J a n .. 1914
T H E JOCR.VAL O F I S D C S T R I A L d - V D E N G I N E E RI N G CH E M I S T R Y
Percentage of total export of product from Germany 20.7
59.5 54.4
58.0 26.0 18.9 16.5 31.4 42.1 9.3 23.5 14.0 24.0 55.8 20.0 11.6 27.7 24.0 6.5 15.1
6.8 11.7 39.1 24.4 31.4 11.7 11.6
9.0
12.1 14.3 7.0
7.7 10.8 4.1
L T . $.
I M P O R T S F R O M GERM.4r;Y
Value in L. . S . money Dollars
34.4i8,OOO 4.358. 000 Anilin and other dyes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.253. 000 4.001. 000 3.618. 000 2.83;. 000 Potassium chloride. . . . . . . . . . . . . . . . . . . . . . . 2 . 8 2 0 . 000 2.680. 000 “Abraum” salts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 .683. 000 1.678. 000 1.359. 000 1 .164. 0 0 0 Sulfates of magnesium and potassium. . . . . . . . . . . . . . . . . . . . . 919. 000 Alizarin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Anilin oil and salt . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .904. 000 836. 000 Indigo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .823. 000 730. 000 Essential o i l s . , . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 516. 000 424. 000 Potassium bicarbonate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 381. 000 Palm oil. cocoanut oil and vegetable fats . . . . . . . . . . . . . . . . . . 374. 000 326. 000 Quinine and its salts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 319. 000 Alkaloids (exclusive of quinine). antipyrine and antifebrine . 288. 000 Bronze and chrome colors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 262. 000 . . . . . 226. 000 . . . . 205. 000 Potassium cyanide 195. 000 Lead pencils . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161. 000 Bleaching powder . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178. 000 167. 000 Carbolic acid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126. 000 Zinc ashes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .124. 000 Crude medicinals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119. 000 119. 000 Chemical preparations for medicinal use . . . . . . . . . . . . . . . . . . . 112. 000 Quebracho extract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .112. 000 Barium salts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112. 000 Gelatine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105. 000 T a r t a r emetic and antimony preparations. . . . . . . . . . . . . . . . . 105. 000 102. 000 Salt . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 i . 000 Ozokerite. purified . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83. 000 8 1 . 000 i 4 . 000 Glue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .67. 000 64. 000 64. 000 62. 000 53.000 Matches e t c . , c . n . s . p . f . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ammonia. ammonium chloride and carbonate . . . . . . . . . . . . . . 55. 000 Lithopone. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33. 000 Resins n . s . p . f . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33. 000 Lake colors . . . . . . . . . . . ......................... 31.000 Ultramarine . . . . . . . . . . ......................... 24. 000
u . s. EXPORTS TO
I
Percentage of total import of product into Germany
GERMANY
Refined petroleum . . . . . . . . . . . . . . . . . . . . .
. . . . . . .6 4 . 3
Oleomargarine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Turpentine and other rosin oils . . . . . . . . . . . . . . . . . . . . . . hate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
88.0 80.0 65.0
...................................
79.4
Cotton seed oil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
82 . 7
Lubricating oils . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Beef and mutton tallow . . . . . . . . . . . . . . . . . . . . . . . . . .
3 ; .4 43.9
Stearic and palmitic a c i d s . , . . . . . . . . . . . . . . . . . . . . . . . .
56.7
.4cetate of lime . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Y j .9
Crude oil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
23.5
....................................
40.2
Wood alcohol
Oleic acid. etc
..
16.0
Ethereal oils n . s . p . f
..
11.22
Oils for industrial uses . . . . . . . . .
18. 5
Resins n . s . p . f . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.6
Gums and varnishes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Albuminous material . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1 7 .0 14.7
Dyewood extracts. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 3 . 5 4.0 Earths n . s . p . f . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Zinc ashes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0.4
TOTAL. $38.194. 240
TOTAL, $16,993, 200
total coal-tar dye receipts in t h e U . S. from Germany . T h e acetate of lime account practically balances t h e indigo account . Out of t h e 2 1 classes of chemical products which Germany imported from t h e United States in 1 9 0 4 ~7 of t h e m each made up 6 5 per cent a n d more of Germany’s total importations of those articles; these seven are acetate of lime. oleomarga-
3
rine, refined petroleum. cotton seed oil. turpentine a n d other rosin oils. rosin a n d rock phosphate . Figures later t h a n 1904 are not conveniently available; there is no reason t o suppose t h a t there has as yet. been a great shifting if any. of t h e relative positions of a n y of t h e items herein involved . BERNHARD C . HESSE
.
.
ORIGINAL PAPERS SYNTHETIC RESINS By L
\
R ~ D W 4 ~ NJ
\iEITlf
Rweived October 30 COitde?tSUtiOii
F P BROCK 1913
AND
Y r O d u c t S Oj Phenolic B o d i e s W i t h H e x a met hyleii c T e t v a m i w
T h e condensation products or synthetic resins which
occur when phenolic bodies are heated in a water solution of formaldehyde. their polymers or eyui\-alents are already well known from scientific a n d patent literature. T h e t e r m ‘‘phenolic bodies ” signifies a n y substance containing a benzene nucleus a n d h a \ ing a hydroxyl attached t o t h e ring . e . g., t h e creiols .
4
T H E J O U R N A L OF I N D U S T R I A L A N D E N G I N E E R I N G C H E M I S T R Y
naphthols, thymol, carvacrol, etc., t h e chlor- and bromphenols, nitro-phenols, phenol sulfonic acid, etc. T h e t e r m formaldehyde includes its hydrates or polymers a n d may be replaced in t h e reaction b y acetaldehyde, benzaldehyde, etc., in certain cases, b u t t h e formaldehyde is in every case more reactive t h a n t h e substituted aldehyde, or t h e aldehydes higher in t h e series. The formaldehyde-phenol reaction products’ are formed in every case with t h e elimination of water as a by-product. T h e first step in t h e reaction goes according t h e t h e following equation: C6H50H CHzO ----f HOCH2.CsH40H forming oxybenzyl alcohol. T h e second change may be represented by t h e equation: 2HOCHz.CeHdOH HOCH, . CeH4. OCH,. C 6 H 4 0H HZO in which two molecules of oxybenzyl alcohol unite with t h e elimination of water and t h e formation of saligeno-saligenin. Further reaction may occur in which t h e saligenin molecules unite, forming a saliretin product with t h e further elimination of water. T h e water of reaction is not t h e only water present, since commercial formaldehyde is 60 per cent water a n d in t h e process of formation, this water separates out from t h e newly formed resin. T h e wet process for obtaining t h e synthetic resins has been exploited successfully, commercially, by a number of research chemists who have patented t h e results of their researches. Difficulty is experienced in following t h e r a t e of this wet reaction for, heretofore, phenol has beenzdifficult t o determine in t h e presence of formaldehyde a n d formaldehyde requires a considerable length of time (about 48 hours) for each determination, t h e results being unreliable within 2 or 3 per cent. Consequently, great difficulty is experienced in following t h e progress of t h e condensation. An a t t e m p t has been made b y Jablonowerl t o follow t h e velocity of this reaction in an open system b y measuring t h e r a t e of change in t h e specific gravity of t h e reacting mixtures. T h e reaction between a phenolic body a n d formaldehyde or a polymer is only a p a r t of t h e more general reaction which takes place between a simple or subs t i t u t e d mobile methylene group a n d a n y substance containing in its molecule a benzene nucleus t o which a hydroxyl is generally attached. T h e hydrogen of t h e active methylene group may be substituted b y t h e alkyl radicals, giving t h e higher aldehydes of t h e f a t t y series or by a benzene nucleus giving t h e aromatic aldehydes. The oxygen, t o which t h e active methylene group is attached, i n aldehydes, m a y be replaced b y sulfur or nitrogen. With sulfur t h e malodorous thio-aldehydes are formed; t h e nitrogen forms with active methylene groups hexamethylene tetramine, hydro-benzamide, etc. T h e nitrogen compounds are Colorless, odorless, transparent a n d crystalline, easily soluble i n water a n d sparingly soluble in alcohol. They sublime without -melting and are stable products; when boiled in dilute acids t h e y break
+
*
j
J . A m . Chem. SOC.,S I , 81 1 (1913).
+
Vol. 6, No.
I
down into ammonia a n d t h e corresponding aldehydes; boiling in alkali has no effect upon them. Hereafter, t h e word phenol will be used t o designate all compounds having a hydroxyl group attached t o t h e benzene nucleus, and b y t h e t e r m “methylene b o d y ” shall be understood all compounds containing an active methylene group; t h e hydrogen of t h e group may or may not be replaced b y other bodies or radicals and t h e methylene group may be attached t o oxygen, nitrogen, sulfur or their equivalent which allows t h e methylene t o remain active. HISTORICAL
Condensation of the Salicylates As early as 1853 Gerhardtl showed t h a t an insoluble resin could be produced by t h e dehydrating of sodium salicylate b y means of phosphorus oxychloride, giving as t h e reaction 2(C7He08) C14H1005 HtO. Gerhardt notes t h a t this resin will hydrolyze in K O H solution. Schroder, Prinzhorn a n d K r a u t 2 in 1869 by dehydrating sodium salicylate with POCla produced a resin insoluble in water, alcohol, ether, etc., which hydrolyzed back in t h e presence of K O H t o salicylic acid. Combustion of t h e insoluble resin gave:
*
+
Calculated Calculated Calculated for hepta for octo for nonoObtained by Obtained by salicylosalicylosalicyloSocoloff Prinzhorn salicylic acid salicylic acid salicylic acid
. ...
C.. H . . . . ..
68.94 3.64
68.92 3.44
68.71 3.48
68.85 3.46
68.92 3.44
T h e calculations show for t h e octo- a n d nonosalicylosalicylic acid a better agreement with t h e combustion experiments t h a n does t h e calculated value for hepta-salicylo-salicylic acid. This larger molecular chain agrees with t h e results of Beilstein a n d Seelheim in dehydrating saligenin as recalculated.a T h e probable linking of t h e chain is according t o t h e formula HO. CsH4COO. (CbHaCOO)7. CeHr C O O H Velden in 1 8 7 7 ~showed t h a t salicylic acid, in t h e presence of sodium amalgam a n d acid, would reduce t o an oxybenzylalcohol and t h e n dehydrate i n t o a saliretin body. This same phenomenon has been shown b y Dr. Baekeland5 t o be possible*when salicylic acid is reduced a t t h e cathode by electrolysis. Salicylic acid and pyrogallol boiled in absolute alcohol give a product which is soluble in alcohol, has t h e formula (Cz6H22O9), a n d is probably represented‘ b y a dehydrated polymer of
0
Condensatiolt of Phenols and Higher Aldehydes I n 1871 Baeyer’ showed t h a t benzaldehyde a n d pyrogallol gave a substance insoluble in KOH, t h e re1
Ann. der Chemie, 87, 159 (1853).
