Factors Influencing the Rate of Killing of Escherichia coli
Exposed to Benzalkonium Chloride
JOHN W. KLIMEK
Sterling-Winthrop Research Institute, Rensselaer, New York
Received for publication September 27, 1955
The killing of bacterial cells by chemical agents may
be considered as the resultant of a reaction between two
systems, the cells and their suspending medium and the
germicide and its solvent. The method for testing
sanitizing agents proposed by Weber and Black (1948)
allows evaluation, on the basis of the two systems, of
factors such as ions present in the suspending or dilut-
ing medium, organic matter, time, temperature and the
bacterial cell itself, that may influence the killing of the
test organism. This method, therefore, was used in the
experiments here reported.
MATERIALS AND METHODS
Briefly, the method consists of the preparation of the
two systems, test cells and antibacterial agent, at twice
the concentration at which they will be used in the
test. Mixing of equal volumes of each system results in a
"medication mixture" from which aliquots are removed
at various intervals for determination of the number of
cells surviving the various periods of exposure to the
antibacterial agent. The number of cells exposed is
determined by averaging plate counts on diluted ali-
quots of the stock suspension of cells at the start and
end of the test.
The number of cells surviving exposure to the
germicide is determined by diluting a 1-ml sample of
the medication mixture with 9 ml of inactivator (2.222
g lecithin, 15.6 ml tween-80 per liter of M/4 phosphate
buffer pH 7.2) and making pour plates of 1-ml portions
of this and a further 1:10 dilution in tryptose glucose
extract agar. Incubation is for 48 hours at 37 C. Killing
of all the exposed cells is considered to have occurred
when at least 99.9999 per cent of the inoculum has been
made nonviable.
The effect of calcium and magnesium (hardness),
iron, aluminum, natural waters, organic matter, the
bacteria themselves, time and temperature on the
killing action of the quaternary ammonium germicide,
benzalkonium chloride, was studied. All bactericidal
determinations were made at concentrations of 200
ppm (1:5000 dilution) of the germicide in duplicate at
20 C, except where the effect of temperature was in-
vestigated.
Synthetic hard water was prepared by dissolving
28.41 g CaCl2 and 12.99 g Mg Cl2-6H20 in 10 L
AND JOHN HAYS BAILEY
distilled water. This stock solution, 3200 ppm hardness
as CaCO3, was appropriately diluted to give the de-
sired hardness. In determining the effect of organic
matter a synthetic dirty water devised by Allport
Downloaded from http://aem.asm.org/ on August 18, 2017 by guest
(personal communication) was used. This is based
upon the analysis of water in which the eating utensils
in an industrial canteen had been washed and consists
of: NaCl 0.05 per cent; peptone 0.05 per cent; glucose
0.08 per cent; dried skimmed milk 0.05 per cent; corn
starch 0.03 per cent; teaseed oil, 0.15 per cent.
The test culture, Escherichia coli # 1981 was main-
tained on the nutrient agar prescribed for this purpose
in the United States Department of Agriculture circular
(U.S.D.A., 1931). In studying the susceptibility of the
organism this medium was altered by substituting
other peptones for the peptonum siccum called for in
Circular 198.
Effect of calcium and magnesium. Hard water has been
reported by Ridenour and Armbruster (1948), Mueller
and Seeley (1948), and Seeley (1949) as adversely
affecting the germicidal powers of quatemary am-
monium compounds, the reduction in activity being
roughly proportional to the hardness of the water used
in the test. Several experiments, the results of which
are presented in table 1, were conducted to determine
what the effect of 450 ppm hardness on the germicidal
action of benzalkonium chloride might be. In these
experiments, the two systems comprising the reaction
mixture were prepared with synthetic hard water, and
the number of cells surviving exposure to the germicide
for 30, 60 and 120 seconds determined. To obviate
repetition of essentially identical results only three of
the control tests have been included in the tabulation.
It will be seen that in the control tests all the exposed
cells were killed on exposure to 200 ppm of the germicide
whereas among the eight tests carried out in the
presence of 450 ppm hardness there were four in which
less than 99.9999 per cent of the exposed cells were
killed within 30 seconds' exposure. In these four tests,
however, the number of cells per cc killed ranged from
75,488,220 (out of 75.5 million) to 167,149,285 (out of
167.15 million exposed). On two minutes' exposure all
exposed cells were killed in every instance. These
1 Obtained from Dr. L. A. Black, U. S. Public Health Ser-
vice, Cincinnati, Ohio.
53
54 J. W. KLIMEK AND J. H. BAILEY [VOL. 4
TABLE 1. Germicidal action of 200 ppmt benzalkonium chloride the third in both systems. Control tests were conducted
in 450 ppm synthetic hard water on Escherichia coli 198 in distilled water. The results of these experiments are
at 25 C
presented in table 2.
Number Cells
Surviving Per cent Cells Killed on Exposure
It will be noted that where the germicide system
Hard-
ness
Cells
per ml
Exposure contributed all of the hardness present in the reaction
IX 106
30 sec 60 sec 120 30 sec 60 sec 120 sec
mixture, all of the exposed cells were killed on 30
sec
seconds' exposure regardless of the hardness of the reac-
ppm tion mixture. However, wherever the cell system had
0 81 10 0 0 99.99999 100 100 been prepared in hard water, the number of cells sur-
0 87 0 0 0 100 100 100 viving a two-minute exposure was large. These results
0 152.3 0 0 0 100 100 100
450 75.5 11780 5260 0 99.9864 99.9930 100 show that the effect of Ca++ and Mg++ is exerted upon
450 81 10 0 0 99.99999 100 100 the cells, not upon the germicide. Furthermore, in view
.450 81 40 0 0 99.99995 100 100 of the results of this and the previously mentioned ex-
450 87 100 0 0 99.99989 100 100 periments, the effect of Ca~- and Mg++ upon the cells
450 87 0 0 0 100 100 100 is transient; increasing the period of exposure results

