Ames Laboratory ISC Technical Reports Ames Laboratory

3-1955

Titrimetric determination of zirconium

M. O. Fulda
Iowa State College

J. S. Fritz
Iowa State College

Follow this and additional works at: http://lib.dr.iastate.edu/ameslab_iscreports

Part of the Chemistry Commons

Recommended Citation
Fulda, M. O. and Fritz, J. S., "Titrimetric determination of zirconium" (1955). Ames Laboratory ISC Technical Reports. 95.
http://lib.dr.iastate.edu/ameslab_iscreports/95

This Report is brought to you for free and open access by the Ames Laboratory at Iowa State University Digital Repository. It has been accepted for
inclusion in Ames Laboratory ISC Technical Reports by an authorized administrator of Iowa State University Digital Repository. For more
information, please contact digirep@iastate.edu.
Titrimetric determination of zirconium
Abstract
A brief literature survey on the present gravimetric, colorimetric, spectrographic, and volumetric methods for
the determination of zirconium has been presented.

Keywords
Ames Laboratory

Disciplines
Chemistry

This report is available at Iowa State University Digital Repository: http://lib.dr.iastate.edu/ameslab_iscreports/95

 u••n• ·ASS IFlED
IoW()...
sc.
AL
ISC-
S91P UNITED STATES ATOMIC ENERGY COMMISSION

TITRIMETRIC DETERMINATION OF ZIRCONIUM

By
M. 0. Fulda
J. S. Fritz

March 1955

Ames Laboratory
Iowa State College
Ames, Iowa

Te c h n i c a I I n form at i o n S e r v i c e, 0 a k R i d g e, Te nne 55 e e

UNCLASSIFIED
 Work perf'onned under Contract No. W-7405-Eng-82.

F. H. Spedding, Director, Ames Laboratory

The Atomic Energy Commission makes no representation or warranty

as to the accuracy or usefulness of the Information or statements contained
In this report, or that the use of any lnfonnatlon, apparatus, method or
process disclosed In this report may not Infringe prlvately~wned rights.
The Commission assumes no llablltty with respect to the use of, or for
damages resulting from the use of, any lnfonnatlon, apparatus, method or
process disclosed in this report.

This report bas been reproduced directly fran the best

available copy.

Reproduction of this information is encouraged by the

United States Atcmic Energy Commission. Arrangements for
your republication of this document in whole or in part
should be made with the author and the organization he
represents.

Printed in USA, Price 25 cents. Available from the

Office of Technical Services, Department of Commerce, Wash-
ington 25, D. C.
GPO 36~589
 ISC-596 1

TITIUMETRIC D:b'TERMINATION OF ZIRCONIUM*

by M. 0. Fulda and J. S. Fritz

ABSTRACT

A brief literature survey on the present gravimetric, colorimetric,

spectrographic, and volumetric methdds for the determination of zirconium
has been pr~sented.

A new method for the quantitative determination of zirconium has

been proposed. The method is titrimetric and is carried out in acid
solution with EDTA (disodium dihydrogen ethylenediaffiinetetraacetate),
a stable, soluble complex being formed. A visual indicator gives a
sharp end point. Thorium, molybdenum, bismuth, antimony, and tungsten
constitute important interferences, but most other metals do not affect
the results. Tin and titanium reduce the sharpness of the end point
but do not otherwise interfere. Iron must be complete~ reduced to
the ferrous state before titration. Copper interference is avoided by
complexing it with thiourea. Anion interferences were also studied.
This method constitutes an approach to the determination of zirconium
in more difficult samples.

In order to check the applicability of the proposed method in a

practical manner, several samples of zircon sands arid zirconium-zinc
alloys were analyzed for zirconium by this method and the results
compared with the results of analyses by the method of Kumins (30),
using the precipitation of zirconium mandelate with mandelic acid and
ignition to zirconium oxide. The results of the proposed titrimetric
method agree very well with the results of the gravimetric method.

I. INTROIDCTION

The mineral zircon was discovered by Werner who found it to be a

mineral separate from diamond (36). He named it from the Latl.n,
silex circonius, or the Arabian, zerk, a precious stone. He was not
aware, however, that another element besides iron, aluminum, silicon
and oxygen was present. The German chemist, Klaproth, discovered
zirconia, Zr02, while analyzing a zircon in 1789 (62).

* This report is based on a M.S. thesis by M. 0. FUlda, submitted

.March, 19SS at Iowa State College, Ames, Iowa. This work was performed
under contract with the Atomic Energy Commission.
2 ISC-596

The metal was first prepared by Berzelius in 1824 (37). Now the
Kroll process is used. In the Kroll process zircon ore is heated
with graphite in an electric arc to produce zirconium carbide and
drive off the silicon. The zirconium carbide is treated with chlorine
gas to form zirconium tetrachloride and this is reduced to spongy
zirconium with molten magnesium in an inert atmosphere. The magnesium
and magnesium chloride are distilled off under high vacuum (68).

The possibilities of zirconium as a structural material are

interesting to metallurgists and chemical engineers. Its ability to
withstand high temperatures and stresses is of importance in the field
of jet engines and in its resistance to corrosion it may rival
tantalum. lJVhen freed from hafnium, it has a low absorbing power for
neutrons; this additional observation is important in the design of
nuclear reactors and explains the increased interest in the separation
of these two elements (32). Also, zirconium does not amalgamate with
mercury, a property which is made use of in mercury vapor applicat{ons.
Zirconium alloys are used in armor plate steels, deoxidizers and
scavengers, and where resistance to heat and corrosion is needed.
Zirconia, zirconium oxide, is used in refractories. It is resistant
to the action of heat, slags, and most acids (37).

Zirconium, therefore, is ir.creasing in importance and more rapid

and selective methods of control analysis are needed. Most of the
methods now available are either timeconsuming or not selective
enough. The majority of the methods are gravimetric for macro amounts
and colorimetric for micro or semimicro amounts, while comparatively
few volumetric methods have been offered. The volumetric methods are
usual~ not very selective and are indirect, requiring a preliminary
separation of a precipitate.

