Page images
PDF
EPUB

This simply scales concentration (C) in direct proportion to the area A of specimen involved and in inverse proportion to the chamber volume V. As an example, the gas concentration in a 10,000 ft3 cabin is shown in figure 11 for a series of lines corresponding to surface areas of 10, 100, and 1,000 ft2.

It should be noted that such scaled estimates assume similar (or uniform) distribution of the gaseous components, and large differences may result in the case of active gases and vapors which tend to be adsorbed on surfaces, e.g. HF

and HCl, and gases and vapors which tend to stratify in layers.

Finally, it should be noted that relationships between the indicated concentrations measured in the smoke chamber and physiological or toxicological effects are also outside the scope of this study. The table of toxicological data, assembled from open literature sources has been included for reference purposes only. Information on the combined, or synergistic, effects of several noxious components (including smoke articles) is apI parently very limited.

5. Conclusions

Based upon the tests performed and an evaluation of the results, the following conclusions have been reached:

1. Materials currently used as interior furnishings for aircraft cabins, and those being considered for future use, vary considerably in their production of smoke and potentially toxic products under simulated fire conditions.

2. The laboratory test method for generating smoke and measuring its optical density appears to be a useful tool for the quantitative classification of materials, and for the possible establishment of revised fire safety standards and criteria for controlling smoke production. Optical density is the single most characteristic measure of the visual obscuring quality of a smoke.

3. For evaluating smoke production, both smoldering and active flaming conditions should be considered. For the majority of materials, more smoke was produced during the flaming exposure test. However, certain materials produced significantly more smoke in the absence of open flaming.

4. Within the limitations and assumptions cited, the specific optical density of smoke measured

in the laboratory may be extrapolated to cabin volumes and surface areas of combustible furnishings in order to provide guidelines for cabin area limitations, or to estimate time periods available for escape or defensive action.

5. Indications of the concentrations of potentially toxic combustion products can be conveniently and inexpensively obtained during the smoke production test using calibrated commercial colorimetric tubes; however, these are suitable only where interferences by other gases are absent, and where precision is not of primary importance. The specific ion electrode is also a convenient method of measuring the concentrations of halogen acid gases. Furthermore, if an attempt is made to relate the indicated concentrations measured in the smoke chamber in terms of toxicological limits, caution must be excercised. It is essential that proper consideration be given to (a) scaling of the areas and volumes in the proposed situation, (b) the integrated dosage where concentration varies with time, (c) the synergistic effects of several components (and smoke particles), and (d) the effects of relative humidity, elevated temperature, stratification, adsorption on surfaces, and physiological factors not considered in this study.

6. References

[1] Marcy, J. F., Nicholas, E. B., and Demaree, J. E., Flammability and Smoke Characteristics of Aircraft Interior Materials. Federal Aviation Agency Technical Report ADS-3, Jan. 1964.

[2] Marcy, J. F., A Study of Air Transport Passenger Cabin Fires and Materials, Federal Aviation Agency Technical Report ADS-44, Dec. 1965.

[3] Gross, D., Loftus, J. J., and Robertson, A. F., A Method for Measuring Smoke from Burning Materials, American Society for Testing Materials Special Technical Publication 422, 1967.

[4] a. Scott Draeger Multi-Gas Detector, distributed by Scott Aviation Corporation, Lancaster, N.Y.

b. MSA Colorimetric Gas Detector Tubes, Mine Safety Appliances Co., Pittsburgh, Pa.

c. Kitagawa Precision Gas Detector, Unico Model No. 400, Union Industrial Equipment Corp., Port Chester, N.Y.

[5] Madorsky, S. L., Thermal Degradation of Organic Polymers, 315 pp. (Interscience (Wiley) 1964).

[6] Ausobsky, S., Evaluation of the combustion gases of plastics, (in German), VFDB Zeitschrift 16, 58-66, 1967.

[7] Coleman, E. H. and Thomas, C. H., The products of combustion of chlorinated plastics, J. Appl. Chem. 4, 379-383, 1954.

[8] Fish, A., Franklin, N. H., and Pollard. R. T., Analysis of toxic gaseous combustion products, J. Appl. Chem. 13, 506-9, 1963.

[9] Kusnetz, H. L., Saltzman, B. E., and Lanier, M. E., Calibration and evaluation of gas detector tubes, Am. Ind. Hyg. Assoc. J. 21, 361-373, 1960.

[10] Saltzman, B. E., Preparation and analysis of calibrated low concentrations of sixteen toxic gases, Anal. Chem. 33, 1100-1112, 1961.

