Water Filtration Using Plant Xylem.pdf

Publish in

Documents

153 views

Please download to get full document.

View again

of 8
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Share
Description
Water Filtration Using Plant Xylem Michael S. H. Boutilier., Jongho Lee., Valerie Chambers, Varsha Venkatesh, Rohit Karnik* Department of Mechanical Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts, United States of America Abstract Effective point-of-use devices for providing safe drinking water are urgently needed to reduce the global burden of waterborne disease. Here we show that plant xylem from the sapwood of coniferous trees – a readily availa
Tags
Transcript
  Water Filtration Using Plant Xylem Michael S. H. Boutilier . , Jongho Lee . , Valerie Chambers, Varsha Venkatesh, Rohit Karnik  * Department of Mechanical Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts, United States of America Abstract Effective point-of-use devices for providing safe drinking water are urgently needed to reduce the global burden of waterborne disease. Here we show that plant xylem from the sapwood of coniferous trees – a readily available, inexpensive,biodegradable, and disposable material – can remove bacteria from water by simple pressure-driven filtration.Approximately 3 cm 3 of sapwood can filter water at the rate of several liters per day, sufficient to meet the cleandrinking water needs of one person. The results demonstrate the potential of plant xylem to address the need forpathogen-free drinking water in developing countries and resource-limited settings. Citation:  Boutilier MSH, Lee J, Chambers V, Venkatesh V, Karnik R (2014) Water Filtration Using Plant Xylem. PLoS ONE 9(2): e89934. doi:10.1371/ journal.pone.0089934 Editor:  Zhi Zhou, National University of Singapore, Singapore Received  October 17, 2013;  Accepted  January 23, 2014;  Published  February 26, 2014 Copyright:    2014 Boutilier et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permitsunrestricted use, distribution, and reproduction in any medium, provided the srcinal author and source are credited. Funding:  This work was supported by the James H. Ferry, Jr. Fund for Innovation in Research Education award to R.K. administered by the MassachusettsInstitute of Technology. SEM imaging was performed at the Harvard Center for Nanoscale Systems, a member of the National Nanotechnology InfrastructureNetwork (NNIN), which is supported by the National Science Foundation under NSF award no. ECS-0335765. The funders had no role in study design, datacollection and analysis, decision to publish, or preparation of the manuscript. Competing Interests:  The authors have declared that no competing interests exist.* E-mail: karnik@mit.edu .  These authors contributed equally to this work. Introduction The scarcity of clean and safe drinking water is one of the majorcauses of human mortality in the developing world. Potable ordrinking water is defined as having acceptable quality in terms of its physical, chemical, and bacteriological parameters so that it canbe safely used for drinking and cooking [1]. Among the waterpollutants, the most deadly ones are of biological srcin: infectiousdiseases caused by pathogenic bacteria, viruses, protozoa, orparasites are the most common and widespread health risk associated with drinking water [1,2]. The most common water-borne pathogens are bacteria (e.g.  Escherichia coli  ,  Salmonella typhi  , Vibrio cholerae   ), viruses (e.g. adenoviruses, enteroviruses, hepatitis,rotavirus), and protozoa (e.g. giardia) [1]. These pathogens causechild mortality and also contribute to malnutrition and stuntedgrowth of children. The World Health Organization reports [3]that 1.6 million people die every year from diarrheal diseasesattributable to lack of access to safe drinking water and basicsanitation. 90% of these are children under the age of 5, mostly indeveloping countries. Multiple barriers including prevention of contamination, sanitation, and disinfection are necessary toeffectively prevent the spread of waterborne diseases [1]. However,if only one barrier is possible, it has to be disinfection unlessevidence exists that chemical contaminants are more harmful thanthe risk from ingestion of microbial pathogens [1]. Furthermore,controlling water quality at the point-of-use is often most effectivedue to the issues of microbial regrowth, byproducts of disinfec-tants, pipeline corrosion, and contamination in the distributionsystem [2,4].Common technologies for water disinfection include chlorina-tion, filtration, UV-disinfection, pasteurization or boiling, andozone treatment [1,2,5]. Chlorine treatment is effective on a largescale, but becomes expensive for smaller towns and villages.Boiling is an effective method to disinfect water; however, theamount of fuel required to disinfect water by boiling is severaltimes more than what a typical family will use for cooking [1]. UV-disinfection is a promising point-of-use technology available [1], yet it does require access to electricity and some maintenance of the UV lamp, or sufficient sunlight. While small and inexpensivefiltration devices can potentially address the issue of point-of-usedisinfection, an ideal technology does not currently exist.Inexpensive household carbon-based