Downloaded from http://aem.asm.org/ on August 18, 2017 by guest

450 148.75 105 0 0 99.99992 100 100
450 152.3 275 2915 60 99.99982 99.99808 99.99996 in the killing of all the exposed cells.
450 167.15 715 30 0 99.99968 99.99998 100 Effect of iron. Mueller and Seeley (1948) and Seeley
(1949) reported that iron exerted a deleterious effect
TABLE 2. Effect of hardness in each system on the geimicidal
upon the germicidal action of the quaternary am-
action of 200 ppm benzalkonium chloride on Escherichia coli monium compounds. The amounts of iron used by
198 at 25 C these authors, however, were large compared to the
concentrations found in most municipal, or for that
Hardness T Hardness
Number Exposure*
of Cells Surviving
matter private, water supplies. Accordingly, the germi-
cidal action of benzalkonium chloride was studied in
Cell system system Reaction 30 sec 60 sec 120 sec
the presence of more realistic concentrations of iron,
ppm ppm ppmt although the concentrations employed were still in
o 0 0 0 0 0 excess of those considered unobjectionable in a water
o 0 0 0 0 0 supply.
O 600 300 0 0 The action of iron on each of the two systems of the
0 600 300 10 0 0
600 0 300 134,400 128,000 4,825 reaction mixture was investigated by the addition of
600 0 300 TNCt 89,600 510 appropriate amounts of ferrous sulfate to the distilled
300 300 300 121,600 80,000 30,000 water used in preparing them. The fact that the
300 300 300 147,200 38,000 11,000 majority of the simple salts of iron are insoluble and
* Medication mixture contained 93 X 106 cells per ml. those that are soluble are acid in reaction, complicates
t Too numerous to count. the evaluation of the effect of this ion on the germicidal
action of benzalkonium chloride, particularly when it is
results clearly show that the effect of the calcium and remembered that the quaternary ammonium germi-
magnesium ions necessary to produce the 450 ppm cides are more active in an alkaline than acid men-
hardness is not a reduction in the germicidal property struum. The acidity imparted to the reaction mixture
of the quaternary but a three- to fourfold prolongation by the ferrous sulfate was controlled, in one series of
of the time required to kill all of the exposed cells. experiments, by the addition of 75 ppm bicarbonate to
In the above experiments, the harness was present the distilled water used in preparation of the germicide
in both of the two systems of the reaction mixture. A system. In another series of experiments the reaction
reaction mixture of any degree of hardness in which mixture was allowed to develop a pH dependent upon
the hardness is confined to one of the systems (up to the concentration of ferrous sulfate present. The results
the time of mixing the two) may be prepared by using of these experiments are given in table 3, and demon-
water of twice the desired final hardness for one system strate that iron, at a concentration of 5 ppm or less and
and an equal volume of distilled water for the other. at pH 5 or above, shows no demonstrable effect upon
By employing this procedure, it would be possible to the killing action of benzalkonium chloride for E. coli
demonstrate the effect of calcium and magnesium ions regardless of whether it be in the germicide or cell
on the two systems of the reaction mixture. systems.
Cell systems were prepared in water of 300 and 600 Water supplies containing iron usually possess some
ppm hardness and in distilled water. Appropriate mix- hardness. The concomitant effect of these three ions on
ing of the systems resulted in reaction mixtures con- the killing action of benzalkonium chloride was investi-
taining 330 ppm hardness, one having the hardness in gated in experiments employing water of 450 ppm
the cell system, another in the germicide system and hardness and various concentrations of iron in both
19561 FACTORS IN KILLING ESCHERICHIA COLI 55

