Introduction tc Instrumental Analysis of Water Pollutants. Training ... · sample mixture with an immiscible solvent in which the constituent is much more soluble. C Filtration -

ED 209 110

TITLE

INSTITUTION

FEPORT NOPUB DATENOT!AVAILABLE FROM

DOCURENT RESUME

SE 035 919

Introduction tc Instrumental Analysis of WaterPollutants. Training Manual.Office of Water Progrue Operations (EPA), Cincinnati,Ohio. National Training and Operational TechnologyCenter.EPA-430/1-79-004May 79200p.EPA Instructional Resources Center, 1200 ChambersRd.,, 3rd Floor, Columbus, OH 43212 ($1.00 plus $0.03per page).

EDRS PRICE MF01/PCO8 Plus Postage.DESCRIPTORS *Chemistry; Chromatography: *Instrumentation:

*Laboratory Equipment: *Laboratory Procedures;Postsecondary Education: Science Education;Spectroscopr; *Water Pollution

IDENTIFIERS Analytical Chemistry: AnalyUcal Methods; WaterAnalysis: *Water Quality

ABSTRACTThis course is designed for those requiring an

intrciuction tc instruments commonly used in water pollutionanalyses. Examples are: pH, conductivity, dissolved oxygen meters,spectrophotometers, turbidimeters, carbon analyzer, and gaschromatography. Students should have a basic knowledge of analyticalchemistry. (CO)

*********************************A*************************************Reproductions supplied by EDRS are the best that can be made

from the original document.***********************************************************************

United States Tonal TrainingEnyironrrientel Protectiork d OperationalAgency Technology Center

Cincinnati OH 45268

Water.inmr.m..

EPA Introduction toInstrumentalAnalysis of Water

CI 4Pollutants

EPA-430 1-79-004May 1979

("\n,C.3

sr Training ManualU S DEPARTMENT OF EDUCATION

NATIONAL INSTITUTE OF EDUCATION

EDUCA TONAL RESOURCES INFORMATION

CENTER IERICXT his document has been reproduced asreceived from the person or organizationoriginating itMinor changes have Peen made to tovoye

rep,oduction guelity

Points Of view or opinions staled in this dour

rnent do nit necesSanly represent ogioal N1E

Oie.stio0 ONICV

--

EPA/ 1 -79 -004May 1979

INTRODUCTION TO INSTRUMENTAL ANALYSIS OF WATER POLLUTANTS

This course is designed for those requiring anintroduction to instruments commonly used inwater pollution analyses. Examples are pH,conductivity, and dissolved oxygen meters,spectrophotometers (infrared, atomic absorption,flame photometry and colorimetric measurements),the turbidizneter, carbon analyzer, and gaschromatography. Course applicants should havea basic knowledge about analytical chemistry.

At the conclusion of this course the student willbe familiar with the various instruments employedto analyze water pollutants in water.

U. S. ENVIRONMENTAL PROTECTION AGENCYOffice of Water Program Operations

National Training and Operational Technology Center

3

FOREWORD

These manuals are prepared for reference use of students enrolled in scheduledtraining courses of the Office of Water Program Operations, U. S. EnvironmentalProtection Agency.

Due to the limited availability of the manuals it is not appropriateto cite them as technical references in bibliographies or otherforms of publication.

References to products and manufacturers are for illustration only;such references do not imply product endorsement by the Office ofWater Program Operations, U. S. Environmental Protection Agency.

The reference outlines in this manual have been selected and developed with agoal of providing the student with a fund of the best available virreut informationpertinent to the subject matter of-the course. Individual instructors may provideadditional material to cover special aspects of their own presentations.

This manual will be useful to anyone who has need for information on the subjectscovered. However, it should be understood that the manual will have its greatestvalue as an adjunct to classroom presentations. The inherent advantages of class-room presentation is in the give-and-take discussions and exchange of informationbetween and among students and the instructional staff.

Constructive suggestions for improvement in the coverage, content, and formatof the manual are solicited and will be given full consideration.

4`Y

I

I

I

I

I

I

I

1

I

CONTENTS

Title or Description Outline Number

Methodology for Chemical Analysis of Water and Wastewater 1

2

Inter laboratory Quality Control Studies 3

Specific COnductance 4

Calibration and Use of a Conductivity Meter 5

Dissolved Oxygen - Factors Affecting DO Concentration in Water 6

Statistics for Chemists

Dissolved Oxygen Determination by the Winkler Iodometric Titration- Azide Modification 7

Dissolved Oxygen Determination by Electronic Measurement 8

Laboratory Procedure for Dissolved Oxygen Winkler Method-AzideModification 9

Principles of Absorption Spectroscopy 10

Use of a Spectrophotometer 11

Atomic Absorption Spectrophotometry 12

Atomic Absorption Laboratory (Zinc) 13

Atomic Absorption Laboratory (Copper) 14

Principles of Emission Spectroscopy 15

Flame Photometry 16

Flame Photometry Laboratory (Sodium) 17

Flame Photometry Laboratory (Strontium) 18

Automation of Chemical Analysis 19

Turbidity 20

Calibration and Use of a Turbidimeter (Nephelometer) 21

Trace Organic Contaminants in Water 22

Methods of Recovering Organic Materials from Surface Waters 23

Preliminary Separation of Extracts 24

Laboratory Procedures for the Preliminary Separation of Extracts 25

Thin-Layer Chromatography 26

Title or Description Outline Numbers

Introduction to Gas-Liquid Chromatography 27

Infrared Instrumentation 28

Infrared - Procedure for Use of Infracord 29

Total Carbon Analysis 30

Special Instrumental Methods - Mass Spectrometry and NMR 31

METHODOLOGY FOR CHEMICAL ANALYSIS OF WATER AND WASTEWATER

I INTRODUCTION

This outline deals with chemical methods whichare commonly performed in water qualitylaboratories. Although a large number ofconstituents or properties may be of interestto the analyst, many of the methods employedto measure them are based on the sameanalytical principles. The purpose of thisoutline is to acquaint you with the principlesinvolved in commonly-used chemical methodsto determine water quality.

II PRE-TREATMENTS

For some parameters, a preliminary treatmentis required before the analysis begins. Thesetreatments serve various purposes.

A Distillation - To isolate the constituent byheating a portion of the sample mixture toseparate the more volatile part(s), and thencooling and condensing the resulting vapor(s)'to recover the volatilized portion.

B Extraction - To isolate/concentrate theconstituent by shaking a portion of thesample mixture with an immiscible solventin which the constituent is much moresoluble.

C Filtration - To separate undissolved matterfrom a sample mixture by passing a portionof it through a filter of specified size.Particles that are dissolved in the originalmixture are so small that they stay in thesample solution and pass through the filter.

D Digestion - To change constituents to a formamenable to the specified test by heating aportion of the sample mixture with chemicals.

III METERS

For some parameters, meters have beendesigned to measure that specific constituentor property.

CH. 14b. 11.77

A pH Meters

pH (hydrogen ion concentration) is meas-ured as a difference in potential across aglass membrane which is in contact withthe sample and with a reference solution.The sensor, apparatus might be combinedinto one probe or else it is divided into anindicating electrode (for the sample) and areference electrode (for the referencesolution). Before using, the meter mustbe calibrated with a solution of known pH(a buffer) and then checked for pro, .troperation with a buffer of a different pHvalue.

B Dissolved Oxygen Meters

Dissolved oxygen meters measure theproduction of a current which is proportionalto the amount of oxygen gas reduced at acathode in the apparatus. The oxygen gasenters the electrode through a membrane,and an electrolyte solution or gel acts as atransfer and reaction media. Prior to usethe meter must be calibrated against a knownoxygen gas concentration.

C Conductivity Meters

Specific conductance is measured with ameter containing a Wheatstone bridge whichmeasures the resistance of the samplesolution to the transmission of an electriccurrent. The meter and cell are calibratedaccording to the conductance of a standardsolution of potassium chloride at 25°C.measured by a "standard" cell with electrodesone cm square spiced one cm apart. Thisis why results are called "specific" con-ductance.

D Turbidimeters

A turbidimeter compares the intensity oflight scattered by particles in the sampleunder defined conditions with the intensityof light scattered by a standard referencesuspension.

1-1

Methodology !or Chemical Aanl hi:. ,,t Wane: al d Waste ater

I'/ SPECIFIC ION ELEC rHODES

Just as the conventional glass electrodefor pH develops an electrical potential inresponse to the activity of hydrogen ionin solution, the specific ion electrodedevelops an electrical potential in responseto the activity of the ion for which the electrodeis specific. The potential and activity arerelated according to the Nernst equati on.Simple analytical techniques can be appliedto convert activity to an expression of con-centration.

These electrodes are used with a pH meterwith an expanded mV scale or with a specificion meter. Two examples are the ammoniaand fluoride electrodes.

A Ammonia

The ammonia electrode uses a hydrophobicgas-permeable membrane to separate thesample solution from an ammonium chlorideinternal solution. Ammonia in the samplediffuses through the membrane and altersthe pH of the internal solution, which issensed by a pH electrode. The constantlevel of chloride in the internal solution issensed by a chloride selective ion electrodewhich acts as the reference electrode.

B Fluoride

The fluoride electrode consists of a lanthanumfluoride crystal across which a potential isdeveloped by fluoride ions. The cell may berepresented by Ag/Ag Cl. Cl (0. 3),F (0.001)LaF/test solution/SCE/. It is used in con-junction with a standard single junctionreference electrode.

V GENERAL ANALYTICAL METHODS

A Volumetric Analysis

Titrations involve using a buret to measurethe volume of a standard solution of a sub-stance required to completely react withthe constituent of interest in a measuredvolume of sample. One can then calculatethe original concentration of the constituentof interest.There are various ways t--) detect the endpoint when the reaction is ompliste.

1 -2

1 Colt), change Inds( at ), s

The method may utilize an indicator whichchanges color when the reaction iscomplete. For example, in the ChemicalOxygen Demand Test the indicator,ferroin, gives a blue-green color to themixture until the oxidation-reductionreaction is complete. Then the mixtureis reddish-brown.

Several of these color-change titrationsmake use of the iodometric processwhereby the constituent of interest quan-titatively releases free iodine. Starchis added to give a blue color until enoughreducing agent (sodium thiosulfate orphenylarsine oxide) is added to reactwith all the iodine. At this end point.the mixture becomes colorless.

2 Electrical property indicators

Another way to detect end points is ach.nge in an electrical property of thesolution when the reaction is complete.In the chlorine titration a cell containingpotassium chloride will produce a smalldirect current as long as free chlorineis present. As a reducing agent (phen-ylarsine oxide) is added to reducethe chlorine, the microammeter whichmeasures the existing direct currentregisters a lower reading on a scale.By observing the scale, the end point oftotal reduction of chlorine can bedetermined because the direct currentceases.

3 Specified end points

For acidity and alkalinity titrations, theend points are specified pH values forthe final mixture. The pH values arethose existing when common acidity oralkalinity components have been neutral-ized. Thus acidity is determined bytitrating the sample with a standardalkali to pH 8.2 when carbonic acidwould be neutralized to (CO3)- . Alka-linity (except for highly acidic samples)is determined by titrating the samplewith a standard acid to pH 4.5 when thecarbonate pi Sent has been convertedto carbonic acid. pH meters are used todetect the specified end points.

8

Methodolo :4.16.for Chemical Analysis of Water and Wastewater

B Gravimetric Prot .:,lures

Gravimetric methods involve directweighing of the constituent in a container.An empty container is weighed, theconstituent is separated from the samplemixture and isolated in the container, thenthe container with the constituent is weighed.The difference in the weights of the containerbefore and after containing the constituentrepresents the weight of the constituent.

The type of container depends on the methodused to separate the constituent from thesample mixture. In the solids determinations,the container is an evaporating dish (total ordissolved) or a glass fiber filter disc in acrucible (suspended). For oil and grease,the container is a flask containing a residueafter evaporation of a solvent.

C Combustion

Combustion means to add oxygen. In theTotal Organic Carbon Analysis, combustionis used within an instrument to convertcarbonaceous material to carbon dioxide.An infrared analyzer measures the carbondioxide.

VI PHOTOMETRIC METHODS

These methods involve the measurement of lightthat is absorbed or transmitted quantitativelyeither by the constituent of interest or else bya substance containing the constituent of interestwhich 'ias resulted from some treatment ofthe sample. The quantitative aspect of thesephotometric methods is based on applying theLambert-Beer Law which established that theamount of light absorbed is quantitativelyrelated to the concentration of the absorbingmedium at a given wavelength and a giventhicicness of the medium through which thelight passes.

Each method requires preparing a set ofstandard solutions containing known amountsof the constituent of interest. Photometricvalues are obtained for the standards. Theseare used to draw a calibration (standard) curveby plotting photometric values against theconcentrations. Then. by locating the photo-metric value for the sample on this standardcurve, the unknown concentration in thesample can be determined.

A Atomic .-%.t)."(wprion

Atomic Absorption (AA) instruments utilizeabsorption of light of a characteristic wave-length. This form of analysis involvesaspirating solutions of metal ions (cations)or molecules containing metals into aflame where they are reduced to individualatoms in a ground electrical state. In thiscondition, the atoms can absorb radiationof a wavelength characteristic for eachelement. A lamp containing the element ofinterest as tht. cathode is used as a sourceto emit the characteristic line spectrum forthe element to be determined.

The amount of energy absorbed is directlyrelated to the concentration of the elementof interest. Thus the Lambert-Beer Lawapplies. Standards can be prepared andtested and the resulting absorbance valuescan be used to construct a calibration(standard) curve. Then the absorbancevalue for the sample is located on this curveLo determine the corresponding concentration.

Once the instrument is adjusted to giveoptimum readings for the element of interest,the testing of each solution can be done ina matter of seconds. Many laboratorieswire recorders into their instruments torapidly transcribe the data, thus conservingtime spent on this aspcct of the analysis.Atomic absorption techniques are generallyused for metals and semi-metals in solutionor else solubilized through some form ofsample processing. For mercury, theprinciple is utilized but the absorption oflight occurs in a flameless situation withthe mercury in the vapor state and containedin a closed glass cell.

B Flame Emission

Flame emission photometry involvesmeasuring the amount of light given off byatoms drawn into a flame. At certaintemperatures, the flame raises the electronsin atoms to a higher energy level. Whenthe electrons fall back to a lower energylevel, the atoms lose (emit) radiant energywhich can be detected and measured.

Again standards must be prepared andtested to prepare a calibration (standard)curve. Then the tranimission value of thesample can he located on the curve .0determine its concentration.Many atomic absorption instruments can heused for flame emission photometry.Sodium and potassium are very effectivelydetermined by the emission technique.

1-3

Niethod)tTv ray Chemical _Analysis of Water and WastewaterHowever, tor many element-, absorptionanalysis is more sensitive beLause there are

great number of unexcited atoms in theflame which are available to absorb theradiant energy.

C Colorimetry

Colorimetric analyses involve treatingstandards which contain known Concentrationsof the constituent of interest and also thesample with reagents to produce a coloredsolution. The greater the concentration ofthe constituent, the more intense will bethe resulting color.

The Lambert-Beer Law which relates theabsorption of light to the thiclmess andconcentration of the absorbing mediumapplies. Accordingly, a spectrophotometeris used to measure the amount of light ofappropriate wavelength which is absorbedby the same thickness of each solution.The results from the standards are used toconstruct a calibration (standard) curve.Then the absorbance value for the sampleis located on this curve to determine thecorresponding concentration.

Many of the metals and several otherparameters (phosphorus. ammonia, nitrate,nitrite, etc.) are determined in thismanner.

VII GAS-LIQUID CHROMATOGRAPHY

Chromatography techniques involve a separa-tion of the components in a mixture by usinga difference in the physical properties of thecomponents. Gas-Liquid Chromatography(GLC) involves separation based on a differ-ence in the properties of volatility and solu-bility. The method is used to determinealgicides, chlorinated organic compoundsand pesticides.

The sample is introduced into an injectorblock which is at a high temperature (e. g.210°C), causing the liquid sample to volatilize.An inert carrier gas transports the samplecomponents through a liquid held in pl-ice asa thin film on an inert solid support materialin a column.Sample components pass through the columnat a speed partly governed by the relativesolubility of each in the stationary liquid.Thus the least soluble components are thefirst to reach the detector. The type ofdetector used depends on the class of compoundsinvolved. All detectors function to sense andmeasure the quantity of each sample componentas it comes off the column. The detectorsignals a recorder system which registersa response.1-4

As with other instrunItial methods, standardswith known concentrations of the substance ofinterest are measured on the instrument. Acalibration (standard) curve can be developedand the concentration in a sample can bedetermined from this graph.

Gas-liquid chromatography methods are verysensitive (nomogram, picogram quantities) soonly small amounts of samples are required.On the other hand, this extreme sensitivityoften necessitates extensive clean-up ofsamples prior to GLC analysis.VIII AUTOMATED METHODS

The increasing number of samples andmeasurements to be made in water qualitylaboratories has stimulated efforts to automatethese analyses. Using smaller amounts ofsample (semi-micro techniques), combiningreagents for fewer measurements per analysis,and using automatic dispensers are all meansof saving analytical time.However, the term "automated laboratoryprocedures" usually means automatic intro-duction of the sample into the instrument,automatic treatment of the sample to test fora component of interest, automatic recordingof data and, increasingly, automatic calculatingand print -out of data. Maximum automationsystems involve continuous sampling directfrom the source (e. g. an in-place probe) withtelemetering of results to a central computer.Automated methods, especially those based oncolorimetric methodology, are recognized forseveral water quality parameters includingalkalinity, ammonia, nitrate, nitrite, phosphorus,and hardness.

IX SOURCES OF PROCEDURES

Details of the procedure for an individualmeasurement can be found in reference books.There are three particularly-recognized booksof procedures for water quality measurements.

A Standard Methods(1)The American Public Health Association,the American Water Works Associationand the Water Pollution Control Federationprepare and publish "Standard Methods forthe Examination of Water and Wastewater."As indicated by the list of publishers. thisbook contains methods developed for use bythose interested in water or wastewatertreatment.

B ASTM Standards(2)

The American Society for Testing andMaterials publishes an "Annual Book ofASTM Standards" containing specificationsand methods for testing materials. The"book" currently consists of 47 parts.

1 ()

Methodolo

The part applicable to water was formerlyPart 23.. It is now Part 31) Water.

The methods are chosen by approval of themembership of ASTM and are intended toaid industry, government agencies and thegeneral public. Methods are applicable toindustrial waste waters as well as to othertypes of water samples.

C EPA Methods Manual(3)

The United States Environmental ProtectionAgency publishes a manual of "Methods forChemical Analysis of Water and Wastes."

EPA developed this manual to providemethodology for monitoring the quality ofour Nation's waters and to determine theimpact of waste discharges. The test pro-cedures were carefully selected to meetthese needs, using Standard Methods andASTM as basic references. In many cases.the EPA manual contains completelydescribed procedures because they modifiedmethods from the basic references. Other-wise, the manual cites page numbers inthe two references where the analyticalprocedures can be found.

X ACCURACY AND PRECISION

A Of the Method

One of the criteria for choosing methodsto be used for water quality analysis is thatthe method should measure the desiredproperty or constituent with precision,accuracy, and specificity sufficient to meetdata needs. Standard references, then,include a statement of the precision andaccuracy for the method which is obtainedwhen (usually) several analysts in differentlaboratories used the particular method.

B Of the Analyst

Each analyst should check his own precisionand accuracy as a test of his skill in per-forming a test. According to the U. S. EPAHandbook for Analytical Quality Control(4),he can do this in the following manner.

for Chemical Analysis of Water and Wastewater

I .4.

To check precision, the analyst shouldanalyze samples with four differentconcentrations of the constituent of interest,seven times each. The study should coverat least two hours of normal laboratoryoperations to allow changes in conditionsto affect the results. Then lie shouldcalculate the standard deviation of each ofthe sets of seven results and compare hisvalues for the lowest and highest concen-trations tested with the standard deviationvalue published for that method in the referencebook. An individual should have better valuesthan those averaged from the work of severalanalysts.

To check accuracy, he can use two of thesamples used to check precision by addinga known amount (spike) of the particularconstituent in quantities to double the lowestconcentration used, and to bring an inter-mediate concentration to approximately 75%of the upper limit of application of themethod. He then analyzes each of the spikedsamples seven times. then calculates theaverage of each set of seven results. Tocalculate accuracy in terms of % recovery.he will also need to calculate the average ofthe results he got when he analyzed theunspikol samples. Then:

Avg. of Spiked% Recovery = Avg. of Amt. of X 100

Unspiked Spike

Again, the individual's % recovery should bebetter than the published figure derived fromthe results of several analysts.

C Of Daily Performance

Even after an analyst has demonstrated hispersonal skill in performing the analysis,a daily check on precision and accuracyshould be done. About one in every tensamples should be a duplicate to checkprecision and about one in every ten samplesshould be spiked to check accuracy.

It is also beneficial to participate in inter-laboratory quality control programs. TheU. S. EPA provides reference samples atno charge to laboratories. These samples

1-5

Nlethodolosy for Chemical Analysts of Water and Waste%\atcr

serve as independent checks on reagents,instruments or techniques; for traininganalysts or for comparative analyses withinthe laboratory. There is no certificationor other formal evaluative function resultingfrom their use.

XI SELECTION OF ANALYTICALPROCEDURES

Standard sources"' 2. 3) will. for mostparameters. contain more than one analyticalprocedure. Selection of the procedure to beused in a specific instance involves 2onsider-ahon of the use to be made of the data. Insome cases, one must use specified procedures.In others, one may be able to choose amongseveral methods.

A NPDES Permits and State Certifications

A specified analytical procedure must beused when a waste conmtituent is measured:

1 For an application for a National PollutantDischarge Elimination System (NPDES)permit under Section 402 of the FederalWater Pollution Control Act (FWPCA),as amended.

2 For reports required to be submitted bydischargers under NPDES.

3 For certifications issued by Statespursuant to Section 401 of the FWPCA,as amended.

Analytical procedures to be used in thesesituations must conform to those specifiedin Title 40, Chapter 1, Part 133, of theCode cf Federal Regulations (CFR). Thelistings in the CFR usually cite two differentprocedures for a particular measurement.

The CFR also provides a system ofapplying to EPA for permission touse methods not cited in the CFR.Approval of alternative methods fornationwide use will be published inthe Federal Register.

1-6

B Ambient Water Quality Monitoring

For Ambient Water Quality Monitoring,analytical procedures have not beenspecified by regulations. However, theselection of procedures to be used shouldreceive attention. Use of those listed inthe CFR is strongly recommended. Ifany of the data obtained is going to be usedin connection with NPDES permits, or maybe used as evidence m a legal proceeding,use of procedures listed in the CFR isagain strongly recommended.

C Drinking Water Monitoring

In December, 1975, National InterimPrimary Drinking Water Regulationsto be effective June 24, 1977 werepublished in the Federal Register inTitle 40, Chapter 1, Subchapter D.Part 141. The publication .ncludesspecification of analytical proceduresto be used when determining complianc,with the maximum contaminant levelsof required parameters.

Tiecause of the low concentrations in-volved in the regulations, there is oftenjust one analytical method cited foreach parameter.

Individuals or organizations may applyto EPA fc.. permission to use methodsnot cited in the above. Approval ofalternative methods for nationwidf: usewill be published in the Federal Register.

\H FIELD KITS

Field kits have been devised to performanalyses outside of the laboratory. The kitmay contain equipment and reagents for onlyone test or for a variety of measurements.It may be purchased or put together by anagency to serve its particular needs.

Since such kits are devised for performingtests with minimum time and maximumsimplicity, the types of labware and reagentsemployed usually differ significantly from theequipment and supplies used to perform thesame measurement in a laboratory.

12

Methodolo

A Shortcomings

for Chemical Analysis of Water and Wastewater

Field conditions do not accommodate theequipment and services required for pre-treatments like distillation and digestion.Nor is it practical to carry and use calibratedglassware like burets and volumetric pipets.Other problems are preparation, transportand storage of high quality reagents, ofextra supplies required to test for and removesample interferences before making themeasurement, and of instruments whichare very sensitive in detecting particularconstituents. One just cannot carry andset up laboratory facilities in the field whichare equivalent to stationary analyticalfacilities.

B Uses

Even though the results of field tests areusually not as accurate and precise as thoseperformed in the laboratory, such tests dohave a place in water quality programs.

In situations where only an estimate of theconcentrations of various constituents isrequired, field tests serve well. They areinvaluable sources of information forplanning a full-scale sampling/testingprogram when decisions must be maderegarding location of sampling sites,schedule of sample collection, dilution ofsamples required for analysis, and treat-ment of samples required to remove inter-ferences t.) analyses.

C NPDES Permits and State Certification

Kit methods are not approved for obtainingdata required for NPDES permits or Stateconstruction certifications. If one judgesthat such a method is justifiable for thesepurposes, he must apply to EPA for per-mission to use it.

D Drinking Water Monitoring

The DPD test kit for residual chlorine isapproved in the December, 1975 FederalRegister for monitoring drinking water.

REFERENCES

1 Standard Methods for the Examination ofWater and Wastewater, 14th Edition.1976, APHA-AWIAA-WPCF, 1015 18th Street,N. W. , Washington, D. C. 20038

2 1975 Ann al Book of ASTM Standards,Part 31, Water. ASTM, 1916 Race StreetPhiladelphia, PA 19103.

3 Methods for Chemical Analysis of Waterand Wastes. 1974, U. S. EPA, EMSL,Cincinnati, OH 45268.

4 Analytical Quality Control in Water andWastewater Laboratories. 1972. U. S.EPA, EMSL, Cincinnati, Clio 45268.

This outline was prepared by A. D. Kroner,Chemist, National Training and OperationalTechnology t2enter, MOTD, OWPO, USEPACincinnati, Ohio 45268.

Descriptors: Analysis, Chemical Analysis,Methodology, Wastewater, Water Analysis

1-7

STATISTICS FOR CHEMISTS

I INTRODUCTION

A Statistics may be deftned, for our purpose,as a collection of methods which have beendeveloped for handling numerical datapertaining to samples or portions of entirepopulations.

13 The statistical methods with which we willconcern ourselves deal with the presentationand analysis of numerical data from samples.

U FREQUENCY

A Definitions

1 Frequency - indicates how many timesa particular score occurs in a collectionof data

S

>,Uca)Sc.a)s4

Ir.

ST. 25b. 11. 77

5

4

3 .1

2 a

1

#

2 Frequency table - a tabular arrange-ment of data, ranked in ascending ordescending order of magnitude,together with the correspondingfrequencies

3 Frequency histogram - a set ofrectangles having bases on a horizontalaxis with centers at the given scoresand heights equal to the correspondingfrequencies (See Figure 1)

4 Frequency polygon - a line graph offrequencies plotted against scores 6 A(can be obtained by connecting mid-points of tops of rectangles in thefrequency histogram) (See Figure 1)

Figure 1

Frequency Histogram & Frequency Polygon

I I

98 99I I I

100 101 102

Chloride pg/1

1 4 2-1x

Statistics For Chemists

13 Application

Consider the application of the abovedefinitions to the following set of data,obtained from twelve determinations forchloride in water.

Results (µg/ 1)

number of observations the median isXn + Xn

2- +

2

middle two scores.

, the average of the

4 Mean - arithmetic average of all thevalues in the sample distribution, de-not2d by X. The formula for calcula-

100

101

99

98

101

100

102

101

99

100

100

102

ting the sample mean is

X= Xi + X2 + Xn

nZ Xi

i=1X =

Table 1

Frequency Table

Chloride ( /1) Frequency

98 1

99 2

100 4

101 3

102 2

III MEASURES OF CENTRAL TENDENCY

A Definitions

1 Central tendency - the tendency ofvalues to cluster about a particularvalue in the distribution

2 Mode - that value which occurs mostfrequently

3 Median - midpoint of an array ofscores. If there i3 an odd number ofobservations, n, the median is

Xn +1 Xn. + 1where represents2 2

n+ 1the2

value in the frequency

distribution. If there is an even

2-2

n

E-Zi where there are n numberof values.

B Aids in calculation of the mean

Application of the following two statementscan reduce errors and amount of timespent in calculating the mean of adistribution.

1 Adding or subtracting a constant to orfrom each score in a distribution isequivalent to adding or subtracting thesame constant to or from the mean ofthe distribution. Thus the followingformula:

iCc = X + C where the Xi's are thevalues in the distribution with mean X,and the Xi f C's are the values in thedistribution with mean Xc.

2 Multiplying or dividing each score ina distribution by a constant is equivalentto multiplying or dividing the mean ofthe distribution by the same constant.Thus the following formulas:

(1)Xc =CX

Or

(2) Xc cX where t= whhe Xi ' s are the

values in the distribution with mean X,


and the eXi's or the -x" 's are theC

.y.alues in the distribution with meanxc

C Application

Consider the application of the abovedefinitions to the previously mentionedset of data, obtained from twelve deter-minations for chloride in water, shownin Table 1.

1 Mode = 100X n + Xn

-2- -i + 12 Median =

2

100 + 100 = 1002

_ X6 + X72

3 Mean = ---1n

. 98 + 2 (99) + 4 (100) + 3 (101) + 2 (102)12

= 100.25

4 Aid in Calculation

Consulting Table 1 and observing thatthe values are in the neighborhood of100 we might subtract 100 from eachscore and obtain the following distribution:

Table 2

Frequency Table

Chloride (Rig/1)

<-2

6

1

2

Frequency

1

2

4

3

2

1'1

Denote the mean of the distribution inTable 1 by X . If we add 100 to eachscore in the castribution in Table 2, weobtain the scores in the distribution inTable 1; likewise if we add 100 to themean, X, of the distribution in Table 2,we obtain the mean, 31 , of the distri-bution in Table I. c

Thus i cc= X + 1 0o

R =S-2-c + 100c n

1(-2) + 2(-1) + 4(0)4- 3(1) + 2(2)Xc + 100

12

Tic = .25 + 100 = 100.25

IV MEASURES OF DISPERSION

A Definitions

1 Dispersion - spread or variability ofobservations in a distribution

2 Range - the difference between thehighest value and the lowest value

R = max - min,

3 Average deviation - the sum of thedeviations of the values from theirmean, without regard to sign, dividedby the total number of data values (n)

The formula for calculating the averagedeviation is:

Elxi - XId= ___,---n

2-3


4 Average deviation of the mean (D)the average deviation of individualdata items from the mean (d) dividedby the square root of the number ofdata items (n)

The definition of the average deviationof the mean can be expressed by theformula:

dD = 4-71

5 Variance - the stun of the squares ofthe deviations of the values from theirmean divided by the total number ofdata values (n) minus 1

The definition of the variance can beexpressed by the following formula:

2E(Xi -

s n - i

6 Standard deviation - the square rootof the variance

The definition of the standard deviationcan be expressed by the followingformula:

31)2

n- 1

However, the formula commonly usedbecause of its adaptability to the handcalculator is the following:

s

j =02EXi2 - ---n - 1

where there aren number of values.

7 Standard deviation of the mean (S) - thestandard deviation of individual dataitems (s) divided by the square root ofthe number of data items (n)

The definition of the standarddeviation of the mean can beexpressed by the formula:

S

8 Relative standard deviation - thestandard deviation (s) expressed asa fraction of the mean, s

The relative standard deviation isoften expressed as a percent. It isthen referred to as the coefficientof variation (V):

V = X 100 =

X

The relative standard deviation isparticularly helpful when comparingthe precision of a number of deter-minations on a given substance atdifferent levels of concentration.

B Aids in Calculation

Application of the following statementscan reduce errors and amount of timespent in calculating the variance orstandard deviation of a distribution.

1 Adding or subtracting a constant to orfrom each score in a distributiondoesn't affect the variance or standarddeviation of the distribution.

Thus the following formulas:

(1) 2s

2 =s2c

(2) ss

where the Xi's are the values i95the distribution with variance s'and standard deviation s, and theXi + C's are the values in thedistribution with variances

c2

and standard deviation Sc.


2 Multiplying or dividing each score in 2 Average deviation - d II -CIa distribution by a constant is equivalent

n

to multiplying or dividing the varianceof that distribution by the square of the n- Xi !Xi- XI ni Xi- XI

Jame constant. 1 98 2. 25 2. 25

Thus the following formulas:2 99 1.25 2. 50

4 100 . 25 1. 00

(1) 2 c232c 3 101 .75 2.25

22

act 2 102 1. 75 3. 50-a

where the Xi's are the values inthe distribution with variances 2,

Xi ,and the CX 's or the arethe values in the distrbution withvariance set.

3 Multiplying or dividing' each score in adistribution by a constant is equivalentto multiplying or dividing the standarddeviation of that distribution by thesame constant.

Thus the following formulas:

(1) sc a Csor(2) s -c C

where the Xi's are the values inthe distribution with standarddeviation s, and the act's or the

Xi 'a are the values in theCdistribution with standarddeviation sc.

C Application

Consider the application of the abovadefinitions to the previously mentionedset of data, obtained from twelvedeterminations for chloride in water,shown in II B, Table I.

1 Range a 102 -98 = 4

= 100. 25 11. 50

n11.50

12= . 96

3 Average deviation of the mean -

dD = rri

Using calculations from number 2,

D , d 0.96 = 0.96 = O. 284-17 3.46

4 Variance s =2 E(Xi- TC)2

n- 1

n Xi Xi-Tc (Xi- R)2 n(Xi- TC)2

1 98 -2. 25 5.06 5.06

2 99 -1.25 1.56 3. 12

4 100 - .25 .06 .24

3 101 + .75 .56 1.68

2 102 +1. 75 3. 06 6. 12

15. 22

s 2 Z(Xi- R)2 16.22 1.47n - 1 11

2-5

Ns

Statiltics for Chemists

(3)217 -----

JEXi-(Exi)2

s22 2 12 16.25n 11 11

5 Standard deviation - s =

n

2

4

3

2

X.3 n2a.

98 98

99 198

100 400

101 303

102 204

1203

120321

n - 1

= 1.48

2X._i niX2i,

9604 9604s = EX i i s /1.91 1.22

2 - (EX ) 2=

9801 19602-n

10000 40000 n - 110201 30603

10404 20808 7 Standard deviation of the mean -120617 s

S =r-ii-

8 i2061712 /20617 - 120601

11 11

s Tr/a = 1.21

6 Aid in calculation

Recalling that adding or subtracting aconstant to each score in the distri-bution doesn't affect the variance orthe standard deviation of the distribu-tion we can simplify the computationsby first subtracting 100 from eachscore in the distribution, thus obtain-ing the frequency distribution shownin Table 2.

n Xi-C n(Xi-C) (Xi-C)2 n(Xi-C)2

1 -2 -2 4 4

2 -1 -2 1 2

4 0 0 0 0

3 1 3 1 3

2 2 4 4 8_3 17

2 (EX1)2

32EX

i n

2-6

n - 1

Using calculations from number 6

S=s 1.22=

1.22 = 0.35.T-71 rr2 3.46

8 Relative standard deviation eas a percent (coefficient of

V= 8 X ioo5E

Using calculations froms = 1.22 and from numb3C = 100.25,

V= s ,. 1.22.------31

100.25x

FigurNormal Distr

2

ution Curve

,

xpresseddation)

number 6 forer 2 for

100 = 1.21%

IDQuantity Measured

V INTRODUCTION TO NORMALDISTRIBUTION CURVE

A Statistics deals with theoretical curveswhich are smoother than frequencypolygons, obtained from experiments inreal life. However, frequency distribu-tions or frequency polygons of experimentaldata often approximate a mathematicalfunction called the "normal" distributioncurve. (See Figure 2)


As shown in Figure 3, the frequency polygonfor the 12 determinations for chloride inwater is e fairly good approximation of thenormal curve. If, however, in the chloridedeterminations we had obtained 103 insteadoi. 98 and 104 instead of 99 this distributionwould not have been a good approximation ofthe normal curve, as is shown in Figure 4.

Figure 3

Comparison of Normal Curve and Frequency Polygon

97 98 99 100

Chloride µg/ 1

Figure 4

101 102

Comparison of Normal Curve and Frequency Polygon

99 100 101 102

Chloride 4 g /1

2 "'

103 104

2-7


B If a frequency distribution is a goodapproximation of the normal curve, wegin use some facts about the normalairy. to give us information about thefrequency distribution.

2-8

Figure 5 shows the normal distributionin terms of the population mean andthe standard deviation of the 22 ulFLitgaa

, and gives the percent of area underthe curve between certain points.

Figure 5

Normal Distribution Curve

4 68% ---911

95%

-3c 4 -2ff

99%

4 +1.1

Figure 6

+2cr

Frequency Distribution Polygon

fi1

II

oe° 75%--114i

100%

-2B -1s X

98 99

4 +30'

+Is +2s100 101 102

Chloride sag /1 2 1

We may check the distribution of sampledata to see if it is a "normal" distributionin the following manner. Substitute thevalue of the sample mean (X) for the valueof the midline and substitute the value ofthe san_....al.e standard deviation (a) for thelimits of the value spans where we mightexpect certain percentages of the dataitems to occur. Then we can check thenumber of data items which actually dooccur within these value spans.

Figure 8 demonstrates this applicationusing the chloride data values from Table 1.The data values are marked on the hori-zontal line and the frequency of theoccurrence of each value is marked on thevertical. The midline of the distributionis marked at the value of the sample meanOr 100, See III C 3). The value of thesample standard deviation (a 1.21, SeeIV C 5) is used to mark value areas underthe curve where different percentages ofdata values will probably occur. Thus,for the area X I- la, X - is = 98.79 andX + 1 s 101.21. Therefore, accordingto the normal distribution curve shown inFigure 5, we might expect about, 68% of thedata items to have values between 99 and101. (The values are rounded to wholenumbers since the data values are thusrecorded).

Consulting Table 1, we find that 75% or 9of the 12 data items have values in thisrange. This percentage is shown inFigure 8 by the frequency polygon for thedata shown earlier in Figure 3.

Likewise assuming a normal distribution,we would expect 95% of the observationsto lie within +2 a 's from the populationmean. In fact, 100% of the observationswere within 2 s's from the sample mean.

In both cases the observed percentages arereasonablLclose to the expected percentages.Other testiexist for determining whetheror not a frequency distribution mightreasonably be assumed to approximatethe normal distribution.


It would be good to become as familiar aspossible with the normal distribution sincean underlying normal distribution isassumed for many statistical testa ofhypothesis.

REFERENCES

1 Bennett, C.A. and Franklin, N. L.Statistical Analysis in Chemistry andthe Chemical Industry. John Wiley& Sons, Inc., New York. 1954.

2 Crow, E. L. , Davis, F. A. , and Maxfield,M. W. Statistics Manual. DoverPublications, Inc., New York. 1.960.

3 Multi, W. J. and Massey, F. J.Introduction to Statistical Analysis.McGraw-Hill Book Co., Inc., NewYork. 1957.

4 Ostle, E. Statistics in Research. TheIowa State University Press, Iowa.1963.

5 Youdon, W. J. Statistical Methods forChemists. John Wiley & Sons, Inc.,New York. 1951.

This outline was prepared by L. A. Lederer,Statistician, formerly with AnalyticalReference Service, Training Program,NCUIH, SEC. Revised by Audrey D.Kroner, Chemist, National 'rraining andOpera` tional Technology Center, MOTD, OWPOUSEPA, Cincinnati, Ohio 45268.

Descriptors: Graphic Methods, Quality Control,Statistical Methods, Statistics

2-9

tNTERLABORATORY QUALITY CONTROL STUDIES

I DEFINITION OF INTERLABORATORYSTUDY IN EPA

A It is a between-laboratory evaluation ofexact physical, chemical or biological.:methods which yield a measurement ofdesired properties.

B It is not an interlaboratory evaluation ofraateMs, reagents, or different teatconditions. It is not an initial study of amethod, nor a study to develop methods.

II PURPOSES OF INTERLABORATORYSTUDY

A Selection of an analytical method.

B Evaluation of an analytical method.

C Evaluation of laboratory and analystperformance.

III PREPARATION FOR INTERLABORATORYSTUDY

A Laboratories are in control. Between-replicate deviation will then be verysmall and '2niform in all laboratories.

B Ruggedness testing of a method has beencompleted by the method developer, orby a single qualified laboratory. EffectsM small changes in time, temperature,pH, reagent and sample volume measure-meT, etc., are known and correctedfor:.

C All laboratories and analysts are familiarwith the method to be tested. Any specialequipment or reagents required areknown and available., If all analysts arenot acquainted with the method, a simplepreliminary study is undertaken toaccomplish this.`21

CH. MET. con. 7a. 11.76

IV INSTRUCTIONS FOR THE STUDY

A Exact method write-ups are providedto each analyst. The technicaldescription of an analytical methodis difficult. The language and organiza-tion must be complete, yet simple andunambiguous. These characteristicsare often conflicting.

Requirements in a method may seemtrivial. Yet if these are not recognizedand controlled, the entire study can beuseless.

B Explicit directions for sample preserva-tion, sample make-up, time limitations,sequence of analyses, etc. , are provided.

C Advance notice of tests is given so thatlaboratories can integrate the test intotheir program and realistic deadlinescan be established. Once established,deadlines for agreement to participate,for complecion of analyses and for report-ing are followed closely.

THE TEST SAMPLES

A The sample is carefully designed to reflectthe concentrations found in natural watersand yet be within the best portion of theconcentration range for the method.

B Since precision of almost all methodsvaries with concentration, a compre-hensive study includes several levels ofconcentration.

C Since accuracy of a method is very im-portant, exactly known levels of con-stituents are added both to distilled andnatural waters for testing. Stabilizednatural waters are not used as samples.

3-1

2nterlaboratory Quality Control Studies

AI SAMIFLE FORM AND CONTAINER

A In the EPA Inter laboratory Program,samples are prepared as concentratesfor final dilution at the testing laboratory.Use of concentrates has several ad-vantages. This greatly reduces spacerequirements and thereby greatly re-duces cost of mailing. With use of sealedglass ampules, preservation of con-centrates is easily obtained throughsteam sterilization. Chemical preser-vatives can be used at fairly high levelsin the concentrate and can be removed asinterferences by the dilution of the con-centrate to final sample volume. Theo-retically, a well designed sample proper-ly sealed in glass is stable. inthfinitely.

B Samples can be prepared as dilute sim-ulated or natural water samples to beanalyzed, as is. These samples have anadvantage in that they reach the analystexactly as a routine sample. There isno error in making a dilution and thissource of variance in the method is re-moved.

Some disadvantages of such samples are:the serious logistics problem fr ,rnstorage if a large number of samples isinvolved; a relatively high cost of mailing;and a limited choice of preservationmethods because of the size of containers,the probable use of plastic containers,and only limited testing of water types.

VII INTER LABORATORY STUDY DESIGN

A Sample concentrates are prepared in sim-ilar yet different concentrations at eachof several levels of analysis.

B True values are calculated from theamount of a constituent added, and areused to measure the acct racy of the studymethod.

C Multiple constituents are prepared in asample by related groups of analyses.

For example, a mineral and physicalsample is analyzed for pH, alkalinity/acidity, specific conductance, total hard-ness, sulfate, chloride, fluoride, solids,calcium, magnesium, ,sodium andpotassium. A nutrient sample is analyzedfor ammonia nitrogen, nitrate nitrogen,Kjeldahl nitrogen, orthophosphate andtotal phosphorus.

D The analyst dilutes separate aliquots tovolume with distilled water and withnatural water of his choice.

E Single analyses are made for eachparameter. Replicates are of limitedvalue in evaluating a method for routineanalyses because of data handlingproblems and failure to detect significan,differences by replication.

F Recoveries are compared for distilledand for natural waters. Recovery fromdistilled water should indicate methodbias. Differences in recoveries fromdistilled and natural waters shouldindicate interference by natural watersamples.

VIII DATA EVALUATION AND REPORTING

A Rejection of Outliers-Accuracy.

The T-Test is applied to all data usingthe standard deviation of all data andcompared with a T value at the 99% con-fidence interval. Accuracy is calculatedafter rejection of this data.

B Precision Measures

A statistical summary is developed forcomputerized treatment of data assuggested by Larsen(3). Method pre-cision is reported as standard deviation,95% confidence interval, and coefficientof variation. Method precision is reportedfor each sample since it varies withconcentration.

04A.,,,, 1,

Interlaboratory Quality Control Studies

C Graphic Display

Methods are evaluated using two-samplecharts described by Youde:- TheThevalues reported by an analyst for asample pair became the x and y coordi-nates for a single data point on the chart.Data form ellipsv. around the true valuebecause of systerailic error. Randomerror is shown as th8 perpendiculardistance from the 45 slope. Generalscatter indicates poor method pre-cision. Bias is shown by movement ofthe ellipse away from the true values.Outliers are visible at some distancefrom the major grouping of data. Ex-amples shown on the following pages are:

Figure 1 - Outliers,Figure 2 - Limited precision and

limited accuracy,Figure 3 - Negative bias,Figure 4 - Positive bias,Figure 5 - Systematic error,Figure 6 - Good precision and

accuracy.

1X INTERLARONATORY STUDYREPORT "'

A Reports are slanted toward audience.

B An data are coded by laboratory andanalyst.

C Glossary of terms is necessary.

D Computer use increases statisticalcapabilities tremendously. There is adanger of 'over - evaluation ".

E Increased efficiency in reporting data isachieved by using the computer print-outdirectly in a report. See Figures 7and 8.

REFERENCES

1 Youden, W. J. 1961. ExperimentalDesign and ASTM Committees. MaterialsResearch & Standards, 1( 11):862-867.

2 Preliminary Study, Nutrient Analyses,September 1969, Analytical QualityControl Laboratory, FWPCA,Cincinnati, Ohio.

3 Larson, Kenneth E., The Summarizationof Data, J. of Quality Technology, Vol.1, No. 1, Jan. 1969.

4 Youden, W. J. 1967. Statistical Tech-niques for Collaborative Tests. TheAssociation of Official AnalyticalChemists, Washington, D. C.

5 Winter, J. A. and Midgett, M. R. 1969.FWPCA Method Study 1, Mineral andPhysical Analyses, Analytical QualityControl Laboratory, FWPCA, Cincinnati,Ohio.

6 Winter, J. A. and Midgett, M. R. 1970.Method Study 2, Nutrient Analyses, ManualMethod, EMSL. USEPA. Cincinnati, Ohio.

This outline was prepared by J. A. Winter,Chief, Method & Performance Evaluation,EMSL, USEPA, Cincinnati, Ohio 45268.

Descriptors: Laboratory Tests, QualityControl, Statistical Methods, TestingTesting Procedures. Error Analysis, GraphicMethods

3-3

2.71

2.4

2.2

1.9

1.1

1.0

-

our

!\ L.,-.ao..31or44

1,

FIGURE 1 -AMMONIA NITROGEN, mg N/LITER, OUTLIERS

REJECTED DY HEST AT 01% PROBABILITY. -FROM: METHOD STUDY 2, NUTRIENT ANALYSES,

MANUAL METHODS -

-

-

''''

441.-

-.

' -,

SAMPLE 311111111.74 1.98 .48 2.10

FIGURE 2

KIELDAHL NITROGEN, mg N/LITER, POOR PRECISION.

