BASIC SCIENCE

Nanomedicine: Nanotechnology, Biology, and Medicine

8 (2012) 702 711

Research Article nanomedjournal.com

Improved antibacterial and antibiofilm activity of magnesium fluoride

nanoparticles obtained by water-based ultrasound chemistry
Jonathan Lellouche, MSc a, b , Alexandra Friedman, MSc b , Jean-Paul Lellouche, PhD b ,
Aharon Gedanken, PhD b , Ehud Banin, PhD a,
a
The Biofilm Research Laboratory, The Bar-Ilan Institute of Nanotechnology and Advanced Materials, The Mina and Everard Goodman Faculty of Life
Sciences, Bar-Ilan University, Ramat-Gan, Israel
b
Department of Chemistry, The Bar-Ilan Institute of Nanotechnology and Advanced Materials, Bar-Ilan University, Ramat-Gan, Israel
Received 9 November 2010; accepted 6 September 2011

Abstract

Antibiotic resistance has prompted the search for new agents that can inhibit bacterial growth. We recently reported on the antimicrobial
and antibiofilm activities of nanosized magnesium fluoride (MgF2) nanoparticles (NPs) synthesized in ionic liquid using microwave
chemistry. In this article, we describe a novel water-based synthesis of MgF2 NPs using sonochemistry. The sonochemical irradiation of an
aqueous solution of [Mg(OAc)2(H2O)4] containing acidic HF as the fluorine ion source afforded crystalline well-shaped spherical MgF2
NPs that showed much improved antibacterial properties against two common bacterial pathogens (Escherichia coli and Staphylococcus
aureus). We were also able to demonstrate that the antimicrobial activity was dependent on the size of the NPs. In addition, using the
described sonochemical process, we coated glass surfaces and demonstrated inhibition of bacterial colonization for 7 days. Finally, the
antimicrobial activity of MgF2 NPs against established biofilms was also examined. Taken together our results highlight the potential to
further develop the concept of utilizing these metal fluoride NPs as novel antimicrobial and antibiofilm agents.

From the Clinical Editor: In this article, the authors describe a novel aqueous synthesis of magnesium fluoride NPs using sonochemistry.
These nanoparticles have improved antibacterial and antibiofilm activity compared to their counterparts with traditional synthesis methods.
2012 Elsevier Inc. All rights reserved.

Key words: Magnesium fluoride nanoparticles; Sonochemistry; Antibacterial properties; Biofilms; Aggregation

The increased resistance of pathogenic bacteria to antibiotic
therapy is becoming a serious public health issue, especially in
light of the limited number of new antibiotics expected to enter
the market in the next few year. 1,2 This emerging health threat
has highlighted the urgent need for the discovery of innovative
bactericides and/or novel pathogen control methodologies. 3-5
In this context, nanotechnology-driven approaches are
expected to open new routes to prevent and fight bacteria-
mediated diseases via an extended use of atomic scale, tailored
nanomaterials. Recent examples include: metallic, 6-9 and metal
oxide 10-12 nanoparticles (NPs), and carbon nanotubes 13 that
exert antimicrobial activity against a broad range of pathogenic
microorganisms. Results thus far suggest that the most active
The authors confirm that there are no known conflicts of interest
associated with this publication, and there has been no significant financial
support for this work that could have influenced its outcome (current and the
past five years).
Corresponding author:
E-mail address: ehud.banin@biu.ac.il (E. Banin).
1549-9634/$ see front matter 2012 Elsevier Inc. All rights reserved.
doi:10.1016/j.nano.2011.09.002
antibacterial nanomaterials exhibit an increased chemical
surface-mediated reactivity mainly due to their large surface-
to-volume ratios and defined crystallographic structures. 8,10
Another emerging trend is the development of composite
multiphase materials that combine antibacterial nanomaterials
with known biocompatible biomaterials via matrix impregnation
and surface coating. This nanofabrication allows the develop-
ment of surfaces (e.g., medical implants) with antibacterial
activity. 14-17 The potential market for such materials is
estimated in billions of dollars per annum. 18 This is because
many of the bacterial infections are associated with medical
implants (e.g.,catheters, heart valves, and artificial joints) in
which bacteria can colonize the abiotic surfaces and form
biofilms that are highly resistant to antimicrobial therapy. 2,19
Our group has recently demonstrated the antibacterial and
antibiofilm capabilities of highly crystalline 25-nm-sized
magnesium fluoride (MgF2) NPs. 20 The fluoride anion has
known antibacterial activities, 21 and our initial goal was to study
whether by producing nanometric particles we can improve its
antimicrobial activity. The MgF2 NPs have been prepared using