This work was undertaken to develop a rapid direct titrimetric

method for zirconium which would also be selective. The titration is
carried out in acid solution with EDTA (disodium dihydrogen ethylene-
diaminetetraacetate); a visual indicator gives a sharp end point.
First a discussion of the titration, the factors affecting the
titration, and the treatment of interferences, is given in this
report. Then some practical analyses of samples containing zirconium
are given.
 3

II. REVII!.'W' OF LITER.A'IURE

A. Other Methods

Some excellent reviews of the ana~tical chemistry of zirconium

have been published (2a; 13; 61; 65, p.321). The report of t~e
Technical Information Service of the Atomic Energy Commission
(TID-3010) is especially useful. It covers the entire field of
zirconium chemistry through 1950 (61).

1. Gravimetric
A large number of gravimetric procedures have been proposed for
macro amounts of zirconium. The classical method is a precipitation
as ZrOHP04, zirconyl hydrogen phosphate, from 10 per cent sulfuric
acid solution. This separates zirconium from all metals but quadri-
valent cerium. The precipitate cannot be ignited to a definite
composition, however, and has to be fused with sodium carbonate and
the hydrous oxide precipitated with ammonia from hydrochloric acid
solution (65, p.J21). For small amounts the phosphate precipitate
can be ignited and weighed as ZrP207 (4; 65, p. 321). Willard
determined larger amounts by precipitation from homogeneous solution
by heating with triethyl phosphate ( 66).

Most of the recent- developments are in the use of organic pre-

cipitants and direct ignition of the precipitate to the oxide. The
methods using mandelic acid (18, 23, 30, 38) and para bromo or para
chloromandelic acid (19, 26, 44, 45) are done in this manner, the only
interference being hafnium. The precipitate from the halogen deriva-
tives can be washed with water; the precipitate from the mandelic acid
must be washed with a 2 per cent HCl-5 per cent mandelic acid solution.
Tetrachlorophthalic acid is as specific, but much less reagent is
required (41). Other organic precipitants are ammonium benzoate (22),
benzilic acid (27, 57), tannin (46), benzoic acid (47), paradimethyl-
aminoazophenylarsonic acid (51), fumaric acid (55), metani trobenzoic
acid (58), cinnamic acid (59), hydrazine sulfate (60), rnetacresoxy-
acetic acid (56), and cupferron (2a; 65, p.321). A thermogravimetric
ana~sis using the precipitate with 8-hydroxyquinoline was unsuccess-
ful (3).

Another inorganic precipitant in addition to phosphate is

selenious acid. The precipitate, Zr(Se03)2, is ignited to the
oxide (13). Zirconium can also be separated as BaZrF6 using barium
nitrate-hydrofluoric acid solution as the precipitant (2a; 21, p.l499).
4 ISC-596

2. Colorimetric

For micro amounts of zirconium colorimetric methods of analysis

are used. Being used for a different range of concentrations than the
method proposed in this report, these methods do not compete; therefore
it will be sufficient to merely mention a few of them.

Most of· the authors of colorimetric measurene nts of zirconium

use Alizarin RedS as the color forming reagent (15, 17, 63). Other
reagents used are quercetin (16), thorin (22), hematoxylin (55),
m-(2-hydroxynaphthylazo) mandelic acid (43), and chloranilic acid (54).
The use of fluorescence analysis using flavanol (la) and morin (67) and
a method applying flotation analysis to fluorescence analysis (42) have
been mentioned in the literature. Kiefer and Boltz (24) determined the
excess phosphate from a precipitation of zirconium as the phosphate
spectrophotometrically as the molybdiphosphoric acid complex.

3. Spectrographic
Spectrographic methods are used for extremely small amounts of
zirconium (22), for the determination of zirconium in hafnium (25,
34, 40), and for the analysis of very difficultly soluble alloys such
as zirconium-niobium.

4. Volumetric
The majorit,y of the methods for the volumetric determination of
zirconium which are found in the literature are indirect methods in-
volving precipitation and a determination of the excess precipitant
or dissolution of the precipitate in an excess of a standard solution
and a determination of the excess. The reason for this is that
zirconium has a constant oxidation state and forms a colorless ion in
solution. These indirect methods involve so many operations that it
might be more practicable to ignite the precipitate to the oxide and
weigh it as such.

White (64) developed many indirect methods. In one a zirconia

precipitate is dissolved in potassium fluoride solution and the
basicity titrated. In another zirconium is precipitated with
p-chloromandelic acid. The precipitate is dissolved in excess standard
sodium methylate in methyl alcohol-benzene solution, and the excess
sodium methylate titrated with standard acetic acid with thymol blue
as the indicator. Another method of White is the dissolution of the
zirconia precipitate in excess standard acid and the titration of the
excess acid.
 ISC-)96 5

Sawaya and Yamashita (48) determined zirconium by dissolving the

hydrous oxide precipitate in potassium fluoride and standard 0.1 N
nitric acid and titrating the excess standard acid. The reaction for
dissolution of the hydrous oxide is

Zr(OH)4 ,.. 6 KF t- 4 HNo3~ K2ZrF6 t- 4 KN0 3 r 4 H2o.

Dahr and Das Gupta (9) determined zirconium volumetrically after

isolation of zirconium oxalohydroxamate. The precipitate is dissolved
in dilute hydrochloric acid and hydrolyzed to hydroxylamine. .lxcess
standard titanous chloride is added and back titrated with Fe(III)
solution.

Graham, et al. (14), developed a polarographic method for zirconium

in magnesium base-alloys based on the polarographic reducibility of
m~itrobenzoic acid. Again the precipitate must be isolated and
redissolved.

A few direct titrations have been reported for zirconium.