FROM: METHOD STUDY 2, NUTRIENT ANALYSES,

.16 MANUAL METHODS

.52

.28

+.04

so*

-0.2 i.280:2 +.14

-

SAMPLE 5a I

.52 .16 1.00

1.0

6.0

/ 5.0

4.0

3.0

I 1 1 I 1

FIGURE 5

KIELQA111. MITIJOGEN, mg N/LITER,

GOOD PRECISION WITH SYSTEMATIC

ERROR. FROM: METHOD STUDY 2,

NUTRIENT ANALYSES, MANUAL

. METHODS

I

LaiJ-a.

44

4

20 t I

.

SAMPLE .8a a

2.0 3.0 4.0 5.0 8.0 7.0

FIGURE 6

pH DATA, GOOD PRECISION AND GOOD ACCURACY.

FROM: METHOD STUDY I, MINERAL AND PHYSICAL

ANALYSES

CJ1

t! 1 T-T I

FIGURE 3

AMMONIA NITROGEN, mg N /LITER, LIMITED ACCURACY56 AS NEGATIVE BIAS. FROM: METHOD STUDY 2, NUTRIENT

ANALYSES, MANUAL METHODS

.42

.14

SAMPLE 1

0.0 .14 28 .42

28

24

20

I

.56 10

FiGURE 4

POTASSIUM mg A/LITER, POOR ACCURACY AS

A POSITIVE BIAS, SODIUM INTERFEGENCE.FROM FWPCA METHOD STUDY I. MINERAL

AND PHYSICAL ANALYSES

Ia

-.4O.

4 SAMPLE 5

16 20 24 21

041.14 ..1.10014.111 1440.11.. 44(41411.00 31144 I. .1$60$ 4ft14 4SIS

U M. . 1Ott 0114. 404 4111041411lf

ouaNt 0110C1181111111101 06 ,..$41461 0006 0111144a 4414.

/441116041 1.4

7, dIal 4.40044 COI/. VII. S.laoa VII. 3.3 14414.43 6.6460 161431 1.4644666.4 3.11144 III. Hy. 6.111396 40. 01 a

6.60414 3.10060 61646. 1.1P. 1.171611 in 7 I I6$0404$0 -4. oell 011111It 1111100. 011I41410 III

0114 ik 6101.0144 116046 11106161 IOC

0.1 1.1 11.3 .I1.6 3.4 4.1 1.011111.0 3.0 1.1 4.1714 I1.0 3.1 1.66 3.1101 31.1 3.1 5.44361.1 3.1 .33$01.1 3.8 .41434.4 33 1.16141.0 3.31.0 3.31.1 3.31.6 3.s1.. 3.11.6 3.111.1 3.01.1 3.11.1 3.11.0 3,11.$ 3.1.$ 3.11.$ 3.6I.. 3.44. I..1. I..3.4 3.13.6 3.3.6 3.,3. 3.13. 3.13.4 3.43.11 3.63.0 3.43. 3.1Lb 3.13.0 1.6

.6 U07111 0411.

1111..0411

FIGURE 7

DOD DATA, STATISTICAL SUMMARY. PREPARED BY

COMPUTER FOR DIRECT ISE IN REPURT.

FROM: METHOD RESEARCH STUDY 3, DEMAND ANALYSES

FIGURE 8

AMMONIA NITROGEN DATA, YOUDEN'S TWOSAMPLE CHARTS.

PREPARED FOR REPORT BY COMPUTER PLOTTER

IA0117181 NAM

.

NEIIC

MUM PAIR

. .

I

a

.

SA FtE 33

101101WL 11081 a

a

- E,, 3

2

0

30

1

SPECIFIC CONDUCTANCE

I INTRODUCTION

An electrical conductivity measurement of asolution determines the ability of the solutionto conduct an electrical current. Veryconcentrated solutions have a large populationof ions and transmit current easily or withsmall resistance. Since resistivity isinversely related to conductivity Ka very concentrated solution has a veryhigh electrical conductivity.

1

Electrical conductivity is determined bytransmitting an electrical current througha given solution, using two electrodes. Theresistance measured is dependent principallyupon the ionic concentration, ionic charge,and temperature of the solution althoughelectrode characteristics (surface area andspacing of electrodes) is also critical. Earlyexperiments in standardizing the measurementled to construction of a "standard cell" inwhich the electrodes were spaced exctly 1 cmand each had a surface area of 1 cre. Usingthis cell, electrical conductivity is expressedas "Specific Conductance". Modern specificconductance cells do not have the sameelectrode dimensions as the early standardcell but have a characteristic electrode spacing/area ratio known as the "cell constant".

1 distance (cm) 1 XkEP R area (cm2) K

k cell constant

Specific conductance units are Mhos /cm orreciprocal ohms/ cm. Most natural, freshwaters in the United States have specificconductances ranging from 10 to 1, ooepmicrorahos/ cm. (1 micromho 10 mho).

II CONDUCTIVITY INSTRUMENTS

Nearly all of the commercial specific con-ductance instruments are of a bridge circuitdesign, similar to a Wheatstone Bridge.Null or balance is detected either by metermovement, electron "ray eye" tubes, orheadphones. Since resistance is directlyrelated to temperature, some instrumentshave automatic temperature compensators,although inexpensive models generally havemanual temperature compensation.

Conductivity instruments offer direct specificconductance readout when used with a cell"matched" to that particular instrument.

Electrodes within the cell may becomedamaged or dirty and accuracy may beaffected; therefore, it is advisable tofrequently check the instrument readingswith a standard KC1 solution having a knownspecific conductance.

III CONDUCTIVITY CELLS

Several types of conductivity cells areavailable, each having general applications.Dip cells are generally used for fieldmeasurement, flow cells for measurementwithin a closed system, and pipet cells forlaboratory use. Many modifications of theabove types are available for specializedlaboratory applications; the Jones cells andinductive capacitance cells are perhaps themost common.

Examples of various cell ranges for the RB3- Industrial Instruments model (0-50raicromhos/ cm scale range) are in Table 1.

RelativeCell Conductivity

Number ValueMaximum rangemicromhos cm

Cel VS02Cel VS2Cel VS2O

CH. COND. 2e. 11.77

10100

0 - 5000 - 5000

Table 1

Most accurate rangemicromhos cm

0

20 - 300200 - 3000

4-1

S seine Conductance

IV Computation of Calibration Constant

A calibration constant is a factor to whichscale readings must be multiplied by com-pute specific conductance.

Properties of some inorganic lone in regardto electrical conductivity are shown below:

Micromhos / cmIon per meq/1 conc.

cMKSPa

where Ksp

c

M

a actual specific conductance

Er calibration constant

a meter reading

CalciumMagnesiumSodiumPotassiumBicarbonateCarbonateChloride

52.046.648.972.043.684.675.9

For example, a 0.001 N KC1 solution(147 micromhos/cm standard) may showa scale reading of 147.

147147 al c 147, cs........... U 1.00

147

In this case the cell is perfectly "matched"to the instrument, the calibration constantis 1.00, and the scale reading representsactual specific conductance. A variety ofcells, each covering a specific range,may be used with any one instrument.However, a calibration constant for eachcell must be computed before solutions ofunknown specific conductance can bedetermined.

V RELATIONSHIP OF SPECIFIC CON-DUCTANCE TO IONIC CONCENTRATION

Natural water consists of many chemicalconstituents, each of which may differwidely in ionic size, mobility, and solubility.Also, total constituent concentration andproportions of certain ions in various naturalwaters range considerably. However, itis surprising that for most natural watershaving less than 2,000 mg/l. dissolved .

solids, dissolved solids values are closelyrelated to specific conductance values,ranging in a ratio of .62 to .70. Of coursethis does not hold true for certain watt rshaving considerable amounts of nonionizedsoluble materials, such as organic com-pounds and nonlonized, colloidal inorganics.

4.-.2

VI ESTIMATION OF CONSTITUENTCONCENTRATIONS

Generally speaking, for waters having adissolved solids concentration of less than1,000 mg/1, calcium and magnesium (totalhardness), sodium, bicarbonate andcarbonate (total alkalinity), and sulfate arethe principal or most abundant ions,representing perhaps 90-99% of the totalionic concentration of the water. Specificconductance, total hardness and totalalkalinity are all simple and expedientmeasurements which can be performed inthe field. Therefore, the remaining principalions are sodium and sulfate, and concentrationsof these can be estimated by empiricalmethods. For example, we find that a certainwater has:

Kap 500 micromhos/ cmTotal Hardness a 160 mg /1 or 3.20 meq/1Total Alkalinity is 200 mg/1 or 3.28 meq /l,as bicarbonate.

Next we multiply the specific conductance by*0.011 (500 X0.011 5.50) to estimate thetotal ionic concentration in meq /l.

* This factor may vary slightly fordifferent waters

Cations (me g/1)

CalciumMagnesiumSodium 5.50-3.20 2.30Total Cations 5.50

e) 0t.,

3.4.20

Anions (meg/1)

Carbonate 0.00Bicarbonate 3.28Sulfate 5.50-3.28 2.22Total Anions 5.50

S ecific Conductance

Realizing that several variables are involvedin empirical analysis, application restsentirely upon testing the formula with previouscomplete laboratory analyses for thatparticular water. If correlation is withinacceptable limits, analytical costa may besubstantially reduced. Empirical analysiscan also be used in determination of properaliquots (dilution factor) necessary forlaboratory analysis.

Records of laboratory chemical analysis mayindicate that a particular stream or' lakeshows a characteristic response to variousstrearnflow rates or lake water levels. Ifthe water's environment has not been alteredand water composition responds solely tonatural causes, a specific conductivitymeasurement may be occasionally used insubstitution for laboratory analyses todetermine water quality. Concentration ofindividual constituents can thus be estimatedfrom a specific conductance value.

VII APPLICATIONS FOR SPECIFICCONDUCTANCE MEASUREMENTS

A Laboratory Operations(2)

1 Checking purity of distilled and de-ionized water

2 Estimation of dilution factors forsamples

3 Quality control check on analyticalaccuracy

4 An electrical indicator

B Agriculture

1 Evaluating salinity

2 Estimating Sodium Adsorption Ratio

C Industry(3)

1 Estimating corrosiveness of water insteam boilers

4

2 Efficiency check of boiler operation

D Geology

1 Stratigraphic identification andcharacterization

a geological mappingb oil explorations

E Oceanography

1 Mapping ocean currents

2 Estuary studies

F Hydrology

1 Locating new water supplies

a buried stream channels (See Fig. 1)b springs in lakes and

streams (See Fig. 2)

2 Detection and regulation of sea waterencroachment on shore wells

G Water Quality Studies

1 Estimation of dissolved solids (2)

(See Section V, also Fig. 3)

2 Empirical analysis of constituentconcentrations (See Section VI, alsoreference 2)

3 Quality control check for salt waterconversion studies

4 Determination of mixing efficiencyof streams (See Fig. 4 )

5 Determination of flow pattern ofpolluted currents (See Fig. 4)

8 Identification of signifi..zant fluctuationsin industrial wastewater effluents.

7 Signal of significant changes in thecomposition of influents to wastetreatment plants

4.3

Specific Conductance

is

FIGURE 1

orrsaioN Of MED STREAM CHANNELS

FIGURE 2

DETECTION OF SPRINGS IN LAKES AND STREAMS

MAN LOWERINGCONDUCTIVITY CELL

LAND SURF


c.;

vs

; 5000;xI00 4000iaV

30001.-v30Z 2000vvWv 1000usa." 500

500 1000 1500 2000 2500 3000

DISSOLVED SOLIDS, Jog /liter

3500

Figure 3. SPECIFIC CONDUCTANCE AND DISSOLVED SOLIDSIN COMPOSITES OF DAILY SAMPLES. , GILA RIVERAT BYLAS, ARIZONA, OCTOBER 1, 1944 TOSEPTEMBER 30, 1944.

Geological Survey WaterSupply Paper 1473.

POILVION IMMO

11111"Pir'.,4**


4.6

00Xg 30000v8

aZv 2000Iv=0Z0 1000Vaja".

VW...WI

t

10 . 20 30 40TEMPERATURE, 'C.

Figure 5. SPECIFIC CONDUCTANCE OF A 0.01NORMAL SOLUTION OF POTASSIUMCHLORIDE AT VARIOUS TEMPERATURES.

Geological Survey Water-Supply Paper 1473.

1600

1400

1200

1000

800

600

400

200

oo100 200 300 400 SOO 600 700

POTASSIUM CHLORIDE, rag /liter

Figure 6. SPECIFIC CONDUCTANCE OF POTASSIUMCHLORIDE SOLUTIONS.

Geological Survey Water-Supply Pcper 1473


VIII NPDES METHODOLOGY

A The Federal Register "List of ApprovedTest Procedures" for NPDES require-ments specifies that specific conductancebe measured with a self-containedconductivity meter. Wheatstone bridgetypell) (2) (3)

Temperature directly affects specificconductance values (see Fig. 5). Forthis reason, samples should preferablybe analyzed at 25 C. If not, temperaturecorrections should be made and resultsreported as arches /cm at 25°C.

1 The instrumer should be standardizedusing KC I solutions. (See Fig.6)

2 It is essential to keep the conductivitycell clean.

B The EPA manual specifies using theprocedure as described in Standard

Methods(2)

or in ASTM Standards(3).

These are approved in 40 CFRI36 forNPDES Report purposes.

C Precision and Accuracy(1)

Forty-one analysts in 17 laboratoriesanalyzed 6 synthetic water samplescontaining the following Lisp increments

of inorganic salts: 100. 108. 808. 848,1640 and 1710 micromhos/cm.

The standard deviation of the reportedvalues was 7. 55, 8. 14, 66. 1, 79.6,106 and 119 g mhos/cm respectively.

The accuracy of the reported values was-2.0, -0.8, -29.3. -38.5, -87.9 and-86.5 A mhos/cm bias respectively.

REFERENCES

1 Methods for Chemical Analysis of Waterand Wastes, EPA-AQCL, Cincinnati,Ohio 45268, 1974.

2 Standard Methods for the Examinationof Water and Wastewater. APHA-AWWA-WPCF, 14th Edition, 1976.

3 ASTM Annual Book of Standards, Part 31,1975.

This outline was prepared by John R. Tilstra,Chemist, National ET:trophication ResearchProgram. Corvallis, Oregon with additionsby Audrey D. Kroner, Chemist. NationalTraining and Operational Technology Center.MOTD, OWPO, USEPA, Cincinnati. Ohio 45268.

Descriptors: Chemical Analysis. Concentration,Conductivity, Dissol-fed Solids, ElectricalConductance, Ions, Physical Properties, Salinity,Sodium, Specific Conductivity, Sulfates, WaterAnalysis, Water Supplies.

4-7

CALIBRATION AND USE OF A CONDUCTIVITY METER

I EQUIPMENT AND REAGENTS

A Equipment

1 Sok Bridge conductivity meters

2 Probes

a Cell VSO2

b Cell VS2

c Cell VS2O

3 Thermometers

4 400 mi beakers

B Reagents

1 Standard KC1 solutions

Normality ofKC1 Solution

400010.0010.010.1

Specific Conductanceraicrozahos/ cm.

2 Distilled water

14.9147.0

1413.012900.0

U CHECKING THE INSTRUMENT

A The measurement of specific conductivityas presented in seed cos II and III iswritten for one type of conductivity meterand probe.

B A battery check is made by depressingthe Battery Check switch, and at thesame time pressing the on-off button.The meter needle should deflect to theright (positive) and come to rest in thegreen zone.

C Place a 10,000 ohm resistor in the holesof the electrical contacts on the meter.Turn the temperature knob to read 25°C.Depress the on-off button and bring themeter needle to a reading of 0 by

CH. COND. lab. 3c. 11.77

turning the specific conductance switch.The specific conductance reading shouldbe approximately 200 micromhos/ cm.

III DETERMINATION OF THE CALIBRATIONCONSTANT

A

B

C

Determine the temperature of thestandard KC1 solutions and move thetemperature knob to that value.

Connect probe Cell VS02 to the con-ductivity meter.

Rinse the probe in the beaker ofdistilled water, wipe the excess waterwith a ldrnwipe and place probe in thefirst beaker of KC1 solution (0.0001 N).

D Make certain the cell is submerged toa point at least 1/2 inch above the airhole and that no entrapped air remains.The cell should also be at least 1/2inch from the inside walls of the flask.

E Press and hold down the ON-OFFbutton, simultaneously rotating the mathscale knob until the meter reads zero.Release the button. (If the meter needleremains off scale or cannot be nulled,discontinue testing in that solution.)

F Record the scale reading in Table I. andproceed to KC1 solutions 0.001N,0.01N, 0.1 N using Steps C, D and E.

G Repeat steps C through F using the VS2,then the VS20 probe.

H Compute the cell calibration constant-afactor by which scale readings must bemultiplied to compute specific conductance:K = cMsp

where K I actual specific conductance,c = caligiiition constantMa meter reading

(continued next page)

51

1

Calibration and Use of a Conductivity Meter

TABLE 1 DATA FOR CALIBRATION CONSTANTS

Probe j Cell VS02 Coll VS2 Cell VS20

KC1 I

Solutin 0.0001N O. 001N 0.01N 0.1N 0.0001N 0.001N O. OIN 0.1N 0.0001N 0.001N O. 01N 0.1N

Test /1

C*11

Constanti....

For each cell, calculate the cell constantby using the meter reading closest to the400 - 600 range. The known specificconductance for the corresponding KC1solution can be found in IB Reagents. Recordthe cell constants on Table I.

IV DETERMINATION OF Ksp FOR SAMPLES

Obtain meter readings, M, for samplesA, B and C using Section III C, D and E.Record M in Table 2. See Table I forthe appropriate cell constant, c, tocalculate Kso for each sample whereKsp = cM. Record results in Table 2.

V EPA METHODOLOGY

The current EPA Manual(1) specifiesusing the procedures found in References2 and 3. These procedures have beenadapted for this laboratory session anall are approved in 40CFR136 for NPDESreport purposes.

5-2

TABLE 2.

ACKNOWLEDGMENT

This outline contains certain portions ofprevious outlines by Messrs. J. W. Mandia,and J. R. Tilstra.

REFERENCES

1 Methods for Chemical Analysis ofWater and Wastes, USEPA, AQCL,Cincinnati, OH 45268, 1974.

2 Standard Methods for the Examinationof Water and Wastewater, 14thEdition.. 1976.

3 Book of ASTM Standards, Part 31, 1975.

This outline was prepared by C. R. Feldmann,Chemist, National Training and OperationalTechnology Center, and revised by Audrey D.Kroner, also with National Training andOperational Technology Center, MOTD, OWPO,USEPA, Cincinnati, Ohio 45268.

Descritom Analytical Techniques, Conductivity.Electrical Conductance. Specific Conductivity,Water Analysis.

SPECIFIC CONDUCTIVITY TESTS

Sample A B C

Probe IVS02

ellVS2

CeVS2O

IVS02

Ce IVS2

IVS2O

elVS02

CelVS2 VS2O

1t... .,

CellConstantSp. Cond.gmbos/cm _

39

DISSOLVED OXYGENFactors Affecting DO Concentration in Water

I The Dissolved Oxygen determination isa very important water quality criteria formany reasons:

A Oxygen is an essential nutrient for allliving organisms. Dissolved oxygen isessential for survival of aerobicorganisms and permits facultativeorganisms to metabolize more effectively.Many desirable varieties of macro ormicro organisms cannot survive atdissblved oxygen concentrations belowcertain minimum values. These valuesvary with the type of organisms, stagein their life history, activity, WI otherfactors.

B Dissolved oxygen levels may be used asan indicator of pollution by oxygendemanding wastes. Low DO concen-trations are likely to be associated withlow quality waters.

C The presence of dissolved oxygenprevents or minimizes the onset ofputrefactive decomposition and theproduction of objectionable amounts ofmalodorous sulfides, mercaptans,amines, etc.

D Dissolved oxygen is essential forterminal stabilization wastewaters.High DO concentrations are normallyassociated with good quality water,

E Dissolved oxygen changes with respectto time, depth or section of a watermass are useful to indicate the degreeof stability or mixing characteristicsof that situation.

F The HOD or other respirometric testmethods for water quality are commonlybased upon the difference between aninitial and final DO determination for agiven sample time interval and con-dition. These measurements areuseful to indicate:

WP. NAP. 25a. 11. 774

1 The rate of biochemical activity interms of oxygen demand for agiven sample and conditions.

2 The degree of acceptability(a bioassay technique) for bio-chemical stabilization of a givenmicrobiota in response to food,inhibitory agents or test conditions

3 The degree of instability of awater mass on the basis of testsample DO changes over anextended interval of time

4 Permissible load variations insurface water or treatment unitsin terms of DO depletion versustime, concentration, or ratio offood to organism mass, solids, orvolume ratios.

5 Oxygenation requirementsnecessary to meet the oxygendemand in treatment units orsurface water situations.

G Minimum allowable DO concentrations arespecified in all Water Quality standards,

II FACTORS AFFECTING THE DOCONCENTRATION IN WATER

A Physical Factors:

1 DO solubility in water for anair/water system is limited toabout 9 mg DO/liter of water at20°C. This amounts to about0.0009% as compared to 21% byweight of oxygen in air.

2 Transfer of oxygen from air towater is limited by the interfacearea, the oxygen deficit, partialpressure, the conditions at the

6-1

Dissolved Oxygen

interface area, mixing phenomenaand other items.

Certain factors tend to confusereoxygenation mechanisms ofwater aeration:

a The transfer of oxygen in airto dissolved molecular oxygenin water has two principalvariables:

1) Area of the air-waterinterface.

2) Dispersion of the oxygen-saturated water at theinterface into the body liquid.

The first depends upon the surfacearea of the air bubbles in the wateror water drops in he air; thesecond depends upon the mixingenergy in the liquid. If diffusorsare placed in a line along the wall,dead spots may develop in the core.Different diffusor placement ormixing energy may improve oxygentransfer to the liquid two or threefold.

b Other variables in oxygen transferinclude:

3) Oxygen deficit in the liquid.

4) Oxygen content of the gas phase.

5) Time.

If the first four variables arefavorable, the process of wateroxygenation is rapid until the liquidapproaches saturation. Much moreenergy and time are required toincrease oxygen saturation fromabout 95 to 100% than to increaseoxygen saturation from 0 to about

6-2

95%. For example: An oxygen-depleted sample often will pick upsignificant DO during DO testing;changes are unlikely with a samplecontaining equilibrium amountsof DO.

c The limited solubility of oxygenin water compared to the oxygencontent of air does not requirethe interchange of a large mass ofoxygen per unit volume of waterto change DO saturation. DOincreases from zero to 50%saturation are common in passageover a weir.

d Aeration of dirty water is practicedfor cleanup. Aeration of cleanwater results in washing the air andtransferring fine particulates andgaseous contaminants to the liquid.

e One liter of air at room temperaturecontains about 230 mg of oxygen.A 5 gal carboy of water with 2 litersof gas space above the liquid hasample oxygen supply for equilibrationof DO after storage for 2 or 3 daysor shaking for 30 sec.

f Aeration tends toward evaporativecooling. Oxygen content becomeshigher than saturation values atthe test temperature, thuscontributing to high blanks.

3 Oxygen solubility varies with thetemperature of the water.Solubility at 100C is about twotimes that at 300 C. Temperatureoften contributes to DO variationsmuch greater than anticipated by

Dissolved Oxygen

solubility. A cold water often hasmuch more DO than the solubilitylimits at laboratory temperature.Standing during warmup commonlyresults in a loss of DO due tooxygen diffusion from the super-saturated sample. Sampleswarmer than laboratory tempera-ture may decrease in volume dueto the contraction of liquid astemperature is lowered. The fullbottle at higher temperature willbe partially full after shrinkagewith air entrance around the stopperto replace the void. Oxygen in theair may be transferred to raise thesample DO. For example, avolumetric flask filled to the 1000 mlmark at 300 C will show a waterlevel about 1/2 inch below the markwhen the water temperature isreduced to 200C. BOD dilutionsshould be adjusted to 200C + or -1 1/20 before filling and testing.

4 Water density varies with tem-perature with maximum waterdensity at 40C. Colder or warmerwaters tend to promote stratificationof water that interferes withdistribution of DO because thehigher density waters tend to seekthe lower levels.

5 Oxygen diffusion in a water mass isrelatively slow, hence vertical andlateral mixing are essential tomaintain relatively uniform oxygenconcentrations in a water mass.

6 Increasing salt concentrationdecreases oxygen solubilityslightly but has a larger effectupon density stratification in awater mass.

7 The partial pressure of the oxygenin the gas above the water interfacecontrols the oxygen solubilitylimits in the water. For example.the equilibrium concentration ofoxygen in water is about 9 mg DO/1under one atmospheric pressure of

/1 o1 ....

air, about 42 mg DO/liter incontact with pure oxygen and 0 mgDO/liter in contact with purenitrogen (@ 200 C).

B Biological or Bio-Chemical Factors

1 Aquatic life requires oxygen forrespiration to meet energyrequirements for growth, repro-duction, and motion. The neteffect is to deplete oxygen resourcesin the water at a rate controlledby the type, activity, and mass ofliving materials present, theavailability of food and favor-ability of conditions.

2 Algae, autotrophic bacteria, plantsor other organisms capable ofphotosynthesis may use lightenergy to synthesize cell materialsfrom mineralized nutrients withoxygen released in process.

a Photosynthesis occurs onlyunder the influence of adequatelight intensity.

b Respiration of alga iscontinuous.

c The dominant effect in termsof oxygen assets orliabilities of alga depends uponalgal activity, numbers andlight intensity. Gross algalproductivity contributes tosignificant diurnal DOvariations.

3 High rate deoxygenation commonlyaccompanies assimilation ofreadily available nutrients andconversion into cell mass orstorage products. Deoxygenationdue to cell mass respirationcommonly occurs at some lowerrate dependent upon the nature ofthe organisms present, the stageof decomposition and the degreeof predation, lysis, mixing andregrowth. Relatively high

6-3

Dissolved Oxygen

deoxygenation rates commonly areassociated with significant growthor regrowth of organisms.

4 Micro-organisms tend to flocculateor agglomerate to form settleablemasses particularly at limitingnutrient levels (after availablenutrients have been assimilated orthe number of organisms are largein proportion to available food).

a Resulting benthic depositscontinue to respire as bedloads.

b Oxygen availability is limitedbecause the deposit is physicallyremoved from the source ofsurface oxygenation and algalactivity usually is morefavorable near the surface.Stratification is likely to limitoxygen transfer to the bed loadvicinity.

c The bed load commonly isoxygen deficient and decomposesby anaerobic action.

d Anaerobic action commonly ischaracterized by a dominanthydrolytic or solubilizing actionwith relatively low rate growthof organisms.

e The net effect is to produce lowmolecular weight productsfrom cell mass with a corre-spondingly large fraction offeedback of nutrients to theoverlaying waters. Theselysis products have the effectof ahigh rate or immediateoxygen demand upon mixturewith oxygen containing waters.

f Turbulence favoring mixing ofsurface waters and benthicsediments commonly areassociated with extremelyrapid depletion of DO.

Recurrent resuspension ofthin benthic deposits maycontribute to highly erraticDO patterns.

g Long term deposition areascommonly act like pointsources of new pollution asa result of the feedback ofnutrients from the deposit.Rate of reaction may be lowfor old materials but a lowpercentage of a large mass ofunstable material may produceexcessive oxygen demands.

C Tremendous DO variations are likelyin a polluted water in reference todepth, cross section or time of day.More stabilized waters tend to showdecreased DO variations although it islikely that natural deposits such as leafmold will produce differences relatedto depth in stratified deep waters.

ACKNOWLEDGMENTS

This outline contains significant materialsfrom previous outlines by J. Vv.". Mandia.

REFERENCE

1 Methods for Chemical Analysis ofWater & Wastes, U. S. EnvironmentalProtection Agency, EnvironmentalMonitoring & Support Laboratory,Cincinnati, Ohio, 45268, 1974.

This outline was prepared by F. J. Ludzack,former Chemist. National Training Center,and revised by Charles R. Feldmann,Chemist. National Training & OperationalTechnology Center, MOTD, OWPO, USEPA,Cincinnati, Ohio 45268

Descriptors : Aeration, Aerobic Conditions,Air-Water Interfaces, Anaerobic Conditions,Benthos, Biolcg ical Oxygen Demand, DissolvedOxygen, Water Pollution, Water Quality

DISSOLVED OXYGENDetermination by the Winkler Iodometric Titration Azide Modification

This method is applicable for use with moatwastewaters and streams that contain nitratenitrogen and not more than 1 mg/1 of ferrousiron. Other reducing or oxidizing materialsshould be absent. If 1 ml of fluoride solutionis added before acidifying the sample andthere is no delay in titration, the method isalso applicable in the presence of 100-200mg/1 ferric iron.

The azide modification is not applicableunder the following conditions: (a) samplescontaining sulfite. thiosulfate. polythionate.appreciable quantities of free chlorine orhypochlorite; (b) samples high in suspendedsolids; (c) samples containing organicsubstances which are readily oxidized in ahighly alkaline solution. or which are oxidizedby free iodine in an acid solution; (d) untreateddomestic sewage; (e) biological flocs; and(f) where sample color interferes withendpoint detection. In instances where theazide modification is not applicable, theDO probe should be used.

A Reactions

1 The determination of DO involvesa complex series of interactionsthat must be quantitative to providea valid DC result. The number ofsequential reactions also compli-cates interference control. Thereactions will be presented firstfollowed by discussion of thefunctional aspects.

MnSO4 + 2 KOH--»Mn(OH)2

+ IC2504 (a)

2 lv1n(OH)2

+ 02 -.*-2 MnO(OH)2

(b)

MnO(OH)2

+ 2 H2

SO4 Mn(S0 4)2 + 3H 20 (c)

Mn(S0 4)2 + 2 ICI -4, MnSO4 + IC2504 +12 (d)

12 + 2 Na 2S 203 --c Na2

S406 + 2 Na I (e)

2 Reaction sequence

The series of reactions involvesfive different operational steps in

CH. 0. do. 31e. 11.11

converting dissolved oxygen in thewater into a form in which it canbe measured.

a 02 --0MnO(OH)2

Mn(SO4)2

12-* Thiosulfate (thio) or

phenylarsine oxide (PAO)titration.

b All added reagents are in excessto improve contact possibilitiesand to force the reaction towardcompletion.

3 The first conversion, 0, -MnO(OH),, (reactions a, 6) is anoxygen trtutsfer operation wherethe dissolved oxygen in the watercombines with manganoushydroxide to form an oxygenatedrrianganic hydroxide.

a The manganous salt can reactwith oxygen only in a highlyalkaline media.

b The manganous salt and alkalimust be added separately withaddition below the surface ofthe sample to minimize reactionwith atmospheric oxygen viaair bubbles or surface contact.Reaction with sample dissolvedoxygen is intended to occurupon mixing of the reagents andsample after stoppering thefull bottle (care should be usedto allow entrained air bubblesto rise to the surface beforeadding reagents to preventhigh results due to includingentrained oxygen).

c Transfer of oxygen from thedissolved state to the pre-cipitate form involves a twophase system of solution andprecipitate requiring effectivemixing for quantitativetransfer. Normally a gross

7-1

r

Dissolved Oxygen Determination

excess of reagents is used to limitmixing requirements. Mixing byrapid inversion 25 to 35 times willaccomplish the purpose. Less energyis required by inversion 5 or 6 times,allowing the floc to settle until thereis clear liquid above the floc, repeatingthe inversion, & allowing the floc tosettle about two-thirds of the way downin the bottle. The reaction is rapid;contact is the principal problem inthe two phase system.

d U the alkalint floc is white,no oxygen is present.

4 Acidification (reactions c and d)changes the oxygenated manganichydroxide to manganic sulfatewhich in turn reacts withpotassium iodide to form elementaliodine. Under acid conditions,oxygen cannot react directly withthe excess manganous sulfateremaining in solution.

5 Iodine (reaction e) may be titratedwith sodium thiosulfate standardsolution to indicate the amount ofdissolved oxygen originallypresent in the sample.

a The blue color of the starch-iodine complex commonly isused as an indicator. Thisblue color disappears whenelemental iodine has beenreacted with an equivalentamount of thiosulfate.

b Phenylstr sine oxide solutions aremore expensive to obtain buthave better keeping qualitiesthan thiosulfate solutions.Occasional use, field operationsdad situations where it is notfeasible to calibrate thiosolutions regularly, usuallyencourage use of purchasedPAO reagents.

6 For practical purposes the DOdetermination scheme involves thefollowing operations.

a Fill a 300 ml bottle* underconditions minimizing DOchanges. This means that thesample bottle must be flushedwith test solution to displacethe air in the bottle with watercharacteristic of the testedsample.

*DO test bottle volumes shouldbe checked - discard thoseoutside of the limits of 300 ml+ or - 3 ml.

b To the filled bottle:

1) Add MnSO4 reagent (2 ml)

2) Add KOH, KI, NaN3 reagent(2 ml)Stopper, mix by inversion,allow to settle until there isclear liquid above the floc,repeat the inversion, & allowthe floc to settle about two-thirdsof the way down in the bottle.Highly saline & other test watersmay settle very slowly. In thiscase, allow some reasonable time(e.g. 2 min.) for completion of thereaction.

c To the alkaline mix (settledabout half way) add 2 ml ofsulfuric acid, stopper and mixuntil the precipitate dissolves.

d Transfer the contents of thebottle to a 500 ml Erlenmeyerflask and titrate with 0.0375normal thiosulfate. Eachml of reagent used represents1 mg of DO/liter of sample.

r7I

N:


The same thing applies forother sample volumes whenusing en appropriate titrantnormality; e.g..

1) For a 200 ml sample, use0.025 N Thio

2) For a 100 ml sample, use0.0125 N Thio

7 The addition of the first two DOreagents, (MnSO4 and the KOH, KIand NaN

3solutions) displaces an

equal quantity of the sample. Thisis not the case when acid is addedbecause the clear liquid above thefloc does not contain dissolvedoxygen as all of it should be con-verted to the particulate MnO(OH)2.Some error is introduced by thisdisplacement of sample duringdosage of the first two reagents.The error upon addition of 2 ml ofeach reagent to a 300 ml sampleis

3001... X 100 or 1.33% loss in DO.

This may be corrected by anappropriate factor or by adjust-ment of reagent normality. It isgenerally considered small inrelation to other errors in sampling,manipulation and interference,hence this error may be recognizedbut not corrected.

8 Reagent preparation and pro-cedural details can be found inreference 1.

IV The sequential reactions for theChemical DG determination providesseveral situations where significant inter-ference may occur in application onpolluted water, such as:

A Sampling errors may not be strictlydesignated as interference but have thesame effect of changing sample DO.Inadequate flushing of the bottle con-tents or exposure to air may raise theDO of low oxygen samples or lower theDO of supersaturated samples.

B Entrained air may be trapped in a DObottle by:

1 Rapid filling of vigorously mixedsamples without allowing theentrained air to escape beforeclosing the bottle and adding DOreagents.

2 Filling a bottle with low temperaturewater holding more DO than that inequilibrium after the samples warmto working temperature.

3 Aeration is likely to cool the samplepermitting more DO to be introducedthan can be held at the room orincubator temperatures.

4 Samples warmer than workingincubator temperatures will beonly partially full at equilibriumtemperatures.

Addition of DO reagents results inreaction with dissolved or entrainedoxygen. Results for DO are invalidif there is any evidence of gasbubbles in the sample bottle.

C The DO reagents respond to any oxidantor reductant in the sample capable ofreacting within the time allotted. HOC1or H202 may raise the DO titrationwhile H2S, & SH may react with sampleoxygen to lower the sample titration.The items mentioned react rapidly andraise or lower the DO melt promptly.Other items such as Fe or SOn mayor may not react completely within thetime allotted for reaction. Manyorganic materials or complexes frombenthic deposits may have an effect uponDO results that are difficult to predict.They may have one effect during thealkaline stage to release iodine fromK/ while favoring irreversibleabsorption of iodine during the acidstage. Degree of effect may increasewith reaction time. It is generallyinadvisable to use the iodometrictitration on samples containing largeamounts of organic contaminants or

7-3


benthic residues. It would be expectedthat benthic residues would tend towardlow results because of the reduced ironand sulfur content - they commonlyfavor high results due to other factorsthat react more "rapidly, often givingthe same effect as in uncontrollednitrite interference during titration.

D Nitrite is present to some extent innatural waters or partially oxidizedtreatment plant samples. Nitrite isassociated with a cyclic reaction duringthe acid stage of the DO determinationthat may lead to erroneous high results.

1 These reactions may be repre-sented as follows:

2HNO2

+ 2 HI -+I2 + 4H20 + N2102 (a)

2HNL,2 1- + 1/20 + H2 O+N20

2(b)

These reactions are time, mixingand concentration dependent andcan be minimized by rapidprocessing.

2 Sodium aside (NaN3) reacts withnitrite under acid conditions toform a combination of NI+ N20which effectively blocks tne,cyclic reaction by converting theHN0a to noninterfering compoundsof nifrogen.

3 Sodium aside added to freakalkaline is reagent is adequate tocontrol interference up to about20 mg of NO Witter of sample.The aside is Astable and grad-ually decomposes, If resuspendedbenthic sediments are not detectablein a sampla showing a returningblue color, it is likely that theaside has decomposed in thealkaline la aside reagent.

E Surfactants, color and Fe+++ mayconfuse endpoint detection if presentin significant quantities.

74

F Polluted water commonly containssignificant interferences such as C.It is advisable to use a membraneprotected sensor of the electronic typefor DO determinations in the presenceof these types of interference.

G The order of reagent addition and promptcompletion of the DO determination iscritical. Stable waters may give validDO results after extended delay oftitration during the acidified stage. Forunstable water, undue delay at any stageof processing accentuates interferenceproblems.

REFERENCE

1 Methods for Chemical Analysis ofWater & Wastes, U. S. EnvironmentalProtection Agency, EnvironmentMonitoring & Support Laboratory,Cincinnati, Ohio 45268, 1974.

ACKNOWLEDGMENTS

This outline contains significant materialsfrom previous outlines by J. 'N. Mandia,Review and comments by C. R. Mirth andR. L. Booth are greatly appreciated.

This outline was prepaRia by F. J. Luazack.farmer Chemist. National Training Center,and revised by C. R. Feldmann, Chemist.National Training & Operational TechnologyCenter, MOTT% OWPO, USEPA, Cincinnati,Ohio 45268.

Descriptors: Chemical Analysis, DissolvedOxygen, Oxygen, Water Analysis

4'7

I INTRODUCTION

DISSOLVED OXYGENDETERMINATION BY ELECTRONIC MEASUREMENT

A Electronic measurement of DO is attractivefor several reasons:

1 Electronic methods are more readilyadaptable for automated analysis, con-tinuous recording, remote sensing orportability.

2 Application of electronic methods withmembrane protection of sensors affordsa high degree of interference control.

3 Versatility of the electronic systempermits design for a particular measure-ment, situation or use.

4 Many more determinations per man-hour are possible with a minor expend-iture of time for calibration.

B Electronic methods of analysis imposecertain restrictions upon the analyst toinsure that the response does, in fact,indicate the item sought.

1 The ease of reading the indicator tendsto produce a false sense of security.Frequent and careful calibrations areessential to establish workability of theapparatus and validity of its response.

2 The use of electronic devices requiresa greater degree of competence on thepart of the analyst. Understanding ofthe behavior of oxygen must be supple-mented by an understanding of theparticular instrument and its behaviorduring use.

C Definitions

1 Electrochemistry - a branch of chemistrydealing with relationships betweenetectrical and chemical changes.

2 Electronic measurements or electro-metric procedures - procedures usingthe measurement of potential differencesas an indicator of reactions taldngplace at an electrode or plate.

3 Reduction - any process in which oneor more electrons are added to an atomor an i.m, such as 02 + 2e 4 20The oxygen has been reduced.

4 Oxidation - any process in which oneor more electrons are removed froman atom or an ion, such as Zno - 2e-- Zri+2. The zinc has been oxidized.

5 Oxidation - reduction reactions - in astrictly chemical reaction, reductioncannot occur unless an equivalentamount of some oxidizable substancehas been oxidized. For example:

2H2

+ 02 4-- 2H20

2H2

- 4e 4-- 4E+1 hydrogen oxidized

02 + 4e "4:- 20 2 oxygen reduced

Chemical reduction of oxygen may alsobe accomplished by electrons suppliedto a noble metal electrode by a batteryor other energizer.

6 Anode - an electrode at which oxidationof some reactable substance occurs.

7 Cathode - an electrode at whichreduction of some reactable substanceoccurs. For example in I. C. 3, thereduction of oxygen occurs at thecathode.

8 Electrochemical reaction - a reactioninvolving simultaneous conversion ofchemical energy into electrical energyor the reverse. These conversions are

Note: Mention of Commercial Products and Manufacturers DoesNot Imply Endorsement by the Environmental ProtectionAgency.

4 4'CH. 0. do. 32a. 11. 77

8-1

,...Dissolved Oxygen Determination

equivalent in terms of chemical andelectrical energy and generaiiy arereversible.

9 Electrolyte a solution, gel, or mixturecapable of conducting electrical energyand serving as a reacting media forchemical changes. The electrolytecommonly contains an appropriateconcentration of selected mobile ionsto promote the desired reactions.

10 Electrochemical cell - a device con-sisting of an electrolyte in which 2electrodes are inunert:0,yd connectedvia an external metallic ccoductor.The electrodes may be in separatecompartments connected by tube con-taining electrolyte to complete theinternal circuit.

a Galvanic (or voltaic) cell - anelectrochemical cell operated insuch a way as to produce electricalenergy from a chemical change,suct, as a battery (See Figure 1).

GALVANIC CELL

non 1

2

1

b Polarographic (electrolytic) cell -an electrochemical cell operated insuch a way as to produce a chemicalchange from electrical energy(See Figure 2).

POLAROGRAPHIC CELL

Nen 2

D As indicated in I. C. 10 the sign of anelectrode may change as a result of theoperating mode. The conversion by thereactant of primary interest at a givenelectrode therefore designates terminologyfor that electrode and operating mode.In electronic oxygen analyzers, theelectrode at which oxygen reduction occursis designated the cathode.

E Each cell type has characteristic advantagesand limitations. Both may be usedeffectively.

1 The galvanic cell depends uponmeasurement of electrical energyproduced as a result of oxygen

Dissolved Ox en Determination

reduction. If the oxygen content of thesample is negligible, the measuredcurrent is very low and indicator drivingforce is negligible, therefore responsetime is longer.

2 The polarographic cell uses a standingcurrent to provide energy for oxygenreduction. The indicator responsedepends upon a change in the standingcurrent as a result of electronsreleased during oxygen reduction.Indicator response time therefore isnot dependent upon oxygen concentration.

3 Choice may depend upon availability,habit, accessories, or the situation.In each case it is necessary to usecare and judgment both in selectionand use for the objectives desired.

II ELECTRONIC MEASUREMENT OF DO

A Reduction of oxygen takes place in twosteps as shown in the following equations:

1 02 + 2H 20 + 2e H202 + 20H

2 H202 + 2e --0 20H

Both equations require electron input toactivate reduction of oxygen. The firstreaction is more important for electronicDO measurement because it occurs at apotential (voltage) which is below thatrequired to activate reduction of mostinterfering components (0.3 to 0.8 voltsrelative to the saturated calomel electrode -SCE). Interferences that may be reducedat or below that required for oxygenusually are present at lower concentrationsin water or may be minimized by the useof a selective membrane or other means.When reduction occurs, a definite quantityof electrical energy is produced that isproportional to the quantity of reductantentering the reaction. Resulting currentmeasurements thus are more specific foroxygen reduction.

B Most electronic measurements of oxygenare based upon one of two techniques forevaluating oxygen reduction in line with

equation II. A .1. Both require activatingenergy, both produce a current propor-tional to the quantity of reacting reductant.The techniques differ in the means ofsupplying the activating potential; oneemploys a source of outside energy, theother uses spontaneous energy producedby the electrode pair.

1 The polarographic oxygen sensorrelies upon an outside source ofpotential to activate oxygen reduction.Electron gain by oxygen changes thereference voltage.

a Traditionally, the dropping mercuryelectrode (DME) has been used forpolarographic measurements. Goodresults have been obtained for DOusing the DME but the difficulty ofmaintaining a constant mercury droprate, temperature control, andfreedom from turbulence makes itimpractical for field use.

b Solid electrodes are attractivebecause greater surface areaimproves sensitivity. Poisoningof the solid surface electrodes isa recurrent problem. The use ofselective membranes over noblemetal electrodes has minimizedbut not eliminated electrode con-tamination. Feasibility has beenimproved sufficiently to make thistype popular for regular use.

2 Galvanic oxygen electrodes consist ofa decomposable anode and a noblemetal cathode in a suitable electrolyteto produce activating energy for oxygenreduction (an air cell or battery). Leadis commonly used as the anode becauseits decomposition potential favorsspontaneous reduction of oxygen. Theprocess is continuous as long as leadand oxygen are in contact in the electrolyteand the electrical energy released atthe cathode may be dissipated by anoutside circuit. The anode may beconserved by limiting oxygen availability.Interrupting the outside circuit mayproduce erratic behavior for a timeafter reconnection. The resulting

8-3


current produced by oxygen reductionmay be converted to oxygen concen-tration by use of a sensitivity coefficientobtained during calibration. Provisionof a pulsed or interrupted signal makesit possible to amplify or control thesignal and adjust it for direct readingin terms of oxygen concentration or tocompensate for temperature effects.

III ELECTRONIC DO ANALYZERAPPLICATION FACTORS

A Polarographic or galvanic DO instrumentsoperate as a result of oxygen partialpressure at the sensor surface to pr ducea signal characteristic of oxygen reducedat the cathode of some electrode pair.This signal is conveyed to an indicatingdevice with or without modification forsensitivity and temperature or otherinfluences depending upon the instrumentcapabilities and intended use.

1 Many approaches and refinements havebeen used to improve workability,applicability, validity, stability andcontrol of variables. Developmentsare continuing. It is possible to producea device capable of meeting any reasonablesituation, but situations differ.

2 Most commercial DO instruments aredesigned for use under specified con-ditions. Some are more versatile thanothers. Benefits are commonly reflectedin the price. It is essential to deter-mine the requirements of the measure-ment situation and objectives for use.Evaluation of a given instrument interms of sensitivity, response time,portability, stability, servicecharacteristics, degree of-automation,and consistency are used for judgmenton a cost/benefit basis to select themost acceptable unit.

B Variables Affecting Electronic DOMeasurement

1 Temperature affects the solubility ofoxygen, the magnitude of the resultingsignal and the permeability of the

8-4

protective membrane. A curve ofoxygen solubility in water versusincreasing temperature may be concavedownward while a similar curve ofsensor response versus temperatureis concave upward. Increasingtemperature decreases oxygen solubilityand increases probe sensitivity andmembrane permeability. Thermistoractuated compensation of proberesponse based upon a linear relation-ship or average of oxygen solubilityand electrode sensitivity is not preciselycorrect as the maximum spread incurvature occurs at about 170 C withlower deviations from linearity aboveor below that temperature. If theinstrument is calibrated at a temperaturewithin + or - 50C of working temperature,the compensated readout is likely to bewithin 2% of the real value. Dependingupon probe geometry, the laboratorysensor may require 4 to 6% correctionof signal per 0C change in liquidtemperature.

