xml file error

Page 1 of 1

2 Replies - 1578 Views - Last Post: 12 February 2008 - 09:56 AM Rate Topic: -----

#1 realwish   User is offline

  • New D.I.C Head
  • member icon

Reputation: 3
  • View blog
  • Posts: 47
  • Joined: 29-January 08


Post icon  Posted 11 February 2008 - 11:32 PM

pls help me to find the error in the code


<?xml version="1.0"?>
<!DOCTYPE book SYSTEM "book.dtd"[
<!ENTITY gr1 SYSTEM "ch16gr1" NDATA GIF>
<!ENTITY gr2 SYSTEM "ch16gr2" NDATA GIF>
<!ENTITY eq1 SYSTEM "ch16eq1" NDATA GIF>
<!ENTITY eq2 SYSTEM "ch16eq2" NDATA GIF>
<!ENTITY eq3 SYSTEM "ch16eq3" NDATA GIF>
<!ENTITY eq4 SYSTEM "ch16eq4" NDATA GIF>
<!ENTITY eq5 SYSTEM "ch16eq5" NDATA GIF>
<!ENTITY eq6 SYSTEM "ch16eq6" NDATA GIF>
<!ENTITY eq7 SYSTEM "ch16eq7" NDATA GIF>
<!ENTITY eq8 SYSTEM "ch16eq8" NDATA GIF>
<!ENTITY eq9 SYSTEM "ch16eq9" NDATA GIF>
<!ENTITY eq10 SYSTEM "ch16eq10" NDATA GIF>
<!ENTITY eq11 SYSTEM "ch16eq11" NDATA GIF>
<!ENTITY eq12 SYSTEM "ch16eq12" NDATA GIF>
<!ENTITY eq13 SYSTEM "ch16eq13" NDATA GIF>
<!ENTITY eq14 SYSTEM "ch16eq14" NDATA GIF>
<!ENTITY eq15 SYSTEM "ch16eq15" NDATA GIF>
<!ENTITY eq16 SYSTEM "ch16eq16" NDATA GIF>
<!ENTITY eq17 SYSTEM "ch16eq17" NDATA GIF>
<!ENTITY eq18 SYSTEM "ch16eq18" NDATA GIF>
<!ENTITY eq19 SYSTEM "ch16eq19" NDATA GIF>
<chapter id="ch16">
<parttitle>II M<scp>anipulating</scp> C<scp>ontaminant</scp> A<scp>vailability and</scp> D<scp>eveloping</scp> R<scp>esearch</scp> T<scp>ools</scp></parttitle>
<title>Phytoremediation With Living Aquatic Plants</title>
<stl>Development and Modeling of Experimental Observations</stl>
<author><fnm>Steven P. K.</fnm> <snm>Sternberg</snm></author>
<para>This chapter provides a summary of the mathematical analysis and experimental design of laboratory measurements of the bioremoval potential for living aquatic plants. This process is called phytoremediation, bioremoval, biosorption, or bioaccumulation. The mathematical models are based on the concept of the conservation of mass and include descriptive equations, including adsorption of the metal onto living and growing biomass. The models describe the concentration of metal in solution as a function of time. An example case from previously published data is included to demonstrate the use of the models. The results from the mathematical models can be used to scale up a process, or to answer questions of how long to run an experiment, how much biomass material is required, what the expected level of removal is, and to help set benchmarks to determine how well a process is working. In addition to presenting model equations, a summary of experimental considerations, such as statistical design, choice of variables, and result quantification has been included. The information provided allows good experimental data to be collected such that a maximum amount of information is obtained with the minimum amount of effort.</para>
<title>Key Words:</title>
<keyword>experimental model</keyword>
<keyword>statistical design</keyword>
<keyword>mass balance</keyword>
<sec1 id="ch16s1"><no>1.</no> <title>Introduction</title>
<para>Bioremoval is a process that can remove soluble heavy metal contaminants from aqueous solutions by an adsorption process (<citeref rid="ch16bib1">1</citeref>&ndash;<citeref rid="ch16bib11">11</citeref>). This process is also called phytoremediation, bioremoval, biosorption, bioaccumulation, or phytoremoval. The adsorbant is biomass, often obtained from an aquatic plant, which may be living or dead (<citeref rid="ch16bib12">12</citeref>&ndash;<citeref rid="ch16bib14">14</citeref>). The biomass is capable of accumulating the metal at concentrations many times greater than that of the solution (<citeref rid="ch16bib15">15</citeref>&ndash;<citeref rid="ch16bib21">21</citeref>). The metal is removed from the system by a simple filtration of the biomass. This process has the advantages of being simple, with a low cost to build and to maintain (<citeref rid="ch16bib22">22</citeref>,<citeref rid="ch16bib23">23</citeref>). It may not require changes in water quality, such as pH, temperature, or nutrient loadings (<citeref rid="ch16bib24">24</citeref>,<citeref rid="ch16bib25">25</citeref>). For these reasons, this technology is currently being studied for potential use in waste water pretreatment from industrial facilities (<citeref rid="ch16bib26">26</citeref>&ndash;<citeref rid="ch16bib28">28</citeref>). Some current shortcomings include the need for long retention times (2&ndash;48 h), and the existence of an equilibrium concentration beyond which no removal occurs, even when additional biomass is added. The need for additional research focuses on identification of biomass sources for particular metals, examinations of bioremoval potentials for multiple metal contaminants, and the use of multiple sources of biomass.</para>
<para>This chapter provides a summary of the mathematical analysis and experimental design for laboratory measurements of the bioremoval potential of living aquatic plants. The intended audience is an interested researcher starting a heavy metal bioremoval project. I combined ideas from many engineering and science fields to develop a procedure for the design of experiments and the presentation of collected information. This work by no means provides the final word on how to conduct such experiments or their interpretation; rather it seeks to provide a common starting point.</para>
<para>The main issues this chapter discusses are the development of a general system for modeling a bioremoval reactor, the effective design of experiments using statistics, and the identification of typical dependent and independent variables. Also, there are several decisions to make in determining how to design a system for making measurements of the bioremoval potential: use of living or dead plants, type of reactor (batch or flow), and water pretreatment. This discussion will center on using living aquatic plants in a batch reactor to remove a metal ion from water.</para>
<sec2 id="ch16s2"><no>1.1.</no> <title>Modeling</title>
<para>An important aspect of present-day engineering is the ability to adequately model the behavior of a system before committing to large-scale investment. Developing a good model requires a basic theoretical understanding of the system, and experimental observation and measurement of the system or parts of the system. (<i>See</i> List of Symbols on p. 203.)</para>
<para>This chapter attempts to develop a general model based on standard engineering practice. The model describes the concentration of a toxic metal in an aqueous solution that is being treated with plant biomass for the purpose of removing the metal. In this chapter, I will model the bioremoval of the metal ions as an adsorption phenomenon. The main modeling tool, however, will be the species mass balance:
<deqn id="ch16eq1"><no>(1)</no><math altimg="ch16eq1"></math></deqn></para>
