Soft X-ray charger (SXC) system for use with electrospray for mobility measurement of bioaerosols

Abstract

Electrospray size spectrometry analysis (ESSA), a mobility-based aerosol technique to characterize the size and charge-state of nanoparticles, uses a neutralizer to bring multiply charged particles to an equilibrium charge-state. Typical ESSA setups incorporate a Po-210 neutralizer that has short half-life time and its radioactive nature leads to safety concerns. This work demonstrates that a soft X-ray charger (SXC) effectively neutralizes electrosprayed bioparticles. Calculations show that the SXC generates bipolar ions with number concentrations one and two orders of magnitude higher than conventional Po-210 and Kr-85 sources, respectively. Additionally, we established that an SXC may be safely implemented in ESSA.

Full text

Page 1

Soft X-ray charger (SXC) system for use with electrospray for mobility
measurement of bioaerosols
Luis B. Modesto-Lopez1, Eric M. Kettleson, Pratim Biswas*
Aerosol and Air Quality Research Laboratory, Department of Energy, Environmental and Chemical Engineering, Campus Box 1180, Washington University in St. Louis,
Saint Louis, MO 63130, USA
a r t i c l e i n f o
Article history:
Received 24 March 2010
Received in revised form
21 April 2011
Accepted 29 April 2011
Available online 20 May 2011
Keywords:
Electrospray
Size spectrometry analysis
Photoionization
Soft X-ray charger
Viruses
Proteins
a b s t r a c t
Electrospray size spectrometry analysis (ESSA), a mobility-based aerosol technique to characterize the
size and charge-state of nanoparticles, uses a neutralizer to bring multiply charged particles to an
equilibrium charge-state. Typical ESSA setups incorporate a Po-210 neutralizer that has short half-life
time and its radioactive nature leads to safety concerns. This work demonstrates that a soft X-ray
charger (SXC) effectively neutralizes electrosprayed bioparticles. Calculations show that the SXC gener-
ates bipolar ions with number concentrations one and two orders of magnitude higher than conven-
tional Po-210 and Kr-85 sources, respectively. Additionally, we established that an SXC may be safely
implemented in ESSA.
? 2011 Elsevier B.V. All rights reserved.
1. Introduction
With the emergence of nano-bio hybrid systems constructed
with novel nanometer-sized biological particles such as viruses,
proteins, or natural complexes [1,2], it is essential to develop safer
and practical measurement techniques without complex sample
preparation procedures and that allow for a precise online char-
acterization based on particle size, charge, and shape. These
parameters influence the particle deposition mechanism (diffusion,
impaction, or sedimentation) and therefore also influence the
deposition location of particles in the human respiratory tract [3].
Electrospray size spectrometry analysis (ESSA) is an aerosol tech-
nique that provides information of the particle size and the charge-
state of various particle types (i.e., solid particles and biological
particles) in the submicrometer and nanometer-size ranges [4e8].
Its relative simplicity and the fact that the analysis is carried out
online and in real-time make ESSA an attractive method for online
identification of proteins and viruses [1,4e7]. Several studies have
demonstrated the applicability of ESSA to tasks requiring benign
handling and transport of biological or otherwise fragile particles
such as identifying specific proteins within a protein mixture
[4,5,7], characterizing giga- and mega-dalton complexes [6],
studying the size and morphology of carbon nanotubes [9], and
measuring the size of light-harvesting chlorosomes [1]. In ESSA, an
electrospray generates highly charged, monodispersed particles,
which after a charge-reduction step are characterized in terms of
their electrical mobility. Because the particle electrical mobility is
a function of both, the particle size and the particle charge, singly-
charged particles are desired to simplify their size calculation with
a data reduction algorithm [10]. In ESSA the particle charge-
reduction step is carried out by exposing the electrosprayed
particles to a bipolar ion environment where they reach an equi-
librium charge-state. Theoretically, in this equilibrium charge-state
known as theBoltzmann equilibrium
a majority of the particles carry zero or one (negative or positive)
charge. To produce the neutralizing bipolar ion environment,
a typical ESSA setup uses a Po-210 diffusion neutralizer, which is an
a-ray emitter (i.e., radioactive). Note that bipolar ion emitters are
used as neutralizers when applied to charged particles and as
chargers when applied to neutral particles; in both cases the
mechanism involves collisions of ions with the particles being
analyzed. While Po-210 has been demonstrated to be effective in
neutralizingelectrosprayed particles, there are threemainconcerns
about its applicability. One is a safety concern, because of its
chemical nature, Po-210 sources need special handling to avoid
radioactive exposure and contamination (requiring significant
chargedistribution,
* Corresponding author. Tel.: þ1 314 935 5482; fax: þ1 314 935 5464.
E-mail address: pbiswas@wustl.edu (P. Biswas).
1Current address: Droplet Interfaces and Flows Research Group, Department of
Chemical Engineering, Universitat Rovira i Virgili, Av. Països Catalans, 26, Tarragona
43007, Spain.
Contents lists available at ScienceDirect
Journal of Electrostatics
journal homepage: www.elsevier.com/locate/elstat
0304-3886/$ e see front matter ? 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.elstat.2011.04.014
Journal of Electrostatics 69 (2011) 357e364

