Catalysis Today 348 (2020) 118–126

Contents lists available at ScienceDirect

Catalysis Today
journal homepage: www.elsevier.com/locate/cattod

Kinetics of acetone hydrogenation for synthesis of isopropyl alcohol over Cu- T

Al mixed oxide catalysts
Sanchari Basu, Narayan C. Pradhan
⁎

Department of Chemical Engineering, Indian Institute of Technology, Kharagpur, 721302, India

ARTICLE INFO ABSTRACT

Keywords: Vapor phase hydrogenation of acetone was carried out over Cu-Al mixed oxide catalysts. The catalysts were
Acetone synthesized via co-precipitation method with Cu:Al atomic ratio varied from 0.25 to 1. The characterization of
Isopropanol the catalysts was done by using XRD, BET, FESEM (coupled with EDS), FTIR, H2-TPR, NH3-TPD and CO2-TPD.
Hydrotalcite The performance of the synthesized catalysts was evaluated in a continuous packed bed reactor by varying
Hydrogenation
parameters such as temperature (150 to 225 °C), H2/acetone mole ratio (0.5 to 1.25) and space-velocity (0.09 to
Kinetics
Activation energy
0.145 kmol acetone/kg cat. h). Along with direct hydrogenation, condensation of acetone also occurred due to
the presence of basic and acidic sites in the catalysts. The catalyst having Cu:Al atomic ratio of 0.5 showed
highest selectivity (98%) to isopropyl alcohol at 175 °C. Kinetic study was conducted at different temperatures
by varying space-velocity. LHHW type model rate equations were fitted with the kinetic data obtained at three
different temperatures. The activation energy of the reaction was determined to be 44.3 kJ/mol.

1. Introduction solid catalysts serves as a better option for synthesis of IPA [1].
Hydrogenation of acetone to IPA is a reversible reaction. The for-
Acetone is a byproduct of the cumene process for phenol. The ward reaction is exothermic and the backward dehydrogenation reac-
growing demand of phenol for phenolic resins and polycarbonates leads tion is endothermic. So, the reaction can be used as a potential chemical
to the surplus production of byproduct acetone. The difference between heat pump system [2]. But, to achieve this, a high selectivity of IPA is
global acetone production and demand is estimated as 1.7 million tons. required. The other reactions that occur along with direct hydrogena-
Therefore, a significant upgradation of acetone to valuable products is tion cause the formation of diacetone alcohol (DAA), mesityl oxide
required for better economy of the process. Catalytic hydrogenation of (MO) and methyl isobutyl ketone (MIBK). The reactions are:
acetone in both vapor and liquid phases for synthesis of isopropyl al- Direct hydrogenation of acetone to IPA
cohol (IPA) has been explored since several decades over various cat-
(CH3 )2 CO + H2 (CH3 ) 2 CHOH H = 55 kJ/mol (1)
alysts. Isopropyl alcohol is widely used as a solvent and chemical in-
termediate in various industries. As a solvent, it finds application in Condensation of acetone to diacetone alcohol
paint industry, inks industry, electronic industry, food and pharma- 2(CH3) 2 CO (CH3)C(O)CH2 C(OH)(CH3) 2 H = 23 kJ/mol
ceutical industries. Chemical derivatives such as isopropyl ethers, iso-
(2)
propyl esters, isopropyl amines, etc. are manufactured using IPA.
Moreover, IPA can be used to produce cumene by the alkylation of Dehydration of diacetone alcohol to mesityl oxide
benzene. This will reduce the demand of propylene for phenol pro- CH3 C(O)CH2 C(OH)(CH3)2 (CH3)C(O)CH = C(CH3) 2 H
duction, which in turn will reduce our dependence on petroleum for
= 21 kJ/mol (3)
propylene. The commercial IPA manufacturing process comprises in-
direct and direct hydration of propylene. The indirect hydration process Hydrogenation of mesityl oxide to methyl isobutyl ketone
causes separation and corrosion problems as it uses sulphuric acid as
the catalyst. The direct process is less corrosive but, it requires pure CH3 C(O)CH = C(CH3)2 + H2 (CH3)C(O)CH2 CH(CH3) 2 H
propylene unlike indirect process, which uses dilute propylene from = 21 kJ/mol (4)
refinery stream. Therefore, catalytic hydrogenation of acetone over The catalysts which showed high IPA selectivity are Raney nickel

⁎
Corresponding author.
E-mail address: [email protected] (N.C. Pradhan).