<para>which can be written mathematically as
<deqn id="ch16eq2"><no>(2)</no><math altimg="ch16eq2"></math></deqn></para>
<para>where the ads subscript refers to the metal being adsorbed. The integral in the rate term allows generation to vary in space throughout the reactor. The last term uses a derivative to represent the change in mass in the solution over time. Most batch reactors are well mixed and so the rate term can be considered constant throughout the reactor. This equation will be used to model the change in metal in the water over time, the amount of biomass, and the amount of metal adsorbed to the biomass.</para>
<sec3 id="ch16s3"><no>1.1.1.</no> <title>Batch Reactor Models</title>
<para>The system of the metal solution and biomass can be modeled as a chemical reactor. Time is the key design variable of interest for a laboratory reactor (consisting of a beaker filled with the waste water into which the biomass is introduced). This type of reactor is called a batch reactor.</para>
<para>When time is an important consideration, we need to include a study of the reaction/adsorption kinetics. Kinetics is the study of how fast a reaction (or adsorption) occurs under given conditions (temperature, competing reactions, and catalysts). The equilibrium conditions can be obtained from these models by assuming a large value for the time variable (where the rate of adsorption is zero).</para>
<para>We start by defining the adsorption rate, an intensive variable that depends on temperature and concentration, which describes the mass of metal adsorbed per unit time. An example equation,
<deqn id="ch16eq3"><no>(3)</no><math altimg="ch16eq3"></math></deqn></para>
<para>describes the adsorption rate as a linear function of a rate constant (which may depend on temperature) and the concentration of the adsorbed species. We will explore more interesting forms of this function in the section on adsorption (<i>see</i> <secref rid="ch16s4">Subheading 1.1.2</secref>).</para>
<para>A batch reactor typically consists of a holding vessel into which all components of the reaction or adsorption process are added. It then is allowed to react for a certain period of time, after which it is drained, cleaned, and the process begun again. It works well for slow reactions, low volumes, or new/novel systems. It is time consuming and requires thorough cleaning after each run. It is the typical reactor chosen for laboratory studies. The main consideration for design purposes is the time needed for the reaction.</para>
<para>The mass balance (<eqnref rid="ch16eq2"><b>Eq. 2</b></eqnref>) description for a batch reactor includes no mass in or mass out terms (there is only a single instantaneous input at time &equals; 0, which is modeled mathematically by using an initial condition). The generation/ removal term describes the rate of metal ion adsorption onto the biomass and it is assumed to be the same everywhere. The accumulation term describes the metal ion concentration in solution. Excluding zero terms and assuming that the reactor has a constant volume (which allows simple conversion from mass to concentration, C &equals; M/V):
<deqn id="ch16eq4"><no>(4)</no><math altimg="ch16eq4"></math></deqn></para>
<para>where <i>r<sub>ads</sub></i> has units of mass of metal per volume of solution per unit time. Realizing that the rate of adsorption is dependent on the quantity of biomass, we can modify this equation by relating the rate of adsorption to the rate of adsorption per unit biomass:
<deqn id="ch16eq5"><no>(5)</no><math altimg="ch16eq5"></math></deqn></para>
<para>where <i>m<sub>x</sub></i> is the amount of biomass per unit volume of solution, and <i>r<sub>bio</sub></i> has units of mass of metal adsorbed per unit biomass per unit time. These equations, combined with an equation for the adsorption rate, provide the model equation for the batch reactor experiments.</para>
<para>We are not considering the other types of reactors like the continuous stirred tank reactor or plug flow reactors. These are flow-based reactors as opposed to the time-based batch reactor, <i>see</i> <b>ref.</b> <citeref rid="ch16bib29">29</citeref> for more details on any of these reactor types.</para>
<sec3 id="ch16s4"><no>1.1.2.</no> <title>Adsorption Models</title>
<para>The next consideration in building our system model equations is to describe various adsorption models. These identify the adsorption rate term (<i>r<sub>bio</sub></i>) used in the reactor model equations. This chapter assumes all the removed metal is adsorbed by the plant biomass. The term could be expanded to include adsorption by sediments, additional organisms (each with its own behavior), or the reactor vessel itself.</para>
<para>The models should all describe the potential of the biomass to adsorb a given metal. We would expect this to be fastest when the biomass is first exposed to the metal, then to slow, and to eventually approach zero as equilibrium is reached. How fast the transition occurs can be modeled using adjustable parameters. The adjustable parameters are obtained experimentally. I will develop two example models, each having three adjustable parameters. One of these three parameters is used to describe the equilibrium concentration in the solution, found after long exposure times of the biomass to the metal solution. Note that we expect the rate to depend on the difference between the actual and equilibrium concentration, this may be described as a driving potential for the adsorption. This concentration potential is described with the term (<i>C<sub>ads</sub></i> &minus; <i>C<sub>eq</sub></i>), where <i>C<sub>ads</sub></i> represents the adsorbed metal concentration and <i>C<sub>eq</sub></i> represents the equilibrium concentration for a particular metal and type of biomass.</para>
<para>Next, two adsorption rate models are proposed. Each has the characteristics of starting large in value and becoming zero as the concentration in the solution approaches the equilibrium value. This idea is based on the concept of mass transfer, and it is from this area of study that these models are found (<citeref rid="ch16bib30">30</citeref>).</para>
<para>Model one has two adjustable parameters
<deqn id="ch16eq6"><no>(6)</no><math altimg="ch16eq6"></math></deqn></para>
<para>k1 may be a temperature dependent term, if necessary, and n is the rate power. When n has the value of one, this is called a first order model. Model two also has two adjustable parameters
<deqn id="ch16eq7"><no>(7)</no><math altimg="ch16eq7"></math></deqn></para>
<para>Both of the parameters, <i>k<sub>2A</sub></i> and <i>k<sub>2B</sub></i>, may be temperature dependent. When the concentration term is large the rate is a constant, and when it is small the rate term is also small. <figref rid="ch16fig1"><b>Figure 1</b></figref> shows the general shapes of these functions. For multiple species of metal adsorbing on common sites this equation can be generalized for metal species, <i>i</i>:
