Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Separation of aluminium and beryllium using amyl alcohols Grassie, Vernon Robert 1942

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-UBC_1942_A8 G7 U8.pdf [ 4.76MB ]
JSON: 831-1.0062162.json
JSON-LD: 831-1.0062162-ld.json
RDF/XML (Pretty): 831-1.0062162-rdf.xml
RDF/JSON: 831-1.0062162-rdf.json
Turtle: 831-1.0062162-turtle.txt
N-Triples: 831-1.0062162-rdf-ntriples.txt
Original Record: 831-1.0062162-source.json
Full Text

Full Text

G 7Uz  SEPARATION OF ALUMINIUM AND BERYLLIUM USING AMIL ALCOHOLS  Thesis submitted by: ¥eraon Ri Grassie  i n p a r t i a l fulfilment of the requirements f o r the degree Master of Arts  THE UNIVERSITY OF.. BRITISH COLUMBIA  March 1942  Jbrewor&r The research was carried-..out under the expert direction of Br. J, Allen Harris.  lernon R«, Grassie  Contentst  Introduction - -  —• - - -  -Methods of Analysis (a) Hydrolysis Method  .-  3  —• .  •- -  3  (b) 8-hydroxy quinoline; method  6  (c) Tannin Method- f o r Beryllium  -•-  _ 7  (d) Analysis of Iron and Chromium -  - -  S o l u b i l i t y Determinations. -  ,7  - ~  (a) Beryllium N i t r a t e ; — - —  ..  —, =  (b) Aluminium Nitrate' -  •  (c) Zinc Nitrate  .  (d) Ifegnesium t i t r a t e -  _ .  - -  8 13  -  .—  19 20  (f ) Chromium Nitrate -  _ - . _ _ „ 21  Investigations f o r C o l l o i d a l Propertiesv- — Conclusions -  8  17 .—  (e) Iron Nitrate -  1  —>-  •—. <—. - — -  Suggestions f o r Further Investigation References -- --  - - ^ .  —0O0—  _  •-  21  ~ ~  22 24  _ _ _ 25  (1)  The use of amyl alcohol i n the separation of aluminium and l beryllium i s described by Browning and Kuzurianf the separation depending upon differences i n s o l u b i l i t y of the nitrates i n this solvent. It i s stated that aluminium n i t r a t e i s completely insoluble i n amy! alcohol while beryllium nitrate i s soluble to the extent of 0.02 to 0.045 grams of BeO per ml. The actual method of treatment involves evaporating of the aqueous solutions of the nitrates to a few drops and adding 10-20 ml. of amyl alcoholi The whole i s then brought to the b o i l i n g point {128-150° C) and according to the a r t i c l e , complete dehydration- i s indicated when the fumes burn q u i e t l y a t the mouth of the test-tube. Upon cooling and f i l t e r i n g - the beryllium-nitrate i s contained i n the f i l t r a t e , the aluminium'nitrate being;retained: by;the- filter'. Browning and Kuzurian determined the f i l t r a t e by extraction with water i n a separating funnel, (two treatments with four times the volume of water having been: found satisfactory) and-the beryllium content then estimated by p r e c i p i t a t i o n with NH^OHj f i l t e r i n g , and igniting to BeO. It i s "noted- that "when Be: and A l are"' present'together^ small varying amounts of Be are retained by the A l even-after two treatments. The method i s a t l e a s t recommended by the investigators for the preparation of beryllium  free from aluminium.  In this investigation, s o l u b i l i t i e s of beryllium and: aluminium nitrates were checked i n the four amyl. alcohols (normal, iso, secondary, and t e r t i a r y ) and i n addition, s o l u b i l i t i e s ! of nitrates of several related elements were determined.•In each casey no attempt was made to estimate s o l u b i l i t i e s at.other than: room temperatures. However i t was  (2)  observed that the s o l u b i l i t y of aluminium nitrate remained extremely low i f s olutions were f i l t e r e d while hot; Instead of evaporating aqueous solutions of the nitratesy portions of the s o l i d nitrates were used  t  less reagent being:necessary f o r the  dehydration; In actual practice this.wouldn't be possible, but i t was noted that results obtained by. this procedure checked with; results obtained by evaporating solutions. Evaporation of the solvent was carried out i n 50 ml. erlenmeyer flasks, and i n most.cases evaporation of onehalf of the solution.was considered s u f f i c i e n t to ensure dehydration. It was observed during the investigation that' normal and i s o amyl alcohols were the only isomers suitable f o r this workj the secondary C  and t e r t i a r y compounds having boiling points too low (11£ G and 102° C respectively) to guarantee-complete dehydration of the nitrates. In addition, although the nitrate of beryllium seemed quite soluble i n t e r t i a r y amyl (perhaps because of incomplete, dehydration) i t was found to have very limited s o l u b i l i t y irr secondary amyl. So i n the subsequent y  investigations f o r the s o l u b i l i t y of the other nitrates, only normal and iso amyl. alcohols were considered, and since they proved to be very similar i n their solvent properties, only the i s o compound was actually employed.  .  >  As previously mentioned, Browning and Kuzurian observed that some of the beryllium nitrate was retained by the aluminium. This phase of the investigation was accepted without checking, but disregarded since treatment three or four times with the solvent would probably be s u f f i c i e n t to remove a l l but;the l a s t traces of beryllium. However the s o l u b i l i t y of aluminium nitrate i n the presence of beryllium was determined and found to be considerably higher;than f o r pure aluminium  (s)  nitrate. The aluminium nitrate -is c o l l o i d a l i n this case and f i l t r a t i o n through.a thick asbestos f i l t e r seems sufficient to retain most of the colloidal nitrate.  