ARTICLE IN PRESS

Water Research 37 (2003) 40384044

Removal of cadmium and nickel from wastewater using

bagasse y asha sugar industry waste
Vinod K. Guptaa,*, C.K. Jainb, Imran Alib, M. Sharmaa, V.K. Sainia
a

Department of Chemistry, Indian Institute of Technology, Roorkee, Roorkee 247 667, India
b
National Institute of Hydrology, Roorkee 247 667, India
Received 25 March 2003; accepted 6 May 2003

Abstract
The bagasse y ash, an industrial solid waste of sugar industry, was used for the removal of cadmium and nickel from
wastewater. As much as 90% removal of cadmium and nickel is possible in about 60 and 80 min, respectively, under the
batch test conditions. Effect of various operating variables, viz., solution pH, adsorbent dose, adsorbate concentration,
temperature, particle size, etc., on the removal of cadmium and nickel has been studied. Maximum adsorption of
cadmium and nickel occurred at a concentration of 14 and 12 mg l
1 and at a pH value of 6.0 and 6.5, respectively. A
dose of 10 g l
1 of adsorbent was sufcient for the optimum removal of both the metal ions. The material exhibits good
adsorption capacity and the adsorption data follow the Langmuir model better then the Freundlich model. The
adsorption of both the metal ions increased with increasing temperature indicating endothermic nature of the
adsorption process. Isotherms have been used to determine thermodynamic parameters of the process, viz., free energy
change, enthalpy change and entropy change.
r 2003 Elsevier Science Ltd. All rights reserved.
Keywords: Adsorption; Sugar industry; Bagasse y ash; Cadmium; Nickel

1. Introduction
The presence of trace metals in the aquatic environment has been of great concern because of their toxicity
and non-biodegradable nature [13]. Cadmium and
nickel are toxic and relatively widespread in the
environment [4]. These metals are used in a wide variety
of industries such as plating and cadmiumnickel
battery, phosphate fertilizers, mining, pigments, stabilizers and alloys [5], and nd their way to the aquatic
environment through wastewater discharges. Therefore,
a systematic study on the removal of cadmium and
nickel from wastewater is of considerable signicance
from an environmental point of view.

*Corresponding author. Tel.: +91-1332-274458; fax: +911332-273560.

E-mail address: vinodfcy@iitr.ernet.in (V.K. Gupta).

A number of methods are available for the removal of

metal ions from aqueous solutions. These are ion
exchange, solvent extraction, reverse osmosis, precipitation and adsorption. Activated carbon adsorption is a
well-known method for the removal of heavy metals [6
8], but the high cost of activated carbon restricts its
large-scale use for the abatement of heavy metal
pollution in developing countries.
A number of workers have used different adsorbent
systems, developed from various industrial waste materials, for the removal of heavy metals [916]. There still
exists a need to develop a low cost and efcient
adsorbent for the removal of cadmium and nickel from
wastewater.
The sugar industry is one of the most important agro
based industries in India. The bagasse y ash, an
industrial solid waste generated from the sugar industry,
causes a great disposal problem and it is currently being
used as a ller in building materials. Our earlier efforts

0043-1354/03/$ - see front matter r 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0043-1354(03)00292-6

ARTICLE IN PRESS
V.K. Gupta et al. / Water Research 37 (2003) 40384044

on the utilization of solid waste materials being

generated in some prime industries for the removal of
toxic metals have been very successful [1719]. In the
present work, an attempt has been made to develop an
inexpensive adsorbent system for the removal of
cadmium and nickel from wastewater using bagasse
y ash.

4039

different concentrations of metal ions. After prescribed

contact times, the solutions were ltered and the
concentrations of metal ions were determined by atomic
spectrometry. The effect of various interfering ions
(sodium, potassium, calcium, magnesium, chloride,
sulphate, nitrate, phosphate) have also been studied on
the adsorption of two metal ions on bagasse y ash.
2.3. Kinetic studies