3
Page 5. Johresbericht, 6, 37 (1877).
* Ibid., 160, 1 (1869). 4
TRIs JOURNAL,
~, ,37
(1912).
Patern0 Gazetta, Chem. ~ t ~ l 8 ,i 1 ~(1872). ~ ~ , 7 Berichlc, I, 25 (1871); 6, 280 (1872).
E
Jan., 1914
T H E J O U R N A L OF I N D U S T R I A L A N D E N G I N E E R I N G C H E M I S T R Y
+
action going according t o t h e equation 2(C7HaO) 2(CeH608) = (C~sH220,) H2O. The resin thus produced, reddens on oxidizing and bleaches on reducing, and by heating a t 2 0 0 ’ C. passes with t h e loss of hydrogen t o a substance of t h e formula CzeH160,. Phenol a n d benzaldehyde, with H2S04 as the condensing agent, give a red resin which was soluble with a red color in concentrated sulfuric acid, water, or alkali, a n d gives, in alkali, a beautiful violet color. Formaldehyde, phenol and concentrated sulfuric acid give a pasty mass which dissolves in K O H solution Trzcinski in 1 8 8 4 ~working with benzaldehyde a n d @-naphthol produced a resin which is insoluble in alkalies and which may be represented by 4CloHs0.4 C 6 HE.C H 0- ( j Hz 0) I n 1886 Claus a n d Trainer2 showed t h a t aldehydes a n d ethyl alcohol condensed with HC1 gave alkali insoluble products. They showed further t h a t by boiling 2 mols of phenol and I mol of acetaldehyde in ether a substance was formed which resembled very closely a higher saliretin in its qualities and gave as its possible formula HzC(CaH,OH)2: @-naphthol a n d acetaldehyde in HC1, gave products insoluble in KOH, while a-naphthol gave a resin soluble in alkali. Their formula suggested for &naphthol resin is C2H402(C10H7)2 and for t h e a-naphthol resin C ~ H ~ ( C I O H ~ O H ) ~ . Condemaz’ion of Oxybenzyl Alcohols (Saligenin) . Beilstein a n d Seelheim3 in 1861 produced a resin from oxybenzylalcohol by dehydrating this substance with acetic anhydride or ethyl iodide which on analysis gave:
+
Expt. 1 C . . . . . . . . . . . . . . 78.02 H . . ............. 5.98
Expt. 2 77.44 5.93
Expt. 3 77.80 5.60
Calculated for 8C7HaOr7HaO 77.59 5.77
a n d Kraut4 in reviewing this work suggests the formula Ca6HsoOp = 8(C7Hs02) - 7HzO which he names heptasaligeno-saligenin. T h a t would be for a substance where c = 77.59 H = 5.77 which agrees fairly well with results for Expt. 2; b u t Expt. I agrees better with rI(CaHs0.J - IOHZO = C77HesOlzsince for this formula C = 7 8 . 0 2 , H = 5.74; a n d Expt. 3 may be represented by g(C7HsO*) 8 H ~ 0 = (CeaH~010)where C = 7 7 . 7 7 , H = 5.76 From these results, it seems not unreasonable t o conclude t h a t the insoluble resin formed is variable a n d is formed by simply lengthening the molecular chain a n d may proceed indefinitely. The chain of t h e tvDe
t h e more nearly will the proportions of phenol t o t h e methylene group be I : I. Moitessier,’ in 1866,pointed out t h a t saligenin dehydrated and with t h e loss of one water passed over into a saliretin resin.
Condensation of Benzene Nuclei and Formaldehyde As early as 1871 Baeyer2 published a monograph on t h e condensation of phenols with aldehydes in which he concludes “ D a s s sich alle aldehyde mit allen Phenolen zu Korpern verinigen.” And Manasse in 1894, working with synthetic resins made from phenols notes t h a t “Esist keine neue Beobachtung das beide Componenten in Verhaltnisse I : I zusammentreten.” Baeyer3 demonstrated also t h a t formaldehyde could be replaced by chloral or the ammonia aldehydes in water and t h a t the phenols such as pyrogallol, resorcin, benzoic acid. gallic acid, etc., would act similarly t o phenol. S ~ h o t t e nin , ~1878,produced resinous substances with formaldehyde and phenols in which the hydrogen of the ring was replaced by such groups as -CHa, --NOz, -COOH, etc. Tollens,” in 1874, produced a rather remarkable product from the reaction of formaldehyde on aniline according t o t h e reaction CGHaNH2 CHzO e C6Hi,--N=CHz. I n 1891 Kleebergs made a resin by adding I O grams phenol t o 20 cc. formalin and adding, further, with cooling, concentrated HC1. The result was a resin insoluble in alkali and from this purified resin he could obtain no concordant combustion analysis. With gallic acid and formaldehyde, a substance C16H13010 was obtained. Abel,’ in 1892,made from a - and @-naphthol a n d formaldehyde, in acetic acid, soluble resins which, on treating with alkaline halogens, became quite insoluble going over t o t h e substance
+
which could be readily reduced with zinc dust in acid solution. Treating t h e dinaphthol methanes with hydroxylamine produced an insoluble substance
J~C1oH-T \CloH6-0
Abel also pointed out t h a t thymol and guaiacol gave condensation products. In 1892 Hosaeuss showed t h a t phenol, resorcin, pyrogallol or phloroglucin heated with dilute formalin in t h e presence of rather strong HzS04 or HC1 gave insoluble resins. Jahresberichle. 1866, 677. (1871); 6, 280 (1872) * Ibid., 6, 280 (1872). 4 Ibid., 11, 784 (1878). 8 Berichte. 17, 653 (1884). 8 Ann. der Chemie, 263, 283. 7 Berichte, 26, 3477 (1892). Ibid., 26, 3213 (1892). 1
* Berichte, 6, 25
continuing t o grow indefinitely. 1 2
4
The longer t h e chain
Bnichlc, 17, 499 (1884). Ibid., 19, 3004 (1886). B. and S., Ann. der Chemic. 117, 87 (1861). Kraut. Ibid.. 166, 123.
5
T H E J O C R N A L OF I N D C S T R I A L A N D ENGINEERING CHEMISTRY
6
Mariasse, 1894, produced a resin by taking I mol phenol I mol 40 per cent formalin I mol caustic soda and allowing t h e reaction t o take place in t h e cold; the odor of formaldehyde disappears, and orthoxybenzyl alcohol is the chief product. Wohl a n d Mylo,' in 1912, working with acrolein suggest t h e polymerization of formalin and phenol according t o t h e equation Hz
00+ =
CHPO+ H P C C >
= 0
+ H20
T h a t this does t a k e place t o a certain extent may account for t h e quinone coloring of t h e resins, being green in alkali, a n d red in acid. Lebach2 (1909-1913) deals, a t length, with t h e application of these resins t o specific industries. The resins are produced in every case from t h e boiling together of phenols and aqueous formaldehyde in t h e presence of a condensing agent. Dr. Baekeland3 has published a number of original papers on t h e practical application of these resins which are formed from aldehydes and phenols. A great variety of commercial articles has been produced by t h e use of these resins as binders, glues, lacquers, varnishes, shellacs, a n d solid products imitating amber, horn, bone, vulcanized rubber, etc. He showed t h a t this reaction, which is too violent when large quantities of catalyzers are added, a n d too slow as a rule when no catalyzer is present, may be controlled effectively by adding less t h a n one-fifth of a formula weight of a basic substance for each mol of phenol present. He pointed out also t h a t t h e resins formed when a basic substance is present a r e more insoluble in ordinary solvents t h a n those resins formed in t h e presence of acids. Litterschied and Thimme,4 in 1904, produced insoluble phenolic resins by substituting monochlor for formaldehyde. methyl ether, C1CH2-O-CH3, Condensation of A r o m a t i c Hydrocarbons and Formaldehyde A very remarkable polymerization5 takes place when mesitylene (symmetrical trimethylbenzene) is treated with formaldehyde, glacial acetic acid and concentrated sulfuric acid. T h e reaction is according t o t h e equation 2CgHa(CH3)3 CHzO + ( C H ~ ) S C ~CH2. H ~ C6H2(CH3)3 . -t H2O and this substance is always formed no matter what proportions of t h e original substances are used. A similar reaction takes place when benzene and chloral react according t o the equation C6He CCla.CH0 (CGHS)z.HC.CC18,which yields diphenyl trichlor ethane. This remarkable condensation of benzene nuclei with methylenes in t h e absence of hydroxyls was noted by Baeyer " E s scheinen sich
+
*
+
1 Berichte, 45, 2046 (1912). 2 Lrbach, Z e d f i i r a n g m Chemie, 22, 1598 (1909); J o u r . Soc. Chem. Ind., 32, 559 (1913) 3 Trans. A m . Electrochem. SOL., 16, 149 (1909; THISJ O U R N A L , 1,
149, 545 (1909); 3, 932 (1911); 4, T37 (1912); 5 , 506 (1913). 4 Chem. Cenlr., 1904, Bd. 949. 5 Baeper, Berichte, 5 , 1094 (1872).
Yol. 6 , No.