Please cite this article as: J. Lellouche, A. Friedman, J.-P. Lellouche, A. Gedanken, E. Banin, Improved antibacterial and antibiofilm activity of magnesium
fluoride nanoparticles obtained by water-based ultrasound chemistry. Nanomedicine: NBM 2012;8:702-711, doi:10.1016/j.nano.2011.09.002
 J. Lellouche et al / Nanomedicine: Nanotechnology, Biology, and Medicine 8 (2012) 702711
a unique one-step, microwave-assisted decomposition of the
ionic liquid BIMBF4 (1-butyl-3-methylimidazolium tetrafluor-
oborate) in the presence of magnesium acetate ([Mg(OAc)2(-
H2O)4]). In this process, the ionic liquid acted as a solvent,
fluorine ion source (BF4- anion decomposition), and as a NP-
precipitation medium enabling a straight NP isolation/purifica-
tion procedure. 20 The MgF2 NPs disclosed a strong biological
activity against Escherichia coli (E. coli) and Staphylococcus
aureus (S. aureus) at a millimolar level. Our results revealed that
the NPs penetrate the cells, causing a disruption in the cell
membrane potential, enhancing lipid peroxidation and DNA
binding. 20 This preliminary study of MgF2 NPs drove us to
design a more robust process, mainly due to the recurrent issues
associated with ionic liquid thermal decomposition. Apparently,
the high reaction temperatures that are rapidly obtained by
irradiating polar materials in a microwave field may lead to
decomposition of the ionic liquid to form carbon on the NP
surface, which may affect its antimicrobial activity. Furthermore,
the formation of a small grain of carbon causes the temperature to
increase further because of the thermal runaway phenomenon,
which creates hot spots in the liquid. 22-24 These conditions
preclude the use of this process for the deposition of MgF2
particles onto temperature-sensitive polymers. Therefore, we
explored the potential use of a water-based sonochemical
processing for MgF2. It is well known that sonochemistry
enabled the generation of novel materials that possess uncon-
ventional properties. 25 This is due to the chemical activation
capability arising from the high-energy collapse of cavitation
bubbles. The continuous formation, growth, and implosive
collapse of sonochemically generated bubbles leads to a local
and extremely high increase in temperature (N 5000 K), pressure
(N 200 MPa), and cooling rates (N 10 7 Ks -1) that drives the
chemical synthesis of NPs. 26,27
In this article, we demonstrate that the fabrication of
crystalline nanosized MgF2 particles in H2O can be accom-
plished and that these particles have improved antibacterial
activity. We developed a new, improved protocol to coat glass
coupons in a one-step process and to show that the coated
surfaces can restrict bacteria colonization and biofilm formation
for up to 7 days. In addition, the antimicrobial activity of MgF2
NPs against established biofilms was examined. Finally, the
improved control in the size distribution of the NPs provided by
the sonochemical technique allowed us to explore the correlation
between the NP size and its antibacterial activity. Taken together,
our results further support the development of MgF2 NPs as
potential antimicrobial and antibiofilm nanosized agents.
Methods
MgF2 NP synthesis
Magnesium (II) acetate tetrahydrate ([Mg(Ac)2(H2O)4]),
Aldrich, Rehovot, Israel, 99% purity) and concentrated hydro-
fluoric acid (HF, 32% weight aqueous solution, ACS grade,
BioLab, Jerusalem, Israel) were dissolved in double-distilled water
(ddH2O, 100 mL) at a 1:2 equivalent ratio for all the prepared
MgF2 NPs. More specifically, three MgF2 NP samples varying in
the NPs' size were prepared by decreasing the HF concentration as
703
follows: 0.1M HF (MgF2-2), 0.01M HF (MgF2-3) and 0.001M HF
(MgF2-4). Keeping the Mg:F molar ratio as 1:2 for all three
samples. The sample marked MgF2-1 refers to a commercial
amorphous MgF2 salt (Aldrich, 99% purity). During the NP
fabrication, each separate mixture was irradiated with a high-
intensity ultrasonic horn (Ti-horn, 20 kHz, 750 W cm -2, 60%
power modulation) under argon (60 minutes, ~25C). The
resulting precipitating products were washed thoroughly with
ddH2O (3 10 mL), absolute ethanol (EtOH, 2 10 mL), and
dried in vacuum (10 -2 mm Hg) in an inert glove box (O2 b 1 ppm).
The experimental details on the microwave-assisted fabrication
procedure of MgF2 NPs have already been reported. 21 This
method used a short-time microwave irradiation (2 minutes, 2.45
GHz, Kenwood 900W) of an ionic liquid (1-butyl-3-methylimi-
dazolium tetrafluoroborate) and magnesium (II) acetate tetrahy-
drate ([Mg(Ac)2(H2O)4]).
MgF2 NP characterization
In this research. x-ray diffraction (XRD) measurements were
carried out on a Bruker D8 diffractometer using Cu K radiation
( = 1.5418). Peak fitting and lattice parameter refinement were
computed using the EVA program (Bruker Analytical X-Ray
Systems, Rehovot, Israel). The NPs' sizes were calculated from
the XRD pattern by employing the Debye-Scherrer equation and
as measured by dynamic light scattering (DLS, Beckman Coulter
N-4 particle size analyzer, Nyon, Switzerland). The NP surface
area was measured using a Micromeritics analyzer (Gemini
2375, micrometrics, Norcross, Georgia) in the linear part of the
BET plot of N2 adsorption/desorption isotherms of each separate
MgF2 sample (all measurements were done in triplicate). The NP
morphology was characterized using a high resolution TEM
(HR-TEM, JEOL-2010 HR-TEM apparatus, accelerating voltage
200 kV, JEOL Ltd., Tokyo, Japan). NP samples for HR-TEM
analysis were prepared in absolute EtOH (ultrasonic dispersion),
deposited onto a copper-coated grid (drop deposition), and then
dried under vacuum (10 -2 mm Hg) before sample processing.
The elemental carbon analysis of MgF2 NP samples (i.e.,
microwave and ultrasonic syntheses) was determined using an
automatic Elemental Analyzer (EA, EA1110, CE Instruments,
Wigan, United Kingdom).
Bacterial cultures and growth conditions
E. coli C600 and S. aureus FRF119 20 were grown at 37C in
Tryptic Soy Broth (TSB, Difco) or Tryptic Soy Broth 66%,
supplemented with glucose 0.2% (TSB-Glu) media, respectively.
These media were chosen based on their ability to promote,
respectively, robust E. coli and S. aureus biofilm formation. 28-30
Antibacterial test
The antimicrobial activity of MgF2 NPs, was examined both
on logarithmic and stationary phase cultures (the former are
known to be more resistant to antimicrobial treatment). To
examine the antimicrobial activity on logarithmic phase cultures a
modified macrodilution assay was utilized. Overnight cultures of
tested bacteria were diluted (1:100) in fresh TSB or TSB-Glu and
