Cold Plasma Rapid Decontamination of Food

Contact Surfaces Contaminated with

Salmonella Biofilms
Brendan A. Niemira, Glenn Boyd, and Joseph Sites

Cross-contamination of foods from persistent pathogen reservoirs is a known risk factor in processing
environments. Industry requires a rapid, waterless, zero-contact, chemical-free method for removing pathogens from
food contact surfaces. Cold plasma was tested for its ability to inactivate Salmonella biofilms. A 3-strain Salmonella culture
was grown to form adherent biofilms for 24, 48, or 72 h on a test surface (glass slides). These were placed on a conveyor
belt and passed at various line speeds to provide exposure times of 5, 10, or 15 s. The test plate was either 5 or 7.5
cm under a plasma jet emitter operating at 1 atm using filtered air as the feed gas. The frequency of high-voltage
electricity was varied from 23 to 48 kHz. At the closer spacing (5 cm), cold plasma reduced Salmonella biofilms by up
to 1.57 log CFU/mL (5 s), 1.82 log CFU/mL (10 s), and 2.13 log CFU/mL (15 s). Increasing the distance to 7.5 cm
generally reduced the efficacy of the 15 s treatment, but had variable effects on the 5 and 10 s treatments. Variation of
the high-voltage electricity had a greater effect on 10 and 15 s treatments, particularly at the 7.5 cm spacing. For each
combination of time, distance, and frequency, Salmonella biofilms of 24, 48, and 72 h growth responded consistently with
each other. The results show that short treatments with cold plasma yielded up to a 2.13 log reduction of a durable form
of Salmonella contamination on a model food contact surface. This technology shows promise as a possible tool for rapid
disinfection of materials associated with food processing.

Keywords: biofilm, cold plasma, food safety, Salmonella, sanitizer

Pathogens such as Salmonella can form chemical-resistant biofilms, making them difficult to
remove from food contact surfaces. A 15 s treatment with cold plasma reduced mature Salmonella biofilms by up to 2.13
log CFU/mL (99.3%). This contact-free, waterless method uses no chemical sanitizers. Cold plasma may therefore have
a practical application for conveyor belts, equipment, and other food contact surfaces where a rapid, dry antimicrobial
process is required.

Practical Application:

Introduction
Foodborne contamination by bacterial pathogens is an ongoing concern for producers and consumers (Sivapalasingam and
others 2004; Mandrell 2009). Contamination of foods may occur
through direct contact with food contact surfaces such as containers, conveyor belts, knives, and so on (Doyle and Erickson 2008).
Pathogens such as Salmonella are known to persist on these surfaces, often as durable biofilms (Ryu and Beuchat 2005; Reisner
and others 2006). These close conglomerations of cells are more
resistant to removal and inactivation than free living planktonic
cells. This limits the effectiveness of conventional antimicrobial
chemical treatments in treating biofilm contamination (Niemira
and Solomon 2005; Ryu and Beuchat 2005). Compounding the
significance of biofilms for impact on human health, the native microflora that are capable of forming biofilms are not only widely
distributed in fresh-produce processing environments, but can support communal biofilms that include human pathogens (Jahid and
Ha 2012; Liu and others 2013).
MS 20131340 Submitted 9/20/2013, Accepted 12/23/2013. Authors are
with Food Safety and Intervention Technologies Research Unit, Eastern Regional Research Center, U.S. Dept. of Agriculture, Agricultural Research Service, Wyndmoor, PA 19038, U.S.A. Direct inquiries to author Niemira (E-mail:
brendan.Niemira@ars.usda.gov).

