Diesel soot combustion on potassium promoted

hydrotalcite-based mixed oxide catalysts

Zhaoliang Zhang

*

, Zonggang Mou, Pengfei Yu, Yexin Zhang, Xianzi Ni

College of Chemistry and Chemical Engineering, University of Jinan, Jiwei Road 106, Jinan 250022, PR China

Received 26 August 2006; received in revised form 11 January 2007; accepted 11 January 2007

Available online 16 January 2007

Abstract

KNO

3

or K

2

CO

3

supported Mg–Al hydrotalcite-based mixed oxide catalysts were investigated for the catalytic combustion of diesel

soot. The activity of the catalysts before and after reactions with soot was conducted under the flow of air and high purity He gas. It has
been found that K shows a great promotion of the catalytic activity, and deactivation was not detected after the reaction with soot in a
muffle furnace in a static air atmosphere. The active phases were examined by XRD. The high activity is found to be due to an interaction
between K and Mg(Al), which may weaken the Mg(Al)–O bonds, thus facilitating the mobility of the O species.
Ó 2007 Elsevier B.V. All rights reserved.

Keywords: Soot combustion; Hydrotalcite; Potassium; NO

x

storage

1. Introduction

Diesel engine cars have been generally accepted in the

world, especially in Europe owing to their low fuel con-
sumption, which results in the decrease of CO

2

emission

to the atmosphere. Although modern diesel engines have
both CO and HC outlet concentrations much lower than
those of spark-ignition engines, soot and NO

x

are predom-

inant pollutants that must be removed using different tech-
nologies

from

traditional

three-way

catalysts

[1]

.

Improvements of the technological level of the engines
alone will not meet the more and more stringent legisla-
tions. The suitable after-treatment technologies were neces-
sary. With respect to the reduction of soot emissions,
catalytic oxidation on diesel particulate filters seems to be
most promising although the catalyst is not resolved

[2]

.

In the respect of NO

x

, NO

x

storage and reduction (NSR)

technology has been commercially viable

[3]

. Compara-

tively, simultaneous removal of soot and NO

x

is still a

great challenge

[4]

. A bifunctional concept of combining

the NO

x

catalytic trap with the soot combustion for filter

regeneration was projected

[5]

. Recently, Toyota Corpora-

tion has developed Diesel Particulate-NO

x

Reduction

(DPNR) system, based on NSR catalysts, which can reduce
NO

x

to N

2

and simultaneously oxidize soot to CO

2

[6]

.

Lietti et al.

[7]

reported the potentiality of a typical Pt–

Ba/Al

2

O

3

NSR catalyst in the simultaneous removal of

soot and NO

x

. Makkee et al.

[8]

reported soot oxidation

activity and deactivation of NSR catalysts containing Pt,
K, and Ba supported on Al

2

O

3

. Just before, hydrotalcite-

based NSR catalysts were reported to show better perfor-
mances in the resistance to deactivation by SO

2

and NO

x

storage than Pt–Ba/Al

2

O

3

Toyota-type NSR catalysts at

reaction temperatures lower than 250

°C

[9]

. Thus, MgAlO

mixed oxides derived from hydrotalcite precursors are con-
sidered to be promising materials for a new generation of
NSR catalysts

[10]

. Normally, potassium is another NO

x

storage component, especially in conjunction with Ba

[11]

. Furthermore, potassium is well known to be the pro-

moter for soot combustion catalysts

[12]

. Therefore, in this

communication, potassium promoted Mg–Al hydrotalcite-
based mixed oxide catalysts for soot combustion were pre-
sented

[13]

.

1566-7367/$ - see front matter

Ó 2007 Elsevier B.V. All rights reserved.

doi:10.1016/j.catcom.2007.01.010

*

Corresponding author.
E-mail address:

[e-mail]

(Z. Zhang).

www.elsevier.com/locate/catcom

Catalysis Communications 8 (2007) 1621–1624

2. Experimental

Mg and Al hydrotalcite with molar ratio of 79:21 was

prepared by co-precipitation from aqueous solutions of
metal nitrates of Mg(NO

3

)

2

Æ

6H

2

O and Al(NO

3

)

3

Æ

9H

2

O.

The mixture of metal nitrates was added to a vigorously
stirred solution containing a slight excess of Na

2

CO

3

.