https://doi.org/10.1016/j.cattod.2019.07.051
Received 24 May 2019; Received in revised form 12 July 2019; Accepted 29 July 2019
Available online 30 July 2019
0920-5861/ © 2019 Elsevier B.V. All rights reserved.
S. Basu and N.C. Pradhan Catalysis Today 348 (2020) 118–126

[3–5], nickel-copper alloys [6], supported noble metals (Pt, Pd, Ru, Rh, calculated from the nitrogen adsorption isotherms using
Ir) [7–12] and copper chromite [13]. There are certain demerits of the Brunauer–Emmett–Teller (BET) method. The Barret–Joyner–Halenda
preferred catalysts such as pyrophoric nature of Raney nickel (causing (BJH) method was applied to determine average pore diameter. The
handling problems), expensive noble metals and toxic nature of chro- analysis was done at −196 °C on 200 mg of sample. Before analysis, the
mites (environmental hazard). A lot of research work has been carried sample was degassed for 2 h under high vacuum at 200 °C.
out to explore promising hydrogenation catalysts that are less harmful. H2-TPR (Temperature programmed reduction) was conducted on a
In recent years, researchers have identified many potential compounds Quantachrome instrument (Model no. CHEMBET 3000) equipped with
showing hydrogenation activity. A lot of work has been done with si- a thermal conductivity detector (TCD). Initially, 15 mg of sample was
lica, alumina and hydrotalcite supported metals [14–20]. Recently, Yan loaded in a quartz tube and degassed at 250 °C under nitrogen flow. In
et al. [21] investigated hydrogenation of furfural and levulinic acid the next step, the temperature was increased at a rate of 10°/min from
over copper based hydrotalcites. Zhou et al. [22] prepared a series of 30 °C to 1000 °C, in a flow of 5% H2/95%N2 (80 ml/min) and the TCD
Cu-Mg-Al hydrotalcites and studied hydrogenation of furfural in aqu- signal was recorded simultaneously.
eous phase. However, acetone hydrogenation to IPA over copper based The FT-IR spectra of various samples were obtained with
mixed oxides derived from hydrotalcites is not reported so far. The PerkinElmer (Model-Spectrum 100) instrument. For FTIR analysis, very
hydrotalcite catalysts can be prepared from cheap precursors. They are thin disks comprising sample and KBr have been used in the medium IR
having remarkable properties, which include appreciable surface area, region (4000–450 cm−1).
tailorable acidic and basic properties, high metal dispersion, stability Temperature programmed desorptions (TPD) of NH3 and CO2 were
against sintering, etc. The present work was, therefore, undertaken to performed to determine the acidic and basic sites. The analysis was
explore the efficacy of bimetallic Cu-Al mixed oxide in converting done with the same instrument (Quantachrome instrument, Model no.
acetone to IPA by hydrogenation. A further objective of this work is to CHEMBET 3000) as used for H2-TPR analysis. In a quartz tube, the
establish the optimum copper loading of such Cu-based hydrogenation samples (8 mg) were loaded, degassed at 250 °C under He flow and
catalyst. cooled to room temperature. The ammonia (20% NH3 in He) and
carbon dioxide was adsorbed at room temperature for 20 min. The
2. Experimental NH3/CO2 desorption was observed by increasing the temperature from
30 °C to 900 °C, in a flow of He (80 ml/min) while the TCD signal was
2.1. Materials recorded.

