CENTER FOR DRINKING WATER OPTIMIZATION
DRAFT
08/08/01
This SOP for the measurement of trihalomethanes (THMs) follows U.S.EPA Method
551.1.
1 It is required that Method 551.1 be studied along with this SOP prior to performance of the analysis. This SOP applies to samples of surface water, ground water, finished water and water at any intermediate drinking water treatment stage.
Trihalomethane Analyte
Chloroform
CAS No.
67-66-3
Bromodichloromethane
Bromoform
Dibromochlormethane
75-27-4
75-25-2
124-48-1
The determination of THMs in water samples is accomplished by liquid-liquid extraction and gas chromatography with micro electron capture detection. A 30 mL volume of sample is extracted with 3 mL of methyl tert-butyl ether (MtBE). The target analytes are identified and measured by capillary column gas chromatography using a micro electron capture detector (GC/

ECD). Analytes are quantified using procedural standard quantification.
Vial Preparation
Use a clean and labeled 40 mL amber EPA vial with a TFE-lined screw cap. A clean vial has been acid washed, rinsed with MQ and then baked. Add ~65 mg of ammonium chloride to the clean vial. This amount produces an ammonium chloride concentration of 1625 mg/L in the sample.
Sample Collection
Collect all samples in duplicate. Completely fill the vial with sample (headspace free) to just overflowing, but take care not to flush out the dechlorination reagent. No air bubbles should be trapped in the sample when the vial is sealed. This can be best achieved by filling the sample vial just above the top and slip the septa on from the side. Cap vials tightly.
Sample Storage
Until analysis, water samples must be stored at 4°C with minimal exposure to light and atmosphere. The sample storage area must be free of organic solvent vapors. Store
Page 1

CENTER FOR DRINKING WATER OPTIMIZATION
DRAFT extract samples in a freezer until analysis. Extract all samples within 14 days of collection and analyze within 14 days following extraction.
08/08/01
Apparatus
Sample containers/extraction vials: 40 mL amber EPA vial with a TFE-lined screw cap.
Autosampler Vials : 2.0 mL vials with screw or crimp cap and a teflon-faced seal.
Pasteur pipets : Glass, disposable.
Solvent Repipetor : up to 10.0 mL delivery range
Volumetric flasks: 2mL, 10 mL.
Micro syringes : 10 µL, 25 µL, 50 µL, 100 µL, 250 µL and 500 µL.
Transfer pipets: 10 mL pipetor with disposable tips.
Orbital Mixer
Balance: analytical, capable of weighing to 0.001 g.
Gas chromatograph : analytical system complete with gas chromatograph equipped for electron capture detection, split/splitless capillary or direct injection, temperature programming, differential flow control, and with all required accessories including syringes, analytical columns and ultra high purity nitrogen gas. A data system is recommended for measuring peak areas. An autoinjector is recommended for improved precision of analyses. The gases flowing through the electron capture detector should be vented through the laboratory fume hood system.
Primary GC column : DB-1 [fused silica capillary with chemically bonded
(Dimethlypolysiloxane)] column, 30m x 0.45
 m ID, 1.27 µm film thickness.
Reagents
Reagent water: Reagent water is defined as a water in which an interference is not observed > than the MDL of each analyte of interest. A Millipore Milli-Q water system is used to generate deionized reagent water.
Methanol: High purity, demonstrated to be free of analytes and interferences. Maintain records of the reagent’s manufacturer.
THM standards: The stock standard solution used is purchased as an ampulized solution in methanol from Supelco (Catalog # 48140-U).
Surrogate: The stock standard solution (1,2,3-trichloropropane) used is purchased as an ampulized solution in MTBE from AccuStandard Inc. (# M-552.1-IS).
Ammonium chloride, NH
4
Cl: ACS reagent grade, used to convert free chlorine to monochloramine.
Sodium chloride, NaCl – ACS reagent grade (granular, anhydrous). If interferences are observed, it may be necessary to heat the sodium chloride in a shallow tray at 400°C for up to 4 hours to remove phthalates and other interfering organic substances. Store in a capped glass bottle rather than a plastic container.
Methyl tert-butyl ether: High purity, demonstrated to be free of analytes and interferences. Maintain records of the reagent’s manufacturer.
Page 2