<deqn id="ch16eq8"><no>(8)</no><math altimg="ch16eq8"></math></deqn></para>
<fig id="ch16fig1" image="ch16gr1"><no><b>Fig. 1.</b></no><caption><para>Rate of adsorption models.</para></caption></fig>
<para>where <i>C<sub>i</sub></i> is concentration of the species of interest, <i>C<sub>a</sub></i> is the concentration of species A, and <i>C<sub>b</sub></i> is the concentration of species B. The temperature dependence of the constants is usually described with an equation of the following type
<deqn id="ch16eq9"><no>(9)</no><math altimg="ch16eq9"></math></deqn></para>
<para>Where <i>A</i> is some pre-exponential frequency factor, <i>E</i> is an activation energy, <i>R</i> is the ideal gas constant, and <i>T</i> is the absolute temperature.</para>
<para>These adjustable models can describe many simple types of relation between the adsorption rate and the concentration difference. The shapes all monotonically decrease as the concentration difference becomes smaller, and include linear, concave-up, and concave-down forms. A comparison of the different models, for a dataset as described in the example data section, is shown in <figref rid="ch16fig1"><b>Fig. 1</b></figref>.</para>
<sec3 id="ch16s5"><no>1.1.3.</no> <title>Biomass Growth Rate</title>
<para>Using living biomass, such as <i>Lemna</i> (<citeref rid="ch16bib31">31</citeref>,<citeref rid="ch16bib32">32</citeref>) or a macro alga (<citeref rid="ch16bib33">33</citeref>) can provide an advantage during the bioremoval process&mdash;it can increase in mass, thereby increasing the total amount of metal that could be adsorbed. However, this additional biomass does not change the equilibrium amount of metal in the solution (<citeref rid="ch16bib34">34</citeref>). The growth of the biomass may be modeled using the mass balance (<eqnref rid="ch16eq2">Eq. 2</eqnref>). Again, for a batch reactor there will be no in or out terms, other than the initial amount present at the time zero.
<deqn id="ch16eq10"><no>(10)</no><math altimg="ch16eq10"></math></deqn></para>
<para>where <i>m<sub>x</sub></i> represents the amount of biomass per unit volume and <i>r<sub>x</sub></i> is the biomass growth rate with units 1/time. The simplest model describes the growth rate as a constant, <i>k<sub>bio</sub></i>. This is acceptable for short time periods and single species batch reactors. More interesting and realistic models have been described elsewhere (<citeref rid="ch16bib35">35</citeref>). For the initial condition <i>m<sub>x</sub></i> &equals; <i>M</i><sub>0</sub> at time zero, this equation becomes
<deqn id="ch16eq11"><no>(11)</no><math altimg="ch16eq11"></math></deqn></para>
<sec3 id="ch16s6"><no>1.1.4.</no> <title>System Model</title>
<para>Now we can combine our reactor models with the adsorption rate equations and biomass growth equation to derive a relationship between the reactor design variable (time for batch, volume for flow) and the concentration of metal in solution.</para>
<para>Combining the mass balance (<eqnref rid="ch16eq5"><b>Eq. 5</b></eqnref>), the adsorption rate laws (<eqnref rid="ch16eq6"><b>Eqs. 6</b></eqnref> and <eqnref rid="ch16eq7"><b>7</b></eqnref>), and the biomass rate model (<eqnref rid="ch16eq10"><b>Eq. 10</b></eqnref>) yields the following solutions:</para>
<para>Model 1</para>
<para>Model 1&ndash;1 case one: with <i>n</i> &equals; 1
<deqn id="ch16eq12"><no>(12)</no><math altimg="ch16eq12"></math></deqn></para>
<para>Model 1&ndash;<i>n</i> case two: with <i>n</i> &gt; 1
<deqn id="ch16eq13"><no>(13)</no><math altimg="ch16eq13"></math></deqn></para>
<para>where <i>C<sub>I</sub></i> is the initial concentration. The parameter <i>k<sub>bio</sub></i> can be found by measuring the amount of biomass. It may not be possible to measure the mass during the experiment, so only the initial and final values will be known. <eqnref rid="ch16eq11"><b>Equation 11</b></eqnref> only requires two mass measurements to be known to determine a value of <i>k<sub>bio</sub></i>. Case one only requires one parameter to be fit with experimental data, <i>k<sub>1</sub></i>. The values of <i>k<sub>1</sub></i> and <i>n</i> can be found using nonlinear best-fit statistics between the model and the measured values of concentration over time. The best model parameters will minimize the sum of squares difference (SSD).
<deqn id="ch16eq14"><no>(14)</no><math altimg="ch16eq1"></math></deqn></para>
<para>Model 2:
<deqn id="ch16eq15"><no>(15)</no><math altimg="ch16eq15"></math></deqn></para>
<para>Again, the parameters M<sub>0</sub> and C<sub>eq</sub> can be obtained directly from the experimental measurements, <i>k<sub>bio</sub></i> from biomass measurements and <eqnref rid="ch16eq11"><b>Eq. 11</b></eqnref>, and <i>k<sub>2A</sub></i> and <i>k<sub>2B</sub></i> can be found using nonlinear best-fit statistics by minimizing the SSD. Comparison between these three models for a set of batch reactor data is included at the example data section.</para>
<para>When using dead biomass or biomass that does not grow (<i>k<sub>bio</sub></i> &equals; 0), the previously listed equations may be modified by making the following substitution:
<deqn id="ch16eq16"><no>(16)</no><math altimg="ch16eq16"></math></deqn></para>
<sec3 id="ch16s7"><no>1.1.5.</no> <title>Metal Adsorbed on Biomass</title>
<para>The next part of the system to model is the interaction between the contaminant metal and the biomass. Specifically we need to know how much metal is adsorbed onto a unit mass of the biomass. This calculation is relatively straight-forward. From the mass balance we know (assume) that any metal ion mass that leaves the water must be adsorbed onto the biomass:
<deqn id="ch16eq17"><no>(17)</no><math altimg="ch16eq17"></math></deqn></para>
<para>where <i>C<sub>I</sub></i> is the initial concentration and Volume represents the total liquid volume of the batch reactor. We can calculate the concentration of metal in the biomass by dividing this value by the amount of biomass:
<deqn id="ch16eq18"><no>(18)</no><math altimg="ch16eq18"></math></deqn></para>
<para>Finally, this allows the calculation of the concentration factor
<deqn id="ch16eq19"><no>(19)</no><math altimg="ch16eq19"></math></deqn></para>
<para>where &rho; is the solution density and CF has units of mass of solution per unit mass of biomass. It can be described as &ldquo;1 g of biomass contains as much metal as CF grams of solution.&rdquo; The best biomass treatment systems will maximize this number.</para>
<sec2 id="ch16s8"><no>1.2.</no> <title>Example Data</title>