Aluminium nitrate nonahydrate [Al(NO3)3.9H2O], Copper nitrate 2.4. Catalyst activity test
trihydrate [Cu(NO3)2.3H2O] and Sodium carbonate were supplied by
Merck (India) Ltd., Mumbai, India. Sodium hydroxide was purchased Acetone hydrogenation in vapor phase was conducted in a packed
from Qualigens Fine Chemicals Limited. Acetone was purchased from bed reactor. Fig. 1 displays the schematic of the set-up for hydrogena-
Merck Specialities Pvt. Ltd., Mumbai, India. All the chemicals used in tion experiments. The set-up had been purchased from Texol Engg. Pvt.
this study were of analytical reagent grade. Ltd. (Pune, India). It consists of a pre-heater for vaporizing acetone and
a reactor-furnace assembly in the upstream section and the downstream
2.2. Catalyst preparation section comprises a condenser and a gas liquid separator. The reactor
(10 mm ID) is heated electrically. Firstly, the reactor was loaded with
Co-precipitation method was used to prepare the catalysts. Initially the catalyst (1 g) at the center and inert packing materials at the top and
a solution comprising copper nitrate and aluminum nitrate in 100 ml bottom regions. After loading, the temperature was increased till 350 °C
distilled water was prepared. The salt solution was added dropwise to (reduction temperature) under nitrogen flow (40 ml/min). After re-
aqueous solution of sodium carbonate at 65 °C. This was followed by duction for 1 h in the presence of pure hydrogen (50 ml/min), reaction
dropwise addition of sodium hydroxide for maintaining pH of around was started. The product stream after reaction was condensed and se-
10. The final mixture was kept under stirring for 1 h at 65 °C. After parated. The liquid product was obtained from the bottom and analyzed
stirring, the sample was aged at room temperature for 2 h. The ageing by SOLGEL wax capillary column in Gas Chromatograph equipped with
process was followed by filtration. The slurry obtained was washed FID. Nitrogen was used as the carrier gas. The injector and detector
with hot distilled water several times to remove sodium. The slurry was temperatures were 200 °C and 250 °C, respectively. The column tem-
dried at 100 °C overnight. After drying, the solid mass was crushed and perature was programmed from 40 °C to 120 °C in the following way:
calcined at 400 °C for 4 h. The samples of four different Cu:Al atomic
40 oC (0.5 min) 80 oC (2 min) 120 oC (10 min)
ratios (0.25, 0.5, 0.75, 1.0) were prepared and labeled as CuAl0.25, 10 oC/min 40 oC/min
ramp ramp
CuAl0.5, CuAl0.75 and CuAl1.0.
The acetone conversion and IPA selectivity were calculated as fol-
2.3. Catalyst characterization lows:
FC3H6O, in FC3H6O, out
The crystalline phases present in the catalysts were identified by Conversion of acetone (%) = × 100,
FC3H6O, in (5)
analyzing Powder X-ray diffraction (XRD) patterns obtained with a
diffractometer (Model: PW-3050/60 PAN Analytical). CuKα radiation Where, FC3H6O,in = moles of acetone entering per min
was used at 30 mA and 40 kV with 100 to 800 scanning angle. The peaks FC3H6O, out = moles of acetone leaving per min
obtained were compared with standard patterns presented in Joint
IPA (moles)
Committee on Powder Diffraction Standards (JCPDS) files-1996. IPA Selectivity (%) = × 100
IPA (moles) + MIBK (moles) (6)
The FESEM images were taken at an acceleration voltage of 5 kV by
using JEOL instrument (Model: JSM-7610 F). The compositional ana-
lysis was done with EDS at an accelerating voltage of 15 kV by using 3. Results and discussion
Oxford instrument (Model: 51/XNX1146). All the samples were sub-
jected to platinum coating in a vacuum sputter coater before scanning. 3.1. Catalyst characterization
Nitrogen physisorption technique was performed in a
Quantachrome AUTOSORB 1 instrument to determine average pore The XRD patterns of calcined catalysts are shown in Fig. 2a. The
diameter, pore volume and surface area. The specific surface area was crystalline phases obtained are confirmed by standard JCPDS files. The

119
S. Basu and N.C. Pradhan Catalysis Today 348 (2020) 118–126

Fig. 1. Schematic diagram of experimental set-up.

