Hydrated Titanium Oxide Thermal Precipitation from Titanium Sulfate Solutions ARTHURW. HIXSOKAND WALTERW. PLECHNER Department of Chemical Engineering, Columbia University, New York, S. Y, A survey of the literature showed that there exists no quantitative data on the hydrolysis of titanium sulfate solutions other than that concerned with its hydrolytic separation in quanlitative analysis wherein excessively dilute solutions are used. It was thought therefore that, since the hydrolysis of titanium sulfate solutions is of great importance in the industry, a quantitative study of some of the factors affecting the precipitation reactions would prone of interest and importance. This invest igation was carried out with solutions cocering the range of titanium oxide concentrations and of acidities used in the industry in order to determine the injluence of titanium concentration and acidity on the course of precipitation. Pure solutions of titanium sulfate were prepared by hydrolyzing redistilled titanium tetrachloride in the presence of titanous chloride and oxalic acid, dissolving the hydrated titanium oxide so obtained in sulfuric acid and crystallizing titanyl sulfate dihydrate which was dissolsed in water acidified with sulfuric acid to form the desired solutions. These solutions of titanium sulfate, sarying in concentration f r o m 1 to 10 per cent of titanium oxide by weight and of selected acid concentration, were hydrolyzed by boiling under a reflux condenser, or in some cases heating in sealed jars at C., and the progress of the hydrolysis ob100 serned analytically.
I
Precipitation, in the titanium sulfate solutions studied, does not increase regularly with dilution. T.l/ith increasing concentration, precipitation in a gieen time period decreases to a m i n i m u m value; with further increase in concentration, precipitation increases to a m a x i m u m and then decreases again. The maximum and minimum poinfs in the precipitation occur at specijk concentrations of titanium oxide at each acidify used. I n the industrial precipitation of titanium oxide by hydrolysis it is necessary to regulate the acidity as well as the titanium oxide concentration in order to obtain high yields economically. With increased acidity, the titanium oxide concentration qf the titanium sulfate solutions in which maximum precipitation occurs is lower. The m a x i m u m percentage precipitated in the time periods studied decreases with increased acidify. Titanium sulfate solutions may be more than 95 per cent precipitated by boiling for less than 8 hours when such contain 6 to 10 per cent of titanium oxide, and the ratio of excess sulfuric acid to titanium oxide is one or less than one. Titanyl sulfate dihydrate is crystallized by heating solutions of titanium sulfate in which the titanium oxide concentration is more than 6 per cent and the ratio of excess sulfuric acid to titanium oxide is 2 or more, It is possible i n this way to prepare relatively pure water-soluble crystalline titanyl sulfate on a n industrial scale.
NDICATIVE of the increasing importance of titanium is the fact that the United States Department of Commerce has published a review which summarizes the titanium industry, with special reference to its economic aspects (17). By far the largest use of titanium at the present time is in pigments for paints, lacquers, enamels, rubber, paper, cosmetics, and soaps. The first use of titanium compounds as pigments (17) was about 1870 by Overton ( I S ) who mixed rutile with a bituminous vehicle for use as a resistant paint on ships’ bottoms, and by J. W.Rylands, a Birmingham varnish maker, who used powdered ilmenite ore as a black pigment. I n 1893 Kidwell obtained a patent covering a paint containing rutile (‘7). I n 1908 the Xorwegian Government made researches for the purpose of extracting titanium from titanium ore and of finding a use for it. Titanium oxide was extracted (by means of sulfuric acid) and found to have high hiding power. Further research into titanium white as a pigment was stimulated a t about this time by legislation in Europe against the use of white lead, However, it was not until 1915 that Barton secured his patent covering the first chemically prepared, substantially iron-free titanium pigment (1). I n the preparation of titanium pigments, whether of pure
titanium oxide or a composite, all processes involve the hydrolysis of a titanium solution to precipitate, either alone or on an extender, a hydrated oxide which is dried and calcined to form the pigment (6). I n spite of the long period of time (some 140 years) that titanium has been known, and of its present increasing importance, there exist no quantitative data on the hydrolysis of titanic sulfate solutions other than those concerned with its hydrolytic separation in quantitative analysis. Therefore, this investigation was carried out with solutions covering the range of titanium oxide concentrations and of acidities commonly used in the industry. While the practical application of the hydrolysis of titanium sulfate solutions for the industrial large-scale production of titanium oxide, or, more exactly, hydrated titanium oxide which is calcined to produce the oxide, may be found in the patent literature, very little fundamental information is contained therein. Rossi and Barton (14) describe the hydrolysis of titanic sulfate solutions which contain 0.5 to 3 per cent of titanium oxide and about 0.15 t o 0.2 per cent of free sulfuric acid. The product obtained by boiling such a solution is called a basic titanium sulfate and contains titanium oxide (70 t o
262
March, 1933
I N D L S T R I A L. A N D E K G I N E E R I K G C H E h l I S T R Y
80 per cent), sulfuric anhydride ( 5 to 10 per cent), and combined water (15 to 20 per cent). They state that the composition of the product nil1 vary a little depending on conditions of precipitation, particularly as regards the titanium oxide and free acid concentration of the solution; excessive dilution yields products lower in sulfuric anh3,dride. Fladniark (!) states that precipitation of titanium hydrates from aqueous solutions is thought t o depend upon the hydrolytic decomposition of the titanium salts, resulting in a complete or partly colloidal solution of the titanium hydrates. The precipitation is then based upon the fact that the colloidal titanium hydroxide is caused to coagulate, whereby it is precipitated in a comparatively insoluble condition. As a result, the hydrolytic equilibrium is disturbed and a new quantity of colloidal titanium hydroxide is produced by deconiposition of the remaining titanium salts. This new quantity of hydroxide is again coagulated, and so on. Solutions of 80 to 350 grams titanium oxide per liter and from 82 to 410 grams per liter of sulfuric acid, not bound t o other ions than titanium, are convenient. Precipitates obtained from such solutions show from 15 to 25 per cent combined water, from 3 to 7 per cent sulfuric anhydride, from a trace t o 2 per cent ferrous oxide, and the remainder as titanium oxide. Blumenfeld ( 2 ) describes the preparation of a titanium sulfate solution containing 15 to 25 per cent titanium oxide. Such a solution, on being heated below the boiling point (95" t o 130" C.), precipitates titanyl sulfate (TiOS04.2Hz0) in the form of needle-like microscopic crystals, Mecklenburg ( 1 1 ) states that, whenever titanium salt solutions are subjected t o hydrolysis, a certain lag or induction period is observed between the commencement of the operation and the first visible precipitation; it is believed that during this induction period colloidal particles of hydrated titanium oxide form in the solution, which particles serve as centers for the accumulation of additional hydrolyzed particles and result in the formation of aggregates which produce a visible precipitate. I n support of this he found that, if titanium hydroxide was precipitated by neutralization of a solution with sodium hydroxide and a small amount of the suspension was added t o a hydrolyzable titanium solution, the mixture appeared turbid until the temperature had been raised above 50" to 60' C. At about this temperature the turbidity disappeared. When heated a t about TABLE1. ORIGISAL7 S O L N. CONCN. 1 hr. 2 hr.
