See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/230970690
Low pressure plasma discharges for the sterilization and decontamination of
surfaces
Article in New Journal of Physics · November 2009
DOI: 10.1088/1367-2630/11/11/115017
CITATIONS
READS
76
155
5 authors, including:
Francois Rossi
Ondřej Kylián
353 PUBLICATIONS 6,250 CITATIONS
132 PUBLICATIONS 1,726 CITATIONS
European Commission, Joint Research Centre, Petten, The Netherlands
SEE PROFILE
Charles University in Prague
SEE PROFILE
Hubert Rauscher
Nina Hasiwa
130 PUBLICATIONS 2,526 CITATIONS
38 PUBLICATIONS 1,184 CITATIONS
European Commission
SEE PROFILE
Atax-Advice
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
NanoReTox View project
Nanotechnology View project
All content following this page was uploaded by Francois Rossi on 10 June 2014.
The user has requested enhancement of the downloaded file.
Home
Search
Collections
Journals
About
Contact us
My IOPscience
Low pressure plasma discharges for the sterilization and decontamination of surfaces
This article has been downloaded from IOPscience. Please scroll down to see the full text article.
2009 New J. Phys. 11 115017
(http://iopscience.iop.org/1367-2630/11/11/115017)
View the table of contents for this issue, or go to the journal homepage for more
Download details:
IP Address: 139.191.135.230
The article was downloaded on 02/08/2011 at 09:20
Please note that terms and conditions apply.
New Journal of Physics
The open–access journal for physics
Low pressure plasma discharges for the sterilization
and decontamination of surfaces
F Rossi1,3 , O Kylián2 , H Rauscher1 , M Hasiwa1 and D Gilliland1
1
European Commission, Joint Research Centre, Institute for Health and
Consumer Protection, Via E. Fermi 2749, 21027 Ispra (VA), Italy
2
Faculty of Mathematics and Physics, Charles University, V Holešovičkách 2,
Prague 8, 180 00, Czech Republic
E-mail: francois.rossi@jrc.ec.europa.eu
New Journal of Physics 11 (2009) 115017 (33pp)
Received 20 March 2009
Published 26 November 2009
Online at http://www.njp.org/
doi:10.1088/1367-2630/11/11/115017
The mechanisms of sterilization and decontamination of surfaces are
compared in direct and post discharge plasma treatments in two low-pressure
reactors, microwave and inductively coupled plasma. It is shown that the removal
of various biomolecules, such as proteins, pyrogens or peptides, can be obtained
at high rates and low temperatures in the inductively coupled plasma (ICP)
by using Ar/O2 mixtures. Similar efficiency is obtained for bacterial spores.
Analysis of the discharge conditions illustrates the role of ion bombardment
associated with O radicals, leading to a fast etching of organic matter. By
contrast, the conditions obtained in the post discharge lead to much lower
etching rates but also to a chemical modification of pyrogens, leading to their deactivation. The advantages of the two processes are discussed for the application
to the practical case of decontamination of medical devices and reduction
of hospital infections, illustrating the advantages and drawbacks of the two
approaches.
Abstract.
3
Author to whom any correspondence should be addressed.
New Journal of Physics 11 (2009) 115017
1367-2630/09/115017+33$30.00
© IOP Publishing Ltd and Deutsche Physikalische Gesellschaft
2
Contents
1. Introduction
2. Principles of low-pressure plasma-based sterilization and decontamination
2.1. Bacterial spore sterilization . . . . . . . . . . . . . . . . . . . . . . . . .
2.2. Destruction of pyrogens . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.3. Elimination of proteins . . . . . . . . . . . . . . . . . . . . . . . . . . .
3. Experimental
3.1. Plasma treatments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.2. Coating of proteins and amino acids . . . . . . . . . . . . . . . . . . . .
3.3. Coatings of bacterial endotoxins . . . . . . . . . . . . . . . . . . . . . .
3.4. Bacterial spores . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4. Results obtained with the ICP discharge
4.1. Model case of BSA . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.2. Application of Ar/O2 plasma for elimination of other biological species .
4.3. Concluding remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5. Results obtained with the MW discharge
5.1. Etching of proteins . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.2. Bacterial endotoxins . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.3. Concluding remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6. Conclusions
Acknowledgments
References
. . .
. . .
. . .
.
.
.
.
.
.
.
.
.
.
.
.
. . .
. . .
. . .
. . .
. . .
. . .
2
4
4
5
5
6
6
9
10
11
12
12
21
22
25
25
26
28
29
31
31
1. Introduction
In the US, approximately 46.5 million surgical procedures and even more invasive medical
procedures—including approximately 5 million gastrointestinal endoscopies—are performed
each year. Each procedure involves contact between a surgical instrument and a patient’s sterile
tissue or mucous membranes. These procedures present a major risk of contamination by
the introduction of pathogens during the operation. Failure to properly disinfect or sterilize
equipment not only carries a risk associated with the breach of host barriers but also a risk
of person-to-person transmission (e.g. hepatitis B virus) and transmission of environmental
pathogens (e.g. Pseudomonas aeruginosa).
Multiple studies in different countries have documented a lack of compliance with
established guidelines for disinfection and sterilization. Failure to comply with scientifically
based guidelines has led to numerous outbreaks. In 2002, the estimated number of Hospital
Acquired Infections (HAI) in the US, was approximately 1.7 million. The estimated deaths
associated with HAIs in US hospitals were about 10 000 [1] at a cost of the order of 5–7 billion
dollars. Similar figures were also reported for the UK, France and Italy, with 100 000 HAI
leading to about 5000 deaths in 2000 for the UK [2], 4000 for France [3] and 5000–7000 in
Italy [4].
Although these cases cannot all be directly attributed to contamination present on
surgical instruments, it is generally recognized that disinfection and sterilization are essential
New Journal of Physics 11 (2009) 115017 (http://www.njp.org/)
3
for ensuring that medical and surgical instruments do not transmit infectious pathogens to
patients. These operations are normally realized by placing the instrument in a mechanical
washer/disinfector and subjecting them to a pre-wash at room temperature to prevent blood
coagulation and adhesion of proteins. The instruments are then cleaned and sonicated with an
alkaline enzymatic detergent, and finally, at high temperature, washed and rinsed. They are then
visually inspected after drying and packaged for sterilization (in most cases by autoclaving, but
also by use of chemical reagents such as hydrogen peroxide or EtO). These operations normally
have to follow strict rules such as the ones described in the norm ISO EN 15883.
However, recent studies made in the UK have shown that visual inspection is not enough
to prevent significant quantities of residues to be left on surgical instruments after complete
cleaning operations [5]–[8]. In these studies made on real instruments decontaminated in a
Sterile Service Unit, all items tested were showing residual contamination, sometimes severe
with levels larger than 4 µg mm−2 locally. These residues were composed of salts, proteins and
undefined organic matter. This observation raises serious concerns since it has been shown that
prion protein can accumulate in peripheral and skeletal tissues of patients having the different
forms of Creutzfeld Jacob Disease (CJD) and could be potentially transmitted iatrogenically by
contaminated medical devices.
Another concern linked to these residues is related to pyrogens coming from both gramnegative (lipopolysaccharides (LPS)) and gram-positive (peptidoglycans or lipoteichoic acids)
bacteria. These pyrogens can cause sepsis when in contact with the patient bloodstream and
provoke septic shock, which is a major cause of death among hospitalized patients. They are,
however, extremely resistant to temperature (e.g. [9, 10]) and difficult to remove by conventional
sterilization procedures (e.g. [11]).
Therefore new methods of decontamination and sterilization, which could ensure complete
removal of bacteria, pyrogens and proteins, are urgently desired. As we will see, the application
of non-equilibrium plasma discharges currently appears as an interesting option. This is mainly
due to the capacity of plasma processes to remove organic material with high efficiency, while
working at low temperature and using non-toxic gases, thus reducing both environmental
impacts and safety risks. These key advantages naturally result in increased interest in the
investigations of the plasma interactions with substances of biological origin and microorganisms.