<para>The previously mentioned equations are used to model a set of experimental data (<citeref rid="ch16bib28">28</citeref>) in which the macroalga, <i>Cladophora parriaudii</i>, was used to remove cadmium from synthetic waste water in a batch reactor. The information was obtained by measuring concentration and biomass over time in a batch reactor in five simultaneous trials with identical initial conditions. The model parameters that are determined from these initial conditions used the average value from all five trials. The two parameters that are statistically fit were determined by minimizing the SSD over all five trials, with the reported SSD being the average of the five sets. <tableref rid="ch16tbl1">Table 1</tableref> summarizes the information used and the final values of the parameters for the two models, including both cases for model 1. The experimental data and the three model equations are shown in <figref rid="ch16fig2">Fig. 2</figref>. All three models appear to fit the data well, though a comparison of the SSD values for each shows that, of the three models, model 1&ndash;<i>n</i> provides the best fit.</para>
<fig id="ch16fig2" image="ch16gr2"><no><b>Fig. 2.</b></no><caption><para>Model comparisons.</para></caption></fig>
<table frame="none" id="ch16tbl1"><title>Table 1 Experimental Data and Model Parameters</title>
<tgroup cols="4">
<colspec colnum="1" colname="col1" align="left" colwidth="*"/>
<colspec colnum="2" colname="col2" align="center" colwidth="*"/>
<colspec colnum="3" colname="col3" align="center" colwidth="*"/>
<colspec colnum="4" colname="col4" align="center" colwidth="*"/>
<thead valign="bottom">
<entry align="left"><para>Model</para></entry>
<entry align="center"><para>1&ndash;1</para></entry>
<entry align="center"><para>1&ndash;<i>n</i></para></entry>
<entry align="center"><para>2</para></entry>
<tbody valign="top">
<entry><para>Parameter 1</para></entry>
<entry><para>M &equals; 1</para></entry>
<entry><para>m &equals; 1.75</para></entry>
<entry><para>K<sub>2</sub>A &equals; 0.1</para></entry>
<entry><para>Parameter 2</para></entry>
<entry><para>k<sub>1</sub> &equals; 0.078</para></entry>
<entry><para>k<sub>1</sub> &equals; 0.05</para></entry>
<entry><para>k<sub>2</sub>B &equals; 0.05</para></entry>
<entry><para>Initial biomass</para></entry>
<entry><para>1.03 g</para></entry>
<entry><para>1.03 g</para></entry>
<entry><para>1.03 g</para></entry>
<entry><para>Sum of squares</para></entry>
<entry><para>Bioremoval percent</para></entry>
<sec1 id="ch16s9"><no>2.</no> <title>Materials</title>
<para>The three primary materials used in these experiments are the water source, the form of the metal contaminant, and the type of biomass (<i>see</i> <secref rid="ch16s18">Notes 1</secref>&ndash;<secref rid="ch16s18">3</secref>). Any experiment must choose these carefully, with full appreciation of how the final results will be used.</para>
<sec2 id="ch16s10"><no>2.1.</no> <title>Experimental Trials</title>
<para><lst type="num">
<item><para>1 L Plastic beakers. Polyethylene or polypropylene has been found to not interact with the water, metals, or plants used in these experiments.</para></item>
<item><para>Plastic stir bar. Polyethylene or polypropylene.</para></item>
<item><para>Time recorder.</para></item>
<item><para>Micropipets with disposable tips. Volume range should include 100 &mu;L to 1 mL. These are used to add metal solution, to collect samples for metal analysis, and to dilute samples to the operating range of the metal analysis equipment. These will also be used for the metal analyses.</para></item>
<sec2 id="ch16s11"><no>2.2.</no> <title>Sample Metal Analysis</title>
<para>For atomic adsorption spectrophotometer, either flame or graphite furnace mode. Refer to the manufacturer for operation procedure. Details of methods, preparation of standards, and discussion of matrix modifiers will, in general, be different for each metal (<citeref rid="ch16bib36">36</citeref>) (<i>see</i> <secref rid="ch16s18">Note 4</secref>).</para>
<sec2 id="ch16s12"><no>2.3.</no> <title>Biomass Metal Analysis</title>
<para><lst type="num">
<item><para>Filtration equipment. Use weighed, dried ashless filter papers to collect biomass.</para></item>
<item><para>Porcelain crucibles with lids, approximate volume of 50 mL.</para></item>
<item><para>Hot plate with chemically resistant top.</para></item>
<item><para>Drying oven, temperatures of 105&deg;C for 48 h.</para></item>
<item><para>Muffle furnace, capable of temperatures to 500&deg;C for 2 h.</para></item>
<item><para>Concentrated nitric acid, metal-analysis grade.</para></item>
<item><para>Concentrated hydrochloric acid, metal-analysis grade.</para></item>
<item><para>Small glass (Pyrex) beaker to mix 25 mL Aqua Regia solution.</para></item>
<sec1 id="ch16s13"><no>3.</no> <title>Methods</title>
<para>This section discusses the procedure used to determine the bioremoval potential and concentration factor, and good experimental technique for developing the information needed to understand the bioremoval processes. The first section provides a list of steps for performing the experiments. The second section is a set of general guidelines for the statistical design of experiments. The third section provides some ideas and suggestions for developing good experimental data. An understanding of the system model equations helps determine which variables to measure.</para>
<sec2 id="ch16s14"><no>3.1.</no> <title>Experimental Trials</title>
<para>The experimental trials can be commenced once the experimental design has been finalized, and the type of water, form of metal, and type of biomass have been chosen. The experiments are relatively straightforward:</para>
<para>For each trial:
<lst type="num">
<item><para>Fill a 1L container with 700 mL source water (<i>see</i> <secref rid="ch16s18">Note 1</secref>).</para></item>
<item><para>Add metal source to container (<i>see</i> <secref rid="ch16s18">Note 2</secref>), and allow to equilibrate, usually 2 h. Sample this solution to obtain the time zero data.</para></item>
<item><para>Add the appropriate amount of weighed wet biomass (<i>see</i> <secref rid="ch16s18">Note 3</secref>).</para></item>
<item><para>Sample the solution at the required time intervals (<i>see</i> <secref rid="ch16s18">Note 5</secref>). Liquid samples may be stored for short times (days) by slight acidification with dilution water (2 wt percent solution of nitric acid).</para></item>
<item><para>Measure the concentration of metal in water as per standard methods (<citeref rid="ch16bib36">36</citeref>). Atomic adsorption with flame or graphite furnace will allow very low concentrations (milligrams per liter to micrograms per liter) to be measured.</para></item>
<item><para>After completion of the solution sampling, collect the biomass for metal analysis.</para></item>
<sec3 id="ch16s15"><no>3.1.1.</no> <title>Biomass Metal Analysis</title>
<para><lst type="num">
<item><para>Filter the biomass from the solution at the conclusion of the experimental trial.</para></item>
<item><para>Record wet weight.</para></item>
