kne

J.

cien

cked bed membrane reactors, has been compared for ultra-pure

profiles have been compared. The extent of mass and heat transfer limitations in the

different reactors has been evaluated, and strategies to decrease (or avoid) these limita-

different process steps such as feed gas preheating and pre-

tion) unit is used to achieve the desired hydrogen purity. In

view of thermodynamic limitations and the high endo-

thermicity of steam reforming, heat transfer at high temper-

atures (850950 C) is required, where excess of steam is usedto avoid carbon deposition (typical feed H2O/CH4 molar ratios

process units can be strongly decreased and the total required

based membrane reactors for pure hydrogen production

have been proposed in literature for different reaction

systems such as methanol reforming [8,9], ethanol reforming

[10], and autothermal reforming [11]. In particular, Gallucci

and Basile [9] have demonstrated the feasibility of packed bed

* Corresponding author. Tel.: 31 53 489 2370; fax: 31 53 489 2882.

Avai lab le at www.sc iencedi rect .com

w.

i n t e r n a t i on a l j o u r n a l o f h y d r o g e n en e r g y 3 5 ( 2 0 1 0 ) 7 1 4 2 7 1 5 0E-mail address: F.Gallucci@tnw.utwente.nl (F. Gallucci).treatment (for example hydrodesulphurisation), primary and

secondary reformers (often multi-tubular fixed-bed reactors)

and high and low temperature shift converters, CO2 removal

and methanation units. Often a PSA (Pressure Swing Adsorp-

reactor volume can be largely reduced, while higher methane

conversions and hydrogen yields beyond thermodynamic

equilibrium limitations can be achieved, at lower tempera-

tures and with higher overall energy efficiencies [47]. Pd-The recent advances in Polymer Electrolyte Membrane Fuel

Cells (PEMFC) for small or medium scale applications make

the production of ultra-pure hydrogen a challenging topic in

energy conversion. On an industrial scale, most of the

hydrogen is currently produced via steam reforming of

methane (SRM). The traditional SRM process consists of

complex heat integration and the associated uneconomical

downscaling make this route inefficient. A high degree of

process integration and process intensification can be

accomplished by integrating hydrogen perm-selective

membranes in the steam reformer [2,3]. Via the integration

of hydrogen perm-selective membranes, the number ofAccepted 11 February 2010

Available online 15 March 2010

Keywords:

Hydrogen production

Membrane reactors

Fluidized bed

Heat and mass transfer limitations

1. Introduction0360-3199/$ see front matter 2010 Profesdoi:10.1016/j.ijhydene.2010.02.050tions have been proposed for the fluidized bed membrane reactor concept. The results

show that the packed bed membrane reactor requires in some conditions double

membrane area with respect to the fluidized bed membrane reactor.

2010 Professor T. Nejat Veziroglu. Published by Elsevier Ltd. All rights reserved.

25) [1]. For the production of ultra-pure hydrogen for small

scale application, the large number of process units withReceived in revised form

8 February 2010hydrogen production via methane reforming. Using detailed theoretical models, the

required membrane area to reach a given conversion and the prevailing temperatureReceived 16 November 2009 fluidized bed and paArticle history: In this theoretical work the performance of different membrane reactor concepts, bothTheoretical comparison of pacmembrane reactors for metha

Fausto Gallucci*, Martin Van Sintannaland,

Fundamentals of Chemical Reaction Engineering Group, Faculty of S

Enschede, The Netherlands

a r t i c l e i n f o a b s t r a c t

journa l homepage : wwsor T. Nejat Veziroglu. Pued bed and fluidized bedreforming

A.M. Kuipers

ce and Technology, (IMPACT) University of Twente,

e lsev ie r . com/ loca te /heblished by Elsevier Ltd. All rights reserved.

self supportedmembrane reactors for different fuels. Simakov

and Sheintuch [11] have developed a small scale autothermal

membrane reformer by coupling an exothermic reaction

(carried out in a separate compartment of the reactor) with the

endothermic methane reforming with hydrogen recovery

through Pd-based membranes.

Both packed bed membrane reactors [12,13] and fluidized

bed membrane reactors [1416] have already been presented

in literature for the reforming ofmethane and advantages and

disadvantages of both concepts have already been discussed.

However a direct quantitative comparison of the two concepts

at the same conditions is lacking. In this paper a direct

comparison between the two concepts is performed for ultra-

pure hydrogen production via methane reforming using

detailed theoretical models. The extent of mass and heat

transfer limitations in the different reactors is evaluated, and

strategies to decrease (or avoid) these limitations are

proposed.

