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

net/publication/273458798

Molecular dynamics simulation on evaporation of water and aqueous droplets

in the presence of electric field

Article  in  International Journal of Heat and Mass Transfer · June 2014

DOI: 10.1016/j.ijheatmasstransfer.2014.02.034

CITATIONS READS

23 293

4 authors, including:

Bing Bing Wang Xiao-Dong Wang

Northeast Electric Power University North China Electric Power University
13 PUBLICATIONS   114 CITATIONS    222 PUBLICATIONS   4,290 CITATIONS   

SEE PROFILE SEE PROFILE

Yuan-Yuan Duan
Tsinghua University
253 PUBLICATIONS   3,325 CITATIONS   

SEE PROFILE

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

Manipulating spectral near-field radiative transfer using microstructures View project

Thermal radiative and optical properties of ultrathin films with simple planar structures View project

All content following this page was uploaded by Bing Bing Wang on 07 October 2020.

The user has requested enhancement of the downloaded file.

This article appeared in a journal published by Elsevier. The attached
copy is furnished to the author for internal non-commercial research
and education use, including for instruction at the authors institution
and sharing with colleagues.
Other uses, including reproduction and distribution, or selling or
licensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of the
article (e.g. in Word or Tex form) to their personal website or
institutional repository. Authors requiring further information
regarding Elsevier’s archiving and manuscript policies are
encouraged to visit:
http://www.elsevier.com/authorsrights
 Author's personal copy

International Journal of Heat and Mass Transfer 73 (2014) 533–541

Contents lists available at ScienceDirect

International Journal of Heat and Mass Transfer

journal homepage: www.elsevier.com/locate/ijhmt

Molecular dynamics simulation on evaporation of water and aqueous

droplets in the presence of electric field
Bing-Bing Wang a,b, Xiao-Dong Wang a,b,⇑, Yuan-Yuan Duan c, Min Chen d,⇑
a
State Key Laboratory of Alternate Electrical Power System with Renewable Energy Sources, North China Electric Power University, Beijing 102206, China
b
Beijing Key Laboratory of Multiphase Flow and Heat Transfer for Low Grade Energy, North China Electric Power University, Beijing 102206, China
c
Key Laboratory of Thermal Science and Power Engineering of MOE, Tsinghua University, Beijing 100084, China
d
Department of Engineering Mechanics, Tsinghua University, Beijing 100084, China

a r t i c l e i n f o a b s t r a c t

Article history: The study on evaporation of water droplet with dissolved salt is important for the understanding of elec-
Received 7 August 2013 trospinning and electrospray mass spectrometry experiments. In this work the dynamics of evaporation
Received in revised form 22 January 2014 of water droplet and aqueous droplet with dissolved either LiCl, NaCl or KCl salt in the absence and pres-
Accepted 13 February 2014
ence of a homogeneous electric field are comparatively studied by molecular dynamics simulations. The
Available online 14 March 2014
results show that (1) the evaporations of water and aqueous droplets are all accelerated by applying an
electric field; (2) the aqueous droplet evaporates slower than the water droplet at low electric field of
Keywords:
0.4 V nm1, however, its evaporation rate is enhanced significantly and exceeds the water droplet at high
Evaporation
Electric field
electric fields of 0.6 and 0.8 V nm1; (3) aqueous droplet with dissolved KCl molecules evaporates the
Aqueous droplet fastest, and then followed by dissolved NaCl and LiCl molecules; (4) an optimal salt concentration is
Electrospinning observed for the aqueous droplet, above which the evaporation rate is reduced significantly and may
Molecular dynamics cause a electrospinning failure. The present results are qualitatively agree with the observations in the
electrospinning experiment and are attributed to the fact that the electric field applied causes the droplet
deformation, directional arrangement of water molecules along the electric field direction, acceleration of
water molecules due to ions acceleration, as well as hydration effect.
Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction
The evaporation of the droplet with dissolved salts in the pres-
ence of a high-voltage electric field has been applied in the elec-
trospinning [1–10] and electrospray ionization process [11–15].
Electrospinning is a straightforward method to produce continu-
ously polymer fibers. The standard setup for the electrospinning
is composed of a syringe needle, connected to a high-voltage
power supply, a syringe pump, and a grounded metal plate [16].
When a strong electrostatic force is applied to the syringe needle,
the polymer solution is ejected from the syringe needle to form a
thin spraying stream, finally, the polymer deposits and solidifies
as a non-woven fibrous structure on the grounded metal plate after
the solvent evaporates almost completely. Because the fibers pre-
pared by the electrospinning has nanoscale diameter and a large
⇑ Corresponding authors. Address: State Key Laboratory of Alternate Electrical
Power System with Renewable Energy Sources, North China Electric Power
University, Beijing 102206, China. Tel./fax: +86 10 62321277 (X.-D. Wang). Tel.:
+86 10 62773776; fax: +86 10 62795832 (M. Chen).
E-mail addresses: wangxd99@gmail.com (X.-D. Wang), mchen@mail.tsinghua.
edu.cn (M. Chen).
surface-to-volume ratio, the fibers can be used as filter media, sep-
aration membranes, wound dressing materials, artificial blood ves-
sels, sensors, composite reinforcements [17–20].
Many experimental investigations [1–10] have shown that the
morphology of fibers prepared by the electrospinning is strongly
related to the spinning conditions such as polymer concentration,
needle-to-plate distance, applied electric field strength, and sol-
vent evaporation. Some investigations [7–10] also demonstrated
that adding the ionic salts into the spinning solution can improve
the solution spinnability remarkably. Qin et al. [7] found no fibers
were produced by electrospinning under the electric voltage of
5 kV when the Polyacrylonitrile (PAN) polymer solution did not
dissolve salt, however, adding a small amount of LiCl salt into
the solution increased the solution spinnability significantly due
to enhancement of solution conductivity. Zong et al. [8] studied
the effect of three kinds of salts (NaH2PO4, KH2PO4, NaCl) on elec-
tro spinning of biodegradable amorphous poly (D, L-lactide) (PDLA)
membrane. They found that the smallest, moderate, and the largest
average diameters of fibers were produced when 1 wt% NaCl, 1 wt%
NaH2PO4, and 1 wt% KH2PO4 were respectively added into the
30 wt% PDLA solution as the spinning solutions. The dependence