Kalthoff and Johnson (28) did a direct amperometric titration of
solutions containing no interfering ions with a mean error of t- 5.3
per cent and a deviation of ,.. 0.8 per cent. Larsen and Grammil
(29, p.)69) did electrometric titrations in their calculation of the
solubility prpduct of zirconium hydroxide. The usual strong acid-
base curves were obtained, but the end point was calculated from an
extrapolation because precipitation and coagulation obscured the
inflection point. Cozzi (7) determined zirconium by reduction at the
dropping mercury electrode. Although these types of titration are
direct, they are usually not very rapid.

Recently EDTA, ethylenediaminetetraacetic acid, has been used by

Milner and Phennah (39) to titrate zirconium in binary alloys with
uranium after dissolving the mandelic acid precipitate in perchloric
acid. Excess EDTA is added to tne sample in acid solution and back
titrated with Fe(III). Salicylie acid is the indicator which is
used. The accuracy claimed for 100 milligrams of zirconium is 1 per
cent.

From this review it can be seen that a direct and rapid method
for the determination in the presence of interfering cations is
needed. This is the work which this report reports.

Ethylenediaminetetraacetic acid, EDTA, was first reported b;}'

Fick and Ulrich (ll) in 1936. They prepared it by reacting ethylene-
diamine, hydrogen cyanide, and formaldenyde, and saponifying the
6 ISC-596

product to EDTA. As shown by Schwarzenbach and Ackermann (50) i t is

essentially a dicarboxylic acid because the first two pK's are so
close (pKl ::: 1.996 , pK 2 = '2..672, pK3 = =
6.161, pKLl 10.262). This
compound has the ability to form very stable, water soluble complexes
with many metal ions. The log of the formation constant for the
reaction

where M•n represents the metal ion, y-4 the ~ffi'A ion, and MY(n-4) the
complex, was reported for many ions by Cabell (6) and compared with
thqse values for other complexing agents. The author was unable to
find a value for the log K for the zirconiwn complex with £lJI'A in the
literature.

The reagent is not specific; however, with the proper choice of

conditions and indicators, the reagent can be made specific.

EJJI'A has found application in many other fields besides analyti-

cal chemistr,r. Two extensive bibliographies on the uses of EDTA are
available. One is entitled, 11Sequestrene, 11 and is published by Al;rose
Chemical Co. (lb). It covers the literature to June, 1952. The other
is entitled, "Bibliography of Ethylenediaminetetraacetic Acid," and is
published by IJersenes Incorporated, formerly Bersworth Chemical Co. (2b).

III. EDTA TITRATION OF ZIROONIUM

A. Factors Affecting the Titration

l. Indicators
The general reactions involved in the titration are

Zr(IV) .- H2Y-2 =ZrY +- 2H .. (1)

Zrina ~ H2Y-2 .- (ab - 2)H+- : ZrY +- aHbln, (2)

where Y and In represent the EDTA and indicator radicals, respectively.

After the bulk of the zirconium has reacted according to reaction 1,
the highly colored zirconium-indicator complex is destroyed
(reaction 2), marking the end point. In addition to being highly
colored, the zirconium-indicator complex must be less stable than the
 ISC-596 7

~irconium-EDTA complex. Furthermore, reaction 2 must be fairly rapid,

so that the end point will not be overrun.

Several indicators were tried. Alizarin Red S has been used for
the colorimetric determination of zirconium (63) but is useless for
the EDTA titration because of the very slow reaction of the zirconium-
indicator complex with EDTA. Carminic acid, chloranilic acid, and
Chrome Azurol S also are unsatisfactory. Both Alizarol Cyanone RC and
Eriochrome Cyanine RC give sharp and vivid end points. The Alizarol
Cyanone RC end point is often slow if a direct EDTA titration is
attempted, but it is fast if a slight excess of EDTA is added and this
excess back titrated hot with standard zirconium. Either direct or
back titration is possible with Eriochrome Cyanine RC, but the end
point in the direct titration should not be approached too rapidly.

2. Effect of pH

With either Alizarol Cyanone RC or Eriochrome Cyanine RC indi-

cator, satisfactory end points can be obtained in the pH range 1.0 to
2.0. Best accuracy is obtained, however, if the final pH is between
1.3 and 1.5 Results for zirconium are high above pH 1.5 and low
below pH 1.3. .

newell (8), using ion exchange studies, found that zirconium ions
in solution chan.ged to complex ions and formed colloids above pH 2.0
according to the reaction

This would explain why no end point was observed by the author above
pH 2.0. In EDTA solution zirconium begins to precipitate at pH 3.0;
this was found by Cabell (6) and confirmed by the author. .

3. Effect of heat

Heating the solution to 70° to 90°C. speeds the reaction and

permits a rapid back titration of excess EDTA with zirconium. A
direct titration in hot solution is not possible. When the solution
is heated a:t pH to 1.3 to 1.5 without excess EDTA, no end point is
observed even with 100 per cent excess EDTA. When the solution is
heated for 15 minutes without excess EDTA, a jel+y-like substance,
possibly the nydrous oxide of zirconium, precipitates and absorbs
8 ISC-596

the indicator. Perhaps the forna tion of complex ions and colloids
which Dewell (8) described as happening above pH 2.0 occurs at a
lower pH at higher temperatures.

4. Extraction attempts

Fritz and Ford (12) extracted thorium prior to a titration vnth

EDTA according to the method of Levine and Grimaldi (33). In this
extraction mesityl oxide is the organic solvent and aluminum nitrate
is the salting out agent. This separated thorium from the rare
earths and large amounts of fluoride and phosphate. A separation such
as this for zirconium would be very useful because fluoride and
phosphate are important interferences in the tiDTA titration (see
Interference Study). Many alloys if zirconium are soluble only in
hydrofluoric acid, and the phosphate precipitation is, as mentioned
earlier, a very good separation scheme for zirconium.