TABLE 3. Effect of iron and pH on the germicidal action, of 200 trations up to 5 ppm, is without effect upon the germi-
ppm benzalkonium chloride on Escherichia coli 198 at 25 C cidal action of benzalkonium chloride for E. coli.
Ppm Iron in No. Cells Surviving Exposure to Germicide Effect of aluminum. Experiments similar to those used
to study the effect of iron were conducted. Hardness
Cell Germi- Reaction Buffered* Unbufferedt was present in either or both of the systems, the result-
System system mixture pH$ 30 sec 60 sec pHt 30 sec 60 sec
ing reaction mixture having a hardness of 450 ppm in
0 0 0 7.7 20 0 6.6 0 0 all cases except where distilled water was used in both
0 2 1 7.4 0 10 6.3 0 0 systems. Aluminum, when used, was added to the
0 5 2.5 7.0 0 0 6.0 0 0 germicide system at 10 ppm, resulting, when diluted
0 10 5 6.8 85 10 4.9 0 0
2 0 1 7.4 0 0 6.4 0 0 with the cell system, in a reaction mixture containing
5 ppm A1. One series of tests was conducted in the
5 0 2.5 7.2 0 0 6.5 0 0
10 0 5 6.9 25 10 5.3 5 10 presence of 75 ppm bicarbonate (added to both sys-
tems); in another series the pH was uncontrolled. The
*
Reaction mixture contained 94.43 X 106 cells per ml.
results of these tests are presented in table 5.
t Reaction mixture contained 77.44 X 106 cells per ml.

