954
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
adding the remainder of the prepared menstruum and then sufficient diluted alcohol to make the product measure one thousand milliliters.
It will be noticed t h a t t h e new method calls for a preliminary extraction with alcohol which contains not less t h a n 94.9 per cent of alcohol by volume; in other words, Cologne spirits” of commerce. Alcohol of this strength removes a resinous extractive from t h e beans which is precipitated in t h e form of a persistent cloud when t h e menstruum is diluted and which is not subsequently removed by percolation or b y any ordinary method of filtration. This colloidal material adds nothing t o t h e flavor of t h e extract, b u t on t h e other hand renders i t unsightly and unsalable. Another obvious disadvantage of t h e proposed method is the loss of alcohol which its use entails. Vanilla beans of average moisture content will retain I O per cent of the alcohol used when placed upon a filter t o drain, which is lost if t h e directions are followed t o expose t h e drug “to t h e air until all of t h e alcohol has evaporated.” The method would seem t o be of greater academic interest t h a n of practical value, which is unfortunate, inasmuch as t h e authorities having t h e enforcement of t h e food and drug laws in charge naturally attach great importance t o official methods of procedure. BAKEREXTRACT COMPANY SPRINGFIELD, MASS.
THE DETERMINATION OF BROMIDE IN MINERAL WATERS AND BRINES By W. F. BAUGHMAN AND W. W.SKINNER The chief sources of bromine in t h e United States are natural and artificial brines in which it is present as bromide associated frequently with small amounts of iodide. Bromine is obtained by appropriate treatment of t h e mother liquor, or “bittern,” obtained as a by-product in t h e manufacture of common salt. Owing t o t h e increased demand for bromine resulting in a greatly increased market value during t h e period of t h e war, search has been made for new supplies of brine rich in bromide. The authors had occasion t o examine a number of samples of brines and desired t o determine their bromide content. The colorimetric methods have been studied by t h e Association of Official Agricultural Chemists2 and b y Sweeney and W i t h r ~ w but , ~ are not entirely satisfactory. I n these methods t h e bromide solution is treated with chlorine water or some other oxidizing agent and t h e liberated bromine absorbed in carbon disulfide, chloroform, carbon tetrachloride, etc., and compared with suitable standards. Where t h e bromine present in the sample taken for t h e determination is greater than 5 mg., however, only approximate results can be obtained by these methods. A search of the literature revealed no method which could confidently be relied upon t o give correct results for bromine associated with t h e other constitu1 Read before the Division of Water, Sewage, and Sanitation of the American Chemical Society, a t the 57th Meeting, Buffalo, April 7 to 11, 1919. 1 J . A . 0.A . C., [I] 1 (1915), 97. a THIS JOURNAL, 9 (1917), 671.
Vol.
II,
No.
IO
ents usually found in brines, and i t was necessary t o attempt t h e development of a satisfactory method. The literature contains many methods for t h e selective oxidation of bromides in t h e presence of chlorides and t h e subsequent removal of the liberated bromine by steam distillation or b y aspiration. Vortman1 recommended t h e use of lead peroxide (PbOn) and acetic acid, t h e liberated bromine being removed by distillation. Cavazzi2 recommended barium dioxide (BaOz) and dilute sulfuric acid. Enge13 made t h e assertion t h a t ammonium persulfate ((NHJ2SPOs) and sodium acetate (NaC2H302) will liberate bromine and not chlorine a t a temperature of 70-80’ C. Berglund4 used a mixture of potassium bisulfate ( K H S 0 4 ) and potassium permanganate (KMn04) in t h e cold, and removed t h e bromine b y aspiration. He found, however, t h a t if sodium chloride were present in excess of I g., a small amount of chlorine would be s e t free along with t h e bromine. To overcome this difficulty he recommended two aspirations, the first t o concentrate t h e bromine content in a suitable absorbing solution from which t h e bromine is again liberated b y potassium bisulfate and potassium permanganate and removed b y aspiration, resulting in pure bromine being obtained. Baubigny and Rivals5 stated t h a t copper sulfate (CuSO4) and potassium permanganate (KMnOl) will liberate at room temperature bromine and not chlorine unless chloride is present in too large amount. Wysss used ferric sulfate (Fez(S04)~) and potassium permanganate (KMn04). White’ used aluminum sulfate (&(SO4)3) and potassium permanganate ( K M n 0 4 ) . Jannasch and Aschoff8 used acetic acid and potassium permanganate (KMn04). B ~ g a r s k y ,Benedict ~ and Snell,lO and Andrewsll suggested t h e use of iodic acid. Gooch and Blurnenthall2 used selenic acid, and Gooch and Cole13 telluric acid. The possibility of obtaining a clean separation of bromine and chlorine b y t h e use of most of these methods is dependent upon t h e Concentration of t h e chloride or of t h e oxidizing agent, or of t h e acidity of t h e solution. They are effective, therefore, only within narrow limits of concentration of the reacting solution and the possibility always exists t h a t either owing t o t h e addition of too much acid or oxidizing agent, or from t h e solution becoming too concentrated during distillation, some chlorine will be set free, or, t h e contrary conditions prevailing, all of t h e bromine will not be liberated. The double aspiration recommended b y Berglund disposes of t h e possibility of bromine being contaminated with chlorine, b u t i t seems t o have been overlooked or not approved by later investigators who . anal. Chem., 25 (1886), 172. Gam. chim. ital., 13, 174. 8 Compt. rend., 118 (1894), 1263. 4 Z . anal. Chem., 24 (1885), 184 6 Compt rend., 125 (1897), 527, 607. 5 Repert. anal. chem., 5 (1885), 238. 7 Chem. News, 67 (1888), 233. 8 2. anorg. Chem., 1 (1892). 144. 9 I b i d . , 10 (1895), 387. . 10 J . A m . Chem. SOC.,25 (1903), 809. 11 I b i d . , 29 (1907), 275. 18 A m . J . Sci., 36 (1913). 5 4 See also Cole, I b i d . , 38 CL914) 265. 18 Ibid., 37 (1914), 257. 1Z
2
Oct., 1919
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
by t h e methods proposed seem t o have sacrificed accuracy and reliability of results t o rapidity of procedure. Bugarskyl has pointed out another possible source of error in case t h e bromide is removed by distillation caused by t h e formation of a small amount of hydrobromic acid (HBr) from steam and bromine. Consequently, if t h e liberated bromine is driven o u t by distillation, it should be determined gravimetrically a n d not iodimetrically. W Y S S , ~recognizing the uncertainty of results obtained by methods already recommended, suggested a method which he thought would dispose of t h e possibility, theoretical or practical, of results being vitiated by t h e liberation of chlorine. He makes t h e statement t h a t in t h e presence of free chromic acid in excess, “chlorine in t h e native state will be found as chromyl chloride which a t room temperature is not volatile a n d not dissociated,” and asserts t h a t if a n y chlorine is set free it will combine with t h e chromic acid. His procedure is substantially as follows: Chlorides a n d bromides, which should be contained in a minimum of solution, are treated with a n excess of chromic anhydride (Cr03) (about 1 5 g.) a n d a few cubic centimeters of hydrogen peroxide (HzOz), and t h e liberated bromine removed by aspiration a t room temperature, absorbed in a potassium iodide ( K I ) solution, a n d titrated with thiosulfate. His recommendation, however, is supported by only seven experiments which are considered not sufficiently comprehensive t o substantiate it. It was, therefore, decided t o subject this proposed method t o a more thorough investigation. REAGENTS
(NaC1)-This was purified as described in a previous paper on t h e determination of i~dide.~ POTASSIUM BROMIDE (KBr)-A quantity of potassium bromide (C. P. reagent) was recrystallized several times f r o m distilled water and carefully dried. A solution was prepared by dissolving a n accurately weighed amount in distilled water. Upon standardizing gravimetrically b y precipitating and weighing as silver bromide, a figure for t h e bromine content was obtained which agreed within a few tenths of a milligram with t h a t calculated from t h e quantity of potassium bromide weighed out. CHROMIC ANHYDRIDE (Cr03)-This was a C. P. reagent which contained a small quantity of sulfuric acid, Check analyses were made a n d i t was determined t h a t t h e small quantity of sulfuric acid usually found in chromic anhydride does not affect t h e accuracy of t h e result. H Y D R O G E N P E R O X I D E (HzOz)-The ordinary 3 per cent U. S. P. article of commerce was used (free of acetanilide). POTASSIUM IODIDE-The salt used for t h e absorption solution was also a C. P. reagent which was tested a n d found t o be free from iodate. SODIUM
1 LOG.
CHLORIDE
cit.
f
Med. Klindk, 24 (1910), 288.
8
Tlus JOURNAL, 11 (1919), 563.