Methods of.Analysis; Since a very.large number of determinations were to be madej and since the s o l u b i l i t i e s : to be determined vrere not to be necessarily absolute, the. most:rapid'and least specific:method:was: sought, even with the  s a c r i f i c e of a certain degree of -accuracy. (a) The.method f i r s t considered.was that of Ivanov, i n which easily  hydrolyzed s a l t s (such as nitrates) of A l , Be, Zn, Fe, etc. are-determined by hydrolysis i n the presence of sodium.thiosulphate, potassium iodide, and potassium iodate. HNOg from the hydrolysis of the nitrates releases iodine from the KI and KlOg which i n turn reacts with an excess of thiosulphate•* The excess thiosulphate i s then; found-by t i t r a t i o n with standard iodine solution* The method, of coursej depends upon amount of hydrogen ion formed as the hydrolysis proceeds, and presence of either hydrogen or hydroxyl ions such as may arise from impurities i n the reagents used, w i l l lead to inaccurate results. That i s to say, the actual s a l t being determined must be 100$ pure-. This represented no disadvantage i n i t s e l f as fas as this investigation was concerned, because compounds being used were i n the pure-state. In the case of A l and Be, the combined reactions are as followss 2A1(N0 ) +KIOg+SKI^SHgOtSNagSgOg S  —->  3  Alg(0H)gV 6KNQg,-«- 3NagS Q * 6KaI 4  3Be(l0 )g + KIO3 + 5KI <-3HgO + 6NagS 0 5  2  g  ————$»  3Be (0H) 4- 6KN0 2  6  5  SBtegS^ + 6NaI  (4)  The procedure i s not recommended f o r amounts exceeding 0.05 grams of BeO or AlgO . To 200 ml of solution, 10 ml of 5% KlOg and 10 ml of g  1% KI are added and then a known excess of: tenth-nornial NagSgOg. The whole i s then boiled f o r not more than 5 minutes, the flask being covered o  with a watch-glassj cooled to.below 20 C, and t i t r a t e d with tenth-normal iodine> using:starch-indicator* From the equations i t i s calculated that 1 ml tenth-normal sodium thiosulphate i s equivalent toj.0.001704; gm. AlgQg or to 0.001255 gm. BeO. To check the accuracy of t h i s method, standard solutions of -  aluminium and beryllium nitrates were prepared and calibrated using the oxine method: and-tannin methods respectively (these methods w i l l be discussed l a t e r ) and various amounts taken f o r analysis. 1  The following results were obtained; •  Table: 1.  BerylliumBee added Thio.used-H/10. BeO found 0.0032 0.0065:  2.58 • 5i2i  : .  D i f f . gm.  0.0032  0.0000  0.0065  0.0000  0.0096  -0.0001  3»  0.0097  4.  0.0130  10.39  0.0130.  0.0000  5.  0.0162  12 a 90  0.0162  0.0000  6.  0.0324  25.59  0.0321,  -0.0003  7.  0.0487  38; 97  0.0489  * 0.0002  8.  0.0649 .  50.96  0.0639  -0.0010  7*68v  (5)  Aluminium AlgOg added  Thio used N/10  AlgOg found  D i f f . gm.  1.  0.0021  1.49  0.0025  0.0004  8.  0.0042  2.84-  0.0048  -+- 0.0006  3.  0.0065  4.36  0.0074  + 0.0009  4  0.0087  5.41  0.G092-  + 0.0005  5.  0.0109  7 « 55..  0.0125  0.0016  6a  0.0218  14.37  0.0245  + 0.0027  7.  0.0527  20.60  0.035L-  +. 0.0024  8. •  0.0456  0.0467  +• 0.0031  9.  0.0545  34.71  0.0591  + 0.0046  10.  0.0654  41.45  0.0706  + 0.0052  8  ©38'  The method thus proved t© be very accurate i n the case of beryllium at least f o r amounts- of Be0 as high as.0.05 gm, but was very poor f o r the determination-of aluminium. Since the s o l u b i l i t y measurements made are. a l l carried out i n the presence of a certain amount of amyl alcohol, i t was then advisable to check the a p p l i c a b i l i t y of the hydrolysis method i n the. case--of Be i n the presenee of the solvent. Known amounts of Be(N0g)g were dissolved with b o i l i n g i n i s o amyl alcoholj extracted with water i n a separating funnel, and determined, by Iv&nov's method. Solutions were boiled f o r various lengths of time to observe any effect this prolonged 'heating might have on.the,condition of the nitrate., Apparenty there i s some sort of decomposition which as the following tabulation shows, leads to inaccurate results} but the.nature.of the reaction has not been examined. Prolonged b o i l i n g with the .solvent has no marked: effect over*and above that produced by boiling merely:long enough:to obtain dehydration*  (6)  Table 2. Be8 added  Thio used N/10  0.0179  9.25  b.  0.0144  7.43  Co  0.0107  . 5*54  d.  0.0072.  e.  BeO found . Diff.. gm„.  Comments:  -0.0063  Nitrate solns.  0.0093.  - 0.0051  i n i s o amyl  0.0070-  - 0.0057  boiled just  3.71  0.0047  - 0.0025  long, enough f o r  0.0056  1.86  0.0025  - 0.0015  dehydration.  a.  0.0179  8; 52-  0.0107  - 0.0072  Solns. boiled  b.  0.0144  7.43.  0.0093  - 0.0051  to ensure  c.  0.0107  5*50  0.0069  - 0.0058  dehydration  d.  0.0072  3;62  0.0045  - 0.0027  then refluxed  e.  0.0056  1.84  0.0023  - 0.0013  f o r 50 min.  .0.0116  It may be seen from these values that.the hydrolysis method i s 1  useless f o r the purpose. Even- worse discrepancies were obtained f o r A l , treated i n a similar' manner, and consequently the p o s s i b i l i t y of using t h i s a n a l y t i c a l scheme'was abandoned; (b) The method involving the use of 8-hydroxy quinoline was decided upon f o r the'analysis, the procedure, being quite rapid, of high accuracy, and widely applicable because.of i t s non-specificity. Thus s 4 s oxine was employed i n the determination of Al,.Zn, Mg, and could be applied equally as v/ell to Ni, Co, and other Third' Group metals the s o l u b i l i t i e s of the nitrates of which were not determined-in this research. Due to the f a c t that-oxinates of the> metals considered are soluble to varying extents i n a l i p h a t i c alcoholsj and since a l l of the determinations were:made i n the presence- of a certain amount of dissolved amyl alcohol, i t was necessary to check the oxine procedure by the p r e c i p i t a t i o n of the oxinates i n solutions containing the solvent.  (7) -  Values of accuracy found i n every case, w i l l be g i v e n i n the section dealing, with actual s o l u b i l i t y determinations. •(c) The oxine method' cannot be applied to the analysis of Be, so unfortunately -it was use a slower/gravimetric method. 7  The guanidine carbonate method, involving a t least 12 hours.of digesting, was prohibitively slow. The tannin procedure was then decided upon f o r beryllium and used throughout this investigation. As i n other, methods of analysis considered, the accuracy of the-tannin procedure was determined i n the presence of amyl alcohol. A standard.beryllium.nitrate solution was prepared, and varying'amounts were then taken-for analysis i n presence of amyl alcohol. In each case, 100 ml of water saturated with the i s o compound were added and the determination carried" out i n the usual manner.  The following values i l l u s t r a t e the a p p l i c a b i l i t y of - the.methods • BeO.added  Table 3, • ' • BeO- foundD i f f . gm.  1.  0.0052  0.0032  0.0000  £ »'  0.0065  0.0067  0.0002  5.  0.0130  0.0129  0.0001  4.  0.0162  0.0162  0.0000  5.  0.0227  0.0230  0,0003  6.  0.0324  0.0325  0.0001  7.  0.0487-  0.0485 -  8.  0.0649  0.0653  —  0.0002 0.0004  It i s apparent from this that presence of the solvent has no marked effect of any kind.on the determination, and i t was hardly to be expected, t h a t ' i t would* 9  (d) Analysis of i r o n was carried out using hexamethylene tetramine.  (8) lo Chromium was analyzed, by the persulphate method..Methods are described i n the section- dealing:;with actual- s o l u b i l i t y determinations. S o l u b i l i t y Determinationsr 1. Beryllium; nitrate-i' Varying weights of the - crystals were treated with.about 50 ml of the solvent, and the whole evaporated to a volume of about 15 ml. In the case of secondary and t e r t i a r y compounds> evaporation was carried out with a second portion of the solvents, to permit dehydration; For secondary amyl there was.a crystalline residue-indicating that the s o l u b i l i t y of the nitrate i n this particular:solvent"had been exceeded. Measurements made with- large and.small amounts of. beryllium nitrate indicated a more- or less.constant value f o r the s o l u b i l i t y ; however, i n the case o f normal, i s o .and t e r t i a r y compounds}, no limiting, s o l u b i l i t y could be ascertained. The value of 0.045 gm of BeO per ml as determined and reported by Browning and Kuzurian was f a r exceeded, the resulting solutions merely becoming increasingly viscous. Accordingly,.for these investigations> v i s c o s i t y of the solutions was estimated by measuring efflux time from a glycerine-water calibrated v i s c o s i t y tube, and the v i s c o s i t y plotted against concentration of BeO. Eventually with higher concentrations the solution becomes too gummy to permit f i l t r a t i o n , even though the extent of the s o l u b i l i t y . i s i n d e f i n i t e i so. i n the accompanying graphs the estimated maximum concentration (and hence the maximum v i s c o s i t y ) which would permit e f f i c i e n t f i l t r a t i o n through coarse f i l t e r paper i s noted. Beryllium nitrate solutions were.not f i l t e r e d (except i n the case of the secondary amyl) since-the compound was found to be completely soluble i n each determination, but ordinarily i n a separation procedure, f i l t r a t i o n would be badly limited by the v i s c o s i t y . As was  (9) mentioned beforej normal and iso amyl alcohols were found to be by f a r the most satisfactory of - the isomers f o r this work, and consequently i n subsequent determinations only the solvent properties of these two are to be considered. Since•normal and iso compounds behaved very s i m i l a r l y in-case of beryllium^ -only the iso amyl was actually employed i n making s o l u b i l i t y measurements of the other nitrates. It i s to be t a c i t l y assumed that normal amyl behaves i n an.identical manner. Upon dissolving the portion of beryllium n i t r a t e , the resulting solution i s cooled to room temperature, exactly 10 ml removed with a pipette-and -transferred to a 100 ml. separating-.funnel f o r extraction with four 90 ml portions- of water. The extraction i s then made,up to 500 ml i n a calibrated f l a s k and- suitable amounts of the-diluted.solution taken f o r analysis by the tannin method as described by Nichols and Schempf. Slight modifications made i n this tannin procedure are as follows: i n the f i r s t place, the pH i s adjusted p r i o r to digestion to the i s o e l e c t r i c point of Be(0H)g (pH of 7.5) by means of a potentiometer with"glass and calomel electrodes, rather than by the use of litmus indicator; and secondly, 15 it grams. of.NH^NOg i s added to each solution i n a t o t a l volumes of 300  ml,  before the pH i s adjusted by addition of NH^OH. The ammonium-: nitrate seemed to improve and- hasten the f i l t r a t i o n of the bulky precipitate by removing a l l tendency of the precipitate to pass-through the . f i l t e r . Precipitates f i l t e r e d on Whatman f i l t e r paper No. 50, using suction, were pre-dried i n an e l e c t r i c oven at,HQ°G f o r £ hours or more, and ignited i n platinum crucibles to BeO i n the usual manner. It was  shown (table 3) that- influence of as much as 100 ml of  iso-amyi-saturated  water .