I
uberhaupt alle Aldehyde unter geeigneten Umstanden direkt mit den aromatischen Kohlenwasserstoffen zu verbinden in dessen treten dabei haufig Harze auf." P A T E N T LITERATURE.-FOr fifteen years efforts have been made t o commercialize these synthetic phenolic resins. Smith,l in 1899, patented a product a n d process obtained by heating one mol of phenol with I mol of formaldehyde and strong HC1, adding wood alcohol a n d amyl alcohol t o retard t h e reaction. The material was then dried in sheets a t 100' C. for 12-30 hours. Luft,2 in 1902, in his improved process added such substances as glycerine a n d camphor t o make t h e synthetic resins more suitable for molding. B l ~ m e r in , ~ t h e same year, produced a resin suitable as a shellac substitute b y using tartaric acid in large amounts as a condensing agent, e . g . , 2 mols of phenol, 2 mols of formalin and I mol of tartaric acid. Fayolle,4 1903-4, in his French patents uses sulfuric acid as a condensing agent and adds large quantities of glycerine, pitch. oils, etc., as organic fillers. Story,5 in 190j, omitted condensing agents altogether; after boiling the phenol a n d formaldehyde for 8-10 hours a t 100' C. t h e resin is dried out a t 80" C. I n 1905 DeLaire6 introduced caustic as a condensing agent. He used t h e alkali in equimolecular pro-. portions with t h e phenol and precipitated t h e resin f r o m solution by acid. DeLaire' advanced t h e a r t further b y resinifying phenol alcohols with heat a n d reduced pressure. Fried. Bayer & C O . , 1907, ~ patented a process of making odorless shellac substitutes by using o-cresol in place of t h e ordinary phenols. Helm,9 1907, introduced a m i n e s and a m m o n i u m salts as condensing agents or catalyzers. The agents he uses in almost equimolecular proportions. A series1° of patents were taken out in 1907 b y Knoll 13Co. in Germany a n d by Wetter in Great Britain for H . Lebach in which t h e condensing agents were acid or alkaline salts; sodium sulfite is specially mentioned. I n Brit. P a t . 28,009, 1907, occurs t h e first mention of t h e possible use of hexamethylenetetramine. However, t h e patent text shows clearly t h a t this reaction is no other t h a n a water process, t o which a condensing agent has been added. Grognot," in 1908, patented a process for making phenolic resins in the presence of large quantities of glycerine and after t h e reaction has proceeded, t h e desired length distilling off t h e excess water a n d glycerine. Dr. Baekelandlz has patented processes for proBrit. P a t . 16,247, 1899; Ger. P a t . 112,685, 1899. Ger. Pat. 140,552, 1902; U. S. Pat. 735,278. 3 Brit. P a t . 12,880, 1902. 4 Fr. P a t . 33,584, 1903; 2,414, 1904; 2,485, 1904; 341,013, 1904. Brit. P a t . 8.875. 1905. 6 F r . Pat. 361,539, 1905. Brit. Pat. 204,811, 1907; Ger. Pat. 189,262, 1905. Ger. Pat. 201.261, 1907. "Brit. P a t . 25,216, 1907. 10 Brit. P a t . 28,009, 1907; 24,072, 1908; SwissPat. 40,994, 1 9 0 i ; F r . Pat. 395,657; Belg. Pat. 204,811, 1907. 11 U. S. P a t . 3Y1,436, 1908. 13 U. S. Pats. 939,966, 941,605, 942,699, 942,700, 942,808, 942,809, 942,852, 949,671, 954,666, 957,137, 982,230, 1,018,385, 1,019,406, 1,019,407, 1,019,408. 1 2
Jan., 1914
T H E J O U R X A L O F I.ZiDCSTRIdL A -VD ELVGIhTTEERISGC H E M I S T R Y
7
ducing synthetic‘ resins from phenols a n d formaldehyde have very great commercial possibilities on account in t h e presence of less t h a n ‘ j j of a mol of basic con- of their uni for mi t y , ch e mi c a1 inertness , dielectric densing agents per mol of phenol. T h e reaction m a y properties, mechanical strength, high refractive index be kept well under control if t h e condensing agent be (lustre), plasticity a t certain stages a n d t h e cheapness present in small amounts a n d t h e resins are ordinarily a n d supply of t h e raw materials from which these more insoluble if basic, rather t h a n acid-condensing resins are manufactured. agents be used. T H E A ?TH Y D R 0 C S RE ACT I O S B E T TY E E S H E X A U E T H T L E I i E Dr. Baekeland’ warns against t h e use of conTETRAIIINE AND PHEKOL densing agents in excess of l;’j of a formula weight HE~:ameth?leiietetvai~ai)ie I’vipheizol per mol of phenol for two reasons: (I)t h e rapid inT h e simplest product which forms when phenol crease in t h e reaction caused b y large amounts of t h e condensing agent; ( 2 ) t h e trouble which t h e presence re acts mi t h he xa methylene t et r a mine is the a d di t io n of t h e condensing agent m a y cause in the final product; product hexamethylenetet,raminetriphenoIl which crysfor example, he warns us t h a t “if a large amount of tallizes out of a cold water solution. Lebach,? 1909, states t h a t hexamethylenetetramine ammonia be used hexamethylene tetramine is formed triphenol when heated b y itself, i. e . , in t h e dry, goes which is a crystalline body of definite chemical comover with t h e evolution of ammonia into “einen position.” Further “ I t is therefore essential t h a t t h e gelben unloslichen und unschmelzbaren Korper uber, proportion of base should not exceed certain definite limits, a n d t h e maximum permissible proportion has der nichts anderes ist als Bakelit.” The reason for been found t o be less t h a n of t h e equimolecular this conclusion doubtless lies in the fact t h a t t h e inproportion of t h e phenolic body present. If larger soluble material has t h e same yellow, transparent, proportions of base be used, there are formed in t h e glossy appearance as Bakelite which has been made mass, such amounts of disturbing bodies a s serve t o from phenol a n d formaldehyde using ammonia as t h e render t h e product technically inferior or worthless condensing agent. However this may be, t h e material has not a t all t h e same composition as Bakelite. for t h e purpose of this invention.” Other patents’ have been granted recently t o Dr. Baekeland gives as his conclusion from his experiments t h a t Bakelite consists of a maierial in which .%ylsw-orth in which t h e resin, resulting from t h e heating of z mols of phenol a n d one mol of formaldehyde, is 6 phenols3 have united with 7 methylene groups. Hexamethylenetetraminetriphenol has a compohardened b y heating in t h e dry b y t h e addition of a sition of 6 phenols €or every 1 2 methylencs, or 41.7 sufficient amount of hexamethylenetetramine. PROBLEM.-FrOm a review, then, of t h e scientific a n d per cent more methylene t h a n is necessary for making Bakelite. p a t e n t literature i t is evident t h a t there remains t o be Another argument which must be brought in opposistudied t h e reaction a n d the products which are formed tion t o this conclusion lies in t h e fact t h a t Dr. Baekewhen hexamethylene tetramine is made toreact in t h e d r y land represents t h e addition of t h e seventh formaldecondition with anhydrous phenol. All t h e previous hyde group directly to t h e molecule without t h e literature has h a d t o do with phenols and active methylenes in water solution to which condensing elimination of t h e oxygen in t h e formaldehyde as shown in t h e equation‘ with t h e elimination of water. Now agents have generally been added. The quantity and in t h e dry this final reaction can not possibly take kind of condensing agent has been shown b y Dr. place. S o oxygen or mater is present t o form with t h e Baekeland t o be of prime importance in these reactions. methylene group, formaldehyde. Indeed t h e only l l e n t i o n of t h e use of hexamethylene tetramine as a oxygen present in t h e reaction is contained in t h e substitute for formaldehyde h a s been made by hydroxyl of t h e phenol a n d as we have shown in t h e b u t it is reasonable t o presume! from t h e context of his patents, t h a t it was t o be used in a water solution a n d case of anisol and phenetolj when the hydroxyl group in t h e presence of a condensing agent. All these becomes inactive no reaction takes place between previous processes require t h e freeing of t h e resin from hexamethylenetetramine a n d t h e compound containing t h e water contained in t h e 40 per cent formaldehyde t h e benzene nucleus? i t seems reasonable, therefore, t o assume t h a t t h e oxygen of t h e hydroxyl is not intersolution a n d also t h e water which is a by-product fered with as this reaction proceeds normally a n d of t h e reaction. The resins prepared in this way have produces t h e insoluble resin. We h a r e shown elsealso t o be washed free of t h e condensing agent. However, in t h e case of anhydrous phenol a n d dry where6 t h a t in t h e reaction between phenol a n d hexahexamethylenetetramine, there is no mater present methylenetetramine no water is evolved. IFr h en hex a met h y 1en et e t r a mine t ri p h e n ol is he a t e d , in t h e ingredients. Neither is there a n y water formed there escapes not only ammonia a n d a small amount as a by-product a n d no condensing agent is present, of phenol b u t also there is present a strong fishlike which must later be freed from t h e resins b y washing. or mouse odor of methylamines. T h e latter byT h e only by-product is ammonia and i t escapes not present when t h e phenol is increased product is readily on account of its volatility. Such a resin with I n t h e s t u d y of this anhydrous reaction t h e condi- t o 6 mols of phenol t o methylenes. 1 Moschatos and Tollens, .4?t?t. der Chemie, 272, 280 tions are always under exact control a n d t h e reactions 2 Z r i t . f i i r ongeu,. Chemie, 2 2 n d year, 1600-1909. are definite a n d easily follon-ed. T h e resins formed 3 THISJ O U R K A L , 1, Mar., 1909; 5 , June, 1913. 2
U. S. Pat. 942,809, 1909. U. S. Pat. 1,020.593 and 1,029,737,
5
3
LOG.c i t .
6
1
Page 13. See page 12. See pages 1 I and 12.
8
T H E J O U R N A L OF I N D U S T R I A L A N D E N G I N E E R I N G C H E M I S T R Y
twice I t h e amount of necessary methylene would be not only costly b u t useless both in t h e quality of t h e resin a n d t h e offensive methylamine odor which i t possesses. With t h e formation of different by-products, t h e absence of water or oxygen t o form methyleneglycol or formaldehyde, thereby preventing t h e hardening process from taking place according t o Dr. Baekeland's formula a n d t h e proportions of phenol t o methylenes being 6 : 1 2 instead of 6 : 7 it seems untenable t o hold t h a t t h e yellow material which forms when hexamethylenetetramine triphenol is heated in t h e d r y , is Bakelite. T h e resin aggregate resulting from t h e heating of d r y hexamethylenetetramine triphenol a t 175 O C. or higher for I hour, with t h e evolution of ammonia a n d methylamines will swell, soften a n d partly dissolve or disintegrate in boiling phenol with a strong evolution of ammonia. This indicates t h e very evident fact t h a t hexamethylenetetramine or some of its degradation products are present in some form in t h e resin. A further proof of this lies in t h e fact t h a t t h e aggregate swells i n dilute HCI. No other resin which we have made swells in acid. All other resins, even t h e alcohol-soluble " Novalak remains quite unaffected i n a n acid solution. T h e swelling indicates t h a t there is present in t h e aggregate a material which is soluble in acid. I n t h e wet process it is impossible t o say t h a t t h e substance which is reacting in solution t o produce t h e resin is hexamethylenetetramine triphenol. There are no methylamines given off as by-products a n d t h e excess hexamethylenetetramine remains in t h e water solution while t h e resin which forms has a ratio of phenol t o methylene 13 : 1 2 i n t h e early stages of heating a n d t h e final product has a ratio of 6 : 6.
Hexamethylenetetramine altd Phenol (Wet Process) When I mol of hexamethylenetetramine a n d 6 mols of phenol are dissolved in 500 cc. of water a n d t h e solution boiled, a light yellow colored, transparent, amber-like, viscous liquid begins t o separate out which on continued heating becomes very viscous a n d finally changes t o a brittle amber solid when cold. This reaction is probably identical with t h a t which takes place when I mol of formaldehyde and I mol of phenol are heated together, using ammonia as a condensing agent t o hasten t h e reaction. T h e general conditions have been presented which occur when a n active methylene group reacts with a phenolic body in t h e presence of water. A very different set of reactions takes place when no water is present during t h e reaction.
