This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
A type of sample collection and a fraction of water to be sample depend primarily on the analytical procedures, which is, in turn, determined by the purposes of the monitoring projects. Therefore, the purpose of the monitoring project should be defined before undertaking a sampling program.
In this report, only Cadmium (Cd) will be monitored. In particular, only dissolved Cd will be subject to examination since the presence of particulate Cd is not of interest for drinking water or wastewater and the criteria for particulate Cd for both types of water is not found in any guideline. Therefore, the term Cd hereafter will always refer to dissolved Cd, unless otherwise stated. The purposes of the sampling program are described below.
For drinking water: The water samples will be used to examine the weekly average total concentration of Cd at the consumers' taps. The variability of Cd concentration at different sites and times will NOT be concerned. According to ADWG 2006, the amount of Cd in drinking water must not exceed 0.002 mg/L.
For wastewater: The main sampling objective is to inspect the compliance status of permitted municipal wastewater treatment plant (assuming the secondary treatment level). The role of analysts is to ensure the satisfied quality of wastewater before discharging to a receiving environment. In term of Cd, a typical consent limit is 0.1 mg/L.
The loss of cadmium (Cd) during the sample collection is a major concern. The main factors affecting the extent of Cd loss (and also other trace metals) during sampling and storage include PH, types of containers and the length of storage time. Some studies and general advices pertaining to loss of Cd under various conditions of these factors may be found in Subramanian et al. 1978. According to their finding, as far as Cd preservation is concerned, a linear polyethylene container is a suitable option for storing water samples at PH < 8.0 (while the Pyrex glass containers should always be avoided). Note that quartz- or Teflon-based sample containers are actually the best options in term of quality but they require a higher capital cost (ed.)Eaton et al. 1995.
There are other important considerations that must be aware of before and during collecting samples either from drinking water or wastewater. Some general precautionary measures (See, (ed.)Eaton et al. 1995) to preserve dissolved Cd (also applying to most trace metals) in the samples are listed in the followings:
All equipments that may contact with water must be cleaned by acid before undertaking sample collection. A recommended procedure is an overnight stagnation cleaning by 10% HNO3.
All equipments to come into contact with water must be made of materials that do not neither adsorb nor release Cd or any other metals
Linear polyethylene (or polypropylene) containers should be used for the above reason. Teflon (PTFE) containers should be used if their (high) cost is not of concern.
Since dissolved Cd is the main concern, the sample should be filtered at the time of collection by a membrane filter of a 0.45 Âµm pore diameter.
Following filtration, the filtrated sample should be immediately acidified with >69% HNO3 to pH<2.
After acidifying, samples should be stored in collection containers and refrigerated at approximately 4oC until time of extraction (maximum storage is 6 months).
The field blanks should be prepared using the same acid and containers as the sample for the purpose of quality control, which will be explain in Section 5.
The minimum sample size is 500 mL for determination of most metals.
Apart from the above general advices, the procedures for sampling of drinking water and wastewater are quite different and their advices are separately presented below.
Sampling of drinking water
In the situation of drinking water, water samples will be taken from multiple points at the consumers' taps (preferably, a tap in a kitchen). The composite sampling is suitable for the purpose of determining the average concentration. In fact, in view of cost efficiency, it is more favourable than the grab sample, which may require more visits to sampling sites; hence higher labour cost and transportation. The details of below procedures can be found in Corfitzen & Albrechtsen 2008.
The samples will be collected using plastic split-valves, which feature a manual switching between a one- or two- stream flows; one is the main stream (90-96%) for consumers' ingestion and the other is a small separated-off stream (4-10%) for sampling collection. The choice of plastic material is preferred over metals for the above reason. The container should have two caps. One of the cap will be attached to a polyethylene tube to lead water from the split valve to the container while the other cap is attached to the filtration tube to allow pressure equalisation during the sampling as well as prevent contamination from surrounding particles.
This method requires consumers' cooperation to activate the split-valves whenever they ingest the tap water at which point a small fraction of separated-off stream will flow to the containers by gravity. The collection period will be one week and hence a container should have a capacity at least 4 litres based on an assumption that 40-50 litres of water is drawn for ingestion in a week and that water from the kitchen tap made up less than 10% of the total water for weekly ingestion.
Sampling of Wastewater
In the case of wastewater, grab samples of final effluent will be collected at the effluent channel. Sampling personnel will coordinate with plant authorities to ensure that the sampling locations: (1) have no bypass capabilities. (2) are accessible at all times. (3) provide reliable representative of the discharge.
