Journal of Cleaner Production 434 (2024) 140333

Contents lists available at ScienceDirect

Journal of Cleaner Production

journal homepage: www.elsevier.com/locate/jclepro

Enhanced low-temperature H2-selective catalytic reduction performance

and selectivity of Pt/ZSM-5–TiO2 washcoated monolithic catalysts for
NOx reduction
Kyungseok Lee a, Kyoungbok Lee a, Byungchul Choi b, Kwangchul Oh a, *
a
Advanced Powertrain R&D Department, Future Powertrain Technologies Research Laboratory, Korea Automotive Technology Institute, 303, Pungse-ro, Pungse-myeon,
Dongnam-gu, Cheonan-Si, Chungnam, 31214, Republic of Korea
b
School of Mechanical Engineering, Chonnam National University, 77 Yongbong-ro, Buk-gu, Gwangju, 61186, Republic of Korea

A R T I C L E I N F O A B S T R A C T

Handling Editor: Jin-Kuk Kim Hydrogen-selective catalytic reduction (H2-SCR) is a promising technology for reducing nitrogen oxide (NOx)
emissions at low temperatures, but the selectivity of this process is limited by the generation of N2O. To achieve
Keywords: enhanced denitrification (deNOx) performance, H2-SCR catalysts with mixed zeolite and TiO2 supports wash­
Hydrogen-selective catalytic reduction coated on monolithic cordierite were developed. The catalyst support, active metal type and content, and
Nitric oxide
monolithic cordierite cell density significantly influenced the H2-SCR performance. The xPt/TiO2 catalysts
Washcoated monolithic catalyst
(where x is the metal content, wt.%) achieved >90% NOx conversion 135–180 ◦ C, whereas the xPd/TiO2 cat­
Platinum/ZSM-5
Titanium dioxide alysts exhibited poor H2-SCR activity, with <30% NOx conversion. In contrast to 0.5Pt/TiO2, the 0.5Pt/zeolite
catalysts achieved high NOx conversion at 140–150 ◦ C (91% at 140 ◦ C for 0.5Pt/ZSM-5 > 89% at 140 ◦ C for
0.5Pt/BETA >85% at 150 ◦ C for 0.5Pt/SSZ-13), suggesting that zeolite supports significantly promote H2-SCR
activity at low reaction temperatures. For the 0.5Pt/ZSM-5 washcoated monolithic catalyst, increasing the cell
density from 400 to 900 cpsi significantly improved the NOx conversion from 52% to 90% at 100 ◦ C owing to the
increased geometric surface area and open frontal area of monolithic cordierite. For catalyst with mixed supports
(0.5Pt/yZSM-5zTiO2) changing the ZSM-5/TiO2 ratio (y:z) from 25:75 to 90:10 increased the low-temperature
NOx conversion from 39% to 89% at 100 ◦ C, indicating that the deNOx characteristics can be tuned. With NO
as the reactant, the 0.5Pt/90Z10T catalyst produced the lowest N2O share (8–17%) within the temperature
window of maximum NOx conversion, resulting in an increased N2 share (83–88%). Thus, the synergetic effect of
mixed ZSM-5/TiO2 catalyst supports can enhance low-temperature NOx conversion and mitigate N2O formation.

1. Introduction greenhouse gas that contributes to climate change.

Internal combustion engines (ICEs) currently account for ~10% of
Rapid population growth and industrialization have led to unprec­ global greenhouse gas emissions (Onorati et al., 2022; Reitz et al., 2020).
edented increases in the global energy demand. The total primary en­ Diesel engines, which produce power through lean-burn combustion,
ergy consumption in 2050 is projected to be 57% higher than that in are the most efficient source of power for transportation and critical
2015, which will strain global energy resources and exacerbate global industries (Duan et al., 2022). Despite numerous advantages over stoi­
warming (Hajjari et al., 2017; MIT Joint Program on the Science and chiometric spark-ignition engines, including superior torque, fuel effi­
Policy of Global Change, 2014). Recent statistics show that fossil fuels ciency, durability, and lower CO2 emissions (Hosseini et al., 2023),
such as biofuels, natural gas, and petroleum account for the largest share diesel engines produce high levels of nitrogen oxide (NOx) and partic­
of global final energy consumption (66.3%), followed by electricity ulate matter. Owing to the environmental impact of these emissions,
(19.7%) (IEA, 2021). Thus, the current energy landscape is character­ various legislative strategies have been implemented to gradually
ized by a reliance on fossil fuels. However, the burning of carbon-based replace ICE-powered vehicles with fuel-cell and battery-electric vehicles
fossil fuels releases significant quantities of carbon dioxide (CO2), a (Dong et al., 2019). For example, the European Union has announced a

* Corresponding author. Advanced Powertrain R&D Department, Future Powertrain Technologies Research Laboratory, Korea Automotive Technology Institute
(KATECH), Republic of Korea.
E-mail address: [email protected] (K. Oh).

https://doi.org/10.1016/j.jclepro.2023.140333
Received 19 October 2023; Received in revised form 6 December 2023; Accepted 19 December 2023
Available online 21 December 2023
0959-6526/© 2023 Elsevier Ltd. All rights reserved.
K. Lee et al. Journal of Cleaner Production 434 (2024) 140333

Table 1
Summary of active catalysts and experimental conditions in the H2-SCR process.
Catalyst Preparation Reactant feed condition NOx conversion Catalyst type References
method

NOx O2 (vol H2 (vol H2O (vol GHSV

(ppm) %) %) %) (h− 1)

Pt/HZSM-5 Impregnation 1000 10 0.5 n/a 36,000 90% at 110 ◦ C Granular Wang et al. (2019b)
Pt/Zr–SiO2 Impregnation 480 5 0.8 n/a n/a >95% at Powder Park et al. (2012)
50–300 ◦ C
Pt/SSZ-13 Rotary 1000 10 0.5 n/a 50,000 81% at 100 ◦ C Sieved pellet Hong et al. (2019)
evaporation
Pt/Beta Hydrothermal 1000 4 0.5–1.0 n/a n/a 90% at 90 C
◦
Sprayed on carbon Cao et al. (2020)
paper
PtW/TiO2 Impregnation 2500 5 1 5 53,000 88% at 125 ◦ C Powder Liu et al. (2016)
PtIr/TiO2 Impregnation 2000 5 1 n/a 52,000 >95% at 150 ◦ C Sieved pellet Zhang et al. (2022)
Pt/WO3// Impregnation 500 5 0.5 10 70,000 60% at 150 ◦ C Monolithic Schott et al. (2009)
ZrO2 honeycomb
Pt/ Impregnation 1000 10 0.5 n/a 48,000 86% at 100 ◦ C Granular Liu et al. (2019)
Nb2O5–ZrO2
Pt/MnOx Incipient wetness 480 5 1 n/a 78,000 ~70% at Powder Park et al. (2011)
100–125 ◦ C
Pd/ Impregnation 1000 6 0.4 n/a 20,000 89% at 200 ◦ C Sieved pellet Patel and Sharma
ZrO2–CeO2 (2021)
Pd/ZSM-5 Ion-exchange 500 5 2 5 120,000 ~97% at 150 ◦ C Sieved pellet Moon et al. (2023)
Pd/TiO2 PEG reduction 2000 5 0.8 n/a 35,000 ~80% at 175 ◦ C Sieved pellet Zhang et al. (2019)
Pd/FeTi Impregnation 2000 5 0.8 n/a 60,000 ~95% at 125 ◦ C Sieved pellet Zhang et al. (2021)
Pd–Ni/TiO2 Impregnation 200 1.5 0.2 n/a n/a 99% at 200 ◦ C Powder Hu et al. (2020)
Pd/CeO2 Wet-impregnation 340 2.2 1 n/a 40,000 ~90% at 220 ◦ C Sieved pellet Savva et al. (2021)
Cu/SiO2 Wet-impregnation 2000 n/a 3 n/a 2250 100% at Powder Gioria et al. (2019)
300–500 ◦ C
Co–AlPdO4 Sol-gel auto- 1000 2 1 3 23,400 ~95% at 250 ◦ C Powder Xu et al. (2017)
ignition

