ANALYTICAL CHEMISTRY Am. SOC.Testing Materials, "Standards on Petroleurn Products" (1945).
Bartholomew, W. H., Ph.D. thesis, Pennsylvania State College, 1941.
Leer, C ; . SI.. and Taitt, A . H., Petroleum Times, 49,5-6, 31-3 (Jan. 6, 1945). McCluer, R'. B., and Fenske, SI. K.,I d . Eng. Chem., 24,1 3 7 1 4 (1932).
Carnahan, F. L., Hersh, R. E., and Fenske, M .R., I d . Eng. Chem., 36, 333-5 (1944).
Cleaves, A. P., and Carver, M. E., p&per presented before Division of Petroleum Chemistry, 111th Meeting of AM. CHEM.Soc., Atlantic City, 1947. Dow, R . B., McCartney, J. S., and Fink, C . E., J . Inst. I-'etroleum, 27,301-9 (1941). Fenske, M. R., Carnahan, F.L., Breston, J. N., Caser, A. H., and Rescorla, A. R., Ind. Eng. Chem., 34,638-46 (1942). Foehr, E. G., Ph.D. thesis, Pennsylvania State College, 1944. Hersh, R. E., Fisher, E. K., and Fenske, M.R., Ind. Eng. Chem., 27, 1441-6 (1935). Irvin, H. B., M.S. thesis, Pennsylvania State College, 1945.
Pennsylvania Grade CI ude Oil .hariation, private communication. Pfister, K.J., Ph.D. thesis, Pennaylvania State College, 1942. Poulter, T. C.. Snyder, 5'. D., and Putscher, R. E., paper present,ed before Synipoaium on High-pressure Technology, Division of Industrial and Engineering Chemistry, 110th Meeting of AM.CHEM.Soc., Chicago, 1946. \lugter, J. O., Waterman, H. I., and Van Westen, H. -4..J . Inst. Petroleum Tech., 21,661-76 (1935). Welshans, L. M., M.S.thesis, Pennsylvania State College, 1942. RECEIVED M a y 2, 1947. Previous article8 on this subject have been yuhlished in Indualrial and Engineering C h e m i s t r y ( 4 , 7 ) .
Composition of Rosin Size Precipitates Analyses of Standard Size Precipitates DOXY4 P R I C E Hercules Experiment Stution, Hercules Powder Compuny,
Petroleum ether, in contrast to other solvents, did not change the composition of fresh rosin size precipitate extraction residue, even when used in the presence of excess water, and fresh size precipitates contained little or no oxidized resin acids that are insoluble in petroleum ether. Therefore, this solvent was used to study the composition of rosin size precipitates. Extraction with it isolated material of the composition of aluminum diresinate
A
(4)mentioned that there is no satisfictory method available for the analysis of rosin size precipitate. RIost attempts a t analysis have been confined to a determination of the ash, and to solvent, extraction of the precipitate (1, 6 ) . The ash values reported have been 4.78 to 5.69% of the size precipitate, which has frequently been interpreted to mean that the material rvas essentially aluminum triresinatr, for which the theoretical ash is 5.5yc. I n bhe present work more complete analysis vias obtained by considering the carbon-aluminum ratio (C/Ail). This value is independent,of the moisture content of the precipitate, and it has served as a guide in determining the composition of procipitates from sizes of various free rosin contents, as well as in identifying sizing materials extracted from papers. Petrolwm cther was used as a solvent for extraction of freshly prepared precipitatt,s because it, did not, change the precipitate residue composition, as other solvents did, and because the fresh precipitates seemed to contain practicdly no oxidisd wsin arids that arc1 insoluble in this solvcnt. Based on data obtained from extractions ,and analyses of various size precipitates, a'mechanism for the reaction between size and aluminum sulfate is suggested. PRECEDISG paper
EXPERIMEYTAL
The prrparation o f standard size precipitates \vas descrihrd ill the first paper of this series ( 4 ) . Floes obtained by this standardized procedure Tvere concentrated by decant,ation and then filtered in a Buchner funnel. The precipitate was washed in the funnel three times with distilled water, air-dried, then ground to pass a 60-mesh screen and dried in a vacuum desiccator. The high dilution of the standard preparation corresponds to size concentrations in standard papermaking ( 6 ) . For largt, of the large samples-e.g., 20 grams of precipitate-filtration volumes required several days. However, several pairs of analyses showed that the additional time of exposure of the precipitate t o the liquor of the system in lvhirh it was prepared did not de-
W ilminpton
YY, Del.
