. Alexa Fluor 488 Maleimide of
excitation/emission wavelength 493/516 nm was used for fluor-
escence imaging purposes. For one dedicated set of experiments,
the fluorophore was dissolved in dimethyl sulfoxide at a concen-
tration of 1 mM, of which 10 µl were introduced to 10 ml of
artificial saliva solution prior to the printing process. Sample
images were captured after 5 days of drying, using an inverted
fluorescence microscope (Olympus IX71).
2.5. Contact angle goniometer
An optical tensiometer (Theta Flow, Biolin Scientific, UK) was
used to measure both advancing and receding contact angles
(CAs) of artificial saliva on all five solid substrates, for which
droplet of 4 µl was produced by the associated micro-syringe.
To quantify the surface free energy (SFE) of these substrates,
2 µl diiodomethane (apolar) or water (polar) were placed on the
substrates, and equilibrium CAs were recorded subsequently.
Owens/Wendt theory describes the surface energy of a solid as
having two components, a dispersive component and a non-
dispersive (or polar) component. Mathematically, the theory is
based on the combination of two fundamental equations
(Good’s equation and Young’s equation) that describe inter-
actions between solid surfaces and liquids [34,35]. This
equation has a linear form y=mx+b, wherein:
y¼
s
L:(cos
u
þ1)
2:
s
D
L
1=2;x¼
s
P
L
1=2
s
D
L
1=2;m¼
s
P1=2
S;b¼
s
D1=2
S:ð2:1Þ
Diiodomethane has a relatively high overall surface tension of
50.8 mN m
−1
but no polar component, so that
s
L¼
s
D
L¼
50:8mNm
1, while water has a surface tension sP
L¼46:4mNm
1
for the polar component and
s
D
L¼26:4mNm
1for the disperse
component. The slope mof that line is used to calculate the
polar component of the surface energy of the solid
s
P
Sand
the intercept bis used to calculate the dispersive component of the
surface energy of the solid
s
D
S, the overall SFE of the solid being
defined as SFE ¼
s
D
Sþ
s
P
S:
2.6. Atomic force microscopy
Surface topography and characteristics of the dried droplets were
examined using an atomic force microscope (Dimension 3100,
Bruker). All imaging and force curve requisition was performed
with silicon nitride cantilevers (PNP-TR-Au, spring constant 0.08
Nm
−1
,ApexProbesLtd,UK).Allcantileversweresubjecttothe
same cleaning routine (rinsing with ethanol followed by exposure
to UV-ozone for 15 min) before the first use and between every
two samples. Force measurements on the dried artificial saliva
nuclei were acquired with a maximum loading force of 5 nN. At
least 100 repetitions, in the form of a 7 × 7 matrix (200 nm apart
in both xand ydirections), were acquired at 2–3randomlocations
within the area of interest. Young’s moduli of the dried products
were estimated by fitting the approaching component of the force
curves with Sneddon model (conical indenters):
F¼E
1v22tan
a
p
d
2,ð2:2Þ
where Eand νare the Young’s modulus and the Poisson’sratio
(0.5 for organic compound) of the materials being indented, res-
pectively; αis the semi-opening angle of the atomic force
microscope (AFM) tip which is 35° in the present study; δis the
indentation depth which is calculated by subtracting the cantilever
deflection from the measured sample height.
3. Results and discussion
3.1. Microdroplet generation
The Ricoh GH2220 print head deployed in the present work
was chosen due to its highly adaptable performance with
different types of fluid. Although the viscosity of the artificial
saliva was well below the ideal operating point for the print
head, there is a wide latitude to modify the print head driv-
ing signals. To measure and adjust droplet size, the camera of
the Jetxpert drop watcher is triggered in synchronization with
the GH2220 drive electronics (Meteor Inkjet Ltd, UK). This is
a standard method to capture the flight trajectory of individ-
ual droplets as they are ejected from the inkjet head. The goal
was to find the optimal pulse shape and voltage amplitude to
actuate the piezoelectric elements of the print head, so as to
eliminate satellite droplets present and reach the desired dro-
plet size. Figure 1 presents a series of images captured,
showing the trajectory of a single artificial saliva droplet of
approximately 5 pl volume being ejected from the print
head. We were able to capture the entire process, from the
moment when the droplet was detached from the nozzle
(far left) and the subsequent movement in the air without
any satellite drops.
The two rows of nozzles of the GH2220 enabled printing
in a single pass with a pixel addressability of 300 dpi, which
could produce droplets as close together as 85 µm. To mini-
mize the possibility of any uncontrolled merging of
adjacent drops across the range of surfaces studied, a print
image using an ordered dither of approximately 23% cover-
age was implemented, resulting an average droplet
separation of approximately 175 µm. With these settings,
and belt speed of just 5 m min
−1
, the print frequency was
kept low (less than 1 kHz), thus minimizing the chance of
nozzle failure due to the sub-optimal rheology. A camera
was positioned next to the solid substrate to capture the
drying of the deposited droplets from the point the conveyor
came to rest, approximately 1 s after printing.