<item><para>Dry in drying oven. This will take 24&ndash;48 h, until a consistent weight is obtained.</para></item>
<item><para>Record dry weight.</para></item>
<item><para>Place in ashing crucible, and incinerate at 500&deg;C in muffle furnace for 2 h, until all organic mater is removed and a consistent weight is obtained.</para></item>
<item><para>Record ash weight.</para></item>
<item><para>Digest ash in Aqua Regia (1:3 mixture of nitric acid and hydrochloric acid).</para></item>
<item><para>Dissolve in Aqua Regia; will at first be a dark mixture.</para></item>
<item><para>Heat mixture over a chemically resistant hot plate to evaporate water; never allow mixture to completely dry.</para></item>
<item><para>Add more Aqua Regia and heat.</para></item>
<item><para>Repeat until solution is clear.</para></item>
<item><para>Resuspend solution in dilute hydrochloric acid.</para></item>
<item><para>Measure concentration of this solution to obtain the mass of metal adsorbed by the biomass, using the same technique as trial samples.</para></item>
<sec2 id="ch16s16"><no>3.2.</no> <title>Statistical Design of Experiments</title>
<para>The most effective experiments are those that result from a conscious and well-thought out plan to determine a specific objective. The commonly accepted tool for creating such experiments is the statistical design of experiments (<citeref rid="ch16bib37">37</citeref>&ndash;<citeref rid="ch16bib39">39</citeref>). The statistical designs will maximize the amount of information obtained from any experiment while minimizing the amount of effort needed to complete the project.</para>
<para>Generally accepted requirements for a good experimental design require that the project first has a clearly defined objective (thesis). This includes an understanding of what may and what may not be learned from the experimental data, as well as what size of an effect can be expected to not be overlooked (relative to the experimental error). Second, the planned experiments must estimate the precision of the results. Without this estimation, there is no way to determine if the observed effects are real or just the byproduct of random errors. This estimation is obtained by replication of each and every experimental trial (<i>see</i> <secref rid="ch16s18">Note 6</secref>). Enough replications must be performed to prove that effects that are large enough to have practical significance will be statistically significant. Third, each variable should have a control case, to ensure that the observed effect is actually correlated with the design variable. With the objective in mind, the experiments may be designed by considering the independent variables (experimentally controlled and varied); the dependent variables (measured responses); the types and frequency of measurements; the relationship between the measured variables and the objective; cost of the experiment; precision of the measurements; and what prior knowledge exists. These will be discussed in more detail in the second section of this topic.</para>
<para>Two types of precision are discussed, as they are all too frequently confused. First is the precision of a measurement. This is determined by performing multiple analysis on a given sample from an experimental trial. It is assumed that a given sample is homogeneous and that each portion of a sample will yield similar results. This is important in demonstrating the value of the analysis technique. The second type of precision involves replication of each trial of an experiment. These may be done in different physical containers simultaneously or at different times. This precision demonstrates the variability in the overall process, and it represents the quantification of the importance of each independent variable. If the measurement precision is poor, then it may be difficult to determine whether or not the experimental precision is significant. Excellent measurement precision will not improve the experimental precision if the independent variables do not significantly affect the dependent variables. An ideal experiment uses a measurement that is very precise to clearly establish the importance of each independent variable on each dependent variable.</para>
<para>Controls are special trials added to an experiment to verify certain hypothesis used in developing the experimental design. The use of a control should establish, within the experimental precision limits, that the considered independent variable does cause the observed response in the dependent variable (<i>see</i> <secref rid="ch16s18">Note 7</secref>). An example may illustrate the point: to prove that biomass is indeed removing metal in an experimental trial, a control experiment might include everything in the trial, except the biomass. The expected response would be that no removal is observed during the experiment in the control. Actual measurement should confirm this. If it is not true, then something else must be causing the removal (such as adsorption onto the walls of the experiment container).</para>
<para>An experiment can be designed once the objectives have been discovered; the independent and dependent values are identified, and the variable levels of the independent variables are chosen. There are many types of experimental design (<citeref rid="ch16bib39">39</citeref>). I will discuss one very simple type called a factorial design. The number of experimental trials is determined by multiplying the number of levels of each independent variable together. For example, if there are three variables, two of which will be examined at two levels and one at three levels, then the number of trials is 2&times;2&times;3, or 12 trials. Additionally, each trial must be replicated in at least duplicate to provide the information needed to determine the experimental precision. Triplicates are often used to help improve the estimate of the experimental precision. Triplication of the trials in this example would lead to 36 total experimental trials. Additional trials for the controls also need to be included and replicated. It is important to consider the total number of trials when choosing the number of variables and levels. In general, only two levels of a variable should be considered unless it is strongly suspect that the relationship between the variables is very nonlinear (perhaps going through a maximum or minimum within the range of values). Analysis of the results from a factorial experiment is straightforward, though tedious. Many statistical software packages will perform this analysis, such as SAS institute Inc. JMP (<citeref rid="ch16bib40">40</citeref>), or Minitab (<citeref rid="ch16bib41">41</citeref>). This method has a very large advantage over one at a time variable testing, as it can identify the interactions between variables with the least amount of experimental work.</para>
<sec2 id="ch16s17"><no>3.3.</no> <title>Bioremoval Experiments</title>
