HETEROGENEOUS AZEOTROPIC

DEHYDRATION OF ETHANOL TO OBTAIN

A CYCLOHEXANE-ETHANOL MIXTURE
Chemical Engineering Department
University of Alicante (Spain)
e-mail: vgomis@ua.es

Vicente Gomis
M Dolores Saquete
Alicia Font
Ricardo Pedraza
Victoria Pastor-Matea

OBJECTIVE

Study the viability of cyclohexane in

the ethanol dehydration to obtain an
ethanol + cyclohexane mixture from
an azeotropic distillation column.

EQUIFASE 2009

INTRODUCTION

Key renewable energy policy documents of the EU

1997 White Paper "Energy for the future"
Doubling the share of renewable energy from 6%
(1997) to 12% (2010)

2003 EU Biofuels Directive (2003/30/EC)
Target for biofuels in transport: 2% by 2005,
5.75% by 2010

2009 Directive on the promotion of the use of
energy from renewable sources (2009/28/EC)
Overall EU target : 20% renewable energy in
gross final energy consumption in 2020
Target of 10% renewable energy in transport in
2020 for all member states
EQUIFASE 2009

INTRODUCTION

Ethanol Production
Sectored emissions in Europe
A griculture
8%

Waste
2%
Industry
8%

Benefits of biofuels
Reduce GHG emissions
Improve air quality

Energy
48%

Reduce petroleum dependence

Transpo rt
34%

Improve energy security

EQUIFASE 2009

INTRODUCTION

Ethanol Production

EQUIFASE 2009

INTRODUCTION

Ethanol dehydration

Pressure Swing Adsorption

EQUIFASE 2009

Azeotropic distillation

INTRODUCTION

Azeotropic distillation
TERNARY
AZEOTROPE (E)

1 00

Ethanol

D
G
(M)
FEED

25

75

(N)

(G)

50

50

Ethanol/Water (D)
+
Benzene (B)

25

75

ABSOLUTE
(C)
ALCOHOL

Heterogeneous region
0

100

Water

EQUIFASE 2009

B
100

75

50

25

Benzene

INTRODUCTION

Possible entrainers
Benzene (Young, 1902)
Acetone
Hexane
Heptane
Toluene
Isooctane
Cyclohexane
EQUIFASE 2009

Gasoline components

Pentane

INTRODUCTION

Conventional Process

Raw
materials

Ethanol
Production

Mixing
with
gasoline

Fuel

Proposed Process

Raw
materials

EQUIFASE 2009

Ethanol + Gasoline
Production

Fuel

INTRODUCTION

Cost diminution of:

- Mixing

Advantages
- Transportation

- Storage
EQUIFASE 2009

EXPERIMENTAL DESIGN
Study in an experimental
semi-pilot plant column

EQUIFASE 2009

Simulation of the industrial process

Semi-Pilot Plant Column study

Cyclohexane

EQUIFASE 2009

Semi-Pilot Plant Column study

Operation Variables
Feed 1: pure cyclohexane. Temperature = 66 1C
Flow rate = 41.00 g/min
Feed 2: water + ethanol mixture (94% wt. of ethanol). Temperature: 63 1C
Flow rate = 4.38 g/min
Condenser: Temperature = 35C
Heat exchanger 3: Temperature of the stream leaving HE-3 = 66 1C

Simulation Variables
 Simulated in Chemcad 6
 Rigorous calculation using the SCDS module (simultaneous correction
method for rigorous fractionation simulation)
 Thermodynamic model: UNIFAC

EQUIFASE 2009

Semi-Pilot Plant Column study

Bottoms Product

1.0

Weight Fraction

0.8

0.6

The trends
observed in the
experimental
results resemble
their simulated
counterparts

Ethanol Simulation
Cyclohexane Simulation
Cyclohexane
Ethanol

0.4

0.2

Optimal

0.0
0

50

100

150

200

250

Reboiler Heat Duty (W)

The ethanol concentration depends on the heat duty

Only values ranging from 80-100 W permit ethanol concentrations close to 5 % wt.

EQUIFASE 2009

Semi-Pilot Plant Column study

Bottoms Product
0.005

Water Simulation
Water

Weight Fraction

0.004

The trends
observed in the
experimental
results resemble
their simulated
counterparts

0.003

0.002

0.001

Too high

< 50ppm

0.000
0

50

100

150

200

250

Reboiler Heat Duty (W)

The concentration of water in the residue stream does vary considerably with respect to
the reboiler heat duty
As the heat duty increases, the concentration of the water gradually decreases, reaching
values lower than 50 ppm.

EQUIFASE 2009

Semi-Pilot Plant Column study

Aqueous phase

Water Simulation
Ethanol Simulation
Water Simulation
Water
Ethanol
Cyclohexane

1.0

Weight Fraction

0.8

0.6

0.4

0.2

0.0
0

50

100

150

Reboiler Heat Duty (W)

200

250

The simulation adequately

reproduces neither the flow
rate values of the bottom
product and aqueous layer
obtained experimentally nor
the composition of the
streams

The composition of the aqueous layer is also dependent on the heat duty
The composition tends to approach that of the plait point of the system.

EQUIFASE 2009

Semi-Pilot Plant Column study

Flows
50

Flow (g/min)

40

30

Simulation
Aqueous decant
Bottoms Product

20

10

0
0

50

100

150

Reboiler Heat Duty (W)

200

250

The simulation
adequately reproduces
neither the flow rate
values of the bottom
product and aqueous
layer obtained
experimentally nor the
composition of the
streams

The flow rate of the residue is always higher than that of the aqueous layer
Both flow rates become more similar when the reboiler heat duty increases.

EQUIFASE 2009

Semi-Pilot Plant Column study

10
0

Ethanol

UNIFAC non isothermal binodal curve

Experimental non isothermal binodal curve

75

25

50

50

75

25

UNIFAC phase split prediction

Plait Point

0
10

Water 100

75

50

EQUIFASE 2009

25

Cyclohexane

Semi-Pilot Plant Column study

100

Etanol
Experimental
UNIFAC
UNIFAC Dortmund
UNIFAC LLE

75

25

50

50

25

75

0
10

Agua

100

75

EQUIFASE 2009

50

25

Ciclohexano

Semi-Pilot Plant Column study

10
0

Etanol
Experimental
UNIQUAC
NRTL constante
NRTL variable

75

25

50

50

25

75

0
10

Agua

100

75

EQUIFASE 2009

50

25

Ciclohexano

CONCLUSIONS
 It is possible, through azeotropic distillation, to obtain a mixture of cyclohexane
+ ethanol with concentrations of water lower than 50 ppm without the need
to distill absolute ethanol beforehand. Afterward, the mixture could be directly
employed as a carburant in car engines with no further modifications.
 The most critical parameter of the process is the reboiler heat duty. At lower
values, this produces a mixture of cyclohexane + ethanol with excessive
amounts of water. Whereas, at higher values the azeotropic distillation column
does not work properly, since the top stream condenses giving only one liquid
phase.
 Significant differences in some values are encountered between experimental
and simulated data which can be attributed to the calculation of the liquid-liquid
equilibrium. It is therefore necessary to improve the correlation of the
experimental equilibrium data for determined regions of the ternary system
diagram.

EQUIFASE 2009

CONCLUSIONS

The production of dry mixture of

ethanol + cyclohexane seems to be
technically and economically viable

EQUIFASE 2009