Downloaded from http://aem.asm.org/ on August 18, 2017 by guest

t Of reaction mixture.
systems of the tests. In one series of tests the acidity of
the reaction mixture was partially controlled by the
addition of sodium bicarbonate to give a final bicar-
bonate concentration of 37.5 ppm; in another series no
control of pH was attempted. The results of these
experiments are given in table 4.
In the tests conducted at an alkaline pH and in the
presence of iron all cells exposed to the germicide were
killed within one minute; where only hardness was
present an exposure of three minutes was required.
These tests are in striking agreement with the results
reported in table 1 and demonstrate that in the presence
of 450 ppm hardness at an alkaline pH iron in a con-
centration of 5 ppm or less does not effect the killing
action of benzalkonium chloride for E. coli. In the series
of tests containing no bicarbonate the pH of the reac-
tion mixture was dependent upon the amount of iron
salt present. At concentrations of iron sufficiently low
to result in a pH above 6 a delay in the time required to
kill all the exposed cells was observed, similar to that
seen with hard water. At higher concentrations of iron
and the resultant decrease in pH, a more pronounced
delay in the killing of the exposed cells occurred. This
increase in the delay in killing all the exposed cells is a
pH effect as shown by the buffered tests.
In view of the results obtained in distilled and hard
water with iron it is apparent that this ion, in concen-
TABLE 4. Effect of iron and pH in hard water on the germicidal action of 200 ppm benzalkonium chloride on Escherichia coli 198 at 25 C
Reaction Mixture*
Aluminum, in an alkaline environment, has no effect
upon the germicidal action of benzalkonium chloride;
in an acid menstruum, however, it markedly delays the
killing action of the germicide. At pH 6.6 with no
aluminum present, all cells were promptly killed by the
benzalkonium chloride; at 6.7 in the presence of
aluminum an insufficient number of the exposed cells
were killed to permit the counting of those surviving
the germicide.
Effect of natural waters. It should be borne in mind
that the synthetic hard water used in the experiments
herein reported did not, of necessity, contain those
constituents and/or contaminants, some known, others
not, that characterize a natural water; it resembled a
natural hard water only in that it contained calcium
and magnesium. To assay the effect of natural water
upon the germicidal action of benzalkonium chloride a
well water of 328 ppm hardness and Rensselaer tap
water (56 ppm hardness) were used. The results of two
tests on these waters are shown in table 6, where it will
be seen that a delay in the killing of all the cells exposed
to the germicide occurred with both waters, but the
exposed cells were all killed on 3 minutes' exposure.
Furthermore, the most pronounced delay, as measured
by the number of cells surviving 30 seconds' exposure to
the quaternary, occurred in the less hard Rensselaer
tap water, demonstrating that the constituent causing
the delay is not hardness. The nature of the interfering
constituent (or contaminant) has not been identified.
No. Cells Surviving Exposure to Germicide

Buffered Unbuffered
Hardness Iron
pH 30 sec 60 sec 120 sec pH 30 sec 60 sec 120 sec
ppm p pm
0 0 7.1 0 0 0 7.1 0 0 0
450 0 8.0 11780 5260 0 6.8 4920 3515 95
450 1 8.0 20 0 0 6.1 2650 180 60
450 2.5 7.8 1310 0 0 4.6 140,800 36,800 720
450 5 7.7 10 0 0 3.9 TNC TNC TNC
* Reaction mixture contained 75.75 X 106 cells per ml
56 J. W. KLIMEK AND J. H. BAILEY [VOL. 4
TABLE 5. Effect of aluminum and pH in hard water on the germicidal action of 200 ppm benzalkonium chloride on Escherichia coli 198
at 25 C
Hardness No. Cells Surviving Exposure to Germicide*

Germicide Reaction
Al+++Buffered Unbuffered
Cell system systems Reaction
pH 30 sec 60 sec 120 sec pH 30 sec 60 sec 120 sec
ppm PPm pPm ppm
0 0 0 0 7.9 0 0 0 6.6 0 0 0
900 0 450 0 8.7 115 0 0 8.35 115 0 0
0 900 450 0 7.6 10 0 0 6.6 365 0 0
450 450 450 0 8.4 10 0 0 7.5 40 0 0
0 0 0 5 7.2 0 0 0 4.1 TNC TNC TNC
900 0 450 5 8.0 1180 20 0 6.6 TNC TNC TNC
0 900 450 5 7.15 485 10 40 4.2 TNC TNC TNC
450 450 450 5 7.8 30 0 0 6.7 TNC TNC TNC