The ability to consistently and rigorously generate micro-
droplets of uniform size provides a substantial advantage
over the other techniques in generating aerosol droplets
such as atomizer and nebulizer. It offers the assurance to
compare droplet nuclei formed on different substrates, with
minimal variation in the droplet characteristics.
Figure 1. Mosaic of images captured from 37 separate ejection events by the
Jetxpert camera to show the flight trajectory of a single artificial saliva dro-
plet from the moment of ejection (far left). Diameter of the droplet is 10 µm.
royalsocietypublishing.org/journal/rsfs Interface Focus 12: 20210044
3
3.2. Evaporation of the deposited droplet
Droplets of 10 µm diameter (volume is approximately 4.8 pl)
were deposited on five different surfaces, namely: glass,
PTFE, SS, ABS and melamine. The time it took for the droplet
to evaporate on the substrates, upon deposition, was esti-
mated using the videos captured and is presented (figure 2)
as a function of the corresponding SFE established based
on deionized water and diiodomethane. The drying time esti-
mated appears to correlate with the SFE of the substrates: the
greater the SFE is, the longer it takes for the droplet to
evaporate.
Evaporation kinetics, including time, of microlitre liquid
droplets on solid substrate have been studied extensively at
millimetre scale [18,19]. It is widely accepted that there are
two limiting drying scenarios: (i) the contact line of the
liquid/solid interface remains constant throughout the
drying while CA decreases, called constant contact radius
(CCR) mode and (ii) droplet radius reduces while the CA
remains the same, called constant contact angle (CCA)
mode, which is often observed on hydrophobic substrates
[36]. It is, however, worth noting that variations in both the
physical and chemical characteristics of the substrate could
help to enhance contact line pinning, although strongly
hydrophobic substrates could be exceptions [37].
For droplets of picolitre volume, Talbot et al. [38] investi-
gated the evaporation kinetics of water and ethanol droplets
on surfaces with different wettability and thermal conduc-
tivity. The droplet volume studied ranges from 4 to 65 pl,
and the drying time for water droplets was around 4 s for
the surfaces used. This is in a very similar time frame to
that observed in the present work, which is not surprising
because the artificial saliva used in our work consists of
97% water. It was not possible to measure the CA of the pico-
litre droplet in the present investigation, but previous studies
suggest that it would be several degrees less than that
measured using microlitre droplet [38]. Although drying
and evaporation kinetics of picolitre droplets are beyond
the scope of this work, the observed correlation between
drying time and SFE does not seem to agree with the pre-
vious finding that evaporation on hydrophilic substrates is
faster than on hydrophobic substrates. This inconsistency
could be attributed to the substrates used, or the surface
adsorption of mucin molecules upon deposition. We high-
light that the variations in the actual drying time of
artificial saliva microdroplets of 10 μm are within 2 s
(figure 2), which indicates that transfer of aerosol droplet in
liquid state is unlikely, and suggests that the effect of surface
hydrophobicity on the drying time of microdroplets (less
than 10 µm) is not a significant factor in the context of surface
transmission. The short time (a few seconds) for these biologi-
cal picolitre droplets to reach full dryness, as opposed to the
length scale (minutes to tens of minutes) it takes for micro- to
macro-sized droplets, underlines the importance of focusing
on the physico-chemical and virological characteristic of the
dried nuclei of such, rather than solely on fresh respiratory
deposition in terms of surface transmission.
3.3. Advancing and receding contact angle
measurements
Advancing and receding CAs of artificial saliva (2 µl ) were
measured on glass, SS, PTFE, ABS and melamine, as pre-
sented in figure 3. Of the five substrates investigated, glass
shows the smallest advancing CA (42.0°) and the second
smallest receding CA (15.2°) of the range, while PTFE results
in the greatest ones (115.2° and 29.5° respectively), which cor-
relates with their SFE values: 67.5 mN m
−1
and 27.3 mN m
−1
for glass and PTFE, respectively. SS, ABS and melamine pre-
sent intermediate values, in terms of SFE and CA of artificial
saliva, showing a similar correlation between advancing and
receding CA values. The advancing CA on SS is noticeably
less than that on ABS or melamine.
The differences in SFE, advancing and receding CAs
between the five substrates surveyed are likely due to the syner-
gistic effect of surface characteristics, e.g. chemical nature,
topography and roughness [39,40]. Glass substrate shows the
lowest roughness (R
a
), 1.1 ± 0.1 nm (obtained by AFM images
over an area of 50 × 50 µm), followed by ABS and melamine
with 3.8 ± 0.7 nm and 3.7 ± 0.8 nm respectively, while PTFE
and SS present great roughness with R
a
=19.4±3.3nm and
33.6 ± 6.1 nm, respectively. Although SS presents a high surface
roughness, both advancing and receding CAs on SS are less than
those on ABS and melamine, which is likely attributed to the
hydrophilic nature of SS, while ABS and melamine are less
polar. This supports our observation of artificial saliva CAs on
the solid substrates and is consistent with our previous exper-
imental work in which SS samples of different surface
3.0
2.5
2.0
1.5
1.0
surface free ener
gy
(mN m–1)
10 20 30 40 50 60 70
evaporation time (s)
glass
PTFE
SS
ABS
melamine
Figure 2. Evaporation time of artificial saliva microdroplet on five different
substrates as a function of their corresponding SFE that was calculated using
diiodomethane and deionized water. Error bars are of similar size to the
dimension of the symbols used, and hence not shown.
