ANALYTICAL CHEMISTRY
1586 tive method of test for smoke point of jet fuels (1). Repeatability is good; duplicate determinations by the same operator do not differ by more than 1 mm. Different operators using different lamps agree within 2 mm. The sensitivity of the new lamp was compared with those of the Institute of Petroleum and Factor lamps on widely differing fuels (Figure 7 ) . Blends of No. 2 fuel oils provided low smoke points. No. 1 fuel oils, kerosines, jet fuels, and gasolines covered the intermediate range. Individual alkanes provided high smoke points. The new lamp is much more sensitive to fuel quality than the Institute of Petroleum lamp. It doubles the spread a t low values and quadruples it a t intermediate values. The Institute of Petroleum lamp does not differentiate among fuels of high smoke point, as it was designed for the lower range. The new lamp is slightly more sensitive than the Factor lamp. The spread is increased by about 10%. The dimensions of the Factor lamp were evidently selected with some knowledge of how they would affect smoke point. However, the increased precision of the new lamp gives it a major advantage. CONCLUSION
The increased emphasis on reducing carbon deposition in jet engines has greatly increased the need for better methods of specifying fuel burning quality. The improved lamp will help meet this need and should prove useful in specifying burning quality of kerosine and home-heating oils. If this lamp finds
acceptance in the industry, arrangements will be made to make it available through normal suppliers under license. ACKNOWLEDGMENT
The valuable assistance of George Hajduk in this study is acknowledged. LITERATURE CITED
Am. Soc. Testing Materials, Philadelphia, “Standards on Petroleum Products and Lubricants,” p. 753, 1954. ( 2 ) Clark, T. P., “Influence of External Variables on Smoking of Benzene Flames,” Natl. Advisory Comm. Aeronautics,
(1)
NACA RM E52G24 (1952). (3)
Clarke, A. E., Hunter, T. G., Garner, F. H., J . Inst. Petroleum 32,627-42 (1946).
Hunt, R. h.,I n d . Eng. Chem. 45, 602 (1953). ( 5 ) Institute of Petroleum, London, “Standard Methods for Testing Petroleum and Its Products,” p. 434, 1955. ( 0 ) Kewley, J., Jackson, J. S., J . Inst. Petroleum Technol. 13, 364 (4)
(1 927). (7) Alilitarg Specification MIL-F-5624C,
18 May 1955, Fuel, Aircraft Turbine and Jet Engine Grades JP-3, JP-4, Jp-5. (8) Schalla, R. L., bIcDonald, G. E., Ind. Eng. C h m . 45, 1497-500 (1953). (9) Terry, J. B., Field, E., IND.ENG.CHEX.,A N A L . ED. 8, 293 (1936). (10) Woodrow, W. A,, V o r l d Petroleum Congr. (London), Proc. 2 , 732 (1933). (11) Worrall, G. I., Ind. Eng. Chem. 46, 2178 (1954). RECEIVED for review February 8, 1956. Accepted June 21, 1956. Division of Petroleum Chemistry, 129th Meeting, ACS, Dallas, Tex., April 1956.
Identification of Synthetic Fibers by Micro Fusion Methods DONALD G. GRABAR and RITA HAESSLY Central Laboratory, lndurtrial Rayon Corp., Cleveland, Ohio
A scheme for the identification of synthetic fibers by the use of micro fusion methods is based upon the melting point of the fiber, the eutectic temperature of the fiber with p-nitrophenol as a reference compound, and the characteristic behavior observed during the heating and cooling of the fibers. Observations are made using a hot stage on a polarizing microscope. Reproducible melting points are obtained by using a silicone oil as a mounting liquid for the fibers to exclude air from the fibers while heating and to improve the microscopic image. Tabulated micro fusion data are given for thirteen synthetic fibers.
