Utilization of Ultraviolet Radiation of Cold Hollow Cathode Glow Discharge Plasma for Water Disinfection
I.A.Soloshenko1, V.Yu.Bazhenov1, V.A.Khomich1, V.V.Tsiolko1,
1 Institute of Physics of National Academy of Sciences of Ukraine, 46 Nauki Ave., 03028, Kiev, Ukraine,
2 Institute of Colloid and Water Chemistry of National Academy of Sciences of Ukraine, 42 Academician
Vernadsky Blvd., 03680, Kiev, Ukraine.
Abstract. In this proceeding the results of experimental studies of peculiarities of Escherichia coli water suspension
inactivation by ultraviolet radiation of hollow cathode discharge plasma in different gaseous media are presented. It is
shown that efficiency of the inactivation by the discharges on oxygen, mixtures of oxygen with deuterium, and water
vapor is essentially higher than that by the discharge on air, as well as the discharges of low and medium pressure
Keywords: Ultraviolet, plasma, hollow cathode, water disinfection.
PACS: 87.50.Gi, 52.25.Os.
In the last decade ultraviolet (UV) radiation attains growing application as antimicrobial agent at disinfection of
water, air, sterilization of packing materials for foodstuffs, etc. Inactivation of microorganisms under action of UV
radiation occurs, first of all, due to DNA damage in result of a set of photochemical reactions . Main advantage of
UV sterilization and disinfection method, as compared to those using chemical agents (chlorine, hydrogen peroxide,
ozone, etc.), consists in fact that at the use of UV radiation no toxic and/or mutagenic residuals are formed. Modern
UV systems commonly use low and medium pressure mercury lamps. Advantages of mercury lamps consist in their
high energy efficiency, reaching 40% for low pressure lamps, and long service life – up to 10 thousands hours.
However, there is essential drawback of using such lamps due to high toxicity of mercury vapor. It results in extra
expenses for safety assurance in production and recycling of those lamps. In the experiments on sterilization of
medical articles by gas discharge plasma  it has been shown that the efficiency of sterilization of Bacillus subtilis
spores by broad-band (200÷300 nm) UV radiation of hollow cathode discharge (HCD) oxygen plasma is essentially
higher than that by monochromatic radiation at 254 nm of low pressure mercury lamp.
Due to that, in the present proceeding the possibility of use of ultraviolet radiation of HCD plasma for water
disinfection was studied. For that purpose comparative experiments on the influence on Escherichia coli water
suspension produced by ultraviolet radiation of both mentioned discharge plasma, and standard low and medium
pressure mercury lamps were performed.
DESCRIPTION OF SETUP AND METHODS OF MEASUREMENTS
The experiments on study of the efficiency of water disinfection by UV radiation of HCD plasma were performed at
the setup schematically shown in Fig.1. Walls of cylindrical chamber 1 having 400 mm length and 50 mm internal
diameter simultaneously served as the discharge cathode. The end of the chamber was closed by window 3 made of
quartz of KU-1 type with 4 mm thickness and low bandpass bound about 175 nm. At a time of the experiments, Petri
dish made of 2 mm thick KU-1 quartz with water suspension of the test microorganisms was placed on the
window 3. Discharge plasma 5 served as volume source of UV
radiation. Air, oxygen, deuterium, mixture of oxygen and deuterium
with various ratios, and water vapor were used as working
gases/mixtures. Working pressure in the chamber was varied in range
from 0.1 to 1 Torr depending on type of gas media. Power introduced
into the discharge was varied in range of 250÷350 W.
At carrying out comparative experiments, low pressure (LP)
mercury lamp of DB-30 type generating practically monochromatic
UV radiation with ≈ 254 nm wavelength, and medium pressure (MP)
mercury lamp of PRK-400 type producing broadband UV radiation in
200÷300 nm wavelength range were used. In the experiments mercury
lamps were placed inside devices consistent with requirements
described in [3, 4]. Petri dishes were placed at the ends of collimating
tubes having 300 mm length and 40 mm internal diameter closed by
windows made of quartz of KU-1 type.
Spectroscopic measurements were performed with the use of
automated setup based on 0.6 m monochromator of MDR-23 type with
1200 lines/mm diffraction grating (inverse linear dispersion
1.3 mm/nm). Radiation intensity at the monochromator output was
measured by photomultiplier tube of FEU-39A type, and the signal
was supplied to the input of measuring-processing system developed
by the authors.