<para>A typical objective for bioremoval experiments may be to determine the kinetic and equilibrium model parameters for the adsorption of a metal by some type of biomass. The important dependent variables include the equilibrium concentration of metal in solution, the bioremoval potential, and the concentration factor of metal adsorbed on the biomass. Additional dependent variables to consider are the growth rate (or lack thereof) of the biomass, and time required to reach equilibrium. Determination of these quantities requires measurement of the concentration of metal in solution at several times (<i>see</i> <secref rid="ch16s18">Note 3</secref>), the amount of biomass, and the amount of metal in the biomass at the beginning and end of the experiment. The standard procedures for the measurements of metal concentrations can be found in <b>ref.</b> <citeref rid="ch16bib36">36</citeref>. It is especially important to properly prepare all standards (<i>see</i> <secref rid="ch16s18">Note 8</secref>).</para>
<para>The experimental or independent variables may include the amount and type of biomass, the amount and type of metal contaminants, and water quality (especially pH and temperature). These variables may be actively controlled or just monitored during the experiments. Only the actively controlled variables will be considered when designing the experiment. For each of the controlled variables, the experimenter chooses two or more levels at which to investigate it. The levels should be chosen far enough apart so that the measured response can be expected to show a significant difference if it exists. The levels may be found by considering the measurement precision. All of these variables must also be measured, usually at the beginning of the experiment.</para>
<para>A simple batch experiment for examining the bioremoval of a single metal using a single plant will include three types of control trials, each replicated for determination of precision. The first control uses a trial without any metal or plant. This control may show if additional sources of metal or plant are present or it may detect contamination from poor laboratory technique. The second control uses a trial with just the biomass. This control will allow determination of reduction in growth rates because of the metal. The third control uses a trial with just metal. This control will expose any additional removal sources, such as adsorption to the container walls, or a chemical or physical precipitation not associated with the biomass. Each control helps determine if the assumptions used to build the model are adequate. Finally, the experiment trials will include both metal and plant perhaps at two levels each. The results from the controls will allow the experimenter to determine the significance of the controlled variables (<i>see</i> <secref rid="ch16s18">Note 9</secref>).</para>
<sec1 id="ch16s18"><no>4.</no> <title>Notes</title>
<para><lst type="num">
<item><para>Water: typical sources of water include lab water (distilled/deionized ), synthetic waste water, and field water. Each has advantages and disadvantages. Deionized water will provide the results with the best precision and reproducibility; however, it is a poor choice for growing biomass in and will probably be the least useful choice in providing information for a particular field site or engineering bio-removal project. Synthetic waste water is made in the lab and will contain many of the additional impurities that field site waste water would contain, though in carefully controlled and measured quantities. Results using this type of water will more closely match the field site, but it may contain too many variables to allow consideration of each one. Field water is a sample taken directly from the source to be treated. It is an excellent choice for developing site-specific data. It will suffer from unknown materials and its composition may vary almost randomly over time. There is no one best choice and the selection should be based on how the results will be used.</para></item>
<item><para>Metal: source and type of metal contaminant include specific metal compounds such as soluble salts or organometallic complexes. Simple salts provide easy-to-reproduce results, which do not fully model the extreme variability in water chemistry of field waste water. More complex sources will be more realistic, but may be difficult to fully quantify in a reproducible manner. There are many possible combinations. Any choice must be based on consideration of how the results will be used.</para></item>
<item><para>Biomass: source and type of biomass used is probably the most important, and least understood, variable in this work. Many different plants have the ability to adsorb metals from aqueous solution and the adsorption can occur on living or dead biomass. The overall process is very similar to ion exchange using beads or resin, except that there is no regeneration step. Instead, the plants either are replaced or simply continue to grow in the reactor. The metal is removed by harvesting an appropriate amount of the biomass over time. A few concerns when choosing a biomass source include: use of dead or living biomass, bioremoval potential, concen tration factor, availability of supply, ease of removal from the solution (harvesting), growth rates and ease of growth, nutrient, light requirements, synergistic/antagonistic effects with multiple metals or multiple plants, and interaction with any sediments. The biomass can be grown separately from the waste stream and the harvested material can be placed in the waste-removal stream, with no concern for keeping it healthy and alive, or it can be grown directly in the waste stream, if the waste-stream conditions are not too harsh. Living biomass will replace itself over time if provided the proper conditions for growth. Dead biomass can be used in waste streams too harsh for the biomass to live in. It is noted that there may be no need to kill, dry, or cure the biomass before using its removal potential. Particle size of the biomass does not appear to be an important variable (<citeref rid="ch16bib5">5</citeref>). Sediment interaction may be especially important for emergent plants&mdash;those with roots in the sediment&mdash;where they may help remove metal from sediments, but may cause problems when harvested. Synergistic/antagonistic effects relate to how the biomass interacts with other biomass and how it interacts with multiple species of metals in the system. Sometimes multiple component systems can do more together than the individual parts, sometimes less. These effects are difficult to predict without experimental testing.</para></item>