from lahorator) preparations of rosin size precipitate, as well as from rosin-sized papers. For the analj ses, the ratio carbon-aluniinum was used instead of the more customarj ash determination. A knowledge of the free rosin content of the original size and the use of a proposed size-aluminum sulfate reaction mechanism accounted for all size precipitate compositions obser\ed. Preparations were made f r o m sizes of 0, 20, and 7570 free rosin.
t)Iy affect its composition. For example, a sample of prrripitate exposed for 4 hours and a sample of the same precipitate exposrd for 24 hours gave respective carbon-aluminum valUPS of 34.4 and 34.2. Solvent extractions were carried out in a glass-stoppered, 250nil. Erlenmeyer flask. Approximately a 5-gram sample of precipitate was placed in the flask, shaken with the solvent, and allowed to settle. The solvent wa$changed 7 to 10 times over a period of sevtlral weeks. The suspended material in t,he extract n-as rt.nioved and rinsed by centrifuging; it was then added to the residue in the flask. The residue of the extraction \vas airdried and then dried t o constant (or only slightly increasing) nx4ght3at 100" C. The extracted material was recovered by lowtcmprrature evaporation of the solvent, and was finally dried at 100" C. \lost extract solutions became cloudy either with age or upon heating. The, apparatus used for the continuous petroleum ether extraction of large samples of paper consisted of a distilling flask in tvhich the extract was distilled and a large flask containing the paptlr sample through nhich the distillate was continuously circulated. The extract from the treatment flask was returned to thr tli.*tilling flask. Cork stoppers sealed n-ith sodium silicate were u s i d in this apparatus. The paper sample was cut into narrow .trips for extraction, and for each load the continuous process was rarried on for 48 hours.
Size C \vas an emulsion-type size of 40% solids; it contained 7 , i r i free rosin based on the solids. The aluminum sulfate used was iron-free papermakers' alum, (assentially A1,(S04)a.18H,O. The ethyl ether used was Nerck, c.P., which had been dried and stored ovw metallic sodium. The pet,roleum ether was a commercial grade with specific gravity 0.6460 ( 2 0 " / 2 0 " )and refractive index 1.3710 (nk0)and contained no nonvolatile component. .ill other reagents used were C . P . grade. Analyses. An analysis of size precipitates for total reiin content ivas attempted by a procedure similar t o that used in the paper analysis (8). Although size precipitate in t,he acidified alcohol specified by this method readily gave a clear solution, the residur by evaporation of this solution could not, be completely
445
V O L U M E 2 0 , N O . 5, M A Y 1 9 4 8 t,stracted with ether within a reasonable time. Instead of an analysis for total resin, therefore, the per cent carbon and per wtit hydrogen of the precipitates were determined by the semimicroprocedure described by Fieser (2). The ash determinations were, unless othern-ise indicated, dir,c,ct ashes a t 1200" C. by bot,h a macro and a semimicromethod, according t o the size of the sample available. This high ignition t vmpcmt urc xas necessary t o deconipose any aluminum sulfate to alumina. "Brnzene-insoluble" material in rosin size precipitate was detcsrmined in a ft.lv instances by treating a 2-gram sample with 50 rill. of warm anhydrous benzene, filtering the mixture through a elas cruciblc,, drying tht, crucihlc at 100" C., and rr-
Tahle I. Theoretical composition of lluminum Resinates A1 f C ' m FixuOr ( O H ) ? .I1(C?N>II ,sOr)?OH hl(C~H?~0i)s
'l'ahle II.