Indeed, it has been shown that many pathogens can be destroyed by employing nonequilibrium plasma discharges (e.g. [12]–[14]) and the nature of plasma interactions with microorganisms or biomolecules has been extensively studied in the past few years, with aim (i) to
identify the dominant processes leading to the desired effect, (ii) for treatment optimization
and (iii) validation of the process. However, it has to be noted that this is a rather challenging
objective: plasma discharges are capable of producing high fluxes of various neutral or ionized
active species as well as energetic photons, which all interact with treated biological systems via
different mechanisms that contribute to the overall efficiency of the treatment. The exact role of
each of these processes naturally depends on many parameters such as e.g. process conditions
(pressure, applied power, gas flow and excitation frequency), geometry of the system as well
as on the kind of target to be inactivated; this considerably complicates the comparison of the
results obtained in different experimental configurations and leads to controversy concerning the
mechanisms at play during the process. However, as will be outlined in the following section,
recent results allow us to draw certain general conclusions regarding possible strategies of
plasma-based sterilization and decontamination of surfaces. Subsequently, some of the aspects
New Journal of Physics 11 (2009) 115017 (http://www.njp.org/)
4
related to the proposed approaches will be discussed on the basis of experiments employing
low-pressure plasma discharges and involving various kinds of biological systems, with the
objective to demonstrate their advantages as well as limitations and drawbacks.
2. Principles of low-pressure plasma-based sterilization and decontamination
2.1. Bacterial spore sterilization
The main mechanisms of spore destruction or deactivation by plasma discharge have been the
object of several reviews (e.g. [14]–[19]) that the reader should refer to for more details. In
brief, two main mechanisms can be invoked for spore inactivation or destruction employing
low-pressure plasma discharges4 :
• The first one is related to the inactivation of spores by interaction of the UV photons emitted
by the discharge with the DNA (e.g. [20]–[22]). It was found that the maximum efficiency
of the UV photons is typically in the wavelength range 200–300 nm [22, 23], and that there
exists a minimum dose necessary for preventing DNA repair [24] that leads to sterilization.
• The second mechanism is linked to spore etching and erosion by the radicals and active
species produced by the plasma discharge (e.g. atomic oxygen, atomic nitrogen and OH
radicals or fluorine atoms [25]–[32]). The etching leads to the direct killing of spores by
destruction of their membrane and to the removal of the material that shields the spores
from exposure to UV radiation. Both effects result in an acceleration of the sterilization
rate.
Nevertheless, it is clear that in common situations, both processes may act simultaneously
during the treatment and thus contribute to the killing/deactivation of spores. The rates of the
two mechanisms are often markedly different, which leads to the two phases of kinetics reported
in the literature (e.g. [12, 15]). Moreover, the significance of these two processes is strongly
dependent on the local plasma conditions, and as a consequence, on the position of treated
object with respect to the plasma discharge: whereas position in the active plasma favors spore
etching promoted by ion bombardment, placing samples in the discharge afterglow increases
the relative role of UV radiation since ions are normally absent in these conditions.
However, it must also be emphasized that basing sterilization on the action of UV only has
two serious drawbacks: the first one is connected to the distribution of the spores on the surface:
in practical cases, the spores might be stacked and mixed with protein residues or biological
film. With the UV photons flux decreasing exponentially with the thickness of the film they
go through, the time needed for reaching the minimum dose necessary for spore inactivation
increases exponentially. This makes sterilization difficult to achieve in practical cases with this
approach and therefore some etching that uncovers shielded spores is normally necessary for
acceptable treatment durations [14]. The second limitation is that UV photons have little or no
effect on other contamination sources such as pyrogens or proteins, which results in a treatment
that addresses only a part of the problem.
This is why we deliberately favor the approach where etching contributes significantly to
the decontamination treatment. We will see that this approach has the advantage of sterilizing
4
It has to be noted that also other mechanisms have been suggested as for instance electrostatic disruption of the
outer cells walls [18] or their killing by ozone [19]. However, the significance of these processes is more important
for discharges operated at higher pressures than the ones reported here.
New Journal of Physics 11 (2009) 115017 (http://www.njp.org/)
5
the surface even when significant soiling is present, and it also removes the other contaminants
such as pyrogens or proteins at low temperatures and with limited treatment durations.
2.2. Destruction of pyrogens
As mentioned before, another surface contamination, overlooked most of the time, is
represented by pyrogens. The outer coats of spores and bacteria contain endotoxins (e.g. LPS,
peptidoglycans or lipoteichoic acids, etc), which are pyrogenic substances. They act as potent
modulators of the human immune system, and their presence in the blood stream leads to
physiological events such as fever, swelling or sepsis and, at higher doses, to death [33]. UV
radiation in the 200–300 nm range has little effect on such substances, as we have already
demonstrated [34]. Moreover, after a plasma treatment based on UV radiation only, the surface
treated can have an increased pyrogenic activity due to the endotoxins liberated by the spores,
and can consequently provoke an unexpected immunological response [35]. Nevertheless, it
has been recently demonstrated that distinct kinds of pyrogenic substances can be inactivated
readily by plasma treatment (e.g. [14, 34, 36, 37]).
2.3. Elimination of proteins
Another group of possible contaminants is constituted by protein residues, which might contain
pathogens, such as the infectious prions associated with the transmission of the CJD. Unlike the
case of bacterial spores, prions do not contain genetic material and UV radiation is ineffective
for their destruction. Moreover, prions have been found to be extremely resistant toward the
conventional sterilization and decontamination techniques (e.g. [38, 39]) because of their unique
and stable secondary and ternary structure that cannot be easily altered.
Although the possibility to remove prions by means of non-equilibrium plasma discharges
has already been demonstrated by Baxter et al [40], the studies focusing on their removal
mechanisms are rather limited. This is primarily because of the necessity of high containment
laboratories due to the biohazard connected with such proteins. Therefore, in order to understand
the mechanisms of plasma–prions interactions, the effects of non-equilibrium plasma discharges
on non-pathogenic models of proteins were studied (e.g. [14], [41]–[48]) using different
methods for monitoring their elimination. The results of these studies reveal general important
facts regarding the action of low-pressure plasma discharges on proteins, and show the
possibility to remove them by oxygen-containing discharges, which induce their fragmentation
and volatilization after oxidation. Nevertheless, there is still a lack of systematic analysis related
to the protein removal efficiency under different experimental conditions as well as to the
identification of the principal mechanism of their elimination.
The results published in the literature lead us to the conclusion that low pressure plasma
processes, operated in conditions leading to high etching rates of organic matter, are the route
to pursue for the development of a treatment able to both sterilize and decontaminate surface of
medical devices, with a number of caveats that will be discussed at the end of this paper.
In order to get a better insight into the possibilities and limitations of low pressure plasma
discharges, two types of processes (direct and post discharge) will be presented and their effect
on different types of contaminations will be discussed.
New Journal of Physics 11 (2009) 115017 (http://www.njp.org/)
6
Figure 1. Schematics of the ICP reactor.
3. Experimental
3.1. Plasma treatments
Two different types of plasma reactors have been used in the present work.
The first type of discharge used is a planar double coil inductively coupled plasma (ICP)
source schematically depicted in figure 1. This plasma reactor allows igniting plasma discharges
with a power between 100 and 500 W, in a pressure range from 0.1 to 20 Pa and gas flows from
1 to 20 sccm.
The processing chamber has a volume of approximately 5 liters and is evacuated by primary
and turbomolecular pumps. The gas mixture used for the discharge is controlled by MKS mass
flow controllers. The biological samples were placed 20 mm from the bottom quartz window.
The second experimental set-up used is a microwave plasma reactor schematically depicted
in figure 2. It consists of a cylindrical stainless steel vacuum chamber having a volume of
approximately 12 liters, equipped with several diagnostics windows and one port for sample
introduction. The processing chamber is fed from a gas handling system composed of mass
flow controllers attached to the process gas lines and is evacuated by a primary pump and a roots
blower. The pressure in the chamber during plasma operation is regulated by an MKS butterfly
valve. The plasma discharge is generated at a total gas flow of 100 sccm at a pressure of 16 Pa
by a microwave source (excitation frequency 2.45 GHz, applied power 1000 W) introduced into
the plasma chamber through a silica window placed at the end of a circular 100 mm waveguide.