<item><para>The preparation of standards of low concentration metals may be significantly altered by adsorption onto the glass walls of labware. All standard solutions should be acidified, as discussed in standard methods (<citeref rid="ch16bib40">40</citeref>). A typical solution for standards is 2&percnt; by weight nitric acid.</para></item>
<item><para>The growth and care of the plant biomass may be the most difficult part of this research. All plants need light, nutrients, air, and space. Water quality may be especially important for some species. If harvesting from a local source, measure and test as many water quality parameters as possible. If possible, find someone who has experience with the particular class of plants. In terms of ease of care, the <i>Lemna</i> species are very robust and capable of handling some neglect. Macroalga can also be easy to care for, though they need constant attention and may quickly die when neglected for even short times (days). Microalga are very easy to grow but are extremely difficult to harvest (filter) from a solution. They also tend to clog any pumps, airlines, or other water treatment equipment. They are extremely unpopular with waste water-treatment personnel.</para></item>
<item><para>There is no one best plant and several species may be used either together or individually. True removal occurs when the plant is harvested from the solution. The metal will have been concentrated by 100- to 10<sup>6</sup>-fold. The plant mass may need to be treated as a hazardous waste, though it may have much less volume than a chemical precipitation sludge, especially if the biomass volume can be further reduced by composting or air-drying. For single metal species removal, some consideration may be given to recovery of the metal (<citeref rid="ch16bib42">42</citeref>,<citeref rid="ch16bib43">43</citeref>).</para></item>
<item><para>Have the greatest concentration measurement frequency early in the run, as it is at the early times that the most significant changes in concentration occur. I recommend samples at times of 0,2,4, 8,12, and 24 h, and then once per day afterwards. It is only necessary to run until you determine what the equilibrium concentration is. This may occur between 48 and 168 h.</para></item>
<item><para>Always provide replication of all measurements and all experiments. Unreplicated data is worthless.</para></item>
<item><para>Use the control trials to ensure that there are no additional sources or sinks for the metals or biomass. These surprises can cost significant time when found late in the experimental trials.</para></item>
<ref id="ch16bib1"><no>1.</no> <author><snm>Muramoto</snm>, <fnm>S.</fnm></author> and <author><snm>Oki</snm>, <fnm>Y.</fnm></author> (<year>1993</year>) <articletitle>Removal of some heavy metals from polluted water by water hyacinth</articletitle>. <jnltitle>Bull. Envir. Contam. Toxic</jnltitle> <vol>30</vol>,<pages><fpage>170</fpage>&ndash;<lpage>177</lpage></pages>.</ref>
<ref id="ch16bib2"><no>2.</no> <author><snm>Axtell</snm>, <fnm>N. A.</fnm></author>, <author><snm>Sternberg</snm>, <fnm>S. P. K.</fnm></author>, and <author><snm>Claussen</snm>, <fnm>K.</fnm></author> (<year>2003</year>) <articletitle>Lead and nickel removal using microspora and <i>Lemna minor</i></articletitle>. <jnltitle>Bioresource Technol.</jnltitle> <vol>89</vol>,<pages><fpage>41</fpage>&ndash;<lpage>48</lpage></pages>.</ref>
<ref id="ch16bib3"><no>3.</no> <author><snm>Sobhan</snm>, <fnm>R.</fnm></author> (<year>1997</year>) <booktitle>Cadmium Removal Using Living Aquatic Plants.</booktitle> <misctext>MS thesis</misctext>, <publisher><pubname>University of North Dakota</pubname>, <location>Grand Forks, ND</location></publisher>.</ref>
<ref id="ch16bib4"><no>4.</no> <author><snm>Haq</snm>, <fnm>N.</fnm></author> (<year>1998</year>) <booktitle>In-situ Bioremediation of Aqueous Lead and Cadmium Using Plants</booktitle>. <misctext>MS thesis</misctext>, <publisher><pubname>University of North Dakota</pubname>, <location>Grand Forks, ND</location></publisher>.</ref>
<ref id="ch16bib5"><no>5.</no> <author><snm>Dorn</snm>, <fnm>R.</fnm></author> (<year>1998</year>) <booktitle>Cadmium Removal Using <i>Chladophora</i> in a Flow Reactor</booktitle>. <misctext>MS thesis</misctext>, <publisher><pubname>University of North Dakota</pubname>, <location>Grand Forks, ND</location></publisher>.</ref>
<ref id="ch16bib6"><no>6.</no> <author><snm>Gardea-Torresdey</snm>, <fnm>J. L.</fnm></author>, <author><snm>Gonzalez</snm>, <fnm>J. H.</fnm></author>, <author><snm>Tiemann</snm>, <fnm>K. J.</fnm></author>, and <author><snm>Rodriguez</snm>, <fnm>O.</fnm></author> (<year>1998</year>) <articletitle>Biosorption of cadmium, chromium, lead, and zinc by biomass of <i>Medicago sativa</i> (Alfalfa)</articletitle>. <jnltitle>J. Haz. Mat.</jnltitle> <vol>57</vol>, <pages><fpage>29</fpage>&ndash;<lpage>39</lpage></pages>.</ref>
<ref id="ch16bib7"><no>7.</no> <author><snm>Kratachvil</snm>, <fnm>D.</fnm></author> and <author><snm>Volesky</snm>, <fnm>B.</fnm></author> (<year>1998</year>) <articletitle>Advances in the biosorption of heavy metals</articletitle>. <jnltitle>Trends Biotechnol.</jnltitle> <vol>16</vol>, <pages><fpage>291</fpage>&ndash;<lpage>300</lpage></pages>.</ref>
<ref id="ch16bib8"><no>8.</no> <author><snm>Adou</snm>, <fnm>C.</fnm></author> (<year>1999</year>) <booktitle>Bioremediation of Zinc and Nickel Using Living Aquatic Plants</booktitle>. <misctext>MS thesis</misctext>, <publisher><pubname>University of North Dakota</pubname>, <location>Grand Forks, ND</location></publisher>.</ref>
<ref id="ch16bib9"><no>9.</no> <author><snm>Roditi</snm>, <fnm>X.</fnm></author>, <author><snm>Hudson</snm>, <fnm>A.</fnm></author>, <author><snm>Fisher</snm>, <fnm>X.</fnm></author>, <author><snm>Nicholas</snm>, <fnm>S.</fnm></author>, <author><snm>Sanudo-Wilhelmy</snm></author>, and <author><snm>Sergio</snm> <fnm>A.</fnm></author> (<year>2000</year>) <articletitle>Field testing a metal bioaccumulation model for zebra mussels</articletitle>. <jnltitle>Environ. Sci. Technol.</jnltitle> <vol>34</vol>, <pages><fpage>2817</fpage>&ndash;<lpage>2825</lpage></pages>.</ref>
<ref id="ch16bib10"><no>10.</no> <author><snm>Omar</snm>, <fnm>H. H.</fnm></author> (<year>2002</year>) <articletitle>Bioremoval of zinc ions by <i>Scenedesmus obliquus</i> and <i>Scenedesmus quadricauda</i> and its effect on growth and metabolism</articletitle>. <jnltitle>Internat. Biodet. Biodeg.</jnltitle> <vol>50</vol>, <pages><fpage>95</fpage>&ndash;<lpage>100</lpage></pages>.</ref>
<ref id="ch16bib11"><no>11.</no> <author><snm>Dursun</snm>, <fnm>A. Y.</fnm></author>, <author><snm>Uslu</snm>, <fnm>G.</fnm></author>, <author><snm>Tepe</snm>, <fnm>O.</fnm></author>, <author><snm>Cuci</snm>, <fnm>Y.</fnm></author>, and <author><snm>Ekiz</snm>, <fnm>H. I.</fnm></author> (<year>2003</year>) <articletitle>A comparative investigation on the bioaccumulation of heavy metal ions by growing <i>Rhizopus arrhizus</i> and <i>Aspergillus niger&#

