542
INDUSTRIAL AND ENGINEERING CHEMISTRY
cellulose derivatives and blends. This would appear to be due to two factors: Others do not seem to have carried fractionation far enough to secure fractions of decidedly greater uniformity than the starting material; and they have placed most reliance on the stress-strain curves, for which it is necessary to run a great number of samples in order to show small differences. The work of Medvedev (22) would have been of more value if his fraction IVB had been compared with unfractionated and blended nitrocelluloses of the same viscosity, The writer had enough material available only for six checks on each point for the fold test, and for three to six checks on the stress-strain curves. It is doubtful if other investigators had samples as large as this. Differences were more pronounced for the unplasticized than for the plasticized films. This is believed to be generally true. There seems to be no commercial advantage in securing a greater degree of uniformity for products to be used in protective coatings and plastic masses. One property, adhesion, seems to be improved by blending high- and low-viscosity material. Certain of the blends lifted the surface of the glass on which films were dried in the same manner as gelatin.
Acknowledgment The author would like to express his appreciation to R. F. Kingery for his assistance in this investigation.
Literature Cited (1) Beck, A., ClBment, L., and RiviBre, C., Chimie et industrie, 24, 1068 (1930). ( 2 ) Berl, C., and Hefter, O., Cellulosechem., 14,65 (1933). (3) Breguet, M. A.,thesis, Lyon, 1924; Rev. gen. colloides, 3,200-6, 230-5 (1925). (4) Caille, A., Chimie et industrie, 19, 402 (1928).
VOL. 30, NO. 5
Ibid., 25, 276-85 (1931). Clark, J., and Miles, F. D., Trans. Faraday SOC.,27, 757-67
(1931). Deschiens, M., Chimie et industrie, 20, 1023 (1928). Duclaux, J., and Barbihre, J.. Bull. SOC. chim., 53,564 (1933). Duclaux, J., and Nodsu, R., Rev. gen. colloides, 7 , 241-50, 38592 (1929). Duclaux, J., and Wollman, E., Bull. S O C . chim., 27, 414 (1920). Elod, E., and Schmid-Bielenberg, H., 2.physik. Chem., B25,27 (1934). Glikman, S. A.,Plasticheskie Massy, No. 21 (1934); J . chim. phys., 31,458 (1934). Herz, W., Cellulosechem., 15, 95 (1934). Herzog, R. O., and Deripasko, A., Cellulosechem., 13,25 (1932). Iwasaki, S.,J. SOC. Chem. I n d . J a p a n , 34, Sugpl. Binding, 9 (1931). Kats, J. R., “Die Rontgenspektrographie aIs Untersuchungsmethode bei hochmolekularen Substanzen,” Berlin, Urban & Schwarzenberg, 1934. Kraemer, E. O., and Lansing, W. D., J. Phus. Chem., 39, 156 (1935). Kumichel, W., Kolloidchem. Beihefte, 26, 161-98 (1928). McBain, J. W., and Scott, D. A., IND. ENG.CHEM.,28, 473 (1936). MoNally, J. G., and Godbout, A. P., J. A m . Chem. SOC.,51, 3095 (1929). Mardles, E. W. J., S. Chem. Soc., 123, 1951 (1923). Medvedev, A. J., Kunststofle, 23, 249-51, 273-6 (1933); J . Applied Chem. (0.5. S.R.), 6, 880 (1933). Obogi, R., and Broda, E., Kolloid-Z., 69, 172 (1934). Okamura, I.,Cellulosechem., 14, 135-8 (1933). Rocha, H.J., Kolloidchem. Beihefte, 30,230 (1930). Bogovin, 2. A.,and Glazman, S., J. Applied Chem. (U. S. 9. R.) 8, 1237-49 (1935). Schneider, G., Canadian Patent 360,989 (Oct. 6, 1936). Sheppard, S. E., J. IND.ENG.CHEM.,13,1017 (1921). Ulmann, M.,Ber., 68, 134-45, 1217-24 (1935). RECEIVED January 14, 1938. Presented before the Division of Cellulose Chemistry at the 94th Meeting of the American Chemioal Society, Rooheeter, N. Y . , September 6 to 10, 1937.
Suitability of Plastics xor Airplane Dopes r e
GORDON M. KLINE AND C. G. MALMBERG National Bureau of Standards, Washington, D. C.
