THE SPECTROPHCTOMETRIC DETERMINATION OF METALS

WITH CHLOROINDAZON DS AND CHLOROINDAZON C

by

Cham Shi-Fai

A thesis submitted in partial fulfilment of the

requirement far the degree of

Master of Philosophy in

The Chinese University of Hong Kong

1977

Thesis Committee:

Dr. 0. W. Lau, Chairman

Dr. P. K. Hon

Dr. W. K. Li

Prof. T. S. Ma, External Examiner

( 覃 仕 輝 )

ACKNOWLEDGMENT

I wish to express my gratitude to Dr. 0. W. Lau for her

guidance and discussion during the course of this research.

Department of Chemistry

The Chinese University of

CRAM Shi-FaiHong Kong

May, 1977.

1

ABSTRACT

This thesis is concerned with the evaluation of two azo

compounds as spectrophotometric reagents for the determination of

trace amounts of metal ions.

The basic principles of spectrophotometry and a brief descripticn

of the apparatus for making spectrophotometric measurement are

presented in Part I of this thesis.

Part II of this thesis deals with the spectrophotometric

determination of cobalt with 1-(6 t _chloro_indazole-3' -rlazo) -2--hydroxy-.

naphthalene-3,6-disulphonic acid (Chloroindazon DS). Chloroindazon DS

dissolves in ammonia and forms a green-coloured complex with

cobalt., which has an absorption maximum at 638 nm. About 100 pg

of cobalt in 100 ml of solution was found to require 45 ml of

0.01% Chloroindazon DS solution for complete reaction. When

Chloroindazon DS is present in less than one fold of the amount

required in excess, no interference was found to be caused by the

reagent. Reaction between Chloroindazon DS and cobalt requires a

long time for comp-2.etion. Raising the temperature to 60°C and 100°C

could reduce the time to twenty minutes and five minutes respectively

for complete colour development. The molar absorptivity of the

complex determined is 3.25 x 1041-mol-1-cm`1 at 638 nm and the

Sandell sensitivity is 0.00181 pg/cm2. The complex is a 1:3 one and.

was determined by the Job's method. Beer's law is obeyed within the

concentration range of 0-5-p.p.rn, of cobalt. When the pH is below

4.6, no complex is formed with cobalt. However, the green complex

once formed is stable at all pH, and shows maximum absorbance at

pH 6.5 and 11.42 respectivety. The relative

2

standard deviation for seven replicate determination of 1p.p.m.

of cobalt is 5.39 x 10-3, and the recovery for cobalt proved

quantitiative.

Among the cations, only copper (II) and iron (II) cause serious

interferences. Nickel (II) also interferes, hut its interference

can be eliminated by adding 0.1 gm of sodium citrate to the solution,

zinc (II), cadmium (II), mercury (II), calcium (II), magnesium (II),

iron (III), aluminum (III) and chromium (III),do not interfere when

the pH of the solution is at 6.5. As for the anions, EDTA and cyanide

destroy the complex completely, but tartrate and citrate show

no interference.

Other metals which also form coloured complexes with

Chloroindazon DS are: nickel (violet), cadmium (maga.nta), zinc

(violet), copper (blue), and mercury (I) and mercury (II) (red-

violet). Except for nickel and copper, Chloroindazon DS cannot

be used for the spectrophotometric determination of these metals

since the reagent has strong absorption at wavelengths close to

the absorption maxima of the complexes formed between the reagent

and these metals. The reagent will still interfere with the

determination of copper and nickel.if excess reagent is present.

The third and remaining part of this thesis deals with the

spectrophotometric determination of calcium with 1-(6 -Chloroindazole-

i

3 -ylazo)-2-hyrdroxynaph-thalene-3-carboxylic acid (Chloroindazon C).

Chloroindazon C dissolves in ethanol and forms a red- violet

complex with calcium, which has an absorption maximum at 586 nm.

About 100 ig of calcium requires 25 ml of 0.C1% ethanoic Chloroindazon C

solution for complete complex formation. Full colour development

3

takes only ten minutes at room temperature. The molar extinction

coefficient and the Sandell sensitivity of the calcium complex at

586 nri are 1.89-x 104 1-mol- 1 -cm -1 and 0.00212 ug/cm2 respectively.

Beer's law is obeyed within the concentraiion range of 0-1 p.p.m.

o calcium. The calcium complex is only stable at pH between

70 to 11.8 since no complex is formed at pH below 10, and yet at

pH above 11.8, the complex is precipitated immediately. Maximum

absorbance occurs around pH 10.8 to 11.5.. Colour development of

the calcium complex was found to depend on the amount of ethanol

present, the order of mixing, and the nature of the buffer solution.

The relative standard deviation for eight replicate determination

of 1p.p.m. of calcium is 1.30 and the recovery for calcium proved

quantitiative.

The interference by aluminum and stronium can be eliminated

by adding 0.05 gm tartrate to the solution. Nickel (II), copper (II),

cobalt (II), zinc (II), cadmium (II) and mercury (II) are masked

by potassium cyanide. Large amount of barium present will cause

the precipitation of the complex. Magnesium and chromium (III)

cause serious interference, however, chromium in quantities up to

50 jig can be tolerated. As for the anions, EDTA and excess of

phosphate and citrate must be absent, but tartrate did not cause

interference.

Other metal ions which form coloured complexes with Chloroindazon C

are: cobalt (green), nickel (violet), zinc (violet), cadmium (maganta),

mercury (violet), copper (blue), stronium (orange), magnesium (orange

red) and aluminum (orange). Among these ions, only cobalt, zinc,

copper and nickel can be conveniently determined by Chloroindazon C

si ectrophotometrically since it absorbs stronly at wavelengtls close

to the absorption lx, ,xima of the complexes which it forms with the other

metals.

4

CONTENTS

PAGE

ABSTRACTi-iii

PART I

INTRODUCTION 1-2

CHAPTER 1Principles of Spectrophotometry 3-6

REFERENCES7

PART II

SPECTROPHOTOMETRIC DE'llERMINATION OF METALS

WITH CHLOROINDAZON DS

CHAPTER 2 Research Plan 8-10

CHAPTER 3 Experimental 11-14

CHAPTER 4 Results and Discussion 15-51

REFERENCES52-53

PART III

SPECTROPHOTOMETRIC DETERMINATION OF METALS

WITH CHLOROINDAZON C

CHAPTER 3 Research Plan 54-55

CHAPTER 6 Experimental 56-58

CHAPTER 7 Results and Discussion59-87

REFERENCES 88

1

PART I

INTRODUCTION

The selective absorption of electromagnetic radiation as

it passes through a solution causes the emerging beam to differ

from the incident one. In the case of visible radiation, this

difference is frequently obvious to the naked eye. A large body

of analytical techniques is based on the ability of substances to

emit or absorb electromagnetic radiation. Many of these methods

are of fairly recent origin, although the general principles of

the processes have been understood for many years. A relation-

ship may be established between the ability of a substances to

absorb radiation of a given wavelength and the concentration of

the substance in the matrix involved. Methods of quantitative

.analysis based on relationship involving absorptivity are, in

general, identified as absorptiometric methods.

Spectrophotometric methods are absorptiometric method

involving light of definite, wavelength( not exceeding, say,

1-10R in band-width) extending to the ultra-violet reagion of

the spectrum. The present thesis is concerned with the evaluation

of two azo compounds, namely, Chloroindazon DS and Chloroindazon C

as spectrophotometric reagents for the determination of trace metals.

It was found that the former reagent is particularly suitable for

the analysis of cobalt and the latter good for the determination of

calcium.

2

At the outset, a brief description of the principles and

apparatus of spectrophotometry will be given. Following this

introductory chapter, description of the synthesis and application

.of Chloroindazon DS for the determination of cobalt and other metal

ions will be given in chapter 2-4, which constitute the second part

of this thesis. The third and remaining part, consisting of three

chapters will deal with the application of Chloroindazon C for the

analysis of calcium and possibly several other metal ions.

3

CHAPTER 1

PRINCIPLES OF SPECTROPHOTOMETRY

10 Lambert-Beer's 1aw

When radiation is incident on matter and if the frecency,

of the radiation is the same as the energy required to rise

the substance from its ground state to the allowed higher

energy level, absorption will occur. Sometimes, the absorption

is selective for a particular wavelength, but in general,

the selectivity is not so good and a wide range absorption:

will occur.

Regardless of wavelength region, the principles and

laws governing the absorption of radiation are the same. Quite

generally, absorption measurements involve determination of

the reduction in intensity suffered by a beam of light as a

consequence of passing through the absorption medium.

When a beam of parallel monochromatic radiation traversing

a solution contained in a cell of thickness b, the rate of

decrease of the incident radiant power P is given by

equation (1)1,

P= Po exp(-k'b) (1)

where Po is the incident radiant power, P is the radiant power

after transmitted through thickness b. k' is the proportionality

constant, which is dependent on wavelength, concentration,

solvent and also temperature. However, for a solution of

known concentration at a given temperature, k' depends on

wavelength only. This is usually called the Lambert's law

or is also known as the Bouguer's law.

4

The relationship between the concentration of an absorbing

species and the extent of absorption was formulated by Beer in

1859. Beer's law is analogous to Lambert's law in describing an

exponential decrease in transmitted radiant power with an arithmetric

increase in concentration.

P= Po exp(-k"c) (2)

The proportionality constant k" depends on wavelength only

if the temperature and path length are fixed.