filters are not effective atremoving pathogens and can be used only when the water isalready biologically safe [1]. Sand filters that can removepathogens require large area and knowledge of how to maintainthem [1], while membrane filters capable of removing pathogens[2,4] suffer from high costs, fouling, and often require pumping power due to low flow rates [6] that prevents their wideimplementation in developing countries. In this context, newapproaches that can improve upon current technologies areurgently needed. Specifically, membrane materials that areinexpensive, readily available, disposable, and effective at patho-gen removal could greatly impact our ability to provide safedrinking water to the global population.If we look to nature for inspiration, we find that a potentialsolution exists in the form of plant xylem – a porous material thatconducts fluid in plants [7]. Plants have evolved specialized xylemtissues to conduct sap from their roots to their shoots. Xylem hasevolved under the competing pressures of offering minimalresistance to the ascent of sap while maintaining small nanoscalepores to prevent cavitation. The size distribution of these pores – typically a few nanometers to a maximum of around 500 nm,depending on the plant species [8] – also happens to be ideal forfiltering out pathogens, which raises the interesting question of whether plant xylem can be used to make inexpensive waterfiltration devices. Although scientists have extensively studied plantxylem and the ascent of sap, use of plant xylem in the context of  PLOS ONE | www.plosone.org 1 February 2014 | Volume 9 | Issue 2 | e89934  water filtration remains to be explored. Measurements of sap flowin plants suggest that flow rates in the range of several liters perhour may be feasible with less than 10 cm-sized filters, using onlygravitational pressure to drive the flow [7].We therefore investigated whether plant xylem could be used tocreate water filtration devices. First, we reason which type of plantxylem tissue is most suitable for filtration. We then construct asimple water filter from plant xylem and study the resulting flowrates and filtration characteristics. Finally, we show that the xylemfilter can effectively remove bacteria from water and discussdirections for further development of these filters. Materials and Methods Materials Branches were excised from white pine growing on privateproperty in Massachusetts, USA, with permission of the owner andplaced in water. The pine was identified as  pinus strobus   based onthe 5-fold grouping of its needles, the average needle length of 4.5inches, and the cone shape. Deionized water (Millipore) was usedthroughout the experiments unless specified otherwise. Redpigment (pigment-based carmine drawing ink, Higgins Inks) wasdissolved in deionized water. Nile-red coated 20 nm fluorescentpolystyrene nanospheres were obtained from Life Technologies.Inactivated, Alexa 488 fluorescent dye labeled  Escherichia coli   wereobtained from Life Technologies. Wood sections were insertedinto the end of 3/8 inch internal diameter PVC tubing, sealedwith 5 Minute Epoxy, secured with hose clamps, and allowed tocure for ten minutes before conducting flow rate experiments. Construction of the Xylem Filter 1 inch-long sections were cut from a branch with approximately1 cm diameter. The bark and cambium were peeled off, and thepiece was mounted at the end of a tube and sealed with epoxy.The filters were flushed with 10 mL of deionized water beforeexperiments. Care was taken to avoid drying of the filter. Filtration and Flow Rate Experiments  Approximately 5 mL of deionized water or solution was placedin the tube. Pressure was supplied using a nitrogen tank with apressure regulator. For filtration experiments, 5 psi (34.5 kPa)pressure was used. The filtrate was collected in glass vials. For dyefiltration, size distribution of the pigments was measured for theinput solution and the filtrate. Higgins pigment-based carminedrawing ink, diluted , 1000 6 in deionized water, was used as theinput dye solution. For bacteria filtration, the feed solution wasprepared by mixing 0.08 mg of inactivated  Escherichia coli   in 20 mLof deionized water (  , 1.6 6 10 7 mL 2 1  ) after which the solution wassonicated for 1 min. The concentration of bacteria was measuredin the feed solution and filtrate by enumeration with ahemacytometer (inCyto C-chip) mounted on a Nikon TE2000-Uinverted epifluorescence microscope. Before measurement of concentration and filtration experiments, the feed solution wassonicated for 1 min and vigorously mixed. Imaging Xylem structure was visualized in a scanning electron micro-scope (SEM, Zeiss Supra55VP). Samples were coated with gold of 5 nm thickness before imaging. To visualize bacteria filtration,5 mL of solution at a bacteria concentration of   , 1.6 6 10 7 mL 2 1 was flowed into the filter. The filter was then cut longitudinallywith a sharp blade. One side of the sample was imaged using aNikon TE2000-U inverted epifluorescence microscope and theother was coated with gold and imaged with the SEM. Opticalimages were acquired of the cross section of a filter following filtration of 5 mL of the dye at a dilution of  , 100 6 . Particle Sizing Dynamic light scattering measurements of particle