2. Reactor configurations

order to increase the hydrogen permeation through the

membranes. With the fluidized bed membrane reactor

a virtually isothermal condition can be achieved and bed-to-

membrane mass transfer limitations are largely avoided. On

the other hand, bubble-to-emulsion phase mass transfer

limitations and the extent of gas back-mixing could deterio-

rate its performance. In particular, the use of membranes

inside the reactor could decrease the extent of back-mixing

and can also help in decreasing the bubble diameter,

enhancing the bubble-to-emulsion phase mass transfer. With

the help of a two-phase phenomenological reactor model, the

effect of bubble-to-emulsion phase mass transfer limitations

on the temperature profiles and reactor performance. The

influence of the reactor and particle dimensions is investigated.

i n t e r n a t i o n a l j o u rn a l o f h y d r o g e n en e r g y 3 5 ( 2 0 1 0 ) 7 1 4 2 7 1 5 0 71432.1. Fluidized bed membrane reactor concept

A schematic representation of the considered fluidized bed

membrane reactor configuration is shown in Fig. 1. The

methane steam reforming takes place in a fluidized bed

operated in the bubbling regime. Pure hydrogen is recovered

via dead-end Pd-based membranes inserted into the fluidized

bed. The hydrogen recovery through the membranes shifts

both the methane reforming and water gas shift reactions

towards the products resulting in higher conversion and

hydrogen yield compared with a conventional reformer.

A pressure difference between the reaction side (fluidized bed)

and the permeation side (membrane lumen) is applied inFig. 1 Scheme of the fluidized bed membrane reactor.3. Reactor models

In both reactor concepts the reactions considered are the

following:

CH4 H2O5CO 3H2 (1)

COH2O5CO2 H2 (2)Where the reaction rate expressions are taken from Numa-

guchi et al. [17]:

r1 k1pCH4pH2O p3H2pCO=Keq;1

p1:596H2O

(3)and gas back-mixing are quantified and a possible strategy to

decrease these limitations is proposed.

2.2. Packed bed membrane reactor concept

A typical tube-in-tube packed bed membrane reactor config-

uration is considered (see Fig. 2). The catalyst is packed in the

tube side of the membrane while pure hydrogen is recovered

in the shell side of the reactor. Also in this case, a pressure

difference between the reaction side and the permeation side

is applied.

The reactor has been studied with both a 1D model and

adetailed2Dmodel inorder to identify theextentofwall-to-bed

heat transfer limitations and the bed-to-membrane mass

transfer limitations (concentrationpolarization) and their effectFig. 2 Scheme of the packed bed membrane reactor.

the bubble phase, distributed according to the local bubble

fraction. The gas extracted from the emulsion phase is

Bubble phase component mass balances1

s sXnc

(8)

i n t e r n a t i on a l j o u r n a l o f h y d r o g e n en e r g y 3 5 ( 2 0 1 0 ) 7 1 4 2 7 1 5 07144subsequently instantaneously replenished via exchange

from the bubble phase (to maintain the emulsion phase at

minimum fluidization conditions) (following Deshmukh

et al. [19,20]).

A uniform temperature is assumed throughout an entiresection of the fluidized bed, assuming no heat losses to the

surroundings (adiabatic conditions) and no heat transfer

limitations between the bubble phase and the emulsion

phase [21,22].

The mass and heat balance equations are as follows:

Total mass balance

usb;n1ATrb;n1 usb;nATrb;n use;n1ATre;n1 use;nATre;n

Xnc n

f00membranei;mol Mw;iAmembraneeb;nr2 k2pCOpH2O pH2pCO2=Keq;2

pH2O

(4)

The contribution by homogeneous gas phase reactions can

safely be neglected for this reaction system.

Themembranes considered in this study are Pd-based thin

dense layer supported membranes prepared by electroless

plating by ECN (Energy research Center of the Netherlands).

The hydrogen permeation rate through the palladium

membranes follows the Richardson equation:

JH2 P0e$eEa=RT$hplumenH2

0:5pshellH2

0:5i(5)

where the values of the apparent activation energy Ea and pre-

exponential factor Pe0 (in agreement with [18]) are

12,540 Jmol1 and 2.21 1003mol s1m2 Pa0.5, respectively.

3.1. Fluidized bed membrane reactor model

A typical 1D two-phase model for a membrane assisted

fluidized bed reactor has been used for the simulation of the

fluidized bed membrane reactor based on the following main

assumptions:

Dead-end hydrogen perm-selective membranes are inte-grated in the reactor.

The reactor consists of two-phases, viz. the bubble phaseand the emulsion phase.

The gas flowing through the emulsion phase is consideredto be completely mixed in each section and at incipient

fluidization conditions.

The bubble phase gas is assumed to be in plug flow (i.e., largenumber of CSTRs), where the bubble size and the bubble rise

velocity change for each section.

The heterogeneous reactions (methane steam reformingand water gas shift reactions) take place only in the emul-

sion phase, assuming that the bubble phase is free of cata-

lyst particles.