The Lambert and Beer laws may be combined into a single

relationship, which is commonly just called Beer's law and may

be written as

(3)

or written with the Brisian logarithms as

(4)

where k"= 2.3a, inverting the log term to change the sign yields-

(5)

PoThe term log is called the absorbance and given the symbol A.P

The term a is called absorptivity. However, it is called the

molar absorptivity when the concentration c is expressed in terms

of moles of absorber per liter and the path length is given in

centimeters, and is usually designated as E. Equation (5) is

sometimes written as

A = E cl (6)

where c is the molar concentration and 1 the path length in cm.

According to equation (6), a plot of absorbance versus

molar concentration will give a straight line passing through

the origin. However, for rriany absorbing substances this is true

only up to a certain concentration, and as discussed in the next

section deviatiors from linearity are quite frequently encountered.

5

2. Components of Instruments for Spectral Measurements and Deviation

from Beer's Law

Any photometer or spectrophotometer must have four basic

components, which are shown schematically in Fig 1. These

components are source of radiation, monochromator, detector and

output system. The detailed description of

Source Monochromator Sample Read outDetector

Fig 1. Block Diagram showing components of a single beam

spectrophotometer.

these components can be found in many texts on analytical

chemistry.

The linear relationship between absorbance and path length

at a fixed concentration of absorbing substances is a general-

ization for which no exceptions are known. On the other hand,

deviations from the direct proportionality between measured

absorbance and concentration are quite frequently encountered.

The deviations are often classified as the instrumental, chemical

and real deviation from Beer's law.

The instrumental deviations arise because of the use of

polychromatic or non-parallel incident radiation2.

The chemical deviation from Beer's law of a system for

chemical reactions occuring in the solution. A classic example

of a chemical deviation is observed with unbuffered potassium

dichromate solutions, in which the following equilibria exist:

6

The total absorbance of and Cr(VI) solution is dependent on the

ratio of concentration between the dimeric and monomeric forms.

This ratio changes markedly with dilution and causes a propounced

deviation from linearity totween the absorbance and the total

concentration of chromium.

Futhermore, Beer's law as it is generally stated is actually

the limiting case of a more exact relation:

in which n is the refractive index of the solution3. This

expression takes into account the light reflected at the solution

interface as a function of the refractive index. Ordinarily n

changes so very little from one concentration to another that

the coefficent may be disregarded. Thus this class

of deviation is not likely to be encounted in analytical chemistrye

Therefore, it is necessary to check before assuming Beer's law

to hold for a particular chemical system with a particular

instrument.

7

REFERENCES

(1) R. B. Fischer and D. G. Peters, "Quantitative Chemical

Analysis", 3rd edition, Sanlanders, 1968, P.621- 625.

(2) Kolthoff and Elving, "Treatise on analytical chemistry",

John Wiley and Sons, 1964, volume 5, part 1, P.2767- 2772.

(3) G. Kortum, "Das optische Verhalten geloster Elektrolyte",

Enke, Stuttgart, 1936, P.65.

PART II

SPECTROPHOTOMETRIC DETERMINATION OF METALS

WITH CHLOROINDAZON DS

CHAPTER 2 RESEARCH PLAN

In recent years, numerous reagents for spectrophotometry

determination of cobalt have been proposed Some of them either

offer high sensitivity at an unfavourable wavelength

phenanthroline or suffer from

numerous interferences, -nitroso naphthol

bipyridine

dithizone

oxine etc. Other reagents

offer high sensitivity and small interferences, but they cannot

dissolve in water or form coloured complexes with cobalt soluble

in water. Examples are ethanolic or dimethylformamide solutions

of 3-hydroxy-picolinealdehyde

salicylaldehyde and

picolinealdehyde razines

and dimedone dioxime' solution in ethyl alcohol

Only cyanide ion interferes with the

determination of cobalt using these reagents, but they are all

spaingly soluble in water. The use of these reagents is not

very convenient in practice.

the datum in these and subsequent brackets refer to the molar

absorptivity of the coniplex formed between the reagent and

cobalt.

Many azo compounds are also used as sensitive and selective

reagents for spectrophotometric determination of cobalt.

The reagent 3- (3?3- dichloro-2-pyridyl) azoj -2,4—diaminotoluene

was reported to form a purple complex with cobalt. The molar

absorptivity was 1.38 x 10 1-mol -cm at 390nm. Beer's law was

obeyed between 0.01-0.4-p.p.m. of cobalt, and the Sandell sensitivity

was 0.42 ngcm. Interference was caused by•palladium, potassium

cyanite and EDTA. Another reagent, 3-(3-chloro-2-pyridyl) azoj-

2,6-diaminopyridine (3-C1-PADAPT), formed a deep blue water-

soluble complex with cobalt. The molar absorpitivity was 36 x 10

1-mol -cm at 620 nm. Beer's law was obeyed between 0.1-1.2p.p.m.

of cobalt. Only iron (III) and large amount of copper were found

to interfere. Although the above two azo compounds have high

sensitivity and selectivity for cobalt, unfortunately, they are

also insoluble in water themselves.

The azo compound, 1—(6 -Chloroindazole-3 -ylazo)-2-hydroxy-

naphthalene-396 disulphonic acid (Chloroindazon DS) has the

following structure:

10

It forms coloured complex with cobalt (green), copper (blue),

nickel (violet), zine (violet), cadmium (magenta) and mercury

(red-violet). Chloroindazon DS is soluble in alkaline aqueous

solution. Preliminary investigations show that this reagent is

very promising for the determination of cobalt. It is, therefore,

the object of this work to investigate the possibility of developing

Chloroindazon DS into a specific and sensitive reagent for the

spectrophotometric determination of cobalt.

In 1975, Molch 24 has used Chloroindazon DS on the spectrophotometric

determination of Cobalt. The results are compared with the present

work. Also, he used the same reagent as a spectrophotometric reagent

to determine nickel in the presence of chromium in chromium-nickel

thin layers12, and to determine copper in coating on quartz crystal

oscillators13.

The optimum experimental conditions for the determination

of cobalt spectrophotometrically with this reagent, the sensitivity

of the method and the effect of foreign ions will all be studied.

The possibility of using Chloroindazon DS to determine other

metal ions will also be investigated.

11

EXPERIMENTALCHAPTER 3

preparation of Chloroindazon DS

(1) Diazotizat on

H. Schweppe8 suggested a method to prepare this reagent by

dissolving 16.7 gm 3-amino-6-Chlorindazole in 300 ml 1M hydrochloric

acid. Small pieces of ice (250 gm) were added to cool down the

solution at about 3°C. Then 40 ml of 20% aqueous solium nitrite

were added to the solution slowly. The precipitate of diazonium

chloride was obtained.

(2) Coupling

2-Hydroxynaphthalene -3,6 disulfonic acid (34.8 gm) was dis-

solved in 300m1 of 1M sodium hydroxide. The solution was cooled

to about 5°C by addition of small ice pieces. Then this solution

was added to the suspension of diazonium chloride with stirring.

The pH of the reaction mixture was adjusted to about 7. A large

amount of the reagent in the form of its sodium salt was precipitated.

The precipitate was collected by suction and dried at 100°C, and

ground into powder.

2. Reagents

(1) Chloroindazon DS solutions

Chloroindazon DS (100mg) was dissolved in 5ml ammonia (density =

0.885) and dilute to the mark with distilled water in a 100ml volumetric

flask. This solution was marked as a 0.1% Chloroindazon DS. Ten

millilitre of the 0.1% Chloroindazon DS soluticn was diluted to 100ml

1.

12

to prepare the 0.01% Chloroindazon DS solution. As the reagent solution

may be oxidized on standing for more than a week, it was stored in a

refrigerator to prevent air oxidation.

(2) Standard cobalt (II) solutions

A stock solution of 100p.p.m. was prepared by dissolving 476.8 mg

ofdcobalt sulfate heptahydrate (BDII, A.R. Grade) in distilled water

and dilute to one liter in a standard flask. From this stock solution,

standard solutions of 60,50,40,30,20,10,8,6,4, and 2p.p.m. were

prepared.

(3) Buffer solutions

The following buffer systems were prepared:

(a) Potassium hydrogen phthalate (0.082M) - hydrochloric acid (0.017M)

(pH = 3.40)

(b) Potassium hydrogen phthalate (0.085M) - sodium hydroxie (0.015M)

(pH = 4.50)

(c) Potassium dihydrogen phosphate (0.078M) - sodium hydroxide (0.22M)

(pH = 6.50)

(d) Borax (0.0197M) - hydrochloric acid (0.021M) (pH= 8.60)

(e) Borax (0.021M) - sodium hydroxide (0.015M) (pH= 9.50)

(f) Disodium hydrogen phosphate (0.041M) - sodium hydroxide (0.018M)

(pH = 11.5)

All the buffer solutions were prepared from analytical wade reagents.

the concentration inside the brackets after each compound indicates

its concentration-in the resulting buffer solution prepared

according to the "Handbook of Chemistry and Physics", 56th edition,

1975-76, CRC Press. D134-135.

13

(4) Standard solutions of other metal ions

Stock solutions of 100p.p.m. were prepared by dissolving

separately copper sulfate pentahydrate (196.4 mg), nickel

sulfate heptah/drate (239.1 mg), zine sulfate heptahydrate

(219.8 mg), cadmium sulfate (114.1 mg), magnesium sulfate

heptahydrate (506.8 mg), calcium acetate (197.2 mg), ferrous

ammonia sulfate hexahydrate (350.9 mg), aluminum sulfate

(583.9 mg), chromium sulfate (318.4 mg), mercury chloride

(67.7 mg) and ferric chloride hexahydrate (241.9 mg) in distilled

water and diluting to 500 ml in standard flasks. All the reagents

were analytically pure.