110
90
70
50
30
recedin
g
contact an
g
le (°)
10 20 30 40 50
advancing contact angle (°)
glass
PTFE
SS
ABS
melamine
Figure 3. Advancing CA of artificial saliva on the five substrates as a function
of the corresponding receding CA.
royalsocietypublishing.org/journal/rsfs Interface Focus 12: 20210044
4
finishing were investigated [41]. It is worth noting that substrates
such as SS, ABS and melamine, whether smooth or rough, can
present high advancing CA but low receding CA, suggesting
that it is possible for those less polar substrates to have an
increased interaction with the artificial saliva once the surface
is already wet, while it would not have been the case when
dry. At the opposite, PTFE only presents high advancing and
high receding CA. The surface characteristics discussed at the
macrodroplet scale, including both polarity and roughness, are
still important consideration at the microdroplet scale.
3.4. Droplet nuclei characterization
Morphology of the microdroplet nuclei with microscopic and
nanoscopic spatial resolution was subsequently investigated
using fluorescence microscopy and atomic force microscopy,
respectively. Artificial saliva mixture containing Alexa Fluor
488 Maleimide was deposited on a set of substrates using
the same printing conditions and imaged by a fluorescence
microscope afterwards. The acquired images are shown in
figure 4 where the array of droplet nuclei is distinctively
identifiable on some of the substrates, in particular on
the glass substrate (dark dots presented in figure 4a,b). The
series of solid particulates confirm the robustness of
the method in preparing the microdroplet nuclei. They (the
bright dots) could be seen on PTFE (figure 4c) and SS
(figure 4d), but are no longer noticeable on substrates such
as ABS (figure 4e) and melamine ( figure 4f), which is likely
due to the signal fluctuation from the background.
To establish fine details of the microdroplet nuclei, AFM
was deployed to survey different regions of the solid sub-
strates, of which representative images are shown in figure 5.
Two distinctive morphological features were observed: large
crystals surrounded by small solid residues (islands),
as shown on glass, ABS and melamine (figure 5b,h,j), or
large crystals without any noticeable neighbouring residue
(figure 5d,f ). As a result of the evaporation process of artificial
saliva, such notable variation could be solely attributed to the
characteristics of the solid substrate.
As explained in a recent study by Lieber et al. [42], respirat-
ory fluid such as saliva consists of a range of inorganic salts
and proteins, with a major fraction of water that will evaporate
under typical ambient conditions. They used an acoustic
levitator to study the temporal evolution of saliva droplets
that underwent evaporation in the air and reported that the
ratio between the equilibrium and initial diameter of a droplet
is 20%, assuming an initial combined mass concentration of
salts and proteins of 0.8%. Although this ratio is unlikely to
be applicable for the droplets deposited on a substrate, their
study highlights the significant difference between the evapor-
ation characteristics of water and saliva, and suggests that both
precipitation and crystallization could take place during the
evaporation of saliva droplets.
Concerning the drying process of a surface deposited
macroscopic droplet, a capillary flow induced by the evapor-
ation of water molecules at the air/liquid interface generates a
convective mass transfer phenomenon and liquid at the ridge
is replenished by liquid from the interior [43]. The convective
flow could carry the dispersed solutes to the solid/liquid con-
tact line, resulting in a circle of solid deposit, known as the
coffee ring effect [43]. This flow would be countered by Maran-
goni flowsthat redistribute the particles back to the centre of the
droplet [44]. In the present work, no coffee ring effect was
observed on any of the substrates, suggesting that either the
evaporation time was not sufficient to drive the solutes, such
as proteins, to the edges of the microdroplet, or the Marangoni
flows were strong enough to counter the convective mass trans-
fer. Instead, solid particles with crystallization features were
seen on all substrates surveyed, which is consistent with the
results reported by Vejerano & Marr [22] whostudied the trans-
formation of a mixture of mucin, salt and surfactants, 1,2-dihex-
adecanoyl-sn-glycero-3-phosphocholine (DPPC), as a function
of relative humidity. Using fluorescence labelled mucin and
DPPC, they were able to demonstrate that the droplet exhibited
an initial core–shell structure, with a great concentration of
mucin at the shell (air/liquid interface). Considering the fast
evaporation kinetics observed in the present work, it is very
probable that mucin molecules were kept at the shell while
the inorganic salts underwent the crystallization process in the
centre as the water molecules evaporated.