955
THIOSULFATE soLuTIoN-This was standardized against pure iodine which had been resublirned twice from a small quantity of potassium iodide.
kind of a suitable bottle a t hand.) In t h e first, t h e bromine is liberated a n d its removal by aspiration is facilitated by filling the cylinder t o about half its capacity with glass beads. It was found
0
\I i
‘
3 C
FIG. A APPARATUS
intake t u b e down AyReaction Cylinder. B and C-Absorpbeads tion Cylinders. E-Rubber Connections through the t o make t h e end of the tube somewhat pointed. A small funnel was joined b y a rubber connection to t h e outside vertical end of t h e intake t u b e so as t o make possible t h e introduction of liquids after t h e cylinder was closed. The other two cylinders contain t h e liquid for absorbing t h e bromine. The lower end of t h e intake t u b e of each cylinder was blown into a little ball, t h e horizontal circumference of which was pierced by fine holes in order t o break up t h e bubbles. It was found t h a t only a very small p a r t of t h e bromine escapes from the first absorption cylinder, so i t is safe t o connect i t with t h e second only by a rubber tube, provided t h e edges of t h e two tubes are close fitting. I n a former paper’ concerning t h e determination of iodide in mineral waters a n d brines, t h e authors suggested a method for removing iodide from t h e sample t o be used for t h e bromide determination. I n t h e present paper i t is assumed t h a t iodide has either been removed, as previously indicated, or t h a t i t is n o t present in amounts sufficient t o interfere with t h e bromide determination. T h e latter assumption often proves t o be a fact in t h e case of brines. EXPERIMENTAL WORK
T h e action of chromic anhydride on a bromide i n solution was investigated first. This reaction is represented by t h e following equation: zCr03 6HBr = Cr203 3Hz0 f 3Br2 T h e reaction cylinder was charged first with a layer of glass beads about I in. in thickness, t h e n
+
1
Lac. C i t .
+
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
956
with 1 5 g. of dry chromic anhydride, and finally enough glass beads t o half fill the cylinder. The first absorption cylinder contained I O g. of potassium iodide dissolved in 2 0 0 cc. of water, and t h e second, 3 or 4 g. potassium iodide in a like volume. After charging, t h e cylinders were joined together and a slow current of air drawn through. A measured volume of t h e standard potassium bromide solution was then added t o the reaction cylinder through t h e small funnel mentioned and t h e funnel washed with a small amount of water until t h e total volume added (including t h e potassium bromide solution) was about 2 5 cc. The velocity of t h e air current was then increased t o from 1/2 t o s/4 1. per min. and the aspiration continued for about I hr. The absorption cylinders were then emptied and t h e contents titrated with thiosulfate. By aspirating into a fresh potassium iodide solution, i t was determined whether or not all t h e bromine had been removed. Vol. of EXPT.CrOs Soln. No. G. Cc. 1 15 25 2 15 25
NaCl Present
G.
None None
TABLE I Br Time Standard2 Present Aspira- NazSzOs Bromine as KBr tion Soh. Found Error G. Hrs. Cc. G. G. 2 12.62 0.1003 0.1000 -0.0003 1 25.33 0.2006 0.2012 +0.0006 0.07 1 1 0.0602 7.50 0.0602 0.0000 0.10 2 7.60 0.0602 0.0602 0.0000 3/4 1 0.00 0.0802 10.12 0.0802 0.0000 a/a 1 0.00 0.0000 3 Faint trace 3 0.0000 Faint trace 3 0.0000 Faint trace 1 0.0602 7.60 0.0602 0.0000 2 0.00 1 0.0602 7.60 0.0602 0.0000 3 0.00 1 0.0602 7.62 0.0603 +O.OOOl 3 0.00 1 0.0602 7.63 0.0604 +0.0004 1 0.00 1 0.0602 7.65 0.0607 +0.0005 0.01 2 1 0.0602 7.68 0.0612 +O.OOlO 2 0.05
3
15
25
None
4
15
25
None
5
15
25
None
6 7 8 9
15 15 15 15
25 25 25 25
2.0 5.0 10.0 0.5
10
15
25
0.6
11
15
25
0.6
12
15
25
1.0
13
15
25
2.0
14
15
25
3.0
15
15
25
4.0
0.0602
16
15
25
5.0
0.0602
1 2 1 11
17
15
25
6.0
0.0602
18
15
25
8.0
0.0602
19
15
25
10.0
0.0602
20
15
25
5.0
0.0201
21
15
25
5.0
0.0281
22
15
25
5.0
0.0401
2 1 1 1 1 11
1
2 1' / z 1' 2 2 2 11/8 1' 2 3 1 2 2
2 2
2 2 2 15 24 2 25 5.0 0.0160 2 1 Stood over night before this aspiration. a 1 cc. = 0.00792 g. bromine. 23
15
25
5.0
0.0080
7.55 0.18 5.75 1.60 0.36 0.00 3.32 2.40 1.08 1.10
0.0612
+0.0010
0.0611
+0.0009
0.0634
f0.0032
0.0630
4-0.0028
0.0602
0,0000
0.10
4.40 1.40 2.10 0.03 0.03 3.00 1.05 3.30 0.15 0.10 2.24 0.30 3.50 0.08 4.90 0.20
1 .oo
0.03 1.92 0.10
0.0201
0.0000
0.0284
+0.0003
0.0404
+0.0003
0.0082
+0.0002
0.0160
+O.OOOO
The results obtained are t h e first five in Table I. It will be noted t h a t they are very satisfactory, those of Expts. 3, 4, and 5 agreeing exactly with the theory. The bromine was usually removed completely by aspirating for about I hr. Expts. 6, 7, and 8 show t h e action of chromic anhydride on a solution of sodium chloride. A very
Vol.