R

C 2014 Institute of Food Technologists

doi: 10.1111/1750-3841.12379
Further reproduction without permission is prohibited

Cold plasma (also known as cool plasma or nonthermal plasma)

is a relatively new antimicrobial process being developed for applications in the food industry (Niemira 2012a). The terminology
used in the literature (cold, cool, nonthermal, and so on) refers
to the physical parameter of operation at or near room temperature, rather than the thermal plasmas found in electric arc welders,
combustion tools or other high-temperature applications (Niemira
and Gutsol 2010). This technology has shown promise as a direct
treatment for fresh and fresh-cut fruits and vegetables, as well as
for nuts and other foods. Generally, high-voltage electricity or
other energy inputs are used to ionize gas molecules, thereby
imparting reactive properties. Cold plasma is waterless, uses no
antiseptic chemicals, and is contact-free. Given the reactive nature
of cold plasma chemistry, gas plasmas have long been used for surface treatment of such materials as electronics, textiles, polymers,
and print surfaces (Niemira and Gutsol 2010). Therefore, cold
plasma may be applicable to food contact surfaces susceptible to
contamination with human pathogens. Little data are available on
the antimicrobial efficacy of cold plasma applied directly to active
mature biofilms.
The objectives of this study are to determine the efficacy of very
rapid cold plasma treatments for inactivating Salmonella biofilms,
and to determine the impact of (a) distance from cold plasma emitter head, (b) frequency of cold plasma generation, and (c) biofilm
growth time on antimicrobial efficacy.
Vol. 79, Nr. 5, 2014 r Journal of Food Science M917

M: Food Microbiology
& Safety

Abstract:

Cold plasma on surfaces . . .