The dimension and morphology of these droplet residues
exhibit explicitly a compliance with the corresponding SFEs of
the substrates on which the picolitre droplets landed and evap-
orated. AFM images in figure 5 suggest that, upon the initial
contact with the solid substrate, the proteinaceous microdroplets
had a maximal contact area on substrates with high SFE, but
500 mm500 mm500 mm
500 mm500 mm500 mm
(c)(b)(a)
(f)(e)(d)
Figure 4. (a) Bright field and (b) fluorescence images of artificial saliva microdroplet nuclei on glass. Fluorescence images of droplet nuclei on (c) PTFE, (d) SS,
(e) ABS and (f) melamine upon deposition of artificial saliva added with fluorophore (Alexa Fluor 488 Maleimide). Scale bar corresponds to 500 µm.
royalsocietypublishing.org/journal/rsfs Interface Focus 12: 20210044
5
kept a minimal contact with the ones with low SFE. For instance,
the droplet residue on glass spans awidth of 35 µm, which is the
largest of the five substrates, and a peak height of approximately
1.5 µm, which is the smallest among the substrates. By contrast,
the droplet nuclei on PTFE measure only approximately 10 µm
in diameter but are 3 µm high, which is the tallest in this range.
This observation echoes the results reported previously concern-
ing the evaporation process of picolitre droplets [38] that water
droplets followed a pinned contact line on glass but a moving
contact line on PTFE, which results in droplet nuclei of different
morphology and crystalline phase.
Glass, ABS and melamine possess minimal surface rough-
ness and low receding CAs (figure 4), on which the residue of
artificial saliva microdroplets was found in the form of small
dendrites or islands across the microdroplet residue region,
next to the large crystal (figure 6). This correlates with the
low receding CA measured on the same substrates at the
millimetre scale, which evidences that CCR mode is appro-
priate for the evaporation of respiratory microdroplets on
substrates with high SFE. While the initial spreading of a
microdroplet on the solid substrate is reflected by the advan-
cing CA, it is the receding CA that plays a critical role in the
subsequent evaporation process. Although SS possesses low
receding CA (figure 4), only a large crystal was found in
the droplet region, which is probably due to either its exces-
sive roughness in comparison with the other substrates tested
or the anisotropic nature of the surface finishing. PTFE has a
less surface roughness than SS, and a similar morphology
(crystal only) can be seen in figure 5c,e—this is because
PTFE has a low SFE and high receding CA (figure 4).
We can safely conclude that CCA mode of evaporation is
applicable here.
(a)1.6 mm
10 mm
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0
(b)
1.6 mm
0.0 mm
y: 52 mm
x: 52 mm
(c)
10 mm
3.0 mm
2.5
2.0
1.5
1.0
0.5
0
(d)
(e)
10 mm
5.0 mm
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0
(f)
3.8 mm
0.0 mm
y: 50 mm
x: 50 mm
(i)
10 mm
2.1 mm
1.8
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.1
(j)
2.0 mm
0.0 mm
y: 50 mm
x: 50 mm
(g)
10 mm
3.0 mm
2.5
2.0
1.5
1.0
0.5
0
(h)
2.5 mm
0.0 mm
y: 50 mm
x: 50 mm
2.9 mm
–0.0 mm
y: 50 mm
x: 50 mm
Figure 5. Morphology and the corresponding three-dimensional reconstructed images of artificial saliva droplet nuclei formed on (a,b) glass, (c,d) PTFE, (e,f ) SS,
(g,h) ABS and (i,j) melamine.
royalsocietypublishing.org/journal/rsfs Interface Focus 12: 20210044
6
The contrast between evaporation modes of respiratory
fluid on solid substrates not only demonstrates the effect of
surface characteristics on drying, but could have significant
implications on virus stability in the context of surface trans-
mission and virus detection. Evaporation of water in such a
fast timeframe could drastically change the phase and local
concentration of virus, protein and salt. Some previous
studies suggest that mucin could act as a protective matrix
to help virus surviving several days in a completely dried
condition [45]. Similar findings were reported for both
SARS-CoV and MERS-CoV that could be recovered from
inanimate substrates after days, weeks or months [46,47].
200 nm
180
160
140
120
100
80
60
40
0
200 nm
180
160
140
120
100
80
60
40
0
200 nm
180
160
140
120
100
80
60
40
0
2 mm
2 mm
2 mm
(b)
(a)
(c)
Figure 6. Small islands of nuclei formed aside the large crystal on (a) glass, (b) ABS and (c) melamine. Scale bar is 2 μm.
royalsocietypublishing.org/journal/rsfs Interface Focus 12: 20210044
7
The effect of fast transformation with the microenvironment
of each respiratory droplet on the structural integrity of a
virus is beyond the scope of this work. However, our work
shows that appropriate antiviral surface strategies could be
developed to inactivate or disrupt virus encapsulated in
microdroplet nuclei by modulating the evaporation process
on a surface. For example, surface deposited surfactants
could be used to influence the Marangoni flow and conse-
quently the evaporation mode. It is equally possible to
adjust SFE of a substrate once we find out whether the crys-
talline or amorphous phase is more effective in disrupting the
virus envelope.