ban on the sale and distribution of new gasoline and diesel automobiles, technologies that exclusively produce H2O and N2 as byproducts.
beginning in 2035 (European Parliament, 2022; Frost, 2023). Although not yet industrially mature, H2-SCR can initiate reactions at
Recently, hydrogen (H2) has received considerable interest as a fuel temperatures as low as 150 ◦ C, whereas urea- and hydrocarbon-based
source, primarily because of its potential to significantly reduce the SCR systems have thresholds of 200 ◦ C (Kim et al., 2023; Shelef et al.,
emission of carbon-containing compounds such as CO, CO2, and par­ 1971). The H2-SCR reaction mechanism involves three distinct pro­
ticulate matter. H2 can also achieve high energy efficiencies and be cesses: NO adsorption and dissociation, NO oxidation and reduction, and
generated from renewable energy sources (Sharma and Ghoshal, 2015). dual-function reactions. The overall process can be summarized as fol­
In ICEs fueled by H2 (H2-ICE), the combustion process only produces lows (Guan et al., 2021; Hu and Yang, 2019; Savva et al., 2021):
water (H2O). However, the high flame propagation rate of H2 increases ( / )
2NO + 4H2 + O2 → N2 + 4H2 O ΔH0 = − 574 kJ molNO (1)
the heat generation rate in the combustion chamber, and the resulting
temperature increase in this confined space can produce engine-out NOx ( / )
emissions (Rouleau et al., 2021). NOx formation during combustion, 2NO + 3H2 + O2 → N2 O + 3H2 O ΔH0 = − 412 kJ molNO (2)
which is ascribed to a thermal formation mechanism (Koch et al., 2020), ( / )
is a major contributor to environmental pollution phenomena such as 2H2 + O2 →2H2 O ΔH0 = − 242 kJ molNO (3)
photochemical smog, acid rain, and the greenhouse effect (Fang et al., ( / )
2020; Lee et al., 2021). 2NO + 5H2 → 2NH3 + 2H2 O ΔH0 = − 755 kJ molNO (4)
Promising technologies for abatement of NOx in the exhaust emis­ The desired H2-SCR reaction (Equation (1)) is environmentally
sion of H2-ICEs include three-way catalysts, selective catalytic reduction benign, yielding N2 and H2O. However, unanticipated side reactions
(SCR), and three-way NOx storage catalysts (Sterlepper et al., 2021). involving N2O generation (Equation (2)) and H2 oxidation (Equation
Although, the urea-SCR has emerged as the predominant aftertreatment (3)) can detrimentally affect the N2 selectivity and catalytic efficiency.
technology for reducing NOx emissions, practical applications are Hence, suppressing these reactions during the H2-SCR process is critical.
limited by a restricted exhaust temperature range for efficient operation. The formation of NH3 via Equation (4) is an intermediate reaction (NH3
Typically, a light-off temperature of ~200 ◦ C with an optimal operation adsorption on Brønsted acid site → 4NH+ 4 + 4NO + O2 → 4N2 + 6H2O +
window of 250–450 ◦ C is required to decompose the injected urea so­ 4H+), which plays a vital role in H2-SCR catalytic activity (Hong et al.,
lution into NH3 (Börnhorst and Deutschmann, 2021). However, in 2018; Liu et al., 2020b).
hybrid electric vehicles with lean-burn H2-ICEs, low exhaust tempera­ Extensive investigations on noble-metal-based catalysts, particularly
tures under cold-start conditions can hinder activation of the urea-SCR Pt, Pd, and Rh, have revealed a hierarchy in catalytic activity, with Pt/
system. In the H2-ICE exhaust, several volume percent of unburned H2 aluminum oxide (Al2O3) exhibiting higher activity than Pd/Al2O3 and
can be emitted through the vehicle’s tailpipe during cold-start and Rh/Al2O3 (Granger et al., 2006). Pt exhibits remarkable activity for the
intermittent stop-and-go operations. Thus, the incorporation of a H2-SCR process at lower temperatures (Gholami et al., 2020; Guan et al.,
low-temperature H2-SCR catalyst upstream of the urea-SCR system can 2021; Hu and Yang, 2019; Zhang et al., 2020). The performance of
reduce NOx emissions during re-ignition events, stop-and-go traffic, and Pt-based catalysts depends on multiple factors, including the type of
intermittent engine operation. support material (Alghamdi et al., 2020; Li et al., 2015; Satsuma et al.,
Following the groundbreaking work (Shelef et al., 1971), which 2003; Zhang et al., 2020), Pt loading (Zhang et al., 2024), metal–support
demonstrated the viability of H2-SCR for NOx mitigation, H2-SCR cata­ interactions (Liu et al., 2020b), Pt valence state (Cao et al., 2020; Kim
lysts have emerged as alternative clean and eco-friendly deNOx et al., 2023), and preparation method (Hong et al., 2018; Wang et al.,

2
K. Lee et al. Journal of Cleaner Production 434 (2024) 140333

Fig. 1. Experimental apparatus for testing catalytic reaction of H2-SCR process.

2019a). Compared with Pt-based catalysts, Pd-based catalysts show min). The as-prepared catalysts were denoted as xPt/TiO2 and xPd/
superior N2 selectivity, but their applicability is limited by a narrow TiO2, where x is the theoretical Pt or Pd loading (0.5–1.5 wt%).
temperature range. Poor N2 selectivity hinders efficient NOx removal in To prepare Pt/Al2O3, the required quantity of the Pt precursor
the dynamic temperature range typically encountered in exhaust sys­ (H2PtCl6⋅xH2O) was introduced into a mixture of distilled H2O and
tems (Duan et al., 2015). Attempts to enhance the activity of Pd-based γ-Al2O3 nanopowder (particle size: <50 nm; Merck 544833-50G). All
catalysts have focused on manipulating the Pd valence states using the subsequent steps were similar to those employed for the preparation
diverse support materials (Patel and Sharma, 2021; Peng et al., 2017) or of Pt/TiO2. The Pt loading was 0.5 wt%, and the resulting catalyst was
incorporating catalytic additives (Hu et al., 2020, 2021; Zhang et al., denoted as 0.5Pt/Al2O3.
2023a). Although noble-metal-based catalysts have impressive capa­ Three different Pt/zeolite catalysts were synthesized via a wet-
bilities for reducing NOx emissions at lower temperatures, the decrease impregnation method using a Pt precursor (H2PtCl6⋅xH2O) and H-SSZ-
in selectivity caused by N2O formation remains an issue (Li et al., 2015; 13 with CHA topology (Si/Al ratio = 80, AHA International), H-ZSM-5
Liu et al., 2018, 2020b). with MFI topology (Si/Al ratio = 80, AHA International), or H-BETA
In this study, we developed catalysts based on noble metals and with BEA topology (Si/Al ratio = 40, AHA International). After adding
various catalyst supports, with the aim of improving the low- the Pt precursor to a mixture of the required zeolite and distilled H2O,
temperature deNOx performance of H2-SCR while maintaining good the following steps were similar to those in the previously described
N2 selectivity. Most reported H2-SCR catalysts have been assessed in methods. The Pt loading was 0.5 wt%, and the as-prepared catalysts
powder or pellet form (Table 1). To enhance their practical applicability, were denoted as 0.5Pt/yZzT, where y and z represent the mass ratios of
our catalysts were instead applied as washcoatings on monolithic hon­ H-ZSM-5 and TiO2, respectively. The methods employed for catalyst
eycomb substrates. The influence of various aspects, including the metal characterization are detailed in the Supplementary Data.
loading in Pd- or Pt-based catalysts supported on TiO2, the type of
catalyst support (γ-Al2O3, SSZ-13, ZSM-5, and BETA zeolite or mixed 2.2. Preparation of washcoated monolithic cordierite substrate
supports with varying ZSM-5/TiO2 ratios) for Pt-based catalysts, and the
use of monolithic cordierite with varying cell densities and wall thick­ Washcoated monolithic catalysts offer many advantages, including
nesses, on catalytic performance were experimentally assessed. numerous narrow channels and inner pores, which facilitate efficient
Furthermore, an improved understanding of the influence of the reac­ mass and heat transfer, excellent thermal and mechanical durability.
tant gas composition and gas hourly space velocity (GHSV) on perfor­ Compared with powder and pellet catalysts, washcoated monolithic
mance was obtained, which can aid in advancing the development of H2- catalysts provide lower pressure drops and larger surface areas and can
SCR catalysts. also withstand carbon and dust deposition during catalytic reactions
(Huang et al., 2023; Wang et al., 2019b). For practical applications, the
2. Materials and methods H2-SCR catalysts were affixed to a ceramic block of monolithic cordierite
(2MgO⋅2Al2O3⋅5SiO2) with regular square-shaped channels. The
2.1. Catalyst preparation as-prepared catalyst powder was applied as a washcoating on the
open-structured monolithic cordierite substrate with a cell density of
Pt/TiO2 and Pd/TiO2 catalysts were synthesized via a wet- 400 cells per square inch (cpsi) and wall thickness of 6 mil. The substrate
impregnation procedure using chloroplatinic acid hydrate had an external diameter of 19.3 mm and lengths of 27.3, 13.7, 9.1, and
(H2PtCl6⋅xH2O; Pt 37–40%, Merck 520896-5G), palladium(II) chloride 6.8 mm, resulting in volumes of 8.0, 4.0, 2.66, and 2.0 cm3, respectively.
(PdCl2; Pd 59–60%, Merck 520659-5G), and a titanium(IV) oxide (TiO2) The substrate length was varied to investigate the effect of the GHSV on
nanopowder (particle size: ≤21 nm, Merck 718467-100G). The required catalytic performance. A dilute slurry was prepared by adding 10 wt%
quantities of the noble metal precursors were dissolved in 80 mL of powdered catalyst to distilled H2O without a binder and stirring for 24 h.
distilled H2O in the presence of TiO2 as a support. The mixture was Subsequently, the substrate was dipped in the slurry, dehydrated in an
vigorously stirred using a polytetrafluoroethylene hexagonal magnetic electric muffle furnace at 250 ◦ C for 15 min, and calcined at 500 ◦ C for 1
stirring bar at 350 rpm for 1.5 h under ambient conditions. Excess H2O h (Lee and Choi, 2021). The quantity of washcoat applied was 180 g per
was removed through overnight evaporation at 85 ◦ C in a drying oven unit volume of monolithic cordierite. Within the washcoat, the Pt con­
(BF-24VDO, BNF Korea). The dehydrated solid was finely ground using tent ranged from 0.9 g/L (0.5 wt% Pt loading) to 2.7 g/L (1.5 wt% Pt
an agate mortar and pestle, placed in a disposable porcelain boat, and loading). Furthermore, the influence of the cell density of the monolithic
then calcined in an electric muffle furnace (cylindrical ceramic heater, cordierite substrate on the H2-SCR performance was investigated.
DAONRS) at 500 ◦ C for 5 h (heating rate: 5 ◦ C/min; air flow rate: 2.5 L/