Downloaded from http://aem.asm.org/ on August 18, 2017 by guest

* Medication mixture contained 81.6 X 106 cells per ml.
TABLE 6. Effect of two natural waters on the germicidal action Effect of organic matter. The germicidal action of
of 200 ppm benzalkonium chloride on Escherichia coli 198 benzalkonium chloride, in common with many other
at 25 C
germicides, is adversely affected by organic matter.
Type of
Wty ero
Hardness
Reaction
Number of Cells Surviving Exposure Number
cells/ml
The Allport synthetic dirty water, containing 3980
Water Mixture 30 sec 60 sec 120 sec 180 sec X 106
ppm total solids, was used as a source of organic matter
ppm. in the experiments designed to study this effect. In ex-
periments using organic matter in but one of the systems
Distilled... 0 0 0 0 _ 186
Distilled.. 0 0 0 0 115 a double strength (7960 ppm) Allport water was em-
Rensselaer. 56 TNC 52000 140 0 186 ployed. Addition of the second system prepared in
Well. 328 8100 103 120 0 186 distilled water then gave a medication mixture con-
Rensselaer. 56 76800 5500 0 0 115 taining the standard 3980 ppm total solids of the All-
Well. 328 38400 11910 30 0 124
port formula. Experiments were conducted at both 25
and 45 C; the results are summarized in table 8.
Although what this constituent may be is as yet un- At 45 C organic matter had no effect upon the germi-
known, the results presented in table 7 show that it cidal action of benzalkonium chloride, practically all
acts primarily upon the cell system. In these experi- the exposed cells being killed by 30 seconds' exposure to
ments the reaction mixtures were composed of (a) cell the quaternary. At 25 C, however, organic matter
and germicide system in Rensselaer tap water; (b) caused a prolongation of the time required for the
cell system in Rensselaer tap water, germicide in buf- killing of all the cells. This prolongation was most pro-
fered distilled water; (c) germicide in Rensselaer tap nounced when the organic matter had been added to
water, cell system in buffered distilled water and, (d) the germicide system, thus indicating that the action
both systems in buffered distilled water. Where the of organic matter is not on the cells, as is the case with
cell systems contained Rensselaer tap water a marked calcium and magnesium ions and natural waters, but
delay in the killing of all the cells occurred and was most upon the germicide. Furthermore, the attenuating
pronounced where both systems of the test were pre- action of the organic matter can be negated by a tem-
pared in Rensselaer tap water. perature of 45 C.

TABLE 7. Effect of Rensselaer tap water in each system on the germicidal action of 200 ppm benzalkonium chloride on Escherichia coli
198 at 25 C
System Number of Cells Surviving Exposure* Per cent Cells Killed on Exposure

Cell Germicide 30 sec 60 sec 120 sec 30 sec so sec 120 sec

BDWt BDW 10 0 0 99.9999 100 100

RTt RT TNC§ 4860 1845 99.9943 99.999
RT BDW 7560 2400 10 99.9912 99.9972 99.9999
BDW RT 440 180 30 99.9995 99.9998 99.9999
* Medication mixture contained 85.8 X 106 cells per ml.
t BDW: Buffered distilled water.
t RT: Rensselaer tap.
§ TNC: Too numerous to count.
19561 FACTORS IN KILLING ESCHERICHIA COLI 57
TABLE 8. Effect of organic matter in each system on the germi- TABLE 9. Effect of organic matter and hardness in each system
cidal action of 200 ppm benzalkonium chloride on Escherichia on the germicidal action of 200 ppm benzalkonium chloride
coli 198 at 25 C and 45 C on Escherichia coli 198 at 25 C and 45 C
Organic Matter Number of Cells Surviving Exposure*
System System Reac- Number of Cells Surviving Exposure
tion
Cell Germicide Reaction 25 C 45 C Mixture 25C45C
system system mixture Hr- Hard- Hard- 2
30 sec 60 sec 30 sec 60 sec nsness ness
ness ness
Z ' Hrd- I ard- 30 sec 60 sec 30 sec 60 sec
0 0 0 0 0 0 0 ppm ppm ppm
At A A TNC$ 340 20 0
0 2A A TNC TNC 30 0 A* 300 A 300 300t TNCt TNC 51,635 32,950
2At 0 A 2080 280 140 0 D§ 0 A 600 300¶ TNC TNC TNC 1,820
A 600 D 0 300T TNC TNC 5,567 340
* Medication mixture for 25 C tests contained 141.4 X 106
cells per ml; for 45 C, 164.5 X 106 per ml.
*
A = Regular Allport dirty water.
t 2A: Double strength Allport dirty water; A = regular t Medication mixture contained 96.7 X 106 cells per ml.
strength. t Too numerous to count.