grown for 4 hours at 37C (shaking, 250 rpm) to allow the cells to
reenter logarithmic phase. Following this the bacteria were diluted
704 J. Lellouche et al / Nanomedicine: Nanotechnology, Biology, and Medicine 8 (2012) 702711
3
again to 10 colony forming units (CFU)/mL in the appropriate
growth media. Next, 100 L of the tested cell suspension was
then added to each well of a 96-well plate and MgF2 particle
samples at various concentrations (0.0001 to 1.0 mg of MgF2/mL)
were also added. The plates were than incubated for 24 hours at
37C. Cell growth was monitored by viable cell count.
To test the effect of MgF2 NPs on survival of stationary phase
cultures we used an overnight culture (approximately 1 10 9
and 1 10 8 CFU/mL for E. coli and S. aureus, respectively).
The cultures were centrifuged; the cells were washed twice and
resuspended in a volume of fresh medium equal to the original
culture volume. The cells were then incubated at 37C for an
additional 6 hours; this procedure ensures that the culture renters
the stationary phase. 31 Following this MgF2 NP (0.0001 to 1.0
mg of MgF2/mL) was added and cell viability was monitored
over time by plating serial dilutions and determining CFU.
Bacteria not exposed to MgF2 NP served as controls.
MgF2 NPs aggregation assays
MgF2 samples (1.0 mg of MgF2/mL) were incubated in both
TSB and TSB-Glu media at 37C for 6 hours. NP sampling for
DLS (DLS, Beckman Coulter N-4 particle size analyzer, Nyon,
Switzerland) measurements were carried out at regular intervals
of 1 or 2 hours. Finally, at the completion (6 hours) of incubation,
NP samples (triplicate measurements) were shortly irradiated
with a high-intensity ultrasonic horn (2-minute-long ultrasound
pulse, Ti-horn, 20 kHz, 750 W cm -2, 35% power modulation)
to check the stability of the aggregates and examine whether they
will disperse.
Surface-coating procedure and antibiofilm assay
Glass surfaces were coated by placing standard microscope
glass slides directly into the sonochemical reaction medium
according to a methodology described previously. 11,32 After
completion of sonication, the sonochemically coated slides were
washed with ddH2O (3 10 mL), absolute EtOH (2 10 mL),
allowed to dry in vacuum (10 -2 mm Hg), and imaged by
scanning electron microscopy (SEM, FEI, QUANTA 200F,
Hillsboro, Oregon). The elemental analyses of the coated layer
and depth profiles were analyzed by time-of-flight ion mass
spectrometry (TOF-SIMS, PHI Model 2100 TRIFT II system,
Chanhassen, Minnesota). UV visible light transmittance proper-
ties of MgF2 NP-coated glass slides were checked using a Cary
100 Scan UV-Vis spectrophotometer (Varian, spectral band-
width of 200 800 nm, Palo Alto, California).
For the microscopic examination of biofilm formation, a flow
cell system was inoculated with a 0.3 OD595 cell culture of either
E. coli or S. aureus (this approximately corresponds to 3 10 8
CFU/mL). The flow was initiated after 1 hour incubation at room
temperature (flow rate of 10 mL/h) and 1% TSB or 1% TSB-Glu
(diluted in ddH2O) was used as growth medium for E. coli and S.
aureus, respectively. Confocal Laser Scanning Microscope
(CLSM) imaging of the resulting biofilms employed a CSLM
(Leica SPE, San Diego, California). Glass slides coated with
MgF2 NPs were removed from the experimental flow cell and
stained using the Live/Dead BacLight kit (Molecular Probes,
Invitrogen [Carlsbad, California], per manufacturer's protocol).
A SYTO9/propidium iodide mixture stain was dissolved
in a mixture of 3% dimethylsulfoxide (DMSO) and ddH2O
(15-minute incubation). Viable bacteria with intact cell mem-
branes are stained in green, whereas dead bacteria with damaged
membranes are stained in red. Both excitation/emission
maxima for these two dyes are 480/500 nm for the SYTO9
stain, and 490/635 nm for the propidium iodide. Obtained
images were further processed by the Imaris Image Analysis
software (Imaris v.6.0, Bitplane Scientific Software, South
Windsor, Connecticut) and represent the general trend seen in
three independent experiments.
Established biofilm assay
Overnight cultures of E. coli and S. aureus, grown in TSB
and TSB-Glu, respectively, were diluted to a final concentration
of approximately 3 10 8 CFU/mL. Next, 200 L of this cell
suspension was added to each well of a 96-well tissue culture
plate (Greinier, Monroe, North Carolina). The plates were than
incubated at 37C with shaking (100 rpm) for 1, 3, 5, and 7 days
(to allow the biofilm to establish) at the indicated time points.
After 1, 3, 5, or 7 days of growth, the various wells were washed
twice with 200 L of phosphate-buffered saline (PBS) per well to
remove nonadherent bacteria. The biofilm-containing wells were
than treated either with MgF2 NP samples synthesized by
microwave or sonochemistry (1 mg/mL). The biofilms were
incubated with the NPs for 24 hours with gentle shaking at 37C;
untreated biofilms served as controls. To determine the number
of biofilm-viable cells following the treatment, wells were
washed twice with PBS to remove planktonic bacteria and then
biofilm cells were detached by exposure to low-energy
sonication using a sonication water bath (TRANSSONIC 460,
Elma GmbH, Singen, Germany) for 1 minute. Following this
stage, the plates were centrifuged (centrifuge 5810 R, Eppendorf,
Athens, Greece) at 4000 rpm for 5 minutes to form a pellet of
cells. The number of viable cells was determined by resuspend-
ing the pelleted cells in sterile media, plating serial dilutions on
Luria Bertani (LB, Difco, BD, Franklin Lakes, New Jersey)
plates and determining the CFU after 24 hours of incubation.
Results
Characterization of MgF2 NPs
To obtained MgF2 NP, we utilized a sonication process
containing an aqueous solution of [Mg(OAc)2(H2O)4] and acidic
HF. The powder XRD analysis of the NP showed a clear
crystalline pattern (Figure 1, A). The XRD pattern matched well
the reflection peaks of the tetragonal MgF2 phase (JCPDS card
No. 00-041-1443) characterized by diffraction planes (110),
(101), (111), (210), (211), (220), (002), (310), (301), (311), and
(222). No additional diffraction peaks of any impurity were
detected, demonstrating the high purity of the product. In addition,
the average size diameter of crystallites calculated by the Debye-
Scherrer equation afforded a value of 26 (0.8) nm, which is
similar to the average size measured by HR-TEM (i.e., 25 (0.6)
nm, Figure 1, B). The fabricated MgF2 NPs showed spherical,
well-shaped nanostructure morphology. Figure 1, B (insert) also
 J. Lellouche et al / Nanomedicine: Nanotechnology, Biology, and Medicine 8 (2012) 702711 705