phases detected in all the calcined samples are Cu (I) oxide (Cu2O) and method and obtained porosity values in the range 10–25 nm. The
Cu (II) oxide (CuO). The Cu (I) oxide phase belongs to the cubic crystal average pore diameter is sufficient for the reactants and products to
system and shows reflection at 29.58° (JCPDS 782076). The Cu (II) pass through.
oxide phase belongs to the monoclinic crystal system and shows re- H2-TPR analysis was carried out to determine the presence of re-
flections at 35.46° (0 0 2) (JCPDS 89–2529), 38.66° (1 1 1) (JCPDS ducible species. As can be observed in Table 3, the hydrogen con-
89–2529), 48.70° (2 0 2) (JCPDS 89–2530), 58.33° (2 0 2) (JCPDS sumption increased with copper loading. Fig. 4displays the TPR pro-
89–5899), 61.40° (1 1 3) (JCPDS 89–2530), 66.07° (3 1 1) (JCPDS files. The reduction peak for all catalysts except the one having highest
89–2530) and 67.97° (1 1 3) (JCPDS 89–2530). As the calcination copper loading appeared in the temperature range of 550–600 °C. The
temperature is 400 °C, the formation of Al2O3 or well defined Cu-Al catalyst having highest copper loading gave reduction peak at 500 °C.
mixed metal oxide phases is of least possibility [23,24]. Due to low The peak shift happened due to decrease in aluminium content. There is
calcination temperature, the aluminium species and Cu-Al mixed oxide strong interaction between copper and aluminium species. The strong
species are likely to be present in amorphous form. In Figure 2b, a interaction might be the cause for high reduction temperature when
comparison has been made for the phases present in CuAl0.5 at dif- aluminium content is high. The peak at around 200 °C corresponds to
ferent conditions. Before calcination, the different phases present are the reduction of CuO species [26].
copper hydroxide nitrate (29.45° (0 2 0), 44.60° (1 2 2), JCPDS Fig. 5 shows the FTIR spectra of the calcined samples. A broad peak
81–1697), copper aluminum carbonate hydroxide hydrate (35.67° (2 1 at 3500 cm−1 has appeared. It can be assigned to the stretching vi-
1), JCPDS 46-0099), aluminum hydroxide (Gibbsite) (38.40° (1 2 1), brations of absorbed water. The band at about 600 cm-1 corresponds to
JCPDS 76–1782) and copper hydroxide carbonate (Malachite) (53.01° Cu-O stretching [27] whereas the bands at 800 cm−1 and 1300 cm−1
(2 5 1), JCPDS 75–1163 and 48.18° (3 4 1), JCPDS 76-0660). All the indicate the presence of residual carbonate anions [23,28].
phases decomposed during calcination. After reaction, the spent cata- The NH3-TPD profiles are shown in Fig. 6. A single desorption peak
lyst consists of only reduced copper phase (43.31° (1 1 1), 50.44° (2 0 has been obtained for all the samples. The quantity of ammonia des-
0), 74.12° (2 2 0), JCPDS 89–2838). The reduction temperature (350 °C) orbed is shown in Table 4. There is a shift in the peak for high copper
was high enough for complete reduction of CuO to Cu phase [25]. loadings. The incorporation of CuO phase enhances the Lewis acidity
The FESEM images of calcined samples were obtained in order to resulting into peak shift. Highest acidity was obtained for CuAl0.25.
determine surface morphologies. The images are shown in Fig. 3(a, b, c The acidity reduced when the Cu:Al atomic ratio increased to 0.5. The
& d). The images were taken at a magnification of 40,000 and a re- acidity dropped probably due to the reduction in aluminium content.
solution of 100 nm. The sheet like morphology is clearly visible in the There is an increase in acidity for further higher loading; 22.5 mmol
picture. The agglomeration increases as the copper content increases. NH3/g cat. for CuAl0.75. This might have happened due to increase in
The EDS result in Table 1 shows that the actual atomic ratio of copper CuO phase. In order to evaluate the presence of basic sites, CO2-TPD
and aluminum of different sample is close to the targeted ratio. was conducted. The CO2-TPD profiles are shown in Fig. 7. A quantita-
The average pore diameter, pore volume and surface area of the tive estimation of carbon dioxide desorbed is presented in Table 5.
calcined samples are given in Table 2. It is found that when the copper There is a broad desorption peak in the temperature region of
loading increases the surface area decreases. The sample CuAl0.25 has 400–750 °C for all samples. The basicity of the catalyst increased a bit
the highest surface area. At higher copper loading, dispersion is less. with copper content. The desorption peak shifted slightly as the copper
Therefore, the surface area of the samples decreased. All the prepared content increased. The presence of acidic and basic properties in Cu/
samples have mesoporous texture having porosity in the range 6–9 nm. Al2O3 was also reported in the literature [29].
The porosity range for mesoporous materials is 2–50 nm, as per Inter-
national Union of Pure and Applied Chemistry (IUPAC). The presence of 3.2. Effect of catalyst composition
mesoporosity is also reported by Alejandre et al. [23]. They have stu-
died the preparation of Cu-Al mixed oxides using co-precipitation The effects of Cu:Al atomic ratio on conversion of acetone and IPA

120
S. Basu and N.C. Pradhan Catalysis Today 348 (2020) 118–126

mesityl oxide. Also, the hydrogenation of C]C double bond is ther-

modynamically favored over C]O double bond. Therefore, at high
metal content, the IPA selectivity decreases. As the highest selectivity
for IPA was displayed by CuAl0.5 catalyst, it was chosen for further
study.

3.3. Effect of temperature

The influence of temperature was investigated by conducting the

experiments at 150, 175, 200 and 225 °C, respectively. The conversion
of acetone increases with temperature as shown in Fig. 10. The se-
lectivity of IPA, however, increases first, attains a maximum at 175 °C
and then decreases sharply with temperature (Fig. 10). The dehydration
of acetone condensation product (diacetone alcohol) to mesityl oxide is
an endothermic reaction. Therefore, a rise in temperature favors MIBK
formation. Similar kind of behavior was observed in another work using
Cu-on-MgO catalyst as reported by Chikan et al. [30]. They observed
that formation of MIBK increases with temperature and at low tem-
perature direct hydrogenation to IPA dominates. This causes the con-
version to increase with temperature at the cost of IPA selectivity. At
low temperature, the activity of the catalyst is low; so, conversion and
selectivity are less. The IPA selectivity was found to increase with time
(Fig. 11). The deactivation of acidic and basic sites of catalyst with time
causing less MIBK production is responsible for this behavior. For more
corroboration of this hypothesis, one experiment was conducted
without catalyst at 175 °C. Though the conversion was very less (6%),
IPA was obtained as the only product of the reaction. This proves that
the presence of basic and acidic sites in the catalyst promotes the other
reactions leading to MIBK. As highest selectivity for IPA was obtained at
175 °C, this was considered as the optimum temperature.