Each monitoring usually takes 2-5 days. One sample will be collected each day with a 2-Litre linear polyethylene containers. The sample will be brought back to the laboratory on ice. Other consideration will follow the aforementioned advices.
Available techniques for sample extraction
The sample extraction (or preconcentration) of trace metals in water is usually required to reduce interferences from matrix constituents in samples and to concentrate the sample analyte to a level detectable by analytical methods. The organic compound could interfere the analytical procedures, for example, by interacting with electrode materials or by creating complexes with ions. The sample extraction/preconcentration can alleviate this problem and hence enhance the limit of detection for many methods.
The following techniques are widely used for sample extraction for trace metals, see Ferriera et al. 2007.
Liquid-liquid extraction (LLE)
LLE (also known as solvent extraction) involves a process of removing a metal from aqueous matrix samples by bringing it into a contact with another immiscible liquid phase. The solute (i.e. the analyte) will be more concentrate in the solvent, in which it has a high solubility, as an organic phase. Meanwhile, the interfering species remains in the aqueous phase. The desirable solvent should: (1) be able to extract the target metal chelate; (2) be immiscible with the aqueous solution; (3) not form emulsion, see Loon 1986.
For the LLE, the choices of buffer and chelating agents are important. A desirable buffer should be stable, have a high buffering capacity and not permit metal contamination. A choice of chelating agents depends on the quantity of target metals it can extract and the range of pH over which metals can be equally extracted, a wider range is always preferable since pH adjustment is a common error in many routine applications.
Analytical procedures utilizing LLE for Cd determination in water are listed in Table A1, see Ferriera et al. 2007.
Cloud point extraction (CPE)
CPE can be considered as an unconventional variation of LLE. The principle of CPE procedures for metal determination is based on the phase separation phenomenon whereby under certain conditions (e.g. temperature) an aqueous solution of surfactant attains the cloud point and become turbid, at which point it separates into two phases, see Bezerra et al. 2005 and Pereira & Arruda 2003. Roughly speaking, this phase separation is implemented in the following sequence: (1) A chelating agent is added to the matrix solution to form metal chelates. (2) A suitable surfactant solution is added into the aqueous solution. To allow micellar formation, a concentration of surfactant must be above critical micelle concentration (CMC). (3) When formed, the metallic chelate will be associated with micellar cores where it is ``trapped". (4) The temperature is then raised to the cloud point temperature (CPT) at which point micellar phase segregate resulting in a high surfactant concentration (small volume) while the other phase has a low concentration. This phase separation can be accelerated by centrifugation. The schematic of the this method is illustrated in Figure 1.
Figure 1: (A) Metals is present in the original solution at low concentration. (B) A Chelate reagent is introduced to form metal chelates. (C) Micellar system is formed by adding surfactant, which allows metal chelates to be ``trapped" with the micelle core. (D) Phase separation occurs: a high density surfactant-rich phase and a low density bulk phase.
In practice, the bulk aqueous phase can be discarded by ice bath (since the surfactant-rich phase in a form of hydrophobic complexes has a high viscousity), see Bezerra et al. 2005. The typical choice of surfactant for Cd determination is Triton X-114, which is commercially available in high purity, inexpensive, low CPT and non-toxic. Preconcentration by CPE is also widely used for Cd determination in several literature as listed in Table A2.
Solid-liquid extraction (SLE)
SLE is predominantly used for preconcentration of Cd and other metals since it greatly improves sensitivity and/or selectivity of atomic spectrometry methods. Ion-exchange resins are mostly employed due to its high surface area and porosity to water, which enhance a metallic adsorption potential. Ion-exchange resins comprises of an insoluble polymer (e.g. polystyrene) lattice with attached functional group. Common types of ion-exchange resins methods are of anion exchangers, cation exchangers and chelating resins. The chelating resin (e.g. Chelex 100) is commonly used since it has a high-selectivity ligand to the metal ion. Nevertheless, cation exchangers containing functional group -SO3H are also widely used.
According to Garg et al. 1998, SLE based on chelating resin (anchored to polymeric support) is used to pre-concentrate metal ions in the following steps: (1) Chelating agent is introduced to the sample under appropriate pH ranges to form complexes with metal ions. (2) A sample containing metal ion is passed over a chelating resin where metal ions are sequestered since they form chelates with chelating agents on the polymeric support while impurities (i.e. interfering substances in matrix) just pass through. (3) A suitable eluent is then used to extract the metal ions from chelate sites on the resin.