2. Experimental methodology
All chemicals and reagents used were of analytical
grade and were obtained from E. Merck, India. Stock
solutions of cadmium and nickel were prepared using
cadmium nitrate and nickel nitrate in deionized water.
A pH meter (Hach, USA) was used for pH measurements. X-ray measurements were made using a Phillips
X-ray diffractometer employing nickel-ltered CuKa
radiations. The surface area of the adsorbent was
measured by a surface area analyser (Quantasorb Model
QS-7). IR spectra of the samples were recorded on an
infrared spectrophotometer (FTIR Perkin Elmer Model
1600). The porosity and density of the adsorbent were
determined by mercury porosimeter and specic gravity
bottles, respectively.
2.1. Adsorbent development
Baggase y ash, a solid waste material of the sugar
industry, was collected from a sugar factory at Iqbalpur,
UP, India. The material was treated with hydrogen
peroxide at 60 C for 24 h to oxidize the adhering organic
matter. It was then washed with de-ionized water, dried
at 100 C, powdered, and sieved to produce desired
particle size (100150, 150200 and 200250 mm). The
material was stored in a vacuum desiccator for further
use.
The stability of the adsorbent was determined by
keeping the material over night in different solvents
(water, dilute acids and bases of 0.10.2 N) and
determining the presence of its constituents in the
solvents. The loss of treatment was determined using
hydrogen peroxide. The constituents of the adsorbent
were determined following standard methods of chemical analysis [20].
2.2. Adsorption studies
Batch adsorption experiments were carried out in a
series of Erlenmeyer asks of 100 ml capacity covered
with Teon sheets to prevent contamination. The effect
of contact time (0150 min), concentration (2.0
20.0 mg l
1), solution pH (2.09.0), adsorbent dose
(2.020.0 g l
1), particle size (100150, 200250 and
300350 mm), and temperature (30 C, 40 C and 50 C)
were studied. Isotherms were obtained by adsorbing

A series of Erlenmeyer asks of 100 ml capacity

containing desired amount of metal ion solutions of
known concentrations were kept in a thermostatic water
bath shaker. A known amount of the adsorbent was added
to each ask and the asks were agitated. At prescribed
time intervals, the solutions were ltered and analysed for
metal ion concentration by atomic spectrometry.
2.4. Metal ion analysis
The concentrations of metal ions were determined by
atomic absorption spectrometer (Perkin Elmer Model
3110) using air acetylene ame. The metal standards
prepared were checked with standard reference material
obtained from the National Bureau of Standards (NBS),
USA before each metal analysis and the deviation found
was insignicant. Average values of ve replicates were
taken for each determination. Quantication of metals
was based upon calibration curves of standard solutions
of metals. The detection limits for cadmium and nickel
were 0.0005 and 0.004 mg/l, respectively. The precision
of the analytical procedures, expressed as the relative
standard deviation (rsd) ranged from 5% to 10%. The
precision for the analysis of standard solution was better
than 5%.

3. Results and discussion

The bagasse y ash was found to be stable in water,
dilute acids and bases. The composition of the adsorbent
was SiO260.5%; Al2O315.4%; CaO2.90%,
Fe2O34.90%, MgO0.81%. The loss on ignition
was found to be 16.0% by weight. The density and
porosity were found to be 1.01 g cm
3 and 0.36%
fraction, respectively. The presence of mullite, haematite, kaolinite, g-quartz, gamma alumina and geolite was
indicated by the X-ray diffraction pattern of the bagasse
y ash (Fig. 1). The surface area (BET) of the adsorbent
of particle size 200250 mm was found to be 450 m2 g
1
[21].
3.1. Effect of operating variables
The adsorption data for the uptake of cadmium and
nickel versus contact time for a xed adsorbent dose of

ARTICLE IN PRESS
4040

V.K. Gupta et al. / Water Research 37 (2003) 40384044

Fig. 1. X-ray diffraction for activated bagasse y ash.