The pH was maintained constant (10.0 ± 1) by dropwise
NaOH addition. Precipitates were kept in suspension at
70

°C for 30 min under stirring, then were filtered, thor-

oughly washed with distilled water and dried overnight at
120

°C. Finally, the prepared hydrotalcite was calcined at

600

°C for 4 h to get MgAlO mixed oxide.

The potassium-promoted samples were prepared by

impregnation method using KNO

3

or K

2

CO

3

as precur-

sors. The slurries of MgAlO and the aqueous solution of
K salts were stirred at 70

°C until drying. The powder

was then dried at 120

°C overnight and calcined at

600

°C for 4 h. These potassium promoted hydrotalcite-

based mixed oxide catalysts were denoted as nS/MgAlOa,
in which S represents KNO

3

or K

2

CO

3

and n indicates

the weight loading (wt.%), whereas a means after the reac-
tion, see

Table 1

.

X-ray powder diffraction (XRD) patterns were recorded

on a Rigaku D/max-rc diffractmeter employing Cu K

a

radiation.

The catalytic tests towards the combustion of soot are

studied by TGA technique with a Perkin Elmer Diamond
apparatus. Printex-U from Degussa is used as model soot.
The soot was mixed with the catalyst in a weight ratio of
1:9 in an agate mortar for 30 min, which results in a tight
contact. The mixture was heated from room temperature
to 750

°C at a heating rate of 10 °C/min in the flow of air

or high purity He at 100 ml/min. The performance of the
catalysts was evaluated by comparing characteristic tem-
peratures of TG-DTA curves of soot combustion with or
without catalysts

[14]

. As showed in

Fig. 1

, T

i

, defined as

soot ignition temperature, was determined as the extrapo-
lated onset of the thermoxidation process. The temperature
of the maximum of the DTA curve corresponding to the
highest combustion velocity is named T

m

. The complete

conversion temperature of soot, T

f

, is deduced at the end

of weight loss. While the difference between T

f

and T

i

,

DT

(=T

f

–T

i

), represents combustion reaction velocity.

3. Results and discussion

Fig. 1

shows the representative TG-DTA-DTG curves

of soot combustion. The characteristic temperatures are
summarized in

Table 1

. Compared to un-catalyzed soot,

MgAlO decreases T

m

from 619 to 572

°C. A little further

decrease is achieved after calcination at 900

°C for 4 h

(MgAlO900). On the other hand, loading potassium
resulted in marked decreases in all characteristic tempera-
tures, suggesting the activity of MgAlO is highly promoted
by potassium. It is important to note that there is only one
weight loss stage and corresponding to an exothermic peak
in DTA curve, characteristic of un-catalyzed soot combus-
tion. Contrarily, all hydrotalcite-based samples show three
stages in the TG-DTG curves

[15]

.

Data in parentheses in

Table 1

gives the activity of the

potassium promoted samples after reactions with soot in
a muffle furnace in a static air atmosphere at the tempera-
tures a little higher than T

f

, respectively. The activity of

5KNO

3

/MgAlO and 6.83K

2

CO

3

/MgAlO were nearly

retained after reactions. However, 10KNO

3

/MgAlO shows

Table 1
The characteristic temperatures of soot combustion for K promoted MgAlO

Catalysts

T

i

(

°C)

T

m

(

°C)

T

f

(

°C)

D

T (T

f

-T

i

) (

°C)

Printex-U soot

564

619

661

97

MgAlO

539

572

613

74

MgAlO900

a

530

567

595

65

5KNO

3

/MgAlO

388 (389)

1

410 (424)

1

439 (441)

1

51 (52)

1

10KNO

3

/MgAlO

376 (418)

2

399 (447)

2

427 (469)

2

51 (51)

2

6.83K

2

CO

3

/MgAlO

b

395 (400)

3

422 (436)

3

446 (452)

3

51 (52)

3

1,3

After the reaction with soot (9:1, weight ratio) at 450

°C for 5 h in a muffle furnace in a static air atmosphere.

2

After the reaction with soot (4:1, weight

ratio) at 430

°C for 0.5 h in a muffle furnace in a static air atmosphere.

a

Hydrotalcite calcination at 900

°C for 4 h.

b

6.83 wt.% K

2

CO

3

contains the same K content as 10KNO

3

/MgAlO.