http://dx.doi.org/10.1016/j.ijheatmasstransfer.2014.02.034
0017-9310/Ó 2014 Elsevier Ltd. All rights reserved.
 Author's personal copy
534 B.-B. Wang et al. / International Journal of Heat and Mass Transfer 73 (2014) 533–541
of fiber diameter on the salt type was attributed to the charge den-
sity carried by jet and the ion size. Qin et al. [9] investigated the
internal structure of electrospinning polyacrylonitrile (PAN) nanof-
ibers by adding different content of LiCl (1–4 wt%). The results
showed that the added salts enhanced the PAN fiber spinnability,
however, did not change the chemical composition of PAN, and
showed that higher LiCl content reduced the diameter of PAN.
Kim et al. [10] investigated the effect of adding the various concen-
trations of NaCl (0, 0.01, 0.1, and 1 M) on the morphology of Poly
(acrylic acid) (PAA) nanofibers. The smallest diameter of the PAA
nanofibers was prepared by PAA solution with 0.01 M NaCl, how-
ever, the nanofibers could not be produced when the concentration
of NaCl increased to 1 M.
Though the evaporation of the solvent is only the concomitant for
the electrospinning process, fast evaporation is essential to formation
of ultrafine fibers with nanoscale diameters [16]. If the evaporation of
the solvent is too slow, it will cause a failure of electrospinning be-
cause the polymer solution (not polymer fiber) is deposited on the
metal plate [10]. Therefore, higher electric field strength and/or addi-
tion of the salt into the solution can improve the solution spinnability
may be attributed to the enhanced evaporation rate of the solvent.
Unfortunately, the responsible mechanisms for enhanced evapora-
tion rate at thus conditions are not yet understood well.
Molecular dynamics attempts to simulate the real behavior of
Nature by identifying each atom and following their motion in
time through the basic laws of classical mechanics. Molecular
dynamics simulating an evaporation process has no need of some
assumptions made by computational fluid dynamics, so this meth-
od was adopted extensively to study the evaporation of liquid
droplet [21–28] or thin liquid film [29–31] in the absence of an
electric field. In the recent decade, molecular dynamics simulations
were also used to investigate the behavior of water droplet with
dissolved ions [11,12,14,15], macroions [13], or biological macro-
molecules [32] for understanding of electrospray mass spectrome-
try experiments. The main motivation for these investigations [11–
15,32] is to analyze biological macromolecules with high accuracy
and throughput. By the usage of electrospray ionization, nucleic
acids, proteins, carbohydrates, and synthetic polymers have been
extensively analyzed with ultrahigh accuracy [15]. These investi-
gations just focused on how to produce completely desolvated
macromolecules, but did not concern the effect of addition of ions
on the droplet evaporation.
The objective of this work is to understand which parameters af-
fect the solvent evaporation and how to accelerate the solvent evap-
oration in the electrospinning for fabricating high-quality nanoscale
fibers. Because water has often been used as a solvent in the electros-
pinning, water droplet is selected in this work. The dynamic behavior
and evaporation rate of water droplet (1120 water molecules) with-
out or with dissolved LiCl, NaCl or KCl (20, 30, 40, 80, 100, 120 salt
molecules) in the absence or presence of a homogeneous electric
field (0.4, 0.6, 0.8 V nm1 along x-direction) are comparatively stud-
ied by performing molecular dynamics simulations, which can give
some insight into mechanism of enhanced evaporation rate. The
droplet is heated in 600 K nitrogen gas atmosphere to activate its
evaporation. Because there are only non-covalent bonds between
water and most of macromolecular polymer used in the electrospin-
ning [32], the macromolecular polymer has a small effect on the
evaporation rate, and hence does not be included in the droplet.
2. Molecular dynamics simulations
2.1. Simulation cases
Molecular dynamics simulations are performed for evaporation
of the water droplet and the aqueous droplet with dissolved either
NaCl, LiCl or KCl in the absence or presence of a homogeneous
E
y
z x
Fig. 1. Model of initial evaporation system: green ball is N, white ball is H, red ball
is O, blue ball is positive ion, and purple ball is chloride ion. (For interpretation of
the references to colour in this figure legend, the reader is referred to the web
version of this article.)
electric field. The droplet is surrounded and heated by nitrogen
gas during the evaporation. The initial configuration of the evapo-
ration system is shown in Fig. 1. The droplet and nitrogen gas are
placed in a cubic box with 12 nm side length. Ions, nitrogen mole-
cules, and water molecules are distinguished with different colors.
The radius of the initial droplets is 2 nm, and the number of water
molecules in the droplet is 1120 for all simulation systems, which
corresponds to the pure water density of 1 g cm3. The number of
nitrogen molecules is 600 corresponding to 16.47 kg m3 gaseous
density. Three kinds of salts (120 NaCl, LiCl or KCl molecules) are
respectively added into the droplet to analyze the effect of salt type
on the droplet evaporation, and 20, 30, 40, 80, 100, and 120 LiCl
molecules are respectively added into the droplet to analyze the ef-
fect of salt concentration on the droplet evaporation. A homoge-
neous electric field E along x-axis direction is applied to the box,
so an additional force Fi = qiE is imposed on each charge qi. Okuno
et al. [16] estimated the electric field around the tip of syringe nee-
dle during the electrospinning and proposed that the value was in
the range of about 0.7 to 100 V nm1 for pure water as spinning
solution. The electric field value of 0.4, 0.6, and 0.8 V nm1 are
adopted in the present work, two smaller values are selected in or-
der to understand why the solution spinnability is worse for the
electric field lower than the threshold value of 0.7 V nm1.
2.2. Potential function
In recent decade, some studies [33,34] have compared the prop-
erties of water and/or aqueous solution predicted by various water
models. More recently, Vega and Abascal [35] conducted a compre-
hensive comparison between various water models. In their
comparison, a certain number of points between zero (bad agree-
ment) and ten (good agreement) was given for the predictions of
each model and property. The results indicated that TIP3P, TIP5P,
TIP4P, SPC/E and TIP4P/2005 obtained an average score of 2.7,
3.7, 4.7, 5.1, and 7.2, respectively. The SPC/E model for water mol-
ecules was originally proposed by Berendsen et al. [36], and the
HO-distance and H–O–H angle for the SPC/E model are 0.1 nm
and 109.47°, respectively. Using the SPC/E water model, Chowdh-
uri and Chandra [37] performed a series of molecular dynamics
simulations of aqueous NaCl and KCl solutions. Their results of
the concentration dependence of self-diffusion coefficients were
found to be closer to the corresponding experimental values. The
study objects of the present work are also aqueous droplets with
dissolved salts, so that the SPC/E water model is chosen. Nitrogen
 Author's personal copy

