Recommended Methods For Purification Solvent PDF
Recommended Methods For Purification Solvent PDF
Recommended Methods For Purification Solvent PDF
INTRODUCTION
The classical method for the production of methanol, the destructive distillation of wood,
has long ago been replaced by synthetic methods. These methods yield a quite pure product,
and the change has caused one of the more important impurities in the methanol employed in
the older literature, namely acetone, to be no longer a pressing problem. The commercial
grade A of methanol is nowadays sufficiently pure for practically all synthetic uses and for
its use as a solvent, except for very specific requirements. The specifications of manufac-
turers for their guaranteed reagents generally conform to those in the American Chemical
Society "Reagent Chemicals" (ref. 1), see Table 1.
Other specifications
The physical and thermodynamic properties of methanol have recently been critically reviewed
(ret. 2). Selected other properties are given
in a book dealing with organic solvents
(ref. 3). Values for selected properties of methanol taken from these and other sources are
listed in Table 2.
856
Purification of solvents and tests for impurities: methanol 857
Methanol, being a protic solvent but having also electron-pair donation capabilities , is an
amphoteric solvent , similar in many respects to water . It is capable of solvating both cat-
ions and anions , and although its dielectric constant (relative permittivity) is modest
(32.7 at 25 00), in comparison with water (78.5 at 25 00), it supports the ionic dissociation
of most electrolytes. On the other hand, it is well miscible with most organic solvents,
except the higher alkanes and perfluoroalkanes, and is a good solvent for organic compounds
in general. These properties have made methanol a very extensively utilized and studied
solvent.
The low freezing point of methanol (-97.6 C) permits the study of electrolytes by, e.g.,
n.m.r. techniques, at low temperatures, where line widths are much reduced and solvent ex-
change is considerably slowed down compared with room temperature. The relatively low normal
boiling point of methanol (64.7 00) can be turned into an advantage, when it permits the
ready removal of the sOlvent from temperature-sensitive reaction products. The low viscosity
of methanol is an advantage too.
The main feature of thestructure of liquid methanol is its association by hydrogen bonding.
Unlike water molecules, which can both donate and accept two hydrogen bonds and therefore
produce an extensive three- dimensional associated network, methanol molecules can donate one
and accept two hydrogen bonds. As a consequence, chain-like associates are formed in the
liquid. It is remarkable, though, that the angular correlation factor g has the value 0.99,
i.e ,nearly unity, which is characteristic of non-associated liquids (compare with g = 2.67
858 COMMISSION ON ELECTROANALYTICAL CHEMISTRY
:inwater and g = 3 . 01 in ethanol , all the values pertaining to 25 C) (ref . 12) . On the other
hand, the entropy of vaporization at the normal boiling point , S' 102 . 06 JK'rro1' , is
definitely higher than the expectation value fran Louton' s rule , 86 JX mo1 , and this
is in agreement with the associated nature of the liquid.
t
The infrared and n.m.r. spectra of methanol , too , indicate its associated nature , in that
few "free" OH groups, but many hydrogen-bonded ones, are seen (ref.. 13). The shifts produced
by solutes, whether polar aprotic solvents or anions, in the relative populations of these
classes of OH groups are typical for the interactions.
Methanol is a good solvent for polar organic substances. Nonpolar substances, as far as the
solubiity can be judged from the solubiity parameter o '[p(" RT)/M]*) = 29.3 J.
-3 /2 (ref. 12), should be dissolved in methanol to at least 1 mole % for substances which
have solubiity parameters above 18.6. The practical limit seems to be a little lower, say
16.5. This limit excludes the alkanes and perfluoroalkanes, which indeed are not well soluble
in methanol, but includes haloaJJzanes and arometic hydrocarbcns, which are generally well
soluble in, or completely miscible with, methanol.
Methanol is a good solvent for electrolytes. Its dielectric constant and solvating power for
ions are adequate for producing complete ionic dissociation of a dissolving electrolyte,
although some ion pairing nay occur. For 1:1 electrolytes ion pairing will occur (ref. 14)
when the sum of the radii of the olvated ions of opposite charge is less than 0.857 mm. The
radii of solvated ions in methanul are generally in the range 0.3 to 0.5 mm, so that ion
pairing should not be a rare occurrence.
The solvation power of methanol tcwards ions is sinilar to that of water. The standard Gibbs
free energy of transfer of monovalent ions from water to methanol is generally <15 kJ rnoL .