T
HE technical literature abounds in descriptions of various techniques for the identification of textile fibers. These methods are based principally on microscopical observations of morphological characteristics visible in cross section and longitudinal views, together with supplementary tests such as staining, solubility, and refractive index (1, $, 4). Although in favorable cases each of these techniques may afford positive identification in itself, in the majority of cases one or more of the tests must be used to confirm the results obtained by another. The method described here, using micro fusion methods, is selfsufficient in most instances, but in some cases also must be confirmed by one or more of the other tests. However, it presents advantages over previously described techniques in being more widely applicable to both dyed and undyed, bright and dull, and filament and staple fibers. It is also generally faster and more
positive in distinguishing between chemically similar fiberse.g., Nylon type 6 and Kylon type 66. Although the fibers used in the test are destroyed, only very small samples are required. Micro fusion methods have heretofore been applied only to monomeric systems. The general technique was originally developed by the Koflers in Germany ( 3 ) ,and extended and promoted in this country mainly through the efforts of McCrone (6) and coworkers. The identification of an unknown is based upon the use of a hot stage and microscope for determining (a)the melting point of the compound, ( b ) the eutectic melting point of the compound with a reference compound, (c) the refractive index of the melt, and ( d ) characteristic behavior observed during the heating and cooling of the compound. The determination of refractive index is unnecessary in the application of the technique described herein. Melting points have been generally considered of limited use in fiber identification for two reasons: synthetic fibers which melt usually do so nonreproducibly over an appreciable temperature range, and many of the synthetic fibers decompose completely before their melting point is reached. The first objection arises primarily from the common method of measuring fiber melting points, the copper block method. By this method a sticking temperature or softening point is observed, which, although useful in indicating use properties of the fibere.g., maximum ironing temperature-is of little analytical value High polymeric fibers invariably are incompletely crystallized When the fiber is heated, this incomplete crystallization causes a softening of the amorphous portions before actual melting of the crystalline parts occurs. Furthermore, a range of crystalline
V O L U M E 28, NO, 10, O C T O B E R 1 9 5 6
1587
perfection exists which extends the range over which true melting takes place. In addition, most fibers are susceptible to Some degree of decomposition when heated in the presence of air, and this decomposition also contributes to the nonreproducible me& ing points obtained by the usual methods. However, by use of 8 hot stage on a polarieing microscope, analytically useful melting points and much supplementary information can be obtained by actually observing the fibers as they are being heated. Although the true thermodynamic melting point may be different, the apparent beginning and ending of the melting of the crystalline portions of the fibers can be observed easily. The degree of decompo8ition eraused by heating the fibers in the presence of air is minimized by using an inert silicone oil. The second objection, that many fibers decompose below their melting points, is a valid one, but i t can be circumvented. Just as with low molecular weight compounds in which such decomposition occurs, the melting points of many synthetic fibers can be depressed below their decomposition temperatures by the addition of a suitable second component. The euteotic melting paint with this second component, as observed on a microscope hot stage, then becomes a8 useful for identification purposes a8 the melting point of the original fiber. A suitable second component for use as a reference compound with synthetic fibers is p-nitrophenol. Those fibers which neither melt nor show eutectic melting with p-nitrophenol exhibit characteristic, reasonably reproducible behavim patterns, which serve to distinguish them when they are heated and cooled on a microscope hot stage EXPERIMENTAL
Microscopic observations are made a t 100 to 2OOx using bath ordinary transmitted light and polarized light. A &&order red compensator is used to determine the sign of elongation, and is also sometimes helpful in observing the disappearance of birefringence which marks the final melting of the fibers. A commercially available Kofler hat stage was used for the present work, but other carefully calibrated commercial or homemade hot stages would work 8.9 well. I n such cases, however, temperatures different from those recorded in Table I might be obtained because of the characteristics of the hot stage used. Therefore, the preparation of a table matched to the hot stage in use is advisable. Melting points of synthetic fibers are obtained much more acourately and precisely if the fibers are heated in the absence of air. This is done most conveniently by immersing the fibers in silicone oil while determining their melting points. This procedure has the added advantage of the im roved image quality of a liquid mount over a %ry mount. The silicone oil appears to be chemically inert with respect t o the fibers (although a slight swelling OCOUIB in some cases) and ha8 a low enough volatility so that condensation in the hot stage is no problem. Dow Corning 710 fluid wa8 used in the present work. For the determination of the melting point and ohitritoteristio behavior of the fibers while they are being heated, a few filaments are placed on half slide, covered with a drop of silicone oil and a cover glass, and placed in the hot stage. Usually lengths of 1 to 2 mm. are most oon-
RESULTS
The results of examination of all common synthetic fibers are tabulated in Table I . The temperatures given for melting points of the fibers themselves are the ranges from the first sign of melting to the disappearanoe of the last trace of birefringence. B
A
”..”.