(equivalent to about £.7 gm.  of the alcohol) was  negligible, so no correction was necessary f o r this factor i n the  (1G)  s o l u b i l i t y measurements made. Correction f o r the volume occupied by the solute, however, was considered advisable. This correction though, i s at best only an approximation* since density of the nitrate considered as ;  anhydrous, i s estimated by extrapolation of nitrate densities from the Periodic table;  The series employed- was: Ba-nitrate, Sr-nitrate,  Ga-nitrate| densities of the anhydrous nitrates plotted against atomic weights. The r e s u l t f o r Be(NQg)g was density: 1.55 gm. per cc. Also, i t i s assumed that say 1 ml of s o l i d Be(N0g)g w i l l continue to occupy 1 ml i n the dissolved state, this never being exactly the case. The various arithmetical adjustments necessary are obvious, and are only summarized i n tabulations to follow-; ''• (a) normal amyl: Efflux time 20 deg. .,  ..Table. 4 Viscosity .Vol. tafcen c.p. for anal;  BeO  Soly. Gorr. TJncorr.; Vol.  Gorr. SOly.  1.  59  5.2  100  0.0075  0.0057  0.987  0.0038  2.  75  7.0  100  0.0146  0.0075  0;975  0.0075  5;  97  9.5  100'  0.J327G  0.0l35  0.954  0.0142  ••4r.  180  21.5  100  0.0546  0.0273  0.906  0.0301  5e  275  34;5  50  0.0551  0.0331  0.887  0.0574  6.  390  49.5  ' 50  0.0402  0.0402 '0.862  0.0466  'K -  500  62.7  50  0.0459  0.0459  0.843  0.0544  •8s-  720  85.5  50  0.0530  0.0530  0.818  0.0648  9. 10.  1180  122.0  •50."  0.0606  0.0606  0.792 .0.0765  2064  200 #  50  0.0777  0*0777  0.734 . 0.1060  (#- approximate-viscosity)  .  In above tabulation, e f f l u x i s measured i n secondsj viscosity i n centipoisesj and "Corrected Volume" column refers to corrected unit  solute volume. S o l u b i l i t y i s i n grams of BeO per ml. (b) iso-amyls Efflux time  Viscosity Vol. taken c.p. f o r anal.  Soly., Corr. Uncorrr Vol.  BeO  Corrected Solubilii  I...  61  5.8  100  0.0089  0.0045  0.985.  0.0046  z.  82  8.0  100  0.0170  0.0085  0.971  0,0087  3.  107  11.0  100  0.0262  0.0132  0.957  0.0158  4. •  117  12.0  100  0.0286  0.0143  0.951  0.0150  5m  180 .  21.5  50  0.0239  0.0259- . 0.918  0.0261  6.  285  35.5  50  0.0550  0.0350  0.880  0.0398  7.  390  50.0  50  0.0585. 0.0385  0.868  0.0444  8.  452  56.5  50  0.0407  0.0407  0.860  0.0473  9.  635  77.0  50  0.0488  0.0488  0.833  0.0586  10.  819  94.0  50  0.0536  0.0536  0.816  0.0656  (g) secondary-amylr In t h i s case* s o l u b i l i t y can be exceeded, leaving a c r y s t a l l i n e residue, a f t e r boiling three times with the-solvent (total volume used being about 50 ml) down to a volume of about 15 ml. Solutions were f i l t e r e d on Whatman #1 f i l t e r papery and filtrate-analyzed as before. Viscosities of f i l t r a t e s were measured, and f o r the small amount which did dissolve, there- was negligible increase i n viscosity over and above that of the pure^ solvents Varying weights of the crystals were taken, and the f o l l o w i n g r e s u l t s obtaineds Efflux time 20 deg  Viscosity Vol. taken c.p. f o r anal.' BeO  Soly, Corr. Uncbrr. Vol.  Gorreeted Solubility  1.  22  2.0  50  0.0121  0.0121  0.958 .0.0127  2.  23  2.0  50  0.0116  0,0116  0.960  0.0121  3.  22  2.0  50  0.0117  0.0117  0«,9"6&  0.0122  j j  (12) The average value obtained f o r the s o l u b i l i t y i n secondary amyl i s 0.0125 gm. BeO per ml. (&).tertiary-amyl; Efflux, time 20 deg.  Viscosity Vol. taken. Soly. Corr. o*p«. ' f o r anal. . • BeO Uncorrb Vol.  Corrected Solubility  1.  63  5.9  100 :  0.0066- 0.0033  0.989  0.0034  2.  87  8.2  100  0.0156  0.0078  0.973  0.0080  5.  132  14.0  100  0.0178  0.0089  0.963  0.0092  4. -  155  17.0  100  0.0218  0.0109  0.961  0,0113  5.-  262  33.0-  50  0.0215  0.0215  0.926  0.0232  6.  599:  50.2.  50  0.0511  0.0311  0.893  0.0348  7.  596  72.5  50  0.0594  0.0394  0.865  0.0455  The curve f o r c a l i b r a t i o n of the v i s c o s i t y tube i s incorporated as f i g . 1» I t i s to be pointed out, without delay,, that the v i s c o s i t y values given i n the preceeding tables are by no means accurate. They are probably correct to within 5% i n the range up to about 50 c.p«, but beyond that value-, they are merely estimations. The curves drawn ( f i g . 2) therefore, would require complete revision i n any future work concerned with the application of the v i s c o s i t y method to quantitative analysis. However, i t i s rather apparent from results obtained, that there i s • a consistent r e l a t i o n between v i s c o s i t y and the amount of beryllium nitrate dissolved. Not included i n this report are measurements of the v i s c o s i t i e s of beryllium nitrate solutions taken at approximately one day intervals over a period of one. week. I t was found that there.was no change: i n v i s c o s i t y of the solutions i n this length of time* providing of course that the containers were kept stoppered to prevent  evaporation.  No explanation i s advanced f o r the shape or the differences i n the curves plotted i n f i g . 2.  1  DRTR _  _LZTi  : Substance . S p e c G>ca\/Watar  GVyc.-v/ator  .LOO  in  \£0__ EFFlux (sec.')  0-9999  l&  V. 17GO  V<ot  U9o4  244  1-2.1-2.1  577  c 2nVv p o i - s e s  From  t  30.3 loo  critical  1ab\«s-  < </>  SO  n 0 tP  •• • • •  •  So  •• • .  -H -< -  /•*» «  <bo  •  _  -+•  '  S y  •  0  40  49  o  (>  K >0  Z< >0  3« >0  >o .  4<  EFFLUX  T \ r \ E  6  >o  8< >o  O IOO  9c  3  (se  2 5 dec). C . 1  8  do  s_  v..  o  "w or eg co oo  Ci  E E e o- cr <r ^ 0 0  3,  N  Ccerdipo'vSes)  v --^  VtSCOS T V  3  I I  V  V V  v • \ \ V  \  c- : 0  HI  E It t i o  \ Y \\  _ -0_  . 0_  O.Q54-  V \l  v.  (13) (2 )Alumiriium Nitrate t The treatment formeasuring s o l u b i l i t y of.Al(KOg)g and other compounds consider.ed,was precisely the same as that f o r Be(NOg)g| the methods of analysis of the solutions, however, were varied f o r the particular case. In every instance, where:filtration was necessary, Whatman• 0. f i l t e r paper "was used to remove the excess solute. Aluminium was determined by the oxine method as mentioned e a r l i e r i n this a r t i c l e | no modifications being made i n the standard procedure. Since determinations were- carried out i n the presence of the amyl alcohol 1  the effect of the solvent upon the s o l u b i l i t y of Al-oxinate was f i r s t noted; Following analyses were made; i n the presence of 100 ml of water saturated with isoamyl alcohol:at room temperature. Tablev 5 AlgQ added s  •'  KBrQg-KBr N/10  A 1 0 found P  5  D i f f . gm.  0;0021  2.83  0.0012  - 0.0009  0.0042  8.00  0.0034  • - 0.0008  3e  0.0065  13.58  0.0057  - 0.0008  4.  • 0.0087  19.98  0.0085'  - 0.0002  5.  0.0109  25.15  0.0107  - 0.0002  6.  O.OL53  35.95  0.0155  0.0000  7.  0.0218  51.48  0.0219  •i-0.0001  8.  0.0327  78.04  0.0332.  '+• 0.0005  (1 ml KBrOg -KBr N/10 = 0.0004254 gm. A ^ 0 ) 5  For the larger amounts, of AlgQg (e.g; #7 and #8) i t appears that the bulky precipitate of Al-oxinate tenaciously absorbs small amounts of the excess oxine which cannot be removed even with prolonged washing with hot water. These absorbed: amounts of oxine tend to offset the loss-due to  (14) s o l u b i l i t y of the precipitate i n i s o amyl alcohol, and the net r e s u l t i s positiveo However, f o r the amounts of A l which were determined i n the following s o l u b i l i t y measurements (less than 1 milligram), i t was taken that the loss due to the presence of amyl alcohol was of the order of 1 milligram. This correction i s incorporated i n the tabulations. Amounts of A l were so small that correction; for the volumes they occupied was unnecessary. Reagents Na^SgOg and KBrOg-KBr both N/30 were employed. T h e - f i l t r a t e from-each b o i l i n g was extracted with water i n a separating funnel as-before, and the entire.extraction taken f o r analysis. Table  6  (a) .normal SMJII Vol. f i l t r a t e  KBrO -KBr N/50  5  1*83  0.00026  0.00126  5'  1.65  - • G.00025  0.00125  0*00025  .1. 2.  g  AlnCL found  Corr-.  Soly. 0.00025  :  3.  "5  1.22  0.00017  0*00117  0.00023  4..  5  1.30  0,00017  0.00117  0.00025  (b) i s o amyl; Vol.  filtrate  KBrOg-KBr N/50. AlgQ  ?  found. Corr.  Soly.  1.  5.  1.70  0.00024  0„00124  2.  5-  1.30  0*00017  0,00117 " 0.00025  3.  5  1.95  0.00028  0.00128  0.00026  4  5  1.14  0.00016  0.G0116  0.00023  S  0.00025  Very similar values were found f o r secondary and- t e r t i a r y alcohols, measurements being made a t room temperature. (Filtered a t room temp.) It i s to be emphasized that above results which seem to indicate a s o l u b i l i t y of \~ of 1 milligram- of A^Og per ml., are very approximate. The correct value i s doubtless of this order of magnitude, though, and  (15) the figures given serve as a basis of comparison of s o l u b i l i t i e s of Be and A l nitrates, and indicate: that a separation using this scheme can very easily be effected. Measurements of s o l u b i l i t i e s were then made i n the presence of varied amounts of Be(NOg)^ i n solution. The concentration of BeO i n each case was determined using tannin. A marked increase i n s o l u b i l i t y was observed| and the Al(H0g)g r e s i s t i n g f i l t r a t i o n was p l a i n l y c o l l o i d a l . i n nature; f l o c c u l a t i n g within about three hours standing, and s e t t l i n g to the bottom of the vessel as a f i n e white residues F i l t r a t i o n through a Whatman #1 f i l t e r paper•immediately after boiling and cooling thus involved a passing of considerable amounts of A l through the f i l t e r . Since the value of this method of separation would l i e i n i t s rapidity, the actual measurements made, f o r purposes of comparison? were on the basis-of f i l t r a t i o n immediately a f t e r cooling. Again i t i s to be noted that a f i n e r f i l t e r ; perhaps asbestos and glass.wool, or a Gooch crucibley or even finer f i l t e r paper using suction; would result i n a better separation. I t was impossible to use suction i n any such determination, since evaporation of the f i l t r a t e resulted, and the primary purpose was to jjeasure the actual s o l u b i l i t y . In an application of the method, suction.could be employed, f o r either qualitative or quantitative analysis, and the residual aluminium nitrate could be washed with HgO-free (freshly boiled) amyl alcohol, to remove Be(Npg)g adhering. The following tabulation i l l u s t r a t e s the effect of the presence of beryllium nitrate upon the 'solubility of aluminium n i t r a t e . F i l t r a t e s exactly 10 ml i n volume were extracted and made up to 500 ml; and then i n aliquot portions, both Be-content and Al-content were found. Summary of the tabulation ^appears i n f i g . 5.  (16) Table  7  isoamyl alcohol; BeO cone.  