Vol. 6 , No.
'
I
hydrous phenol be mixed to'gether a n d t h e mixture heated a t 60' C. or higher a reaction begins with t h e evolution of ammonia, a n d a m m o n i a i s the only byproduct as shall be shown later. If t h e phenol be present in excess, a n d if t h e mass be heated t o t h e boiling point, t h e reaction is very rapid. Practically all t h e ammonia is given off in t h e first ten minutes of t h e boiling a n d t h e reaction is quite complete a n d all t h e nitrogen has been evolved as ammonia within z hours from t h e beginning of t h e reaction. T h e reaction takes place probably in a series of steps. Accepting th8 formula of hexamethylenetetramine as N E E ( C H r N = CHz)a this reaction may t a k e place a s follows: z C ~ H S O H N-(CH-N = CHz)a ----) HO. CeH4.0. CH1. CeH4. C Hz. NHz H N = (CH-N
+
+
= CHZ)?
a n d a-second reaction m a y occur 2CpHsOH
+ HN=(CH--N=CHz)z + HzN-CHr-N ----)
NHz.CHz.CeH4.0.CHz.CeH40H
= CH2
This process continues until all t h e C H Zgroups have combined with t h e phenol a n d ammonia remains as t h e by-product. T h e intermediate product, aminosaligeno-saligenin, has been isolated a n d will be described in a later paper. No nitrogen a n d no methylamines are given off in measurable quantities a t any stage i n t h e process. No water is formed' a n d t h e nitrogen contained in t h e hexamethylenetetramine is evolved quantitatively as ammonia a s will be shown laterSg This reaction between a phenolic body a n d an active methylene group m a y be followed b y simply measuring t h e amount of ammonia which has been evolved. This is especially t r u e if t h e phenol is in excess, e. g., if 24 mols of phenol be heated with z mols of hexamethylenetetramine t h e final product is a resin in which t h e methylene groups have united with t h e phenols in t h e ratio of 13 phenols t o 1 2 methylenes.* T h e remaining I I mols of phenols are present a s free phenol a n d may be separated easily from t h e resin by dissolving t h e mass in dilute sodium hydroxide after t h e ammonia has all been evolved, neutralizing t h e solution or making it acid with HC1 a n d filtering off t h e precipitated resin. THE RESINS IN E,XCESS PHENOL
T h e resin which forms when hexamethylenetetramine is heated in t h e presence of excess anhydrous phenol seems always t o be t h e same material a n d resembles very closely Nov01ak.~ It is soluble in alcohol, acetone, phenol, caustic: etc., a n d is insoluble and precipitated from its solution b y acids. Heated, it By-products of the D r y Reaction does not harden rapidly b u t remains a liquid a t high When formaldehyde is used the by-product formed temperatures melting a t 105-1 10' C. a n d remaining is water, t h e reaction taking place with t h e dehydration liquid at 180' C. for many hours. If t h e heating is of t h e product as follows: continued, a flexible dark red skin forms over t h e which is only partly soluble in alcohol. A z H O C H Z + CeHaOHH O . C ~ H ~ . C H ~ O C ~ H ~ . C H Zsurface OH 1 Baekeland, THIS JOURNAL, 8, 932 (1911); 4, 741 (1912). T h e reaction continues with increasing size of t h e * Sep pages 1 1 and 12. molecule a n d further elimination of water. a See page 11. If, however, dry hexamethylenetetramine a n d a n 4 See analysis of Novolak, page 10.
.
Jan., 1914
T H E J O U R N A L OF I N D U S T R I A L A N D ENGINEERING C H E M I S T R Y
peculiar gas is given ,off during t h e heating which has t h e odor of burning phenol. When cold t h e resin hardens t o a glossy material which is more brittle than common rosin a n d apparently quite useless as a commercial product. This resin is not a solution of the final product in phenol b u t is a n individual having definite chemical properties a n d heating does not readily polymerize it. A variation in t h e amount of excess phenol present does not alter t h e characteristics or t h e amount of t h e synthetic resins formed when a definite amount of hexamethylenetetramine is used; e. g., if 18 mols of phenol a n d I mol of hexamethylenetetramine be heated until t h e ammonia has ceased t o evolve, t h e same amount of resin is formed a n d t h e resin has t h e same characteristics as when 1 2 mols of phenol are heated with I mol of hexamethylenetetramine until ammonia ceases t o evolve. As t h e amount of excess phenol is gradually decreased t h e mass which is left in t h e flask after t h e ammonia has ceased t o come off changes i t s physical properties, for all combinations of phenol with hexamethylenetetramine; those above 1 2 mols of phenol t o one mol of hexamethylenetetramine remain a yellow liquid whether hot or cold a n d are in reality a solution composed of t h e soluble resin described above a n d t h e unused phenol. But as t h e phenol is decreased below the ratio of 1 2 rnols of phenol t o I mol of hexamethylenetetramine, t h e liquid thickens and finally, as t h e excess of free phenol decreases, t h e resin changes t o a n infusible insoluble transparent yellow or red solid at all temperatures. Nine mols of phenol t o I mol of hexamethylenetetramine give a n aggregate which is quite solid a n d brittle in t h e cold a n d a tacky rubbery mass when hot (170' C.), Eight mols of phenol t o one of hexamethylenetetramine give a mass which is solid a t all temperatures, not as brittle in t h e cold as t h e g : I resin b u t rubbery when hot, like polymerized tung oil. The more t h e phenol is reduced t h e less elastic a n d resilient t h e mass becomes when h o t a n d at t h e ratio of (j.00-6.5)mols of phenol t o one mol of hexamethylenetetramine t h e material is quite solid a n d hard a t all temperatures; a t room temperatures i t is a solid which is not brittle b u t hard, dense, transparent a n d tough, with a tensile strength ranging from 4,00c-j,200 lbs. per sq. inch. When hot i t can be dented about like a hard filled rubber. Decreasing t h e phenol below j.j-6.5 mols t o one hexamethylenetetramine again increases t h e brittleness of t h e resin due t o t h e excess of t h e crystalline hexamethylenetetramine. The product, however, does not melt at a n y temperature. RESIN FORMATIOh-
Statements have been made - t o t h e effect t h a t t h e addition of t h e proper amount of hexamethylenetetramine t o phenols or phenolic resins t o form a 6 . I resin produces, on heating, a n exceedingly rapid evolution of ammonia a n d leaves a very porous, spongy mass. This is t r u e in only a limited sense. T h e rate of t h e evolution of ammonia may or may not be rapid, depending upon t h e method of c a r r y h g out t h e experiment, If heated on a water b a t h a t 100' C.
9
for 2 5 hours, less t h a n 50 per cent of t h e total ammonia will be evolved a n d t h e mass is a viscous liquid when h o t , a n d a brittle solid when cold. On t h e other hand, if a 6 : I mixture is heated rapidly a t 180' C. a spongy material soon forms which m a y have a volume twenty times t h a t of t h e original materials. This porous form of t h e resin is very advantageous when t h e material is t o be pulverized later, since i t powders in t h e ball mill in less t h a n one-tenth t h e time required t o pulverize solid lumps of t h e resin. This finely ground material, when pressed in hot molds under a pressure of 4-6 tons t o t h e sq. in., becomes a homogeneous, transparent, solid of maximum mechanical strength, highest dielectric properties, a n d chemical inertness. T h e grinding a n d molding of t h e powdered material in h o t molds, under pressure, is only one of a number of methods which we have in use in this laboratory for producing transparent solid goods. By a simple manipulation of t h e material, during heating, i t is possible t o produce large pieces of t h e final tra,nsparent, insoluble resin without t h e use of external pressure. We have, a t present, in our laboratory rods of this material 2 feet long a n d I ~ / P inches in diameter which have been produced by simply pouring t h e material while liquid into open molds a n d allowing it t o harden under suitable heat treatment without the application of external pressure. These rods in t h e final condition are homogeneous, almost water white transparent a n d free from fractures or gas bubbles. I N T E R M E D I A T E A N D BY-PRODUCTS
It might a t first be supposed t h a t some oxybenzyl alcohols or substituted saligenin would be present at t h e end of t h e reaction. This, of course, is impossible when i t is remembered t h a t no water is present a n d all t h e ammonia has been evolved: oxybenzyl alcohol, saligeno-saligenin, oxybenzylamine or an amid saligenin are also impossible since there is present in the remaining mass no water or oxygen necessary t o form t h e alcohols a n d no nitrogen remains t o form t h e corresponding amines. Such a compound, therefore, as
which is probably t h e second product formed when phenol a n d formaldehyde react, is impossible when phenol unites with hexamethylenetetramine in t h e dry. The intermediate products which form m a y be NHz.CHz.CeH40H, NH2.CH2.CaH4.0CH2.C6H40H, C6H5.0CHS.CsH40H, etc., b u t a t t h e end of t h e reaction t h e first two products can not be present as all t h e nitrogen present in t h e hexamethylenetetramine has been evolved as ammonia. And as we shall now show t h e final product which forms in excess phenol is a definite compound. This compound has little or no tendency t o polymerize or harden a n d become insoluble on heating, b u t when mixed with
T H E JOURNAL OF I N D U S T R I A L A N D ENGINEERING CHEMISTRY
IO
active methylene compounds transforms readily into t h e infusible resins. .Combustion Analysis of the End Product Formed in Excess Phenol T h e analysis of a resin which we have prepared by boiling a mixture of 1 2 mols of dry phenol and I mol of hexamethylenetetramin- until all t h e nitrogen has been evolved as ammr,nia, dissolving t h e resulting thick, yellow liquid in j per cent K O H a n d precipitating with zN HC1, t h e precipitate filtered off into Norton cones and washed with z N HC1 a n d dried a t joo C. in a gas oven gave results as follows:
Vol. 6, No.
