Journal of Microbiological Methods 54 (2003) 177 182

www.elsevier.com/locate/jmicmeth

Quantitative evaluation of antibacterial activities of metallic

oxide powders (ZnO, MgO and CaO) by conductimetric assay
J. Sawai *
Department of Applied Chemistry, Kanagawa Institute of Technology, 1030 Shimo-Ogino, Atsugi, Kanagawa 243-0292, Japan

Received 1 November 2002; received in revised form 20 January 2003; accepted 20 January 2003

Abstract

Antibacterial activities of metallic oxide (ZnO, MgO and CaO) powders against Staphylococcus aureus and Escherichia coli
were quantitatively evaluated by measuring the change in electrical conductivity of the growth medium caused by bacterial
metabolism (conductimetric assay). The obtained conductivity curves were analyzed using the growth inhibition kinetic model
proposed by Takahashi for calorimetric evaluation, and the metallic oxides were determined for the antibacterial efficacy and
kinetic parameters. The parameters provide some useful indicators for antimicrobial agents, such as the dependence of
antibacterial activity on agent concentration, and the affinity between the agent and the bacterial cells. CaO was the most
effective, followed by MgO and ZnO, against E. coli. On the other hand, ZnO was the most effective for S. aureus and was
suggested to have a strong affinity to the cells of S. aureus.
D 2003 Elsevier Science B.V. All rights reserved.

Keywords: Antibacterial activity; Calcium oxide; Conductimetric assay; Magnesium oxide; Zinc oxide

1. Introduction the use of ceramic with inherent antimicrobial activity

In recent years, the use of inorganic antimicrobial
agents has attracted interest for the control of
microbes (Okouchi et al., 1995; Wilczynski, 2000).
The key advantages of inorganic antimicrobial agents
are improved safety and stability, as compared as
organic antimicrobial agents. At present, most anti-
bacterial inorganic materials are TiO2 (Shirashi et al.,
1999; Huang et al., 2000) and the ceramics immobi-
lizing antimicrobial metals, such as silver and copper
(Kourai, 1993; Wang et al., 1995). On the other hand,
* Tel./fax: +81-46-291-3193.
E-mail address: sawai@chem.kanagawa-it.ac.jp (J. Sawai).
has recently attracted interest (Isshiki et al., 1994;
Okouchi et al., 1995). The present authors have
already evaluated the antibacterial activity of 26
ceramic powders, and 10 were found to inhibit bacte-
rial growth. Among these active powders, MgO, CaO
and ZnO exhibited strong antibacterial activity (Sawai
et al., 1995, 1998, 1999, 2000). However, to date
there have been few studies involving quantitative
evaluation of the antibacterial activity of these metal-
lic oxides (Okouchi et al., 1995). Also, no study on
the affinity between the agent and the bacterial cells,
which is an important parameter when the agent is
applied to food and the environment, is made.
Since these materials are insoluble or slightly
soluble, the conventional methods, such as halo test