100

200

300

400

500

80

85

90

95

100

T

f

TG

DTA (

μV)

TG (%)

Temperature (

o

C)

-15

0

15

30

DTA

Exo

T

m

T

i

DTG

Fig. 1. TG-DTA-DTG curves for soot combustion under the flow of air
for 5KNO

3

/MgAlO.

1622

Z. Zhang et al. / Catalysis Communications 8 (2007) 1621–1624

a large decrease of activity. It is also seen in

Table 1

that

the values of DT decrease in the presence of catalysts. In
fact, DT is about 97

°C for un-catalyzed soot, whereas, it

is about 51

°C for potassium promoted samples. Nearly

the same values of DT were observed after reactions.

Fig. 2

shows the XRD patterns of the hydrotalcite-based

samples after calcinations. For comparison, the pattern of
10KNO

3

/MgAlOa is also present (

Fig. 2

d). An ill-crystal-

line MgO-type MgAlO (JCPDS 74–1225) phase with peaks
at 36.86, 42.82, and 62.17

° is seen in all the patterns. After

the Mg-Al hydrotalcite calcination at 900

°C for 4 h, the

spinel

MgAl

2

O

4

(JCPDS

21-1152)

phase

appeared

(

Fig. 2

f). In addition to the MgAlO phases, 5KNO

3

/

MgAlO and 10KNO

3

/MgAlOa samples show peaks of

hydrotalcite structure (JCPDS 22-0700), indicating that
these samples are partially regenerated through the ‘‘mem-
ory effect’’

[16]

. It is also found that the crystalline KNO

3

(JCPDS 71-1558) remains in the 10KNO

3

/MgAlO sample

after calcination at 600

°C for 4 h. In contrast, there are

no characteristic XRD peaks associated with potassium
on 5KNO

3

/MgAlO and 6.83K

2

CO

3

/MgAlO samples after

calcination at 600

°C for 4 h, suggesting a high dispersion

of potassium species. Worth notable is the absence of the
KNO

3

phase in the XRD pattern of 10KNO

3

/MgAlOa

(

Fig. 2

d). Similar to 6.83K

2

CO

3

/MgAlO, 10KNO

3

/MgA-

lOa shows weak peaks of the spinel MgAl

2

O

4

phase.

It is long known that Al

2

O

3

is a non-active sample

[17]

and MgO has a low activity

[12]

for soot combustion. Their

mixed oxide MgAlO derived from hydrotalcite was found
to inherit the performance of MgO, no effect from Al.
Potassium introduction as KNO

3

or K

2

CO

3

increases sig-

nificantly the catalytic activity in soot combustion, not only
for the fresh catalysts but also for the catalysts after reac-
tions, as shown in

Table 1

. In order to elucidate the mech-

anism, the samples for soot combustion were conducted
under the flow of high purity helium (O

2

< 1 ppm), the rep-

resentative is shown in

Fig. 3

. While no exothermic peak

was observed for MgAlO, the potassium promoted samples
gave an exothermic peak even in the absence of gaseous O

2

.

This evidence indicates unambiguously that bulk O

2

is

incorporated in soot combustion in potassium promoted
samples. T

i

increase and the broad range of the exothermic

peaks of soot combustion in the absence of gaseous O

2

than in it presence may be due to low O

2

concentrations

and diffusion rates

[18]

. MgAlO did not show any catalytic

activity in the absence of gaseous oxygen, while in the pres-
ence of air it decreased soot combustion temperature
slightly. This suggests that, in the presence of MgAlO,
the reaction occurs through the adsorption of gaseous O

2

on the catalyst surface, generating an active species that
is capable of oxidizing soot

[12]

.