3.4. Effect of feed composition

The effect of feed composition was studied by varying hydrogen to

acetone molar ratio at 175 °C. The results are presented in Fig. 12. The
conversion of acetone increased with mole ratio. However, the se-
lectivity of IPA was observed to decrease at hydrogen to acetone mole
ratios > 1. Due to the presence of multifunctional sites, there always
exists a competition between direct hydrogenation and mesityl oxide
formation (and hydrogenation). At higher H2/acetone mole ratios, the
acetone partial pressure becomes low. At low acetone partial pressure,
the metallic sites cannot counter the activity of basic and acidic sites
which induce condensation and dehydration reactions leading to the
formation of mesityl oxide. Again, as the reduction of C]C double bond
is easier than that of C]O double bond, hydrogenation at higher hy-
drogen concentration produces more MIBK than IPA. A similar behavior
has been reported by Rodrigues and Monteiro [31]. As highest IPA
selectivity was obtained at the mole ratio of 1, this was considered as
the optimum mole ratio for the present catalyst.

3.5. Effect of space-velocity

The effect of space-velocity was studied at 150 °C, 175 °C and 200 °C
Fig. 2. a XRD patterns of samples after calcination. (•) CuO. 2b XRD patterns of
with hydrogen to acetone molar ratio of unity. The results are shown in
uncalcined CuAl0.5, calcined CuAl0.5 and spent CuAl0.5. (•) CuO, (□) Cu, (+)
Figs. 13 and 14. The conversion decreases with space-velocity as ex-
Copper aluminium carbonate hydroxide hydrate, (o) Gibbsite Al(OH)3, (*)
Malachite Cu2CO3(OH)2, (Δ) Cu2(OH)3NO2.
pected. The selectivity of IPA increases with space-velocity. The trend is
same for all temperatures. This indicates that the rate of direct hydro-
genation of acetone to IPA is enhanced with space-velocity. This also
selectivity were studied at 175 °C. The conversion of acetone was found corroborates the observation from the run without any catalyst in the
to increase with increase in concentration of Cu in the catalyst (Fig. 8). reactor. At low space-velocity, other reactions are more likely to occur
A maximum conversion of 25% was achieved with sample having Cu/Al that lead to the formation of MIBK.
atomic ratio of unity. This is due to the fact that as the metal content
increases, the hydrogenation activity also increases. In case of IPA se- 3.6. Kinetic modelling
lectivity (Fig. 9), the catalysts did not follow the same trend. CuAl0.5
showed highest selectivity (98%). The high metal loading and presence For development of suitable kinetic model, experiments were con-
of acidic as well as basic sites favors the formation of MIBK through ducted under conditions free from any diffusional limitations, at three

121
S. Basu and N.C. Pradhan Catalysis Today 348 (2020) 118–126

Fig. 3. FESEM images of (a) CuAl0.25, (b) CuAl0.5, (c) CuAl0.75, (d) CuAl1.0.

Table 1 Table 3
Cu:Al (atomic ratio) of different catalysts. Hydrogen consumption in H2-TPR.
Catalyst Targeted Actual Catalyst H2 Consumption (mmol/g)

CuAl0.25 0.25 0.18 CuAl0.25 5.4

CuAl0.5 0.5 0.46 CuAl0.5 6.1
CuAl0.75 0.75 0.67 CuAl0.75 6.9
CuAl1.0 1.0 0.81 CuAl1.0 7.0

