Microdetermination of carbon dioxide

Microdetermination of carbon dioxide

ANALYTlCAL 3, 464-471 (1982) BICCHEMISTRY Microdetermination JOHN From the of Carbon T. VAN BRUGGEN Depnrtmsnt of Biochemistry, Portland, AN...

447KB Sizes 2 Downloads 59 Views

.ANALYTlCAL

3, 464-471 (1982)

BICCHEMISTRY

Microdetermination JOHN From

the

of Carbon

T. VAN BRUGGEN

Depnrtmsnt

of Biochemistry, Portland,

AND JEAN University Oregon

Received November

of

Dioxide’ C. SCOTT Oregon

Medical

School,

20, 1961

Since carbon dioxide is a major metabolite of many systems, the need often arises to determine as COz the total of H&O,, HCO,-, and CO,= in a biological system. In addition, it is often desirable to be able to determine the isotopic concentration, i.e., radioactivity, of the same sample. When the carbon dioxide is present in micro amounts and the solution volumes containing the COz are large, i.e., 50-100 ml, the accurate, convenient determination of concentration and radioactivity of CO2 has been difficult. Martin and Green (1) have reviewed the methods available for CO, determination. These authors and others (2,3) have combined the use of spiral absorbers and conductivity measurements for the determination of COz. These techniques, however, are not well adapted to the requirements above. Holm-Jensen (4) has described an elegant, simple, spiral absorption device that not only provides for the determination of small amounts of CO, in air but also provides a means of concentrating the CO, in a small volume of absorbent. The description of the spiral absorber and the accompanying details given by Holm-Jensen (4) are directed toward the measurement of CO, in air, and not toward the measurement of CO? in fluids. In addition, the measurements reported by him were done manually with equipment not generally available. This paper presents the details of the adaptation of the spiral absorber of Holm-Jensen to the determination of CO, present in solution at the level of 0.13 to 2.6 pmoles per sample. Instrumentation is described that allows for the continuous, automatic recording of the conductivity of the measuring solution. The concomitant determination of the radioactivity of the sample is also described. APPARATUS

The spirals used in the initial ’ This work tion of Oregon Arthritis and Public Health

work were obtained

at the University

was supported in part by a grant from the Medical Research Foundaand iu part by research grant A-2305 from the National Institute of Metabolic Diseases, National Institutes of Health, United States Service. 404

M ICRODETERMINATION

OF CU2

465

of Aarhua,2 subsequent ones have been made locally.3 The electrodes of the spirals are made essentially as described by Holm-Jensen (4). With 0.005 N KC1 bathing the electrodes, the solution shows a resistance of 17,000 ohms; a 0.01 N KC1 solution shows 8,500 ohms, both as measured with a Serfass conductance bridge A. H. Thomas No. 3965. The conductance of the solutions can be measured manually with the Thomas bridge but in the present studies the conductance of the solutions has been measured with a Leeds and Northrup Conductivity Monitor Model 4958 with a calibrated scale 0 to 100 micromhos cm-l fitted with an automatic temperature compensator L and N No. 152022. When the L and N Monitor is used with conventional conductivity cells, the scale reading (0 to 100) is multiplied by the cell constant to obtain the true micromhos per centimeter. In the present studies, the meter reading alone is recorded and used in the calculations. With the “range adjustment” of the L and N fully advanced, 0.005 N KC1 gave a reading of 83. The automatic temperature compensator probe is placed in an area between the spiral and the S tube (Fig. 1) to sense temperature changes and to relate all measurements to a base temperature of 77°F. The 0 to 10 mv (across 100 ohms) output of the monitor is fed to a Photovolt Corporation Varicord 43 linear/log self-balancing, potentiometer, strip chart, recorder. The recorder has a 12-in. width paper and is used with its calibrated lo-mv range. Manipulation of the range adjuster switch of the monitor permits the use of strontium hydroxide absorber solutions over the range of 1.3 to 4.3 pmoles/ml and permits the use of the full sensitivity of the recorder assembly. CHEMICALS