Cd

Amount adsorbed, mg/g

10 g l
1 are shown in Fig. 2. The initial concentration of

14 and 12 mg l
1 for cadmium and nickel, respectively
were taken for the experiment. These plots indicate that
the remaining concentration of metal ions becomes
asymptotic to the time axis such that there is no
appreciable change in the remaining metal ion concentration after 60 min in the case of cadmium and 80 min
in the case of nickel. These represent the equilibrium
time at which an equilibrium metal ion concentration is
presumed to have been attained.
The plots showing the effect of initial concentration
on the adsorption of two metal ions are shown in Fig. 3.
These plots showed that the total metal ion adsorbed
increased sharply in the beginning and then slowly
towards the end of the run. A perusal of the results
clearly indicated that adsorption of cadmium increased
from 1 to 14 mg l
1 and that of nickel from 1 to
12 mg l
1, and then became constant indicating that the
maximum adsorption occurred at 14 and 12 mg l
1 for
cadmium and nickel, respectively.
The plots showing the effect of pH on the adsorption
of two metal ions are shown in Fig. 4. The initial
concentration of 14 and 12 mg l
1 for cadmium and
nickel, respectively, were taken for the experiment. It
was evident from the plots that the maximum uptake of
cadmium and nickel occurred at pH 6.0 and 6.5,
respectively. The pH of the solution was measured
before and after the adsorption and no change was
noticed in the pH. The adsorption increased from
0.70 mg g
1 at pH 2.0 to 1.20 mg g
1 at pH 6.0 in the
case of cadmium and from 0.24 mg g
1 at pH 2.0 to
about 1.00 mg g
1 at pH 6.5 in the case of nickel and
then started to decrease in both the cases. The decrease
in adsorption at higher pH may be attributed to the

Ni

1.60
1.20
0.80
0.40
0.00
0

30

60
90
Time, min

120

150

Fig. 2. Effect of contact time on adsorption of cadmium and

nickel.

hydrolysis of cadmium and nickel. Similar behavior has

been reported by Sen and De [22] for adsorption of
Hg(II) and by Gupta et al. [18] for the adsorption of
lead.
The effect of the adsorbent dose on the removal of
cadmium and nickel is shown in Fig. 5. The adsorption
increasesd from 0.64 to 1.18 mg g
1 with increase in
adsorbent dose from 2 to 10 g l
1 in the case of cadmium
and then became almost constant. The increase in
adsorption was found to be from 0.40 to 0.96 mg g
1
in the case of nickel for a similar increase in adsorbent
dose. The results indicate that a dose of 10 g l
1 of
adsorbent is sufcient for the optimum removal of both
the metal ions.
The percent adsorption of cadmium was found to be
99.9%, 90.0% and 88.0% for particle size of 100150,
200250 and 300350 mm, respectively. The adsorbent of

ARTICLE IN PRESS
V.K. Gupta et al. / Water Research 37 (2003) 40384044

Ni

30 deg C

Amount adsorbed, mg/g

1.20
0.80
0.40
0.00
0

4
8
12
16
Initial concentration, mg/l

Amount Adsorbed, mg/g

0.80
0.40
0.00
2

10

pH
Fig. 4. Effect of pH on adsorption of cadmium and nickel.

Amount adsorbed, mg/g

Cd

Ni

1.60
1.20
0.80
0.40
0.00
0

1.00
0.50
0.00

12

16

12

16

20

Fig. 6. Effect of temperature on the adsorption of cadmium.

30 deg C

1.20

1.50

Ni

1.60

50 deg C

Initial concentration, mg/l

Fig. 3. Effect of initial concentration on adsorption of

cadmium and nickel.

Cd

40 deg C

2.00

20

20

Dose, g/l
Fig. 5. Effect of adsorbent dose on adsorption of cadmium and
nickel.

the particle size 200250 mm was selected for further

adsorption studies due to its ease of preparation.
The effect of temperature on the adsorption of the
two metal ions is presented in Figs. 6 and 7. It was clear
from these plots that the adsorption was almost constant

Amount adsorbed, mg/g

Amount adsorbed, mg/g

Cd
1.60

4041

40 deg C

50 deg C

2.00
1.50
1.00
0.50
0.00
0

4
8
12
16
Initial concentration, mg/l

20

Fig. 7. Effect of temperature on the adsorption of nickel.

at lower concentrations while it increased at higher

concentrations of adsorbates. The adsorption followed
the order 30 C>40 C>50 C for both the metal ions.
The increase in adsorption with increasing temperature
indicated endothermic nature of the adsorption process.
The effect of various interfering ions on the adsorption of two metal ions indicated that a slight decrease in
the scavenging efciency of the adsorbent occurred in
the presence of these ions. The decrease was 10.0%,
8.5%, 7.7%, 6.8%, 4.0%, 4.0%, 4.2% and 3.0% in the
case of cadmium adsorption in the presence of sodium
(100 mg/l), potassium (50 mg/l), calcium (100 mg/l),
magnesium (50 mg/l), chloride (300 mg/l), sulphate
(50 mg/l), nitrate (50 mg/l) and phosphate (10 mg/l),
respectively. The decrease was found to be 9.8%,
10.0%, 9.7%, 8.0%, 4.3%, 4.2%, 4.1% and 3.7% in
the case of nickel adsorption in the presence of sodium,
potassium, calcium, magnesium, chloride, sulphate,
nitrate and phosphate, respectively. The effect of ionic
interaction is due to the competitive adsorption and can
be explained using the rate of adsorption capacity of
the ion in a single component system to that in a