3
K. Lee et al. Journal of Cleaner Production 434 (2024) 140333

Table 2
Metal-element analysis of the H2-SCR catalysts analyzed by XRF.
Catalysts Pt (wt. Pd (wt. Ti (wt. Al (wt. Si (wt. Si/Al
%) %) %) %) %) ratio

0.5Pt/TiO2 0.51 n.d. 99.49 n.d. n.d. n.d.

1Pt/TiO2 1.08 n.d. 98.92 n.d. n.d. n.d.
1.5Pt/TiO2 1.59 n.d. 98.41 n.d. n.d. n.d.
0.5Pd/TiO2 n.d. 0.45 99.55 n.d. n.d. n.d.
1Pd/TiO2 n.d. 0.98 99.02 n.d. n.d. n.d.
1.5Pd/TiO2 n.d. 1.41 98.59 n.d. n.d. n.d.
0.5Pt/Al2O3 0.59 n.d. n.d. 99.41 n.d. n.d.
0.5Pt/SSZ- 0.53 n.d. n.d. 1.28 98.19 76.7
13
0.5Pt/ZSM- 0.47 n.d. n.d. 1.31 98.22 75.0
5
0.5Pt/BETA 0.52 n.d. n.d. 2.62 96.86 37.0
0.5Pt/ 0.51 n.d. 91.22 0.11 8.16 74.2
10Z90T
0.5Pt/ 0.54 n.d. 77.23 0.29 21.94 75.7
25Z75T
0.5Pt/ 0.49 n.d. 52.40 0.59 46.52 78.8
50Z50T
0.5Pt/ 0.54 n.d. 26.41 0.97 72.08 74.3
75Z25T
0.5Pt/ 0.53 n.d. 9.27 1.12 89.08 79.5
90Z10T

n.d.: not detected.

2.3. Catalytic performance tests

The H2-SCR reaction was performed in a vertical fixed-bed reaction

apparatus under atmospheric pressure (Fig. 1 and S1). The experimental
apparatus comprised a mass flow meter (HFC-202, Teledyne Hastings),
digital flow controller (KRO-4000), gas-washing bottle (Pyrex, Sigma-
Aldrich) to provide H2O, quartz reactor (diameter: 19.5 mm; height:
300 mm), ceramic electric tubular furnace (ARF-40KC, ASH), Dimroth
condenser (Pyrex) that employed antifreeze circulation at − 25 ◦ C to
capture vapor, and FTIR gas analyzer (Midac I-2004). The cylindrical
quartz reactor was encircled by an electric furnace for precise control of
the catalyst inlet temperature, which was increased from 100 to 350 ◦ C
at a rate of 5 ◦ C/min using a PID programmable controller (NP200,
HANYOUNG nuX) connected to a K-sheath thermocouple located ~10
mm in front of the washcoated substrate. The standard gaseous feed was
composed of 500 ppm NO, 5 vol% O2, 1 vol% H2, and 5 vol% H2O, with
the balance consisting of N2. To evaluate the effects of the O2 and H2
concentrations, the O2 feed was varied from 5 to 15 vol%, and the H2
feed was varied from 1 to 2 vol%. The total feed stream was delivered at
a rate of 2000 mL/min, which corresponds to a GHSV of 15,000 h− 1. The
influence of the GHSV in the range of 15,000–60,000 h− 1 was assessed
by varying the volumes of washcoated cordierite. The feed reactants
were supplied through a 1/4-inch stainless-steel tube covered with
Fig. 2. XRD patterns of (a) xPt/TiO2 and xPd/TiO2, (b) 0.5Pt/zeolite, and (c)
precision heat tape (Clayborn Lab) to maintain a temperature at 150 ◦ C
0.5Pt/yZzT catalysts.
and prevent H2O condensation. To determine the reactant and product
concentrations, data recorded every 10 s using the FTIR gas analyzer
were analyzed using AutoQuant Pro 4.5. where, [NO]in and [NO]out are the NOx concentrations at the inlet and
outlet, respectively.
The effectiveness of the H2-SCR reaction was evaluated based on the
2.4. Estimation of catalytic performance
proportion of NO that reacted to form N-containing products, including
NO2, N2O, NH3, and N2. The catalytic selectivity of the H2-SCR reaction
Owing to the presence of O2, NO undergoes partial oxidation to NO2
was determined using the following equation:
in the gas phase at all reaction temperatures. Therefore, the H2-SCR
catalytic activity was determined based on NOx conversion (ηNOx) rather [NO2 ]out
NO2 share (%), SNO2 = × 100 (7)
than NO conversion (Zhao et al., 2015). The catalytic conversion of NOx [NO]in − [NO]out
was determined as follows:
2[N2 O]out
NOx concentration, [NOx] = [NO] + [NO2 ] (5) N2 O share (%), SN2O = × 100 (8)
[NO]in − [NO]out
[NOx]in − [NOx]out
ηNOx (%) = × 100 (x = 1, 2) (6) [NH3 ]out
[NOx]in NH3 share (%), SNH3 = × 100 (9)
[NO]in − [NO]out