B.-B. Wang et al. / International Journal of Heat and Mass Transfer 73 (2014) 533–541 535

Table 1 molecules are maintained the fixed bond length and bond angle
Values of potential parameters [37–39]. by the method of constraints [45].
Particle r (Å) e (kJ mol1) q (e)
O 3.169 0.6502 0.8476 2.3. Equilibrium preparation
H 0.000 0.0000 +0.4238
Na+ 2.583 0.4184 +1.0000 The numerical scheme adopted to solve Newtonian equations of
Li+ 1.505 0.6904 +1.0000
K+ 3.331 0.4184 +1.0000
motion for every particle is based on the Velocity–Verlet algorithm
Cl 4.400 0.4184 1.0000 [46]. Before the onset of droplet evaporation, a well-defined sys-
N 3.710 0.6990 +0.0000 tem composed of the droplet and the surrounding nitrogen has
to be prepared. The initial velocities of particles in both gaseous
phase and droplet are assigned to be Maxwell–Boltzmann distribu-
molecules and ions are modeled as two-centers-Lennard–Jones
tions with the desired temperature of 298 K. Periodic boundary
particles [38] and charged Lennard–Jones particles [37,39], respec-
condition is applied for all the three coordinates, and the electric
tively. So the intermolecular interactions of particles are expressed
field strength, E, is set to 0 V nm1 during the equilibrium run.
as a sum of Coulombic and Lennard–Jones 12–6 potentials [37,39],
Time step length dt = 1 fs is used for all simulations. The droplet
that is,
and the surrounding nitrogen are simulated in an NVT ensemble.
" 12  6 #
qi qj rij rij The velocities of the particles in gaseous phase and the droplet
U ij ¼ þ 4eij  ð1Þ are separately rescaled to keep a constant temperature of 298 K
r ij rij r ij
for every time step. The system reaches an equilibrium state after
where qi, rij, ri and ei denote the charge of ith particle, the distance 100,000 time steps, and only a very small amount of water mole-
between ith and jth particles, the minimum energy and the zero en- cules escape from the droplet and occur in the surrounding vapor
ergy separation distance, respectively. The values of the potential as shown Fig. 2a.
parameters qi, ri and ei of water, ions and nitrogen are summarized
in Table 1 [37–39]. The potential parameters between different par- 2.4. Droplet evaporation
ticles are described by the mixing rules, i.e., rij = (ri + rj)/2, and
eij = (ei + ej)1/2. A spherical truncated distance for short-range forces After the preparation of equilibrium system, the value of the
is taken as 10 Å. Several methods, such as the Ewald method [40], homogeneous electric field is set to 0, 0.4, 0.6 or 0.8 V nm1 for dif-
the Lekner method [41,42], and the PPPM method [43,44], have ferent evaporation processes, in order to investigate the effect of
been used to evaluate the long-range dipolar interactions. In our the electric field strength on the droplet evaporation. To trigger
preliminary simulations, the Ewald method and the PPPM method the droplet evaporation, the nitrogen temperature is abruptly
are used to respectively calculate the evaporation of the droplet raised to 600 K to heat the droplet, and this temperature is kept
with dissolved 20 or 30 LiCl molecules at E = 0.8 V nm1. The results by velocity-rescaling method [21] for every time step during the
demonstrate that the two methods predict almost the same evapo- whole evaporation process. The instantaneous positions and veloc-
ration rates, however, the PPPM method holds much faster calcula- ities of particles are recorded every 1 ps and all quantities of inter-
tion speed than the Ewald method. Moreover, Daub et al. [44] est are calculated statistically for 10 recorded results to reduce
investigated electrowetting of a salty nanodroplet on a solid surface statistical fluctuations. Water molecules are considered to be in
by molecular dynamic simulations. Their simulation system also in- the droplet if their neighbor molecule number within a distance
cluded a droplet–vapor interface, and non-truncated electrostatic of 4.34 Å, nneighbor P 4, and in the vapor phase if nneighbor < 4 [28].
interactions were treated by the PPPM method. Their simulation re- The snapshots of evaporation for the water droplet with dis-
sults were found to be in good agreement with the experimental solved 120 NaCl molecules and the number density of water mol-
measurements. Therefore, the PPPM method is chosen here to treat ecules, Na+ and Cl ions at different evaporation instants for
the long-range electrostatic interaction. Nitrogen and the water E = 0.8 V nm1 are presented in Fig. 2. The number density of the

(a) Cl
- (b) - (c) Cl
- (d) Cl
-
Cl
+ +
Na Na
+
Na
+ Na
water water water water

0 20 40 60 80 100 120 0 20 40 60 80 100 120 0 20 40 60 80 100 120 0 20 40 60 80 100 120