T
HE problem of producing a fire-resistant doped fabric to replace the hazardous cellulose nitrate product was investigated in 1933 a t the National Bureau of Standards for the National Advisory Committee for Aeronautics ( 2 ) . It was established that none of the synthetic resins then available would produce films of satisfactory tautness when used alone on the fabric, and that their addition to either cellulose nitrate or cellulose acetate dope resulted in a corresponding lowering of the fabric tension. Data were also obtained on the comparative rates of burning of fabrics doped with various cellulose nitrate and cellulose acetate compositions, both with and without the addition of fireretarding salts to the fabric. It was shown that cellulose acetate dope yields a relatively nonhazardous product when applied to untreated fabric and that, when applied to fabric treated with boric acid-borax mixture, a product which is nonflammable under ordinary conditions is obtained. However, the investigation was not continued a t that time to include a study of the effect of varying the ingredients of a cellulose acetate dope on the tautening characteristics and particularly on the moisture sensitivity of the doped fabric. Cellulose acetate has now become available commercially in several grades, representing materials of different degrees
of acetylation and viscosity. Likewise, other cellulose derivatives, possessing characteristics which indicate t h a t they might make satisfactory dope bases, are being manufactured. Among these more recently developed products are ethylcellulose, cellulose acetobutyrate, cellulose acetopropionate, and cellulose nitroacetate. These products are also available in various grades denoting differences in chemical composition and in viscosity. The proper formulation of dopes based on these new materials to obtain the requisite balance of solvents, diluents, and plasticizers, both as to types and amounts, can be accomplished only by a thorough laboratory study of the various factors involved. The Bureau of Aeronautics of the United States Navy Department has, therefore, requested the National Bureau of Standards to undertake such an experimental study in order to develop a dope, based on these comparatively nonflammable cellulose derivatives, which will compare favorably with or surpass cellulose nitrate dope with respect to the effect of high relative humidity on the tautness of the doped fabric. I n the first phase of this work a variety of cellulose derivatives and synthetic resins were applied to airplane fabric to study the relation of tautness to the type of plastic base used in the dope, the percentage of acyl or alkyl substitu-
MAY, 1938
INDUSTRIAL AND ENGINEERING CHEMISTRY
tion in the various cellulose derivatives, and the viscosity of these derivatives. (The term “viscosity” is used in this paper to denote certain flow characteristics of solutions of the various plastics as measured in the manner described in the footnote to Table I.) These panels are also providing data on the relative resistance to weathering of the various doped fabrics, while tests arc in progress to determine the effect of solvents, diluents, and plasticizers on the tautening property and to establish the moisture relations of the various doped fabrics.
Test Materials and Measurements PROPERTIES OF PLASTICS. The materials tested were furnished by various manufacturers whose interest and cooperation in this investigation have been very helpful. They also supplied data regarding the viscosity and chemical composition of the compounds shown in Table I. I n the
FIGURE 1. CONSTRUCTION OF EXPOSURE PANEL
543
A number of different cellulose derivatives of potential interest as film-forming constituents for airplane dopes have recently become available commercially. An investigation has been undertaken at the National Bureau of Standards to determine the fundamental factors involved in the formulation of dopes containing these new derivatives to obtain optimum tautness and durability. Data are presented in this paper relative to the effect of varying the acyl or ethoxyl content and the viscosity of cellulose esters and ethers on the tautness of fabrics doped with them. It is concluded that the solvents and diluents govern to a large extent the tautening properties of the dope and the durability of the film deposited on the fabric.
are shown in Table 11. Formula I was used in general for the cellulose esters and formula VI for the ethylcelluloses. The other formulas were introduced because of the special solubility characteristics of certain derivatives, such as cellulose triacetate, or to compare them with the solvent compositions recommended in the literature or by the manufacturer.