CENTER FOR DRINKING WATER OPTIMIZATION
DRAFT
08/08/01
Initial Demonstration of Capability (IDC)
The IDC is performed at least once, by each analyst, before analysis of any sample, to demonstrate proficiency to perform the method. The following procedures are based on
Section 6 of the DBP/ICR Analytical Methods Manual 2 .
Initial Demonstration of Low System Background: Analyze a Method Blank (MB) to verify that no contamination exists above ½ the minimum reporting levels (MRLs). The
MRL for each of the THMs is 1.0 µg/L.
Initial Demonstration of Precision: Analyze a total of five samples of reagent water fortified with each of the trihalomethanes at a concentration of 20 µg/L. These five samples must be analyzed on five separate days and must be of separate extraction and analysis batches. The relative standard deviation (RSD) must be less than 20%.
Initial Demonstration of Accuracy: Calculate the average recoveries of the replicates in the Initial Demonstration of Precision . The averages must be within +/- 20% of the theoretical amount.
Method Detection Limit (MDL) Determination
Prepare a sample with a fortifying concentration of 0.5
 g/L for each of the trihalomethanes. The following procedure is based from Standard Methods 1030 C 1 .
Analyze seven portions of this solution over a period of at least 3 d to ensure that MDL determination is more representative than measurements performed sequentially.
Calculate the standard deviation to the replicate analysis. From a table of the one-sided t distribution select the value of t for 7 – 1 = 6 degrees of freedom and at the 99% confidence level. This value is 3.14. The product of 3.14 times the standard deviation is the MDL.
Method Blank
Each time a set of samples is extracted or reagents are changed, a method blank (MB) must be analyzed. If the MB (Milli-Q) produces an interference peak within the retention time window of any analyte that would prevent the determination of that analyte or a peak of concentration greater than ½ the MRL for that analyte, the analyst must determine the source of contamination and eliminate the interference before processing samples. Field samples of an extraction set associated with an MB that has failed the specified criteria are considered suspect.
Matrix Spike
Chlorinated water supplies will usually contain significant background concentrations of several method analytes. The concentrations may be equal to or greater than the fortified concentrations. Relatively poor accuracy and precision may be anticipated when a large background must be subtracted. For many samples, the concentrations may be so high that fortification may lead to a final extract with instrumental responses exceeding the linear range of the electron capture detector. If this occurs, the extract
Page 3

CENTER FOR DRINKING WATER OPTIMIZATION
DRAFT
08/08/01 must be diluted. In spite of these problems, sample sources should be fortified and analyzed as described below. By fortifying sample matrices and calculating analyte recoveries, any matrix induced analyte bias is evaluated.
The laboratory must add known concentrations of analytes to 10 % of the samples. The concentrations should be equal to or greater than the background concentrations in the sample selected for fortification. Acceptable fortification concentrations according to the
ICR Methods Manual are 20 µg/L and 40 µg/L.
Calculate the mean percent recovery, R, of the concentration for each analyte, after correcting the total mean measured concentration, A, from the fortified sample for the back-ground concentration, B, measured in the unfortified sample, i.e.:
R = 100 (A - B)/C where C is the fortifying concentration. U.S.EPA Method 551.1 requires the recoveries of all analytes must fall within 80-120%. If a recovery falls outside of this acceptable range, a matrix induced bias can be assumed for the respective analyte and the data for that analyte must be reported to the data user as suspect.
Surrogate
The surrogate standard is fortified into the aqueous portion of all samples and laboratory reagent blanks. The surrogate is a means of assessing method performance from extraction to final chromatographic performance. U.S.EPA Method 551.1 lists acceptance criteria of 80-120%.
When surrogate recovery from a sample, blank or calibration check is outside these limits check (1) calculations to locate possible errors, (2) standard solutions for degradation, (3) possible sources for contamination, and (4) instrument performance. If those steps do not reveal the cause of the problem, reanalyze the extract. If the extract reanalysis meets the surrogate recovery criterion, report only data for the reanalyzed extract. If the extract reanalysis fails the recovery criterion, the analyst should check the calibration by analyzing the most recently acceptable calibration check standard. If the calibration check fails criteria, recalibration is in order. If the calibration check is acceptable, it may be necessary to extract another aliquot of sample. If the sample reextract also fails the recovery criterion, report all data for that sample as suspect.
Duplicate Analysis
Ten percent (10%) of samples will be analyzed in duplicate. Duplicate results must not reflect a relative percent difference (RPD) greater than 25% for any one analyte and the
RPD for 90% of the analytes being determined must be less than 20%. If this criteria is not met the analysis must be repeated.
Solvent Blank
Each analysis run must be started with an MTBE solvent blank. This is a check on the extraction solvent as well as on the instrument system.
Page 4

