926

Ind. Eng. Chem. Res. 1988,27, 926-929

desorbed as C4 olefins. It is calculated that the energy

levels of four types of C4Hg+a t standard temperature are
tert-C4Hg+< i-C4Hg+ (7 kcal/mol) < sec-C4Hg+ (21
kcal/mol) < n-C4Hg+(37 kcal/mol), where the energy
values in parentheses are the energy differences between
each carbonium ion and tert-C4Hg+at room temperature,
respectively. The most predominant species, tert-C4Hg+,
should be desorbed mostly as isobutene because of its low
reactivity. The isobutene should then be converted to
isobutane via the hydrogenation. In fact, about 70% of
the C4 hydrocarbon was isobutane.
The low selectivity of C5 and C6hydrocarbons can also
be attributed to the low reactivity of tert-C4H9+and
tert-C5HI1+.

Conclusions
(1)The high yield of hydrocarbons on the hybrid catalyst could be attributed to the removal of the thermodynamic limitation of the methanol formation from synthesis
gas by the quick conversion of methanol to hydrocarbons
on the zeolite.
(2) The hybrid catalyst containing Y-type zeolite gave
C3and C4paraffins selectively from synthesis gas, whereas
the methanol conversion on DAY alone gave significant
amounts of aromatic hydrocarbons and coke. The reduced
formation of coke and aromatics could be attributed to the
quick hydrogenation of olefins which are the intermediate
products from methanol to aromatics or coke on the
methanol synthesis catalyst.
(3) The physical properties of zeolite which enable the
short residence time of molecules in the zeolite, for instance, the small particles composed of small-sized crystals
with large pores, are favored by the reduced coke and
aromatics formation.

(4)The high selectivity of C, and C4 paraffins could be

attributed to the high reactivity of C1 and C2carbonium
ions and the lower reactivities of the isopropylcarbonium
ion and especially, the tert-butylcarbonium ion.
Registry No. Pd, 7440-05-3;Cu, 7440-50-8; Zn, 7440-66-6; Cr,
7440-47-3; CO, 630-08-0; C4H10,106-97-8 i-C4H10,
75-28-5; C3H8,
74-98-6; CSHs, 115-07-1.
Literature Cited
Barrer, R. M. Molecular Sieve Zeolite ZI; Flanigan, E. M., Sand, L.
B., Eds.; Academic: New York, 1971; Chapter 1.
Chang, C. D.; Bell, W. K. Catalysis of Organic Reactions; MarcelDekker: New York, 1981; p 73.
Chang, C. D.; Lang, W. H.; Silvestri, A. J. J. Catal. 1979, 56, 268.
Chang, C. D.; Miale, J. N.; Socha, R. F. J. Catal. 1984,90,84.
Fujimoto, K.; Kudo, Y.; Tominaga, H. Nippon Kagakkai shi 1982,
206.
Fujimoto, K.; Kudo, Y.; Tominaga, H. J. Catal. 1984, 87, 136.
Fujimoto, K.; Saima, H.; Tominaga, H. J. Catal. 1985, 94, 16.
Hoogendoorn, J. C. Prepr. ZGT Symp. 1973, 1.
Inui, T.; Hagiwara, T.; Tadegami, Y. J.Jpn. Pet. Znst. 1984,27,228.
Kodama, S.; Tarama, K.; Mishima, A.; Fujita, K.; Yasuda, M. Kogyo
Kagaku Zasshi 1943, 46, 69.
Morrison, G. H.; Boid, R. N. Organic Chemistry, 4th ed.; Allen and
Bacon: Boston, 1983; Chapter 5.
Murchison, C. B.; Murdick, D. A. Hydrocarbon Process. 1981, 160.
Olive, G. H.; Olive, S. Angew. Chem. Znt. Ed. Engl. 1976, 15, 136.
Saima, H.; Fujimoto, K.; Tominaga, H. Bull. Chem. SOC.
Jpn. 1985,
58, 795.
Shimomura, K.; Ogawa, K.; Oba, M.; Kotera, Y. J. Catal. 1978,52,

191.

Weisz, P. B. Advances in Catalysis; Eley, D. D., Selwood, P. W.,

Academic: New York, 1962; Chapter 3.