ARTICLE IN PRESS
V.K. Gupta et al. / Water Research 37 (2003) 40384044

4042

multicomponent system [23]. The presence of interfering

ion would reduce the uptake of cadmium and nickel. As
the decrease was not very appreciable, the adsorbent can
be used for the removal of cadmium and nickel in the
presence of these ions.
3.2. Adsorption isotherms
The results obtained on the adsorption of cadmium
and nickel were analysed by the well-known models
given by Langmuir and Freundlich.
3.3. Langmuir isotherm

Table 1
Langmuir and Freundlich constants for the uptake of cadmium
and nickel
Langmuir constants

Freundlich constants

Qo (mg/g)

KF

Cadmium
30oC
40oC
50oC

1.24
1.67
2.00

5.67
6.21
7.14

8.93
10.12
12.50

1.02
1.11
1.26

Nickel
30oC
40oC
50oC

1.12
1.35
1.70

6.42
7.21
8.66

10.04
12.31
14.00

0.90
1.05
1.17

The Langmuir isotherm has been used by various

workers for the sorption of variety of compounds. The
model assumes uniform energies of adsorption onto the
surface and no transmigration of adsorbate in the plane
of the surface. The linear form of Langmuir isotherm is
given by the following equation:

0.10

where qe is the amount adsorbed (mg g
1), Ce the

equilibrium concentration of the adsorbate (mg l
1) and
Qo and b are the Langmuir constants related to
maximum adsorption capacity and energy of adsorption, respectively. When 1=qe was plotted against 1=Ce ;
straight lines with slope 1=bQo were obtained (Fig. 8),
which show that the adsorption of cadmium and nickel
followed the Langmuir isotherm. The Langmuir constants, b and Qo ; were calculated and the values of these
are given in Table 1. The values of constants indicate
favourable conditions for adsorption.

y = 0.11x + 0.01

Cd

0.05
log qe

1=qe 1=Qo 1=bQo Ce ;

0.15

0.00

Ni

-0.05
-0.10
-0.15
-1.0

y = 0.1x - 0.04
-0.5

0.0
log Ce

0.5

1.0

Fig. 9. Freundlich isotherm for adsorption of cadmium and

nickel.

3.4. Freundlich isotherm

The adsorption data of cadmium and nickel is
also analysed by Freundlich model. The logarithmic form of Freundlich model is given by the

1.60
y = 0.16x + 0.90

1/qe

1.40

Ni

1.20
1.00

Cd

0.80

following equation:
log qe log KF 1=n log Ce ;

2

1

where qe is the amount adsorbed (mg g ), Ce the

equilibrium concentration of the adsorbate (mg l
1) and
KF and n are Freundlich constants related to adsorption
capacity and adsorption intensity, respectively [24]. The
plots of log qe against log Ce ; for the adsorption data of
cadmium and nickel are given in Fig. 9, which clearly
show that the data is not tting very well to the
Freundlich model. However, the Freundlich constants,
KF and n; were calculated from the best-t lines and the
values of these at three different temperatures are given
in Table 1.

y = 0.13x + 0.81

0.60

3.5. Kinetic studies

0.0

0.5

1.0

1.5

2.0

2.5

1/Ce
Fig. 8. Langmuir isotherm for cadmium and nickel.

