(560ff) Core-Shell Pt/Al2O3@Cu/ZSM-5 Catalyst for Ammonia Slip Catalyst in Diesel Aftertreatment

Core-Shell Pt/Al₂O₃@Cu/ZSM-5
Catalyst for Ammonia slip catalyst in Diesel Aftertreatment

Rajat
Subhra Ghosh*,
M. P. Harold*, Thuy Le*, J.D. Rimer*, and D. Wang**

*Dept. of Chemical and Biomolecular
Engineering, University of Houston, Houston, TX 77204-4004, USA

**Cummins Inc., 1900 McKinely
Av., MC50197, Columbus, IN 47201, USA

Introduction

The ammonia slip
catalyst (ASC) is an essential aftertreatment unit in modern diesel vehicles.
The purpose of ASC is to prevent the release of unreacted NH₃ leaving
a selective catalytic reduction (SCR) unit. The state-of-the-art ASC for
selective oxidation of NH₃ to N₂ has a dual-layer
architecture comprising a Pt/Al₂O₃ bottom layer (PGM) and
a Cu/Fe-exchanged zeolitic (M-Z) top layer [1]. The dual-layer structure enables
a high N₂ selectivity at high NH₃ conversion. It does so
by converting NO, one of the main oxidation products of the PGM, through
reaction with counter-diffusing NH₃. While recent studies
demonstrate the effectiveness of the dual layer design [2], further advances
are needed to reduce the PGM loading and ASC volume while enhancing low
temperature activity.

In this work, the dual-layer concept
is scaled down to the level of a single catalyst particle with the intent to
meet the above-mentioned challenges. A core-shell (CS) particle comprising a
PGM core and a M-Z shell, specifically Pt/Al₂O₃@Cu/ZSM-5
(Figure 1a), was synthesized. Reaction testing of the CS catalyst showed
enhanced low temperature activity without compromising N₂
selectivity on comparing with a physical mixture of catalyst having same Pt and
Cu loading. The CS catalyst had comparable light-off to a Pt/Al₂O₃
catalyst having 3x Pt loading. The enhanced activity is attributed to the
de-stabilization of platinum oxide (PtO_x)
exposing the more active metallic platinum in the CS catalyst, supported by
independent temperature programmed reduction (H₂-TPR) measurements.

Materials and methods

Core-shell
particle synthesis: A Pt/Al₂O₃[Pt(0.05 w/w)/Al₂O₃] catalyst serving
as the core particle was synthesized by incipient wetness impregnation using tetraamine platinum nitrate.
Synthesis of the shell involves seeding of core particles, followed by secondary
growth [3]. The CS particle was sequentially ion exchanged with NH₄NO₃
and Cu(NO₃)₂ to obtain the
Cu-ZSM-5 shell layer.

Reaction
testing: A bench flow fixed-bed reactor, described
elsewhere [2], was used to evaluate the CS powder catalyst. The feed containing
500 ppm NH₃ + 5% O₂ (balance Ar)
had a space velocity of 280k h-1. A FT-IR was used to measure effluent NO, NO₂,
N₂O, and NH₃.

Results and Discussion

Catalyst
characterization: The Si/Al ratio (SAR) of the
zeolite obtained from XPS and EDX data was found to be ~30. Figure 1b shows a SEM
image of prepared CS particles which are 40-50 microns in diameter with a shell
thickness of 1-1.5 microns. From EDX, the Cu weight percent in the shell is ~
2.9. Powder XRD confirmed the MFI lattice structure of the shell.