66.3 i4.3 77.4
174..819
4 7.1 48 6
187..888
3.49
2.91
2ti.(i
Effect o f \loisture on Solvent Extraction of Size Precipitates .Inalyrib of Keiidiie a f t p r
_ _ _ _ Extraction _
__
3Ioiiture content v a s clstiniated in tv-o ways: by drying at 100' (1. for 16 hours, and by the Karl Fischer dtlterniination. Thi. mcxthoxyl tic.tc,i,iniriatiorin-as maderminedon the ash of a 0.5-gram sample. Hence, for this determination, the ash was subjected to a sodium carbonate fusion, subsequent acid solution, and precipitation of the aluminum. Precipitations of aluminum as the hydroside, as the basic succinate, and by the &hydroxyquinoline method were investigated. Of these, the first was most satisfactory, and the standard Rz03 method was adopted for this determination. Because of the many operations involved and tmhe difficulty of handling a gelatinous precipitate such as aluminum hydroside i t is evident that the aluminum deterininat,ion is less accurate than the ash, part,icularly for small amounts of ash. Hence, xhile the carbon-aluminum value of 63 for the precipitate from size C in Table IV is appreciably lower than might be expected for a 75% free rosin size, it is considered more valid than the value of 300 obtained from direct aluminum analysis of a b similar precipitate. Furt,her confirmation that the reaction involved in the formation of size precipitate is one between aluminum sulfate and the sodium resinate of the size was found in the alum requiremenk of the three sizes when used in papermaking. Typical data appear in Table V; they show that t,he alum requirement increased with increased sodium resinate content, of the size or decreased free rosin content', as would be espected. The data for petroleum ether estract,ion of two precipitates are given in Table VI. I n both cases, the composition of the residue was approximately that of aluminum diresinate, despite the original free rosin content of the size. The material extracted obviously contains free rosin and neutral bodies, as well as some aluminum compound, possibly soluble diresinate. The amount
Table I\-.
Composition of Precipitates from Various Rosin Sizes
5
Size Used B
ti
B type''
1
.4
Size Precipitate
Free Rosin in Size
%
C.
5;
.4sh A1 ( 1 2 0 ~ ° C . ) ,(Calcd.), % Yo 5.46 5.43
C/Al
0
i1.9 73.0 73.0 73.2 72.8
5.44 5.58 5,54
2.88
2z.3
1 8
73.0 74.4 74.5
5.56 4.40 4.40
2.94
24.8
20
4.40 4.54 4.64
2.33
31.9
4.59 3.95 3.97
2.43
31.2
2.10
34. I
-
7
.I
2n
74.5 75.0 75.6
8
A
20
75.3 71.7 71.6
__
-
20
71.6 72.4 72.0 12.2
3.96 4.07 4.02
-
-
72.2 75.7 75.7
4 05 4.16 4.19
2.14
33.7
75.7 74.8 74.8
4.18 2.20 2.30
2.21
34.2
74.8
2.25
1.19
62.7
9h
A
10
A
20
11c
C
-I J
Prepared from resin arids of acid S o . 185; no rosin neutral bodies Ijresent. h Required 12.6 ml. of 10% alum per 3 grams of sjze. C Required 3.3 ml. of 1070 alum per 3 grams of size.
447
V O L U M E 20, NO. 5, M A Y 1 9 4 8 Table 7'.
Aluminum Sulfate Required by Yarious Sizes in Papermaking
[Conditions. 3.07, size. Burgess standard bleached sulfite pull> a t 750 * 10 Schopper-Riegler. Distilled water pH 5.3 t o 5.6 t o which nonalkaline salts were added t o obtain total hardness of 450 p.p.m. CaC08. gtandard papermaking procedure ( 6 ) ] 10% AI2,(S0~)aAdrled Size t o Sizing Crock" €'rep Rosin in Size
.M1. h
17 19 13 Average over eight hard waters used.