4
K. Lee et al. Journal of Cleaner Production 434 (2024) 140333

Table 3 Table 4
Estimated d-spacing and crystallite size of H2-SCR catalysts determined by Textural characterization for BET specific surface area and pore properties of H2-
powder XRD patterns (TiO2 at 2θ = 25.3◦ , γ-Al2O3 at 2θ = 67.1◦ , SSZ-13 at 2θ = SCR catalyst analyzed by N2-isotherms.
9.6◦ , ZSM-5 at 2θ = 7.9◦ , BETA at 2θ = 22.4◦ for estimation of d-spacing and Catalysts SBET SM (m2/ VT VM VS DP
crystallite size). (m2/g)a g)b (cm3/ (cm3/ (cm3/ (nm)f
Catalysts Relative crystallinity d-spacing (nm) Crystallite size g)c g)d g)e
(%) (nm) TiO2 64.4 0 0.45 0.01 0.45 37.1
TiO2 100 0.352 18.4 0.5Pt/TiO2 65.4 0 0.45 0 0.44 34.1
0.5Pt/TiO2 88.2 0.352 21.2 1Pt/TiO2 67.2 0 0.44 0 0.43 32.3
1Pt/TiO2 83.7 0.352 20.5 1.5Pt/TiO2 67.7 0 0.41 0 0.41 32.4
1.5Pt/TiO2 82.7 0.353 20.9 0.5Pd/ 63.1 0 0.43 0 0.42 27.3
0.5Pd/TiO2 83.1 0.352 21.0 TiO2
1Pd/TiO2 76.3 0.352 20.4 1Pd/TiO2 60.9 0 0.44 0 0.44 28.8
1.5Pd/TiO2 75.9 0.351 20.5 1.5Pd/ 57.7 0 0.41 0 0.41 28.4
γ-Al2O3 100 0.140 0.9 TiO2
0.5Pt/Al2O3 87.9 0.140 0.8 γ-Al2O3 141.2 20.8 1.23 0.01 1.22 33.3
SSZ-13 100 0.922 87.4 0.5Pt/ 130.7 11.7 0.99 0.01 0.97 30.9
ZSM-5 100 1.120 47.0 Al2O3
BETA 100 0.396 22.2 SSZ-13 782.7 736.3 0.35 0.27 0.07 3.1
0.5Pt/SSZ-13 92.9 0.922 95.3 ZSM-5 422.7 322.9 0.28 0.13 0.12 3.4
0.5Pt/ZSM-5 99.6 1.117 51.0 BETA 624.7 463.3 0.82 0.19 0.59 3.1
0.5Pt/BETA 87.5 0.397 28.7 0.5Pt/SSZ- 762.4 707.3 0.35 0.26 0.07 3.0
0.5Pt/ 83.1 Z: 1.114, T: Z: 23.8, T: 20.9 13
10Z90T 0.352 0.5Pt/ 435.4 325.8 0.31 0.13 0.14 3.1
0.5Pt/ 78.7 Z: 1.111, T: Z: 47.9, T: 20.7 ZSM-5
25Z75T 0.352 0.5Pt/ 600.1 415.6 0.70 0.17 0.47 3.4
0.5Pt/ 71.6 Z: 1.116, T: Z: 59.8, T: 20.9 BETA
50Z50T 0.352 0.5Pt/ 82.7 12.6 0.43 0.01 0.41 32.9
0.5Pt/ 53.4 Z: 1.114, T: Z: 62.7, T: 21.2 10Z90T
75Z25T 0.352 0.5Pt/ 141.6 73.9 0.46 0.03 0.42 31.1
0.5Pt/ 29.2 Z: 0.112, T: Z: 53.2, T: 21.0 25Z75T
90Z10T 0.352 0.5Pt/ 244.1 169.8 0.37 0.07 0.28 3.4
50Z50T
0.5Pt/ 338.7 228.7 0.42 0.09 0.29 3.4
75Z25T
N2 share (%), SN2 = 100 − (SNO2 − SN2O − SNH3 ) (10) 0.5Pt/ 391.1 302.2 0.31 0.12 0.16 3.1
90Z10T
3. Results and discussion a
SBET (BET specific surface area): calculated by BET method.
b
SM (micropore area): calculated by t-plot method.
3.1. Catalyst characterization c
VT (total pore volume less than 300 nm in diameter): calculated by BET
method at P/P0 = 0.99.
d
3.1.1. Apparent metal loading VM (micropore volume): calculated by t-plot method.
e
The contents of metallic elements impregnated in the prepared H2- VS (mesopore and macropore volume): calculated using adsorption curve of
SCR catalysts, as determined by X-ray fluorescence (XRF), are summa­ BJH method.
f
DP (average pore diameter): calculated using adsorption curve of BJH
rized in Table 2. The apparent loadings of metallic Pt and Pd in all the
method.
H2-SCR catalysts are in good agreement with the theoretical loading
amounts, indicating that the Pt and Pd precursors were successfully
impregnated into the TiO2, Al2O3, and zeolite (SSZ-13, ZSM-5, and The XRD patterns of the 0.5Pt/zeolite catalysts are shown in Fig. 2b.
BETA) supports without any losses. Additionally, the Si/Al ratios of the For 0.5Pt/SSZ-13, the diffraction peaks at 9.6◦ , 13.1◦ , 16.2◦ , 18.0◦ ,
0.5Pt/zeolite catalysts are similar to those of the parent zeolites, with 20.9◦ , 25.3◦ , 26.3◦ , and 31.1◦ are characteristic of the well-ordered
0.5Pt/SSZ-13 and 0.5Pt/ZSM-5 having Si/Al ratios of 76.7 and 75.0, chabazite structure of SSZ-13 zeolite (Hong et al., 2019). For
respectively, and 0.5Pt/BETA having a Si/Al ratio of 37.0. Each 0.5Pt/ 0.5Pt/ZSM-5, diffraction peaks corresponding to the MFI structure in
yZzT catalyst has a Pt content of the same order of magnitude as the ZSM-5 appear at 7.9◦ , 8.8◦ , 23.1◦ , 24.0◦ , and 24.4◦ (Moon et al., 2023).
theoretical loading and a Si/Al ratio of 74.2–79.5, confirming that Pt Similarly, for 0.5Pt/BETA, diffraction peaks characteristic of the BEA
was impregnated into the mixed H-ZSM-5/TiO2 supports at the desired topology of BETA at 7.7◦ , 9.8◦ , 13.5◦ , and 22.4◦ (Lee et al., 2019). The
loading amount. XRD peaks corresponding to ZSM-5 in the 0.5Pt/yZzT catalysts increase
in intensity with an increase in the ZSM-5 ratio from 0.5Pt/10Z90T to
3.1.2. Crystalline structure 0.5Pt/90Z10T, whereas the TiO2 peaks become gradually weaker
X-ray diffraction (XRD) was used to obtain quantitative structural (Fig. 2c). Similar to 0.5Pt/TiO2, no Pt species are observed for the
information about the H2-SCR catalysts (Fig. 2 and Fig. S2). The XRD 0.5Pt/zeolite and 0.5Pt/yZzT catalysts because of the high dispersion of
patterns of Pt/TiO2 and Pd/TiO2 (Fig. 2a) exhibit sharp diffraction peaks Pt on the zeolites and the mixed ZSM-5/TiO2 supports. The decrease in
at 25.3◦ , 37.8◦ , 48.1◦ , 54.1◦ , 55.1◦ , 62.8◦ , 70.4◦ , and 75.2◦ , corre­ the relative crystallinity of TiO2 with decreasing TiO2 content is
sponding to the anatase phase of TiO2. The minor peaks at 27.5◦ , 36.1◦ , consistent with the above explanation for the 0.5Pt/yZzT catalysts
and 68.9◦ reveal the presence of the rutile phase of TiO2 (Duan et al., (Table 3). The d-spacings of the H2-SCR catalysts and parent supports are
2019; Zhang et al., 2019), consistent with the parent P25 TiO2, which similar, indicating that no significant lattice expansion, contraction, or
comprised 80% anatase and 20% rutile phases (Liu et al., 2020a). The collapse occurred after impregnation with the metallic species.
crystalline phase of the TiO2 support is maintained after impregnation
with Pt or Pd. However, no diffraction patterns corresponding to Pt or Pd 3.1.3. Textural characteristics
species are observed, suggesting that the noble metals were highly Table 4 presents the textural properties of the parent supports and
dispersed on the TiO2 support or the loading amount was too low to be H2-SCR catalysts, as derived from N2-isotherms (Fig. S3). The BET sur­
detected. face area (SBET) and pore volumes (VT, VM, and VS) do not change

5
K. Lee et al. Journal of Cleaner Production 434 (2024) 140333

Fig. 3. Elemental analysis of high-resolution scanning transmission electron microscope-energy-dispersive X-ray spectroscopy (TEM-EDS) mapping: (a) 0.5Pt/TiO2,
(b) 0.5Pd/TiO2, (c) 0.5Pt/Al2O3, (d) 0.5Pt/ZSM-5, (e) 0.5Pt/10Z90T, (f) 0.5Pt/50Z50T, and (g) 0.5Pt/90Z10T.