2 Increasing pressure tends to increaseelectrode response by compressionand contact effects upon the electrolyte,dissolved gases and electrode surfaces.As long as entrained gases are notcontained in the electrolyte or underthe membrane, these effects arenegligible.

Inclusion of entrained gases results inerratic response that increases withdepth of immersion.

3 Electrode sensitivity changes occur asa result of the nature and concentrationof contaminants at the electrode sur-faces and possible physical chemical orelectronic side reactions produced.These may take the form of a physicalbarrier, internal short, high residualcurrent, or chemical changes in themetal surface. The membrane isintended to allow dissolved gas pene-traticti but to exclude passage of ionsor particulates. Apparently some ionsor materials producing extraneous ionswithin the electrode vicinity are ableto pass in limited amouniz which


become significant in time. Dissolvedgases include 1) oxygen. 2) nitrogen,3) carbon dioxide, 4) hydrogen sulfide,and certain others. Item 4 is likely tobe a major problem. Item 3 may pro-duce deposits in alkaline media; mostelectrolytes are alkaline or tend tobecome so in line with reaction II.A. 1.The usable life of the sensor varieswith the type of electrode system,surface area, amount of electrolyteand type, membrane characteristics,nature of the samples to which thesystem is exposed and the length ofexposure. For example, galvanicelectrodes used in activated sludgeunits showed that the time betweencleanup was 4 to 6 months for electrodesused for intermittent daily checks ofeffluent DO; continuous use in the mixedliquor required electrode cleanup in 2to 4 weeks. Each electrometric cellconfiguration and operating mode hasits own response characteristics.Some are more stable than others.It is necessary to check calibrationfrequency required under conditionsof use as none of them will maintainuniform response indefinitely. Cali-bration, before and after daily use isadvisable.

4 Electrolytes may consist of solutionsor gels of ionizable materials such asacids, alkalies or salts. Bicarbonates,KCI and K1 are frequently used. Theelectrolyte is the transfer and reactionmedia, hence, it necessarily becomescontaminated before damage to theelectrode surface may occur. Electro-lyte concentration, nature, amount andquality affect response time, sensitivity,stability, and specificity of the sensorsystem. Generally a small quantity ofelectrolyte gives a shorter responsetime and higher sensitivity but also maybe affected to a greater extent by agiven quantity of contaminating sub-stances.

5 Membranes may consist of teflon,polyethylene, rubber, and certainother polymeric films. Thicknessmay vary from 0.5 to 3 mils (inches X1/1000). A thinner membrane will

aecrease response time and increasesensitivity but is less selective andmay be ruptured more easily. Thechoice of material and its uniformityaffects response time, selectivity anddurability. The area of the membraneand its permeability are directlyrelated to the quantity of transportedmaterials that may produce a signal.The permeability of the membranematerial is related to temperature andto residues accummulated on themembrane surface or interior. Acloudy membrane usually indicatesdeposition and more or less loss ofsignal.

6 Test media characteristics control theinterval of usable life between cleaningand rejuvenation for any type ofelectrode. More frequent cleanup isessential in low quality waters than forhigh quality waters. Reduced sulfurcompounds are among the moretroublesome contaminants. Salinityaffects the partial pressure of oxygena+ any given temperature. This effectis small compared to most othervariables but is significant if salinitychanges by more than 500 mg/l.

7 Agitation of the sample in the vicinityof the electrode is important becauseDO is reduced at the cathode. Underquiescent conditions a gradient indissolved oxygen content would beestablished on the sample side of themembrane as well as on the electrodeside, resulting in atypical response.The sample should be agitatedsufficiently to deliver a representativeportion of the main body of the liquidto the outer face of the membrane.It is commonly observed that noagitation will result in a very low ornegigible response after a short periodof time. Increasing agitation will causethe response to rise gradually untilsome minimum liquid velocity is reachedthat will not cause a further increasein response with increased mixingenergy. It is important to checkmixing velocity to reach a stable highsignal that is independent of a reasonablechange in sample mixing. Excessive

8-5

,7


mixing may create a vortex and exposethe sensing surface to air rather thansample liquid. This should be avoided.A linear liquid velocity of about 1 ft/secat the sensing surface is usuallyadequate.

8 DO sensor response represents apotential or current signal in themilli-volt or milli-amp range in ahigh resistance system. A high qualityelectronic instrument is essential tomaintain a usable signal-to-noise ratio.Some of the more common difficultiesinclude:

a Variable line voltage or low batteriesin amplifier power circuits.

b Substandard or unsteady amplifieror resistor components.

c Undependable contacts or junctionsin the sensor, connecting cables, orinstrument control circuits.

d Inadequately shielded electroniccomponents.

e Excessive exposure to moisture,fumes or chemicals in the wrongplaces lead to stray currents,internal shorts or other malfunction.

C Desirable Features in a Portable DOAnalyzer

1 The unit should include steady stateperformance electronic and indicatingcomponents in a convenient but sturdypackage that is small enough to carry.

2 There should be provisions for additionof special accessories such as bottleor field sensors, agitators, recorders,line extensions, if needed for specificrequirements. Such additions shouldbe readily attachable and detachableand n -Lintain good working characteristics.

3 The instrument should include asensitivity adjustment which uponcalibration will provide for directreading in terms of mg of DO/liter.

4 Temperature compensation and temp-erature readout should be incorporated.

5 Plug in contacts should be positive,sturdy, readily cleanable and situatedto minimize contamination. Waterseals should be provided wherenecessary.

6 The sensor should be suitably designedfcCrthe purpose intended in terms of

/ sensitivity, response, stability, andprotection during use. It should beeasy to clean, and reassemble for usewith a minimum loss of service time.

7 Switches, connecting plugs, and con-tacts preferably should be located onor in the instrument box rather thanat the "wet" end of the line near thesensor. Connecting cables should bemultiple strand to minimize separatelines. Calibration controls should beconvenient but designed so that it isnot likely that they will be inadvertentlyshifted during use.

8 Agitator accessories for bottle useimpose special problems because theyshould be small, self contained, andreadily detachable but sturdy enoughto give positive agitation and electricalcontinuity in a wet zone.

9 Major load batteries should berechargeable or readily replaceable.Line operation should be feasiblewherever possible.

10 Service and replacement parts avail-ability are a primary consideration.Drawings, parts identification andtrouble shooting memos should beincorporated with applicable operatinginstructions in the instrument manualin an informative organized form.

D Sensor and Instrument Calibration

The instrument box is likely to have someform of check to verify electronics,battery or other power supply conditionsfor use. The sensor commonly is notincluded in this check. A known reference

J

sample used with the instrument in anoperating mode is the best availablemethod to compensate for sensor variablesunder use conditions. It is advisable tocalibrigke before and after daily use undertest conditions. Severe conditions, changesin conditions, or possible damage call forcalibrations during the use period. Thereadout scale is likely to be labeled -calibration is the basis for this label.

The following procedure is recommended:

1 Turn the instrument on and allow it toreach a stable condition. Perform therecommended instrument check asoutlined in the operating manual.

2 The instrument check usually includesan electronic zero correction. Checkeach instrument against the readoutscale with the sensor immersed in anagitated solution of sodium sulfitecontaining sufficient cobalt chloride tocatalyze the reaction of sulfite andoxygen. The indicator should stabilizeon the zero reading. If it does not, itmay be the result of residual or straycurrents, internal shorting in theelectrode, or membrane rupture.Minor adjustments may be made usingthe indicator rather than the electroniccontrols. Serious imbalance requireselectrode reconditioning if the electronicchick is 0. K. Sulfite must be carefullyrinsed from the sensor until the readoutstabilizes to prevent carry over to thenext sample.

3 Fill two DO bottles with replicatesamples of clarified water similar tothat to be tested. This water shouldnot contain significant test interferences.

4 Determine the DO in one by the azidemodification of the iodometric titration.

5 Insert a magnetic stirrer in the otherbottle or use a probe agitator. Startagitation after insertion of the sensorassembly ana note the point ofstabilization.


a Adjust the instrument calibrationcontrol if necessary to comparewith the titrated DO.

h If sensitivity adjustment is notpossible, note the instrumentstabilization point and designateit as ua. A sensitivity coefficient,

is equal to 50 where DO is thetitrated value for the sample onwhich ua was obtained. An unknown

DO then becomes DO = lla This

factor is applicable as long as thesensitivity does not change.

6 Objectives of the test program and thetype of instrument influence calibrationrec.: irements. Precise work mayrequire calibration at 3 points in theDO range of interest instead of at zeroand high range DO. One calibrationpoint frequently may be adequate.

Calibration of a DO sensor in air is aquick test for possible changes insensor response. The difference inoxygen content of air and of water istoo large for air calibration to besatisfactory for precise calibrationfor use in water.

IV This section reviews characteristics ofseveral sample laboratory instruments.Mention of a specific instrument does notimply USEPA endorsement or recommendation.No attempt has been made to include all theavailable instruments; those described areused to indicate the approach used at onestage of development which may or may notrepresent the current available model.

A The electrode described by Carrit andKanwisher (1) is illustrated in Figure 3.This electrode was an early example ofthose using a membrane. The anode wasa silver - silver oxide reference cell witha platinum disc cathode (1-3 cm diameter).The salt bridge consisted of N/2 KC1 and

8-7

Dissolved Ox Determination

KOH. The polyethylene membrane washeld in place by a retaining ring. Anapplied current was used in a polarographicmode. Temperature effects were relativelylarge. Thermistor correction was studiedbut not integrated with early models.

Lytatienne Oise Ltlectraire Layer

Figure 3

B The Beckman oxygen electrode is anotherillustration of a polarographic DO sensor(Figure 4). It consists of a gold cathode,a silver anode, an electrolytic gel con-taining KC1, covered by a teflon membrane.The instrument had a temperature readoutand compensating thermistor, a sourcepolarizing carrent, amplifier with signaladjustment and a readout DO scale withrecorder contacts.

SENSOR ELECTRONICS

rye*, ..C.

71Orat101

.

'1C f 10.111 aH

17f. olfCf

-V;;3r f10Cov 161 f1riffle 10111

010 C 1110111

11C0101111 USIO

Figure 4, THE BECKMAN OXYGENSENSOR

8

C The YSI Model 51 (3) is illustrated inFigure 5. This is another form ofpolarographic DO analyzer. The cellconsists of a silver anode coil, a goldring cathode and a KC1 electrolyte witha teflon membrane. The instrument hasa sensitivity adjustment, temperature andDO readout. The model 51 A has temp-erature 'compensation via manual presetdial. A field probe and bottle probe areavailable.

'O' Ring

Matinbrano

KC1 Solution

Anode Coil

YS1 Model 51 DO Sensor

Mar.

D The Model 54 YSI DO analyzer (4) is basedupon the same electrode configuration butmodified to include automatic temperaturecompensation, DO readout, and recorderjacks. A motorized agitator bottle probeis available for the Model 54 (Figure 6)

VP 8161 M 11.1 411Cre


E The Galvani.,:. Cell Oxygen Analyzer (7, 8)employs an indicator for proportional DOsignal but does not include thermistorcompensation or signal adjustment.Temperature readout is provided. Thesensor includes a lead anode ring, anda silver cathode with KOH electrolyte(4 molar) covered by a membrane film(Figure 7).

Preeisti OeivNie Coll °xylem 'VIA,"

1111v.r Cathd

\ I

Cet. L..4

PIythyleft Mentlornnpra

Thormalstr CamlaIttelevise

?aowed Soule.hint SOO Settles

Plastic *womb, IMO

tetalor *l.

teed Awed. Ring

F The Weston and Stack Model 300 DOAnalyzer (8) has a galvanic type sensorwith a pulsed current amplifier adjustmentto provide for signal and temperaturecompensation. DO and temperaturereadout is provided. The main powersupply is a rechargeable battery. Thesensor (Figure 8) consists of a lead anodecoil recessed in the electrolyte cavity(50% 111) with a platinum cathode in the tip.The sensor is covered with a teflon mem-brane. Membrane retention by rubberband or by a plastic retention ring may beused for the bottle agitator or depthsampler respectively. The thermistorand agitator are mounted in a sleeve thatalso provides protection for the membrane.

G The EIL Model 15 A sensor is illustratedin Figure 9. This is a galvanic cell withthermictor activated temperature com-pensation and readout. Signal adjustmentis provided. The illustration shows anexpanded scheme of the electrode whichwhen assembled compresses into a sensorapproximating 5/8 inch diameter and 4 inchlength exclusive of the enlargement at theupper end. The anode consists of com-pressed lead shot in a replaceable capsule(later models used fine lead wire coils),a perforated silver cathode sleeve aroundthe lead is covered by a membrane film.The electrolyte is saturated potassiumbicarbonate. The large area of leadsurface, silver and membrane providesa current response of 200 to 300 micro-amperes in oxygen saturated water at200C for periods of up to 100 days use (8).The larger electrode displacement favorsa scheme described by Eden (9) forsuccessive DO readings for BOD purposes.

V Table 1 summarizes major characteristicsof the sample DO analyzers described inSection IV. It must be noted that an ingeniousanalyst may adapt any one of these for specialpurposes on a do-it-yourself program. Thesample instruments are mainly designed forlaboratory or portable field use. Thosedesigned for field monitoring purposes mayinclude similar designs or alternate designsgenerally employing larger anode, cathode,and electrolyte capacity to approach betterresponse stability with some sacrifice inresponse time and sensitivity. The electroniccontrols, recording, telemetering, andaccessory apparatus generally are semi-permanent installations of a complex nature.

A =IOW LEDGMEN'TS:

This outline contains certain materials fromprevious outlines by D. G. Ballinger,N. C. Malof, and J. W Mandia Additionalinformation was provided by C.R. (-firth,C. N. Shadix, D. F. Krawczyk, J. Woods,and others.

9

Dissolved Determination

WESTON & STACKDO PROBE

CORD

V'--- CORD RESTRAINER

SERVICE CAP

PROBE SERVICE CAP

ELECTROLYTE FILL. SCREW

PROBE BODY

PLA T IN U M CATHODE

CONNECTOR PINS

PIN HOUSING

LEAD ANODE

REMOVABLE PROBE SHIELDAND THERMISTOR HOUSING

Figure 8

5-;

Cable SealingNut

A 15017

CableConnection

CoverA15016A

Cable

'0' RingR385

5J

"/Menrats) Securing

'0' Ring\-1317

'0' RingR524

pedal AM_ ELECTRODE COMPONENT PARTS

Lead AnodeComplete

(A15024A)

'0' Ring Si:ver-CathodeR389 A15013A

Membrane Securing'0' Ring

R 317

Filler Screw1471

AnodeContact

A150140(With Sleeve S24)

AnodeContactHolder

A15015A

'0' RingR622

'0' RingR 612

End CapA15011A

'0' RingR622

ca0Hma

Note: Red wire of cable connects to Anode Contact Holder

Black wire of cable connects to Anode Contact t73

Membrane not shown E. I. L. part number T22

Figure 95 J


TABLE 1

CHARACTERISTICS OF VARIOUS LABORATORY DO INSTRUMENTS

Anode Cathode Elec Type Membr

DOSig.Adj.

Temp.Comp.Temp. Rdg.

Carrit &Kanwisher

silver-silver ox.rin

Ptdisc

KC1KOHN/2

pol* polyeth no no

Beckman Aqring

Audisc

KCIgel

pol teflon yes yesyes

Yellow Springs51

Agcoil

Auring

KC1solnsat.

pol teflon yes no*yes

Yellow Springs54

II IT II IT If yes yesyes

PrecisionSci

Pbring

silverdisc

KOH4N

galv `*polyeth no noyes

Weston &Stack300

Pbcoil

Ptdisc

KI40%

galv teflon yes yesyes

EIL Pb Ag KHCO3

galv teflon yes vf.

yesDelta

75Lead Silver

discKOH1N

galv teflon yes yesno

Delta85

Lead Silverdisc

KOH1N

galv teflon yes yesyes

*Poi - Polarographic (or amperometric)4-Ga1v - Galvanic (or voltametric)

REFERENCES

1 Carrit, D.E. and Kanwisher, J.W.Anal. Chem. 31:5. 1959.

2 Beckman Instrument Company. Bulletin7015, A Dissolved Oxygen Primer,Fullerton, CA. 1962.

3 Instructions for the YSI Model 51 OxygenMeter, Yellow Springs InstrumentCompany, Yellow Springs, OH 45387.

4 Instructions for the YSI Model 54 OxygenMeter, Yellow Springs InstrumentCompany, Yellow Springs, OH 45387.

8-12

Accessories forwhich designed

Recording temp.& signal adj. selfassembledrecording

field and bottleprobe

recording fieldbottle & agitator_probes

a git . probedepth sampler

recording

field bottle &agitator probefield bottle &agitator probe

5 Technical Bulletin TS-68850 PrecisionScientific Company, Chicago, IL 60647.

6 Mancy, K. H. , Okun, D. A . and Reilley,C.N. J. Electroanal. Chem. 4:65.1962.

7 Instruction Bulletin, Weston and StackModel 300 Oxygen Analyzer. Roy F.Weston, West Chester, PA 19380.

8 Briggs, R. and Viney, M. Design andPerformance of Temperature Com-pensated Electrodes for OxygenMeasurements. Jour. of Sci.Instruments 41:78-83. 1964.

Go

9 Eden. R. E. BOD Determination Usinga Dissolved Oxygen Meter. WaterPollution Control. pp. 537-539. 1967.

10. Skoog, D.A. and West, D. M. Fundamentalsof Analytical Chemistry. Holt,Rinehart & Winston, Inc. 1966.

i ' 1u i


Li Methods for Chemical Analysis ofWater & Wastes, U. S. EnvironmentalProtection Agency, EnvironmentalMonitoring & Support Laboratory,Cincinnati, Ohio 45268, 1974

This outline was prepared by F. J. Ludzack. formerChemist, National Training Center. MOTD. OWPO.USEPA, Cincinnati, OH 45268 and NateMalof, Chemist, USEPA, OWPO, NationalField Investigations Center, Cincinnati, OH

Descriptors : Chemical Analysis, DissolvedOxygen, Dissolved Oxygen Analyzers,Instrumentation, On-Site Tests, Water Analysis.Analysis, Wastewater, Oxygen

8-13

r

LABORATORY PROCEDURE FOR DISSOLVED OXYGENWinkler Method-Azide Modification

I APPLICABILITY

A The azide modification is used formost wastewaters and streamswhich contain nitrate nitrogenand not more than 1 mg of ferrousiron/1. If 1 ml 40% KF solution isadded before acidifying the sample andthere is no delay in titration, the methodis also applicable in the presence of100-200 mg ferric iron/1.

B Reducing and oxidizing materialsshould be absent.

C Other materials which interfere withthe azide modification are: sulfite,thiosulfate, appreciable quantities offree chlorine or hypochlorite, highsuspended solids, organic substancesreadily oxidized in a highly alkalinemedium, organic substances readilyoxidized by iodine in an acid medium,untreated domestic sewage, biologicalflocs, and color which may interferewith endpoint detection. A dissolvedoxygen, meter should be used whenthese materials are present in thesample.

II REAGENTS

Distilled water is to be used for thepreparation of all solutions.

A Manganous Sulfate Solution

Dissolve 480 g MnSO: 4H 0 (or 400 gMnS0

42H20, or 3644g MilSO4 H2O) in

water and dilute to 1 liter.

B Alkaline-Iodide-Azide Solution

Dissolve 500 g sodium hydroxide (or700 g potassium hydroxide) and 135 gsodium iodide (or 150 g potassium iodide)in water and dilute to 1 liter. To thissolution add 10 g of sodium azidedissolved in 40 ml water.

C Sulfuric Acid, Conc.

The streng1h of this acid is 36 N.

CH. O. do. lab. 3d. 8.78

Ct̂i

D Starch Solution

Prepare an emulsion of 10 g of solublestarch in a mortar or beaker with asmall quantity of water. Pour thisemulsion into 1 liter of boiling water,allow to boil a few minutes, and letsettle overnight. Use the clear supernate.This solution maybe preserved by theaddition of 5 ml per liter of chloroformand storage in a refrigerator at 10°C.

E Sodium Thiosulfate Stock Solution 0.75 N

Dissolve 186.15 g Na2S9035H20 in boiledand cooled water and dilute to 1 liter.Preserve by adding 5 ml chloroform.

F Sodium Thiosulfate Standard Titrant 0.0375N

Dilute 50.0 ml of stock solution to 1 liter.Preserve by adding 5 ml of chloroform.

G Potassium Biiodate Solution 0.0375N

Dry about 5 g of KH (101)2 at 103°C fortwo hours and cool in a 3desiccator.Dissolve 4.873 g of the solid in water anddilute to I liter. Dilute 250 ml of thissolution to 1 liter.

H Sulfuric Acid Solution 10%

Add 10 ml of conc sulfuric acid to 90 mlof water. Mix thoroughly and cool.

1 Potassium Iodide Crystals

III STANDARDIZATION OF THE TITRANT

A Dissolve 1-3 g of potassium iodide in100-150 ml of water.

B Add 10 ml of 10% sulfuric acid and mix.

C Pipet in 20 ml of the 0.0375N potassiumbiiodate and mix. Place in the dark for5 minutes.

D Titrate with the 0.0375N sodiumthiosulfate standard titrarit to theappearance of a pale yellow color.

9-1

Laboratory Procedure for Dissoled 0

Mix the solution thoroughly during thetitration.

E Add 1-2 ml of starch solution and mix.The solution is now blue in color.

F Continue the addition of the titrant.with thorough mixing, until thesolution turns colorless.

G Record the ml of titrant used.

H Calculate the N of the sodiumthiosulfate standard titrant. It will beapproximately 0.0375.

N = (ml x N) of the biiodateml of titrant

= 20.0 x 0.0375ml of titrant

= 0.75ml of titrant

IV PROCEDURE

A Addition of Reagents

1 Manganous sulfate and alkalineiodide-azide

To a full BOD bottle (300 ml 3 ml),add 2 ml manganous sulfate solutionand 2 ml alkaline-iodide azide reagentwith the tip of each pipette below thesurface of the liquid.

2 Stopper the bottle without causingformation of an air bubble.

3 Rinse under running water.

4 Mix well by inverting 4-5 times.

5 Allow the precipitate to settle untilat least 200 ml of clear supernatehave been produced.

6 Repeat steps 4 and 5.

7 Add 2 ml conc. sulfuric acid with thetip of the pipette above the surface ofthe liquid.

8 Storper the bottle without causingformation of an air bubble.

9 Rinse under running water.

10 Mix by inverting several times todissolve the precipitate.

11 Pour contents of bottle into a wide-mouth 500 ml Erlenmeyer flask.

B TITRATION

1 Titrate with 0. 0375N thiosulfate to apale yellow color.

2 Add 1-2 ml starch solution and mix.

3 Continue the addition of the titrant,with thorough mixing, until thesolution turns colorless.

4 Record the ml of titrant used.

C CALCULATION

mg DO/1 = ml titrant x N titrant x 8 x 1000ml sample

If the N of the titrant A actly = 0,0375,

mg DO/1 = ml titrant x 0.0375 x 8 x 1000300

ml titrant x 1

= ml titrant

REFERENCEMethods fnr Chemical Analysis of Water& Wastes, U.S. Environmental ProtectionAgency, Environmental Monitoring &

Support_Laboratory, Cincinnati, Ohio 45268, 1974

This outline. was prepared by C. R. Feldmann,Chemist, National Training and Operational VPO,Technology Center, MOTD, OWPO, USEPA,Cincinnati, Ohio 45268

Descriptors: Analytical Techniques,Chemical Analysis, Dissolved Oxygen,Laboratory Tests, Oxygen, Water Analysis

Laboratory Procedure for Dissolved Oxygen

DATA SHEET

ml of titrant =

N of titrant =

ml of sample =

mg =ml titrant x N titrant x 8 x 1000

ml sample

=

=

x x 8 x 1000

9-3

PRINCIPLES OF ABSORPTION SPECTROSCOPY

I INTRODUCTION

In any system employing principles ofabsorption spectroscopy, there are threebasic components.

A SOURCE of Radiant Energy

B MEDIUM (Sample) which absorbsRadiant Energy

C DETECTOR to measure the RadiantEnergy transmitted by the Sample

4r

11411fAINT IIHNINIf WNW 12111111CTOR

Figure 1. Basic Components of AbsorptioSpectroscopy System

II RADIANT ENERGY

A Wave Nature

1 The various forms of radiantenergy have been arranged in asingle schematic diagram referredto as the electromagnetic spectrum(see Figure 2). Allot the energieswhich snake up this spectrum maybe represented graphically aswaves. All waves move throughspace (and for most purposes air)at a constant velocity, 3 X 1010cintsec.

2 Three variable characteristics ofindividual waves serve to differ-entiate each from all other wavesin the spectrum.

CH. MET. as. 5c. 1. 76

a The Wave Length

- The linear distance be-tween the crests of twoadjacent waves. (Units:distance/wave. )

b The Frequency

- The number of waveswhich pass a given pointin a unit of time. (Units:waves/time unit.)

c The Wave Number

- The number of waveswhich occur in a givenlinear distance. (Units:waves/ distance unit.)

2 It is evident that more waves ofshort wavelength will "fit" into agiven linear distance than wouldwaves of a greater wavelength.Thus, waves having short wave-lengths will have higher wavenumbers. Mathematically, wavelength is the reciprocal of wavenumber, if the same units of linearmeasurement are used in eachexpression. Since the velocity ofall waves is equal and constant, itis also apparent that a gi eaternumber of waves of short wave-length can pass a given point in aunit of time than waves having alonger wavelength.

B Particle Nature

Planck conducted certain experimentswhich indicated that light has a particleas well as a wave nature. Energy rayscan be said to consist of particles witha definite amount of energy. Theseparticles or packets are referred toas photons or quanta. The energy (E)of each minute packet is given byPlanck's equation.

10-1

Principles of Absorption Spectroscopy

E by (5)

Where E. = Radiant Energy in ergs

h = Planck's proportionalityconstant (6.6 X 10-27erg sec.)

is Frequency in waves persecond

Thus, it can be seen that the energy ofa given photon is directly proportionalto the frequency of the given RadiantEnergy.

FREQUENCYCycNs(er aravess)/sec.

WAVELENGTHContisatorsjcs.)

Nanow.ters (non)

WAVE NWAIIER1 (Or waves)/css.

III ABSORPTION OF ENERGY BY ATOMSAND MOLECULES

A Absorption of energies of given fre-quencies by atoms and molecules canbe used as abasis fortheir tualitativeidentification. Absorption spectro-scopy is based on the principle thatcertain displacements of elertr-ms oratoms w ithin a molecule are per-missible according to the quantumtheory. When radiant energy of thesame energy required to bring aboutthis permissible change is supplied tothe molecule, the change occurs andenergy is absorbed.

20 16 10._ 8 104

fi52 102 6 10 14

1010 106 102 fi52 fo6

Cosmic raysX raysWtrevieletVisiblsInharaal

RadarTolovisionRadio waysElectric currant

U111111111

.11101111. Mb

1111111416011111M1

FIGURE 2

X. I I Crest

ÊINIIMINEW----1111111=11Et-

J\f\Al\Aldt Trough

0 5

(1) WAVE LENGTH

(2)

(3)

X FREQUENCY

distancewave

(4) ). (cm/wave)Fizure 3.

/4- 2

Xwavestime

X v (waves/se'

10 13

CONSTANT

distance = VELOCITY

C = (3 X 1010 cm/sec)Relationship of Wave Length and Frequency

t'rtU

Principles of Absorpon Spectroscopy

WAVELENX-Rays Ultraviolet Visible Near IR Far IR

rCalories per moleneeded for change

o

142, 000 71, 000

0 0 0 tInner Ionization Outer

Electron Shift Electron Shift

35, 000

-111.

0 °Vibration

Figure 4. Electromagnetic Spectrum Showing Energy Ranges andCorresponding Electronic, Vibrational and Rotational Motions

1, 400

0 : oRotation

I Displacement of electrons is a per-missible change which can occurwhen energy of ultraviolet an dvisible frequencies strikes certainatoms and molecules.

a Inner electron shiftElectrons located in the innerorbit of an atom may, when theproper frequency of radiantenergy is available, shift to anorbit farther removed fromthe nucleus. This shift repre-sents a change from a lowerenergy to a higher one. If thisnew position is unstable, theelectron may revert to someposition nearer the nucleus;the energy which was gainedmay then be emitted from theatom as part of its emissionspectrum. The number ofenergy changes possible withinan atom is a function of thenumber of electrons and thenumber of changes each mayenter. Each possible changegives rise to a new spectralfrequency. Since the frequencyof radiation needed to accom-plish such changes is of a highorder of magnitude, the energyused is considerable in quantity.

Molecular aggregations oftendisintegrate in such circum-

U

stances; thus, these higherfrequencies are used mainlyfor work with elements orvery stable compounds.

b Ionization

Under a specific frequency ofradiation, an electron may bephysically separated from itsparent atom. This processhas been termed ionization. Achange of energy level of thismagnitude requires-lessenergy than the inner electronshift. Such changes are characteristic of those of the rareearths, inorganic ions, tran-sition elements and m anyorganic compounds un derfrequencies within the ultra-violet range.

c Outer electron shiftThe various orbital electronsin an atom may vary in theamount of energy required toshift them outwardly from thenucleus. For example, it re-quires less energy to shift anelectron from a position moredistant than it does, to shift anelectron outwardly from theinner orbit. Outer electronshifts occur readily in coloredorganic molecules for which

10-3

Principles of Absorptjon Spectr3copy

10-4

CJ

telectronic transitions a r emade easier by the presenceof chromophore groups whichparticipate in resonance.Thus, the excitation of thedelocalized outer electrons(pi electrons) i s relativelyeasy and requires energy inthe visible range.

2 Vibration of atoms within moleculesis a permissible' change which canoccur when energy of near infraredfrequency strikes certain organicmolecules.

The atoms within a molecule areheld together by attractive bondingforces. Atoms within a moleculeare constantly moving toward andaway from other atoms, but f o rpurposes of theory can be said tohave a certain "average" position.

The change in position of an atomin relation to another atom is calledvibration. The mechanics of vibra-tion require energy; the mannerand rate of vibration of the atomsdepend upon frequencies of electro-magnetic radiation which strikesthem. Therefore, a specific partof a molecule may absorb significantquantities of certain spectral fre-quencies. Such absorption will bereflected in the absorption spectrumof the compound. The energy re-quirements for this type of energychange are of a lower order ofmagnitude than those above;therefore, we would expect that thefrequency required would be lowerand the wave length longer. Suchchanges occur in organic com-pounds under infrared radiation.

3 Rotation of molecules is a permis-sible change which can occur whenenergy of far infrared frequencystrikes lertain organic molecules.

A molecule rotates around its sym-metrical center. The manner andrate of rotation again depends uponthe energy supplied to it.

Specific spectral frequencies ofelectromagnetic radiation can beemployed to increase the rate ofrotation. The used radiation is,in effect, absorbed and reflectedin the absorption spectrum.Organic molecules utilize infra-red radiation while varying theirrate and manner of rotation.

B The Lambert-Beer Law provides thebasis for quantitative analysis b yabsorption spectroscopy. It is a com-bination of the Bouguer (or Bouguer-Lambert) and Beer Laws.

1 Bouguer (or Bouguer =Lambert) Law

When a beam of monochromaticradiation passes through an ab-sc,..bing medium, each infinitesi-mally small layer of the mediumdecreases the intensity of thebeam by a constant fraction.Mathematically:

-di k db (6)

On integration and converting basee to base 10 logarithms,

log Io = A = Kb (7)

-ai = increment by which in-cident monochromaticradiation is decreased(or absorbed) by themedium.

I = intensity of the radia-tion emerging from theabsorbing medium.

k = proportionality constantwhose value depends onthe wavelength and thenature of the medium;i.e., the solvent usedif the absorbing medi-um is a solution , andthe temperature.

db increment thickness ofthe absorbing medium.

Io= radiation entering the medium.

log a- A = absorbance(optical density)

K = 2.303kb = length of radiation passing

t h rough the medium (i. e. ,the width of the cell, gener-ally express in cm.)

2 Beer's LawEach molecule of an absorbingmedium absorbs the same fractionof radiation incident upon it regard-less of concentration.Mathematically:

-dlI

3 k' dc (8)

On integration and converting basee to base 10 logarithms,

log 9 = K' ck' = a proportionality constant

whose value is governed bythe same factors which de-termine the value of k.

(9)

dc * increment concentration ofthe absorbing medium.

K' * 2.303 k'c = concentration of the absorb-

ing medium (in the case of asolution c is generally ex-pressed in moles/liter.

3 Lambert-Beer Law

Aalog =ebc (10)to1T-

e 2 a constant obtained by com-bining K plus K'. When b isexpressed m cm and c inmoles/liter, e is called themolar absorptivity.

4 The term transmittance is some-times used to express how muchradiation has been absorbed by amedium.

,.....aulai p lig,.L.2f3ortrosconv

Transmittance (T) *I

%o Transmittance (%T) =

T^

I

lo100 (12)

The relationship between absorbanceand transmittance is given by the ex-pression:

A = log --I-T

5 The application of the Lambert-Beer Law to a nroblem involvingquantitative ..,nalysis is made bythe use of a calibration curve(or graph). See Figure 5.

Several standard solutionscontaining known concentrationsof the material under analysis - --.,are "read in the spectrophoto-meter. Figure 5 is prepared bygraphing concentrations vs. cor-responding absorbance readings.If a straight line is obtained, thematerial is said to follow Beer'sLaw in the concentration iangeinvolved. The absorbance of thesample is then "read" and thecorresponding concentration ob-tained from the calibration curve.

IncreasingConcentraqon

Figure 5.

ACKNOWLEDGMENT:

This outline contains certain portions froma previous outline by Betty Anr Punghorst,former Chemist, National airing Center.

10-5

Principles of Absorption Spectroscopy

w"t

RANGE

SOURCE OFRADIANTENERGY

ABSORPTION BY SAMPLE DETECTION OFI1A DIANT ENERGY

TRANSMITTEDCHEMICAL NATURE

OF SAMPLETYPE OF SAMPLE

CELL USED

Ultraviolet Hydrogen Arc Inorganic ions andOrganic Molecules

Quartz Fluorite

--N ,

(

Photoelectric Cells

PhotographicPlates

Visible IncandescentTungsten Bulb

Colored Inorganic andOrganic Molecules.

Glass 'Eye PhotographicPlates

Photoelectric Cells

Infrared Nernst Glower organic MoleculesGlobar Lamp

Sodium Chloride orPotassium Bromide

Thermocouple

REFERENCES

1 Delahay, Paul. Instrumental Analysis.The MacMillan Co., New York.1957.

2 Mellon, M. G. Analytical AbsorptionSpectroscopy. John Wiley & Sons,Inc. , 1950.

3 Willard, Merritt & Dean. InstrumentalMethods of Analysis. D. van No-strand Co. , Inc., 1958.

10-6

4 Dyer. Applications of AbsorptionSpectroscopy of Organic CompoundsPrentice-Hall, Inc. 1965.

This outline was prepared by C. R. Feldmann.National Training Center, MOTD, OWPO,JSEPA, Cincinnati, Ohio 45260.

Descriptors: Chemical Analysis, Water Tests,Spectroscopy, Spectrophotometry

I

USE OF A SPECTROPHOTOMETER

I SCOPE AND APPLICATION

A Colorimetry

Many water quality tests depend on atreating a series of calibration standardsolutions which contain known concentra-tions of a constituent of interest, and alsothe sample(s) with reagents to produce acolored solution. The greater the concen-tration of the constituent. the more intensewill be the resulting color. A spectro-photometer is used to measure the amountof light of appropriate wavelength which isabsorbed by equal "thicknesses" of thesolutions. Results from the standardsare used to construct a calibration (standard)curve. Then the absorbance value for thesample is located on the curve to determinethe corresponding concentration.

B Lambert Beer Law

States the applicable relationships:

A= ebc

1 A = absorbance

2 e = molar absorptivity

3 b= light path in cm

4 = concentration in moles/liter

II API- ARATUS

A Requirements

Are giver, as part of the method write-up

1 The applicable wavelength isspecified. The unit used isnanometers (nm).

2 The light path (cell dimension)is often open-ended. e. g.. "onecm or longer. " One must know

CH. IN. sp. 1. 11.77

the light path length in theavailable spectrophotometer,because it is inversely relatedto the usable cententrations inthe test. (Longer path lengthsdetect lower concentrations).

3 /12M: For National PollutantDischarge Elimination System(permit), or for Drinking WaterRegulations test requirements.check with the issuing/ reportagency before using light paths(cells) that differ from the lengthspecified in the approved method.If you have permission to use analternate path length, concentra-tions for the test can be adjustedaccordingly. These adjustmentsare discussed in IV and in VII (below).

III PREPARATION OF THE SPECTRO -PHOTOMETER

A Phototube/ Filter

1 May have to choose a phototube foruse above or below a particularwavelength.

2 A filter may be required,

3 If the available instrument involvesa choice, check that the phototube(and filter, if applicable, ) reviredfor the wavelength to be used is inthe instrument.

4 Always handle and wipe off thephototube and/or filter with tissueto avoid leaving fingerprints.

B Cell compartment

1 This area must be kept clean anddry at all times.

7 I_A_ 11 -1

USE OF A SPECTROPHQMIaIEB..,

C Cells

1 A set must "match" each otherin optical properties. To checkthis, use the same solution atthe same wavelength, arri verifythat the absorbance value is thesame for each cell.

2 Alternatively, a single cell canbe used if it is thoroughly rinsedafter each reading.

3 Instrument cells should be freeof scratches and scrupulouslyclean.

a Use detergents, organicsolvents or 1:1 nitric acid-water.

b Caustic cleaning compoundsmight etch the cells.

c Dichromate solutions arenot recommended becauseof adsorption possibilities.

d Rinse with tap, then distilledwater.

e A final rinsing and drying withalcohol or acetone beforestorage is a preferred practice.

D Warm-Up

1. Plug in the power cord.

2 Turn the power switch on and giveit an additional half-turn to keepthe needle from "pegging."

3 Wait to use until the recommendedwarm-up time has passed. Any-where from 10 to 30 minutes maybe required.

4 If the instrument drifts duringzeroing, allow a longer time.

11-2

E Wavelength Alignment

An excellent point is the-known, maximumabsorption for a dilute solution of potassiumpermanganate which has a dual peak at526 nm and 546 nm. Use 2 matched cells for thefollowing steps:1 Prepare a dilute solution of

potassium permanganate (about10 mg/1).

2 Follow the steps in VI A. ZeroingOperation, using a wavelength of510 nm, and distilled water as a"reagent blank." Keep the water in the cellduring this entire procedure.

3 Rinse the matched cell two timeswith rap water, then two times withthe permanganate solution.

4 Fill the cell three-fourths fullwith the permanganate solution. Keep thepermanganate solution in this cell during thisentire procedure.

5 Thoroughly wipe the cell witha tissue. Hold the cell by thetop edgeq.

6 Open the cover and gently insertthe cell, aligning it to the ridgeas before.

7 Close the cover.

8 Record the wavelength and theabsorbance reading on a sheet ofpaper.

9 Remove the cell of permanganatesolution and close the cover.

10 Set the wavelength control at thenext graduation (+ Snm).

11 If the needle is not at infinite(symbol co) absorbance, use theleft knob to re-set it.

12 insert the cell containing distilledwater using the techniques noted In5. 6 and 7 above.

13 If necessary, use the right knobto re-set the needle at zero absorbance.

14 Remove the cell and insert the cellof permanganate solution using thetechniques noted in 5, 6 and 7 above.

15 Record tne wavelength and theabsorbance reading.

16 Repeat steps 9 through 15 aboveuntil absorbance readings arerecorded at 5 nm incrementsfrom 510 nm through 560 nm.

17 Make a graph plotting absorbancereadings against wavelengths.With very good resolution, therewill be two peaks - one at 525 amand one at 545 nm. A single flattopped "peak" between these twowavelengths is acceptable.

18 If the maximum absorbances(Peak(sioccur below or above 526 nm or546 rim and at a number of scaleunits different from the statedinstrument accuracy, the wavelengthcontrol is misaligned. To compensatefor this until the instrument can beserviced, add or subtract the errorscale units when setting wavelengthsfor subsequent tests.

IV CALIBRATION STANDARDS

A Requirements

A set of calibration standards is required.with concentrations usable in the availablespectrophotometer cell (light path length).

1 The method reference may providea table of light path lengths andthe corresponding applicable con-centration range for calibrationstandards, so one can choose therange required for his instrumentcell or sample concentration.

2 The method reference may givedirections for preparing one rangeof concentrations for a given lightpath length. If your cell providesa different length, your concentrationrequirements can be easily calculatedby recalling that the light path lengthis inversely related to concentration.Thus, if your cell is twice the givenpath length, you need the given con-centrations divided by two.

3 The method reference may givedirections for preparing only onerange of concentrations for thecalibration standards, and thennot be specific about the associated


path length. You will have to testif the range is applicable to yourinstrument by preparing the givenconcentrations, obtaining absorbancevalues for them and checking theresults according to section VII(below).

B PreparationThe calioration standards required forspectrophotometric measurements are sodilute, that they are commonly prepared bydiluting stronger solutions. These are des-cribed in general terms below. Weights andvolumes involved in preparing these solutionsfor a specific test can be found in the methodwrite-up.

1 Stock Solutions

a Prepare by weighing or measuringa small amount of a chemicalcontaining the constituent ofinterest and dissolving it to aone liter volume.

b Common stock solutions haveconcentrations in the range of0.1 to 1.0 mg /mi.

c Most are refrigerated for storageand some are further treated byadding a preservative. Many arestable up to six months.

2 Standard Solutions

a Prepare by diluting a stocksolution (at room temperature).Common volumes are 10.0 or 20.0 mlof stock diluted to one liter.

b Resulting standard solutions haveconcentrations in the range of1.0 to 10.0 ug/nal.

c Although some standard solutionsmay he stable for a period of time,it is a recommended practice toprepare them on the day of use.

3 Calibration (Working) Standard Solutions

a Prepare by diluting a standardsolution. Usually a set of cali-bration standards is requiredso that resulting concentrationsgive five to seven results withinthe sensitivity limits of the instru-ment. Common volumes are 1 to 10 mlof standard solution diluted to 100 ml.

11-3

USE OF A SPECTOPHOTOMETER

b Resulting solutions mighthave concentrations in therange of 0.01 and 1.0 ugiml.

c A reagent blank (distilledwater) should be includedin the set of standards.

4 Adjusting Concentrations

a You may find it necessaryto adjust preparation quantitiesgiven in the method write-up,because your cell (light pathlength) differs from theexample.

b These adjustments arediscussed in A Requirements(above), and are usuallyapplied to the Standard(intermediate) Solution.

C Chemical Treatment

1 Most colorimetric methodsrequire, that the calibrationstandards (including thereagent blank) are to betreated as the sample. Thus,they are to be processedthrough pretreatments andthrough the test as if theywere samples. Then anytest effects on sample resultswill be compensated by thesame effects on results ob-tained for the treated standards.

2 One should be aware that pH isa critical condition for mostcolorimetric reactions.Ordinarily, a pH adjustmentis included in the test methodand reagents include chemicalsto control pH. Thus,the pro-cessed standards correspond tothe samples regarding pH, andthus they correspond in degreeof color development. If stand-ards are processed in someother manner, they must bepH adjusted to correspond tothe samples at the time ofcolor development.

11-4

V SAMPLE DILUTIONS

A Concentration Limits

The concentration of the sample mustresult in an absorbance within the rangeof the calibration standards, i.e., accu-rately detectable in the light path providedby the instrument. A dilution before analysismay be required to accomplish this. It isnot accurate to dilute a sample after pro-cessing in order to obtain a usable absorb-ance reading.

1 Record dilution volumes so a dilutionfactor can be calculated and appliedto results.

2 An analyst often has a good estimate ofthe expected concentration of a sampleif s/he routinely tests samples fromthe same source. In this case, a singledilution, if any, is usually sufficient.

3 If a sample is from an unknown source,the analyst has several choices.

a Process the sample. If the readingshows it is too concentrated, diluteit until you get a value in the usablerange. This result is not accurateenough to report, but you now knowhow to dilute the sample to processit through the test to get usable re-sults.

b Prepare at least a 50% dilution andanalyze it plus an undiluted aliquot.

c Prepare a variety of dilutions.

d Use some other analytical methodto get a rough estimate of the ex-pected concentration.

B Final Volumes

1 Dilute to a final volume sufficient to rinsethe measuring glassware and provide thetest volume cited in the referenced method.

2 Save any undiluted sample.

4

I

I

I

I

I

USE OF A SPECTOPHOTOMETER

VI PROCEDURE FOR USING A SPECTRO-PHOTOMETER

A Zeroing Operation

The following steps have been writtenfor spectrophotometers used in thiscourse. Check the manual for theavailable instrument for the stepsapplicable for your work.

1 Set the wavelength control tothe setting specified for thestandards you are testing.Approach the setting by be-ginning below the numberand dialing up to it.

2 If a cell is in the holder, re-move it.

3 Close the cell holder cover.

4 Turn the power switch/zerocontrol (left) knob until theneedle reads infinite (symbol 03 )absorbance (on the lower scale).

5 Rinse a cell two times withtap water, two times withdistilled water, then twotimes with the reagent blanksolution.

8 Fill the cell about three-fourths full with reagentblank solution.

7 Thoroughly wipe the outsideof the cell with a tissue toremove fingerprints and anyspilled solution. Hold thecell by the top edges.

8 Open the cell holder cover andgently slide the cell down intothe sample holder.

9 Slowly rotate the cell untilthe white vertical line onthe cell is in line with theridge on the edge of thesample holder.

limurirarm...............................r..-.......m.. 111.1.

10 Close the cover and turn the lightcontrol (right) knob until the needlereads zero absorbance (on thelower scale).

11 Record an absorbance of zero forthis zero concentration solution ona data sheet. (See next page).

12 Slowly remove the cell and close thecover. (No solution should spill insidethe instrument). Keep the solution inthe cell.

13 The needle should return to the infiniteabsorbance setting. It it does not:

a Reset the needle to the co absorbancemark using the power switch/zero(left) control knob.

b Re-test the reagent blank solutionusing steps 7 through 12 above.

c If the needle does not return to thecx) absorbance mark, another settingas noted in a. and b. is required.Additional warm-up time may benecessary before these settings canbe made.

B Reading Absorbances

Using a single cell in the spectrophotometersused in this course

1 Discard any solution in the cell.

2 Rinse the cell two times with tap water,and two times with distilled water. Thenrinse it two times with the lowest concen-tration standard remaining to be tested.or with processed sample.

3 Fill the cell about three-fourths full withthe same standard or sample.

4 Use a tissue to remove any fingerprintsfrom the cell and any droplets on theoutside. Hold the cell by the top edges.

5 Open the cell holder cover and gentlyslide the cell down into the sample holder.

11-5


6 Slowly rotate the cell until thewhite vertical line on the cellis in line with the ridge on theedge of the sample holder.