Received for review April 2, 1987

Revised manuscript received December 29, 1987
Accepted January 23, 1988

Calcination of Nickel/Alumina Catalysts

Ienwhei Chen,* Shyan-Yeu Lin, and Dar-Woei Shiue
Department of Chemical Engineering, Tatung Institute of Technology, Taipei, Taiwan, R.O.C.
The objective of this study is to examine particle sizes and their variation of nickel oxide and nickel
of NiO/A1203 and Ni/A1203 catalysts pretreated by various conditions of calcination, respectively.
A series of NiO/A1203 catalysts were prepared with impregnation followed by calcination ranging
from 673 to 973 K. This study was carried out by using a Lindberg box furnace, X-ray diffractometer,
and thermogravimetric analyzer. The results indicate that the particle sizes of nickel oxide and nickel
are the functions of nickel salt, calcination temperature, nickel loading, heating rate of calcination,
amount of water for impregnation, and oxygen flow rate. The ratios of particle size of nickel to that
of nickel oxide are presented and discussed.
Nickel catalysts find applications in a number of important industrial processes such as hydrogenation of unsaturated organics, steam reforming, and methanation
(Peterson, 1977). An important stage in the preparation
of Ni/A1203catalysts is calcination of alumina-supported
nickel salts to alumina-supported nickel oxide. The calcination may have several purposes. One is to eliminate
extraneous material such as binders and die lubricants, as
well as volatile and unstable anions and cations, which are
not desired in the final catalyst (Satterfield, 1980). A
substantially elevated temperature is usually needed to
increase the strength of the final pellet by incipient sintering, but excessive sintering will reduce the catalyst
activity by reducing surface area and may also cause diffusion limitations due to reduction of pore size (Bartholomew and Farrauto, 1976).
0888-5885/88/2627-0926$01.50/0

The effect of calcination temperature on physical

properties such as surface area and crystallite size distribution has been dealt with (Richardson and Crump,
1979). Metal particle size and metal surface area change
with time because of catalyst degradation (i.e., sintering,
poisoning, and coking). The ratio of metal oxide to alumina is also important since much of it may be in the form
of an aluminate with a low concentration of metal, but
most of the oxide form may be retained at high concentrations (Preparation of Catalysts, 1976). However, there
is little information dealing with the chemistry involved
in calcination of nickel/alumina catalysts. The final
particle size of supported nickel is related, as expected, to
calcination conditions of supported nickel salt. Particle
sue of supported nickel oxide would be a function in terms
of calcination temperature, calcination time, nickel loading,
0 1988 American Chemical Society

Ind. Eng. Chem. Res., Vol. 27, No. 6, 1988 927

heating rate, oxygen flow rate, nickel salt, and amount of
water for impregnation.
Previous studies of catalyst particle size and various
methods for estimating this property have been summarized and reviewed (Whyte, 1973; Farrauto, 1974). The
particle sizes of catalysts have been studied by techniques
including X-ray diffraction (XRD), hydrogen chemisorption, and transmission electron microscopy (TEM). It is
clear from recent literature that magnetic susceptibility
measurements can also be used to measure catalyst crystallite size and size distribution (Richardson, and Dubus,
1978; Primet et al., 1977). It is pointed out that X-ray
diffraction was considered to be a more useful technique
for determining the particle size of a catalyst.
The present study was undertaken to (i) determine and
compare the particle size of nickel oxide of NiO/A1203
pretreated with various techniques (Le., calcination temperature, calcination time, metal loading, heating rate,
oxygen flow rate, nickel salt, and amount of water for
impregnation) and (ii) simulate characterization of alumina-supported nickel from that of alumina-supported
nickel oxide.

Experimental Section
Impregnation of y-alumina powder (Merck) with an
aqueous nickel salt solutions is used to prepare aluminasupported nickel catalysts. Nickel salts were nickel nitrate
(Wako), nickel chloride (Kokusan), nickel sulfate (Kokusan), and nickel formate (Wako). Amount of water for
impregnation is from 0.35 to 2.5 mL/g of A1203. After
impregnation, the samples were dried in air at 333 K for
24 h followed by calcination in air at temperatures ranging
from 673 to 973 K for 15-90 min and subsequent reduction
in flowing hydrogen at 973 K for 2 h. Heating rate is from
5 to 25 K/min. Flowing rate of oxygen is from 0 to 135
mL/min. By the described manner, a series of aluminasupported nickel catalysts with 10, 15, 20, and 25 w t %
Ni were obtained.
Both the identification of crystalline compounds and the
determination of average crystallite size were carried out
by examination of X-ray diffraction spectrum using a
Rigaku D/max-A system diffractometer with Fe Ka radiation.
The mean crystallite size (D)is determined by the
Scherrer formula
D = K X / Bcos e
The size (0)
was defined as (volume)li3leading to a value
K = 0.95 (Klug and Alexander, 1954) when B is defined
as the half-maximum line width. The half-maximum line
width from NiO(ll1) and N i ( l l 1 ) reflecting planes was
employed for alumina-supported catalysts.