[11] Saltzman, B. E. and Gilbert, N., Am. Ind. Hyg. Assoc. J. 20, 379-386, 1959.

[12] Rechnitz, G. A. and Kresz, M. R., Anal. Chem. 38, 1786, 1966.

7. Appendix 1. Gas Analysis

7.1. Colorimetric Indicator Tubes

The manufacturer provided general information on the detector tubes regarding their measuring range, interfering reactions, reuse and the effects of temperature and relative humidity. The upper and lower limits of the measuring ranges of these tubes and some references to the toxicological limits of these gases are summarized in table 2. With good quality control during manufacture and frequent calibration, specific tubes can give meaningful results. However, certain shortcomings may be noted. These include:

1. Variation of packing density within the tube and nonuniformity of indicator gel among the tubes. Since the adsorption rate of a sample gas by the gel depends primarily on the reacting surface area available per length of tube, variable packing density would affect reproducibility.

2. Certain gases and vapors are not adsorbed by the precleaning layer but react similarly with the indicator as the gas of interest to produce an unexpected interference.

3. The transition zone of the discolored stain front makes it difficult to judge the exact demarcation line and thus introduces errors.

These shortcomings can be minimized, for example, by frequent calibration to establish probable sources of errors, by knowing the specific interfering gases in the sample not absorbed by the precleaning layer and the sensitivity of the tube to these gases, and by determining the concentration of the interference gas, if any, found in the sample. With cumulative experience on using the tubes both during calibration and sampling, the error in judging the line of demarcation of the discolored section by an operator can be minimized. The merit of the colorimetric tubes as in any other analytical method should be judged by performance on a specific gas. Sensitivity, accuracy, and interference effects depend on the chemical system used in the tube and they are

obviously different for different gases. An extensive review of some of the techniques and problems associated with these tubes is given by Kusnetz, et al. [9].

The advantages of the indicator tubes are convenience and simplicity, yielding immediate results with the avoidance of transfer vessels and other sampling problems. In the hands of an experienced operator, reasonable accuracy can be attained.

Of the colorimetric tubes used, tubes for four compounds have been calibrated and examined for interferences and temperature effect. For calibration purposes low concentrations of HCl or HCN were prepared from a flow dilution system suggested by Saltzman [10]. The system consists of an asbestos plug which serves as a flow-limiting device [11], and a mixing chamber as shown in figure 12. Tubing to the asbestos flowmeter is 1 mm i.d. polytetrafluoroethylene tubing to minimize dead volume. The pressure regulating cylinder was filled with concentrated H2SO4. Flows were calibrated by attaching a graduated 0.1 ml pipet to the meter outlet and timing with a stopwatch the movement of a drop of mercury past the graduations. Flow rates as low as 0.01 cm3/min can be achieved with good long term stability.

The degree of dilution of pure HCl from the tank was controlled by the asbestos plug and the diluting gas metered by a rotameter. Mixture concentration could be varied from 10 to 1000 ppm. A needle valve controlled the flow rate to the indicator tube. The pressure drop across the colorimetric indicator during calibration was balanced by applying an appropriate vacuum at the other end of the tube. This arrangement avoids creating any disturbance to the diluting system when the tube is inserted to start a calibration.

Low concentrations of HCN were generated by aeration of a 4.6 molar solution of KCN in a midget impinger. A thermostated water bath surrounding the bubbler and air supply condenser maintained a temperature of 30 °C (86 °F). The system produced an output of 100 ppm and further dilution was necessary for lower concen

TABLE 2. Measuring range of colorimetric indicator tubes and toxicological data for selected gases

[merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][ocr errors][ocr errors][merged small][merged small][merged small][merged small][merged small][merged small][subsumed][merged small][merged small][merged small][ocr errors][merged small][merged small][merged small][merged small][merged small][merged small][ocr errors][merged small][merged small][merged small][ocr errors][merged small][merged small][ocr errors][merged small][merged small][subsumed][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][ocr errors][subsumed][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small]

Henderson, Y., Haggard, H. W.: Noxious Gases (Reinhold Publishing Corp., New York 1943).
Elkins, H. B.: The Chemistry of Industrial Toixiology (John Wiley & Sons, Inc., New York 1959).
American Conference of Governmental Industrial Hygienists, Document of Threshold Values, Cincinnati, Ohio 45202 (1966 edition).

** Maximum average atmospheric concentration for 8-hr daily exposure adopted by American Conference of Governmental Industrial Hygienists, 1966.