Chelating materials can be prepared either from synthetic or natural sorbents. However, a majority of works reported in literatures use the synthetic materials. Among a number of synthetic sorbents, fullerene (e.g, buckyball C60, single-wall and multi-wall carbon nanotube) is reported with a remarkable preconcentration factor for Cd, see Baena et al. 2002. Other commonly used sorbents for Cd preconcentration include Zeolite, Polyurethane foam, Silica gel and Amberlite.
There are two general approaches to introduce chelating agents into sorbents: (1) the chemical bonding of chelating agents on existing sorbent; and (2) the physical binding of chelating ligand on the sorbent. The latter is more typical in practices for its simplicity to implement. Nonetheless, chemical bonding offers an advantage of a strong bonding of ligands and support polymers that prevents a possible flush of ligan molecule during eluation step. Table A3 lists some of commonly used sorbents and chelating agents along with their preconcentration factor and techniques used for Cd determination in the water matrix, see Ferreira et al. 2007.
Comparison of sample extraction methods
Advantages and disadvantages for metal extraction (or preconcentration) using above techniques are listed in Table 1. It is clear that in spite of its high preconcentration factor, LLE has became less popular in recent years because an advent of ``green" and inexpensive techniques; CPE and SLE, which do not requires a use of toxic reagents. In fact, it was demonstrated by Pereira & Arruda 2003 that from 1996 to 2001, 60.5 % of metal preconcentration prior to atomic spectrometry has been reported using SLE and 13.5 % using either LLE and CPE.
Table 1. Comparisons of various techniques for sample extraction (pre-concentration)
High efficiency for removing potential matrix interferences
Expensive and slow
High consumption of toxic organic compounds (used as solvents)
May need re-extraction to reach the desired concentration
Low cost and simple to perform
Not require uses of toxic reagents
Surfacants are non-toxic
High preconcentration factor
Relatively low partition coefficients of several metal species with determinate chelate
Avoid using toxic solvents
Rapid and high selectivity
High preconcentration factor
Need only a small amount of chelating resin and sorbents
References: 1; See Ferreira et al. 2007, 2; See Bezerra et al. 2005, 3; Garg et al. 1998
Current techniques for sample analyses
There are a number of currently available techniques for quantifying Cd and each has its own advantages/disadvantages over the others. The principles of these techniques will be briefly described and contrasted in this section.
Roughly speaking, this technique utilizes specific spectral characteristic of each metal; in particular, the adsorption, emission or fluorescence of different metals (in a form of ionized atom) will occur at different wavelength. The precision and accuracy of this method depend heavily on the atomization step in which the reduction of metals into ionized atoms is performed.
In the all types of atomic spectroscopy methods, the atomization step is very similar. The fundametal type is an atomization by flame. An aqueous analyte sample is first introduced into the nebulisation chamber via a continuous suction to be transformed into a stream of fine aerosol. The aerosal will be mixed with oxidant and fuel before introduction into the flame where it is atomized. The constituents of this mixture is crucial in the determination of the flame temperature and hence the efficiency and sensitivity of the atomization. The acetylene-air flame (2100-2400 oC) is typically recommended for determination of Cd, see (ed.)Eaton et al. 1995. In case that an ultra-low concentration of Cd is to be determined, a sample preconcentration may be employed prior to aspiration into the air-acetylene flame.
After atomization, several detection techniques can be used to estimate the concentration of the atomized analyte.
Atomic adsorption spectroscopy (AAS)
In the (flame) AAS method, the hollow cathode light source is required to contain the same metal as the analyte. This light is radiated to the flame where the atomic adsorption by the analyte (only those in the elemental form) takes places at particular wavelengths, which are specific spectral characteristics of the analyte. Afterward the photon will be re-emitted and scatter in all directions. The photon emerging from the flame will pass through the monochromator whereby only photons not absorbed by the atomized analyte (those that are not re-emitted photons) are selectively allowed to reach the photomultiplier tube at the other end. The concentration of the analyte will be proportional to the degree of adsorption. The entire ASS processes are summarized in Figure 2.
This method suffers some drawbacks including:
Chemical interferences: This refer to the problem where the adsorption by atoms during the atomization is deficient, which results in Type II error in measurement. This problem usually occurs when the flame temperature is too low to disassociate the molecules or the disassociated atoms of the element of interest is oxidized with other compounds in sample, which inhibit it from further disassociation at the flame temperature.