0167-7012/03/$ - see front matter D 2003 Elsevier Science B.V. All rights reserved.
doi:10.1016/S0167-7012(03)00037-X
178 J. Sawai / Journal of Microbiological Methods 54 (2003) 177182
and turbidometry, are unsuitable for evaluation of
the antibacterial activity. In the previous study
(Sawai et al., 2002), we examined the applicability
of the conductimetric assay for evaluation of anti-
bacterial activity of antibiotics. The results obtained
by the conductimetric assay were closely consistent
with those by the turbidity method. The conducti-
metric assay made it possible to evaluate the anti-
bacterial activity of antibiotics. The conductimetric
assay detects bacterial growth and metabolism via
changes in electrical conductivity (Eden and Eden,
1984). The key advantage of the conductimetric
assay is its applicability to highly turbid samples,
such as foods and insoluble materials. Takahashi
(1990) proposed the inhibition kinetic model for
calorimetric analysis, which can provide some useful
indicators for antimicrobial agents, such as the
minimum inhibitory concentration (MIC), the
dependence of antibacterial activity on agent con-
centration, and the affinity between the agent and
the bacterial cells. In this study, we measured the
antibacterial activities of the metallic oxide powders
(ZnO, MgO and CaO) against Staphylococcus aur-
eus and Escherichia coli by conductimetric assay
and tried to apply the model to evaluate quantita-
tively the activity of these materials.
2. Materials and methods
2.1. Microorganisms
S. aureus 9779 and E. coli 745 were obtained from
the Tokyo Metropolitan Research Laboratory for
Public Health. The strains of bacteria were stored in
20% glycerol solution at
85 jC. They were thawed
and incubated in brain heart infusion broth (Eiken
Chemicals, Tokyo, Japan) at 37 jC for 20 h. The cells
were in stationary phase and washed once in sterile
saline (0.85 w/v%), and then resuspended in saline at
approximately 103 CFU/ml. The tube containing the
bacterial suspension was immersed in ice water before
use in the experiments.
2.2. Preparation of powder slurry
ZnO, MgO and CaO powders (Kishida Chemical,
Osaka, Japan) were used, with mean particle sizes of
2.6, 3.7 and 2.7 Am, respectively. A powder slurry
was prepared by suspending the powders in sterile
saline.
2.3. Conductimetric assay
For measurement of the conductivity change
caused by bacterial growth, a BactometerR microbial
monitoring system model 64 (bioMerieux, Tokyo,
Japan) was employed (Sawai et al., 1995). The stand-
ard module for the Bactometer was divided into 16
individual wells, and a pair of electrodes was inserted
into each well. In the conductimetric assay, liquid
growth media can be also used. It was reported that
better quality conductivity curves could be obtained
by using solid media than by using liquid media (Eden
and Eden, 1984). Modified plate count agar (yeast
extract 20 g, dextrose 4 g, tripton 20 g, agar 10 g, H2O
1000 ml: bioMerieux) was poured into the well until
the electrodes were covered (0.5 ml). After the agar
had solidified, the bacterial suspension (0.1 ml) and
powder slurry (0.1 ml) were pipetted into the well.
The module was capped tightly and set in the incu-
bator of the Bactometer. Five wells were used for each
dilution. The conductivity change with bacterial
growth was monitored during incubation at 35 jC
for 48 h.
3. Results and discussion
3.1. Kinetic analysis of antibacterial activity
Fig. 1 shows the effect of ZnO powder concen-
tration on the conductivity change with growth of S.
aureus. The conductivity curves were shifted toward