Page 2

protocols for acquiring the source). The second issue is its short
half-life time (120 days) that requires frequent replacement. An
alternative to solve the short half-life time constraint of Po-210 is to
use a Kr-85 source, a b-ray emitter, which has a half-life of 10.8
years. However, previous studies demonstrated that the bipolar ion
generation rate of a Kr-85 charger is one order of magnitude
smaller than the Po-210 [11], which significantly affects the accu-
racy of the ESSA. Moreover, with the Kr-85 charger the particles’
attainment of the equilibrium charge-state depends on their resi-
dence time in the charger (i.e., the carrier gas flow rate) and such
charge-state is achieved only at longer residence times [11], as
compared to a Po-210 source. The third concern is particle losses by
deposition onto system walls due to ineffective reduction of the
charge, particularly when the particle is in the nanometer-size
range. This charge-reduction inefficiency may arise from two main
factors. First, the number of charges on electrosprayed particles
scales with the square of particle diameter and thus as the particle
size increases a higher bipolar ion concentration is needed to
reduce the particle charge-state. Second, the probability of ion
collisionwith a particle decreases with decreasing particle size; this
probability is particularlylowwhen the gas mean free path is larger
than the size of the particles. More specifically, when neutral
particles are exposed to a bipolar ion environment the fraction of
particles that acquire a single charge by diffusion is much smaller
than the fraction of particles carrying no charge. Similarly, for
initially charged nanoparticles, the opposite phenomenon occurs
and only a small fraction of them carry no charge.
A potential alternative to the Po-210 diffusion neutralizers is
photoionization with UV radiation or with soft X-rays [12,13]. The
photoionization process generates a sufficiently high, symmetrical
ion number concentration and it is more controllable (by turning
the generating device on and off) that the common radioactive
sources. Although both, UV radiation and soft X-rays, generate
bipolar ions with symmetrical ion number concentration, the
present work focuses on the applicability a soft X-ray charger (SXC)
device to reduce the charge-state of electrosprayed particles. An
SXC has several practical advantages over conventional diffusion
neutralizers: (1) direct photoionization of carrier gas molecules and
of aerosol particles (by photoemission of a surface electron) occurs
simultaneously resulting in the generation of a large number of
positive and negative ions with nearlyequal number concentration,
this in turn may enhance the particle neutralization efficiency by
diffusion [13e15]; (2) the SXC generates bipolar ions at a rate one
orderof magnitude higher thana-rayemitters (i.e., Po-210); and (3)
the ion generation is readily controlled by turning the device on
and off, thus it does not require both special handling and frequent
replacement [13]. Jiang et al. [14,16] demonstrated in a nanoparticle
charging model, accounting for simultaneous photoionization and
diffusion charging, that the photoionization yield coefficient, the
parameter that governs photoionization charging, depends only on
particle size, while the ion attachment coefficient, which governs
the diffusion charging mechanism, depends on both the particle
size and the initial charge-state. Thus, photoionization may be
more effective to neutralize multiply charged particles of a given
size.
Previous SXC studies, however, focused on using photoioniza-
tion to charge particles and collect them using electric fields
[15,17,18] or to investigate the achievement of the equilibrium
charge-state of aerosol particles with low charge-state [13,19,20].
To the best of our knowledge, photoionization with an SXC has
never been applied to neutralize (i.e., reduce the charge-state)
highly charged particles generated by an electrospray, where the
number of charges carried by particles scales with the square of
particle size [21]. In the present study, the application of an SXC to
ESSA of polystyrene latex (PSL) particles, bacteriophage MS2, and
ovalbumin (OVA) protein is reported. PSL particles are routinely
used as standards for size calibration in various instruments. MS2 is
commonly used as a surrogate of biothreat viruses [22], its nucleic
acid sequence and crystal structure are well-characterized, and it
has a size in the range of 24e27 nm, depending on the character-
ization technique [6,22e25]. Ovalbumin is a protein with a size of
approximately 6 nm, it has been studied by high-performance
liquid chromatography with inductively coupled plasma mass
spectrometric detection [26], and has been employed in animal
inhalation studies. Moreover, PSL, MS2, and OVA particles have
been characterized using the electrospray and electrical mobility-
based techniques, thus they are suitable to demonstrate the func-
tioning of an SXC in ESSA.
2. Experimental section
2.1. Materials
Standard polystyrene latex (PSL) particles of 60 nm in diameter
were purchased from Duke Scientific and diluted in deionized
water to a final particle concentration of 13 vol.%. Bacteriophage
MS2 (ATCC 15597-B1) was propagated as described previously [27].
Following propagation, phage suspensions were concentrated and
purified using an Amicon Centriplus YM-10 filtration device (Mil-
lipore Corp., Bedford, MA). Suspensions were centrifuged at 3000g
for 8 min, and then washed in 20 mM ammonium acetate to dilute
out impurities. This process was repeated three times to sufficiently
concentrate and purify the phage suspensions. An ovalbumin (OVA,
SigmaeAldrich) solution of 0.005 wt.% was prepared in 20 mM
aqueous ammonium acetate.
2.2. Diffusion neutralizers
Aerosol diffusion neutralizers typically consist of a radioactive
source that emits a or b particles that, upon collision with the
molecules of the carrier gas, generate a bipolar ion environment.
The bipolar ions then collide with the aerosol particles, resulting in,
depending on their polarity, an increase or a reduction of the
number of charges on the particles. In the present study, Po-210 (a-
ray emitter, 0.5 mCi, TSI, model 348002) and Kr-85 (b-ray emitter,
2 mCi, TSI, model 3077) sources were used as diffusion neutralizers.
Po-210 is the most common bipolar ion source employed in ESSA
[4e6,24]. The Po-210 sourcehas a shape of a tablet with diameterof
3.2 cm and thickness of 0.97 cm; on one side its has a circular area
(active area) covered by a mesh, through which a-rays are emitted,
with diameter of 1.78 cm. In the longer, cylindrical b-ray emitter
neutralizer, inert Kr-85 gas is sealed completely inside a stainless-
steel tube that has a wall thickness of 0.05 mm. The Kr-85 gas
containing-tube is shielded by a metal outer housing (also made of
stainless-steel, inner diameter ¼ 3.89 cm) with inlet and outlet
diameters of 0.64 cm. The aerosol passes through the space
between the Kr-85 gas containing-tube and the outer housing [28].
The Kr-85 neutralizer’s maximum operating flow rate is 5 L min?1.
The Po-210 and Kr-85 sources used in this study were 4-months-
old and 9-years-old, respectively.
2.3. Photoionizer
The SXC device (Hamamatsu Photonics Ltd., Japan, Model L7113)
used in this study operates with an input potential of 12 V and
generates X-rays with an intensity of 3.5e9.5 keV and wavelength
of 0.12e0.41 nm, from a circular beryllium window of approxi-
mately 1.5 cm in diameter. A detailed description of the SXC can be
found in previous publications from our group and other authors
[13,15,17,20].
L.B. Modesto-Lopez et al. / Journal of Electrostatics 69 (2011) 357e364
358