Downloaded from http://aem.asm.org/ on August 18, 2017 by guest

§ D = Allport dirty water, double strength.
t Too numerous to count. ¶ Medication mixture contained 141.4 X 106 cells per ml.

If calcium and magnesium as well as organic matter TABLE 10. Effect of peptone on the sensitivity of Escherichia
were present in the reaction mixture, a more pronounced coli 198 in 450 ppm hard water to the killing action of 200
effect upon the germicidal action might be expected. ppm benzalkonium chloride at 25 C
Under these circumstances, interfering substances, one
Number of Cells Surviving Exposure Cells/ml
acting upon the cells, the other upon the germicide, Peptone exposed
would be present. The experimental results shown in 30 sec 60 sec 120 sec 180 sec X 106
table 9 demonstrate this to be the case. At 25 C so Bacto ........... 60 0 0 0 58.6
many cells survived a one-minute exposure to the Proteose ........ 10 0 0 0 70.3
benzalkonium chloride they could not be counted; at Proteose #2 ... 550 0 0 0 79.3
45 C the number surviving the same exposure period Thiotone........ 30 0 0 0 62.8
ranged from 32,950 to 340 out of 96.7 million. The Tryptone ....... 1,475 20 0 0 74.7
Neopeptone..... 325 0 0 0 68.5
greatest number of cells were killed in the reaction Evans .......... 125 0 0 _ 120.6
mixture prepared with the cell system containing the Gurrs ......... 125 10 0 - 76.1
organic matter. Oxo ........ 50 0 0 71.7
Effect of peptone. Those factors discussed above that British Drug
influence the time required to kill all the cells exposed House ........ 130 30 0 62.8
to benzalkonium chloride have been extrinsic, acting Casitone ........ TNC* TNC 16,000 1,700 78.5
Peptonum
upon cells maintained as closely as possible in a constant siccum ........ 13,520 11,666 460 0 75.75
(or "standard") condition by controlling the medium Phytone ........ 34,500 11,450 3,740 300 32.6
upon which they were grown and scrupulous adherence Polypeptone. TNC 185,600 51,200 5,780 76.1
to the transfer and incubation routine prescribed in the * Too numerous to count.
Weber and Black procedure. Such organisms may be
considered as possessing attributes conserved, if not
induced, by the management they receive in the labora- of the inoculum. The cells were harvested after 20 hours'
tory. Under certain specific conditions, the resistance incubation and their susceptibility determined by the
of these cells to benzalkonium chloride, as shown by a Weber and Black procedure using 450 ppm synthetic
prolongation of the time required to kill all of them, is hard water for the preparation of both the cell and
one of the attributes of these cells. germicide systems.
It can be demonstrated that this attribute of relative The results obtained with these cells are presented in
resistance is induced by the medium upon which the table 10, where it will be seen that media containing
cells are grown. This was done by preparing media that either casitone, phytone, polypeptone or the official
differed from that called for in the Weber and Black peptonum siccum for Disinfectant Testing (U. S. D. A.,
procedure only in the peptone used. Bactopeptone, 1931) produced cells of considerably greater resistance
casitone, neopeptone, phytone, polypeptone, proteose to the quaternary than did any of the other 10 peptones
peptone, proteose peptone # 2, thiotone, tryptone, the tested. The sensitivity of E. coli 198 (at least) is ap-
peptones sold by British Drug House, Evans, Gurrs, parently a function of the medium upon which it is
Oxo, and, for comparison, peptonum siccum were em- grown, increasing or decreasing depending upon the pep-
ployed. These media were inoculated with E. coli 198 tone used in the medium. This was illustrated in the
grown on the medium routinely used for the production following experiments. Ten consecutive transfers of E.
58 J. W. KLIMEK AND J. H. BAILEY
coli 198 were made on medium containing bactopeptone
except for the ninth, which was made on peptonum
siccum medium. Cells from the first and eighth trans-
fers were killed by exposure for one minute to 200 ppm
benzalkonium chloride, whereas those from the ninth