11,
No.
IO
faint trace of iodine was liberated in the potassium iodide solution which, after 3 hrs. aspiration, was equivalent t o not more t h a n 0.1 mg. of bromine. When, however, t h e solution of sodium chloride and chromic anhydride was allowed t o stand over night and then aspirated, using a fresh potassium iodide solution, slightly more iodine was liberated, an amount equivalent t o about I mg. of bromine. T h a t this was due t o t h e action of chromic anhydride on sodium chloride and not t o chromic acid carried over mechanically was proven b y aspirating a solution containing only chromic anhydride. Not a trace of iodine was liberated. Therefore, if t h e sodium chloride were free from bromine, and it is believed from the precautions taken it can be assumed t h a t it contained no trace of bromine, then chromic anhydride will, a t room temperature, liberate a trace of chlorine from chloride, especially if allowed t o act for a considerable length of time. I n Expts. 9 t o 2 4 , mixtures of sodium chloride and potassium bromide were used. A measured volume of t h e potassium bromide solution was added t o a weighed quantity of sodium chloride in a small dish, water added if necessary t o dissolve t h e sodium chloride, and t h e solution then evaporated nearly t o dryness. A few glass beads were added t o t h e reaction cylinder, then t h e mixture of sodium chloride and potassium bromide scraped in as completely as possible, and then beads added until the cylinder was half full. After connecting t h e cylinders and starting t h e aspiration, 1 5 g. chromic anhydride dissolved in I O t o 1 2 cc. water were added t o t h e reaction cylinder, followed by washings from t h e evaporating dish which had contained t h e mixed chloride and bromide sufficient t o bring t h e total volume added t o about 2 5 cc. as before. Bromine was determined in the presence of 0 . j or 0.6 g. sodium chloride with very satisfactory results. But when the quantity of sodium chloride was increased t o I g. or more, the results for bromine were too high, showing t h a t some chlorine was liberated. The results show t h a t more chlorine is liberated from a mixture of chloride and bromide t h a n from solutions of pure sodium chloride, t h a t is, the presence of bromide decreases t h e stability of sodium chloride toward chromic anhydride. This is similar t o t h e observation of Berglund' concerning t h e action of potassium bisulfate and potassium permanganate on solutions of mixtures of chloride and bromide and t o our experience concerning the action of ferric sulfate on solutions of mixtures of iodide and br0mide.l I n t h e latter case we showed t h a t while ferric sulfate would not liberate bromine from a pure bromide solution below a certain concentration, yet if free iodine were present, bromine would be liberated and t h a t free iodine decreased the stability of bromide toward ferric sulfate. Berglund explained his results by assuming t h a t some chlorbromide was formed and this seems plausible, b u t is difficult t o prove experimentally. With increasing concentration of chloride, the rate of evolution of bromine decreases. I n t h e absence 1 LOC.
cit.
T H E J O U R N A L OF I N D U S T R I A L A N D EiVGINEERING C H E M I S T R Y
OCt., 1919
of a chloride, i t is possible t o remove all t h e bromine in I hr. or less. I n Expt. 17 only about 89 per cent of t h e bromine was removed by 3 hrs. aspiration from a solution containing 6 g. of sodium chloride, assuming t h a t all the halogen evolved was bromine, although it is probable t h a t some of it was chlorine; in Expt. 18, about 87 per cent after 3l/2 hrs. aspiration from a solution containing 8 g. of sodium chloride; and in Expt. 19, about 81 per cent after 31/2 hrs. aspiration from a solution containing I O g. sodium chloride. If t h e mixture of chromic acid, bromide, and chloride is allowed t o react without aspirating for a considerable length of time, as over night, and then aspirated, t h e bromine comes over more rapidly. This is shown in Expts. 2 5 t o 31, Table 11. Also, more chlorine is liberated. TABLS11-MIXTURE
ALLOWED TO S T A N D OVER NIGHT ASPIRATION Br Time of Standard1 of NaCl Present Aspira- NazSzOa Bromine Soln. Found EXPT. CrOa S o h . Present as KBr tion G. G. Hrs. No. G. Cc. G. Cc. 15 25 10 0,0802 1 9.45 0.0816 25 3 0.6 13/4 0.25 0.0824 ' 26 20 20 ll/r 11.62 10 0.0802 7 n 05 27 15 20 5 0.0602 la/, 8.64 0.0684 28 20 20 10 0.0602 11/4 8.54 0.0676 29 15 20 10 0.0000 1 Trace 30 20 16 10 0.0000 3 0.05 0.0004 31 15 25 8 0.0602 ll/z 8.18 0.0649
BEFORB
VOl.