% by 100 101 102 103 104 105 106 107 108 109
EXPERIhIESTAL PROCEDURE
All chemicals used were the best c. P. reagent grade chemicals obtainable, with the exception of titanium sulfat'e. Distilled water was used throughout. Iron-free titanium sulfate is not commercially available. The so-called c. P. titanium sulfate imported by chemical supply companies has a n appreciable iron content and is not completely soluble. It was therefore necessary to prepare iron-free titanium sulfate in this laboratory. Titanium tetrachloride was used as the starting material. This was redistilled once at 134" to 138' C. About 400 cc. of the titanium tetrachloride were dissolved in 1400 cc. of wat'er. The resulting solution of titanium tetrachloride was reduced by means of stick zinc to a light brown color, and 100 cc. were further reduced to a deep violet color. The solution was reduced until distinctly colored by the presence of titanous ion in order to insure that all the iron present was in the ferrous state to prevent hydrolysis and precipitation of ferric iron with the titanium. Dilute oxalic acid solution also prevents precipitation of any iron with the titanium (10, 16). The highly reduced titanium tetrachloride solution was added drop by drop to a boiling solution of 10 grams of oxalic acid in 10,000 cc. of water contained in a round-bottom flask. Then 800 cc. of the lightly reduced solution were added dropwise. Boiling was continued for a total period of 6 hours, hot water being added from time to time to maintain constant volume. These precipitation conditions were arrived a t after considerable experimentation. Comparatively small changes in initial concentration of titanium or oxalic acid, too rapid addition of the titanium tetrachloride to t,he boiling solution, or change in the total volume result in the precipitation of a colloidal unfilterable form of metatitanic acid. The precipitate of metatitanic acid, obtained as described, Tvas filtered on several large Buchner funnels and washed with boiling hot water until the washings were free from chloride. The precipitate was sucked dry and transferred to a beaker, and about 1.25 times its weight of concentrated sulfuric acid added. The mixture was gently boiled until crystallization of the mass took place. After cooling, an equal weight of water was added and the whole allowed to stand for several days with occasional stirring. The resulting solution was clarified by filtration with Super-Cel which had been previously washed free from iron with hot hydrochloric acid and then free from chloride with hot water. The solution so prepared was free from iron (ammonium thiocyanate test) and chloride (silver nitrate test). Portions of this solution were then concentrated to 15-18 per cent titanium oxide by evaporation below 60' C. under reduced pressure.
8 hr.
UOI.
13.72 12.21 10.76 9.44 8.14 6.69 5.37 4.08 2.74 1.44
from titanium sulfate solutions nor that such may be a component part of the molecule-i. e., the product precipitated may be a basic sulfate as claimed by Rossi and Barton ( I , $ ) or a mixture of metatitanic acid and basic titanium sulfate.
PRECIPIT~4TIONO F ~ I T A N I U MSULFATE SOLUTIONS WITH A4CIDITY
PERCES :TAGE Ti02 B Y VOLUME3 hr. 4 hr. 5 hr. 6 hr. 7 hr.
12.59 11.82 12.36 11.09 IO.92 10.39 10.85 10.97 10.07 9.16 8.68 8.40 10.01 9.30 8.71 8.05 7.21 5.34 9.16 8.25 7.78 5.10 0.724 0.980 7.51 7.02 5.46 0 , 7 8 4 0 . 7 3 1 0.910 5.82 3.11 1.76 1.96 2.19 2.24 1 . 2 1 2.04 3.47 3.33 2.98 3.11 2.85 3.56 2.28 2.17 2.23 2.48 1.28 1.10 1.54 0.953 1 . 0 1 0.696 0,599 0.491 0.440 0.398 0.380 0,322
9.95 7.68 6.03 0.374 1.01 2.24 2.97 1.96 0.585 0.359
9.41 7.27 1.24 0.253 0.935 2.22 2.98 1.73 0.408 0.250
100" to 105' C. for 3 hours, the titanium oxide was almost entirely precipitated, the yield being in excess of 95 per cent. Under the same conditions, a n unseeded solution yielded no more than 35 to 70 per cent titanium oxide. As little as one per cent of seed (based on the weight of seed titanium oxide as compared to the precipitated titanium oxide) was sufficient to produce the result. Mecklenburg kielieves that the titanium hydroxide added is not actually dissolred on heating but is comerted to some colloidal form which promotes the hydrolysis of titanium sulfate solutions. Outside the patent literature there does not seem to have been any recognition of the sulfuric anhydride content of the product obtained by thermal precipitation of hydrated titanium oxide
263
ORIGINAL
7 -
C O N C N . 1 hr. % b y weight 9.91 8.24 9.37 11.14 8.35 6.99 7.51 2.92 6.67 7.79 5.64 13.03 4.58 35.39 3.66 44.16 2.55 46.91 1.32 56.29
FACTOR 0.52
-
Ti02 PRECI lPIT. 1.24 sulfate of the o r i g i n a l v $1.22 hydrolysis products may 1.20 h a r e b e e n a l t e r e d in U 5 1.18 w a s h i n g them free of u adhering solution. Hyxl.16 drated titanium o x i d e u, 1.14 precipitated from a tita1.12 nium tetrachloride solu1.10 tion b y boiling m a y 1.08 readily be washed free of c h l o r i d e w i t h h o t 1.06 w a t e r ; on t h e o t h e r hand, when a s i m i l a r 1 8 3 4 5 6 7 product prepared from Percent T i t a n i u m Oxide a titanium FIGURE 4. SPECIFICGRAVITYAT tion is washed with hot 2 j o C. O F T I T . ~ N I U M S U L F A T E w a t e r , the s u l f a t e is SOLUTIONS WITH ACIDITYFACTOR OF 0.51 washed out very slowly, and even after prolonged washing (for a week or longer) small amounts of sulfate are present in the washings. Probably combined sulfate is gradually removed by continued washing. It may be noted, however, that the statement of Rossi and Barton (14) that excessive dilution during hydrolysis yields products lower in sulfuric anhydride is partially confirmed, in that the sulfuric anhydride content increased with increasing dilution of the original solution from 4.68 per cent sulfuric anhydride in the precipitate from the solution which originally contained 9.91 per cent titanium oxide, t o 8.28 per cent sulfuric anhydride in the precipitate from the solution which originally contained 4.58 per cent titanium oxide; but the precipitated titanic acid from the lowest concentration (originally 1.32 per cent titanium oxide) shows a lowered sulfuric anhydride content-i. e., 6.00 per cent.
hlarch, 1933
INDUSTRIAL A N D ENGINEERING CHEMISTRY
PRODUCTS OF TITABLE111. ANALYSISOF PRECIPITATION T.4SIUM SULFATE S o L U T l O N s WITH ACIDITYFACTOR 0.5 Ti01 IS OnrcIxiaL SOLS.