PREPARATION OF DOPED PANELS.Wooden frames (Figure l), manufacturers’ reports the chemical compositions were exconstructed in accordance with specification SP-16 of the Naval Aircraft Factory, were used in these tests. Cotton airplane pressed as percentages. The number of equivalents of each fabric, weighing approximately 4 ounces per square yard, was substitution group present in the compounds and also of placed over the frame, and a 50-pound weight was attached to free hydroxyl groups were calculated from these percentage each side. In this way uniform tension in both directions of the figures. The latter values are listed in Table I and indicate weave of the cloth was produced. The fabric was tacked to the frame with No. 4 upholsterer’s tacks, spaced approximately 1 inch immediately the degree of hydrolysis, which is closely related apart. Four coats of clear dope and, in the case of the pigmented to the solubility (1) and moisture-absorbing (4) properties panels, two coats of aluminum-pigmented dope were brushed on of the cellulose derivative. the panels. The average weight of the clear doped fabrics after COMPOSITION OF DOPES. Inasmuch as it was desired to drying was 7.2 ounces and of the pigmented doped fabrics, 8.7 ounces per square yard. compare the behavior of various cellulose derivatives with TESTING THE FABRICS. Tautness measurements were made on one another. the same solvent mixture and plasticizer, the fabrics conditioned a t antriphenyl phosphate, were proximately 70” F. and 65 pkr employed in a majority of cent relative humidity. The the dopes. I n most cases measurements were made with the s p r i n g -lo a de d tautness the proportions of the latter meter (Fi re 2) described elsewere 1 part of plasticizer where ( S y This instrument to 9 parts of cellulose demeasures the d e f l e c t i o n in thousandths of an inch (mils) rivative. Preliminary tests of the fabric under load. The indicated that varying the less the deflection under a given concentration of the cellulose load, the greater is the tautbase in the solvent mixture ness. For doped fabric the between the limits of 5 and loadused is 1 pound, but for the cloth alone a 2-ounce load10 p e r c e n t h a d comparaing is employed t o prevent extively little effect on the cessive s t r e t c h i n g . The tautness produced. A conaverage deflection observed for centration of about 6.4 per the undoped fabrics with the 2-ounce load was 45 mils, with cent was therefore used for an average variation of *2.5 most samples, a l t h o u g h it and a maximum difference of was necessary to dilute some + 6 mils. A 1 - p o u n d l o a d of the aluminum-pigmented deflects such fabrics approximately 260 mils. dopes to obtain a solution After drying for at least one of s u i t a b l e v i s c o s i t y for week in the laboratory, the brushing. C o m p l e t e data panels were placed in a room on the formulas of the dopes kevt at 70” F. and 65 ver cent FIGURE 2. SPRING-LOADED TAUTNESS METER
INDUSTRIAL AND ENGINEERING CHEMISTRY
544
VOL. 30, NO. 5
TABLEI. VISCOSITY AND COMPOSITION OF TEST MATERIALS Material Cellulose acetate
Sample
No. C A-a 1
CA-a2 CA-a3 CA-a4 CA-a5 CA-a6 CA-a7 CA-a8 CA-a9 CA-a10 CA-bl CA-b2 CA-b3 CA-b4
Viscositya 9 . 4 sec. 6.2 10.5 28 44 56 81 83 94
-
--
Percentage basis3 9 . 1 acetyl ...... 40.0 40.9 ...... 40.4 40.0 ...... ...... 38.4 ...... 38.7 ...... 38.4 ...... 39.1
......
......
39.0 40.3 39.2 38.1
‘2 33 64
77
...... ...... ...... ......
Composition 2 . 3 8 acetyl 2.48 2.57 2.51 2.48 2.32 2.35 2.32 2.38 Ca. 3 . 0 2.37 2.51 2.39 2.29
CAP-a1 CAP-bl CAP-b2 CAP-b3 CAP-cl CAP-cZ CAP-c3
31 sec. 3 4 . 5 cp.6 470 cp. 650 cp. 35 cp. 42 cp. 90 cp.
1 7 . 1 acetyl 15.5 15.5 15.5 14 11 9
26.9 propionyl 31.3 31.3 31.3 34 34 33
0.74 0.58
CAB-bl CAB-b2 CAB-b3
36 315 665
3 2 , s acetyl 32.2 32.0
1 5 . 4 butyryl 15.2 16.1
2 . 3 1 acetyl 2.27 2.28
Cellulose nitroacetate
CNA-bl
Medium
Cellulose nitrate
CN-cl CN-c2 C N-c3
0 . 5 sec. 1 0 , 5 sec. 84 sec.
....
Methylcellulose
MC-bl
Low
....
Ethylcellulose
EC-bl EC-b2 EC-b3 EC-dl EC-d2 EC-d3 EC-d4 EC-d5 EC-d6
6 . 2 sec. 22 sec. 66 sec. 20 cp. 30 op. 35 cp. 75 CP. 150 cp. 210 cp.
47.5 ethoxyl 47.7 48.2 48.3 52.0 48.7 48.0 48.3 49.5
BC-bl CR-bl MM-cl
Medium 125 cp.
....
Cellulose acetopropionate
Cellulose acetobutyrate
Beni ylcellulose Chlorinated rubber Methyl methacrylate
cp. cp. CP.