From XRD results, it is known that, in

Fig. 1

, the first

two stages corresponding to endothermic peaks are associ-
ated with the thermal decomposition of regenerated hydro-
talcites. Only the third stage corresponding to exothermic
peak in DTA curves is characteristic of soot combustion.
It is deduced from literatures

[19,20]

that 5KNO

3

/MgAlO,

10KNO

3

/MgAlO and 6.83K

2

CO

3

/MgAlO showing one

soot combustion peak, were due to a single active species.
Although 10KNO

3

/MgAlO consists of KNO

3

and MgAlO

phases, after the reaction with soot in air, KNO

3

phase

diminished in correspondence of the activity decrease.
The activity before and after reactions is in reasonable
range for 5KNO

3

/MgAlO and 6.83K

2

CO

3

/MgAlO, imply-

ing that the reaction mechanism might be unchanged. The
presence of MgAl

2

O

4

in MgAlO phase, as for MgAlO900,

results in a slight increase in activity. However, one soot
combustion peak was identified in TG-DTA-DTG curves
of MgAlO900, meaning MgAlO and MgAl

2

O

4

function

through a common active species, for example, Mg(Al)–
O, as proposed above for MgAlO.

There are two possibilities for the absence of potassium-

containing phase in XRD patterns of 5KNO

3

/MgAlO,

10KNO

3

/MgAlOa and 6.83K

2

CO

3

/MgAlO: (1) KNO

3

or

K

2

CO

3

is highly dispersed; (2) KNO

3

or K

2

CO

3

decom-

posed and potassium incorporates into the lattice of

10

20

30

40

50

60

70

80

Intensity

2

θ (degree)

#

#

#

f

e

d

c

b

a

*

*

*

*

*

*

o

o

o

+

+

Fig. 2. XRD patterns of MgAlO (a), 5KNO

3

/MgAlO (b), 10KNO

3

/

MgAlO (c), 10KNO

3

/MgAlOa (d), 6.83K

2

CO

3

/MgAlO (e), MgAlO900

(f). + : MgAlO; o: KNO

3

; #: hydrotalcite;

*

: MgAl

2

O

4

.

0

100

200

300

400

500

600

700

40

50

60

70

80

90

100

DTA (

μV)

TG

TG (%)

Temperature (

o

C)

-40

-30

-20

-10

0

DTA
DTG

Fig. 3. TG-DTA-DTG curves for soot combustion under the flow of high
purity helium for 10KNO

3

/MgAlOa.

Z. Zhang et al. / Catalysis Communications 8 (2007) 1621–1624

1623

MgAlO. Provided the first condition existed, NO

3

and/or

CO

2
3

may have entered the interlayer space of the re-

formed hydrotalcites

[16]

, and then in the following reac-

tions with soot, the partially re-formed hydrotalcite would
decompose, as shown in

Fig. 1

. This, in turn, may give rise

to the formation of Mg(Al)–O–K species.

From the results presented here, the promotion effect

of potassium lies in the increase of mobility of bulk O.
In fact, the diffusion of oxygen is difficult at low temper-
atures and the oxidation reaction occurs slower than at
higher temperatures. The DT values for 5KNO

3

/MgAlO,

10KNO

3

/MgAlO

and

6.83K

2

CO

3

/MgAlO

decreased

greatly compared to that of MgAlO, which suggests that
the potassium loading is favorable for an easier diffusion
of oxygen at low temperatures (400–450

°C)

[21]

. As pro-

posed for K/Al

2

O

3

, Pt/K/Al

2

O

3

[8]

and KOH/MgO

[12]

,

a mechanism can thus be proposed

[22]

: soot oxidation

takes place through lattice and adsorbed oxygen transfer
to the soot surface, which can be replenished by the
gas-phase oxygen. K addition generates an interaction
between K and Mg (Al), which may weaken the
Mg(Al)–O bonds, thus facilitating the mobility of the O
species. Accordingly, the soot combustion activity for
5KNO

3

/MgAlO,

10KNO

3

/MgAlOa

and

6.83K

2

CO

3

/

MgAlO with different types and contents of K are very
close. The higher activity for 10KNO

3

/MgAlO than

10KNO

3

/MgAlOa comes from KNO

3

. As the melting

point of KNO

3

is 334

°C

[23]

, the residual KNO

3

on

10KNO

3

/MgAlO made a more tight contact between soot

and catalyst, which brought an additional activity.

Considering the development of a DPNR catalyst,

10KNO

3

/MgAlO is illustrative and meaningful. During

longer lean period, NO

x

is stored as KNO

3

or Mg(NO

3

)

2

,

as shown by 10KNO

3

/MgAlO, which has a high soot com-

bustion activity in the flow of air. Under a short rich per-
iod, KNO

3

or Mg(NO

3

)

2

decomposed, leaving K species

in MgAlO mixed oxide, as shown by 10KNO

3

/MgAlOa,

which also possesses activity in the flow of He. Therefore,
soot catalytic combustion proceeds not only on lean but
even on rich conditions. It is reported that in an atmo-
sphere containing CO

2

, potassium will convert to K

2

CO

3

[24]

. In this work, it is tested that a certain amounts of

K

2

CO

3

on MgAlO mixed oxide show likewise promotion

to soot combustion.

