Journal of Environmental Treatment Techniques  
2021, Volume 9, Issue 1, Pages: 335-341  
J. Environ. Treat. Tech.  
ISSN: 2309-1185  
Journal web link: http://www.jett.dormaj.com  
https://doi.org/10.47277/JETT/9(1)341  
Development of an Electrolytic Pilot Plant for the  
Production of Chlorine Gas “In Situ” in the  
Disinfecting Water Process  
2
Ruben Guerra *, Marlene Ballestas , Alonso Barrera , Vladimir Pinzón , Henry Herrera  
1
1
2
1
1
Department of engineering, ITSA University, Soledad, Colombia  
2
Department of basic sciences, De la Costa University, Barranquilla, Colombia  
Received: 19/08/2020  
Accepted: 23/11/2020  
Published: 20/03/2021  
Abstract  
Technology of gaseous chlorine in water treatment has shifted to the dosage of sodium hypochlorite or calcium hypochlorite because of  
their greater biocide power; require less contact time with the microorganisms in the pretreated water, and the pH of slightly affect water.  
The generation of chlorine gas in place is based on the principles of electrolytic dissociation and laws of electrolysis by Faraday. The  
equipment corresponds to an electrolytic cell of three (3) compartments , a central one (anode) where chlorine gas emerges , and two side  
(
cathodes) where sodium hydroxide is produced in the central compartment must be refilled chloride solution sodium consumed due to the  
electrochemical reaction , the chlorine evolved being of gaseous nature ascends the column of sodium chloride and is captured by the vacuum  
venture system , who to put in direct contact with target that has been previously subjected to the processes of uptake , coagulation ,  
sedimentation and filtration. Compartments anode-cathode - anode are physically separated by porous diaphragm, whose purpose is to permit  
selective flow of sodium and chloride ions , avoiding side reactions recombination , as additional product is the formation of sodium  
hydroxide in the cathode chambers, which should be removed from the cell through side pipes. Statistical models were used to optimize the  
performance and operation of the prototype.  
Keywords: Saline electrolysis; chlorine gas; electrolytic reactor  
1
built (1), which is loaded with concentrated brine (6) to which  
1
A Headings are the primary heading type  
electric current supplied by a DC power rectifier source is applied  
3), the mixture of gases generated in the anode are extracted by  
One of the most common diseases among the population of  
Latin America are related to lack of water and contaminated water  
1]; taking millions of lifes per year [2]. In most large cities in the  
(
the suction created by the Venturi system (2), the pump (5)  
circulates a working volume (4) of 200 lt at a flow rate of 15,3  
Lt/min (9,18 mt3/h) , passing through the Venturi and the shunt  
[
country with more than 10000 inhabitants, chlorination has been  
made reasonably reliable using gaseous chlorine dosing  
technology supplied in cylinders, however, chlorination of  
potable water supply systems that serve smaller populations  
usually remains unreliable and intermittent [3]. To help solving  
this problem, many technological alternatives applicable to  
disinfection have been investigated [4 and 5]. These include  
various technologies suitable for ozonation [6], iodine [7],  
ultraviolet radiation [8], as well as the use of various methods for  
performing gas chlorination, chloramines, chlorine dioxide and  
methods for generating disinfectants in situ [9 and 10]. The latter  
method being the one that has proven to be the most promising of  
all the technologies used [10].  
(
7). As seen in the Fig. 1 and Fig. 2.  
2.2 CAD design and construction of an electrolyte cell with a  
capacity of 15 A  
The reactor design used in the tests consists of an electrolyte  
cell loaded with 20 liters of brine at concentrations of 45 and 68  
gr NaCl/lt respectively, which produces oxidizing gases (chlorine  
and oxygen species) in the anode compartment, and sodium  
hydroxide and hydrogen in the cathode compartment, which is  
shown in the fig. 4. These two compartments are separated by a  
semipermeable polysulfonated membrane usually made of Nafion  
@
, General purpose material in the manufacture of car batteries,  
the prototype body was made of 5mm thick transparent acrylic  
sheets resistant to the action of chlorine, ozone and NaOH. The  
cathode of the unit is made of stainless steel and a matrix of 22  
cylindrical graphite electrodes 10mm in diameter by 300mm of  
length.  