7 Close the cover.

8 Record the concentration oft he standard and its absorbanceon a data sheet. (For a sample,record its identification code andits absorbance on the data sheet).

DATA SHEET

Concentrationmg/liter

Absorbance

0.00

SAMPLE

SAMPLE

9 Repeat steps 1 through 8 (above) foreach standard and sample to betested. If a large number of meas..urements are to be made, check theinstrument calibration every fifthreading.

a Use another aliquot of asolution already tested tosee if the same reading isobtained. If not, repeat thezeroing operation in A (above).

b Alternatively, you can use theblank, if supply permits, andrepeat the zeroing operationin A. (above).

//- 6

10 When all the readings have beenobtained, discard any solutionremaining in the cell and rinsethe cell with tap water. Cleanthe cell more thoroughly, (III C.3),as soon as possibli:

11 If no other tests are to be done,turn off the instrument, pull outthe plug and replace any protectivecovering.

VII CHECKING RESULTS

A Readings Greater Than 0.70

On our instrument, these are consideredto be inaccurate. Check the mania/ foryour instrument or check the scale divi-sions to determine the limit for othermodels.

1 Do not use readings greater than0.70 to develop a calibration curve.

2 From five to eight points (countingzero) are recommended for constructinga calibration curve. If you have fewerthan five usable values, you should notdraw a curve.

3 To prevent excessively high valuesin future tests, decrease the cellpath length, if possible, by usingan adapter and smaller cell.

4 If you cannot decrease the cell pathlength. you can at least obtainenough values to construct a curve.Prepare standards with five to eightconcentrations ranging from zero tothe concentration of the standardhaving an absorbance nearest to 0.70.This gives you more values for a curve,but it reduces the applicable range ofthe test. Usually the sample can bediluted before testing so the resultwill fit on the standard curve.

B Highest Reading is Less Than 0.6

1 Increase the cell path length by usinga larger cell. A higher reading results.

2 Prepare a different set of standardswith greater concentrations.


VIII CONSTRUCTING A CALIBRATIONCURVE

If you have from five to eight usableabsorbance values. you can constructa concentration curve.

A Graph Paper

Should be divided into squares ofequal size in both directions

B Concentration Axis

1 Labeling

The longer side should belabeled at equal intervalswith the concentrations ofthe calibration standardsmarked from 0.0 to at leastthe highest concentrationrecorded for the standardson the data sheet.

2 Units

a It is most convenientto express these con-centrations in the unitsto be reported. Otherwise,a unit-conversion factorwould have to be appliedto obtain final, reportablevalues every time you usethe curve.

b Example: If you dilute astandard solution to make100.0 ml volumes of cali-bration standards, youhave a choice in expressingthe resulting concentrations.You can use weight/ 100 ml.or you can calculate weight/1 liter. If you are to reportresults as weight/liter, butyou construct your curveusing weight/ 100 ml, you willhave to, multiply every sampleresult from the curve by1000 or 10 to obtain the re-pciMable value. It is mucheasier to convert the originalcalibration standard concentra-tions to the desired units and touse these as labels on the graph.

7

C Absorbance Axis

1 Labeling

The shorter side should belabeled at equal intervalswith absorbance numbersmarked from 0.00 to at least0.70 absorbance units.

D Plotting the Curve

1 Use the absorbances recordedfor each standard concentrationto plot points for the curve.

2 The points should fall in areasonably straight line.

3 Use a straight-edge to drawa line of best fit through thepoints. If the points do notall fall on the line, an acceptableresult is an equal number ofpoints falling closely above,as well as below the line.Experience provides a basisfor judging acceptability.

4 It is not permissable to extra-polate the curve.

IX USING THE CALIBRATION CURVE

A Finding concentration of the sample

1 Use the absorbar ce value(s)recorded for the sample(s), andthe calibration curve to find theconcentration(s). If the concentra-tion units differ from those requiredfor reporting results, apply a unitconversion factor (VIIIB. 2).

2 If more than one dilution of a samplewas tested, use the result that fallsnearest the middle of the curve.

B If a sample was diluted, calculate thedilution factor and apply it to the con-centration you find for the sample fromthe calibration curve.

11 -7


1 Dilution Factor=

final dilution volumeml sample used in dilution

Example: You diluted 10 mlsample to 50 ml. The con-centration found by using acalibration curve was 0.5mg/ liter.

Then .

constituent. mg/1 =

2.5 mg /liter

11-8

50m1 x 0.5 itig10 ml liter

This outline was prepared by Audrey D.Kroner, Chemist, National Training andOperational Technology Center, MOTD.OWPO, USEPA, Cincinnati, Ohio 45268

Descriptors: Analytical Techniques,Chemical Analysis, Colorimetry,Laboratory Tests, Spectrophotometry

ATOMIC ABSORPTION SPECTROPHt7TOMETRY

I INTRODUCTION

Atomic absorption spectroscopy has beenwell known to physicists and astronomersfor more than 100 years. In 1850, Kirchofftook light from the sun and collimated itwith a lens through the flame of an ordinarylaboratory burner, and then passed thelight through a prism which dispersed itinto the characteristic visible spectrum ,cwith which we are all familiar. He thentook a platinum spoon containing a sodiumsalt and introduced it into the flame Heobserved that the yellow light that waspresent in the spectrum disappeared andin its place appeared the characteristicresonance lines of sodium. Since thenastronomers have used the technique todetect and measure the concentration ofmetals in the vapors of stars. In 1953,Walsh 1j recognized its potential advantageover emission spectroscopy for tracemetal analyr A. Be designed and built ananalytically useful atomic absorptioninstrument. Shortly thereafter the ad-vantages of atomic absorption instrumen-tation were recognized in the United States.

II THEORY

The basis of the method is the measurementof the light absorbed at the wavelength of aresonance line by the unexcited atoms ofthe element. Elements not themselvesexcited to emission by a flame may bedetermined in a flame by absorption pro-vided that the atomic state is capable ofexistence.At the temperature ofoa normal air-acetylene flame (2100°C) only about oneper cent of all atoms is excited to emissionin a flame; therefore absorption due to atransition from the ground electronicstate to a higher energy level is virtuallyan absoiutc measure of the number of atomsin the flame, and the concentration of theelement in the sample. Electrons willabsorb energy at the same characteristicwavelength at which they emit energy.This is the principle upon which the tech-nique of atomic absorption spectroscopyis based.

CH.MET.aa. 2.12.71

The advantages of atomic absorptionspectroscopy as compared to emissionspectroscopy are: (1) that atomicabsorption is independent of the excitationpotential of the transition involved and(2) that it is less subject to temperaturevariation and interference from extraneousradiation and interference from extraneousradiation or energy exchange betweenatoms.Atomic absorption analytical apparatus(Figure 1) consists of a suitable sourceof light emitting the line spectrum of anelement, a device for vaporizing thesample, a means of line isolation(monochomator or filter) and photo-..lectric detecting and measuring equipment.

If the detector is placed to receive onlythe resonance line of the element from thelight source, measurement can be made ofthe absorption of resonance-line radiationon its passage through the vaporizedsample. The magnitude of this absorptiongives a measure of the concentration offree ground-state atoms of the element inthe vapor and when referred to a calibra-tion curve, provides a means of determiningthe concentration of the element in thesample.

M INSTRUMENTATION

The general arrangement of an absorptionflame photometer is no different from anemission flame photometer except for theaddition of a light source. An aerosol isintroduced into a flame which is placed onthe optical axis between the entrance slitof the monochromator and the monochromaticlight source. Energy of the wavelengthebsorbed by the sample is provided by asource lamp whose emitting cathode ismade of that element. This energy ispassed through the flame and then througha dispersing-device. A detector measuresthe absorbed and unabsorbed excitingradiation.

A Light Source

For the more volatile elements such asthe alkali metals, mercury and thallium,

12-I

Atomic Absorption Spectrophotometry

VAPORIZING MONOCHROMATURNOLLOW-CATNODE ROTATING SYSTEM

FLAME'SLIT

lAtSAMPLE IN FUEL IN

FIGURE I

the most convenient source ib the spectralvapor lamp, which consists of a closedglass or silica tube, into which are sealedoxide-coated electrodes, containing ors ofthe rare gases and some of the appropriatemetals. For line sources of the less-volatile elements, hollow-cathode dischargetubes have been found the most satisfactory.These consist of an anode and hallow cylin-drical cathode (either composed of or linedwith the appropriate metal) mounted in asealed glass tube cor..aining one of therare gzscs (helium or argon).Lamps are operated at low currents toimprove lineari..y of response and maintainnarrow emission lines.

B Vaporization of Sample

Atomic-absorption methods have beenapplied almost exclusively to theanalysis of solutions and for this pur-pose flames, fea with a fine spray ofthe sample solution, similar to thoseemployed in flame photometry are used.

The burner has two principal functionsto perform:

a It must introduce the sample intothe flame.

b It must reduce the metal to theatomic state.

A,Z .2

PHOTO-DETECTOR

SLIT

ELECTRONICSAND

READOUT

2 Burners can be classified as:

0 f)

a Total-consumption

Those which introduce the samplespray directly into the flame.(Figure 2)

tSAMPLE

FIGURE 2


b PremixThose which introduce the spray intoa condensing chamber for removinglarge droplets. (Figure 3)

3 Flame shape is important. The flameshould have a long path length (but anarrow width, such as a fishtail flame)so that the source traverses an increasednumber of atoms capable of contributingto the absorption signal.

4 The effective length of the flame maybe increased by multiple passagesthrough the flame with a reflectingmirror system, or by alignment ofseveral burners in'series."

5 The flame temperature need only behigh enough to dissociate molecularcompounds into the free metal atoms.

AUXIUARYFUEL

AIR

FROMSAMPLE

Typical flame temperatures are shownin Table I.

Table I

Fuel-CAdant ApproximateTemp.. °C

Nitrous Oxide - acetylene

Hydrogen - air

3000

2100

Hydrogen - oxygen 2700 - 2800

Acetylene - oxygen 3100

Acetylene - atr 2000 - 2200

Propane - oxygen 2700 - 2800

Illuminating gas - oxygen 2800

Cyanogen - oxygen 4900

FLAME

ORIFICE

CAPILLAR

t MI11110111111,

ASPIRATINGMR

DRAIN

Figure 3

SPOILERSMIXING CHAMBER

4i-- -1

Atomic Absorption Spectr<ophotomet

C Line Isolation

1 The use of a line spectrum of the ele-ment being determined, rather than acontinuous spectrum, makes possiblethe use of monochromators of lowresolving power or even filters. Whena spectral lamp is used as a light source,it is only necessary to isolate theresonance line from neighboring linesof the light source or vaporized sample.The resolution of the method is iir "licitin the width of the emission and absorp-tion lines.

2 To realize the full potentialities of themethod, the strongest absorption linemust be used. For elements withsimple spectra, the resonance linearising from the lowest excited state isusually the line exhibitinestrongestabsorption.

3 Calibration curves depart from linearityat much lower concentrations in absorp-tion work as compared with emissionwork. Curvature results partly fromincreased pressure broadening as theconcentration of salt rises, but alsodepends on source characteristics,particularly self-abscrption, and onthe nature and homogeneity of the flame.

D Detection

1 Photo-electric detectors used in atomicabsorption analysis need be no moresensitive than those used in emissionanalysis, since in the atomic absorptiosr..method, concentration of an element is ".determined by measuring the reductionin intensity of the resonance line emittedfrom a sonrce of high intensity.

2 Single-or double-beam circuits may beadopted for work with a single beaminstrument, reeulte are directly depend-ent upon source and detector stability.roth must be powered by separate.power supplies. In a double-beamsystem small variations in the sourcesignal are compensated automatically.

12-4

IV EVALUATION

A Sensitivity

1 For an air-acetylene flame of length2 or 3 cm the lower limits of detectionof elements having low resonance-lineexcitation potential (eg Na-K) areapproximately equal in a single-beamatomic absorption and emission methods.

2 For elements having highly reversedresonance lines or resonance lines ofhigh excitation potential, the atomic -absorption method has decidedadvantages over emission methods.Examples of elements in these categoriesare 2n, Mg, Fe and Mn.

3 A disadvantage of the atomic-absorption.method, when compared with flameemission, is the lack of a qui,* andsimple method of varying sensitivityto deal with solutions of widely varyingelement concentrations. The sensitivityof an atomic -z ;sorption instrument isdetermined almost entirely by flamecharacteristics, notably length of lightpath through the flame.

4 A comparison of sensitivity obtained byemission and adsorption techniques isgiven in Table II.

1:3 Precision

---- --,

1 Precision of a single-beam atomicabsorption instrument is primarily afunction of the stability of light outputfrom the spectral lamp. This in turnis dep\endent on the stability of the mainsupply and inherent stability of the lamp.The largest fluctuations are only + 2percent for the hollow cathode tube andsodium spectral vapor lamp. A double-beam instrument significantly reducesthis error.

2 In common with flame-emission methods,atomic absorption is subject to "noise"from,the flame and the detector. Changesin absorption caused by fluctuations in

Atomic- Absorption Spectrophotometry

Element

AluminumAntimonyArsenic

Sensitivity mg/1Flew A. A

2 0.50. 21. 0

Barium 0.3 1.0Beryllium 25 0.05Bismuth 0. 2Cadmium 2 0.01Calcium 0.003 0.01Cesium 0.05Chromium 0. i 0.01Cobalt 0. 15Copper 0.01 0.005Gallium 1.0Gold 1

!radium 0. 5Iron 0.2 0.05Lead 2 0. 15Lituum 0.002 0,005Magnesi.im 0. 0.003Manganese 0 01 0.01Mercury 10 0.5Molybdenum O. 2Nickel 0 05Palladium 1. 0Platinum 0.Potassium 0.001 0.005Rhodium 0. 3Rubidium 0. 02Selenium 1. 0Silver 0.05 0.02Sodium 0 002 0.005Strontium 0.01 0.02Teilurtum 0, 5Titanium 1. 0Thallium 0. 2Tin 2.0Vanadium O. 5Zinc 200 0.005

Table IIflame temperature are much less thanthose in emission because\the strengthof the absorption line is principallydependent on Doppler width whereasine intensity of emission from the flameis much more sensitive to temperature.

C .Accuracy

This is -"iown by the types of interferencefounts in :lame emission and atomic ab-sorption spectroscopy. There are threetypes

1 PhysicalCollision of atoms and electrons oratoms and .nolecules will transferenergy thus causing an enhancement ordepression of analysis-line emission.

0.12

0.10L6I

< 0.08cece0re 0.06

0.04

0.02

This has a large effect on flameemission analysis but has only anegligible effect on atomic absorption.

2 Radiative

Light from elements other than the onebeing measured pass the line isolatingdevice (monochromator or filter). Thisoccurs in flame emission work, forexample, the interference of calciumand magnesium in sodium determinations.This interference is also encounteredin atomic absorption using a D.C.system but is very small because of thelarge signal from the hollow-cathodetube. Radiative interference is elimina-ted in an A.C. system.

3 Chemical

Emission from an element in the flameis depressed by the formation of com-pounds, which are not dissociated atflame temperatures. This also affectsabsorption because the formation oftemperature - stable compounds m theflame causes proportionate reductionin the population of ground-state andexcited atoms.

Investigations to date suggest chemicalinterference is confined, almostentirely, to the alkaline-earth elementsand that calcium absorption is more sub-ject to this interference than is magne-sium absorption.Typical calibration curves are shownin Figure 5.

50 100 150 200 250METAL, ppb

Figure 5

300 350 400


REMOVAL OF INTERFERENCES ANDCONCENTRATION OF SAMPLE

A Removal of Interferences

1 The methods for overcoming these inter-ferences in atomic absorption are sirnilarto those used in flame emission, namely,either separation of interfering ions orsuppression of tl interference by ad-dition in excess of a substance that willprevent formation of compounds betweeninterfering ions and the element beingdetermined.

B Concentration of Sample

1 Organic separations can be used toconcentrate a sample. Interferencesare removed, as seen above, and alsothe organic solvent enhances the absorp-tion.

2 Ion exchange has also been used success-fully for concentrating samples for atomicabsorption.

VI CONCLUSIONS

Atomic absorption methods are as goodas or better than emission methods, fo7elements to which they can both be applied,in sensitivity, precision and accuracy..They can'be applied to a far wider rangeof elements than dan emission analysis.The addition. l. cost of hollow cathode dis-charge tubes is compensated by the greaterrange of analyses and greater reliabilityof results.

VU INSTRUMENTS AVAILABLE

A Perkin Elmer

1 Model 303 - double beam, AC - $5, 920.00

1 Model 290 - single beam, AC -$2,900.00

B Beckman attachments for existingspectrophotometers

1 Use with model D. U. and D.U. -2-single beam, DC - $2, 135. 00

12 -6

2 Use with" model D. B. - single beam.AC - $2,495.00

C Jarrell-Ash

1 Dual atomic absorption flame spectrometer-single beam, AC - $5, 800. 00

D E. E. L.

1 Atomic absorption spectrophotometer -single beam, AC - $2, 850. 00

ACKNOWLEDGEMENT

Certain portions of this outline containtraining meterial from a prior outline byNathan C. Malof.

REFERENCES

1 Walsh, A. Spectrochim, Acta. 7, 108.1955.

2 Kahn, Herbert and Slavin, Walter. AtomicAbsorption Analysis. InternationalScience and Technology. November 1962.

3 David, D. J. The Application of AtomicAbsorption to Chemical Analysis. TheAnalyses. 85:779-791. 1960.

4 Platte, J. A. , and Marcy, V. M. A NewTool for Water Chemicals. IndustrialEngineering. May 1965.

5 Biechler, D. G. Determination of TraceCopper, Lead, Zinc, Cadmium, Nickel.and Iron in Industrial Waste Water byAtomic Absorption SpectrophotometryAfter Ion Exchange on DorveA A-1.Analytical Chemistry, 37:1054- 1055 1965.

8 Elwell, W. T. , and Gidley, J. A. F. Atomic -absorption Spectrophotometry. The Mac-Millan Company. New York. 1962.

7 Willard. H. H. , Merritt, L. L. , and Dean,J. A. Instrumental Methods of Analysis.D. Van Nostrand Co., Inc. N. Y. 1965.

This outline wps prepared by P. F. Hallbach,Chemist, rational Training Center, MOTD,OWPO, USEPA, Cincinnati, Ohio 45268.Descriptors: Analytical Techniques, AtomicAbsorption, Instrumental Analysis, MetalsAnalysis

ATOMIC ABSORPTION LABORATORY

ZINC ANALYSIS

Preparation of Standards:

A stock standard zinc solution was prepared by dissolving 1.000grams of pure zinc metal in nitric acid and diluting to 1 liter withdistilled water for a concentration of 1 ml 1 mg Zn. Appro-priate dilutions were made to provide 0, 0.1, 0. 3, 0. 5, 0. 8, and1.0 mg zinc/1 standards.

Operating Conditions for P-E 302:

Wavelength 214

Range UV

Slit 5

Source 10 ma, hollow cathode

Burner P-E premix

Air 20 psi. (30 psi. on tank)

Auxiliary Air 9.0 on flowmeter

Acetylene 8 psi; 9.0 on flowmeter

Flame clear and non-luminescent

Sample Uptake Rate 2-5 ml/ min.

CH. MET. aa. lab. 1.12.71 13-1

_

Zinc AnalAnal sis

13-2

ATOMIC ABSORPTION LABORATOL,

Work Sheet

pig &ill % Absorption Absorbance

0. 1

0. 3

0. 5

0. 8

1. 0

unknown

Concentration of unknown = mg Zn/1

Scale expansion =

Suppression (meter response) =

This outline was prepared by R. J. Lishka,

Research Chemist, Bureau of WaterHygiene, EPA Cincinnati, OH 45268.


COPPER ANALYSIS

Preparation of Standards:

A stock standard copper solution was prepared by dissolving1. 0000 gram of pure copper metal in nitric acid and dilutingto 1 liter with distilled water for a concentration of 1 ml .1 1

mg Cu. Appropriate dilutions were made to provide 0, 0. 1,

0. 3, 0. 5, 0. 8, and 1.0 mg copper/1 standards.

Crpera Conditions for P-E 303:

Wavelength 325

Range UV

Slit 4

Source 15 ma., hollow cathode

Burner P-E premix

Air 20 psi. (30 psi. on tank)

Auxiliary Air 9.0 on flowmeter

Acetylene 8 psi. ; 9. 0 on flowmeter

Flame clear and non-luminescent

Sample Uptake Rate 2-5 nil/ min.

CH. MET.aa. lab.2. 12,71 14-1

)

Copper Analysis

mg Cu/1

0. 1

0. 3

0. 5

0. 8

1. 0

unknown


Work Sheet

% Absorption Absorbance

Concentration of unknown = mg Cu/1

Scale expansion =

Suppression (meter response) =

14-2

This outline was prepared by R. J. Lishka,Research Chemist, Bureau of WaterHygiene, EPA, Cincinnati, OH 45268.

40

1i

ABSORBANCE rr-, assonsommummuli Immummommlemor ammo as

I 1

1

4

0 O 0 CD 0 I t

iv

r. O O

CD

-0)

110

I

I I

einiosummummommantelmemsnessummmisim mom mow RE O

MUM

4

Copper Analysis Copper Analysis

TABLE Ill VALUES OF ABSORBANCE FOR PER CENT ABSORPTION

To convert per cent absorption (%A) to absorbance, find she per cent absorption to the newest whole digit

in the left.hand column; read *cross to the column located under the tenth of o per cent desired, and recd

the value of absorbance. The value of absorbance corresponding to 26.8% absorption is thus 0.1355.

%A .0 .1 .2 .3 .4 .S 6 .7 .1 .9

0.0 .0000 0004 .0009 .0--013 .0017 .0022 .0026 .0031 .0035 .0039

1.0 .0044 .0048 .0052 .0057 .0061 .0066 .0070 .0074 .0079 0083

2.0 .0088 .0092 .0097 .0101 .0106 .0110 .0114 .0119 .0123 .0128

3.0 .0132 .0137 .0141 .0146 .0150 .0155 .0159 0164 0168 .0173

4.0 .0177 .0182 .0186 .0191 .0195 .0200 .0205 .0209 .0214 .0218

5.0 .0223 0227 0232 0236 .0241 .0246 0250 .0255 0259 .0264

6.0 .0269 0273 .0278 0283 .0287 .0292 .0297 .0301 .0306 .0311

7.0 0315 0320 .0325 .0329 .0334 0339 .034: 0348 .0353 .0357

8.0 .0362 .0367 .0372 .0376 0381 .0386 .0391 .0395 .0400 .0405

9.0 .0410 .0414 .0419 .0424 .0429 .0434 .0438 .0.143 .0143 .0453

10.0 .(458 0462 .0467 .0472 .0477 .0482 04117 0491 .0496 0501

11.0 .0506 .0511 .0516 .0521 .0526 .0531 .0535 0540 .0545 .0550

12.0 0555 .0560 0565 0570 .0575 .0530 .0585 .0590 .0595 .0600

13.0 .0605 .0610 .0615 .0620 .0625 .0630 .0635 0640 .0645 0650

14.0 .0655 .0660 0665 .0670 .0675 .0680 .0685 .0691 .0696 .0701

15.0 .0706 .0711 .0716 .0721 .0726 .0731 .0737 0742 .0747 0752

16.0 .0757 .0762 .0768 .0773 .0778 .0783 .0788 .0794 .0799 0304

17.0 0809 .0814 .0870 .0825 .093C .0835 .0841 .08.16 0851 0857

18.0 .0862 .0867 .0872 0878 .0883 .0888 .0894 .0899 .09C' .0910

19.0 .0915 .0921 .0926 0931 .0937 0942 .0947 .0953 .0958 0964

20.0 .0969 .0975 .0990 .0985 .0991 .0996 .1002 .1007 .1013 .1018

21.0 .1024 .1029 .1035 .1040 .1046 .1051 1057 1062 .1068 1073

22.0 .1079 .1085 .'.090 .1096 1101 1107 1113 .1118 .1124 .1129

23.0 .1135 .1141 .1146 1152 .1158 .1163 .1169 .1175 1180 .1186

24.0 .1192 .1198 .1203 .1209 .1215 .1221 .1226 1232 .1738 .1244

25.0 .1749 1255 .1261 .1267 .1273 .1278 .1784 1290 1296 1302

26.0 .1308 .1314 1319 .1325 .1331 .1337 1143 .1349 .12,55 1361

77.0 .1367 .1373 .1379 .1385 .1391 .1397 .1403 .1409 1415 1421

28.0 .1427 .1433 .1439 .1445 .1451 .1457 .1463 1469 1475 1481

29 0 .1487 .1494 .1500 .1506 1512 .1518 1524 .153n 1537 1543

30.0 .1549 1555 -51 .1568 .1574 .1530 1586 .1593 .1599 1605

31.0 .1612 1618 .1624 .1630 .1637 .1643 1649 .1656 .1662 .1669

32.0 .1675 .10E1 .1688 .1694 .1701 .1707 .1713 .1720 .1726 .1733

33.0 .1739 .1746 .1752 .1759 .1765 .1772 .1778 .1785 .1791 :798

34.0 .1805 .1811 .1818 .1824 .1831 .1838 1844 1851 .1858 1864

35.0 .1871 .1f178 .1894 .1891 .1898 .1904 .1911 .1918 .1925 1931

36.0 .1938 .1945 .1952 .1959 .1965 .1972 1979 .1986 1993 .2000

37 0 2007 .2013 .2020 .2027 .2034 .2041 .2048 .2055 2062 2069

38.0 .2076 .2083 .2090 .2097 .2104 .2111 .2118 .2125 .2132 2140

39.0 .2147 .2154 .2161 .2168 .2175 .2182 2190 2197 .2204 2211

40.0 .2218 .2226 2233 2240 .2248 2255 2262 2269 .2277 .2284

41.0 .2291 .2299 2306 .2314 2321 2328 2336 .2343 2351 2358

14-4

Copper Analysis

TABLE III CONTINUED

r;N 0 1 2 .3 .4 .5 .6 .7 8 .9

42.0 .2366 2373 2381 .2388 .2396 2403 .2411 2418 .2426 2434

43 0 2441 2449 .2457 .2464 .2472 .2480 .2487 2495 .2503 .2510

44.0 .2518 .2526 .2534 .2541 .2549 .2557 .2565 .2573 2581 .2588

45 0 2596 .2604 .2612 .2620 .2628 .2636 2644 .2652 .2660 .2668

46 0 2676 2684 .2692 .2700 .2708 .2716 .2725 2733 .2741 .2749

47 0 .2757 .2765 .2774 .2782 2790 .2798 .2807 .2815 .2823 .2832

480 2840 .2848 .2857 .2865 .2874 2882 .2890 .2899 .2907 2916

490 2924 2933 .2941 .2950 .2958 2967 .2976 .2984 2993 3002

500 .3010 3019 .3028 .3036 .3045 .3054 3063 .3072 3090 3089

51,0 .3098 .3107 .3116 3125 3134 .3143 3152 .3161 .3170 .3179

52.0 3188 .3197 .3206 .3215 .3224 .3233 .3242 .3251 3261 3270

53.0 .3279 .3288 .3298 .3307 .3316 3325 .3335 3344 .3354 3363

54 0 3372 .3382 .3391 3401 .3410 3420 .3429 .3439 .3449 .3458

55.0 3468 .3478 .3487 .3497 3507 .3516 3526 .3536 3546 3556

560 .3565 3575 3585 .3595 .3605 3615 .3625 3635 .3645 3655

57.0 .3665 3675 3686 .3696 3706 3716 .3726 .3737 3747 3757

58.0 .3768 377R .3788 3799 .3809 .3820 3620 .3840 3851 386'

590 .3R72 3883 .3893 3904 .3915 3915 3936 3947 3958 3969

60.0 .3979 3990 .4001 401? 4023 4u.:4 4045 4056 .4067 4078

61.0 .4089 .4101 .4112 .4123 4134 .4145 4157 .4168 .4179 4191

62.0 4202 4214 4225 .4237 .4248 .4260 4271 4263 4295 4306

63 0 .4318 4330 4342 .4353 .4365 .4377 4389 4401 4413 4425

64 0 .4437 .4449 4461 .4473 .4485 .4498 4510 .45.22 .4535 4547

65.0 .4539 4572 4584 .4597 .4609 .4622 4634 4647 .4660 4672

66 0 .4685 .4698 4711 .4724 .4737 4750 4763 4776 471sti 4802

,!7.0 4815 41128 4841 .4855 .4868 4881 4995 4906 4921 4935

68.0 .4448 4962 4976 .4989 .5003 5017 .5031 .5045 .5058 .5072

69.0 .5080 .5100 .5114 .5129 .5143 .5157 .5171 .5186 5200 .5214

70 0 5229 .5143 5258 .5272 .5287 .5302 .5317 .5331 5346 5361

71 0 .5376 .5391 5406 5421 .5436 5452 .5467 .5482 5498 .5513

72.G .5543 5544 .5550 5575. .5591 5607 5622 5638 5670

73 0 .5686 5702 .5714 .5735 5751 .5768 5784 5800 .5817 5834

74.0 5850 5367 S884 5901 .5018 .5935 .595? .5969 5986 6011

75 0 .6021 .60:3 .6055 .60" .6041 .6108 .6126 6144 6161 6130

76 0 .6108 .6213 .6234 6253 .6271 .6289 6308 6326 6345 06477 0 6383 6402 5421 .6440 .6459 .6478 .6498 6517 6536 65..S

78 0 6576 .6596 6615 6635 .6655 .66'6 .6696 .67'6 6737 6757

79.0 6778 .6799 6819 .6640 .6861 .6852 6904 .6925 6946 6961.

80 0 6090 7011 .7033 .7055 .7017 .710C .712/ .7144 7167 7190

81 0 7212 7235 .7258 .7282 .7305 .'328 7352 73/5 7309 7423

82 0 .7447 7471 7496 .7520 .7545 7570 7505 .7620 7645 7670

93.0 .7696 .721 .7747 .7773 .7799 7825 .7857 .7878 .7905 .7932

84.0 79:,9 .7986 .8013 .8041 .8069 .8097 .8125 .8153 8182 .8210

85 0 8239 .8268 .8297 .8327 8356 .8386 8416 8447 8177 8508

86.0 .8537 .8570 .8601 8633 .8665 8697 8729 8761 9794 8827

87 0 8861 8894 8928 .8962 8996 ,9031 .9066 .9101 9136 9172

88.0 .9208 9245 .9281 9318 .9355 9393 .9431 9469 9508 9547

89.0 .9586 .9676 9666 9706 9747 97'38 9830 .9872 9914 9957

14-5

PRINCIPLES OF EMISSION SPECTROSCOPY

I INTRODUCTION

The term "spectroscopy", in its broadestsense, refers to the study of the radiationsof the electromagnetic spectrum. As nosingle instrument exists which will separateradiation tromallparts of the spectrum, itis divided into regions which are related tothe different types of instruments capableof producing or measuring waves of variouslengths, A diagram of the spectral distri-bution of energy is shown in Figure 1.

Emis on spectroscopy is concerned withthat r gion wherein radiation can be sort :dout into a spectrum by means of prisms orgratings, and includes the near infrared,the visible and the ultraviolet.

Emission spectroscopy has become an in-dispensable part of the modern chemist'sanalytical methods. With it he can analyzea wide vuriety substances, both quzlita-tive4 aid gut nti+attively, for trace elements,-egardless of valence states.

X 10 106

II THEORY

A Origin of Spectra - Excited atoms andatomic ions emit light of definite wave-lengths. The excitation of multipleelements in a sample results in thesimultaneous production of the spectraof all of these elements.

During excitation by a thermal orelect.ical source, the outer orbitalelectrons of an element absorb energyand rise to higher energy levels.These electrons then return spontane-ously to their normal or ground stateby a single jump or by a series ofjumps. The energy emitted with eachjump produces a spectral line of char-acteristic wave length and fTzquencyfor the particular chemik-al element.The combination of lines produced bythe excited atoms of Oa element thusprovides the emission spectrum ofthat element.

FIGURE 1

SPECTRAL DISTRIBUTION OF ENERGY

10' 102 10° 10-2 10'A BLrrj 104 lo" 10-7 METERS

1 _1_ 1 1 1 1 1

io' 5 7 11 13

LINFRA-

-ULTRA SONICRADIO

HERTZIAN WAVESI ED

A

ULTRA17

716LE-li

VISIBLE

1---X-RAYSH

I 10 PER SEC.

GAMMA COSMICRAYS RAYS

,rTFIFITATE177171-171F5177:5117I nm 2 METERS x 10'

WAVELENGTH k 1500 75°600500400 300 250 200 1 nm = 10 ANGSTROMS

REGION INCLUDED IN EMISSION SPECTROSCOPY

CH. MET. es. lc. 12.71 15-1

-111p:ti

Principles of Emission Spectroscopy

B Types of Spectra - There are threetypes of emission spectra,.

1 The line spectra produced by highlyexcited atoms or alomic ions.

2 Band spectra which originate withhighly excited molecules.

3 Continuous spectra which resultwhen light is emitted by incandes-cent solids.

III INSTRUMENTATION

A The emission spectrograph consists ofessentially f 013 7 distinct functionalparts: the excitation source, the slit,the optical system, and the recordingsystem. A diagram of a Littrow typespectrograph is shown in Figure 2.

FIGURE 2

SCHEMATIC DIAGRAM OF ALITTROW SPECTROGRAPH

COLLI MATOR-CAMERA LENS

1 The production of line spectra forthe detection or the determinationof the elemental constituents in asample requires a n exc i t at i o nsource, such as an arc, a spark,or a flame.

2 The slit of a spectrograph permitsonly a narrow beam of light ofmixed wave lengths to enter theinstrument.

3 The optical system consists of aseries of lenses and either a prismor grating to separate light raysof different wave lengths into thespectrum.

4 The recording system can bea An eye piece such as used in

a spectroscope.b A photographic plate such as

used in a spectrograph.

EXCITATIONSOURCE

SLIT

MIRROREDBACK

15-2

TOTALREFLECTING

PRISM

30*-60"-.90°QUARTZ PR!SM

CAMERA

Princi lea of Emission Sp_ectroscopy

c A photoelectric cell like thatused in direct reading spectro-graphs.

IV SELECTION OF EQUIPMENT ANDCOST CONSIDERATIONS

A A decision concerning the advisabilityof installing a laboratory for spectro-chemical analysis..whether in an indus-trial plant, research la boratory orhealth department, m i g h t rest uponfour main considerations.

1 Whether the analytical problemsencountered have a satisfactorysolution other than spectroscopy.,

2 The time urgency of the work.

3 The expected volume and diversity.

4 The cost of the initial investment,together with operating expenses.

V PROBLEMS AND TECHNIQUESASSOCIATED WITH EMISSIONSPECTROSCOPY

A After the equipment has been se'c upproperly and is ready for operE.ting,there are many parameters to be de-termined before the first samples areanalyzed.

1 Source - arc or sparkThere are several types of exci-tation sources available, as shownin Table 1. Generally, the type ofsample and the precision and accu-racy required for the analysis willgovern which excitation methodwill be employed.

2 Electrode

The spectrographic electrode playsan important part in the analysis.Some important features aredensity, shape and purity.

Table 1. EXCITATION SOURCES

TYPE PRECISION SENSITIVITY RANGE USE

D. C. ARC Fair Excellent 0.00001 -1.0 %a

Basic sources for generalqualitative analysis - soils,ores, oxides, slags, ashes,etc. Highest sensitivity ofdetection for trace elements.

HIGH VOLTAGEA . C. SPARK

Excellent Fair 0.01 - 30.0%Most stable. Use for alloyingconstituents in metals. So lu-tion techniques.

LOW VOLTAGEA. C. ARC Good Good 0.001 - 1.0%

Quantitative determination ofresidual impurities for lowalloying constituents in metal:

SPARK IGNITEDUNI -ARC

Good Good 0.001 - 1.0%Combines precision of thespark and the sensitivity ofthe D.C. arc.

HIGH VOLTAGEA. C. ARC

Fair Excellent 0.00001 -1. 0%

Steadier than D.C. arc -Conductors and nonrefractorymaterials.

15-3


a Density - The current and tem-perature obtainable are depend-ent on the density of the graphite.

b Shape - The nature ofthe sam-ple being analyzed, will usuallydetermine the shape and typeof electrode.

c Purity - Trace analysisrequires that the electrodes be

. .

of very high purity.

3 Exposure conditions

Current, voltage, exposure time,etc., play an important part.

4 Photographic plates

The photographic emulsion mostgenerally used for quantitativeanalysis is the S.A.#1. Thisemulsion responds to the spectralrange from 2400 - 4400 A. It hashigh contrast, low backgrounddensity, and low granularity.The S.A .#2 emulsion is also de-sirable for ;',race analysis w h enlower contrast, higher speed, andwide latitude are needed. S. A. #2also covers the range from 2400 -4400A. Other photographic emul-sions covering the wave le n gt hrange from 2400 - 12, 000 A areavailable, each having specificfeatures. The processing of thesephotographic plates is extremelyimportant and the directions of themanufacturer should be followedclosely.

B Qualitative Analysis

In spectroscopy, qualitative analysis isa relatively simple process, although itcan be more time consuming than aquantitative determination, especiallyi f a complete qualitative analysis i. srequired. The presence o r absenceof more than sixty of the chemicalelements can be determined readily bya simple inspection of the resultingpattern of spectral lines.

All elements give specific lines whensufficiently excited, which is the basisof qualitative spectrochemical analysis.

lir Quantitative Analysis

Quantitative spectrochemical analysisis based on the fact that the amount oflight emitted by an element present atvery low concentration is directly pro-portional to the number of its excitedatoms present, if another factors arekept constant. The intensity of thespectral line of the analysis element(or degree of blackening of the photo-graphed image of the line) in an unknownsample is compared to the intensityof the corresponding line in a standardsample to provide an estimate of theconcentration or the element producingthat line.

D Preparation of Standards

1 Since the major constituents of asample are important factors af-fecting burning qualities of the arcand intensity of resulting spectrallines, it is important that the com-position of the standards approxi-mate the samples as closely aspossible. T his is accomplishedby preparing a synthetic matrixto approximate the averagecomposition of the samples minusthe analysis element or elements.

2 Increasing concentrations of theanalysis elements are added to thesynthetic matrix and are sparkedunder the identical conditions usedfor the unimown samples for cali-bration purposes.

3 It s hould be emphasized that inquantitative spectrographic anal-ysis, the best possible standardsmust always be prepared. A goodspectrographer is only as good ashis analyses, and his analyses areonly as go od a s his standards.Spectrographic standard solutionsare prepared from Reagent Gradechemicals. Oxides are preparedwhere possible; however, nitratesor chlorides will usually suffice.


Occasionally, it is necessary tuse special spectroscopically puregrades of certain metals. Nitricand hydrochloric acids should beredistilled; double distilled wateris also suggested where possible.Borosilicate glasswr.re, cleanedwith both chromic and nitric acids,is used throughout.

4 Internal standards

In quantitative analysis, the errors=used by temperature fluctuationsand wandering of the are are mini-mized by the use of t he internalstandard technique. This consistsof comparing the intensity of asuitable line of analysis elementin a standard and in an unknownsample to a certain selected lineof another element whose concen-tration is fixed in all samples.

Since both the unknown and theinternal standards are parts of thesampe sample, variations in timeexposure, plat e characteristics,and developing conditions will notaffect the relative density of the twolines which are equal in intensityin the light source.

Working curves are established byplotting the intensity ratio of theanalysis line to the internal stand-ared line vs. the concentration ofthe analysis element.

E Accuracy

The average error in quantitativeemission spectroscopy, using thephotographic process, is generally be-tween +5 - 10%. In the direct readingprocess, however, where photomulti-plier tubes are substituted for thephotographic plate, an average errorof +2% is possible. This accuracy ismore than satisfactory, consideringthat determinations of trace elementsare being made in the parts per billionrange.

VI APPLICATION OF .EMISSION SPEC-TROSCOPY TO WATER ANA LYSES

A Trace elements in water originate froma variety of sources, but can beclassified into three principal groups.

1 Elements contributed b y solublematerials chemically weatheredfrom rocks and soil.

2 Elements that are selectively con-centrated by vegetation and findtheir way to surface watersfollowing decay and run-off.

3 Industrial sources, especiallythose devoted to mining, alloying,and cleaning and plating of metals.

Such sources may contribute signi-ficant quantities of trace elementsto surface waters in areas bothpopulated and unpopulated. The de-tection and measurement of thesetrace elements are difficult withconventional analytical proceduresbecause they are not adapted tolarge numbers of samples, and insome caves they are not sensitiveenough. The use of spectrographicprocedures for routine monitoringof raw waters, however, is ad-mirably suited to the pur posesince a large number o f elementsmay be determined simultaneouslywith a high degree of accuracy.

B Trace elements, whether in finished orraw waters, are generally present inconcentrations too low to mea s u r edirectly with the spectrograph. There-fore, a m e an s of concentration isnecessary before the examination canbe completed. This can be,accom-plished in severalways (e. g., evapor-ation, pr ecipitation, ion exchange, etc. )

1 The method employed by theEnvironmental Protection Agency,Research and Monitoring Office,consists of evaporating a filteredvolume of sample at low heat ona hot plate.

15-5


A volume of sample is chosen Con-taining 100 mg of dissolved solidsand evaporated to a volume of 5.0ml or 20,000 mg of solids/1. (Thedissolved solids consist for the mostpart of salts of sociiunt, potassium,calcium and magnesium; these arethe materials for which correctionsmust be made later. ) All samples,therefore, had the same final saltconcentration, but the procedureproduces varying levels of sensi-tivity for different sources becauseof the dissimilar initial volumes.Thus, the limit of detection for traceelements in the Columbia River islower than for the Missouri Riverbecause the low dissolved solids inColumbia River waters permit alarger volume of sample to beevaporated.

a This might b e explained bestby examining the detectionlimits shown in Table 2. Forconvenience, w e have arbi-trarily chosen 400 mg/1 as anaverage TDS content of surfacewater, and the detection limitsas shown are based on thisfigure. For w at e rs of only200 mg TDS/1, the limit wouldbe halved; conversely, if theTDS is 800 mg/1, the limit isdoubled.

Table 2. DETECTION LIMITS*, 1.1101

Zinc 20 Copper 10

Cadmium 20 Silver 2

Arsenic 100 Nickel 20

Boron 10 Cobalt 30

Iron 10 Lead 40

Molybdenum 40 Chromium 10

Manganese 10 Vanadium 40

Aluminum 40 Barium 2

Strontium 4 Beryllium 0. 1

*In water having 400 rag TDS/1.

156

1 00

b The term "detection limit" Itsused to indicate the lowestconcentration of the materialthat can be distinguished frombackground noise or interfer-ence. The limits, as shownin the table, are based onpractical observations and arenot mathematically contrivedfigures based on signal tonoise ratios or other formulas.

c When an element has not beenpositively identified in a sample,the detection limit, precededby a "less than" sign (4), i sused to report the analysis. Itshould be emphasized that thefailure to detect a particularelement does not mean that theelement is absent; it may bepresent in the original samplebut in a concentration belowthe stated detection limit. Thecomplete spectrographic pro-cedure has been publishedpreviously. Table 3 lists theresults from analyses of over1500 samples. The frequencyof detection, the range, and themean concentrations are given.

2 Accuracy

Recoveries were run with a compos-ite of samples from the MissouriRiver. The composite was analyzedto determine what trace elements,if any, were present and at whatconcentrations. To a series ofaliquots of this composite w er eadded increasing concentrationsof the analysis elements. T h esamples were acidified, evaporatedon a hot plate, transferred to cyl-inders, and finally sparked underidentical conditions. Scaler countswere converted to concentrationsby using the data obtained fromthe synthetic matrix and the percentrecoveries were calculated. Astatistical treatment of this datashowed a range of 80% to 113% atthe 90% confidence level.


TABLE 3. SUMMARY OF TRACE ELEMENTS IN WATERS or THE UNITED STATES.

ELEMENT-14TC ." '1.

Occurrescu..Frequency

Of Detection, %

76.5

Observed Values. u 11W.L.1MaTii,7----Msui2 1113 64Zinc 1207

Cadmium 40 2.5 1 120 9.5

Arsenic 17 5.5 5 336 64

Boron 1546 96.0 1 5000 101

Phosphorus 747 47.4 2 5040 120

lroa 1191 75.6 1 4600 52

Molybdenent 511 32.7 2 1500 66

Manganese 810 51.4 0.3 3230 56

A bataiimm 456 31.22 1 2760 74

Beryllium 85 S . 4 0.01 1.22 0.19

Copper 1173 74.4 1 280 15

&War 104 6.6 0.1 36 2.6

Nickel 254 16.2 1 130 19

Cobalt 44 2.9 1 46 17

Lead 305 19.3 2 140 23

Chromium 155 24.5 112 9.7

Vanadium 54 3.4 2 300 40

Barium 1568 99.4 2 340 43

Strontium 1571 i_ 99.6 000 217

1577 Samples (Octohor 1. 1482 - September 30, 1967)14434 Aluminum Samples

TABLE 4. RECOVERIES OF TRACE ELEMENTS ADDED TO RIVER WATER

Element LowPercent Recovery

High Average

Zn 88. 107. 94.1

Cd 100. 106. 103.5

As 80. 100. 93.5

B 82. 97. 88.9

P 70. 110. 91.8

Fe 85. 108. 94.1

Mo 75. 100. 83.7

Mn 94. 105. 99.1

Al 98. 125. 106.7

Be 101. 117. 109.2

Cu 96. 113. 100.8

Aga

Ni 90. 118. 103.0

Co 95. 103. 99.3

Pb 60. 83. 83.6

Cra

V

80.

100.

105.

110.

86.3

104.0

Ba 70. 97. 81.7

Sr 81. 112. 95.1

a Silver precipitated and was not included.

101


Table 4 lists the range of these percent recoveries along with the aver-age recovery for each element.

C Analytical Reference Service Results

A sample containing 8 trace metals atknown concentrations was analyzed by66 participating agencies using a num-ber of analytical methods. The meanresults of all 66 laboratories, in addi-tion to those laboratories employingspectrographic procedures, are shownin Table 5. It is apparent from thissummary that the spectrograph can beused to 11 plendid advantage on watersamples.

SUMMARY

The theory and instrumentation of emissionspectroscopy are presented. Considerationis given to the problems an d techniquesassociated with trace element analysis inwater using the Emission Spectrograph.

For the present time, atomic absorptionis the method of choice for metals analysesin the EPA Methods Manual. (8)

When broad spectrum analyses are required,the Emission Spectrograph may be usedwith comparable accuracy and preOision.

Table 5. SUMMARY OF RESULTS OBTAINED BY SPECTROGRAPHICPROCEDURES ON ARS SAMPLE: WATER METALS, NO. 2.

Amount added Mean ofAMOUNT RECOVERED mg/1

Element mg/1 66 Labs Lab #71126 NWQN Lab Lab #1615

Al 1.80 2.21 1.80 2.30 3.10

Cr 0.18 0.14 0.18 0.13 0.17

Cu 0.42 0.43 0.29 0.38 0.41

Fe 0.62 ** 0.44 0.46 0.40 0.75

Mn 0.25 0.28 0.25 0.23 0.38

Cd 0.24 0.26 0.25* 0.30 0.50

Zn 0.90 0.94 0.68 0.94 0.88

Pb 0.18 0.20 0.27 0.19 --

**Iron value was adjusted to mean value as suggested on Page 47 of ARS report.*This result was erroneously included under Lab #7112A in the ARS report.