3.0

The thermodynamic parameters for the adsorption of

cadmium and nickel were calculated by using the
following equations, and the values of thermodynamic

ARTICLE IN PRESS
V.K. Gupta et al. / Water Research 37 (2003) 40384044

4043

Table 2
Thermodynamic parameters for the uptake of cadmium and nickel
Adsorbate

Cadmium
Nickel

DG (kJ mol
1)

DH  (kJ mol
1)

DS (J mol
1)

30 C

40 C

50 C

30 C

40 C

50 C

30 C

40 C

50 C

4.37
4.69

4.60
4.98

4.95
5.44

7.18
9.28

11.74
15.41

9.39
12.19

9.27
15.15

22.81
33.32

14.55
20.89

Table 3
Values of B; Di ; Do ; Ea and DS# for the uptake of cadmium and nickel
B 104 (s
1)

Cd
Ni

Di 108 (cm2 s
1)

30 C

40 C

50 C

30 C

40 C

50 C

7.29
3.74

7.36
3.80

7.41
3.87

2.96
1.52

2.99
1.54

3.01
1.57

parameters are given in Table 2:

o

DG
RT ln b;

DH o
RT2 T1 =T2
T1  ln b2 =b1 ;

DS o
DG o
DH o =T;

where b; b1 and b2 are the equilibrium constants at

temperatures T; T1 and T2 ; respectively, obtained from
the Langmuir isotherms. A perusal of Table 2 indicated
that the free energy decreased with an increase in
temperature thereby indicating an increase in adsorption
at higher temperature and endothermic nature of the
adsorption.
To establish the practical utility of adsorption, kinetic
data has been treated by the models given by Boyd et al.
[25], which is valid under the experimental conditions
used. This is in accordance with the observations of
Reichenberg [26] as given by the following equations:
a
X
F 1
6=p2
1=n2 exp
n2 Bt;
6
n1

and B pDi =r2 ;

where F is fractional attainment of equilibrium at time t;

Di the effective diffusion coefcient of adsorbates in the
adsorbent phase, r the radius of adsorbent particle
assumed to be spherical, and n is an integer.
The fractional attainment of the equilibrium can be
determined by the equation
F Qt =Qa ;

where Qt and Qa are the amounts adsorbed after time t

and after innite time (24 h), respectively. For every
calculated value of F ; corresponding values of Bt are
calculated from Eq. (6). The linearity tests of Bt versus

Do 10
8 (cm2 s
1)

Ea (kJ mol
1)

DSx (J K
1 mol
1)

3.14
1.51

66.67
100.0

35.93
38.01

time plots (not given) are employed to nd out the value

of B (slope). The values of Di are calculated using
Eq. (7). At lower concentrations, the Bt versus t plots
did not pass through the origin, signifying the adsorption to be lm diffusion, but at higher concentrations,
the plots were linear and passed through the origin up to
a certain concentration, indicating the adsorption to be
particle diffusion [27] in nature.
The value of energy of activation (Ea ) for the
adsorption process is calculated from the following
equation:
ln K2 =K1 Ea =RT2
T1 =T1 T2 ;

where K2 and K1 are rate constants at temperature T2

and T1 : The value of Ea was also calculated from the
slope (slope=
Ea =2:303 R) of the plot of ln K versus
1=T (plot is not given), and it was found that the values
were in good agreement with those obtained from
Eq. (9).
The value of pre-exponential constant, Do; (analogous
to Arrhenius frequency factor) and the entropy of
activation, DS x ; are calculated from the following
equations:
Di Do exp
Ea =RT;

10

Do 2:72 d 2 kT=h expDS# =R;

11

where d is the average distance between the successive

exchange sites and is taken as 5Ao : R; h and k are
the gas, Plank and Boltzman constants, respectively.
The values of Ea ; Di ; Do ; and DS x are given in Table 3.
The negative values of DS# reect that no signicant
change occurs in the internal structure of bagasse y ash
during the adsorption of cadmium and nickel.

ARTICLE IN PRESS
4044

V.K. Gupta et al. / Water Research 37 (2003) 40384044

4. Conclusions
Bagasse y ash obtained from the local sugar industry
is an inexpensive and effective adsorbent for the removal
of cadmium and nickel from wastewater. The removal of
the two metal ions takes place by a partial diffusion
mechanism. The adsorption data t very well to the
Langmuir model in comparison to the Freundlich
model. The adsorption was found to be endothermic
in nature.

Acknowledgements
Authors are thankful to the Council of Scientic and
Industrial research (CSIR), India, for providing nancial support to undertake the work.