Before heating the stage, it is helpful t o note by observing the Becke line whether the refractive index itrallel to the fiber axis is greater or less than &at of the silicone oil. Commercial fibers which have a refractive index parallel to the fiber axis greater than that of the oil are Dacron, Nylon type 66, Nylon type 6, Kurslon, Verel, Vicars, and saran. In this manner a preliminary grouping of the fibers can be made. The stage is then heated a t a temperature rise ofabout3” C.perminute,andthefollowingpoints of behavior and correspondmg tempemtures are observed: (a) abrupt longitudinal shrinkage, ( b ) discoloration (due to decomposition), ( e ) change in polarization colors, ( d ) transverse swelhng which usually precedes 01 marks ( e ) the beginning of melting, and (f) end of melting
Figure 1. Eutectic melting of Nylon Type 6 with p-nitmphenol 0.
611.0’ C.
b. 69.5“ C.
C.
71.0’C.
d. 73.0°C.
ANALYTICAL CHEMISTRY
1588 Each of these temperature limits is reproducible to within about &2O C. Eutectic melting points are the temperatures a t which definite melting is first detected. Because of the greater difficulty in observing eutectic melting, these temperatures are given as ranges of 3" to 4" C. The temperatures a t which swelling and shrinkage take place are less reproducible than the melting temperatures, and all temperatures are dependent to some degree upon the heating rate, which must be carefully regulated. Polyethylene is not included in the table because i t is seldom encountered in fiber form except as large monofilaments. These and the bulk polymer, such as in molded articles, can be identified using this technique by their final melting point and the fact that, when melted n-ith p-nitrophenol, they form immiscible melts. The common type of polyethylene melts a t 110-14' C. Recently, homver, higher molecular weight materials made by the Ziegler-type process have become available R hich may have melting points as high as 135" C. DISCUSSION
Persons familiar with the use of fusion methods should esperience little difficulty in this application to synthetic fibers. For those with no prior experience in observing melting phenomena under the microscope, practice runs with known fibers are advisable before an unknonm is undertaken. For example, i t is sometimes difficult to distinguish betveen eutectic melting and shrinkage of the fibers, because the fibers when shrinking often drag along clumps of p-nitrophenol crystals, giving the impression of flow.
p-Nitrophenol sublimes above 80' C. when heated a t the prescribed rate. Above about 90' C. it sometimes condenses on the bottom surface of the cover glass in droplets of liquid which, if not recognized as such, may cause some confusion. Therefore, nith those fibers whose eutectic temperature lies above 90" C., indications of eutectic melting must be looked for in the fiber rather than in the p-nitrophenol. Table I shows that the technique can distinguish between most fibers conclusively. For example, the fibers n-hich melt by themselves can be immediately set apart from those which do not. Among those n-hich melt, only Xylan type 66 and Dacron cannot be unequivocally distinguished by their melting points alone, and with this pair of fibers the eutectic temperatures are adequate to tell them apart. In the group of fibers which do not melt by themselves there is only one pair of fibers which have nearly the same eutectic temperatures and behavior characteristics-namely, Orlon and Scrilan. I n this case, although their behavior is sufficiently characteristic for an experienced operator to distinguish one from the other, confirmatory tests are indicated. Rayon, as well as the natural fibers cotton and n-001, exhibits no visible interaction mith p-nitrophenol. The technique is applicable to staple and filament yarns and to bright and delustered fibers. ITith dyed fibers polarization colors are obscured, but othernise there is usually no interference. Bulk polymers and hlms-for example, Mylar film-may likewise be identified n i t h this method by using thin slices shaved off with a razor blade.
Table I. Micro Fusion Data for Synthetic Fibers Eutectic &felting Point,
Melting Point, Fiber
Behavior When Heated Alone
' c.
Originally fiber is nearly isotropic b u t shows definite ( f ) birefringence by 60-70' C. Polarization colors fade from about 290' C., then fiber melts abruptly. Fibers discolor. Polarization colors fade from about 240-50' C. and fiber begins t o discolor. Isotropic a t 260-5' C., melts a t 265-75' C. and is dark brown from decomposition. Does not melt when heated in air.
9t5-100
N o significant change before melting. Following first abrupt melting fibers go to uniform low order gray polarization colors until final melting. Shrinks and polarization colors fade at about 236' C. Becomes isotropic a t 240-5' C. Same a s Nylon 66.
109-12 71-5
c.
Arne1
297-300
Vicara
265-75
Dacron Nylon 66
252-70 261-64
Acetate
23545
Nylon 6
215-26
Saran
170-6
Shrinks noticeably a t about 160' C., and polarization colors change to low order gray ( - ) until melting occurs.