(a) A stock solution of C.P. Sr (OH)z approximately 50 ~moles/ml is prepared and is saturated with SrCO,. Aliquots of this solution are used with CO,-free water to prepare absorber solutions in the range of 1.3 to 4.3 pmoles/ml. The choice of the strength of solution is dictated by the sensitivity desired and the amount of COz to be measured. In general it is inadvisable to use up more than one half of the capacity of the absorber, i.e., a 2.5 ymoles/ml solution is adequate for the analysis of up to 1.3 pmoles of CO,. It is also desirable to use an absorber solution ‘The authors are deeply indebted to I. Holm-Jensen of the Pharmacological Institute, University of Aarhus, Denmark, for his kindness and help in the use of the spirals. Mr. Chorfitz Peterson of the University of Aarhus is thanked for his preparation of the original spirals used. Dr. K. Zerhan of the University of Copenhagen suggested the use of the S-shaped tube for the liberation of the CO2 and his kind assistanre is acknowledged. ’ Available from Howard Benson, 1801 N. C.sllowav St ., McMinnvillr. Oregon

466

VAN

BRUGGEN

AND

SCOTT

that will give an adequate change in conductivity with small amounts of CO,. In this regard, we generally use a Sr(OH)2 with not more than &fold the capacity of the amount of CO2 to be analyzed. (b) CO,-free water is easily obtained from the proper use of the Research Model Cartridge De-Ionizer of the Illinois Water Treatment Co., Rockford, Illinois, and Gardner Laboratories, Bethesda, Maryland. (c) CO,-free air is obtained from water-pumped Nz cylinders. The gas is passed over Ca (Cl) 2 for drying and then through two Drierite Gas Drying Unit absorbent towers in series, filled with soda lime. An additional source of CO,-free air is desirable for use in the flushing of containers and syringes prior to their use. (d) Sr(OH), solutions are added to the spiral via l-ml syringes fitted with a needle which in turn is extended by the use of small sized polyethylene tubing. The high CO2 content of respired air makes the use of a pipetting technique extremely hazardous. DESCRIPTION

OF

THE

APPARATUS

Figure 1 illustrates the components of the assembly. The lower vessel, the S-shaped acidification tube, is made from glass tubing of such a diameter, and the arms of the tube of such a length, that the volume of fluid to be acidified and analyzed will fill only some t.wo thirds of the total vo1ume.4 This will assure a short turnover time for the air space of the tube. CO,-free air (or air to be analyzed) is introduced through A ; stopcock B is used as an off-on control! the fine control of the air flow rate being more conveniently done with the aid of a microneedle valve in the CO? free air line. Tube C is of a convenient size, i.e., 8 mm, and is closed with a vial closure permitting repeated additions of solutions via syringe and needle. The outlet of the S tube at D is connected to the spiral by Tygon tubing. Tygon is used as a flexible connector because CO, does not readily penetrate its walls. To facilitate the connection of the tubing to the glassware, the Tygon tube is fitted at each end with a short piece of glass tubing and a connector piece of rubber t,ubing. Approximation of the ends of the glass tubing with the openings on the apparatus minimizes loss of CO, or contamination of the air stream from external CO?. The spiral is similar to that described by Holm-Jensen (4). In the present studies the spirals were made of 6 turns of 8 mm tubing and were capable of handling 15 ml/min of air flow with 1 ml Sr(OH), solution. The outlet of the spiral, G, is connected to a coil of Tygon or glass tubing which serves as a CO*-free dead air ‘The .‘i tube unlikely

air space above the fluid in the S tube, the presence of a sharp bend at the spiral end, and the very slow flow of air through the system the contamination of the spiral by air entrainment, of acid, etc.

in the makr

.MICRODETERMINATION

FIG.