Ti02 HYDROLYZED I S
SOLS. 8 H o o n s % bu 5% - wt. I _
100 103 106
109 a
9.91 7.51 4.58 1.32
31.4 97.3 44.4 81.8
MOISTURE
% ._ 6.67 9.15 7.92 7.11
TiOP
SOaa
H20a
%
%
%
%
4.68 7.31 8.28 6.00
7.61 7.65 8.28 7.4Ll
99.50 99.50 99.69 100.00
87.21 84.54 83.13 86.51
ToT.AL~
redistilling distilled water once from alkaline permanganate and once from a dilute phosphoric acid solution through a block tin condenser.) The pycnometer full of the solution was placed in a thermostat a t 2.5" =t 0.05' C. for 2 hours, the volume was adjusted, and the whole was then weighed. The specific gravities and densities so obtained are given in Table IV. TABLE Iv. SPECIFIC 'GRAVITY 4 N D DENSITYOF T I T A S I ~ M SULFATE SOLUTIOSS WITH ~ ~ C I D I T FACTOR Y 0.51
On d r y basis
Deviation of the physical properties of a series of solutions from the l a v of mixtures may be indicative of chemical change. Dennison (3) states that, if on the mixing of two substances no chemical reaction occurs, the mixture law holds for the physical properties of the mixture; I. e., the physical propert i e s of a m i x t u r e a r e additive. H e also showed that, if any compound formation takes p l a c e on m i x i n g , the deviation from the mixture l a x is proportional to the amount of compound formed. At any composition of the mixture at which the deviation becomes a m a x i mum, the a m o u n t of compound formed must reach a maximum. If a chemical c o m p o u n d is formed, the number of I 2 3 4 5 6 7 8 9 10 moles of compound is a t Percent Titanium Oxide FIGURE5 . DEJIATIOUOF OB- a maximum w h e n t h e m i x t u r e has the same SERTED FROM C4LCULATED D E n S I T Y FOR TITANIUM S U L F ~ TSOLUTIOR~S E composition as the comWITH ACIDITYFACTOR 0.51 pound; i. e.. the maximum d e v i a t i o n of the property curve occurs a t the composition of the mixture representing the formula of the compound. The position of the maximum in property-composition curves shifts with changes in temperature. However, if the deviations of the property measured from that calculated according to the law of mixtures are plotted against the composition, the positions of the maxima are the same a t all temperatures. As previously noted, it is possible for the titanium in the solutions used to exist in various chemical combinations with the sulfuric acid and water present. Variations in the form of chemical combination, or the predominating form of combination, might account for the difference in the course of precipitation in the various dilutions which was observed. Measurement of the specific gravities of a series of solutions of different titanium oxide concentrations and a constant acidity factor of 0.5 was planned to determine whether the formation of compounds or hydrates in these solutions could be confirmed by maximum deviations in this physical property curve. -4 stock solution was prepared for these determinations which showed on analysis: TiOi, % by weight T o t a l Hzs04, % Equivalent HzSOd,
267
%
9 87 29.29 24.23
Excess HsSOi, Acidity factor
yo
5.06 0.51
This solution was then diluted to the desired titanium concentrations. The titanium oxide concentrations were determined gravimetrically so as to insure the greatest possible accuracy. The specific gravity of each solution was determined by weighing in a 25-cc. glass-stoppered Regnault pycnometer. (This was previously calibrated a t the temperature to be used with triple-distilled water, prepared by
Ti02
SP G n
DESSITY
% by w e i g h t 9,909 9.176 8.374 7.530 6.670
1.3819 1.3449 1,3084 1.2727 1.2358
1.3774 1.3109 1,3045 1.2689 1.2321
TiOz
SP GR.
5.750 4.737 3.65 2.49 1.29
1.1986 1.1602 1.1215 1.0821 1.0416
vu by w e i g h t
DEKSITY 1.1950 1.1568 1.1181 1.0788 1,03&5
I n order t o check these results and to secure more data, a second series was run, adjusting the concentrations of titanium oxide so that they would fall between those of the above group. The same original solution, pycnometer bottles, and thermostat v-erp used. The values obtained are given in Table T'. TABLEv. SPECIFICGRAVITYA S D DESRITT OF TIT.%SIT-.\I SULF.kTE S O L U T I O S S WITH -4CIDITY FACTOR 0.51 Ti02
SP.G R .
DENSITY
7uby weight 9,909 9.504 8.767 7.965 7.112
TiOz
SP. G R .
DENSITY
% by w e i g h t 1.3815 1.3621 1,3270 1.2915 1.2547
1.3774 1.3580 1.3230 1.2876 1.2509
6.204 5.230 4.224 3.04 1.88
1.2170 1.1792 1.1410 1.1018 1.0618
1.2133 1.1757 1.1376 1.0985 1.0566
I n Figure 4 the specific gravity values are plotted as the ordinates and the titanium oxide concentrations of the solutions as the abscissas. The curve so obtained for the titanium sulfate solutions rises as a straight line from 1.3 per cent titanium oxide to 4 per cent tita.032 nium oxide a n d 330 from that point the ,028 specific gravity in,026 creases more rapidly ,024 with increasing concentration as indi,022 cated in the figure. .9 ,020 Such behavior is to + ,018 be expected, as only z.016 in an ideal solution ,014 would the specific x c.012 gravi ty-concentra67 tion curve continue ,010 as a straight line be,008 yond t h e most ,006 dilute solutions. ,004 The deviation ,002 of the d e n s i t y obs e r v e d from that 0 3 6 9 12 15 18 21 24 27 30 calculated a c c o r d Percent Sulphuric Acid i n g t o t h e l a w of FIGURE6 . DEVI-~TIOS OF OBSERVED m i x t u r e s w a s de- FROM CALCULATED DENSITYFOR SULFURIC ACID SOLUTIOSS termined, using as t h e densities a t 25' C. of the components: titanium oxide 3.9, sulfuric acid 1.8255, and water 0.9971. The densities of water and sulfuric acid were obtained from the International Critical Tables. The value of 3.9 for titanium oxide represents the average of many determinations made for the Titanium Pigment Company, Inc. Recently, the figure 3.862 has been published by Roth and Becker as the density of titanium
.p
I N D U S T R I A L A N D E N G I N E E 11 1 N G C I1 E M I S T 11 Y
268
oxide at 21" C. (15). The deviatious so obtained (Table VI) were plotted against the titanium oxide concentration (Figure 5) in application of the theory of Dennison (9) that the deviation of a property from the mixture law indicates compound formation, and the point of maxinium deviation the presence of a compound identical in composition with that of the solution regardless of the temperature at which the measurements may he made. For purposes of comparison the density deviation from the calculated for sulfuric acid solutions in a eoncentration range covering tlle concentrations of total acid prescnt in the titanium sulfate solutions is given in Table VI1 and Figure 6. TABLE \-I. h Y l h T l O X OF OBsenvED DENSlTU FKOM CAI,CULATED x'oa SOLUTIOXB WITH ACIDITYFACTOR 0.5
O ~ a v n . CATXD.DnvinD~~~~~~D ~ ;
nol %.bv
waroh6
9.909 9.504 9.176 8.767 8.374 7.965 7.530 ?.I12 8.670 6.204
1.3774 1.3580 1.3409 1.3230 1.3015
I 6'274 I . 5087 1.4881 1.4663 I . 4446 1.4233
1.2876 1,2689 1.40(10 1.280!1 1.3777 1.23?1 1 3540 1.2133 1 . a288
0 . 1,m> 0.1477 0.1472 0.1433 0.I401
0.1357 0.1311 0,1188 (I. 1?19 0.ll55
4
5
2
b
9 I0
11
12 13 14 15
tmwht 5.750 5.230 4.737 4.224
3.65 3.04 2.49 1.88 1.29
1.I!I50 I .?!47R 1,1757 I.Yi8I 1.1568 1.2508 1.1376 1.2231 I.1181 1 . 1 9 2 4 1.0085 l . M O 7 1.0788 1,1303
N
0 102s 01024
0.0938
o.nnss 0.0745 0,0012 0.0515
1.0977 0.0391 1.0385 1.0661 0.0?76 1.0588
9%
9% I 2 B
Oirauo. CALCD,DavwT ~ ~ O , ncivsrTr ~ ~D ~~ T ~~~ % "2
1.0038
1.0104
1.0169 1.0234 1.0300
1.0367
1.0434 1.050? 1.0571 1.0640 1.0710 1.0780 1.0850
1.0922 1.0D94
1.0054
1.0137
0.0016
0,0033
0.0051 ll.0068 i.a,n:fi 0.0085 0.0101 I.046R
1.0220 1.0502
16
17 18 19 20 21
1.1017 1.1141 1.1215 1.1290
1.1305
I . U ~ S I o.0117 1.0634 1.0717 1.0799 1.0882
0.0132
22
1.1441 1.1917
23 24
1.1671
1.0065 1.1046 1.1132 1.1212
0.0185 0.0198
0.0146 0.0159 0.0172 0.0210 0.0218
25 26 27 28 29 30
1.1594 1.1750
1.1829
1.1sn9
1.1296 1.1379
0.0229
1.1645 1.1628 1.1711
0.025s
1.1462
1.1793 1.1876 1.1959
1.2042 1.2126
i.1!289
1.2207 1.2220
1.2150
....
