Surface and groundwater are the main sources of potable water in South Africa.
Groundwater contributes 9% to the countrys water supply while the rest
is sourced from surface water (77 from surface water, 14% from sewage and effluent
purification and 9% from ground water) (Van Vuuren, 2009).
Groundwater is naturally of good microbiological and chemical quality when compared
to surface water. However heterotrophic and coliform bacteria have been reported
in groundwater. Sources of pollution of surface and ground water include municipal
solid waste, sewage sludge and industrial waste and effluent (Momba
et al., 2003; Theron et al., 2008).
These pollution sources carry with them chemical pollutants and microorganisms
including enteric pathogens such as Escherichia coli, Shigella
sp., Salmonella sp. and Vibrio sp. which are causative agents
of various diseases that can result in death (Alcamo, 2001).
Consequently, the disinfection of surface and ground water is of public health
Conventional drinking water disinfection methods employed in water treatment
have their own disadvantages. Widely used disinfectants include chlorine, chloramine,
ozone and UV light (Freese and Noziac, 2004). Each of
these can have some problems, for example chlorine dissolves in water to form
hypochlorous acid, which in turn reacts with natural organic matter (NOM) (e.g.,
humic and fluvic acids) to form numerous disinfection by-products (DBPs), including
trihalomethanes (THMs), chlorinated biphenyls and other halogenated hydrocarbons
(Genthe and Kfir, 1995). These compounds are either endocrine
disruptors (EDCs) or carcinogenic (cancer causing) in humans. Such disinfection
by-products and their ill health effects challenge research into developing
other methods of disinfecting water, which do not result in toxic by-products
while ensuring the good quality of water is not compromised (Schutte
and Focke, 2007).
Novel nanomaterials offer the potential for the treatment of surface water
and groundwater contaminated with inorganic, organic and bacterial pollutants.
In our laboratories, cyclodextrin polymers containing a small percentage of
carbon nanotubes exhibited the ability to decrease the concentration priority
organic pollutants in water at concentrations as low as parts per million (ppm)
(Salipira et al., 2007). Likewise bioactive silver
nanoparticles are emerging to be good candidates for antibacterial use because
of their large surface area (Morones et al., 2005).
Silver nanoparticles have been immobilized on carbon materials such as activated
carbon and carbon aerogels and have been used as bactericidal materials (Ibarra
et al., 2007; Zhang et al., 2004).
The use of silver as an antibacterial agent dates back to ancient times when
silver vessels were used to keep and purify water, wine and vinegar (Liau
et al., 1997).
From this point of view, this study was designed to prepare cyclodextrin polyurethanes containing a small amount of carbon nanotubes supporting silver nanoparticles. The synthesized polyurethanes were then used for the disinfection of model (laboratory) and environmental water samples.
MATERIALS AND METHODS
The experimental study was conducted in the months of May 2007 to July 2008, using both environmental waters and sterile distilled water seeded with pathogenic bacteria. Microbiological analyses were performed in aseptic conditions under a biohazard safety cabinet Class II (Vivid Air Filtration and Ventilation Suppliers, SA) in the laboratory. Tests for turbidity, temperature, pH and conductivity of environmental water samples were also considered as these factors play an important role in the disinfection process. All microbiological tests were performed in triplicates before and after disinfection while physiochemical tests were done only before disinfection.
Sampling of environmental test water: Surface and ground water were
used as test waters for this study. The main purpose of our sampling was to
obtain environmental samples to evaluate the antibacterial activity of the silver
impregnated polymers in a realistic scenario. Surface water was collected from
two treatment plants which were the Magalies Klipdrift treatment plant and the
Magalies Temba treatment plant while ground water was collected from two randomly
selected boreholes situated in Delmas. Surface water samples collected from
two points along the flow path of the treatment plants. Samples were collected
from raw water inlets and after filtration. Magalies Klipdrift treatment plant
receives its raw water from the Pienaars River and the Rooderplot Dam. Magalies
Temba treatment plant receives its raw water from the Leeukraal Dam. For each
sampling points, three water samples were considered. Sampling collection was
performed according to standard procedures for microbial testing (APHA,
1989) and analyses were done within 4 h of collection.
Physico-chemical analyses of environmental test water: Temperature and conductivity were measured using potable meter (Eutech, model Cyberscan PC 300), while pH was measured using a potable pH meter (WinLab Data Line). Also, turbidity was measured using an electronic turbidity meter (HACH, model 2100N). All physicochemical parameters were measured on site following the manufacturers instructions.