15-8 0

REFERENCES

Emission Spectroscopy

1 Machtrieb, Norman H. Principlesand Practice of SpectrochemicalAnalysis. McGraw-Hill BookCompany, Inc. 1950.

2 Clark, George L. The Encyclopediaof Spectroscopy. Reinhold Pub-lishing Corp. 1960.

Water

1 Kroner, Robert C. and Kopp, John F.Trace Elements in Six WaterSystems of the United States.J,A.W.W.A. 57:150. 1965.

2 Kopp, John F. and Kroner, Robert C.A Direct Reading SpectrochemicalProcedure for the Measurement ofNineteen Minor Elements inNatural Water. Applied Spec-troscopy 19:155. 1965.

Kopp, John F. and Kroner, Robert C.Tracing Water Pollution with anEmission Spectrograph. JournalWater Pollution Control 39:1659.1967.

10-t...)


4 Haffty, Joseph. Residue Method forCommon Minor Elements. U. S.Dept. of Interior Geological SurveyWater Supply Paper No. 1540A.1960.

5 Silvey, W. D. and Breman, R.Concentration Method for Spec-trochemical Determination ofSeventeen Minor Elements inNatural Water. Anal. Chem. 34:784. 1962.

6 Skougstad, M.W. and Horr, C. A.Occurrence and Distribution ofStrontium in Natural Water.U. S. Geological Survey WaterSupply Paper No. 1396 C. 1963.

7 Clarke, F. W. The Composition ofthe River and Lake Waters of theUnited States. U. S. GeologicalSurvey Papers No. 135. 1924.

8 Methods for Chemical Analysis of ,

of Water & Wastes, EPA-AQCL,Cincinnati, OH' 45268.

This outline was preivired by J. F. Kopp,Chief, Metals Analyses Unit. Analyticalr4uality Control Laboratory, NERC, EPA,Cincinnati, OH 45268.

Descriptors : Heavy Metals, Instrumentation,Metals, Spectroscopy, Water Analysis

15-9

FLAME PHOTOMETRY

I PRELIMINARY

Flame photometry is the art and science ofapplying thermal energy (heat) to elementsin order to effect orbital shifts which producemeasurable characteristic radiations. Thecolor GI t h e emission and the intensity ofbrightness of emission permit both qualita-tive and quantitative identification.

The application of a very hot flame (2000° Cor more)produces excitation of the element,caused by t he raising ;if an 'electron to ahigher energy level and is fo,llowed by t heloss of a small amount of energy in the formof radiant energy as the electron falls backinto its original position or to a lower energylevel.

COLLIMATINGMIRROR

YELLOWRED

,..

=1*- FUEL

4'Pm-OXYGEN

SAMPLE

ATOMIZER BURNER

PRISM BLUE SLIT

GREENPHOTODETECTOR

II INSTRUMENTATION

The s ix ecsential parts .of a flame photo-meter are: pressure regulators an d flowmeters 'tor the fuel gases, atomizer, burner,optical system, photosensitive detector andan instrument.f or indicating or recordingoutput of the detector. These componentsare schematically shown in Figure 1.

A Atomizer and Burner 1.4

Numerous variations in atomizer an dburner designs have been used. Figure2 depicts the integral aspirator- burnerused in Beckman instruments. Thesample is introduced through theinnermost concentric tube, a vertical

FIGURE 1. SIMPLIFIED DIAGRAM OF A FLAME PHOTOMETER

1i,''L 4

CH. MET. 16c. 12.71

METER

16-1

Flame Photometry

palladium capillary. A concentricchannel provides oxygen, and its tip isconstricted to form an orifice. Oxygenis passed from this orifice causing thesample solution to be drawn up to thetip of-the inner capillary. There, theliquid is sheared off and dispersed intodroplets. All droplets are introduceddirectly into flame, with a sample con-sumption of 1 - 2 ml per minute.

The main requirement of the burner isproduction of a steady flame when sup-plied with fuel an d oxygen or air atconstant pressures. In the Beckmanaspirator-burner, a concentric channelprovides oxygen to operate the atomizerand the flame, The additional concen-tric channel provides fuel for the flame.

B Optical System, Photosensitive DetectorAnd Amplifier

The optical system must collect thelight from the steadiest part of the flame,render it monochromatic with a prism,grating or filters, and then focus it on-to the photosensitive surface of the de-tector. Use of filter photometers is leastdesirable due to their limited resolution.Flame spectrophotometers imp roveapplication as they will separate emis-sions in a mixture of metals, s u c h asmanganese lines at 403.3 nm and thepotassium lines at 404.6 nm P la ce-rn ent of a concave mirror behind theflame so that the flame is at the centerof the curvature increases intensity offlame emission by a factor of 2.

Any photosensitive device may be usedin a flame photometer. The detectormust have a response in the portion ofthe spectrum to be used and have goodsensitivity. The photomultiplier tubeis the preferred detector for f la m espectrophotometers.

The amplifier increases the signal fromthe phototube and improves resolutionbetween close spectral lines. It alsop e r m it s identification of elementspresent in samples when the concen-tration is very small.

16-2

Sample

Figure 2. DETAILED DIAGRAM OFBURNER-ATOMIZER

III APPLICATIONS OF FLAMEPHOTOMETRY TO WATER ANALYSES

Measurement of sodium and potassium in thepast has been confined to complex, tediousand time-consuming gravimetric procedures.The flame technique enables the analyst toperform these determinations in a matterof seconds. If these metals alone were theonly elements capable of measurement byflame photometry the use of the instrumentcould st ill be justified in a great manylaboratories.

Other cations which may be detected andmeasured in waters and waste materialsare calcium, magnesium, lithium, copper,and others. Table 1 includes those elementswhich may be measured with commerciallyavailable equipment/including ultra-violetand photomultiplier accessories.

Table 1 does not include wavebands whichoccur in the infrared spectrum. Sodium,for example, has an emission band at 819 nmwhich is not detectable w it h the commoninstruments.

Many other metals, including the rare earths,can be measured using the flame techniquebut they are not included in the table because

Flame Photometry

Table 1

WavelengthApproximateSensitivity

mg/1Wavelength

ApproximateSensitivity

mg/1

Aluminum 484.2 2 Lead 405, 2

467.2 3 368'9 2

396.2 4 364 3

Barium 553.5. 0.3 Lithium 670.8 . 0.002493 0.4

Magnesium 371, 0.1Beryllium 471 25 383' 0.1

510 100 285.2 0.2

Boron 548 1 Manganes, 403 0.01521 2 279* 1

495 3 561 2

Cadmium 326.1' 2 Mercury 235. 7* 10

228. 8* 40Potassium 766.5 0.001

Calcium 422.7 0.003 404.6, 0.2622 0.004 i9 344.7' 3

554 0.01Silver 338.3 ?

0.05Chromium 425.4 0.1 328.1 0. 1

.360? 0. 1520 0.1 Sodium 589 3/ 0.002

330.3' 1

Copper 324' 0.01327' 0.01 Strontium 460. 7 0.02

681 0.01Iron 372 0.2 407.8 0.5

386 0.2373 0.3 Zinc 213.9* 500

500 200

* bltra-violet spectrum

? = Doubtful detection in visible spectrum

/C 3

e"

Flame Photometry

the necessity for their measurement in wateris a rare occurrence.

IV INTERFERENCES

A Spectroscopic Interferences

Energy at othe r wave lengths or fromother elements than those intended t obe measured may reach the detector.This problem is related to the resolutionof the instrument and slit widths used.

Many of the instrvmental difficulties arerelated to reproducibility of the flame.The quality and composition of the fuel-affect the constancy and temperature ofthe flamc which in turn influences t heenergy of emission. Likewise, slightvariations in fuel pressures and ratiosaffect the reproducibility of the flamewith reference to shape, temperature,background, rate of sample consumption,etc. In some cases, the temperatureJf the flame is the limiting factor in de-termining the presence of a metal. (Thealkaline earth metals emit radiationsat "low" temperatures, whereas othermetals require very "hot" flames.)

Table 2 indicates temperatures obtain-able with different fuel-oxidant mixtures.

Table 2.Approximate Temperatures of

Fuel-Oxidant Mixtures forFlame Photometer Use

Fuel-Oxidant pprox a eTema. ° C

Hydrogen - air 2100

Hydrogen - )xygen 2700 2800

A cetylene - oxygen 3100

A cetylene - air * 2000 2200

Propane - oxygen 2700 2800

Illuminating gas - oxygen 2800

Cyanogen - oxygen ** 4900

* Undesirable because of carbon deposits.** Used in research problems.

16-4

Emiss'-n reading of spectral linesalways includes any contribution fromthe flame background emission on whichthe line is superimposed. When thephotometer includes a monochromator,it is possible to read the backgroundradiation in the presence of the testelement. First, the line +backgroundintensity is measured in the normalmanner at the peak or crest of the bandsystem. Next, the wave length dial isrotated slowly until emission readingsdecrease to a minimum at a wave lengthlocated off to one side or the other ofthe emission line or band. It is usuallypreferable to read the background at alower wave length than the peak. Back-ground reading is subtracted from theline + background reading.

Products of combustion may affect thecharacteristics of the flame or mayaffect the. optical system by fogging orcoating of lenses and mirrors.

B Factors Related to the Compositionof the Sample

A ri element may be self-absorbing --a phenomenon in which the energy ofexcitation is not proportional to the con-centration of the element. As previouslydiscussed, exictation is followed byloss of energy in the form of radiationas the electron falls back to its originalposition or to a lower energy level.During passage of radiant energy throughthe outer fringes of the flame, thisenergy is subject to absorption throughcollision with atoms of its own kindpresent in the ground energy leve 1.Absorption of radiant energy weakensthe strength of the spectrum line. Usingthe emission line at 589 nm for sodium,Figure 3 indicates that the line ceasesto be linear at 13 mg/l. As the sodiumcor.centration increases, the self-absorption effects become m or epronounced. Sample dilution to pdrmitreading on linear portion of the curveis often practiced.

Two or more elements present in thesample may produce radiant energy at

the same, or near the same wavelength.For instance, calcium at 423 nm a n dchromium at 425 nm could in t e r f e rewith each other by additive effect. Thecorrection may be to dilute out the un-wanted metal or measure one of theemissions at a different wavelength.

The emission energy of one element maybe enhanced or depressed by energiesfrom other elements. This phenomenon(radiation interference) occurs when oneelement causes another to modify its ac-tual emission intensity in either a neg-ative or positive mariner. Correctionis obtained by dilution or by controlledinterference addition.

Other types of difficulties encounteredare too numerous to list here. In gen-eral, they may be overcome by improvedinstruments (high resolution, narrowerslit openings, optics, flame adjustment)or possibly by special techniques.

Some inexpensive instruments, designedfor limited use, may employ illuminatinggas with air or propane with air as amatter of economy or convenience.

STANDARD CURVE FOR SODIUM

."'

."'

Ik) -1-2b

I 4

mg Sodium/1Figure 3

Flame Photometry

TECHNIQUES

The following techniques are intended toserve as examples of current proceduresin use for routine samples and for specialsamples where corrective procedures areindicated.

A Emission Intensity vs. Concentration

This is the classical procedure in flamephotometry. Solutions (standards)containing known concentrations of testelements are compared with an unknownsample. This technique i s applicableonly when no interference is present.

B Radiation Buffers

For measurements of alkaline earthmetals (sodium, potassium, calcium,magnesium) radiation buffers are pre-pared as solutions saturated withregard to each metal, respectively. Apotassium buffer, for examr,., is pre-pared by saturating distillea water withsodium, calcium, and magnesiumchloride. A calcium buffer in turn issaturated with sodium, potassium andmagnesium chloride.

C Preparation of Radiation Buffers

For a sodium measurement, the buffersolution is added equally to samplesand standards so that the interferencesare alike for all readings, therebycancelling each other (see Table 3).

D Instrument Improvement

Potassium emits energy bands at 766,405, and 345 zip The bands are atopposite ends of the spectrum and the405 and 345 bands are not usable in thevisible spectrum. The 766 line alsoloses sensitivity because of its prox-imity to the infrared region. Use of ared sensitive phototube or photomulti-plier, however, permits measurementwith an ordinary instrument at concen-trations as low as 0.1 mg/1, or less.This approach i s applicable to otherelements also.

16-5

Flame Photometry

E Standard Addition

Equal volumes of the sample are addedto a series of standard solutions con-taining different known quantities oftest element, all diluted to the samevolume (see Table 4). Emission in-tensities of the resulting solutions arethen determined at the wavelength ofmaximum emission and at a suitablepoint on the flame background. Aftersubtracting the background emission,the resulting net emissions are plottedlinearly against the concentration ofthe increments of the standard solutionsthat were mixed with the unknown. Thepercent transmission of the mixturecontaining unknown sample and zerostandard (distilled water) i s doubledand the concentration correspondingto this point on the graph will be theconcentration of the undiluted unknownsample. This can be explained alge-braically in conjunction with Figure 4.

F Internal-Standard Method

The method consists of adding to eachsample and standard a fixed quantity ofinternal standard element. The elementmust be one not already present in thesample. Lithium is usually the inter-nal standard used. Thies method is mostconvenient when using instrumentshaving dual detectors. The emissior.intensities of standards and samplesare read simultaneously or slicces-ively depending upon instrumentation.

G Separation of Interferences

In cases where certain elements inter -fere, they may be physically removed,or the interference may be "blockedout" by reading the emission at differentwavelengths. To measure lithium, forexample, calcium, barium, andstrontium are precipitated as carbon-ates of the metals. The lithium isretained u-. the filtrate and measuredat a wavelength of 671 rum.

t

NaC1 KC1 CaC12 MgC12

Sodium Buffer

Potassium Buffer

Calcium Buffer

Magnesium Buffer

-

+

+

+

+

-

+

+

+

+

-

+

+

+

+

-

r_...

Conc. of standards 0.0 mg/1 5.0 mg/1 10.0 mg/1

Volume of standardadded to sample 10.0 ml 10.0 ml 10.0 ml

Volume of sample used 10.0 ml 10.0 ml I 10.0 ml

Concentration of elementin each portion of mixture

x2- + 0 mg/1 x-2 + 2.5 mg/1 x-2 + 5 mg/1

Table 4

I I i (-.li

16-6

50

40

36

30

2C

10

Flame Photometry

5 7

()2--a+2 5) (2 3 5)

Concentration-mg/1

Let x = concentration of element in . . 2Y = ; .5 (from the example inFigure 4)unknown sample.

Then Y = 310 transmission of an equalmixture of unknown sampleand zero standard, or

x 0

2Y =

2+ which simplifies to 2Y = x

Figure 4

xby substitution, x = 7 5 + 3 5

3 5

x = 7 mg/1

Flame Photometry

BIBLIOGRAPHY

1 Kingsley, George R. and Schaffert,Roscoe R. Direct Mierodeter-mination of Sodium, Pottssiumand Calcium in a Single biologicalSpecimen with the Model Du FlameSpectrophotometer and Photo-multiplier Attachment.Anal.Chem.25:1937-41. 1953.

2 Gilbert, PaulT. Jr. Flame Photo-metry - New Precision in ElementalAnalysis. Industrial LaboratoriesBeckman Reprint R-56. Aug. 1952.

3 Detection Limits for the Beckman ModelDu Flame Spectrophotometer. DataSheet 1. BeckmanPublication,April 1952.

4 Baker, G. L. and Johnson, L. H.Interference of Anions on CalciumEmission in Flame Photometry.Anal. Chem. 26:465-568. 1954.

5 West, P. W. , Folse, P. andMontgomery,. D. The Applicationof Flame Spectrophotometry toWater Analysis. Anal. Chem. 22:667.Beckman Reprint R-40. Model10300. 1950.

18-8

6 Scott, R. K. , Marcy, V. M. andHromas, J. J. The FlamePhotometer in the Analysis ofWater and Water-formed Deposits.ASTM Bulletin. Model 10300.(Abs. ) May, 1951. page 12.

7 Burriel, F., Marte and Ramirez, J.Flame Photometry. MunozElsevier Pub. Co. , N.Y. 1957.

8 Chow, T. J. and Thompson, T. G.Standard Addition Method.Anal. Chem. 27:18 -21. 1955.

TEXTS

1 Willard, H. H., Merritt, L. L. andDean, J.A. InstrumentalMethods of Analysis. D. vanNostrand Co., Inc., N.Y. 1958.

112 Dean, J.A. Flame PhotometryMcGraw-Hill Book Co. , N. Y. 1960.

3 Clark, G. L. The Encyclopedia ofSpectroscopy. Reinhold Pub-lishing Corp. , N. Y. 1960.

This outline was prepared by R. C. Kroner,Chief, Physical and Chemical Methods,Analytical Quality Control Laboratory,NERC EPA, Cincinnati, OH 45268.

Descriptors: Chemical Analysis, Water Tests,Flame Photometry, Spectrophotometry,Spectroscopy

J. 1

FLAME PHOTOMETRY LABORATORY (SODIUM)

I REAGENTS

A Deionized Distilled Water

To be used for the preparation of all reagents,calibration standards, and as dilution water.

B Stock Sodium Solution

Dissolve 2.542 g of NaC1, previouslydried at 140°C, in deionized distilledwater and dilute to 1000 ml.

1.00 ml = 1.00 mg NO

C Intermediate Sodium Solution

Dilute 10.00 ml of the stock sodium solutionto 100.0 ml with deionized distilled water.Use this solution for preparing thecalibration curve in the sodium range of .

1-10 mg/1.

1.00 ml = 100 µ g Na+

D Standard Sodium Solution

Dilute 10.00 ml of the intermediatesodium solution to 100 ml with deionizeddim.illed water. Use this solution forpreparing the calibration curve in thesodium range of 0.1-1,0 mg/l.

1.00 ml = 10.0 gg No.+

II INTERFERENCE CONTROL

Refer to the cited reference

III STANDARDS

Standards may be prepared in any of theseapplicable ranges: 0 - 1.0, 0 - 10, or0 - 100 mg/1.

CH. MET. lab. 2c. 5.76

IV INSTRUMENT OPERATING CONDITIONS

Theoretical wavelength 589 nm

Fuel pressure 7.5 lbs/in2

Oxygen pressure 10 lbs/in2

For all other conditions needed, consult themanufacturer's instrument manual.

V PROCEDURE

1 Number the six plastic cups provided 0,2, 4, 6, 8, and 10.

2 Fill them 314 full with the appropriatesodium standards; e.g., 0 mg/1 scarriardinto cup 0, etc.

3 Fill a 7th plastic :,.up 3/4 full with theunknown.

4 Fill an 8th plastic cup with distilled water.

5 The power toggle switch (on left side ofinstrument) is already turned on.

6 Set the sensitivity knob to the standbyposition.

7 Set the wavelength knob to the theoreticalvalve of 589 nm (the scale is at the top ofthe instrument).

8 The fltr shtr - open knob is in the shtr(closed) position.

9 Open the main valve on the oxygen cylinder;all other oxygen gauges are already set.

10 Open the main valve on the hydrogencylinder; all other hydrogen gauges arealready set.

1E17 -1

Flame Photometry Laboratory (Sodium)

11 Raise the door on the right side of theburner housing (behind instrument).

12 Cautiously bring a lighted match to thetip of the burner in the housing.

13 Close the door on the burner housing.

14 Caution: Do not place any part of your bodyover the coils on the top of the burnerhousing. Hot gases are escaping.

15 Raise the silver lever between the instru-ment and the burner housing to the verticalposition.

18 Place the plastic cup containing distilledwater in the cup holder which is nowexposed at the right aide of the burnerhousing.

17 Push the silver lever clockwise so thatthe cup holder swings into the burnerhousing and the wateeis aspirated. Adistinct difference in sound will be noticedwhen water or a sample is being aspirated.If at any time during the determinationthis sound again changes. it will indicatethat all of the liquid has been aspiratedfrom the cup. Simply move the silverlever and refill the cup with the appropriateliquid.

18 Do not allow air to be aspirated for morethan about 15 seconds. If there is anydelay, aspirate distilled water until theproblem causing the delay has beencorrected.

19 Turn the sensitivity knob to position 1.

20 Turn the dark current knob until the needlereads 0 on the percent transmittance scale.

21 Turn the sensitivity `_mob to position 4.

22 Repeat step 20.

23 Swing the cup of distilled water out of theburner housing and replace it with cup 10.Swing this cup back into the burner housing.

24 Turn the fltr - shtr -open knob to the openposition (this opens the shutter).

17-2

25 Turn the wavelength knob slowly to the left;the needle will move to the left.

26 At some point the needle will suddenlyswing toward the right. It will probably benecessary to make adjustments with the slitknob in order to keep the needle op-scalewhile finding the point at which the needleswings back to the right. Record this wave-length. It is the peak wavelength. Do notchange this setting until indicated in theinstructions.

27 If the point at which the needle swings backto the right is overshot, turn the wavelengthknob about 1/4 turn to the right and repeatsteps 25 and 28.

28 Make the needle read 10010 transmittanceby turning the slit knob. Record the slitmm reading. Do not change this setting.100 is the peak transmittance readingfor this solution.

29 Turn the fltr - shtr - open knob to the shtrposition (this closes the shutter).

30 Replace cup 10 with cup 8.

31 Open the shutter.

32 Record the percent transmittance readingand close the shutter.

33 Repeat steps 30, 31, and 32 using cups6, 4, 2. 0, and the unknown, in turn.

34 Aspirate distilled water and using the darkcurrent knob make the needle read 0 percenttransmittance.

35 Aspirate cup 10.


37 Slowly turn the wavelength knob to the left.The needle will move to the right and atabout 1/4 - 1 percent transmittance willmove no further to the right. Record thewavelength reading. This is the backgroundwavelength. Do not change this setting.The percent transmittance is the backgroundreading for this solution.

i1 rlI ..

Flame Photome Laborator (Sodium)

38 If the point at which the needle moves nofurther to the right is overshot, turn thewavelength about 1/4 turn to the right andrepeat step 37.

39 Close the shutter.



42 Record the backgtound transmittancereading for this solution.


44 Repeat steps 40. 41. 42. and 43 usingcups 6. 4. 2, 0, and the unknown, in turn.

45 Aspirate distilled water for about 15 sec.

46 Turn the sensitivity knob to the standbypoisition.

47 Close the main valve on the hydrogencylinder.

48 Close the main valve on the oxygen cylinder.

49 Empty the eight plastic cups and discardthem.

50 Leave the power toggle switch (on leftside of instrument) on.

51 For each of the 6 volutions subtract thebackground percent transmittance readingfrom the peak percent transmittancereading.

52 Using the graph paper provided in themanual, plot the 6 differences vs. theappropriate concentrations. Draw theline to best fit connecting the 6 points.This is the calibration graph.

53 Find the difference percent transmittancefor the unknown on the percent trans-mittance axis.

54 Draw a straight line to the right until itintersects the calibration line.

55 From the point of intersection draw a linestraight down to the concentration axis.

56 This is the concentration of the unknown.

REFERENCE

Standard Methods for the Examination of Waterand Wastewater, 13th Edition, page 317.Method 153A. 1971.

17-3

7


0. 0 mg/1

2.0 mg/1

4. 0 mg/1

8.0 mg/1

8.0 mg/1

10.0 mg/1

Sample

Peak Wavelength

Percent Transmission Readings

Background Peak Difference

nm

Background Wavelength nm

Slit mm

17-4

I 1 '''"I , d

SUuill1U.uISlI.I.IUUUUUlUIuuI.uUI:t P 'liii' lIS.'lPI:;ipI(,IgI4IJ a ffiUPiuiIuiISiiiIPi a ia.,aa,a.u..su..au.u.u..u..iu....u.i;;;;:;;;;;;;:::..ps..esmssauu.::--aa*aaiau....i.iusaui I.v.......Iu1a.....a....II.a.I...,.a.Iu...auIMI.a.....,...I.Ia,.....uu.. ..au.u.i ....u..mu. uIsu.auuuuIuuu.uuu.auuIauIuuDumau....uuulauii.su..0 ..iuii.ii iiuiuiuu IUIRIIIUUU$tlUUNUIU.IUUIIUUI lIUI1UIISlUUUUIuui1UuNla.i.i...,.....i...i. U.USU.aU.U..U.lU.US....I.lI.IUi.i..SIUIUUUIUiiU.U.....lU.I.IuPI...IU.. mu.0 .SU. .s uuu ... iuu ..im u.u..u.0 u .usuua u....i. u uuuiuuuul ...umuuuuUUIUuIU.IIul.UIINUI.I.l.II.Ia.uu..m.uu.uuiuuiu.uauu .i.sua...u....l.,ui....au..asa IuuauuUuuUuupuu.aIuuIIuuu...... USuuUauI*IIa*.IuU.INuIUuI..uuI.uuUUuBuI.lUUUuU..uU.lSIIU.II1U....Iuu.I.0a.uu..u.ap.a.usi.u.i ....Iuu.IuuIuI.uUuISu..IIIUuuuuIuIUuluRuIulIauuR.s.I..S.$u.s.p.uuuusauuIuuuuesIususIuuIIuIuuaauuuaIsuIIuapuuuUauuupuuupuuuIIuuuuupuu. ....uuu..us...ui...au...ui..u....uuIm.uiu..ua.ul.am....usuu.ps....,u...I.uuuuuuiauuussusais i...I. ia..iisaui...ui..ii.as.asi.i isisisiusi issisilsus lu...u.s.i.....I....u.i.u....u. u...u.....uI..a..uuiiU.Uu...uu...uuusu...I.uu..uiva.uuuiuiuIuiUS..uuu.uu.ul.uuuiaIIiuuiuuuuuiuuuSumuiueuuuuuuuuiuuIiuuiuUSiuuuuuSu...i..au.....i.. .uuu.a........uu.jSiuUiuUSuuiiuiRiuiu$UIIUUUIUUUUISuSuuuiu.u. uuu......uuu..uuu..amua......uu.uuu..uumu.0 miuuiuiuuu iuuuu..iuiuua...uim.u.u.iu..a.u..i.ui uuuiuu.uuiu.a.aui.uuuusuuuu uuaIuuuuuuisuI..Iuau..i.m.. uauiuusu euuiuusu u.iu...s...a.uuui.u.usi..uiuuu.u..u.u.i...0 lull..u .IuussuIu....uI.u..uuuuSuIu.u.....u.I..u.u..R.uIuuI.Iu.....u.SuS..uu.u5.u.ulIU..U2.Ui.li.uiului..UuU.IIIIUU.uu$Uu.I.Ui.lUIpS.ili.luuR.UUuuS... luluSuuul mliisuuuuuueuuuuiu uuuuau u.uu.uuiu...uuuu.uu.ua.... muuuuuuuuuuuuuuuuusuuuuuuiuuu .iu....u...uuuuu...e..u..uumu.u..u..0 uuuuu.uu uiuuiu. ..uuu..uuuuu. uuusuu.uu u.u.uuuuuu.aumI.u..uu..usuuuiIuuuIuuu uuauuuuiuuu.u.uu..uuuui.uuu..uuu.uuuu.....u..u...u. uiuiu.u.uuuuuuusuuuuuuiiuuusuisuuueuuuu..uuuuuuIUUUIIUUUIUIIIUUIU ...s.uuuu.u.uuu.uuuiuu..ump.uuuuuuuuuuuuuiu u.u..uuuumuuuuuI.uiuuuu.uu.u....u.uu.uuu..iuuuuulmiuuuupuuuuuuuuiuuuuuuIuluI.iiuu..u.uuu u.uuuumu....u..uuu uiuu.uuuuuiuuuu.uu.uu..uuu.uu....uuiuuu.u......uiu.luuuumuuSuuuu..Iu..uuu...ua.uu.uuuu.uuu.uiuuuuulu mu..u..uuu.....u.uum.au..uuuuuuu uuauuuuu uuuuuuu.uuuuIuuu.Iuuuu.uuu..u.IR.u....uuu u.u luau.a...uum.s..u.p..uu.u...u..u.uuu.a.u.u...uu..u...u.us...0 u.uu..u ..u....u.. usuuuu.u.uuu..Iu.u.. u....u.u.....u...u...i..u..a.uu..u.ui uImuu.uuu...us.ui.uumu.u.a.uau'uuuuu...uauu...uuuuuuiu u u..u....u..uuuuuu ia usa iuuuu u uiuu uivaa .uuuu i uuiuuuu uu u..u...i. ... .ui. isi uuui us u.uui usu.u.auu.ua..u.uuuuuuIuuuul.uulu.u.uu..uuauuuu.uuuu.uuu.uauuu.uus.u.a.u*uuuus.uauuuuuuu.uu...uu suau.u.aiuuu.uuuuuuuuauusuuauuuaaau uuuuu upuuiausuusau.u....uu..u.uuu.uu...uuus....iuui..a...suu...u....a.uauuuuu.s..u. uuusassuus.auuu.uUUUIIUIUUIUUUUUS.IUUUaIu.SIU.PUI..UUU5.uSUU.IUUuUUISUUIIUIU.uuIIUSI.UIIUIISUuuU..u.ua.e u.s uuuu ii us .iuu .s.u.....uu uuusuuuuu su usuuiu. .a. .u. ui uuu.m.. us..s.u.ua.u.u.s.au..uuu....auuuuu.u.u.uu..iu.u...u.uu..suu..au...s..su.....su.suiuusu a.u......u.uu...u.uuj.u..umsuauu....ui..u...uu.uuu.u....uu...s.....a uu.u..ua.u..uUUISUUUISUIIUUSIUUIUPIIIUUUIUUUiUIUISUUUiUiUiUUS 15UUUUIU u.uuuuu..iu.au........u.ua...aisus..u.pass.upsgui.,.uu.s..a.u..u..s.iuusu.u.uu.u.aiva..u..s.u....u..i........s..0 isuiuuuiuiuuuuuuiuiauusuisiauuiiiiiuuauivauuiiuuiuuuu

.U

UI

S

UI

UU.UU

.a


This outline was prepared by C. R. Feldmann, Descriptors: Chemical Analysis, LaboratoryNational Training Center, MOTD, OWPO, Tests, Water Analysis, Sodium, Alkali Metals,USEPA, Cincinnati, Ohio 45268. Metals

17-6

1 (/ ,.:.,

FLAME PHOTOMETRY LABORATORY (STRONTIUM)

I GENERAL

This procedure is listed as being "tentative"in the cited reference. Also. strontium isnot listed in Table I of the Federal Register,Volume 38. Number 199, Tuesday, October 18.

1973; i.e.. as of October 16, 1973, strontiumis not included in the National PollutantDischarge Elimination System.

II

A

B

C

REAGENTS

Fifty percent by volume hydrochloric acid

SH4CH, 3N

Stock Strontium Solution

Weigh 1.685 g of anhydrous SrCO3 andplace it in a 500 ml Erlenmeyer flask.Place a small funnel in the neck of theflask and add 50% HC1 slowly until all ofthe SrCO3 has dissolved. Add 200 ml ofdistilled water and boil for a few minutesto expel CO2. Cool and add a few dropsof methyl red indicator. Adjust to theintermediate orange color by adding 50%by volume HCl or 3N NH4OH. Transferquanta, ìvely to a 1 liter volumetricflask and dilute to the mark with distilledwater.

I.00 ml = 1.00 mg Sr+ 2

D Standard Strontium Solution

Dilute 25.00 ml of stock strontium solutionto 1000 nal with &stilled water. Use thissolution for preparing Sr standards in the1- 25 mg/1 range.

1.00 ml = 25. 0,mg Sr+ 2

III INTERFERENCE CONTROL

The radiation effect of possible interferingsubstances is equalized throughout thestandards by use of the standard additiontechnique.

CH. MET. lab. 3 d. 5.76

IV INSTRUMENT OPERATING CONDITIONS

Theoretical wavelength 460. 7 nm

Fuel pressure 7.5 lbs/in2

Oxygen pressure 10 lbs/in2

For all other conditions needed, consult the

manufacturer's instrument manual.

V PROCEDURE

1 Number the five flasks provided 0, 5, 10,

15, and 20.

2 Into each of the five, pipette 10.0 ml of the

unknown.

3 Using a clean pipette, add 10.0 ml of the0 mg /1 standard into flask 0.

4 Again using a clean pipette, add 10.0 mlof the 5 mg /1 standard into flask 5.

5 Proceed in a similar manner using the 10,15, and 20 mg /1 standards and flasks 10.15, and 20.

8 Stopper and shake all five flasks.

7 Mark five plastic cups 0, 5, 10, 15, and 20.

8 Fill the plastic cups about 3/4 full with theappropriate solutions from theflasks.

9 Fill a 6th plastic cup with distilled water.

10 The power toggle switch (on left side ofinstrument) is already turned on.

11 Set the sensitivity knob to the standbyposition.

12 Set the phototube voltage knob (on right side

of instrument) to position C.

13 Set the wavelength knob to the theoreticalvalue of 461 nm (the scale is at the top ofthe instrument).

18-1

Flame Photometry ',taboret° (Strontium

14 Set the fltr - shtr - open knob in the shtrposition (closed position).

15 Open the main valve on the oxygen cylinder;all other oxygen gauges are already set.

18 Open the main valve odithe hydrogencylinder; all other hydrogen gauges arealready set.

17 Raise the door on the right side of theburner housing (behind instrument).

18 Cautiously bring a lighted match to the tipof the burner in the housing.

19 Close the door on the right side of theburner howling.

20 Caution: Do not place any part of yourbody over the coils on the top of the burnerhousing. Hot gases are escaping.

21 Raise the silver lever between theinstrument and the burner housing tothe vertical position.

22 Place the plastic cup containing distilledwater in the cup holder which is nowexposed at the right aide of the burnerhousing.

23 Push the silver lever clockwise so thatthe cup holder swings into the burnerhousing and the water is aspirated. Adistinct difference in sound will be noticedwhen water or a sample is being aspirated.If at any time during the determinationthis sound again changes, it will indicatethat all of the liquid has been aspiratedfrom the cup.. Simply move"the silverlever and refill the cup with the appropriateliquid.

24 Do not allow air to be aspirated for morethan about 15 sec. If there is any delay,aspirate distilled water until the problemcausing the delay has been corrected.


26 Turn the dark current knob until theneedle reads 0 on the percent transmittancescale.

18-2


28 Repeat step 26.

29 Swing the cup of distilled water out of theburner housing and replace it with cup 20.Swing this cup back into the burner housing.

30 Turn the fltr - shtr - open knob to the openposition (this opens the shutter).

31 Turn the wavelength knob slowly to theleft; the needle will move to the left.

32 At some point the needle will suddenlyswing toward the right. It will probablybe necessary to make adjustments withthe slit knob in order to keep the needleon-scale while finding the point at whichthe needle swings back to the right. Recordthis wavelength. It i5 the peak wavelength.Do not change this setting until indicatedin the instructions.

33 If the point at which the needle swings backto the right is overshot, turn the wave-length knob about turn to the right andrepeat steps 31 and 32.

34 Make the needle read 100% transmittanceby turning the slit knob. Record the slitmm reading. Do not change this setting.100 is the peak percent transmittancereading for this solution.

35 Turn the fltr - shtr - open knob to the shtrposition (this closes the shutter).

36 Replace th cup 20 with 15.


38 Record the percent transmittance reading.


40 Repeat steps 36, 37, 38, and 39, usingcups 10, 5, and 0 in turn.

41 Aspirate distilled water and using the darkcurrent knob make the needle read 0 percenttransmittance.

42 Aspirate cup 20.

Flame Photometry Laboratory (Strontium)


44 Slowly turn the wavelength knob to the left.The needle will move to the right and atabout 5 -30 percent transmittance willmove no farther to the right. Record thewavelength reading. This is the back-ground wavelength. Do not change thissetting. The percent transmittance isthe background reading for this solution.

45 If the point at which the needle moves nofarther to the right is overshot, tarn thewavelength about turn to the right andrepeat step. 44.




49 Record the background transmittancereading for the solution.


51 Repeat steps 47. 48. 49, and 50 usingcup 10. 5, and 0 in ti.rn.

52 Aspirate distilled water for about 15 sec.

53 Turn the sensitivity knob to the standbyposition.

54 Close the main valve on the hydrogencylinder.

55 Close the main valve on the oxygen cylinder.

56 Empty all six plastic cups and discard them.Empty the flasks.

57 Leave the power toggle switch (on left sideof instrument) on.

58 For each of the 5 solutions, subtract thebackground percent transmittance readingfrom the peak percent transmittancereading.

59 Using the graph paper provided in themanual. plot the 5 differences vs. theappropriate concentrations. Draw astraight line connecting the five points.This is the calibration graph.

60 Double the difference value obtained for thesolution in cup 0.

61 Find the value on the percent transmissionaxis.

62 Draw a straight line to the right until itintersects the calibration line.

63 From the point of intersection, draw a linestraight down to the horizontal axis.

84 This is the concentration of the unknown.

REFERENCE

Standard Methods for the Examination of Waterand Wastewater, 13th Edition, page 328.Method 155A. 1971.

18-3


10.0 ml unknown+

10.0 ml 0.0 mg/I std.

10.0 ml unknown+

10.0 nil 5.0 mg/1 std.

10.0 ml unknown+

10.0 ml 10.0 mg/1 std.

10.0 ml unknown+

10.0 nil 15.0 mg/1 std.

10.0 ml unknown+

10.0 ml 20.0 mg/1 std.

Peak Wavelength

Percent Transmission Readings

Background Peak Difference

Background Wavelength

Slit

18-4

nm

nm

mm

1 *v., ....,

I

U..I..U..I.'

I.

uuuuuiuuuiuiaua

aiiiaiiiusiiuuumuuaiiiiiiu..'.......'u.....ui.....iuU....auiuuuui.i.eiu..uiivaiiiuuiuiicuuius.uuu.........a....usiusisuauiiiuii.iuuu.auaiu....ii..i.uiuiI.I....Ui..i.uuu..iu.i.ii..uiu.iuuu.uuUu..i.....uuuiuuuuiiuu....u.II...u.aI...0 .umuuIuuu.u.Iu....II....Ii.ari.u.uuuiiuuI...U..IIIsu..Uuuuiiiuuu

L'-''R &I1s UI à'1,_ioiiIaUaiIiiUuuiiiUauIuUuuaIuiUUUiiIUUIUUu........uu.u.....u..........u......U...U..iiuuuaiuiieiia.......u........u....u uiiuiii.. uU......iuiiuuuuiuauauuuuiuuuiusisiuuiiiuuuiui. uuuuimui aiiuuawuiuiuu.iuu.sI. ....iiuiii.i.u..uuiuuuuuuuiausi.a.......................u.....'..u.u....uii......a........'....'us.....'.......u'....uiiiuuuiauuiiiuuiu uiiuuuuuuiusaiiiuuaiuiiuiuu I...u......uumu.u...auuuuisu........um.ua.ii uiiiiiu u u uuiuii u..i.a..im iiiUI u.iiuiuuusuuuuauuia.auuiuuiui u uaiui*iasuaiiiuuuuiuuuuuiu.iiuu.uuuuuu.u..iu..u.i..ua.auuuuiuivaiuu.iIiIUUUIIUIUIIUUUNUIUIIIIIUUIUUIUUU

U...U...'U.....

uuiaa.uiuuuusauuau.uiuiiui..uuuuuiII I.1.,diUUUUUUUIa. *aau..a.u.iui..IIRUI.slaIuuu.I.a.us.auuaii.su.iuuuuivaaauuuaua...uiuu.uia.uu.s.iau.iuua.aiu.u.uiiiuivauiiuiuaiuauuuauauuuaauuuauaiiauiuivaaaiaIUUUIUIUUUUIUUIaiuasuu.aau...iiiuuauiuiuuiuu.........au...'auuaauauauuuaaaaaa.....u.au..uu..'.uuaaiiiiiisuiiiuiaaauaauau ..ui......uua..as.a..mauu.aaaa uuuuaaia .........i..a..gi...ISUUUUUUUIUUUIUUUUUUIUIUUUIUUUSIIIIUUUIIUUUUUUIUUUUUUII

uasaiauiumiuiiuuiuiuauauaaaauaisaiiauusauiauuuauiuui...u.....I.I.. ivauaaiaiUuuailull UIIIIUUUUUIIUUUUIUIUIUIIIIUIuuUUulUIUlluulllllilIllfllUUllUiuiaauuuiiaaauaaaiauivauaiUllI IUUUUIIIUI UUUIIIUUIIUUIIUUIUUIUI1IUUIUUUIUUIIUUUIIIUUUIIUUl.l..lu.....a...U.I...11....l.u.I.lIIIIIIIUIIIIII I.IIRIUI.0 iuaiiu.l.I.l.Ululu.l...l..'..u..I..lU...UUIUUIUlIUUUIUUIUIIUUUIUUUUUUII..IIUU.UUI.II.UI.I..II.U..UIIu....u'...ul.l...U..'..u..uIl.l'aIUUUIUUIUIUIIIUUUUUIUUIUUUUUUlUllU...iU...U...U...I...a...U...S...

I.U.


This outline was prepared by C. R. Feldxaann,National Training Center, MOTD. OWPO.USEPA, Cincinnati, Ohio 45268.

18-6

Descriptors: Chemical Analysis, LaboratoryTests. Water Analysis, Strontium, AlkalineEarth Metals. Metals

1

1

1

1

AUTOMATION OF CHEMICAL ANALYSIS

I INTRODUCTION

Environmental pollution control and monitor-ing have become, in very recent years, thefocal point of national and internationalattention. The analytical chemist has animportant role in matters regarding pollution,for be determines not only the amount ofpollution in air or water but also the efficiencyof prevention and treatment methods. Theincreasing number of samples and of meas-urements to ba made requires him to employevery means possible to keep abreast withthe demands.

The need for instrumental and automatedanalytical techniques is well recognized.Automated sensors with continuous readoutfor the measurement of key constituentsassociated with pollution would be ideal.

II DEGREES OF AUTOMATION

A Maximum automation - continuous samplingdirect from source, with telemetering ofresults to central computer.

B Medium automation - grab samples inlaboratory with multiple simultaneousinstrumental analysis, recording, cal-culating and digital print-out of data.

C Minimum automation - maximum sim-plicity and efficiency of nianual analyticalmethods.

Possible considerations are:

1 Automatic pipettes, burettes andreagent dispensers

2 Choice of glassware - precision andaccuracy of graduated test tubes andcylinders vs. volumetric flasks shouldbe compared to the accuracy of theanalytical method.

3 Electrical "on and off" timers on hotplates (C. O. D. ) and automatedinstruments (Technicon).

CH. MET. 23a. 1.741 2G

4 Use of semi -micro methods:

a TEN - digest and diatill 50 mlsample in place of 500 ml.

b B.O. D. - incubate 125 ml samplesin place of 300 ml.

5 Combining reagents:

a Nitrite - reagent contains buffer,diazotization and coupling reagent.

b Phosphorus - reagent contains acid,molybdate and reluctant.

6 Laboratory management - eliminationof needless repetitious work

a Lab's sample log number vs.engineer's coding system

b Distributing samples to instrumentalstations in permanent labeledcontainers

c Permanent data sheets vs. individualslips of paper for record keeping.

d Washing glassware - cleaning acidvs. soap; labor costs

e Documented laboratory procedures,techniques, and practices to avoidvariations.

III SENSORS AND INSTRUMENTATION

In order to measure the amount of any con-stituent in a liquid or gaseous medium on anautomated system, a relationship betweenthe parameter and an electrical signal mustbe established. When such a relationship isestablished and is not affected by uncon-trollable variables, electronic technologyalready available can be applied to the sensor'ssignal for digital print-out of concentrationor for telemetering and conversion of thesignal to meaningful data through computer-ization.

19-I

Automation of Chemical Analysis

Various types of sensors and instrumentalmethods for water quality parameters areshown in Table 1. Parameters whosemeasurement is based on an electrochemicalprinciple are very desirable because of thepossibility of developing an electrode orprobe. In such a case, no intermediatechemical reaction is needed for the ion inthe sample to produce a signal.

A Probes for many specifilo ions are avail-il-757,7-wwever, most are affected byinterferences and variables. The activityof the specific ion, sensed by the electrodes,is related to its concentration by theactivity coefficient, which depends otn thetotal ionic strength of the solutiont4,.By increasing the ionic strength in allthe samples, the effect of individualvariation between samples can be elim-inated.

With some specific ion probes, a pHadjustor and a complexing agent must alsobe added to the samples to eliminateinterference from hydrogen, hydroxyland metal ions.

Specific ion probes are valuable in someapplications but must be evaluated forsuch use. Most of the automated con-tinuous monitoring systems for waterquality that are now available use sensorswhich require no intermediate chemicalreaction and only measure parameterssuch as temperature, pH, conductivity,solar radiation, turbidity, chlorides and'dissolved oxygen.

B Automatic Titrators with preset end nointare applicable for many measurementsusing various types of electrodes. Up to10 samples can be titrated withoutattendance.

C Atomic Absorption - 'Ibis method isapplicable for determining the concentrationof many metals. The instrument isusually used only in a laboratory. It isadaptable to:

1 an automatic samples2 automatic reagent additions3 automatic extractions

19-2

D Photometric Methods - With automatedaddition of reagents, the end product canbe measured with flow through cells in:

1 Visible rangePas

2 Ultra violet range

E Organic Carbon Analyzer - Manual in-jection of samples vs. automated samplingsystem is dependent on amount of sus-pended solids in samples.

IV PRESERVATIVES AND QUALITY CONTROL

A Preservatives added to samples must becompatible with the automated method tobe used. For example, sulfuric acid andmercury may interfere with colorimetricmethods.

With automated methods, the preservativeshould be added to samples, standardsand control blanks.

B Routine standardization of automatedinstrument - After every 10 or 20 samples:

1 Check baseline with a control blank.

2 Check sensitivity with a standard

The laboratory should also establish aprogram for routine interlaboratory andintralaboratory quality control.

AUTOMATED WET-CHEMICAL SYSTEMS

Many chemical parameters for water qualityare defiled by the conditions in the chemicalanalytical methodology of Standard Methods,e. g. C, O. D. , chlorine, phenolics and TKNdigestion. All new automated methods shouldbe compared to the manual Standard Methodfor accuracy.

A Robot Chemist - individual sample andreaction vessel, syringe type measurementof reagents, micro-switches and values,complex mechanics. Many are not verywell accepted by chemists.

1`''j

B Mecolab - individual sample and reactioncontainers; maximum 15 samples withoutmanual change

1

C Technicon AutoAnalyzer - a continuousflow through system propelled by de-pression of various inside diameters oftyjjon tubing using constant speed rollers.The modular system is capable of almostall manual chemical techniques, such assampling (up to 200 samples), filtration,dilution, addition of reagents, mixing,heating, dialysis, steam distillation,extraction with an immiscible solventand digestion. The concentration of theend products is measured with a photo-meter or fluorometer and recorded. Theversatility of the Auto Analyzer enablesthe automation of almost any manualprocedur e.