Gas removed from the fluidized bed via membranes isassumed to be extracted from both the emulsion phase andi1

f00membranei;mol Mw;iAmembrane1 eb;no 0 61 Note that

Transfer term

Q use;n1ATre;n1 use;nATre;n Xnci1

f00membranei;mol Amembrane1 eb;n

Xnci1

Kbe;i;nVb;nrb;nwb;i;n we;i;n

9where

use;nAT ue;nAT1 eb;nusb;0AT utotATeb;0use;0AT utotAT1 eb;0

Energy balance (in case of energy supply inside the reactor)

Xnci1

HTfeedi

usb;n0ATrb;i;n0 use;n0ATre;i;n0

Xnci1

HTouti

usb;nNATrb;i;nN use;nNATre;i;nN

(Xnc

i1HTouti

f00membranei;mol Mw;iATeb;n

f00membranei;mol Mw;iAT1 eb;n)

E 0 10

where E depends on the kind of energy supply used (see e.g.

Gallucci et al. [15]). All the parameters used are described

elsewhere [15].

3.2. Packed bed membrane reactor 1D and 2D models

The axial temperature and concentration profiles in both

reaction and permeation compartments were modeled with

a 1D reactor model. The mass and energy conservation

equations read:

Mass conservation equations:

vrsgu

sg

vz

CMSD 2pris2cell pr20

J (11)ub;n1ATrb;n1 ub;nATrb;n i1

Kbe;i;nVb;nrb;n wb;i;n we;i;n

Xnci1

f00membranei;mol Mw;iAmembraneeb;n

we;i;nSFQ wb;i;nSFQ 0 7Emulsion phase component mass balances1

use;n1ATre;n1 use;nATre;n Xnci1

Kbe;i;nVb;nrb;nwb;i;n we;i;n

Xnci1

f}membranei;mol Mw;iAmembrane1 eb;n

0@Xnrxn

j1nj;irj

1AVe;nrp;n1 ee we;i;nSFQ wb;i;nSFQ 0SFx x if x > 00 if x 0 :

0 lz

i n t e r n a t i o n a l j o u rn a l o f h y d r o g e n en e r g y 3 5 ( 2 0 1 0 ) 7 1 4 2 7 1 5 0 7145vrtgu

tg

vz

CMSD2riJ (12)

rsgusg

vwsj;gvz

vvz

rsgD

sz

vwsj;gvz

rsj;g (13)

rtgutg

vwtj;gvz

vvz

rtgD

tz

vwtj;gvz

!61 3tg

dtp

jtj

1wsj;g

CMsD

2riJ jtj rtj;s 0

(14)

Energy conservation equations:

3sgr

sgC

sp;g rsbulkCsp;g

vTsvt

rsgusgCsp;gvTs

vz vvz

lseff

vTs

vz

Xj

rsj;gHsj;g

2pr0s2cell pr20

astwTtw Ts (15)

3tgr

tgC

tp;g rtbulkCtp;g

vTtvt

rtgutgCtp;gvTt

vz vvz

lteff

vTt

vz

Xj

rtj;gHtj;g

2riattw

Ttw Tt (16)

rtwg Ctwp;g

vTtw

vt vvz

ltweff

vTtw

vz

2r0r20 r2i

astwTs Ttw

2rir20 r2i

attwTt Ttw (17)

The remaining constitutive correlations needed to close the

model are summarised elsewhere (see Smit et al. [23])

The 2D model consists of a pseudo-homogeneous, 2D

reactor model consisting of the total gas phase continuity and

NavierStokes equations augmented with gas phase compo-

nent mass balances and the overall energy balance (see e.g.

Tiemersma et al. [12]). The model is based on a standard

dispersion model which describes the gas phase mass and

energy transport as convective flowwith superimposed radial

and axial dispersion.

The following assumptions have been made in this model:

The gas bulk can be described as an ideal Newtonian fluid. The porosity profiles have been accounted in the model.

The model equations in 2D axisymmetrical cylindrical co-

ordinates are:

Continuity equation

v3rg

vt V$

3rgu

0 (18)

Total momentum balance equation

v

vt

3rgu

V$

3rguu

3Vp b3rgu V$

3sg 3rgg (19)

Friction coefficient

b 1501 32

33

mg

rgd2p 1:751 3

333jujdp

(20)

wherejuj u2r u2z

q(21)where source terms equals:

Sr;i 1 3rSXnrj1

rjDHj for j 1;2;.;nr (28)