3. Apparatus

All absorption spectra were recorded with a 323 Hitachi

Recording Spectrophotometer using matched 1-cm silica cells.

A Radiometer PHM28 pH-meter with saturated calmoel- glass

electrode system was used for all pH measurement.

4. Recommended Procedure for the Determinztion of Cobalt

A suitable aliquot of sample solution containing up to

500 Ug of cobalt was pipetted into a 100-ml volumetric flask.

This was followed by the addition of 30 ml of potassium dihydrogen

phosphate- sodium hydroxide buffer solution to adjust the pH

to 6.50. If nickel was known to be present in the sample solution,

0.1g of sodium citrate should be added to the solution before

adding the buffer solution. Then 22.5 ml of 0.1%. Chloroindazon DS

solution was added. The solution was heated to 60°C and maintained

at this temperature for twenty minutes or heated to 100°C for

five minutes. After cooling to room temperature, the solution

14

was diluted to the mark with distilled water. The absorbance of

the cobalt complex was measured at 638 nm in matched 1-cm silica

cells against water as a reagent blank. The concentration of

cobalt could be deduced from a calibration graph obtained in a

similar manner from solutions containing 0-500 ug of cobalt

in 100ml.

15

RESULTS AND DISCUSSIONCHAPTER 4

Assessment of Optimum Experimental Conditions for the Spectrophotometric

Determination of Cobalt with Chloroindazon DS

(1) Effect of pH on Chloroindazon DS

The absorption spectra of Chloroindazon DS (0.01% solution)

at four different pH values were recorded and shown in Fig. 4-1.

It can be seen that these spectra depend markeclly on pH, and the

absorption maximum is shifted to longer wavelengths at higher pH.

When the pH is at 5.82 (Fig. 4-1a), the absorption maximum is at

464nm with a shoulder at 420nm. At pH 11.5 (Fig. 4-1b), the absorption

maximum is shifted to 500 n.. with a shoulder at 463 nm. At pH 12.5

(Fig. 4-1c), the absorption maximum is at 514 nm, and finally at

pH close to 13 (Fig. 4-1d), the absorption maximum is at 526 nm.

No matter at what pH, Chloroindazon DS has strong absorption in

the wavelength region from 340.nm to 600 nm, and absorbs relatively

weakly beyond 600 nm.

(2) Optimum wavelength for analysis

Two solutions were prepared to contain 1p.p.m. of cobalt,

a suitable amount of Chloroindazon DS, and at phi 6.45 and 8.71

respectively. The absorption spectra of these two solutions were

recorded and shown in Fig. 4-2 and Fig. 4-3 respectively. The

cobalt complex was found to be green and have two absorption

maxima at both pH. At pH 6.45, the maxima were at 638 nm and

418 nm respectively, while at pH.8.71, they were at 638 nm and

405 nm respectively.

16

2.0

Absorbance

Fiq. 4-1 The visible absorption spectrum of 0.01% Chloroindazon DS solution

at different pH: (a) pH 5.62, max at 464 nm (b) pH 11.5, max at 500 nm

(c) pH 12.5, max at 514 nm (d) pH 12.95, max at 526 nm

(a)

(c)(b)

1.0

(d)

0

600 700400 500340

Wavelength, nm

17

1.0

Absorbance

Fig. 4-2 The visible absorption spectrum of the cobalt complex of Chloroindazon D

at pH 6.45 : lp.p.m. cobalt

max at 638 nm and 418 nm

0.5

O

600 700400 500340

Wavelength, nm

18

1.0

Absorbance

Fic. 4-3 The visible absorption spectrum of the cobalt complex of Chloroindazon DS

at pH 8.71: 1p.p.m. cobalt

max at 638 nm and 405 nm

0.5

O

500400340 600 700

Wavelength, nm

19

We can say that the wavelength of the absorption maxima change

only slightly with pH. Since the reagent absorbs quite strongly

around 400 nm but rather weakly beyond 600 nm, the obvious choice

of the optimum wavelength for the analysis of cobalt is at 638 nm.

(3) Effect of excess reagent

Since Chloroindazon DS still has a small absorption beyond

600 nm, it is necessary to find out whether excess Chloroindazon DS

interferes with the spectrophotometric determination of cobalt

at 638 nm. Therefore, solutions containing a fixed amount of

cobalt (1p.p.m.) but various amount of Chloroindazon DS at pH 10.3

were examined at 638 nm. The result obtained were tabulated in

Table 4-1, and a plot of absorbance versus reagent concentration

was constructed and shown in Fig. 4-4.

Table 4-1

Effect of reagent concentration on the determination of cobalt

at room temperature

Volume (ml) of 0.01% Chloroindazon DS

Absorbance atadded to 100ug cobalt

(Total volume = 100ml, pH= 10.30) 638 nm

10 0.128

20 0.265

0.39730

0.45735

40 0.471

45 0.478

0.48050

0.47970

80 0.487

85 0.485

20

Absorbance

Fie. 4.4 The plot of absorbance at 637 nm of solutions containing 1 p.p.m.at 638 nm

of cobalt at pH 10.3 versus concentration of 0.01% Chloroindazon DS

0.5

0.4

0.3

0.2

0.1

3010 700 20 40 60 8550 80

volume of 0.01% Chloroindazon DS, ml

21

Fig.4r.4shows that 45 ml of. 0.01% Chloroindazon DS is sufficient

to complex up to 100ug of cobalt. With excess reagent, the

absorbance was found to remain essentially constant. Therefore,

the reagent does not interfere with the determination of cobalt.

when it is less than one fold in excess. Therefore, Chloroindazon DS

can be used as a complexing agent for the spectrophotometric

determination of cobalt.

(4) Rate of color development, effect of temperature and stability

of the colour

When the reagent was added to a cobalt solution, the green

colour developed within 1-2 minutes at room temperature, but the

reaction required a much longer time for completion. However,

raising the reaction temperature to 60°C could reduce the time of

complete colour development to twenty minutes (see Table 4-2).

The colour remained stable for at least one week.

Table 4-2

Effect of temperature on the rate of reaction between 100ug of

cobalt and 45 ml of 0.01% Chloroindazon DS at pH 10.65 in a

total volume of 100 ml

Absorbance at 638 nm

Time (minutes)

room temperature 60°C 100°C

5 0.501

10 0.391 0.489 0.500

20 0.420 0.499 0.501

0.430 0.50130

80 0.443

overnight 0.499

22

C5) Optimum pH for colour development

The effect'of pH on the colour development was studied by

making absorbance measurements at'638 nm on a series of solutions

with fixed cobalt and reagent concentrations but with pH ranging

from 3.60 to 11.55. The results obtained were tabulated in

Table4-3,and the plot of absorbance versus pH is shown in Fig.4-5.

It can be seen from Fig.4-5that the absorbance increases

steeply in the pH range of 4.5-6, and much less so between pH 9-11.

The absorbance had a maximum value at pH 6.5 and 11.2. The most

suitable pH for the spectrophotometric determination of cobalt

with Chloroindazon DS was suggested to be 6.5. Because at this

pH, the Chloroindazon DS had small absorbance at 638 nm and the

cobalt complex had a maximum absorbance value. On the other hand,

other metal ions, such as zinc, cadmium and mercury did not form

complex with Chloroindazon DS at this pH.

When the pH was below 4.6, very little cobalt complex was

formed as evidenced by the spectra shown in Fig.4-6and Fig. 4-70

However, it is worthy to note-that if the green cobalt complex

was allowed to be formed before the pH was adjusted, the colour

was found to be stable at.all pH.

23

Absorbance atFig. 4-5 Effect of pH on the completeness of colour formation

638 nmof cobalt with Chloroindazon DS : 1 p.p.m. cobalt

0.6

0.5

0.4

0.3

0.2

0.1

0 123 4 5 106 117 9

pH

8

24

1.0

Absorbance

Fig. 4-6 The visible absorption spectrum of solution containing 1 p.p.m. cobalt

mixed with 4.5 ml 0.1% Chloroindazon DS at pH 4.b0

0.5

0

400340 500 600 700

Wavelength, nm

25

1.0

Absorbance

Fig. 4-7 ThP vsible absorption srectrum of solution containing 1 p.p.m. cobalt

mixed with 4.5 ml 0.1% Chloroindazon DS at ph 3.60

0.5

0

400 500340 600 700

Wavelength, nm

26

Table 4-3

Effect of pH on the absorbance of solutions containing 100ug of

cobalt and 45 ml of 0.01% Chloroindazon DS in a total volume of

100 ml which were heated to 60°C and maintained for twenty

minutes, then cooled to room temperature

Absorbance at 638 nmpH

0.0213.60

4.60 0.114

0.5245.82

6.50 0.552

0.5327.01

0.5277.37

0.5107.70

8.71 0.483

0.4879.64

10.48 0.482

10.90 0.531

11.45 0.530

11.55 0.500

(6) Order of addition

It is immaterial whether to add the reagent before or after

pH adjustment with the potassium dihydrogen phosphote - sodium

hydroxide buffer, especially when the reaction mixture Tas heated

to 60°C for twenty minutes.