The success of each automated procedureis primarily dependent on the manifolddesign. Effort should be made to achievethe maximum rate of each chemical re-action by determining the optimum con-ditions of time, temperature, pH andreagent concentration. When this isobtained, there is generally an increasein sensitivity, thus requiring a smalleraliquot of sample with less chance ofinterfering ipne, a shorter flow streamthrough the system which will produce afaster response and wash out (shortersampling time) and smoother, steadierplateaus.

The Teohnicon AutoAnalyzer is nowinternationally used for automating wet-chemical procedures in the laboratory.The Analytical Quality Control Laboratoryof EPA includes Technicon procedures inthe manual(9) of methods for Office ofWater Programs laboratories.

VI CONTINUOUS MONITORING SYSTEM

A Several criteria or objectives have beenestablished for using the AutoAnalyzer forcontinuous monitoring of a stream or atreatment plant influent or effluent:

1 9 '

Automation of Chemical Analysis

1 Capability of continuous operation forone week with a minimum amount ofmanual manipulation (less than 4 manhours per week).

2 At least a + 5% accuracy at the 50%(0.30 absorbancerabsorption con-centration.

3 Reagents and base line should remainstable for a week.

4 Reagent consumption should be lessthan 4 liters per week.

5 There should be a minimum ,distanceof reaction-flow stream for a fastresponse and equilibrium of the system.

6 It is desirable that colorimetricchemistry obeys Beer's Law for rapidcalculations.

REFF1ENCES

1 McDermott, J. H. , Ballinger, D. G. ,Sayers, W. T. The Future of Instrument-ation in Water Pollution, presented at the14th National Symposium, AnalyticalInstrument Division, Instrument Societyof America, May 21, 1968, Philadelphia, PA.

2 Elving, P. J. the Need for Instrumental andAutomated Analytical Techniques, JWPCF,39, 12. 1967.

3 Middleton, F. M. , He ckroth, C. W . Where doWe Go From Here? J. Water & WasteEng., May, 1971.

4 Orion Research, Inc. Analytical MethodsGuide, October, 1971.

5 Mancy, K. H., Instrumental Analysis for WaterPollution Control, Ann Arbor SciencePublishers, 1971.

6 Reilley, C. H., Advances in AnalyticalChemistry and Instrumentation, Vol. 5,Interscience Publishers, 1986.

White, W. L., Erickson, M. M., Stevens,S.C. Practical Automation for theClinical Laboratory, C. V. Mosby Co. 1988.

(Continued on Page 35-5)19-3

TABLE I

INSTRUMENTAL METHODS FOR ANALYSIS OF WATER QU,A LITY PARAMETERS

l ARAMETER SENSOR MODE OF MEASUREMENT USES* EVALUATION

TEMPERATURE 'robe thermacoule P F L C excellent

.H electrode potentiometric P, F, L, C good

CONDUCTIVITY èlectrode conductivity

photometric

P, F, L, CP1 F, L, C

good

excellentSOLAR RADIATION photocell

DISSOLVED QXYGEN electrode polirographic P, F, L, C good

CHLORIDE electrode conductivity

BRIABOACric

P, Ft L. CE, L, C

fair

wet-chemical

TURBIDITY hotocell (scatter) hotometric P F LC ood

COLOR photocell (absorbance) photometric F, L, C good

SUSPENDED SOLIDS photocell photometric

ALKALINITY wet- chemical potentiometric Lwet-chemical photometric L

ACIDITY wet-chemical potentiometric Lwet-chemical photometric L

CALCIUMelectrode potentiometric P. L wilM.

flamep,----

atomic absorption L ....wet-chemical photometric L

MAGNESIUM flame atomic abso tion Lwet-chemical slictometric L

AMMONIA electrode potentiometric P, L, C(?)wet-chemical P, L, C(?) good

NITRITE wet-chemical photometric P, L, C(?) _good

NITRATE probe ootentiometricwet-chemical photometric P, L, C good

PHOSPHORUS wet-chemical photometric Pi L, C good

SULFATE wet - chemical photometric L

SULFIDE wet- chemical photometric L

12D

.

130

TABLE 1 CONTINUED

INSTRUMENTAL METHODS FOR ANALYSIS OF WATER QUALITY PARAMETERS

PARAMETER SENSOR MODE OF MEASUREMENT USES* EVALUATION

TOTAL ORGANIC CARBON wet - chemical infrared Lwet-chemical flame ionization L

CYANIDE electrode P......e_potentiometricwet-chemical photometric P, L, C(?

PHENOLICS wet-chemical photometric

SPECIFIC ORGANICS gas - chromatography electron- capturephotometric

LL ---lictuid-chromotography

BACTERIOLOGICAL wet-en zyamatic ...photometricwet- fluorescence fluorometric i

*F-Field, monitoring streamsP-Treatment Plants,' monitoring influent or effluentL- Laboratory, with individual or automated sampling systemC-Continuous operation, for monitoring streams or treatment plants for operational control.

REFERENCES (CONTINUED)

8. Kamphake, L.. J., and Williams, R. T. Automated Analysis forEnvironmental Pollution Control. In Press. November 1971.Taft Water Reseach Center, Cincinnati, OH.

9 Methods for Chemical Analysis of Water & Wastes, 1971, EPA.Analytical Quality Control Laboratory, Cincinnati, OH.

10 Buggie, F. D. Recent Advances in Automated Multiple Analysisof Water Quality in Industry. Tecticon Industrial Systems,June 1971.

This, outline was prepared 'by Lawrence .1. Kamphake,Research Chemist, Waste Identification & AnalysisProgram, AWTRL, RATWRC, NERC, EPA,Cincinnati, OH 45268.

Descriptors: Analysis, Automation, ChemicalAnalysis, Instrumentation, Water Analysis

1 3 2

TURBIDITY

I INTRODUCTION

'rtarbidity as a water quality index refersto the degree of cloudiness present.Conversely, it is an index of clarity.

A Definition(1)

Turbidity is an expression of the opticalproperty that causes light to be scatteredand absorbed rather than transmitted instraight lines through samples of waster.

B Relationship to Suspended Solids

This optical property, turbidity, iscaused by suspended matter. The size.shape and reflection/absorptionproperties of that matter (not its weight;determine the degree of optical effects.It is very possible to have water withhigh turbidity but very low mg/1 sus-pended solids. Thus one cannot useturbidity results to estimate the weightconcentration and specific gravity of thesuspended matter.

C Causes

1 clay. sand

2 silt, erosion products

3 microscopic and macroscopicorganisms

4 finely divided organic products

5 others

D Effects on Water Quality(2)

Turbidity is an indicator of possiblesuspended matter effects such asimpeding effective chlorine disinfectionand clogging fish gills. However, thefollowing list is limited to those effectsassociated with the optical (clarity)nature of turbidity.

CH. TUBB. 2.11.771 )J )%.

1 Reducing clarity in water

a drinking water qualityb food processingc industrial processesd fish (seeing natural food)

e swimming/water sports

2 Obscuring objects in water

a submerged hazardsb water sports

3 Light penetration

a zffects depth of compensationpoint for photosynthetic activity(primary food production).

4 Thermal Effects

High turbidity causes near surfacewaters to become heated because ofthe heat absorbancy of the particulatematter.

a Results in lower rate of oxygentransfer from air to water.

b Stabilizes water column andprevents vertical mixing.

1 decreases downward dispersionof dissolved oxygen

2 decreases downward dispersionof nutrients

E Criteria for Standards(2)

1 Finished Drinking Water - Maximumof one unit where the water enters tie)distribution system. The proposed'standard is one unit monthly average andLive units average of two consecutive days.Under certain conditions a five unitmonthly average may apply at state option.

2 For Freshwater Aquatic Life andWildlife - The combined effect of colorand turbidity should not change the

20-i

Turbidity

compensation point more than 10%from its seasonall establishednorm, nor should such a changeplace more than 10% of the biomass ofphotosynthetic organisms below thecompensation point.

3 Turbidity Criteria Used by Industries:

a Textiles - 0.3 to 5 units

b Paper and allied products - Rangesfrom 10 to 100 units, depending ontype of paper.

c Canned, dried and frozen fruits andvegetables - Same as for finisheddrinking water (1 turbidity unit).

F Processes to Remove Turbidity (Solids)

1 Coagulation

a pre-chlorination enhances coagulation

2 Sedimentation

3 Filtration

4 Ae

5 Others

II VISUAL METHODS TO ESTIMATETURBIDITY

A Early Efforts

In the early 1900's, Whipple and Jacksonmeasured turbidity and developed a calibra-tion scale for turbidity instruments.

B Jackson Candle Turbidimeter

Later Jackson developed apparatuswhich utilized the same "extinction"principle as the instrument devisedearlier with Whipple.

20-2

1 Instrument

SprimplooddCyliade

Glatt Tub

meal Tube

Candle

Figure I JACKSON CANDLETURSIDIMETER

The,sample was poured into a flat -.boftomed, graduated glads tube

held over a special candle. Aturbidity reading was taken whenthe operator, observing from thetop of the tube, saw the image of thecandle flame disappear into a uniformglow. The reading related the finaldepth of sample in the tube with tubecalibrations obtained from a standardsuspension solution.

2 Standard Suspension

The.standard was a suspension ofsilica prepared from Fuller's ordiatomaceous earth. This wasdiluted to prepare a series ofstandard suspensions to graduatethe turbidimeter. Graduations on allJackson turbidimeters are made inconformity to this original data.Other suspensions are standardized byusing the pre-calibrated turbidimeter anddiluting accordingly.

3 Unit Used

Jackson Turbidity Unit (JTU) - partsper million suspended silica turbidity.

4 Standardization of Apparatus

The current edition of Standard

Methods(1) contains specificationsfor the three essential components,i.e.. the calibrated glass tube, thecandle and a support.

5 Current Standard Suspension

Solutions (1)

1 Natural turbid water from the samesource as that tested gives bestresults. Determine turbidity withthe instrument, then dilute to valuesdesired.

2 The supernatant of a settled solutionof kaolin is also used as a standard.

6 Limitations of Method

a Apparatus - difficult to exactlyreproduce flame as to intensity andactual light path length. In general,it is a rather crude instrument withseveral variables that affect accuracy.

b Very fine suspended particles do nottend to scatter light of the longerwavelengths produced by the candle.

c Very dark and black particles canabsorb enough light in comparisonto the scattering of light to causean incorrect reading of imageextinction.

d Turbidities below 25 JTU cannotbe direc ly measured. For lowerturbidities (as in treated waters),indirect secondary methods arerequired to estimate turbidities.

C Hellige Turbidirneter(4)

This instrument utilizes the sameextinction principle as the JacksonCandle Turbidirneter.

1 Equipment

An opal glass bulb supplies the lightwhich is reflected (usually through a

Turbidity

filter) upward through the samplewhich is contained in a glass tube.The entire system is enclosed in ablack metal box. The operator viewsthe sample by looking downward throughan ocular tube screwed into the top ofthe box and adjusts the brightness of acentral field of light by turning acalibrated dial on the outside of theapparatus. The point of uniform lightintensity occurs when a black spot inthe center of the field just disappears.

2 Range of Applicability

The equipment offers a choice of bulbs,filters and volumes of sample tubes.The variety affords a means to directlymeasure turbidity ranging from 0 through150. The ranges can be extended bydilution.

3 Results

The final reading from the dial istranslated into ppm silica turbidity unitsby using a graph corresponding to thebulb, filter and volume of sample used.

4 Standard Suspension Solution

Standardizing suspensions are not usedby the operator. The graphs aresupplied by the company for eachinstrument.

D Secchi Disk(5)

This is a very simple device used inthe field to estimate the depth of visibility(clarity) in water.

1 Equipment

The disk is a weighted circular plate,20 cm in diameter, with opposing blackand white quarters painted on the surface.The plate is attached to a calibrated lineby means of a ring on its center to assurethat it hangs horizontally.

2 Readings

The disk is lowered into water untilit disappears, lowered farther, then

20-3

Turbidit

raised until it reappears. Thecorresponding visibility depth (s)are determinea from thecalibrated line. Some read bothdepths and average them. Someread only the reappearance depth.

3 Standardizing the Procedure

There are many variables (positionof sun and of observer, roughness ofbody of water, etc.) that affectreadings. However, the sameobserver using a standard set ofoperating conditions can provideuseful data to compare the visibilityof different bodies of water.

4 Application of Results

Limnologists have found itconvenient to establish a Secchidisk "factor" for estimating thephotic depth where light intensityis about 1 per cent of full sunlightintensity. The true photic depth isdetermined by use of a submarinephotometer and at the same timethe observer takes a series ofSecchi disk readings to obtain anavetage. Dividing the true photicdepth by this average gives afactor which can be used to multiplyother disk readings for an approxima-tion of photic depth.

E Status of Visual Methods for ComplianceMonitoring

The Federal Register(6) "List ofApproved Methods" does not includeany of these visual methods for NationalPollutant Discharge Elimination System(NPDES) requirements. The visualmethods are not recognized in theFederal Register(3) issue on InterimDrinking Water Regulations, either.

20-4

III NEPHELOMETRIC MEASUREMENTSFOR COMPLIANCE MONITORING

The subjectivity and apparatusdeficiencies involved in visual methodsof measuring turbidity make eachunsuitable as a standard method.

Since turbidity is an expression of theoptical property of scattering orabsorbing light, it was natural thatoptical instruments with photometerswould be developed for this measurement.

The type of equipment specified for

compliance monitoring(3, 6) utilizesnephelometry.

A Basic Principle(7)

The intensity of light scattered by thesample is compared (under definedconditions) with the intensity of lightscattered by a standard referencesolution (formazin). The greater theintensity of scattered light, the greaterthe turbidity. Readings are made andreported in NTUs (NephelometricTurbidity Units).

B Schematic

Lamp

Meter /1

PhotocAsi

Lens

a

TUIblditY ",meinStitt.. Ugh'

Sample Loll(Top V.410

Figure 2 NEPHELOMETER(90° Scatter)

Light passes through a polarizing lensand on to the sample in a cell.Suspended particles (turbidity) in thesample scatter the light.

Turbidity

Photocell (s) detect light scattered bythe particles at a 90° angle to the pathof the incident light. This lightenergy is converted to an electricsignal for the meter to 'measure.

1 Direction of Entry of IncidentLight to Cell

a The lamp might be positioned asshown in the schematic so thebeam enters a sample horizontally.

b Another instrument design has thelight beam entering the sample(in a flat-bottom cell) in a verticaldirection with the photocellpositioned accordingly at a 90°angle to the path of incident light.

2 Number of Photocells

The schematic shows the photocell (s)at one 90 degree angle to the path ofthe incident light. An instrumentmight utilize more than one photocellposition, with each final position beingat a 90 degree angle to the sample liquid.

3 Meter Systems

a The meter mighNnasure thesignal from the scattered lightintensity only.

b The meter might measure thesignal from a ratio of the scatteredlight versus light transmitteddirectly through the sample to aphotocell.

4 Meter Scales and Calbration

a The meter may already becalibrated in NTUS. In this case.at least one standard is run ineach instrument range to be usedin order to check the accuracy ofthe calibration scales.

b If a pre-calibrated scale is notsupplied, a calibration cube isprepared for each range of theinstrument by using appropriatedilutions of the standard turbiditysuspension.

C EPA Specifications for Instrument Design(7)

"V

Even when the same suspension is usedfor calibration of different nephelometers,differences in physical design of theturbidimeters will cause differences inmeasured values for the turbidity of thesame sample. To minimize such differences.the following design variables have beenspecified by the U. S. Environmental ProtectionAgency.

1 Defined Specifications

a Light Source

Tungsten lamp operated at not lessthan 85% of rated voltage and at not ;,more than rated voltage.

b Distance Traveled by Light

The total of the distance traversedby the incident light plus scatteredlight within the sample tube shouldnot exceed 10 cm.

c Angle of Light Acceptance of theDetector

Detector centered at 90° to theincident light path and not to exceed

30q from 90°.

(Ninety degree scatter is specifiedbecause the amount of scatter varieswith size of particles at differentscatter angles).

d Applicable Range

The maximum turbidity to bemeasured is 40 units. Several rangeswill be necessary to obtain adequatecoverage. Use dilution for samples if

their turbidity exceeds 40 units.

2 Other EPA Design Specifications

a Stray light

Minimal stray light should reach thephotocell (s) in the absence of turbidity.

20-5

Turbidity

20-6

Some causes of stray light reachingthe photocell (s) are:

1 Scratches or imperfections inglass cell windows.

2 Dirt, film or condensation on theglass.

3 Light leakages in the instrumentsystem.

A schematic of these causes isshown in Figure 3.

Ught Leakagefrom Lam System

Lame Lan,

Photeceilisi

Light Lsokeg. ft..Tansmithrd Light

1011Mh7010LiallAIP 11=1121711111111

111411/4mom NomLIINmEnsw.IMIIMVAW

I.VAPP.

Ught Scow, by Om tubs(TOP Viol.)

Figur 3 NEPHELOMETER

SOURCES OF STRAY LIGHT

Stray light error can be as muchas 0.5 NTU. Remedies are closeinspection of sample cells forimperfections and dirt, and gooddesign which can minimize theeffect of stray light by controllingthe angle at which it reaches thesample.

b Drift

The turbidimeter should be freefrom significant drift after a shortwarm-up period. This is imperativeif the analyst is relying on a manu-facturer's solid scattering standardfor setting overall instrumentsensitivity for all ranges.

Sensitivity

In waters having turbidities less thanone unit, the instrument should detect

turbidity differences of 0.02 unitor less. Several ranges will benecessary to obtain sufficientsensitivity for low turbidities.

3 Examples of instruments meeting thespecifications listed in 1 and 2 aboveinclude:

a Hach Turbidimeter Model 2100 and2100 A

b Hydroflow Instruments DRT 100, 200,and 1000

4 Other turbidimeters(12) meeting thelisted specifications are alsoacceptable.

D Sources of Error

1 Marred Sample Cells

a Discard scratched or etched cells.

b Do not touch cells where lightstrikes them in instrument.

c Keep cells scrupulously clean, insideand out. )

1 Use detergent solution.

2 Organic solvents may also beused.

3 Use deionized water rinses.

4 Rinse and dry with alcohol oracetone.

2 Standardizing Suspensions(7)

1 `1

a Use turbidity - free water forpreparations. Filter distilled waterthrough a 0.45 p ru pore size membranefilter if such filtered water shows alower turbidity than the distilled water.

b Prepare a new stock suspension ofFormazin each month.

c Prepare a new standard suspensionand dilutions of Formazin each week.

3 Sample Interferences

a Positive

1 Finely divided air bubbles

b Negative

1 Floating debris

2 Coarse sediments (settle)

3 Colored dissolved substances(absorb light)

(7)E Reporting Results

NTU

O. 0-1. 0

1-10

10-40

40-100

100-400

400-1000

Record to Nearest:

0.05

0. 1

1

5

10

50

>1000 100

F Precision and Accuracy (7)

1 In a single laboratory (MDQARL).using surface water samples atlevels of 26, 41, 75 and 180 NTU,the standard deviat4ons were + 0.60,+ 0. 94, + 1. 2 and + 4. 7 units,respectively.

2 Accuracy data is not available atthis time.

IV STANDARD SUSPENSIONS AND RELATED

UNITS(9)

One of the critical problems in measuringturbidity has-been to find a material whichcan be made into a reproducible suspen-sion with uniform sized particles. Variousmaterials have been used.

1 3

\A atural Materials

1 atomaceous earth

2 Fuller's earth

3 Kaolin

4 Naturally turbid waters

Turbidity__

Such suspensions are not suitableas reproducible standards because thereis no way to control the size of thesuspended particles.

B Other Materials

1 Ground glass

2 Microorganisms

3 Barium sulfate

4 Latex spheres

Suspensions of these also provedinadequate.

C Formazin

1 A polymer formed by reacting hydrazinesulfate and hexamethylenetetraminesulfate.

2 It is more reproducible than previously-used standards. Accuracy of + one percent for replicate solutions has beenreported.

3 In 1958, the Association of AnalyticalChemists initiated a standardized systemof turbidity measurements for the brewingindustry by:

a defining a standard formula for makingstock Formazin solutions and

b designating a unit of measurementbased on Formazin, i.e.. the FormazinTurbidity Unit (FTU).

4 During the 1960's Formazin was increasing-ly used for water quality turbidity testing.It is the currently recognized standard forcompliance turbidity measurements.

20-7

Turbidity

D Units

1 At first results were translatedinto Jackson Turbidity Units(JTU). Hoi ever, the JTU wasderived from a 'visual measurementusing concentrations (mg/li:_r) ofsilica suspensions prepared byJackson. They have no directrelationship to the intensity oflight scattered at 90 degrees ina nephelometer.

2 For a few years, results ofnephelometric measurementsusing.specified Formazinstandards were reported directlyas Turbidity Units (TU's).

3 Currently, the unit used is namedaccording to the instrument used formeasuring turbidity. SpecifiedFormazin,standards are used tocalibrate the instrument and resultsare reported as NephelometricTurbidity Units (NTUs).

V TURBIDITY MEASUREMENTS FORPROCESS CONTROL

B Applications of Control Data (10, 11)

1 Coagulation Processes

a To check the effectiveness ofdifferent coagulants.

b To check the effectiveness ofdifferent dosages.

c To regulate chemical dosages byautomating chemical feed controls.

2 Settling Processes

a To determine intermittent need forsettling processes.

b To control the sludge blanket heightin activated sludge treatment processes.

c To activate removal and re-cycling ofvery high density sludge from settlingtanks.

d To monitor effectiveness of settlingprocesses.

3 Filtration ProcessesThe schematic and design characteristicsdiscussed above for nephelometric a To determine intermittent needinstruments is the required method for for filtration.measuring turbidity for compliancepurposes. Turbidity data is also widely b To facilitate high rate filtrationused to check water for process design procesLes.purposes and to monitor water for processcontrol purposes. The nature of the c To prevent excessive loadings forliquids to be monitored, and the degree of filtration systems.sensitivity required for signalling theremedy to be applied have led to the d To check the efficiency of filtrationdevelopment of monitoring instruments.- systems.tion that differs in design or in principlefrom the instrument previously described. a To regulate filter backwash operations.

A Users of Control Data

1 Potable Water Treatment Plants

2 Municipal Wastewater Treatment Plants

3 Industrial Processers

20-8

4 Rust in Water Distribution Systems

a To locate sc.xrces of contamination.

b To monitor intermittent occurrences.

5 Steam Boiler Operations

a To detect corrosion products inboiler water.

Turbidity

b To detect evidences of corrosionin condensates.

c To determine the effectiveness ofcorrosion treatment measures.

C Varieties of Instrumentation

1 Surface Scatter Nephelometers

In forward - scattering instruments,the angle of the incident light isadjusted to illuminate the' surfaceof a smooth flowing liquid at an angleof about 15 degrees from horizontal.rather than beamed through a glasscell of the liquid as described for anephelozneter earlier in this outline.A photocell is located immediatelyabove the illuminated area so thatvertically scattered light fromturbidity in the sample reaches it.

1

ne

I ITurbidity Panicles t 1-" Photeteltsi1.......,.1

\ ' I , .

Scooter loiht

Siowea Wafer Surface

To awn

Figure 4 NEPHELOmETER (Surface Scatter)

Variations of the methodologyinclude sidescatter and backs catterdesigns.

a Advantages

I No glass sample cells areused. Attendant problems ofcleanliness and condensationare eliminated. The surfaceof the liquid provides a nearperfect optical surface which

t

is difficult to achieve in glass cells.

2 Stray light effects on the photocellare minimized because the simplerdesign eliminates some of the sourcesof stray light.

3 Since flowing sample is used,interferences from air bubbles and/or floating materials are quicklyeliminated.

4 This design is sensitive to the presenceof larger suspended particles.

b Disadvantage

As turbidity becomes high, penetrationof incident light decreases to cause afalling off of response.

2 Absorption Spectrophotometry

The incident light is beamed through asmooth, flat stream of sample and thetransmitted light (in contrast tonephelometric scattered light) is measuredby a spectrophotometer. A schematicis shown in Figure 5.

Turbidity Particles Absorb lists

FlowingWm*,

l\ I

1

ePlifoocoll(0 Neer

Figure 5 ABSORPTION SPECTROPHOTOMETRY

a Advantages

1 No glass sample cells are used.

2 The simpler design eliminatessources of stray light.

3 Applicable to measure highturbidities. e.g., in sludges.

b Disadvantage

I Low sensitivity for many applications.

2 Color constituents interfere.

20-9

Turbidity

VI SUMMARY

Turbidity measurements represent theoptical property of light scattering bysuspended solids. NTUs are an indexof the effects of the size. etc. , ofsuspended particles but cannot be usedto indicate ms /1 quantities of thoseparticles. The parameter is requiredfor finished potable water and is extreme-iy useful in reference to aesthetic quality(clarity), photic conditions and thermaleffects In bodies of water. It is alsowidely applied for process control ofwater and wastewater treatment andof industrial processes.

There have been difficulties in developinga satisfactory standard method for thismeasurement. Early methods dependedon a subjective judgement of an extinctionpoint where transmitted light balancedscattered light in rather crude apparatus.Although the apparatus was refined andstandardized to a large extent, thesubjectivity of these visual methods wasstill an unsatisfactory element of suchmethodology,

Eventually, optical instrumentation wasdeveloped to eliminate subjectivityfrom the measurement. Nephelometry(scattering) was chosen for the standardmethod and U. S. EPA has specifiedseveral instrument design criteria tofurther promote standardization of themeasurement.

Finding a suitable (reproducible)standard suspension has also been aproblem. Currently, Formazin isspecified as the standard because. todate, it is more reproducible thanother suspensions proved to be.

Establishing a meaningful unit progressedalong with development of instrumenta-tion and agreement on a standardsuspension. The current unit (NTU) isderived from the method of measurement,nephelometry, and use of a standardFormazin suspension.

Even with the efforts to standardize

20-10

instrument design, to find a suitablestandard suspelision, and to agreeon a meaningful unit, there are stillproblems about this measurement.Instruments meeting the designcriteria and standardized withFormazin suspensions can giveturbidity readings differingsignificantly for the same sample.

Another problem area is associatedwith sample dilutions. Work hasindicated a progressive error onsample turbidities in excess of 40 units)so such samples are to be diluted.However, obtaining a dilution exactlyrepresentative of the originalsuspension is difficult to achieve.Thus dilutions often significantly failto give linearly decreased resultswhen re-measured,

REFERENCES

1 APHA, kWWA. WPCF, Standard Methodsfor the Examination of Water andWastewater, 14th ed., APHA,1976.

2 National Academy of Sciences, NationalAcademy of Engineering, 1974 EPArevision of Water Quality Criteria,1972, EPA, GPO, Washington, DC20402, # 5501-00520.

3 U. S. Government, Code of FederalRegulations, Title 40, Chapter 1,Part 141 - National Interim PrimaryDrinking Water Regulations, publishedin the Federal Register, Vol 40, No. 248,Wednesday, December 24, 1975.

4 He llige, Inc.. Graphs and Directions forHe llige Turbidinieter, Garden City,NY, Technical Information #8000.

5 Lind, 0. T. , Handbook of CommonMethods in Limnology, Mosby, 1974.

6 U. S. Governinent, Code of FederalRegulations, Title 40, Chapter 1,Part 136 - Guidelines EstablishingTest Procedures for the Analysis ofPollutants, published in the FederalRegister, Vol 41, No. 232, Wednesday,December 1, 1976.

Turbidity

7 Methods for Chemical Analysis ofWater and Wastes. EPA - AQCL,Cincinnati, Ohio 45268, 1974.

This outline was prepared by Audrey D.Kroner, Chemist, National Training and

8 Analytical Quality Control in Water andOperational Technology Center, MOTD,OWPO, USEPA, Cincinnati, Ohio 45268

Wastewater Laboratories. EPA -AQCL, Cincinnati. Ohio. 1972.

9 News and Notes for the Analyst. HachChemical Company, Ames, Iowa.Volume 1, No. 3.

Descriptors: Chemical Analysis,Instrumentation, Secchi disks, Turbidity,Wastewater. Water Analysis

10 Sawyer and McCarty, Chemistry forSanitary Engineers, McGraw-Hill. t

NY, 1967.

1 1 Turbidimeters, Hach Chemical Company,Ames, Iowa. Fifth Revised Edition.1975.

12 Bausch and Lomb andTurner of California areadditional examples (9/77)of Models meeting thelisted specifications.

20-11

CALIBRATION AND USE OF A TURB1DIMETER (NEPHELOMETER)

I REAGENTS(1)

A Turbidity - free water - Pass distilledwater through a 0.45 um pore sizemembrane filter if such filtered watershows a lower turbidity than the originaldistilled water.

B Stock Turbidity Suspension - 400 units

1 Directions are for preparing 100.0 ml.Larger volumes may be required.

2 Solution, 1: Dissolve 1.00g hydrazinesulfate, (N112)2* H2SO4, in turbidity

free water and dilute to 100ml in avolumetric flask.

3 Solution 2: Dissolve 10.00ghexamethylenetetramine in turbidity-free water and dilute to 100m1 in avolumetric flask.

4 Suspension: Mix 5.0ml Solution 1with 5. Oml Solution 2 in a 100 mlvolumetric flask. Allow to stand for24 hours at 25 + 3°C; Dilute to themark and mix.

5 Stability: Prepare a new stocksuspension each month.

C Standard Turbidity Suspension - 40 units

1 Dilute 10.00m1 stock turbiditysuspension to 100m1 with turbidity -free water in a 100 ml volumetric flask.The turbidity is defined as 40 units.

2 Stability: Prepare a new standardsuspension each week.

D Dilute Standard Turbidity Suspension -4 units

1 Dilute 1.0 ml stock turbiditysuspension to 100 ml with turbidity-free water in a 100 ml volumetricflask. The turbidity should be i units.

2 Stability: Prepare a new standardsuspension each week.

E Secondary Standards

1 Solutions standardized with Formazincan be purchased from the manufacturerof the instrument.

2 Date such solutions. Store under theconditions specified. Discard andreplace when flocculation in thesolution is observed or when it failsa periodic check with a Formazin Standard.

(2)(3)

II PREPARATIONS FOR MEASUREMENTS

A Suspensions

1 Check date of preparation and pre-pare fresh solutions if required.

B Sample Cell

1 Cells should be cleaned immediatelyafter use as described in V. B. below.

2 Inspect cells for cleanliness. Ifnecessary. clean them using V. B. below.

3 Check cells for scratches and etching.Discard those with imperfections.

C Instrument

1 Scale - If a scale is inserted, checkthat it is in the correct position. Ifthe scale is blank, construct acalibration scale for each range onthe instrument. (See III B).

2 Zero - Adjust meter needle to zeropoint on scale as directed by manufactuer.

3 Lens - Check for cleanness. If required,follow raarilacturer's instructions forremoving and cleaning the lens.Accurate re-positioning of the lens iscritical for accurate measurements.

CH.TURB.lab. 1.11.77 1 1 -121-1

,

on andVse of a Tualhidimeter (Nephelometer)

141 -411PP"*"4 4 WeiltnUp Period - Follow

fasitsfacturer's instructions.ontinuous running is often suggested

because of the photomultiplier tubes.

Ir

5 Focus - Use template or methoddescribed by manufacturer to checkand. if necessary, set the focus.

D Determine Range of Sample Turbidity

1 Use steps 5 through 16 in III A belowEXCEPT Step 12 which should be:Obtain a turbidity readinirfigm thescale.

2 Note which Range (instrument scale)best "brackets" the turbidity of eachsample.

3 If the turbidity of a sample exceeds40 units, use the higher scales coo-yided to determine the dilution requiredso the final reading will be below40 units. For final measurements, usethe Range (Scale) appropriate for thediluted sample.

III INSTRUMENT CALIBRATION AND

MEASUREMENTS (1)

A Pre-Calibrated Scale

1. Each day, prepare at least onestandard for the required instrumentrange (s) (as determined in II. D. above)by diluting one of the Formazinsuspensions described above in I. Reagents.The table below gives "EXAMPLEDILUTIONS".

2 Setthe instrument RANGE knob atthe first range to be tested. (Theinstrument should be ON, warmed up,zeroed. etc., as in II C above).

3 Make any instrument adjustment specifiedby the manufacturer to use this RANGE.

4 Rinse the SAMPLE CELL 3 times withthe appropriate suspension (or sample).

5 Shake the suspension to thoroughlydisperse the solids. (For secondarystandards, check the manufacturer'sinstructions for this step).

EXAMPLE DILUTIONS

NO.

EXAMPLEINSTRUMENTRANGES VOLUME

TURBIDITYSTANDARD

FINALDILUTION

FINALTURBIDITY

1 0 - 0. 1 2.5 ml 4 unit 100.0 nil 0. 1

2 0 - O. 2 5.0 m1 4 unit 100.0 ml 0.2

3 0 - 0. 3 7.5 ml 4 unit 100.0 nil 0. 3

4 0 - 1 2.5 nil 40 unit 100.0 ml 1

5 0 - 3 7.5 ml 40 unit 100.0 ml 3

6 0 - 10 25.0 ml 40 unit 100.0 nil 10

7 0 - 30 7.5 ml 400 unit 100.0 nil 30

0 - 100 2 5 ml 400 unit 100.0 nil 100 .--9 0 - 300 75 ml 400 unit 100.0 ml 1-- l

100.0 mil

300

400IO

TT

0 - 400 100 ml 400 unit

0 - 1000 100 ml 400 unit 100. 0 nil 400

121-2

Calibration and Use of a Turbidir-eter (Nepnelometer)

6 Wait until air bubbles disappear inthe suspension.

7 Pour the suspension into theSAMPLE CELL up to the levelspecified by the manufacturer.CAUTION: Always hold the cellabove the area from which lightscattering is measured.

8 If applicable, screw cap on cell.

9 Wipe the outside of the cell with alint-free tissue.

10 Examine the suspension in the cellto check for air bubbles. If airbubbles are present. eliminatethem:

a by inserting the cell in the sampleholder and waiting a few minutesso bubbles rise above photo-multiplier tube. CAUTION: Morebubbles can form if a temperaturerise occurs.

b by holding the cell at the top and:

1 flicking side with your finger or

2 dipping the end of the cell into anultrasonic cleaning bath or

3 centrifuging the filled cell incups with rubber cushions andsurrounded with water.

4 NOTE: After any of theseremedies, again wipe theoutside of the cell. When airbubbles are gone, insert thecell in the sample holder.

11 Place the LIGHT SHIELD according tothe manufacturer's instruction.

12 Use the STANDARDIZING control toobtain a meter reading correspondingto the turbidity of the standardsuspension.

13 Remove the LIGHT SHIELD.

1 c,

14 Remove the SAMPLE CELL.

15 Discard the standard suspension.

16 Rinse SAMPLE CELL 3 times withturbidity - free water.

17 Use Steps 4 through 11 for eachsample (or diluted sample) to betested in this range. For samples.step 12 should be : Record theturbidity reading for the sample.Then do Steps 13 through 16 asabove.

NOTE: The final reading for samplesshould not exceed 40 NTU. If thisreading is exceeded for a sample.dilute it and repeat the calibration/measurement procedure above usingthe appropriate range and standard.(Selection of the range as describedin II. D. above should make thisunnecessary at this stage of theprocedure).

B Non-Calibrated Scale

Prepare a series of standards andmake a calibration scale for eachrange of the instrument.

a The instrument should be ON,warmed up, zeroed, etc., asin II C above.

b Prepare enough standards togive several points on eachscale so estimated readings canbe reasonably accurate.

c Use the table of EXAMPLEDILUTIONS in III A above toprepare the highest standardfor each instrument range.The rest of each calibratingseries can also be preparedby dilutions based on the in-formation in the table.

2 Self-prepared scales should alsobe calibrated each day using theprocedure given in III A above forpre-calibrated scales.

21-3

41,

.1...'_lion and Use of a Turbidimeter (Nephelometer)

3 Self-prepared scales are used forsamples in the manner described inIII A above for pre-calibrated scales.

IV CALCULATION/RESULTS(1)

A Diluted Samples

Multiply final sample readings by theappropriate dilution factor.

B Reporting Results

Report results as follows:

NTU Report to Nearest:

0.0 - 1.0

1 - 10

10 - 40

40 - 100

100 - 400

400 - 1000

> 1000

C Precision and Accuracy

0.05

0. 1

1

5

10

3 Observe stability times noted foreach in I. above.

B Sample Cells

1 Discard cells with scratches oretching.

2 Clean cells immediately after mewith this order of treatments

a detergent

b organic solvents. if required

c deionized water

d alcohol or acetone rinses todry

e lint-free tissue, if required

3 Store in a manner to protect thecells from scratches.

C Instrument

1 A line operated instrument should50 be permanently located so moving

it often is not necessary.100 2 Turbidimeters should be protected

from dust. especially the lens system.

1 In a single laboratory (ivIDQARL).using surface water samples atlevels of 26. 41, 75 and 180 NTU.the standard deviations were + 0.60.+ 0. 94, + 1. 2 and + 4.7 units.....respectively.

2 Accuracy data is not available at thistime.

STORAGE

A Standard Suspensions

1-kStore in glass containers at roomtemperature.

2 Excess light or heat may affectstability.

21-4

3 Store any removable parts asdirected by manufacturer.

4 Close any access doors.

5 Because of the photomultiplier tubes.the manufacturer may suggest con-,tinuous running of the instrument toinsure maximum accuracy for mea-surements. Frequency of use candetermine the actual routine for warm-uptime.

6 Follow any other storage directionsin the manufacturer's manual.

1 4 '-1.2: g

REFERENCES

C

Calibration and Use of a Turbidimeter (Nephelometer)

i Methods for Chemical Analysis ofWater and Wastes, EPA-MDQARL,Cincinnati. Ohio 45288, 1974.

2 Laboratory Turbidimeter, Model 2100A,Hach Chemical Co.. Ames. Iowa.1973.

3 Turbidimeters. Information CircularNo. 373A. Fisher ScientificCompany, Pittsburgh, PA

4 Analytical Quality Control, EPA-AQCL, Cincinnati, Ohio 45288,

1972

1 4

This outline was prepared by Audrey D.Kroner, Chemist, National Training andOperational Technology Center, MOTD,OWPO, USEPA, Cincinnati, Ohio 45268.

Descriptors: Analytical Techniques,Laboratory Tests, Turbidity

21-5

TRACE ORGANIC CONTAMINANTS IN WATER

I INTRODUCTION

The subject of trace organic contaminantsin water continues to receive increasingamounts of attention. Concurrently, thesources of these refractory materials arebecoming more varied and complex. Theproblems associated with these substancesare likewise increasing and satisfactorywater treatment is becoming more and moredifficult.

II SOURCES

A Man Made

1 Domestic wastes - In variousstages of sewage treatment, dis-charged into rivers and streams.

2 Many complex industrial wastes,which have been aczidentally ordeliberately spilled or dumped.

3 Carrier solvents, such as thoseused in pesticide formulations.

4 Chemicals, such as fertilizers,which are applied directly to theland and water.

Petrochemicals.

I3 Natural

1 Extracellular products of algae by

a Diffusion of metabolic inter-mediates;

b By-products of metabolism;

c Hydrolysis of capsular materials.

2 Actinomycetes - "microorganismspresent in rivers/streams by theirgrowth and decomposition cycles.

3 Bacterial decomposition of organicmaterials.

WP.ORG. 2a. 1. 74

4 Run-off of land vegetation and soil.

5 Sediments, through decompositionand subsequent exchanges.

III CONCENTRATIONS

A Waste outfalls contain from a few ppmup to 13% of organic contaminants.

B Surface waters contain organic contam-inants in concentrations ranging fromless than ppb to several ppm.

C Organic refractory substances causeserious problems in comparative traceamounts.

IV PROBLEMS

A Taste and odor are usually the firsteffects noticed.

B Increased chlorine and carbon demandinterfere with coagulation and increasetreatment costs.

C Fouling of ion exchange resins and mal-odors in food and beverage industries.

D Adverse effects on aquatic forms thatsupport higher aquatic life, off flavorsin fish flesh, and direct toxic effectson fish.

E Potentially long-term toxic effects andpossible carcinogenic effects on humans.

COLLECTION, ISOLATION AND

A Sample Collection

IDENTIFICATION TECHNIQUES

1 Grab samples from 1 to 5 litersare adequate in many cases.

22-1

Contaminants in Water

The carbon adsorption method (CAM)is the most common concentrationtechnique4

The megasampler, a scaled-upversigi of the CAM, is us e d inspec iff studies where large quan-t ities of organic materials areneeded.

B Isolation

The carbon samples are sequentiallyextracted with chloroform and ethylalcohol to desorb the organic ma-terial from the carbon.

After extraction, the excess solventis removed, and the samples arebrought to dryness to yield:

a) Carbon chloroform extract(CCE).

b) Carbon alcohol extract (CAE).

2 The CCE is separated into broadclassical groups by techniquesbased on solubility differences.

3 Further separations are made bysuch techniques as adsorption,paper, and thin layer chromato-graphy.

C Identification

1 Tentative identification is normallymade using gas chromatography,paper chromatography, thin layerchromatography, and fluorescentspectroscopy.

2 Positive identification g en erallyrequires the use of infrared spec-troscopy, mass spectroscopy,mass spectroscopy in conjunctionwith gas chromatography, and nu-clear magnetic resonance (NMR);sample sizes of a few mg are re-quired for NMR.

22-2

VI SPECIFIC COMPOUNDS DETECTEDAND IDENTIFIED USING THE CAM

A These compounds have been detectedand identified.

a -conidendrin; o-nitro-chlorobenzene;phenyl ether; ABS detergents; non-ionic detergents; 2-ethylhexyl phthalate;dieldrin; endrin; DDT; DDE; DDD;aldrin; heptachlor; heptachlor epoxide;BHC; parathion; methyl parathion;malathion; sevin; thiodan; chlordane;telodrin; technical and 7-chlordane;bis-(2-chloroethyl) ether; bis-(2-chloroisopropyl) ether; naphthalene;tetralin; styrene; acetophenone; ethylbenzene; 2 -ethyl hexane; di isobutylcarbinol; phenyl methyl carbinol, 2-methyl -5 -ethyl pyridine.

B These compounds have been detectedbut not identified specifically.

Nitriles; amines; phenols; acids;ketones; aldehydes; alcohols.

VII TREATMENT AND REMOVALPRACTICES

A Plant treatment procedures, such ascoagulation, sedimentation and filtra-tion, are generally not too effective.

B Chemical Treatment

1 Copper sulfate used to control algae.

2 Oxidizing agents, such as chlorine,chlorine dioxide, ozone, and po-tassium permanganate, are usedwith varying degrees of success.

3 Activated carbon treatment re-moves organic substances byadsorption.

C Biological Tr eatment

1 Natural degradation in unsaturatedsoils and streams.

2 Biological oxidation of organicmaterials both in streams andacclimated systems.

Trace Or

VIII SUMMARY

The problems associated with trace organiccontaminants in water are becoming moreapparent as -Jur needs and usage of waterincrease. The continued growth of thechemical industry, our increasing popula-tion, and the publie- demand for morepalatable water, emphasize even more theurgency of these problems. Developmentshave been made in the detection of theserefractory materials and in their removalfrom water supply sources. It is apparent,however, that further advances in thecollection, identification, and removal ofthese pollutants are needed to insure thepublic of high-quality water and waterresources.

REFERENCES

1 Anon. Tentative MChloroform ExtraWater. JAWWA 54

2 Ryckman, D. W. , Band Edgerley. X. Jr: , ,,5

niques for the EvaluitPollutants. Ibid. 58:97

3 Ettinger, M. B. Developratection of Trace Organtinants. Ibid 57:453. IOW

4 Robeck, G. G. , Dostal, K. A. , Cohen,J. M. , and Kreissl, J. F. Effecttiveness of Water TreatmentProcesses in Pesticide RemovalIbid. 57:181. 1965.

5 Anon. Taste and Odor ControlChemicals and Methods. Taste& Odor Cont. J. 31(1):1. 1965.

This outline was prepared by R. L. Booth,General Analysis Group, AnalyticalQuality Control Laboratory, NERC, EPA,Cincinnati, OH 45258.

odor, organic compounds, organic matter,taste, water pollution, water quality

22-3

METHODS OF RECOVERING ORGANIC MATERIALS FROM SURFACE WATERS

Methods of recovering organic materialsfrom water may be classified according to thedegree of concentration required before thedesired analyticalprocedure can be applied.

I CONCENTRATED SOLUTIONS

When the concentration is high and/or themethods are sufficiently sensitive (i.e..when only a minor degree of concentrationis involved), the following methods shouldbe considered.

A Liquid-liquid extraction usually involveswater and an immiscible organic solvent.Solvents should be investigated in theseries of increasing polarity:

Aliphatic hydrocarbonsAromatic hydrocarbonsEthersChlorinated compounds

(Cc14, CHC13, CH2C12)EstersAlcohols, amines, acids, etc.

Because of hydrolysis or decompositon,pH may be of critical importance; thispoint will be discussed at length in thediscussion of Analytical Procedures.Inorganic salt concentration may alsobe important.Remember the concept of the partitioncoefficient:

K = awhere K is the partition coefficient,

Cs is concentration in the ex-tracting solvent

Cw is the concentration in water.

Continu44 extractors must be usedwhere K alue is not favorable tothe ext g solvent. Continuous batchand counter-current extractors may beused.Separatory funnels are available in avariety of sizes. Batch continuous

CH. OTS, 37b. 74

extractors are commercially ain sizes up to one liter or so,. andbe readily assembled in largerContinuous counter-current extraare convenient for up to 10-20

B Steam distillation can be useda sample of small amounts oforganic compounds; in thisdistillate. usually must becentrated by liquid-liquid extractThe organic compound should have atleast a moderate vapor pressure at1000C and should be almost insolublein water.A variation of this method is simpleevaporation to concentrate non-volatileorganic compounds.

C Precipitants may be used: silver saltsof acids, chloroplatinates or tetraphenylboron, derivatives of amines, phenyl-hydrazine derivatives of ketones an daldehydes, etc.

D Ion exchange may be used to concentrateacids and bases. The only limit to thevolume of water to be filtered is theamount of inorganic salts in the water;these inorganic salts usually use up treeexchange capacity too rapidly to makethis method practicable.

II VERY DILUTE SOLUTIONS

To concentrate extremely dilute solutions,the carbonfilter is the most useful method.The organic matter is adsorbed from aqueoussolution and desorbed by an organic solvent.The great advantage is the large amount ofwater that can be put through a small filter;the disadvantages lie in the lack of quanti-tative adsorption and desorption.

A The Adsorption Process

1 The adsorption process involves anequilibrium in sJli.'ion.

Adsorbed Unadsorbed

23-1

,N13

of Recovering Organic Materials from Surface Waters

2 The Freundlich Isotherm is often use-'' 'Agin evaluating adsorption behavior.

- 1;w.. x

; Veit. may be chosen for convenience,1

1,,I Mier the --n-, the stronger theLion. The constant k gives

21-value at unit concentration.

a, weight of adsorbed compoundweight of carbon

k constant s x/m when c is unityequilibrium concentration oforganic compound in the liquidphase; it is measured directly.constant determined by theslope of the isotherm

o

1

3 General Rule: Non-polar compoundsare strongly adsorbed and polar com-pounds weakly adsorbed from water(e.g., mineral oil is more stronglyadsorbed than glycine).