1.2009
1.2373
0.02PS 0.0247 u.0283 0.0270
0.0276 0.0282 0.0287 0.0292
0.0298
o.om
0.0301 0,0304 0.0806
The density deviation curve for the titaiiiuni sulfate solutions, acidity factor 0.5 (Figure 5), docs not &ow R m&mum in the concentration range studied, and therefore the composition of any compounds which are present in these solutions is not indicated by the density deviation. The density deviation curve plotted for sulfuric acid shows a distinct maximum a t 33 per cent sulfuric acid which would tho presence of the indicate, according to Morgan (M), compound HBO~11H.O. These data are included in order
Vol. 2.5, No. 3
to denronstrato that the deviation curves for the titanium su1fat.e solutions and for sulfuric acid solutions corresponding in concentration t o the total acidity of the titanium solutions are not parallel. However, one pecu1iarit.y of the density deviation curve of the titanium sulfate solutions may be pointed o u t At lower conccntratiom of titanium oxide tlie curve is a straight line up to a concentration of about 4.5 per cent titanium oxide, and above this concentration deviates sharply from tlic straight line. At this concentration-i. e., about 4.5 per cent titanium oxide-is also the point a t which a minimum occurs in the precipitationconcentration curve (Figure 2). It seems probable that there is some change in the composition of the solution at this concentration, evidenced by cliarrge in direction of the physical property curve as well as tlie precipitation curve. However, since~the knowledge available concerning the cnmpo~ ~ T sition and characteristics of titanium solutions is extremely limited, it is not possible to state at the present time the nature of the change which occurs. Since the coursc of precipitation %'a$ tlie primary study of the investigation, tlie question arose as to whether this irregular precipitation cnme is characteristic of titanium sulfat,e solutions in general, or is peculiar to solutions with an acidity factor of 0.5. A series of investigations of solutions with acidity factors ranKing from 0.1 to 3 and with a hydrolysis period ranging from 1 to 24 hours in the more acid solutions was planned. The stock solutions and dilutions were prepared and adjusted in the same manner as those of the preccding series. I n tliose series when the acidity factor was greater than one, lonRer hydrolysis periods were used, and the precipitation mas carried out in air-tight, sealed glass jars in an oven a t 100" C . For the series with an acidity factor of one or less, the procedure was exactly the same as that previously described for the hydrolysis of the solutions with an acidity factor of 0.5.
IT. TITANIUM SULPATE SOLUTIOSS,ACIDITY FACTOR 0.1 The stock solutiori used in this series gave the following a~ialysk % b!, weight O ~ ~ m s / l i t e r TiO* Total ILSOI
10.01
25.52
130.4
329.2
% bv wucighl Equivalent HISO. E~ecsaI h S O .
Acidity isotoi
24.50 1.02 0.10
Ten dilutions were made with distilled water, keeping the acidity factor constant at 0.1 and varying the titanium oxide concentration from approximately 1 to 10 per cent
INDUSTRIAL AND ENGINEERING CHEMISTRY
March, 1933
269
tating to make the total amount precipitated 97 per cent of the original titanium oxide content. This mode of precipitation parallels closely that shown in curve 103, Figure 1. -4s in the previous series, it is in this concentration for which the precipitation has been plotted-i. e., 9.3 per cent titanium oxide-that maximum precipitation in the 8-hour period occurred. The precipitation of the 5.6 per cent titanium oxide solution (curve 205, Figure 7), which is representative of the sulfate solutions of 3.6 to 6.6 per cent titanium oxide concentration, is peculiar in that after the second hour of boiling the hydrated titanium oxide which has been precipitated apparently redissolved until about the sixth hour, after which the percentage precipitated remained very nearly constant. This solution is also peculiar in that, of the whole series of titanium sulfate solutions of acidity factor 0.1, a minimum in the precipitation-concentration curve (Figure 8) occurs in this solution with a concentration of 5.6 per cent titanium oxide. The same observation was made in the previous series, in which a titanium sulfate solution of acidity factor 0.5 and 4.6 per cent loo titanium oxide concen90 tration occurred a t a ,, minimum in the pre2 3 70 cipitation-concentra.E tion c u r v e , and also 'i6O gave the same type of k 50 precipitation-time $40 curve. u Curve 200 of Figure 30 7 s h o w s the a m o u n t 20'1 2 3 4 5 6 7 8 9 1 0 Percent Titanium Oxide precipitated with time for t h e m o s t concen- FIGURE 8. INFLUENCE OF TITANIUM ON PERCENTtrated solution of this OXIDE CONCENTRATION series-i. e., per cent AGE PRECIPITATED IS 8 HOURSAT BOILING POIUT titanium oxide. (Acidity factor of titanium sulfate solutions, The percentage pre0.10) cipitated a t the end of 8 hours did not have any direct relationship to the titanium oxide concentration of the original solution (Figure 8). With increasing titanium oxide concentration above 1.3 per cent, there was a decrease in the percentage precipitated until a concentration, in the original solution, of 5.6 per cent titanium oxide was reached. Beyond this point the percentage precipitated rapidly rose with increased concentration up to 9.3
by weight. These solutions were precipitated by boiling under a reflux condenser as were those of series I. Samples were withdrawn a t the end of hourly periods of boiling with a total boiling time of 8 hours. The results of the hyd r o l y s i s of series I1 a r e summarized in Table VIII. 100 Figures 7 and 8 are 90 p l o t t e d from the data in 80 Table VIII. I n Figure 7 T1 3 70 in which the p e r c e n t a g e e p r e c i p i t a t e d is p l o t t e d .-u 60 against the time d u r i n g 2 50 a which hydrolysis has pro40 ceeded, for this series, as in oi 30 series I, only typical curves 2 eo are plotted. With this series of solu10 tions, as with series I, the 0 course of precipitation over 1 2 3 4 5 6 7 8 T i m e in H o u r s a g i v e n p e r i o d of t i m e FIGURE 7. INFLUENCEOF varied with different conTIMEOF HYDROLYSIS ON PERc e n t r a t i o n s of titanium CEKTAGE PRECIPITATED oxide, The differences in (Acidity factor of titanium sulfate concentration between the solutions, 0.10) dilutions were small enough Curve 200. 10.01% Ti02 Curve 201. 9.3077 TiOz so that two or more of the Curve 205. 5 . 6 1 4 Ti02 Curve 209. 1.29% Ti02 solutions used always fell within each concentration range that had a specific effect on the course of precipitation. I n plotting the data a typical curve in each range was selected as was done in series I. I n Figure 8 the percentage of titanium oxide precipitated a t the end of 8 hours of boiling was plotted against the titanium oxide concentration. The concentrations of titanium oxide plotted as showing the typical precipitation with time were also those which fell a t maximum or minimum points in the precipitationconcentration curve, as well as the highest and lowest concentrations used. I n the most dilute solution studied, with an original titanium oxide concentration of 1.29 per cent by weight (curve 209, Figure 7) slightly more than half the total precipitation during 8 hours took place in the first hour, after which the rate of precipitation decreased. This closely resembles the course of precipitation for the most dilute solution of series I-i. e., 1.32 per cent titanium oxide, acidity factor 0.5 (curve 109, Figure 1).