+ 
Fig. 2. Snapshots of the simulation box and corresponding number densities of water molecule, Na and Cl ions at different instants: (a) t = 0 ps; (b) t = 100 ps; (c) t = 200 ps;
(d) t = 1000 ps.
 Author's personal copy

536 B.-B. Wang et al. / International Journal of Heat and Mass Transfer 73 (2014) 533–541

Water droplet particles at a distance x from origin is measured by the molecule

1400 -1
E=0.0 V nm , E=0.4 V nm
-1
number in a cubic shell with 1 Å width divided by the volume of
-1
E=0.6 V nm , E=0.8 V nm
-1
the shell. The aqueous droplet is stretched to be an approximate
1200
Aqueous droplet with 120 NaCl column and its volume gradually decreases with the time.
-1 -1
E=0.0 V nm , E=0.4 V nm
1000 -1 -1
E=0.6 V nm , E=0.8 V nm
3. Results and discussion
800
N

3.1. Effect of electric field

600
Fig. 3 shows the temporal evolution of the water molecule num-
400 ber in the water droplet and the aqueous droplet with dissolved
120 NaCl molecules at the electric fields of 0, 0.4, 0.6, and
200 0.8 V nm1. Evaporation rates of the water and aqueous droplets
all increase with the increased electric field, which agrees with
0
0 200 400 600 800 1000 1200 1400 1600 the earlier studies of Okuno et al. [16] and Vaitheeswaran et al.
[47], they have found that the evaporation of thin water film is en-
t (ps)
hanced by applying of an electric field. Fig. 3 also shows that the
Fig. 3. Temporal evolution of the water molecules number in the water droplet and evaporation of the aqueous droplet is slower than the water drop-
the droplet with dissolved 120 NaCl molecules at various electric fields. let at low electric fields of 0 and 0.4 V nm1. The evaporation rate

Pure water E=0.4 V/nm

a

0 ps 100 ps 200 ps 1000 ps

b Pure water E=0.8 V/nm

0 ps 100 ps 200 ps 1000 ps

NaCl 120 E=0.4 V/nm

c

0 ps 100 ps 200 ps 1000 ps

d NaCl 120 E=0.8 V/nm

0 ps 100 ps 200 ps 1000 ps

Fig. 4. Snapshots of droplets at different evaporation instants.

 Author's personal copy

B.-B. Wang et al. / International Journal of Heat and Mass Transfer 73 (2014) 533–541 537

Fig. 5. Snapshots of local particle distributions at t = 120 ps and at E = 0.8 V nm1: (a) water droplet; (b) aqueous droplet with dissolved 120 NaCl molecules.

Table 2
5 Water droplet Average interaction energy between water molecules and ions (Na+ and Cl) for
E=0.0 V nm
-1 aqueous droplet with 120 NaCl molecules at E = 0.8 V nm1 at different instants.
-1
E=0.4 V nm Time (ps) 0 40 80 120 160
4 E=0.8 V nm
-1
Average interaction energy 47.95 45.80 43.04 42.17 37.89
(kJ mol1)
N water

2
Aqueous droplet with 120 NaCl
-1
16 (a) Aqueous droplet with 120 NaCl
E=0.0 V nm
-1
E=0 V nm
-1
1 E=0.4 V nm
-1
14 E=0.4 V nm
-1
E=0.8 V nm
-1
E=0.8 V nm
Ek,Na ×10 (J)

0 12
17

0 100 200 300 400 500 600

t (ps) 10
+

Fig. 6. Temporal evolution of the coordination number of water molecules.

8

of the aqueous droplet is elevated remarkably as the electric field is 6

increased. At E = 0.6 V nm1, the evaporation of the aqueous drop-
0 20 40 60 80 100 120 140 160 180 200
let is still slower than the water droplet at t < 500 ps, the rate is in-
t (ps)
creased sharply at t > 500 ps and the droplet almost completely
evaporates at t = 640 ps. At E = 0.8 V nm1, the evaporation of the 16
aqueous droplet is further enhanced and the lifetime of the droplet (b) Aqueous droplet with 120 NaCl
-1
is only 220 ps far less than 920 ps of the water droplet. The results 14 E=0 V nm
-1
above imply that adding the ions into the droplet cannot always E=0.4 V nm
-1
enhance the droplet evaporation when the electric field strength E=0.8 V nm
12
is lower than the threshold value of electrospinning, which is in
Ek,Cl ×10 (J)

good agreement with the observation in the experiments [7].

17

10
Fig. 4 presents snapshots of evaporation process of the water
and aqueous droplets at t = 0, 100, 200 and 1000 ps for E = 0.4
-

8
and 0.8 V nm1. At E = 0.4 V nm1, the water and aqueous droplets
almost preserve initial spherical shape throughout the whole evap-
oration process due to weak electric field force, and the size of 6
aqueous droplet is slightly larger than the water droplet at the
same evaporation instant. However, when the electric field is in- 4
0 20 40 60 80 100 120 140 160 180 200
creased to 0.8 V nm1, strong electric field force induces the defor-
t (ps)
mation of the droplets, the both droplets are stretched to be the
cylindrical shape along the electric field direction. Larger surface Fig. 7. Temporal evolution of Ek,ion in the aqueous droplet at different electric fields:
areas for the elongated droplets prompt the collision between (a) sodium ions; (b) chloride ions.
nitrogen molecules and water molecules, and hence enhance the
droplet evaporation at high electric field. It is worth noting that
the applied electric field also causes the directional arrangement where oxygen atoms are closer to sodium ions and hydrogen
of water molecules in the elongated droplets. Almost all of oxygen atoms are closer to chloride ions, since the Coulombic forces be-
atoms align along the opposite direction of the electric field for the tween water molecules and ions are stronger than the electric field
water droplet (Fig. 5a), which agrees with the previous report of force (Fig. 5b).
Okuno et al. [16]. However, the directional arrangement is de- Fig. 6 shows the temporal evolution of the coordination number
stroyed in the vicinity of Na and Cl ions for the aqueous droplet, of water molecules for the water and aqueous droplets at E = 0, 0.4,
 Author's personal copy