a longer detection period and the slopes became
gentler as the ZnO power concentration increased.
Similar tendencies were obtained for the other pow-
ders and bacteria. For further quantitative evaluation,
the model proposed by Takahashi (1990) is applied
to a conductimetric assay for the evaluation of
antibacterial activities of these metallic oxide pow-
ders. They proposed two models and successfully
used these models to evaluate the antimicrobial
activities of drugs and other chemicals for calori-
metric analysis. One is the noncompetitive inhibition
model for determining kinetic parameters of growth
 J. Sawai / Journal of Microbiological Methods 54 (2003) 177182
Fig. 1. Effect of ZnO addition on conductivity change of growth
medium for S. aureus.
inhibition. The other is the antibacterial efficacy
model for determining the minimum inhibitory con-
centration.
At first, the antibacterial activity of the metallic
oxides was analyzed using the noncompetitive inhib-
ition model. If bacterial growth is assumed to follow
the Michaelis Menten reaction, a viable cell V takes
up the substrate S to form an intermediate VSn that
produces a new viable cell and a metabolic by-product
P, as described by
V nSVVSn ! 2V P
When an antibacterial agent exists as an inhibitor I,
the following schemes are considered.
V mIVVIm
VSn mIVVSn Im
The antibacterial agent inhibits the viable activity of
the cells by forming nonviable states VIm and VSnIm,
where m is the apparent stoichiometric number of the
antibacterial agent necessary to inhibit the reproduc-
tion of new cells. The microbial growth kinetics
proposed by Monod (1949) give the following equa-
tion from analysis of Eqs. (1) (3).
n m m
li =l0 Ks =S 1 I =KdV 1 I =Kd 
1 plotted against the ZnO powder concentration in Fig.
179
where Ks is a dissociation constant of the substrate,
and Kd and KdVare dissociation constants of the
inhibitors. These constants are given by
Ks V Sn =VSn  5
Kd VSn Im =VSn Im  6
KdV V Im =VIm  7
In Eq. (4), li is the growth rate constant at the
inhibitor concentration of I = i, and l0 represents the
maximum growth rate constant, corresponding here to
that at I = 0. Since the substrate exists in excess
concentration in actual measurements, Eq. (4) can be
reduced to the following given Ksb[S]n,
li =l0 1 Im =Kd 
1 8
The decrease in growth activity in the conductimetric
assay is given as the time lag with respect to the
conductivity curve of the control. Here, the time
required for the conductivity change to reach by 5%
is defined as detection time (s). As shown in Fig. 1,
the delay of the growth rate is postulated to be the
time lag between s0 and si at the antibacterial con-
centrations zero and i, respectively. The bacterial
concentration at the conductivity change of 5% was
1 approximately 109 CFU/ml. Five wells were used for
each dilution, and the SE of si was less than 10%. The
conductivity change (CC) induced by bacterial cells in
the exponential growth phase is assumed to approx-
imate to the following.
2
CCt>AN0 expflt
tL g 9
3
where N0, tL and A are the initial viable number,
duration of lag phase, and an empirical constant,
respectively. When s and si were substituted into
Eq. (9), li/l0 becomes equal to (s0
tL)/(si
tL). If
tLbt, li/l0 in Eq. (8) can be replaced to s0/si, as
follows (Takahashi, 1990).
s0 =si 1 Im =Kd
1 10
The decrease in the growth rate of S. aureus, (s0/si), is
2. Five wells were used for each dilution, and the SE
4 of the average value remained below 15%. The values
180 J. Sawai / Journal of Microbiological Methods 54 (2003) 177182