,,,
2
OF TITANIUM SULFATE SOLUTIONS WITH ACIDITYFACTOR 0.1 TABLEVIII. PRECIPITATION
ORIGINAL SOLN.CONCN. 1 hr. % bu t o l . I
200 201 202 203 204 205 206 207 208 209
13.04 11,681 10.550 8.968 7.838 6,481 5,203 3.919 2.652 1.325
11.40 10.49 9.29 8.26 7.05 3.32 2.92 1.31 1.34 0.723
2 hr.
PERCENTAGE TiOz BY VOLUME 3 hr. 4 hr. 5 hr. 6 hr. 7 hr.
.
8 hr.
ORIGIAAL Coma. 1 hr.
PERCENT.4GE
I
TiOz
PRECIPITATED
2 hr.
3 hr.
4 hr.
5 hr.
6 hr.
7 hr.
8 hr.
10: eo 8.45 38.15 35.85 54.36 22.37 20.79 57.76 67.69
31.35 22.06 39.42 58.32 72.92 33.96 27.23 50.01 61.80 76.37
22.97 36.31 67.72 63.29 67.73 25.91 29.90 53.74 65.91 79.54
33.87 47.87 83.54 72.15 37.89 21.86 39.15 58.15 68.70 76.30
33.41 64.64 86.00 73.61 53.96 28.60 42.07 59.73 67.27 84.52
47.74 95.61 93.14 74.65 57.50 29.53 44.96 70.50 69.53 87.70
64.35 96.83 93.32 72.32 63.84 28.84 49.41 73.82 75.72 90.11
% by weight 9.g5 9.10 9.66 6.39 3.74 5.55 2.12 5.03 2.96 4.28 4.04 3.79 3.10 1.96 1.01 1.12 0.428 0.314
lb:4i
9.04 7.44 3.40 3.29 2.53 4.80 3.75 1.81 0.904 0.271
8.62 8.68 6.81 4.13 0.512 6.09 1.47 0.723 1.74 2.51 2.47 2.27 3.33 4.87 3.61 4.57 5.06 4.63 2.86 3.17 3.01 1.16 1.64 1.58 0.850 0,868 0,808 0.314 0.205 0.163
4.64 0.370 0.705 2.48 2.83 4.61 2.63 1.03 0.644 0.131
Curve 201, Figure 7, represents the course of precipitation of the sulfate solutions of 7.4 to 9.3 per cent titanium oxide, and is plotted for the 9.3 per cent concentration. There was apparently a t first a lag or induction period up to the third hour of boiling, during which nuclei or centers of precipitation were probably set up, the percentage precipitated rising to only 20 per cent. During the next 4 hours the precipitation was the most rapid, the percentage precipitated rising from 20 to 95 per cent. Finally, during the last hour observed, the precipitation again proceeded rather slowly, only 2 per cent more of the titanium oxide precipi-
10.01 9.297 8.470 7.378 6.604 5.610 4.540 3.580 2.497 1.287
12.60 10.20 11.99 7.90 10.00 48.84 43.92 66.55 49.66 45.43
per cent titanium oxide a t which concentration optimum precipitation for the concentration and time range studied occurred. It is only a t this concentration that complete precipitation from the industrial point of view-i. e., with recovery of more than 90 per cent of the titanium oxide values of the solutions-took place. At higher concentrations the amount precipitated again decreased. For time periods of less than 8 hours, similar curves are obtained, but here, as in series I, the marked minimum and maximum are not as sharply defined. The concentration of the solution in which maximum precipitation occurred varied with the length of
I r(; 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
270
the hydrolysis period. The longer the period over which hydrolysis takes place, the higher is the concentration of the solution in which maximum precipitation occurs, This is illustrated in Table IX, using data selected from Table VIII. It may be seen from the table that the shift of the optimum precipitation concentration toward more concentrated solutions markedly decreases after the fourth hour, indicating that a longer heating period would not give maximum precipitation in still higher concentrations of titanium sulfate solutions of this acid content. TABLEIX. OPTIMUMPRECIPITATIOX OF TITANIUM OXIDE TIME OF hIAX. PPTX.
Hours 1 2 3
4
SOLN.OF
Ti02 % by weight 3.5 5.6 6.7
6.7-8.5
T I M E OF hf.4X. P P T N . Hours 5
6
k
SOLS. O F
Ti02
Yob y weight 8.5 8.5 9.3 9.3
Comparison of the figures and the data in the tables summarizing the course of precipitation of titanium sulfate solutions of acidity factors of 0.1 and 0.5 shows that the course of precipitation is not altered by this change in acidity factor. However, maximum precipitation occurs in solutions containing a higher concentration of titanium oxide when the acidity factor is lower; i. e., with an acidity factor of 0.1, a maximum precipitation of 97 per cent at the end of 8 hours occurred in a solution containing 9.3 per cent titanium oxide, whereas with an acidity factor of 0.5 maximum precipitation (again 97 per cent a t t h e e n d of 8 h o u r s ) occurred in a solution containing originally 7.5 per cent titanium o x i d e . Similarly, the minimum in the precipitation- c o n c e n t r a t i o n curve occurs at a higher concentration of titanium oxide when the acidity factor is 0.1 than when the a c i d i t y factor is 0.5; the solution containing 5.6 per cent titanium oxide gave the minimum precipitation in 8 h o u r s in a s e r i e s with an acidity factor of 0.1, FIGURE 9. INFLUESCEOF whereas a solution containing TIMEOF HYDROLYSIS ON PER- 4.6 per cent titanium oxide CENTAGE PRECIPITATED gave minimum precipitation (Acidity factor of titanium sulfate in 8 hours when the acidity solutions, 1.0) factor was 0.5. Consequently C u r v e 400. 9.98’3’ Ti02 C u r v e 404. 6.4 d TiOn the precipitation-concentraCurve 407. 3.3 % TiOz tion curves of the two series Curve 409. 1.3 % Ti03 for any given time period are similar in shape, but show maximum and minimum points a t different concentrations of titanium oxide. If the curves showing the course of precipitation with time for the two series are compared, it will be noted that the curves representing the same concentration of titanium oxide are not similar in shape, but that those which represent concentrations of titanium which fall at corresponding positions in the precipitation-concentration curve are similar in shape, For example, when the percentage precipitated in the solution originally containing 5.6 per cent titanium oxide with an acidity factor of 0.1 is plotted against time, the curve does not resemble that for a solution containing 5.6 per cent titanium oxide with an acidity factor of 0.5 but is very similar to that for the solution containing 4.6 per cent titanium oxide with an acidity factor of 0.5; the 5.6 per cent titanium oxide solution with acidity factor of 0.1 and the 4.6 per
Vol. 25, No. 3
cent titanium oxide solution with acidity factor of 0.5 both fall a t the minimum point in their respective precipitationconcentration curves. The acidity factors 0.1 and 0.5, of the solutions thus far studied, did not cover a very wide range of a c i d i t y . This might account for the similarity in the course of precipitation in these series of solutions. I n order t o d e t e r m i n e whether or not a higher acidity might make a m a r k e d difference in the precipitation curve and also to include more solutions of practical interest, s e v e r a l series of titanium sulPercent Titon’urn Oxide fate solutions of higher FIGURE10. INFLUENCE OF TITAON PERacidities were investi- NIUM OXIDECONCENTRATIOS PRECIPITATED IN 8 HOURS gated. A series (series CENTAGE AT BOILING POINT 111) of solutions with ( I c i d i t y factor of titanium sulfate solutions. a c i d i t y factor of 1.0, 1.0) twice that of the most acid solutions previously studied and ten times that of the least acid, was prepared, and the hydrolysis carried out as before.