Several combinations of mesityl oxide and salting out agents

were tried. Aluminum nitrate, zinc nitrate, calcium nitrate, lithium
nitrate, and magnesium nitrate were tried; but no quantit·ative
separations were obtained. Scadden and Ballou (49) used a mixture of
dibutyl ether, dibutylphosphoric acid, and monobutylphosphoric acid
and on~ extracted 98 per cent of the zirconium present. This was the
most quantit~tive extraction that could be found in the literature.

5. Duffer attempts

It was mentioned earlier that the best pH to use is 1.3 to 1.5.

The pH change during the titration is very slight for titrations of
10 milliliters or less; however, to ~aintain such a close pH tolerance
it would seem advisable to use a buffer system. Sodi. urn periodate-
=
periodic acid (pKa 1.64), sodium oxalate-oxalic acid (pKa : 1.19),
sodium saccharide-saccharin (pKa = 1.60), and sodium dichloroacetate-
dichloroacetic acid (pKa = 1.30) buffer systems were tried, but all of
them interfered in the. titration. No suitable buffer was found.

D. Reagents and Apparatus

1. Indicators

a. Alizarol Cyanone RC. A 0.4 per cent aqueous solution of

this eye, obtained from National Aniline Division, Allied Chemical
and Dye Corp., was used as an indicator.
 ISC-.596 9

b. Eriochrome Cyanine RC. A 0.4 per cent aqueous ~olution of

this dye, obtained from Geigy Co., was used as an indicator.

c. Eriochrome Black T. A l per cent solution of this qye,

obtained from Hach Chemical Co., in triethanolamine (10) was used as
an indicator.

2. Standard solutions

a. EDTA. Reagent grade disodium dihydrogen ethylenediamine-

tetraacetrc-icid, available commercially from Hach Chemical Co., was
used for this solution. To prepare a 0.0.5 M solution 70 grams of the
salt were dissolved in 3.5 liters of water. This solution was
standardized against pure zinc metal as described by Fritz and Ford
(12) and Brown and Hayes (5) or against the 0.0.5 M standard ~irconium
solution.

b. Hafnium Chloride. A 0.05 M solution in (5 ~ 9.5 ) hydrochloric

acid was prepared by dissolving 11.4 grams of HfO(OH)2 in 50 milli-
liters of concentrated hydrochloric acid and diluting to one liter.
This solution was standardized by evaporating an aliquot and igniting
the residue to the oxide. The zirconium content was 2.1 per cent by
weight as determined by spectrographic analysis.

c. Thorium Nitrate. A 0. 05 M solution in 0.02 to 0.03 M

nitric acid was prepared and standardized as described 'by Fritz and
Ford (12). The standardization is done gra~~metrically by precipita-
tion of thorium oxalate and ignition to thoria.

d. Zirconyl Chloride. Zirconyl chloride octahydrate containing

less than 100 parts per million hafnium was used for t his solution.
A 0.05 M solution in (5 t- 9.5) hydrochloric acid was prP-pared by dts-
sol ving 16 grams -::£ the reagent in 50 milliliters of concentrated
hydrochloric acid and diluting to one iiter.

3~ Miscellaneous

All other chemicals used in this work were reagent grade unless
otherwise specified. The zinc amalgam was 20 mesh zinc lightly
amalgamated as directed qy Kolthoff and Sandell (29, p. 569).
10 ISC-S96

4. Apparatus
All pH measurements were made with a Beckman Model G pH meter
using a glass-calomel electrode system. Magnetic stirrers uere used
for all of the titrations. Other apparatus were normal laboratory
equipment.

C. Procedures

1. Procedure for direct titration

The zirconium concentration of the solution before the titration

was O.lS to 0.2S millimoles of zirconium per So milliliters of solu-
tion. The pH was adjusted to 1.4 with dilute ammoniurn hydroxide (1 ~ 9)
or dilute perchloric acid (1 ~ 9), two drops of the indicator solution
were added, and ~he solution was titrated with O.OS M EDTA to the dis-
appearance of tbe pink oolor of the zirconium-indicator complex.

2. Procedure for direct titration in the presence of iron

If iron was present, 10 grams of amalgamated zinc was added to

SO oillili ters of the solution which contained O.lS to 0. 2S millimoles
of z irconi urn and up to 1. 0 millimole of iron. The pH had to be less
than 1.0 in order to insure complete reduction of the iron. The
mixture was stirred about five minutes, then the pH was adjusted to
1.2 to 1.4 with dilute a~~6nium hydroxide (1 ~ 9). Two drops of the
indicator solution were added, and the solution was titrated with
O.OS M EDTA to the disappearance of the pink color of the zirconium-
indicator complex while bubbling nitrogen through the solution by
means of a gas washing tube. Magnetic stirring was used.

3. Procedure for back titration

The same procedure was used as for the direct titration, and a
slight excess of EDTA was added. When necessary, the pH of the solu-
tion was readjusted to 1.4 with dilute ammonium hydroxide (1 ~ 9).
The solution was heated almost to boiling and back titrated with O.OS M
zirconyl chloride to the first permanent pink color.
 ISC-596 11

D. Stoichiometry

Since aqueous solutions of zirconium polymerize above pH 2,

there was some doubt as to whether the titration with EDTA could be
made stoichiometric. A test for stoichiometry was made by titrating
aliquots of a zirconyl chloride solution which had been standardized
by evaporation and ignition to the oxide and by mandelic acid precipi-
tation and ignition to the oxide. The 3DTA solution used to titrate
the zirconium was standardized against pure zinc metal by methods
reported in the literature (5, 12). Stoicniometric results w·ere
obtained only when a correction was applied for the approximately 2
per cent hafnyl chloride present in the reagent grade zirconyl chloride.
This assumes that the hafnium was titrated with the zirconium. The
hafnium content of the solution was determined independently b,y the
spectrographic laboratory ~nd reported as the Hf/Zr ratio. The
calculation of the metal ion molarity is