I
C6H,.0CH2.C6H40H'as shown b y boiling-point analysis This is quite impossible for t h e reason t h a t such a resin is not a n individual b u t may be dissolved up in K O H a n d separated into two constituents b y t h e addition of HC1, t h e precipitate containing approximately 5 0 per cent a n d t h e filtrate holding in solution approximately 50 per cent of t h e original phenol used. T h e rise i n t h e boiling point given by Aylesworth is no doubt caused b y t h e 50 per cent free phenol a n d shows t h a t t h e molecule of t h e resin is so large in t h e precipitate as t o have very little effect upon t h e boiling point of t h e solvent employed as t h e rise in t h e boiling point may be accounted for Calculated for Found Calculated for by t h e free phenol present. Ci04HnzOic C90H80014 A further proof t h a t t h e molecule consisting of 1 2 C = 78.16% C = 77.98% 77.96% C = 78.04% methylenes and 1 3 phenols is from t h e fact t h a t t h e H = 6.17 H = 5.81 6.23 H = 5.88 0 = 15.85 15.81 0 = 16.16 0 = 16.03 total amount of resin formed in a phenol solution, a n d which is precipitated by t h e acid as described (1) C~orHazOlemay be written 14(C7HsO).CeHbO.HzO (2) CsoHaoOla may be written 12(CrHsO).CeII,O.H~0 above, corresponds in weight t o 6l/2 phenols t o each or C ~ H ~ . O C H Z . C ~ H ~ . O C H IOCHzCaH4OH. .C~H~. 6 methylenes added or 13 phenols t o 1 2 methylenes. Dr. Baekeland's published analyses of certain soluble This is in agreement with t h e combustion analysis, products which he has obtained b y heating excess and with t h e highest number obtained in t h e rise of ph&ol in t h e presence of formaldehyde solution' boiling points. T h e formula t h e n which we propose a n d by heating oxybenzyl alcohol and phenol are as for this soluble resin is C ~ H ~ O C H Z ( C ~ H ~ O C H ~ ) ~ ~ . C ~ and following Kraut'sP suggested nomenclature we follows : Found for Found for have called this material phenyl-endeka-saligenoCalculated for Phenol + CH20 Saligenin + Phenol saligenin Cio4HszOin _L__T h a t this resin is a n intermediate product which C = 78.1670 78.04Y0 77.9570 78,16y0 78.24y0 later is transformed b y being mixed with formaldchyde H = 5.81 5.79 5.97 5.65 5.75 0 = 16.03 16.17 16.08 16.19 16.01 or hexamethylenetetramine a n d heated into t h e T h a t t h e end of t h e molecu!ar chain in t h e indi- insoluble infusible methylene-phenol condensaticn prodvidual, made as described above, can not have a ucts of the highest dielectric a n d tensile strength a n d -CH20H group attached to t h e benzene ring is evident chemical inertness, is almost certain from t h e fact t h a t from t h e f a c t t h a t t h e reaction between anhydrous a n aggregate can be made consisting of I O mois of phenol phenol and hexamethylenetetramine could not produce to I mol of hexamethylenetetramine a n d this heated such a substance, especially since, as will be shown until all t h e ammonia has been evolved which results presently, t h e only by-product from t h e reaction is i n a heavy, viscous yellow transparent liquid when ammonia. The other end of t h e molecule is hydroxyl- hot a n d semi-solid when cold, consisting of over 63 ated, probably, as t h e substance is readily soluble in per cent phenyl-endeka-saligeno-saligenin(novolak), t h e rest being water-soluble phenols ; hexamethylenealkalies. T h e H 2 0 may be either a water of hydration which tetramine is then added in such proportions as t o attached itself t o t h e molecule during the process of make the total phenol .to methylme in t h e ratio of solution a n d precipitation, or, which seems more I I and heated until the mass has become a yellow reasonable, t h e process is one of oxidation, for t h e porous sponge, t h e mass is then nowdered a n d pressed individual changes its color from yellow t o green when in molds a t zooo C. a n d I Z , O C O lbs. per sq. in. for i t dissolves in alkali a n d from green t o red when it j-10 mins. a n d the resin becomes a transparent amberis precipitated by acid. The drying of t h e p p t d , material like material with a dielectric strength of 5*90,000 a t soo C. changes t h e color further to a reddish brown. volts per inm , a n ohmic resistance of z X 1 0 9 per This possible oxidation is described at further length cm3 a n d a tensile strength of 4000-4 500 lbs. per sq. inch. T h a t a substance consistitig of more t h a n 60 per on page 14. Such a formula would have a niolecuAaligenin a n d t h e remain~ phenyl-endeka-saligeno ~ ~ ~ ~ lar weight for C ~ O ~ H ~=Z 1O5 5I2~ or for C ~ O = H cent 1384. The highest formula weight calculated from der a water-soluble phenol is transformed by the adt h e rise in boiling point which Dr. BZekeland* has dition of hexamethylenetetramine and further heating into an insoluble substance of such high dielectric a n d found for novolak dissolved in acetic acid is 1 2 1 6 . Ayleswortha in his U. S. patent 1,029,737 lays claim tensile strength seems t o preclude t h e possibility of t o t h e production of a soluble phenolic resin formed this resin remaining unchanged or transforming into from phenol a n d formaldehyde b y boiling z mols a n y other product than a n insoluble infusible resin of phenol in I mol of aqueous formaldehyde until of highest dielectric a n d tensile strength For i t must be a resin is produced having t h e formula C6H5.0CH2.- remembered t h a t t h i s resin, of itself, is more brittle t h a n cheap rosin and soluble in all t h e ordinary solvents. 1 ..I H. Baekeland. THIS JOURNAL, 1, 545 (1909).
* THISJOURNAL 8
1, 545 (1909).
I. W. Aylesworth, Brit. Pat. 4,396,1911,
1 2
x.
Lebach, Jour. SOC.Chem I n d , 82, 561 (1913). A n n der Chcmie, 166, 123.
J a n . , 1914
T H E J O U R N A L OF I N D U S T R I A L A N D ENGINEERING C H E M I S T R Y
Indeed we have found this intermediate substance present i n all t h e materials we have made of whatever variety, soluble or insoluble, whether heated for a long or short period a n d whether a t a high or relatively low temperature. Low temperature, a short time of heating a n d excess phenol or what is t h e same thing, not enough hexamethylenetramine t o make a 6 mols of phenol t o I mol of hexamethylenetetramine resin gives a greater percentage of this soluble resin. It is also present in small proportions in 6 : I resins t h a t have been heated for m a n y hours a t zooo C. T h e following table gives t h e result of a few analyses of resins as described:
The flask containing t h e hexamethylenetetramine a n d phenol was then heated carefully. Great care was necessary as t h e reaction is rapid, being strongly exothermic so t h a t it goes a t a n increased rate even
u
).
When t h e soluble, brittle, fusible resin is mixed, powdered a n d heated for I hour with hexamethylenetetramine i n t h e proportions of 6 mols of novolak t o I mol of hexamethylenetetramine, assuming one mol of novolak, C g ~ H ~ 0 0 = 1 4 1384, t h e mass becomes largely insoluble in all ordinary solvents. caustic alkalies, etc. It swells or gelatinizes in boiling phenol b u t does not dissolve a n d exhibits generally the properties of t h e insoluble products. The final resin formed from t h e novolak a n d hexamethylenetetramine is much darker t h a n t h e resin formed directly from phenol a n d hexamethylenetetramine. This is, no doubt, due t o darkening which takes place when t h e resin is separated from t h e excess free phenol. By-products Formed in the Reaction lietiueen Anhydrous Phenol and D r y Hexa?neth~lencretra?na~ae=
An apparatus was arranged as shown in t h e accompanying diagram. A flask with a long side delivery t u b e was sealed off after t h e hexamethylenetetramine a n d phenol were added. The delivery t u b e led into a reflux condenser, thence into a safety t r a p . sulfuric acid Marchand U-tubes a n d finally through a CaCls t u b e a n d safety bulb into a gas t r a p over water. These experimental results were obtained in our laboratories by Mr. Weston Carpenter.
I - HZP. -
2-G a s t r a p . 3 - S o f e t y bulb
7-H~Soet r a p . c o o l e 4
4 - c u 4 t u b e t o keckcyoerimantdry 6-H,Soq Safety coo/ed 8- C o o d e n s e y .
9- D i s t , l l a n g
/O-(;/ass
5-GQ..tubc
D r y phenol and hexamethylenetetramine, heated on a water bath 14 hrs., later, heated in an air bath a t 100' C. for 221 h r s . . . . . . . . . . . . . . . . 8 : 6 68.4 8.9 22.7 A Bakelite pipe stem bought in the 94.4 .. 4.2 market.. . . . . . . . . . . . . . . . . . . . . . . . . ? 94.0 .. 3.2 Bakelite lacquer,No. 5 , satin finish(a). . ? 0.0 32.5(b) 5 2 . 1 6 : 1 r e s i n h e a t e d 1 5 h r s . a t 1 2 5 ° C ( c ) ,. 6 : A 80.0 1.0 20.0 6 : 1 resin, containing 7 per cent naphthalene, heated 10 hrs. a t 122' C . . . . 6 : 6 68.4 6.3 19.5 6 : 1 resin heated 1:rtt.r in mold for 1 6 : 6 84.7 1.3 18.7 hr . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.96 7.64 90.0 6 : 1 resin heated for 5 hrs. a t 154' C . . 6 : 6 (a) T h e lacquer was 22 per cent resin and i 8 per cent solvent. The percentages in the table are calculated on the resin ( b ) This is doubtful as the figure was obtained b y the repilar bromination method and the results calculated for free phenol. T h e real values are probably higher. (c) 6 : 1 resin is formed b y heating together 6 mols phenol and I mol hexamethylenetetramine.
I1
inezp.
flask.
wool.
after t h e flame is removed if t h e initial heating has not been sufficiently slow t o allow t h e 'first of t h e reaction t o t a k e place a t a moderate rate until t h e ammonia has been partly evolved. Checks were kept upon t h e reaction in four ways: (I) The loss in weight in t h e flask. ( 2 ) T h e increase in weight in t h e Marchand tubes. (3) Titration of t h e ammonia which came over by boiling i t off from t h e (SH4)2S04 on t h e addition of KOH. (4) Direct titration of t h e ammonia in duplicate experiments. The following table shows t h e result of three experiments:
., , .
I..
,
2..
,,,,
.
3.......