In contrast to the ICP set-up, the samples to be treated were placed outside the active plasma
zone at a distance of 300 mm from the quartz window, i.e. in the near post discharge, where
plasma densities are low, as indicated in table 1.
New Journal of Physics 11 (2009) 115017 (http://www.njp.org/)
7
Figure 2. Schematics of the microwave plasma reactor.
Table 1. Electron density and electron temperature at samples position in used
reactors (MW: 1000 W, 100 sccm, 16 Pa; ICP: 350 W, 20 sccm, 10 Pa).
Ne (cm−3 )
Gas mixture
Reactor
Ar
O2
N2
O2 /N2 (50 : 50)
Te (eV)
ICP
MW
ICP
MW
(8.3 ± 0.1) ⇥ 1011
(2.0 ± 0.3) ⇥ 1010
(1.8 ± 0.2) ⇥ 1010
(1.8 ± 0.1) ⇥ 1010
(2 ± 0.1) ⇥ 109
(2.5 ± 0.4) ⇥ 109
(1.2 ± 0.1) ⇥ 109
(1.5 ± 0.3) ⇥ 109
1.7 ± 0.1
2.7 ± 0.2
2.6 ± 0.2
2.6 ± 0.1
1.03 ± 0.03
1.1 ± 0.2
0.55 ± 0.04
0.8 ± 0.2
In order to characterize the plasma discharges produced in both reactors, different
diagnostics methods were used, namely a Langmuir probe (SmartProbe™; Scientific Systems
Ltd) placed at the samples positions, optical emission spectroscopy (Avantes AVS-PC2000
monochromator equipped with a 2048-element linear CCD array) and mass spectrometry
(HAL4, Hiden Analytical). Moreover, the substrate temperature during plasma treatment has
been determined by IR pyrometry (Raynger MX4, Raytek).
The two reactors differ strongly by the conditions produced and the processes applied.
In the microwave (MW) reactor, the samples are placed outside of the active plasma zone and
submitted mostly to a flux of neutrals and radicals, while in the discharge in the ICP reactor, they
are also subjected to a significant ion bombardment. In other words, although in both reactors
comparable fluxes of active species reach the samples, which can be demonstrated for instance
by actinometric measurements of O atom density (see figure 3), the fluxes of charged particles
to the sample differ enormously, as can be seen from figure 4 and table 1, where the plasma
characteristics measured at the samples’ position are given for both reactors.
Moreover, the two reactors cause markedly different heating of the treated samples. For
instance, whereas temperature of Si wafer after 1 min of plasma duration in Ar was found to
be 180 ◦ C in the ICP reactor, only a slight increase of the substrate temperature was observed
in the MW reactor even after 10 min of plasma duration (see figure 5). Such a difference in
the heating of the samples could affect significantly the kinetics of the etching of biological
New Journal of Physics 11 (2009) 115017 (http://www.njp.org/)
8
Figure 3. Emission spectra of Ar/O2 9 : 1 discharges (MW: 1000 W, 100 sccm,
16 Pa; ICP: 300 W, 20 sccm, 10 Pa).
Figure 4. Langmuir probe characteristic measured in Ar discharges at the
samples’ position (MW: 1000 W, 100 sccm, 16 Pa; ICP: 300 W, 20 sccm, 10 Pa).
samples or the sterilization process in general (e.g. [49]–[51]). Therefore, in order to limit
this side effect, preliminarily tests were undertaken to find out a treatment time allowing us
to neglect the heating influence. Based on these experiments performed at a variable treatment
duration, it has been found that the onset of the nonlinearity of the etching rate with treatment
time, which can be attributed to the temperature effect, starts when the substrate is heated to the
temperature of approximately 100 ◦ C, that corresponds to the treatment duration around 30 s for
ICP discharge operated at 300 W (see figure 6). Therefore, for all the etching tests performed in
the ICP reactor the treatment duration was kept below 30 s (i.e. in the case of longer treatment
times, the plasma treatment was performed in several consecutive steps).
New Journal of Physics 11 (2009) 115017 (http://www.njp.org/)
9
Figure 5. Temporal evolution of substrate temperature in Ar and O2 plasma
(MW: 1000 W, 100 sccm, 16 Pa; ICP: 300 W, 20 sccm, 10 Pa).
Figure 6. Etching of BSA for plasma treatment having variable duration (ICP
reactor, Ar/O2 20 : 1 mixture, 10 Pa, 300 W). The numbers indicate substrate
temperature reached after end of each of the treatment steps.
3.2. Coating of proteins and amino acids
The samples were prepared by spotting an aqueous solutions (0.1% wt) of different selected
substances, namely bovine serum albumin (BSA), lysozyme, ubiquitin as representatives of
proteins and poly-L-histidine, on one-side polished, cleaned Si wafers. This substrate has
been chosen to allow a precise analysis of the decontamination mechanisms, but represents
an ideal case as compared with stainless steel substrates that present many surface defects
New Journal of Physics 11 (2009) 115017 (http://www.njp.org/)
10
Figure 7. 2D map of ellipsometric angle 9 of the BSA sample.
after repeated usage. After deposition, the samples were dried overnight in a common flow
hood. During the drying of the droplet, the deposit forms a thin central part with a thickness
of about 100 nm, surrounded by a thicker ring (coffee ring effect) as can be seen in figure 7
for the BSA sample. Both treated and untreated samples were subsequently examined by
various surface diagnostic methods: stylus profilometry (Alpha-step IQ, KLA-Tencor), atomic
force microscopy (AFM type Solver P47H, NT-MDT Co.) and ToF-SIM spectroscopy (TOFSIMS IV, IONTOF). Finally, the samples were visualized by imaging ellipsometry, which is a
diagnostic method based on the determination of changes in polarized light upon its reflection
from a scanned surface and allowing us to detect deposited material having thickness as low as
a few nanometers. Ellipsometry measurements were performed using a variable angle multiwavelength imaging ellipsometer (EP3, Nanofilm Surface Analysis GmbH) in air at room
temperature at a wavelength of 554.3 nm at an angle of incidence of 42◦ and a field of view of
2000 µm ⇥ 2000 µm. A conventional PCSA (Polarizer–Compensator–Sample–Analyzer) nullellipsometric procedure is used to obtain two-dimensional (2D) maps of ellipsometric 1 and 9
angles [52]. In the following, the evolution of the 1 and 9 angles are used as a semi-quantitative
indication of the organic layer removal.
3.3. Coatings of bacterial endotoxins
The efficiency of the plasma treatment in terms of etching of pyrogenic biomolecules was
studied with LPS, peptidoglycan (PGN) and lipid A, the latter constituting the major pyrogenic
component of LPS. The samples were prepared following the same protocol used for deposition
of proteins and poly-L-histidine, i.e. by spotting small droplets of 0.1% wt aqueous solution
of these substances on polished Si wafers. The untreated samples as well as samples exposed
to plasma discharges were examined by the same surface diagnostic methods used for proteins
and poly-L-histidine. However, it has to be noted that in contrast to the protein samples, the
deposits of bacterial endotoxins were rather spatially inhomogeneous (see figure 8), which made
profilometric removal measurements unreliable.
New Journal of Physics 11 (2009) 115017 (http://www.njp.org/)
Y [µm]
11
Figure 8. 2D map of ellipsometric angle 9 of the lipid A sample.