-
1
1
-.--
0.02
Error
G.
$0.0014 +0.0022 +0.0082 +0.0074
solution begins t o deepen rapidly. T h a t this is due t o bromine (or chlorine) liberated by t h e chromic anhydride and hydrogen peroxide was proven b y treating a solution of chromic anhydride which contained no bromide or chloride with hydrogen peroxide in a similar manner. No iodine was liberated in t h e absorbing solution. The strong oxidizing influence is evidently due t o t h e nascent oxygen. An inspection of t h e records of Expts. 32 t o 38, inclusive, will show t h a t t h e addition of hydrogen peroxide increases t h e rate of evolution of bromine and also t h e amount of chlorine liberated. EXPT.32-8 G. iVaC1, 00602 G. Br A S KBr, 15 G. CrOs, 18 Cc HzO Aspirated 1 hr., two 2-cc. portions HzOz added a t 20 min. intervals Standard NazSzOa soln. required = 7.55 cc. Added 2 cc HzOz, aspirated 2 hrs. Standard NazSzOa soln. required = 0.15 Added 2 cc. HzOl, aspirated 1 hr. Standard NazSzOs soln. required = 0.05 Bromine found = 0.0614 g.
It is evident from these last experiments t h a t free bromine can be readily removed b y aspiration from a concentrated sodium chloride solution a n d t h a t after t h e bromine is once set free, sodium chloride does not have a retarding influence on its removal. It is more likely t h a t t h e decrease in t h e rate of evolution with increasing amounts of chloride present is due to t h e sodium chloride removing some of t h e chromic anhydride from t h e sphere of action, possibly b y t h e formation of chromyl chloride (Cr02C12). Thus, due t o t h e decreased concentration of t h e chromic anhydride, t h e bromide would be oxidized less readily. From a solution containing in 2 5 cc. j g. chromic anhydride and 0.0600 g. bromine as potassium bromide, only z z per cent of t h e bromine was evolved after 31/2 hrs. aspiration. However, after standing over night, t h e remainder of t h e bromine was removed in about z hrs. It is, therefore, clear t h a t t h e use of chromic acid alone does not solve t h e problem of separating bromine and chlorine. The influence of hydrogen peroxide as recommended by Wyss was next studied. Upon t h e addition of hydrogen peroxide t o a solution of chromic anhydride, a rapid evolution of oxygen takes place and t h e chromic anhydride is reduced t o t h e insoluble chromic oxide (Crz03) according t o the equation 2HzCrOe -I-~ H z O = Z Crz03 -I- 2 0 2 -I- 5HzO. When hydrogen peroxide is added t o a solution of chloride, bromide, and chromic anhydride which is being aspirated into a potassium iodide solution, an effervescence occurs due t o t h e oxygen formed and in a few seconds t h e iodine color in t h e potassium iodide
Total = 7 . 7 5 ~ ~ . Error = +0.0012 g.
EXPT.33-10 G. NaCl, 0.0602 G. Br AS KBr, 15 G. CrOa, 18 Cc. H20 Aspirated 1 hr., two 2-cc. portions HzOz added a t 20 min. intervals Standard NazSzOa soln. required = 7.60 cc. Added 2 cc. Hz01, aspirated 2 hrs. Standard NazSzOa s o h . required = 0.23 Added 2 cc. HzOz, aspirated 1 hr. Standard NazSnOa s o h . required = 0.15
-
+0.0004 +0.0047
1 cc. = 0.00792 g. bromine.
957
Bromine found = 0.0632 g.
Total = 7.98 cc. Error = +0.0030 g.
EXPT.34-10
G. NaCI, 0.0602 G. Br AS KBr, 15 G. CrOa, 20 Cc. Ha0 Mixed and allowed t o stand over night in reaction cylinder Aspirated 2 hrs., three 2-cc. portions HzOz added a t 20 min. intervals Standard NmSzOa soln. required = 8.40 cc. Aspirated 1 hr. Standard NazSzOs s o h . required = 0.05 Added 2 cc. HzOz, aspirated '/z hr. Standard NazSzOa s o h . required = 0.10
-
Bromine found = 0.0677 g.
Total = 8.55 cc. Error = 4-0.0075 g.
EXPT.35-10
G. NaC1, 0.0802 G. Br AS KBr, 15 G. CrOa, 20 Cc. H10 Mixed and allowed t o stand over night in reaction cylinder Aspirated 2 hrs., three 2-cc. portions HzOz added a t 20 min. intervals Standard NazSzOs soln. required = 11.85 cc. Aspirated 1 hr. Standard NazSzOa s o h . required = 0.00 Added 2 cc. HzOz and aspirated 1/z hr. Standard Na2SzOs soln. required = 0.10
Bromine found = 0.0946 g.