Materials and Methods

M: Food Microbiology
& Safety

inverted and incubated at 37 C for 24 h. Plates were examined

with an automated plate counter and populations enumerated.
Microorganisms
The plate counts were compared to those of the untreated conAll isolates utilized in this study were from the USDA-ARS- trol slides in order to determine log CFU/mL reduction. ExperERRC culture collection. Human outbreak strains of Salmonella iments performed in duplicate, with 3 plates per treatment, per
included S. Anatum F4317, S. Stanley H0558, and S. Enteri- replication.
tidis PT30. The isolates were maintained in tryptic soy broth
(TSB, Difco, Detroit, Mich., U.S.A.). Biofilms were grown using Statistics
the method of Niemira and Solomon (2005), with the following
For each combination of biofilm age, frequency, time, and dismodifications. Sterile glass slides, 7.62 cm 2.54 cm, were placed tance, values for reduction from control were pooled. Analysis of
into 30 mL of sterile TSB in 50 mL tubes. This was inoculated variance (ANOVA) was used to establish the significance of the
with 100 L stock solution of each Salmonella culture. The tubes various treatment factors, singly and in combination. Data were
were incubated at 37 C for 24, 48, or 72 h to allow biofilm evaluated using ANOVA and the StudentNewmanKeuls test
formation. On the day of treatment, biofilm-coated slides were (SigmaPlot 11, Systat Software, Chicago, Ill., U.S.A.).
aseptically removed from the tubes, rinsed for 10 s under a stream
of sterile distilled water to remove unattached cells and placed Results
in a biohood. They were allowed to air dry 10 min before cold
All short-duration treatments with cold plasma significantly replasma treatment. For each day, an untreated control slide was sim- duced Salmonella biofilms, with maximal reductions of 1.57 log
ilarly air-dried, held, and subjected to recovery and enumeration CFU/mL (5 s), 1.82 log CFU/mL (10 s), and 2.13 log CFU/mL
(described below).
(15 s). For 5 cm spacing treatments, reductions generally increased
with exposure time (Figure 1, 2, and 3). Treatments of 5 s were
Cold plasma treatment
consistently lower than those of 10 or 15 s. For most (but not
The equipment used was a modified version of a Dyne-A-Mite all) pulse frequencies examined, 15 s treatments gave higher reHP (Enercon Corp., Menomonee Falls, Wis., U.S.A.), described ductions than 10 s treatments. There was no clear association of
previously (Niemira 2012b). Briefly, this AC plasma jet device pulse frequency with treatment time in these samples. Increasing
is based on a form of gliding arc plasma. An ionizing potential the distance to 7.5 cm generally reduced the efficacy of the 15 s
is established across a 1 cm gap between 2 shaped electrodes, treatment, but had a variable effect on the 5 and 10 s treatments.
generating a plasma arc within a Teflon cowling. A flow of feed The observed reductions are not consistent with a regular pattern
gas (dried air) at 60 psi drives the plasma arc outward, expanding of response.
and cooling it. The version of the device used in the present study
Across all pulse frequencies examined, the obtained inactivation
has been modified to allow for variation of the electrical pulse was more consistent for the 5 s treatment than for the 10 and 15 s
frequency. It has been previously noted that varying the pulse treatments (Figure 1, 2, and 3). For the longer treatment times,
frequency can significantly impact the survival of bacterial exposed increasing the pulse frequency from 24 to 48 kHz led to a patto cold plasma (Alkawareek and others 2012). Various frequencies tern of reductions in which the most extreme responses (greatest
from 23 to 48 kHz were evaluated. Power consumption at these reductions compared with least reductions) were significantly diffrequencies ranged from approximately 522 to 549 W.
ferent from each other, but not consistently correlated with the
The remaining variables examined in this study were exposure frequency. The frequencies at which the extreme responses were
time (5, 10, or 15 s) and distance from the plasma jet emitter head obtained were not necessarily the same for each of the time and
(5 or 7.5 cm). These distances were chosen to place the biofilms distance combinations.
within the zones of active plasma (5 cm) or quenched plasma
However, for the 5 s treatment, increasing the pulse frequency
(7.5 cm). These correspond to areas closer to the electrodes and resulted in a pattern of reduction. At the 5 cm spacing, higher
within the plasma plume where most gas molecules are fully ion- frequencies led to lower inactivation of Salmonella biofilms, with
ized or in a region farther from the electrodes and beyond the the reduction at 48 kHz significantly less than that obtained at
plasma plume, where the most reactive plasma species have re- 24 kHz (Figure 1A). In contrast, at the wider 7.5 cm spacing,
combined during time-in-fight before interacting with the target higher frequencies resulted in greater inactivation, with the re(Niemira and Sites 2008; Niemira 2012a). A photographic pre- ductions at 35, 38, 42, and 48 kHz significantly greater than that
sentation of the shape and size of the plasma plume has previ- obtained at 24 kHz (Figure 1B).
ously been published (Niemira 2012b). Glass slides coated with
The age of the Salmonella biofilms had little significant effect on
Salmonella biofilms were placed on a mini-conveyor belt (Dorner the pattern of inactivation across combinations of treatment time,
Mfg., Hartland, Wis., U.S.A.). Time of exposure under the cold distance, and pulse frequency. Some specific combinations showed
plasma emitter head for each side of the glass slide was controlled slight variations, resulting in a significant difference among treatby varying the speed of the conveyor belt.
ment times, for example, 42 kHz at 7.5 cm for 24 h (Figure 1B)
compared with the same combination for 48 h (Figure 2B) or
Recovery and enumeration
72 h (Figure 3B). However, these differences were not large in
Cells were recovered and enumerated using the method of an absolute sense and did not conform to a consistent pattern of
Niemira and Solomon (2005). After treatment, the slides were response. Untreated controls gave plate counts of 7.3 to 7.6 log
placed in a fresh 50 mL capped tube with 10 mL buffered peptone CFU/mL across all biofilm ages examined.
water (BPW). The tubes were vigorously shaken for 60 s to detach
and recover cells from the surface of the glass. The resuspended Discussion
cells were serially diluted in BPW and spread plated onto TSA
Contamination of food by bacterial human pathogens continplates. Dilutions were chosen based on preliminary studies, with ues to be a significant concern, with Salmonella, Escherichia coli
a minimum detection limit of 2.0 log CFU/mL. The plates were O157:H7, and other shigatoxigenic E. coli, Listeria monocytogenes,
M918 Journal of Food Science r Vol. 79, Nr. 5, 2014