n
C
"
Table VI. Petroleum Ether .Extraction of Size Precipitates C,
%
Alaterial I'recipitate 11
74.8
Residue of estraction (23%)
04.9
w
%
Siae C 2.25
c 'AI
1.19
62.7
3.57
18 1
75.1
0.75 0.26 0.25
__
75.2
0.26 Size A 4.40 8.20
(I
05.0 ~
65.0
Extracted (78%)
Ash AI (1200O C.), (Calcd.),
75.4
--
Precipitate 2
74 3
Residue of extraction (31%) Extracted ( i . j ? )
70 5 73 4 73.3
Data
In
Table 111
6 . 70 6.80
__
14
339
2.33
31.9
Table I11
4 38
16 2
Tahle V I 1
1 43
51.2
2.70
-
2.69 -_
73.4
2
in
o f this material in the precipitate of size C is very sniall because size C is a high free-rosin size. The ash indicates only about 3 5 solublc, aluminum diresinate in the original precipitate. In rontraet t80this, the material extracted from size -1precipitate amounts to 26%jcalculated as soluhle aluminum diresinate. .ALU>fINU&I DIRESINATE IN SIZE PRECIPIT.411.:
I t has been shown that the residue of petroleunl (Ither extraction of size precipitate (\\-cator dry) has a composition suggest,ive of aluminum diresinate, A1(Cy,H.qO,)?OH. llorcovcr, the first niiiterial insolubilizcd and precipitated from a petroleum ether extract of size precipitate also had this composition. These findings suggest that an appreciahk portion of size pwcipitate consists of this particular compound, and that in the fresh precipitxte, a portion of thv aluminum dirwiriate is "soluble" (or at. Ivast dispersible) in petroleum t~tkiw. Of thiY soluhle portion a fraction tiecame insolubilixed, possibly by oxidation of the twin acids, as n-ell as hj- sonic othci, mechanism such as c*oapulaI ion. Furt,her evidence of the presericc. of aluminum dircsinate in size precipitates was obtaincd by thr continuous petroleum &her (bxtraction of large samples of frwhly prepared rosin-sized papers. Both laboratory papers (standard preparation, 5 ) and comhiercial papers !\-ere used aud the extractions were earricd out on 900gram samples. In both extractions a preripitatr appeared in the extract at the end of 3 to 4 hours' operation. This material \vas removed and washed with solvrnt hy centrifuging. -1fter h i n g dried in the usual manner, it was analyzed (Table 1-11). Comparison of the value forthc carbr>n-aluminum ratio with theoretical values for the aluminum resinates (Table I) shows that this material has the composition of thc diresinate. In both casrs the amount obtained was about 0 . 1 4 5 of the bone-dry paper or about 10% of the size on thc paper. However, a portion of' the size precipitate remained on the paper, and another portion (ash-containing) remained soluble in the petroleum ether. Hence, the amount recovered in this fashion is no indication of the total amount present.
The aluminum diresinate recovewd in this manner did not seem to be affected by alcohol or cold acidified alcohol. but i t was somexhat swollen by carbon tetrachloride, and fused when it was heat,ed in a flame. After removal of the insoluhilized aluminum diresinate, the clear extract was concentrat,ed on a steam bath. It remained clear during this process, but formed a gelatinous precipitate when diluted with fresh solvent. This precipitate disappeared upon further concentration. The extract residue, a stiff past?, was dried to constant weight at 83" C. h u t , even so, it probabl:, rontainrd some petroleum ether residues. Tho dried paste was partially soluble in petroleum ether and alcohol, apparently completely so in carbon tetrachloride and in a 50-50 xylenealcohol mixture. The inconiplete solubility in petroleum ether might be caused by the oxidation and consequent, insolubilizatiou of sonic of the material or by coagulation of colloidal niattrr. The residue from the extract of the waterleaf paper had a consistency similar to soft wax and w-as completely soluble in thc ether. Analyses of the materials extracted f ~ u mthe papers, as \vel1 as analyses of the paper before and after extraction (shown in Table VIII), confirm the expected result t8hat petroleum ether extraction renioved a certain amount 3f free rosin and most. of the nonacid ro3in materials from thr sized paper. The increase in the acid number of the resinous material (8) of the petroleum etherextracted paper over that from the unextracted paper indicated that the petroleum ether removed so111e nonacidic material from thr resin in the paper. The methoxyl content of roPini: is lo^, whereas that of natural pulp resins and of rosin neutral bodies is high. Hence the high niethoxyl content of material extracted from the sized paper (corrected for pulp re-ins) also indicated the removal of rosin neutral bodies by petroleunl ether extract,ion. I t also removed some aluminum compound which remained soluble in the petroleum ether extract during evaporation (see ash values in lower half of Table YIII). This is analogous t o results oblained for the extraction of size precipitates. In fact, the carbon-aluminum ratio of material extracted from paper the same as that for mamade with size -4was appt~oximat~ely terial extract,ed from size .\ pwcipitate ( s w Tahle VI). _~__ 'lahle YI I .