concentration at which the agent makes the bacterial

multiplication rate to zero. The decrease in growth
rate, (1
s0/si), is assumed to be proportional to the
antibacterial agent concentration (I) to the power of b,
i.e.,
1
s0 =si aIb 12

where a and b are constants. b represents the depend-

ence of antibacterial activity on agent concentration.
Because the [I]100 corresponds to the agent concen-
tration in the case of si = l, it can be defined as
follows,
I100 Isi l 1=a1=b 13

The constants a and b can then be determined; here

Fig. 2. Antibacterial curves for ZnO against S. aureus. Dotted line:
calculated from parameters m and Kd in Table 1; solid line:
calculated from parameters a and b in Table 1.
of Kd and m are determinable by a regression analysis
on the basis of Eq. (10). The dotted line was drawn by
fitting the parameters (Kd and m) to the experimental
values. The antibacterial agent concentration at which
the growth rate is suppressed by 50% ([I]50) can be
calculated using the equation
1=m
I50 Kd
[I]50 for ZnO against S. aureus was 0.67 mg/ml.
On the other hand, the minimal inhibitory concen-
tration (MIC), used as a conventional indicator of
antimicrobial activity, was determined by the separa-
ted model described above. Since the antibacterial
activity of these metal oxides is as pointed out
remarkably large as compared with those of antibi-
otics, the use of the term MIC may not be proper.
Therefore, as the indicator of antibacterial activity
equivalent to MIC, [I]100 is defined as the minimum
the values are 1.01 and 1.90, respectively. The solid
line in Fig. 2 was drawn based on these parameters,
and the [I]100 of ZnO powder against S. aureus
calculated from this result is 0.99 mg/ml. This is in
agreement with the experimental results, as demon-
strated by the absence of conductivity change at
powder concentrations between 0.8 and 1.0 mg/ml
(Fig. 1).
3.2. Action mechanism of metallic oxides
11
The same analysis was performed for the other
powders and E. coli, and antibacterial parameters were
determined for ZnO, MgO and CaO (Table 1). ZnO
had a markedly small value of Kd against S. aureus. Kd
shows the equilibrium state between VSnIm, VSn and I,
and S. aureus was found more readily to form the
nonviable state with ZnO than with other powders. The
plasters and ointments containing ZnO have been used
successfully to treat wounds and various dermatoses
(Mitchnick, 1992; Lansdown, 1993). ZnO has recently
received attention for its effectiveness as a treatment