538 B.-B. Wang et al. / International Journal of Heat and Mass Transfer 73 (2014) 533–541

and 0.8 V nm1. The coordination number is defined as the average
number of water molecules in a sphere with 0.35 nm radius around
a water molecule. The value of 0.35 nm chosen here is based on the
fact that it is a standard length to determine the formation of
hydrogen bonds between water molecules [48]. Small coordination
number means weak interactions between water molecules, thus,
water molecules should escape from the droplet more easily. At
(a) Aqueous droplet with 120 NaCl
t = 0 ps, the electric field is not applied to the droplets, the coordi-
nation numbers are 4.47 for the water droplet and 3.71 for the
aqueous droplet. Small coordination number for the aqueous drop-
let is attributed to the strong attractive force between water mol-
ecules and ions (hydration effect). Our previous study [28]
demonstrated that the hydration effect produced by adding the
ions into the droplet had a dominant influence on the droplet evap-
oration compared with decreased interactions between water mol-
ecules, and the hydration effect became stronger and stronger
during the evaporation, so that the aqueous droplet evaporated
gCl -H(r) t=0 ps
-

slower than the water droplet in the absent of the electric field.
120 gCl -H(r) t=50 ps
-

At low electric field, the hydration effect is still the dominant factor
gCl -H(r) t=100 ps
-
to determine the evaporation, hence, the evaporation of the aque-
100
gCl -H(r) t=150 ps
- ous droplet is still slower than the water droplet at E = 0.4 V nm1.
gCl -O(r) t=0 ps However, the hydration effect reduces during evaporation for
80 -

gCl -O(r) t=50 ps

E = 0.8 V nm1 (Table 2), which implies that the negative effect
g (r)

on evaporation produced by adding ions into the droplet weakens

60 gCl -O(r) t=100 ps
-

at high electric field. Thus, the coordination number of water mol-

gCl -O(r) t=150 ps
-
ecules can be used as a good indicator to determine the evapora-
40
tion rate. Fig. 6 shows that the coordination number becomes
less for the water and aqueous droplets at E = 0.8 V nm1 than
20 those at E = 0 and 0.4 V nm1 due to the alignment of water mole-
cules under action of high electric field. This is the second reason
0 for faster evaporation of the water and aqueous droplets at high
0 1 2 3 4 5 6 7 8
electric field compared to that without electric field or at low elec-
r (Å) tric field. Okuno et al. [16] investigated the evaporation of thin
water film on a solid wall at a high electric field of 1 V nm1, they
(b) Aqueous droplet with 120 NaCl gNa -O(r) t=0 ps
+ also found that the interactions among water molecules reduced
300 gNa -O(r) t=50 ps
+
significantly due to the existence of high electric field.
gNa -O(r) t=100 ps
+
Fig. 7 shows the average kinetic energy of sodium and chloride
250 ions in the aqueous droplet at E = 0, 0.4 and 0.8 V nm1. At low
gNa -O(r) t=150 ps
electric field of 0.4 V nm1, Ek,ion is almost the same with that of
+

gNa -H(r) t=0 ps

E = 0 V nm1 due to weak electric field force. However, Ek,ion is in-
+

200
gNa -H(r) t=50 ps
+
creased significantly as the electric field is increased to 0.8 V nm1.
g (r)

gNa -H(r) t=100 ps

+
The acceleration of ions can be explained by the following two rea-
150
gNa -H(r) t=150 ps
+ sons: strong electric field force is imposed to ions by high electric
field, the separation distance between sodium and chloride ions
100
becomes large due to elongated droplet, which reduces their
attractive force.
50
The radial distribution functions of g Cl O ðrÞ, g Cl H ðrÞ, g Naþ H ðrÞ,
and g Naþ O ðrÞ for the aqueous droplet at E = 0.8 V nm1 are shown
0
0 1 2 3 4 5 6 7 8 in Fig. 8. The first peak values of g Naþ O ðrÞ, g Naþ H ðrÞ, g Cl H ðrÞ, and
g Cl O ðrÞ occur at r = 1.35, 1.95, 2.35, and 3.25 Å, respectively. It
r (Å)
indicates that most of oxygen atoms are closer to sodium ions,
Fig. 8. Radial distribution functions at various instants for the aqueous droplet at while hydrogen atoms are closer to chloride ions due to Coulombic
E = 0.8 V nm1: (a) g Cl O ðrÞ and g Cl H ðrÞ; (b) g Naþ O ðrÞ and g Naþ H ðrÞ.

1200
14 Water droplet 1080 Water droplet
E=0.4 V nm
-1 Aqueous droplet
13 960 with 120 KCl
-1
E=0.8 V nm
12 840 with 120 NaCl
Aqueous droplet with 120 NaCl
with 120 LiCl
×10 (J)

-1
E=0.4 V nm 720
11
17

-1
E=0.8 V nm
N

600
10
k, water

480
9
E

360
8
240
7 120
6 0
0 20 40 60 80 100 120 140 160 180 200 0 100 200 300 400 500 600 700 800 900 10001100
t (ps) t (ps)

Fig. 9. Temporal evolution of Ek,water in the water and aqueous droplets at different Fig. 10. Temporal evolution of the water molecule number in the water and
electric fields. aqueous droplets at E = 0.8 V nm1.
 Author's personal copy

B.-B. Wang et al. / International Journal of Heat and Mass Transfer 73 (2014) 533–541 539