size distri-butions were performed using a Malvern Zetasizer Nano-ZS. Results Xylem Structure and Rationale for use of Conifer Xylem The flow of sap in plants is driven primarily by transpirationfrom the leaves to the atmosphere, which creates negative pressurein the xylem. Therefore, xylem evolution has occurred undercompeting pressures of providing minimal resistance to the flow of sap, while protecting against cavitation (i.e. nucleation) and growthof bubbles that could stop the flow of sap and kill the plant, and todo this while maintaining mechanical strength [7]. The xylemstructure comprises many small conduits that work in parallel andoperate in a manner that is robust to cavitation [7,8] (Figure 1). Inwoody plants, the xylem tissue is called the sapwood, which oftensurrounds the heartwood (i.e. inactive, non-conducting lignifiedtissue found in older branches and trunks) and is in turnsurrounded by the bark (Figure 1b,c). The xylem conduits ingymnosperms (conifers) are formed from single dead cells and arecalled tracheids (Figure 1c), with the largest tracheids reaching diameters up to 80  m m and lengths up to 10 mm [7]. Angiosperms(flowering plants) have xylem conduits called vessels that arederived from several cells arranged in a single file, having diameters up to 0.5 mm and lengths ranging from a fewmillimeters to several meters [7]. These parallel conduits haveclosed ends and are connected to adjacent conduits via ‘‘pits’’ [8](Figure 1d,e). The pits have membranes with nanoscale pores thatperform the critical function of preventing bubbles from crossing over from one conduit to another. Pits occur in a variety of configurations; Figure 1d,e shows torus-margo pit membranes thatconsist of a highly porous part shaped like a donut (margo) and animpermeable part in the center called torus, occurring in conifers[8]. The porosity of the pit membranes ranges in size from a fewnanometers to a few hundred nanometers, with pore sizes in thecase of angiosperms tending to be smaller than those ingymnosperms [8,9]. Pit membrane pore sizes have been estimatedby examining whether gold colloids or particles of different sizescan flow through [8,10]. Remarkably, it was observed that 20 nmgold colloids could not pass through inter-vessel pit membranes of some deciduous tree species [10], indicating an adequate sizerejection to remove viruses from water. Furthermore, inter-tracheid pit membranes were found to exclude particles in the200 nm range [8], as required for removal of bacteria andprotozoa.Since angiosperms (flowering plants, including hardwood trees)have larger xylem vessels that are more effective at conducting sap,xylem tissue constitutes a smaller fraction of the cross-section areaof their trunks or branches, which is not ideal in the context of filtration. The long length of their xylem vessels also implies that alarge thickness (centimeters to meters) of xylem tissue will berequired to achieve any filtration effect at all – filters that arethinner than the average vessel length will just allow water to flowthrough the vessels without filtering it through pit membranes(Figure 1a). In contrast, gymnosperms (conifers, including softwood trees) have short tracheids that would force water toflow through pit membranes even for small thicknesses (  , 1 cm) of xylem tissue (Figure 1a). Since tracheids have smaller diametersand are shorter, they offer higher resistance to flow, but typically agreater fraction of the stem cross-section area is devoted to Water Filtration Using Plant XylemPLOS ONE | www.plosone.org 2 February 2014 | Volume 9 | Issue 2 | e89934  conducting xylem tissue. For example, in the pine branch shown inFigure 1b used in this study, fluid-conducting xylem constitutes themajority of the cross-section. This reasoning leads us to theconclusion that in general the xylem tissue of coniferous trees – i.e.the sapwood – is likely to be the most suitable xylem tissue forconstruction of a water filtration device, at least for filtration of bacteria, protozoa, and other pathogens on the micron or largerscale.The resistance to fluid flow is an important consideration forfiltration. Pits can contribute a significant fraction (as much as 30– 80%) [7,8] of the resistance to sap flow, but this is remarkablysmall considering that pit membrane pore sizes are several ordersof magnitude smaller than the tracheid or vessel diameter. The pitsand pit membranes form a hierarchical structure where the thin,highly-permeable pit membranes are supported across themicroscale pits that are arranged around the circumference of the tracheids (Figure 1a). This arrangement permits the pitmembranes to be thin, offering low resistance to fluid flow.Furthermore, the parallel arrangement of tracheids with pitsaround their circumference provides a high packing density for thepit membranes. For a given tracheid with diameter  D   and length L  , where pit membranes occupy a fraction  a  of the tracheid wallarea, each tracheid effectively contributes a pit membrane area of  p  DL  a /2, where the factor of 2 arises as each membrane is sharedby two tracheids. However, the nominal area of the tracheid isonly  p  D  2 /4, and therefore, the structure effectively amplifies thenominal filter area by a factor of 2 a (  L  /  D   ) (Figure 