(ref. 15). The solubiity of sparingly soluble 1:1 electrloytes should therefore, be at most
400 times smaller than in water, and, for example, in the case of silver chloride the factor
is 53 (ref. 16). Electrolytes which are well soluble in water are generally also soluble in
methanol, with some noteworrthy exceptions, such as sodium fluoride, potassium chloride, and
the sulfates. The solubiities of some salts of the alkali metals are shown in Table 3.
Methanol is less acidic than water. In aqueous solutions at 25 C pK= 15.5530.016 (one
standard deviation) (ref. 19) (Ka in mol/dm3). A more recent but less precise estinate is
pKa = 15.20.3 (ref. 20). The autoprotolysis constant has been reported as pK = 16.70
(ref. 21 and 22), and a more recent determination has given pK5 = 16.91 (ref. 23) (Ks in
mel2 dm6). Since methanol is a typical ampholytic protic solvent it has a levelling effect
on the reactivities of acids and bases.
TABLE 3. Molal solubiities of some salts of the alkali metals, (mel/kg
solvent) at 25 C (ref. 17)
F Cl Br I NO3 003 SO,
In view of the fact that corrmercial methanol is now sufficiently pure for most applications,
the purification methods that have been utilized in the past for precise laboratory rk are
generally no longer required. These methods have been listed in ref. 3 and 24 and need to be
applied only under exceptional circumstances.
Still, one of the more cainnon impurities, water, is present in cainrrcial pure methanol to
an appreciable extent (ca. 500 ppm, see Table 1). This amount of
water is not expected to
have a profound effect on dilute solutions of solutes in methanol, in view of the hydroxylic
nature of this solvent itself and its low
molar volume, hence high concentration of hydroxyl
groups. There are 24.55 mol/dm3 of OH graips in methanol, and only 0.022 mol/dm3 of water
molecules (at the level of 500 ppm). For those applications where this impurity is still.
critical, following is a reccnnnended method of purification.
Step 2 . Pass the methanol slowly through a column of dry molecular sieves (Linde type LA),
store the methanol over a lump of calcium hydride. As long as the calcium hydride retains
its lump form, the water content is <7 ppm.
Step 3 . For a reduction of the water content to 0 . 5 ppm, the methanol is distilled twice in
.anall-glass still frc*n metallic sodium. Freshly cut pieces , 1-2 g per L methanol are used,
the first few tens of mL of distillate are discarded, and the rest, boiling within 0.01 00
of the boiling point , is used inunediately (ref . 26).
Caution: Anhydrous magnesium perchiorate is not suitable for the drying of methanol : danger
of EXPLOSION! Nor are drierite or calcium oxide effective drying agents for methanol . Calcium
metal is apt to contaminate the methanol with ammonia from its calcium nitride npurity.
Procedure II (conductivity grade) (ref. 27)
Step 1. Reflux 4.5 L methanol for 2th over 50 g magnesium. Distill 4 L and reflux these over
silver nitrate for 24h , under exclusion of atmospheric moisture and carbon dioxide.
Step 2 . Distill the methanol and shake it with activated alumina for 24h . Filter through a
sintered glass filter into a vessel which contains an atmosphere of pure nitrogen, from which
the methanol is distilled again in an all-glass system.
Specific conductances of 5x10 S are readily obtained with this method. With a conductivity
cell sealed between
specific conductance observed was 1 . 5x10 S cin .
the condenser and receiver in the last distillation step , the mininal
Procedure III (removal of bases and heavy metal$) (ref. 28 and 29)
Step 1 . Condition a dry macroreticular (e .g. Amberlyst 15) or macro-porous sulfonated-
polystyrene-type cation exchange resin in the hydrogen-ion form for 24h with methanol to pre-
swell it and eliminate the removable water.
Step 2 . Prepare a 1 : 2 mixture (by volume) of a strongly acidic cation exchange resin (e.g.,
Duolite C-20 , pretreated successively with 2 NHC1 1, . 5N NaOH , and 2N HC1) and a strongly
basic anion exchange resin (e.g., Duolite A-162, pretreated successively with 1. 5N NaOH,
2N HC1, and 1. 5N NaOH), wash with water, dry and preswell it in methanol for 24h.