4.000 4.000 4.000
41,730 1.943 1.954 1.930 1.937 + 0 . 5 6 .-0.6 -0.3 40.370 1,943 1 . 9 4 3 1,943 1.950 0.0 0.0 +0.3 41.510 1.943 1.954 1 , 9 3 0 1 , 9 4 0 4-0 56 --0.6 -0.1
The results show t h a t the loss in t h e flask agrees very closely with t h e calculated yield of ammonia from t h e hexamethylenetetramine used. The titration of the ammonia here, as in many other experiments, has practically always been I O O per cent of t h e nitrogen present a n d t h e gain in t h e concentrated sulfuric acid just equals t h e calculated N H 3 a n d agrees also with t h e loss from t h e flask. Care was taken t o include, in t h e final weight of t h e flask, t h e weight of phenol found in the washings from t h e reflux condenser a n d t h e t r a p . This generally amounted t o less t h a n 0.0j gram of phenol. T o prevent t h e escape of a n y ammonia in t h e rubber joints t h e glass tubes were placed end t o end, over which was drawn very thick walled rubber tubing which previously h a d been given a heavy coat of pyroxylin varnish. I n t h e h'Iarchand U-tubes some care was needed t o prevent t h e sulfuric acid from splashing as t h e ammonia absorbed rapidly a n d the sulfuric acid struck back. This was overcome by absorbing p a r t of t h e ammonia in a small amount of sulfuric acid in t h e first t u b e a n d the remainder of t h e ammonia in t h e
T H E J O U R N A L OF I N D U S T R I A L A N D ENGINEERING C H E M I S T R Y
I2
second tube. Glass wool plugs prevented t h e splashing and t h e results are fairly concordant. Simple as t h e reaction would seem t o be, t h e very rapid evolution of t h e ammonia in t h e early part of t h e experiment made a number of experiments necessary before t h e present apparatus was devised a n d t h e ammonia could be absorbed i n i t s rapid evolution without allowing a n y of i t t o escape. We concluded t h e n from these experiments t h a t n o water is formed in t h e reaction or i t would be evolved during t h e boiling which takes place a t 180' C. for t w o hours a n d t h e loss i n t h e flask would not equal t h e calculated loss for ammonia a n d t h e actual weight of ammonia found by titration a n d b y weighing as ( N H 4 ) 2 S 0 4 . It was not considered sufficient t o simply weigh t h e ammonia coming off, as methylamines if present would increase t h e weight and give useless results. Nor was t h e titration alone sufficient as i t did not distinguish between ammonia a n d methylamines. T h e t w o experiments were necessary in order t o check up t h e results a n d t o make certain t h a t no other product t h a n NH,is evolved during this reaction between phenol a n d hexamethylenetetramine. The resin which formed in t h e flask was t h e soluble, fusible phenyl-endeka-saligeno-saligenindissolved in t h e excess phenol. THE
ACTIVE
HYDROGEN
ON
THE
BENZENE
NUCLEUS
T h e hydrogen which is replaced in t h e ring is apparently very largely t h e ortho-hydrogen although either t h e meta- or t h e para-hydrogen may possibly be replaced. This reaction is similar t o t h e formation of orthoxybenzylalcohol when phenol a n d formaldehyde are treated in t h e presence of NaOH. T o test this statement, three experiments were undertaken with t h e cresols: '5.1 grams of each of t h e pure cresols were weighed out separately into test tubes a n d t o each 1.1 grams of hexamethylenetetramine were added. T h e three test tubes were placed in a t h i n beaker which rested on a block of wood, in a n electric oven at 135' C. I n a few minutes t h e metacresol began t o react giving off ammonia a n d passing over into a n insoluble, yellow, scoriaceous resin which, when hot, was rubbery for a very short time, b u t soon transformed into a hard non-rubbery resin a t 135' C. T h e para-cresol acted in t h e same way as did t h e meta-cresol although slightly slower. The two resins were identical in appearance a n d general properties. T h e ortho-cresol acted very differently. After three d a y s of heating t h e reaction h a d proceeded only far enough t h a t t h e mass was a viscous, yellow liquid when hot (135' C.) a n d a brittle solid when cold. T h e reaction was very slow and did not evolve ammonia rapidly at a n y time. When heated a t 200' C. for 3 0 minutes i t formed into a rubbery, resinous mass when h o t a n d a hard rather brittle solid when cold. On t h e whole, then, t h e para- a n d meta-cresols acted rapidly a n d similarly while t h e ortho-cresol acted very slowly a n d did not form a resin which when hot was hard a n d solid This delayed reaction of t h e ortho-cresol may be explained as a steric hindrance for the hydroxyl. It m a y also be explained as due t o a substitution
Vol. 6 , No.
I
of '/%t h e active (ortho) hydrogen in t h e ring b y a nonreactive group. T h e latter seems t h e more probable. The fact t h a t t h e meta- a n d para-cresols are very rapid in reaction seems t o indicate t h a t t h e metaa n d para-hydrogens are not t h e ones which are replaced during t h e formation of t h e synthetic resin. THE
ACTIVITY
O F T H E HYDROGEN I N T H E HYDROXYL GROUP
Of t h e proposed formula for these condensation products, t h e more reasonable seems t o be t h e straight chain formula with t h e phenol rings linked together group i n between, t h e attachment with t h e -CH20being made a t t h e hydroxyl of t h e phenol a n d a t t h e ortho hydrogen of t h e ring as follows:
T h e reason for assuming t h a t t h e reaction is closely connected with t h e hydroxyl group lies in t h e fact t h a t if t h e hydrogen of t h e hydroxyl or t h e hydroxyl itself be replaced by a non-reactive group, no apparent reaction takes place after prolonged heating either in t h e dry or in wet processes. For example, if anisol be heated a t t h e boiling point with hexamethylenetetramine in t h e absence of water for one week continuously, no appreciable reaction takes place. T h e anisol remains a colorless, transparent, mobile liquid with t h e anisol odor a n d t h e hexamethylenetetramine settles out on cooling as a crystalline salt. The salt was tinged slightly yellow b u t further t h a n t h a t no reaction was noticeable. If t h e hydroxyl be replaced by hydrogen, e . g., if naphthol be replaced b y naphthalene, no reaction takes place after prolonged heating in t h e dry or wet while with P-naphthol in t h e absence of water t h e reaction is very rapid. T H E REPLACEMENT O F T H E HYDROXYLS I N T H E R E S I N S BY N O N - R E A C T I V E
GROUPS
The fact t h a t phenol-formalin resins, made u p into lacquers a n d heated in thin films insoluble i n alcohol, acetone, etc., are attacked slowly b y caustic potash leads one t o t h e conclusion t h a t whatever substance is formed i t very probably contains a percentage of free hydroxyl. So long as t h e hydroxyls are present i t is t o be expected t h a t t h e substance remains a t least partly or slowly soluble in free caustic. Working upon t h e assumption t h a t t h e reaction continues as an open chain of benzene rings linked together b y t h e group -0-CH2which does not finally form a closed ring or an inner anhydride, it was desirable t o fill the hydroxyl on t h e end of t h e long chain with a n inert group, or t o replace t h e hydroxyl entirely. T h e similarity of anisol a n d phenetol t o phenol a n d the inertness of their alkylated hydroxyls suggested their use in small quantities in making caustic insoluble resins. A resin was made up consisting of 6 mols phenol, I mol anisol and I mol hexamethylenetetrafiine. T h e
J a n . , 1914
T H E JOURNAL OF INDUSTRIAL A N D ENGINEERING CHEMISTRY
mixture was heated for a short time in a n open beaker a n d cooled while still liquid. The resin was a light, transparent, amber-yellow color, solid a t ordinary temperatures a n d soluble in alcohol a n d acetone although more slowly soluble t h a n t h e ordinary resins made u p without t h e addition of anisol. When made into a lacquer a n d cooked a t 170’ C. on brass sheets i t gave a light transparent film which was insoluble in acids, alcohol, acetone a n d N caustic potash. It required 7 2 hours t o loosen t h e film with t h e caustic. All lacquer films from resins made of phenol a n d a n active methylene, whether made in t h e presence or absence of water with or without t h e ordinary condensing agents, are attacked b y N caustic within 23 hours, t h e lacquer coating reddening in a short time after t h e caustic is added. T h e film from resins compounded of 6 mols of phenol a n d I mol of anisol or phenetol remained unreddened a n d quite unattacked b y t h e iV caustic potash a t t h e end of t h e first 24 hours a n d a t t h e end of 7 2 hours was only slightly attacked along t h e edges of t h e coated brass strips a n d a t no time was a n y reddening of t h e lacquer film noticeable. T h e explanation for this increased insolubility of t h e resin a n d of i t s greater resistance t o color changes seems t o lie in t h e absence of free hydroxyl. This is best illustrated for naphthalene from t h e following equations: CsH60H
+ --CHtNH-
+ CioHs *
. C~H~.OCH~.C~O NH3 H~
+
THE
F I N A L PRODUCT-THE
INSOLUBLE
I3 INFUSIBLE
RESIKS
T h e transformation of t h e phenyl-endeka-saligenosaligenin into t h e final, insoluble, infusible resin is brought about b y the further condensation of i t s molecules b y active methylenes. Dr. Baekeland concludes t h a t t h e final product is not obtained through t h e polymerization of novolak as a n intermediate product, b u t t h a t t h e growth of t h e chain stops when six phenols have entered a n d t h a t a n extra hydrated formaldehyde forms with t h e six phenol chain a n inner anhydride as shown by t h e following formula : C6Hr.OCHz C ~ H ~ . O C H ~ . C ~ H ~ . O COCHZ.CBH~.OCHZ.CBH~ HZ.C~H~ 1L ~ O - ~ ~ ~ ~ C H Z - - - ~ I 0-J This inner anhydride, he assumes, polymerizes into t h e insoluble, infusible resin. T h a t this can not be t h e mechanism of t h e reaction in t h e case of t h e production of these synthetic resins b y means of anhydrous hexamethylenetetramine, is evident from t h e fact t h a t no water or oxygen is present which could possibly form such a n inner anhydride as above described. If t h e reaction takes place with t h e formation of t h e larger ring i t would be according t o t h e formula
C6Ha.OCH: C ~ H ~ . O C H Z . C ~ H ~ . O C H ~ . C ~ H ~ . O C H Z . C ~ H ~ . O I
L---O----CHZ------
I
a n d would lack a t least one oxygen i n t h e molecule which t h e inner anhydride formula requires. This N Hz. C ~ H ~ . O C H ~ . C ~ Ooxygen H, might, in certain cases, be replaced b y nitrogen although this is i’mprobable as t h e final product never CGHsOH hTH,.CH2.C6H4.0CHzCioH7 + yields more t h a n a trace of nitrogen a n d even this Cs Hs. 0.CH2. CsH4.OCH2.CioH7 “3, a m o u n t m a y be mechanically locked as ammonia in etc. t h e final product. a n d this reaction m a y continue indefinitely. This Further i t m a y be mentioned t h a t t h e larger moleresin would no longer have t h e properties of a sub- cule of phenyl-endeka-saligeno-saligenin, i. e . , 13 moles stance possessing free hydroxyls, b u t i t would display of phenols : 1 2 moles of methylenes, could n o t , without t h e inert characteristic of a n ether-like anisol which hydrolyzing into a shorter chain, yield such a molecule is n o t broken down in t h e presence of boiling caustic as Dr. Baekeland proposes. potash. T h a t such a reaction is possible m a y be Recently, Dr. Raschigl has proposed certain formulas inferred from Baeyer’s‘ work with mesitylene a n d for t h e condensation of phenols a n d active methylenes formaldehyde giving, according t o t h e equation, which, apparently, are not identical with a n y of t h e substances isolated as products of this reaction. His formulas show t h a t in passing from “Bakelite I” t o “11” t h e carbon content increases from 7 7 per cent t o 81.4 If t h e hydrogen i n a mesitylene ring is active in the per cent while in actual results all t h e combustion presence of CH20 a n d forms water with t h e oxygen analyses of Bakelite which Dr. Baekeland has pubof t h e formaldehyde leaving a substance formed by lished a n d t h e analyses which we have carried out of dehydration behind, i t is not unreasonable t o assume these synthetic resins, show t h a t t h e carbon content t h a t a t high temperatures (180’ C.) a hydrogen on t h e decreases from 78.1 per cent t o 79 per cent or lower naphthalene ring m a y be sufficiently active t o form as t h e resins pass from t h e soluble t o t h e insoluble ammonia with a n amine group, a n d leave a resin as a state. As these proposed formulas evidently do not by-product which has no free hydroxyl at t h e e n d agree with actual analyses t h e y will n o t be considered of t h e molecular chain as described above. T h e further. permanence in t h e yellow color of t h e resin or lacquer I n arriving a t a n empirical formula for t h e final a n d their insolubility in normal caustic, as well as all resin, a series of combustion analyses were carried o u t other ordinary solvents, point t o t h e absence of free on products prepared as described i n t h e followhydroxyls in t h e substance. ing table: Cp,Hb.OCHZ.CloH7
+ -CH,TU”-
+
1
Berichte. 5, 1094 11872)
+
1 Z.
fur ungew. Chem , 26, 1945 (1912).