A sensitive to the NPs than was E.coli, and no growth was

observed at a concentration of 0.1 mg MgF2/mL. The differences

(111)
250 in E. coli and S. aureus strains' sensitivity to the MgF2 NP
(110) MgF2 Nps
treatment can be attributed to differences in the composition of
200 the gram-negative and gram-positive membranes. 32 NPbacte-

(211)
rial interactions are influenced by interfacial forces, especially
Intensity (a.u.)

150 electrostatic, that can control the interaction between NPs and the
bacterial surface. 32
100
Bacteria in the stationary phase are known to be more
(210)

(301)
resistant to stress and antibiotic treatment; thus we examined

(220)
(101)

(002)
the antimicrobial activity of the MgF2 against stationary phase
50

(310)

(311)
cultures (Figures 2, B and D). The highest concentration of

(222)
NPs tested (1.0 mg MgF2/mL) was found to reduce the
0 viability of E. coli from 2.8 10 9 CFU/mL to a concentration
below 10 2 CFU/mL (Figure 2, B). S. aureus (Figure 2, D)
20 30 40 50 60 70 80
2 (degree)
was again more sensitive to the NP treatment than was E. coli,
B and the bacteria were completely killed at a concentration of
0.01 mg/mL.
We re-exposed the E. coli culture that survived the treatment
with the highest MgF2 NP concentration (1 mg MgF2/mL, Figure
2, B) to examine whether the bacteria developed resistance to
MgF2 NPs (Figure S1). The exposed bacteria showed the same
sensitivity and growth yield to bacteria that were not pre-exposed
to the NP. This suggests that the bacteria that survived the first
treatment are not resistant to the NPs.
Finally, it is important to emphasize that the antimicrobial
activity of the NPs synthesized by sonochemistry was more
potent than the NPs synthesized by the previously described
microwave chemistry protocol (Figure 2).

MgF2 NP aggregation

A major obstacle in NP utilization that could also affect the

antimicrobial activity is the tendency of the particles to
aggregate in water-based solutions. All the tested MgF2 NPs
aggregated in a time-dependent manner (Figure 3), regardless of
Figure 1. MgF2 NP characterizations. MgF2 NP characterizations. (A) the preparation method (i.e., microwave versus sonochemical).
Powder X-Ray diffraction (XRD) analysis of the crystalline NP. The XRD However, a clear difference in aggregation diameter and
pattern matched the reflection peaks and relative Miller indices of tetragonal
stability was observed between the preparation methods. The
MgF2. (B) HR-TEM images of MgF2 NPs and (insert in B) characteristic
lattice fringes of the crystalline phase. average aggregate diameter was approximately 400 nm for the
ultrasound synthesized NP and approximately 1450 nm for NPs
prepared using microwave radiation in both culture media (i.e.,
disclosed characteristic lattice fringes of the crystalline phase. The TSB and TSB-Glu) (Figure 3). These values of aggregate sizes
measured interfringe distance of 3.32 matches perfectly the were obtained by DLS measurements, which also showed that
(110) interplanar distance (JCPDS card No. 00-041-1443). the NPs prepared by the ultrasound method afforded the
maximal value of aggregate diameters (406/404 nm for TSB and
Antimicrobial activity of MgF2 NPs TSB-Glu, respectively) already after a 2-hour incubation time.
In contrast, the NPs prepared by the microwave method showed
The antimicrobial activity of fabricated MgF2 NPs was a constant time-dependent increase in aggregate size. Moreover,
examined by using two common bacterial pathogens, E. coli both types of NPs behave differently, following a short
(gram negative) and S. aureus (gram positive) (Figure 2). ultrasound pulse that was implemented to evaluate the
Logarithmic cultures of both strains were inoculated with an reversibility nature of the NP aggregate. The ultrasonically
increasing dosage of MgF2 NPs and growth was monitored over fabricated NPs broke into small 38- or 40-nm-sized aggregates
a period of 24 hours. Increasing concentrations of NPs (0.0001 to in TSB and TSB-Glu, respectively. This likely corresponds to
a 1.0 mg MgF2/mL) inhibited the growth of E. coli and S. aureus the irreversible association of two primary crystallite NPs. In
progressively in a dose-dependent manner. E. coli bacterial contrast, microwave-prepared NPs afforded irreversibly associ-
growth was completely inhibited at the highest concentration of ated aggregates that remained in a submicrometer diameter size
NPs (1.0 mg/mL, Figure 2, A). S. aureus (Figure 2, C) was more range of 479 nm in TSB and 616 nm in TSB-Glu (Figure 3).
706 J. Lellouche et al / Nanomedicine: Nanotechnology, Biology, and Medicine 8 (2012) 702711

Figure 2. Antimicrobial affect of MgF2 NPs. Killing curves of (A and C) mid-log and (B and D) stationary phases of E. coli (AB) and S. aureus (CD). For
the mid-log assays (A and C), inocula with an initial concentration of 10 3 CFU/mL of each strain was exposed to variable concentrations (0.0001 to 1 mg/mL) of
MgF2 NP synthesized by sonochemistry (solid lines). For the stationary phase assays, stationary phase E. coli (B) and S. aureus (D) were exposed to different
concentrations of MgF2 NPs (0.0001 to 1 mg/mL) synthesized by sonochemistry. Arrows in B and D refer to the time point at which the MgF2 NPs were added
(see Methods section for details). Killing was determined by viable counts and was expressed in CFUs. Dashed lines represent the killing curves of E. coli (in
2AB) and S. aureus (in 2CD) exposed to the highest concentration tested (1 mg/mL) of MgF2 NPs synthesized by microwave chemistry (marked with MW).
All other tested NPs preparations were synthesized by sonochemistry. In all experiments, untreated bacteria served as a negative control. Error bars represent the
standard deviation (SD) of three independent experiments.