M = Hf/Zr x~ Zr/Hf X~
Hf/Zr ~ 1 221.4 Zr/Hf ~ 1 113.1
v

Hf/Zr =spectrographic ratio

g =weight in grams of combined oxides
from evaporation and ignition

V =volume in liters of aliquot evaporated

M = metal ion molarity of solution

To check the stoichiometry further a standard zirconyl chloride

solution was prepared from zirconyl chloride containing less than
100 parts per million hafnyl chloride, and this solution was titrated
with two different standard ED'l'A solutions. The results given in
Table 1 show that at least within ordinary titration error the method
is stoichiometric, a one to one zirconium-EDTA complex being formed.
12 ISC-596

Table 1. Direct Titration of Zirconium

Zr taken Theoretical Em'A Experimental EDTA Difference

(mg) (ml) (ml) (rnl)

22.8 5.24 5.24 0.00

22.8 5.24 5.24 0.00
22.8 5.24 5.25 t-0. 01

22.8 9.47 9.47 0.00

22.8 9.47 9.48 ,.0. 01
22.8 9.47 9.49 ~o.o2

Some hafnium usually accompanies zirconium and will be titrated

as zirconium unless the hafnium-zirconium ratio is determined and
taken into account in the calculations as indicated. If it is not
convenient to determine this ratio, it can usually be assumed to be
constant in a series of samples. The ~A can then be standardized
against a sample containing a known weight of zirconium plus hafnium,
the concentration of the EDTA solution being expressed, as the zirconium-
hafnium titer.

E. Interference Study

Table 2 lists the interfering and non-interfering ions. The fact

that ions such as copper(II), nickel(II), and lead(II) do not inter-
fere although their EDTA complexes are very stable (log K formation
is approximately 18 (6))indicates that both the zirconium-EDTA and
zirconium-indicator complexes are very strong.

The direct titration procedure must be followed if aluminum is

present. This is because once the zi.rconiu.m-indicator complex has been
broken by adding excess EDTA, the indicator is free to react with the
aluminum. The direct titration succeeds because the aluminum reacts
with the indicator too slowly to interfere, especially at room
temperature.
 ISC-596 13

Table 2. Interference Study

Direct titration Back titration

EDTA EDI'A EUTA EDTA
Theoretical Experimental Theoretical Experimental
Ion addeda (ml) (ml) (ml) (ml)

Ag'" 5.24 interferes 5.21 5.25

Al+-+-+- 5.24 5. 24 5.21 interferes
Ba+-+- 5.24 5.~4 5.21 5.20
Be+-+- 5.24 5.23 5.21 5.25
Bit-+-+- 5.24 interferes 5.21 interferes
ca+-+- 5.24 5.24 5.21 5.20

Cd'"'" 5. 24 5. 22 5.21 5.21

cet-+-+- 5.24 5.24 3.50 3.51
Cot-+- 5.24 5.26 5.21 5.20
cr+-+-+- 5.24 5. 22 3.50 3.50
eu ... +- 5.24 interferes 4.50 4.51b
Cut-+- 4.47 4.46b 3.50 3.51
Fe+-+- 5.33 5.34 5.21 interferes
Fe4-+-+- 5.24 interferes 5.21 interferes

H~'"'" 5.24 .5 .24 4.17 4.18

K 5.24 5.22 4.50 4.50
La+-+-+- 5. 24 5 . 24 3.50 3.51
Li'" 5.24 5.22 5.21 5.20
Mg+-+- 5.24 5. 22 5.41 5.39
Mn+-+- 5.24 5.22 5.41 5.41

aApproximatelY equal to zirconium concentration.

bThiourea (0.5 grams) added before the titration.
14 ISC-596

Table 2 (Continued)

Direct titration Back titration

EDTA EDTA EDTA EDTA
Theoretical Experimental Theoretical Experimental
Ion added (ml) (ml) (ml) (ml)

Mo._., .. 5.24 interferes 5.41 interferes

Na+- 5.24 5.24 5.69 5.70
Ni~~ 5.24 5.36 5.21 5.25
Nll4 .. 5.24 5.24 5.21 5.21
Pb~~ 5.24 5.22 5.21 5.18
Sb+-+-+- 5.24 interferes 5. 21 interferes
Sr+-+- 5.24 5.23 5.21 5. 23
sn+-+- 5.24 interferes 5 .21 interferes
sn+-+-.. +- 5.24 interferes 5.41 5.38
Ti+-+-+- 5.24 interferes 5~35 interferes
Ti+-+-+-+- 5.24 interferes 5.41 5.37
Th+-+-+-+- 5.24 interferes 5.41 interferes
uo 2 ~ .. 5.24 5.24 3.50 3.50
vo+-+- 5.24 5.33 4.17 4.18
Zn+-+- 5.24 5. 24 5.21 5.23
C2H3o2- 5.41 5.38 S.69 5. 70
F- 5.41 interferes 5.21 interferes
Mandelate 5.24 5.25 4.17 4.18

MoO~
= 5.24 interferes 5.41 interferes
so4- 5.41 interferes 5 .21 interferes
Tartrate 5.24 .interferes 5.21 interferes
V03- 5.24 5.21 5. 21 5. 21
WO
4 = 5.24 interferes 5~23 interferes
 ISC-596 15

Two methods can be used to get stoichiometric results if copper(II)

is present. Copper interferes in a direct titration, but no inter-
ference is noted if excess EDTA is added and the solution is heated
several minutes before back titration. Appd.rently the copper-EDTA
complex is almost as strong as the zirconium-EDTA complex. If 0.5 gram
of thiourea is added to the solution before the titration, a strong, .
colorless copper-thiourea complex forms which removes the copper inter-
ference completely. The solution can then be titrated directly or by
the back titration procedure.

Although iron(III) forms a very strong complex with EDTA

(log K = 25), iron(IT) forms a comparatively weak complex (log K = 14. 7)
(6). Interference from iron can be avoided by a preliminary reduction
with zinc amalgctm. The titration must be carried out in an inert
atmosphere because iron(II) is very easily oxidized to iron(III) in the
presence of EDTA causing the end point to fade. Table 3 shows a stuqy
of iroh interference.