>laterial Precipitated f r o m Petroleum Ether Extracts of Kosin-Sized Papers
Paper Prepared with Size A 1,uboratory sample
73.4 72.9
-
Mill samplt.
7.64 7.fi8
.-
73.2
7.66
73.4 73.3
7.463 7.45
4. Ori
18 0
IThile the petroleum ether extractions described above iridieate experimentally the occurrence of aluminum diresinato in size precipitate, there is a theoretical reason for especting such a compound. The formation of ~ ~ ~ J C ~+HSOa~ I I ~would ~ O ~ ) ~ account very readily for the positive charge of rosin size preeipitate a t papermaking pH's. Moreover, such a compound would be expect,ed to hydrolyze wit,h water-\vashing to give hl(C2oHzsO?),OH, which rvould no longer have a positive charge, and which seems to be the material' obtained, from the size precipit'ates prepared in the standard manner. SrZE PRECIPITATE COMPOSITION AND RIECHANIS.\I OF REACTION OF SIZE WITH ALU>\I
The present information about the composit~ionof standard size precipitate may be summarized as follows: 1. .4luniinum triresinate, the normal salt, hydrolyzt'j inimediately upon contact with watcr t o th(, dirtasinatt. and resin acid
ANALYTICAL CHEMISTRY
448 Table Y I I I .
P e t r o l e u m E t h e r E x t r a c t i o n of Sized Paper
Paper Analyses ~~~1~~~ Acid *so l'etroleiini IliSOi ash ExExtracted Corrected Ether tractgbiefJ, by TAPPI for K a t e r (1200' l'di>el Extrarted c; lC Xethod leaf Waterleaf T O (I 37 0.40 7 [no A l t ( S 0 4 1 ~ 0.35 Y __ __ used] 0.38 8 Waterleaf Yes 0.37 0.35 21 0.3i 0.23 21 [no .41~(90r), _ _ __ used] 0 37 0.30 21 No 11 n4 1.90 112 Sized paper 1 ,83 118
c.)c',
-_
Yeb
1.88 1.30 1.20
0.53 0 . 63
115 129 131
142
1ti.i
_-
__
_-
0 .?I3
1 2a
130
> l a t e r i a l i n Clear P e t r o l e u m E t h r r Extract LIaterial Extracted (Based .Ish Neth.\I C , (1P0Oo C.), o ~ y l , ( C a k d . ) , on Pull)), From \Vat erle af [no A h ( S 0 ~ ) Qidtvl
paper
%E
'4
7c
7c
0 114
79 6 80 2
0.77 0 80
0 7i
76.8 76 9
0.43 0.46
c
c, AI
~_
1.49 0.06 1.68 Sized paper (corrected for waterleaf) 1' Paper moistened with concentrated &SO1 before ashing. h TAPPI Method T408-111-44. < For grim rosin.-, 0.10 t o 0.14';'c niethoxyl. 76.8 76.3
2.82 3.28
0.44 0.38C
( 7 ) . Therefore, the normal salt cannot be expected in size precipitate. 2 . Size precipitate contains no alumina [Al(OH),] as such. If present,, the amount is negligible ( 4 ) . 3. It contains a material of the composition of aluminum diresinate which can be isolated from sized paper. 4. It, contains free resin acids and rosin neutral bodies. This follows from items 1 and 3 above, and the fact that precipitatr from neutral size has the over-all composition of aluminum triresinate. This is also indicatrd by the composition of matrrial extrarted by pet,roleum ether. 5 . The amount of aluminum sulfate required to form the precipitate is determined only by the amount of sodium resinate in t,he size: the free resin acids are carried into the precipitate un,changed. [Bialkowsky ( 1 ) has shown that sodium resinate undergoes rcq- little hydrolysis at beater concentrations and at niodcrate temperatures (30" C.), and that the presence of pulp aflects the hydrolysis of sodium resinate. The effect of cellulosic inaterials on size precipitate composition will be. considered in a subscquent paper.] 6. The residues from prxtrolrum ether extractions of size prceipitates have a composition approximating that) of aluminum dircwnatc..