This difference in aggregation may indeed explain the reduced
antimicrobial activity that was observed with the microwave-
synthesized NPs.
Characterization and antibiofilm properties of MgF2
NP-coated surfaces
Our next step was to examine whether we can utilize the
sonochemical procedure to coat surfaces with MgF2 NPs and
provide antibiofilm properties to the surface similar to those
observed in microwave chemistry. A sonication reaction with
conditions typical for the formation of MgF2 NPs was applied.
However, in this experiment the synthesis was carried out in the
presence of microscope glass slides that were inserted in the
sonication cell (see Methods section for details). The advantage
of this one-step process is that the ultrasonic fabrication synthesis
of the NP is followed by microjets that throw the newly formed
NPs onto the glass slide and embed the NPs in the glass. 11,33
SEM measurements of the transparent coated slides were
performed, and the results are depicted in Figure S2, A. The
picture reveals the presence of MgF2 NPs on the surface. The
average NP size was found to be 28 4 nm (Figure S2). We also
investigated the penetration depth profile of the adsorbed NPs
into the glass slide by conducting TOF-SIMS measurements; the
penetration of magnesium atoms was observed up to a 90-nm
depth within the glass (Figure S3). The coating of the
ultrasonically deposited NPs has also been characterized by
measuring the UV visible light transmittance feature in the 200
750 nm wavelength range (visible domain). The functional NP
coating caused a reduction of light transmittance of less than 2 %
(200 750 nm domain) in agreement with the low refractive
index of bulk crystalline MgF2 (spectrum not shown). 34
The MgF2 NP-coated slides were then tested for their ability
to restrict the bacterial colonization of E. coli and S. aureus.
Figure 4 presents the corresponding CLSM images that depict
biofilm development following a 7-day bacterial exposure. The
untreated surfaces supported massive biofilm formation in
comparison with MgF2 NP-coated surfaces. The results suggest
that the MgF2 NP-coated slides effectively inhibited bacterial
adhesion and biofilm formation for at least 7 days. As with the
results obtained with the planktonic cultures, S. aureus seems to
be more sensitive to the MgF2 NP-coated surfaces in comparison
 J. Lellouche et al / Nanomedicine: Nanotechnology, Biology, and Medicine 8 (2012) 702711 707

Figure 3. Analysis of aggregation of MgF2 NPs. MgF2 NPs (1 mg/mL) synthesized by microwave or ultrasonic syntheses were incubated in TSB (A) or TSB-
Glu (B) at 37C. Aggregation was monitored every 2 h by DLS. Following this incubation, the ability to disperse the aggregates was evaluated by applying a
short ultrasonic pulse (2 min). Number over bars represents the average aggregate diameter. Error bars represent the SD of three independent experiments.

with E. coli. No S. aureus cells were observed on the coated
surfaces during all 7 days, whereas a small number of E. coli
cells were detected on the surface, some of which were dead
(stained in red, Figure 4). These results suggest that a few E. coli
cells likely succeed in the initial adhesion process, but the MgF2
NP-coated surface triggers the inhibition of further development
and/or detachment. The exact mechanism by which the MgF2
NPs mediated these processes is still unclear and requires
additional study.
Antimicrobial activity of MgF2 NP against mature biofilms
Next, we explored the ability of MgF2 NPs to display
antibiofilm properties against already established mature bio-
films. We exposed 1-, 3-, 5-, and 7-day-old biofilms to MgF2 NP
(1 mg MgF2/mL) for 24 hours (Figure 5). One-day-old E. coli
biofilms showed 87% decrease in biofilm viability, and 7-day-
old biofilms showed only 44% decrease in viability (Figure 5, A;
light gray bars). S. aureus displayed the same trend. One-day-old
biofilms that were treated with NPs showed 92% decrease in
biofilm viability whereas 7-day-old biofilms showed a 69%
decrease in biofilm viability (Figure 5, B; light gray bars). In
addition to, and similar to, the results observed with planktonic
cultures, the microwave synthesized particles less effectively and
showed no activity on 7-day-old biofilms of either bacteria
(Figure 5; dark gray bars).
The effect of MgF2 NP size on aggregation and
antimicrobial activity
A major advantage of the sonochemical-based synthesis
procedure is the ability to control the MgF2 NP size. This is
achieved by varying the concentrations of both [Mg(Ac)2(H2-
O)4] and HF components but still maintaining their relative
molar ratio at 1:2 during the ultrasonic process. Using that
strategy, we were able to synthesize three groups of MgF2 NPs
that vary in diameter. We are able to change the average
diameter from 300 nm (marked as MgF2-2) to 26 nm (marked as
MgF2-3), and further to 5 nm (marked as MgF2-4), as
determined by the application of the Debye-Scherrer formula
(Figure S4, B). XRD patterns obtained for all the groups of NPs
confirmed their crystallinity and size-dependent nature (peak
enlargement with a decreasing NP size, Figure S4, A). DLS
measurements of the same products in EtOH afforded similar
hydrodynamic diameters of 290 5 nm (MgF2-2), 25 10 nm
(MgF2-3) and 3 6 nm (MgF2-4) (Figure S4, B). In addition,
commercially available 2.5 micrometer-sized MgF2 particles
(marked as MgF2-1) were also examined. As expected, the
708 J. Lellouche et al / Nanomedicine: Nanotechnology, Biology, and Medicine 8 (2012) 702711

Figure 4. Antibiofilm activity of MgF2 NP coatings. E. coli and S. aureus were grown for 7 days in flow cells on uncoated and MgF2 NP-coated glass surfaces
(experimental setup and growth conditions described in the Methods section). Green and red staining represents live and dead bacterial cells, respectively.
Images obtained by CLSM represent the general trend seen in three independent experiments.