3.5. Nanomechanical properties of the droplet nuclei
Nanomechanical measurements were performed to further
evaluate the proposed ‘core–shell’structure, during which a
nanoscopic tip (radius in the region of 10 nm) made contact
with the microdroplet nuclei and retracted subsequently at
a given frequency (1 Hz in the present work). Since the exper-
iments were carried out in ambient, capillary force between
the AFM tip (silicon nitride) and the nuclei dominates the
surface adhesion (recorded as the hysteresis between
approaching and retraction curves in figure 7a), which can
be described as
Fcap ¼4
p
R
g
Lcos
u
,ð3:1Þ
where Ris the AFM tip radius, γ
L
is the surface tension of
water and θis the CA of water on the two surfaces in contact.
It is clear that the capillary force is determined by the local
SFE of the nuclei that engage with the AFM tip. Used in
the past to evaluate the SFE of mineral such as calcite with
nanoscopic spatial resolution [48], this method was used in
the current study to survey various locations across the
microdroplet residue regions, including both the large crys-
tals and the small islands, found on all five inanimate
substrates. Equally, AFM-based force spectroscopy could be
used to identify any surface heterogeneity since polar chemi-
cal moieties would attract water molecules, which
consequently result in an increased surface adhesion [49].
At least 100 force curves, with a representative one shown
in figure 7a, were collected from each point examined to
ensure statistical robustness.
Averaged surface adhesion values are presented in
figure 7b. It shows that very similar magnitudes of surface
adhesion were acquired from the microdroplet nuclei depos-
ited on most of the substrates, except SS. The consistent
values of surface adhesion on glass, PTFE, ABS and melamine
confirm that the SFE of droplet nuclei is very similar, support-
ing that our proposed ‘core–shell’model is applicable on these
surfaces. The absolute values (around 20 nN) reported agree
with a previous study whereby surface adhesion between a
silicon AFM cantilever and a glass slide was measured [50]
as well a recent work by ourselves concerning surface
adhesion on hair fibres in ambient conditions [51]. The
nuclei present on SS resulted in a surface adhesion that is
approximately three times more than those on the other inan-
imate substrates, suggesting that they possess high SFE. We
speculate that this might be attributed to the exposure of
hydrophilic region of mucin on the surface as the result of
evaporation of the proteinaceous microdroplets.
By indenting the AFM cantilever into the entities present
on the solid substrate (droplet nuclei in the present work) for
tens of nanometres, it is possible to quantify the viscoelasti-
city of the nuclei and to evaluate their structural
characteristics. Young’s moduli of the nuclei, both large crys-
tal and small islands regions, are presented in figure 8. The
locations surveyed, across all five substrates, show values of
close range (approx. 4 MPa), except the small islands on
ABS. The Young’s modulus values are consistent with those
acquired on polymer film [52], which further supports the
proposed structure that mucin molecules present as the
shell for the microdroplet nuclei. Although it is not possible
to eliminate the influence of the underlying substrate on the
approaching
10
0
–10
–20
retraction
0200 400
displacement (nm)
600
force (nN)
(a)
surface adhesion (nN)
glass PTFE SS
large crystal
(b)
small islands
2 m
ABS melamine
80
60
40
20
0
Figure 7. Surface adhesion forces acquired on both the large crystals and the
small nuclei area present on all five different substrates. No small islands
were observed on PTFE and SS surfaces.
Young’s modulus (MPa)
g
lass PTFE SS
large crystal
small islands
2 m
ABS melamine
8
6
4
2
0
Figure 8. Young’s modulus as a function of surface materials receiving dro-
plets. Both large crystals (filled columns) and small nuclei (empty columns)
were surveyed.
royalsocietypublishing.org/journal/rsfs Interface Focus 12: 20210044
8
Young’s modulus, we suggest that the high value acquired
from the small islands on ABS is more likely attributed to
the variation in the mucin concentration.
4. Summary
An advanced inkjet printing method was used to generate
microdroplets of respiratory fluid on five common substrates.
We show that surface characteristics of the inanimate sub-
strate have a substantial impact on not only the drying
kinetics of respiratory fluid microdroplets, but also the prop-
erties of the resulting nuclei. The evaporation kinetic of
artificial saliva follows CCR mode on substrates with low
SFE or great roughness, but CCA mode on those with high
SFE. This results in two distinctively different morphologies
of the microdroplet nuclei. Atomic force microscopy-based
methods, force spectroscopy and nanoindentation were
deployed to investigate the nuclei on all five substrates,
which could be an invaluable approach in the future studies
of respiratory droplets. The nanomechanical measurement
results support a core–shell structure of the microdroplet
nuclei due to the fast evaporation process of microdroplet,
which could have significant implication on the surface
viability/transmissibility of viruses as the crystals might pre-
serve the integrity of viruses, or disrupt the virus structure.