Page 3

2.4. Characterization of particle size
The experimental plan is summarized in Table 1. In a first set of
experiments, a commercial electrospray aerosol generator (EAG,
Model 3480, TSI), with a capillary inner diameter of 40 mm, was
operated in the cone-jet mode toaerosolize aqueous suspensions of
PSL particles, bacteriophage MS2, and an OVA solution as depicted
in Fig. 1(a). In the EAG a standard centrifuge vial containing
a sample solution is placed inside a cylindrical pressure vessel. The
vessel accommodates a capillary and a high-voltage platinum wire,
both of which are immersed in the solution. Maintaining a differ-
ential pressure moves the solution through the capillary. The liquid
flow rate through the capillary was kept constant at 0.2 mL min?1.
An electric field induces a charge on the solution at the capillary tip
and draws the liquid from the tip into a conical jet from which
ultrafine charged droplets are emitted. The highly charged droplets
are then mixed with a sheath gas flow of air and CO2that carries
them from the electrospray zone to a neutralization chamber. The
total sheath gas flow rate used in this work was 1.1 L min?1
(1.0 L min?1air þ 0.1 L min?1CO2). CO2in the sheath gas flow is
Table 1
Experimental plan to compare the performance of an SXC with that of diffusion (radioactive) neutralizers.
TestElectrospray
system
Capillary
inner diameter
[mm]
40
ChargerNeutralization
chamber
volume [m3]
2.8 ? 10?5
Objective
1
2
EAGPo-210
SXC
To compare the efficacy of an SXC with the commonly used Po-210
charger in electrospray size spectrometry, using a well-characterized commercial setup.
The electrospray generation part and the neutralization chamber are a built into a sole piece.
To compare the efficacy of an SXC with that of a Kr-85 charger in electrospray
size spectrometry. The capillary diameter is larger than in Tests 1 & 2 because the geometry
of the entire setup is different (i.e., the Kr-85 neutralization chamber is cylindrical
and not built-it with the electrospray source, see the text for details).
3
4
In-house 160Kr-85
SXC
10 ? 10?5
1.3 ? 10?5
Fig. 1. Experimental setup for size spectrometry (a) commercial electrospray setup (EAG, Tests 1 and 2) and (b) in-house electrospray setup (Tests 3 and 4).
L.B. Modesto-Lopez et al. / Journal of Electrostatics 69 (2011) 357e364
359