Cold plasma on surfaces . . .

argon gas at 1 atm, Bermudez-Aguirre and others (2013) was able
to reduce E. coli on lettuce and tomato by 1.6 log CFU/mL after a
10 min treatment at 12.83 kV. On apples treated for 3 min with a
15 kV gliding arc cold plasma system using air at 1 atm, Salmonella
was reduced by up to 3.7 log CFU/mL and E. coli was reduced
by up to 3.5 log CFU/mL (Niemira and Sites 2008). A study
of inoculated almonds employing the same apparatus used herein
showed reduction of E. coli and Salmonella of up to 1.34 and 1.16
log CFU/mL, respectively (Niemira 2012b). Reductions in that
study were obtained after rapid treatments of 20 s.
The mode of action for cold plasma inactivation is
dependent on the specific technology used and the processing
conditions, but are not yet clearly understood (Alkawareek and
others 2012; Niemira 2012a). Different kinds of cold plasma systems can operate at atmospheric pressures or under low pressure.
Overall, UV light and direct interactions with reactive chemical species are the primary mechanisms. However, the relative

Figure 1Cold plasma reduction of 24 h

biofilms of Salmonella, treated for 5 s
(black circle), 10 s (gray triangle), or 15 s
(white square). Treated surfaces were 5 cm
(A) or 7.5 cm (B) from the plasma emitter
head. Within each plasma pulse frequency
(kHz), reductions with the same letter are
not significantly different (P < 0.05,
ANOVA). Bars indicate standard error.

Vol. 79, Nr. 5, 2014 r Journal of Food Science M919

M: Food Microbiology
& Safety

and Shigella spp. being of chief concern (Sivapalasingam and others

2004; FDA 2009; CSPI 2010). Costs associated with foodborne
illness have been estimated to be between $34 and $39 billion (IFT
2004; Scharff 2010). There is a pressing need for rapid, effective
treatments that can eliminate these pathogens from foods and food
contact surfaces. Such a technology must also be practical for commercial implementation with respect to worker safety, economics
of operation, and capital costs.
Cold plasma has emerged in recent years as a promising technology for use in the food processing industry (Niemira 2012a).
Different cold plasma systems vary in their electrode configuration, voltages, feed gas compositions, physical conformations, and
gas pressures. The wide range of technical aspects of different cold
plasma systems can make comparisons difficult (Niemira 2012a).
Nevertheless, demonstrations of antimicrobial efficacy using these
varied systems suggest potentially useful applications of these technologies. For example, using a needle-mesh electrode array, with

Cold plasma on surfaces . . .

antimicrobial contribution of each varies based on the type of
technology tested (Niemira 2012a). UV is a key factor in lowpressure cold plasma systems (Lassen and others 2005; Tran and
others 2008; Sureshkumar and Neogi 2009), whereas ambient
pressure systems have reported little or no effect from UV (Gweon
and others 2009; Machala and others 2010). The application of
cold plasma to foods directly must balance antimicrobial efficacy
against the potential for sensory or quality impacts resulting from
the treatment. However, an important area for potential contamination is food contact surfaces within the processing environment
(Doyle and Erickson 2008). Therefore, the use of cold plasma
as a potential antimicrobial process for treatment of food contact
surfaces holds promise.
In the present study, cold plasma was able to rapidly inactivate
Salmonella biofilms on a model food contact surface, using a conveyor belt treatment system. The treatment times used in this study
were chosen to be consistent with those that may be most practical in a commercial field harvest or food processing operation.
Reductions of up to 2.13 log CFU/mL were obtained after a single 15 s pass-through treatment at a spacing of 5 cm between the