Table 3. Iron Interference Study

Mole ratio Theoretical EDTA Experimental EJ1l'A Difference

Fe/Zr (ml) (ml) (ml)

1/1 5.33 5.36 t-0.03

2/1 5 -~ ~, 5.34 t-0.01
4/1 5.33 5.33 0.00
5/1 2.13 2.15 t-0. 02
10/1 2.13 2.13 0.00
15/1 2.13 2.13 0. )0

Interference by: tin(IV) is avoided by adding excess EDTA to a rather

acidic sample (pH about 0.5) and then back titrating at room temperature
with standard zirconyl chloride solution. Ti tanium(IV) interference is
avoided by a similar procedure, but EDTA must be added in excess of the
amount equivalent to the combined zirconium and titanium. The solution
is then heated almost to boiling and back titrated (at pH 1.2 to l. 4)
with standard zirconyl chloride solution. The presence of either tin
or titanium decreases the sharpness of the end point.
16 ISC-596

When nickel is present in an atmunt equimolar to the zirconium

concentration, a reproducible error of 2 per cent is encountered in
the direct titration; but no interference is encountered in the back
titration. The reasons for the behavior in the direct titration are
not apparent.

No interference is caused by perchlorate, chloride, nitrate, or

acetate.· Anions which complex zirconium strongly or precipitate
zirconium, such as photphate, molybdate, sulfate, fluoride, and
oxalate, interfere seriously.

Hafnium and thorium interferences were investigated quantitatively.

Mi-xtures of known amounts of zirconium and hafnium and zirconium and
thorium were prepared and titrated. The results tiven in Table 4
confirm the quantitative nat~re of these two interferences.

Table 4. Titration of Mixtures

Theoretical EDTA Experimental EDTA

(ml) (ml)

:gquivalent to Zr 3.66
Equivalent to Th 2.59
J.~qui valent to sum 6.25 6;26, 6.22, 6.22

Equivalent to Zr 2. 7l
Equivalent to Hf 2.64
Equivalent to sum 5.35 5.32, 5.33, 5.32

IV. ANALYS~S OF SM!PLES CONTAINING ZIRCONIUM

A stuey of a new method of analysis would not be complete without

a few practical analyses of unknown samples and a comparison of the
results with those of another method. Samples of zircon sands and
zirconium-zinc alloys were available. The method used for comparison
was the mandelic acid precipitation of zirconium proposed by Kumins(30),
this being the most selective and commercially available reagent for
zirconium and hafnium.
 ISC-596 17

Three spectrographic determinations, one on reagent grade zirconyl

chloride octaqydrate, one on a zirconium zinc alloy, and one on a zircon
sand sample showed a Hf/Zr ratio of 2.0 ~ 0.1. This means that about
2 per cent of the zirconium in the samples was actually hafnium. Using
this factor of 2 per cent, the author found that the "molecular weight"
of zirconium to be used in the volumetric determination should be 92.97
instead of 91.22, an increase of 1.92 per cent. The value for the
"molecular weight" of zirconium oxide is 124.97 instead of 123.22, an
increase of 1.42 per cent. When not taken into account, this difference
causes discrepancies when re::>ul ts for the titrine tric procedure are
compared with those of the gravimetric procedure.

A. Zircon Sand Samples

When received, these samples were in the form of a powder from a

sodium hydroxide fusion. The samples were leeched with water to remove
some of the sodium hydroxide, dissolved in hydrochloric acid and diluted
to volume in a volumetric flask so that a 10-milliliter aliquot
contained 0.15 to 0.25 millimoles of zirconium. The zirconium was
determined ti t.rimetrically by the back titration procedure and gravi-
metrically by the mandelic acid procedure of Kumins (30). Table 5
shows a comparison of results by the two procedures.

Table 5. Analyses of Zircon Sands for Zirconium

Ti trimetric determination Gravimetric Deviation

Range No. Average detennination from gravimetric
% Zr detns. %Zr %Zr %Zr

19.99 - 20.89 2 20.44 20.55 -0.11

21.48 - 22. 27 2 21.86 i2.57 -0.71
21.66 - 21.77 3 21.71 21.97 -0.26
21.51 - 21.62 3 21.55 21.76 -0.21

39.78 - 40.00 3 39.85 40.52 -0.67

39.55 - 39.55 3 39.55 40.16 -0.61
34.63 - 34.76 3 34.70 35.66 -0.96
34.69 - 34.78 3 34.74 35.57 -0.83
lB ISC-596

Some hydrous oxide of silica slowly precipitated in the volumetric.

flask. As little of this precipitate as possible was allowed in the
aliquots by carefully pipetting aliquots from above the precipitate.
The precipitate does not interfere in the ti trimetric procedure in
small amounts; in large annunts it adsorbs the indicator. The pre-
cipitate· does interfere in the gravimetric procedure, but it can be
removed by evaporation with hydrofluoric acid after the zirconium
mandelate precipitate has been ignited to the oxide. The fact that
all of the titrimetric determinations in Table 5 deviate negatively
fr~m the gravimetric determilliitions indicates that perhaps all of the
silica was not removed in the gravimetric determinations.