N
Table 2
Physical properties of the catalysts. F= (rcali rexpi )2
i= 1 (7)
Catalyst BET surface area Average pore diameter Pore Volume (cm3/
(m2/g) (nm) g) where, rcali = Calculated rate, rexpi = Experimental rate.
For establishing L-H-H-W model, the following mechanism was
CuAl0.25 116.6 6.04 0.22 proposed with dissociative adsorption of hydrogen and competitive
CuAl0.5 99.0 6.88 0.17
CuAl0.75 69.8 7.64 0.08
adsorption of hydrogen and acetone [5,9]:
CuAl1.0 37.4 8.56 0.01 k1
1.H2 + 2S 2H.S
k1

temperatures namely, 150, 175 and 200 °C, keeping hydrogen to k2

acetone molar ratio constant at unity. The space-velocity was varied 2.A + S A.S (A: Acetone )
from 0.09 to 0.145 kmol acetone/ kg cat. h. The kinetic data obtained
k2

are presented in Tables 6–8. The kinetic data were used in developing k3
Langmuir-Hinshelwood-Hougen-Watson (L-H-H-W) type model. To es- 3. A.S + H.S AH.S + S
k3
timate the unknown parameters, nonlinear regression based on Le-
venberg-Marquardt algorithm was used in the POLYMATH software. k4
The objective function was minimized by using the following equation: 4. AH.S + H.S B.S + S(Rate-determining step)

122
S. Basu and N.C. Pradhan Catalysis Today 348 (2020) 118–126

Fig. 4. H2 TPR profiles of (a) CuAl0.25, (b) CuAl0.5, (c) CuAl0.75, (d) CuAl1.0.
Fig. 6. Ammonia TPD profiles of (a) CuAl0.25, (b) CuAl0.5, (c) CuAl0.75, (d)
CuAl1.0.

Table 4
Ammonia desorbed in TPD analysis.
Catalyst NH3 desorbed (mmol/g cat)

CuAl0.25 34
CuAl0.5 16.2
CuAl0.75 22.5
CuAl1.0 25

Fig. 5. FTIR spectra of (a) CuAl0.25, (b) CuAl0.5, (c) CuAl0.75, (d) CuAl1.0.

k5
5. B.S + S B + 2S(B: Isopropyl alcohol)
k5

Initially, rate equations were derived considering each elementary

step one by one as rate-determining step. As the conversion of acetone
is low, the partial pressure of product has been neglected. The final
forms of rate equations are presented below.
Model 1: Dissociative adsorption of H2 is rate controlling

k PH2
r1 =
(1 + K2 PA )2 (8)
Fig. 7. Carbon dioxide TPD profiles of (a) CuAl0.25, (b) CuAl0.5, (c) CuAl0.75,
Model 2: Adsorption of acetone is rate controlling (d) CuAl1.0.

k PA
r2 = (k K2 PA K1 PH2
1 + K1 PH2 (9) r3 =
(1 + K2 PA + K1 PH2 )2 (10)
Model 3: First surface reaction is rate controlling

123
S. Basu and N.C. Pradhan Catalysis Today 348 (2020) 118–126

Table 5
Carbon dioxide desorbed in TPD analysis.
Catalyst CO2 desorbed (mmol/g cat)

CuAl0.25 6.52
CuAl0.5 7.79
CuAl0.75 9.04
CuAl1.0 8.35

Fig. 10. Effects of temperature on acetone conversion and IPA selectivity.

Conditions: Cu:Al (atomic), 0.5; Hydrogen: Acetone (molar), 1; Space-velocity,
0.126 kmol acetone/kg cat.h)

Fig. 8. Hydrogenation activity of Cu-Al mixed oxides at different Cu/Al atomic

ratio.
Conditions: Temperature, 175 °C; Hydrogen: Acetone (molar), 1; Space-velo-
city, 0.126 kmol acetone/kg cat. h

Fig. 11. Variation of IPA selectivity with time at different temperatures.

Conditions: Cu:Al (atomic), 0.5; Hydrogen: Acetone (molar), 1; Space-velocity,
0.126 kmol acetone/kg cat.h

Therefore, second surface reaction was found to be the rate-determining

step. The evaluated rate constant, k and equilibrium constants (K1, K2
and K3) are presented in Table 9. It can be observed that the rate
constant increases with temperature and the equilibrium constants
decrease with temperature as expected. From Arrhenius plot (Fig. 15),
Fig. 9. Selectivity of IPA as a function of time for different Cu:Al atomic ratio. activation energy and frequency factor were calculated to be 44.3 kJ/
Conditions: Temperature, 175 °C; Hydrogen: Acetone (molar), 1; Space-velo- mol and 40134.83 kmol/kg cat. h. Sen et al. [9] obtained around 60 kJ/
city, 0.126 kmol acetone/kg cat.h mol of activation energy over platinum based catalysts at 30 °C and
atmospheric pressure.
Model 4: Second surface reaction is rate controlling