Other impurities are either not obnoxious and can be tolerated at the levels at which they
are present in high grade commercial methanol, or, like carbon dioxide and other acidic im-
purities, are removed by procedures I or III. For the removal of other impurities, such as
acetone or formaldehyde, from technical grade methanol, consult ref. 3, 24, or 26. Purified
methanol should be prevented from being oxidized to formaldehyde by exclusion of air.
DETECFION OF IMPURITIES
Gas chromatography is the generally useful method for the determination of the residual impur-
ities in methanol. High sensitivity detectors, such as flame ionization, are requirod for the
determination of the impurities at the <10 ppm level. As a stationary phase, a mixture of 67%
polyethylene glycol 4000 and 33% diethyl sebacate, applied as 5% on silanized kieselguhr, has
been recommended; as the operating temperature 60 C (ref. 32). See also ref. 33 for a study
of several column packings for the gas chromatography of methanol, and ref. 34 for the analy-
sis of methanol by high performance liquid chromatography.
Ethanol can be detected in methanol down to a level of 3 ppm by gas chromatography (ref. 35).
Amines and other reactive impurities cai-i be detected by means of a copper(II)-ion-selective
electrode (ref. 37).
Water i generally determined by Karl-Fischer titration down to <500 ppm (ref. 32). For a
determination at the 70 ppm level, by measurements in the UV of the displacement of the keto-
ketal equilibrium after acidification with HBr, see ref. 36. For determinations by means of
physico-chemical measurements at the <10 ppm level (ebulliometry and critical solution tem-
perature of cyclohexane) see ref. 26.
INTRODUCTION
Ethanol is available commercially in quite pure form, either as the azeotrope with water
(containing ca. 5% water) or as so-called "absolute" ethanol, containing only 0.2% water. The
specifications of manufacturers for their purest grades generally conform to those in the
American Chemical Society "Reagent Chemicals" (ref. 1), see Table 1.
The physical and thermodynamic properties of ethanol have recently been reviewed (ref. 2).
Selected other properties are given in a book dealing with organic solvents (ref. 3) and in a
few other sources. Values for selected properties of ethanol taken from these sources are
listed in Table 2.
Other specifications
Boiling range, C 2 1
% mansmittanceh
at2lOnm 35 30
at220nm 55 52
at23Onm 72 73
at250nm 90 90
at >270 mm 98
860
Purification of solvents and tests for impurities: ethanol 861
there but specified for AnalaR grade reagent which also has a boiling range
of only 0.5C); eMn, Co, Ni, Cu, Zn, Cd, Pb, Hg (most stringent specifica-
tion for ARISTAR grade reagent: Hg 0.002 ppm, Pb 0.01 ppm, Cd 0.005 ppm);
fMg, Al, 1<, Ca, but for Na the linit is 0.5 ppm; Bracketing the normal
boiling temperature of 78.3C; light transmittance in 1 cm light-
path cells; aThe 30 ppm liniit applies specifically to isopropyl acohol;
the test for fusel oil is the absence of odor, after 10 cm3 ethanol has been
mixed with 5 cm3 water and 1 m3 glycerol, the mixture placed on a filter
paper and left to evaporate; 'The pink color should persist for >5 mm. when
20 cm3 are treated with 0.1 cm3 of 0.1 N KNnO at 15C.
Ethanol, being a protic solvent but having also electron-pair donating capabilities, is an
amphoteric solvent, and is capable of solvating both cations and anions. It has only a
modest dielectric constant (relative perrnittivity), 24.55 at 25 C, hence most electrolytes,
when not too dilute, will be extensively ion-pafred in it. Ethanol is miscible with most
organic solvents , including alkanes, and is a good solvent for' organic compounds in general.
Mixtures of ethanol with water are better solvents for electrolytes than neat ethanol, and
have been used extensively for the investigation of many properties of electrolytes in mixed
aqueous solvents.
The physical properties of ethanol, its properties as a solvent, and its ready availability
in pure form contribute to its general utility as a solvent. The vapors are non-toxic even
862 COMMISSION ON ELECTROANALYTICAL CHEMISTRY
at rrderate1y high concentrations, at which they are readily detectalle by their odor
(ref. 3) . The physiological effects of ingestion, of course, need not be described here, but
they place some legal restrictions on the free use of ethanol.