Background correction: Light scattering and molecular adsorption by solid particles in the flame can result in higher adsorption; hence Type I error in measurement.
Figure 2: A simplified diagram of AAS processes
Atomic emission spectroscopy (AES)
In the (flame) AES method, the process is very similar to AAS except that the photons at particular wavelengths emitted by excited electron returning to the ground state are detected instead. This method does not require the hollow cathode lamp since the flame is used as an excitation source. The photomultiplier tube utilizes this metal-specific spectrum to estimates the analyte concentration, which is also proportional to the degree of emission.
Graphite furnace atomic adsorption/emission spectroscopy (GF-AAS/AES)
In GF-AAS/AES methods, the atomization source is constructed from a furnace, which is an electro-thermal source, instead of using the nebulisation chamber. The sample is introduced into the furnace tube in which the temperature is gradually increased to allow the analyte to slowly evaporated. Then the temperature is increased rapidly to create atomization.
Major drawbacks of using graphite furnace includes:
Molecular adsorption: This occurs when the elements of interest have a high volatility. A slow heating when using graphite furnace will caused them to volatize during atomization, resulting in adsorption of broad-spectral range.
Matrix effects: This mainly refers to the event that the sample components other than the element of interests interfere the formation of free atomic analyte during the atomization process.
Specific element interference: In case of Cd, the use of HCI during digestion or other processes can reduce the sensitivity of the method, see EPA method 200.9 revision 2.2.
Inductively coupled plasma (ICP-AAS/AES)
With the ICP method, atomization at a very high temperature (>7500 oC) is achieved by the plasma created from argon gas. At this temperature, atomization is almost complete, i.e., almost all elements are ionized to the single charge form. The ICP technique can be equipped with either AES or AAS for a detection system; hence ICP-AES and ICP-AAS, respectively.
The main interferences to ICP includes.
Spectral interference: Generally, spectral interferences refers to the disturbance of light emission from spectral sources other than that of the element to be determined.
Physical interference: The presence of physical interferences is usually associated with sample nebulisation, built-up substances for earlier run or transport processes.
Inductively Coupled Plasma-Mass Spectrometry (ICP-MS)
The detection system of this method uses a mass spectrometry (MS), commonly a quadruple MS, that provides a gain of 3 to 4 folds to a limit of detection. MS is usually coupled with the ICP as one of standard analytical methods. MS separates the singly-charged ions based on their mass-to-charge ratio. After sample analytes are atomized to single positively-charges ions by ICP, they will be extracted through the interface region (called cone) to a high vacuum enclosure where these single positively-charged ions will pass through the electron multiplier whence the mass spectrometer. At the electron multiplier, a number of electrons will be repeatedly amplified via the impact of positively-charged ions on the cascade of negatively-charged dynodes; hence a stronger signal for MS to read and a significantly lower limit of detection.
The key advantage of MS is its superior sensitivity (i.e. low limit of detection). However, this method can suffer from several interference sources including:
Isobaric interference: are caused by isotopes of different elements (in a form of singly- or doubly- charged ion) that have the same mass-to-charge ratio of the targeted element which results in undistinguishable isotope if using a low-resolution MS. Fortunately, Cd of 111-isotope is free from of this interference and hence recommend for use in ICP-MS (see USEPA method 200.8 Revision 5.4 1994, Table 4).
Isobaric polyatomic ion interferences: are caused by a recombination of sample analyte with some matrix elements, resulting in ions with more than one atom of which the mass-to-charge ratio is the same as that of the targeted isotope. Fortunately, Cd does not suffer from this type of interferences since its oxide interferences (ZrO and MoO) are not commonly found in drinking water or wastewater (according to USEPA method 200.8 Revision 1994, Table 2).
Other interferences: includes physical and memory interferences. The former is typically associated with the transmission or transport of ions in plasma at the interface whereas the latter involves the effect of the deposition of previous samples to the signal measure for a new sample.
Comparison of methods
Every method described here inherits some advantages/disadvantages over the others and a comparison is given in this section. The first criteria for selecting a suitable method is the limit of detection (LOD), which will be discussed in later section. The estimated LOD for Cd at 228.8 nm is reported in (ed.)Eaton et al. 1995 as the following: LOD (Âµg/L); 25 for AAS, 0.1 for GF-AAS, 4 for ICP-AAS, 0.04 for ICP-MS (see, Turiel et al. 1999). Apparently, only GF-AAS and ICP-MS has a sufficient capability for determination of Cd in drinking water(LOD<2 Âµg/L ). However, an application of a suitable preconcentration technique to AAS can remarkably improve the LOD, as shown in Table A4. Clearly, a couple of AAS with CPE or SLE makes it a potential candidate.