111.
TITANIUnr
SULFATE SOLUTIONS, ACIDITY FACTOR 1.0
A solution of titanium sulfate in dilute sulfuric acid was prepared from crystallized titanyl sulfate as has been already described. This solution analyzed: %,bv weight
Ti02 Total H ? S O I Equivalent H 6 O i
9.77 33.90 23.91
Grams/ liter 135.56
470.25
332.12
Excess HgSOi Acidity factor
% bli weight 9.99 1.02
Grama/ liter 138.13
... .
Ten dilutions of this solution were prepared containing titanium oxide from about l to 10 per cent by weight with the acidity factor constant. -4s in the previous series a sample of each dilution was retained for analysis, and the bulk of the solutions immediately hydrolyzed by boiling under a reflux condenser. As in the previous series, samples were withdrawn a t the end of hourly boiling periods, with a total hydrolysis period of 8 hours. These withdrawn samples were decanted, or filtered, and analyzed for titanium oxide. The results of the analyses and the percentage precipitated calculated therefrom are given in Table X. Some of the data from Table X are graphically presented in Figures 9 and 10. I n Figure 9 is plotted the percentage precipitated against the time of hydrolysis for four selected titanium sulfate solutions. The percentage precipitated a t the end of the hourly boiling periods for the 9.8 per cent titanium oxide solution (the most concentrated of this series) is plotted as curve 400. It may be seen that the rate of precipitation was fairly constant over the 8-hour period (very nearly a straight line being obtained) and was quite low, little more than 10 per cent of the titanium oxide present having been precipitated. The same was true for the solutions of 8.2 and 8.9 per cent titanium oxide except that the precipitation rate was somewhat greater, 33 and 16 per cent, respectively, of the titanium oxide having precipitated in 8 hours. The precipitations of solutions of 5.6 to 7 . 3 per cent titanium oxide are represented by curve 404 which has been
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 hl I S T R Y
272
Vol. 25, No. 3
as may be observed by comparison of their precipitation- have been shown to be important factors in the precipitatime curves. The effect of acidity factor, between the tion of the titanium sulfate solutions thus far studied. It limits of 0.1 to 1.0, is shown somewhat more clearly in Table was desired to Iearn whether the effects of increased acidity XII. The table indicates the percentage precipitated in 8 noted above extended t o solutions of still greater acidity, hours for the most dilute and most concentrated solutions Therefore such solutions were prepared and hydrolyzed. used, for those solutions in which maximum precipitation Since with increased acidity it had been found that maxioccurred, and for those solutions which were a t the mini- mum precipitation, in an 8-hour period, occurred in solumum in the precipitation-concentration curve. tions of lower concentration of titanium oxide, it was thought TABLEXII. EFFECT O F ACIDITY FACTOR ON PERCENTAGE advisable to hydrolyze for a longer period of time in order PRECIPITATED IN 8 HOURSAT BOILIKGPOIKTO F TITANIUM to determine whether maximum precipitation might occur SULFATESOLUTIONS in a solution of higher concentration. For convenience Acidity factor: -0.1--0.2--.-0.5-~ --1. h these precipitations were carried out in glass-topped pre% % % % serving jars sealed with a rubber ring or gasket, which were % HYDRO- % HYDRO-% HYDRO- % H Y D n O placed in a Freas electric oven a t 100" C. SOLN. Ti02 LYZED Ti01 LYZED TI02 LYZED Ti02 LYZED 1. 2. 3. 4.
Dil. Min. Max. Coned.
1.3 5.6 9.3 10.0
90.1 28.8 96.8 64.4
1.4 4.6 9.1 10.2
86.4 36.8 96.5 49.3
1.3 4.6 7.5 9.9
81.8 44.4 97.3 31.4
1.3 3.5 6.4 9.8
78.6 49.6 96.9 12.0
Considering the most dilute solution in each series-horimay be seen that the zontal column 1 in Table XII-it amount precipitated decreased with increase in acidity. It is also to be noted that in every series these dilute solutions all precipitated to the greatest extent during the first hour, after which the rate of precipitation was comparatively constant and low. Increase in the acidity factor caused greater precipitation in the 8-hour period, in those solutions which were a t the minimum in the precipitation-concentration curve of each series. This increased precipitation with increased acidity factor is not peculiar if it is considered that the minimum in the precipitation-concentration curve occurs in more dilute solutions with the higher acidities. We know, from a consideration of the precipitation of the titanium solutions within each series, that with increased dilution of the titanium solutions below that concentration a t the minimum in the precipitation-concentration curve, the amount of precipitation was very much greater. The hydrolysis of all these solutions was characterized by considerable precipitation in the first hour, followed by re-solution of the precipitate during the next few hours and only slight further hydrolysis or re-precipitation during the remainder of the boiling period. Maximum precipitation occurred in progressively more dilute solutions as the acidity was increased, as indicated in horizontal column 3 of Table X I I . Thus maximum precipitation took place in a solution containing 9.3 per cent titanium oxide of acidity factor 0.1; whereas, in the series of solutions of acidity factor 1.0, maximum precipitation was found to have been in the solution of only 6.4 per cent titanium oxide. The percentage precipitated was very nearly the same in all cases. It is evident from this that in the industrial precipitation of titanium oxide, in which it is desirable to obtain yields of 90 to 95 per cent, it is necessary to regulate the acidity factor as well as the titanium oxide concentration, and to select the proper acidity for the titanium concentration which can be most readily and economically obtained. The precipitation of these solutions all proceeded in the same way. During the first few hours of boiling, the rate of precipitation was quite slow; this was then followed by a period of rapid precipitation, and finally in the last few hours the rate rapidly decreased and there was but little further precipitation. The course of precipitation of the most concentrated solutions used in the series resembled that of the most dilute solutions, except that the amount, or rather percentage, precipitated was very much less. As in the dilute solutions, the rate of precipitation over the 8-hour boiling period was very nearly constant, and likewise with increased acidity t h e amount precipitated was very much less (horizontal column 4 of Table XII). Both the acidity and the titanium oxide concentration
Iv.
SULFATESOLUTIOKX, ACIDITY FACTOR 2.0 An acid titanium sulfate solution was prepared which had the composition: TITANIUM
%bu
Ti02 Total HzSOl Equivalent H z S O I
weight 9.96 44.82 24.35
%tu
Grams/ liter 144.69 652.08 354.07
Excess HzSOi Acidity factor
wezoht 20.47 2 .06
Grams/
liter 298.01
..,
,
Ten dilutions of this solution were prepared for precipitation. These solutions were placed in glass-topped preserving jars, sealed by means of a rubber ring or gasket, and placed in a Freas electric oven a t 100" C. At intervals of 4, 6, 8, 12, and 24 hours, one jar containing each dilution was removed from the oven and allowed to cool, and the decanted or filtered solution was analyzed for titanium oxide. The results obtained are summarized in Table XIII. O F TITANIUM SULFATE SOLUTIOKS TABLEXIIl. PRECIPITATIOK WITH ACIDITY FACTOR 2.06 SOLN.
b
%,by wezoht 9.96 9.33 8.49 7.46 6.67 5.87 4.76 3.96 2.66 1.39
-PERCENTAQE 4 hr. 6 hr.