CENTER FOR DRINKING WATER OPTIMIZATION
DRAFT
Second Source Calibration
08/08/01
An additional calibration standard is extracted and analyzed alongside every calibration curve that is to be used. This additional calibration standard is purchased from a supplier different from the one used to purchase the standards used in preparation of the calibration curve. Percent recoveries are calculated for the second source calibration standard. These recoveries must fall within the range of 70-130 % in order for the initial calibration curve to be considered valid.
Calibration Check
If an initial calibration is not run on a particular analysis day, the calibration check must be run and matched to the previous calibration curve to ensure that the instrument is still within calibration. In order for the calibration check to be considered valid, recoveries must fall between 50% and 150% for all the target analytes when quantified using the previous calibration curve. Furthermore, a calibration check must be analyzed after every tenth sample analysis, after the final sample analysis and recoveries must fall between 80% and 120% for all the target analytes to be considered acceptable. At least one calibration check must be extracted with each set of samples. Calibration check standards need not necessarily be different extracts but can be injections from the same extract as long as the holding time requirements are met.
Preparing a Primary Standard
The primary dilution standard is prepared by diluting the ampulized stock standard solution containing 2,000,000
 g/L concentrations for each of the THM species in methanol.
1. Rinse syringe with small amount of high concentration standard.
2. Dispense 100 µL into a 10 mL volumetric flask partially filled with methanol.
Ensure the needle is submerged in methanol at bottom of flask.
3. Fill the flask with methanol to the 10 mL mark.
4. Mix by inversion.
The analyte concentration is 20,000
 g/L in the primary dilution standard. Methanol is used due to its miscibility with water.
Preparing Secondary Standards
Secondary standards are prepared by the addition of specific volumes of the primary dilution standard to separate ~72 mL volumes of reagent water. These aqueous standards are treated, including extraction, in the same manner as the samples.
Bottles do not exactly contain 72 mL (range between 72-74 ml). So to determine exact volumes, weigh the capped labeled bottle (w/ septa) before and after adding reagent water and then determine the actual contained volume by using the density of H
2
O.
Page 5

CENTER FOR DRINKING WATER OPTIMIZATION
DRAFT
08/08/01
1. Add ~109 mg of ammonium chloride to a clean, empty and labeled 72 mL bottle.
2. Weigh the capped labeled bottle (w/ septa) containing the dechlorination reagent.
3. Fill bottle with Milli-Q reagent water to top so no headspace is present. This can be best achieved by filling the sample bottle just above the top and slip the septa on from the side.
4. Cap bottles.
5. Reweigh water-filled bottle.
6. Place a
5/8” – 25 guage needle in the septa of the bottle.
7. Determine the primary standard dose:
Primary Standard to add (
 l) = Total Volume (
 l) x Desired concentration / 20,000
 g/L
The following table lists the required primary standard volumes assuming the total volume in the bottles are 72 mL.
Desired Standard
Concentration (
 g/L)
0
Standard Volume
(mL)
72
Primary Standard
Volume to add (