In the case of item 6 in three out of the four residurs examined, the carbon-aluminum ratio was lower t,han the theoretical value; rt'asons for this have lx~c~n discussed above. Because small amounts of sodium sulfate, alumina, or aluminum monoresinatc (computed as 0.3, 0.3, a.nd 3Yc, respectively) in the original precipitate, as n-ell as oxidat,ion, could account for t,his low carbon-aluminum value,, the residues are assumed to be the diresinate. The material removed by the petroleum ether contained free resin acids and neutral bodies of the rosin. It has heen assumed that the carbon content of these materials will be about the same as for resin acids, and their ashes zero. This view is supported by rosin analyses, by the analysis of resinous materials removed from pulp (Table VIII), and by the nearly identical analyses of precipitates from size B and from a size containing no neutral bodies (Table IV). The aluminum-containing material removed from t,he size precipitates by petroleum et'her could not be the normal salt'
or alumina. d s aluminum monorcsinate nould hrs c+qxrted to be even less soluble in petroleum ether than the diresinate, a furt,her assumption was made that this material was soluble (or dispersible) aluminum diresinate. The absence of aluminum monoresinate was confirmrd by t h t , analysis of the matrrial first iiis.)luhilizcd upon concc.ntrntiiig thr, c3xtrac.t (analysi9 ot' thc diresinate). Assumptions. T o cxplain tho mechanism of the aluminum sulfates-sizc leaction, the following assumptions w r made ~ ~ on nt thc h q i s of p i ~ ~ w itifoi~niatimi:
1. I k i d u c ~ sof pctroleum t,ther estractioii of siw prwipitatc, art: aluminum dirc~sinntc~. 2 . Resin acids a n d rosin ncwtral bodies h a w ahout t h e sanir p r crnt carbon and no ash. * 3. The aluminuin-containii~gmaterial removed from size preripitate by pctroleu~nether i- a soluble (or dispersed) aluminum diresinate. assumptions point t o the The pwscnt infoi~iiiationaiid 111 react ion HIhs J. .Il-+3 I ~ c ~ s - H?') +.IIRPs?OH J. (1) as thc nic~chanisni rvhcrehy ,iize precipitate is formed. (The symbol Res is used t o repi'esent, t.he various win acid radicals present in the rosin size solution.) Sonarid material and free rosiv, originally prvsent in thr sizr, nil1 he carried down unchangrd with the reaction products of Equation 1. The preripitate formcd by the rcac'tion will be 9 7 . 5 7 hv \\right of the sodium resinate addrd, and will consipt of 32cG rrsin acids and 68c; aluminum diresinate, 113. w i g h t . This corresponds very closrly to the composition arid ~nrclianismsuggested by Seugebaucxr (3') 25 years ago. Tahlc I S contains the theoretical values of carh-n-aluminum for various mirtures. ohtained by the use of assumption 2 and Equation 1. Tahle I S indicatrs that in thvery case the carbon-aluminum ratio is a good approximation t o (though slightly lower than) the theoretical value based on Elquat,ion 1. The small discrepancy is prohahly d u t to inadequate n-ashing of the .precipitate.
+
+
+
Tahle 1X. Coniposition of I I i x t u r e s of Resin Acids a n d A l u m i n u n i Diresinate
+
Aluminum Resin Acids Diresinate, Neutral Bodieh,
9%
7c
0 10 20 30 40 50 60 70 80 90 100
100
90
80
70 fin 50
40 30 20 in 0
C,
AI,
79.5 79.0 78 5 77.9 77.4 76.9 76.4 75.9 75.3 74.8 74.3
0 0.418 0.836 1.26 1.67 P 09
%
%
".dl
2.93 3.34 3 76 4.18
CIA1
...