4. Results and discussion

All the simulations for both the FBMR and PBMR have been

performed without sweep gas and considering vacuum in the

permeation side. A comparison between the two reactor

concepts is carried out based on the membrane area required

for a target conversion, because it is anticipated that

membrane costs is the most important parameter in deter-

mining the reactor investment costs. A first comparison has

been made between the fluidized bed membrane reactor

model and the 1D packed bed reactor model at ideal condi-

tions: isothermal conditions and absence of mass and heat

transfer limitations, i.e., the number of grid cells of the 1D

model is set equal to number of CSTRs in theMAFBmodel. The

results show that as expected in these conditions the two

reactors give identical performance in terms of membrane

area required for a given conversion. In this way it has been

verified that the two models are working properly. The

following simulations have been performed with a heat flux

through the reactor walls. The main difference between the

fluidized bed and the packed bed membrane reactors is

related to the heat management. In fact, for the fluidized bed

membrane reactor it is well known that a virtually isothermal

condition can be achieved while for the packed bed

membrane reactor a temperature drop in the first part of therg Mgp

RTgideal gas (22)

Newtonian fluid

sg lg 23mg

V$uI mg

hVu VuT

i(23)

Porosity profile (Hunt and Tien [24])

3r 30 1 30exp 6r0 r

dp

(24)

Component mass balance

v

vt

3rgwi

V$

3rguwi

V$

rgDi$Vwi

Sr;i

with Di Dr;i 00 Dz;i

(25)

where source terms equals:

Sr;i 1 3rSMiXnrj1

ni;jrj for i 1;2;.;nc (26)

Energy balance

3rgCp;g 1 3rSCp;S

vTvt

Cp;gV$3rguT

V$li$VT Sh with li

lr 0

(27)reactor is always observed irrespective of the profile of

temperature at the reactor wall.

z, m

0. .2 0. .6 0. .0 1. .4 1.0 0 4 0 8 1 2 1 6 1.8 2.0

Tem

peratu

re, K

860

880

900

920

940

960

980

dp = 0.0005 m

dp = 0.0015 m

dp = 0.0025 m

Fig. 3 Axial temperature profile in a packed bed

membrane reactor.

z, m

0. .2 0. .6 0. .0 1. .4 1.0 0 4 0 8 1 2 1 6 1.8 2.0

Tem

peratu

re, K

860

880

900

920

940

960

980

Wall temperatureReaction temperature

Pre-reforming zone

Fig. 5 Axial temperature profile in a packed bed

membrane reactor with pre-reforming zone.

i n t e r n a t i on a l j o u r n a l o f h y d r o g e n en e r g y 3 5 ( 2 0 1 0 ) 7 1 4 2 7 1 5 07146A typical result for the axial temperature profile in the

packed bed membrane reactor is reported in Fig. 3, indeed

showing a temperature drop of 80100 K in the first part of the

reactor which can give stability and sealing problems for the

membrane. In fact, the membrane material should withstand

a large axial temperature gradient, which might cause the

detachment of the Pd-based layer from the support with

consequent loss in perm-selectivity. Moreover, the first part of

the membrane is not effectively used since it is working at

a relatively low temperature which, following Richardsons

equation, results in a lower hydrogen permeation flux. The

decrease in temperature at the beginning of the reactor also

gives a decrease in the reaction rate. The result is an increase

of the membrane area required for a specified conversion

compared to the case of constant temperature. In particular,

the membrane area required increases by about 21% ifcompared with an isothermal operation (which is only

possible in a fluidized bed membrane reactor). The effect of

the particle size on the temperature profile is negligible as also

z, m

0. .2 0. .6 0. .0 1. .4 1.0 0 4 0 8 1 2 1 6 1.8 2.0

CH

4 co

nversio

n, -

0.0

0.2

0.4

0.6

0.8

1.0

dp = 0.0005 m

dp = 0.0015 m

dp = 0.0025 m

Fig. 4 Methane conversion for different particle size in

a packed bed membrane reactor.indicated in the same figure. By changing the particle size the

combination of the temperature change and the change in the

effectiveness factor is counterbalanced so that the final

conversion is practically the same for different particle

diameters (see Fig. 4).

A way to overcome the problem of the temperature drop is

the use of a pre-reforming zone (in our case about 2025 cm)

where no membrane is applied. In this case (pre-reforming

section 25 cm) the membrane is used at an almost constant

temperature (maximum temperature difference 28 K) so that

the stability problems are prevented and the membrane is

effectively used resulting in a lower membrane area needed

for a given conversion (i.e., slightly longer packed bed, but

smaller membrane area). The results are shown in Fig. 5. In

these conditions the increase of membrane area with respect

to an ideal fluidized bed membrane reactor is 13%.

Another difference that can occur between a packed bed

and a fluidized bed is mass transfer limitations between thebed and the membrane wall which are present in the packed

bed but not in the fluidized bed. To investigate the extent and

r/R, -

0. .2 0. .6 0.0 0 4 0 8 1.0

H2 w

eig

ht fra

ctio

n, -

0.01

0.02

0.03

0.04

0.05

0.06

z/L = 0.1

z/L = 0.5

z/L = 0.8

Fig. 6 Radial profile of the H2 weight fraction for the

isothermal operation mode.