1 . 1 3 acetyl 1.08 1.08 1.08 1.00
>
Equivalents basis 0.62 hydroxyl 0.52 0.43 0.49 0.52 0.68 0.65 0.62
...... ...... ......
0.63 0.49 0.61 0.71
...... ...... ...... ......
0.68
...
1 . 3 4 propionyl 1.64 1.64 1.64 1.83 1.72 1.60 0.66 butyryl
0.65
0.69
....
....
...... ...... ......
...... ...... ...... ... ...... ......
...... ...... ...... ...... ......
......
0 . 5 3 hydroxyl 0.28 0.28 0.28 0.17 0.54 0.82 0 . 0 3 hydroxyl
0.08
0.03
...... ......
...... ...... ......
....
......
......
2.43 ethoxyl 2.44 2.48 2.49 2.77 2.52 2.47 2.49 2.58
0.57 hydroxyl
...... ...... ...... ......
.... ....
...... .. .. ..
......
0.... .56
0.52 0.51 0.23 0.48 0.53 0.51 0.42
...... ...... ...... ......
...
The viscosity values were obtained in the manufacturers’ laboratories b y the following variety of methods (all parts by weight unless otherwise nobed). Cellulose acetate series CA-a: A. 9. T. M. falling ball method using 20% solutions in acetone. (A. S. T. M. Standards, P a r t 11, Non-Metallic ’ Materials 1936.) Cellulosd acetate sefies CA-b: A. S. T. M. falling ball method using 20% solutions in a mixture of 90 parts acetone and 10 parts ethyl alcohol. Cellulose acetoproplonate series CAP-a: A. 9. T. M. falling ball method using 20% solutions i n acetone. Cellulose acetopropionate series CAP-b a n d CAP-c, and cellulose acetobutyrate series CAB-b: Capillary visoometer using 10% solution in acetone a t 250 C . Cellulose nitrate series CN-c: A. S. T. M.falling ball method. Ethylcellulose series EC-b: A. S. T. M. falling ball method using a 20% solution in a mixture of 80 parts toluene and 20 parts ethyl alcohol Ethylcellulose series EC-d: Capillary viscometer method using a 5% solution in a mixture of 80 parts toluene and 20 parts absolute ethyl alodhol by volume a t 25’ C. ChIorinated,rubber CR-b: Capillary viscometer using a 20% solution in toluene a t 25’ C. h Cp. = centipoises.
a
relative humidity. Three measurements of tautness were made in this room at intervals during a period of one week. The panels were placed upon the roof of the Industrial Building of the National Bureau of Standards on August 14, 1937, on racks inclined at an angle of 45” and facing south. Measurements of tautness and brittleness were made at intervals under prevailing conditions of temperature and humidity. Brittleness of the film-i. e., tendency to “ringworm”-was determined by pressing the thumb firmly into the fabric covering at each corner of the panel. Data obtained on the doped panels in the conditioning room and on the roof are presented in Table 111.
Effect of Various Factors on Initial Tautness TYPEOF PLASTIC.The tautness values of the fabrics doped with the different plastic materials, measured in the conditioning room a t 70” F. and 65 per cent relative humidity, are summarized in Table IV. Most of the cellulose plastics produced tautness values ranging between 82 and 97. Cellulose triacetate and the cellulose acetobutyrate, which was also practically a triester, had average tautness values of 60 and 77, respectively. Great difficulty was experienced in obtaining a taut fabric with benzylcellulose. Plasticizer was omitted from most of the ethylcellulose dopes because of the comparatively poor tautening qualities of this type of cellulose derivative and its reported characteristic of form-
ing flexible films without the plasticizer. However, it was necessary to add plasticizer to the high-ethoxyl compound (panels 74 and 75) because the unplasticSed films of this material ringwormed spontaneously one day after doping. The resinous materials, such as chlorinated rubber and methyl methacrylate polymer, did not tighten the fabric as well as did most of the cellulosic materials. Solutions of butyl methacrylate and methyl Cellosolve methacrylate polymers failed to tighten the fabric when applied in the same solvent mixture used for the methyl derivative. ACYLOR ALKYLSUBSTITUTION.The data in Table I11 indicate that varying the acetyl content of cellulose acetate between the limits of 2.29 and 2.57 equivalents did not have an apparent effect on the ability of the compound to tighten the fabric. Likewise the results obtained with ethylcellulose containing from 2.43 to 2.77 ethoxyl equivalents did not show a correlation between ethoxyl content and tautening property. There was not sufficient variation in the acyl composition of the cellulose acetopropionates and cellulose acetobutyrates to indicate its effect on tautness. Tests to determine this relation are now underway on materials of this type having a greater variation in acyl content. VISCOSITY. The absence of any effect of the size of the cellulose molecules, as indicated by the viscosity of solutions of these materials, on the tautening property is shown by the