The integral intensity of UV radiation in the plane of sample
irradiation was measured by DAU-81 radiometer before and after each
record of spectrum intensity distribution, and respective average value allowed to link together the sets of
experimental data obtained in spectroscopic and medical-biological researches.
Escherichia coli 1257 strain received from Scientific Research Institute of Standardization and Control of
Medical Biological to. 3 ml of Escherichia coli water suspension with density of 106 CFU/ml was placed into the
Petri dish with 32 mm diameter resulting in layer thickness of about 3 mm. After UV irradiation for predetermined
time, required aliquot of the suspension or its dilution was introduced into Endo medium, survived bacteria were
incubated for 18-24 hours at 370 C temperature, and colony count was performed. On a basis of obtained data,
bacteria survival curves, that is dependencies of a number of survived bacteria on UV radiation doze, were built.
Due to fact that experimental studies of sterilization efficiency of Escherichia coli water suspension were
performed with the use of UV radiation with essentially different spectrum shape, in this proceeding the method of
determining effective irradiation dose for studied sample was used which enabled correct comparison of the results
obtained with the use of mentioned UV sources.
Essence of the method consisted in “weighing”
spectrum distributions of intensity absolute values
for each used type of UV radiation.
At determination of “weighing” function, first of
all, results of works [5-7] devoted to studies of DNA
molecule absorption and efficiency of bactericidal
action of UV radiation on the microorganisms in
dependence on the radiation frequency were taken
into account. It has been shown in  that DNA
absorption spectrum in considered wavelength range
(≈180÷300 nm) represents superposition of broad
absorption bands having maxim at about 190 nm and
260 nm (bands II and I, curve 1 in Fig.2) due to
electron excitation of diene and triene fragments of
DNA molecule chain. In experiments with Bacillus
subtilis spores  and Escherichia coli bacteria  it
has been shown that at long enough wavelength (λ>
230 nm) curves of dependencies of absorption and
bactericidal efficiency on λ practically coincide, and
FIGURE. 2. 1 – DNA absorption spectrum; 2 – transmission
curve of BaF2 filter with 3 mm thickness; 3 – “weighing”
curve, obtained by multiplying DNA absorption spectra and
transmission curve of BaF2 filter.
FIGURE. 1. Scheme of experimental setup.
1 – chamber-cathode, 2 – anode, 3 – window
of KU-1 quartz, 4 – Petri dish of KU-1 quartz,
5 – discharge plasma.
both of them reach maximum at λ≈260 nm. At shorter
wavelength (λ ≤ 220 nm) the dependencies diverge –
absorption curve goes up with wavelength decrease, whereas
bactericidal efficiency rapidly decreases. (Such difference is
most likely due to strong absorption of short wavelength
radiation by the structures surrounding DNA – cell walls,
plasma membranes, cytoplasm, etc.). In the present
experiments it was founded that bactericidal action of
radiation of the discharge used by us on studied Escherichia
coli culture remained completely unchanged at introduction
of additional filter made of BaF2 with low pass boundary
λ = 215 nm, that is, in our case radiation with wavelength
λ ≤ 215 nm did not provide essential effect on vital functions
of the microorganisms. On another side, this fact enabled
expression of “weighing” function in form of a product of
spectrum dependencies of factors of BaF2 filter
transmission(curve 2 in Fig.2) and DNA absorption obtained
in . Resulted “weighing” function normalized by 1 at
λ=254 nm is presented by curve 3 in Fig.2. At short λ it was
actually determined by factor of the filter transmission, and at
longer λ – by factor of DNA absorption. Thus, we used the
portion of radiation transmitted by the filter and absorbed by
DNA for defining irradiation dose at survival curves of the
microorganisms presented below.