27

C7) Effect of foreign ions

The criterion for an interference was an absorbance varying

±5% from the expected value. In order to determine the effect

of the foreign ions that might possibly be encountered in the

determination of cobalt, solutions were prepared to contain 10ml

of 10p.p.m. cobalt, 45 ml of 0.01% Chloroindazon DS solution and

varying concentrations of each ion to be tested in a total volume

of 100 ml and at pH 6.5. These solutions were analysed for cobalt

following the recommended procedure. A large number of cations

including copper (II), nickel (II), zinc (II), cadmium (II),

mercury (II), calcium (II), magnesium (II), iron (III), iron (II),

aluminum (III), and chromium (III), and several anions including

tartrate, citrate, cyanide and EDTA were examined. Results of

the interference studies are given in Table4-4 and 4-5.

It is evident from Table4-4 that there was no interference

from 500 ug of most of the cations examined on the determination

of 100 ug of cobalt in 100 ml. Only copper, nickel and iron (II)

were found to cause interference. However, interference from

nickel can be eliminated by the addition of 0.1 gm of sodium

citrate to the solution. Since iron (III) does not interfere

with the determination, interference from iron (II) can be

eliminated by prior oxidation to iron (III). Copper in quantities

up to 50 ug can be tolerated.

At high pH, zinc, cadmium and mercury also form complexes

with Chloroindazon DS, which seriously interfere with the deter-

mination of cobalt. However, formation of these complex were

found to be prevented when the pH of solution was kept at 6.50,

so that at this pH interferences from these ions are practically

eliminated.

28

Table 4-4

Effect of foreign cations on the determination of 100 ug of cobalt,

at pH 6.5 (total volume = 100 ml)

Metal % errorAmount added (ug) Cobalt found (ug)

Cu+2 -1.3250 98.68

200 110.18c +10.18

108.65a +8.65500

108.65a,b +8.65500

Ni+2 -16.7683.24a500

101.35b +1.35500

Zn+2 -2.43500 97.57

Cd+2 100,72500 +0.72

+2Hg 104.72 +4.72500

Ca +2100.18500 +0.18

Mg+2 100.00500 0

Fe+2 113.16 +13.16500

Fe+3 101.99500 +1.99

Al+3 101.09 +1.09500

0Cr+3 100.00500

a: the shape of the curve was changed.

b: 0.1 gm of sodium citrate was added to the solution before

addition of the buffer and reagent.

c: 0.1 gm of Potassium Sodium (+) tartrate was added to the

solution before addition of the buffer and reagent.

29

Results in Table4-5 show that tartrate and citrate did not

interfere with the determination, so that they could be used as

masking agents for the determination of cobalt with Chloroindazon DS.

However, serious interference was caused by small amount of cyanide

or EDTA. Interference from these ions can be eliminated by prior

seperation with anion exchange resins.

Table 4-5

Effect of anions on the determination of 100.0 ug of cobalt at

pH 6.50 (total volume = 100 ml)

Salt Amount added (m) % errorCobalt found (ug)

Potassium

-0.80.452 99.2

Sodium (+)

1.131 -1.498.6tartrate

Sodium 0.577 -1.198.9

citrate 1.444 -1.498.6

Na2EDTA 0.568

Potassium

cyanide 0.50

30

2. Beer's law, Molar absorptivity and Sensitivity

A series of solutions were prepared to contain varying amounts

of cobalt and enough of 0.01% Chloroindazon DS solution in a volume

of 100 ml and at pH 6.50 (for,0.2-1.0p.p.m. cobalt) and pH 10.42

(for 1-6p.p.m. cobalt), and their absorbance were measured at

638 nm to test whether Beer's law was obeyed. The results obtained

are tabulated in Table4-6 and4-7,from which the Beer's law plots

were constructed and shown in Fig. 4-8 and Fig. 4-9 respectively.

From these two figures it can be seen that Beer's law was obeyed

over the ranges from 0-1p.p.m. and from 1-5 p.p.m. of cobalt.

Thus, effectively the linear range is from 0-5 p.p.m.

The molar absorptivity for the cobalt complex at 638 nm

was calculated from data in-Table4-6 using the equation

Table 4.6

Absorbance at 638 nm of a series of cobalt solutions at pH 6.50

containing enough of reagent for complex formation.

Absorbance at 638nmCobalt concentration (p.p.m.)

0.2 0.109

0.4 0.221

0.6 0.330

0.8 0.441

1.0 0.552

31

Fig. 4-8 The plot of absorbance at 638 nm versus

0.6concentration of cobalt at pH = 6.50

(Beer's law plot

0.5

0.4

0.3

0.2

0.1

0.4 0.60.2 0.8 1.00

concentration of cobalt (p.p.m.)

Absorbanceat638nm

32

Fig. 4-9 The plot of adsoroance at 638 nm versusAbsorhance

concentration of cobalt at pH 10.42at 638 nm

(Beer's law plot)

3.0

2.0

1.0

0 6421 3

Concentration of cobalt (p.p.m.)

5

33

Table 4-7

Absorbance at 638 nm of a series of cobalt solutions at pH 10.42

containing enough of reagent for complex formation

Cobalt concentration (p.p.m.) Absorbance at 638 nm

1 0.483

0.9902

1.4693

1.949

5 2.427

2.8446

The sensitivity of a colour reaction, according to Sandell9,

represents the number of microgram of an element, converted to

the colored product, which in.a column of solution having a cross

section of 1 cm2 shows an extinction 0.001 (i.e. absorbance = 0.001).

Expressed in terms of an element, the senstivity is E, where e

is the molar absorptivity of the coloured product, and M its atomic

weight, n is the number of atoms of the element in a molecule of

the compound. Therefore, the Sandell sensitivity for the reaction

between cobalt and Chloroindazon DS was found to be 0.00181 ug/cm2.

4

34

3. Nature of the cobalt complex

The cobalt : ligand ratio at pH 6.50 was measured by the

continuous variation method 10,11 using three different wavelengths.

The results obtained are collected in Table4-8 and the plots of

absorbance against the mole fraction of ligand are shown in Fig.4-10.

Table 4-8

Data for the determination of cobalt to ligand ratio: 1.519x10-4M

Co (II) and pH 6.50

Mole fraction Absorbance at Absorbance at Absorbance at

of ligand 638 nm 660 nm 680 nm

0.2 0.259 0.244 0.184

0.3 0.410 0.373 0.260

0.4 0.549 0.501 0.340

0.5 00697 0.634 0.414

0.6 0.839 0.751 0.480

0.7 0.982 0.844 0.529

0.8 0.854 0.715 0.447

0. 0.454 0.368 0.230

From Fig.4-10,maximum absorbance was found to occur at the mole

fraction of ligand equal to 0.7575, which was independent of

wavelength used for measurement. Therefore, the mole ratio plot

showed the formation of a 1:3 complex.

35

Absorbance

Fig. 4-10 The plot of ahsorhance versus the mole fraction

of Chloroindazon DS at pH 6.50

1.0

0.9

0.8

638 nm

0.7

0.6

660 nm

0.4

680 nm0.3

0.2

0.1

0.50.4 0.6 0.90 0.1 0.30.2 0.7 0.75175 0.8 1.0

mole fraction of Chloroindazon DS

0.5

36

The structure of cobalt-Chloroin.dazon DS complex can he deduced

from its composition. According to the metal to lizand ratio

is 1:3, the following figure shows the suggested structure of

the complex. Bepause of steric interference, the nitrogen-

nitrogen double bond of Chloroindazon DS must be a trans double

bond. Cobalt forms an octahedral complex with Chloroindazon DS

as ligand. The most probable donar atoms are the hydroxy oxygen

and the diazo-nitrogen atom farther away from the hydroxy oxygen

such that six-membered rings are formed.

H

NCI

N=NSO3

Co

O

SO3

N

37

k. Precision

The precision of the procedure was checked by measuring the

absorbance of seven samples, each of which contained 1.00 p.p.m.

of cobalt. The results obtained are collected in Table 4-9.

Table 4-9

Absorbance of the cobalt complex with Chloroindazon DS measured

at 638 nm: 1p.p.m. cobalt and pH 6.50

Cobalt added Absorbance at Cobalt found

(p.p.m.) 638 nm (p.p.m.)

1.00 0.555 1.005

1.00 0.555 1.005

1.00 0.551 1.000

1.00 0.551 1.000

1.00 0.5+9 0.993

1.00 005+$ 0.991

1.00 0.552 1.000

Mean= 0.999

Standard Deviation= S_-g x 1n-3

The standard deviation was found to be+ 0.539 per cent at the

1 p.p.m. of cobalt level.

38

5. Determination the Cobalt Content in Tap water

Into 100 ml volumetric flask, 25 ml of tap water were added

followed by the addition of known amount of cobalt, 30 ml of

buffer solution (pH = 6.50), 0.1 g sodium citrate and a suitable

amount of 0.1% Chloroindazon DS solution. The solution was

heated to 60°C and maintained at this temperature for 20 minutes.

After cooling to room temperature, the solution was diluted to

the mark with distilled water. The absorbance at 638 nm was

measured against distilled water as a blank, however, when no

cobalt was added, the absorbance was measured against the reagent

blank. The results obtained are shown in Table4-10 and a plot of

absorbance versus the amount of cobalt added in Fig 4-11.

Table 4-10

Determination of Cobalt in Tap water

Cobalt added Absorbance at Cobalt found (p.p.m.)