Microbiological analyses of environmental waters: Water samples were
analyzed using standard methods (APHA, 1989). Escherichia
coli and Vibrio cholerae and were detected after the membrane filtration
technique using filters with 0.45 μm pore size and 47 mm diameter (Millipore).
Different volumes were filtered depending on the type of water used. Saline
water was used as a diluent for the 10 mL volumes to spread the bacteria evenly
over the filter membrane. The filter membranes were placed onto Thiosulphate
Citrate Bile Sucrose agar (TCBS) and chromocult media (Biolab) for the isolation
of V. cholerae and E. coli respectively and incubated at 37°C
for 24 h. Microbiological analyses were carried out in triplicates for each
Preparation of spiked water samples: Pathogenic Escherichia coli (ATCC 25925) were obtained from the American Type Culture Collection. The strain was then reconfirmed by cultural, morphological and biochemical tests. The manipulation of the bacterial strain was done under a biohazard safety cabinet Class II (Vivid Air Filtration and Ventilation Suppliers, SA).
Two loops of E. coli ATCC 25925 were grown in sterile nutrient broth
(200 cm3) under constant aeration on a rotary shake incubator for
24 h at 37°C at a speed of 117 rpm. Cells from the nutrient broth were then
centrifuged at 4000 rpm for 10 min and the supernatant was disposed. The cells
were then washed using Phosphate Buffered Solution (PBS) for several times (Momba
and Cloete, 1996). The washed cells were then re-suspended in tetra sodium
pyrophosphate, a surfactant to prevent clumping of cells. The concentration
of the cells harvested was determined by first serially diluting the initial
biomass suspension in sterile saline solution. The enumeration of the initial
concentration was done using the colony count method. The spiked water samples
were prepared by adding the biomass suspension into appropriate volumes of sterile
distilled water in separate 1 L sample bottles. Counts for viable E. coli
in spiked water were confirmed by serially diluting the spiked water followed
by culturing using the spread plate method (APHA, 1989).
Disinfection of test waters: Silver impregnated carbon nanotube co-cyclodextrin polymers (Ag MWNT/CD) were synthesized in our laboratories by initially impregnating carbon nanotubes with silver nanoparticles. Silver impregnation was done by the reduction of a silver salt using a weak reducing agent in a polyvinylpyrrolidone solution. The silver impregnated carbon nanotubes were then polymerized into cyclodextrin polyurethanes using a bifunctional linker. Characterization of these materials was carried out by transmission electron microscopy, scanning electron microscopy, energy dispersive x-ray. Metal content of the polyurethanes was done by acid digestion of the polyurethanes and analyzing the metal using atomic absorption spectroscopy. These polyurethanes contained 0.019% (by weight) of silver. For comparative studies, multiwalled carbon nanotubes and cyclodextrin polymers (MWNT/CD), containing no silver, were also evaluated for their antibacterial character. The polymers all contain approximately 1% by weight of carbon nanotubes.
The multi-walled carbon nanotubes polymer cyclodextrin (MWNT-CD) (0.3 g) and
silver impregnated carbon nanotubes cyclodextrin polymer (Ag-MWNT-CD) (0.3 g)
were packed in separate empty solid phase extraction (SPE) cartridges (Salipira
et al., 2008). The cartridges were connected to a separation funnel
which contained collected environmental samples (250 mL) and a collecting flask.
The water samples were allowed to filter through by gravitational force at an
average flow rate of 5 mL min-1. Prior use, plastic ware were sterilized
by soaking them in a 5% bleach solution for 24 h (Momba and
Cloete, 1996). The bleach was washed off using sodium thiosulphate and rinsed
with sterile distilled water. Glassware was sterilized by autoclaving (Momba
et al., 2006). The final viable count was again determined using
the colony count method.
Nanoparticle leaching studies: Silver impregnated carbon nanotubes and cyclodextrin polymers were packed in SPE cartridges and were connected to the disinfection set up.
Distilled water (100 mL) was passed through the polymers at flow rates of 5,
30 and 85 mL min-1. The eluted water was then analysed for the presence
of leached out silver using atomic absorption spectroscopy (Ibarra
et al., 2007).
Statistical analysis: Statistical analysis was done using the SPSS computer statistical software (version 13.0). Test of significance was carried out using the Students Independent T-Test at 95% confidence interval.
Physiochemical quality of samples: Table 1 illustrates the physiochemical characteristics of environmental samples. The temperatures ranged from 12.5-14.3°C, conductivities from 16.4-79.1 mS m-1, pH from 7.13-8.23 and turbidity readings ranged from 1.1-7.4 NTU.