Ethanol is an associated liquid, and forms short chains by hydrogen bonding. Ethanol mole-
cules can donate one but accept two hydrogen bonds , hence the aggregates formed are
irregular. Evidence for the association arises fran many properties of liquid ethanol . For
example, the entropy of vaporization at the normal boiling temperature is
&SV 110.18 JKmol', much higher than the value given by Trouton' s rule for nonassocia-
ted liquids, "86 JIC' irol '. The angular correlation factor g has the value of 3.01 at 25C,
ccznpared with 2.67 for water (ref. 12), typical of highly associated liquids (nonassociated
ones have g 1).
The association of ethanol has also been studied spectroscopically, meinly by infrared
(ref. 38) and n.m.r. (ref. 39) methods. The relaxation time of the ethyl group has been
found to be shorter than that of the hydroxyl group, since the latter is hydrogen bonded to
other similar groups.
Many organic liquids are completely miscible with ethanol, including the alkanes on the one
hand and highly polar liquids such as - gyopl on the other. The solubility parameter of eth --
anol, calculated as [p(iHV - RT)/M} 26,1 J cm3 /2, is sufficiently near those of
nonpolar substances"(e.g., tin tetrachlride, 6 17.8 J an3/2 or iodine,
(S 28.8 J cur3 /2) to make it a good solvent for these.
The ability of ethanol to solvate both cations and anions is related to its polarity as mea-
sured by the donor number D A 30.4 kcal mo11 and the acceptor number AN 37.1, (see Table
2) which makes it a fairly good solvent for electrolytes. The standard Gibbs free energy of
the transfer of monovalent ions fran water to ethanol is 5 to 20 kJ mol' (ref. 15), so that
the solvation of the ions by ethanol is only to that extent worse than by water. Hcwever,
the solvated ions are apt to associate in ethanol to solvent-shared and contact ion pairs,
since the sum of the radii of the ions rarely exceeds the maximal distance of 1.14 nm for
two oppositely single charged ions to be counted as an ion pair (ref. 14), in this solvent
having c 24.55 at 25C.
These considerations are relevant to the solubility of electrolytes in ethanol. For electro-
lytes that have an appreciable solubility in water, the ionic interactions are too complica-
ted for the present state of the theory to prnnit a prediction of what the solubility should
be in ethanol. For electrolytes kf low solubility such predictions are more nearly feasible.
For example, silver chloride shi6uld be 139 times less soluble in ethanol than in water
(ref. 16) whereas the actual ratio is 149 (ref. 40). The solubilities of some salts in eth-
anol are given in Table 3.
Ethanol is less acidic than water, the equilibrium constants for the reactions
C2H5OH + H,O = C2H5OH2 + H2O and C2H5OH + 0H C2H5O + H20 being 6 x 10' and 63, respec-
tively (ref. 21). The acidity constant has been estimated as pKa = 16 for ethanol in aqueous
solutions (ref. 19). The autoprotolysis constant has been givers as p1<5 19.1 (ref. 21),
18.95 (ref. 22), 18.75 or 19.28(ref. 23) and 18.67 (K in rnol2 kg2) and 18.88
(K in mo12 dm-6) at 25t (ref. 42).
Purification ofsolvents and tests for impurities: ethanol 863
PURIFICATION OF ETHANOL
Ethanol is available ncwadays in highly pure form, see Table 1 , so that the methods that have
been used :in the past (ref. 3 , 24) for the purification of thj solvent for most laboratory
work are generally no longer required.
Of the impurities remaining in the best reagent grade ethanol are the other alcohols (metha-
nol, propanol) , acetone, and water. The last mentioned is the most reactive and likely to
interfere, and procedures for its removal have been recommended ( see below) . Benzene is an
impurity not nonilly encountered, but it may. be present in a sample dried by ternary azeo-
tropic distillation. It can be renoved frun dry ethanol by careful fractional distillation,
forming an azeotrope with a normal boiling temperature of 68 . 2C (ref. 2L) . Aldehydes , if
present, can be removed by the passage of the ethanol through a column of a strong base anion
exchange resin in the HSO3 form (ref. 3).
Procedure I (removal of organic impurities) (ref. 44)
Step 1. Add carefully concentrated sulfuric acid to the ethanol, and distill the ethanol.
The H, SO1 acts as an oxidizing agent for aldehydes, and this step also reduces the content
of amines.
Step 2. Redistill the ethanol from solid sodium hydroxide, to remove entrained sulfuric acid
and acidic impurities.