In addition, the biological activity of the surface (i.e. its pyrogenicity) was evaluated by
the ‘whole blood test’, which measures the production of interleukin IL-1β, induced by the
contact between the contaminated surface and blood cells [53, 54]. To create a controlled
surface contamination by pyrogens, 24 well plates were incubated with 100 µl of pyrogen
solution (LPS, lipid A and PGN diluted in a range of 0.01–10 ng ml−1 for LPS and lipid A,
and 0.1–10 µg ml−1 for PGN). The plates were dried overnight in a flow hood, exposed to the
plasma discharge and afterwards tested by the whole blood incubation the next day in order to
estimate remaining biological activity of the deposit. The surface biological activity was then
evaluated before and after treatment by measuring the IL-1β cytokine release from human whole
blood sample placed in contact with the surface treated. An enzyme-linked immunosorbent
assay (ELISA) is used to measure the release of IL-1β, which is detected with commercially
available antibody pairs (RD systems, Space Import-Export srl.). Detection of the biotinylated
antibody is quantified by streptavidin-peroxidase (Biosource, Prodotti Gianni SpA) and its
substrate TMB (3,30 ,5,50 -tetramethylbenzidine; Sigma). Recombinant cytokines were used as
standards (National Institute of Biological Standard and Control (NIBSC), South Mimms). Full
details on the procedure can be found in [54].
3.4. Bacterial spores
The effect of plasma discharges on bacterial spores was studied using stainless steel coupons
covered with a known number (about 2.5 ⇥ 106 ) of Geobacillus stearothermophilus spores
(Raven Biological). As can be seen in figure 9, the spores on coupons are not disposed in a
monolayer but are stacked and shielded by organic matter, which has a major impact on the
treatment efficiency and kinetics.
Moreover, it has to be stressed that since the main aim of the present paper is to evaluate
the capability of different discharges to erode or etch biological contamination, the treatment
efficiency reported here refers solely to the degree of morphological changes of spores exposed
to plasma, which was determined by scanning electron microscopy (SEM LEO 435VP). To the
reader interested in further details concerning biological tests, we can recommend extensive
literature devoted to this topic [12, 13, 15, 16, 24].
New Journal of Physics 11 (2009) 115017 (http://www.njp.org/)
12
Figure 9. SEM image of untreated G. stearothermophilus.
4. Results obtained with the ICP discharge
The main objective of this section is to present a comprehensive study of plasma interaction with
biological systems with emphasis to their etching processes. In order to meet this objective, the
section is subdivided.
First, the experiments focused on the conditions leading to the fastest reduction of
deposited biological matter as well as on the identification of possible mechanisms contributing
to this process are presented. In these experiments, the etching rates of BSA selected as a model
substance are measured under different operational conditions (discharge mixture composition,
power and pressure). These results are then compared with the properties of the discharges
determined by in situ plasma diagnostics, which enables estimation of roles of different agents
leading to BSA removal.
Subsequently, the operational conditions identified to be the optimal for BSA elimination
are used for the plasma processing of other biological samples to test the universality of such
treatment on a wider set of different biological materials.
4.1. Model case of BSA
4.1.1. Screening tests. After the plasma treatment with an Ar/O2 ICP discharge, visible
modifications of the protein samples were observed indicating gradual removal of the protein
deposit. An example of this behavior is given in figure 10, where the ellipsometric 2D maps of
the 1 angles of a BSA sample before and after plasma treatment are presented.
Moreover, the etching rate of the deposit, evaluated by surface profilometry performed on
the rings of the BSA spot, was found to be strongly dependent on the gas mixture used. This is
demonstrated in figure 11, where the results of screening tests performed in pure argon, and its
binary and ternary discharge mixtures with hydrogen, nitrogen and oxygen are summarized. It
can be clearly seen that Ar, Ar/H2 and Ar/N2 discharges lead to approximately the same rate
of BSA elimination (around 200 nm min−1 ), whereas BSA is removed markedly faster in Ar/O2
plasma at otherwise identical operational conditions (i.e. pressure, power and gas flow). The
removal rates obtained in this last case were higher than 600 nm min−1 , showing importance
of oxygen for fast volatilization of proteins. Moreover, it can be seen as well that etching
efficiency of Ar/O2 discharge is not enhanced either by nitrogen or by hydrogen addition. This
also indicates that other molecular radicals produced in such plasmas, like OH or NO, whose
presence has been confirmed by OES (figure 12), do not contribute significantly to the process
of BSA elimination.
New Journal of Physics 11 (2009) 115017 (http://www.njp.org/)
13
Figure 10. 2D maps of the 1 angles of BSA (upper row) and corresponding
cross-sections (lower row) before treatment (a) and after 15 s (b) and 120 s (c) of
treatment in Ar/O2 20 : 1 mixture—protein dilution 1 mg ml−1 , Si wafer.
Figure 11. BSA etching rates—protein dilution 1 mg ml−1 , Si wafer, plasma
treatment 10 Pa, 200 W, 22 sccm.
New Journal of Physics 11 (2009) 115017 (http://www.njp.org/)
14
Figure 12. Optical emission spectra of Ar/O2 /N2 20 : 0.5 : 0.5 (upper) and
Ar/O2 /H2 20 : 0.5 : 0.5 (lower) plasma discharges (10 Pa, 200 W, 22 sccm).
4.1.2. Ar /O2 plasma discharge—process optimization. As demonstrated above, the BSA
removal rate has been found to be the highest in Ar/O2 mixture as compared with the discharges
sustained in other gas mixtures. In order to identify the optimal treatment conditions, the etching
rate was measured as a function of argon/oxygen ratio, pressure and applied RF power.
Regarding the dependence of removal rate on discharge mixture composition, it has been
found that the addition of oxygen into argon leads initially to a strong enhancement of the
BSA removal rate. Nevertheless, further increase of oxygen fraction in the Ar/O2 mixture
above approximately 10–15% results in a gradual decrease of removal efficiency as depicted
in figure 13. Moreover, it can be seen that the pressure dependence of BSA etching rate changes
with discharge mixture composition: whereas in argon-dominated mixtures increase of pressure
causes increase of etching rate, the opposite trend has been observed in oxygen-rich mixtures.
Finally, the etching rate has been found to rise substantially with the applied RF power. For
instance, a power increase from 200 to 300 W resulted in almost double the BSA removal rate
in Ar/O2 20 : 1.
4.1.3. Ar /O2 ICP plasma discharge—study of etching mechanism
4.1.3.1. Influence of O atoms. As mentioned above, the capability of ICP discharges to
eliminate BSA protein is strongly dependent on the presence of oxygen in the discharge mixture.
Therefore, the main mechanism leading to the protein removal could be chemical etching. In
order to test this hypothesis, O atom density in Ar/O2 discharges has been measured by mass
New Journal of Physics 11 (2009) 115017 (http://www.njp.org/)
15
Figure 13. Etching rate of BSA as a function of Ar/O2 discharge mixture
composition (total gas flow 20 sccm).
Figure 14. O atom density and BSA removal rate in Ar/O2 (300 W, 10 Pa,
20 sccm).
spectroscopy and the obtained results were compared with BSA etching rates. It can be seen
in figure 14 that the density of O atoms increases monotonically with rising O2 content in
the discharge mixture in contrast to the etching rate, which shows a well-defined maximum
for Ar/O2 ratio around 9 : 1. This finding therefore suggests that pure chemical etching is not
dominant, or at least is not the only mechanism that causes BSA removal.
However, changing the composition of Ar/O2 discharge mixture not only causes changes
in the O atom density, but also in the density of positive ions as will be discussed later in the
text. Therefore, to exclude the possible effect of ions, additional experiments were performed
at fixed plasma density and variable O atom density, i.e. conditions achieved by simultaneously
decreasing the argon over oxygen ratio and increasing the applied RF power. The results of
New Journal of Physics 11 (2009) 115017 (http://www.njp.org/)
16
Figure 15. O atom density and etching rate at constant Ni (10 Pa, total gas flow
20 sccm).
Figure 16. Ion density and BSA removal rate in the Ar/O2 discharge (300 W,
total gas flow 20 sccm).
these experiments revealed that the rate of BSA removal follows concentration of O atoms in
the discharge (figure 15), which clearly shows that there exists a correlation between these two
quantities.