Fig. 8 reports the relative H2 weight fraction (defined as the

actual H2 weight fraction divided by the H2 weight fraction at

the catalyst center) as a function of the radial position. The

results reported in the Fig. 8 suggest that at higher membrane

permeabilities mass transport limitations to the membrane

Fig. 9 Schematic representation of the membrane reactor

concept with bubble increasing in size.

r/R, -

0. .2 0. .6 0.0 0 4 0 8 1.0

H2 w

eig

ht fractio

n, -

0.025

0.030

0.035

0.040

0.045

0.050

0.055

0.060

Original H2permeation rate

2 * Original H2

permeation rate

5 * Original H2

permeation rate

z/L = 0.1

Fig. 7 Radial profile of the H2 weight fraction for the

isothermal operation mode at different hydrogen

i n t e r n a t i o n a l j o u rn a l o f h y d r o g e n en e r g y 3 5 ( 2 0 1 0 ) 7 1 4 2 7 1 5 0 7147the influence of the concentration polarization a 2D model

was used.

First the radial H2 concentration profiles have been calcu-

lated at different axial positions at isothermal conditions. As

can be seen in Fig. 6, radial concentration profiles are present

but not really pronounced. It can be concluded that for the

present membranes and for small membrane diameters (1 cm

in the simulation shown in the figure), the bed-to-wall mass

transfer limitations have a negligible influence on the

membrane area.

In view of further developments and optimization of Pd-

based membranes, higher membrane fluxes will become

possible in the near future. Whether concentration polariza-

tion will occur with increased permeability was investigated

numerically. Simulation results where the membrane

permeability was increased with a factor of 2 and 5. As shown

permeabilities.in Figs. 7 and 8, an increase of hydrogen permeability causes

a significant increase of the concentration polarization, even

in membrane tubes with a diameter as small as 1 cm.

r/R, -

0. .2 0. .6 0.0 0 4 0 8 1.0

Re

la

tiv

e H

2 w

eig

ht fractio

n, -

0.0

0.2

0.4

0.6

0.8

1.0

Original H2

permeation rate

2 * Original H2

permeation rate

5 * Original H2

permeation rate

z/L = 0.1

Fig. 8 Relative H2 weight fraction for the isothermal

operation mode at different hydrogen permeabilities.wall will negatively affect the reactor performance resulting in

an increased H2 slip through the reactor exhaust. The 2D

model can be further applied to quantify the effects of mass

Factor multiplying the mass transfer coefficient, -

1e+0 1e+1 1e+2 1e+3 1e+4 1e+5

CH

4 co

nversio

n, -

0.945

0.950

0.955

0.960

0.965

0.970

0.975

0.980

Mass transfer limitations calculated

as Fluidized Bed without internals

No mass transfer limitations

Fig. 10 Effects of bubble-to-emulsion phase mass transfer

limitations on the conversion (FBMR).

itself increases by increasing the reactor length as schemati-

cally indicated in Fig. 9.

As a result of this bubble increase, themethane conversion

decreases as indicated in Fig. 10. The figure shows that the

methane conversion decreases by increasing the mass

transfer limitations. In case of mass transfer limitations

calculated as a fluidized bed reactor without internals (worst

case) the methane conversion decreases tremendously as

compared with the case without mass transfer limitations

(previously indicated as ideal condition for fluidized bed

Membrane area, m2

3 4 5 6 7 8

CH

4 co

nversio

n, -

0.950

0.955

0.960

0.965

0.970

0.975

0.980

Fig. 11 Membrane area needed for a given conversion in

Table 1 Comparison between staged fluidized bed and1D packed bed.

T, C P, bar Fluidized bed (5 stages) Packed bed (1D)

700 20 3.24 3.94

i n t e r n a t i on a l j o u r n a l o f h y d r o g e n en e r g y 3 5 ( 2 0 1 0 ) 7 1 4 2 7 1 5 07148transfer limitations at different membrane diameters, and

different operating conditions.

Concerning the fluidized bed membrane reactor, an

important transfer limitation affecting its performance is the

mass transfer limitation between the bubble phase and the

emulsion phase. In fact, the gas transported inside the bubbles

should be exchanged with the emulsion phase to react. A high

mass transfer limitation (low mass transfer coefficient)

between the bubble phase and the emulsion phase results in

a larger gas slip via the bubble phase and a lower conversion

degree. In our fluidized bed membrane reactor model the

bubble-to-emulsion phases mass transfer coefficient is

calculated with the equations derived for a fluidized bed

without internals. Although the internals (solid membranes)

should enhance the mass transfer characteristics of the bed,

at the moment, reliable equations for bubble-to-emulsion

phases mass transfer coefficient for fluidized bed with

inserts are not available.