Ti02 PPTD.-8 hr. 12 hr. 24 hr.
% by
sol. 14.47 13.01 11.71 10.10 5.70 7.31 5.79 4.41 2.92 1.46
1.6" 55.1'3 3.8' 3:20 3.50 4.55 4.8'' 10.1b 6.4 4.2 5.2 5.0 50.7 88.0 8.5 43.3 50.0 72.4 29.9 Precipitate completely soluble in water.
600 601 602 603 604 605 606 607 608 6 09 0
ORIWXALCONCN.
11.10 10.10 20:9b
..
.... .. ....
51.1" 43.10 4.8b 23.30 4.9 5.6 10.3 69.3 44.0 62.6
80.0' 48.0" 6.6) 2.3b 73.0b 6.3 5.0 43.3 53.9 52.2
Precipitate partially aoluble in water.
Perhaps the most notable feature in this series is the fact that, in the higher concentrations of titanium oxide, crystallization of titanyl sulfate occurred alone or together with precipitation of metatitanic acid. I n the solutions of 6 to 9 per cent titanium oxide, precipitation of the meta acid as well as crystallization took place as indicated by the fact that the precipitate formed was only partially soluble in water. I n the two solutions of greater concentration than 9 per cent titanium oxide, only crystallization of titanyl sulfate occurred, inasmuch as the precipitate which formed was distinctly crystalline in character and completely soluble in water. The results obtained for this series are not as regular as were the results for the precipitation of the solutions of lower acidity previously studied. This is probably because of the additional complication of crystallization of a water-soluble titanyl sulfate, in addition to the precipitation of relatively insoluble, hydrated titanium oxides. One point of interest is the fact that it was possible to crystallize titanyl sulfate with a reasonably high yield-. g., 80 per cent in 24 hours a t 100" C.-from solutions of considerably lower concentrations of titanium oxide than those specified by Blumenfeld (2) wherein the conditions for the
March, 1933
I SIT R Y INDUSTRIAL AND ENGINEERIN G CHE&
crystallization of titanyl sulfate are stated to be: a temperature of 110" t o 130" C., a titanium oxide concentration of 200 grams per liter, sulfuric acid content of 500 grams per liter, and a specific gravity of 1.45 t o 1.615. The most concentrated solution used in the above series and found to give a reasonably good yield of crystalline titanyl sulfate was heated a t 100' C. and contained only 145 grams per liter of titanium oxide, although the sulfuric acid concentration of 652 grams per liter was greater, and the specific gravity was a b o u t QC 80 1.46. However, it should be noted that 70 for those solutions of this series which '0 3 60 were of lower titanium sulfate concenep 50 tration than those in which crystalliza0 tion occurred, the course of precipitaE+ 40 tion closely r e s e m b l e d that of the g 30 U & 20 other series studied. Examination of IO the precipitation-concentration curve (Figure 11) shows the maximum and o ~ e ~ c ~ n + ~ , + ~ & m m i n i m u m p o i n t s that seem to be FIGURE ll. INFLUcharacteristic of the course of precipiENCE OF TIT.&NIUM tation of these solutions. In 6 hours OXIDE CONCENTRA- at about 100" C., maximum precipitaTIoN ON PERCENTAGE tion occurred in a solution originally PRECIPITATED IN 6 o n t a i n i n g 3.96 per cent titanium AT looo c, coxide. Among the solutions of acidity ~ - ~ ~ ~ l t ~ , " , l : factor " , ~ ~ 1.0, ~ ~ maximum ~ ~ ~ ~ ; precipitation 2.06) occurred in the 6.4 per cent titanium oxide solution This is in agreement with the previous finding that, with increased acidity, niaximum precipitation occurs in more dilute (as respects titanium oxide concentration) solutions. When the hydrolysis period was extended to 12 hours the maximum precipitation again occurred in the same concentration and not in a solution of higher titanium oxide content, but in the 4-hour period maximum precipitation occurred in the solution which originally contained 2.7 per cent titanium oxide. This too is in agreement with the findings in previous series that the concentration of the solution which gives maximum precipitation increases with increased time of hydrolysis up to 6 hours, and then does not increase further with a longer precipitation period.
SUMMARY AND COSCLU~IONS The course of precipitation of pure titanium sulfate solutions ranging from 1 to 10 per cent by weight of titanium oxide, with acidity factors varying from 0.1 to 2.9 and with a precipitation period from 1 to 24 hours has been studied. These concentrations and acidities cover the range commonly used in the industry. I n the first four series the course of precipitation at the boiling point of solutions of titanium sulfate of concentrations from 1 t o 10 per cent titanium oxide with a constant acidity factor in each series, and over time intervals of 1 hour for 8 hours was studied. The acidity factors used were 0.1, 0.2, 0.5, and 1.0. In such a series the rate of precipitation does not bear a direct relationship to the time or to the concentration of titanium sulfate. With increasing concentration of titanium sulfate the precipitation decreases to a minimum and again rises until a maximum point is reached, and then with still higher concentration the precipitation again shows a decrease. This is true regardless of the length of the hydrolysis period for those solutions and time periods studied. However, the concentration at which the maximum precipitation occurs increases with increase in the hydrolysis period up to 6 hours; if the hydrolysis period is longer than 6 hours, maximum
273
precipitation occurs in the solution of the same concentration as that which showed the greatest precipitation in 6 hours. The relationship of the course of precipitation to increased time depends upon the concentration of the titanium sulfate solution. With different acidities the precipitation-time curves for solutions of the same titanium oxide concentration are not similar, but those representing concentrations of titanium oxide which fall at corresponding positions in the precipitation-concentration curves are similar in shape. In the last two series the precipitation of the solutions of higher acidity was carried out over longer time periods, and in a n oven a t 100' C. The acidity factors used were 2.0 and 2.9. Here again the course of precipitation shows no direct relation to time or concentration. I n every series the precipitation-concentration curve s h o w the marked maximum and minimum point characteristic of the course of precipitation of these solutions, and the concentration in which maximum precipitation occurs increases with increased length of the hydrolysis period u p to a definite time and is then constant. I n these series, however, titanyl sulfate crystallized from solutions containing more than 6 per cent titanium oxide. It is also true in these series that the form of the precipitation-time curve is dependent upon the concentration of the titanium sulfate solution. For all of the series studied it may be observed that over a definite hydrolysis period, as the acidity factor is increased, the titanium oxide concentration of the titanium sulfate solutions in which maximum precipitation occurs is lolver (Table XIV). The minimum point in the precipitationconcentration curve also occurs in a solution of lower concentration of titanium oxide as the acidity is increased. It may also be noted in Table X I V that the maximum percentage precipitated decreased with increased acidity. It is also true that in the most dilute and the most concentrated solutions studied, the percentage precipitated in a given time decreased with increased acidity. TABLE XIV. INFLUENCE OF ACIDITY FACTOR o x PRECIPITATION ACIDITY FACTOR 0.1 0.2 0.5 1.0 2.0 2.9 ~~
HYDROLYSIS COYCN.G I V I N Q MAX. TIME PPTN.OF Ti02 Hours % b y weight % by 1101. 8 8 S 8 6 & 12 8
9.3 9.1 7.5 6.4 4.0
...
>fAX. PERCENTAQB PPTD.
