DETERMINATION OF BLANK

The calibration procedure described in the next section is also used for the blank determination except that no iodate is added. If analytical reagent quality chemicals are used there should be no blue colour with starch. If a slight colouration results a blank correction may be ascertained by titrating with thiosulphate until the solution is colourless. If this blank correction exceeds 0.1 ml, the reagents are suspect and should be prepared afresh. The potassium iodide or manganous reagent is generally the cause of the trouble. If no blue colour is formed on adding starch, check that a blue colour does result when 0.1 ml or less of 0.01 N iodate is added.

This guards against the presence of reductants. The blank testing should be under-taken when each new batch of reagents is prepared.

H. CALIBRATION

1. APPROXIMATELY 0.5 N THIOSULPHATE SOLUTION (Prepare only if a microburette titration is desired.)

Dissolve 145 g of good grade sodium thiosulphate, Na 2S203 .5H20 and 0.1 g

26 A PRACTICAL HANDBOOK OF SEAWATER ANALYSIS

of sodium carbonate Na2CO3 in 1 liter of water. Add 1 drop of carbon bisulphide, CS2 , per liter as a preservative. The thiosulphate solution should be prepared many liters at a time and is stable for many months in a dark well-stoppered bottle kept below 25 C. The factor (see below) will slowly decrease.

1.0 ml of 0.5 N thiosulphate = 0.25 mg-at

2. APPROXIMATELY 0.01 N THIOSULPHATE SOLUTION

Prepare as above using only 2.9 g of sodium thiosulphate per liter.

Note: Dilute thiosulphate solution in a burette and the tubing connecting an automatic burette to a stock of the reagent will deteriorate quite rapidly and the burette and tubing should be flushed several times with new solution before titrations are recommenced after any shutdown period exceeding a few hours.

3. EXACTLY 0.100 N IODATE SOLUTION

(Prepare only if a microburette titration is desired.)

Dry a little analytical reagent quality potassium iodate, KI0 3 , at 105 C for 1 hr.

Cool and weigh out exactly 3.567 g. Dissolve the salt in 200-300 ml of water, warming slightly. Cool, transfer to a 1000-ml measuring flask and make to the mark with distilled water. This solution is stable indefinitely.

4. EXACTLY 0.0100 N IODATE SOLUTION

Prepare as above using exactly 0.3567 g of KI0 3 .

5. DETERMINATION OF THE FACTOR f

Fill a 300-ml BOD bottle with sea water, or with distilled water, and add 1.0 ml of concentrated sulphuric acid and 1.0 ml of alkaline iodide solution and mix thoroughly. Finally add 1.0 ml of manganous sulphate solution and mix again.

Withdraw approximately 50-ml aliquots into the titration flasks. Use one or two flasks for blank determinations if necessary (see G above), and to the other add 5.00 ml of either 0.100 N or 0.0100 N iodate from a clean 5-ml pipette which should be calibrated as a check on the volume. Allow the iodine liberation to proceed for at least 2 min but not for more than 5 min, during which time the solution should be at a temperature below 25 C and out of direct sunlight. Titrate the iodine with the appropriate thiosulphate solution. If y is the titration in millileters then

1.00

f for the 0.5 N thiosulphate or

5.00

f ^— for the 0.01 ^N thiosulphate.

The mean value of f should be found from at least three and preferably five replicates.

1.4. DETERMINATION OF CARBONATE, BICARBONATE, AND FREE CARBON DIOXIDE FROM PH AND ALKALINITY MEASUREMENTS

DEFINITIONS

TOTAL ALKALINITY

Total alkalinity is the number of milliequivalents of hydrogen ion that are neutralized by 1 kg of sea water when a large excess of acid is added. This approxi-mates closely to the acid required to titrate a solution to the pH inflection point near pH 4.5. The total alkalinity is often expressed per liter (20 C) rather than per kilogram. A volume basis is used in the tables and equations in the present method.

SPECIFIC ALKALINITY

This is defined as either of the ratios:

total alkalinity ( kilogram basis) total alkalinity ( liter basis)

Cl%o Cl/liter(20 C)

and is about 0.123. The ratio may vary appreciably, however, according to the nature of the water mass concerned and must be determined experimentally for the most precise work.