[merged small][ocr errors]

1. Gas of Interest 2. Pressure Regulator

trations. Both HCl and HCN systems were very stable and consistent.

A static method using an FEP polytetrafluoroethylene 5-mil-thick collapsible bag was used to generate low concentrations of nonreacting gases. Under this arrangement, the sample gas was deposited by a gas-tight microsyringe and diluted with air or other gases from a 1-liter syringe. This method is not applicable to HCl or HCN because of losses resulting from adsorption, but gave satisfactory results with CO from 10 to 1000 ppm.

7.2. Specific Ion Electrode

A permeable membrane electrode for chloride ions (after Pungor) described recently [12] was

[ocr errors]

3. Asbestos Plug Rotameter

4

5. Diluting Gas 7. Colorimetric Tube
6. Mixture Waste 8. Vacuum Source

used in a system to determine the HCl concentration in a gas sample potentiometrically. This method has higher accuracy, range, and reliability than that of colorimetric indicator tubes. Its working range is between 20 and 20,000 ppm for a 100 cm3 gas sample. For lower concentrations, a larger sample must be used.

In practice, the highly water-soluble HCl gas and vapor in the 100 cm3 sample was totally absorbed when the sample flowed at a rate of 100 cm3/min through polytetrafluoroethylene tubing (5.3 mm i.d.) containing about 40 mg of loosely packed glass wool wetted with 0.1 cm3 water. The exposed glass wool was carefully transferred to a polytetrafluoroethylene cup of small

[blocks in formation]

FIGURE 13. Measured emf as a function of HCl concentration in gas phase based on HCl in 100 cm3 sample absorbed by 1 cm3 of water.

[blocks in formation]

**The lack of mutual interference among HCl, HCN and CO for these tubes for concentration up to 1000 ppm HCl, 100 ppm HCN, and 1000 ppm CO was confirmed by NBS results.

internal volume. Water was added to make a total solution of 1 cm3 before insertion of the specific ion electrode, and a low-leakage, smalldiameter-tip, conventional calomel-KCl reference electrode. A high impedance differential voltmeter or an expanded scale pH meter may be used to measure the emf between the electrodes. The specific electrode has a sensitivity limit of 10-5 mole per liter for chloride ion in solution and an equilibrium response time of about 1 min. It consists essentially of a polymeric silicone rubber membrane impregnated with particles of silver chloride precipitate. The membrane covers the tip of a small diameter glass tube filled with a chloride solution. Figure 13 shows the calibration curve of emf in mV and HCl concentration in ppm calculated on the basis of a 100 cm3 gas sample absorbed in a 1 cm3 solution. The curves were based on measurements made with solutions of known HCl concentrations.

Known interferences of bromide or iodide ions may be considered negligible if their concentrations are less than one-tenth of the chloride ion concentration [12]. In most fire gas or smoke chamber analyses no interference would be expected. In cases where the concentration of bromide ions is likely to be the same order as that of chloride, a bromide specific electrode can be used. This electrode is not affected by chloride ion concentrations as high as 50 times that of bromide.

Table 3 shows the type of indicator reagents used in the detector tubes. It also lists the known components and concentrations which would cause sufficient interference to give erroneous readings.

Basic data (supplied by Mfr.)

The precleaning layer serves to remove the interfering components and the table shows the maximum concentrations that can be tolerated. The data are based on information furnished by the tube manufacturer as well as NBS data showing the lack of mutual interference among the major components of HCl, HCN and CO. Except for H2S which apparently poisons the reactive surface in the HCl tube, other interferences did not significantly alter the usefulness of those colorimetric tubes in the present smoke chamber study.

Table 4 shows some of the basic and calibration data for the colorimetric indicator tubes used. Included are the concentration ranges for which the tubes are rated and the sample volume and measured sampling rate for which the predetermined scale calibration holds. The length of indicating layer compared with the maximum of the concentration range indicates the resolution of the tube. The transition zone is a subjective estimate of the length between complete color change to no change which affects the reading error. The calibration ratios were based on the average of three separate runs for each of the stated concentrations. The method of preparation of an actual concentration of a single component in an atmospheric air mixture was given in the previous section. Unlike a previous study where several disinterested observers were asked to judge the demarcation front of the color change [9], the present results were based on the observation of one individual only. With the exception of the type B HCl tube which was +90 percent in error, all other errors fell within a ±20 percent range.

[blocks in formation]
[merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][ocr errors][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][subsumed][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][ocr errors][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small]

Concentration range extended by use of individual calibration (not furnished by the manufacturer). b Average readings of three separate tubes.

« PreviousContinue »