MAY, 1938
INDUSTRIAL AND ENGINEERING CHEMISTRY
543
TABLB 11. FORMULAS OF EXPERIMENTAL DOPES Formula
I clear
Plastic Baae
Plasticizer
%
%
Aluminum Powder Solvent
%
%
6.4
Composition of Solvent Parts by weight Acetone 48, ethyl acetate 20, methyl ethyl ketone 20, diacetone alcohol 5 Acetone 50, ethyl acetate 15, methyl ethyl ketone 15, diacetone alcohol 8
I1 pigmented IIA IIB
0 . 6 triphenyl phosphate None 93.0 1 . 2 triphenyl phosphate 4 8 88.0 I with plasticizer omitted 0 . 6 triphenyl phosphate None 93.0 Acetone 50, ethyl alcohol 10, toluene 40 6.0 1 . 2 triphenyl phosphate 4.8 88.0 Same as I1 clear Same as I1 with only 5 % plasticizer (based on nonvolatiles) in both clear and pigmented dopes Same as I1 with plasticizer omitted
I11 clear I11 pigmented
10.4 6.0
1 . 0 benzyl alcohol 0 . 6 benzyl alcohol
None 4.8
88.6 88.6
Acetone 16.5, ethyl alcohol 8.5, benzene 36.6, toluene 18.0, xylene 9.0 Same as I11 d e a r
IV clear IV pigmented
6.4 6.0
0.6 benzyl alcohol
None 4.8
93.0 88.6
Same as I11 clear Same as 111 clear
VI clear VI igmented clear VIA pigmented
VIAP
6.4 6.0 6.4 6.0
0 . 6 Dow No. 6
None None
1 . 2 Dow No. 6
None 4.8 None 4.8
93.6 89.2 93.0 88.0
Acetone 25, Same as VI Same as VI Same as VI
VI1 clear
6.4
0 . 6 triphenyl phosphate
None
93.0
4.8
88.0
Ethyl acetate 25, methyl ethyl ketone 25, butyl acetate 25, m e t h y l Cellosolve 25 Same as VI1 clear
6.0 ifgmented Same I1 clear 6.4
VI1 pigmented VIIA
as
0 . 6 benzyl alcohol
1 . 2 triphenyl phosphate 6.0 Same as VI1 with plasticizer omitted
ethyl acetate 20, toluene 45, diacetone alcohol 10 clear clear clear
VI11 clear VI11 pigmented
6.4 6.0
0 . 6 triphenyl phosphate 1 . 2 triphenyl phosphate
None 4.8
93.0 88.0
Methyl Cellosolve Methyl Cellosolve
I X clear I X pigmented
3.8 2.4
0 4 triphenyl phospliate 0 . 5 triphenyl phosphate
None 1.9
95.8 95.2
Chloroform 70, tetrachlorethane 30 Same as I X clear
X clear X pigmented
7.7 6.0
None None
None 4.8
92.3 89.2
Toluene Toluene
X I clear X I pigmented
7.6 6.0
None None
None 4.8
92.4 89.2
Toluene 33.1 xylene 22.3 Cellosolve 22.3 methyl amyl ketone 22.3 Toluene 26.8: xylene 24.4: Cellosolve 24.4: methyl amyl ketone 24.4
X I 1 clear XI1 pigmented
4.5 4.5
None None
None 3.6
95.5 91.9
Water Water
data in Table 111. Cellulose acetates varying in viscosity, as measured by the falling ball test, from 2 to 94 seconds did not differ markedly in their tautening ability. The same was true for ethylcelluloses whose 5 per cent solutions in a mixture of 80 parts of toluene and 20 parts of ethyl alcohol gave viscosity values of 20 to 210 centipoises. SOLVENTCOMPOSITION.The results obtained with the different solvent formulas tested in this investigation lead to the rather startling conclusion that the most important single factor involved in the tautening property of a dope is the solvent composition employed. I n the term “solvent composition” are included those organic liquids that form colloidal solutions of the cellulose derivatives, those that have only cosolvent and swelling action, and those that are merely diluents in the blend. It has been the usual practice to formulate the solvent composition on the basis of such requirements as drying time, antiblushing characteristics, solvent property, and economy, without an adequate appreciahion that the choice of the solvent composition governs not only the tautening property of the dope but also the flexibility (conversely the brittleness) of the film deposited on the fabric. Table V presents the data obtained on the effect of various solvent compositions on the tautness of the doped fabric. The difference in the .tautness values of the two cellulose nitrate panels is much greater than the tautness differences shown in Table IV for the various types of cellulose derivatives or f or compounds varying only in viscosity and percentage substitution. Since these panels were prepared, the effect of various solvents and mixtures of solvents and nonsolvents on the shrinkage and flexibility of films of cellulose derivatives has been studied in detail. The solutions are poured into glass Petri dishes with and without a cellophane lining. The film in contact with the glass provides information on the adhesive property of the composition and is also used to follow the rate of drying. The film in contact with cellophane is used to obtain an estimate of the amount of shrinking characteristic of the dope; in only a few instances has tenacious adherence of the film to cellophane been noted.