Results and Discussion

Thermogravimetric Analysis. To better understand
the chemical changes of alumina-supported nickel nitrate
during the calcination, the thermal decompositions of
unsupported Ni(N0J2.6H20 and y-AlZO3 were studied by
means of a thermogravimetric analyzer (TGA). The weight
loss versus temperature curves of the decompositions are
shown in Figure 1. It is observed that the decomposition
of nickel nitrate to NiO is complete above 673 K by weight
loss of 74.7% as compared with a calculated value of 74.3%
from stoichiometry for conversion of Ni(NO3),.6H20 to
NiO. y-Alumina is used as support in this experiment
because ita weight remains reasonably constant within the
range of temperatures from 473 to 1123 K.
Effect of Nickel Salt on Particle Size of Nickel
Oxide. The particle size of nickel oxide of each catalyst

156

656

856

1056

Temperature ( K )

Figure 1. Thermogravimetric analysis of (A) y-Al,O, and (B) Ni(N03)2-6H20.

OS

901

? I
Z

30c
I

2
I

30

60

90

Calclnatlon Time ( min )

Figure 2. Effect of calcination time on particle size of nickel oxide

over alumina.
Table I. Effect of Nickel Salt on Particle Size of NiO over
Alumina
nickel salt
particle size of NiO, A
40

78
127

is determined by X-ray diffraction as shown in Table I.

Four different catalysts containing 10 wt % Ni were prepared by impregnation of alumina with different nickel
salts including nickel nitrate, nickel chloride, nickel sulfate,
and nickel formate, followed by calcination at 673 K for
1 h. It is observed that the particle size of nickel oxide
prepared from nickel nitrate is significantly smaller than
that prepared from the other nickel salts. This result
suggests that nickel nitrate is better used for preparation
of NiO/A1203in order to meet the need of high dispersion
and great surface areas of supported nickel oxide. From
examination of their X-ray diffraction spectra, the nickel
sulfate still survived after calcination at 673 K for 1 h. The
nickel sulfate will lose SO3 to be NiO if nickel sulfate is
calcined at 1113 K.
Effect of Calcination Time on Particle Size of
Nickel Oxide. The effect of calcination time at 673 K
on particle size of nickel oxide for alumina-supported
nickel nitrate with 10 wt % Ni is shown in Figure 2. The
data indicate that the particle size of nickel oxide remains
constant within the range of calcination time from 15 to
90 min at 673 K. In other words, calcination at 673 K
would not be able to cause any sintering of supported
nickel oxide.
Effect of Calcination Temperature on Particle Size
of Nickel Oxide. The particle sizes of nickel oxide are
shown in Figure 3 as a function of calcination temperatures

928 Ind. Eng. Chem. Res., Vol. 27, No. 6, 1988

.a

500

0
673
823
973
Calcination Temperature ( K )

Figure 3. Effect of calcination temperature on particle size of nickel

15 wt % nickel loading; (0)20 wt % nickel
oxide over alumina: (0)
loading.

-0

200 I
0

*a 1601

-p:

Figure 5. Effect of heating rate on particle size of nickel oxide over

alumina.

I
0

/"

14
20
26
Heating Rote ( K l m i n )

= I

\i.

z
c

Ln
N

601

10

20
30
Nickel Loading (wto/o)

Figure 4. Effect of nickel loading on particle size of nickel oxide

over alumina.