B. Zirconium-Zinc Alloys

When received, these samples were in the massive form. Some

coUld be broken and ground up with a mortar and pestle; others had to
be cut into smaller pieces with an abrasive cutter. Samples contain-
ing 250 to 400 milligrams of zi:r:conium were weighed and dissolved in
bromine and ethyl acetate. The solution was evaporated almost to
dryness, and the residue was dissolved in 10 milliliters of concentrated
hydrochloric acid with heating. The alloys dissolved very readily in
bromine and ethyl acetate; however, an alternate solvent mixture was
nitric and perchloric acids. Dilute nitric acid (1 ~ 9) was added
until the vigorous bubbling caused by the dissolving zinc subsided;
then 10 milliliters of concentrated perchloric acid was added, and the
solution was heated for one to three hours or until solution was
complete~ The solutions were transferred to 200-milliliter volumetric
flasks and diluted to volume. Ten-milliliter aliquots were used for
the titrimetric determination by the back titration procedure, and
50-milliliter aliquots were used for the gravimetric determination by
the mandelic acid precipitation method of Kumins (30). Table 6 shows
a comparison of the results by the two procedures.
 ISC-596 19

Table 6. Analyses of Zirconium-Zinc Alloys for Zirconium

Volumetric determination Gravimetric Deviation

Range Average dete:rmination from gravimetric
%Zr %Zr %Zr %Zr

24.34 -
24.47 24.45 24.59 -0.14
2.80 2.82
- 2.81 2.90 -0.09
8.80 -
9.04 8.91 9.07 -0.16
12.53 -
12.59 12.58 12.60 -0.02
13.07 -
13.16 13.12 13.37 -0.25
2.44 - 2.47 2.46 2.34 +-0.12
10.16 - 10.31 10.24 10.35 -0.11

V. SUGGESTIONS FOR FUTURE WORK

Work must be done to avoid the interferences of fluoride,

phosphate, sulfate, and other ions. Hydrofluoric acid and sulfuric
acid are good solvents for ores and alloys of zirconium, and phosphate
is a good quantitative precipitant for the separation of zirconium~
Possibly an extraction procedure such as that used by Fritz and Ford
(12) for thorium can be worked out.
For titrations involving 25 to 50 milliliters of EDTA solution
the pH change during the titration may be considerable. The use of a
buffer which does not interfere in the titration would be an improve-
ment on the proposed method.

From a more general outlook titration with complexing agents is

an approach offering wide opportunity for rapid and selective deter-
minations. Perhaps the applications can be extended to non-aqueous
systems.
 20 ISC-596

VI. LITERATURE CITED

la. Alford, W. C., Shapiro, 1., and White, C. E., Anal. Chern., 23,
1149 (1951).

lb. Alrose Chemical Company. 11 Sequestrene,li Providence, Rhode

Island, Alrose Chemical Co. 1952 .

2a. Barnard, R. L. and Telford, R. E. Zirconium. In Rodden, C. J.

"Analytical Chemistry of the Manhattan Project." New York,
McGraw-Hill Book Co., Inc. 1950.

2b. Bersworth Chemical Company. "Bibliography of Ethylenediamine-

tetraacetic Acid. 11 Framingham, Mass., Bersworth Chemical
Company, now Versenes Inc. 1953.

3. Barrel, M. and Paris, R., Anal. Chim. Acta, _!±, 267 (1950).

4. Brady, E. 1. and Eng1ekemeir, D. Vl. In Coryell, C. D. and

Sugarman, N., eds. ''Radiochemical Studies: The Fission
Products." Book 3. p . 1491. Nevi York, Mc Gr aw-Hill Book
Co., I .n c. 1939 .

S. Brown, E. J. and Hayes, T. J., Anal. Chim. Acta, _2., 1 (1953).

6. Cabell, M. J., Analyst, 77, 859 (1952).

7. Cozzi, D., Proc. Intern. Congr. Pure Applied Chern. London, 11,
57 (1947) (In Italian). (Original not available for examina-
tion; abstracted in ·rechnical Information Service, AEC 'report
TID-3010. p. 12).

8. Dewell, E. R. Unpublished M. S. Thesis. Ames, Iowa, Iowa State

College Library. 1951.

9. Dhar, S. K. and Das Gupta, A. K., J. Sci. Ind. Research (India),

11, 500 (1952). (Original not available f"'Or' examination;--
abstracted in Q· A., 47, 6306 (1953) ).

10. Diskant, E. M., Anal. Chern., 24, 1856 (195~).

11. Fick, R. and Ulrich, H., German Patent 638,071 (19·3 6). Abstracted
in~~' 31, 1043 (1937). ·

12. Fritz, J. S. and Ford, J. J., Anal. Chern., 25, 1640 (1953).
 ISC-596 21

1). Furman, N. H. ed. ''Scott's Standard Methods of Chemical Analyses."

Vol. 1 New York, D. Van Nostrand Co., Inc. 1939.
14. Graham, R. P., Van Dalen, E., and Upton, A. M. C., Can. J. Chern.,
30, 1069 (1952). (Original not available for examination; - -
abstracted in Q· !·, 48, 6909 (1954) ) •
15. Green, D. E., Anal. Chern., 20, 370 (1948).
16. Grimaldi, F. S. and White, C. E., Anal. Chem., 25, 1886 (1953) ..
17. Guenther, R. and Gale, R. H. Knolls Atomic Power Laboratory
report KAPL-305, 1950.
18. Hahn, R. B., ~. Chern., 21, 1579 (1949).
19. , Anal. Chern., 23, 1259 (1951).
20. Horton, A. D., Anal. Chern., 25, 1331 (1953).
21. Hume, D. N. In Coryell, C. D. and Sugarman, N. eds.
"Radiochemical Studies: The Fission Products." Book J.p. 1499.
New York, McGraw-Hill Book Co., Inc. 1951.
22. Jewsbury, A. and Osborn, G. H., Anal. Chim. Acta, l' 642 (1949).
23. Jonckers, M. D. E., Chim. Anal., 32, 207 (1950). (Griginal not
available for examination; abstracted in Technical Informa-
tion Service, AEC report TID-3010. p. 65).
24. Kiefer, E. W. and Boltz, D. F., Anal.~., 24, 542 (1952).
25. Kingsbury, G. W. J. and Temple, R. B.,~· Applied Chern.,!' 406
(1951).
2b. Klingenberg, J. J. and Papucci, R. A., Anal. Chern., 24, 1~61
(1952).
27. Klingenberg, J. J., Vlannes, P. N., and Mendel, M. G., Anal.
Chern., 26, 754 (1954).
28. Kalthoff, I. M. and Johnson, R. A., J. Electrochem. Soc., 98, 138
(1951).
29. Kalthoff, I. M. and Sandell, E. B. "Textbook of Quantitative
Inorganic Arralysis. 11 3rd ed. New York, The Macmillan Co.
1952.
22 ISC-596