These tests are still in progress, but they indicate that in order to obtain a maximum tautening effect it is necessary to formulate a dope so that a minimum of active solvent will be present during the final drying stage. Some solvent action is necessary in this stage to prevent precipitation (blushing) of the cellulose derivative. The selection of this solvent is an important factor in avoiding the formation of a brittle film, which Sheppard and Newsome (6) found to be associated with the presence of large crystallites. Thus, cellulose acetate dissolved in methyl Cellosolve gave the maximum tautness for this cellulose derivative (Table V), but Table I11 shows that the film is brittle and ringworms after a few days of exposure on the roof. Likewise, cellulose acetobutyrate in formula VI1 (panel 5 2 ) , which contains 25 per cent methyl Cellosolve, tautens well but yields a brittle film. However, although the tests are not yet completed, present indications are that it will be possible to formulate dopes with these cellulose derivatives which will tauten the fabric satisfactorily without yielding brittle films.
Effect of Weathering on Tautness of Doped Fabrics I n order to obtain comparative information on the effect of variations in temperature and humidity on the tautness of fabrics doped with various cellulose derivatives, it would be desirable to have all the panels a t about the same initial tautness. This was not achieved with the panel tests reported in Table 111. However, the tautness values obtained with the cellulose compounds, using the solvent blend listed as formula VII, were more nearly alike than with any other composition. The variation in the tautness values obtained for these panels under differentweather conditions is shown in Figure 3. The cellulose acetate panels were not markedly affected by the heat of the sun but did slacken during periods of rain to a greater extent than any of the other cellulose derivatives. T h e cellulose acetopropionate panels also showed marked slackening in rainy weather after 3 months of exposure on the roof. The cellulose acetobutyrate and
546
INDUSTRIAL AND ENGINEERING CHEMISTRY
p 4 e
:. :. :. :. : . .: :. :.
VOL. 30, NO. 5
MAY, 1938
INDUSTRIAL AND ENGINEERING CHEMISTRY
:ggge 3-31
547
:. :. :. : . : . .: .: :.
gg2zg e :. : . :. .: .: .: :. i 3 1 1 3
gGggg a : .: :. :. :. .: .: :. 133-3
........ . . . . . . .. .. .. .. .. ..
$%L : :
g25ga:
. . . . . .: :. :. :::::
.. .. ..
.. .. ..
.. .. .. +%
.. .. ..