189 93.9 62.0 46.2 36.8 30.4 23.9 22.5 19.9 17.8
P r e c i p i t a t e C o m p o s i t i o n According to Proposed M e c h a n i s m ' J Aliiminuni Diresinate. Others, .Ictual C/A1 Found 7c Size 24.8-25.3 37 n 63 31.2-34.2 .j0 50 .I 62.7 83 17 c
oc
CONCLUSIORS
Size precipitate n-as easily dispersible in many solvents. S o true solution should be assumed, despite visual clarity, unless the material can be shown to be a solut,ion by ultramicroscopic examination. Many solvents changed the composition of the size precipitate extraction residue, possibly by hydrolysis caused by traces of water. Et.hyl ether and alcohol showed this effect. Fresh size precipitates, prepared by a standard method, tontained lit.tle or no oxidized resin acids. The standard preparation was made by adding 25 ml. of 3.0Gc size to 1 liter of dis-
449
V O L U M E 20, NO. 5, M A Y 1 9 4 8 tilled water and adding aluminum sulfate to obtain a final pH of 4.5. Pet,roleuni ether extraction did not change the co~npositiori of t,he precipitate residue, even in the presence of a water phase. ii material of the composition of aluminum diresinate was isolated froni size precipitates and sized papers by prtroleum ether est~raction. Pet,roltwni ether wtrartioii of size precipitates prepawd from sizes of 0 tlJ 7Scc frw rosin lrft a residue of the approxi~nate composition of duniinum diresinate in earh case. The reaction AI-++ 3 RcsH 2 0 +AIIit~spOH4 HI& 4
+
+
+
and the free rosin rontent of tht, original sizes used can account for th(1 size prwipitatr. compositions obscJrved.
LITERATURE CITED
(1) Bialkowaky, H . 1933).
K., Papcr Tradt J.,97, No. 13, 29-46 (Sept. 28,
(2) Fieser, L. E., "Experiments in Organic Chemistry," pp. 350-9, Boston, D. C. Heath and Co., 1935. (3) Keugehauer, E. L., Papier-Fabr.. 10, 1308--12 (1912). (4) Price, D., I n d . Eng. Chem., 3 9 , 1143 (1947). (5) Price, D . . and Camei,on, D. D., P u l p P a p e r dIag. Can., 47, No. 3, 142-8 iConwiition Issue., 19461 : Pa,wer Trade J... 123.. No. 25. ~
I
,
I
35-41 (Dee. 19, 1946). (6) Hobinson, J.J., Ihid., 103, No. i . 104-12 (Aug. 13, 1936). ( i )Srurlin, H., and Vandenberg. E. J.,Hercules Powder Po., unpubfX)
lished work. Tec-lr.X 9 h o r . Pulp and Paper Makers, Method T408-111-44.
KECBIVEDAugust 28, 1947. Presented before the Division of Cellulose Cheinistry a t t h e 111th hleeting of tlir . i \ I E R I C A S C H E M I C I L SOCIETY, A t h n t i c C'ity, S . .J.
Quantitative Separation of Calcium, Barium, and Strontium SlLVE KALLRI WN, Lerloux & Co., 155 S i x t h Aae., New York, 4 niethod for the quantitathe separation of calcium, hariuni, antl strontium is exclusively based on the difference in soluhilitj of the chlorides of the three allialine earths in n-hut) 1 alcohol containing hydrogen chloride and i n h)drochloric acid containing n-biit)l alcohol. From the hutjl alcohol solution of the perchlorates of calcium, harium, and s t r o n good method. 1 t'oi, t h r s quantitative separation of ralriuni, baiium, and strontium from r m h other arc' c k >( ribed in the litri aturc. From the anhydi~oui:nitrates calciuni ma>-be extracted with ah.wlutc1 alwhol (12). or pix+~ably, with a mixture of (qual volumtzs of absolute alcohol and anhydrous ether (5,10) 01' with c*onct~ntr:itednitric acid (9). As these tiictthods are ba t~stizction,i~treatnitmtof thc filtered niti,atrs of barium antl,!or strontium, \vith intermittent solution in water, is r q u i r d in clc~alingwith l a r p r a n i ~ u n t sof the alkaline, earths. Far superior therrfore is the precipitation method ,suggt,stod t)>. \Yillarti and Goodspeed ( 1 3 ) in which harium and sti~ontiuni i i i ~ prt4pitnttd as nitratcs by the addition of roncentratcd nitt,ic:ic*itl.to thv aqueous solution of the mixed nitrates. I3arium nitrate appctms to be practically insoluble in 7 6 5 nitric acid, :iffording a clean-cut separation from calcium. Strontium nit i x t c , beirig appreciably soluble in the 7 6 5 nitric acid, requii 8OC, : i d wiicrntration: this m c e d t a t w repreripitation of the s-trontiuniriitr,ate in dealing with more than 25 mg. of calcium. atisf:ictory are the methods for i;cpar:itinp barium arid
Table 1. Solubility of Barium Chloride in a 4 to 1 Mixture of 10.6 ,V HyJrochloric Acid and Ether in Presence of Strontium Chloriclt*at 20' C. H.,(-I
~
'l'liLe1, 61.0I h
0 .03J cr.0
,,,
. .