surface area obtained by the BET method substantiated the size
dependence of the samples. The surface area increased with the
decrease in NP diameter from approximately 13 (MgF2-1) to
186 (MgF2-4) m 2/g (Figure S4, B).
Next, we examined how the changes in NP size and surface
area affect the antimicrobial activity. The results presented in
Figure 6 clearly show a reverse correlation between the NP size
and the antimicrobial activity for both E. coli and S. aureus (0.01
mg MgF2/mL); i.e., the antimicrobial activity increases as the NP
size decreases. In fact, the smallest NPs tested may even allow
the internalization of NPs via porins and channels, thus
improving penetration. We found it interesting that the
micrometer-sized commercial particles were completely inactive
against both strains.
To understand the relationship between particle size and the
antimicrobial activities, we performed DLS measurements of the
different MgF2 (1-4) samples. All the tested MgF2 particles
(samples 14) displayed similar aggregation behavior in
biological media (Figure S5). A clear time-dependent aggrega-
tion of MgF2 samples was observed. The average aggregate size
was similar in the two biological media, and the maximal value of
aggregate diameters was observed after 2 hours of incubation.
The average aggregate size varied and was dependant on the
initial particle size with the smallest aggregates corresponding to
the smallest particles. In both media, the diameters measured after
6 hours of incubation were around 150, 450, 1600, and 4500 nm
for MgF2 4,3,2, and MgF2 1, respectively (Figures S5, AB).
To deepen our understanding of this phenomenon, we
evaluated the nature of the aggregates. We observed a direct
correlation between the numbers of particles that constitute the
aggregates and the initial particle size (Figures S5, AB). In the
smallest particle sample (i.e., MgF24) the aggregates (150 nm
 J. Lellouche et al / Nanomedicine: Nanotechnology, Biology, and Medicine 8 (2012) 702711
Figure 5. MgF2 NP antimicrobial activity against established biofilms.
MgF2 NP antimicrobial activity against established biofilms. Viability of
(A) E. coli and (B) S. aureus pre-grown biofilms. Biofilms were grown
for a period of 1, 3, 5, and 7 days before the treatment and exposed to
MgF2 NP samples obtained by sonochemistry and microwave syntheses at
a concentration of 1 mg/mL for 24 h (for details see Methods section). The
percent viability compared with untreated control is presented in the left y-
axis. The dashed lines (right y-axis) refer to the number of viable cells in
the pre-grown biofilm without treatment. Error bars represent the SD of
three independent experiments.
on average) constituted approximately 30 individual particles
(Figures S5, CD). These results can be compared with the ratio
observed in the other MgF2 samples (MgF2-3, 2, and 1), in which
a smaller number of particles compose the aggregate (Figures
S5, CD). The best example to illustrate this phenomenon was
observed in the micronic MgF2-1 sample. In this sample,
aggregates with an average size of 4500 nm were composed of
only two distinct particles (Figure S5, C). Moreover, all types of
MgF2 samples behaved similarly, following a short ultrasound
pulse that was implemented to evaluate the reversibility qualities
of the MgF2 aggregates (Figures S5, CD). All MgF2 samples
broke up and provided an aggregate size to particle diameter
coefficient of approximately 1, which suggest that aggregates
broke down to single particles (Figures S5, CD).
Discussion
Crystalline MgF2 NPs (25 nm on average) have been
previously produced using a one-step, microwave-assisted
thermal decomposition process of room-temperature ionic liquid
BIMBF4 in the presence of [Mg(Ac)2(H2O)4]. The NPs showed
good antimicrobial and antibiofilm activities. However, this
709
Figure 6. Impact of size of MgF2 NPs on the antimicrobial activity. Growth
yields of (A) E. coli and (B) S. aureus grown in the presence of MgF2 NP
(0.01 mg/mL) suspension at different sizes (samples 14) for 24 h at 37C.
The initial inoculums were 10 7 CFU/mL (i.e., OD595 of 0.01) and growth
yield was monitored by measuring the optical density at 595 nm. E. coli and
S. aureus grown in TSB and TSB-Glu, respectively. Untreated bacteria
served as a control. Error bars represent the SD deviation of three
independent experiments.
method presents a few obvious limitations, mainly, the extreme
experimental temperature conditions that may reach the BIMBF4
boiling point (i.e., 212C). Furthermore, this indicated temper-
ature range is also close to the highest application-limit
temperature measured for the IL by long-term isothermal
thermogravimetric analysis (i.e., 240C). The high temperature
during microwave synthesis might also produce carbonaceous
impurities that are formed by the thermal decomposition of
BIM +1 onto the surface of the NPs. This was indeed
demonstrated, e.g., in the thermal microwave decomposition of
[Fe(III)NO39H2O] in BIMBF4 which afforded 300 nm-sized
FeF2 bar-shaped particles covered by a nanometer-thick adlayer
of amorphous carbon. 35 In agreement with these results,
elemental analyses performed on microwave-assisted synthe-
sized MgF2 NPs showed the same trend, i.e., the formation of
contaminating carbonaceous impurities (1.4% weight, Table S1).
Such carbonaceous impurities may also have indirect effects on
the physical properties of the NPs (e.g., aggregation) as well as
on their antimicrobial activity, as addressed below. An additional
downside associated with microwave synthesis is the fact that the
temperature range precludes the use of this process for the
deposition of MgF 2 particles onto temperature-sensitive
polymers. 36 Finally, the relatively high cost of ionic liquids
may make its use unattractive for commercial applications.
Based on these disadvantages associated with the microwave
technique, our study focused on the development of a room
temperature, water-based synthesis that will be more applicable