2
Plant Development  
2
.1 CAD design and construction electrolytic pilot plant for gas  
chlorine production  
For the purpose of evaluating the performance of chlorine-  
generating equipment in situ an electrolytic cell was designed and  
*
Corresponding author: Ruben Guerra, Department of engineering, ITSA University, Soledad, Colombia.  
335  
Journal of Environmental Treatment Techniques  
2021, Volume 9, Issue 1, Pages: 335-341  
you to adjust the intensity of the vacuum and create an alternate  
path to water, decreasing the pump load and increasing its flow  
rate.  
Figure 1: Schematic diagram of the test team  
Figure 4: 3D design of chlorine gas mixing with Venturi  
2
.4 Design and construction of a 25A DC electric power rectifier  
source  
For the supply of current to the electrolytic cell, a DC voltage  
source with a capacity of 25A was implemented, which is shown  
in Fig. 6. The source construction is part of the maximum current  
and voltage values required by the cell.  
Figure 2: 3D design of the test team  
Figure 5: DC source electronic design  
2
.5 Construction of the sodium chloride solution dosing system  
The level control in the reactor brine is based on the  
application of an operational amplifier working as a voltage  
comparator circuit, this is shown in the fig. 7.  
Figure 3: 3D design of the Electrolytic Cell generating oxidizing gas mixture with  
capacity of 15 A  
2
.3 CAD design and construction of a chlorine gas mixing  
system with water by Venturi principle  
The chlorination system applied in the project was developed  
with Venturi tube which is able to generate a suction of emptying  
in the anode chamber, enough to extract the production of  
oxidizing gases including chlorine gas, and injecting it directly  
into the pump-driven water stream, while serving as a mixing  
chamber between chlorine and water. A shut-off valve was  
arranged by pass with the Venturi, as shown in the fig. 5, it allows  
Figure 6: Electronic design of sodium chloride solution dosing system  
336  
Journal of Environmental Treatment Techniques  
2021, Volume 9, Issue 1, Pages: 335-341  
2
.6 Grafcet program for the automation of the chlorination  
3.2 Experimentation methodology  
plant  
This study case falls within the analysis of experiment 2  
through 4, in which there are 4 main factors, with 2 possible levels  
for each, the main interest of these tests being to determine which  
factors have the greatest significance in the response of the model  
studied, and to rule out the effects of non-significant factors, and  
is evidenced in Table II. Statistical model 2 through 4 includes 16  
possible combinations of treatments for which the response may  
vary, each combination of treatment indicates the behavior of the  
output of the system or variable evaluated against random  
changes in the inputs. The tests were carried out in a Bach-like  
system. The production of free chlorine in the tank with 200Lt  
was estimated by applying the titration technique with chemical  
reagents, using the HTH product KIT. Measurements were made  
The control of the equipment is clearly sequential, as shown  
in fig. 8, which objective is to turn outputs on and off (DC Source,  
Pump and Valve) coordinatedly, depending on the information of  
the entries (Start, Stop, Level R, Level T). While in its initial  
stage the program moves towards simultaneous activation of  
stages 2 and 3 that activate the DC Source and the pump after  
checking the minimum level in the tank and a call to start with the  
memory %M5.  
3
by extracting 8 cm of the treated water every 5 minutes and  
adding 5 drops of the ortho-tolidine reagent HTH ® which causes  
2
the sample to take yellow hues, from the clearest to 0.5 PPM Cl  
2
to the most intense with 5 PPM Cl determining concentration by  
visual comparison.  
Table 2:  
Figure 7: For the automation of the chlorination plant  
3
Checking and Verifying the Proposed System  
Using Statistical Models  
The previous study for the design of an electrolyte cell to  
produce chlorine gas in situ assumes that the most important  
factors affecting chlorine gas production and its Cf concentration  
(
ppm) in the chlorine gas dissolution tank are:  
Current: It is the energy supplied by the DC voltage source  
continuous current) and which can break down the  
(
molecular structure of the NaCl sodium chloride and thus  
produce the chlorine gas.  
Time: Duration of the process being tested and measured in  
minutes.  
Concentration: Number of grams of sodium chloride dissolved  
per liter of NaCl grams/liter solution.  
PH: It is a measure of acidity or alkalinity of a dissolution The  
+
PH indicates the concentration of hydronium ions (H  
3
O ).  