Vol. e: anal.  0.0075  150  5.15  0.00134  0.00234 05,007870*0008  2.  0.0098  150  3.20  G.00157  O;00257  0*0080•  0.0008  3.  0.0120  150  4.00  0.0017-1 0.00271  Q.TOG907  0.0009  4.  0.0152  150  4.75-  0.00202  0*0101  0.0010  5.  0.0207  100  8*29  0.00555. . 0,00453  6.  0.0266  100  10.84  0.00461  0.00561  0^0281  0.0028  To  0.0384  100  12.20  0.00519  0.00620  0.0310  0.0031  8.  0,0444 .  100  16.19  0.00689  0.00790  0.0595  0.0040  0.00502  0.0227 ;  0.0023  # this' s o l u b i l i t y i n grams per- ml i s not corrected f o r volume occupied by either solute. I t merely represents', then, the actual amount of AlgOg associated with given concentration of BeO. The second column refers to the-volume of the extract analyzed} the f i f t h column represents the correction-for s o l u b i l i t y of Al-oxinate i n isoamyl alcoholj assumed to be constant throughout the range. It i s to be concluded from this investigation that use of larger amounts of solvent would undoubtedly lower the amount of Al-nitrate carried throughj due to the decrease i n concentration of BeO. The process i n an actual separation would necessarily involve f i l t r a t i o n through a" f i n e f i l t e r using suction, and washing the residue as previously prescribed. I t would be advisable to l e t the solutions stand f o r over three hours, i n any other case, i n order to permit the c o l l o i d a l aluminium nitrate to flocculate«TFiltration could then be effected i n the manner adhered to i n this work.  d  o o  6 8  —J to—0i_ 0 Ci  0  ID  O «> CD CC  > S V • &. ftrt> (TO*  Solobi \'»ty  \  "X,  c-  g icr_  lo  is 0 0  1  HV /mV. a o  6'  01  8  \  3Q cp.  \ ©  O.OS4 ••• V/>  o  «4  <v£>  0  (17) (5) Zinc Nitrate; Zinc behaved i n exactly the same beryllium,, the nitrate being seemingly i n d e f i n i t e l y soluble i n isoamyl alcohol, the only one of the four dsomers employed as a solvent. The v i s c o s i t y of solutions increased greatly as the concentration of Zn(NO -)o was increased, and v  again viscosities.were approximated, and plotted against  concentrations.  (See f i g . 4). •The method-/.of analysis, .after boiling;.and-extraction i n the manner previously described f o r A l and Be> was the oxine method. In the f i r s t place, e f f e c t of 50 ml of-isoamyl saturated water on the analysis was determined* • Table ZhO added  8  KBrOg-KBr-N/10  ZnO found  D i f f . gm.  -L«:  0..QG63  5.60  0.0057  2. .  0.0127  11.69  0.0119  —0.0008  5.  0.0190  18.00  0.0183  — 0.0007  4.  0.0255  24.55  0.0250  -0.0005  5,  0.0318 '  30.45  0.0310  -0.0008  6.  0.0382  36.63  0.0575  -0.0009  7.  0.0510  48.82'  0.0497'  -0.0013'  8.  0.0637  61.49  0.0626  -0.0011  (1 ml KBrOg-KBr N/10 S 0.001018 gm. ZnO)  '  -0.0006  - 0.0008  Average.  In this case there seemed to be no indication of absorption of excess oxine by-large precipitates of Zh~oxinate>(values 6, 7, and 8), counter-acting the e f f e c t of obvious s o l u b i l i t y i n presence of the i s o amyl solventj so the average of the eight-values'was taken, and the figure 0.0008 used as the correction i n the work to follow.  (18) Solutions of Zn(N0 )g i n isoamyl, of varying concentrations, were 3  then prepared i n the usual manner$ and their v i s c o s i t i e s determined. Also, as before, 10 ml samples were extracted and made up to 500 ml. Correction i s made f o r the volume occupied by the dissolved nitrate the density of the anhydrous nitrate being estimated from the Periodic Table (Series: Hg, Cd, Zn,:Mg), to be 2.50 gnu per cc. Table  9  Efflux Vise, Vol. f o r KBrOg-KBr 20 deg c .p. analysis N/10 ZnO  ZnO Gorr.  Soly. Corr. Uneorr. Vol.  Gorr. Soly.  1.  71  7.0  50  24.81  0,0252  0.0260  0.0260  0.970  0.0268  2.  96  9.5  40  25.06  0.0235  0.0243  0.0304  0.965  0.0316  3.  122  15.0  30  27.06  0.0275  0.0283  0.0473  0.945  0.0501  4.  125  13.0  30  28.81  0*0295  0.0501  0.0502. 0.942  0.0555  5.  145  15.5  20  0.0228  0.0234  0.0585  0.952  0.0627  24.0  20  28.05  0.0285  0.0293  0.0733  0,915  0.0802  6. 200 7.  246  31.-5  10  16.95  0.0172  0.0180  0.0900  0.895  0.1000  8,  408  52.0  10  21.04  0.0214  0.0222  0.1110  0.872  0,1271  9.  560  69.5  10  23.55  0.0239  0.0247  0.1255  0.856  0.1442  782  91.0  10  26.25  0.0267  0.0275  0.1375  0.840  0.1657  10.  - In this tabulation, s o l u b i l i t y i s again expressed i n grams of ZnO per ml. Again i t i s to be emphasized that these viscosity values are only approximate. I f the " v i s c o s i t y method" of analysis i s applicable i n the case of beryllium, then i t would also be feasible f o r analysis of zinc i n view of the similar behavior. A given concentration of BeO produces f a r greater increase i n v i s c o s i t y than does the same concentration of ZnO, comparing f i g . 2 and f i g . 4. For a concentration of ©.06 gm per ml of each, we get a value of appx. 80 cp.for BeO i n isoamyl, and appx. 17 cp f or ZnO i s isoamyl.  c  c  0.i 0 ¥>  A-  .Nt~ ..  . i  | cr  ii  :  0 ' 0 ? 0 C *• «0 ID '" . IJ 0  - ™. , .  0 0  0  0  _  0  'N •  •• '00  a  V 1  V '• V  o  \  '  \  \ \ \  \ ' \\\ ' ..  \\ : . .. s  :  .• '  ses),  0  \ •' \ V .  \  a «  \  A  •-I  •  0  > «|m-/ml. Z n O -  !  >  ft  \"  '. '  '1 \  A  : -V V  \  <\  \ •:'  \ 'A  "•V.  \ \  V .  • ••• V  A  v• • \ \ -\ A•  0  6  0  <f  0  0  8  0 6  6  N  ' V:  SI  V  • IV'  1  Fve. 4-  c< 0 o  o  (19) (4) Magnesium Nitrates The Mg(NO )g behaved i n the same manner-as nitrates of Be and s  Zn. Analysis was by the oxine method; Effect of 50 ml of isoamyl saturated water upon the analysis i s given as follows: Table  10  MgO added,  KBrOg-KBr N/10  MgO found  D i f f . gm.  1.  