For a fair comparison with the packed bed membrane

reactor, the fluidized bed has been simulated with the same

case of mass transfer limitations (FBMR).membrane area but also with the same bed length. As

a matter of fact, the bubble-to-emulsion phase mass transfer

limitation increases with increasing bubble diameter, which

Number of stages

2 4 6 8 10 12

CH

4 co

nversio

n, -

0.945

0.950

0.955

0.960

0.965

0.970

0.975

0.980

Fig. 12 Conversion reached for a given area in case of

mass transfer limitations for different stages (FBMR).reactor). The figure also shows that by improving the mass

transfer by a factor of 10 results in a conversion close to the

ideal case without mass transfer limitations.

In order to achieve the same conversion degree of a fluid-

ized bed membrane reactor without mass transfer limitations

the membrane area installed in the reactor needs to be

increased as indicated in the following Fig. 11.

The membrane area required in case of mass transfer

limitations increases 2.4 times with respect to the case

without limitations as reported in the figure.

However, it has to be pointed out that the use ofmembranes

inside the bed leads to a decrease of the bubble size (both

because of gas extraction through themembranes and because

of break-up of bubbles by the solid membrane tubes) and

a consequent decrease of the mass transfer limitations. Fig. 10

shows that a decrease of 10 times in the mass transfer limita-

tions is enough to reach the limit conversion required. Thus,

a more detailed experimental work on the determination of

the bubble-to-emulsion phase mass transfer coefficient in

a fluidized bed with internals should be carried out.

On the other hand, even considering the worst case

(bubble-to-emulsion phase mass transfer coefficient equal to

a fluidized bedwithout internals) themass transfer problem in

the fluidized bed can be easily circumvented. In fact, themass

transfer resistance is higher when the bubble diameter

H4 C

on

ve

rs

io

n, -

0.7

0.8

0.9

1.0

Staged Fluidized bed

2D isothermal modelz/L, -

0.05 0.10 0.15 0.20 0.25

C

0.5

0.6

Fig. 13 Comparison between a staged fluidized bed and

a packed bed with 2D model. 5 bar reaction pressure.

i n t e r n a t i o n a l j o u rn a l o f h y d r o g e n en e r g y 3 5 ( 2 0 1 0 ) 7 1 4 2 7 1 5 0 7149becomes larger, and the bubble diameter increases with

increasing of the bed height, so that we can reduce the bubble

diameter by inserting stagers such as meshing wires at

different reactor heights (i.e., staging the fluidized bed reactor).

InFig. 12 the conversion inafluidizedbedwithmass transfer

limitations is shown for different numbers of stages. For these

simulations the axial position of the stagers has not been opti-

mized. This means that the distance between two stages is

constant for one simulation and it is given by dividing the total

lengthof thereactorby thenumberofstages. Ina futureworkan

optimization of the axial position for stagers will be carried out.

The area used in this simulation was kept the same as was

needed in the case of no mass transfer limitations. From the

figure it can be seen that the conversion required can be

achieved already with 34 stages. Thus, dividing the reactor in

different stages completely circumvents the problems ofmass

transfer limitation for the fluidized bed membrane reactor.

A direct comparison (in terms of membrane area required)

between theperformancesof a stagedfluidizedbedmembrane

reactor and a packed bed membrane reactor simulated in

1D (no effects of bed-to-wall mass transfer limitations) is

reported in Table 1 (for a 50 Nm3/h hydrogen production).

The packed bed membrane reactor needs around 22%

larger membrane area if compared with a staged fluidized bed

membrane reactor. Moreover, the membrane in the packed

bed is exposed to an axial temperature variation of about 100 K

in the first 25 cm of reactor length, with possible stability and

sealing.

The fluidized bed becomes even more advantageous

compared topackedbedmembranereactorwhen theeffects of

concentration polarization on the performances of packed bed

are considered. A comparison in terms ofmethane conversion

between the staged fluidized bed membrane reactor and the

packed bed membrane reactor simulated with the 2D

isothermal model (effect of bed-to-wall mass transfer) is

reported in Fig. 13. The effect of bed-to-wall mass transfer

limitations in the packed bed reactor results in a decrease of

methane conversion compared to the fluidized bedmembrane

reactor. An overview of themembrane reactor increase due to

the effects of bed-to-wall mass transfer limitations is reported

inTable 2. In theworst case (complete conversion requiredand

1 bar reaction pressure) the packed bed membrane reactor

Table 2 Increase of membrane area with respect to theFBMR due to bed-to-membrane mass transfer limitation.

Conversiondegree

Preaction 1bar

Preaction 5bar

Preaction 10bar

0.975 85.4% 64.5% 60.0%

1 98.7% 81.3% 79.5%requires almost double themembrane areawith respect to the

staged fluidized bed membrane reactor.

Finally, we can state that the more evident advantages of

a fluidized bed reactor with respect to a packed bed membrane

reactor are: constant temperature along the reactor and better

heat integration (see Gallucci et al. [14,15]), no mass transfer

limitationbetweenthefluidizedbedand themembranesurface.