significantly with increasing Pt or Pd loading on the TiO2 support. average pore diameter of 27.3–34.1 nm.
However, the BET surface areas of the 0.5Pt/yZzT catalysts gradually
increase as the ZSM-5 ratio increases because ZSM-5 has a higher BET 3.1.4. Metal distribution and surface morphology
surface area (422.7 m2/g) than TiO2 (64.4 m2/g). The high BET surface The distribution of the impregnated metals in the H2-SCR catalysts
areas of the zeolite-supported catalysts were attributed to the presence was revealed by transmission electron microscopy (TEM)-energy-
of micropores, which have high micropore areas and small average pore dispersive X-ray spectroscopy (EDS) elemental mapping (Fig. 3 and S4)
diameters. The relatively low BET surface areas of the xPt/TiO2 and and comparison with the parent supports without loaded metals
xPd/TiO2 catalysts were attributed to the absence of micropores, (Fig. S5). For the xPt/TiO2 and xPd/TiO2 catalysts, the major elements
resulting in a large proportion of mesopores and macropores with an are Ti and O, and the distribution density of Pt or Pd increases with

6
K. Lee et al. Journal of Cleaner Production 434 (2024) 140333

Fig. 4. Field emission scanning electron microscope (FE-SEM) micrographs of H2-SCR catalyst.

increasing loading from 0.5 to 1.5 wt%. In the 0.5Pt/zeolite catalysts, Al, distributed on the catalyst support would not be detectable.
Si, and O are derived from the zeolites, which are composed of TO4 The morphologies of the catalyst surfaces in the field emission
tetrahedra (T = Si or Al) connected by corner-shared O atoms (Price scanning electron microscopy (FE-SEM) micrographs are similar to those
et al., 2021). Increasing the ZSM-5 ratio from 10 to 90 in the 0.5Pt/yZzT in the TEM micrographs (Fig. 4 and S6). The morphology of each sup­
catalysts induces a significant morphological transition from TiO2 to port is retained after impregnation with the Pt or Pd species, suggesting
ZSM-5, with the Ti content on the catalyst surface gradually decreasing that metal loading did not induce morphological transformation of the
and the Si content gradually increasing. For the 0.5Pt/10Z90T catalyst, supports. This lack of transformation was also supported by the negli­
Si and Al atoms are not detected due to the lowest mixing ratio of ZSM-5 gible variation in the d-spacings of the metal-impregnated catalysts, as
support. TEM-EDS analysis revealed that all the elements were homo­ determined by XRD analysis. As the ZSM-5 ratio in the 0.5Pt/yZzT cat­
geneously dispersed on the catalyst surfaces; however, the particle sizes alysts increases, the surface morphology of ZSM-5 becomes more prev­
of the impregnated metals could not be determined because the particles alent, and uniformly mixed ZSM-5 and TiO2 morphologies are observed
were too small for TEM analysis. These results are reasonable and in the SEM micrographs. Owing to the low metal loading, the Pt and Pd
consistent with the lack of Pt and Pd species observed in the XRD particles on the catalyst surfaces are indistinguishable, suggesting that
analysis, as small metallic clusters of Pt and Pd species uniformly many nanoscale metallic particles were dispersed on the supports

7
K. Lee et al. Journal of Cleaner Production 434 (2024) 140333

Fig. 5. XPS spectra of Pt 4f and Pd 3d for H2-SCR catalysts: (a) xPt/TiO2, (b) xPd/TiO2, (c) 0.5Pt/zeolite and 0.5Pt/Al2O3, and (d) 0.5Pt/yZzT (Pt0 ratio = (Pt0/(Pt0
+ Pt2+ + Pt4+)) × 100, Pt2+ ratio = (Pt2+/(Pt0 + Pt2+ + Pt4+)) × 100, Pt4+ ratio = (Pt4+/(Pt0 + Pt2+ + Pt4+)) × 100, Pd0 ratio = (Pd0/(Pd0 + Pd2+)) × 100, and Pd2+
ratio = (Pd2+/(Pd0 + Pd2+)) × 100).

without severe agglomeration. 0.5Pt/Al2O3 and 0.5Pt/zeolite catalysts could be identified (Fig. 5c). The
Pt 4f7/2 and Pt 4f5/2 peaks are located at 72.6 eV and 72.4 eV, corre­
3.1.5. Valence states of impregnated metals sponding to Pt2+, and 76.7–76.8 eV, corresponding to Pt4+, for
The valence states of the active metal species of the catalyst surfaces 0.5Pt/SSZ-13 and 0.5Pt/BETA, respectively. The Pt2+/Pt4+ ratio is not
were determined using X-ray photoelectron spectroscopy (XPS). The Pt affected by the zeolite support. For 0.5Pt/ZSM-5, a new peak at 71.7 eV
4f spectra of all the xPt/TiO2 catalysts (Fig. 5a) contain deconvoluted corresponds to metallic Pt. The ratios of Pt0 and Pt4+ are approximately
peaks at 71.7–71.8 eV (Pt 4f7/2), 72.6–72.7 eV (Pt 4f7/2) and 75.8–75.9 52.1% and 49.7%, respectively. As shown in the Pt 4f spectra of the
eV (Pt 4f5/2), and 74.5–74.6 eV (Pt 4f7/2) and 77.7–77.9 eV (Pt 4f5/2), 0.5Pt/yZzT catalysts (Fig. 5d), increasing the ZSM-5 ratio from 10 to 75
corresponding to Pt0, Pt2+ (as in PtO), and Pt4+ (as in PtO2) species, decreases the Pt2+ ratio from 68.4% to 63.0%, with a concomitant in­
respectively (Kim et al., 2023; Zhang et al., 2022). Increasing the Pt crease in the Pt4+ ratio from 31.6% to 37.0%. However, for the
content from 0.5 to 1.5 wt% induces a decrease in the Pt0 ratio from 0.5Pt/90Z10T, with the highest ratio of ZSM-5, Pt4+ and Pt0 peaks
14.7% to 10.4%. Deconvolution of the Pd 3d5/2 and Pd 3d3/2 spectra overlap with the Al 2p peak, likely because of the higher content of
(Fig. 5b) gives peaks at 335.5–335.8 eV and 340.8–341.1 eV, charac­ ZSM-5 in the support. The Pt 4f and Al 2p peaks of 0.5Pt/90Z10T are
teristics of Pd0 species, and 336.9–337.1 eV and 342.2–342.3 eV, similar to those of 0.5Pt/ZSM-5, and the Pt0 and Pt4+ ratios are
characteristic of Pd2+ species (as in PdO) (Patel and Sharma, 2021; approximately 46.6% and 53.4%, respectively.
Zhang et al., 2023b). The Pd0 ratio also decreases slightly from 41.1%
(0.5 wt% Pd) to 39.0% (1.5 wt% Pd). Although the Pt 4f and Al 2p peaks
are known to overlap, the characteristic peaks of the Pt species in the

8
K. Lee et al. Journal of Cleaner Production 434 (2024) 140333

Fig. 6. Effect of Pt and Pd loading on the H2-SCR performance over TiO2 support: (a)–(d) xPt/TiO2 and (e)–(h) xPd/TiO2 (open symbol: N2O, closed symbol: N2 in (b)
and (f)). Experimental condition: 500 ppm NO, 5 vol% O2, 1 vol% H2, 5 vol% H2O, and GHSV = 15,000 h− 1.