T H E J O U R N A L O F I N D U S T R I A L A N D ENGINEERING CHEMISTRY
I4
PREPARATION O F RESIN T h e insoluble material from a resin hardened with formaldehyde and washed free of substances soluble in normal N a O H . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A 6 : 1 resin containing traces of anisol heated in air at 1 7 5 * C. for 6 hours then pressed in a mold a t 12,000 lbs. per sq. in. for 5 min. at 200° C . . . . . . . . . . . . . . . . . .
6 : 1 resins heated at 200° C. for 100 hrs
...............
C
H
75.58 75.34
6.08 6.11
74.56 75.01 74.74 75.32 75.02 78.83(0) 75.58
6.40 4.50 6.35 6.55 6.35 5.47 5.20
(a) Unaccountably high.
T h e average of t h e eight analyses gives C = 75.4 H = 5.9 0 = 18.7 which corresponds t o 77(C7H60).zo(C7 T h e group C 7 H 4 0 2may be explained as a n oxidation i n which 2 hydrogens are replaced by a n oxygen somewhat as follows:
-0-
Simultaneous with this oxidation is a reddening which takes place rapidly a t increased temperatures, i . e . , above 200' C. T h e oxidation probably reaches a n equilibrium a t 4g(C7Hs0).48(C7H402). Reasons j o r Accepting the Chain Formula Oxybenzyl' alcohol condenses by dehydration t o a resin which yields, on combustion analyses, a n empirical 8(C7H8Oz) - 7 H 2 0 or formula represented . by I I ( C , H ~ O ?) IOHZO. This condensation is most easily represented by t h e chain constitution, H 0 C Hz. CsH4. C H20. CeH4. C H20. . . . . . Cs H40 H, a n d since t h e resin is soluble in caustic potash t h e hydroxyls are free as t h e formula indicates a n d not a n inner anhydride. T h e close resemblance of t h e phenolmethylene condensation t o t h e condensation of oxybenzyl alcohol is apparent. T h e salicylate2 condensations which are closely allied t o t h e dehydration of oxybenzyl alcohol pass, with t h e loss of water, t o a resin represented by 9(CeHdOH.COOH) - 8 H 2 0 which is most easily represented by a chain formula, 0 0
II
have ample proof t h a t it does, then so large a molecule as novolak could not possibly pass t o t h e final inner anhydride formula which Dr. Baekeland suggests for Bakelite C, unless t h e larger molecule previously hydrolyzed i n t o small aggregates, a reaction which is improbable. At best, t h e final resin formed from phenol a n d hexamethylenetetramine cannot be formed b y t h e mechanism which Dr. Baekeland suggests for t h e formation of Bakelite C which consists in t h e addition of C H 2 0 t o a shorter chain t o form a n inner anhydride. CHzO is not a n d cannot be present in a reaction between anhydrous phenol a n d hexamethylenetetramine. If water be formed, which, later, decomposes t h e hexamethylenetetramine t o form CH20, 2 . 7 per cent would be required t o produce enough CHzO for t h e above reaction t o t a k e place, a n d we have shown t h a t t h e total by-products formed in this reaction are a t least 99.5-99.7-100.0 per cent pure ammonia, of t h e water necessary leaving on a n average about for t h e reaction, i. e., if our gross error in determining ammonia be considered as entirely due t o t h e formation of water (a supposition which has little probability behind it). A further reason for holding t o t h e chain formula is from t h e fact t h a t t h e final resins whether formed from phenol a n d formaldehyde or hexamethylenetetramine are slowly soluble in normal caustic potash. This fact points t o t h e presence of a free hydroxyl in t h e molecule of t h e final resin. T h e hydroxyl may be removed or filled with groups which make t h e resin inert toward alkalies. For t h e sake of completeness, t h e following tables of constants for these resins have been added: PHYSICAL AND DIELECTRIC PROPERTIES
Tensile Strength T h e tensile strength of t h e pure resin is between 2000 a n d 4500 pounds per Square inch depending upon t h e conditions under which it is made, such as temperature, pressure a n d time of heating. T h e influence of temperature a n d time of heating upon t h e tensile strength are given in t h e following tscble: TABLE I-TENSILE STRENGTH
HO.CaH4CO.CsH4.CO. . . . . . CeH4.COOH
1 2
Kraut, Beilstein, Seelheim, loc. cit. Schroeder, Prinzhorn a n d Kraut, loc. cit.
O F THE RESINS
I n t h e study of these resins for possible commercial uses t h e following constants have been determined:
II
This resin not only dissolves in a solution of K O H b u t is hydrolyzed by i t back into salicylic acid. T h e fact t h a t t h e soluble resin which we have named phenyl-endeka-saligeno-saligeninand which is similar t o t h e resin which Dr. Baekeland calls novolak a n d which we have shown t o possess t h e probable formula CsHs.(OCH2.C6H4)110CH2.C6H40H passes with t h e further absorption of methylenes t o t h e insoluble resin, is t h e strongest argument in favor of t h e material being a chain molecule of indefinite length. If t h e resin novolak passes t o t h e insoluble resin of greatest chemical inertness, tensile strength, etc., a s we
Vol. 6 , NO. I
NO.
4 8 12 5 9 13
6
Time heated Hours '/2
I/% l/a
1 1
1 11/2
10 14 7 11
11/2 2 2
15
i
11/2
Temperature
C. 123 146 181 129 155
179 128 150 181 127 150 183
Aggregate Tensile average showing strength effect in time Lbs. per sq. in. Lbs. per sq. in. 1600 1420 1920 1900 1540 2710 2000 2750 4400 1630 3600 2920
.. 1646
..
.. 2050
..
.. 3050
.. 2716
..
By adding t h e values corresponding t o a given temperature, t h e effect of temperature upon t h e tensile strength can be determined.
T H E JOURNA4L OF I N D U S T R I A L A N D ENGINEERING C H E M I S T R Y
J a n . , 1914
TABLE11-TEMPERATURE Temperature
Average total value in lbs. per sq. in.
C.
125 150 180
2377 3103 3650
These values show t h a t t h e higher t h e temperature a t which the material is heated, t h e stronger it becomes, in a given length of time. T h e influence of t h e molding pressure upon t h e tensile strength is shown in Table 111. TABLE111-MOLDING PRESSURE NO.
Molding pressure Lbs per s q in
Tensile strength Lbs. per sq. in.
2,860 5,700 8,560 11,400
800 937 3200 3970
16 17 18 19
T h e material used in these experiments consisted of a mixture of t h e rubbery product a n d t h e final product. It was desirable t o know what mixture gave t h e highest tensile strength. T h e following table shows t h e influence of t h e composition upon t h e tensile strength: TABLE IV-(RUBBERY N O
AND
FINAL PRODUCT) TENSILE STRENGTH
Per cent rubber product Final product
37 38 39 41
10 20 30 50
90 80 70 50
Tensile strength Lbs. per sq in. 4720
45 10 2880 2120
This table shows t h a t i t is best t o use only a few per cent of t h e rubber product. This is t o be expected since t h e rubber product by itself is comparatively brittle. Crushing Strength T h e crushing strength of t h e resin is much higher t h a n t h e tensile strength, which is usually t h e case with comparatively brittle materials, a n d varies from 2 0 , 0 0 0 t o 30,000 pounds per sq. in. (ebonite 2 2 0 0 a n d porcelain ~ j , o o o ) . Table V gives t h e relation of t h e crushing strength, t i m e of heating a n d temperature of heating a n d tensile strength. TABLEV Tensile strength Lbs. per No. sq. i n . 4 10 13 14 20 24
1600 2750 2710 4400 2700 3100
Time Hours =/a
1I/* 1 11/2
11/2 11/2
Temp.
c.
123 150 179 181 183 184
Crushing strength Lbs. per sq. in. 7,456 20,800 22,250 28,100 30,400 32,000
For higher temperatures t h a n those given in t h e table there is no increase in strength in t h e resin a n d for temperatures above 300' C. t h e resin begins t o weaken. Dielectric Strength Discs of t h e resin ' / 2 a millimeter thick gave voltage puncture tests of from 40,000 t o 5 0 , 0 0 0 volts. These tests were made for us through t h e kindness of t h e Crocker-Wheeler Company. By way of comparison, hard rubber has a dielectric strength of 38.000 volts per millimeter a n d ebonite 30,000.