500
Aqueous droplet
400 with 120 KCl
with 120 NaCl
g(r) cation-O

300
with 120 LiCl
200 t = 0 ps t = 100 ps t = 200 ps

100

0
1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5
r (Å) r (Å) r (Å)

Fig. 11. The radial distribution functions of g(r)cation-O at different evaporation instants (cation denotes K+, Na+, or Li+).

attractive action, which agrees with the result observed in Fig. 5b.
The first peak positions of g Cl O ðrÞ, g Cl H ðrÞ, g Naþ H ðrÞ, and 1200
Aqueous droplet
g Naþ O ðrÞ are almost unchanged throughout the whole evaporation
with 20 LiCl
process, which implies that the water molecules in the close vicin- 1000
with 30 LiCl
ity of ions do not change their relative positions. Thus, it can be ex-
with 40 LiCl
pected that water molecules will be accelerated due to increased 800 with 80 LiCl
kinetic energy of ions driven by high electric field. Fig. 9 shows
with 100 LiCl
the temporal evolution of the average kinetic energy of water mol- with 120 LiCl
600
N

ecules in the water and aqueous droplets at different electric fields.

At the same high electric field of 0.8 V nm1, Ek,water for the aque-
ous droplet is far larger than that for the water droplet at 400
t > 100 ps. As a result, water molecules can escape from the droplet
more easily, which results in faster evaporation of the aqueous
200
droplet than the water droplet at high electric field.

3.2. The effect of salt type 0

Fig. 10 shows the temporal evolution of the water molecule
number in the water droplet and the aqueous droplet with dis-
solved 120 LiCl, NaCl or KCl molecules at E = 0.8 V nm1. Fig. 10
again confirms that adding a small amount of salt significantly en-
hances the droplet evaporation at high electric field. The lifetime is
about 200 ps for KCl droplet, 220 ps for NaCl droplet, and 420 ps for
LiCl droplet, which are all far less than 920 ps for the water droplet.
This result agrees with the experimental observation [8] and con-
firms that the salt type has significant effect on the solution
spinnability.
The radial distribution functions of g(r)cation-O for LiCl, NaCl, and
KCl aqueous droplets at evaporation instants of 0, 100 and 200 ps
are shown in Fig. 11. The peak values of g(r)cation-O for LiCl droplet
are the largest, and then followed by NaCl and KCl droplets. Thus,
LiCl droplet has the strongest hydration effect, since more water
molecules are bound around lithium ions. The peak position of
gðrÞLiþ O , gðrÞNaþ O , and gðrÞKþ O is 1.95, 2.35, and 2.85 Å, respec-
tively, so the strongest attractive force occurs between water mol-
ecules and lithium ions. The two factors are responsible for the
slowest evaporation of LiCl droplet among the three aqueous
droplets.
3.3. The effect of salt concentration
The temporal evolution of the water molecule number in the
aqueous droplets with dissolved 20, 30, 40, 80, 100, and 120 LiCl
molecules at E = 0.8 V nm1 is shown Fig. 12. As the number of LiCl
molecules increases, droplet evaporation is first enhanced and then
reduced, the optimal number is 30 which corresponds to 0.0588
LiCl mass fraction. The optimal salt concentration was also ob-
served in Kim et al.’s experiment [10]. Fig. 13 shows the average ki-
netic energy of lithium ions, chloride ions, and water molecules in
the aqueous droplets with different LiCl concentrations at
0 50 100 150 200 250 300
t (ps)
Fig. 12. Temporal evolution of the water molecule number in the aqueous droplet
with various salt concentrations at E = 0.8 V nm1.
E = 0.8 V nm1. With high LiCl concentration, the attractive force
between lithium ions and chloride ions increases, which sup-
presses the acceleration effect of the electric field on ions
(Fig. 13a and b), and hence reduces the average kinetic energy of
water molecules. However, as LiCl concentration increases, more
water molecules are driven by lithium and chloride ions, which
has positive effect to accelerating water molecules. Therefore, the
average kinetic energy of water molecules is determined by these
two competitive factors, and its maximum value occurs when 30
LiCl molecules are added into the droplet (Fig. 13c). As a result,
the aqueous droplet with dissolved 30 LiCl molecules has the fast-
est evaporation rate.
4. Conclusion
The effect of droplet evaporation dynamics on electrospinning
has not been understood very well. This work comparatively stud-
ies the evaporations of the water droplet and the aqueous droplet
with dissolved NaCl, KCl or LiCl molecules in the absence and pres-
ence of a homogeneous electric field by molecular dynamics simu-
lations. The main conclusions are as follows:
(1) Applying an electric field enhances the droplet evaporation,
and the enhancement is more significant for the aqueous
droplet than for the water droplet. The enhancement mech-
anism can be attributed to the following two reasons. First,
the spherical-shape droplet is stretched to be a long cylinder,
increasing the heat transfer area between the droplet and
 Author's personal copy

540 B.-B. Wang et al. / International Journal of Heat and Mass Transfer 73 (2014) 533–541

(4) LiCl aqueous droplet has the strongest hydration effect and
16 (a) Aqueous droplet
the strongest attractive force between water molecules
with 20 LiCl
14 with 30 LiCl and lithium ions, and hence has the slowest evaporation rate
with 40 LiCl compared to NaCl and KCl droplets.
12 with 80 LiCl (5) As the number of LiCl molecules increases, the droplet evap-
with 120 LiCl oration is first enhanced and then reduced, indicating that
Ek, Li+×10 (J)

10 there is an optimal LiCl concentration, this is because

17

increasing LiCl concentration has both positive and negative

8
effects on the enhancement of the average kinetic energy of
6 water molecules.