The work highlights the importance of length scale on the
drying of droplets and consequently the possibility of surface
transmission of virus-containing small droplets.
Data accessibility. This article has no additional data.
Authors’contributions. A.K.P. was involved in conceptualization, method-
ology, validation, formal analysis, investigation, data curation and
writing the original draft; S.Q. was involved in validation, formal analy-
sis, investigation, data curation and writing the original draft; O.S. was
involved in validation, formal analysis, investigation, data curation and
writing the original draft; N.R. was involved in investigation and data
curation; M.B. was involved in methodology, validation, formal analy-
sis, investigation, data curation and writing the original draft; P.J.F. was
involved in conceptualization, methodology, validation, writing, visual-
ization, supervision and funding acquisition; Z.J.Z. was involved in
conceptualization, methodology, validation, writing, visualization,
supervision, project administration and funding acquisition.
Competing interests. We declare we have no competing interests.
Funding. This work was supported by the Engineering and Physical
Science Research Council (grant no. EP/V029762/1). Z.J.Z. thanks
the Royal Academy of Engineering for an Industrial Fellowship
(award no. IF2021\100).
Acknowledgements. We would like to thank Dr Huaiyu Yang, Department
of Chemical Engineering, Loughborough University, for discussion
related to surface crystallization phenomena. DuPont Teijin Film UK,
FiberLean Technologies and Innospec are thanked for their generous
support to the overall research project. We would like to thank Drs
Richard Thomas, Maurice Walker, and Jack Vincent, Defence Science
and Technology Laboratory, for fruitful discussion.
References
1. Leung NHL. 2021 Transmissibility and transmission
of respiratory viruses. Nat. Rev. Microbiol. 19,
528–545. (doi:10.1038/s41579-021-00535-6)
2. 2020 COVID-19 transmission—up in the air. Lancet
Respir. Med. 8, 1159. (doi:10.1016/S2213-
2600(20)30514-2)
3. van Doremalen N et al. 2020 Aerosol and surface
stability of SARS-CoV-2 as compared with SARS-
CoV-1. N. Engl. J. Med. 382, 1564. (doi:10.1056/
NEJMc2004973)
4. van Doremalen N et al. 2021 Surface-aerosol
stability and pathogenicity of diverse MERS-CoV
strains from 2012–2018. bioRxiv. (doi:10.1101/2021.
02.11.429193)
5. Tang JW, Marr LC, Li Y, Dancer SJ. 2021 Covid-19
has redefined airborne transmission. BMJ 373,
N913. (doi:10.1136/bmj.n913)
6. Goldman E. 2020 Exaggerated risk of transmission
of COVID-19 by fomites. Lancet Infect. Dis. 20, 892.
(doi:10.1016/S1473-3099(20)30561-2)
7. Greenhalgh T, Jimenez JL, Prather KA, Tufekci Z,
Fisman D, Schooley R. 2021 Ten scientific reasons in
support of airborne transmission of SARS-CoV-2.
Lancet 397, 1603. (doi:10.1016/S0140-
6736(21)00869-2)
8. Goldman E. 2021 SARS wars: the fomites strike
back. Appl. Environ. Microbiol. 87, e00653-21.
(doi:10.1128/AEM.00653-21)
9. Onakpoya IJ, Heneghan CJ, Spencer EA, Brassey J,
Plüddemann A, Evans DH, Conly JA, Jefferson T.
2021 SARS-CoV-2 and the role of fomite
transmission: a systematic review version 1; peer
review: 2 approved with reservations.
F1000Research 10, 233. (doi:10.12688/
f1000research.51590.1)
10. Ijaz MK, Nims RW, McKinney J. 2021 Indirect
transmission of severe acute respiratory syndrome
coronavirus virus 2 (SARS-CoV-2): what do we know
and what do we not know? Infect. Control Hosp.
Epidemiol. 1–2. (doi:10.1017/ice.2021.57)
11. Tang JW et al. 2021 Dismantling myths on the
airborne transmission of severe acute respiratory
syndrome coronavirus-2 (SARS-CoV-2). J. Hosp.
Infect. 110, 89. (doi:10.1016/j.jhin.2020.
12.022)
12. Bake B, Larsson P, Ljungkvist G, Ljungstrom E, Olin
AC. 2019 Exhaled particles and small airways.
Respir. Res. 20, 8. (doi:10.1186/s12931-019-0970-9)
13. Morawska LJGR, Johnson GR, Ristovski ZD,
Hargreaves M, Mengersen K, Corbett S, Chao CYH,
Li Y, Katoshevski D. 2009 Size distribution and sites
of origin of droplets expelled from the human
respiratory tract during expiratory activities.
J. Aerosol Sci. 40, 256. (doi:10.1016/j.jaerosci.2008.
11.002)
14. Fabian P, McDevitt JJ, DeHaan WH, Fung RO,
Cowling BJ, Chan KH, Leung GM, Milton DK. 2008
Influenza virus in human exhaled breath: an
observational study. PLoS ONE 3, e2691. (doi:10.