0.0030  5.75  0.0029  - 0.0001  2.  0.0060  11.31  0.0057  - 0.0005  5.  0.0120'  23.85  0.0120  0.0000  4.  0.0240  47.22  0.0238  - 0.0002  5©  0.0361  71-.23.  0.0359  — 0,0002  6.  0.0481'  96.23  0.0485  4-0.0004  Differences observed were very slight and not consistently positive or consistently negative, so i n the work to follow, no correction was made f o r s o l u b i l i t y of Mg-oxinate i n isoamyl .alcohol. Table Efflux 20 deg  Vise, cp.  11  Vol. f o r analysis  KBrOg-KBr N/10  MgO  Soly. Uncorr,  Gorr; Vol.  Gorr. Soly.  1.  113  12.0  20  8.43  0.00424  0.0106  0.979  0.0108  2.  112  12.0  20  12.18  0.00614  0.0154  0.970  0.0159  3.  162  19.5  20  16.23  0.00817  0.0204  0.960  0.0213  4.  245"  31.0  20  22.70  0.01142  0.0285  0.944  0.0302  5.  294^  37.5  20  24.11  0.01215  0.0304  0.940  0.0324  6.  330  42.5  20  51.96  0.01609  0.0402  0.920  0.0437  7.  798  93.0  10  26.45  0.01534  0.0667  0.868  0.0769  8.  962  10  28.31  0,01428  0.0711  0.859  0.0829  107.0  .(1 ml KBrOg-KBr N/10 = 0.000504 g MgO) Density of anhydrous Mg-nitrate i s extrapolated as 1.98 g. per ml.,  8 0  o  8  01  6  0  MA.  W. N  £ 5-  aJ—^e-  —E0 0  •0 c  <J 0 v  t  X  d  o: v5  \  0-  o O  V \  A  .0  -JOB  O-OSgfe gcr>./crA. H q O  s  Tig.  S  (20) and this value i s used f o r "corrected volume" calculations. It w i l l be seen from f i g ; 5 that f o r a concentration of 0.06 gm; per ml of MgO, we get a v i s c o s i t y of appx. 65 cp; as compared with Values of 17 cp f o r ZnO and 80 cp f o r BeO. This phenomenon w i l l be discussed latere Summary of the viscosity^concentration curves f o r Be, Zn, and Mg appears i n f i g . 6. (5) Iron Nitrate; The s o l u b i l i t y of Fe(NOg)g i n isoamyl was found to be very small; a f t e r the c o l l o i d a l solution formed upon boiling was allowed to flocculate over a period of 24 hoursi After this length of time, solutions were f i l t e r e d through Whatman#l f i l t e r paper. F i l t r a t i o n before the suspended p a r t i c l e s settled out, resulted i n p r a c t i c a l l y the entire Fe-content passing through;the coarse f i l t e r . F i l t r a t e s "(10 ml) were extracted and Fe determined Fe(0E)g  by precipitating  with hexamethylene tetramine. ••• (CHg)gN 4- 10H20«Z-f6HeH0. 4-4NH 0H 4  4  Fe(N0 ) -i- 3NH 0H 5  5  4  > Fe(0H)g  4Fe = 12NH 0H=3(C%)gN 4  3NH N0g 4  4  A small excess of 10% aqueous hexamethylene tetramine solution was added i n each case to the solutions, volume.200 ml, o r i g i n a l l y s l i g h t l y acid and containing 20 gm of NH G1. The precipitated hydroxide i s f i l t e r e d 4  on Whatman#50 f i l t e r paper using suction* and ignited to FegOg. Some f i l t r a t e s (the entire extraction being taken f o r analysis) gave no test a t a l l f o r Fe, not even with KSCN reagent; however a small amount of Fe was always retained by the amyl alcohol' a f t e r extraction, leaving i t orange-red In colour. In the -tabulation to follow (table 12) eases where no.f Fe^Ojawas.. found s i g n i f y that no test occurred with KSSN*' •  -  - < V/ii  0 N  g  0  8  0  6 0  ....  . ;  . -...  l> o • ~  <L oL d. V  <^  *s  6> o" c O t N  o o o  u o <• r  UJ  8 >  \ASCOS\TV 0  V • V'  \.  x ; 'V ' N V  0  ( c e n t i po»se 0  \ \  \ \ •'\ \  .  0  \  \  _ \ S V  \ 0  \ \ A  \  \  •  • '\ • V  •  v  3\\ >'  \v VV.  \\  » I  %  w  I i J |  1  o  ID  0 " o  0  6  o  o. 0  n  0  \\  6 \*  I  1  »\ 1 \\  V  0  cr  o <j  g;  -0  •• j  o  (21) Table Vol. f i l t r a t e  12  Fe^Qg found  FegCy soly.  1.  10  nil  2.  10  nil  3.  10  0.0015  0.00015  4.  10  0.0007  0.00007  5.  10  0.0007  0.00007  6.  10  nil  It i s estimated, then, that the s o l u b i l i t y of Fe(N0g)g i n i s o amyl alcohol i s of the order of 1/10 of 1 milligram of Feg03 per ml. It would not be use this method f o r separating Be Mg and Zn from Fe, due to the uncertainty of the c o l l o i d a l nature of the Fe, and due: to-the length of time required.for i t to f l o c c u l a t e . (6) Chromium Nitrate; No actual values f o r s o l u b i l i t y i n the case-of Cr were found. Solutions of Cr(N0g)g i n isoamyl were prepared, howevery and the behavior was very similar to that i n the case of Fe(N0g)g. C o l l o i d a l suspensions resulted, probably composed of Chromium oxide, which would not flocculate on seven days standing; and after.this length of time, continued to pass completely through Whatman #1 f i l t e r paper. The method, therefore, would be of no use i n separation of Be, Mg, and Zn from Cr.  Investigation f o r Colloidal Properties by Centrifuging Due t o the obvious c o l l o i d a l nature of solutions of Fe and Cr, and A l i n the presence of Be, i t was strongly suspected that the solutions of nitrates of Be. Mg, and Zn were c o l l o i d a l . Iron solutions were centrifuged and i t was observed-qualitatively that s e t t l i n g o f , c o l l o i d a l Fe^Og was quite rapid. No d e f i n i t e evidence was obtained, however, f o r c o l l o i d a l  (22) tendencies by centrifuging i n cases of Be, Mg and Zn. Solutions of nitrates of these three elements i n isoamyl alcohol were prepared and standardized. Samples were.then eentrifuged at appx. 1500 r.p.m. f o r definite lengths of time, the top portions of eentrifuged solutions removed using:a pipettey and analyzed by the- methods previously described. The p o s s i b i l i t y exists that centrifuging at higher speeds would produce the evidence sought. . Results of centrifuging are. as follows: BeO Concentration Time of Centrifuging  6.0181 -0  0.0178 . 