transfer required more than two minutes' exposure for
the same result. Cells from the tenth transfer were as
sensitive to the quaternary as those of the first or
eighth.
DISCUSSION
The killing of bacterial cells by a chemical agent is
herein considered as the resultant of the reactions be-
tween two systems: the cells and their suspending
medium and the germicide and its diluent. Our results
show that if the reaction mixture be composed of
benzalkonium chloride, Escherichia coli 198 and
distilled water, all (99.9999 per cent) the exposed cells
are killed at a rapid rate (ca. 30 seconds), whereas if
certain ions, especially calcium and magnesium, be
present the time required for the germicide to kill all
the exposed cells is increased to approximately three
minutes. This would indicate that these ions are exert-
ing some effect upon either the cells or the cationic
quaternary. As far as the latter possibility is concerned,
it is difficult to develop a concept wherein ions of like
charge would react. One is thus lead to the assumption
that the action of calcium and magnesium ions is upon
the cell system, the validity of which assumption is
demonstrated by the fact that E. coli 198 suspended in
distilled water and exposed to benzalkonium chloride
diluted in hard water are killed as promptly as when
distilled water is used in both systems.
What the ultimate factor or factors that cause this
delay in the killing of a relatively small portion of the
bacteria exposed to benzalkonium chloride may be, re-
mains unanswered. The experiments here reported
clearly show that the peptone of the medium upon
which the cells are grown determines whether or not
calcium and magnesium ions will bring about a delay
in the killing of all the exposed cells. It may be inter-
esting but it is futile to speculate as to what factor or
factors may be present or absent in those peptones that
produce cells showing this phenomenon. The fact that
a majority of the peptones tested produced cells that
did not show the hard water effect, strikingly empha-
sizes this phenomenon as a special feature of a few
peptones. It follows that the delay in the killing of all
the cells exposed to benzalkonium chloride in the
presence of calcium and magnesium ions should be
regarded as an artifact encountered only in the labora-
tory where the cells have been grown on a medium
containing a specific peptone. Were it not for the fact
that the peptone specified in the FDA, and hence the
Weber and Black method for testing of disinfectants, is
[VOL. 4
one of those causing the delay in killing by benzal-
konium chloride the retardant action of hard water
would probably have escaped notice.
The apparent discrepancy between those evaluation
studies showing a lack of prompt killing by benzal-
konium chloride of all the exposed cells in hard water
and those of Mallman et al. (1946) that showed a highly
satisfactory performance in the field in hard water
areas, disappears when one recognizes that the cells
Mallman was killing with benzalkonium chloride were
naturally occurring bacteria and not those propagated
on a medium rendering them resistant to the killing
action of the quaternary.
Whereas the effect of calcium and magnesium ions on
Downloaded from http://aem.asm.org/ on August 18, 2017 by guest
benzalkonium chloride is nil, the action of organic
matter, as represented by proteins and lipids, is marked.
This is clearly demonstrated in the experiments with
the Allport dirty water where preparation of the
germicide system with dirty water has a more pro-
nounced effect upon the rapidity with which the cells
were killed than when the other system contained the
organic matter.
Of the 3980 ppm solids of the Allport water 2680 are
contributed by protein, starch and lipid. The molecules
of the first two substances offer large surfaces for ad-
sorption and, at least with the protein, possibility of
chemical reaction with the quaternary; the lipid is
capable of both adsorption and solution of the benz-
alkonium chloride whence it is held in the suspended
fat droplets which tend to rise to the surface of the