4.1.3.2. Influence of charged particles. The removal of biological material from surfaces could
be attributed to sputtering of the deposit by charged particles. However, this mechanism does not
seem to be the major cause of the observed effects: the density of charged particles dramatically
decreases with increasing O2 fraction in the discharge mixture (see figure 16), which is in
contrast to the evolution of the etching results. Furthermore, the energies of ions, measured by
ion mass spectrometry, are too low to enable effective sputtering of BSA protein (figure 17):
based on the preliminary experiments using ion beams with variable ions energy, effective
organic material sputtering requires ion energy beyond 150–200 eV [55, 59].
New Journal of Physics 11 (2009) 115017 (http://www.njp.org/)
17
Figure 17. Energy distribution of positive ions (Ar/O2 20 : 1, 10 Pa, 200 W).
Figure 18. Comparison of ion density and the rate of BSA removal at constant
O atom density (10 Pa, total gas flow 20 sccm).
Analogous to the case of evaluating the effect of O atoms, further experiments were
performed at constant O atom density and variable plasma density. Such conditions were
achieved by increasing applied RF power at fixed Ar/O2 mixture composition and pressure,
as verified both by mass spectrometry and OES. The results obtained in this set of experiments
are summarized in figure 18, which shows that the rate of BSA removal follows plasma density.
Since O atom density variations in these experiments are negligible, this result indicates the
important contribution of charged particles to the process of protein elimination.
4.1.3.3. Combined effect of O atoms and ions. As demonstrated in the previous sections, the
variation of the BSA removal rate with composition of the Ar/O2 discharge mixture cannot be
fully explained either by pure chemical etching connected with atomic oxygen or by sputtering
New Journal of Physics 11 (2009) 115017 (http://www.njp.org/)
18
Figure 19. Optical emission spectra of Ar/O2 20 : 1 plasma discharge with and
without BSA sample in the ICP reactor acquired after 10 s of plasma duration
together with their difference.
of the protein by energetic ions: it has been found that atomic oxygen alone does not contribute
significantly to the etching of the films, while Ar sputtering is observed only at energies much
higher than those measured in our experiments. Nevertheless, it was found that O atoms and
ions together contribute to the process and therefore it can be assumed that their synergetic
action is behind the observed behavior of the BSA removal rate. One possible reaction scheme
combining the effects of O etching and impact of ions is the process of chemical sputtering. In
this reaction scheme, the role of ions is a cleavage of bonds in the protein structure. This occurs
by the displacement of atoms caused by impact of energetic ions, which leads to the creation
of dangling bonds and thus to the rise of radical sites close to the protein surface. Such created
sites react with impinging oxygen atoms leading to the formation of volatile products (such as
CO, NO and OH) that are indeed observed in the emission spectra in the initial stage of plasma
treatment (figure 19). These two sequences have to occur simultaneously because the passivation
of an open bond by O species needs to happen faster than recombination in the cascade volume.
In other words, cleavage of bonds in the protein structure by ions facilitate surface reaction of
O atoms and enhance their natural capability to volatilize organic substances.
Obviously, the efficiency of this process is governed by fluxes of both ionic species and
oxygen atoms to the protein surface, and the ratio of ion/atom fluxes need to be high to observe
this effect. Whereas O atom density, and thus the flux of O atoms, increases with increasing O2
in the discharge mixture (figure 14), leading to an enhancement of the removal of proteins, the
density of charged particles steeply decreases at the same time (figure 16). The latter reduces
the creation of defects in the protein film and consequently lowers the number of active reaction
sites, thereby limiting the etching rate of the film. These two opposing effects can explain the
appearance of a well-defined maximum of the etching rate at the conditions representing an
optimal trade-off between the radicals’ production and ion density.
This reaction scheme has also been observed in experiments performed on bacterial
spores [56]–[58] and proteins [59] with Ar ions and O neutral beams applied for mimicking
plasma effects. These experiments demonstrated the synergetic effect of O radicals and ions on
New Journal of Physics 11 (2009) 115017 (http://www.njp.org/)
19
Figure 20. AFM images of 1 ⇥ 1 µm section of BSA surfaces (a) before treat-
ment and after 1 min treatment in (b) argon and (c) Ar/O2 20 : 1 mixture (plasma
treatment 10 Pa, 200 W).
decontamination of surfaces, further supporting the role of chemical sputtering in the removal
of biological films by low pressure plasma discharges.
4.1.4. Detailed characterization of treated samples. In order to better characterize changes of
chemical composition and morphology of protein samples, further experiments were performed
using atomic force microscopy (AFM) and time-of-flight secondary ion mass spectroscopy
(ToF-SIMS) on protein films before and after treatment.
It has been found that the etching of the protein films is not homogenous but accompanied
by a strong increase of their roughness (see figure 20) indicating a spatial inhomogeneity of the
etching process.
Moreover, a detailed investigation of temporal evolution of root mean square (rms)
roughness of BSA samples exposed to plasma discharges revealed three distinct phases (see
figure 21):
• Initially the rms roughness increases relatively slightly with the amount of removed
material (denoted as phase I).
• After removal of a sufficient quantity of the protein film, a second phase starts, which is
characterized by an enhanced increase of the rms roughness with increasing amount of
removed BSA.
• Finally, as soon as the amount of removed material reaches a value close to the initial
thickness of protein deposit, the surface roughness starts to decrease with further protein
removal.
Whereas the transition between phases II and III, observed also in previous study [60], can
be easily identified as the point at which underlying Si substrate is locally reached by plasma
treatment, the transition between phases I and II remains unclear. This phenomenon has already
been observed in other studies, for instance, during the plasma etching of polymer resist, where
it was attributed to an amplification of initially random defects created at the surface, related
to a local preferential sputtering due to the change of angle of incidence of ions impinging the
defects’ surface when the surface topography evolves.
Another possibility can be related to the presence of inorganic compounds or impurities
in the protein native structure or preparation (sulphur, sodium, calcium, etc). Plasma discharge
in the first phase volatilizes carbon-containing species, but due to the presence of inorganic
New Journal of Physics 11 (2009) 115017 (http://www.njp.org/)
20
Figure 21. Variation of rms roughness with the total amount of etched material.
Figure 22. ToF-SIMS 2D maps of distribution of C2 H6 N and Ca on samples
treated in Ar/O2 20 : 1 mixture (10 Pa, 200 W).
compounds that are distributed in the deposit, certain parts of protein surface will be enriched
in such compounds, which are much more difficult to remove. Subsequent plasma action will
therefore remove preferably the parts not covered by these inorganic residuals. The delayed
onset of the phase II can be—under this hypothesis—caused by time needed for the formation
of local sites with high density of non-volatile residuals.
Although another series of dedicated experiments are still necessary for identification of
process leading to the observed time evolution of surface roughens of treated samples, the
presence and densification of non-organic compounds with treatment time has been confirmed
by means of ToF-SIMS analysis. As is demonstrated in figure 22, the organic compounds
New Journal of Physics 11 (2009) 115017 (http://www.njp.org/)
21
Figure 23. Temporal variation of height decrease of BSA, lysozyme, ubiquitin
and poly-L-histidine (left panel) and corresponding removal rates (right panel)
(Ar/O2 20 : 1, 10 Pa, 200 W).
characteristic for proteins (C, H, O and N containing compounds) gradually disappeared from
mass spectra with plasma treatment, the fraction of inorganic compounds increases and finally
such compounds represent the only remaining material.
4.2. Application of Ar/O2 plasma for elimination of other biological species
In order to demonstrate the capability of Ar/O2 ICP plasma discharge optimized for BSA to
remove other biological deposits, further experiments were performed using different proteins,
as well as poly-L-histidine, lipid A and bacterial spores. To limit the heating of the samples
during their plasma treatment, the applied RF power was decreased to 200 W.
Regarding the plasma etching efficiency of different proteins and poly-L-histidine, i.e. one
elemental building blocks of proteins, we observe only relatively small differences in terms of
their etching rates as depicted in figure 23. However, these differences can be caused by variable
presence of inorganic compounds—for instance poly-L-histidine solution was prepared from
HCl-containing powders—whose presence could influence the removal efficiency. Moreover,
it should be noted that a recent study performed on different homopolymers of amino acids
revealed that their etching rates are very similar in spite of their distinct chemical properties [61].