0.0179 30 min* 150 min.  MgO Concentration Time of Centrifuging  0.0264 0  0.0267 30 min.  0.0267 150 min.  ZnO-Concentration Time of Centrifuging  0.0208 0  0.0208 60 min.  0.0206 180 min.  —oOo—  CONCLUSIONS 1. The method involving the use of isoamyl or. normal amyl alcohols i s feasible as a means of separating aluminium.and beryllium, the aluminium nitrate being very sparingly soluble, and beryllium nitrate being seemingly i n f i n i t e l y soluble. The s o l u b i l i t y value of 0.045 gm, per ml of BeO as given by Browning: and Kuzurian was found to be a very conservative figure, t h i s concentration of BeO being  far.exceeded.  2. Solutions of Be(NO )g, Mg(N0g)g and Zn(N0g) i n isoamyl.(each nitrate g  2  being i n f i n i t e l y soluble i n the solvent), become increasingly viscous as the-concentration of solute increases-and-this property could probably be u t i l i z e d - i n a rough quantitative scheme* 3. I t was concluded that the.boiling points of secondary and.tertiary amyl alcohols were too low to ensure "dehydration" of the solute,  (23) (considerable amounts of Al(N0g)g being dissolved i f water i s present), and consequently these-isomers are not recommended* 4. F i l t r a t i o n must be carried out using a very fine f i l t e r i n cases of mixtures of A l and Be nitrates, due to the tendency of the Al(NOg)g to become c o l l o i d a l . Allowing the solution to stand f o r several hours obviates- this d i f f i c u l t y . 5. F i l t r a t i o n i s limited, although i n an a r b i t r a r y way,  by the viscosity of  the solution, i n cases of Be, Mg. and Zn; referring p a r t i c u l a r l y to the ease of Be j, i n separating this element, from A l . The l i m i t a t i o n i s a r b i t r a r y because i t depends upon the type of f i l t e r employed* The v i s c o s i t y l i m i t f o r laboratory f i l t r a t i o n i s 30 cp. On an i n d u s t r i a l scale, use of large suction f i l t e r s could accomodate media of higher v i s c o s i t y than 30 cp,, thus increasing the effectiveness of a given volume of amyl alcohol i n extracting Be(N0g)g from a mixture of the nitrates>of A l and  Be.  6. Upon b o i l i n g the nitrate-with the solventj the amyl alcohol takes on a brown color i n the cases of Be and- Zn$ suggesting- presence of- oxides of nitrogen, and this i n turn suggesting a certain degree of decomposition of the nitrate»: This phenomenon was not apparent i n the case of Mg. The claemical composition  of the solutes was not determinedi  and f o r  simplicity, throughout this work, they are considered as anhydrous nitrates. 7. The v i s c o s i t i e s observed f o r Be* Mg and Zn nitrate solutions seem to be some function of the molal concentration. For instance, from the curves we have f o r a concentration of 0.06  gm. per ml of the oxides* values of  80 cp f o r Be, 64 cp f o r Mg, and 18 cp f o r Zn. This may be tabulated as follows:  (24)  Oxide coney gm«/ml  Cone mols/liter  Viscosity  Be  0.06  2.40  80  Mg  0.06  1.50  64  Zn  0.06  0.75  18  This p e r i o d i c i t y of properties, although only very roughly indicated i n the above, would probably carry throughout the beryllium-aluminium group* (Group II)* In f i g . 6, crossing of the beryllium l i n e by the magnesium l i n e i s no doubt an error. Further investigation would probably reveal the family of curves to be quite regular. —oOo—  SUGGESTIONS FOR "TURTHER INVESTIGATION ' 1. S o l u b i l i t y of other compounds of these elements: e.g., sulphates, chloridesy etc, i n a m y l alcohol, or similar solvents* 2. S o l u b i l i t y of the nitrates i n higher alcohols of the a l i p h a t i c series. 3. Investigation of the viscosity-concentrationrelationship i n cases of Be, Mg, and A l , with a view towards employing i t as a rapid quantitative analysis method. In this method, the exact volume of the solution being-examined must be known before the v i s c o s i t y i s determined. Choices of methods and use of instruments i s l e f t to the discretion of the investigator. I t may be added,"however, that the simple method employed i n this research would not be s u f f i c i e n t l y r e l i a b l e f o r exact work. 4. Determination of the exact composition of the solutes (Be, Mg and Zn) when dissolved i n amyl alcohols, 5* Investigation of c o l l o i d a l tendencies by centrifuging.  (25) 6. Measurements of viscosity-concentration "relations f o r the remainder of Group I I , Be  I  /*< Ca  Zn  Sr  Gd  I  I  I  I  Ba I  Hg  //  •  7* S o l u b i l i t i e s of nitrates of.other metals of the Aluminium and Iron groups. —-oQo— Referencesj 1. Brovming and Kuzurian:: Orig. Com. 8th Internat. Gongr.- Appl. Chem. 1, 87-90, (1912) 2. Ivanov:: J . Russ. Phys. Chem. Soc. 46; 419-427j (1914) 5. Berg and Teitelbaumtt  Z. anal. Chem. 81j 1, (1930)  4. Hahn and Viewegss: Z. anal-. Chem. 71} 122^130, (1927) 5, Shuman and Berry: s Ind. Eng. Chem. (Anal.' Ed.) 9, 77, (1937) 6. Kolthoff and Sandell:: J. Am. Chem. Soc. 50} 1900f-(192S) 7, J i l e k and: Kotasi Z. anal. Ghem. 87j 422, (1932) 8. Nichols and Schempf:: Ind. Eng. Chem. (Anal. Ed.) 11, 278, (1939) 9, Ray and Chattopadhya% i Z. anorg. allgem. Chem. 169} 99-112, (1928) 10. A.I.Vogel: "Text-book of Quantitative Analysis", Longmans, (1939) 570 11. Ibid?: 537.  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items