Acknowledgement

The authors thank the National Natural Science Foun-

dation of China (grants: 20577015) for the financial sup-
port of this work.

References

[1] A. Bueno-Lo´pez, K. Krishna, M. Makkee, J.A. Moulijn, J. Catal. 230

(2005) 237.

[2] S. Biamino, P. Fino, D. Fino, N. Russo, C. Badini, Appl. Catal. B 61

(2005) 297.

[3] N. Takahashi, H. Shinjoh, T. Iijima, T. Suzuki, K. Yamazaki, K.

Yokota, H. Suzuki, N. Miyoshi, S. Matsumoto, T. Tanizawa, T.
Tanaka, S. Tateishi, K. Kasahara, Catal. Today 27 (1996) 63.

[4] D. Fino, P. Fino, G. Saracco, V. Specchia, Appl. Catal. B 43 (2003)

243.

[5] V.G. Milt, C.A. Querini, E.E. Miro´, M.A. Ulla, J. Catal. 220 (2003)

424.

[6] J. Suzuki, S. Matsumoto, Top. Catal. 28 (2004) 171.
[7] L. Castoldi, R. Matarrese, L. Lietti, P. Forzatti, Appl. Catal. B 64

(2006) 25.

[8] K. Krishna, M. Makkee, Catal. Today 114 (2006) 48.
[9] G. Centi, G. Fornasari, C. Gobbi, M. Livi, F. Trifiro`, A. Vaccari,

Catal. Today 73 (2002) 287.

[10] G. Fornasari, R. Glfckler, M. Livi, A. Vaccari, Appl. Clay Sci. 29

(2005) 258.

[11] T.J. Toops, D.B. Smith, W.P. Partridge, Appl. Catal. B 58 (2005) 245.
[12] R. Jime´nez, X. Garcı´a, C. Cellier, P. Ruiz, A.L. Gordon, Appl. Catal.

A 297 (2006) 125.

[13] Z. Zhang, Z. Mou, P. Yu, Y. Zhang, X. Ni, M. Bao, Chinese Patent

Application, No. CN200510128436.X.

[14] M. Bokova, C. Decarne, E. Abi-Aad, A. Pryakhin, V. Lunin, A.

Aboukaı¨s, Appl. Catal. B 54 (2004) 9.

[15] L. Chmielarz, P. Kus´trowski, A. Rafalska-Łasocha, D. Majda, R.

Dziembaj, Appl. Catal. B 35 (2002) 195.

[16] A.E. Palomares, J.G. Prato, F. Rey, A. Corma, J. Catal. 221 (2004)

62.

[17] P.G. Harrison, I.K. Ball, W. Daniell, P. Lukinskas, M. Ce´spedes,

E.E. Miro´, M.A. Ulla, Chem. Eng. J. 95 (2003) 47.

[18] K. Sato, M. Yamaguchi, S. Fujita, K. Suzuki, T. Mori, Catal.

Commun. 7 (2006) 132.

[19] H. An, C. Kilroy, P.J. McGinn, Catal. Today 98 (2004) 423.
[20] H. An, P.J. McGinn, Appl. Catal. B 62 (2006) 46.
[21] R. Flouty, E. Abi-Aad, S. Siffert, A. Aboukaı¨s, Appl. Catal. B 46

(2003) 145.

[22] G. Mul, F. Kapteijn, J.A. Moulijn, Appl. Catal. B 12 (1997) 33.
[23] A.L. Carrascull, M.I. Ponzi, E.N. Ponzi, Ind. Eng. Chem. Res. 42

(2003) 692.

[24] N. Nejar, J.M. Garcı´a-Corte´s, C. Salinas-Martı´nez de Lecea, M.J.

Illa´n-Go´mez, Catal. Commun. 6 (2005) 263.

1624

Z. Zhang et al. / Catalysis Communications 8 (2007) 1621–1624