4 Adsorption will depend on the type ofcarbon, grain size, pH of solution,polarity of adsorbate, nature of thesolvent, temperature, etc. If pow-dered carbon is not used, the columnlength and contact time must bechosen so as to permit efficient ad-sorption.

5 Desorption is simply t he reversereaction in the adsorption equili-brium and is influenced by the samefactors as adsorption. It is usuallynecessary to use an organic solventfor desorption with an extendedperiod of extraction.

B Setting up the Carbon Filter

23-2

1 In the next section,is a descriptionof the procedure used in setting upand running carbon filters in ourlaboratory. However, it should beunderstood that elaborate apparatusis not required and that excellentresults can be obtained with a pieceof glass tubing closed at either endby a rubber stopper in which is in-serted a piece of smaller glass tubing.

itt

Such apparatus is adequate fortesting small volumes (e.g., sew-age or concentrated industrialwastes), although not convenientfor handling thousands' of gallonsof water.

2 Wherever the investigation is con-cerned only with certain specificcompounds, bench tests in smallcolumns should be carried out inorder to evaluate the performanceto be expected from the filter.

III INSTALLATION OF EQUIPMENT ANDCOLLECTION OF CARBON FILTERSAMPLES

A The carbon filter consists of a piece ofpyrex glass pipe, 3" in diameter and18" in length. The ends are fitted withbrass plates and 3/ 4" galvanized nipples.A stainless steel screen is fixed in aneoprene gasket at both ends.

B Presettling,. Prefilter, and Backwash

River water will frequently clog the car-bon filter before the desired amount ofwater has been sampled. It is necessaryto remove sufficient turbidity to permitthe required amount of water to passthrough the unit. This may require apresettling tank, and a prefilter con-taining sand and gravel. For watershaving less than 100 ppm of turbidity,a presettling tank generally is not re-quired. Tap water, of course, requiresno prefiltering and may be passed di-rectly into the carbon filter.

C Presettling Tank

A standard hot water tank connectedwith the inlet at the bottom and outletat the top, with a clean-out tap at thebottoms, can serve as a presettlingtank. The outlet connects to the pre-filter ing sand and gravel. Thehot wate_ hould be flushed atfrequent int e rvaLlisMe..Rr e vent largeaccumulations of solids. Open settlingtanks can be used if care is taken to

1 ; tI

Methods of Recovering Organic Materials from Surface Waters

prevent long detention times andbiologi -cal action . With open tanks a pump willbe required to move the water throughthe filter.,

D Sand Prefilter

The sand prefilter consists of a standardpiece of steel pipe, 3" in diameter and3 feet long, threaded at both tends andèquippedwith 3"X 1" reducer couplings.Two disks of stainless steel screen arefitted to the inside diameter of the pipe.A simple way to prepare and hold thescreenin place is to cut out a diskaboutthe size of the outside diameter of thereducer and then push it into place sothat is is below the threads. The screenis held tightly against the inside of thereducer. The pipe is packedwith 6" of1/8" gravel, 24 inches of 0. 6 - 0.8 mmsand, and another 6" of 1/8" - 1/4"gravel. No free space is left in the pipe.The gravel should be packed by jarringwhile filling the pipe. This arrangement

provides a strainer rather than a filterwith a movable bed. With such an ar-rangement, backflushing can be donewithout disturbing the filter. The con-struction details are shown in Figure 1.The dimensions may be varied accordingto local conditions.

IV FILTER ARRANGEMENT

A The presettling tank. the sandprefilter,and the carbon filter should be installedat the most convenient source of rawwater. If less than 15 psi pressure isavailable, it maybe necessary to pumpthe water through the system. A sche-matic drawing of a workable system isshown in Figure 2. Exact lengths ofpipes, etc., are not given since thesewill vary with the local situatioa, Boththe sand and carbon filters should beconnected with unions at both ends foreasy removal. A typical installationis shown in Figure 2.

16-18 Mesh (pc3" X 1" Reducer CouplingStainless SteelScreen ------..4"---- 1---

7:0 1/ 8"--1/ 4." Gravele

Threaded Each End

7.'2

Standard 3" Pipe 0.6 to 0.8 mm SandN

/ 8" ' 1 /4" Gravel

16-18 MeshStainless Steel "eScreen 3" X 1" Reducer Coupling

FIGURE 1 - Details of Sand Prefilter1 " 1k.., .,

23-3

-.Ty


B The raw water passes into the bottomof the sand prefilter, around into thebottom of the carbon filter. When therate of flow through-the system fallsbelow 1/4 gpm, backwashing of the sandfilter is necessary, using a high pressuresource of water. A clean hose is con-nected to the top of the sand filter, thevalve to the carbon filter is closed andthe drain valve on the sand filter isopened. The sand is back-flushed un-til the water coming out is clear. Thelength of time between backwashingswill vary. On the Missouri River,' forexample, it has been found that onceevery 24 hours is sufficient. Afterbackwashing, connect the system asbefore and continue the sampling. Besure to disconnect hose at top of sandfilter after backwashing. A pressuregauge is inserted in the system to in-dicate when clogging is taking place inthe carbon filter. Total pressure inthe filter should not exceed 50 psi.

Union -4J?

Sand Prefilter I.-

C A water meter rocated at the end of thesystem to prevent excessive fouling ofthe meter, is used to measure thevolume of water samples. It is goodpractice to disassemble and clean the .meter thoroughly after each run. Adisk-type meter, 5/8"X 3/4", is fairlysatisfactory. A valve following themeter serves to throttle the flow, ifnecessary. A flow-regulating devicemay also be used for this purpose.The complete installation is shown inthe schematic diagram in Figure 2.

D Fine carbon washes out of the filterwhen it is first started. A few gallonsof water can be passed through the topconnection and through the carbon filterdrain, before cutting in the meter tokeep the meter free of carbon. Thehose connection and drain on the carbonfilter can be used to back-flush thecarbon should it become clogged. Thisis a last resort and should only be used,if absolutely necessary.

r_Hose Connection--o.

Pump(If Required) "II

Union

EPZRaw

Valve

Water Draip

PressureGauge

Valv

nionWater

Meter

Carbon Filter

Union

FlowRegulator(1/2GPM)

Figure 2 - Schematic Diagram of PipinglInstallation for SandPrefilter and Carbon Filted (').*5

23-4

Methbda of Recovering Organic Materials from Surface Waters

E The system outlined is not intended toremove all traces of turbidity from thewater before passing through the carbonfilter. Its purpose is to take out grossmaterials and most organisms, and topermit the required volume of waterto pass through the carbon. The valvenearest-the bottom of the sand filteris to be closed when it is desired toobtain a sample of raw water directlyfor other analyses. The drain valveis opened and, after flushing, the nec-essary raw water sample is collected.

F Pumping

1 If adequate pressure is not avail-able for sampling raw water, it isnecessary to pump through thefilter system. It is important thatthe pump used should not contami-nate the sample through oily packingor other sources. Newpumps areoftentimes grease-coated and shouldbe thoroughly cleaned before beingput into service. If a lift of morethan 12 feet is required, it will benecessary to use a jet4type pump.A model suitable for the conditionsshould be selected.

2 Before a new pump is put into serv-ice, it should be thoroughly flushedwith hot water containing a littledetergent. The pump should be oper-ated against the minimum amountof head required to get the waterthrough the filter. This may re-quire by-passing of part of the flow.

G Collection of Sample

1 With fairly concentrated samples(sewage, industrialwastes, etc.).a small filter and a few gallons ofwater may be sufficient. Generally,for river sampling, water should bepassed through a large filter at arate of 1/4 - i/2 gallon per minute,until up to 5000 gallons or more havebeen filtered. With highly turbidwaters clogging may occur earlier.Although a suitable sample can beobtained sometimes with several

hundred gallons of river water, itis desirable to filter a minimum of2000 gallons if at all possible. Aflow-regulating device set at 1/4 or1/2 gpm can beplaced in the sys-tem ahead of the carbon filter. Asuitable unit an be obtained fromDole Valve Company, 6201 OaktonStreet, Morton Grove, Illinois.These devices generally need aminimum pressure of 15 psi forproper operation.

2 Since the purpose is to get a quantita-tive measure of organics in water,it is very important to have ac-curate flow measurements. If dif-ficulties inflow occur, this shouldbe noted on a log sheet. Meterreadings should be recorded dailyon log sheets and should be desig-nated either in GALLONS or CUBICFEET. If a satisfactory meter cannot be obtained, the flow rate for aset volume (1 or 2 liters) can bedetermined at regular intervals.

H Precautions

The purpose of the carbon filter is toadsorb small amounts of impuritiesfrom the water. It is important toavoid contamination of the carbon fromother sources. Hence, the followingshould be observed.

1 New galvanized fittings are usuallycoated with oil or grease. The oilshould be removed with a wash inkerosene followed by a detergentwash before such fittings are usedfor making connection to the filter.

2 Ordinary organic pipe jointing com-pounds should not be used. Red lead(lead oxide) mixed to a paste withwater, can be used for this purpose.

3 Plastic hose is to be avoided. Ifrubber hose is used in any connec-tions it should be flushed thoroughlybefore being connected to the filter.Copper tubing is ideal for con-nections.

23-5


I After the required volume of waterhas been run through the filter, thecarbon is removed, dried, and ex-tracted. The wet carbon is spreadout in a thin layer on a metal sheetand air-dried for several days. Thetime needed for drying increaseswith the thickness of the layer. Warmair can be passed over the surfaceof the carbon to hasten the drying,but with the risk of driving offweakly held volatile substances.

V LABORATORY TREATMENT OF SAMP LESOBTAINED BY TAI CARBON ADSORPTIONTECHNIQUE: EtTRACTION PROCEDURES

The sample is collected by passing approx-imately 5000 gallons of water through a carbonadsorption unit. The unit is shipped to SEC,along with a daily record of the sampling ac-tivity and the carbon treated as describedbelow.

A Preliminary Treatment

1 Records

a Log sample in. List date re-ceived. Assign it a number.

b Send daily record sheet to thelaboratory where pertinent datais recorded on a data card (i.e.,source, location, dates sampled,date received and total flow).

2 Drying

a Remove wet carbon from thetube and dry iton copper, brassor stainless steel trays in anoven at 40°C. (n.b. Air cir-culated through drying cabinetshould be prefiltered throughcarbon to prevent adsorptionof foreign materials.) Dryingrequires about two days.

b Store carbon samples in onegallon paint cans tightly closed.

23-6

B Solvent Extraction of Carbon

It is necessary that blanks be run on allsolvents and on the carbon used for thecollection of the sample.

1 Packing the extractprs

a Filter paptr at bottom ofsoxhlet.

b Glass wool (pre-extracted)3" in depth to prevent carbonfines from patssing into the pot.Add ehloroform to wet the wool.

c Carbon is added and packed.. (n. b. Do not pack too tightly.)

2 Chloroform extraction

a Add chloroform about twocylinder volumes.

b Extract for 35 hours.

c Siphon and blow the bulk of thechloroform into the pot with air.

d Remove the flask containingchloroform solution from thesystem. Concentrate to 250 mlby distillation and filter into a300 ml Erlenmeyer flask. E-vaporate to about 20 ml or, asteam bath with a stream of airand transfer to a tared 5 dramvial. The remaining solventis evaporated at room temper-ature without an air jet and theweight obtained.

3 Alcohol extraction

a Remove residual chloroformfrom the carbon by one ofthe following methods.

1) Blow warnicd (40°C)prefiltered air upwardth.ough the carbon forabout four hours.

Methods of Recoverin

2) Leach residual chlorofoemoff the carbon with ethylalcohol (95%) and distill offlhe chloroform alcohol mix -tur e. Repeat until virtuallypure alcohol.remains.

3) Remove 'the carbon fromthe soxhlet and air dry ontrays. When dry, repackinto soxhlet as before.(Chloroform vapors mustbe disposed of.)

b Add sufficient ethanol (abouttwo cylinder volumes) and ex-tract for 24 hours.

c Concentrate the alcohol solutionas in the case of chloroform,except that the final drying canbe carried out on a steam bathwith a stream of air.

The first method listed is the moatsatisfactory for our work here, mainlybecause a minimum of supervision isrequired and chloroform vapors areexhausted out the hoods.

C Extractable Materials

1 Calculation of concentrations

On most waters, it is convenient tocompute the recovery in parts perbillion (ppb) (i. e., mg/1), usingthe following formula.

ppbgrams recovered X 106

PP gallons filtered X 3.785

2 Infrared spectra

Infrared spectra are routinely runon both the chloroform and alcoholextracts.

15

Or c Materials from Surface,Waters

D Special Applications

Other solvents may find use in per-forming 'extractions. If spe c i a 1applications are needed, a testin gprogram is necessary to establishwhat solvents may best be used.

E Adsorption and Desorption

The effectiveness of adsorption an ddesorption varies for differentmaterials and should be considered ininterpretation of the results.

This outline was prepared by R.H. Burtschell,Chemist, formerly with AWTL, NERC,Cincinnati, OH and J. J. Lichtenberg,Pesticides Identification Group Leader, AQCL,NERC, EPA, Cincinnati, OH 45268

Descriptors: Absorption, Adsorption, CarbonFilters, Filtration, Organic Matter, Separ-ation Techniques, Solvent Extractions, WaterAnalysis

23-7

PRELIMINARY SEPARATION OF EXTRACTS

I INTRODUCTION

As a general principle, it maybe stated thatthe more sensitive modern analytical tech-niques (gas and other forms of chromatog-raphy, spectrophotometry, etc., ) requirefairly pure samples to begin with. Also asa general principle, it may be said that themore widely different the purification stepsemployed, the greater the degree of purifi-cation.Ion exchange, dialysis, crystallization,electrophoresis, etc, , are useful but cannotbe discussed here. We will cover brieflyonly three techniques.

A Distillation

B Partition

C Adsorption

II THEORETICAL BACKGROUND

A Distillation

1 Raoult's Law states that the partialpressure of A(PA) in an ideal mixtureof volatile solvents is

PAmole fraction of A in liquid Xvapor pressure of pure A atthe temperature of the system.

The total vapor pressure (which, ofcourse, sets the boilingpoint) is thenPA + PB + This is the basis forstudies of distillation. The questionof non-ideal solutions is too compli-cated to discuss. Most physicalchemistry texts go into it at length.

2 Partially miscible liquids are a spe-cial case, often discussed as "steamdistillation" although co-distillationis a better term. Quoting fromGlasstone, "As long as the twolayers are present, irrespectiveof their relative amounts, the total

CH. OTS. 4613. 1, 74 C.; It

vapor pressure remains constantand the system bolls at a definitetemperature. "

Besides the familiar case of steamdistillation, non-polar organics canbe co-distilled with polar liquids,e. g. , insecticides with glycerol andpolar compounds with non-polarsolvents (i.e., mineral oil). The co-distillation can be done with super-heated solvents, under vacuum.Where trace quantities of one com-ponent are involved, there may notbe a two -phase system present andthe solvent vapor serves only as acarrier gas. Here the solute canusually be recovered easily, how-ever, so the method is worth consid-eration regardless of the mechanism.

In practical work, the three impor-tant points are the stability of thesample, its volatility, and the easeof recovery from the distillate.

3 Distillation from acid and basicsolutions may alsobe useful. Theprinciple behind this will b ediscussed in the solubility separa-tion method.

B PartitionPartition is the principle utilized inliquid-liquid extraction, paper chroma-tography and gas chromatography. Thefundamental law is the Distribution Law:"A dissolved substance distributes it-self between the layers of a two-layersystem so that at constant temperaturethe ratio of the concentrations is alsoconstant". The Law is properly appliedonly to dilute solutions, but a first ap-proximation to the distribution ratio anbe obtained from the ratio of solubilities.

CsK K

SsS w

24-1

Preliminary Separation of Extracts

B

K is a true constant but apparent dis-crepancies occur where the solute hasdifferent molecular weights in the twophases. Benzoic acid, for instance, islargely 0 COOH plus some benzoate ionin water; in benzene it is the dimer(+COOH)2. K then applies giol,iato thespecies common to both layers,4COOH, and ignores the (Amer.

Most texts on qualitative organic analy-sis contain a great deal of practicalinformation on solubilities, while theusualundergraduate physical chemistrytexts cover the essentials of the theory.

C Adsorption

There are two kinds of adsorption to beconsidered: physical (van der Waals)adsorption, and chemisorption. Thetheory of adsorption from solution is notas well understood as adsorption of gasesbut we can make a few comments.

Physicaladsorption depends on the rela-tively weak vender Waals forces due toelectrostatic attraction of dipoles, itbeing assumed that non-polar moleculesact as oscillating dipoles; i.e., the nucleiof atoms and molecules form oscillatingdipoles with the electrons. Since thebonds are weak, adsorption can be madereversible under proper conditions.

Chemisorption is thought of as actualchemicalreaction, and the resultingbonds are much more difficult to break.Losses due to "irreversible adsorption"may therefore appear; this is particu-larly true of carbon, less so of silicaand alumina.

III SOLUBILITY SEPARATION OFEXTRACTS

In the next section, we will describe a lab-oratory procedure which we have found veryuseful in handling carbon filter extra ct a . Theextracts are split into acid, basic, and neutralfractions, and the neutral fraction is furtherseparated by adsorption chromatography.

24-2

This procedure has been extremely usefulwhere the fractions are weighed and in-frared spectra made. However, wherevolatile compounds are to be looked forby gas chromatography, it is almost im-perative that a steam distillation step beincluded. Otherwise the mass of heavy,probably polymeric, material presentcauses much interference.

Really good clean-up methods are likelyto be highly specific and need to be"tailor-made" to suit the situation. Sincethis lecture is concerned only with the"preliminary" purification, we have omitteda discussion of precise analytical partitionand adsorption columns (thin-layer, ionexchange, paper and gas chromatography,gradient elution methods, etc.) Su c hmethods constitute a second and higherdegree of purification, but ordinarily re-quire that they be preceded by a "rougher"primary purification.

IV NOTE ON LOSSES OF VOLATILECOMPOUNDS

It may not be generally recognized, but itis quite 'true that heavy losses of evenmoderately volatile compounds occur ifone tries to evaporate off all of the solventwhen preparing to weigh an extract. Wehave had losses of over 50% in attemptingto carry out what we thought were verycareful evaporations of ether from 10 to20 mg amounts of phenols.

For this reason, we suggest that no ex-tract or fraction thereof be evaporateddown for weighing unless it is known to becompletely nonvolatile. This precautionbecomes more necessary as the purityof the sample increases, because non-volatile impurities act as "solvents" toreduce the amount of loss (accordingto Raoult's Law).

Preliminar Separation of Extracts

REFERENCES

1 Maron and Prutton. Principles ofPhysical Chemistry (4th Edition)The Macmillan Co. 1965.

2 Moore. Physical Chemistry (3rd Ed.)i Prentice Hall, Inc. 1965.

3 Glasstone and Lewis. Elements ofPhysical Chemistry. D. vanNostrand Co, 1960.

This outline was prepared by R H Burtschell,Chemist, formerly with Waste Identificationand Analysi Program, Advanced WasteTreatment Research Laboratory. NERC, EPA,

Cincinnati, OH 45268.

Descriptors: Absortpion, Adsorption,Distillation, Organic Matter, SeparationTechniques, Water Analysis

1 r ' 24-3

LABORATORY PROCEDURES FOR THE PRELIMINARYSEPARATION OF EXTRA CTS

I STEAM DISTILLATION OF EXTRACTS

Steam distillation need be only a verysimple procedure. Put a gram or two ofcrude or partially purified sample in asmall flask containing 50 - 100 ml ofdistilled water. Set up for ordinary dis-tillation and boil off most of the water.Set up for ordinary distillation and boil offmost of the water. If necessary, add morewater to the flask and continue distillinguntil the distillate has no tract of cloudinessor a second phase as it drips from thecondenser. Transfer the distillate to aseparatory funnel, extract with a suitablesolvent, and concentrate to a convenientvolume. As the product is relativelyvolatile, do not try to evaporate to drynessif it can be avoided. The non-volatileresidue in the distillation flask can also berecovered if necessary.

If it is desired to recover acids (e. g.,phenols in the 4-PAP colotimetric menod),make the solution acid before distilling;the distillate will then contain only neutraland acidic substances. The neutraLs canthen be separated during the extractionstep. t3asic materials remaining in thestill Fa., e.g.. aniline, can then berecovered by making basic and distillingover a second fraction.

Ii SOLliBILITY SEPARATION OF EXTRACTS

A The organic contaminants from watcrae extracted train carbon or concen-trated by liquid extraction ordinarilyform an exceedingly complex mixturefor which there is no one satisfacto;:separation procedure. If the analysisis directed towards a single component,the procedure may be designed for thatpurpose; otherwise a useful and generallyapplicable preliminary separation maybe made on the basis of relative acidities.

CH. OTS, 22c. 1. 74

B By extracting an ether solution of thesample with water, then with dilutehydrochloric acid, then sodiumbicarbonate, and finally sodiumhydroxide, a separation into watersoluble, basic_ strongly acidic, weaklyacidic, and neutral fractions may bemade. The portion insoluble in ethermay also be recovered as an additionalfract. on. This method is obviously notsuitea to very volatile substances, norto further separation of the watersoluble fraction, nor to substancesunstable in the presence of water ordilute acid or base; in addition, it mustbe remembered that substances whosepartition coefficients are not extremelyunfavorable to water may be present in

several fractions in varying amounts.

C This procedure constitutes one of thebest preliminary steps in analyzing anyunknown sample and may oftenprofitably be either preceded orfollowed by steam distillation (at variousph's) fractional crystallization, etc.Simple micro qualitative tests fornitrogen, halogens, sulfur, phosphorus,etc., are also often useful; for a moreextended discussion, see the textsmentioned in the bibliography.

D General

The sample should be substantially freeof solvent. The amount may varyconsiderably but one-half gram is a veryconvenient amount; as little as 100 mgor even less may be used, but in thiscase there may be large percentageerrors. Ethyl ether has proven to bean excellent solvent although often notall the sample will dissolve in it,benzene or chloroform may be usedalthough these solvents are more likely

cause troublesome emulsions.

25-1

Laboratory Procedures for the Preliminary Separation of Extracts

III LABORATORY DIRECTIONS

A Solubikty Separation Procedure

Solution in Ether

1 Weigh previously dried samplecontained in a tared 50 ml beaker(approximately 0.5 gm preferred).Also, weigh a 125 ml flask for theWater Solubles (WS).

2 Dissolve the sample with a smallamount of methanol (about 1 ml).Add the methanol dropwise and stirwith a rigid wire until the samplereaches a syrupy consistency.

3 Add 30 ml ether to the sample.Not all will dissolve. Collect theEther Insolubles (El) on a sinteredglass funnel with suction and A etaside for paragraph 4. Pour thecontents of the suction flask into a125 ml separatory funnel equippedwith a teflon stopcock, (Stopcockgrease will contaminate the sample. )

Ether Insolubles (EI)

4 Dissolve "EI" previously collectedon the sintered glass funnel withmethanol. Wash funnel once witha small volume of chloroform andcollect in the suction flask. Trans-fer filtrate to the original taredbeaker with methanol. Evaporateto dryness on a steam bath andrecord weight for Ether Insolubles,

Sration of the Water Solubles (WS)

5 Shake the ether solution(3) threetimes with 15 ml portions ofdistilled water. Drain the waterlayers into the weighed "WS" flaskafter each shaking. Evaporate theWater Soluble fraction to drynesson a steam bath with a jet of clean,ciry air and record weight forWater Solubles.

2b-2

'7Separation of Weak Acid Fraction (WA)

Separation of Amine Fraction (B)

8 Mark flask "HC1". Shake ethersolution three times with 15 mlportions of 5% HC1 and drainaqueous layers into HC1 flask aftereach shaking. Make this aqueoussolution strongly basic by carefullyadding NaOH pellets (about 30) or25% NaOH. The solution becomesdarker at this point. Set aside forback-extraction.

Separation of Strong Acid Fraction (SA)

7 Restore, original ether volume to30 ml, if necessary. Mark flask"NaI-1C0 3". Shake three timeswith 15 ml portions of 5% NaHCO3and drain aqueous layers intoNaHCO

3flask after each shaking.

Acidify aqueous solution in flaskby adding sufficient concentratedHC1 (dropwise because of liber-ation of CO2) until strongly acid tolitmus paper. About 4 ml issufficient. The solution becon.ascloudy at this point. Set aside forback-extraction.

8 Restore original ether volume to30 ml, if necessary. Mark flask"NaOH". Shake three times with15 ml portions of 5% NaOH anddram aqueous 1,6ers into "NaOH"flask after each shaking. Washonce with 15 ml distilled water anddrain this into "NaOH" flask.

CAUTION: Tilt separatory funnelgently a few times, instead ofshaking the first portion, aser "ilsions err occur.

If a small adiount of the waterwash remains emulsified, afterdraining, add several drops o'saturated Na2SO4 solution andshake. This may help to break

Procedures for the Preliminary Separation of Extracts

Ether olutionF

extract with H2O

WEIGHED SAMPLE

add etheIIr, filter

I

I

Residueevaporate, weigh

ETHER INSOLUBLES (EI)

Ether Layerextract wi+ti HC1

Etter Layerextract with NaHCO3

Water layermake acid

extract with ether

Water Layerevaporate, weigh

WATER SOLUBLES (WS)

I

Water LayerMake basic, extract

with ether

I I

Ether Layer Water Layerdry, evaporate Discard

and weigh

I

BASES (B)


and weigh

STRONGIACIDS SSA)

Ether Layerextract with NaOH

Ether Layerdry, evaporate

and weigh

11

NEUTRALS (N)

Water La1yermake acid

extract with etherI

I


and weigh

i

WEAK ACIDS (WA)

1",.,4' ,l

I

.25-3


the emulsion. Acidify aqueoussolution in flask with concentratedHC1 until strongly acid to litmuspaper. About 8 ml sufficient. Thesolution becomes cloudy at thispoint. Set aside for back-extraction.

Neutral Fraction (N)

9 Mark flask "Nu. Pour the etherlayer remaining in the separatoryfunnel into the flask "N". Addabout 10 gm of anhydrous sodiumsulfate, cap with aluminum foil andallow ,,:o stand overnight to dry theether solution.

B Back Extraction

Strong Acids (SA)

10 Mark flask "SA". Transfer NaHCO3extract from step 7 to separatoryfunnel and shake several times,carefully, relieving pressurecaused by evolution of CO2. Shakethe NaHCO3 extract, previouslymade acid, with a 15 ml portion ofether. Caution: Release CO2during first shaking repeatedly toavoid pressure build-up.

Drain the aqueous layer into theNaHCO3 flask. Pour the etherlayer into the "SA" flask. Returnaqueous layer to separatory funneland repeat extraction two moretimes with 15 ml portions of ether.

After all the ether layers have beencollected in the "SA" flask, addabout 10 gms anhydrous sodiumsulfate, cap with foil and allow tostand overnight to dry the ethersolution. Discard the aqueousNaHCO-3 portion which should bealmost colorless.

Weak Acids (WA)

11 Mark flask "WA". Shake the NaOHextract, previously made acid,three times with 15 ml portions of

25-4

ether and collect in flask "WA" asin step 10. Add sodium sulfate todry, cap with foil and allow tostand. Discard aqueous NaOHportion which should be almostcolorless.

Bases (B)

12 Mark flask "B". Shake the HC1extract, previously made basic,three times with 15 ml portionsof ether and collect in flask "B".Add sodium sulfate to dry, capwith foil and allow to stand.Discard aqueous HC1 portion.

C Transfer

13 Weigh four flasks and mark them''N'', "WA", "SA", and "B.%Record weights of each. Transferthe corresponding dried ethersolutions frc-n the Na2SO4 into theweighed flasks by filtering throughfilter paper.

14 Evaporate the ether from the abovefractions with the aid of clean, drycirculating air and gentle heat ona steam bath. The ether solutionsshould be removed from the airand heat before totally dry andallowed to dry spontaneously toprevent loss of volatile components.Record the weight of the driedfractions.

D Chromatographic Separation

15 Weigh a 10 ml beaker and recordweight. Dissolve the Neutrals inabout 5 ml anhydrous ether,transfer most of the Neutrals tothe beaker and place the remainingfew drops in a vial for infraredanalysis. Wash the (N) flask withether and pc.ur washings into vialalso. Dry the Neutrals containedin the beaker by spontaneousevaporation and record weight.

I

1

I

I

I

1


16 Weigh a set of 125 ml Erlenmeyerflasks and mark them Aliphatics,Aromatics and Oxys. Recordweights of each.

17 Fill the chromatographic columnwith 4 1/2" of silica gel and tapdown to about 4". Place theAliphatics flask under the column.Adu about 20 ml of iso-octane to

wet the column.

Aliphatics

18 Before the iso-octane reaches thelevel of the silica gel, add theNeutrals which have been pre-viously ack arbed onto a smallamount of silica gel by stirring inthe beaker with a rigid wire.Elute with 85 ml iso-octane.Rinse the beaker with a smallportion and add to the column first.Allow this volume to enter silicagel before adding the remainder ofthe iso-octane. Collect the eluentin the Aliphatics flask, evaporateto dryness with the aid of gentleheat and circulating air.

1

1

Elute' withIso-octane

ALIPHATICS

Aromatics

19 Place the Aromatics flask underthe column. Elute column with85 ml of Benzene-rinsing thebeaker with a small portion andadding immediately as the previouseluent reaches the level of thesilica gel. Allow this volume toenter silica gel before addingremainder of benzene. Evaporatewith the aid of gentle heat andcirculating air. The Aromaticsfraction should be removed fromthe air and heat before totally dryand allowed to dry spontaneouslyto prevent loss of volatilecomponents.

Oxys

20 Place the Oxys flask under thecolumn. Elute with 85 ml of a 1:1methanol chloroform mixture -rinsing the beaker with a smallportion and proceed as before.Evaporate to dryness with the aidof gentle heat and circulating air.

CHROMATOGRAPHIC SEPARATION OF NEUTRA LS

NEUTRA LS

Adsorb on Silica gel Column

Elute withBenzene

3

Elute wttnChloroform/ Methanol

(1 1)

AROMATICS OXYGENATED COMPOUNDS

25-5

Laborato Procedures for the Pre liraina Se tion of Extracts

21 Record weights of the driedAliphatics, Aromatics and Oxys.

NOTE: (All chromatographicsolvents should be re-distilled before use)

REFERENCES

1 Shriner, Fuson and Curtin. TheSystematic Identification of OrganicCompounds. 4th ed. John Wiley& Sons. New York. 1956.

2 Cheronis and Entrikin. SemimicroQualitative Organic Analysis.Thomas Y. Crowell Co. New York.1947.

3 Cheronis. Micro and SernimicroMethods. Vol. VI of the seriesTechniques of Organic Chemistry.Arnold Weissberger, Ed. :Inter-science. New York. 1954.

25-6

4 Schneider. Qualitative OrganicMicroanalysis. John Wiley &Sons. New York. 1946.

These books deal principally with chemicalmethods of identification; Shriner, Fusonand Curtin and Cheronis and Entrikin arerecommended for those new to the field.For instrumental analysis and quantitativework, see standard works in those fields.

This outline was prepared by R. H.Burtschell, Chemist, formerly with AWTL,NERC, Cincinnati, OH and J. J. Lichtenberg,Pesticides Identification Group Leader,AQCL, NERC, EPA, Cincinnati, OH 45268.

Descriptors : Laboratory Tests, OrganicMatter, Separation Techniques, Sol-gentExtractions, Water Analysis

THIN-LAYER CHROMATOGRAPHY

I INTRODUCTION

A Historical

1 Adsorption chromatography was dis-covered by Tswett in 1903. Tswettfound that plant pigments (e. g.. chloro-phylls) can be separated by filtering asolution of them in petroleum etherthrough a column of calcium carbonate.He noticed that yellow and green zoneswere formed in the column.

2 Adsorption chromatography was intro-duced in 1931 when the method wasused for the preparative chemistry ofthe polyene pigments, for examplethe separation of carotene into itscomponents.

3 Partition chromatography for hydro-philic compounds was introduced in1941.

4 Thin-layer chromatography was intro-duced by Stahl in 1956. He describedan ingenious and practicable device forpreparing layers of silica gel about 250microns thick with a plaster of parisbinder.

5 Thin-layer chromatography has provenuseful for the quantitative estimation ofpesticides and for the "clean-up" ofsample aliquots prior to the estimationof their components by gas chroma-tography.

B Theory

1 Thin-layer chromatography is ananalytical technique which has provenuseful for the separation of componentsin a given system and tb evaluate thecomponents in that system.

2 All of the techniques of chromatographyare based upon the same simple prin-ciple. They involve a moving systemof some type (liquid or gas) which is in

CH. MET. cr.3b. 1.74

equilibrium with a stationary phase.When the stationary phase is a solidand the forces acting between it andthe mixture are adsorptive in nature,the technique is called adsorptionchromatography.

3 When the stationary phase is a simpleliquid or a liquid held on some type ofsupport, the chromatography is con-sidered to be partition chromatography.

4 Diagnostic or qualitative thin-layerchromatography is used to determinethe number of components in a system.

5 Preparative thin-layer chromatographyis used to separate a mixture into itscomponents in such a way that reason-able amounts can be isolated andstudied.

6 Quantitative thin-layer chromatographyis used to measure the amount of thecomponents present. This is done byestimating the density of the spots ascompared to standard spots.

II DEFINITION OF TERMS

A Adsorbent. The finely divided powderwhich makes up the stationary phase inadsorption chromatography, or holds thestationary liquid in partition chromatography.

B Layer. The thin layer of adsorbent bound

or unbound, deposited on a glass place.

C Spotting. The application of the substanceto the thin layer adsorbent.

D Development. The passing of a liquidthrough the layer to effect a separation.

E Developer. The liquid used for thedevelopment.

F Eluent. The complete removal of a sub-stance from the adsorbent layer.

26-1

Thin-Layer Chromatography

G

H

Visualization. The rendering visible ofthe results of a developed chromatogram.

Ratio facter "Rf ". The distance traveledby a given substance divided by the distancetraveled by the solvent front. Bothmeasured as the origin.

M ADSORBENTS

A Si licic Acid or Silica Gel

Chemically identical, but the gel is pre-pared in such a manner as to enhance itsadsorptive power.

B Aluminum Oxide or Alumina

C Diatomaceous Earth

D Powdered Cellulose

E Essentially any substance which hasdesirable adsorptive and chemical charac-teristics can be successfully used in thin-layer chromatography.

IV PREPARATION OF PLATES

A Methods

1 Spreading or spraying of slurries

2 Dipping in a slurry

B Equipment

1 Kirchner type

2 Stahl type

3 Microchromatoplates

C Layer thickness

1 For qualitative analysis thin layers(250 microns) are used.

2 Formarative thin layer thickerlayers may be used.

26-2

D Moisture Content /r1 Trace quantities of water greatly

reduce the activity of various adsorbents.

2 Normally plates are activated bydrying at 110°C for one hour and thenstored in a desiccator.

TECHNIQUE

A Spotting the Plate

1 The activity of the layer is affected byhumidity and temperature.

2 The sample is applied so that thedirection of the development is oppositeto the direction in which the layer hasbeen applied.

3 Use of spotting template

4 Use of micro-pipette and micro-litersyringe

5 Size of sample spots

6 Use of standards

B Solvent System

1 Investigation of single and multi-component systems

2 Carbon tetrachloride, hexane, andn-heptane as solvents for separationof pesticides

C Development of the Plate

1 Development tank

2 The plate is placed in a glass tank con-taining the solvent. When the solventfront reaches the upper reference line(10 cm) the plate is removechnd thesolvent is allowed to evaporate.

1 0,1I lit)

Thin-La er Chromato, ra s h

VI VISUALIZATION

A Rhodamme B

1 Spray reagents - Rhodamine B, spiritsoluble, 0.1 mg per ml in ethanol

2 Method of spraying

3 Advantages of Rhodamine B

a The pesticide is not destroyed.

b The exact position of the pesticidecan be determined.

c The use of a selective solventpermits the pesticide to be elutedfrom the adsorbent while theRhodamine B is retained.

B Impregnation of Plates with AgNO3

1 Increased sensitivity for quantitativeestimation

2 Elimination of spraying

C Exposure to UV Light )4

1Pesticides appear as quenched areas in

the fluorescent background.

2 Recovery studies using Rhodamine Bspray are comparable to those usingsilver nitrate spray. The followingpesticides may be seen at the 10 1.1g

level under UV light after RhodammeB spray:

endrindieldrinDDTaldrinlindanechlordaneparathionmethyl parathion

heptachlorheptachlor epoxidemethoxychlortoxapheneDDEovextedion

D Suggested systems for adsorptionchromatography are given in Table 1.

E Recovery

1 Elution of pesticides from the thinlayer with a mixture of ethyl andpetroleum ethers. .

2 Recovery zones as shown in Figure 1.

3 Silica gel collection assembly -Figure 2.

4.1.=11


26-4

10.0 cm

8.0 cm

6.0 cm

6.9

3.8 cm

I.0 cm

SPOTTING 0LINE if

SECTIO

SECTI

3.3Z

SECTION

SECTION

It

SECTK/N

SPOT I 2 3 4 5 6 7 8 9 10 i 1 IZ

1.0 cm

ZONE FOR:

SPOT I SPOT 2 SPOT 3

4-- SOLVENT LINE

7.8 ALDRIN-------- 75 DDE

7.3 DDT

1L

6.5 HEPTACHLOR

-5.5 y-CHLORDANE---------5.2 DOD

-4.2 LINDANE

- -- - - --- 3.4 HEPTACHLOR E X PDX IDE

2.4ENDRIN, DIELDRIN

Figure 1 Diagram of Designation of Sections in the Cleanup and Separationof CCE-Aromatics on Silica Gel Layers.


VACUUM

EYE DROPPER

I

'HOSE --GLASS WOOL

SILICA GEL COLLECTION ASSEMBLY

FIGURE 2

I "I #,..

26-5


TABLE 1.

SUGGESTED SYSTEMS FOR ADSORPTION CHROMATOGRAPHY

Compounds Adsorbent Developer Visualization

1. Long-chain (overeight carbons)aliphatic ketones

silica gel G

2. Aliphatic and aluminum oxidearomatic aldehydesand ketones

3. Aromatic aldehydesand ketones

4. Vanillin and othersubstituted benz -aldehydes

5. 2, 4-Dinity<i-phenylhydrazonesof aldehydes

6. Miscellaneousalkaloids

7. Alkaloids andbarbiturates intoxicology

8. Strong-baseamines

9. Sugar acetatesand inositolacetates

26-8

silica gel G

silica gel G

silica gel G

silica gel G

aluminum oxide

silica gel G

unbound alum-inum oxide

silica gel G

a.b.c.

a.b.

c.d.e.

benzene -EtOEt mixtures phosphomolybdictoluene-Et0Et mixtures acid (2, 4-DNPH) .pet. ether -Et0Et mixturesbenzenebenzene -EtOH(98:2) (95:5) (90:10)CHCLIEt0Efpet. ether- benzene (1:1)

hexane -EtOAc (4:1) (3:2)

a. pet. ether -EtOAc (2:1)b. hexane -EtOAc (5:2)c. CHCLI-Et0Ac (98:2)d. decalrn-CH 2C12-Me0H

(5:4:1)

a. benzene-pet. ether(3:1) for aliphatic

b. benzene -EtOAc (95:5)for aromatic

a. CHCLa -acetone-dietillamine (5:4:1)

b. CHC13-diethylamine

c. (c9y:c11)ohexane-CHCL, -diethylamine (5:4:n

d. cyclohexane-diethylamine (9:1)

e. benzene- EtOAc-diethylamine (7:2:1)CHC1cycloKexane-CHC13(3:7) plus 0.05%diethylamine

a. Me0Hb. CHC13-Et0Et (85:15)

f.g.

a. acetone-heptane (1:1)b..- CHC13/NH,

(set. at 2240)-EtOR(96 %) (30:1)

benzene with 2 -10% Me0H

I I'lf kr,

2, 4- DNPH

(U.V. on phosphors)

(U.V. on phosphors)

colored

( Dra gendorff ' sReagent)

(Dragendorff'sReagent)

I vapor -U. V.(Dra gen dorff' sReagent)

H2O

I

I

1

4:

Thin-La er Chromato ra

TABLE 1 (ccritinued)



10. Aldose 2, 4-DNPH's

11. Cardiac glycosides

12. Dicarboxylic acids

13. p- Hydroxybenzoicacid esters

14. Sulfonamides

15. Food dyes

16. Misc, essentialoils

17. Coumarins

18. Alkali maflas-1+Na KMgt

+

19. 24 ferrocene

a. aluminumoxide G

b. silica gel G

silica gel G

silica gel G

silica gel G

silica gel G

silica gel G

silica gel G

silica gel G

purified silicagel G

silica gel G

20. Iviisc. lipids silica gel G

21. Fatty acid methylesters

silica gel G

22. Glycerides silica gel G

a. toluene -Et0A c (1:1)

b. toluene -Et0A c(3:1) (1:1)

a. CHC13-MeOH (9:1)b. CHC13- acetone ;65:35)

a. benzene -MeOH -HOAc(45:8:4)

b. benzene-dioxane-HOAc (90:25:4)

pentane-HOAc (88:12)

CHCLi-Et0H- heptane(1:1:1)

a. CHC1.1 -aceticanhyorride (75:2)

b. benzenec. methyl ethyl ketone -

HOAc-MeOH (40:5:5)

benzene-CH23 (1:1)

a. pet. ether-EtOAc(2:1)b. hexane -Et0A c (5:2)

Et0H-HOAc 1100:1)

a. benzeneb. benzene -EtOH (30:1)

(15:1)c. propylene glycol-

Me0H (1:1)d. propylene glycol-

chlorobenzene-Me0H (1:1:1)

a.b.c.

EtOEtisopropyl etherisopropyl ether -HOAc(98.5:1.5)

hexane -EtOEt mixtures(up to 30% EtOEt)

a. Skellysolve F-EtOEt (70:30)

a. colored com-pounds

b. NaOH

H2SO4 and heat

bromphenol blueacidified with citricacid

(U. V. on phosphors)

p-climethylamino-benzaldehyde,acidified (U. V. onphosphors)

SbC13

in CHC13

Sb C13 in CHC13

(U.V. on phosphors)

acid violet (1.5%aqueous soln.)


(H2SO4 and heat)

a. I,b. H2SO4 and heat

H2 SO4 and heat

26-7


TABLE 1 (continued)



23. Long-chainaliphatic alcohols silica gel G

24. Phenols

25. Phenols andphenolic acids

26. Phenols

27. 3, 5-Dinitroben-zoates of alcoholsand phenols

28. Steroids

a. starch-bolind talkieacid Ideeel-guhr (1:1)

starch-boundsilicic acidwith phosphors

plaster -of-Parisbound silica gel

silica gel G

silica gel G

29. 19-Nor-steroids silica gel G30. Blood cholesterol silica gel G

and cholesterolesters

31. Triterpenoids

26-8

silica gel G

b. (10:90) for mcnoglyceridesc. (70:30' for diglyceridesd. (90:10) for triglyceridese. (60:40) (35:65) (85:15)

a. pet. ether -EtOEt -HOAc(90:10:1)

b. pet. ether -EtOEt(20:80) (10:90) (70:30)

a. xyleneb. xylene-CHC13 (3:1)

(1:1) (1:3)c. CHC13

a. Skellysolve B- EtOEt(3:7)

b. (Sk1e3) 11y solve B- Et0A c.

c. Skellysolve B-acetone(3:1)

a. hexane -Et0A c(4:1) (3:2)

b. benzene -EtOEt (4:1)c. benzene

H2SO4 and heat




a. benzene-pet. ether (1:1) colored compoundsb. hexane -EtOAc (85:15)

(75:25)c, toluene-EtOAc (90:10)

a. benzeneb. benzene -Et0A c

(9:1 (2:1)c. cyclohexane -EtOAc

(9:1) (19:1)d. 1, 2 - dichloroethane

Et0Ac-cyclohexane mixtures

a.b.c.d.

P.

b.c.d,e.

benzenebenzene -EtOAc (9:1)1, 2 -dichloroethaneCHC13

isopropyl ether-acetone (5:2) (19:1)isopropyl ethercyclohexanebenzeneULz

M2

1 `*1 t-4 li

SbC13

in CHC13

SbC13 in CHC13

SbC13

in CHC13

(H2

SO4 and heat)

I

I

I

I

I

I

I

I

I

I

I

I

I

I

1

I


TABLE 1 (continued)

StGGESTED SYSTEMS FOR ADSORPTION CHROMATOGRAPHY


32.

33.

34.

Carotenes and fat -soluble vitamins,A, D, E, K

Thiophere deriva-tives

Plasticizers(phthalates, phos-phates and otheresters)

unbound alumi-num oxide

a. silica gel Gb. aluminum

oxide G

silica gel Gwith phosphors

a.b.c.

a.b.c.a,

b.c.

Me0HCC14xylene

benzene-CHC13 (9:1)Me0Hpet. ether

isooctane-Et0Ac(90:10)benzene -EtOAc (95:5)butyl ether - hexane

H2SO4 and heat

(1.7.V. on phosphors)

a. (U.V. on phosphors)b. 12

(80:20)

REFERENCES

1 Smith, D. and Eichelberger, J. Thin-Layer Chromatography of CarbonAdsorption Extracts Prior to GasChromatographic Analysis for Pesti-cides. J. Water Poll. Control Fed.37, 77, (1965).

2 Damaska, W. H. F. D. A. CincinnatiDistrict, Private Communication.October, 1964.

3 Randerath, Kurt. Thin-Layer Chroma-tography, Academic Press. NewYork. 1963,

4 Bobbitt, J. M. Thin-Layer Chromato-graphy. Theinhold, New York. 1963.

This outline was prepared by P. F. Hallbach,Chemist, National Training Center, WaterPrograms Operations, EPA, Cincinnati,OH 45268.

Descriptors: Adsorbents, Chromatography,Organic Matter, Separation Terhmques,Water Analysis

26-9

INTRODUCTION TO GAS-LIQUID CHROMATOGRAPHY

Part I

I INTRODUCTION

A Definition

Gas-liquid chromatography is an analyticalmethod for the separation and identificationof a mixture of volatile (usually organic)components in a sample. As with anychromatographic technique the columnconsists of two phases, the immobile orstationary phase (a liquid on an inert solidsupport), and the mobile phase (an inertgas). Thq column functions to separatethe sample components because they havevarying vapor pressures and affinities forthe stationary phase. In many ways thecolumn behavior resembles that offractional distillation. The partition whichoccurs between the mobile and immobilephases will thus cause the components toproceed through the column at varyingrates. The separation is recorded andquantitated by the detector system.

B Advantages

1 GLC can be used to separate come mdsof similar boiling points which canzioteasily be separated by distillation.(See Table 1)

2 GLC can be extremely sensitive; forexample, using the electron capturedetector it is possible to "see""icogram (10-12) quantities.