548
INDUSTRIAL AND ENGINEERING CHEMISTRY
VOL. 30, NO. 5
f o r m u 1a t i o n which yields a flexible and durable film with one derivative CLEAR RAIN may produce a brittle and unstable CLEPiR film with another cellulose compound CLOUDY CLEAR because of differences in their solubility. However, certain trends can RAIN be observed in the data in Table 111. CLEAR The c e l l u l o s e a c e t a t e films are, in C L O u ~ ~ general, very stable to sunlight. T h e RAIN only panels which gave evidence of embrittlement were three covered with clear dope made from low-viscosity cellulose acetate (panels 1, 5, and 25), 80 100 80 100 80 100 80 100 12D one panel covered with clear cellulose 60 80 100 80 100 triacetate dope (23), and one panel DEFLECTION I N MILS covered with aluminum-pigmented dope IN TAUTNESS OF DOPED FABRICS EXPOSED ON THE ROOF FIGURE 3. VARIATION made by dissolving cellulose acetate of Formula VI1 was used for all of the dopes. S indicates slackness of t h e fabric. Weather conditions medium viscosityin methyl c~~~~~~ on the days when measurements were made were as follows: Days on Temp., Relative Days on Temp., Re!ative Days on Temp., Relative (14), in which case the embrittlement Roof F. Humidity, 7% Roof ' F. Humidity, % Roof F. Humidity,% is directly attributable to the solvent 2 90 40 17 92 52 78 54 41 employed. The clear dope films made 5 94 52 24 72 74 93 51 43 63 97 31 74 35 121 105 54 30 96 56 with cellulose acetopropionates of vari9 98 56 52 63 10 64 ous viscosities failed on five of the seven 11 65 99 67 50 90 128 43 43 13 86 62 panels exposed (36, 38, 42, 44, and 46). One panel covered with cellulose acetopropionate of low viscosity and containing aluminum pigment TABLEIV. SUMMARY OF TAUTENING PROPERTIES OF VARIOUS (43) failed in 17 days on the roof. However, a sample of TYPES O F h A s T I C s USED IN DOPES medium-viscosity cellulose acetopropionate of more recent PIGMENT manufacture than the rest showed good stability (33, 34, and Av. DeSamples No. Formula under flection 135). The low-viscosity cellulose acetobutyrate dopes, both Material Tested Used Lb. Load Variation clear and pigmented (48 and 49), failed in 17 days on the roof, Mils Mzls as did also a panel covered with medium-viscosity cellulose Cellulose acetate (hydro13 I 86 82-90 acetobutyrate which was applied in a solvent mixture conlysed) 1 I X 60 Cellulose triacetate taining 25 per cent methyl Cellosolve (52). Cracking of the I 91 89192" Cellulose acetopropionate 7 3 Cellulose acetobutyrate I 77 83 76-79 films of clear cellulose nitroacetate (55) and cellulose nitrate 1 Cellulose nitroacetate 3 I 90 86195 (57, 59, and 62) was observed within 1 to 2 months of exCellulose nitrate XI1 97 1 Methylcellulose posure. All twelve panels covered with clear ethylcellulose I1 88 83195 8 Ethylcellulose Most panels 111 103 1 Benz yloellulose dopes (66, 68, 70, 72, 74, 76, 78, 80, 82, 84, 86, and 89) failed slack 1 X 99 ... after relatively short exposure to the weather. Two panels Chlorinated rubber XI 97 ... Methyl methacrylate resin 1 covered with aluminum-pigmented dopes consisting of low0 One panel read 76. viscosity ethylcelluloses dissolved in a mixture of acetone, ethyl acetate, toluene, and diacetone alcohol (73 and 75) failed in less than 2 months, whereas panels prepared with cellulose nitrate panels behaved about the same; they below-, medium-, and high-viscosity ethylcelluloses dissolved came less taut when removed from the conditioning room in this same formula (67, 69, 71, 77, 79, 87, and 90) had not into the sunlight and, in general, showed a slight additional cracked after 3-month exposure. On the other hand, three decrease in tautness in rainy weather. The ethylcellulose panels Of pigmented medium-viscosity ethylcellulose and benzy~ce~lu~ose panels were considerably poorer in initial tautness than those covered with the cellulose esters. Both of the cellulose ethers underwent a decrease in tautness in sunlight. During periods of rain the ethylcellulose panels showed additional slackening, whereas the benzylcellulose PIGMENT panels had a tendency to tighten. Cellulose Cellulose 1% should be emphasized that the behavior of these cellulose Cellulose AcetoAcetoCellulose EthylWEATHER
~
films under various weather conditions with respect to both tautness and moisture absorption is dependent to a large extent upon the type and amount of plasticizer employed. For the above tests triphenyl phosphate, present to the extent of 10 per cent of the total nonvolatiles, was used throughout. I n the course of the phase of this investigation dealing with the effect of plasticizers on tautness, both initially and upon exposure, compositions should become available which will have greateruniformityof tension under varying weather conditions. ~
f
of Weathering f ~ ~ on Flexibility ~ of D~~~ ~
i
The results of exposure tests on panels doped with the different cellulose derivatives (Table 111) must be considered as only tentative evidence of their relative stability. A
~
~
Blend No, I &,V$Cello-
Acetate 84 72
solve
67
2;:
..
..
,
propionate 92 79
..
.. ..
"
T
butyrate 76 65
Nitrate 86 61
.. .. ..
..
.. ..