This finding, i.e. similarity of the etching rates of various biomolecules, supports the proposed
mechanism of chemical sputtering. For the initiation of the destruction of biomolecules, active
sites have to be formed on their surfaces. In the case of pure chemical etching in oxygen
containing mixtures, this is achieved by hydrogen abstraction from the polymeric structure,
which is a process highly dependent on the exact chemical structure of treated substance. In
contrast, in the case of chemical sputtering, the active sites are generated by bond cleavage
caused by impact of energetic ions, which is rather insensitive to the chemical composition of
treated biomolecules in this energy range. This in turn makes the exact chemical structure of
treated substance irrelevant for resulting etch rate as observed in our experiments.
As mentioned above, the samples of bacterial endotoxins were highly spatially
inhomogeneous and therefore it was not possible to estimate their removal rate directly.
New Journal of Physics 11 (2009) 115017 (http://www.njp.org/)
22
Figure 24. 2D maps of ellipsometric angle 1 and corresponding cross-sections
of lipid A sample (a) untreated (b) treated 10 s and (c) sample treated 270 s in
Ar/O2 20 : 1 plasma (10 Pa, 200 W).
However, imaging ellipsometry showed that similarly to the previous cases, lipid A and PGN
can be readily removed from the surface by Ar/O2 plasma as can be seen in figures 24 and 25,
which show 2D maps of ellipsometric angles 1 of untreated and plasma treated samples together
with their typical cross-sections.
Finally, it has been found that bacterial spores can also be significantly eroded by Ar/O2
plasma discharge as demonstrated in figure 26. This finding clearly demonstrates, contrary to
the statements made previously [62], that etching can significantly contribute to the sterilization
process; etching can explain fast inactivation of bacterial spores in those plasma discharges
where only limited intensity of UV radiation is produced.
Nevertheless, our results also revealed a significant difference between plasma treatment of
biomolecules and spores: namely temporal nonlinearity of the etching rate and the much lower
values obtained in the case of spores (see figure 27). This is most likely due to much higher
fraction of inorganic compounds in the spores’ outer walls than in the protein samples, which
reduces the efficiency of the process.
4.3. Concluding remarks
It has been demonstrated that the ICP plasma discharge is capable of effectively removing or
eroding a wide range of biomolecules having distinctly different chemical structures as well as
bacterial spores representing highly resistant types of micro-organisms. Regarding the rates of
New Journal of Physics 11 (2009) 115017 (http://www.njp.org/)
23
Figure 25. 2D maps of ellipsometric angle 1 and corresponding cross-sections
of PGN sample (a) untreated (b) treated 10 s and (c) sample treated 270 s in
Ar/O2 20 : 1 plasma (10 Pa, 200 W).
these processes, their highest values can be achieved in an Ar/O2 discharge mixture. It has been
shown that this is a result of not only high fluxes of O atoms reaching the treated samples but
also by the action of ions bombarding their surfaces and thus promoting creation of surface
active sites. Moreover, it should be noted that this discharge mixture was already found to
inactivate effectively bacterial spores (e.g. 6 log reduction of surviving spores was reported to be
achieved within 40 s of plasma treatment under similar operational conditions [61]). Therefore,
the plasma treatment based on the Ar/O2 discharge mixture has all the pre-conditions to be a
highly universal sterilization and decontamination method.
In spite of these highly encouraging results, certain drawbacks of the process based solely
on extensive etching of pathogens have to be mentioned.
Firstly, it has been demonstrated that Ar/O2 plasma is not well suited for volatilization
of inorganic compounds. This results in severe implications for application of this approach in
a real situation, when the pathogens are for instance mixed with salts originating from body
fluids or contain such compounds per se, as in the case for bacterial spores. The markedly
lower etching rates connected with presence of inorganic compounds implies the necessity of
longer treatment duration that consequently represents risk of degradation of treated objects
by their extensive heating. The effect of heating can be significantly reduced either by pulsing
the discharge, or—as suggested recently for the case of bacterial spores—by application of
ternary Ar/O2 /N2 [63, 64], or Ar/NO [65] mixtures providing not only high etching rates but
also intense UV radiation capable of inactivating spores effectively. However, the persistence of
New Journal of Physics 11 (2009) 115017 (http://www.njp.org/)
24
Figure 26. SEM images of (a) untreated G. stearothermophilus spores and spores
treated for (b) 30 s (c) 60 s and (d) 120 s (Ar/O2 20 : 1, 10 Pa, 200 W).
Figure 27. Comparison of etching of BSA and G. stearothermophilus spores
(Ar/O2 20 : 1, 10 Pa, 200 W).
New Journal of Physics 11 (2009) 115017 (http://www.njp.org/)
25
Figure 28. Comparison of BSA removal rates in ICP and MW reactors (ICP:
300 W, 10 Pa, 20 sccm; MW: 1000 W, 13 Pa, 100 sccm).
remaining inorganic material, whose relevance for the safety of medical instruments still needs
to be investigated, seems to be the limiting factor of the process based on oxygen-containing
mixtures.
Secondly, the conditions suitable for effective etching of biological samples can cause
etching also of surfaces of polymer-based instruments, or at least could induce significant
alterations of their chemical and physical properties. For instance, recent comparison of the
effect of different sterilization methods on polyethylene glycol coatings revealed that their
exposure to plasma leads to a much faster decrease of their non-fouling character as compared
with traditional sterilization techniques [66]. Therefore, other approaches have to be developed
to limit these undesirable effects.
5. Results obtained with the MW discharge
As mentioned above, basing sterilization and decontamination of surfaces solely of etching
has certain drawbacks, connected mainly with undesirable etching of polymeric substrates.
Moreover, it has been demonstrated that for fast etching of biomolecules, high flux of ions on
their surfaces is necessary. Therefore in order to reduce the possible impact of plasma processes
on the treated objects, further experiments were performed using a MW plasma reactor with
samples placed downstream of the active discharge.
5.1. Etching of proteins
As expected, placing samples outside the active plasma zone has a strong impact on the etching
of biomolecules: whereas in the case of the ICP reactor the BSA etching rates were typically
higher than 200 nm min−1 , the values obtained in the case of a post discharge were not greater
than several nanometers per minute as can be seen in figure 28, which compares BSA etching
rates in different gas mixtures.
New Journal of Physics 11 (2009) 115017 (http://www.njp.org/)
26
Figure 29. O atom density and ion density in O2 /H2 MW discharge (1000 W,
13 Pa, 100 sccm).
Moreover, the optimal discharge mixture in terms of BSA removal differs for the two
employed plasma discharges: whereas in the case of ICP discharge, the fastest BSA removal
was achieved in Ar/O2 mixture having low fraction of oxygen, the discharge offering highest
efficiency of removal of BSA in the MW reactor is the one sustained in an O2 /H2 mixture.
However, unlike the case of ICP discharge, the increased etching rate after hydrogen addition
into oxygen discharge cannot be explained by chemical sputtering, since both O atom density
and ion density decrease with increasing amounts of hydrogen as can be seen in figure 29.
Therefore, the observed tendency has to be explained purely by action of chemically active
radicals produced under such conditions. In order to volatilize organic compounds, active sites
have to be created on the deposit surface first, which can be done either by impact of ions as
described in the previous section, or by abstraction of hydrogen from the polymeric structure
of treated material. This can proceed by chemical reaction with atomic oxygen, but also by
reaction with H atoms or highly reactive OH radicals, both effectively produced in O2 /H2
plasma discharge (see figure 30), which can in turn explain the observed results.
5.2. Bacterial endotoxins
Similarly to BSA protein, the fastest removal rate of bacterial endotoxins can be achieved in
the MW reactor in O2 /H2 mixture as reported previously [14]. However, the rate of this process
is again markedly lower than that obtained in the ICP reactor. Nevertheless, biological tests
performed on untreated and treated samples showed significant decrease of their bio-activity
after their exposure to the plasma discharge. Concretely, as can be seen in figure 31, the
bioactivity of 1 ng of coated LPS decreased after 5 min of the treatment to the value lower
than that corresponding to the pyrogenicity of 0.001 ng of the initially deposited material.