Table 1:  
Factors  
Current  
Time  
Low Level  
10A  
High Level  
15A  
Multifactor variance analysis of data was developed with  
Statgraphics Centurion XV Version 16.1.11 Software. The Anova  
Multifactorial result for a Maximum Order of Effect of 2  
20 min  
45 gr/lt  
40 min  
68 gr/lt  
(
considering the effect of The Main Factors and Interactions only)  
Concentration  
PH:  
shows the results in Table III. In this case, 3 effects and two  
interactions have a P-value less than 0.05, indicating that they are  
significantly different from zero with a confidence level of 95.0%.  
The R-Square statistic indicates that the model explains 98.503%  
of the variability in Cf (PPM). The adjusted R-square statistic,  
which is best suited for comparing models with different number  
of independent variables, is 98.4254%. The Pareto diagram shows  
graphically in Fig. 9, the magnitude of the effects of factors on the  
final concentration Cf (ppm) in this case Factor A: Current;  
6.5(H  
3
O
+
)
3 +  
7.5(H O )  
3
.1 Selecting the response variable  
The response variable needed by the designer of an electrolyte  
cell for chlorine gas production is the final concentration of  
chlorine dissolved in parts per million of Chlorine (PPM Cl ).  
2
337  
Journal of Environmental Treatment Techniques  
2021, Volume 9, Issue 1, Pages: 335-341  
Factor B: Time; Factors C: Concentration; and factor A  
interactions with C and the interaction of factor B with C. The  
ANOVA result is confirmed by the Normal Probability chart for  
the Cf (ppm) response where it is observed that Factor A: Current;  
Factor B: Time; Factor C: Concentration; and the interactions of  
A and C; B and C are the most relevant in the system, because  
they move away from the normal probability line. The  
distribution of errors is normal indicating that Fig. 10 is a straight  
line.  
experiment.  
Table 5:  
Table 3:  
Figure 9 reveals the good correlation between the results of  
free chlorine measurements against regression values.  
Figure 8: Normal Probability Chart  
Table 4:  
Figure 9: Experimental free chlorine Vs estimated (regression No1)  
The number of levels was expanded to three (3) for each  
significant factor in order to obtain a more reliable regression  
curve tailored to the model, with a factorial design 33 of 27  
treatments, as recorded in TABLE VI. As a result of regression  
analysis No2, applied to Table 5, a curve is shown in Tables 7 and  
8, a curve that only contemplates the most important factors  
(
Current-Time and Concentration).  
This analysis will only consider FactorS A: Current; Factor  
B: Time; Factor C: Concentration and interactions of A with C  
and B with C, shown in Table IV. The No.1 regression equation  
obtained is: Cf (ppm) = - 65,9022 + 5,48696* Current - 0,216304*  
Time + 0,847826* Concentration - 0,0630435* Current*  
Concentration + 0,00869565* Time* Concentration. (See table  
V). It also yields Cf (PPM) values close to the model and  
Regression Model Nº2  
Cf (I, T, C) = - 0,56052632+0,27397661*Current + 0,13055556*  
Time + 0,02202643* Concentration.  
Noting that the regression coefficient associated with the  
Concentration factor is very low (0.02202643), a new regression  
was found that only considers the most significant factors  
338  
Journal of Environmental Treatment Techniques  
2021, Volume 9, Issue 1, Pages: 335-341  
(
Current and Time) resulting in a regression model Nº3 simpler  
Table 10:  
of the process, which is evidenced in the tables 9 y 10, fig. 12.  
Regression Model No. 3  
Coefficients  
P Value Probability  
1.62817E-13  
1.04431E-10  
5.27966E-21  
1.56206E-06  
Interception  
-3.310526316  
0.273976608  
0.13055556  
0.130555556  
Table 6:  
Variable X 1 (Current)  
Variable X 2 (Time)  
Variable X 3 (Concentration)  
Figure 10: Regression graphs (Experimental, Nº2 y Nº3)  
3
.3 Residual analysis  
There is no characteristic pattern, this indicates that the  
residues are independent or randomly distributed. Fig. 13 does not  
see any pattern or funnel shape, so the data can be considered to  
satisfy the linearity criterion. The debris fits quite well to a  
straight. The points (errors) fall close to the straight line,  
distributed both above and below it, i.e. there is no characteristic  
pattern, indicating that the residues are independent and randomly  
distributed, in conclusion, there is normality in the errors.  