N
oellulose
s5
94
86
in a mixture of toluene, ethyl alcohol, and acetone (81, 83, and 85) failed in 17 days; this particular formula, then, apparently tends to lay down a film which is inherently brittle. The clear benzylcellulose film (91) cracked and yellowed soon after exposure. The pigmented benzylcellulose films l ~ (92, 93, and 94) showed evidences of cracking after about 2 months of exposure, although the fabric was not tautened sufficiently for the typical ringworm type of failure to occur. The panels covered with methylcellulose (64 and 65) showed
MAY, 1938
INDUSTRIAL AND ENGINEERING CHEMISTRY
remarkable tautness during the first week of exposure, but the methylcellulose was completely removed by the first rain. This action was, of course, anticipated, but the rapidity with which the aluminum-pigmented methylcellulose film was dissolved by the rain was rather surprising in view of the comparatively slow rate of solution of methylcellulose in the laboratory a t ordinary temperature. The film of clear chlorinated rubber of low viscosity (95) yellowed and disintegrated rapidly upon exposure to sunlight. The pigmented film (96) cracked in a manner similar to the benzylcellulose films after about 2-month exposure. The pigmented methyl methacrylate film (97) did not show any evidences of cracking after 3 months on the roof.
Summary and Conclusions 1. The tautnesses of airplane fabric doped with various plastics dissolved in a variety of solvent mixtures were determined. It was observed that the most important single factor involved in the initial tautening property of a dope is the solvent composition. I n order to obtain a maximum tautening effect, it is necessary t o formulate a dope so that a minimum of active solvent will be present during the final drying stage. The selection of this solvent is also an important factor in avoiding the formation of a film which is initially brittle or which rapidly becomes brittle upon exposure out of doors. 2. The highest initial tautness values were obtained with cellulose triesters, such as cellulose triacetate and a practically completely acylated cellulose acetobutyrate. Varying the acyl or ethoxyl content of partially hydrolyzed cellulose derivatives did not have a pronounced effect on the ability of the compounds to tighten the fabric. The initial tautening property is also apparently independent of the size of the cellulose molecule as indicated by certain flow characteristics of solutions of these materials. The tests for the majority of the cellulose esters were made with films containing 10 per cent of triphenyl phosphate. 3. I n exposure tests the cellulose acetobutyrate and cellulose nitrate panels behaved quitesimilarly,slackening somewhat when removed from the conditioning room into the sunlight
549
and, in general, showing a slight additional decrease in tautness in rainy weather. The cellulose acetate panels slackened during periods of rain to a greater extent than any of the other derivatives. The cellulose acetopropionate panels also showed marked slackening in rainy weather after 3 months of exposure on the roof. The ethylcellulose and benzylcellulose panels were considerably poorer in initial tautness and slackened still further upon exposure. However, the results of these exposure tests must be considered as only exploratory and preliminary to the testing of dopes formulated to develop optimum tautness, flexibility, and moisture resistance. Before dopes having these characteristics can be formulated, it will be necessary to obtain detailed information on the effect of various solvents, diluents, and plasticizers on the properties of the film-forming plastics.
Acknowledgment This investigation was sponsored by the Bureau of Aeronautics, United States Navy Department, and the results are published by permission of the chief of that bureau. The authors wish to express their appreciation of the interest and suggestions during the course of this work of Lieutenant Commander C. F. Cotton and J. E. Sullivan of the Bureau of Aeronautics, and of C. W. Woodrow of the Naval Aircraft Factory.
Literature Cited (1) Coltof, W., J. SOC.Chem. Ind., 56, 363-75T (1937). (2) Kline, G. M., J. Research Natl. Bur. Standards, 14, 575-87 (1935) ; Research Paper 788. (3) Kline, G. M., and Malmberg, C. G., Ibid., 20 (May, 1938). (4) Sheppard, S. E., and Newsome, P. T., S.Phys. Chem., 33,1817-35 (1929). Chem. Ind., 56, (5) Sheppard, S. E., and Newsome, P. T., J. SOC. 256-61T (1937). RECEIVEDJanuary 28, 1938. Presented before the Division of Paint and Varnish Chemistry a t the 94th Meeting of the American Chemical Society, Rochester, N. Y., September 6 t o 10, 1937. Publication approved by the Director of the National Bureau of Standards, U. 9. Department of Commerce.
INTHE SEMICURED, SEMIHARDENED STATE,SIMILAR TO RUBBER, CATALINMAY BE REMOVED FROM SLICEDON THIS MACHISE WHENUNUSUALTHICKNESSES ARE REQUIRED
THE
OVEX AXD