3.2. Influence of metal loading on H2-SCR activity et al., 2018; Liu et al., 2016). Although the xPt/TiO2 catalysts achieve
>90% NOx conversion at 135–180 ◦ C, the xPd/TiO2 catalysts exhibit
The effects of metal loading (0.5–1.5 wt%) on the H2-SCR perfor­ poor H2-SCR activity (<30% NOx conversion). Changing the Pd loading
mance of the xPt/TiO2 and xPd/TiO2 catalysts were evaluated as a from 0.5 to 1.5 wt% increases the maximum NOx conversion from 16%
function of reaction temperature in the range of 100–350 ◦ C (Fig. 6). The to 29% at 160 ◦ C. However, at the initial temperature of the H2-SCR test,
xPt/TiO2 catalysts exhibit superior NOx conversion compared with xPd/ the NOx conversion is negative because NOx species are desorbed from
TiO2 catalysts (Fig. 6a and e). For the xPt/TiO2 catalysts, NOx conver­ the catalyst and then reduced by H2.
sion increases sharply from 120 ◦ C, reaching a maximum of ~97% at N2O was the main by-product of the H2-SCR process on both xPt/
140–170 ◦ C, which is a typical volcanic-type profile. Subsequently, NOx TiO2 and xPd/TiO2, with the formation of <1 ppm NH3 and unreacted
conversion gradually decreases to ~5% at 350 ◦ C because the unselec­ NO2 being oxidized by NO in the presence of coexisting O2. As shown in
tive combustion of H2 with coexisting O2 depletes the amount of H2 Fig. 6b, the formation of N2O gradually increases with NOx conversion
available for NOx reduction as the reaction temperature increases (Hong on the xPt/TiO2 catalysts, reaching a maximum of 33–35 ppm at

9
K. Lee et al. Journal of Cleaner Production 434 (2024) 140333

Fig. 7. Comparison of H2-SCR performance over 0.5Pt/TiO2 and 0.5Pt/Al2O3: (a) NOx conversion, (b) N2 and N2O formation (open symbol: N2O, closed symbol: N2),
(c) NO2 concentration and NO2/NOx ratio, and (d) NO share in reaction route and reacted NO ratio. Experimental condition: 500 ppm NO, 5 vol% O2, 1 vol% H2, 5
vol% H2O, and GHSV = 15,000 h− 1.

~150 ◦ C and then leveling off at 270 ◦ C. Notably NO is oxidized by O2 to loading of 0.5 wt% (Fig. 7). As shown in Fig. 7a, NOx conversion over
produce NO2, which then participates in the H2-SCR reaction, as sup­ 0.5Pt/Al2O3 only occurs above 140 ◦ C, reaching a maximum of 89% at
ported by the observation that NO2 was completely reduced at tem­ 160 ◦ C. The NOx conversion then decreases sharply to 20% at 250 ◦ C,
peratures between 130 and 170 ◦ C (Fig. 6c), where the xPt/TiO2 and no H2-SCR activity is observed above 300 ◦ C. 0.5Pt/Al2O3 exhibits
catalysts achieved maximum NOx conversion, irrespective of the Pt significantly lower H2-SCR activity than 0.5Pt/TiO2 over the entire
loading amount. A similar phenomenon was observed for the xPd/TiO2 temperature range. In addition, more N2O is formed over 0.5Pt/Al2O3,
catalysts (Fig. 6g). with a peak concentration of 43 ppm occurring at 160 ◦ C (Fig. 7b). Thus,
The ratios of reacted NO to N-containing products, N2, NO2, N2O, 0.5Pt/Al2O3 has a stronger tendency to induce side reactions, resulting
and NH3, were evaluated to assess the effectiveness of the H2-SCR pro­ in high N2O share of 17% at 160 ◦ C (Fig. 7d). The high N2O production
cess. Fig. 6d and h shows the ratios of NO consumption for 0.5Pt/TiO2 and inferior deNOx performance of 0.5Pt/Al2O3 suggest that this cata­
and 0.5Pd/TiO2 relative to the competing reaction routes for NO lyst is not suitable for emission control applications. The NO2 profile of
reduction to N2, N2O, and NH3 (Equation (1), (2), and (4)) and NO 0.5Pt/Al2O3 is similar to those of the Pt/TiO2 and Pd/TiO2 catalysts,
oxidation to NO2 in the H2-SCR process. More than 55% NO2 from with a significant decrease in the NO2 concentration occurring between
reacted NO did not participate in the H2-SCR reaction over 0.5Pt/TiO2 140 and 160 ◦ C during the light-off period (Fig. 7c). Previous studies
below 120 ◦ C because the catalyst was not activated at this temperature have shown that H2 spillover, a crucial step in NOx reduction, is
(Fig. 6d). However, little NO2 was observed between 140 and 180 ◦ C, involved in enhancing the performance of H2-SCR catalysts (Hu and
leading to an increase in the N2 share from 23% at 100 ◦ C to 87% at Yang, 2019; Liu et al., 2020b). H2 spillover is believed to occur on
160 ◦ C. Increasing the reaction temperature and NOx conversion reducible supports (e.g., TiO2), where active hydrogen atoms migrate to
increased N2O formation by undesirable side reactions, causing the N2O the metal–support interface (Hu et al., 2020; Messou et al., 2021).
share to increase gradually from 11% at 100 ◦ C to 17% at 140 ◦ C and However, the occurrence of H2 spillover on non-reducible supports (e.g.,
then plateau at ~13% at 160–220 ◦ C. A relatively low N2O share was γ-Al2O3) is controversial, with some studies reporting that this process is
observed at 160 ◦ C, where NOx conversion over 0.5Pt/TiO2 was maxi­ unfavorable (Beaumont et al., 2014). Therefore, H2 spillover at the
mized. As the ratio of reacted NO for 0.5Pd/TiO2 (<20%) was signifi­ metal–support interface could enhance NOx conversion over
cantly lower than that for the xPt/TiO2 catalysts (Fig. 6h), NOx 0.5Pt/TiO2.
conversion over 0.5Pd/TiO2 did not exceed 20% (Fig. 6e). The N2O
share in reacted NO was higher than that of 0.5Pt/TiO2 between 180 ◦ C 3.4. Influence of H2 and O2 on the catalytic temperature window
(17%) and 300 ◦ C (29%). A similar trend was observed for the xPd/TiO2
catalysts (Figs. S7c and d). Considering the catalytic activity, by-product To investigate the effects of the H2 and O2 feed concentrations on the
formations, and N2 share after the H2-SCR reaction, Pt/TiO2 is more H2-SCR performance, 0.5Pt/TiO2 was selected as a reference catalyst
suitable than Pd/TiO2 as a deNOx aftertreatment catalyst. because its H2-SCR performance was acceptable despite having the
lowest Pt loading. Increasing the H2 feed concentration shifts the onset
3.3. Comparison of TiO2 and Al2O3 supports temperature for NOx conversion to lower temperatures (from 130 ◦ C at
1 vol% H2 to 110 ◦ C at 1.5 and 2 vol% H2, Fig. 8a). Although the
The effect of the catalyst support on H2-SCR activity was investigated maximum NOx conversion of ~98% was not affected by the H2 con­
by comparing 0.5Pt/TiO2 and 0.5Pt/Al2O3, which have a fixed Pt centration, NOx conversion increased at temperatures below 150 ◦ C

10
K. Lee et al. Journal of Cleaner Production 434 (2024) 140333

Fig. 8. H2-SCR performance of 0.5Pt/TiO2 catalyst at different reactant concentrations: (a)–(c) effect of H2 reductant in feed stream (reaction condition: 500 ppm
NO, 5 vol% O2, 1–2 vol% H2, 5 vol% H2O, and GHSV = 15,000 h− 1), (d)–(f) effect of coexisting O2 concentration (open symbol: N2O, closed symbol: N2 in (b) and
(e)). Reaction condition: 500 ppm NO, 5–15 vol% O2, 1 vol% H2, 5 vol% H2O, and GHSV = 15,000 h− 1).