Table 1. SEPARATIONS BY GLC

Compounds Reference

3 - Methylcyclohexene(B.P. 1040C) and4-Methylcyclohexan(B.P. 1030.C)

Cyclohexane (B.P.80.813 C) and Benzene(B.P. 80.2,3C)

CH. MET. cr. 5b. 1.74

A erograpb ResearchNotes (Spring 1964)

Chromosorb News-letter (FF-104)

C Disadvantages

1 Due to the extreme sensitivitypossible it is often necessary toapply extensive cleanup techniques.

2 The many variables of the techniquerequire a skilled analyst.

II COMPONENTS OF A GASCHROMATOGRAPH (See Figure 1)

A Gas Supply

The mobile phase (carrier gas) transportsthe sample components through thecolumn into the detector. The type. ofgas used varies with the detector.(SeeTable 2)

B Injector

Liquid samples are manually introducedinto the heated injector block through arubber septum by means of a syringe.Automatic liquid injectors as well asinjection systems for solid and gaseoussamples are commercially available.

C Column

The vaporized sample enters the columnwhich can be glass or metal and of varyinglength (1' - 20') and diameter (1/8" - 1/4").The column is packed with the tationary(immobile) phase and contained within aconstant temperature oven,

1 Solid support

The solid support should have a largesurface area yet be inert so that activesites will not cause adsoiption ofsample components. Diatomaceousearths, teflon and glass beads havebeen used. (See Table 3) .

27-1

lutroduction to Gas - Liquid Chromatography

AUTOMATIC STRIPCHART RECORDE

DETECTORr - --21 --- ".1 OVENi

DE0 -ITECT

SAMPLEINJECTION

PORT PRESSUREI REGULATOR

GLASSWOOL

z2=-saV

1

X:..1

06,It01.-itft6a0.311

LW

s--z.1tmZ0LI

L- - -- - .. ..........1

a) SILICONERUSSERSTOPPERS

T 1/16" DIAMETERS. S. TUBING

lFigure 1. COMPONENTS OF A GAS CHROMATOG2APH*

Table 2. CARRIER GASES

Detector Carrier gas

Thermal conductivity

Mier ocoulometric

Flame ionization

Electron capture

Helium (Purified,Grade A)

Helium (Purified,Grade A)

Hydrogen (Purified)

Nitrogen or a mixtureof 95% argon and 5%methane (Purified)

T le 3. SOLID SUPPORTS

urface area (m /gm)

4. 8

3u

Chromosorb P(Diatomaceous Eart

Chromosorb W(Diatomaceous Eart

Ohromosorb G(Diatomaceous Eart

Chromosorb T(Teflon)

27-2

1.2

0. 5

7.0 - 8.0

2 Stationary liquid

The separation and partition occurringin the column is directly affected by thechoice of stationary liquid. For ex-ample, in the separation of benzene(B. P. 80. 1 °C) and cyclohexane (B. P.80. 8°C), the choice of a non-polar phasesuch as hexadecane results in benzenepreceding cyclohexane off the coluMn.However, if a more polar phase such asbenzylbiphenyl is chosen cyclohexaneprecedes benzene. Table 4 shows sometypical stationary liquids and their uses.(NOTE: One requirement for any liquidis that it have a high boiling point so thatit wild not boil off the column)

D Detector

ThP detector or brain of the gas chroma-tograph senses and measures the quantityof sample component coming off the column.The detector should be maintained at atemperature higher than the coiumn so thatcondensation does not occur in the detectorblock. Several types of detectors are inuse today.

*ReproduCec1Tvith permission) from Chemistry.(37:11, p. 13. November 1964).

1 .-1 nI ..j

1

1

I

1

I

I

I

I

I

I

I

I

I

I

I

I

Introduction to Gas-Liquid Chromatography

Table 4. STATIONARY LIQUIDS

Stationary liquid Used to separate

Silicone oils QF -1, Dow Corning 200,and Dow 11, OV-1, OV-3, OV-17,

Chlorinated hydrocarbonspesticides

OV-Z10, OV-225

Silicone oil SE-30 Homologous series of n-alkanes

Ben zyl-Cyanide -Silver Nitrate Homologous series of olefins

Polyethylene Glycol Amines

Cyano Silicone Steroids

1 Thermal conductivity

This detector consists of a Wheatstonebridge two arms of which are thermalconductivity cells each containing asmall heated element. When only carriergas is flowing through both the samplecell and reference cell, the resistanceof the heated element is constant it bothcells. The bridge remains balanced andbaseline is recorded. However, whencarrier gas plus sample component enterthe sample cell, the thermal conductivityin that cell changes thus also producinga change in the resistance of the heatedelement. The bridge becomes unbalancedand a peak is recorded. The main dis-advantage of the TC cell in water pollutionwork is its lack of sensitivity.

2 Ionization detectors

a Flame

This detector consists of a flamesituated between a cathode and anode.As carrier gas alone burns, someelectrons and negative ions are pro-duced which are collected at theanode and recorded as baseline.When carrier gas plus sample com-ponent are burned, more electronsand negative ions are produced whichresult in a peak on the recorder.The detector is capable of "seeing"nanogram quantities of organic com-pounds; however, the detector issensitive to all org ic compounds.This lack of specifi ty produces dis-advantages in the a alysis of water

1'Yr)tjI

extracts which contain a variety ofnaturally occurring organics.

b Electron capture (See Figure 2)

This detector consists of a radiationsource (e.g., tritium) - capable ofproducing slow electrons in a carriergas such as nitrogen. The electronscollected at the anode 'are recordedas baseline. When sample com-ponents which have an electronaffinity (e.g., chlorinated hydro-carbons) enter the detector, electronsare "captured." The subsequent de-crease in current is recorded as apeak. The detector has the advantagethat it is extremely sensitive (pico-gram range) and is somewhat selective.

c Thermionic

A recent adaptation of the flameionization detector shows promisefor the specific analysis of organicphosphorus compounds. An alkalisalt is incorporated into the designof a conventional flame ionizationdetector so that the salt heated bythe flame produces an ion current.When compounds containing phos-phorus emerge from the column,they give 600X the response with thisdetector as with the conventionalflame.

3 Microcoulo-.11,)1.r

Although less sensitive (by approximatelya factor of 10) than electron capture.

27-3

Introduction to Gas - Liquid Chromato raphy

RADIOAC TI VETRITIUM .

FOIL

COLUMNGAS- -

FLOW

GAS EXHAUST

TO

ELECTROMETER

ANODE

90*

COLUMN

+ N2 ";"" a SLOW

*SLOW X.

ELECTROMETER

t.OSS --V. REDUCES CURRENT

Figure 2. ELECTRON CAPTURE DETECTOR (Wilkens Instrument Company)

this detector is finding wide use inpesticide analysis. This highly specificdetector consists of titration cells forthe measurement of chloride-containingand sulfur-containing compounds. Thesample component emerging from thecolumn is combusted to produce HC1or SO,,, respectively. HCl. is continu-ouslylitrated by silver ions present inthe cell, the amount of current requiredto regenerate these silver ions isrecorded as a peak. The system forsulfur containing compounds is analo-gous except that SO2 produced is con-tinuously titrated by L., which issubsequently regenerated.

Arther mlcrocoulometric detector hasrecently been applied to the specificdetermination of nitrogen. It is basedon the reduction of nitrogen-containingcomtiounds to N11,4 which is then titratedby H in the titration cell.

4 Coulson Electrolytic Conductivity

This detector was primarily developedfor the detection of organic halides,organic nitrogen compounds, andorganic sulfur compounds.

21-4

The unit consists of pyrolyzer with aseparately heated inlet block, watercirculating and purification system,detector cell and dc conductivitybridge. The sample is oxidized orreduced and reaction products formelectrolytes when dissolved in thedeionized water. Changes in con-ductivity between two platinumelectrodes are measured by the dcbridge. (Figure 2 a)

E Recorder

The recorder system registers theresponse of the detector to samplecomponents. In the case of ionizationdetectors, it i. often necessary to employan electrometer in order to amplify thesmall current changes.

Expensive integration and digital read-outequipment is also available to facilitatemeasurement of peak areas.

fib!,


0 CIA,C6L

as S-L 'Epps9 SEAallrca

GAS lcoy TaC,04

COP4,1,JC T ,nrrtCELL , sk;fP SI,LiqW0

Ptj

rtiSsACI

(Aug r2TUBE

S4ACELCK

.YAfTTEE

CAPILLOIT

li

as TEs EESi...3701a as 314 P'..P

Iit

"1

IC V EZC

Figure 2 a. FLOW DIAGRAM OF ELECTROLYTIC CONDUCTIVITY DETECTOR

(Coulson Instruments Company)

ac4:3 SAMPLE 5 of Sta- nda-r-, Pesttc.de

tare ( :no! eacS Pestactue /5,11

COI, ChaN Length -6' x6 mmStationary Phase - ", DC 2C0

OC Analtrom ABS (9u / Ir.,0 Mesh)Mobile Phase - Z80 mt 'man N2Temperature - 2'0°C

DETECTOR, Electron Cap ure

e.e.

O

6.4

0 5 7 11

TIME IN MINUTES)13 15

Figure 3. GAS CHROMATOGRAM OF PESTICIDE MIXTURE

1 C ?.) 27-5


III QUALITATIVE ANALYSIS

A Retention Time

The retention time of a sample componentis defined as the time it takes for thatcomponent to travel through the column.There are a number of variables whichaffect the retention time of a compound.

1 Physical parameters of column''> operation

a Column length

b Column temperature

c Carrier gas flow rate

2 The nature and amount.of stationaryliquid itself

For a given set of column conditions, aspecific compound will have a specificretention time (See Figure 3 and Table 5).Various column and detector combinationscan be used to confirm identification.

B Retention Volume

Retention volume is defined as the totalolume of gas required to move a com-ponent through the column.

RETENTIONVOLUME (R `T)

II

.27-6

=RETENTION FLOWTIME (RT) X RATE

C Relative Retention Times and Volumes

It is possible to interpret data more easilyby reporting retention data relative to aparticular compound (e. g., aldrin as inTable 5).

IV QUANTITATIVE ANALYSIS

A Measurement of Peak Area

v

The quantity of sample component presentis directly proportional to the area underits peak. (NOTE: This assumption canonly be made if it has been previouslydetermined that a linear response isobtained in the range under study. ) Thefollowing are a few of the ways in whichthis area can be measured.

1 Planimeter

2 Triangulation

'.". 3 Peak height X half-width (see dieldrmpeak in Figure 3)

AREA = peak height X peak half-width

4 Disc integrator

B Measurement f Peak Height,,.

With the electro capture detector it maybe possible to use peak height for quanti-ctive measurements where the followingonditions are met.

Table 5. RETENTION DATA FOR FIGURE 3

Pesticide Retention Time (RT) Relative RetentionTime .,

Heptachlor 3.3 minutes - 0.79

A ldrin 4.2 1.00

Heptachlor Epoxide 5.3 1.26

Dieldrin 7.7 1.84

,inmeommellIslonlml411111111111111111011111111111111111111111111M

et

Introduction to Gas - Liquid Chromate: a

1 A steady basline is obtained.

2 Retention times can be reproducedfrom one injection to the next.

V SUMMARY

The basic components of a gas chromatographhave been described. Elementary aspects ofquantitative and qualitative analysis arepresentea.

BOOKS

1 Dal Nogare, S. and Juvet, R. S. , Jr,Gas-Liquid Chromatography. NewYork: Inters cience. 1962.

2 Littiewood, A.B. Gas Chromatography.New York: Academic Press. 1962.

lZ

1 c '

NEWSLETTERS

1 Aerograph Gas Chromatography Newsletter.Wilkens Instrument and Research, Inc.,P.O. Box 313 Walnut Creek, California.

2 F & M Gas Chromatography Newsleter.F & M Scientific Corporation. StarrRoad and Route 41, Avondale, Pa.

3 Gas-Chrom Newsletter. Applied ScienceLaboratory; Inc. State College, Pa.16801.

This outline was prepared by B. A. Punghorst,Chemist, .formerly with National TrainingCenter, MDS, WPC) EPA,Cincinnati, OH 45268.

Descriptors: Adsorbents, Chromatography,Gas Chromatography, Organic Matter,Separation Techniques, Water Analysis

1

ti

27-7

INFRARED INSTRUMENTATION

I THEORY

A The Effect of Infrared Radiation on theMolecule

1 All molecules are made up of atomsconnected by chemical bonds whichact very much like springs.

2 Atoms or atomic groups are ir. continu-'ous motion with respect to each other.Each molecule has specific vibrationalfrequencies for these motions.

3 A molecule which has the same vibra-tional frequency as the infrared radia-tion striking it will absorb that energy.

4 If the radiant frequency differs fromthe characteristic frequencies of themolecule, the radiation passes throughthe molecule undiminished.

5 The characteristic frequencies forparticular molecules are determinedby the masses of their atoms, their

atial geometry and the strength oftheir connecting bonds.

B Measuring Infrared Radiation

1 The infrared spectrum gives basicinformation about the molecularstructure of a compound.

2' Organic chemicals are largely madeup of different combinations of atomicbrlding blocks called functional groups:-OH, -NH2, -CH3, -CO, -CN,

3 These subgroups have characteristicabsorption bands in certain parts ofthe infrared spectrum.

4 Spectra correlation charts provide akey to the location of characteristicabsorption bands of the most commonfunctional groups.

CH. MET. 26. 1. 74

II THE INSTRUMENT

A Source

1 A length of ceramic tubing heated toabout 1200°C

2 It produces a continuous spectrum ofradiant energy, most of which fallswithin 2.5 to 15 micron region.

B MOnochromator

1 It disperses the radiation into itscomponent wavelengths.

2 Selects the particular wavelengthinterval transmitted to the detector.

3 Maintains, by varying slit widths,approximately constant energy throughthe slits to the detector.

C Wavelength

1 A mirrors rotational position deter-mines the particular wavelength whichemerges from the slit.

2 Linkage with the recorder drum shaft'insures that the wavelength scale onthe drum corresponds exactly to thewavelength selected.

D Resolution

1 A NaCI prism provides the best com-bination of dispersion and transmissionin the operating.range.

2 The sealed and desiccated monochroma-tor protects the prism from moistureand dust.

E Thermocouple

1 A high sensitivity thermocouple con-sists of a blackened gold leaf targetwelded to two pins.

28-1.

Infrared Instrumentation

2 Temperature differences at the junctionsof the leaf and pins set up a thermo-electric potential which is utilized.

IF Optical Attenuator

1 By subtracting energy from thereference beam, the attenuator makesthe amount of energy in the referencebeam precisely the same as that inthe sample heam.

2 This permits the pen to record directlyin terms of sample transmittance.

G Instrument Performance

1 Scanning time

Twelve minutes for total scan, auto-matically programmed for constantrate per spectral slit with for con-stant accuracy over the wavelengthrange.

2 Accuracy .

a Abscissa + O. 03p., ordinate + 1%T.

b Less than 3% scattered light at14.7p. and less than 1% from 2.5p.to 14p..

$

III MICRO INFRARED TECHNIQUE.

A Solution Technique

1 The technique most reliable forquantitative measurements in infraredregion is the solution technique.

2 One drawback to this technique is thescarcity of suitable solvents withsufficient areas of transparency.

28-2

3 Carbon disulfide CS 2is the most

valuable of these because of itstransparency in the "finger print"region of the infrared from 1, 350 to650 cm-1 (7.4p. to 15). Carbon tetra-chloride is a useful solvent from4,000 to 1,650 cm-1 (2.5p. to 6.00.

4 The sample is dissolved in 0. 3 ml ofsolvent. The solution is then trans-ferred to a 5 mm energy-path cavitycell by a hypodermic syringe. Amatching cavity cell is used for sol-vent compensation. i

5 With this technique, sensitivities downto 5-10 lig of strongly absorbing co-n-pound:: are achieved without resortingto scale expansion.

B The Potassium Bromide KBr PelletTechnique

1 Since KBr is transparent throughoutthe commonly used portion of theinfrared region, the entire infraredregion is available as contrasted to theuse of solvents.

2 A 6X beam-condenser is used toincrease the energy transmittedthrough the pellet.

$ By this procedure a reliable sensi-tivity of about 5p.g can easily beattained without stale expansion.

REFERENCES

1 Blinn, R. C. Infrared technique usefulin residue chemistry, AOAC. Vol.48, 1009-1017. 1965.

2 InfraCord, Perkin-Elmer spectrophoto-meter model 137-B Brochure, 1965.

,r

This outline was prepax ed by J. W. Mandia,Chemist, formerly'with NationalTraining Center, MDS; wpo, EPA,Cincinnati, OH 45268.

Descriptors: Chemical Analysts, InfraredRadiation, Instrumentation, Water Analysis

1 w 1%.$ ,,c

I

I

I

I

I

I

1

I

I

1

1

INFRARED

Procedure For Use Of Infracord

1. The "on-off" toggle switch should be in the "on" position, and the instrumentwarmed up for about 15 minutes.

2. \ The "reset-stop-scan" knob should be in the "reset" (extreme counter-clockwise)Position; never'try to manually move the chart drum unless this knob is in the"reset" position.

3. Please do not touch the "slit" or "100%" knobs, or those on the side of, or inthe rear of, the instrument.

4. The pen holder assembly should be drawn back so that the pen is not touchingthe chart drum.

5. With the "reset-stop-scan" switch in the "reset" position, gently remove the. chart drum from the instrument.

6. Place a piece of recorder paper on the drum in such a way that the guidelinesabove and below the hole in the upper left corner of the paper are aligned withthe line inscribed in the drum itself. Replace the drum in the instrument.

7. Gently move the pen holder assembly forward so that the tip of the pen touchesthe paper.

i

8. Lay the two sodium chloride salt plates provided on a tissue. Always hold thesodium chloride salt plates with a tweezer or the finger tips; in the latter case,hold the plates by the ends, or narrow sides. Never hold the plates by the widesides since these are the optical faces.

9. Dip the end of a capillary tube into the liquid sample to a depth of 1-2 mm.The liquid will rise in the tube via capillary action.

10. Holding the capillary tube containing the samplE in a vertical position, touchit lightly several times to the center of a salt plate; a small spot of samplewill be formed.

11. Place the second salt plate on top of the first plate and press down lightly,the spot should spread out soas to cover almost the entire area between thetwo plates.

12. Remove the spring clip holder from the instrument and place the two-plate"sandwich" in it.

13. Place the holder in the sample beam (front) bracket on the instrument.

14. Move the "reset-stop-scan" knob to the scan position; the rest of the processis automatic and the drum will stop revolving when the spectrum of the samplehas been completed (about 13 minutes).

1 Li r-....I ki

CH. MET. as. lab. la. 12.71 29-1

infrared

15. Draw the pen holder assembly back from the chart paper.

le. Move the "reset-stop-scan" knob to the "reset" position and slowly rotate thedrum in a counter clockwise direction until it stops.

17. Remove the drum from the instrument, and then, the chart paper from the drum.

18. Place a new piece of chart paper on the drum and place the drum back in theinstrument.

19. Remove the spring clip holder from the instrument.

20. Remove the salt plat "sandwich" from the holder and separate the two plates.

21. Using a tissue dipped in the beaker of solvent, gently, but thoroughly, wipethe sample from both of the --Ut plates; allow the plates to air dry.

This outline has been prepared by CharlesR. Feldmann, Chemist.. National TrainingCenter, MDS, WPO, EPA,.Cincinnati,OH 45268.

TOTAL CARBON ANA LYSIS

I INTRODUCTION

A History of Carbon Analyses

In the wake of a rapid population growth,and the increasing heavy use of ournatural waterways, the nation, and indeedthe world, is presented with the acuteproblem of increased poLlutional loads onstreams, rivers and other receivingbodies. This had resulted in a growingawareness of the need to prevent thepollution of streams, rivers, lakes andeven the oceans. Along with this aware-ness has developed a desire for a morerapid and precise method of detecting andmeasuring pollution due to organicmaterials.

B The Methods

In the past, two general approaches havebeen used in evaluating the degree oforganic water pollution.

1 The determination of the amount ofoxygen or other oxidants required toreact with organic impurities.

2 The determination of the amount oftotal carbon present in these impurities.

C Oxygen Demand Analyses

The first approach is represented byconventional laboratory tests for deter-mining Chemical Oxygen Demand (COD)and Biochemical Oxygen Demand (BOD).One of the principal disadvantages of thesetests is that they are limited primarilyto historical significance, that is, theytell what a treatment plant had been doing,since they require anywhere from twohours to five days to complete. Since up

CH. MET. 24c. 1.78

to now no faster method has beenavailable, traditional BOD and CODdeterminations have become acceptedstandards of measure in water pollutioncontrol work even though they areessentially ineffective for processcontrol purposes.

Until the introductiotrof the CaitonaceousAnalyzer, all methods taking the secondapproach, the total \carbon method ofevaluating water quality, also provedtoo slow.

II THE ANA LYSIS OF CARBON

A Pollution Indicator

Now the carbonaceous analyzer providesa means to determine the total carboncontent of a dilute water sample inapproximately two minutes. With propersample preparation to remove inorganiccarbonates, the instrument determinesthe total organic carbon content in thesample.

B Relationship of Carbcn Analysis to BODand COD

I c, '~..../

This quantity varies with the structurefrom 27 percent for oxalic acid througha0 percent for glucose to 75 percent for.aethane. The ratio of COD to mg carbonalso varies widely from 0.67 for oxalicacid through 2.67 for glucose to 5.33 formethane. Representative secondarysewIgl effluents have given a ratio ofCOD to carbon content of between 2.5 and3.5 with the general average being 3.0.

Thz BOD, rOD and carbon contents ofthese and some other representativecompounds are summarizes: in the follownig table.

30-1

Total Carbon Analysis

Stearic Acid - C18

H3602

Glucose - C6H1206

Oxalic Acid - C2H204

Benzoic Acid - C7H602

Phenol - C6

H60

Potassium Acid PhthalateICHC8 H404

Salicylic Acid C7H603

Secondary Effluent, Clarified

5-DayBOD-mg/ mg

COD - Carbon

.786 2.91 '/6

.73 1.07 40

.14 .18 27

1.38 1.97 69

05 to 2.1 de- 2.36 77

pending uponconcentration

.95 1 15 47

1.25 1 60 61

13* 75* 21*23* 67* 12*

4* 36* 7*

*In units of mg/1

III THE INFRA -RED TYPE CARBONANALYZER

A Principle of Operation

Basically the infra-red carbonaceous analyzermade by Beckman, consists of threesections - a sampling and oxidizing system.a Beckman Model 315 Infrared Analyzerand a strip-chart recorder.

.s

Infra-rec Carbonaceous Analyzer Soho._ .tic

V

condor',/,'beck

infra:Dianalyzer

A micro sample (20-40 pl) of the water tobe analyzed is injected into a catalytic com-bustion tube which is enclosed by an electricfurnace tLermostated at 950°C. The wateris vaporized and the carbonaceous materialis oxidized to carbon dioxide (CO2) and steamin a carrier stream of pure oxygen or C09-free air. The oxygen flow carries the steamand CO2 out of the furnace where *he steam iscondensed and the condensate removed. TheCO2. oxygen cr air, and remaining watervapor enter an infrared analyzer sensitizedto provide a measure of CO2. The outputof the infrared analyzer is recorded on astrip chart. after which, the curve producedcan be evaluated by comparing peak heightwith a calibration curve based upon standardsolutions. Results are obtained directly inmilligrams of carbon per liter.

B Application

Results(1)

show that the method is applicablefor most, if not all, water-soluble organiccompounds including those that containsulfur, nitrogen, and volatiles.

Nonvolatile organic substances car. bedifferentiated from volatiles, such ascarbon dioxide or light hydrocarbons by


determination of carbon both before andafter the sample solution has been blown .with a carbon-free gas.

C Sample Preparation

The Carbonaceous Analyzer is oftenreferred to as a total carbon analyzerbecause it provides a measure of all thecarbonaceous material in a sample, bothorganic and inorganic. However, if ameasure of organic carbon alone is de-sired, the inorganic carbon content ofthe sample can be removed during samplepreparation.

1 Removal of inorganic carbon

The simplest procedure for removinginorganic carbon from the sample isone of acidifying and blowing. A fewdrops of HC1 per 100 ml of samplewill normally reduce pH to 2 or less,releasing all the inorganic carbon asCO2' Five minutes of blowing witha gas free of carbon sweeps out the CO2formed by the inorganic carbon. Onlythe organic carbon remains in thesample and may be analyzed withoutthe inorganic interference.

2 Volatile -arbonaceous material

Volatile carbonaceous material thatmay be lost by blcwtng is accountedfo: by using a dual channel carbonanalyzer. Beckman's new analyzeras the previr sly detailed nigh

temperature (950°C) furnace plus alow temperature (1500C) one. Usingquartz chins wetted with phosimoricacid, tne low temperature ct:annelsenses only the CO2 ;creed by theacid) in the original sample. Theremaining organics and water areretained in the condenser connectedto this low temperature furnace.None of the organics are oxidized bythe 1500C furnace

By injecting a sample into fie lowtemperature furnace, a peak repre-senting the inorganic carbon is ob-Lined on the strip chart. Injectinga nonacidified sample into the high

temperature furnace yields a peakrepresenting the total carbon. Thedifference between the values detr--mined for the two peaks is the totalorganic carbon.

3 Dilute samples

If the sample is dilute (less than 100mg/liter carbon) and is a true solution(no suspended particles) no furtherpreparation is required.

4 Samples containing solids

If the sample contains solids and/orfibers which are to be included in thedetermination, these must be reducedin size so that they will be able to passthrough the needle which has an openingof 170 microns (needles having largeropenings may be obtained if necessary).In most cases, mixing the sample in aWaring Blender will reduce the particlesize sufficiently for sampling

IV PROCEDURE FOR ANALYSIS

A Interferences

Water vapor, resulting from vaporization ofthe sample, canses a slight interfere.'ice inthe method. Most of the water is trappedout by the air cooled condenser positionedimmediately after the combustion furnace.However, a portion of the waver vaporpasses through tie system into the infrareddetector appears on the s'.rip chari ascarbon. The water 'lank also appears onthe stea,siaro calibration curve, and istherefor3 remove-: fi rn the fir ,1 calcu-lation. L-1 tests of solutiol.s containing -thefollowing anions: NO3, Cl , SO-2, PO4 ,

no interference was encountered with con-centrations up to one percent.

B Precision and Accuracy

The recovery of carbon from standardsolution: is 98.5 - 100. 0 per ''ent. Theminimum detectable concentration usingthe prescribed op-rating instructions is 1

mg/1 carbon. Generally, the data arereproducible to 4- 1 mg/ 1 with a stanaarddeviation of 0.7 mg/ 1 at the 100 mg/ 1 level.


V THE FLAME IONIZATION TYPECARBON ANALYZER

A.n example of a flame ionizationcarbonaceous analyzer is the oneproduced by Doh.rmann.

To determine TOTAL ORGANIC CARBONa 30 ul acidified water sample isinjected into a sample boat containing acobalt oxide oxidizer at room temper-ature. The boat is then advanced to the90° C vaporization zone where H2O,CO, (from dissolved CO2, carbonatesand2bicarbonates) and organic carbonmaterials which are volatile ac 90° Care swept into the bypass column. Herevolatile organic carbon (VOC) is trappedon a Porapak Q column at 60° C whilethe HBO and CO

2are swept through the

.switching valve and vented to atmosphere.

After sample vaporization, the valve isautomatically switched to the pyrolyzepDsition and the boat is .hen advanced tothe pyolysis zone. Residual organi-2-arkion (ROC) materials left in the boatreact with the Co

304 at 850° C to

produce CO2is

At the same time the by-pass column s backflushed at 120°C thussweeping the VOC material through thepyrolysis zone. Both the VOC and theCO2 (from the ROC) are swept by helium

r--I 000

til

IL NEillIZATION

:CTOR

11.0

2

30-4

Ile

It

IMIT1111

CIL$lP

SIITC11111

'II': YALU

TIPORIEAT/11 STEP

R21

CO2DEDT

into the hydrogen enriched nickelcatalyst reduction zone where all carbonis converted to methane at 350° C.

The reduction product is swept throughthe switching valve, the water detentioncolumn and into thr, flame ionizationdetector which responds linearly tomethane. The detector output isintegrated and displayed in milligramsper liter (mg/1) or parts per million(Rim) on the digital meter.

To determine tot2.1 carbon, simply s4,tthe function switch to TOTAL CARBON.and cycle an unacidified sample throughthe vaporize and pyrolyze steps. Theswitching ralve remains in the pyrolyzeposition, di ecting ALL cz rbonaceousmatter to the detector.

TIC TRAPPED

UMWMASS COLIN 11 *C

11 IRO

EDUrie TRILTSle-4""EINE lOsE

1311*C1 t111Ci11

120

CO2

VIC

t=j3-1CO14TAPIR

1,1:

orci

31sT

C /7 /4SAM( ISAI

11L ET I HIP

TIC VILAPLE (RUNIC CAMRIC REVISAL 011ANiC CA1100

Li


VI THE AUTOCLAVE TYPE OFCARBON ANALYZER

Oceanography International makes a TOCapparatus in which the sample is placedin an ampule that contains phosphoric acidand potassuum persulfate. The ampule isflame sealed and autoclaved. The tip ofthe ampule is then broken and the CO2removed by a gas stream that carries theCO2 from the sample to an infrared de-tector. The autoclave digestion approxi-mates a COD in which the COD's H SOis replaced with H PO and K Cr a 4

is replaced by K2S302 84. This insthInent..

does not have auriculty with salts coatinga catalyst, uut has a compr.ntrively hightime requirement.

VII LOW ORGANIC CARBON LEVELS

The organic carbon in solution can bemeasured down to levels as low as 50 ppbusing an instrument manufactured byBarnstead. Samples up to 100 ml involume are introouced into the instrument.The sample is introduced into the systemwith a syringe or pipet. Inorganic carbonin the form of dissolved CO2 is firststripped out of the sample by a stream ofair (acidification is used to bring the pHto 4.2). The air carries the inorganiccarbon (CO2) to a measuring chambercontaining r8 megoln-cm, zero-organicwater. The CO2 dissolves in the purewater and the resultant change in specificresistance is measured and stored inmemory. See the figure on page 17-6.

Following the determination of inorganiccarbon, the ultra-violet lamp within theirradiation chamber is energized. Alldissolved organic compunds - both volatileand non-volatile - are completely oxidized.The resulting CO2 is carried by the airstream to the measuring chamber whereit also dissolves in the water. The unitagain measures the resistivity of the waterwithin the measuring chamber to determinethe total carboe of the sample. The systemnow automatically subtracts the storedinorganic carbon value from the total carbonvalue and displays the difference, which

is the level of organic carbon in thesample.

After the result has been displayed, theunit automatically begins a clean-upcycle in which, all CO., and other contami-nants in the system are removed by anion-exchange cartridge. Once all waterwithin the system has returned to an 18megokun-cm, zero-organic condition, thesystem is ready for the next sample.

Dohrmann has a ultra-violet lamp systemfor low organic carbon levels that connectsalong side of their regular analyzer. TheCO2 from the ultra-violet lamp componentpasses to the regular analyzer and itsflame ionization detector.

Ultra-violet systems will give low resultsif the organic carbon is not in solution.

VIII APPLICATIONS

Several of the many research and industrialapplications of the Carbonaceous Analyzerare listed below:

A Determine the efficiency of various waste-water renovation processes both in-thelaboratory and in the field.

B Compare a plant's waste outlet with itswater inlet to determine the degree ofcontamination contributed.

C Monitoring a waste stream to check forproduct loss.

D Follow the rate of utilization of organicnutrients by micro-organisms.

E To detect organic impurities in inorganiccompounds.

1 9 1

tt

30-5


Valve Iansseonge ValveCarthago

Ultra-violet low organic carbon Barnstead apparatus.

LED Display

A summary of some TOC laboratory instrument:-

Instrument Manufacturer Mode of Oxidation

Beckman 950°C furnace andcatalyst of Co oxides

Oceanography International H3PO4-K25208 plus

autoclave digestion

Astro, sold by Curtin Matheson 850°C combustionchamber

Dohrmann-Envirotech 850° furnace plusreduction to methane

Dohrmann-Envirotech

Barnstead

(Low Organic Carbon Levels)

Ultra-violet lamp.H3PO4

Ultra-violet lamp,H3PO4

Note that there may be other manufacturers of TOC equipment.f 1 '1

30-8

Detector

Infra -red

Infra -red

Infra-red

Flame ionization

Returned to unit withflame ionization

Conductivity


DC ADVANTAGES OF CARBON ANLYZER

A Speed

The Carbonaceous Analyzer's mostimportant advantage is its speed ofanalysis. One analysis can be performedin 2 - 3 minutes for a channel on theBeckman instrument or double that on theDohrman. This speed of analysis bringsabout economy of operation. This isprobably more than the number of COD orBOD teats that can even be started, muchless completed, in the same period of time.

B Total Carbon

Another advantage is that the measure ofcarbon is a total one. The oxidizingsystem of the analyzer brings about com-plete oxidation of any form of carbon. Nocompound has been found to which themethod is inapplicable.

X CONC LUSIONS

The Carbonaceous Analyzer provides arapid and precise measurement of organiccarbon in both liquid and air samples. Itshould be found useful for many researchand industrial applications, a few of whichhave been mentioned.

Because of its rapidity it may be found moreuseful than the more time-consuming BODand COD measurements for monitoringindustrial waste streams or waste treatmentproCesses.

REFERENCES

1 Van Hall, C. E., Safranko, John andStenger, V. A. Anal. Chem. 35,315-9. 1963.

2 Van Hall, C. E., and Stenger, V. A.Draft of Final Report - Phase I - Con-tract PH 86-63-94, Analytical ResearchToward Application of the Dow TotalCarbon Determination Apparatus to theMeasurement of Water Pollution.

3 Van Hall, C. E., Stenger, V. A.Beckman Reprint - R6215. Taken fromPaper Presented at the Symposium onWater Renovation, Sponsored by theDivision of Water and Waste Chemistry.ACS in Cincinnati. Jan. 14-16, 1963.

4 DobIls, R. A. , R. H. Wise and R. B. Dean,Anal. Chem. 39, 1255-58. 1967.

5 Various manufacturers' literature.

This outline was prepared by Robert T.Williams, Chief, and revised by Charles J.Moench, Jr., Waste Identification andAnalysis Section, MERL. USEPA, Cincinnati,Ohio 45268,

Descriptors: Biochemical Oxygen Demand,Carbon, Chemical Analysis, Chemical OxygenDemand, Organic Matter, Organic Wastes,Water Analysis, Instrumentation, Nutrients

30-7

i SPECIAL INSTRUMENTAL METHODS MASS SPECTROMETRY AND NMR

I INTRODUCTION

Mass spectrometry and nuclear magneticresonance spectrometry (NMR) are nowwidely used in organic analytical chemistry,but have. only begun to find application to theanalysis of chemical pollutants in the aquaticenvironment. Now that the cost, and in somecases the complexity of these instrumentshave decreased, they will undoubtedly becomequite important in water pollution analysisduring the next few years. This is especiallytrue of mass spectrometry, and the majorityof this lecture time will be devoted to exarsiplesof applications of this instrumental method towater pollution problems encountered at theSoutheast Water Laboratory. Examples ofNMR applications will also be given, as wellas a brief discussion of the cost and manpowerrequirements, sample requirements, opera-ting principles, and informational output ofboth methods. Inorganic mass spectrometry,a very useful method for elemental analysisof water samples, will not be discussed sincethe operating techniques and sample require-ments are quite different from those fororganic mass spectrometry. ,..

II MASS SPECTROMETRY

A Economic factors

1 Mass spectrometers range in pricefrom $20, 000 to $200, 000 or more.Ordinary general pin-pose instrument-ation begins at between $30, 000 and$35, 000.

2 Mass spectrometers are complicatedinstruments and generally are "down"from 100, to 50 °' of the time, dependingupon the make and complexity of thespectrometer. Maintenance re9uiresmoney for parts and service.

3 Time is an important economic factor.To operate, maintain, and interpretspectra from a mass spectrometer ofm2dium complexity with a reasonablework load, at least one trained personis required full time.

B Sample requirements

I Solid samples can be handled by9directprobe accessories. From 10' to 10-7

CH, MET. ms. 1. 1.74

grams of sample are required, depend-ing upon sample volatility and instrumentsensitivity. Much information can beobtained from a sample of low purity,say 90%, and some useful informationcan be obtained from simple mixtures.Viscous liquids are also generallyhandled in the direct probe.

2 The liquids sample inlet on a massspectrometer is generally used forvolatile liquids. Sample size is be-tween 0. 1 and 1 mg, and, here again,much information can be obtained fromimpure samples or in certain cases,samples in solution.

3 Gaseous samples are handled in specialgas sampling flasks and inlets.

4 Gas chromatographic inlet system

The interfacing of a gas chromato-graph to a mass spectrometer providesa very useful sample introduction mode.Using this technique, any cor,,poundthat will chromatograph is eluted fromthe GC column, passes through a specialsample enricher, and goes into the massspectrometer proper. The enricher,known as a separator, removes the GCcarrier gas preferentially, thus increas-ing the concentration of the eluted samplecompound while lowering the pressure toa level suitable for the mass spectro-meter. The application of this GCintroduction mode to water pollutionanalysis is obvious. An organic ex-tract of a water sample is chrornato-gr aphed under previously establishedconditions, and a rapid scan (1 to 3seconds) is made of each eluting com-ponent of the extract to produce thecorresponding mass spectrum. Samplesize required here is usually on theorder of 0.1µg of each component,although useful spectra can sometimesbe obtained with nanogram quantities.

C Operating principles

Regardless of which sample introductionmode is used, the sample must be in thegaseous state for the protess of ionization.The sample molecules are bombarded bya beam of electrons of variable energy inthe ion source (Figure 1) to produce positive

31-1

S ecial Instrumental Methods - Mass S ectrometry and NMR

ElectrostaticSector

Analyzer

Tube

Magnet

Slits

Filament ----

producing -11-1 IonElectron i% SourceBeam

Molecules

ElectronMultiplier

111..1111111..11th

MS ChartPaper

lo tisector

Amplifier

Visicorder withfast Galvonometers

Figure 1. Block diagram of a double-focusing mass spectrometer.

ions. These ions are swept through aseries of slits and down the analyzertube by a strong electrostatic field. Thesource and tube must be maintained at alow pressure (<10-5 torr) to minimizecollisions of the ions with each other orwith air molecules. The ions pass througha magnetic field where they are deflectedto different degrees based on their massto charge ratios. By sweeping the mag-netic field strength, ions of successivelyincreasing mass are sequentially broughtinto focus at the collector. In a double-focusing mass spectrometer, resolutionis improved by an electrostatic sectorwhich renders the ion beam monoenergeticby velocity focusing before it arrives at themagnetic focusing sector. The electronmultiplier amplifies the signal receivedat the collector and the resulting signal isfurther amplified to drive a set of galvon-°meters which make a trace on a photo-graphic chart paper.

31-2

D Information output

1 All fragment ions of a molecule giverise to peaks of certain intensitieswith positions on the abscissa of thespectrum corresponding to ion masses(m/e values). This information con-stitutes the mass spectrum, and com-parison with the spectra of knowncompounds give definitive compoundidentification. Figure 2 is a plot of atypical mass spectrum, that of bis-(2-ethylhexyl) phthalate, compared withthat of a compound extracted from theTennessee River. They ere practicallyidentical.

2 The molecular weight of a compoundcan often be obtained from its massspectrum. The fragment ion of greatestmass in most cases corresponds to themolecular we4ght of the compound(Figure 2, M =390), having been createdby loss of only one electron duringelectron bombardment.

r1 r7t)

TENNESSEE RIVER SAMPLE.

NO-ON

C -ONsO

1116-(2-11THYLHEXYL) PHTHALATE

AN INDUSTRIAL POLLUTANT

C

9

160

167

-ON

b«9

erc"-S3-:csce117

P

M279 3901

290 290 210 360 3j10 340 340 3i0-1400

0

om I v --04N 7-ON

AI

279 390

190 2L-ffi 260 0 3)20 390 440

Figure 2

O

tp

0.z

19

cial Instrumental Methods - Mass Spectrometry and NMR

3 The masses of important fragment ionsindicate the empirical formulae of theions and hence give clues as to struc-tures of parts of the molecule (Figure 2).

4 Isotope peaks are helpful in determiningif a molecule contains chlorine, bromine,sulfur, mercury, and other atoms whichhave more than one highly abundantnaturally occuring isotope.

5 High resolution mass spectrometry iscapable of providing almost the exactmass (to four decimal places) of theparent ion and of the important frag-ment ions from the molecule. Thisdegree of accuracy of mass determin-ation eliminates all but one or twopossible empirical formulas for theseions, thus providing very importantclues to the molecular structure.

III EXAMPLES OF APPLICATIONS OF MASSSPECTROMETRY TO WATER POLLUTIONPROBLEMS

(Corresponding spectra will be nrojectedand.discussed.

A The identification of S, S, S-tributylphosphorotrithioate in Charleston Harbor,S. C., depended to a large extent onmolecule weight determination by massspectrometry.

B The identification of phenyl mercuricchloride, believed to be the cause of afish kill in Boone Lake, Tennessee,involved the observation of mercuryand chlorine isotope peaks in the massspectrum.

C Diazinon was separated from a TombigbeeRiver water extract by thin layer chrom-atography, renioved from the adsorbent,and analyzed by mass spectrometry.

D The analysis of petroleum rocket fuelresidues extracted from water illustratethe applicability of mass spectrometry tooil pollution analysis.

E Bis- 2- (ethylhexyl) phthalate was identifiedas a pollutant in the Tennessee River withthe aid of mass spectrometry (Figure 2).

F An actinomyeetes metabolite, responsiblefor earthy taste ati.d odor problems in anOhio municipal water supply, WELE ident:fied as 2-methylisoborneol. The mass

31-4

1

spectral fractionation pattern of theunknown compound was important in theelucidation of its structure.

G Nonylphenol has been extracted from atreated woolen mill effluent, collectedupon elution from a gas chromatographiccolumn, and identified by mass spectro-metry using the direct probe accessory.

H Several ter,..enes and terpene derivativesin the treated effluent of a kraft papermill have been identifiedby combinationgas chromatography-mass spectrometry(GC-MS). The complexity of the gaschromatogram illustrates both the greatproblems encountered in waste character-ization and the beauty of the GC-MStechnique; whereby, many of the individualcomponents can be identified.

I Three resin acids and several fatty acidshave been identified in the effluent and inthe foam downstream from a paper mill.The methyl esters of the acids were pre-pared by reaction with drazomethane, andthe volatile esters were analyzed by thecombined GC-MS technique.

IV NUCLEAR MAGNETIC RESONANCESPECTROMETRY (NMR)

A Economic factors

1 NMR instrumentation ranges in pricefrom about $20, 000 for a "desk-top"low resolution (80 MHz) spectrometerto $85, 000 for a 100 MHz instrumentwith common accessories.

2 NMR equipment is not as mechanicallycomplex as mass spectrometry instru-mentation; down time is generally only10% to 20% and a full time-Cperator isnot necessary for a reasonable workload. However, an experienced operatoris necessary for optimum output.

B Sample requirements

1 Conventional NMR samples must be inthe liquid phase, either neat or insolution. Solvents must be deuteratedor aprotic (e. g. , CC14, CS2) since anyprotons present absorb energy to givespectral interferences. Milligramquantities of sample (in solution) arerequired except that with particularlysensitive instrumentation 0.1 mg cansometimes give good spectra, and with

r-- ---} --i-n I 2

bis-(2-ITHYLMIXYL) MITHALATI

I

; i 9

a43

IQ

Figure 3_,

IJJ

Special Instrumental Methods Mass Skectrometrzand NMR

time-averaging computerization toenhance the signal to noise ratio, lessthan 10 micrograms is sometimessufficient. A purity of 90% is usuallyadequate to give useable spectra, andsimple mixtures can often yield usefulinformation.

2 A lack of sensitivity (a factor of 103 to106 times as much sample is requiredas for mass spectrometry) and thelonger time required for analysis arethe principal disadvantages ofThe method is extremely valuable;however, in detailed structural eluci-dation, providing information as to thetypes and environments of hydrogenatoms which cannot be obtained by anyother instrumentation.

C Information output

Hydrogen nuclei absorb energy in theradiofrequency range at certain frequenciesdepending on their environment. Thus,signals at certain frequency areas of thespectrum are characteristic of methyl,methylene, aldehydic, aromatic, etc. ,protons. (Figure 3). The NMR iscapable of integrating these signal areasrelative to each other; therefore, allowingcalculation of the number of each type ofproton in the molecule. In the case ofsimple mixtures, this'feature often allowsthe calculation of percentage compositionof the components.

In addition to absorbtion of radiofrequencyenergy, nuclei with magnetic moments,such as hydrogen, phosphorus, and carbon-13, interact with each other when in thesame proximity to cause "splitting" of theproton signal peaks into a larger numberof smaller peaks (Figure 3). This isknown as "spin-spin coupling" and is avery informative element of the NMRmethod. Splitting rules allow the calcul-ation of the number of peaks resulting fromthe coupling of one proton with other adja-cent protons. Thus, the degree of splittingof a particular hydrogen signal is indicativeof the protonic environment of the hydrogenatoms, and this allows important deductionsto be made regarding the structure of themolecule.

200

31-6

D Examples of applications of NMR towater pollution problems

1 The NMR spectrum of a phthalate esterisolated fromthe Tennessee Riverprovided the prime evidence for thedetailed structure of the molecule,especially of the 2-ethylhexyl moities.It was later shown that this spectrummatched that of bis-(2-ethylhexyl)phthalate (Figure 3).

2 The NMR spectrum of the actinomycetesmetabolite extracted from an Ohio mun-icipal water supply showed the compoundto have four tertiary methyl groups.

3 The frequency of the methylene absorp-tion signals of the NMR spectrum of atoxic compound isolated from CharlestonHarbor, S. C. , indicated the compoundto be the phosphate derivative of a phos-phite ester present in an industrialeffluent discharging into the harbor.

REFERENCES

1 Mc Lafferty, F. W. Interpretation ofMass Spectra, W. A. Benjamin, Inc.,New York, 1966.

2 Silverstein, R. M. , and Bassler, G. C.Spectrometric Identification of OrganicCompounds, John Wiley and Sons, Inc. ,New York, 1963.

3 Willard, H. H., Merritt, L. L. , Jr. , andDean, J. A. Instrumental Methods ofAnalysis, Chapters 6 and 16, FourthEdition, D. Van Nostrand Co. , Inc. ,Princeton, N. J. , 1965.

4 Teasley, J. I. Identification of a Cholin-esterase - Inhibiting Compound from anIndustrial Effluent, Environ. Sci. &Technol. 1(5):411, 1967.

5 Medsker, L. L. , Jenkins, D. , and Thomas,J. F. Odorous Compounds in NaturalWaters, Environ. Sci. & Technol 3(5):476, 1969.

This outline was prepared by Arthur W.Garrison, Research Chemist, SoutheastWater Laboratory, WQO.

Descriptors: Chemical Analysis, Instrumen-tation, Mass Spectrometry, Nuclear MagneticResonance, Spectrophotometry, Water Analysis

Documents

Introduction tc Instrumental Analysis of Water Pollutants. Training ... · sample mixture with an immiscible solvent in which the constituent is much more soluble. C Filtration -