Page 4

required to avoid breakdown in the gas surrounding the spray and
to suppress formation of corona discharge, thereby obtaining
a stable spray. The neutralization chamberhas a circular windowon
one side where a Po-210 neutralizer of 3.2 cm in diameter (active
area diameter ¼ 1.78 cm) is inserted. Following charge neutraliza-
tion, aerosol particles were sent to a scanning mobility particle
sizer (SMPS) system to measure their particle size distribution. The
SMPS is composed of an electrostatic classifier, a differential
mobility analyzer (DMA, TSI, Models 3081 and 3085), and an
ultrafine condensation particle counter (CPC, TSI, 3025A). The 3081
DMA was used for measurements of PSL and MS2 and the 3085
DMAwas used for measurements of OVA. Briefly, the functioning of
the DMA is depicted in Fig. 1(a); singly-charged particles are
introduced from the top of the instrument at avolumetric flow rate,
Qa, the voltage in the central electrode of the DMA increases
gradually, with each voltage value corresponding to a certain
particle electrical mobility. As particles move down, only particles
with the right electrical mobility, at a given voltage, pass through
the slit at the bottom of the DMA and are sent to the CPC to be
counted applying optical methods. The rest of the particles are
carried away by the sheath flow, Qs. A more detailed description of
the DMA and the electrical mobility theory can be found elsewhere
[10]. Note that the electrostatic classifier in the SMPS has a Kr-85
neutralizer built-in; however, in all the experiments it was
bypassed to evaluate only the performance of the bipolar ion
generators upstream in the neutralization chamber. Bypass of the
Kr-85 in the SMPS did not have any significant effect in the particle
size distribution. Initially, experiments were carried out using the
Po-210 source incorporated in the EAG (Test 1), then the Po-210
source was removed and replaced with the SXC device, which
was then attached to the chamber window, and the same set of
experiments were conducted (Test 2). In the case of the SXC, a thin
polyamide film (Kapton? 30HN, DuPont Corp., 30 mm thick) was
used to seal the circular window where the Po-210 source is usually
inserted to protect the X-ray emitter surface from the aerosol flow
and to keep the neutralization chamber airtight [15,18].
In a third set of ancillary experiments, measurements were
carried out in an in-house electrospray system with the Kr-85
source as shown in Fig. 1(b) (Test 3). Although the performance of
Kr-85 sources to reduce the charge-state of electrosprayed particles
is not as efficient as other bipolar ion sources such as Po-210, these
experiments were aimed at providing additional data on the use of
the SXC in ESSA. For the Kr-85 experiments, the suspensions were
fed with a syringe pump, at a flow rate of 0.5 mL min?1, through
a tapered stainless-steel capillary needle of 160 mm inner diameter.
The use of a capillary of larger inner diameter, than in the EAG
(Tests 1 and 2), was constrained by the geometrical dimensions of
the cylindrical Kr-85 neutralizer (i.e., the Kr-85 and the Po-210
neutralizer chambers had different external shapes and dimen-
sions, and both are commercial products). The capillary was set
inside a closed chamber through which particle-free CO2carrier gas
was sent at a rate of 0.5 L min?1(see Fig. 1(b)). The highly charged
particles were then passed through the cylindrical neutralization
chamber containing the Kr-85 source. Subsequently, charge-
reduced electrosprayed particles were sent to the SMPS system.
In the last set of experiments (Test 4) the Kr-85 charger in the in-
house electrospray system was replaced with a soft X-ray neutral-
ization chamber as shown in Fig.1(b) and the same measurements
were carried out.
3. Results and discussion
The efficacy of a bipolar neutralizer in reducing the number of
charges on an electrosprayed particle depends on the ion number
concentration generated by the neutralizer, the particle residence
time in the neutralization chamber, and the particle size
[11e15,19,29]. Table 2 summarizes the ion number concentration
[13,15] generated by each charger for the specific neutralization
chamber volume calculated with:
ni¼
?
Is
aeV
?1=2
(1)
In expression (1), Isis the saturation current, a is the recombi-
nation constant for bipolar ions (1.6 ? 10?12m3s?1), e is the
electron elementary charge (1.6 ? 10?19C), and V is the volume of
the neutralization chamber. Note that ion concentrations in Table 2
are ideal values since the calculations do not account for wall
deposition losses. The maximum energy of a and b particles is
5.3 MeV and 0.695 MeV, respectively [11], three orders of magni-
tude higher than the photon energy of the soft X-rays, 9.5 keV.
However, the bipolar ion number concentration generated with the
SXC is one and two orders of magnitude higher than the Po-210 and
the Kr-85 neutralizers, respectively. Similar results were reported
in charging studies comparing an SXC and an a-ray emitter (Am-
241) source byShimada et al. [13]. A high ion numberconcentration
is desired to increase the ion-particle collision probability and to
effectively bring the particle charge-state to the Boltzmann charge
distribution over the entire size range because: (1) the number of
charges carried by electrosprayed particles is known to scale with
the square of the particle size [21], thus even ultrafine particles
carry multiple charges, and (2) the probability of a particle colliding
with an ion decreases with decreasing particle size. Furthermore,
Liu et al. [29] reported that the saturation current (i.e., the ion
number concentration) of a Po-210 neutralizer decreases as the
source becomes older. For instance, the Isof a 4-month-old Po-210
neutralizer was approximately 15 nA and it dropped to 7 nA and
0.7 nA after 8 and 24 months,respectively[29]. Thus, the values of Is
in Table 2 represent the best-case scenario for the Po-210
neutralizer in this study. Conversely, the SXC is known to have
uniform saturation current during its lifetime and to generate
bipolar ions with nearly the same electrical mobility [13]. After
comparing the ion number concentration generated by each
neutralizer, experiments were carried out to verify the actual
performance of the neutralizers with electrosprayed particles of
different characteristics. The charge neutralization mechanism of
electrosprayed particles using photoionization is depicted in Fig. 2.
Nanoparticles, entering the neutralization chamber, were multiply
charged as a result of the electrospray and the average number of
elementary charges, q, carried by each particle is summarized in
Table 3. In the chamber, an SXC generates an equal number of
bipolar ions by three mechanisms, namely, direct photoionization
of particles, photoionization of carrier gas molecules, and photo-
emission of the chamber wall. The bipolar ions then collide
randomly with the highly charged particles leaving mostly zero or
one elementary charge by the so-called diffusion neutralization
mechanism (i.e., the same mechanism as diffusion charging) that
results from the Brownian motion of the ions and particles.
Table 2
Calculated ion number concentration for each bipolar ion generator.
ChargerSaturation
current, Is[A]a
1.5 ? 10?7
Ion concentration,
ni[m?3]
1.4 ? 1014
2.1 ? 1014
5.3 ? 1013
7.6 ? 1012
Ion production
rate, S [m?3s?1]
3.35 ? 1016
7.23 ? 1016
4.5 ? 1015
9.4 ? 1013
Soft X-ray (EAG)
Soft X-ray (In-house)
Po-210 (EAG)
Kr-85 (In-house)
2.0 ? 10?8
1.5 ? 10?9
aThe Isof the SXC was provided by manufacturer, and for the diffusion chargers
the value was taken from Liu et al. [29].
L.B. Modesto-Lopez et al. / Journal of Electrostatics 69 (2011) 357e364
360