cold plasma emitter head and the biofilm surface. Shorter times
gave lower reductions. In a study using a 6 kV plasma of 99.5%
helium and 0.05% oxygen, Alkawareek and others (2012) used a
15 s treatment at 20 or 40 kHz. Pseudomonas aeruginosa biofilms
were reduced by 0.8 or 1.6 log CFU for the 2 tested frequencies,
respectively. However, in the present study, which used air as the
cold plasma feed gas, no such relationship of frequency and antimicrobial efficacy was observed. This highlights the importance
of the design and operating conditions of the cold plasma equipment. It is likely that optimization of the process for commercial
implementation will require balancing the antimicrobial efficacy
with the speed at which the process can operate. As organic material may be present in a commercial processing environment,
future research should incorporate this as part of the development
toward optimization.
Varying the distance between the emitter head and the treated
surface resulted in a variation in the antimicrobial reduction.
Closer spacing was generally more effective at reducing the
pathogen, with increasing the distance from 5 to 7.5 cm serving to reduce the efficacy of the longer treatments. Also, at

M: Food Microbiology
& Safety

Figure 2Cold plasma reduction of 48 h biofilms

of Salmonella, treated for 5 s (black circle), 10 s
(gray triangle), or 15 s (white square). Treated
surfaces were 5 cm (A) or 7.5 cm (B) from the
plasma emitter head. Within each plasma pulse
frequency (kHz), reductions with the same letter
are not significantly different (P < 0.05,
ANOVA). Bars indicate standard error.

M920 Journal of Food Science r Vol. 79, Nr. 5, 2014

Cold plasma on surfaces . . .

the cold plasma treatment at all frequencies, treatment times, and
distances. Previous studies have shown that biofilms produced using the methodologies used herein can reduce the antimicrobial efficacy of sanitizing compounds and nonthermal processing
technologies. In studies of biofilms of Salmonella (Niemira and
Solomon 2005), E. coli O157:H7 (Niemira 2007) and L. monocytogenes, and L. innocua (Niemira 2010), the response of biofilms to
antimicrobial measures varied significantly based on such factors
as the biofilm age, growth temperatures, and culture conditions.
However, this variability of response was not observed in the
present study. Before treatment, biofilms herein were allowed to
dry in a flow of moving air for 10 min, thus rendering them a more
durable form of the biofilms and thereby presenting what would
be expected to be more a substantial resistance to the cold plasma
process. Nevertheless, the biofilm-associated pathogens were
reduced by multiple log CFU/mL without contact, abrasion, or
other physical augmentation of the process. This suggests that cold
plasma may be an effective tool for treating biofilm-associated
organisms, including those that may be inaccessible to normal

Figure 3Cold plasma reduction of 72 h

biofilms of Salmonella, treated for 5 s (black
circle), 10 s (gray triangle), or 15 s (white
square). Treated surfaces were 5 cm (A) or
7.5 cm (B) from the plasma emitter head.
Within each plasma pulse frequency (kHz),
reductions with the same letter are not
significantly different (P < 0.05, ANOVA).
Bars indicate standard error.

Vol. 79, Nr. 5, 2014 r Journal of Food Science M921

M: Food Microbiology
& Safety

7.5 cm, the interaction between plasma pulse frequency and antimicrobial reduction was more complicated, particularly at the
higher frequencies examined. Therefore, it may be that systems
that rely on active plasma applied directly to the surfaces to be
treated will provide the most consistent results. The distances used
in this study correspond to a zone of active plasma (5 cm) and a
zone where the active plasma has given way to quenched plasma
containing a greater proportion of only the longest lived chemical species (7.5 cm). Active plasma is the most reactive form of
cold plasma and is therefore expected to have the highest antimicrobial efficacy. More research is needed to understand the
specific chemistries involved within the active plasma itself as
well as with the mixture of active/quenched plasma and ambient air. Such research will need to address the high humidity expected to be present in the ambient air in food processing
environments.
In this study, the age of the biofilm was not a significant factor in altering the antimicrobial efficacy of the process. Salmonella
biofilms of 24, 48, and 72 h old responded in similar ways to