...
Dyne1
Acrilan
...
Shrinks slowly over 135' C. Polarization colors fade from about 175' C. Becomes isotropic a t 190' C . Sign of elongation changes from ( - ) t o isotropic a t 16580° C., then t o ( f ) a t 185-200' C. Reverts t o ( - ) when cooled. Same as Orlon.
85-9
Orlon
... ...
Kuralon
...
Verel
...
Darlan
...
...
Behavior When Heated with p-Nitropheno!
XOsignificant change before melting.
Swells noticeably above 90' C. Swelling increases when p-nitrophenol melts b u t does not dissolve. Sign of elongation changes from ( f ) t o isotropic a t 103-8' C., and t o ( - ) above 110150 Sharp and complete melting. Melting is sluggish, b u t complete before melting of p-nitrophenol. Noticeable swelling over 80' C. Tends to retain form when melted. Melting is sharp, and usually complete by 80-5' C. Only very slight shrinkage or swelling until about 140-50O C.. then decomposes with evolution of gas and becomes deep red in color. Fairly sharp melting, usually complete by about 95-100' C. Noticeable swelling above 95-100' C. Melting accompanied by much shrinkage. Melting slow to begin b u t is rapid by 108-10' C. Abrupt shrinking a t 105-10° C., but melting difficult to observe. Dissolves completely in melt of p-nitrophenol.
c.
Slight shrinkage a t 215' C. At 220-5' C., abrupt shrinking, swelling, and most fibers go isotropic or show very low order gray polarization colors. Slight birefringence in few filaments persists as high as 290-300' C. Negative sign of elongation. Polarization colors fade from 1 3 5 4 5 ' C. Isotropic by160-70' C. No definite melting but fibers appear t o fuse together a t around 2000 Negative sign of elongation. At 180-90' C. shrinks abruptly, most fibers go isotropic, and swell. All fibers isotropic by 205-10' C.
c.
88-92 69-73
103-6 106-9 105-10
92-5
No significant change before very sharp melting.
104-7
No significant change before sharp melting.
V O L U M E 2 8 , NO. 1 0 , O C T O B E R 1 9 5 6
It is somewhat surprising that eutectic melting takes place under the circumstances described, in view of the lack of an intimate mixture of the two components. However, the eutectic is very sharp in some cases, as shown by the photomicrographs of Figure 1. This series of pictures shoms unmistakable melting of Xylon type 6 with p-nitrophenol a t approximately 40’ C. below the melting point of p-nitrophenol and about 140’ C. below the melting point of Kylon type 6 . Figure 1, a, shows t x o filaments surrounded by fragments of p-nitrophenol crystals just before melting has begun. I n Figure 1, b, melting has begun as evidenced by the curved, slightly sir-ollen, and irregular outline of the filaments, and the rounding off and coalescing of pnitrophenol fragments. As the temperature continues to rise the melting proceeds faster and more obviously as shonm in Figure 1, c and d. If a t this point the slide is shifted to show a field containing an excess of p-nitlophenol, the fibers can be seen to dissolve completely a t temperatures well belorn7 the melting point of p-nitrophenol I t IS believed that the relatively high vapor pressure of the p-nitrophenol is responsible for the successful eutectic melting, the fibers being bathed with vapor of p-nitrophenol Tithin the confines of slide and cover glass. Of a
1589
number of compounds surveyed as possible reference reagents, only those which exhibited a strong tendency to sublime also exhibited eutectic melting. The results obtained also indicate the potential utility of fusion methods for other studies with synthetic fibers. For example, the change in the sign of elongation that occurs when Orlon is heated indicates structural changes in the nature of a second order transition. LITERATURE CITED (1) Am. Assoc. Textile Chemists Colorists, Tentative Test RIethod 20-53, “Identification and Quantitative Separation of Fibers.” (2) Am. SOC.Testing Materials, D 276-49, “Standard Method of
Identification of Fibers in Textiles.” (3) Kofler, L., Kofler, A., “hlikro-Methoden,” Universitatsverlag Wagner Ges. m.b.H., Innsbruck, 1948. (4) Luniak, B., “Identification of Textile Fibres,” Pitman & Sons, London, 1953. (5) hlcCrone, W. C., Mikrochemie tser. Mikrochim. Acta 38, 476 (1951). RECEIVED for review February 10, 1956. Accepted June 15, 1956. Presented a t Meeting-in-Miniature. Cleveland Section, ACS, February 15, 1956.