Sample

638 nm Total(p.p.m.) Tap water

Untreated 0 0 0.0 0.0

0 0 0.0 0.0

Treated 0.2 0.102 0.20 0.00

0.4 0.203 0.40 0.00

0.6 0.307 0.60 0.00

0.8 0.406 0.80 0.00

1.0 0.508 1.00 0.00

Average of three measurements

39

Absrbance at

638nm

0.5

Fig. 4-11 The plot of absorbance at 638 um

versus the amount of cobalt

added in tap water

0.4

0.3

0.2

0.1

0 0.2 0.4 0.6 1.0Concentration of cobalt (p.p.m.)

0.8

40

Results in Tablek-10 show that no cobalt was detected in the

untreated tap water samples, and this is in agreement with the

result obtained by the method of standard addition. Moreover,

recovery of cobalt from spiked samples proved to be quantitative.

Thus, the determination of cobalt using Chloroindazon DS was

proved to be reliable.

6. Color Reaction with Other Metal Ions

In addition to form a green complex with cobalt (II),

Chloroindazon DS also forms coloured complex with nickel (II),

copper (II), zinc (II), cadmium (II) and mercury (I) and mercury (II).

The absorption spectra of these complexes are shown in Fig 4-12 to 4-17

and their spectrophotometric data are summarized in Table 4-11.

Table 4-11

Spectrophotometric data for the Chloroindazon DS complexes of

Ni (II), Cu (II), Zn (II), Cd (II), Hg (I) and Hg (II) at pH 10.60

Approximate Molar Absorptivity

Metal Colour max (nm)

(1-mol-1-cm-1)

Nickel Violet 2.77 x 104558

2.35 x 104594

Copper Blue 1.78 x 104578

Zinc Violet 3.19 x 104545

Cadmium Magenta 3.95 x 104546

Mercury(I) Red-violet 8.12 x 104538

and (II)

41

1.0

Absorbance

Fig. 4-12 The visible absorption spectrum of the Ni(II) complex-of Chloroindazon DS

at pH 10.60 : 1 p.p.m. nickel

at 558 nm and 594 nm

0.5

0

600 700500400340

Wavelength,nm

max

0.4

Absirbance

0.2

0

340 400 500 600 700

Wavelength,nm

Fig. 4-13 The visible absorption spectrum of the Cu(II) complex of Chloroindazon DS at

pH 10.60: 1 p.p.m. copper

at 578 nmLmax

0.4

Abscrbance

0.2

0

340 400 500 600 700

Wavelength, nm

at 55 nmmax

Fig. k-1k The visible absorption spectrum of the Zn(II) complex of Chloroindazon DS

at pH 10.60: 1 p.p.m. zinc

n.d

AhsorbanceFie: f-15 The visible absorption spectrum of the Cd(Il) complex of Chloroindazon DS

at pH 10.60: 1 p.p.m. cadmium

0.2

o

340 400 50C 600 700

Wavelength, nm

at 5+6 nmmax

1.0

Absorbance Fig. -16 The visible absorption spectrum of the Hg(l) complex of Chloroindazon DS

at oH 10.60 2 1 p.p.m. 'mercury

at 538 nmmax

0.5

0

340 400 500 600 700

Wavelength, nm

1.C

Absorbance

0.5

0

340 400 500 600 700

Wavelength nm

Fie: f-17 The visible absorption spectrum of the Hg(II) complex of Chloroindazon DS

at pH 10.60: 1 p.p.m. mercury

a at 538 am1 max

From Tabled-11, it can be seen that all the absorption maxima for

these complexes were below 600 nm, where Chloroindazon DS absorbs

strongly. It is expected that Chloroindazon DS itself will

interfere with the determination of these metals, and the molar

absorptivities quoted in Tabled-11 could only be considered to

be approximate. Thus, Chloroindazon DS cannot be a good%

spectrophotometric reagent for these metals. Nevertheless,

qualitive analysis of these metals with Chloroindazon DS is still

a sensitive method.

A closer examination of the absorption maxima reveals that

two of them have values close to 600 nm». They belong to nickel

(59 nm) and copper (578 nm). If care is taken not to add too

much excess of the reagent, Chloroindazon DS can still be used

to determine copper and nickel, since the interference from the

reagent at 59 nm and 578 nm is comparatively smaller. In fact,

Diet er Molch has used Chloroindazon DS as a spectrophotometric

reagent to determine nickel in the presence of chromium in

chromium-nickel thin layers, and to determine copper in coating

on quartz crystal oscillators,

Conclusion

The present work shows that Chloroindazon DS is a sensitive

and selective reagent for cobalt. The optimum wavelength for

measurements was found to be at 638 nm, where the reagent absorbs

very weakly so that no interference was caused by the reagent

when it was present in less than one fold in excess. The optimum

pH used is suggested to be 6.30 because at this pH the cobalt

complex has a maximum absorbance whereas the reagent has small

absorbance at 638 nm, and some interfering ions such as zinc,

cadmium and mercury do not form complex with the reagent at this

pH. The reagent can be quite easily synthesized by a simple

coupling reaction using chemicals which are commercially available.

Many reagents for the spectrophotometric determination of

cobalt have been reported and reviewed14-17

Some of them are

compared with Chloroindazon DS in Table 4—12.

From an examination of many references, the molar absorptivities

of the cobalt complexes ranged from about 1x10 to 3.6x10 1-mol -cm,

corresponding to Sandell sensitivities (A=0.001) of about 0.06 to

20.0016 pgcm• Investigation of the molar absorptivity or Sandell

sensitivity of the reagents listed in Table4—12,it can be seen

that most of them are sensitive reagents for cobalt. The sensitivity

and selectivity of Chloroindazon DS compare favourably with other

reagents listed in Table4-12. The molar absorptivity of its cobalt

complex is only smaller than that of 3- (3,3-dichloro-2-pyridyl)azc

2,4—diaminotoluene, and is comparable to these of

3- (5 -chloro-2-pyridyl) azc -2,6-diaminopyridine and quinoxaline

2,3-dithiol

Table 4-12

Cnmnarision of Chloroindazon DS with some other spectrophotometric reagents for cobalt

Wavelength Molar absorptivity Sandell sensitivity

ReagentInterferences Refs.

Dimedone dioxime

3- 5-chloro-2-pyridyl)

azoJ-2,6-diaminopyridin

3-hydroxyp.ico line-

aldehyde

Ferrozine

400

620

543

370

450-520

42- 10 x 10

43. 60 X 10

3.04 x 10

2.66 x 10

4.64 x 103

5.53 X 104

0.0023

0.0016

0.01 3

B- Dithionaphtholic acid 335

5-f (3 95-dichloro-2-pyridyl)

azoj-2,4-diaminotoluene

Quinoxaline-2,3-Oithiol

590

472

1 -(2-pyridylazo)-2-naphthol 620

2,4,6,-Tris(2'-pyridyl)-

S-Triazine

485

Pyridine-2-aldehyde-2-

quinolylhydrazone

Chloroindazon DS

510

638

1 .38 x 1 03

3.56 x 0r

41 .90 x 10

2.80 x 103

43.00 x 10

43.25 X 10

0.00042

0.0017

0.0021

0-0018

If- interference from these ions can be easily eliminated

However, most of the reagent listed in Tablek-12 except ferrozine

are soluble in ethanol or DMF and insoluble in water. Ferrozine,

though soluble in water, forms a cobalt complex with smaller

absorptivity than that of Chloroindazon DS. Although nickel,

copper and iron (II) which often interfere with the determination

of cobalt are no exception in the case with Chloroindazon DS,

yet nickel can be masked by citrate-, and iron (II) can be oxidized

to iron (III), which does not interfere. Therefore, Chloroindazon DS

is, in effect, both a sensitive and selective reagent for cobalt.

In 1975? Molch published a paper on the spectrophotometry

determination of cobalt with Chloroindazon DS, where measurements

were, made at 6k0 nm and the molar absorptivity of the cobalt

complex was found to be 2055 x 10 1-mol -cm at 6k0 nm.

An ammonia chloride-ammonia buffer (pHlO) was used. By using

the Jobs method, the cobalt-Chloroindazon DS complex was found

to be a 1:2 complex at pHlO. However, in this work, the mole

ratio plot showed the formation of a 1:3 complex at pH 6.50. The

discrepancy in results might arise from the different pH values

used. Our experiments show that pHlO is not suitable for the

determination of cobalt, since the absorbance of the cobalt

complex at pH 10.0 is less than that at pH 6.50. This explains

why the molar absorptivity determined by Molch at 6k0 nm is less

than what we found at 638 nm, and in this case no significant

difference should have arisen in using slightly different wave¬

lengths. Futhermore, as mentioned, previously, zinc, cadmium and

mercury will form complex with the reagent at pH 10.0 and will

interfere.

51

Besides cobalt, Chloroindazon DS may be used, if desired,

to determine copper and nickel spectrophotometrically, provided

no excess reagent should be used.

52

REFERENCES

(1) J.L. Bahamonde, D.P. Bendito and F. Pino, Analyst, 1974, 99,

355-359•

(2) J.N. Srivastava and R.P. Singh, Talanta, 1973, 20, 1210-1213.

(3) M.H. Hashmi, A. Rashid, M. Umar and F. Azam, Anal. Chem.,

1966, 38(3), 439-441

(4) A. Garcia De Torres, M.'Valcarcel and F. Pino-Perez,

Analytica Chim. Acta, 1974, 68(2), 466-469.

(5) R. Belcher, S.A. Ghonaim and A. Townshend, Talanta, 1974,

21(3), 191-198.