NH₃ oxidation: NH₃ oxidation was
carried out on 4 samples: Pt(0.15)/Al₂O₃ [P(0.15)-A],
Pt(0.05)/Al₂O₃ [P(0.05)-A], physical mixture of P(0.05)-A
+ ‘x’ wt% Cu(2.9%)-ZSM-5 [CuZ,
x = 10, 20], and the core-shell Pt/Al₂O₃@Cu(2.9%)-ZSM-5 [CS].
The NH₃ oxidation light-off temperatures (T50) were as follows: P(0.15)-A(260°C);
P(0.05)-A(310°C); P(0.05)-A+CuZ(~300°C) and CS (250°C),
as shown in Figure 1d. The T50 of 250°C for the core-shell sample was 50°C
lower than the samples having the same amount of Pt. Remarkably, the CS
light-off temperature was ~5°C less than the sample containing 3x higher Pt(P(0.15)-A).
Moreover, the CS achieved 100% conversion at much lower temperature (325°C)
than P(0.05)-A (450°C). The lower light off for CS
catalyst was due to the increase in NH₃ oxidation activity in the
core of the catalyst. Reaction testing for steady state ammonia oxidation over
different stages of synthesis (coreàseeded coreàCS) as seen in Figure 1f. The seeded core (P(0.05)-AS)
had a light off at 215°C, which is 95°C lower than P(0.05)-A.
The CS catalyst with fully developed shell of ZSM-5 lowered the light off to
250°C due to the diffusion resistance which is still lower than P(0.05)-A. The P(0.05)-A and
P(0.05)-AS have platinum dispersions of 39 and 25 %, respectively. The
dispersion measurement does not explain the lowering of light off. H₂-TPR
showed the formation of H₂O peaks at lower temperature for P(0.05)-AS compared to P(0.15)-A (Figure 1g). The oxides of
Pt are destabilized in seeded core, making it more susceptible to reduction
compared to P(0.05)-A. This exposes more metallic Pt
in the case of seeded core making it more reactive [4]. The destabilization
effect of platinum oxides is due to the modified metal-support interactions
[5,6]. The combined factors of destabilizing effect of platinum oxides,
suppression of platinum oxide formation due to the modification of γ -Al₂O₃
surface leads to the higher activity of the seeded core.
N₂
selectivity: The CS catalyst shows a very high N₂ selectivity,
achieving a maximum N₂ yield of 95% at 300°C, with negligible N₂O
formation (yield < 3%) (Figure 1c). At higher temperature (500°C), N₂
yield drops to ~ 90 % which is much higher than P(0.05)-A
+ CuZ physical mixture (Figure 1e). The high N₂
selectivity of CS catalyst is due to the SCR chemistry taking place in the
Cu-ZSM5 shell [2].

These findings not only confirm
that the dual-layer ASC concept can be achieved at the single particle level,
but there is an apparent added advantage of lower light-off at reduced PGM
loading not obtained previously.

Figure
1: (a)
Core-shell (CS) particle schematic (b) SEM micrograph of synthesized CS
catalyst (c) Steady state NH₃ conversion and product yield for CS
catalyst (d & e )Steady state NH₃ conversion/N₂ yield
for various samples, CS, P(0.05)A, P(0.05)A + 10 wt %
Cu(2.9)-Z, P(0.05)A + 20 wt % Cu(2.9)-Z, P(0.15)A (f)
Steady state NH₃ conversion over different stages of synthesis, core
[P(0.05)-A], seeded-core [P(0.05)-AS] and CS catalysts.[Feed: 500 ppm NH₃,
5 % O₂, balance Ar, 280 K h^-1
and 0.18 g of catalyst] (g) Transient H₂-TPR behavior as a function
of time over P(0.05)-AS (seeded core), P(0.05)-A and Al₂O₃
with 1 % H₂, balance Ar, 30sccm and 100 mg
of catalyst.

References

[1]        S. Shrestha et al., Topics in Catalysis,
56, 182-186 (2013)

[2]        S. Shrestha et al., Catalysis Today,
267, 130-144 (2016)

[3]        Ding W. et al., Chemie-Ingenieur-Technik, 87, 702-712 (2015)

[4]        Völter J. et
al., Journal of Catalysis, 104, 375-380 (1987)

[5]        Huizinga T., et al., Applied Catalysis,
10, 199-213 (1984)

[6]        Yazawa, Y., et al., Applied Catalysis A: General, 237,
149-148 (2002)