water. As a consequence of adsorption and/or solu-
tion, these molecules tend to shunt the germ cide from
the bacteria exposed to it. The fact that benzialkonium
chloride does kill the exposed cells under conditions
militating against contact between it and the bacteria
indicates a selective adsorption by them of the germi-
cide, and a reduced but still effective residue of free
germicide.
A rough idea of the magnitude of this selective ad-
sorption is given by the following: If one assumes the
maximum weight of a living bacterial cell to be 1 X
10- mg (Lamanna and Mallette, 1953) the ratio of
cells to organic matter (protein, starch and lipid) in
these experiments was approximately 1:2.6.
In studying the effect of iron and aluminum the
concentration used was selected as one representing an
excess over that usually present in water supplies. At
higher concentrations these ions prolong the time
necessary to kill all the cells exposed to benzalkonium
chloride. Whether this effect is due to the acidity of the
salt supplying the ion or to the ion itself is more or less
of only academic interest. The important point is that
at concentrations of iron usually found in water supplies
this ion does not affect the germicidal activity of benz-
alkonium chloride in either alkaline or acid (5.0 or
1956] FACTORS IN KILLING ESCHERICHIA COLI
above) pH values. Aluminum in an alkaline medium is
also without effect.
SUMMARY
Calcium and magnesium, iron, aluminum, certain
natural waters and organic matter have been reported
in the literature to adversely affect the germicidal
action of quaternary ammonium compounds. In the
experiments herein reported the effect of these sub-
stances has been measured in terms of the time re-
quired to kill all (99.9999 per cent) the organisms ex-
posed to 200 ppm benzalkonium chloride at 25 C in
their presence.
Under certain conditions calcium and magnesium
ions (hard water) cause a prolongation in the time
necessary to kill all the exposed cells. This prolongation
is, however, an artifact, its occurrence depending upon
the peptone used for the production of the cells. Of 14
peptones tested only 4 produced cells showing this
effect. The action of calcium and magnesium ions in
producing this effect is on the cells.
Organic matter acts upon the germicide to cause a
prolongation in the time required to kill all the exposed
cells. The effect of organic matter can be negated by
increasing the temperature at which the cells are
exposed to the germicide from 25 to 45 C.
Independent of hardness, certain natural waters
bring about a prolongation in the time required to kill
all the exposed organisms. This increase is not, how-
59
ever, great. The constituent or contaminant of these
waters causing the prolongation is unknown.
Iron in a concentration of 5 ppm or less has no effect
upon the germicidal action of benzalkonium chloride.
Aluminum at the same concentration also has no effect
in an alkaline medium. At an acid pH, however, this
ion acts upon the cells to produce a prolongation in the
time necessary to kill all the exposed cells.
REFERENCES
LAMANNA, C., AND MALLETTE, M. F. 1953 Basic Bacteriology,
p. 35. The Williams and Wilkins Co., Baltimore, Mary-
land.
MALLMANN, W. L., KIVELLA, E. W., AND TURNER, G. 1946
Sanitizing dishes. Soap and Sanit. Chemicals, 22, 130-
Downloaded from http://aem.asm.org/ on August 18, 2017 by guest
133; 161.
MUELLER, W. S., AND SEELEY, D. B. 1948 Effect of some
water constituents on a quaternary salt. J. Diary Sci.,
31, 723-724.
RIDENOUR, G. M., AND ARMBRUSTER, E. H. 1948 Some fac-
tors affecting the properties of quaternary ammonium
compounds as sanitizers. Am. J. Public Health, 38, 504-
511.
SEELEY, D. B. 1949 The effect of certain metallic ions on the
germicidal activity of quaternary ammonium compounds.
Ph.D. Thesis, University of Massachusetts.
UNITED STATES DEPARTMENT OF AGRICULTURE 1931 Circular
198.
WEBER, G. R., AND BLACK, L. A. 1948 Laboratory proce-
dures for evaluating practical performance of quaternary
ammonium compounds and other germicides proposed for
sanitizing food utensils.Am. J. Public Health, 38, 1405-
1417.