Table 7: Regression statistics No2  
Regression statistics No2  
Multiple correlation coefficient  
0.992270038  
0.984599828  
0.9825111  
0.165675926  
27  
2
Determination coefficient R  
Adjusted R2  
Typical error  
Observations  
Table 8: Regression statistics No2  
Regression Model No. 2  
Figure 11: Residual analysis  
Coefficients P Value Probability  
-4.5605263 1.62817E-13  
Interception  
3
.4 Response surface  
Three-dimensional response surface where cf (PPM)  
Variable X 1 (Current)  
Variable X 2 (Time)  
Variable X 3 (Concentration)  
0.27397661 7.17884E-15  
0.13055556 5.27966E-21  
0.02202643 1.56206E-06  
production of gaseous chlorine is observed based on current and  
time. In this case there are two influence factors (k-2), the  
response surface is displayed in a three-dimensional space in  
which the third dimension represents the expected production of  
chlorine gas Cf (ppm) on the two-dimensional plane defined by  
the combinations of the current and time factor levels, shown in  
Figure 12.  
Regression Model Nº3  
Cf= -3,31052632 + 0,27397661*Current + 0,13055556*Time  
Table 9: Regression statistics No3  
Regression statistics No3  
Multiple correlation coefficient  
0.978343926  
0.957156837  
0.953586574  
0.27.517697  
27  
2
Determination coefficient R  
Adjusted R2  
Typical error  
Observations  
339  
Journal of Environmental Treatment Techniques  
2021, Volume 9, Issue 1, Pages: 335-341  
containing after this time, this data is shown in TABLE 12 and  
Fig. 16.  
Figure. 12: Response surface  
Figure. 13: Closed circuit - Recirculation performance  
3
.5 Equipment performance in closed-circuit recirculation  
The team's behaviour was assessed by measuring the increase  
in Cl2 concentration by a volume of 200lt that was recirculated at  
a flow rate of 15.3Lt/min, incrementally capturing the gases  
generated through the Venturi. And it is evidenced in Table 11  
and fig. 15. For this test, the cell was loaded with 20 Lt of brine  
at a concentration of 45 gr of NaCl/Lt and a current of 8A in the  
source. It should be noted that these current and concentration  
levels can be considered as the minimum adjustment levels of the  
equipment, being able to work at concentrations of 68 gr of  
NaCl/Lt and 15 A.  
Table 12:  
Chlorine accumulation in a 200Lt water Volume  
Free Chlorine :PPM  
Free Chlorine :PPM  
Time  
(Min)  
Current I:10A  
Time  
Current I:10A  
Concentration:  
45grNaCl/Lt  
(Min)  
Concentration:  
45grNaCl/Lt  
8.5  
0
0.0  
1.0  
1.5  
2.0  
3.0  
3.5  
4.5  
5.0  
6.0  
6.5  
7.0  
7.0  
8
195  
210  
225  
240  
255  
270  
285  
300  
315  
330  
345  
360  
1
3
4
6
7
9
5
0
5
0
5
0
9.0  
9.5  
10.0  
11.0  
11.5  
12.0  
12.5  
13.0  
Table 11:  
Free Chlorine: PPM Current I: 8A  
1
1
1
1
1
1
05  
20  
35  
50  
65  
80  
TIME (Min)  
Concentration: 45grNacl/Lt  
13.5  
14.0  
14.0  
0
5
0.0  
1.0  
1.5  
2.0  
3.0  
4.0  
5.0  
6.0  
8.0  
9.0  
9.5  
10.0  
10  
1
0
5
0
5
0
5
0
5
0
5
0
1
2
2
3
3
4
4
5
5
6
This test reveals that even at minimum adjustment levels the  
equipment could leave a 10 PPM footprint on 200lt of water  
which would amount to saying it would take 2m3 of water to 1  
PPM in an hour.  
Figure. 14: Open circuit - Performance without recirculation  
With these concentration and flow values the equipment can  
carry a flow rate of 12.8 m3 to 1 ppm of chlorine in one hour. The  
measurement of equivalent residual chlorine was carried out as  
follows:  
3
.6 Open circuit - Performance without recirculation  
For this test, the cell was loaded with 20 Lt of brine at a  
concentration of 45 gr of NaCl/Lt and a current of 10A in the  
source. The total time of the experiment was 360 min, during this  
time it is observed the increase in the concentration of the output  
flow from a value of 1 ppm to the five minutes of the test to a  
value of 14 ppm at 360 min of the test, keeping this value  
Water samples were taken every 15 minutes in a specimen,  
added to the Cl meter by comparison.  