Some disadvantages such as the erosion problems and

horizontal membrane sealing should be further studied

experimentally.The authors are grateful to the Dutch Ministry of Economic

affairs for financial support of this work in the EOS program

(Project EOSLT05010).

Table of symbols

AT Area of bed cross section, m2

Amembrane,n Membrane surface area per cell, n, m2

CSTR Continuously stirred tank reactor,

dp Particle diameter, m

Cp Heat capacity, J/(kg K)

D Dispersion coefficient, m2/s

Dg Gas diffusivity, m2/s

eb Bubble phase fraction,

ee Emulsion phase fraction,

Ea Activation energy for hydrogen permeation, J/mol

g Gravitational acceleration (9.81), m/s2Hj Enthalpy of specie j, J/mol

HTi;x Enthalpy of component i at temperature T at position

x, J/mol

J Permeation flux through membrane, mol/(m2 s)

jj Mass flux component j, mol/(m2 s)

ki Reaction rate constant for ith reaction

Kbe,i,n Bubble-to-emulsion phase mass transfer coefficient

for component I in cell, n, s1

Keq,i Equilibrium constant for jth reaction [depending on

the reaction]5. Conclusions

In this work, two different membrane reactor concepts for

the H2 production via methane steam reforming have been

compared via detailed models. It has been shown that both

concepts may suffer from mass transfer limitations. For the

fluidized bed membrane reactor the mass transfer limitations

occur between the bubble phase and the emulsion phase. The

effect of these mass transfer limitations on the membrane

area required is quite significant. However, these mass trans-

fer limitations can be easily circumvented by staging the

fluidized bed with consequent break-up of bubbles and

decrease of mass transfer limitations. For the packed bed

membrane reactor, the mass transfer limitations occur

between the catalytic bed and the membrane area (concen-

tration polarization). These mass transfer limitations cannot

be easily avoided (not even with membrane tube diameters as

small as 1 cm), and the packed bedmembrane reactor requires

in some cases double the membrane area with respect to the

staged fluidized bed operated at the same conditions. More-

over, a 2025%moremembrane area is required by the packed

bed (with respect to the fluidized bed) because of the temper-

ature profile prevailing in the packed bed.With the advance of

the development of H2 perm-selective membranes with ever-

increasing permeabilities, the advantages of fluidized bed

membrane reactor become more and more pronounced.

AcknowledgmentMw[i] Molar mass for component i, kg/mol

CMD Average molar mass, kg/mol

Greek2

i n t e r n a t i on a l j o u r n a l o f h y d r o g e n en e r g y 3 5 ( 2 0 1 0 ) 7 1 4 2 7 1 5 07150a heat transfer coefficient, J/(m K s)

b friction factor,

r Density, kg/m3

3 Porosity,

3e Emulsion phase porosity,

l Thermal conductivity, J/(m K s)

leff Effective thermal conductivity, J/(m K s)

mg Viscosity of gas, Pa s

sg Stress tensor, kg/(m s)

f00membranei;mol Molar flux component i through themembrane percell, mol/(m2 s)

Subscripts

0 Reactor inlet,

b Bubble phase,

e Emulsion phase,

g gas phase,

i Component i,

j Number of reaction,

n Number of CSTRs for emulsion or bubble phase,

r radial co-ordinate,

s solid phase,

z axial co-ordinate,

r e f e r e n c e s

[1] Rostrup-Nielsen JR. Syngas in perspective. Catal Today 2002;71(34):2437.

[2] Adris AM, Elnashaie SSEH, Hughes R. Fluidized bedmembrane reactor for the steam reforming of methane. CanJ Chem Eng 1991;69(5):106170.

[3] Kikuchi E. Palladium/ceramic membranes for selectivehydrogen permeation and their application to membranereactor. Catal Today 1995;25(34):3337.

[4] Gallucci F, Paturzo L, Fama` A, Basile A. Experimental study ofthe methane steam reforming reaction in a dense Pd/Agmembrane reactor. Ind Eng Chem Res 2004;43:92833.

[5] Gallucci F, Paturzo L, Basile A. A simulation study of thesteam reforming of methane in a dense tubular membranereactor. Int J Hydrogen Energy 2004;29:6117.