Table 1
Antibacterial parameters of ZnO, CaO and MgO powders against S. aureus and E. coli
Microorganisms Substance m Kd [I]50 (mg/ml) a b [I]100 (mg/ml)
S. aureus ZnO 2.60 0.35 0.67 1.01 1.90 0.99
CaO 1.53 9.37 4.30 0.07 1.61 5.08
MgO 3.78 141 3.71 0.01 2.82 5.00
E. coli ZnO 1.49 30.2 9.85 0.07 0.75 32.4
CaO 3.49 18.3 2.30 0.06 2.56 3.06
MgO 3.78 167 3.87 0.01 2.62 5.58
 J. Sawai / Journal of Microbiological Methods 54 (2003) 177182
for atopic dermatitis related to S. aureus (Akiyama et
al., 1998). Eczematous skin lesions of atopic dermatitis
are usually densely colonized with S. aureus, and a
correlation between the severity of eczematous lesions
and the density of S. aureus colonization has been
demonstrated (Williams et al., 1990). It was reported
that the ZnO inhibited the fibrin formation by coagu-
lase of S. aureus (Akiyama et al., 1998). The effective-
ness of ZnO for atopic dermatitis may depend on the
strong affinity between ZnO and S. aureus. To inves-
tigate the antibacterial activity of ZnO powder slurry in
detail, the effect of Zn2 + was examined using ZnCl2
solution. The solubility of ZnO to water is 5.2  10
5
mol/l (Rikagaku-Jiten, 1981). For E. coli and S.
aureus, the conductance curves closely matched those
of controls even when Zn2 + existed at a concentration
10 times the solubility of ZnO (data was not shown).
This suggests that the Zn2 + existing in ZnO powder
slurry had no effect on the growth of E. coli and S.
aureus and that the contact of the ZnO powder and
bacterial cell is a very important factor. In the previous
study (Sawai et al., 1996), the generation of hydrogen
peroxide (H2O2) from ZnO powder was observed.
Using four kinds of antibiotics, an investigation was
made to determine whether or not H2O2 generated
from a ZnO powder slurry was related to its antibacte-
rial activity. Changes in the sensitivity of E. coli to the
antibiotics suggested that H2O2 was one of the primary
factors concerned in the antibacterial activity of the
ZnO powder slurry (Sawai et al., 1998).
For MgO and CaO, the values of Kd and [I]100
were almost the same against both E. coli and S.
aureus. In the previous study (Sawai et al., 1995),
MgO and CaO powders were found to act on both
gram-positive bacteria and gram-negative bacteria in a
bactericidal manner. The solutions containing Mg2 +
and Ca2 + at the concentration 10 times the solubility
of MgO and CaO, respectively, did not affect the
growth of E. coli and S. aureus (data not shown). In
the antibacterial activity of MgO and CaO, the contact
between the powders and bacterial cells is an impor-
tant factor. An alkaline effect caused by the hydration
of MgO and CaO is considered to be the primary
mechanisms. However, since the bactericidal actions
of the MgO and CaO powder slurries were greater
than that of an NaOH solution of identical pH, they
are considered to posses other antibacterial mecha-
nisms in addition to that of alkalinity. The antibacte-
181
rial mechanism involving an oxygen radical is
suggested also for the case of MgO and CaO (Sawai
et al., 1999, 2000).
3.3. Application of metallic oxides
In recent years, the development of antimicrobial
agents that have little or no negative impact on the
natural environment has become important. In shell
harvesting districts, large amounts of shell are heaped
near the seaside, creating serious problems such as the
emission of offensive odors and soil pollution from
heavy metals contained in the viscera (Kikuchi, 1998).
The main component of shell is CaCO3, which is
converted to CaO by heating. Dolomite is also an
inexpensive natural mineral composed of CaCO3 and
MgCO3, affording CaO and MgO upon heat treat-
ment. The heated shell and dolomite powders also
exhibited the antibacterial activities, which increased
with heating temperature (Okouchi et al., 1995; Sawai
et al., 2001a,b; Shiga et al., 1999). The heated scallop-
shell was applied to fresh vegetables, shredded cab-
bage, and could reduce the viable bacterial cells
(Sawai et al., 2001c). The use of these materials in
food processing is therefore not only to be a source of
minerals but also to prolong the shelf life of food-
stuffs. Bari et al. (1999, 2002) reported that heated
oyster shell powder treatment successfully decreased
E. coli O157:H7 in fresh radish sprout production and
on the surface of tomatoes. Also, these shell powders
are beginning to be blended to give antimicrobial to
plastic cast, polyethylene film, nonwoven fabric, and
paper (Oshima, 1998).
When exposed to air, CaO and MgO absorb CO2
and return to CaCO3 and MgCO3. Therefore, it can be
said that the safety of these materials is comparatively
high. The application of the materials to Legionella in
a cooling tower is presently considered. When these
materials are applied to food and the environment, the
information related to affinity of the materials and
microbial cells, such as Kd, will be needed; the
conductimetric assay can provide quantitative and
quick evaluation of the antibacterial activity of the
test materials which have slight solubility, such as
ceramic powders. The present quantitative analysis
may be useful for collecting fundamental data related
to the influences of a variety of antimicrobial agents
on environmental microbes.
182 J. Sawai / Journal of Microbiological Methods 54 (2003) 177182