EXPT.3&10
Total = 11.95 cc. Error = 4-0.0144 g.
G. NaCI, 0.0602 G. Br
AS KBr, 15 G. CrOa, 20 Cc. H20, 2 CC. HzOz Mixed and allowed t o stand over night in reaction cylinder Aspirated 2 hrs., two 2-cc. portions HzOz added a t 20 min. intervals Standard hrazsloa s o h required = 7.99 cc. Aspirated 1 hr. Standard NazSzOa soln. required = 0.00 Added 2 cc. HzO,, aspirated 1 hr. Standard Na&Os s o h . required = 0.10
Bromine found = 0.0641 g.
EXPT.37-10
Total = 8.09 cc. Error = +0.0039 g.
G. NaCl, 0.0802 G. Br
A S KBr, 15 G. CrOa, 20 Cc. HIO, 2 CC. HzOi Mixed and allowed t o stand over night in reaction cylinder Aspirated 2 hrs., two 2-cc. portions HzOz added a t 20 min. intervals Standard NazSzOa s o h . required = 10.60 cc. Aspirated 1 hr. Standard NazSzO3 soln. required 0.00 Added 2 cc. HaOz, aspirated 1 hr. Standard NaiSzOa s o h . required =) 0.10
-
-
Bromine found = 0.0847 g.
Total = 10.70 cc. Error = +0.0045 g.
958
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
EXPT. 38-10 G. NaCl, No KBr, 20 G. CrOs, 16 Cc. HzOz Mixed and allowed t o stand over night in reaction cylinder Aspirated 3 hrs. Standard NazSz03 s o h . required 3 cc. HzOz added, aspirated 1/z hr. Standard NazSzOd soln. required Aspirated '/z hr. Standard NazSzOa soh. required 3 cc. HzOa added, aspirated 1 / z hr. Standard NazSz03 s o h required
= 0.05 cc. = 0.10
T h e assertion of Wyss t h a t free chlorine in a solution of chromic anhydride will form chromyl chloride, which at room temperature is not dissociated a n d not volatile, and t h a t , consequently, this principle may be used for separating chlorine and bromine, is not according t o t h e experimental facts as we have found them. T h e method published by Wyss, therefore, is wholly unreliable. T h e difficulties pointed out, however, can be overcome by using t h e double aspiration principle first suggested b y BergIund in his bisulfate and permanganate method. T h e first aspiration serves t o concentrate t h e bromine content in a n absorbing solution which will change t h e free halogen t o halide, t h e absorbing solution being subjected t o a second treatment with chromic anhydride, resulting, on account of t h e small amount of chloride present, in pure bromine being evolved. T h e results in Table 111 were obtained by employing double aspiration. The reaction cylinder was charged as before with glass beads, t h e sample t o be examined, 1 5 g. chromic anhydride, and 2 5 cc. water. T o t h e first absorption cylinder were added 2 0 cc. and t o t h e second 5 cc. of a solution which contained in 2 5 cc. I g. sodium sulfite (Na2S03) and 0 . 2 g. sodium carbonate ( N a ~ C 0 3 ) . Sufficient water was then added t o each cylinder t o bring t h e volume t o about zoo cc. The sulfite reduces t h e bromine t o bromide and t h e reaction may be represented b y t h e following equation: NazS03 2Br H20 = zHBr Na2S04 T h e carbonate neutralizes t h e hydrobromic acid and prevents it from being volatilized during t h e subsequent evaporation. After connecting, t h e cylinders were aspirated for a short time until t h e contents of t h e reaction cylinder were thoroughly mixed. T h e small funnel on t h e reaction cylinder was then removed and t h e inlet tube of the cylinder closed by clamping a rubber t u b e over t h e end. I n order t o guard against a possible escape of bromine a t t h e ground glass stopper of t h e reaction cylinder, t h e pressure in t h e apparatus was reduced by sucking out a little air after closing. T h e apparatus was t h e n allowed t o stand over night. I n t h e morning, it was aspirated for about 3 hrs. and four portions of z cc. each of hydrogen peroxide added a t half-hour intervals. T h e absorbing solution was then evaporated nearly t o dryness, t h e residue dissolved in about 5 cc. of water and added, using 2 0 cc. water t o wash t h e evaporating dish, t o t h e reaction cylinder which had been previously charged with glass beads and chromic anhydride and connected with t h e absorption cylinders filled with potassium iodide solution. Aspiration was continued until all t h e bromine had been evolved (about I hr.), and t h e potassium iodide solution titrated with thiosulfate.