This combined treatment is said to reduce the content of aldehydes and ketones to <1 ppm, and
make the ethanol a solvent suitable for spectroscopic studies of solutes in the ultraviolet
region.
For the removal of water from so-called "absolute" ethanol, containing about 2000 ppm of
water, the following procedures are recommended.
Step 2. Treat a portion of 200 cm3 of the dehydrated ethanol with 3.5 g calcium hydride
crushed to a powder until all of it dissolves by gentle boiling. This treatment produces cal-
cium ethoxide. Distill two th-ixds of the alcohol off, to remove ammonia originating in
calcium nitride possibly present in the hydride.
Step 3. Transfer the remeining solution of calcium ethoxide in ethanol into 2 dm3 of dehy-
drated ethanol in a distilling apparatus provided with a double condenser. Boil the solution
gently for 20 h, allowing 100 cm3 to distill over in a slow current of dried pure hydrogen.
Discard this distillate.
Step 4. Distill the bulk of the ethanol slowly in the current of hydrogen, discarding the
last 1Q0cm3.
Step 2., Pass the dehydrated ethanol slowly through a coli,mn of dry molecular sieves (Linde
type 4A'). This step reduces the water content further (a sinilar treatment reduces the water
content of methanol to 7 ppm).
DETECTION OF IMPURITIES
Trace concentrations of heavy metals can be determined by flameless atomic absorption, using
a graphite furnace and a tantalum boat (ref. 46.), or by anodic stripping voltammetry in the
residue after evaporation (ref. 47).
Organic impurities are generally determined by gas chrartography, with a stationary phase
made up of 2 portions of polyethyleneglycol 1000 to one portion of bis(2-ethThexy1)sebacate,
applied 15% on silanized kieselgi.thr (ref. 47), or of polyethyleneglycol 300, applied 25% on
Chromosorb W (ref. 48). In the latter case the limit of determination is 5 ppm. Flame
864 COMMISSION ON ELECTROANALYTICAL CHEMISTRY
Water can be qualitatively detected dcn to 500 ppm by the formation of a precipitate when
aluminum ethoxide in benzene is added to a sample of the ethanol . Quantitative deteniiination
o water is carpied out by Karl Fischer titration. An autanated coulometric Karl Fischer
titration has been described (ref. 't9) that has a limit of detennination of 5 ppm. Some
corrmercial instruments have by nci an even better performance.
REFERENCES
1 . Arrrican Chemical Society, Reagent Chemicals , American Chemical Society , Washington D.C.,
5th ed., (1974).
2 . R.C. Wilhoit and B.J . Zwolinski , J. Phys . Chem. Ref. Data, 2, Suppl . 1 , 40 , 55 (1973).
3 . J.A. Riddick and W. B . Bunger , Organic Solvents, pp . 145-147 , 638-69 , Wiley-Interscience,
NewYork (1970).
G.W. Marks, J. Acoust. Soc. Amer., 42, 103 (1967).
5 . M .H. Abraham and A. Nasehzadeh , J. Chem. Soc. Faraday Trans. 1 , 7?, 321 (1981).
6 . C . Reichaixit , Solvent Effects in Organic Chemistry, Chemie , Weiriheim (1979).
7. E.M. KosoWer, J. Am. Chem. Soc. 80, 3253, (1958).
8. U. Mayer, V. Gutmann and W. Gerber, Monatsh. Chem., 106, 1235 (1975).
9. V. Gutmann, Coordination Chemistry in Nonaqueous Solutions, Springer, Vienna (1968).
10. M.J. IKanilet, J.L. Aboud and R.W. Taft, J. Am. Chem. Soc., .99, 6027 (1977).
11. R.W. Taft and M.J. IKamlet, J. Am. Chem. Soc., 98, 2886 (1976).
12. Y. Marcus, Introduction to Liquid State Chemistry, Wiley, Chichester (1977).
13 . M. C. R. Symons and V .K. Thomas , J. Chem. Soc. Faraday Trans . 1 , 77, 1891 (1981).
1L. N. Bjerrum, Kgl. Danske Videnskab. Selskab., 7, 9 (1926).