1f). The imagesin Figure 1c indicate that typical  D  , 10–15  m m and  a , 0.2 yieldan effective area amplification of   , 20 for tracheid lengths of 1– 2 mm. Therefore, for a filter made by cutting a slice of thickness , L   of the xylem, the actual membrane area is greater by a largefactor due to vertical packing of the pit membranes. Larger filterthicknesses further increase the total membrane area, but theadditional area of the membrane is positioned in series rather thanin parallel and therefore reduces the flow rate, but potentiallyimproves the rejection performance of the filter due to multiplefiltration steps as shown in Figure 1a. Construction of the Xylem Filter and Measurement of Flow Rate The xylem filter device was constructed by simply peeling off the bark and cambium from a section of the pine branch andinserting it into a tube (Figure 2a). Although a simple tube fastenercould provide a leak-tight seal between the tube and the xylem, weused epoxy to ensure that there was no inadvertent leakage. Whendeionized water was loaded into the tube above the xylem andsubjected to pressure in the 0.5–5 psi (3.45 to 34.5 kPa) range, we Figure 1. Xylem structure.  a) Structure of xylem vessels in flowering plants and tracheids in conifers. Longer length of the vessels can providepathways that can bypass filtration through pit membranes that decorate their circumference. b) Photograph of  , 1 cm diameter pine (  pinus strobus )branch used in the present study. c) Scanning electron microscope (SEM) image of cut section showing tracheid cross section and lengthwise profile.Scale bar is 40  m m. d) SEM image showing pits and pit membranes. Scale bar is 20  m m. e) Pit membrane with inset showing a cartoon of the pit cross-section. The pit cover has been sliced away to reveal the permeable margo surrounding the impermeable torus. Arrow indicates observed hole-likestructures that may be defects. The margo comprises radial spoke-like structures that suspend the torus, which are only barely visible overlaying thecell wall in the background. Scale bar is 1  m m. f) Dependence of area amplification, defined as the pit membrane area divided by the nominal filterarea, on the tracheid aspect ratio  L / D  and fractional area  a  occupied by pit membranes.doi:10.1371/journal.pone.0089934.g001Water Filtration Using Plant XylemPLOS ONE | www.plosone.org 3 February 2014 | Volume 9 | Issue 2 | e89934  found that water readily flowed through the xylem. The flow ratewas proportional to applied pressure (Figure 2b), which allowedfor the extraction of the hydrodynamic conductivity  K   (m 2 Pa 2 1 s 2 1  ) of the filter, defined by Q ~ KA D P l   ð 1 Þ where  Q    is the volumetric flow rate (in m 3 s 2 1  ) under pressuredifference  D P   across the filter, while  l   and  A  are the thickness andthe cross-section area of the filter, respectively. The observedconductivities for three different filters were in the range of   , 5– 6 6 10 2 10 m 2 Pa 2 1 s 2 1 (Figure 2c), or equivalently,  , 0.5–0.6 kg s 2 1 m 2 1 MPa 2 1 when defined with respect to mass flow rate of water.Biologists have performed similar flow rate measurements bycutting a section of a plant stem under water, flushing to removeany bubbles, and applying a pressure difference to measure theflow rate [11,12]. Xylem conductivities of conifers [7] typicallyrange from 1–4 kg s 2 1 m 2 1 MPa 2 1 , which compares very wellwith the conductivities measured in our experiments. Lowerconductivities can easily result from introduction of bubbles [11]or the presence of some non-conducting heartwood. We cantherefore conclude that the flow rate measurements in our devicesare consistent with those expected from prior literature onconductivity of conifer xylem. Filtration of Pigment Dye  After construction of the filter, we tested its ability to filter apigment dye with a broad particle size distribution. The red colorof the feed solution disappeared upon filtration (Figure 3a)indicating that the xylem filter could effectively filter out the dye.Since the dye had a broad pigment size distribution, weinvestigated the size-dependence of filtration by quantifying thepigment size distribution before and after filtration using dynamiclight scattering. We found that the feed solution comprisedparticles ranging in size from  , 70 nm to  , 500 nm, with somelarger aggregates (Figure 3b). In contrast, the filtrate particle sizedistribution peaked at  , 80 nm, indicating that larger particleswere filtered out. In a separate experiment, we observed that20 nm fluorescent polystyrene nanoparticles could not be filteredby the xylem filter, confirming this size dependence of filtration. Figure 2. Xylem filter.  a) Construction of xylem filter. b) Effect of applied pressure on the water flux through the xylem filter. c) Hydrodynamicconductivity of the filter extracted at each measured pressure using the total filter cross-section area and thickness as defined by Equation 1. Errorbars indicate 6 S.D. for measurements on three different xylem filters.doi:10.1371/journal.pone.0089934.g002Water Filtration Using Plant XylemPLOS ONE | www.plosone.org 4 February 2014 | Volume 9 | Issue 2 | e89934
Related Search

Previous Document

Leaflet Diare

We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x