References
[1] Clement RE, Eiceman GA, Koester CJ. Environmental
analysis. Anal Chem 1995;67:221R55R.
[2] MacCarthy P, Klusman RW, Cowling SW, Rice JA.
Water analysis. Anal Chem 1995;67:525R82R.
[3] Zuane JD. Hand book of drinking water quality:
standards and control. New York: Van Nostrand Reinhold; 1990.
[4] Forstner U, Wittmann GTW. Metal transfer between solid
and aqueous phases. In: Metal pollution in the aquatic
environment. New York: Springer; 1979. p. 97269.
[5] Low KS, Lee CK. Cadmium uptake by the moss
Calymperes delessertii and besch. Bioresour Technol 1991;
38:16.
[6] Cheremisinoff P, Ellerbush F. Carbon adsorption hand
book. Michigan: Ann Arbor Science Publishers; 1979.
[7] Lalvani SB, Wiltowski T, Hubner A, Weston A, Mandich
N. Removal of hexavalent chromium and metal cations by
a selective and novel carbon adsorbent. Carbon 1998;36:
121926.
[8] Pollard SJT, Fowler GD, Sollars CJ, Perry R. Low cost
adsorbents for waste and wastewater treatment: a review.
Sci Total Environ 1992;116:3152.
[9] Aly HM, Daifullah AAM. Potential use of bagasse pith for
the treatment of wastewater containing metals. Adsorp Sci
Technol 1998;16:338.
[10] Daneshvar N, Salari D, Aber S. Chromium adsorption and
Cr(VI) reduction to trivalent chromium in aqueous solutions by soya cake. J Hazardous Mater 2002;B94:4961.

[11] Dimitrova SV, Mehandgiev DR. Lead removal from

aqueous solution by granulated blast furnace slag. Water
Res 1998;32:328992.
[12] Ferraiolo G, Zilli M, Converti A. Fly ash disposal and
utilization. J Chem Technol Biotechnol 1990;47:281306.
[13] Periasamy K, Namasivayam C. Process development for
removal and recovery of cadmium from wastewater by a
low cost adsorbent: adsorption rate and equilibrium
studies. Ind Eng Chem Res 1994;33:31720.
[14] Srivastava SK, Tyagi R, Pant N. Adsorption of heavy
metal ions on carbonaceous material developed from the
waste slurry generated in local fertilizer plants. Water Res
1989;23:11615.
[15] Galiatsatou P, Metaxas M, Kasselouri-Rigopoulou V.
Adsorption of zinc by activated carbon prepared from
solvent extracted olive pulp. J Hazardous Mater 2002;B91:
187203.
[16] Mathialagan T, Viraraghavan T. Adsorption of cadmium
from aqueous solutions by perlite. J Hazardous Mater
2002;B94:291303.
[17] Gupta VK, Rastogi A, Diwivedi MK, Mohan D. Process
development for the removal and recovery of zinc and
cadmium from wastewater using slag a blast furnace waste
material. Sep Sci Technol 1997;32:2883912.
[18] Gupta VK, Mohan D, Sharma S,. Removal of lead from
wastewater using bagasse y ash-a sugar industry waste
material. Sep Sci Technol 1998;33:133143.
[19] Gupta VK, Gupta M, Sharma S. Process development for
the removal of lead and chromium from aqueous solutions
using red mudan aluminium industry waste. Water Res
2001;35:112534.
[20] Vogel AI. Text book of quantitative chemical analysis, 5th
ed.. London: ELBS Publications; 1989.
[21] Gupta VK, Ali I. Utilization of bagasse y ash (a sugar
industry waste) for the removal of copper and zinc from
wastewater. Sep Purif Technol 2000;18:13140.
[22] Sen AK, De AK. Adsorption of mercury (II) by coal y
ash. Water Res 1987;21:88591.
[23] Tan TC, Chia CK, Teo CK. Uptake of metal ions
by chemically treated human hair. Water Res 1985;19:
15762.
[24] Weber WJ. Physiochemical processes for water quality
control. Wiley: NewYork; 1972. pp. 208.
[25] Boyd GE, Adamson AW, Meyers LS. The exchange
adsorption of ions from aqueous solution by organic
zeolites. II Kinetics. J Am Chem Soc 1947;69:283648.
[26] Reichenberg D. Properties of ion exchange resins in
relation to their structures III. Kinet Exchangers J Am
Chem Soc 1953;75:58997.
[27] Helfferich F. Ion exchange. New York: McGraw Hill;
1962.