Qualitative considerations will also play a role in a selection of methods, see Table 2.
Table 2. Qualitative comparisons of related methods for trace metal determination
High efficiency for removing potential matrix interferences
Only one element can be determined at a time
Inexpensive, widely used
Chemical interferences (especially for water with high salinity)
Need a background correction
Only one element can be determined at a time
Low cost and simple to perform
Not require use of toxic reagents
Surfacants are non-toxic
High preconcentration factor
Only one element can be determined at a time
Not suitable for element of a high volatility (which includes Cd)
Sample preconcentration may not be ncessary
Many potential interferences
Problematic if measure Cr, Ni, and Ca because of Isobaric interferences
We now make a decision on which method is most suitable for determination of Cd in drinking water and wastewater. The method predominantly used for trace metals analysis of drinking water and wastewater are AAS, GF-AAS and ICP-MS. Our selection will be based on these methods.
AAS: This method benefits from its low cost, high sample throughput, simple operational procedures, and, in case of Cd, it is free from interferences but it suffers from its low sensitivity. However, a preconcentration and extraction (e.g. CPE or SLE) can substantially improve the sensitivity limit of AAS. Yet another drawback of AAS is that it is not multi-elemental. Although a multi-wavelength radiation source is available, it is typically not recommended. Thus it is practically time consuming if determinations of many trace elements are required.
GF-AAS: For GF-AAS, its slow heating process during atomization could be problematic for Cd determination because Cd has a high volatility. So, this method should be put as the last resort among related methods in this report.
ICP-MS: ICP-MS offers a great advantage of high sensitivity and multi-element capability. However, it requires a high capital cost and also suffer from many interferences, which make its operation more difficult. Moreover, in spite of its high sensitivity, ICP-MS may still require a suitable sample extraction and/or preconcentration procedures to reduce concomitant effects in some cases (e.g. saline in seawater could accumulate at the equipment interfaces).
Based on the pros/cons and the LOD criteria, the suitable methods are decided as follows:
In case of drinking water, ICP-MS will be used for Cd determination based on its ultra-low sensitivity and multi-elemental determination. Although the latter advantage is not necessary in this report, in which only Cd is of concern, it is practically an important advantage for most drinking water monitoring program. Its high cost may be compensated from the fact that it may not require a preconcentration/separation processes (e.g. SLE or CPE), which involves extra costs of reagents and sorbents used in these processes. So, for a long term uses, it would require a life-cycle analysis to fairly justify this issue of capital cost; nevertheless, ICP-MS is still expected to bear a higher cost.
In case of wastewater, the concentration of Cd is not as low as that in the drinking water. A use of (flame) AAS in conjunction with a suitable preconcentration method (either CPE or SLE) proves to be sufficient, e.g. see Afzali & Mostafavi 2008 and Mohamadi et al. 2010. The AAS (using the air-acetylene flame) has been a common method for decades, so it is inexpensive and experienced staffs are abundant. The preconcentration and/or separation by SLE is also inexpensive, simple to operate and free from toxic reagent.
A suitable approach to undertake for accurate quantitation
All of methods herein are essentially a sort of ``comparative" method where the signal intensity of analytes in questions is compared with some ``references" to allow accurate quantitation. Typical methods for establishing calibration curve include external standard, internal standard and standard addition. The principles of these methods will be briefly described here, for the details of their implantation referring to (ed.)Eaton et al. 1995.
Principle: The calibration curve is specifically prepared for the analyte element using external calibration standards of known concentration (including calibration blanks of zero standard), each of which will indicate the instrument response to a concentration. The calibration curve is plotted for instrument responses (y-axis) v.s. concentration of the calibration standards (x-axis). In practice, the linear range of a calibration curve should span over the concentration range of interest. The primary instrument response can then be converted to the analyte concentration through linear regression of the calibration curve.
Pros/Cons: When the analytical method is strongly matrix-dependent, it requires the matrix matching between the analyte sample and the calibration standards. This surely becomes problematic if the matrices are unknown. Nonetheless, if this is not the case, this method is attractive for its relatively rapid implementation.