Characteristics of environmental waters before and after disinfection: The enteric pathogens frequently encountered in water sources include bacteria such as Escherichia coli, Vibrio cholerae and Salmonella sp. These are usually transmitted to humans by ingestion of contaminated water or food and cause various diseases. Table 2 shows the initial bacterial profile and CFU counts of the water samples. Bacterial species isolated from environmental samples were E. coli and V. cholerae. The highest E. coli counts was 1.5x104 cfu 100-1 and was detected in Temba raw water while the highest V. cholerae counts was 7.4x103 cfu/100 mL detected in groundwater collected from from Delmas A3 borehole.
||Physio-chemical quality data (results obtained from a set
of three experiments)
|Limits for no risk: Turbidity: 0 to 1 NTU; Conductivity: <70
mS m-1; pH: 6.0 to 9.0; Temperature: 15 to 25°C (SANS 241,
DWAF, 1996). A3 and A7 are borehole water samples
||Bacterial characteristics of environmental waters before and
after treatment with cyclodextrin based polymers (average results after
a set of three sets of results)
||Bacterial counts of spiked water samples before and after
treatment with polymers (after averaging a set of three sets of results)
Physicochemical properties of water samples: Some of the parameters
were within the target water quality ranges while others were above the target
ranges (SANS 241, DWAF, 1996). For example, the conductivities
and pH values for all environmental water samples were within the target ranges.
All water samples collected during the study period had turbidity values ranging
between 1.1 and 7.4 NTU. Raw water collected from Temba and Klipdrift treatment
plants had the highest turbidity values (7.0-7.4 and 5.10-5.4 NTU, respectively)
which exceeded 5 NTU which possesses a possible health risk (DWAF
1996). Turbidity values higher than 1 NTU is a matter of concern in terms
of the performance of the disinfection process. Studies by previous investigators
have reported that the effectiveness of the disinfection process is linked to
the turbidity of the water (Momba et al., 2006).
These investigators pointed out the lower the turbidity (<1 NTU), the higher
the efficiency of the disinfection process. It is important to note that turbidity
in water is caused by the presence of suspended matter, which usually consists
of a mixture of inorganic matter such as clay and soil particles and organic
matter. The latter can be both living matter such micro-organisms and non-living
matter such death algae cells (DWAF, 1996). Measuring
the turbidity of water is therefore a good indication of the concentration of
the suspended matter in water. The temperatures were quite low because the samples
were collected in winter. The highest sample temperature was 14.6°C obtained
from Klipdrft raw water samples whilst the lowest temperature 11.8°C. The
average temperature was 13.1°C (Table 1).
Characteristic of raw surface water before and after disinfection: From results shown in Table 2, it can be noted that Temba treatment plant had 1.5x104 cfu/100 mL of E. coli and 23 cfu/100 mL of Virio cholerae while Klipdrift treatment plant had 1.4x103 cfu/100 mL and only 2 cfu/100 mL of Vibrio cholerae. Upon disinfection with the silver polymer, a 3 log reduction of E. coli from both Temba and Klipdrift raw water was observed. The silver free polymer reduced bacterial counts by 2 logs in Temba water and a 1 log reduction from Klipdrift raw water at the end of the 90 min study period. There was complete removal of V. cholerae by the silver impregnated polymers whilst the multiwalled carbon nanotubes and cyclodextrin polymer recorded 5 cfu/100 mL after disinfection.
Characteristic of filtered surface water before and after disinfection: Filtered surface water collected from the two treatment plants had low bacterial counts when compared to the raw water. Temba plant had 39 cfu/100 mL of E. coli while Klipdrift had 60 cfu/100 mL. V. cholerae was not observed from both plants after filtration. The low bacterial counts or the absence of V. cholerae can be attributed to the treatment processes employed in the water works. These include pre-chlorination in the case of Klipdrift and filtration for both plants. After disinfection with the silver polymer, no bacteria were isolated from both Temba and Klipdrift filtered water samples. The MWNT/CD polymer reduced the counts from 39 to 22 cfu/100 mL for Temba plant and 60 to 10 cfu mL-1 for Klipdrift. These counts exceeded by far the limits allowed by the South African Water Quality standards (SANS 241, 2006). Consequently this study suggests that the MWNT/CD cannot be considered for the disinfection of drinking water in terms of the removal of pathogenic bacteria from surface water.
Characteristic of raw ground water before and after disinfection: Both
Vibrio cholerae and Escherichia coli were isolated from groundwater
samples. From borehole A3, 7.4x103 cfu/100 mL of E. coli was
found while the borehole A7 had 2.2x102 cfu/100 mL of E. coli.