Page 5

3.1. SXC and Po-210 source
Particle size distribution measurements with the EAG are shown
in Fig. 3 for the SXC and the Po-210 source. Note that for both
neutralizers (Tests 1 and 2) the neutralization chamber volume and
the particle residence time were constant; however, the ion
number concentration was higher with the SXC (see Table 2).
Electrospray of 60 nm PSL particles neutralized with the SXC and
with Po-210 resulted in a peak mobility diameter of 63 nm in both
cases (Fig. 3(a)). Nevertheless, the particle number concentration at
63 nm had a two-fold increase with the SXC compared to the Po-
210 neutralizer, and because the size distribution was unimodal,
it can be inferred that the 60 nm PSL particles selected with the
DMA were singly charged. The particle size distribution of MS2
phages, shown in Fig. 3(b), with both the SXC and the Po-210
neutralizer resulted in a narrow mode diameter at 24.1 nm, in
excellent agreement with previous size spectrometry reports
[6,25,30]. Such a value is slightly smaller than the reported size of
27.5 nm using X-ray crystallography [23]; however, the discrepancy
was attributed to the MS2 phage encapsulating water molecules in
the X-ray crystallographic studies and undergoing compression in
the electrosprayed droplet [6,23]. The narrow peak showed that
a major fraction of the MS2 phages carried þ1 charge. Fig. 3(c)
shows the particle size distribution of electrosprayed OVA particles,
after neutralization with the SXC and with the Po-210 source. The
particle size distribution shows three peaks at 6.38 nm, 7.91 nm,
and 8.82 nm that correspond to monomers, dimers, and trimers,
respectively carrying þ1 charge, in very good agreement with
previous reports [7].
These results indicate that photoionization with the SXC was
sufficient to bring a larger fraction of nanoparticles to the Boltz-
mann charge-state (i.e., mostly zero or singly-charged particles),
compared to particles neutralized with the Po-210 source, inde-
pendently of the particles’ initial charge-state. It also means that
photoionization with the SXC reduced the particle loss by deposi-
tion onto system walls as can be observed in Figs. 3 and 4. This fact
was verified by measuring the total particle number concentration
at the outlet of the commercial electrospray system, using
a condensation particle counter (CPC), and the results are
summarized in Table 3. The difference in total particle number
concentration between the SXC and the Po-210 neutralizer was
possibly the result of a larger fraction of particles carrying zero or
one elementary charge with the SXC. Conversely, multiply charged
particles exiting the Po-210 neutralizer would have higher elec-
trical mobility, and consequently deposit at shorter distances
without reaching the DMA slit or the CPC.
The parameters that govern the charging mechanisms by
photoionization and by diffusion are the photoelectric yield coef-
ficient and the ion attachment coefficient, respectively. The first
parameter depends only on particle size, even for very small
particles, because the photon energy of soft X-rays (w7.4 ?10?16J)
is much greater than the work function of most uncharged mate-
rials. On the contrary, the ion attachment coefficient depends on
both the particle size and the initial charge-state of the particle, and
its value decreases as the number of charges on a particle increases
[14]. In the case of highly charged particles, however, the work
function depends not only on the bulk properties of the particles,
which govern the electrochemical potential, but also on the state of
its surface layer that influences the potential difference between
the particles and the gas media. In addition, although the electro-
sprayed particles are positively and highly charged, the photon
energy of the soft X-rays is high enough to eject electrons out of
their surface layer. The maximum charge level a sphere can acquire
Table 3
Elementary charges carried by electrosprayed particles and total number concen-
tration at the output of the electrospray system.
Particle Number of charges
acquired by
electrospray, qa
Measured average total
number concentration
[cm?3]
Particle
lossb[%]
SXC
8.66 ? 104
9.59 ? 104
2.70 ? 105
Po-210
5.03 ? 104
8.37 ? 104
2.08 ? 105
PSL (60 nm)
MS2 (24.1 nm)
OVA (6.38 nm)
674
114
41.9
12.7
23.0
8
aq ¼ 0.23D1:95
bWith respect to the SXC and calculated with 100 ? ([SXC] ? [Po-210])/[SXC].
p
, from Suh et al. [21].
Fig. 2. Mechanism of photoionization of highly charged, electrosprayed particles with an SXC.
L.B. Modesto-Lopez et al. / Journal of Electrostatics 69 (2011) 357e364
361