1371/journal.pone.0002691)
15. Papineni RS, Rosenthal FS. 1997 The size
distribution of droplets in the exhaled breath of
healthy human subjects. J. Aerosol Med. 10, 105.
(doi:10.1089/jam.1997.10.105)
16. Tang JW, Li Y, Eames I, Chan PK, Ridgway GL. 2006
Factors involved in the aerosol transmission of
infection and control of ventilation in healthcare
premises. J. Hosp. Infect. 64, 100. (doi:10.1016/j.
jhin.2006.05.022)
17. Edwards DA, Man JC, Brand P, Katstra JP,
Sommerer K, Stone HA, Nardell E, Scheuch G. 2004
Inhaling to mitigate exhaled bioaerosols. Proc. Natl
Acad. Sci. USA 101, 17383. (doi:10.1073/pnas.
0408159101)
18. Cazabat AM, Guéna G. 2010 Evaporation of
macroscopic sessile droplets. Soft Matter 6, 2591.
(doi:10.1039/b924477h)
19. Stauber JM, Wilson SK, Duffy BR, Sefiane K. 2015
On the lifetimes of evaporating droplets with
related initial and receding contact angles.
Phys. Fluids 27, 122101. (doi:10.1063/1.
4935232)
20. Fedorenko A, Grinberg M, Orevi T, Kashtan N. 2020
Survival of the enveloped bacteriophage Phi6
(a surrogate for SARS-CoV-2) in evaporated saliva
microdroplets deposited on glass surfaces. Sci. Rep.
10, 22419. (doi:10.1038/s41598-020-79625-z)
21. Di Novo NG, Carotenuto AR, Mensitieri G, Fraldi M,
Pugno NM. 2021 Modeling of virus survival time in
respiratory droplets on surfaces: a new rational
approach for antivirus strategies. Front. Mater. 8,
631723. (doi:10.3389/fmats.2021.631723)
22. Vejerano EP, Marr LC. 2018 Physico-chemical
characteristics of evaporating respiratory fluid
droplets. J. R. Soc. Interface 15, 20170939. (doi:10.
1098/rsif.2017.0939)
royalsocietypublishing.org/journal/rsfs Interface Focus 12: 20210044
9
23. Marr LC, Tang JW, Van Mullekom JV, Lakdawala SS.
2019 Mechanistic insights into the effect of
humidity on airborne influenza virus survival,
transmission and incidence. J. R. Soc. Interface 16,
20180298. (doi:10.1098/rsif.2018.0298)
24. World Health Organization. 2020 Surface sampling
of coronavirus disease (COVID-19): a practical ‘how
to’protocol for health care and public health
professionals.
25. Malamud D, Abrams WR, Barber CA, Weissman D,
Rehtanz M, Golub E. 2011 Antiviral activities in
human saliva. Adv. Dent. Res. 23, 34. (doi:10.1177/
0022034511399282)
26. White MR, Helmerhorst EJ, Ligtenberg A, Karpel M,
Tecle T, Siqueira WL, Oppenheim FG, Hartshorn KL. 2009
Multiple components contribute to ability of saliva to
inhibit influenza viruses. Oral Microbiol. Immunol. 24,
18. (doi:10.1111/j.1399-302X.2008.00468.x)
27. Shugars DC. 1999 Endogenous mucosal antiviral
factors of the oral cavity. J. Infect. Dis. 179, S431.
(doi:10.1086/314799)
28. Diaz-Arnold AM, Marek CA. 2002 The impact of
saliva on patient care: a literature review.
J. Prosthet. Dent. 88, 337. (doi:10.1067/mpr.2002.
128176)
29. Dodds MW, Johnson DA, Yeh CK. 2005 Health
benefits of saliva: a review. J. Dent. 33, 223–233.
(doi:10.1016/j.jdent.2004.10.009)
30. Humphrey SP, Williamson RT. 2001 A review of
saliva: normal composition, flow, and function.
J. Prosthet. Dent. 85, 162. (doi:10.1067/mpr.2001.
113778)
31. Aps JK, Martens LC. 2005 Review: the physiology of
saliva and transfer of drugs into saliva. Forensic Sci.
Int. 150, 119. (doi:10.1016/j.forsciint.2004.10.026)
32. Veerman EC, van den Keybus PA, Vissink A, Nieuw
Amerongen AV. 1996 Human glandular salivas: their
separate collection and analysis. Eur. J. Oral Sci.
104, 346. (doi:10.1111/j.1600-0722.1996.tb00090.x)
33. Wong L, Sissons C. 2001 A comparison of human
dental plaque microcosm biofilms grown in an
undefined medium and a chemically defined
artificial saliva. Arch. Oral Biol. 46, 477. (doi:10.