k K1 K2 K3 PA PH2 4. Conclusions
r4 =
(1 + K2 PA + K1 PH2 + K2 K3 PA K1 PH2 ) 2
(11) The developed CuAl mixed oxide catalysts showed good activity for
Model 5: Desorption of IPA is rate controlling hydrogenation of acetone in vapor phase. The experimental runs were
conducted by varying parameters such as temperature, H2/Acetone
k K1 K2 K3 K 4 PA PH2 molar ratio and space-velocity. It was observed that the presence of
r5 =
1 + K2 PA + K1 PH2 + K2 K3 PA K1 PH2 + K1 K2 K3 K 4 PA PH2 metallic, basic and acidic sites on the catalyst play a significant role in
determining the product selectivity. Due to the presence of basic and
(12)
acidic sites, acetone condensation and dehydration also take place apart
The experimental kinetic data fit rate Eq. (11) satisfactorily. from direct hydrogenation catalyzed by metal sites. The condensation

124
S. Basu and N.C. Pradhan Catalysis Today 348 (2020) 118–126

Table 6
Results of kinetic run at 150 °C.
PA x 103 PH2 x 103 Reaction Rate x 103 Conversion (%) W/F
(atm) (atm) (kmol/kg cat. h) (kg cat. h/
kmol)

426.5 573.5 14.38 9.77 6.80

429.7 570.3 14.37 9.09 6.33
431.1 568.8 14.35 8.80 6.13
432.6 567.4 14.34 8.50 5.92
433.6 566.4 14.32 8.27 5.78
435.1 564.9 14.30 7.95 5.56
436.2 563.8 14.28 7.72 5.41
437.4 562.6 14.26 7.50 5.24
438.3 561.7 14.24 7.30 5.12
439.4 560.6 14.22 7.05 4.96

Table 7
Fig. 12. Variation of acetone conversion and IPA selectivity with feed compo- Results of kinetic run at 175 °C.
sition.
PA x 103 PH2 x 103 Reaction Rate x 103 Conversion (%) W/F
Conditions: Temperature, 175 °C; Cu:Al (atomic), 0.5; Space-velocity,
(atm) (atm) (kmol/kg cat. h) (kg cat. h/
0.126 kmol acetone/kg cat. h kmol)

449.8 550.2 14.95 10.16 6.80

453.4 546.5 14.90 9.43 6.33
455.0 544.9 14.88 9.12 6.13
456.6 543.4 14.87 8.80 5.92
457.7 542.3 14.85 8.60 5.78
459.4 540.6 14.83 8.25 5.56
460.6 539.4 14.81 8.01 5.41
461.8 538.2 14.80 7.76 5.24
462.8 537.2 14.78 7.56 5.12
464.0 535.9 14.76 7.32 4.96

Table 8
Results of kinetic run at 200 °C.
PA x 103 PH2 x 103 Reaction Rate x 103 Conversion (%) W/F
(atm) (atm) (kmol/kg cat. h) (kg cat. h/
kmol)

472.9 527.0 15.48 10.52 6.80

476.9 523.0 15.43 9.76 6.33
478.7 521.3 15.41 9.44 6.13
480.4 519.5 15.39 9.11 5.92
481.6 518.3 15.37 8.90 5.78
Fig. 13. Effect of space-velocity on acetone conversion. 483.4 516.5 15.36 8.54 5.56
Conditions: Hydrogen: Acetone (molar), 1; Cu:Al (atomic) = 0.5 484.8 515.2 15.33 8.30 5.41
486.2 513.8 15.32 8.03 5.24
487.2 512.8 15.29 7.83 5.12
488.5 511.4 15.27 7.57 4.96

Table 9
Rate constant and equilibrium constants for rate Eq. (11).
Temperature (oC) k K1 K2 K3 R2
(kmol/kg cat. h) (atm−1) (atm−1)

150 0.145 5.18 2.52 0.97 0.91

175 0.254 1.62 1.91 1.0 0.98
200 0.521 0.24 0.92 2.63 0.96

product is further dehydrated and hydrogenated to MIBK. As the de-

hydration reaction is endothermic, high temperature favored the se-
lectivity of MIBK. The optimum temperature at which highest se-
lectivity of IPA obtained was 175 °C. The catalyst showed maximum
selectivity for IPA at H2/Acetone mole ratio of unity. Beyond this mole
ratio the IPA selectivity declined. The study of variation in space-ve-
locity suggested that a high space-velocity favors IPA selectivity. The
Fig. 14. Effect of space-velocity on IPA selectivity.
best activity was obtained over CuAl0.5. The kinetic study was con-
Conditions: Same as in Fig. 13. ducted with the best catalyst. The rate equations based on LHHW

125
S. Basu and N.C. Pradhan Catalysis Today 348 (2020) 118–126

[7] F. Rositani, S. Galvagno, Z. Poltarzewski, P. Staiti, P.L. Antonucci, J. Chem.