1. Carbon

dioxide

467

OF CO:!

absorption

apparatus.

space to prevent back-diffusion of room air CO,. The spiral is supported by a spring grip clamp (A. H. Thomas) attached to the top horizontal portion of the spiral. Cleaning of the spiral is facilitated by the successive additions of water, dilute acid, water, and acetone to the opening F while water pump suction is attached to G. Draining of these solutions is speeded by swinging the spiral 180” on its spring clamp support. PROCEDURE

FOR CARBON

DIOXIDE

ASSAY

Normal H2SOa (1 ml) is added to the S tube, the CO*-free air stream is attached, the Tygon tube is used to connect the S tube and the spiral, the outlet trap is attached at G, and then CO,-free air is passed through

468

VAN

BRUGGEN

AND

WO’l”1

the assembly to remove the CO, originally present. The conductivity monitor and recorder are turned on and allowed to warm up. The time required for the degassing of COZ depends upon the flow rate of the sweeping air stream. The air flow is stopped, and the spiral is charged with 1 ml standard Sr(OH),, the solution being added at G via syringe and extended needle. The recorder drive is started and the range adjuster of the monitor is manipulated to obtain a deflection of the recorder close to the maximum of the lo-mv range. (This adjustment is conveniently made at a previous time and remains essentially constant for Sr++ solutions of identical strength.) With the establishment of the zero time reading, prior to circulation of the absorber solution, the air stream is now started. After an initial passage of most of the solution in the form of a single column of fluid, the liquid and gas phases should now distribute so that sections of fluid of some 10 mm in length are alternated with air sections some 50 mm in length. The air speed is adjusted so that about 3 set is required for a section of fluid to make one complete turn (360”) of the spiral. This close control over the air stream is best achieved with the use of a microneedle valve in the air line distal to the COZ absorbers in the air line. As described by Holm-Jensen (4) there is an initial small rapid fall in conductivity due to adsorption phenomena, this being followed by a much smaller change. This gradual change seen as the N, stream continues to circulate the fluid is minimal if there is no contamination with CO, and if the N, stream is essentially COz free. When the operator is satisfied that optimum conditions prevail, the air stream is slowed or shut off entirely and the sample is injected at C via syringe and needle, or allowed to flow into the system at A or C. In any event, the flow rate of addition should be controlled so as to not exceed the previous flow rate. After the addition of sample, the air stream is again started and the COZ liberated in the S tube is absorbed in the spiral. Figure 2 represents an actual trace of an experiment, the details of which are given below. Sample ru7t: The initial reading, at point A, is 88.2 this representing the maximum conductivity of the solution, and was obtained prior to circulation of the Sr(OH), through the spiral. The inflections of the trace between A and B represent the time required for the solution in the spiral and gas flow to reach a steady, even rate of circulation. In this period, (A-B) there is also seen the initial rapid dropoff in conductivity attributed by Holm-Jensen (4) to adsorption phenomena. The slope of the line between B and C is minimal and represents the usual slope for the indicated reagent when the air source is low in CO,. At the point C

~ilCRODETERMINATlON

469

OF WI:

80

70

60

50

40

L

I

I

I

I

I

0

5

IO

15

20

TIME FIG.

2.

IN MINUTES

The time course of the conductimetric

measurement

of CO*.

the sample containing 1.25 pmoles COa’ in 10 ml H,O was injected into the S-shaped acidification tube. The interval C-D represents the time required for movement of the front of the liberated gas into the spiral. In less than 5 min the greater part of the CO, had been absorbed. The slope of the line E-F is greater than the B-C slope due in part to traces of CO* yet being absorbed. The E-F slope is now projected toward the zero time axis and the point G fixed as the intercept of the E-F slope with the vertical time line that passes through C, the injection time or point. The calculation of the amount of CO, is made as follows:

470

VAN

BRUGGEN

AND

SCOT.?