with an increase in the H2 concentration. Doubling the H2 concentration formation. In fact, minimizing the H2 feed is necessary to avoid fuel
from 1 to 2 vol% significantly increased the catalytic activity from 9% to economy penalties in H2-ICE applications.
55% at the lowest reaction temperature (100 ◦ C). However, the highest As shown in Fig. 8d, f, and Fig. S8c, increasing the O2 concentration
H2 concentration negatively impacted the catalytic activity between 150 from 5 to 15 vol% enhances the NOx conversion and N2 share during the
and 200 ◦ C, resulting in a remarkable decrease in NOx conversion light-off period between 120 and 150 ◦ C. However, above 150 ◦ C, NOx
compared with 1 and 1.5 vol% H2. Because the H2–NOx and H2–O2 re­ conversion decreases with increasing O2 concentration. These results
actions are strongly exothermic (Equation (1)–(3)), hotspot formations indicate that a high O2 concentration promotes the participation of more
inside the monolithic cordierite substrate can facilitate auto- NO2 in the H2-SCR reaction at low reaction temperatures (Fig. S8d).
acceleration of the H2-SCR reaction (Eβer et al., 2022). The surface of Excess O2 can also facilitate the oxidation reaction between H2 and O2 to
the monolithic cordierite coated with the catalyst, which is active in the form H2O, thereby depleting the amount of reducing agent available for
H2-SCR reaction, has a higher temperature than the observed catalytic the H2-SCR reaction above 150 ◦ C (Liu et al., 2016; Zhang et al., 2024).
inlet temperature because increasing H2 feed concentration promotes This process is evidenced by the decrease in the reacted NO ratio and
the exothermic H2–O2 reaction. Therefore, it can be hypothesized that increase in the NO2 share with increasing reaction temperature and O2
the sudden decrease in NOx conversion at 2 vol% H2 is due to the heat concentration, which are responsible for the reduction in the N2 share at
generated by the H2–O2 reaction, which increases the surface tempera­ higher O2 concentrations (Fig. 8f). In addition, when the O2 concen­
ture of the monolithic cordierite substrate. Notably, the enhanced NOx tration increases from 5 to 15 vol%, N2O formation increases from 16 to
conversion is accompanied increased N2O formation (46 ppm at 2 vol% 28 ppm during the light-off period (Fig. 8e) and then remains relatively
H2 and 37 ppm at 1.5 vol% H2 vs. 33 ppm N2O at 1 vol% H2, Fig. 8b), constant up to 190 ◦ C. Thus, the N2 share is similar (~86%) at both 5 and
which explains the increase in the N2O share at inlet temperatures below 15 vol% O2. The effect of the O2 concentration on the H2-SCR perfor­
140 ◦ C at both 1.5 and 2 vol% H2 (Fig. 8c and Fig. S8a). At 2 vol% H2, the mance indicates that a certain amount of coexisting O2 is critical for
maximum N2 share at 140 ◦ C (83%) is 4%p lower than that at 1 vol% H2 increasing both the NOx conversion and N2 yield, as it can widen the
(87%) owing to the increased formation of N2O at higher H2/NO ratios. active temperature window of the H2-SCR catalyst for deNOx.
These results suggest that an optimal amount of H2 reductant is
important for achieving high NOx conversion and suppressing N2O

11
K. Lee et al. Journal of Cleaner Production 434 (2024) 140333

Fig. 9. Effect of GHSV from 15,000 h− 1 to 60,000 h− 1 on the H2-SCR performance over 0.5Pt/TiO2 at a fixed flow rate of 2000 mL/min: reaction condition of (a)–(d)
= 500 ppm NO, 5 vol% O2, 1 vol% H2, and 5 vol% H2O, and reaction condition of (e)–(h) = 500 ppm NO, 5 vol% O2, 0.5 vol% H2, and 5 vol% H2O.

3.5. Influence of GHSV temperature corresponding to the maximum NOx conversion shifts from
170 ◦ C at a GHSV of 15,000 h− 1 to 230 ◦ C at a GHSV of 60,000 h− 1. This
Fig. 9 shows the effect of the GHSV on the H2-SCR performance of shift is also associated with a delay in the participation of NO2 in the H2-
0.5Pt/TiO2. The GHSV was varied from 15,000 to 60,000 h− 1 by varying SCR reaction during the light-off period (Fig. 9c and g). The maximum
the volume of washcoated monolithic cordierite from 8.0 to 2.0 cm3 NOx conversion significantly decreases to 75% at a GHSV of 30,000 h− 1
(corresponding to a length from 27.3 to 6.8 mm) at a fixed reactant gas because the amount of H2 available as a reductant decreases. The shift in
flow rate. At 1 vol% H2, increasing the GHSV shifts the active temper­ the active temperature window was attributed to a gradual decrease in
ature window to higher temperatures (Fig. 9a). The maximum NOx the residence time (τ) of the reactant gases inside monolithic cordierite
conversions are the same at GHSVs of 15,000 and 30,000 h− 1, but the as its length decreases from 27.3 mm (τ = 240 ms) to 13.7 mm (τ = 120
NOx conversion gradually decreases to 78% upon further increasing the ms), 9.1 mm (τ = 80 ms), and 6.8 mm (τ = 60 ms), which could hinder
GHSV to 60,000 h− 1. Furthermore, the reaction temperature at the light-off and NOx reduction.
maximum NOx conversion shifts from 160 ◦ C at a GHSV of 15,000 h− 1 to The characteristic features of N2O formation were independent of the
210 ◦ C at a GHSV of 60,000 h− 1. A similar phenomenon was observed deNOx temperature window. However, the reaction temperature cor­
upon increasing the GHSV at 0.5 vol% H2 (Fig. 9e). The reaction responding to maximum N2O formation shifts to higher temperatures

12
K. Lee et al. Journal of Cleaner Production 434 (2024) 140333

Fig. 10. H2-SCR performance of 0.5Pt/zeolite catalysts (SSZ-13, ZSM-5, and BETA): (a) NOx conversion, (b) N2 and N2O formation (open symbol: N2O, closed
symbol: N2), (c) NO2/NOx ratio, and (d)–(f) NO share in reaction route. Experimental condition: 500 ppm NO, 5 vol% O2, 0.5 vol% H2, 5 vol% H2O, and GHSV =
30,000 h− 1.

(Fig. 9b and f). Nonetheless, N2O formation decreases as the GHSV in­ (0.5Pt/BETA) > 85% at 150 ◦ C (0.5Pt/SSZ-13)) and then decreasing at
creases owing to reduction in the amount of reduced NO compared to higher reaction temperatures. Notably, the 0.5Pt/zeolite catalysts
that at a GHSV of 15,000 h− 1. As discussed above, the share of the NO exhibited higher NOx conversion than 0.5Pt/TiO2 below 170 ◦ C, sug­
reaction routes indicates that the onset temperature for complete NO2 gesting that zeolite supports can significantly promote H2-SCR activity
consumption and the reacted NO ratio are hindered during the light-off at low reaction temperatures. In particular, 0.5Pt/ZSM-5 exhibited
period with increasing GHSV. This hindrance results in a decrease in higher NOx conversion (52%) at 100 ◦ C than 0.5Pt/BETA (33%) and
NOx conversion and the N2 share, regardless of the feed concentration of 0.5Pt/SSZ-13 (14%), which was attributed to the presence of Pt0 species
H2 (Fig. 9d, h, and Fig. S9). Notably, the GHSV has a negligible impact on 0.5Pt/ZSM-5 (Fig. 5c). Metallic Pt could play an important role in
on the N2 share at the temperature corresponding to maximum NOx activating H2, thereby enhancing the H2-SCR activity at low reaction
conversion, with N2 shares of 87–91% at 1 vol% H2 and 80–87% at 0.5 temperatures (Hong et al., 2018; Kim et al., 2023).
vol% H2, comparable to those at a GHSV of 15,000 h− 1 (87% at 1 vol% In addition to NOx conversion, N2O and N2 formation over the 0.5Pt/
H2; 85% at 0.5 vol% H2). zeolite catalysts was notable different from that over 0.5Pt/TiO2. N2O
was the major by-product of the H2-SCR reaction, and its generation
affected the formation of N2 (Fig. 10b). The NO2/NOx ratio shows a
3.6. Influence of zeolite support on H2-SCR activity dependence on the order of NOx conversion during the light-off period
below 150 ◦ C (Fig. 10c), similar to the results observed for the Pt/TiO2
Previous studies revealed that both metal oxides and zeolites can be catalysts. The NO2 share provides evidence for NO2 involvement in the
used as supports for H2-SCR catalysts (Hong et al., 2019; Wang et al., H2-SCR reaction (Fig. 10d–f). During the light-off period, more NO2
2019a; Zhang et al., 2024). Fig. 10a shows the catalytic activities of the participates in the H2-SCR reaction, resulting in a lower NO2 share for
series of Pt/zeolite catalysts with a fixed Pt loading of 0.5 wt% as a 0.5Pt/ZSM-5 than for 0.5Pt/SSZ-13 or 0.5Pt/BETA. Considering the
function of reaction temperature at 0.5 vol % H2 and a GHSV of 30,000 NOx conversion and N2O and N2 shares, 0.5Pt/ZSM-5 is a promising H2-
h− 1. Volcanic deNOx profiles are observed over all the 0.5Pt/zeolite SCR catalyst for reducing NOx emissions at low temperatures. However,
catalysts, with NOx conversion maximized in the temperature range of further improvements in the reaction efficiency can be achieved by
140–150 ◦ C (91% at 140 ◦ C (0.5Pt/ZSM-5) > 89% at 140 ◦ C

13
K. Lee et al. Journal of Cleaner Production 434 (2024) 140333

Fig. 11. Effect of cell density of washcoated monolithic cordierite substrate on the H2-SCR performance over 0.5Pt/ZSM-5 catalyst: (a) NOx conversion, (b) N2 and
N2O formation (open symbol: N2O, closed symbol: N2), (c) NO share reaction route for 600 cpsi substrate, and (d) NO share in reaction route for 900 cpsi substrate.
Reaction condition: 500 ppm NO, 5 vol% O2, 0.5 vol% H2, 5 vol% H2O, and GHSV = 30,000 h− 1.

cordierite on the H2-SCR activity has not been previously reported.