15. Y. Marcus, Pure Appl. Chem., 55, 977 (1983).
16. Y. Marcus, Rev. Anal. Chem., 5, 53 (1980).
17. C.M. Criss, in Physical Chemistry of Organic Solvent Systems, A.K. Covington and
T. Dickinson , eds . , 78-79 , Plenum, London (1973).
18. E. Price, in The Chemistry of Nonaqueous Solvents, J .J. Lagcwski, ed., 1, 70, Academic
Press (1966).
19. P. Ballinger and F.A. Long, J. Am. Chem. Soc., 82, 795 (1960).
20. E.M. Wooley, J. Tomkins and L.G. Hepler, J. Soln. Chem., 1, 341 (1972).
21. I.M. Koihoff and S. Bruckenstein, in Treatise of Analytical Chemistry, I.M. Koithoff and
P.J. Elving, eds., I, 1, 484 and 523, Interscience, New York (1959).
22. N .A. Izmailov, Elektrokhimiya Rastvorov, lzd. Khimiya, Moscow (1966).
23. N.T. Smolova, Zh. Fiz. Khim., 52, 2905 (1978); Russ. J. Phys. Chem., 52, 1668 (1978).
24. D.D. Perin, W.L.F. Armarego and D.R. Perrin, Purification of Laboratory Chemicals, 2nd
ed., 249251, 320321, Perganon, Oxford (1980).
J
25. J. Inpton and .A. Riddick, Commercial Solvents Corp., Unpublished work, quoted in Ref. 3.
26. L. Gillo, Ann. Chim., 12, 281 (1939); C.A. 34, 2221 (1940).
27. E.C. Evers and A.G. Knox, J. Am. Chem. Soc., 73, 1739 (1951).
28. B.M. Collins, D. Kitchen and T.D. Rees, Chem. md. (London), 1972, 173.
29. Sumimoto Chemical Co. Ltd., Japan. Patent 82 77634; C.A. 97, 127 029x (1982).
30. M. Griesbach and K.H. Lieser, Fresenius Z. Anal. Chem., 302, 181 (1980).
31. M. Foerster and K.H. Lieser, Fresenius Z. Anal. Chem., 309, 352, 355 (1981).
32. Anon., Merck Standards, Merck, Darmstadt, 1971.
33. G. Mohan and S.K. Ghosh, Pert. Technol., 16, 27 (1979); C.A. 92, 190 790c (1980).
34. N.E. Spingem, C.T. Garvie-Gould and L.L. Vuolo, Anal. Chem., 53, 565 (1981).
35. E.T. Shefelev and T.A. Mindrina, Zh. Anal. Chem., 33, 684 (1978).
.36. J. Toullec and N. El-Alaoui, Anal. Chim. Acta,, 109, 187 (1979).
37. B.K. Desmukh and J.F. Coetzee, Anal. Chem., 56, 2373 (1984).
38. F.A. Smith, E.C. Creitz, J. Res. Nat. Bur. Standards, 46, 145 (1951).
39. H. Versmold, Ber. Bunsenges. Phys. Chem., 78, 1318 (1974).
40. R. Alexander, A.J. Parker, J .H. Sharp, and W.E. Waghome, J. Am. Chem. Soc., 94, 1148
(1972).
41. H. Stephen and I. Stephen, Solubilities of Inorganic and Organic Compounds, Pergamon,
Oxford, Part 1 (1963).
42. R. Schaal and A. Thze, Bull. Soc. Chim. France, 1961, 1783.
43. J. Uher, J. Pelchova, V. Krumphanzl, Sb. Vz,'s. Sk. Chem. -Technol. Praze Potravin.zi, E43,
113 (1975); C.A., 84, 41942 m.
44. B. Glabik and H. Sobczyk, Chem. Stosow, 18, 187 (1974); C.A., 81, 145492 m.
45. C.N. Riiber, Z. Elektrochern;, 29, 334 (1923).
46. H.-M. Huang, C.-C. Wu, M.-C. Chou and C.-C. Liu, Pen Hsi Hua Hsueh, 8, 233 (1980); C.A.
94, 166, 952 c.
47. Anon., Merck Standards, Merck, Daxmstadt, 1971.
48. A. S. Lavrenova, I. P. Ogloblina, V .A. Fedorov, N.P. Kuznetsova, Metody Anal. Khim.
Reaktivov Prep., 18, 39 (1971); C.A. 77, 147 319 j.
49. E.A. Bondarevskaya, E.D. Knopotova, T.E. Mashko, Zh. Analit. Khim., 30, 560 (1975).