Page 6

by photoionization before the process reaches a saturation point is
given by Equation (2) [12]:
qmax ¼4p30
Where 30is the permittivity of vacuum (8.85 ? 10?12C2N?1m?2),
hv is the photon energy (7.4 ? 10?16J), FNis the work function of
the particle material, and r is the particle radius. Based on expres-
sion (2), estimates of the maximum number of charges that 6.31-
nm, 24.1-nm, and 60-nm particles acquire by photoionization with
the SXC are 10071, 38465, and 95763 respectively. These values are
much greater than the Gaussian limit of the particles (i.e., the
critical point at which a charged solid particle becomes unstable),
implying that upon charging with direct photoionization, multiply
charged particles with positive polarity become a sink for negative
e2ðhv ? FNÞ r ?3
8
(2)
ions, making it reasonable to assume that these particles reach the
Boltzmann charge-state distribution instantaneously, thereby
resulting in enhancement of the diffusion neutralization rates.
3.2. SXC and Kr-85 source
The SXC and Kr-85 experiments were conducted to provide
results on the efficacy of the SXC as neutralizer. Note that the Kr-85
source is a less efficient neutralizer than the SXC and the Po-210
source because of its strong dependence on gas flow rate [11] and
its low ion production rate. The SXC produces bipolar ions at a rate
three orders of magnitude higher than that of the Kr-85 source (see
Table 2). Furthermore, for these experiments a capillary with
a larger inner diameter (160 mm) was used and a higher volumetric
liquid flow rate was applied, compared to the EAG (see Section 3.1).
Such conditions would lead to a bigger initial droplet (approxi-
mately 2.5 mm, as calculated from the scaling laws of electrospray
[31,32]) and thus facilitate particle agglomeration within the
droplet. In the in-house setup a capillary with a larger diameter
than in the EAG was used because of the geometrical differences
between the Po-210 and Kr-85 sources’ neutralization chamber.
Moreover, the EAG is a commercial product that only allows the use
of certain sizes of capillaries to obtain optimal results.
Fig. 4 shows the particle size distribution of (a) bacteriophage
MS2 and (b) OVA particles electrosprayed with the in-house setup
andneutralizedwithboththeSXCandtheKr-85source.Theparticle
size distribution of MS2 phages shows a mode diameter of 26 nm,
which is in the range of values found in literature for the size of MS2
phages [6,23,24,30]. This value is 2 nm larger than the one obtained
with the EAG in this and previous studies [6,25,30] and closer tothe
Fig. 4. Particle size distribution of (a) bacteriophage MS2 and (b) OVA obtained with
an SXC and a Kr-85 charger in an in-house electrospray setup.
Fig. 3. Particle size distribution of (a) 60 nm PSL particles, (b) bacteriophage MS2, and
(c) OVA obtained with an SXC and a Po-210 charger in a commercial electrospray setup.
L.B. Modesto-Lopez et al. / Journal of Electrostatics 69 (2011) 357e364
362