11.7 11.5 9.4 8.0
97 97 97 97
4.4 3.2
88 85
Concentration of titanium oxide, length of hydrolysis, and acidity factor are therefore all factors which affect the course of precipitation of the titanium sulfate solutions studied. 1. A method of preparing pure titanium sulfate solutions was devised. Titanium tetrachloride was hydrolyzed in the presence of titanous chloride and oxalic acid, the hydrated titanium oxide precipitated was converted t o the sulfate by means of hot concentrated sulfuric acid, this sulfate then dissolved in water and titanyl sulfate dihydrate crystallized by heating. 2. Precipitation is not a regular function of the concentration, and definite concentration ranges have a specific effect upon the course of precipitation when both the time period and acidity factor are constant. 3. In the most dilute solutions used, the rate of precipitation during the first hour is much higher than that of the most concentrated solutions used; after the first hour of precipitation the rate is about the same for the most concentrated and the most dilute solutions. 4. With increased length of hydrolysis time up t o 6 hours, the concentration of the solution in which maximum precipitation occurs is greater; further increase in hydrolysis period gives maximum precipitation in the same concentration of titanium oxide with slight increase in the total percentage precipitated. 5 . With increased acidity factor the titanium oxide con-
I h’ D U S T R I A 1, A N D E N G I N E E R I N G C H E M I S T R Y
274
centration of the titanium sulfate solutions in which maximum precipitation occurs is lower. 6. The maximum percentage precipitated in the time periods studied decreases with increased acidity. 7. The concentration of titanium oxide has more influence on the rate of precipitation than the acidit,y factor. Thus in those solutions a t the minimum of their respective precipitationconcentration curves the percentage precipitated increased in spite of increased acidity factor because these solutions occurring at the minima of the precipitation-concentration curves are more dilute as the acidity factor is raised. 8. In the most dilute and the most concentrated solutions the shape of the precipitation-time curves is the same for all acidities studied but the percentage precipitated in a given time decreases with increased acidity factor. 9. When the acidity factor is 2 or higher, titanyl sulfate is crystallized alone or with metatitanic acid from solutions containing more than 6 per cent by weight of titanium oxide. 10. When the acidity factor is 1 or less than 1, solutions of 6 t o 10 per cent by weight of titanium oxide may be 95 per cent precipitated by boiling for 8 hours.
T’ol. 25, No. 3
LITERATURECITED
A4CKSOWLEDGJlEST
Barton, L. E., U. S . Patent 1,155,462 (Oct. 15, 1915). Blumenfeld, J.,U. S. Patent 1,504,669 (Aug. 12, 1924). Dennison, Trans. Faradag Soc., 8, 20, 35 (1912). Fladmark, BZ.,U. S. Patent 1,288,863 (Dee. 24, 1918). Herz, IT., and Bulla, -L, 2. anorg. Chem., 61, 367 (1909). Hixson, A. IT.,and Plechner, IT. TI’., Chem. Le. M e t . Eng., 36, 76 (1929). Patent 503,124 (hug. 15, 1893). ( 8 ) Kolthoff, I. LI.,J . Phys. Chem., 36, 860 (1932). (9) Kullgren, C., 2. phusik. Chem., 85, 466 (1913). (10) Laugier, A , Ann. chim.p h y s . , 89, 306 (1614): Schzeigger’s J . , 19, 54 (1817). (11) Mecklenburg, IT., U. S.Patent 1,768,528 (May 13, 1930). (12) Morgan, J. L. R., J . An&.Che7rz. Soc., 38, 558 (1916). (13) Overton, J., British Patent 9825 (Dec. 23, 1893). (14) Rossi, A. J.,and Barton, I,. E., U. S. Patent 1,196,030 (Aug. 29, 1916). (15) Roth, IT. A . , and Recker, G., Z . ph~sik.Chem., BodensteinFestband, 55 (1931). (16) Ryan. L. IT-.,U. S.Patent 1,820,987 (Sept. 1, 1931). (17) Toungman, E. P., Bur. Mines, Circ. 6365 and 6386 (1930).
The authors are indebted to the Titanium Pigment Company for the phot,ographs used to illustrate this paper.
RECEIVED October 20, 1932. W. W.Plechner’e present address is Titanium Pigment Company, Inc., New Tork. N. Y .
(1) (2) (3) (4) (5) (6)
Amination by Ammonolysis IV. Design and Construction of Equipment’ P. H. GROGGIIVS, Bureau of Chemistry and Soils, U. S. Department of Agriculture, Washington, D. C.
T
HE preparation of amines
The reaction between organic compounds, containing replaceable substituents, and ammonia generally takes place at elerated temperatures. I t is therefore essential that ammonolysis be carried out either in high-pressure autoclaves or in tubular systems. Autoclaves are adapted .for discontinuous or batch operations, whereas pipe systems are particularly suitable ,for coniinuous high-pressure syntheses. Although direct-jired sysiems were at one time quife prevalent, these have been supplanted by jacketed vessels heated with steam or circulated fluid, or by tubular systems immersed in molten alloys. Such systems procide a safer, more HIGH-PRESSVRE EQUIPMEKT accurate, and practically automatic control of FOR A~LIMONOLYSIS the amination process.
through the reaction between a r o m a t i c c o m pounds and ammonia generally takes place a t e l e v a t e d temp e r a t u r e s . It is t h e r e f o r e essential that the ammonolysis be c a r r i e d out either in highpressure autoclaves or in tubular systems. A u t o c l a v e s are generally adapted for discont i n u o u s or batch operations, whereas pipe systems are particularly suitable for continuous high-pressure syntheses.
Autoclaves may Tary in size or shape according to the particular requirements of a plant or process. They may differ as to the materials of construction, depending on the nature of the reacting materials. Some are provided Kith stirrers, for agitation is essential when the contents of the vessel are not homogeneous; and finally, some are provided with jackets to permit temperature control by means of steam or a circulating liquid. I n general, industrial autoclaves are hollow cylindrical vessels ranging from 100 to 1000 gallons in capacity and are designed and constructed to operate up to 5000 pounds pressure per square inch. The steel shell may be forge- or hammer-welded, electric fusion-welded, or may be forged, rolled, and drawn so as to provide a seamless vessel. Hammer-welded apparatus having a wall thickness of 2 inches is 1 P a r t s I a n d I1 of this paper appeared in INDUSTRIAL AND CHDMIBTRP i n J a n u a r y , a n d P a r t I11 i n February, 1933.
readily o b t a i n a b 1e ; seamless forgings or electric fusion vessels can be p r o d u c e d with walls of a n y desired d i m e n s i o n s . H a m m e r - w e l d e d vessels are being r e p l a c e d by the other t y p e s of apparatus, s i n c e the former are l i m i t e d to a comparatively light wall thickness and i n v o l v e t h e u s e of l o w c a r b o n a n d consequently low tensile strength plate. Table I gives the safe working p r e s s u r e that can be used on electric f u s i o n - w e l d e d autoclaves having a diameter ranging from 1 t o 5 feet, with shell thickness ranging from 1 to 3 inches.
DIRECT-FIRED AUTOCLAVES The use of direct-fired autoclaves for carrying out highpressure reactions was a t one time a general practice. Coke, fuel oil, and gas furnished the means of heating such vessels. Such systems possessed certain inherent defects, particularly in regard to the danger of igniting the inflammable contents of the autoclaves in case of a leaky connection, and in the difficulty encountered in accurately maintaining the optimum operating temperature. When gas or fuel oil is used as a source of heat, it is now possible to obtain accurate and automatic control. Gas is, however, distinctly more costly than fuel oil, and, when the duration of the operating cycle is long or the price of the intermediate is low, its employment on a large scale may be prohibitive. This problem is not so important in localities where natural gas can be obtained a t a ENGINEERIWQlow cost, nor is it of appreciable consequence where the gas for the combustion chamber is manufactured in the plant.