(6) S. Shibata, M. Furukawa and E. Kamata, Anal. Chim. Acta,

1974, 73(1), 107-119.

(7) S. Shibata, M. Furukawa and K. Goto, Talanta, 1973, 20,

426-430.

(8) H. Schweppe, Fresenius' Z. Anal. Chem., 1969, 244(5), 312-314.

(9) E.B. Sandell, "Colorimetric Metal Analysis, 3rd edition,

Interscience, 1959, 83-84.

(10) W.C. Vosburgh and G.R. Cooper, J. Am. Chem. Soc., 1941, 63,

437-442.

(11) R.K. Gould and W.C. Vosburgh, J. Am. Chem. Soc., 1942, 64,

1630-1634.

(12) D. Molch and H. Koenig, Z. Chem., 1974, 14(10), 408-410.

(13) D. Molch, H. Konig and E. Than, Z. Chem., 1974, 14(9),

369-3.70.

(14) E.B. Sandell, "Colorimetric Determination of Traces of Metals

3rd edition. Interscience,'1958, 409-429.

53

15) D.F. Boltz and M.G. Mellon. Anal. Chem.

(a) 1968, 40, 255R-2738.

(b) 1970, 42, 1.52R-168R.

(c) 1976, 48, 216R-232R.

(16) W.J. Williams, Talanta, 1958, 1, 88.

(17) K. Toei and S. Motomizu, Analyst, 1976, 101, No. 1204,

497-511.

(18) S.K. Kundra, M. Katyal and R.P. Singh, Anal. Chem.,

1974, 46(11), 1605-1606.

(19) H. Gorniak and B. Janik, Fresenius' Z. Anal. Chem., 1975,

273(2), 127.

(20) J.A.W. Dalziel and A.K. Slawinski, Talanta, 1968, 15(4),

367-372.

(21) S.P. Singhal and D.E. Ryan, Anal. Chim. Acta, 1967, 370),

91-96.

(22) Hiroto Watanabe, Talanta, 1974, 21(4), 295-302.

(23) M.J. Janmohamed and G.H. Ayres, Analyt. Chem, 1972, 44(14),

2263-2268.

(24) D. Molch, H. Koenig and E. Than, Z. Chemie, Lpz., 1975,

15(9), 361-362.

PART III

SPECTROPHOTOMETRIC DETERMINATION OF METALS

WITH-CHLOROINDAZON C

54

RESEARCH PLANCHAPTER 5

A search of the literature reveals that not many colorimetric

reagents for the determination of calcium are available. Among

them, chloranilate1 is only suitable for milligram amounts of

calcium. The murexide2 method is applicable in the concentration

range of 1 to 3p.p.m. of calcium, but a high concentration of

the reagent must be used to ensure quantitative formation of the

calcium complex. Unfortunately, the reagent itself is unstable

and about 50% decomposition occurs over four hours at room

temperature at the pH of determination. Practically all heavy

metals interfere. The method using 2-chloro-5-cyano-3,6-dihydroxy-

benzoquinone3 (HDDQ) as a reagent is an indirect one, and it is

dependent on an initial complete seperation of the calcium complex

involving a filtration step. Calcichrome4 is subjected to an

electrolyte effect, which reduces the sensitivity by 45 to 50%.

The absorption maximum of Chlorophosphonazo II15 and that of

its calcium complex are not well seperated so that interference

is caused by the reagent itself. Other reagents, such as

2,3,4-Trihydroxyacetophenone6 have small molar absorptivities

which are in the order of 103. Glyoxal-bis(2-hydroxyanil)7 and

its derivates8'9, seem to be the best reagent for calcium up to

now, and they have high molar absorptivity, but unfortunately,

the reagents and its calcium complexes are very unstable. In

short, the majority of the reagents mentioned above are subjected

55

to interferences from other ions, or the calcium complexes or

the reagents themselves are rather unstable, and the sensitivity

is when compared with those obtained from the spectrophotometric

determination cif other metals.

In 1969, H. Schweppe10 synthesized a compound 1-61-Chloroindazole-

3 -ylazo-2-hydroxynaphthalene-3-carboxylic acid (Chloroindazon C)

with the following structure:

N N

CI NH0 COOH

H

It formed colored complexes with many metals, namely, cobalt

(green), nickel (violet), zinc (violet), cadmium (maganta),

copper (blue), mercury (violet), calcium (red-violet), stronium

(orange), magnesium (orange-red)-and aluminum (orange).

Preliminary investigations show that this reagent is very

promising for the determination of calcium.

It is the object of this work to investigate the possibility

of developing Chloroindazon C into a specific and sensitive

reagent to determine calcium spectrophotometrically. The optimum

experimental conditions for the determination of calcium with

this reagent, the sensitivity of the method and the effect of

foreign ions will all be studied. The possibility of using

Chloroindazon C to determine other metal ions will also be investigated.

N

56

EXPERIMENTALCHAPTER 6

1. Preparation of'Chloroindazon C

(1) Diazotization

H. Schweppe10 suggested a method to prepare this reagent by

dissolving 16.7 gm of 3-amino-6-chlorindazol in 300m1 1M

hydrochloric acid. Small pieces of ice (250gm) were added to

cool down the solution to about 3°C. Then 40 ml of 20% aqueous

sodium nitrite were added to the solution slowly. The diazonium

chloride precipitated immediately.

(2) Coupling

2-Hydroxynaphthalene-3-carboxylic acid (18.8gm) was dissolved

in 300 ml of IN sodium hydroxide. The solution was cooled down

to 5°°C by adding pieces of ice. Then. this solution was added to

the suspension of diazonium chloride with stirring. The pH of

the reaction mixture was adjusted. to about 7. The reagent in

the form of the sodium salt was precipitated out. It was

collected by suction and washed three times with 200m1 of

distilled water. Then it was dried at 100°C and ground into

powder.

2. Reagents

(1) Chloroindazon C solution. (0.01%)

Chloroindazon C (100mg) was dissolved in 5 ml of warm

dimethylformamide solution, and 45 ml of ammonia (density= 0.885)

were then added and diluted to one litre with ethanol. The

Freshly prepared solution was orange-red in colour, but on

57

standing at room temperature for two days, the solution turned

yellow. As the reagent solution is not stable at room temperature,

it/was stored in a refrigerator.

(2) Standard calcium solutions

Standard solutions of 30, 20, 10, 8, 6, 4, and 2 p.p.m.

were prepared by appropriate dilution of the stock solution of

100 p.p.m.9 which was previously prepared by dissolving 0.3944 gm

of calcium acetate (dried, BDH, A.R.) in distilled water and

diluting to 1 litre.

(3) Buffer solutions

The following buffer systems were prepared:

(a) Disodium hydrogen phosphate (0.01+6M)*- Sodium hydroxide

(0.0076M) (pH= 11.0)

(b) Sodium bicarbonate (0.035M)- Sodium hydroxide (0.032M)

(pH= 11.0)

(c) Ammonium chloride (0.05M) -Ammonium hydroxide (0.1M)

(pH= 11.0)

(d) Borax (0.017M)- Sodium hydroxide (0.033M) (pH= 10.80)

All buffer solutions were prepared from analytical grade

reagents.

The concentration inside the brackets after each compound

indicates its concentration in the resulting buffer solution

prepared according to Handbook of Chemistry and Physics,

56th edition, 1975-76, CRC Press, D134-135.

3. Apparatus

All the absorption spectra were recorded with a 323 Hitachi

Recording Spectrophotometer using matched 1-cm silica cells.

58

A Radiometer p}I meter Model 28 with a saturated calomel-glass

electrode system was used for all pH measurements.

1+., Recommended Procedure for the Determination of Calcium

A sample aliquot containing up to 300 y-g of calcium was

pipetted into a 100ml volumetric flask. This was followed by

the addition of 75ml 0.01% Chloroindazon C solution in ethyl

alcohol. About 1-2 ml of sodium hydroxide (0.IM) was used to

adjust the pH between 10.8 to 11.5, and the solution was finally

made up to the mark with distilled water. After mixing and

allowing the solution to stand for ten minutes to ensure complete

colour development, the absorbance was measured at 586 nm in

matched 1-cm silica cells against reagent blank. The concentration

of calcium could be deduced from a calibration graph prepared

in a similar manner from solutions containing 0-300 pg of

calcium in 100 ml

59

RESULTS AND DISCUSSIONCHAPTER 7

Assessment of Optimum Experimental Conditions for the

Spectrophotometric Determination of Calcium with Chloroindazon C

(1) Effect of pH on Chloroindazon C

The absorption spectra of Chloroindazon C (0.01%) at five

different pH values were recorded and shown in Fig. 7-1• It

can be seen that these spectra depend markedly on pH. Below

pH 4, the solution has a deep red colour. The absorption

maximum is at 532 nm. At pH 5, its colour is orange red and

at pH 6 to 11, the colour is pale yellowish orange, both of

which exhibit an absorption maximum at 420 nm. At pH between

11 to 13, the absorption maximum is shifted to 470 nm. Finally,

at pH above 13, it turns deep orange yellow, the maximum

absorption wavelength being at 438 nm. No matter at what pH,

Chloroindazon C has strong absorption in the wavelength range

from 340 to 580 nm, and absorbs relatively weakly beyond 580 nm.

(2) Optimum wavelength for analysis

A calcium solution (1p.p.m.) was prepared to contain

Chloroindazon C and at pH 11 and the absorption spectrum was

recorded as shown in Fig.?-2. The calcium complex is red-violet

in colour, and has two absorption maxima which are very close

to each other and at 560 nm and-586 nm respectively. Since

Chloroindazon C absorbs relatively strongly at 560 nm, but only

weakly at 586nm, therefore the-latter wavelength is chosen for

the spectrophotometric determination of calcium.