Microbiological Determination of Nitrate G. B. GARNER, J. S. BAUMSTARK, M. E. MUHRER, and W. H. PFANDER Departments of Agricultural Chemistry and Animal Husbandry, University of Missouri, Columbia, M o . Nitrate determinations have lacked specificity in many methods described in the literature. The specificity of the test reported is based on the reduction of nitrate to nitrite by a microbiologically produced nitrate reducLase. The method is simple and equipment is easily assembled. The nutritional requirements of the microorganisms are met by trypticase. Kitrate may be determined in the presence of many compounds known to interfere in other methods. The range of the test is from 2 to 20 y of potassium nitrate, with a reproducibility of 1 0 . 1 2 y.
The applicability of the enzymic reduction of nitrate for analytical use has been suggested in studies (9,11) on the purified enzyme “nitrate reductase.” Many organisms, such as E. coli, are capable of nitrate reduction and have been used for cell-free nitrate reductase preparations. This laboratory found i t more convenient to use a microorganism isolated from the rumen of a sheep to produce nitrate reductase in each tube than to attempt purification, preservation, and use of the enzyme. This method has been successfully employed in determining nitrate in silage, forage, hay, rumen fluid, and urine. APPARATUS
A
POSITIVE correlation between a qualitative test for nitrate and the toxicity of forages produced in drought areas (8) led t o a need for a rapid, quantitative method for determining nitrate. hlany methods described in the literature are based on determining nitrate nitrogen by difference or by extraction and nitration of organic compounds. The Devarda method ( 1 ) has been extensively used. It requires either a separation of the nitrate nitrogen from the total water-soluble nitrogen or a total nitrogen determination and a standard Kjeldahl determination. The actual distillation step in the Devarda method is rather difficult. The Robertson method ( d ) , like the Devarda method, is not applicable in the presence of cyanamide or urea. The limitation and precautions for these methods have been studied ( 5 ) . Kitration of organic compounds 2,4-xylenol ( 7 ) , and 3,4-xylenol such as phenoldisulfonic acid (4), ( 3 ) in the presence of sulfuric acid has been used. Without extreme precautions, results may be high in the presence of carbonaceous material, if the temperature becomes elevated, but this may be partially overcome by repeated extraction. A polarographic method for the determination of nitrate has been described (6) for concentrations in the range of 1.0 X 10-5 to 2.5 X 10-8 mole per liter. A method (IO) for reducing nitrate to nitrite uses zinc dust in an acid medium. 1-Saphthylaminesulfanilic acid is not so satisfactory as S(1-naphthyl)-ethylenediaminesulfanilic acid as a stable color-producing reagent (12).
The photometer was constructed with a photocell and highgain amplifier for use with narrow band filters. A set of matched test tubes, 13 X 100 mm., having a light path of 11 mm. was selected. The desired wave length of 5500 A. was approached by the use of Corning filters 3384, 5120, and 9780. REAGENTS
All reagents were prepared from analytical grade chemicals in double-distilled water. The color-developing reagent is similar to the reagent described by Saltzman ( I d ) . The reagents are stable for a t least 2 months if refrigerated. Stock Reagent Solution. To prepare 0.1 % 11’( 1-naphthyl)ethylenediamine dihydrochloride, dissolve 0.1 gram of the reagent in 100 ml. of water. Working Reagent Solution. Dissolve 5 grams of sulfanilic acid in 700 ml. of warm water, cool, and add 170 ml. of glacial acetic acid followed by 20 ml. of the above stock solution. Dilute to 1liter. Standard Potassium Nitrate Solution. Standard Stock Solution. Prepare the standard solution in such a manner as to give 0.2 mg. of potassium nitrate per ml. of solution. Working Standard Solution. Dilute 5 ml. of the stock solution to 100 ml. Triple-Strength Trypticase Medium. Dissolve 3 grams of trypticase and 1.5 grams of sodium chloride in 100 ml. of distilled water. Autoclave a t 15 pounds per square inch for 15 minutes. Trypticase may be obtained from the Baltimore Biological Laboratory, Inc. (BBL). 0.2M Phosphate Buffer. Dissolve 1.362 grams of potassium dihydrogen phosphate and 4.157 grams of disodium hydrogen phosphate in 500 ml. of water (pH 7 ) .