I5
Speci,iic Electrical Resistance The specific electrical resistance of t h e final resin as determined b y t h e Bureau of Standards a t Washington varies f r o m 2 . 7 t o 2,800 X 106 megohms per cm3., depending upon t h e method of treating t h e same. TABLEVI-SPECIFIC Heated in water b a t h No. Hours 1. . . . . . . . . . . . 17 2 . . . . . . . . . . . 22'/2 3 . . . . . . . . . . . . 29
6 . . . . . . . . . . . . 17
ELECTRICAL RESISTANCE
Heated in oven Temperature of Specific resistance Hours oven O C . Megohms per cm3. 5 10 10 15 10 5
154 164 135 125 129 123
73 2800 96 240 150 2.:
X 106 X 106 X 108 X 106
x x
106
108
The resins formed b y t h e reaction between hexamethylene a n d phenol in t h e anhydrous state as described in this monograph present all t h e commercial possibilities of t h e resins formed in t h e water processes, a n d have certain other valuable qualities in themselves such as uniformity of product, ease of manipulation in t h e making a n d higher dielectric properties t h a n t h e wet formed resins containing traces of water a n d t h e condensing agents. T h e application of these anhydrous resins t o t h e electrical, paint, varnish, lacquer, glue, molding a n d other industries has been so great as t o require special t r e a t m e n t in a separate monograph. Application for basic a n d process p a t e n t s on these resins a n d their uses have been made b y our firm S. Karpen & Brothers of Chicago, who have given very hearty a n d liberal financial support t o this multiple Fellowship in t h e Industrial Department of Kansas S t a t e University. We wish t o express our very deep t h a n k s t o Dr. R . K. Duncan, a t whose kind suggestion a n d encouragement this research was undertaken a n d completed. BIBLIOGRAPHYOF
PHENOLS, METHYLENESA N D THEIR CONDENSATION PRODVCTS
Claus and Ruppel, J . prakt. Chem., (21 41, 52. Lederer and Manasse, J . prak!. Chem., [ 2 ] 6 0 , 224. Berichte, 27, 2409-1 1 (1894). Kleeberg, A n n dev. Chem.. 263, 283. Greene, A n n . der Chem., 2-19. Baeyer, Berichte, 5 , 25 (1871); 6, 280, 1094 (1872). Claus a n d Trainer, Berichte, 19, 3004 (1886). Moitessier, Jahresbevichle, 1866, 676-7. K r a u t , A n n . Chem.. 156, 123. Gerhardt, A n n . Chem. Phys., [31 7 , 215. Beilstein a n d Seelheim, Ann. Chem., 117, 83. Schotten, Berichlc, 784, 1878. Wohl, Ibid., 19, 1842. Tollens, Ibid., 17, 653. Gambier a n d Brochet, Comptes Rendus. 120, 557. Gerhardt, Ann. der Chem.. 87, 159. Kolbe and Lautem, A n n . der Chem., 115, 196. Manasse, Berichle, 27, 2409 (1894). Gazelle Chim. Italiana, 2, 1 (1872). Meer, Fabinyi, Steiner, Beuichte, 7, 1197-1201; 11, 283--7. O t t o Fischer, 4 n n . de7 Chem.. 206, 147. Michael, Berichle, 17, 20-21. Michael and Comey, Ibid., 19, 1388. Michael a n d Ryder, Amer. Chem Jour., 6 , 338. Trzcinski, Berichle, 16, 2838; 17, 499. Reed, J o u r . prakt. Chem., 34, 160 (on aldehydes and amines). Velden, J . prakt. Chem., [2] 16, 164; Jahuerberichte, 537, 1877. Lebach, Z e i l . angeiei. Chem., 22, 1600 (1912). Raschig, Z e i t . angew. Chem., 26, 1945 (1912). Abel, Berichle, 26, 3477 (1892); 19, 3316; 13, 1954; 2 0 , 140. Claisen, A n n . der Chem., 237, 261. Hosaeus, Berichte, 26, 3213 (1892).
'
T H E JOURNAL OF INDUSTRIAL A N D ENGINEERING CHEMISTRY
16
Meer. Ber., 7, 1200. Schroder a n d Prinzhorn a n d K r a u t , A n n . de7 Chem., 150-1. Tollens, Ber., 17, 6 5 3 (1884). Schiff, I b i d . , 11, 831. Lippmann a n d Strecker, I b i d , 13, 831. Laurent a n d Gerhardt, Jahresberichfe, 1850, 488. e Cech, Ber., 11, 248. Butlerow, .4nn. C h e m . , 116, 322. H o f m a n n , Berichte. 3, 152. Wohl, Bexichte, 19, 1840 (18861. Moschatos a n d Tollens, A n n . der Chem., 373, 271 (1893). Litterschied a n d T h i m m e . L i e h i g A n n . , 334, 49, 67. Chem. Centr., 7 S 2 , 949 (1904). Wohl, Ber., 45, 2046. Andwers and Daecke. Ber., 33, 3374. Lederer. J. prakt. Chem., [21 SO, 225. Lebach, Zeil. angew Chem., 1598-1909; Jour. SOL.Chem. I n d . , 8.3, 559 (1913). Baekeland, THISJOURNAL, 1, 149, 545 (1909); 3, 932 (1911); 4, 737 (1912); 5, 506 (1913). Baekeland, Trans. A m . Electrgchem. Soc., 16, 593 (1909). Jablonower, Jour. A m . Chem. Soc., 36, 811 (1913). PATENT REFERENCES
U. S. Pat. Ger. P a t .
MANASSE 526,786 1894 SPEYER 99,570 1897 SMITH
Brit. P a t . Ger. P a t .
16,247 1899 112.685 1899
GROGNOT
U. S. P a t .
LUST
Ger. P a t . U. S. P a t .
LEBACH-KNOLL 82 CO. Belg. P a t . 204,811 1907 WETTER,KNOLL & CO. Brit. P a t . 28.009 1907 WETTER,KNOLL & CO. Brit. P a t . 24,072 1908
391,436 1908
140,552 1902 735,278 . .
U. S. P a t .
12,880 1902 6 , 8 2 3 1903
Brit. P a t . Belg. P a t .
AYLSWORTH
1,020,593 . . 1,029,737 4 , 3 9 6 1911 232,899 1910
..
ELUMER Brit. P a t . Brit. P a t .
HENTSCHKB
FAYOLLE
Fr. Pat. Fr. Pat. Fr. Pat. Fr. P a t .
335,584 2,414 2,485 341,013
1903 1903 1904 1904
8,875 30,884 210,965 DE LAIRE 361,539 15,517 189.262
1905
STORY
Brit. Pat. Austrian P a t . Belg. P a t
Fr. P a t Brit. P a t . Ger. P a t .
..
1908 1905 1905 1905
FRIED. BAYER & CO.
Ger. Ger. Ger. Brit.
Pat. Pat Pat. Pat.
201,261 1907 285,588 . . 237,786 . . 26,317 1907 HELM
Brit. P a t .
25,216 1907
Ger. P a t .
157.553 E AEKELAND U. S. P a t . 939,966 941,605 942.699 942,700 942,808 942.809 942,852 949,671 954,666 957,137 982,230 1,018,385 1,019,406 1,019,407 1,019,408 LEDERER Brit. Pat. 17,693
KNOLL
Brit. P a t . Swiss P a t . Fr. Pat.
*
.. 1909 1909 1909 1909 1909 1909 1909 1910 1910 1910 1911 1912 1912 1912 1912 1897
NEWTON
28,009 1907 40,994 1907 395.657 . .
Brit. P a t . Ger. P a t .
10,790 1897 99,570 1898 TH 0M P S0N
Ger. P a t .
20,223 1903
DEPARTMENT OF I N D U S T R I A L RESEARCH UNIVERSITY O F KANSAS,LAWRENCE
A METHOD FOR DETERMINING THE AMOUNT OF ZINC CHLORIDE IN TREATED W O O D B y ERNEST BATEMAN Received November 6, 1913
T h e need of a reliable method for determining t h e a m o u n t of zinc chloride i n t r e a t e d wood has already been mentioned in THIS J O U R N A L . It was clearly shown i n a series of wood-impregnation experiments recently conducted b y t h e Forest Service. I n these experiments maple a n d red oak ties were t r e a t e d i n three ways: ( I ) With zinc chloride solution alone; ( 2 ) with zinc chloride solution a n d creosote i n emulsion; (3) with zinc chloride a n d creosote injected i e p a r a t e l y . T h e analytical method used until t h a t time h a d given
Vol. 6, No.
I
excellent results for the soft woods, chiefly hemlock a n d t a m a r a c k , b u t when applied t o t h e h a r d woods, particularly o a k , i t did n o t give concordant results, a n d failed entirely in t h e analysis of h a r d woods t r e a t e d with zinc chloride a n d creosote i n emulsion. T h e method of analysis described i n t h i s p a p e r is one developed t o overcome t h e difficulties encountered. So f a r as known, t h e complete method has not been described before, although, i n its s e p a r a t e p a r t s , i t is formed f r o m a selection of certain well-known methods modified for t h i s particular work. T h e problem t h a t confronted t h e Forest Service i n i t s experimental t r e a t m e n t s necessarily confronts also m a n y commercial t r e a t i n g plants a n d users of t r e a t e d t i m b e r . SEPARATING
INORGANIC
SALTS FROM
ORGAh-IC
MATTER
A choice is offered of t h r e e possible methods of freeing inorganic salt i n wood from t h e organic m a t t e r so t h a t t h e usual inorganic methods of analysis m a y be carried o u t . These methods are: I. Destruction of organic material by b u r n i n g ; estimating inorganic materials f r o m t h e ash. 11. Removal of inorganic material b y leaching or extraction with water, or dilute chemicals: analysis of t h e leaching solution. 111. Destruction of organic material b y chemical m e a n s ; analysis of t h e resulting solution. I.
DESTRUCTION
OF
ORGANIC
MATERIAL
BY
BURNIKG
This is t h e most simple method because i t requires n o special a p p a r a t u s , b u t i t is limited t o those inorganic materials which a r e nonvolatile even a t comparatively high temperatures. A portion of t h e inorganic salt would be lost, of course, if t h e t e m p e r a t u r e reached t h e volatilization point of t h e material. I n spite of t h i s limitation, t h e method is t h e one most generally used, a n d i t is suitable for m a n y inorganic materials. It should be remembered t h a t certain metals, which i n themselves are nonvolatile i n a n ordinary Bunsen burner, a r e volatile when i n t h e form of chlorides. I r o n a n d aluminum are notable examples. These metals a n d all of their salts, except halogen salts, a r e nonvolatile. T h e salts df zinc behave i n t h e s a m e m a q n e r , a n d t h e metal itself is volatile a t a b o u t 1000" C. It is obvious, therefore, t h a t a method of dest r o y i n g t h e organic materials b y burning c a n n o t be used i n t h e estimation of zinc chloride. In spite of these facts t h e idea is somewhat prevalent t h a t t h i s method can be used, a n d t h e following t e s t was therefore m a d e : Known a m o u n t s of a s t a n d a r d zinc chloride solution were added t o 5 grams of d r y sawdust, t h e excess of moisture carefully evaporated i n a drying oven, a n d t h e wood ashed a t as low a t e m p e r a t u r e as possible. T a b l e I gives t h e results of these tests. It is seen t h a t a t least I O per cent less zinc chloride was found i n t h e ash t h a n was a d d e d t o t h e s a w d u s t , a n d i n one case t h e difference was as g r e a t as j g per cent. F u r t h e r , t h e r e is no uniformity in t h e results, so t h a t a f a c t o r for compensating for t h i s loss can n o t be obtained. Any method, therefore, which includes t h e ashing of t h e wood is n o t reliable for zinc chloride determinations.