1016/S0003-9969(01)00016-4)
34. Owens DK, Wendt RC. 1969 Estimation of the
surface free energy of polymers. J. Appl. Polym. 13,
1741. (doi:10.1002/app.1969.070130815)
35. Good RJ, Girifalco LA. 1960 A theory for estimation
of surface and interfacial energies. III. Estimation of
surface energies of solids from contact angle data.
J. Phys. Chem. 64, 561–565. (doi:10.1021/
j100834a012)
36. Picknett RG, Bexon R. 1977 The evaporation of
sessile or pendant drops in still air. J. Colloid
Interface Sci. 61, 336. (doi:10.1016/0021-
9797(77)90396-4)
37. Stauber JM, Wilson SK, Duffy BR, Sefiane K. 2015
Evaporation of droplets on strongly hydrophobic
substrates. Langmuir 31, 3653. (doi:10.1021/acs.
langmuir.5b00286)
38. Talbot EL, Berson A, Brown PS, Bain CD. 2012
Evaporation of picoliter droplets on surfaces with
a range of wettabilities and thermal conductivities.
Phys. Rev. E 85, 061604. (doi:10.1103/PhysRevE.85.
061604)
39. Bonn D, Eggers J, Indekeu J, Meunier J, Rolley E.
2009 Wetting and spreading. Rev. Mod. Phys. 81,
739. (doi:10.1103/RevModPhys.81.739)
40. Sheng YJ, Jiang S, Tsao HK. 2007 Effects of
geometrical characteristics of surface roughness on
droplet wetting. J. Chem. Phys. 127, 234704.
(doi:10.1063/1.2804425)
41. Ávila-Sierra A, Zhang ZJ, Fryer PJ. 2021 Effect of
surface roughness and temperature on stainless
steel–whey protein interfacial interactions under
pasteurisation conditions. J. Food Eng. 301, 110542.
(doi:10.1016/j.jfoodeng.2021.110542)
42. Lieber C, Melekidis S, Koch R, Bauer HJ. 2021
Insights into the evaporation characteristics of saliva
droplets and aerosols: levitation experiments and
numerical modeling. J. Aerosol. Sci. 154, 105760.
(doi:10.1016/j.jaerosci.2021.105760)
43. Deegan RD, Bakajin O, Dupont TF, Huber G, Nagel
SR, Witten TA. 1997 Capillary flow as the cause of
ring stains from dried liquid drops. Nature 389, 827.
(doi:10.1038/39827)
44. Hu H, Larson RG. 2006 Marangoni effect reverses
coffee-ring depositions. J. Phys. Chem. B 110, 7090.
(doi:10.1021/jp0609232)
45. Thomas Y, Vogel G, Wunderli W, Suter P, Witschi M,
Koch D, Tapparel C, Kaiser L. 2008 Survival of
influenza virus on banknotes. Appl. Environ.
Microbiol. 74, 3002. (doi:10.1128/AEM.00076-08)
46. Otter JA, Donskey C, Yezli S, Douthwaite S,
Goldenberg SD, Weber DJ. 2016 Transmission of
SARS and MERS coronaviruses and influenza virus in
healthcare settings: the possible role of dry surface
contamination. J. Hosp. Infect. 92, 235. (doi:10.
1016/j.jhin.2015.08.027)
47. Kampf G, Todt D, Pfaender S, Steinmann E. 2020
Persistence of coronaviruses on inanimate surfaces and
their inactivation with biocidal agents. J. Hosp. Infect.
104, 246. (doi:10.1016/j.jhin.2020.01.022)
48. Sauerer B, Stukan M, Abdallah W, Derkani MH,
Fedorov M, Buiting J, Zhang ZJ. 2016 Quantifying
mineral surface energy by scanning force
microscopy. J. Colloid Interface Sci. 472, 237.
(doi:10.1016/j.jcis.2016.03.049)
49. Butt HJ, Cappella B, Kappl M. 2005 Force
measurements with the atomic force microscope:
technique, interpretation and applications. Surf. Sci.
Rep. 59,1–152. (doi:10.1016/j.surfrep.2005.08.003)
50. Jones R, Pollock HM, Cleaver JAS, Hodges CS. 2002
Adhesion forces between glass and silicon surfaces
in air studied by AFM: effects of relative humidity,
particle size, roughness, and surface treatment.
Langmuir 18, 8045. (doi:10.1021/la0259196)
51. Labarre L, Squillace O, Liu Y, Fryer PJ, Kaur P, Whitaker
S, Marsh JM, Zhang ZJ. Submitted. Hair surface
interactions against different chemical functional
groupsas a functionof environment and haircondition.
52. Sun Y, Akhremitchev B, Walker GC. 2004 Using the
adhesive interaction between atomic force
microscopy tips and polymer surfaces to measure
the elastic modulus of compliant samples. Langmuir
20, 5837. (doi:10.1021/la036461q)
royalsocietypublishing.org/journal/rsfs Interface Focus 12: 20210044
10
Available via license: CC BY 4.0
Content may be subject to copyright.