Principle: The basic idea of this method is to add aliquots of some known analytes, called "internal standard", that do not present in the sample prior to the addition but should have the same physical and chemical property as the analyte of interest. The plot of concentrations of analytes to be determined (x-axis) v.s. the ratio of analyte signal to the internal standard signal (y-axis) will be used as a calibration curve instead.
Pros/Cons: Clearly, this method can effectively eliminate the interferences influencing the internal standard to same extent as the analyte (e.g. those matrix effects that reduce nebulizer efficiency at the introduction system). If that is not the case for suspected interferences, internal standards must be chosen so that they suffer the inferences in a similar extent to the analyte. This may be difficult to achieve if many analytes are to be simultaneously measured by ICP-MS.
Principle: In this method, the analyte response is compared with the response of other samples to which a small volume of known standard solution is added (or technically , called ``spiked"). Typically, an analyte sample should be spiked so that its concentration is increased by a factor larger than 2. The concentrations of each spike contribution are plotted on the x-axis and the corresponding instrument signals are plotted on the y-axis. Hence, the y-interception will be the signal response of analyate. Then, the concentration of the analyte is taken to be the x-interception point (after taking the absolute value), which is obtained by extrapolation, assuming that the instrument responses is linearly dependent to the analyte concentration.
Pros/Cons: This method is suitable when the matrix is unknown and help to eliminate the interferences associated with the temperature reduction by the matrix components during atomization. However, this is a cumbersome method since the accurate spiking is crucial and it would be time-consuming to prepare highly accurate spiked samples.
Isotope dilution (only for ICP-MS)
Principle: In this technique, the sample is fortified with a know amount of an isotope of the analyte (prior to sample preparation for a technical reason). Given that the isotopic ratio of the added isotope to the analyte isotope is known, the sample concentration can be calculated by measuring the isotope ratio of the sample and isotope-spiked sample.
Pro/Cons: This method can eliminate all physical interferences methods but it is not available for monoisotopic element. Also, it cannot resolve the isobaric interference.
Suitable quantitation methods
In the case of ICP-MS used for determination of ultra-low Cd in drinking water, the isotope dilution can be used to eliminate all of the matrix effects, assuming that the isobaric interferences are negligible as previously explained. However, in the case of AAS, the conventional calibration curve by external standard will be used based on its simplicity to implement. This is usually sufficient since matrix effects are not strong when using AAS in conjunction with SLE.
Quality control procedures
In order to have a good quality control procedure, the following processes will be additionally performed during the analytical procedures:
Analysis of calibration blank
Before running any sample, the calibration blank will be analysed.
Analysis of field blank (FB)
The field blank is an aliquot of reagent water or other blank solution that is handled similar to the sample of interest in all aspects, including sampling, storage, preservation and preparation. The purpose of the FB is to monitor the presence of any method analytes or interferences (e.g. spectral background in reagents) that may present in the field or laboratory environment. As done with the samples, the calibration blank should be run prior to analyse the field blank. If the field blank values exceed MDL, the contamination or interferences must be suspected and the correction must be performed to minimize the contamination. As suggested by EPA, the sample must be prepared and analysed again "when FB values constitute 10% or more of the analyte level determined for a sample or is 2.2 times the analyte MDL whichever is greater".
Analysis of laboratory fortified blank (LFB)
The LFB can be prepared by fortifying a known amount of analytes to the field blank. The LFB will be analysed using the same procedure as samples in order to measure the accuracy and bias of the methodology. The analysis of LFB is then reported as the percent recovery (R) calculated by:
where S the concentration equivalent of analyte added to fortify the LBR solution. If R is out of the range of , a speculation to identify the sources of problem is required.
Quality control Sample (QCS)
The QCS is used for the same purpose as LFB except that instead of being prepared in the laboratory, QCS is purchased from the outside source to demonstate a freedom of contaminants in the laboratory. It always comes with a certified range of detection for a comparison.
Typical analytical detection limits
Several detection limits used to validate and control the quality of the methods are described in this section. In fact, some of them have been previously outlined.
Limit of detection (LOD)
LOD is defined in (ed.)Eaton et al. 1995 as "the constituent concentration in reagent water that produces a signal 2(1.645)s", where s is the standard deviation of the concentration measured from multiple replicates. In other word, the probability of false detection (either Type I or II errors) is not greater than 5%. Typically, 3s is used instead of 2(1.645)s for convenience.