References Kokugan, T., Shimizu, M., 1996. Detection of active oxygen

generated from ceramic powders having antibacterial activity.
J. Chem. Eng. Jpn. 29, 627 633.
Akiyama, H., Yamazaki, O., Kanzaki, H., Tada, J., Arata, J., 1998.
Sawai, J., Shoji, S., Igarashi, H., Hashimoto, A., Kokugan, T.,
Effect of zinc oxide on the attachment of Staphylococcus aur-
Shimizu, M., Kojima, H., 1998. Hydrogen peroxide as an anti-
eus. J. Dermatol. Sci. 17, 67 74.
bacterial factor in zinc oxide powder slurry. J. Ferment. Bioeng.
Bari, M.L., Kusunoki, H., Furukawa, H., Ikeda, H., Isshiki, K.,
86, 521 522.
Uemura, T., 1999. Inhibition of growth of Escherichia coli
Sawai, J., Kojima, H., Igarashi, H., Hashimoto, A., Shoji, S., Shi-
O157:H7 in fresh radish (Raphanus sativus L.) sprout produc-
mizu, M., 1999. Bactericidal action of calcium oxide powder.
tion by calcinated calcium. J. Food Prot. 62, 128 132.
Trans. Mater. Res. Soc. Jpn. 24, 667 670.
Bari, M.L., Inatsu, Y., Kawasaki, S., Nazuka, E., Isshiki, K., 2002.
Sawai, J., Kojima, H., Igarashi, H., Hashimoto, A., Shoji, S., Sa-
Calcinated calcium killing of Escherichia coli O157:H7, Salmo-
waki, T., Hakoda, A., Kawada, E., Kokugan, T., Shimizu, M.,
nella, and Listeria monocytogenes on the surface of tomatoes.
2000. Antibacterial characteristics of magnesium oxide powder.
J. Food Prot. 65, 1706 1711.
World J. Microbiol. Biotechnol. 16, 187 194.
Eden, R., Eden, G., 1984. Impedance Microbiology. Research
Sawai, J., Shiga, H., Kojima, H., 2001a. Kinetic analysis of bacter-
Studies Press, Letchworth, UK, pp. 41 62.
icidal action of heated scallop-shell powder. Int. J. Food Micro-
Huang, Z., Maness, P.C., Blakee, D.M., Wolfrum, E.J., Smoliski,
biol. 71, 211 218.
S.L., Jacoby, W.A., 2000. Bacterial mode of titanium dioxide
Sawai, J., Himizu, K., Kojima, H., 2001b. Death rate constants of
photocatalysis. J. Photochem. Photobiol., A Chem. 130,
Escherichia coli by heated dolomite powder slurry. Trans. Mater
163 170.
Res. Soc. Jpn. 26, 1347 1350.
Isshiki, K., Suhara, H., Mizuuchi, K., Tokuoka, K., 1994. Effective-
Sawai, J., Satoh, M., Horikawa, M., Shiga, H., Kojima, H., 2001c.
ness of calcium preparation to control microbial growth in food.
Heated scallop-shell powder slurry treatment of shredded cab-
Nippon Shokuhin Kogyo Gakkaishi 41, 135 140 (in Japa-
bage. J. Food Prot. 64, 1579 1583.
nese).
Sawai, J., Doi, R., Maekawa, Y., Kojima, H., 2002. Indirect con-
Kikuchi, S., 1998. Conversion of marine industrial waste to resour-
ductimetric assay of antibacterial activities. J. Ind. Microbiol.
ces. Shigen Kanri Gijutsu 45, 306 310 (in Japanese).
Biotech. 29, 296 298.
Kourai, H., 1993. Immobilized microbiocide. J. Antibact. Antifun-
Shiga, H., Sawai, J., Kojima, H., 1999. Utilization of heated shell
gal Agents 21, 331 337 (in Japanese).
powder in biocontrol. Trans. Mater Res. Soc. Jpn. 24, 557 560.
Lansdown, A.B.G., 1993. Influence of zinc oxide in the closure of
Shirashi, F., Toyoda, K., Fukinbara, S., 1999. Photolytic smf photo-
open skin wounds. Int. J. Cosmet. Sci. 15, 83 85.
catalytic treatment of an aqueous solution containing microbial
Mitchnick, M.A., 1992. Zinc oxide, an old friend to rescue. Cosmet.
cells and organic compounds in an annular-flow reactor. Chem.
Toiletries 107, 111 116.
Eng. Sci. 54, 1547 1552.
Monod, J., 1949. The growth of bacterial cultures. Annu. Rev.
Takahashi, K., 1990. Application of calorimetric methods to cellular
Microbiol. 3, 371 394.
processes: with special references to quantitative evaluation of
Okouchi, S., Murata, R., Sugita, H., Moriyoshi, Y., Maeda, N.,
drug action on living cells. Thermochim. Acta 163, 71 80.
1995. Calorimetric evaluation of the antimicrobial activities of
Wang, Y.L., Wan, Y.Z., Dong, X.H., Cheng, G.X., Tao, H.M., Wen,
calcined dolomite. J. Antibact. Antifungal Agents 26, 109 114
T.Y., 1995. Preparation and characterization of antibacterial vis-
(in Japanese).
cose-based activated carbon fiber supporting silver. Carbon 36,
Oshima, A., 1998. Application of antibacterial calcium. Food
1567 1571.
Chem. 9, 61 94.
Wilczynski, M., 2000. Anti-microbial porcelain enamels. Ceram.
Rikagaku-Jiten, 1981. 3rd ed. Iwanami Shoten, Tokyo.
Eng. Sci. Proc. 21, 81 83.
Sawai, J., Igarashi, H., Hashimoto, A., Kokugan, T., Shimizu, M.,
Williams, R.E.A., Gibson, A.G., Aitchiso, T.C., Lever, R., Mackie,
1995. Evaluation of growth inhibitory effect of ceramic powder
R.M., 1990. Assessment of a contact-plate sampling technique
slurry on bacteria by conductance method. J. Chem. Eng. Jpn.
and subsequent quantitative bacterial studies on atopic dermati-
28, 288 293.
tis. Br. J. Dermatol. 123, 493 501.
Sawai, J., Kawada, E., Kanou, F., Igarashi, H., Hashimoto, A.,