ranging from 673 to 973 K for alumina-supported nickel

formate with 15 and 20 wt % Ni, respectively. Particle
size of nickel oxide increases drastically with increasing
calcination temperatures from 673 to 873 K, and then it
increases slightly with calcination temperatures from 873
to 973 K. This result shows that the calcination temperature is the main factor to cause sintering of supported
nickel oxide. The sintering becomes less significant if the
calcination temperature rises above 873 K. A possible
explanation is that more nickel aluminate is formed from
strong interaction of nickel oxide and alumina during
calcination at temperatures above 873 K. It is expected
that the forming of nickel aluminate enhances the stability
and resistivity to sintering of nickel oxide on alumina.
From the examination of the X-ray diffraction spectrum
of alumina-supported nickel oxide, which containing 15
w t ?% Ni is calcined a t 873 K, nickel aluminate is found
to be existent. If the calcined alumina-supported nickel
oxide is dissolved in 5 N HC1 solution and the extract
solution is analyzed with atomic absorption spectrophotometer, 36 wt ?% nickel is found to be contained in NiA120r,and 64 wt % nickel exists as NiO. Similar observation has been reported on the preparation of NiO/ aA120,. After calcination at 1643 K, the XRD spectrum and
the chemical analysis showed NiA1204to be the only nickel
compound (Gavalas et al., 1984).
Effect of Nickel Loading on Particle Size of Nickel
Oxide. The effect of nickel loading on the particle size
of nickel oxide at calcination of 673 K is shown in Figure
4. It can be seen that the particle size of the nickel oxide
on alumina increases with increasing nickel loading from
10 to 25 wt 9% nickel, and its increase becomes especially
rapid as nickel loading increases from 20 to 25 wt % nickel.
The possible reason is that the first layer on the alumina
surface would be fully occupied with nickel nitrate as the
nickel loading goes up to 20 wt % , and the degree of sintering during calcination becomes significant with nickel

30

75

120

150

Oxygen Flowing ( m l / m i n i

Figure 6. Effect of oxygen flow rate on particle size of nickel oxide

over alumina.

loading above 20 wt % due to the developing of threedimensional nickel nitrate. In other words, the nickel
utilization is very poor for the increment of nickel loading
above 20 wt %.
Effect of Heating Rate on Particle Size of Nickel
Oxide. Data presented in Figure 5 illustrate the effect of
the rate of temperature rise to desired calcination temperature for alumina-supported nickel nitrate with 10 wt
% nickel on particle size of the nickel oxide. Heating rates
used were 5,8, 10, and 24 K/min, respectively. It is observed that the particle size of nickel oxide increases
slightly with increasing heating rate. It is a possible explanation that, the higher the heating rate, the greater the
producing rate of decomposition products of Ni(NO&.
6H20would be. The decomposition products, which are
water and oxidization agents, are to aid in causing large
particles of nickel oxide during calcination (Bartholomew
and Farrauto, 1976). More sufficient time is allowed to
proceed the intraparticle transportation of the decomposition products at slower heating rates.
Effect of Oxygen Flowing Rate on Particle Size of
Nickel Oxide. Particle size of nickel oxide is shown in
Figure 6 as a function of oxygen flow rate. The data
correspond to different samples of the same aluminasupported nickel nitrate with 15 wt % Ni, each calcined
at 673 K for 1 h. Particle size of nickel oxide decreases
with increasing oxygen flow rates ranging from 0 to 135
mL/min. This results very possibly from (i) increased
efficiency transport of oxygen and decomposition products
and (ii) uniform temperature distribution due to more
effective removal of heat generated from exothermal oxidation at higher oxygen flow rate.
Effect of Amount of Water for Impregnation on
Particle Size of Nickel Oxide. Effect of the amount of
water for impregnation on particle size of nickel oxide is

Ind. Eng. Chem. Res., Vol. 27, No. 6, 1988 929

/O

0/ O
0

1
2
3
Water Amount for Impregnation (ml/gm Alp, )

Figure 7. Effect of the amount of water for impregnation on particle size of nickel oxide over alumina.
Table 11. Relationship of Particle Size between Ni and NiO
over Alumina at Different Calcination Temperatures
particle size
particle size
calcination
of
of Ni, 8,
P
temp, K
Xb
Yc
X
Y
X
Y
673
140
210
236
546
1.7
2.6
773
312
380
374
684
1.2
1.8
873
364
398
473
757
1.3
1.6
380
412
494
973
790
1.3
1.6
9

a r = particle size of Ni/particle size of NiO. b X = 15 wt %

nickel loading. Y = 20 w t 70nickel loading.