+
+
+
11,
No.
IO
T h e results are considered very satisfactory and indicate t h a t bromine may be determined in t h e presence of a s much as I O g. sodium chloride with a negative error of less t h a n I mg. a n d n o possibility of t h e results being vitiated by t h e liberation of chlorine.
= 0.00 = 0.10
Vol.
TABLE111-DOUBLE ASPIRATION EXPT.
No.
Standard Bromine Br Present as KBr NazSzOa Soln.1 Found G. G. cc.
0.0602 0,0642 3 6 0.0602 0.0602 10 0.0802 10 10 0 . 0802 0.0591 5 10 0.0591 10 0.0985 10 0.0788 1 cc. = 0.00792 g. Br.
39 40 41 .. 42 43 44 45 46 47 48 1
NaCl Present G. hTone
7.60 8.00 7.52 7.52 10.10 10.00 7.42 7.40 12.40 9.94
0.0602 0.0634 0.0596 0.0596 0 . oaoo 0.0792 0.0588 0.0586 0.0982 0.0787
Error
G. 0.0000 -0.0008 -0.0006 4.0006 -4.0002 -0.0010 -4.0003 -0.0005 4.0003 -0.000 1
T h e procedure recommended for t h e determination of bromine in a mixture of bromide and chloride, using t h e apparatus already described, is a s follows: Evaporate the sample of water or brine, which should not be acid (if necessary, add small amount of sodium carbonate), to dryness or nearly so. Charge the reaction cylinder by introducing first glass beads to a depth of about I in., then as much of the sample as can be scraped in, and finally enough glass beads to fill the cylinder half full. Make a solution of sodium sulfite and sodium carbonate of such a concentration that 2 5 cc. will contain I g. of sulfite and 0.2 g. of carbonate. Add 20 cc. of this solution to the first absorption cylinder, 5 cc. to the second, and dilute each to approximately 200 cc. Connect the three cylinders and draw through a slow current of air. Add 15 g. chromic anhydride dissolved in IO to 1 2 cc. water to the reaction cylinder, followed by washings from the evaporating dish which contained the sample, sufficient to bring the total volume added to about 25 cc. Aspirate until the contents of the reaction cylinder are in solution and thoroughly mixed, then discontinue, close the inlet tube with a small piece of rubber tubing and a clamp, and reduce the pressure in the apparatus slightly by sucking out some air in order to guard against any possible escape of bromine a t the ground glass stopper. Allow to stand over night, then aspirate with a rather strong current of air (about ' / z to 3/4 1. per min.) for 3 hrs., adding four 2-cc. portions of 3 per cent hydrogen peroxide at 30 min. intervals. Stop the aspiration and evaporate the contents of the two absorption cylinders nearly to dryness. Clean out the reaction cylinder and freshly charge with glass beads and 15 g. chromic anhydride. Into the first absorption cylinder, put I O g. potassium iodide dissolved in zoo cc. of water, and into the second 3 or 4 g. in a like amount of water. Connect the apparatus, draw through a slow current of air, and transfer the contents of the evaporating dish to the reaction cylinder by means of the small funnel, using 2 5 cc. of water. Aspirate with a rather strong current of air until all the bromine is evolved (about I hr.) and titrate the potassium iodide solution with thiosulfate. SUMMARY
Chromic acid in concentrated solution liberates bromine from bromides quantitatively a t room temperature, and t h e bromine may be removed b y aspiration. It liberates only a trace of chlorine from chlorides and forms probably chromyl chloride which remains in solution. When chromic acid acts on a solution of chlorides and bromides, some chlorbromide is formed which is removed with t h e bromine by aspirating.