Limit of quantitation (LOQ)
LOQ is defined in (ed.)Eaton et al. 1995 as "the constituent concentration that produces a signal sufficiently greater than the blank that it can be detected within specified levels under optimal conditions of laboratory and instrument". As a convention, LOQ is set to be the concentration that produces a signal 10s above the reagent water blank signal.
Method detection limit (MDL)
The MDL is defined by EPA as ``the minimum of detectable concentration of analytes that can be measured with 99% confidence that the true analyte concentration is greater than zero". The procedure of MDL calculation may be found in 40 CFR 136. It is worth mentioning that the MDL is the only ``practical" limit in a sense that it takes into account the true performance of instrument at the time of analysis, which depends on the calibration of the laboratory personnel, the (non-optimal) conditions of instrument after using many years and etc.
Instrumental detection limit (IDL)
IDL is defined in (ed.)Eaton et al. 1995 as "the constituent concentration that produces a signal greater than five times the S/N of the instrument". IDL is useful to evaluate instrument noise level, which may be vary over time, the procedure for IDL calculation is given in EPA.
Typical limit for the selected methods
The limits of ICP-MS for determination of Cd in drinking water as reported in Turiel et at. 2000 are listed in Table A5. The isotople line is Cd-111 and two QCSs are used as reference materials; NIST SRM 1643c and AC-E. LOD (3s) is based on ten replicate measurements of the reagent blank. Relative standard deviation (RSD) is based on 12 replicate measurements of samples.
The limit and characteristics of AAS using SLE preconcenration (Multiwalled carbon nanotubes) as reported in Mohammadi et al. 2010 are listed in Table A6. The LOD (3s) is based on 10 replicate measurements. The QC includes field blanks, QCS and LFB, see Table A7.
D. Afzali and A. Mostafavi, Flame atomic absorption spectrometry determination of trace amounts of Ag+ and Cd2+ after simultaneous separation and preconcentration
on to modified clinoptilolite zeolite as a new sorbent, Canadian Journal of Analytical Sciences and Spectroscopy, 53(2): 82-89, 2008
I.A. Ansari, V.K. Dewani and M.Y. Khuhawar, Evaluation of metal contents in Phulleli canal and Hyderabad city sewage by flame atomic absorption spectrometer, J. Chem. Soc. (Pakistan), 21:359-368, 1999
J. R. Baena, M. Gallego, and M. Valcarcel, Fullerenes in the analytical sciences, Trends Anal. Chem., 21: 187-198, 2002
M. A. Bezerra, M. Z. Arruda and S. C. Ferreira, Cloud point extraction as a procedure of separation and pre-concentration for metal determination using spectroanalytical techniques: A review, Applied Spectroscopy Reviews, 40: 269-299, 2005
O. T. Butler, J. M. Cook, C. M. Davidson, C. F. Harrington and D. L. Miles, Atomic spectrometry update. Environmental analysis, J. of Analytical Atomic Spectrometry, 24: 131-177, 2009
J. Chen and K.C. Teo, Determination of cadmium, copper, lead and zinc in water samples by flame atomic absorption spectrometry after cloud point extraction, Anal. Chim. Acta, 450:215-222, 2001
C. B. Corfitzen and H. J. Albrechtsen 2008, Sampling for drinking water quality in drinking water installations regarding metal concentrations: Method Description & Validation of Method, The Agency for Spatial and Environmental Planning, view 19 May 2008, <http://www2.blst.dk/udgiv/Publications/978-87-92256-51-5/html/default_eng.htm>.
A.D. Eaton, L.S. Clesceri, and A. E. Greenberg (editors) 1995, Standard methods for the examination of water and wastewater, American Public Health Association, Washington, DC.
I. Facchin and C. Pasquini, Two-phase liquid-liquid extraction in monosegmented flow analysis. Determination of cadmium with 1-(2_-pyridylazo) naphtol, Anal. Chim. Acta, 308:231-237, 1995
S. L.C. Ferreira, J. B. de Andrade, M. Gracas, A. Korn, and et al., Review of procedures involving separation and preconcentration for the determination of cadmium using spectrometric techniques, J. of Hazardous Materials, 145:358-367, 2007
E.M. Gama, A.S. Lima and V.A. Lemos, Preconcentration system for cadmium and lead determination in environmental samples using polyurethane foam/Me-BTANC, J. Hazard. Mater., 136:757-762, 2006
B. S. Garg, R. K. Sharma, N. Bhojak, and S. Mittal, Chelating Resins and Their Applications in the Analysis of Trace Metal Ions, Microchemical Journal 61: 94-114 , 1999
E. Kenduzler, Determination of cadmium(II) in water and soil samples after preconcentration with a newsolid phase extractor, Sep. Sci. Technol., 41:1645-1659, 2001.