L)
0
Syringe to Use
N/A
0.5
5.0
10
20
72
72
72
72
1.8
18
36
72
5
25
50
100
40
80
100
72
72
72
144
288
360
250
500
500
150 72 540 1000
The 0.5
 g/l standard represents is the concentration used for determining the MDL. The
10 and 20
 g/l standards are specified by the ICR Methods Manual to be used as calibration checks
8. Inject the appropriate dose with the injection needle plunged to the bottom of the sample vial. The liquid injected will rise, due to initial density differences between methanol and H
2
O, so start the injection with the bottle inverted.
9. Then rotate the bottle during injection to ensure better mixing and to ensure that no dosing solution exits the exhaust needle. The dose volume will displace an equal amount of sample out the 5/8” needle.
Page 6

CENTER FOR DRINKING WATER OPTIMIZATION
DRAFT
08/08/01
5/8” exhaust needle
Injection syringe
Surrogate Standard Solution
The surrogate stock standard solution (1,2,3-trichloropropane) is purchased from
AccuStandard (# M-552.1-IS) at a concentration 1,000,000
 g/l. in MTBE. From this stock standard solution, a primary dilution standard is prepared in acetone at a concentration of 125,000 µg/l. A volume of 6 µL of this primary dilution standard is spiked into each blank, sample, calibration standard and QC sample, to a final concentration of 25 µg/l.
Matrix Spiking Standard Solution
A volume of 20 or 40 µL of the primary standard is used in the fortification of the duplicate sample. Consideration should be given to attempting to fortify at concentrations above those expected in the sample.
Calibration of Instrument
Each analysis run should be started with an MtBE solvent blank. This is a check on the extraction solvent as well as on the instrument system. If this run is acceptable, the extracts for the cal ibration curve are analyzed (2 µL injection volume). The
Chemstation Chromatography Software System is used to generate a calibration curve by plotting the areas (A an
) against the concentrations (C an
). The curve can be defined as either first or second order. Correlation coefficients must be greater than 0.990.
1. Remove the samples from storage and allow them to equilibrate to room temperature.
2. Remove the aqueous sample to the prescored 30 mL mark on each sample vial.
Page 7

CENTER FOR DRINKING WATER OPTIMIZATION 08/08/01
DRAFT
3. Add 6

L of the surrogate standard (125,000
 g/L 1,2,3-trichloropropane in acetone).
Mix by slowly and carefully inverting the sample vial once with minimal sample agitation.
4. Add 8.0 g of sodium chloride and 3.0 ml of MTBE (via bottle top pipetor). Sodium chloride is added to increase the ionic strength of the aqueous phase and thus further drive the trihalomethanes into the organic phase.
5. Recap vial and mix for ~30 seconds on an orbital mixer (until most of the salts are dissolved).
6. Let stand for ~ 15 minutes for separation.
7. Extract ~2 ml of the top organic layer with a pasteur pipette.
Organic Layer
8. Transfer extracted sample to GC vial and cap.
9. Analyze samples as soon as possible.
Remove 10 ml
6

L
Surrogate
8.0 g NaCl 3 ml MtBE
1) Cap vial
2) Mix 30 sec
3) Stand 15 min
~2 ml organic layer
~2 ml organic layer to GC vial
Gas Chromatography
1. The temperature program utilized is as follows: 35°C initial temperature, 4 minute hold time, 30°C/minute ramp, 180°C final temperature, 0 minute hold time. This represents a 8.83 minute run time.
2. Calibrate the system daily by either the analysis of a calibration curve or a continuing calibration check.
3.
Inject 2 µL of the sample extract. Use the Chemstation chromatography system to record the resulting peak sizes in area units.
Page 8