1919
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
959
ple I O O g. of the fresh berries were placed in a flask and covered with 300 cc. of 95 per cent alcohol, then allowed t o stand until needed, being shaken two or three times per week. At t h e end of 5 mos. t h e mixWATGRLABORATORY, BUREAUOF CHEMISTRY WASHINGTON, D. C. ture was boiled for 2 hrs. under a reflux condenser, filtered, and t h e treatment repeated two or three times with fresh portions of solvent, t h e last extract THE COMPOSlTION OF THE FRUIT OF THE VIRGINIA obtained being colorless. The combined extracts were CREEPER, AMPELOPSIS QUlNQUEFOLIA evaporated in, vacuo t o a thick syrup and taken up By GEORGED. BEAL AND EDWARD A. GLENZ with water. The solution was washed three times with Received February 28, 1919 Ampelopsis qzti.nqidefolia, family Vitaceae, is a climb- ether t o remove t h e green oily matter and the resulting woody vine found in woods and thickets through- ing cherry-red solution was clarified with lead subout t h e central portion of t h e United States and from acetate and alumina cream. The filtrate was then Quebec t o Manitoba in Canada. It is known variously made ,up t o a volume of 2 0 0 0 cc. and used for t h e as the Virginia creeper, American ivy, five-leaved various sugar determinations, while t h e precipitated ivy, woodbine, and false grape. The fruit consists lead compounds were a second source of material of small, blue, one- t o four-seeded berries, and is not for t h e study of t h e organic acids. The total reducing sugars were determined by t h e edible. Some few cases of poisoning have been ascribed t o it, but proof of its poisonous nature is method of Defren-O’Sullivan both before and after lacking. The fluid extract of t h e leaves has been used inversion, and calculated as dextrose. The reducing sugar before inversion amounted t o 9.97 per cent medicinally. The various methods for t h e extraction of the prox- and after inversion t o I 2 . zg per cent of t h e fresh fruit. imate constituents were carried out according t o The sucrose calculated from t h e increase in reducing Parsons.’ T h e solvents used were benzene, methyl power after inversion amounted t o 2. 26 per cent of alcohol, water, and sulfuric acid. After each extrac- t h e fresh fruit, while t h a t determined by direct and tion t h e residue was dried and weighed t o determine invert polarization in another portion of t h e same extract by t h e Clerget method was found t o be 2 . 2 2 t h e amount of material extracted. per cent. B E K Z E N E EXTRACT-100.6 g. of the air-dried fruit The amount of levulose as determined from polariwere extracted with benzene for 8 hrs. in a Soxhlet extractor. After evaporation of t h e benzene an oily scope readings at zoo and b o 0 , according t o Wiley’s extract remained, having a greenish brown color and formula,l was found t o be 7.00 per cent, and possessing a somewhat aromatic odor. The extract t h e dextrose t o be 3.67 per cent. Only one osaamounted t o 28.91 per cent. This contained a yel- zone could be obtained from t h e solution. When low oil and 0.394 per cent of a waxy substance insolu- heated rapidly so t h a t a temperature of zoo0 was ble in low boiling petroleum ether. The extract reached in 3 min. i t melted a t 204’ and was evidently glucosazone. This would indicate t h e absence of yielded 0 . 1 4 per cent of neutral ash. M E T H Y L ALCOHOL EXTRACT-The residue from the other osazone-forming sugars than dextrose and levubenzene extraction was extracted with methyl alco- lose. hol, sp. gr. 0.848, for 8 hrs. in a Soxhlet extractor. ORGANIC ACIDS-In all, three different samples were This extract had a purplish red color similar t o t h a t used for t h e identification of t h e organic acids. The Qf grape juice, and after standing for some time de- methyl alcohol extract was evaporated t o a syrup, posited some green, solid matter. The total material taken up with water, and t h e acids precipitated with extracted by t h e alcohol amounted t o 2 1 . I per cent lead subacetate and alumina cream. This precipiand was used in qualitative tests for t h e sugars and t a t e and t h e precipitate used in the clarification of organic acids present. the sugar solution were used for t h e qualitative tests, COLI) W A T E R EXTRACT-The residue after t h e alco- reactions being obtained indicative of oxalic, malic, hol extraction was macerated with cold water for 16 tartaric, citric, and tannic acids. For t h e quantitahrs., filtered, and after thorough washing with cold tive determination a weighed quantity of t h e dried water was expressed. A neutral red solution was ob- fruit was digested with z per cent hydrochloric acid tained, the water-soluble extractives amounting t o for I hr. a t t h e boiling temperature, and after filterI . 8 per cent. ing, clarified with boneblack. After concentrating SULFURIC ACID EXTRACT-After t h e water treat- t h e solution t h e method of Barfoed2 was used for t h e ment, t h e residue was boiled under a reflux condenser separation. The results, calculated t o the air-dry for I hr. with I per cent sulfuric acid, and then macera- fruits, were oxalic acid I . 2 1 per cent, and citric acid t e d in t h e cold for z days. The mixture a t no time 0 . 58 per cent. gave a reaction for starch with iodine. A reddish oIL-The o;l was t h e most interesting part of t h e purple, syrupy extract was obtained, containing 14.7 whole fruit. As t h e plant belonged t o the same family per cent of solids. as t h e grape, we expected t o find an oil similar t o grape SUGARS-TO avoid any discrepancies due t o chem- oil. The air-dried fruit was crushed between two ical change, t h e sugars were determined in a portion boards and I O O g. of clean seeds picked out. These of t h e fresh fruit. At t h e time of gathering t h e sam1 Wiley’s “Agricultural Analysis,” 8. 272.
Details of a method making use of t h e principle of double aspiration are given for the determination of bromine in the presence of large amounts of chlorides.
1
A m . Chem. J . , 1 (1880), 377
9
2.anal. Chem., 7 (1868), 403.