V.A. Lemos and P.X. Baliza, Amberlite XAD-2 functionalized with 2-aminothiophenol as a new sorbent for on-line preconcentration of cadmium and copper, Talanta, 67:564-570, 2005
J. C. V. Loon, Selected methods of trace metal analysis: Biological and environmental samples, John Wiley & Sons, New York, 1985.
J.L. Manzoori and G. Karim-Nezhad, Development of a cloud point extraction and preconcentration method for Cd and Ni prior to flame atomic absorption spectrometric determination, Anal. Chim. Acta, 521:173-177, 2004.
S.Z. Mohammadi, D. Afzali and D. Pourtalebi, Flame atomic absorption spetrometric determination of trace amounts of lead, cadmium and nickel in different matrixes after solid phase extraction on modified multiwalled carbon nanotubes, Cent. Eur. J. Chem., 8(3): 662-668, 2010.
I. Narin, M. Soylak, K. Kayakirilmaz, L. Elci and M. Dogan, Preparation of a chelating resin by immobilizing 1-(2-pyridylazo) 2-naphtol on amberlite XAD-16 and its application' of solid phase extraction of Ni(II), Cd(II), Co(II), Cu(II), Pb(II), and Cr(III) in natural water samples, Anal. Lett., 36:641-658, 2003.
Y. Okamoto, Y. Nomura, H. Nakamura, K. Iwamaru, T. Fujiwara and T. Kumamaru, High preconcentration of ultra-trace metal ions by liquid-liquid extraction using water/oil/water emulsions as liquid surfactant membranes, Microchem. J., 65:341-346, 2000
M. G. Pereira and M. Z. Arruda, Trends in Preconcentration Procedures for Metal DeterminationUsing Atomic Spectrometry Techniques, Microchim. Acta 141: 115-131, 2003.
G.J.G.G. Sigit and J.P. Brunette, Liquid-liquid extraction of cadmium and cobalt with mixtures of 1-phenyl-3-methyl-4-stearoyl-5-hydroxypyrazole(HPMSP) and n-dodecylamine (DDA) in toluene, Monatsh. Chem., 129:787-797, 1998
M.A.M. da Silva, V.L.A. Frescura and A.J. Curtius, Determination of trace elements in water samples by ultrasonic nebulization inductively coupled plasma mass spectrometry after cloud point extraction, Spectrochim. Acta, B 55:803-813, 2000
K. S. Subramanian, C. L. Chakrabarti, J. E. Sueiras, and I. S. Maines, Preservation of some trace metals in samples of natural waters, Analytical Chemistry, Vol. 50, No. 3, 444-448, 1978
J.L. FERNÁNDEZ-TURIEL, J.F. LLORENS, A. ROIG, M. CARNICERO and F. VALERO, MONITORING OF DRINKING WATER TREATMENT PLANTS USING ICP-MS, Toxicological and Environmental Chemistry, 74:87-103, 2000
Z.H. Xie, F.Z. Xie, L.Q. Guo, X.C. Lin and G.N. Chen, Thioacetamide chemically immobilized on silica gel as a solid phase extractant for the extraction and preconcentration of copper(II), lead(II), and cadmium(II), J. Sep. Sci., 28:462-470, 2005
Appendix A: List of Tables
Table A1: Liquid-liquid extraction used for determination of Cd in water matrix
(See section 3)
Facchin et al. 1995
Sigit &Brunette 1998
Okamoto et al. 2000
Ansari et al. 1999
APDC: ammonium pyrrolidinedithiocarbamate; PAN: 1-(2-pyridylazo) naphthol;
HPMSP: 1-phenyl-3-methyl-4-stearoyl-5-hydroxypyrazole; DDA: n-dodecylamine;
PC-88A: 2-ethylhexyl phosphoric acid mono-2-ethylhexyl ester; GF-AAS, AAS, see Section 3
Table A2: Preconcentration by CPE for Cd determination, see Ferriera et al. 2007.
Silva et al. 2000
Chen & Teo 2001
Manzoori & Karim-Nezhad 2004
DDTP: O,O-diethyldithiophosphate; TAN: 1-(2-thiazolylazo)-2-naphthol
Table A3: A list of some sorbents and chelating agents for Cd preconcentration by using SLE.