Counts for Vibrio cholerae in both borholes A3 and A7 were 5.3x102
and 8.5x10 cfu/100 mL, respectively. Worth noting is that ground water samples
recorded the highest V. cholerae counts compared with surface water.
Naturally, groundwater is of excellent microbiological quality (Foster,
1995). However, the presence of high numbers of pathogenic bacteria, especially
V. cholerae in ground water suggests contamination of the water bed in
this area by faecal contaminants. Poor sanitation could be the cause of the
recent diarrhoea outbreaks which were suspected to be directly linked to poor
water quality in the Delmas area (Lang, 2007). Many rural
municipalities that rely on ground water as their main water sources have adapted
some measure of disinfection. These municipalities have small water treatment
units which are classified under small water treatment plants. It has been reported
that South African small water treatment plants face challenges such as poor
administration, lack of human resources and insufficient financial capacity
(Obi et al., 2007). Hence the Delmas cases could
have stemmed from improper handling of problems associated with the water quality
Upon disinfection of ground water with silver impregnated polymer, there was a 2 log reduction of E. coli from both borehole A3 and A7 where the resulting counts were 20 and 7 cfu/100 mL, respectively. For the removal of V. cholerae, silver impregnated polymers reduced V. cholerae counts by 2 logs where the silver polymer recorded 7 cfu/100 mL for A3 and 1 cfu/100 mL for A7. However, the silver free polymers had a significantly lower removal of both bacterial species (p≤0.05) from both ground water sources with the exception of E. coli from A3 where a 1 log reduction was observed after treatment. The poor performance of the silver free polymer was also observed during the treatment of groundwater.
Comparative analysis of disinfection techniques: Some of the techniques
used in the disinfection of water include chlorination, ozonation and UV radiation
(Freese and Noziac, 2004). Chlorine is a widely used
disinfectant because of its effectiveness, low capital and running costs and
it is relatively easy to handle (White, 1999). All the
above-mentioned techniques at correct dosages can give 2 to 3 log reduction
of bacterial counts in water (Freese et al., 2003).
This is similar to the reduction obtained by the silver impregnated carbon nanotubes
and cyclodextrin polymers. The removal of bacteria by the silver impregnated
polymers was observed to be significantly higher than the carbon nanotubes and
cyclodextrin polymers (p≤0.05). Hence silver impregnated polymers synthesized
in our laboratories performed competitively with conventional disinfection techniques.
However some optimization studies on the polymers still need to be carried out
to achieve maximum bacterial reductions.
Silver leaching out studies: Some concerns have been raised regarding
the toxicity of engineered nanomaterials towards the environment and humans.
In this study, it was necessary to assess the leachability of silver nanoparticles
into the water treated with the silver polymers. Treated water was found to
contain less than 0.1 mg L-1 of silver because silver levels in treated
water were below the instruments detection limit. According to the World
Health Organizations Water Quality Guidelines for drinking water, silver
levels of 0.1 mg L-1 could be tolerated without any health risk.
This concentration gives a total dose over 70 years of half the human NOAEL
(no- observed- adverse- effect- level) of 10 g (WHO, 2006).
Hence possible leaching of silver into the water when using silver impregnated
polymers dose not pose any significant health risk since levels are below 0.1
Effect of the nature of sample on the disinfection properties of silver: Comparing the performance of silver impregnated polymer with the other polymer at removing bacteria in environmental samples and model samples synthesized in the laboratory, there is a great difference at the removal of bacteria from both environmental and spiked water samples spiked with E. coli (ATCC 25925). From spiked water samples, a higher bacterial removal capacity was observed compared to the bacterial removal capacity in environmental samples. The silver polymer exhibited approximately a 3 logs removal of E. coli when using raw water from Temba. Also, approximately 4 logs and 6 logs of E. coli were removed from Samples A and B, respectively. In this study, the nature of samples was observed to affect the efficacy of the silver polymer. The possible factor that contributed to this could be the presence of other constituents in environmental water samples such as turbidity which is indicative of the concentration of dissolved suspended solids. Turbidity in general has negatively impacted the performance of silver polymer.
A low bacterial removal from raw unfiltered water samples implies that the polymer in not an ideal disinfectant for such waters. However high bacterial removal by the silver impregnated polymers from filtered water samples suggests that the polymers can be positioned after filtration is performed to get maximal removal of bacteria. Silver impregnated polymers promise to be good candidates for complimenting the currently employed disinfection methods
Support from the National Research Foundation (NRF), Minteks Nanotechnology Innovation Centre (NIC) and University of Johannesburg (UJ) and the Tshwane University of Technologys Water Care Department is gratefully acknowledged.