CENTER FOR DRINKING WATER OPTIMIZATION
DRAFT
08/08/01
4. If the response for the peak exceeds the calibration curve or the working range of the detector, dilute the extract and re-analyze. The analyst must not extrapolate beyond the calibration range established.
Identify sample components by comparison of retention times to retention data from the calibration standard analysis. If the retention time of an unknown peak corresponds, within limits, to the retention time of a standard compound, then the identification is considered positive. Calculate analyte concentrations from the calibration curve.
Retention time windows are given as +/- three times the standard deviation obtained in calculating the average retention time of the initial calibration curve. However, the experience of the analyst should weigh heavily in the interpretation of the chromatogram.
Gas purge flow should be left on at all times, with or without GC use to prevent the inflow of oxygen and resulting contamination/oxidation of parts. The flow can be lowered though in times of dormancy. The oven can be shut off if the GC is not used for an extended period of time (i.e. months).
Short term Maintenance (Every time)
Compressed ultra high purity nitrogen gas tanks are checked daily for content (pounds per square inch-psi) and are replaced when the contents fall below 500 psig. Every time a gas tank is changed out, this should be noted on the oxygen trap, which has a limited capacity.
Long term Maintenance (Months)
Glass liner : The glass liner, located in the injection inlet, should be replaced ~ once a year or when used up. When it is used the glass wool inside it will have a darker color
Rubber injection seal : The rubber seal, located at the top of the injection inlet, should be replaced every 3-4 months
Gold seal : The gold seal, located at the bottom of the injection inlet under the glass liner, should be checked ~ every 6 months for carbon build-up. It is accessed from inside the oven.
Column Burn : The column should be burned off every 3-12 months, depending on use.
This elutes off any constituents that may be more difficult to remove. A chromatogram is produced showing eluted constituents. A 5-7 hour burn is sufficient. The burn temp should be ~50 o C less than the max column temperature reported on the box.
Very Long term Maintenance (Years)
Moisture Trap : will need regeneration when exhausted
Oxygen Trap : the high capacity (2 L of oxygen) trap will need replacement after ~18 tanks of gas used
ECD Detector : will need servicing every 3-10 or so years.
Page 9

CENTER FOR DRINKING WATER OPTIMIZATION
DRAFT
Gas flow rates : should be checked annually.
Specific maintenance on an improperly functioning part is recorded on the GC
Maintenance Form.
08/08/01
1. The toxicity or carcinogenicity of each reagent used in this method has not been precisely defined, however, each chemical compound must be treated as a potential health hazard. From this viewpoint, exposure to these chemicals must be minimized.
The laboratory is responsible for maintaining a current awareness file of OSHA regulations regarding the safe handling of the chemicals specified in this method. A reference file of material safety data sheets should also be made available to all personnel involved in the chemical analysis.
2. The toxicity of the extraction solvent, MtBE, has not been well defined. Susceptible individuals may experience adverse affects upon skin contact or inhalation of vapors.
Therefore protective clothing and gloves should be used and MtBE should be used only in a chemical fume hood or glove box. The same precaution applies to pure standard materials.
1. Method interferences may be caused by contaminants in solvents, reagents, glassware and other sample processing apparatus that lead to discrete artifacts or elevated baselines in chromatograms. All reagents and apparatus must be routinely demonstrated to be free from interferences under the conditions of the analysis by analyzing method blanks. Subtracting blank values from sample results is not permitted.
2. The use of high purity reagents and solvents helps to minimize interference problems. Solvent blanks should be analyzed for each new bottle of solvent before use.
An interference free solvent is a solvent containing no peaks yielding data at > MDL and at the retention times of the analytes of interest.
3. Interfering contamination may occur when a sample containing low concentrations of analytes is analyzed immediately following a sample containing relatively high concentrations of analytes. Routine between-sample rinsing of the sample syringe and associated equipment with MtBE can minimize sample cross-contamination. After analysis of a sample containing high concentrations of analytes, one or more injections of MtBE should be made to ensure that accurate values are obtained for the next sample.
4. Matrix interferences may be caused by contaminants that are co-extracted from the sample. The extent of matrix interferences will vary considerably from source to source, depending upon the water sampled.
Page 10