and less expensive.
The process we developed is based on the use of the
ultrasound irradiation of a [Mg(OAc)2(H2O)4] precursor in an
aqueous solution with hydrofluoric acid (HF) in a 1:2 molar ratio
(HF provides the fluorine anion source instead of ionic liquid).
The experimental protocol described herein has been optimized
regarding various experimental multiparameter variations that
included a systematic screening of tested combinations of
reagent ratios and concentrations and of medium temperatures
710 J. Lellouche et al / Nanomedicine: Nanotechnology, Biology, and Medicine 8 (2012) 702711
(data not shown). The optimal sonochemical synthesis protocol
yielded MgF2 NPs that showed size and crystalline properties
similar to those obtained previously with microwave synthesis. 20
The MgF2 NPs also showed good antimicrobial activity and
antibiofilm activity both in blocking biofilm formation or
treating mature biofilms. In fact, their antimicrobial activity
was even more potent than that observed for NPs synthesized
using the microwave process (mg/mL vs. g/mL concentrations
for microwave and ultrasonic syntheses, respectively; Figure 2).
The difference in activities cannot be attributed to the size of the
NPs because both NPs had a similar diameter. We hypothesize
that the reduction in the antimicrobial activity of the microwave-
synthesized MgF2 NPs may be due to surface modifications such
as the deposition of the amorphous carbon mentioned above.
Another possibility is differences in the aggregation pattern of
the respective NPs during incubation. 37-39
To test this hypothesis, we examined the aggregation of the
various MgF2 NPs in liquid growth media (without bacteria).
The growth medium contains proteins, sugars, and salts that may
adsorb physically onto the NP surface and induce aggregation
due to a decrease in their surface energy. 37,38 Aggregation was
measured at the highest NP concentration (1 mg/mL) using DLS.
Our results indicated that the MgF2 NPs that were synthesized
using sonochemistry showed less aggregation over time in
comparison with the NPs synthesized using microwave chem-
istry. The dispersal of the NPs synthesized by sonochemistry was
also better following a short sonication pulse. Such an
aggregation-related behavior of NPs in complex physiological
media may be simply based on the subsequent thermodynam-
ically driven equilibrium: isolated free NPs (f-NPs) aggregat-
ed NPs (a-NPs) [f-NPs a-NPs] where micrometer-sized a-NP
aggregates might co-exist with isolated free suspended f-NPs.
This equilibrium, however, might be readily displaced towards f-
NPs that break off from the aggregate during bacterial
interactions or due to shear forces. A major significance of our
findings is related to the fact that the MgF2 NPs' aggregates
(micrometer size) did not preclude any substantial antibacterial
activity, and that a clear size-dependent antimicrobial activity
was observed with the MgF2 NPs. Furthermore, the smaller
particles also produced the smallest aggregates, with the highest
aggregate-size to NP-size coefficient. Thus the reduction in size
provides aggregates with an increasing number of NPs. Taken
together with the ability of the aggregates to disperse, our results
point out that despite their aggregation mode, the nanometer size
of the individual particle is most likely the crucial factor in
killing the bacteria. This partly contradicts the well-accepted
irreversibility of the aggregation phenomenon of NPs. 37-40 To
the best of our knowledge, and as a significant model of NP
aggregation behavior, it is the first time that the antibacterial
activity of inorganic aggregated MgF2 NPs has been associated
with a reversible equilibrium between loosely aggregated a-NPs
and isolated f-NPs. Future work will be required to further
evaluate this phenomenon.
In conclusion, the present research characterized the anti-
bacterial and antibiofilm activities of crystalline MgF2 NPs
obtained by sonochemical synthesis. Using an ultrasonic
irradiation system, the conditions of the preparation of MgF2
NPs were optimized to obtain NPs that were effective at
millimolar concentrations. Antimicrobial activity was also
strongly dependent on the particle size for both bacteria, with
smaller sized NPs having more efficient antibacterial activity
than larger NPs.
We further utilized the ultrasonic-based synthesis procedure
to effectively coat glass surfaces with MgF2 NPs and
demonstrated that these surfaces can restrict the biofilm
formation of the studied bacteria. The NPs were also active
against mature biofilms. Our results revealed the importance of
the synthesis procedure on the aggregation pattern and its
physicochemical interactions. We demonstrated that the reduc-
tion in antimicrobial activity of the microwave synthesized MgF2
NPs may be due to surface modifications, such as the deposition
of amorphous carbon and strong aggregation. Taken together,
the results of this study emphasize the potential of using metal
fluoride NPs for the design of sterile surface coatings that may be
useful for various medical applications.
Acknowledgments
We thank Dr. Elena Degtyar, Aylana Reiss-Mendel and
Louise Braverman for their helpful discussion and critical review
of the manuscript. This research is part of the requirements for a
PhD thesis for Jonathan Lellouche at Bar Ilan University.
Appendix A. Supplementary data
Supplementary data to this article can be found online at
doi:10.1016/j.nano.2011.09.002.
References
1. Andersson DI, Hughes D. Antibiotic resistance and its cost: is it possible
to reverse resistance? Nat Rev Microbiol 2010;8:260-71.
2. Hoiby N, Bjarnsholt T, Givskov M, Molin S, Ciofu O. Antibiotic
resistance of bacterial biofilms. Int J Antimicrob Agents 2010;35:322-32.
3. Simoes M, Simoes LC, Vieira MJ. A review of current and emergent
biofilm control strategies. Lwt-Food Sci Technol 2010;43:573-83.
4. Ammons CBM. Anti-biofilm strategies and the need for innovations in
wound care. Recent Patents Anti-Infect Drug Disc 2010;5:10-7.