Technol. Biotechnol. 35 1985 234-240.
[8] N.V. Pavlenko, A.I. Tripol’skii, G.I. Golodets, Theor. Exp. Chem. 22 (1987)
667–675.
[9] B. Sen, M.A. Vannice, J. Catal. 113 (1988) 52–71.
[10] J. Cunningham, S. O’Brien, J. Sanz, J.M. Rojo, J.A. Soria, J.L.G. Fierro, J. Mol.
Catal. 57 (1990) 379–396.
[11] K.J. Williams, A.B. Boffa, J. Lahtinen, M. Salmeron, A.T. Bell, G.A. Somorjai, Catal.
Lett. 5 (1990) 385–394.
[12] A. Balouch, A.A. Umar, A.A. Shah, M.M. Salleh, M. Oyama, ACS Appl. Mater.
Interfaces 5 (2013) 9843–9849.
[13] T.M. Yurieva, L.M. Plyasova, O.V. Makarova, T.A. Kreiger, J. Mol. Catal. A Chem.
113 (1996) 455–468.
[14] S. Narayanan, R. Unnikrishnan, J. Chem. Soc. Faraday. Trans. 94 (1998)
1123–1128.
[15] L.M. Gandia, A. Diaz, M. Montes, J. Catal. 157 (1995) 461–471.
[16] P. Jin, Q. Chen, L. Hao, R. Tian, L. Zhang, L. Wang, J. Phys. Chem. B 108 (2004)
6311–6314.
[17] Y.Z. Chen, C.M. Hwang, C.W. Liaw, Appl. Catal. A Gen. 11 (1998) 207–214.
[18] N. Das, D. Tichit, R. Durand, P. Graffin, B. Coq, Catal. Lett. 71 (2001) 181–185.
[19] R.S. Rao, A.B. Walters, M.A. Vannice, J. Phys. Chem. B 109 (2005) 2086–2092.
[20] O.V. Shutkina, O.A. Ponomareva, P.A. Kots, I.I. Ivanova, Catal. Today 218 (2013)
30-40.
[21] K. Yan, J. Liao, X. Wu, X. Xie, RSC Adv. 3 (2013) 3853–3856.
[22] M. Zhou, Z. Zeng, H. Zhu, G. Xiao, R. Xiao, J. Energy. Chem. 23 (2014) 91–96.
Fig. 15. Arrhenius plot for LHHW model.
[23] A. Alejandre, F. Medina, P. Salagre, X. Correig, J.E. Sueiras, Chem. Mater. 11 (1999)
939–948.
[24] D. Redaoui, F. Sahnoune, M. Heraiz, A. Raghdi, Acta Phys. Pol. A 131 (2017)
kinetic models were validated. The activation energy was estimated to 562–565.
be 44.3 kJ/mol. [25] M.H. Yamukyan, Kh. V. Manukyan, S.L. Kharatyan, J. Alloys Compd. 473 2009 546-
549.
[26] R.B. Mane, S.E. Kondawar, P.S. Niphadkar, P.N. Joshi, K.R. Patil, C.V. Rode, Catal.
References
Today 198 2012 321-329.
[27] M. Faisal, S.B. Khan, M.M. Rahman, A. Jamal, A. Umar, Matter. Lett. 65 (2011)
[1] P. Dutia, Chem. Weekly (2012). 1400–1403.
[2] Y. Duan, M. Xu, X. Huai, J. Therm. Sci. 23 (2014) 85–90. [28] A.F. Cunha, Y.J. Wu, J.C. Santos, A.E. Rodrigues, Ind. Eng. Chem. Res. 51 (2012)
[3] S. Kishida, S. Teranishi, J. Catal. 12 (1968) 90–96. 13132–13143.
[4] N.O. Lemcoff, J. Catal. 46 (1977) 356–364. [29] H.P.R. Kannapu, Y.W. Suh, A. Narani, V. Vaddeboina, D.R. Burri, R.R.K. Seetha,
[5] N.S. Chang, S. Aldrett, M.T. Holtzapple, R.R. Davison, Chem. Eng. Sci. 55 (2000) RSC Adv. 7 (2017) 35346–35356.
5721–5732. [30] V. Chikan, A. Molnar, K. Balazsik, J. Catal. 184 (1999) 134–143.
[6] P.B. Babkova, A.K. Avetisev, G.D. Lyubarskii, A.I. Gel’bshtein, Sov. Appl. Mech. 5 [31] A.C.C. Rodrigues, J.L.F. Monteiro, Appl. Catal. A Gen. 362 (2009) 185–192.
1969 309-312.

126