[6] Oklany JS, Hou K, Hughes R. A simulative comparison ofdense and microporous membrane reactors for the steampi Partial pressure for component i, atm

P0e Pre-exponential factor for permeation of Pd

membrane, mol/(s m2 Pa0.5)

R Gas constant (8.3145), J/(mol K)rj Reaction rate for jth reaction, mol/(kgcat s)

ri Inner tube radius, m

r0 Outer tube radius, m

Sh source/sink term for heat balance, J/(m3 s)

Sr,i source/sink term for mass balance, kg/(m3 s)

T Temperature, K

u Mixture velocity, m/s

us Superficial gas velocity, m/s

V Volume, m3

vj,i Stoichiometric coefficient for jth reaction and ith

component,

wz weight fraction, reforming of methane. Appl Catal A Gen 1998;170:1322.[7] Patil CS, van Sint Annaland M, Kuipers JAM. Fluidised bedmembrane reactor for ultrapure hydrogen production viamethane steam reforming: experimental demonstration andmodel validation. Chem Eng Sci 2007;62(11):29893007.

[8] Fu C-H, Wu JCS. Mathematical simulation of hydrogenproduction via methanol steam reforming using double-jacketed membrane reactor. Int J Hydrogen Energy 2007;32(18):48309.

[9] Gallucci F, Basile A. Pd-Ag membrane reactor for steamreforming reactions: a comparison between different fuels.Int J Hydrogen Energy 2008;33(6):167187.

[10] Gallucci F, Van Sint Annaland M, Kuipers JAM. Pure hydrogenproduction via autothermal reforming of ethanol ina fluidized bed membrane reactor: a simulation study. Int JHydrogen Energy 2010;35(4):165968.

[11] Simakov David SA, Sheintuch Moshe. Demonstration ofa scaled-down autothermal membranemethane reformer forhydrogengeneration. Int JHydrogenEnergy2009;34(21):886676.

[12] Tiemersma TP, Patil CS, van Sint Annaland M, Kuipers JAM.Modelling of packed bed membrane reactors for autothermalproduction of ultrapure hydrogen. Chem Eng Sci 2006;61(5):160216.

[13] Gallucci F, Comite A, Capannelli G, Basile A. Steam reformingof methane in a membrane reactor: an industrial case study.Ind Eng Chem Res 2006;45(9):29943000.

[14] Gallucci F, Van Sint Annaland M, Kuipers JAM. Autothermalreforming of methane with integrated CO2 capture in a novelfluidized bed membrane reactor. Part 1: experimentaldemonstration. Top Catal 2008;51(14):13345.

[15] Gallucci F, Van Sint Annaland M, Kuipers JAM. Autothermalreforming of methane with integrated CO2 capture in a novelfluidized bed membrane reactor. Part 2: comparison ofreactor configurations. Top Catal 2008;51(14):14657.

[16] Mahecha-Botero A, Grace JR, Jim Lim C, Elnashaie SSEH,Boyd T, Gulamhusein A. Pure hydrogen generation ina fluidized bed membrane reactor: application of thegeneralized comprehensive reactor model. Chem Eng Sci2009;64(17):382646.

[17] Numaguchi T, Kikuchi K. Intrinsic kinetics and designsimulation in a complex reaction network: steam reforming.Chem Eng Sci 1988;43:2295301.

[18] Lin YM, Liu SL, Chuang CH, Chu YT. Effect of incipientremoval of hydrogen through palladium membrane on theconversion of methane steam reforming experimental andmodeling. Catal Today 2003;82:12739.

[19] Deshmukh SARK, Laverman JA, Cents AHG, Van SintAnnalandM,Kuipers JAM.Developmentofamembraneassistedfluidized bed reactor. 1. Gas phase back-mixing and bubble-to-emulsion phasemass transfer using tracer injection andultrasound experiments. Ind Eng Chem Res 2005;44:595565.

[20] Deshmukh SARK, Laverman JA, Van Sint Annaland M,Kuipers JAM. Development of a membrane assisted fluidizedbed reactor. 2. Experimental demonstration and modelingfor the partial oxidation of methanol. Ind Eng Chem Res2005;44:596676.

[21] Deshmukh SARK, Van Sint Annaland M, Kuipers JAM. Heattransfer in a membrane assisted fluidised bed withimmersed horizontal tubes. Int J Chem React Eng 2005;3:A1.

[22] Patil CS, Van Sint Annaland M, Kuipers JAM. Experimentalstudy of a membrane assisted fluidized bed reactor for H2production by steam reforming of CH4. Chem Eng Res Des2006;84:399404.

[23] Smit J, Van Sint Annaland M, Kuipers JAM. Feasibility studyof a reverse flow catalytic membrane reactor with porousmembranes for the production of syngas. Chem Eng Sci 2005;60(24):697182.

[24] Hunt ML, Tien CL. Non-darcian flow, heat and mass transfer

in catalytic packed-bed reactors. Chem Eng Sci 1990;45:5563.

Theoretical comparison of packed bed and fluidized bed membrane reactors for methane reformingIntroductionReactor configurationsFluidized bed membrane reactor conceptPacked bed membrane reactor concept

Reactor modelsFluidized bed membrane reactor modelPacked bed membrane reactor 1D and 2D models

Results and discussionConclusionsAcknowledgmentTable of symbolsReferences