Absorbance

Fig. 7-1 The visible absorption spectrum of 0.01% Chloroindazon C solution at

different pH: (a) pH 30, Xmax at 532 nm (b) pH 5.°2, %max at 420 nm

(c) pH 8.0, Xmax at 420 nm (d) pH 11.0, Xmax at 470 nm (e) pH 13.95, max at 438 nm

340 400 5 0 0 600 700

Wavelength, nm

(b), (a),

(c)

(e)

(d)

1.

Aborbance

Fig. 7-2 The visible absorption spectrum of the calcium complex of Chloroindazon C

at pH 11.0: 1 p.p.m. calcium

at 586- nm and 50 nmmax

0.5

0

340 400 500 66O 700

Wavelength,nm

(3) Effect of excess reagent

Solutions containing a fixed concentration of calcium (Ip.p.m.)

but various amounts of Chloroindazon C at pH 10.80 were examined

at 386 nm. The results obtained were tabulated in Table 7-1 and

a plot of absorbance versus reagent concentration was constructed

and shown in Fig. 7-4

Table 7-1

Effect of reagent concentration on the determination of calcium

Volume (ml) of 0.01% Chloroindazon C

added to 100jig Calcium

(Total volume= 100ml, pH= 10.80)

Absorbance at 586 nm

5

10

15

20

25

30

35

4-0

50

60

70

80

0.14-8

0.292

0.4-37

0.4-63

0.4-63

0.4-68

0.4-70

0.4-68

0.4-72

0.473

0.4-78

0.4-78

Fig. 7-4- shows that 23 ml of the 0.01% Chloroindazon C solution

is sufficient to complex up to 100jig of calcium. With excess

reagent, the absorbance was found to remain essentially constant,

Therefore, Chloroindazon C has no interference on the

spectrophotometric determination of calcium when it is less than

one fold in excess.

Absorbance

Fig. 7-4 The plot of absorbance at 56 nra of solutions containing 1 p.p.m.

of calcium at pH 10.80 versus concentration of Chloroindazon C

Volume (ml) of 0.01% Chloroindazon C added

(4) Rate of colour development and stability of the colour

The rate of colour development is very high and the reaction

was found to be completed within ten minutes at room temperature

and pH 11 and the colour remained stable for st least ten hours

(see Table 7-2.)

Table 7-2

The rate of reaction between 100jig of calcium and 25 ml 0.01%

Chloroindazon C at pH 11 in a total volume of 100 ml

Time Absorbance at 586 nm

10 minutes

30 minutes

50 minutes

60 minutes

10 hours

0o469

0.470

0.469

0.468

0.469

(5) Optimum pH for colour development

Absorbance measurements were made at 586 nm on a.series of

solutions with fixed calcium and reagent concentrations but

with various pH values ranging from 3 to about 12. The results

obtained were tabulated in Table7-3, an( Fig.7-5is a plot of

absorbance against pH. Fig.7-5shows that the absorbance increased

steeply in the pH range of 9-10. The maximum absorbance occurred

at pH 10.8 to 11.5 When the pH was below 9» only a small amount

of complex was formed and it precipitated out quickly. At pH

between 2 to 6, the calcium complex was found to precipitate

out after mixing, whereas in the range of 7 to 9 the precipitate

Absorbance at

586 nm

-j

Fie. 7-5 Effect of pH on the completeness of colour formation of

calcium with Chloroindazon C: 1 p.p.m. calcium

o,T

04

o.z

0,1

0 i 6 7 j 3 ,0 II

appeared about ten to thirty minutes after mixing. At pH 12,

the complex would precipitate out about five minutes after

mixing. Therefore, the most suitable pH range for the

spectrophotometric determination of calcium with Chloroindazon C

is from 10.8 to 11.5.

Table 7-5

Effect of pH on the absorbance of solutions containing 100pg

of calcium and 25 ml of 0.01% Chloroindazon C in a total volume

of 100ml.

PH Absorbance at 586 nm

3.00

5.00

6.00

7.38

8o35

9o28

9.33

10.25

10.32

10.80

10.92

11.18

11.80

O.266

0.300

0.212

0.229

0.206

0.180

0.152

0.447

0.455

0.465

0.468

0.470

0.468

(6) Order of addition

When the sample solutions contain less than 100 ng of

calcium, the Chloroindazon C solution can be added before

67

dilution with distilled water. If the calcium concentration is

higher than 100 jig, it is better to add distilled water before

adding the Chloroindazon C solution. Otherwise, the calcium

complex was found to precipitate within a few minutes.

() Effect of ethanol

The reaction mixture should best contain one-third of

ethanol by volume, since ethanol concentration above 1+0% by

volume will cause precipitation of the calcium complex within

twenty minutes. If no ethanol is present in the reaction

mixture, the absorption maximum of the calcium complex is at

504 nm, where Chloroindazon C absorbs strongly. (see Fig. 7-3)

(8) Effect of Buffer solution

Many buffer solutions, such as the sodium hydroxide-sodium

bicarbonate buffer, sodium hydroxide-borax buffer, were found

to cause the calcium complex to precipitate within twenty minutes.

Futhermore, the order of addition also affected the complex

formation, as no complex was formed when the buffer solution was

added before the reagent. No suitable buffer solution could be

found.

(9) Effect of foreign ions

The criterion for an interference was an absorbance

varying +5% from the expected value. In order to determine

the effect of the foreign ions that might possibly be encountered

in the determination of calcium, solutions were prepared to

contain 10 ml of 10 p.p.m. calcium, 25 ml of 0.01% Chloroindazon C

solution and varying concentrations of each ion to be tested in

a total volume of 100 ml and at pH 11. These solutions were

analysed for calcium following the recommended procedure, and

the results obtained are shown in Tables 7-4 and 7-5.A

1.0

Abeorbance

0.5

0

340 400 500

Wavelength, nm

600 700

Fig. 7-3 The visible absorption spectrum of the calcium complex of Chloroindazon C

at tdH 10.80 in aqueous solution: 1 p.p.m. calcium

at nmmax

Table 7-f

Effect of foreign cations on the determination of 100.0 p.g of

calcium at pH 11.0

Metal Amount added Calcium found % error

500a

500a

500a

500a

500a

500a

100

500b

50

50b

100

100b

200

500

200

50

100

100°

200b

'50

100b

500b

99.44

96.92

101.40

99.72

96.49

98.25

107.94

99.53

105.70

103.40

108.70

102.10

101.10

precipitation of complex

102.89

93.93

82.01

89.58

61.28

96.21

89.53

no complex formed

-0.56

-3.08

+1.40

-0.28

-3.51

-1.75

+7.94

.-0.43

+5.70

+3.40

+8.70

+2.10

+ 1.10

+2.89

-6.07

-17.99

-10.42

-38,72

-3.79

-10.47

a: 1ml 1F KCN was addedbefore the addition of Cliloroinclazon C

b: 0.05gm tartrate was added before the addition of reagent,

then waited for about thirty minutes for colour development.

c: 500jig tartrate was added.

Table 7-S

Effect of anions on the determination of 100.0 jig of calcium

at pH 11.0 (total volume= 100ml)

Salt Amount added Calcium found % error

Na EDTA25 pg

50 jig

100 jig

0.02 gm

100 jig

300 jig

0.1 gm

100 jig

300 pg

500 jig

0.1 gm

Sodium

phosphate

Sodium

citrate

Potassium

Sodium(+]

tartrate

0.1 gm

93.42

94.06

90.66

no complex formation

97.66

98.09

no complex formation

100.42

97.02

96.18

no complex formatior

105.24

-6.58

-5.94

-9.34

-2.34

-1.91

+0.42

-2.98

-3.82

h3.24

It can be seen from Table7- that although nickel, copper,

zinc, cadmium, cobalt and mercury forms coloured complexes with

the reagent, they could be masked by potassium cyanide and five

fold excess of?these ions could be tolerated. Aluminum and

stronium also interfered with the calcium determination, however,

they could be masked by the addition of 0.05 g of tartrate to

the solution so that five fold excess of aluminum and one fold

excess of stronium could also be tolerated. There was no

interference from 200 jig of barium and iron (III) on the

determination of 100 jig of calcium. However, larger amount of

barium present caused the complex to precipitate. Serious

interference was caused by magnesium and chromium. However,

for chromium, the presence of 50 jig of this element could be

tolerated in the determination of 100 jig of calcium.

From Table7-5 it can be seen that serious interference was

caused by EDTA. There was no interference from 300 jig of sodium

phosphate and 500 jig of sodium citrate on the determination of

100 jig of calcium. However, large amount of phosphate and

citrate could prevent complex formation. Tartrate did not

interfere with the determination of calcium.

2. Beer's law, Molar absorptivity and Sensitivity

A series of solutions were prepared to contain varying

amount of calcium and enough of 0.01% Chloroindazon C solution

in a volume of 100 ml and at pH 11.0 and their absorbance were

measured at 586 nm to the test whether the color system conforms

to Beer's law. The results obtained are tabulated in Table 7-6,

and from these data a plot of absorbance versus concentration

was constructed as shown in Fig. 7-6, Beer's law was obeyed over

the range of 0.0 to 3.0 p.p.m. of calcium for 1-cm cells.