Point A = 88.2 Original Sr(OH)2 at A = 2.42 Point C = 83.2 Sample 10 ml containing 1.25 Point G = 37.0 83.2 X 2.42 = 2.28 pmoles Sr(OH)n/ml present 88.2 37 x 2.28 = 1.01 pmoles Sr(OH)Jml present 83.2 2.28 - 1.01 = 1.27 Ltmoles CO1- in sample = 102y0 recovery APPLICATION

OF THE

TECHNIQUE

AND RANGE

pmoles/ml pmoles CO? at C at G

OF USEFULNESS

In our hands, 1 to 25 ml of solution containing between 0.13 to 2.55 pmoles of CO, have been analyzed with Sr(OH), solutions containing 1.29 to 4.01 ~moles/ml. As stated previously, it is wise to choose the combination of reagents that will allow a maximum sensitivity coupled with the desired range or capacity of reagent. Table 1 records sample data obtained from t,he analysis of standard TABLE 1 RECOVEBY OF CO2 FROM STANDABD SOLUTIONS~ Sr(OH)z used (/moles)

No. of runs

1.29

0.128

4

2.43

0.128

5

1.29

0.26

5

2.43

0.26

5

2.43

0.64

5

4.01

0.64

5

2.43

1.28

10

CO, recovered (rmoles)

Avemee

(%)

0.13 f 0.003b (0.12-0.13) 0.13 * 0.002 (0.13-O. 14)

101.5

0.27 k 0.004 (0.26-0.28) 0.27 f 0.005 (0.26-0.27) 0.67 f 0.007 (0.66-0.68) 0.66 f 0.012 (0.63-0.69) 1.31 f 0.011

104.5

102.4

104.2 104.0 102.4 102.3

(1.24-1.37)

4.01

1.28

5

1.30 f 0.009 (1.27-1.32)

101.4

4.01

2.6

5

2.6 f 0.020 (2.53-2.72)

101.6

0 NazCOs anhyd. A.R. used for standard solutions. b Upper line is mean f standard error, lower line is the range of the values obtained.

MICRODETERMINATION

OF

CO2

471

Na,CO, solutions. It is seen that the limits of accuracy of the method, in our hands, are good. The method may well be even more accurate if additional attention is given to a number of factors. For the general needs of determining the amounts of CO, produced by biological systems respiring in a closed environment, i.e., Warburg flasks, etc., the method is most useful. RADIOASSAY

OF THE

CARBON

DIOXIDE

After the initial determination of the amount of CO, in the sample, the measured CO, is present in the spiral as insoluble SrC03. Determination of the radioactivity of the sample is easily made as follows: The opening F of a second spiral (which need not be fitted with electrodes) is attached to the outlet G of the first spiral by a length of Tygon tubing. The second spiral is charged with 1 ml of absorbing solution. This solution may be a soluble alkali such as NaOH or KOH with which CO, will form a soluble salt. The strength of the solut,ion need be adequate only to absorb the COz and need not contain additional alkali to interfere with the subsequent radioassay of evaporated solutions. CO, is liberated from the first spiral by the addition of dilute acid via syringe and needle, entrance being made through the rubber connector at F (on the first spiral). The air stream should be adjusted to a minimum rate because CO, liberation is rapid. When the CO2 is now trapped as salt in the second spiral, aliquots of the solution of the second spiral are taken via syringe and polyethylene extended needle probing via the outlet G. CONCLUSIONS

A microtechnique has been described that will allow the determination of small amounts of CO, by a conductimetric procedure. Under the conditions herein described, as little as 0.013 /*mole of CO, per milliliter in a lo-ml sample, can be determined with considerable accuracy. With the instruments and technique described, the method will determine up to 2.6 pmoles of CO? per sample. REFERENCES 1. 2. 3. 4.

MAKTIK, W. MC., AND GREEN, J. R., Ind. Eng. Chew, Anal. Ed. DAILLY, D. F., AND ELLIOTT, T. A., J. Chem. Sot. 3, 3398 (19.56) GOODWIN, R. D., Anal. Chem. 25, 263 (1953). HOLM-JENSEN, I., Anal. Chim. Acta 23, 13 (1980).

5, 114 (1933).