....
.... ....
0.0500 0.1000 0,2000 0.2500 0 . ROO0 0.2000
BaCI? Found GTam 0.0490 0,0981 0.2980 0.2971 0.2482 0,2494 0.2513 0.2847 0.2591 0,0309 ~~~
Volu iiie Esclu'ire of TTashings .1J1. 25 50 50 75 75 75 75 75 73 75
Error
Gram -0.0010 -0.0019 -0.0020 -0.0029 -0.0018 -0.0006 +0.0013 +0.0047 +0.0091 +0.0009
N. Y .
tium, a 20% solution of hjdrogen chloride in n-but:l alcohol (Willard and Smithreagent)precipitateschlorides of harium and strontium; calcium chloride is very soluble. From an aqueous solution of the chlorides of harium and strontium, a 4 to 1 mixture of 11.0 X h)drochloric acid and n-butyl alcohol precipitates barium but riot strontium chloride.
strontium fi,om t:avh otht.1. T h r l'avorctl procedure consists in precipitating harium chromatt, at a pH of about 4.6 in a buffered acetic acid-acetatc solution. Iid of a 4 to 1 niixturts o f 10.6 .V hydrochloric acid and ethtxr (4)has scarcrly bren ci.iticall~.tfisrusscd in tht. literatures. This writer carried out a riuni1)c~i~ of tc (Tai)lc I), but \\-as unable to confirm the findiiigs o t (;ooch antl Soderman, ivho olaini that their nirathotl :iffoi,(ts i i rl~~:~ii-cut ity:wation of thc! tivo alkaline earths. I n the first place, thr solubilit>- of 1)ai.iuni c~hloi~icte appears to tic, considwably givater than can bi, gathered floni the data subniittcd hy the authors (.$), \Tho stsitti that the soluhility of bmiuni chlorid(. in 75 ml. of n 4 to 1 mixture of 33"; (10.6 S)hydrocmhloric acid and cxthc.r at 20" (', amounts to l(w than 0.5 mg. I I a r (L7), o n thts oththr hand, ha> shown that thr, solubility of barium chloride i n a 1 to 1 mixtuw o f concent~.iitid (12.0 S ) hydrorhloi,ic~acid arid ethct. at 20" ('. aniwnts to apprbsiniattxly 1.3 mg. of barium chloi.idc1 ptsi' 100 nil. of solution arid statw +hat the "solubility illcrc,ases ~ ( ~ I . ? .,.allidly n-ith tllc diIninuation in strength of acid." From thv data in Table I it would appear that the solubility of barium c*hloiidr in a 4 t o 1 mixture of 10.0 .\- hydrochloric acid and ethc>ramounts to about 4 m e . per 100 nil. of solution, antl that a niarktd tt~ndcncytoward coprecipitntion of sti~>ntiuni chloride with the hxrium chloride may offset part or all of the solubility 1ossc.s of the baiium chloride or may even cituv high bariuni rtLsults. F I L L A R D AKD SMITH KE.4GENT
I n earlier papers (6, 7 , 14) the effrct, of a 20% solution of hydrogen chloride in n-butyl alcohol (Willard and Smith reagent)