Cold plasma on surfaces . . .

sanitation procedures. Microscopy observation of the biofilms preand post-treatment with cold plasma was not within the scope of
this study, but is will be an important area of future research to elucidate modes of action for the responses observed. Also, biofilms
produced using flow reactors, rotating biofilm reactors, or other
methodologies (McLean and others 2004; Liu and others 2013)
would provide additional information on the interaction of biofilm
structure with the susceptibility to cold plasma inactivation. More
research is needed to adapt the technology used herein for the
particular needs of commercial food processing equipment that
may be particularly susceptible to biofilm contamination. Also, as
response to nonthermal interventions can be isolate-dependent,
the response of industrial isolates of Salmonella, as well as clinical
isolates, should be evaluated.
Reductions in pathogen viability were achieved without antimicrobial chemicals, physical contact, or the addition of water
or other solutions. It is likely that combining cold plasma with a
physical scrub or chemical intervention will enhance the reductions seen herein. The arc discharge system in this study presents
a flexible plasma plume, which suggests that it would be suitable
for a variety of food contact surfaces such as belts (unitary and
flexi-belts), equipment, knives, and so on.

M: Food Microbiology
& Safety

Conclusions
Short treatments with cold plasma significantly inactivated
Salmonella biofilms by up to 2.13 log CFU/mL with very short
(15 s) treatment times. Closer spacing generally yielded more
effective inactivation. Variation of pulse frequency in cold plasma
generation resulted in a complex variability of antimicrobial efficacy. The antimicrobial efficacy of the treatments was consistent
regardless of the maturity of the Salmonella biofilms treated. Cold
plasma shows promise as a possible tool for waterless, rapid disinfection of food contact materials associated with food processing.

Acknowledgments
The authors would like to thank Ms. Janysha Taylor for her expert technical assistance, and thank Drs. M. Olanya and A. Sheen
for their reviews of this manuscript. Mention of trade names or
commercial products in this publication is solely for the purpose of
providing specific information and does not imply recommendation or endorsement by the U.S. Dept. of Agriculture. The USDA
is an equal opportunity employer.

References
Alkawareek MY, Algwari QT, Laverty G, Gorman SP, Graham WG, OConnell D, Gilmore
BF. 2012. Eradication of pseudomonas aeruginosa biofilms by atmospheric pressure non-thermal
plasma. PLoS ONE 7(8):e44289. doi:10.1371/journal.pone.0044289.

M922 Journal of Food Science r Vol. 79, Nr. 5, 2014

Bermudez-Aguirre D, Wemlinger E, Pedrow P, Barbosa-Canovas G, Garcia-Perez M. 2013.