Page 7

value reported by Golmohammadi et al. [23] using X-ray crystal-
lography. However, when the distribution obtained with the SXC
was deconvoluted with a multiple-peak Gaussian function, three
peaks (data not shown) with mode diameters at 24.8 nm, 30.1 nm,
and38nmwerefound.Mostlikelythe24.8-nmpeakcorrespondsto
single particles of the MS2 phage, while the other two may result
from agglomerates of biological residues. In the profile with the Kr-
85 source, in the inset of Fig. 4(a), a main peak appears at w26 nm
and a second broader peak appears at w36 nm; however, the MS2’s
characteristic 24.1-nm peak was not observed even after deconvo-
lutionwith a Gaussian function. The 26-nm peak is also observed in
the profile of Fig. 3(b) (with the EAG), but in that figure the peak is
hardly noticeable compared with the main24.1-nm peak. It appears
as though the profile with the SXC was different from that of Kr-85;
and such discrepancy arose from the neutralization efficacy of the
SXC and the physical array of the in-house electrospray setup e
particle measurement system. In other words, the high number
concentration of bipolar ions produced with the SXC resulted in
neutralization of particles of all sizes, including agglomerates
formed in the droplets, thus producing a broad particle size distri-
butionand making it difficult todiscern the MS2 peak (24.1 nm). Liu
and Pui [28] showed that degree of aerosol neutralization that can
be achieved by a specific bipolar ion emitter is determined by the
product of the bipolar ion number concentration in the neutrali-
zationchamberandtheresidencetimeinthechamber.Moreover,in
the in-house setup the distance from the electrospray zone to the
neutralization zone of the Kr-85 neutralizer was approximately
11 cm, sufficiently long to have high particle loss by deposition onto
system walls. Thus, in the case of the Kr-85 source, the peaks
observed in the profile most probably correspond to particle
agglomerates, and were carried by the air flow to the neutralization
zone and then to the measurement system because of their lower
electrical mobility. The particle size distribution of OVA, shown in
Fig. 4(b), shows the characteristic monomer and dimer peaks at
6.8nmand8.2nmrespectively.Again,notethattheparticlenumber
concentrationwas higher with the SXC thanwith the Kr-85 charger.
The apparently low resolution of the SXC’s profile may have been
caused by the similar reasons discussed in the case of the MS2
phages. As a note of caution, and from the results of this section, in
ESSA one must be careful to use capillaries with small diameters to
ensure the production of one particle per droplet and to arrange the
setup so as to avoid particle losses.
4. Conclusions
Photoionization with an SXC was shown to be equally effective
in neutralizing multiply charged particles, in the nanometer-size
range, as the commonly used Po-210 diffusion neutralizer. With the
SXC, direct photoionization of aerosol particles and gas molecules
as well as Brownian diffusion of ions took place simultaneously,
which enhanced the neutralization efficiency, thereby resulting in
instantaneous achievement of the particles’ Boltzmann charge
distribution. Electrospray size spectrometry analysis of PSL, MS2
bacteriophages, and OVA particles neutralized with the SXC
showed excellent agreement with the radioactive Po-210 source
used in the present and previous studies. Total number concen-
tration measurements showed that photoionization with the SXC
was sufficient to bring highly charged particles to the equilibrium
charge-state and reduced particle losses. When compared against
a Kr-85 b-ray emitter in an in-house electrospray system, the SXC
resulted in higher particle number concentration for the two
particles tested, due to both the higher ion number concentration
generated with the SXC and the Kr-85 performance’s dependence
on the carrier gas flow rate. Although, the Po-210 neutralizer is
broadly used in ESSA, the results of this study demonstrate that it
can be replaced for a safer and more controllable technology such
as photoionization with soft X-rays. Nevertheless, an improved SXC
design that could be incorporated into an EAG needs to be further
investigated.
Acknowledgments
Discussions with Dr. Samuel Yenne from Virus Detection
Systems Corporation (Cary, North Carolina, USA) are gratefully
acknowledged. LBML acknowledges partial financial support from
the FeT Research Group, Universitat Rovira i Virgili.
References
[1] L.B. Modesto-Lopez, A.M. Collins, R.E. Blankenship, P. Biswas, Electrospray-
assisted characterization and deposition of chlorosomes to fabricate a light-
harvesting biomimetic deivce, Energy Environ. Sci. 3 (2010) 216e222.
[2] C.B. Mao, D.J. Solis, B.D. Reiss, S.T. Kottmann, R.Y. Sweeney, A. Hayhurst,
G. Georgiou, B. Iverson, A.M. Belcher, Virus-based toolkit for the directed
synthesis of magnetic and semiconducting nanowires, Science 303 (2004)
213e217.
[3] A. Gomez, The electrospray and its application to targeted drug delivery,
Respir. Care 47 (2002) 1419e1431.
[4] G. Allmaier, C. Laschober, W. Szymanski, Nano ES GEMMA and PDMA, new
tools for the analysis of nanobioparticleseprotein complexes, lipoparticles,
and viruses, J. Am. Soc. Mass Spectrom. 19 (2008) 1062e1068.
[5] G. Bacher, W.W. Szymanski, S.L. Kaufman, P. Zollner, D. Blaas, G. Allmaier,
Charge-reduced nano electrospray ionization combined with differential
mobility analysis of peptides, proteins, glycoproteins, noncovalent protein
complexes and viruses, J. Mass. Spectrom. 36 (2001) 1038e1052.
[6] C.J. Hogan, E.M. Kettleson, B. Ramaswami, D.R. Chen, P. Biswas, Charge
reduced electrospray size spectrometry of mega- and gigadalton complexes:
whole viruses and virus fragments, Anal. Chem. 78 (2006) 844e852.
[7] S.L. Kaufman, Analysis of biomolecules using electrospray and nanoparticle
methods: thegas-phaseelectrophoretic
J. Aerosol Sci. 29 (1998) 537e552.
[8] L.B. Modesto-Lopez, J.D. Pasteris, P. Biswas, Sensitivity of micro-Raman spec-
trum to crystallite size of electrospray-deposited and post-annealed films of
iron-oxide nanoparticle suspensions, Appl. Spectrosc. 63 (2009) 627e635.
[9] B. Ku, P. Kulkarni, Morphology of single-wall carbon nanotube aggregates
generated by electrospray of aqueous suspensions, J. Nanopart. Res. 11 (2009)
1393e1403.
[10] E.O. Knutson, K.T. Whitby, Aerosol classification by electric mobility: appa-
ratus, theory, and applications, J. Aerosol Sci. 6 (1975) 443e451.
[11] J.H. Ji, G.N. Bae, J. Hwang, Characteristics of aerosol charge neutralizers for
highly charged particles, J. Aerosol Sci. 35 (2004) 1347e1358.
[12] E. Hontanon, F.E. Kruis, Single charging of nanoparticles by UV photoioniza-
tion at high flow rates, Aerosol Sci. Technol. 42 (2008) 310e323.
[13] M. Shimada, B.W. Han, K. Okuyama, Y. Otani, Bipolar charging of aerosol
nanoparticles by a soft X-ray photoionizer, J. Chem. Eng. Jpn. 35 (2002)
786e793.
[14] J. Jiang, M.H. Lee, P. Biswas, Model for nanoparticle charging by diffusion,
direct photoionization, and thermionization mechanisms, J. Electrostat. 65
(2007) 209e220.
[15] P. Kulkarni, N. Namiki, Y. Otani, P. Biswas, Charging of particles in unipolar
coronas irradiated by in-situ soft X-rays: enhancement of capture efficiency of
ultrafine particles, J. Aerosol Sci. 33 (2002) 1279e1296.
[16] J. Jiang, J.C. Hogan, D.-R. Chen, P. Biswas, Aerosol charging and capture in the
nanoparticle size range (6e15 nm) by direct photoionization and diffusion
mechanisms, J. Appl. Phys. 102 (2007) 034904e034907.
[17] C.J. Hogan, M.H. Lee, P. Biswas, Capture of viral particles in soft X-ray-
enhanced corona systems: charge distribution and transport characteristics,
Aerosol Sci. Technol. 38 (2004) 475e486.
[18] E.M. Kettleson, B. Ramaswami, C.J. Hogan, M.H. Lee, G.A. Statyukha, P. Biswas,
L.T. Angenent, Airborne virus capture and inactivation by an electrostatic
particle collector, Environ. Sci. Technol. 43 (2009) 5940e5946.
[19] B.W. Han, M. Shimada, K. Okuyama, M. Choi, Classification of monodisperse
aerosol particles using an adjustable soft X-ray charger, Powder Technol. 135
(2003) 336e344.
[20] H. Lee, C. Kim, M. Shimada, K. Okuyama, Bipolar diffusion charging for aerosol
nanoparticle measurement using a soft X-ray charger, J. Aerosol Sci. 36 (2005)
813e829.
[21] J. Suh, B. Han, K. Okuyama, M. Choi, Highly charging of nanoparticles through
electrospray of nanoparticle suspension, J. Colloid Interface Sci. 287 (2005)
135e140.
[22] J.H. Thomas, S.K. Kim, P.J. Hesketh, H.B. Halsall, W.R. Heineman, Bead-based
electrochemical immunoassay for bacteriophage MS2, Anal. Chem. 76 (2004)
2700e2707.
[23] R. Golmohammadi, K. Valegård, K. Fridborg, L. Liljas, The refined structure of
bacteriophage MS2 at 2.8 A resolution, J. Mol. Biol. 234 (1993) 620.
mobilityanalyzer (GEMMA),
L.B. Modesto-Lopez et al. / Journal of Electrostatics 69 (2011) 357e364
363