EXPERIMENTAL RESULTS AND
“Weighed” spectrum distributions of UV radiation power
for HCD on used working media and that for PRK-400 lamp
are presented in Fig.3. Radiation spectrum for low pressure
lamp is not shown in the figure because it represents narrow
triplet at λ≈254 nm. As it follows from analysis of these
spectra, at work on air (Fig.3a) main contribution to UV
radiation of the plasma is made by that of γ system NO (A2Σ+
– X2П ), and at work on oxygen (Fig.3b) – by radiation of
secondary negative system (SNS) of molecular ions +
(А2Пu – X2Пg). At work on water vapor (Fig.3c) main
contribution to total UV radiation also is made by SNS of
2 O . The shape of radiation spectrum for the discharge
plasma on mixture of deuterium with oxygen essentially
depends on ratio of these components. At the discharge
glowing on pure deuterium, the spectrum is continuous and
has a maximum at λ≈220 nm. With oxygen added (Fig.3d),
continuous spectrum of deuterium is superimposed by radiation of the bands of SNS of +
2 O . In case of MP mercury
lamp the UV radiation is represented by continuous spectrum in 210÷230 nm range and spectral lines in remaining
part of considered wavelength range (Fig.3e). One can see from the figure that in case of the use of oxygen,
deuterium (or mixture deuterium with oxygen) and water vapor main “weighed” UV radiation power is concentrated
in ≈215÷230 nm spectrum range, whereas at air use it is spread over several bands in 210÷260 nm range. In case of
medium pressure mercury lamp major power of UV radiation is concentrated in 240÷270 nm wavelength range.
In Fig.4 bacteria survival curves are presented which have been obtained at treatment of Escherichia coli water
suspension having density of 2⋅106 CFU/ml by UV radiation of LP and MP mercury lamps and that of hollow
cathode discharge plasma on air, oxygen, water vapor, and mixture of deuterium with oxygen (N0 and NS are initial
FIGURE. 3. Distributions of “weighed” intensity of
UV radiation over spectrum for hollow cathode
discharge plasma on different working media and for
MP mercury lamp. a) air, pressure 0.2 Torr, b) oxygen,
pressure 0.1 Torr, c) water vapor, pressure 0.1 Torr; d)
mixture 0.35 Torr D2 + 0.08 Torr O2; e) MP mercury
number and that of survived bacteria, respectively). Each point in the figure represents data averaged for three to six
separate experiments. Bacteria survival curve for D2+O2 mixture is obtained by averaging separate curves for cases
of the use of pure deuterium and two mixture
deuterium with oxygen. (In spite of certain
difference in UV spectrum intensity distributions
for these media, the bacteria survival curves
One can see from the figure that bacteria
survival curves obtained at the use of plasma UV
radiation from discharges on oxygen, water vapor,
and mixtures of deuterium with oxygen practically
coincide. Curves obtained at the use of UV
radiation of mercury lamps and that of the
discharge plasma on air are also close to each
other, however, they are located above the first set
of the curves. Such essential difference in
behavior of the curves gives undoubted evidence
to fact that bactericidal features of UV radiation in
215÷300 nm wavelength range depend not only on
radiation dose absorbed by DNA (as it was
observed in the case of monochromatic radiation),
but as well on shape of the radiation spectrum.
Comparison of spectra of radiation absorbed by
DNA for the discharges on different gases shows that maximum efficiency is provided by the discharges with
radiation having maximum intensity in 215 to 230 nm spectrum range. High efficiency of UV radiation having
maximum in range λ≈215÷230 nm (which is located at region of II band of DNA absorption, Fig.2) may be
presumably due to:
1) difference in nature of DNA damage caused by UV radiation in mentioned wavelength range, as compared to
that occurring at the use of radiation with wavelength matching I band of DNA absorption;
2) stronger damage caused by radiation in mentioned wavelength range to other biological molecules, particularly,
enzymes which are responsible for reparation of damaged DNA .
1. I. L. Shechmeister, “Sterilization by ultraviolet irradiation”, in “Disinfection, Sterilization and Preservation”. Ed. S.S.Block,
IV Edition, Lea&Febinger, Philadelphia, 1991.
2. I. A. Soloshenko, V. V. Tsiolko, V. A. Khomich et al., Plasma Physics Reports, 26, pp. 792-800 (2000).
3. J. Kuo, Ching-Lin Chen and M. Nellor, J. Environ. Engin., 129, , pp. 774-779 (2003).
4. J. R. Bolton and K. G. Linden, J. Environ. Engin.,. 129, pp. 209-215 (2003).
5. T. Inagaki, R. N. Hamm, E. T. Arakawa and L. R. Painter, J. Chem. Phys.,. 61, pp. 4246-4250 (1974).
6. N. Munakata, M. Saito and K. Hiera, Photochemistry and Photobiology, 54, pp. 761-768 (1991).
7. T. Wang, S. J. MacGregor, J. G. Anderson et al., in Proc. of XV International Conference on Gas Discharge and their
Applications, Toulouse (France), 2004, pp. 729-732.
8. B. F. Kalisvaart Water Science and Technology, 43, pp. 191-197 (2001).
Опубликовано в рубрике Documents