In other words, 5 min of plasma treatment is sufficient to cause decrease of LPS bioactivity
corresponding to more than 3 log lower than initial contamination.
Similar depyrogenation efficiency was observed for other hydrogen containing discharge
mixtures (see figure 32) that offers even markedly lower etching rates of pyrogens than that
New Journal of Physics 11 (2009) 115017 (http://www.njp.org/)
27
Figure 30. Optical emission spectra of O2 /H2 50 : 50 MW discharge (1000 W,
13 Pa, 100 sccm).
Figure 31. LPS pyrogenicity before and after its treatment in O2 /H2 50 : 50 MW
discharge (1000 W, 13 Pa, 100 sccm).
obtained in O2 /H2 mixture [14]. This is a finding of high importance, since it clearly shows
that depyrogenation of surfaces can be reached not only by physico-chemical elimination of
pyrogenic substances from surfaces as in the case of ICP discharge but also by alteration of
their chemical structure.
In order to evaluate the chemical changes leading to the suppression of the pyrogens’
bioactivity, ToF-SIMS analysis was performed on treated and untreated Lipid A, i.e. pyrogenic
center of LPS. As can be seen in figure 33 and in detail in figure 34, treatment in hydrogen
containing mixtures cause significant changes in the ToF-SIMS spectra, namely rapid decrease
New Journal of Physics 11 (2009) 115017 (http://www.njp.org/)
28
Figure 32. Lipid A pyrogenicity before and after 60 s of its treatment in O2 /H2
50 : 50 MW discharge (1000 W, 13 Pa, 100 sccm).
of the peaks originating from the fatty acid chains (e.g. C12 H23 O2 and C14 H27 O2 ) as well as
substantial alterations of phosphoryl groups (PO, PO2 , PO3 and PH2 O4 ), i.e. two parts of lipid
A structure suggested to govern its bioactivity.
However, the mechanism leading to such modification is still not clear, and it should be
noted that a similar effect has been observed after irradiation of lipid A samples by intense
VUV radiation emitted by a deuterium lamp having the peak of radiation intensity at 162 nm
(see figure 35), which shows that VUV radiation removes peaks corresponding to the whole
lipid A molecule as well as its major fragments. Therefore, one possible explanation refers to
the degradation of lipid A induced by intense VUV radiation emitted by hydrogen-containing
plasma discharges.
5.3. Concluding remarks
As was demonstrated on the example of the MW reactor, placing samples to the plasma
afterglow reduces drastically the removal rate of the biomolecules. This effect is produced
predominantly by the significant decrease of ion flux reaching surface of the samples, which in
turn leads to the lowering of the importance of chemical sputtering that plays a major role in the
active discharge. However, this effect can be seen from two different points of view, depending
on the particular application. It is obvious that lowering etching rates of organic materials slows
down the efficiency of the sterilization and decontamination process and thus represents a clear
disadvantage as compared with the treatment done in the active plasma zone. In contrast, this
approach lowers adverse impact of the treatment on the integrity and properties of processed
instruments. Moreover, it has been demonstrated as well that even under conditions leading
to almost negligible physico-chemical removal of biomolecules from surfaces, it is possible to
find out conditions leading to the fast decrease of their bioactivity. Nevertheless, concerning this
issue, it is important to stress that optimal conditions for inactivation of biological pathogens are
strongly dependent on their nature. As an example of the selectivity of plasma post-discharge,
New Journal of Physics 11 (2009) 115017 (http://www.njp.org/)
29
Figure 33. ToF-SIMS spectra of untreated lipid A (a) and lipid A treated 10 s
in (b) Ar/H2 50 : 50 and (c) O2 /H2 50 : 50 MW discharges (1000 W, 13 Pa,
100 sccm).
the inactivation of bacterial spores requires high intensity of UV radiation that cannot be
achieved in hydrogen-based mixtures found to be advantageous for depyrogenation of bacterial
endotoxins. Therefore, the process based on the inactivation lacks universality of the process
based on the extensive etching of biological systems. Moreover, it is still not clear whether
chemical inactivation is possible for all types of pathogens.
6. Conclusions
While the use of plasma discharges for the sterilization of bacterial spores has been studied in
detail in the past few years, relatively fewer results have been published concerning the removal
of proteins and pyrogens from surgical instruments. Recent studies revealed, however, that the
limited effect of the present decontamination and sterilization procedures pose a real public
health problem. Therefore, the possibility to eliminate such biomolecules by two types of lowpressure plasma treatment was addressed in this study.
New Journal of Physics 11 (2009) 115017 (http://www.njp.org/)
30
Figure 34. Part of ToF-SIMS spectra of untreated lipid A and lipid A treated 10 s
in different MW discharges (1000 W, 13 Pa, 100 sccm).
Figure 35. (a) Geometry of grid placed on the lipid A sample and corresponding
2D chemical maps of samples irradiated by VUV lamp, (b) whole lipid A image,
(c) C12 H23 O2 peaks and (d) C14 H27 O2 peak (the squares correspond to lipid A
exposed to VUV and destroyed by the radiation).
It was demonstrated, using model proteins and peptides, that biomolecules having
distinctly different properties can be efficiently removed by directly exposing the surface to
be treated to low pressure ICP discharges. Analysis of various discharges shows that Ar/O2
mixtures are the most efficient for a fast removal of these residues by a mechanism related
to chemical sputtering involving both radicals and ions. This mechanism also intervenes for
bacterial spores that are inactivated and destroyed by etching as well, which is advantageous
New Journal of Physics 11 (2009) 115017 (http://www.njp.org/)
31
mainly with respect to the universality of this approach. Nevertheless, the etching of
biomolecules could be accompanied by etching of substrate materials.
Different conclusions can be drawn concerning treatment performed in a post discharge.
Although in this case it is still possible to sterilize bacterial spores laying on a surface by
the UV radiation emitted in post-discharge or inactivate pyrogens by the chemical reaction of
radicals with their surfaces, the protein removal rates observed in a post discharge are much
lower as compared with a direct plasma treatment. This result illustrates the problem related to
the treatment of real surgical instrument surfaces: since the practical experience obtained from
hospitals shows that protein residues can be found on a significant proportion of surgical tools
after the decontamination, the strategy relying on a post discharge treatment can be questioned,
since no assurance can be given that these residues will be harmless after treatment. These
findings have a serious implication on the type of material that can be treated without serious
deleterious effects.
However, our results have to be confirmed by experiments on more relevant types of
contamination (involving inorganic compounds, lipids, blood, etc), since complex matrices can
significantly reduce the etching rates obtained on pure proteins.
Acknowledgments
This work has been supported by the FP6 2005 NEST project ‘Biodecon’. We thank
K Stapelmann and B Denis for technical help during experiments.
References
[1] Klevens R M, Edwards R, Richards C L, Horan T C, Gaynes R P, Pollock D A and Cardo D M 2007 Public
Health Rep. 122 160–6
[2] Kmietowicz Z 2000 Br. Med. J. 320 534
[3] Anonymous. Délégation Générale de la Santé. http://www.sante.gouv.fr/htm/pointsur/nosoco
[4] Anonymous. Dipartimento di Sanita Pubblica, Università degli Studi di Firenze. http://www.anio.it/
file/15 infezioni ospedaliere.pdf
[5] Lipscomb I P, Sihota A K, Botham M, Harris K L and Keevil C W 2006 J. Hosp. Infect. 62 141–8
[6] Lipscomb I P, Sihota A K and Keevil C W 2008 J. Hosp. Infect. 68 52–8
[7] Baxter L, Baxter H C, Campbell G A, Grant K, Jones A, Richardson P and Whittaker G 2006 J. Hosp. Infect.
63 439–44
[8] Murdoch H, Taylor D, Dickinson J, Walker J T, Perrett D, Raven N D H and Sutton J M 2006 J. Hosp. Infect.