Page 8

[24] C.S. Kaddis, S.H. Lomeli, S. Yin, B. Berhane, M.I. Apostol, V.A. Kickhoefer,
L.H. Rome, J.A. Loo, Sizing large proteins and protein complexes by electro-
spray ionization mass spectrometry and ion mobility, J. Am. Soc. Mass Spec-
trom. 18 (2007) 1206e1216.
[25] J. Thomas, B. Bothner, J. Traina, W.H. Benner, G. Siuzdak, Electrospray ion
mobility spectrometry of intact viruses, Spectrosc. Int. J. 18 (2004) 31e36.
[26] K. Takatera, T. Watanabe, Determination of sulfhydryl-groups in ovalbumin by
high-performance liquid-chromatography with inductively-coupled plasma-
mass spectrometric detection, Anal. Chem. 65 (1993) 3644e3646.
[27] C. Hogan, E. Kettleson, M. Lee, B. Ramaswami, L. Angenent, P. Biswas, Sampling
methodologies and dosage assessment techniques for submicrometre and
ultrafine virus aerosol particles, J. Appl. Microbiol. 99 (2005) 1422e1434.
[28] B.Y.H. Liu, D.Y.H. Pui, Electrical neutralization of aerosols, J. Aerosol Sci. 5
(1974) 465e472.
[29] B. Liu, D. Pui, B. Lin, Aerosol charge neutralization by a radioactive alpha-
source, Part. Part. Syst. Char. 3 (1986) 111e116.
[30] R.M. Eninger, C.J. Hogan, P. Biswas, A. Adhikari, T. Reponen, S.A. Grinshpun,
Electrospray versus nebulization for aerosolization and filter testing with
bacteriophage particles, Aerosol Sci. Technol. 43 (2009) 298e304.
[31] J.F. de la Mora, On the outcome of the coulombic fission of a charged isolated
drop, J. Colloid Interface Sci. 178 (1996) 209e218.
[32] L.B. Modesto-Lopez, P. Biswas, Role of the effective electrical conductivity of
nanosuspensions in the generation of TiO2agglomerates with electrospray,
J. Aerosol Sci. 41 (2010) 790e804.
L.B. Modesto-Lopez et al. / Journal of Electrostatics 69 (2011) 357e364
364