2
Adding 5 drops of Orthotolidine Hydrochloride.  
The color of the sample was compared to the meter standards  
340  
Journal of Environmental Treatment Techniques  
2021, Volume 9, Issue 1, Pages: 335-341  
by colorimetric comparison of the sample. 0,5 – 5 PPM.  
When the concentration exceeded 5 PPM dilutions were made  
in which 10 ml of the sample was taken and taken to 20ml, 30ml,  
3
4
5
6
7
8
9
Carrasco IR, Morales HS. Importancia de la cloración del agua: sitios  
de abastecimiento con presencia de bacterias patógenas.  
Enfermedades Infecciosas y Microbiología. 2019 may 15.  
González L, Saavedra N, Saavedra A. Tecnologías e innovaciones  
para la purificación del agua. Teoría y praxis investigativa. 2018 july  
4
0ml, 50ml, 60ml, with distilled water and this solution samples  
were taken for comparison. Dilutions were taken into account to  
calculate the total amount of residual chlorine in treated water.  
1
2.  
Bolisetty S, Peydayesh M, Mezzenga R. Sustainable technologies for  
water purification from heavy metals: review and analysis. Chemical  
Society Reviews. 2019 jan 03.  
Sánchez GM. (2019). Agua ozonizada, antecedentes, usos en  
medicina y bases preclínicas. Revista Española de Ozonoterapia.  
4
Conclusions  
The prototype meets expected expectations with an efficiency  
greater than 85%.  
2
019 jan 15.  
The operation and maintenance of the equipment is very  
simple. The steps for this procedure would be: Cell washing  
after each operation, brine supply for a new chlorine gas  
production, recirculation pump ignition and level control  
system, automated controller ignition.  
Song K, Mohseni M, Taghipour F. Application of ultraviolet light-  
emitting diodes (UV-LEDs) for water disinfection: A review. Water  
research. 2016 march 03.  
Wang L, Wang Z, Zhang L, Hu C. Enhanced photoactivity of  
Bi2WO6 by iodide insertion into the interlayer for water purification  
under visible light. Chemical Engineering Journal. 2018 aug 28.  
Gea-Izquierdo E. Métodos de desinfección del agua y su implicación  
en la legionelosis. Tecnología y ciencias del agua. 2018 march 02.  
The equipment can work with batteries powered by solar cells.  
This would be applied in regions where there is no power  
supply of the electric fluid.  
The equipment can treat flow rates greater than 300 m3/day  
with a concentration of 1 ppm; this is possible since the  
electrolyte cell produces other oxidizing gases such as  
ozone, this increases the ability to water to almost double  
what is calculated.  
10 Givirovskiy G, Ruuskanen V, Ojala L, Kokkonen P, Ahola J. In situ  
water electrolyzer stack for an electrobioreactor. Energies. 2019 may  
1
8.  
If we assume a consumption of 100 liters per person per day;  
3000 people can be served, at a cost of $27.40 person/month.  
The equipment avoids the need to handle and import cylinders  
with chlorine gas or liquid sodium hypochlorite drums.  
Aknowledgment  
The authors thanks the journal of environmental treatment  
techniques for ginvig this space to develop the sciences.  
Ethical issue  
Authors are aware of, and comply with, best practice in  
publication ethics specifically about authorship (avoidance of  
guest authorship), dual submission, manipulation of figures,  
competing interests and compliance with policies on research  
ethics. Authors adhere to publication requirements that submitted  
work is original and has not been published elsewhere in any  
language.  
Competing interests  
The authors declare that there is no conflict of interest that  
would prejudice the impartiality of this scientific work of  
Development of an electrolytic pilot plant to produce chlorine gas  
“
in situ” in the disinfecting water process.  
Authors’ contribution  
All authors of this study have a complete contribution for data  
collection, data analyses, manuscript writing and transalate  
process.  
References  
1
Lara H, García EM. Prevalencia de enfermedades asociadas al uso de  
agua contaminada en el Valle del Mezquital. Entreciencias: diálogos  
en la sociedad del conocimiento. 2019 oct 18.  
2
Rodríguez JP, García CA, García, J. C. (2016). Enfermedades  
transmitidas por el agua y saneamiento básico en Colombia. Revista  
de salud pública. 2016 july 22.  
341