Is This A Good Question/Topic? 0
  • +

Replies To: xml

#2 girasquid   User is offline

  • Barbarbar
  • member icon

Reputation: 108
  • View blog
  • Posts: 1,825
  • Joined: 03-October 06

Re: xml

Posted 12 February 2008 - 08:36 AM

What is the error you are getting?

Try checking to make sure that all of your quotes match up properly.
Was This Post Helpful? 0
  • +
  • -

#3 no2pencil   User is offline

  • Professor Snuggly Pants
  • member icon

Reputation: 6644
  • View blog
  • Posts: 30,937
  • Joined: 10-May 07

Re: xml

Posted 12 February 2008 - 09:56 AM

View Postrealwish, on 11 Feb, 2008 - 11:32 PM, said:

pls help me to find the error in the code

Normally you won't get your 'work' done for you, but geeze, just load the xml file into firefox & it'll tell you what the error is!


XML Parsing Error: undefined entity
Location: file:///C:/test.xml
Line Number 54, Column 164:<para>Bioremoval is a process that can remove soluble heavy metal contaminants from aqueous solutions by an adsorption process (<citeref rid="ch16bib1">1</citeref><citeref rid="ch16bib11">11</citeref>). This process is also called phytoremediation, bioremoval, biosorption, bioaccumulation, or phytoremoval. The adsorbant is biomass, often obtained from an aquatic plant, which may be living or dead (<citeref rid="ch16bib12">12</citeref><citeref rid="ch16bib14">14</citeref>). The biomass is capable of accumulating the metal at concentrations many times greater than that of the solution (<citeref rid="ch16bib15">15</citeref><citeref rid="ch16bib21">21</citeref>). The metal is removed from the system by a simple filtration of the biomass. This process has the advantages of being simple, with a low cost to build and to maintain (<citeref rid="ch16bib22">22</citeref>,<citeref rid="ch16bib23">23</citeref>). It may not require changes in water qua

Was This Post Helpful? 0
  • +
  • -

Page 1 of 1