shown in Figure 7 for the preparation of alumina-supported nickel nitrate with 15 wt % Ni followed by calcination at 673 K for 1 h. It can be shown that the particle
size of nickel oxide increases slightly with increasing
amount of water. A possible explanation is that the more
water there is for impregnation, the more difficult the
penetration of nickel nitrate into the pore of alumina would
be. Therefore, more nickel nitrate accumulates on the
mouth of the pore. This results in a larger particle size
of nickel oxide after calcination.
Relationship between Particle Size of Nickel Oxide
and That of Nickel. The particle sizes of nickel oxide
and nickel are shown in Table I1 as a function of calcination temperature for alumina-supported nickel formate
with 15 and 20 wt 7% Ni. Each sample was reduced at 973
K for 2 h after 1-h calcination. It is observed that the
particle size of nickel oxide and nickel is increased from
140 to 380 A and from 236 to 494 A, respectively, for 15
wt % nickel loading samples; it is also increased from 210
to 412 A and from 546 to 790 A, respectively, for 20 wt %
nickel loading samples as the calcination temperature is
increased from 673 to 973 K. The particle size of nickel
is larger than that of nickel oxide in the whole range of
calcination temperatures. The lower the calcination temperature, the more significant the particle size difference
of nickel and nickel oxide would be. A possible reason is
that the sintering and poor redispersion of nickel oxide and
nickel would occur during reduction. The greater the
temperature difference between calcination and reduction,
the greater the sintering of nickel oxide and nickel would
be. The ratio of particle size of nickel to that of nickel

oxide is shown in Table 11, is found to be decreased with

increasing calcination temperature up to 873 K, and then
reaches a constant at 1.3 and 1.6 for catalyst samples with
15 and 20 w t % Ni, respectively. This results suggest that
the particle size of alumina-supported nickel could be
predicted from that of alumina-supported nickel oxide due
to the establishment of the constant ratio between them
if calcination is undertaken above 873 K.

Conclusions
The particle size of the nickel oxide of NiO/A1,0, is
found to be significantly influenced by nickel salt, calcination temperature, nickel loading, heating rate, oxygen
flow rate, and amount of water for impregnation.
Nickel nitrate is a better choice for preparing a higher
dispersion of nickel oxide on y-alumina.
The particle size of nickel oxide remains constant within
the range of calcination times from 15 to 90 min at 673 K.
The particle size of nickel oxide increases with increasing
calcination temperatures up to 673 K and heating rates
up to 10 K/min and then increases only slightly at temperatures above 673 K and heating rates above 10 K/min.
The increase of the particle size of nickel oxide with
increasing nickel loading up to 20 wt % is slower as compared to that with nickel loading above 20 wt 70.
A slower heating rate to the desired calcination temperature and a higher oxygen flow rate during calcination
of alumina-supported nickel nitrate increase the transportation of oxygen and decomposition products and
prevents the occurrence of a hot spot, which may badly
sinter the catalyst, to obtain better dispersed nickel oxide
on alumina.
The particle size of nickel oxide is slightly increased with
increasing amount of water for impregnation.
The constant ratio of particle size of nickel to that of
nickel oxide is obtained if the calcination temperature is
above 873 K in the process for preparation of aluminasupported nickel catalyst.
Acknowledgment
The authors thank the Tatung Company and the Chinese National Science Council for financial aid.
Registry No. Ni, 7440-02-0; NiO, 1313-99-1.

Literature Cited
Bartholomew, C. H.; Farrauto, R. J. J. Catal. 1976, 45, 41.
Farrauto, R. J. AICHE S y m p . Ser. 1974, 70, 143.
Gavalas, G. R.; Phichitkul, C.; Voecks, G. E. J. Catal. 1984,88,54.
Gessner, G. H. Condensed Chemical Dictionary; Van Nostrand
Reinhold: New York, 1981.
Klug, H. P.; Alexander, L. E. X - R a y Diffraction Procedures; Wiley:
New York, 1954.
Peterson, R. J. Hydrogenation Catalysis; Noyes Data: Park Ridge,
NJ, 1977.
Preparation of Catalysts; Delmon, B., Jacobs, P. A., Poncelet, G.,
Eds.; Elsevier: Amsterdam, 1976.
Primet, M.; Dalmon, J. A.; Martin, G. A. J. Catal. 1977, 46, 25.
Richardson, J. T.; Dubus, R. J. J. Catal. 1978, 54, 207.
Richardson, J. T.; Crump, J. G. J. Catal. 1979,57, 417.
Satterfield, C. N. Heterogeneous Catalysis; McGraw-Hill: New
York, 1980.
Whyte, T. I. J. Catal. Rev. 1973, 8, 117.
Received for review June 12, 1987
Accepted January 15, 1988