2 Acknowledgments
0 20 40 60 80 100 120 140 160 180

t (ps) This study was partially supported by the National Natural Sci-
15 ence Foundation of China (Nos. 51276060 and 51076078), by the
(b) Aqueous droplet 111 Project (No. B12034), by Program for New Century Excellent
14
with 20 LiCl Talents in University (No. NCET-11-0635).
13 with 30 LiCl
with 40 LiCl
12 References
with 80 LiCl
11 with 120 LiCl
×10 (J)

[1] J.S. Choi, S.W. Lee, L. Jeong, S.H. Bae, B.C. Min, J.H. Youk, W.H. Park, Effect of
17

10 organosoluble salts on the nanofibrous structure of electrospun poly (3-

hydroxybutyrate-co-3-hydroxy-Valerate), Int. J. Biol. Macromol. 34 (2004)
-
k, Cl

9
249–256.
E

8 [2] K.P. Matabola, A.R. de Vries, A.S. Luyt, R. Kumar, Studies on single polymer
composites of poly(methylmethacrylate) reinforced with electrospun
7 nanofibers with a focus on their dynamic mechanical properties, Express
Polym. Lett. 5 (2011) 635–642.
6 [3] Y.M. Shin, M.M. Hohman, M.P. Brenner, G.C. Rutledge, Experimental
5 characterization of electrospinning: the electrically forced jet and
0 20 40 60 80 100 120 140 160 180 instabilities, Polymer 42 (2001) 9955–9967.
[4] S.A. Theron, E. Zussman, A.L. Yarin, Experimental investigation of the
t (ps) governing parameters in the electrospinning of polymer solutions, Polymer
45 (2004) 2017–2030.
[5] V.E. Kalayci, P.K. Patra, Y.K. Kim, S.C. Ugbolue, S.B. Warner, Charge
13 (c) Aqueous droplet consequences in electrospun polyacrylonitrile(PAN) nanofibers, Polymer 46
with 20 LiCl (2005) 7191–7200.
12 with 30 LiCl [6] D.H. Reneker, I. Chun, Nanometre diameter fibres of polymer, produced by
with 40 LiCl electrospinning, Nanotechnology 7 (1996) 216–223.
11 [7] X.H. Qin, Y.Q. Wan, J.H. He, J. Zhang, J.Y. Yu, S.Y. Wang, Effect of LiCl on
with 80 LiCl electrospinning of PAN polymer solution: theoretical analysis and
Ek, water×10 (J)

with 120 LiCl experimental verification, Polymer 45 (2004) 6409–6413.

