Article pubs.acs.org/est
Virus Removal in Ceramic Depth Filters Based on Diatomaceous Earth Benjamin Michen,*,†,‡ Fabian Meder,† Annette Rust,§ Johannes Fritsch,∥ Christos Aneziris,‡ and Thomas Graule†,‡ †
Laboratory for High Performance Ceramics, EMPA, Swiss Federal Laboratories for Materials Science and Technology, Ueberlandstrasse 129, 8600 Duebendorf, Switzerland ‡ Institute for Ceramics, Glass and Construction Materials, Technical University Bergakademie Freiberg, Agricolastrasse 17, 09596 Freiberg, Germany § Bachema AG, Ruetistrasse 22, 8952 Schlieren, Switzerland ∥ Technology and Management Faculty, University of Applied Sciences Ravensburg−Weingarten, Doggenriedstrasse, 88241 Weingarten, Germany S Supporting Information *
ABSTRACT: Ceramic filter candles, based on the natural material diatomaceous earth, are widely used to purify water at the point-of-use. Although such depth filters are known to improve drinking water quality by removing human pathogenic protozoa and bacteria, their removal regarding viruses has rarely been investigated. These filters have relatively large pore diameters compared to the physical dimension of viruses. However, viruses may be retained by adsorption mechanisms due to intermolecular and surface forces. Here, we use three types of bacteriophages to investigate their removal during filtration and batch experiments conducted at different pH values and ionic strengths. Theoretical models based on DLVO-theory are applied in order to verify experimental results and assess surface forces involved in the adsorptive process. This was done by calculation of interaction energies between the filter surface and the viruses. For two small spherically shaped viruses (MS2 and PhiX174), these filters showed no significant removal. In the case of phage PhiX174, where attractive interactions were expected, due to electrostatic attraction of oppositely charged surfaces, only little adsorption was reported in the presence of divalent ions. Thus, we postulate the existence of an additional repulsive force between PhiX174 and the filter surface. It is hypothesized that such an additional energy barrier originates from either the phage’s specific knobs that protrude from the viral capsid, enabling steric interactions, or hydration forces between the two hydrophilic interfaces of virus and filter. However, a larger-sized, tailed bacteriophage of the family Siphoviridae was removed by log 2 to 3, which is explained by postulating hydrophobic interactions.
1. INTRODUCTION More than 15% of the world’s population does not have access to safe drinking water. This might be due to poor water accessibility, a lack of sanitation, and inadequate water treatment.1 Proper water management may overcome these problems. One aspect of water management is the treatment of biologically and chemically contaminated water. The main drinking water risks in developing countries are associated with microbial pollution and thus, waterborne diseases lead to millions of deaths; according to the World Health Organization (WHO), diarrhea is the cause of 1.6 to 2.5 million deaths per year, with the affected being mainly infants under the age of five years.2 While various technologies are available with which a tremendous number of contaminants in water can be controlled, such methods often rely on multistage treatment, electric energy supply, and trained personnel, making water treatment too expensive for less developed countries.3 Moreover, in regions where no intact distribution system is available, © 2011 American Chemical Society
for instance, in many urban and rural areas in less developed countries or in catastrophic areas, only decentralised or pointof-use (POU) water treatment is thought to be efficient.4 To meet the requirements of POU technologies applied in developing countries, such interventions must be easy to operate and of low-cost while providing a sufficient quantity of safe drinking water. According to these criteria, Sobsey et al.5 identified ceramic and biosand household water filters as the most effective POU technologies, and demonstrated that ceramic filters were associated with the highest prevention of diarrheal diseases. Hunter6 confirmed ceramic filters as the most effective household water treatment in developing countries. Field studies in which ceramic water filters were Received: Revised: Accepted: Published: 1170
September 2, 2011 December 8, 2011 December 15, 2011 December 15, 2011 dx.doi.org/10.1021/es2030565 | Environ. Sci. Technol. 2012, 46, 1170−1177
Environmental Science & Technology
Article
smaller model would minimize the size exclusion effect which may contribute to virus removal in porous material. Viruses in water, as with other colloids, carry a specific surface charge which is pH-dependent. The certain pH at which the surface charge is zero is termed the isoelectric point (IEP). At pHvalues below the IEP, the colloid usually possesses a positive net charge and at pH-values above it are found to be negative in nature. The IEPs of viruses have been reviewed17 and are found to be in the pH range from 1.9 to 8.4; most frequently, they are measured in a band of 3.5 < IEP < 7. In order to represent the spectra of virus IEPs, it is logical to select at least two kinds of bacteriophages that have different IEPs, one with an acid IEP and another possessing a net zero charge in the neutral pH range. The dimensions of the two phages must be similar in order to exclude the possibility that the difference in phage retention is due to deviations in their size. In this way, the evaluation of the effect of the virus’ surface charge on the adsorption is thought to be feasible. Addressing the above evaluated criteria, two bacteriophages, namely Enterobacteria phage MS2 (MS2; IEP = 3.5, diameter = 25 nm) and Enterobacteria phage PhiX174 (Phi; IEP = 6.6, diameter = 26 nm) have been chosen to investigate virus adsorption in this work (for more detail, see Supporting Information). In addition to MS2 and Phi, an isolated wild-type bacteriophage was used which belongs to the family Siphoviridae. This phage, we will refer to it as the “Siphophage”, has an icosahedral head with a diameter of 60 nm and a noncontractile, flexible tail with a length of about 160 nm. The Siphophage has an IEP =2.7 and a hydrodynamic diameter (dHydro) of 94 nm, see Supporting Information. We used liquid propagation methods for all phages including purification of virus stocks by washing, centrifugation, and dialysis. The enumeration took place by counting plaques on lawns of bacteria in Petri dishes, details on propagation, purification, and enumeration are also reported in the Supporting Information. 2.3. Filtration Experiment. In order to quantify the filter retention performance, we use the log reduction value (LRV), see eq 1. The LRV gives a logarithmic expression of the fractional retention (R), eq 2.
distributed to groups in communities in Bolivia, Colombia, and Cambodia showed about a 50% reduction in diarrhea when compared to the control groups.7−9 This partial reduction in illness might be explained by alternative routes of infection and/or due to the passage of microorganisms through the ceramic filter. Recently, it was shown that colloidal particles with diameters