Using the absorbance data in Table 7-31 the molar absorptivity,

at 586 nm of the calcium-Chloroindazon C complex was calculated

to. be 1.89 x 10 1-mol -cm using the equation

This corresponds to a Sandell sensitivity, log

Table 7-6

Absorbance at 586 nm of a series of calcium solutions of pH 11.0

containing enough of reagent for complex formation.

Calcium (p.p.m.) Absorbance at 586 nm

0.2

0.4-

0.6

0.8

1.0

2.0

3.0

0.081

0.182

0.273

0.368

0.68

0.938

1.03

Absorbance

at 586 nm

Fig. 7-6 The plot of absorbance at 586 nm versus

concentration of calcium (Beer's law plot)

Concentration of calcium (p.p.m.)

3. Precision

The precision of the procedure was checked by measuring

the absorbance at 586 nm of eight samples containing 100 jig of

calcium, 25 ml of 0.01% Chloroindazon C in a total volume of

100 ml and at pH 11.0. The results are collected in Table 7-7.

Table 7-7

Determination of calcium in synthetic samples

Added (jig) Absorbance at 586 nm Found Ca (ug)

100

100

100

100

100

100

100

100

0.465

0.469

0.468

0.468

0.467

0.466

0.472

0.469

98

100

100

100

99

98

102

100

99.6

1.30

Mean=

Standard deviation=

The standard deviation was found to be

the 1 p.p.m. level of calcium.

1.30 per cent at

k. Color Reaction with other Metal Ions

In addition to forming a red-violet complex with calcium,

Chloroindazon C also form coloured complexes with zinc (II),

cadmium (II), cobalt (II), nickel (II), copper (II), mercury (II),

stronium (II), magnesium (II) and aluminum (ill). The spectra

of these complexes are shown in Fig. 7-7 to Fig. 7-15, and their

spectrophotometry data are summarized in Table 7-8.

Table 7-8

Spectrophotometric data for the Chloroindazon C complexes of

Zn (II), Cd (II), Hg.(ll), Co (II), Ni (II), Cu (II), Sr (II),

Mg (II) and A1 (III) at pH 11.0

Metal Colour Amax (nm)

Approximate molar absorptivity

Zinc

Cadmium maganta

Mercury violet

Cobalt green

Nickel violet

Copper blue

Stronium orange

Magnesium orange red

Alumium orange

552

580

52

556

644

54o

578

588

448

442

436

2.88 x let

2.73 x let

2.47 x let

Ll

6.57 X 10

2.35 x 10

3.46 x 10

3.05 x 10

1.51 x 10

2.94 x 10

9.77 x 105

41.09 x 10

Abanrbance.

Wavelength, nm

Fig. 7-7 The visible absorption spectrum of the Zn(ll) complex of Chloroindazon;C~at

Abs Drbance

Wavelength, nm

Fig. 7-8 The visible absorption spectrum of the Cd(ll) complex of Chljoroindazon p at

1.0

Absorbanpe

0.5

0

400 500 600 700

at 556 nmmax

Fie- 7_q The-visible!absorption spejctrum ;_of „the_Xg.CIl)._Pbniple3 ..of _Chlpxqih_daori_C_

at pH 11.0•: 1 p.p.m. mercury

1.0

Absorbance

0.5

0

340 400 500 600 700

Fig 7-10 The visible absorption spectrum of. the Co(II) complex of Chior.oindaz.oh

pH 11.0: 1 pp.m. cobalt

at 644 nmmax

Wavelength, nm

Fig. 7-11 The viaible anbaorption apectrum of the NiCllD cofflDlex f of ChXoroindazon;XT at

pH 11.0:1 p.p.m. nickel

max at 540 nm and 578 nm

Wauainnnt nm

0.4

Abstorbance

Fig. 7-12 The visible absorption spectrum of the Cu(II)I complex of Chloroindazon C at

oH ll.o:1 p.p.m.copper

at 588 nmmax.

0.2

0

340 400 500 600 700

Wavelength,nm

0.4

Absorbance

0.2

O

34C 400 500 6oo

Wavelength ,nm

Fig. 7-15 The visible absorption spectrum of the Sr(ll)1 complex of Chlorjsindaton C at

pH 11.0: 1 p.p.m. stronium

at 448 nmmax

A'hc:rir'h«nr t

0.2

r

L i—_cnn

Wavelength, nm

600 700

at 442 nmmax

14 The visible aborption spectrum of the %g(II) complex of Chloreindarom C at

pH 11.0 .1. pp.m. magnesium

1 .n

AUp n n q

0.5

C

dOO HOn

Wavelength, nm

600 70C

i at nmlmax

Fig. 7-15 The visible absorption spectrum of the Al(III) complex of Chloroindazon C

at pH 11.0: 1 p.p.m. aluminum

Many of these coloured metal complexes have their absorption

maxima below 580 nm so that Chloroindazon C cannot be used for

the spectrophotometry determination of these metals since the

reagent absorbs -jstrongly below 50 nm and will cause serious

interference. However, Chloroindazon C can serve as a sensitive

reagent for the qualitive analysis of these metals. On the other

hand, Chloroindazon C can be a good spectrophotometric reagent

for zinc, cobalt, nickel and copper.

86

Conclusion

The present work shows that Chloroindazon C is a sensitive

and rather selective reagent for calcium. The optimum wavelength

for measurements was found to be of 586 nm, where the reagent

has practically zero absorbance, so that no interference was

caused by the reagent when-it was present in less than one fold

in excess. The optimum pH lies in the range of. 10.8-11.5, which

can be easily adjusted by adding 1-2 ml of 0.114 sodium hydroxide,

and the desired accuracy was attainable without the use of buffer

solutions. The reagent can be quite easily synthesized by a

simple coupling reaction using chemicals which are commercially

available.

Chloroindazon C is better than many other spectrophotometric

reagents for calcium in that the absorption maximum for measurement

is well seperated from those of the reagent itself such that

Chloroindazon C will not interfere-with the determination. Many

reagents such as Chlorophosphonazo III', murexide2 and Calcichrorne4

absorb strongly at the absorption wavelength of their calcium

complexes. The rate of reaction between Chloroindazon C and

calcium is fast and the absorbance remained constant for at least

ten hours. Thus the stability of the calcium- Chloroindazon C

complex solution is much better than those of the calcium complexes

of murexide2, glyoxal bis-(2-hydroxyanil)7 and its derivates8'9.

A comparision of Chloroindazon C with some other reagents

is included in Table 7-9.

Tahl p 7 —Q

Comparision of Chlorindazon C with other reagents for calcium

Reagent Wavelength (nm)

Molar absorptivity

Interferences

7

2,3, --Tr ih.y4roxy-?

acetophenone

CChloroohosuhonazo III''

A TTT12,13Arsenazo III'

1 kAntipyrylazo III

2Murexide

Calcinhrnmfi

Crlvo-ya! hi—

(2-hydroxyanil)'

Chloroindazon C

540

669

667.5

600

650

605

506

510

615

533

586

3.35 x 103

6.4 x 10

l 1.46 x 10

k2 x 10

2.8 x 1(h

2.15 X 10

L1 x 10

4.5 x 103

7.6 x 103

1.5 X 10

1.89 x 10

Sr, Ba, Fe(lII)

Sr, Ba, Li

Fe(lll), A1

Sr, Ba, Mg

Mg, Sr, Ba, Li

Mg, ionic

strength affects

the absorbance

very much

Ba, Si

Mg, Cr

From Table7-9 the molar absorptivity of Chlorindazon C is

comparable to if not better than the other sensitive reagents for

calcium except Chlorophosphonazo III. However, Chlorophosphonazo II!

itself interferes with the determination of calcium. The selectivit;

is quite good for Chloroindazon C comparing with other reagents

listed in Table7-9i if the interference of the reagent is also

taken into account.

Besides calcium, the reagent can be developed into a

spectrophotometrie reagent for zinc, cobalt, nickel and copper.

88

REFERENCES

(1) R.E.U. Frost-Jones and J.T. Yardley, Analyst, 1952, 77, 468,

(2) Max B. Williams and James H. Moser, Anal. Chem., 1953, 25,

1414-1417.

(3) R.E. Rehwoldt, B.L. Chasen and J.B. Li, Analyt. Chem.,

1966, 38(8), 1018-10190

(4) M.H. Lancing and T.S. West, Analyt. Chem., 1963, 35,

2131-2135.

(5) J.W. Ferguson, J.J. Richard, J.W. O'Laughlin and C.V. Banks,

Anal, Chem., 1964, 36(4), 796-799.

(6) N.V. Deshpande and A.J. Mukhedkar, Microchem. J., 1975,

20(2), 165-172.

(7) J.R.W. Keer, Analyst, 1960, 85, 867-870.

(8) F. Lindstrom and C.W. Milligan, Anal. Chem., 1964, 36(7),

1334-1338.

(9) Carl W. Milligan and F. Lindstrom, Anal. Chem., 1972,

44(11), 1822-1829.

(10) H. Schweppe, Z. Analyt. Chem., 1969, 244(5), 310-3120

(11) E.B. Sandell, Colorimetric Metal Analysis, 3rd edition,

Interscience, 1959, 83-84.

(12) V. Michaylova and P. Ilkova, Analytica Chim. Acta, 1971,

53(1), 194-198.

(13) V. Michaylova and N. Kouleva, Talanta, 1974, 21(6), 523-532.

(14) B.W. Budesinsky, Analytica Chim. Acta, 1974, 71(2), 343-3470