10
17

[8] X.H. Zong, K. Kim, D.F. Fang, S.F. Ran, B.S. Hsiao, B. Chu, Structure and process
relationship of electrospun bioabsorbable nanofiber membranes, Polymer 43
9
(2002) 4403–4412.
[9] X.H. Qin, S.Y. Wang, Interior structure of polyacrylonitrile (PAN) nanofibers
8
with LiCl, Mater. Lett. 62 (2008) 1325–1327.
[10] B. Kim, H. Park, S.H. Lee, W.M. Sigmund, Poly(acrylix acid) nanofibers by
7 eletrospinning, Mater. Lett. 59 (2005) 829–832.
[11] W.D. Luedtke, U Landman, Y.H. Chiu, D.J. Levandier, R.A. Dressler, S. Sok, M.S.
6 Gordon, Nanojets, electrospray, and ion field evaporation: molecular dynamics
simulations and laboratory experiments, J. Phys. Chem. A 112 (2008) 9628–
0 20 40 60 80 100 120 140 160 180 9649.
t (ps) [12] J.K. Chung, S. Consta, Release mechanisms of poly(ethylene glycol) macroions
from aqueous charged nanodroplets, J. Phys. Chem. B 116 (2012) 5777–5785.
[13] I.L. Kanev, N.K. Balabaev, A.V. Glyakina, V.N. Morozov, Computer simulation of
Fig. 13. Average kinetic energy at different evaporation instants for aqueous
gas-phase neutralization of electrospray-generated protein macroions, J. Phys.
droplets with different LiCl concentrations at E = 0.8 V nm1: (a) Li+; (b) Cl; (c)
Chem. B 116 (2012) 5872–5881.
water molecule. [14] D.D. Christopher, M.C. Natalie, How are completely desolvated ions produced
in electrospray ionization: insights from molecular dynamics simulations,
Anal. Chem. 83 (2011) 8372–8376.
[15] V. Znamenskiy, I. Marginean, A. Vertes, Solvated ion evaporation from charged
water nanodroplets, J. Phys. Chem. A 107 (2003) 7406–7412.
surrounding nitrogen. Second, the electric field induces the [16] Y. Okuno, M. Minagawa, H. Matsumoto, A. Tanioka, Simulation study on the
directional arrangement of water molecules, weakening influence of an electric field on water evaporation, J. Mol. Struct. – Theochem.
the interactions between water molecules. 904 (2009) 83–90.
[17] J.S. Kim, D.H. Reneker, Polybenzimidazole nanofiber produced by
(2) Due to stronger hydration effect, water molecules are bound electrospinning, Polym. Eng. Sci. 39 (1999) 849–854.
around ions, adding ions into the droplet cannot always [18] I.D. Norris, M.M. Shaker, F.K. Ko, A.G. Synthetic, Electrostatic fabrication of
enhance the droplet evaporation for the electric field lower ultrafine conducting fibers: polyaniline/polyethylene oxide blends, Synth. Met.
114 (2000) 109–114.
than the threshold value of electrospinning.
[19] J.S. Kim, D.S. Lee, Thermal properties of electrospun polyesters, Polym. J. 32
(3) At high electric field, the average kinetic energy of water (2000). 616-613.
molecules is increased significantly due to acceleration of [20] J.M. Deitzel, W. Kosik, S.H. McKnight, T.N.C. Beck, J.M. DeSimone, S. Crette,
ions driven by electric field, so that usage of the aqueous Electrospinning of polymer nanofibers with specific surface chemistry,
Polymer 43 (2002) 1025–1029.
droplet has distinct advantage over the water droplet for [21] S. Sumardiono, J. Fischer, Molecular simulations of droplet evaporation by heat
electrospinning. transfer, Microfluid. Nanofluid. 3 (2007) 127–140.
 Author's personal copy
B.-B. Wang et al. / International Journal of Heat and Mass Transfer 73 (2014) 533–541
[22] E.S. Landry, S. Mikkilineni, M. Paharia, A.J.H. McGaughey, Droplet evaporation:
a molecular dynamics investigation, J. Appl. Phys. 102 (2007) 124301.
[23] L.N. Long, M.M. Micci, B.C. Wong, Molecular dynamics investigation of droplet
evaporation, Comput. Phys. Commun. 96 (1996) 167–172.
[24] P.E. Mason, Molecular dynamics study on the microscopic details of the
evaporation of water, J. Phys. Chem. A 115 (2011) 6054–6058.
[25] L. Consolini, S.K. Aggarwal, S. Murad, A molecular dynamics simulation of
droplet evaporation, Int. J. Heat Mass Transfer 46 (2003) 3179–3188.
[26] C. Si, X.D. Wang, M. Chen, Y.Y. Duan, Effect of wall adsorption on the nano-
droplet evaporation in a nano-channel: a molecular dynamics investigation,
Colloids Surf. A: Physicochem. Eng. Aspects 436 (2013) 450–458.
[27] A.P. Bhansali, Y. Bayazitoglu, S. Maruyama, J.K. Little, Molecular dynamics
simulation of an evaporating sodium droplet, Int. J. Therm. Sci. 38 (1999) 66–
74.
[28] B.B. Wang, X.D. Wang, M. Chen, J.L. Xu, Molecular dynamic simulations on
evaporation of droplets with dissolved salts, Entropy 15 (2013) 1232–1246.
[29] Z.J. Wang, M. Chen, Z.Y. Guo, C. Yang, Molecular dynamics study on the liquid
vapor interfacial profiles, Fluid Phase Equilib. 183 (2001) 321–329.
[30] V.P. Carey, A.P. Wemhoff, Disjoining pressure effects in ultra-thin liquid films
in micropassages-comparison of thermodynamic theory with predictions of
molecular dynamics simulations, J. Heat Transfer 128 (2006) 1276–1284.
[31] J.P. Yu, H. Wang, A molecular dynamics investigation on evaporation of thin
liquid films, Int. J. Heat Mass Transfer 55 (2012) 1218–1225.
[32] M.Z. Steinberg, K. Breuker, R. Elber, R.B. Gerber, The dynamics of water
evaporation from partially solvated cytochrome c in the gas phase, Phys.
Chem. Chem. Phys. 9 (2007) 4690–4697.
[33] P. Mark, L. Nilsson, Structure and dynamics of the TIP3P, SPC, and SPC/E water
models at 298 K, J. Phys. Chem. A 105 (2001) 9954–9960.
[34] P.T. Kiss, A. Baranyai, Sources of the deficiencies in the popular SPC/E and
TIP3P models of water, J. Chem. Phys. 134 (2011) 054106.
[35] C. Vega, J.L.F. Abascal, Simulating water with rigid non-polarizable models: a
general perspective, Phys. Chem. Chem. Phys. 13 (2011) 19663–19688.
[36] H.J.C. Berendsen, J.R. Grigera, T.P. Straatsma, The missing term in effective pair
potential, J. Phys. Chem. 91 (1987) 6269–6271.
View publication stats
541
[37] S. Chowdhuri, A. Chandra, Molecular dynamics simulations of aqueous NaCl
and KCl solutions: effects of ion concentration on the single-particle, pair, and
collective dynamical properties of ions and water molecules, J. Chem. Phys.
115 (2001) 3732–3741.
[38] F. Kohler, N. Quirke, J.W. Perram, Perturbation theory with a hard dumbbell
reference system: I. Application to liquid nitrogen, J. Chem. Phys. 71 (1979)
4128–4131.
[39] S. Chowdhuri, A. Chandra, Hydration structure and diffusion of ions in
supercooled water: ion size effects, J. Chem. Phys. 118 (2003) 9719–9725.
[40] M. Mecke, J. Fischer, J. Winkelmann, Molecular dynamics simulation of the
liquid–vapor interface of dipolar fluids under different electrostatic boundary
conditions, J. Chem. Phys. 114 (2001) 5842–5852.
[41] J. Lekner, Simulation of Coulomb fields in computer-simulated disordered
systems, Phys. A 176 (1991) 485–498.
[42] S. Lishchuk, Lekner summation of dipolar interaction in quasi-two-
dimensional simulation reference number s, Mol. Phys. 100 (2002) 3789–
3793.
[43] J.V.L. Beckers, C.P. Lowe, S.W. de Leeuw, An iterative PPPM method for
simulating coulombic systems on distributed memory parallel computers,
Mol. Simul. 20 (1998) 369–383.
[44] C.D. Danb, D. Bratko, A. Luzar, Electric control of wetting by salty nanodrops:
molecular dynamics simulations, J. Phys. Chem. C 115 (2011) 22393–22399.
[45] J.P. Ryckaert, G. Ciccotti, H.J.C. Berendsen, Numerical integration of the
cartesian equations of motion of a system with constraints: molecular
dynamics of n-Alkanes, J. Comput. Phys. 23 (1977) 327–341.
[46] W.C. Swope, H.C. Andersen, P.H. Berens, K.R. Wilson, A computer simulation
method for the calculation of equilibrium constants for the formation of
physical clusters of molecules: application to small water clusters, J. Chem.
Phys. 76 (1982) 637–649.
[47] S. Vaitheeswaran, H. Yin, J.C. Rasaiah, Water between plates in the presence of
an electric field in an open system, J. Phys. Chem. 109 (2005) 6629–6635.
[48] C. Caleman, D.V.D. Spoel, Evaporation from water clusters containing singly
charged ions, Phys. Chem. Chem. Phys. 9 (2007) 5105–5111.