5. Richards JJ, Melander C. Controlling bacterial biofilms. Chembiochem
2009;10:2287-94.
6. Yoon KY, Byeon JH, Park JH, Hwang J. Susceptibility constants of
Escherichia coli and Bacillus subtilis to silver and copper nanoparticles.
Sci Total Environ 2007;373:572-5.
7. Hernandez-Sierra JF, Ruiz F, Cruz Pena DC, Martinez-Gutierrez F,
Martinez AE, de Jesus Pozos Guillen A, et al. The antimicrobial
sensitivity of Streptococcus mutans to nanoparticles of silver, zinc oxide,
and gold. Nanomed-Nanotechnol 2008;4:237-40.
8. Dae Hong J, Jun Sung K, Eunye K, Kyeong Nam Y, Jong-Ho K, Sung
Jin P, et al. Antimicrobial effects of silver nanoparticles. Nanomed-
Nanotechnol 2007;3:95-101.
9. Lee C, Kim JY, Lee WI, Nelson KL, Yoon J, Sedlak DL. Bactericidal
effect of zero-valent iron nanoparticles on Escherichia coli. Environ Sci
Technol 2008;42:4927-33.
10. Applerot G, Lipovsky A, Dror R, Perkas N, Nitzan Y, Lubart R, et al.
Enhanced antibacterial activity of nanocrystalline ZnO due to increased
ROS-mediated cell injury. Adv Funct Mater 2009;19:842-52.
11. Applerot G, Abu-Mukh R, Irzh A, Charmet J, Keppner H, Laux E, et al.
Decorating parylene-coated glass with ZnO nanoparticles for antibacter-
ial applications: a comparative study of sonochemical, microwave, and
 J. Lellouche et al / Nanomedicine: Nanotechnology, Biology, and Medicine 8 (2012) 702711
microwave-plasma coating routes. ACS Apll Mater Interfaces
2010;2:1052-9.
12. Fu GF, Vary PS, Lin CT. Anatase TiO 2 nanocomposites for
antimicrobial coatings. J Phys Chem B 2005;109:8889-98.
13. Kang S, Pinault M, Pfefferle LD, Elimelech M. Single-walled carbon
nanotubes exhibit strong antimicrobial activity. Langmuir 2007;23:
8670-3.
14. Zilberman M, Elsner JJ. Antibiotic-eluting medical devices for various
applications. J Control Release 2008;130:202-15.
15. Vasilev K, Cook J, Griesser HJ. Antibacterial surfaces for biomedical
devices. Expert Rev Med Devic 2009;6:553-67.
16. Streicher RM, Scbmidt M, Fiorito S. Nanosurfaces and nanostructures
for artificial orthopedic implants. Nanomedicine-UK 2007;2:861-74.
17. Shi ZL, Neoh KG, Kang ET, Wang W. Antibacterial and mechanical
properties of bone cement impregnated with chitosan nanoparticles.
Biomaterials 2006;27:2440-9.
18. Sandhiya S, Dkhar SA, Surendiran A. Emerging trends of nanomedicine -
an overview. Fundam Clin Pharmacol 2009;23:263-9.
19. Costerton JW, Montanaro L, Arciola CR. Biofilm in implant infections:
Its production and regulation. Int J Artif Organs 2005;28:1062-8.
20. Lellouche J, Kahana E, Elias S, Gedanken A, Banin E. Antibiofilm activity
of nanosized magnesium fluoride. Biomaterials 2009;30:5969-78.
21. Gazzano E, Bergandi L, Riganti C, Aldieri E, Doublier S, Costamagna C,
et al. Fluoride effects: the two faces of Janus. Curr Med Chem 2010;17:
2431-41.
22. Liang R, Yang MR, Zhou QX. Thermal stability, equilibrium vapor
pressure and standard enthalpy of vaporization of 1-butyl-3-methylimi-
dazolium tetrafluoroborate. Acta Phys-Chim Sin 2010;26:1468-72.
23. Valderrama JO, Robles PA. Critical properties, normal boiling
temperatures, and acentric factors of fifty ionic liquids. Ind Eng Chem
2007;46:1338-44.
24. Rao KJ, Vaidhyanathan M, Ganguli M, Ramakrishnan PA. Synthesis of
inorganic solids using microwaves. Chem Mater 1999;11:882-95.
25. Mastai Y, Gedanken A. Sonochemistry and other novel methods
developed for the synthesis of nanoparticles. Chem Inform 2006;37:
200-50.
26. Gedanken A. Preparation and properties of proteinaceous microspheres
made sonochemically. Chem-Eur J 2008;14:3840-53.
27. Bang JH, Suslick KS. Applications of ultrasound to the synthesis of
nanostructured materials. Adv Mater 2010;22:1039-59.
711
28. Shanks RMQ, Donegan NP, Graber ML, Buckingham ME, Zegans AL,
Cheung AL, et al. Heparin stimulates Staphylococcus aureus biofilm
formation. Infect Immun 2005;73:4596-606.
29. Souza Antunes AL, Trentin DS, Bonfanti JW, Ferreira Pinto CC,
Rodrigues Perez AJ, et al. Application of a feasible method for
determination of biofilm antimicrobial susceptibility in staphyllococci.
APMIS 2010;118:168-92.
30. Dewanti R, Wong ACL. Influence of culture conditions on biofilm
formation by Escherichia coli O157:H7. Int J Food Microbiol
1995;26:147-64.
31. Banin E, Lozinski A, Brady Keith M, Berenshtein E, Butterfield PW,
Moshe M, et al. The potential of desferrioxamine-gallium as an anti-
Pseudomonas therapeutic agent. Proc Natl Acad Sci U S A 2008;105:
16761-6.
32. Neal AL. What can be inferred from bacterium-nanoparticle interactions
about the potential consequences of environmental exposure to
nanoparticles? Ecotoxicology 2008;17:362-71.
33. Applerot G, Perkas N, Amirian G, Girshevitz O, Gedanken A. Coating of
glass with ZnO via ultrasonic irradiation and a study of its antibacterial
properties. Appl Surf Sci 2009;256:S3-8.
34. Murata T, Ishizawa H, Motoyama I, Tanaka A. Preparation of high-
performance optical coatings with fluoride nanoparticle films made from
autoclaved sols. Appl Opt 2006;45:1465-8.
35. Jacob DS, Bitton L, Grinblat J, Felner I, Koltypin Y, Gedanken A. Are
ionic liquids really a boon for the synthesis of inorganic materials? A
general method for the fabrication of nanosized metal fluorides. Chem
Mater 2006;18:3162-8.
36. Wen J. Heat capacities of polymers. In: Mark J, editor. Physical properties
of polymers handbook, 2nd ed, Vol. 1. Springer; 2007. p. 145-54.
37. Pacek AW, Ding P, Utomo AT. Effect of energy density, pH and
temperature on de-aggregation in nano-particles/water suspensions in
high shear mixer. Powder Technol 2007;173:203-10.
38. Li D, Kaner RB. Shape and aggregation control of nanoparticles: not
shaken, not stirred. J Am Chem Soc 2006;128:968-75.
39. Li D, Kaner RB. How nucleation affects the aggregation of
nanoparticles. J Mater Chem 2007;17:2279-82.
40. Fedorenko SV, Bochkova OD, Mustafina AR, Burilov VA, Kadirov
MK, Holin CV, et al. Dual visible and near-infrared luminescent silica
nanoparticles. Synthesis and aggregation stability. J Phys Chem C
2010;114:6350-5.