63 432–8
[9] Tsuji K and Harrison S J 1978 Appl. Environ. Microbiol. 36 710–4
[10] Ludwig J D and Avis K E 1990 J. Parenter. Sci. Technol. 44 4–12
[11] Moesby L, Hansen E W, Christensen J D, Hoyer C H, Juhl G L and Olsen H B 2005 Eur. J. Pharm. Sci. 26
318–23
[12] Moisan M, Barbeau J, Moreau S, Pelletier J, Tabrizian M and Yahia L’H 2001 Int. J. Pharm. 226 1–21
[13] Lerouge S, Wertheimer M R and Yahia L’Y 2001 Plasmas Polym. 6 175–88
[14] Rossi F, Kylián O and Hasiwa M 2006 Plasma Process. Polym. 3 431–42
[15] Moisan M, Barbeau J, Crevier M-C, Pelletier J, Philip N and Saoudi B 2002 Pure Appl. Chem. 74 349–58
[16] Laroussi M 2002 IEEE Trans. Plasma Sci. 30 1409–15
[17] Laroussi M 2005 Plasma Process. Polym. 2 391–400
New Journal of Physics 11 (2009) 115017 (http://www.njp.org/)
32
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33]
[34]
[35]
[36]
[37]
[38]
[39]
[40]
[41]
[42]
[43]
[44]
[45]
[46]
[47]
[48]
[49]
[50]
[51]
[52]
[53]
[54]
[55]
[56]
Laroussi M, Mendis D A and Rosenberg M 2003 New J. Phys. 5 41
Eto H, Ono Y, Ogino A and Nagatsu M 2008 Appl. Phys. Lett. 93 221502
Feichtinger J, Schulz A, Walker M and Schumacher U 2003 Surf. Coat. Technol. 174–175 564–9
Boudam M K, Moisan M, Saoudi B, Popovici C, Gherardi N and Massines F 2006 J. Phys. D: Appl. Phys.
39 3494–507
Halfmann H, Denis B, Bibinov N, Wunderlich J and Awakowicz P 2007 J. Phys. D: Appl. Phys. 40 5907–11
Lerouge S, Fozza A C, Wertheimer M R, Marchand R and Yahia L’H 2000 Plasma Polym. 5 31–46
Philip N, Saoudi B, Crevier M-C, Moisan M, Barbeau J and Pelletier J 2002 IEEE Trans. Plasma Sci.
30 1429–36
Nagatsu M, Terashita F, Nonaka H, Xu L, Nagata T and Koide Y 2005 Appl. Phys. Lett. 86 211502
Lerouge S, Wertheimer M R, Marchand R and Yahia L’H 2000 J. Biomed. Mater. Res. 51 128–35
Kylián O, Sasaki T and Rossi F 2006 Eur. Phys. J. Appl. Phys. 34 139–42
Cousty S, Villeger S, Sarette J P, Ricard A and Sixou M 2006 Eur. Phys. J. Appl. Phys. 34 143–6
Hayashi N, Guan W, Tsutsui S, Tomari T and Hanada Y 2006 Japan. J. Appl. Phys. 45 8358–63
Vujošević D, Mozetič M, Cvelbar U, Krstulović N and Milošević S 2007 J. Appl. Phys. 101 103305
Vratnica Z, Vujošević D, Cvelbar U and Mozetič M 2008 IEEE Trans. Plasma Sci. 36 1300–1
Vicoveanu D, Popesco S, Ohtsu Y and Fujita H 2008 Plasma Process. Polym. 5 350–8
Beutler B, Hoebe K, Du X and Ulevitch R J 2003 J. Leukoc. Biol. 74 479–85
Kylián O, Hasiwa M and Rossi F 2006 IEEE Trans. Plasma Sci. 34 2606–10
Williams D 2003 Med. Device Technol. 14 8–11
Kylián O, Hasiwa M and Rossi F 2006 Plasma Process. Polym. 3 272–5
Hasiwa M, Kylián O, Hartung T and Rossi F 2008 Innate Immun. 14 89–97
Taylor D M 1999 J. Hosp. Infect. 43 S69–76
Lipscomb I P, Sihota A K and Keevil C W 2006 J. Clin. Microbiol. 44 3728–33
Baxter H C, Campbell G A, Whittaker A G, Jones A C, Aitken A, Simpson A H, Casey M, Bountiff L,
Gibbard L and Baxter R L 2005 J. Gen. Virol. 86 2393–9
Mogul R, Bol’shakov A A, Chan S L, Stevens R M, Khare B N, Meyyappan M and Trent J D 2003 Biotechnol.
Prog. 19 776–83
Ceccone G, Gilliland D, Kylián O and Rossi F 2006 Czech. J. Phys. 56 B672–7
Baxter H C, Campbell G A, Richardson P R, Jones A C, Whittle I R, Casey M, Whittaker A G and Baxter R
L 2006 IEEE Trans. Plasma Sci. 34 1337
Bernard C, Leduc A, Barbeau J, Saoudi B, Yahia L’H and De Crescenzo G 2006 J. Phys. D: Appl. Phys.
39 3470–8
Deng X T, Shi J J, Chen H L and Kong M G 2007 Appl. Phys. Lett. 90 013903
Deng X T, Shi J J and Kong M G 2007 J. Appl. Phys. 101 074701
Kylián O, Rauscher H, Gilliland D, Brétagnol F and Rossi F 2008 J. Phys. D: Appl. Phys. 41 095201
Kylián O, Rauscher H, Sirghi L and Rossi F 2008 J. Phys.: Conf. Ser. 100 062017
Moreau S, Moisan M, Tabrizian M, Barbeau J, Pelletier J, Ricard A and Yahia L’H 2000 J. Appl. Phys.
88 1166–74
Villeger S, Ricard A and Sixou M 2004 Eur. Phys. J. Appl. Phys. 26 203–8
Cvelbar U, Vujošević, Vratnica Z, and and Mozetič M 2006 J. Phys. D: Appl. Phys. 39 3487–93
Tompkins H G and Irene E A (ed) 2005 Handbook of Ellipsometry (Berlin: Springer)
Mazotti F, Hartung T and von Aulock S 2007 J. Biomed. Mater. Res. A 80 276–82
Hasiwa M, Kullmann K, von Aulock S, Klein C L and Hartung T 2007 Biomaterials 3 1367–75
Eckstein W 1991 Computer Simulation of Ion Solid Interactions 1st edn (Springer Series in Materials
Science) (Berlin: Springer)
Opretzka J, Benedikt J, Awakowicz P, Wunderlich J and von Keudell A 2007 J. Phys. D: Appl. Phys.
40 2826–30
New Journal of Physics 11 (2009) 115017 (http://www.njp.org/)
33
[57] Raballand V, Benedikt J, Wunderlich J and von Keudell A 2008 J. Phys. D: Appl. Phys. 41 115207
[58] Benedikt J, Flötgen C, Kussel G, Raballand V and von Keudell A 2008 J. Phys.: Conf. Ser. 133 012012
[59] Kylián O, Benedikt J, Sirghi L, Reuter R, Rauscher H, von Keudell A and Rossi F 2009 Plasma Process.
Polym. 6 255
[60] Rossi F, Kylián O, Rauscher H, Gilliland D and Sirghi L 2008 Pure Appl. Chem. 80 1939–51
[61] Kylián O, Rauscher H, Denis B, Ceriotti L and Rossi F 2009 submitted
[62] Halfmann H, Bibinov N, Wunderlich J and Awakowicz P 2007 J. Phys. D: Appl. Phys. 40 4145–54
[63] Stapelmann K, Kylián O, Denis B and Rossi F 2008 J. Phys. D: Appl. Phys. 41 192005
[64] Kylián O and Rossi F 2009 J. Phys. D: Appl. Phys. 42 085207
[65] Hueso J L, Rico V J, Frías J E, Cotrino J and Gonzáles-Elipe A R 2008 J. Phys. D: Appl. Phys. 41 2002
[66] Bretagnol F, Rauscher H, Hasiwa M, Kylián O, Ceccone G, Hazell L, Paul A J, Lefranc O and Rossi F 2008
Acta Biomater. 4 1745–51
New Journal of Physics 11 (2009) 115017 (http://www.njp.org/)
View publication stats