Effect of atmospheric pressure cold plasma (APCP) on the inactivation of Escherichia coli in
fresh produce. Food Control 34:14957.
Center for Science in the Public Interest (CSPI). 2010. Behind CSPIs outbreak data: a
look at the produce outbreak numbers. Available from: http://www.cspinet.org/foodsafety/
produce_data.pdf. Accessed 2013 August 1.
Doyle M, Erickson MC. 2008. Summer meeting 2007 the problems with fresh produce: an
overview. J Appl Microbiol 105:31730.
Gweon B, Kim DB, Moon SY, Choe W. 2009. Escherichia coli deactivation study controlling the
atmospheric pressure plasma discharge conditions. Curr Appl Phys 9:6258.
Institute of Food Technologists (IFT). 2004. Bacteria associated with foodborne diseases. J Food
Sci 58:201.
Jahid IK, Ha SD. 2012. A review of microbial biofilms of produce: future challenge to food
safety. Food Sci Biotechnol 21:299316.
Lassen KS, Bolette N, Reinar G. 2005. The dependence of the sporicidal effects on the power
and pressure of RF-generated plasma processes. J Biomed Mater Res B: Appl Biomater
74B(1):5539.
Liu NT, Lefcourt AM, Nou X, Shelton DR, Zhang G, Lo YM. 2013. Native microflora in
fresh-cut produce processing plants and their potentials for biofilm formation. J Food Prot
5(6):82732.
Machala Z, Chladekova L, Pelach M. 2010. Plasma agents in bio-decontamination by DC
discharges in atmospheric air. J Phys D: Appl Phys 43:2220018.
Mandrell RE. 2009. Enteric human pathogens associated with fresh produce: sources, transport
and ecology. In: Fan X, Niemira BA, Doona CJ, Feeherry FE, Gravani RB, editors. Microbial
safety of fresh produce Ames, Iowa: Blackwell Publ. p 541.
McLean RJC, Bates CL, Barnes MB, McGowin CL, Aron GM. 2004. Methods of studying
biofilms. In: Ghannoum M, OToole GA, editors. Microbial biofilms. Washington, DC: ASM
Press. p 379413.
Niemira BA. 2007. Irradiation sensitivity of planktonic and biofilm-associated Escherichia coli
O157:H7 isolates is influenced by culture conditions. Appl Environ Microbiol 73(10):
323944.
Niemira BA. 2010. Influence of growth temperature on irradiation sensitivity of planktonic
and biofilm-associated Listeria monocytogenes and L. innocua. Food and Bioprocess Technol
3(2):25764.
Niemira BA. 2012a. Cold plasma decontamination of foods. Annu Rev Food Sci Technol
2012(3):12542.
Niemira BA. 2012b. Cold plasma reduction of Salmonella and Escherichia coli O157:H7 on
almonds using ambient pressure gases. J Food Sci 77:M1715.
Niemira BA, Gutsol A. 2010. Nonthermal plasma as a novel food processing technology.
In: Zhang HQ, Barbosa-Canovas G, Balasubramaniam VM, Dunne P, Farkas D, Yuan
J, editors. Nonthermal processing technologies for food. Ames, Iowa: Blackwell Pub.
p 27188.
Niemira BA, Sites J. 2008. Cold plasma inactivates Salmonella Stanley and Escherichia coli O157:H7
inoculated on golden delicious apples. J Food Prot 71(7):135765.
Niemira BA, Solomon EB. 2005. Sensitivity of planktonic and biofilm-associated Salmonella to
ionizing radiation. Appl Environ Microbiol 71:27326.
Reisner A, Krogfelt KA, Klein BM, Zechner EL, Molin S. 2006. In vitro biofilm formation of
commensal and pathogenic Escherichia coli strains: impact of environmental and genetic factors.
J Bacteriol 188:357281.
Ryu J-H, Beuchat LR. 2005. Biofilm formation by Escherichia coli O157:H7 on stainless steel:
effect of exopolysaccharide and curli production on its resistance to chlorine. Appl Environ
Microbiol 71:24754.
Scharff RL. 2010. Health-related costs from foodborne illness in the United States.
Produce Safety Project. Available from: http://www.producesafetyproject.org/admin/
assets/files/Health-Related-Foodborne-Illness-Costs-Report.pdf-1.pdf. Accessed 2013
September 20.
Sivapalasingam S, Friedman CR, Cohen L, Tauxe RV. 2004. Fresh produce: a growing cause
of outbreaks of foodborne illness in the United States, 1973 through 1997. J Food Prot
67:234253.
Sureshkumar S, Neogi S. 2009. Inactivation characteristics of bacteria in capacitively coupled
argon plasma. IEEE Trans Plasma Sci 37(Suppl. 2):234752.
Tran N, Amidi M, Sanguansri P. 2008. Cool plasma for large scale chemical-free microbial
inactivation of surfaces. Food Aust 60:3447.
U.S. Food Drug Admin (FDA). 2009. Potential for infiltration, survival and growth of human pathogens within fruits and vegetables. Available from: http://www.fda.gov/Food/
GuidanceRegulation/HACCP/ucm082063.htm. Accessed 2013 August 1.