30. Kumins, C. A., Anal. Chern., 19, 376 (194?).

31. Larsen, E. M. and Gammill, A. M., J. Am. Chern. Soc, 72, 3615
(1950).
32. Larsen, E. M. and Terry, G.,.!!:· Am. Chern. Soc., 75, 1560 (1953).
33. Levine, H. and Grimaldi, F. S. Atomic Energy Commission report
AECD-3186, 1950.
34. Litton, F. B. Atomic Energy Commission file no. NP-3276;
ATI -97923 report no. 8. (Original not available for ex~ina­
tion; abstracted in Technical Information Service, AEC
report TID-3010. p. 42).
35. Mayer, A. and Bradshaw, G., Analyst, 77, 476 (1952).
36. Mellor, J. "Comprehensive Treatise on Inorganic and Theoretical
Chemistry. 11 Vol. 7. London, Longmans, Green, and Co. 1927.

37. Miller, G. L. "Zirconium." New York, Academic Press, Inc. 1954.

38. Mills, E. C. and Hermon, S. E., Analyst, 78, 256 (1953).
39. Milner, G. W. C. and Phennah, P. J. Atomic Energy Research
Establishment report AERE-C/R-1354.
40. Mortimore, D. M.. and Romans, P. A.,~· Optical Soc. Am., 42, 673
(1952).
41. Murty, P. S. a.nd Rao, B. S. V. R., ~· Anal. Chern., 141, 93 (1953).
42. Nishikawa, T. and Machida, Y., J. Chern. Soc. Japan. Pure Chern.
Sect., 67, 102 (1946). (Origin~ not available for ex~in~
tion; abstracted in Technical Information Service, AEC
report TID-3010).
43. Oesper, R. E., Dunleavy, R. A., and Klingenberg, J. J., Anal.
Chern., 24, 1492 (1952).
44. Oesper, R. E. apd Klingenberg, J. J., Anal. Chern., 21, 1509 (1949).
45. Papucc::4 R. A., Fleishman, D. M., and Klingenberg, J. J., Anal.
Chem. ,. 25, 1758 (1953).
46. Purushottam, A. and Rao, B. S. v; R., Analyst, 75, 555 (1950).
 ISC-.596 23

47. Rao, C. L., Venkataramaniah, H., and Rao, B. S. V. R., J. Sci.

Ind. Research (India), 10, 1.52 (19.51). (Original not -
available for examination; abstracted in c. A., 46, 2957
(19.52) ) • - -

48. Sawaya, T. and Yamashita, M., J. Chem. Soc. Japan. Pure Che~.
Sect., 72, 3.56 (19.51). (Original not available for examina:-
tion; abstracted in Q· !·' 46, 8.54 (19.52) ).
49. Scadden, E. M. and Ballou, N. E., Anal. Chern., 2.5, 1062 (19.53).

50. Schwarzenbach, G. and Ackermann, H., Hel v. Chim. Acta, 30, 1798
(1947) •
.5L Stehney, A. F. and Safrinski, L. W. Argonne National Laboratory
report AECD-3097.
52. Steinberg, E. P. In Coryell, C. D., and Sugarman, N., eds.
"Radiochemical Studies: The Fission Products. rr Book 3.
p. 149.5. New York, McGraw-Hill Book Co., Inc. 19.51.

53. Taketatsu, T., ~· Chern. Soc. Japatt. l?ure Chern. Sect., 74, lOll
(19.53). (Original not available for examination; abstracted
in Q· !·' 48, .5727 (1954) ).
54. Thamer, B. J. and Voigt, A. F., J. Am. Chern. Soc., 73, 3197 (19.51).

55. Venkatararnaniah, M. and Rao, B. S. V. R., Anaiyst, 76, 107 (19.51).

56. ___ and _ _ _ , Anal. Chern., 23, .539 (19.51).

57. and , J. :rndian Chern. Soc., 28, (19.51). (Original

not available for examination; abstracted in-Q. !·' 46,
603.5 (19.52) ).

58. _ _ _ and _ _ _ , ~· Anal. _Chern., 133, 248 (1951).

59. Venkateswarlu, C. and Rao, B. S. V. R., J. Indian Chern. Soc., 28,

3.54 (19.51). (Original not available for-examination;
abstracted in Q· ~., 46, 603.5 (19.52) ).
60 . and , J. Indian Chern. Soc., 27, 395 (19.50).
(Original not available for examination; abstracted in
Technical Information Service, AEC report TID-3010. p. 81).
24 ISC-596

61. Voreas, H. E. and Croxton, F. E. Technical Inlormation Service,

AEC report TID-3010. 1951.
62. Weeks, M. E. "The Discovery of the Elements." Sth ed. baston,
Penn., Journal of Chemical Educatior. 1945.

63. Wengert, G. B., An~-· Chern., 24, 1449 (1952).

64. White, J. C. Oak Ridge National Laboratory re~ort ORNL-1515, 1953.

65. Willard, H. H. and Diehl, H. "Advanced Quantitative Analysis."

New York, D. Van Nostrand Co., Inc. 1944.

66. Willard, H. H. and Hahn, R. B., AnaL. Chern., 21, 293 (1949).

67. Zacharias, J. R., Gilliland, E. R., Hubbard, M. M., and Foster, C. C.

Massachusetts Institute of Technology Progress report no. 17
for July 1, 1949 - July 1, 1950. AEC file no. NP-1777.
(Original not available for examination; abstracted in
Technical Information Service, AEC report TLU-3010. p. 83).

68. Zirconium. '!'he Iron Age, 170, 286 (1952).