CENTER FOR DRINKING WATER OPTIMIZATION
DRAFT
08/08/01
1. SURROGATE ANALYTE -- A compound which is added to a sample aliquot in known amount(s) before extraction or other processing and is measured with the same procedure used to measure other sample components. The purpose of the surrogate is to monitor method performance with each sample.
2. LABORATORY DUPLICATES -- Two aliquots of the same sample designated as such in the laboratory. Each aliquot is extracted, derivatized and analyzed separately using identical procedures. Analysis of these duplicates indicates the precision associated with laboratory procedures, but not with sample collection, preservation or storage procedures.
3. FIELD DUPLICATES -- Two separate samples collected at the same time and place under identical circumstances and treated exactly the same throughout field and laboratory procedures. Analysis of these duplicates gives a measure of the precision associated with sample collection, preservation and storage, as well as with laboratory procedures.
4. METHOD BLANK (MB) -- An aliquot of reagent water or other blank matrix that is treated exactly as a sample including exposure to all glassware, equipment, solvents, reagents, internal standards, and surrogates that are used with other samples. The MB is used to determine if method analytes or other interferences are present in the laboratory environment, the reagents, or the apparatus.
5. MATRIX SPIKE (MS) -- An aliquot of an environmental sample to which known quantities of the method analytes are added in the laboratory. The MS is analyzed exactly like a sample, and its purpose is to determine whether the sample matrix contributes bias to the analytical results. The background concentrations of the analytes in the sample matrix must be determined in a separate aliquot and the measured values in the MS corrected for background concentrations.
6. STOCK STANDARD
– Neat material or a commercially purchased concentrated solution used in the calibration process
7. PRIMARY DILUTION STANDARD -- A solution of the target analyte(s) prepared in the laboratory from stock standard solutions and diluted as needed to prepare secondary standards.
8. SECONDARY STANDARD -- A solution prepared from the primary dilution standard and used to calibrate the instrument response with respect to analyte concentration.
9. SECOND SOURCE STANDARD
– A solution of method analytes of known concentration which is used to fortify an aliquot of reagent water. It is obtained from a source different from that of the standards used to calibrate and is used as a check on the accuracy of standards being used to calibrate the analytical instrumentation.
Page 11

CENTER FOR DRINKING WATER OPTIMIZATION
DRAFT
08/08/01
10. METHOD DETECTION LIMIT (MDL) -- The minimum concentration of an analyte that can be identified, measured and reported with 99% confidence that the analyte concentration is greater than zero.
11. MATERIAL SAFETY DATA SHEET (MSDS) -- Written information provided by vendors concerning a chemical's toxicity, health hazards, physical properties, fire and reactivity data including storage, spill, and handling precautions.
XI INSTRUMENT SOFTWARE AND OPERATION
The Agilent GC Chemstation software (Rev. A.08.03 [847]) is used for data acquisition.
Method files are named as follows: THM MMDD .M
Sequence files are named as follows: MMDD T.S
Data files are named as follows: MMDD T00X.D where MM = Month, DD = Day, X represents consecutive numbering
1 APHA, AWWA and WEF ( 1998 ) Standard Methods for the Examination of Water and
Wastewater 20 th Edition , Washington, D.C.
2 USEPA ( 1996 ) DBP/ICR Analytical Methods Manual , Office of Water, Cincinnati, OH.
3 USEPA ( 1995 ) Method 551.1: Determination of chlorination disinfection byproducts, chlorinated solvents, and halogenated pesticides/herbicides in drinking water by liquidliquid extraction and gas chromatography with electron-capture detection. In Methods for the Determination of Organic Compounds in Drinking Water-Supplement III , EPA/600/R-
95/131. Cincinnati, Ohio: National Exposure Research Laboratory, USEPA Office of
Research and Development.
Page 12