CARBONATE ALKALINITY

The carbonate alkalinity is the number of milliequivalents of hydrogen ion that are neutralized by 1 kg of sea water ( or 1 liter at 20 C in the present method) in converting carbonate and bicarbonate ions to carbon dioxide by the reactions:

C032- + H+ ;=^ HC031-HC031- + H+ ;=^: H2CO3

In all natural sea waters with a salinity exceeding about 10%0, the total alkalinity will arise from the ions CO32-, HCO31-, H2BO31-, OHl- and, in a negative sense, H+. The effect of other ions can be neglected. Therefore, by definitions, if

represents concentration in millimoles per liter,

Total alkalinity = [HCO3]' + 2 [CO3]2- + [H2BO3]1- + [OH]1- - [H]+

and whence:

Carbonate alkalinity = [HCO3]' + 2 [C03]2-,

Carbonate alkalinity = total alkalinity - [H2BO3]1- - [OH]1- + [H]+ . The terms [OH]1- and [H]+ may be neglected in the pH range 5.3-8.7. The term

[H2BO3]1- may only be neglected at pH values less than 7.3 which will be rarely, if ever, encountared.

The entire carbonate "balance" in a sample of sea water can be evaluated from a knowledge of the total alkalinity and pH, provided we also know the disso-ciation constants of carbonic acid and boric acid in sea water of various salinities and temperatures.

The relevant data for correcting laboratory-measured pH values to give results 27

28 A PRACTICAL HANDBOOK OF SEAWATER ANALYSIS

at in situ depths and temperatures and for calculating concentrations of the various carbonate species have been published (Buch and Gripenberg, J. Conseil, Conseil Perm. Intern. Exploration Mer, 7: 233, 1932; Buch and Nynâs, Acta. Acad.

Àboensis Math. Phys., 12: (3), 1939; Buch, Havsforskn. Inst. Skr., Helsingf., (151), 1951). An excellent account is given by Harvey, drawing from these sources, and is to be found in his book The Chemistry and Fertility of Sea Waters (Cambridge University Press, 1957). The following presentation stems largely from this source but the pH range for some tables has been extended and we have also used more recent values for the second dissociation constant of carbonic acid. These values obtained by Dr John Lyman (Ph.D. Thesis, University of California, Los Angeles, 1957), differ significantly from the Buch values and led us to recompute some tables.

The corrections made for the presence of boric acid are based on the assumption that free boric acid is always present in sea water at a constant ratio to the chlorinity.

This is not necessarily the case, especially in some inshore waters, as a variable amount of the boron may be tied up as organic complexes of differing buffering powers (e.g. Gast and Thompson, Anal. Chem., 30: 1549, 1958) but the errors in carbonate calculations thus incurred are probably rarely important as long as the pH of a sample is less than 8.25.

The method for determining total alkalinity that is described here is taken from Anderson and Robinson (Id. Eng. Chem. [Anal. Edition] 18: 767, 1946) and is sufficiently precise for most purposes. The method acts as a check of the specific alkalinity. As the latter is near 0.123 in most sea areas it is only necessary to make alkalinity measurements on samples taken at great depths or at inshore locations where there is marked land drainage or where active carbonate precipitation is suspected. If changes of specific alkalinity are required as a delicate indicator of differing water masses or of biological activity the Anderson—Robinson technique is not sufficiently precise and very careful direct titration experiments are essential (a good example is given by Cooper, J. Marine Biol. Assoc. U.K., 19: 747, 1934).

For pH measurements to have much value in such a buffered medium as sea water they must be carried out with great care using internationally accepted standard buffers as reference points on a glass-electrode pH scale.

The treatment below is divided into two main subdivisions, 1.4.1. and 1.4.11.

In the first, we describe the experimental details for the determination of pH and total alkalinity. In the second, we describe the calculations necessary for the full evaluation of the carbon dioxide system.

1.4.1. DETERMINATION OF EXPERIMENTAL QUANTITIES 1.

pH

DETERMINATION

A. CAPABILITIES

Range: pH 7.3-8.6 PRECISION

The

equipment

and conditions

that can be used

are

so varied that

no

general statement

of

precision can be

made. If the pH

is used

as a

general indicator

of

water

masses and

water

structure a

precision

of

71- 0.03

unit

is probably sufficient

and

should be easily obtainable.

For

measurements designed to be used

in the

calculation

of

the

total

carbon dioxide

content of

sea water

a

precision

of

-± 0.02

unit

should be attempted

and

can be obtained under sea-going

conditions if

care is taken.

For

measurements designed to be used

in the

calculation

of the relative

amounts

of the

various forms

of

carbon dioxide (CO2— , HCO32— ,

and free

CO2 )

the

highest

possible

precision is required

in pH

measurement. Ideally

a

precision

of

-± 0.005

unit

should be attempted

but

under most

conditions

it is doubtful whether better than

--t0.01 can be achieved

and

then

the

greatest care

in

sampling

and

standardization is required.

Im Dokument A PRACTICAL HANDBOOK OF SEAWATER ANALYSIS (Seite 37-41)