Table 5
Fig. 11 shows the effect of the monolithic cordierite substrate on the
Physical properties of monolithic cordierite substrate with different cell density
H2-SCR performance at 0.5 vol% H2 and a GHSV of 30,000 h− 1 with 180
and wall thickness.
g/L of washcoated 0.5Pt/ZSM-5. Increasing the density from 400 to 900
Specifications 400/6 600/4.3 900/2.5 cpsi has a positive effect on NOx conversion (Fig. 11a). In particular,
Cross-section image NOx conversion below 140 ◦ C is enhanced, with a significant improve­
ment in NOx conversion at 100 ◦ C from 52% (400 cpsi) to 90% (600
cpsi) and 92% (900 cpsi). Additionally, NOx conversion exceeds 90% up
to 140 ◦ C at cell densities of both 600 and 900 cpsi. This positive effect is
related to the increase in cell density, which determines the number of
Cell density [cpsi] 400 600 900
Bulk density [g/L] 395 324 271
channels, open frontal area (OFA), and geometric surface area (GSA) of
Wall thickness [mm] 0.15 0.11 0.06 the monolithic cordierite substrate. As summarized in Table 5, the
Heat capacity 200 ◦ C [J/ 166 193 161 apparent OFA increases gradually from 77.4% to 85.6% upon increasing
(K⋅L)] the cell density from 400 to 900 cpsi, with a concomitant increase in the
Hydraulic diameter [mm] 1.11 0.93 0.78
GSA from 27.7 to 43.7 cm2/cm3. Thus, a higher cell density leads to
Coefficient of thermal 4.58 6.07 3.27
expansion [ × 10− 7/◦ C] greater GSA and OFA availability, which provides more active sites for
Cell spacing [mm] 1.27 1.04 0.85 reactant molecules to participate in the H2-SCR reaction. Although
Open frontal area [%] 77.4 80.0 85.6 increasing the cell density improves NOx conversion at low reaction
Geometric surface area 27.7 34.5 43.7
temperatures, N2O production also increases from 32 ppm (400 cpsi) to
[cm2/cm3]
42 ppm (600 and 900 cpsi) at 100 ◦ C (Fig. 11b). However, both a lower
N2O share and higher N2 share are attained during the light-off period at
suppressing N2O formation and converting the reacted NO to N2. 600 and 900 cpsi because the amount of reacted NO increases (Fig. 11c
and d vs. Fig. 10d).
Compared with the reactant feed mixtures shown in Fig. 11 (500
3.7. Optimal cell density of monolithic cordierite ppm NO and 0.5 vol% H2), the NOx conversion of washcoated cordierite
with cell density of 400 cpsi was highly enhanced at 250 ppm NO and
As demonstrated by the effect of the amount of H2 reductant in the 0.25 vol% H2, reaching a level similar to those with 600 and 900 cpsi
feed, increasing the H2 concentration enhanced the catalytic activity and (Fig. 12a). Although the [NO]:[H2] feed ratio is identical, a NOx con­
widened the active temperature window. However, this approach could version of approximately 85% is achieved at 100–160 ◦ C over the
lead to increased fuel consumption in the H2-ICEs because it requires the washcoated cordierite with 400 cpsi. The reduced NO concentration in
introduction of additional H2 fuel upstream of the aftertreatment system the feed stream is responsible for the high NOx conversion at a GHSV of
to compensate for insufficient H2 as a reductant in the exhaust gas. The 30,000 h− 1. NO in the feed reacted with 0.25 vol% H2 and could be
catalytic performance can be influenced by the cell density (Heck et al., sufficiently reduced on the active sites even at this high GHSV.
2001; Hosseini et al., 2020). At 0.5 vol% H2, the H2-SCR performance Furthermore, the formation of N2O is significantly reduced at all cell
was evaluated over the 0.5Pt/ZSM-5 catalyst washcoated on monolithic densities. At 900 cpsi, the maximum N2O concentration is reduced by
cordierite with different cell densities and wall thicknesses. To the best half from 42 ppm (Fig. 11b) to 22 ppm (Fig. 12b). Because more NO
of our knowledge, the effect of the physical properties of monolithic

14
K. Lee et al. Journal of Cleaner Production 434 (2024) 140333

Fig. 12. Effect of cell density of washcoated monolithic cordierite substrate on

the H2-SCR performance over 0.5Pt/ZSM-5: (a) NOx conversion and (b) N2O
formation. Reaction condition: 250 ppm NO, 5 vol% O2, 0.25 vol% H2, 5 vol%
H2O, and GHSV = 30,000 h− 1.

reacts and little N2O is formed, the N2 share is relatively high (79–86%)
up to 160 ◦ C over all the washcoated monoliths (Figs. S10b–d). How­
ever, an increase in the unreacted NO2 content (Fig. S10a), which was
oxidized by NO with O2, decreases the N2 share at high reaction tem­
peratures compared with the results in Fig. 11.

3.8. Influence of mixed supports on H2-SCR activity

The deNOx performance of H2-SCR catalysts was highly dependent

on the catalyst support, type and loading amount of active metal, and
cell density of monolithic cordierite. In particular, the TiO2 and ZSM-5
Fig. 13. Effect of mixed supports with different ZSM-5/TiO2 ratios on the H2-
catalyst supports significantly influenced the active temperature win­ SCR performance over 0.5Pt/yZzT catalysts washcoated on 600 cpsi substrate:
dow for deNOx. The development of catalysts that can simultaneously (a) NOx conversion, (b) N2O formation, and (c) NO share in reaction route.
reduce NOx emissions and N2O formation is challenging, because these Reaction condition: 500 ppm NO, 5 vol% O2, 0.5 vol% H2, 5 vol% H2O, and
reactions compete for the same catalytic active sites. Therefore, the ef­ GHSV = 30,000 h− 1.
fect of mixed supports of ZSM-5 and TiO2 on the H2-SCR performance
was investigated by varying the ratios of these supports. The catalysts conversion over the entire reaction temperature range is highest for
were washcoated onto 600 cpsi monolithic cordierite with a coating 0.5Pt/50Z50T (55.8%), followed by 0.5Pt/75Z25T (52.4%), 0.5Pt/
amount of 180 g/L. NOx conversion on the 0.5Pt/yZzT catalysts was 90Z10T (50.5%), 0.5Pt/25Z75T (47.3%), and 0.5Pt/10Z90T (30.3%).
compared with that on 0.5Pt/TiO2 and 0.5Pt/ZSM-5 (Fig. 13a). At an Although 0.5 Pt/50Z50 exhibits the highest NOx conversion, it also
ZSM-5/TiO2 ratio of 10:90, the NOx conversion profile is identical to produces more N2O during the H2-SCR process (Fig. 13b). Considering
that of 0.5Pt/TiO2 below 160 ◦ C but follows that of 0.5Pt/ZSM-5 at that the deNOx performance below 150 ◦ C is similar for all the 0.5Pt/
higher temperatures. Additionally, a maximum NOx conversion of 66% yZzT catalysts, except 10Z90T and 25Z75T, increasing the proportion of
is attained at 150 ◦ C, which is the lowest among all the 0.5Pt/yZzT ZSM-5 in the mixed support could suppress N2O formation at low re­
catalysts. However, increasing ZSM-5 ratio in the mixed support from 25 action temperatures below 150 ◦ C. Notably, 0.5Pt/75Z25T and 0.5Pt/
to 90 significantly improves the low-temperature NOx conversion from 90Z10T exhibit lower N2O formation and higher NOx conversion than
39% to 89% at 100 ◦ C, indicating that the mixed supports can alter the 0.5Pt/ZSM-5. As illustrated in Figs. 13c and 0.5Pt/90Z10T produces the
deNOx characteristics of the 0.5Pt/yZzT catalysts. The average NOx

15
K. Lee et al. Journal of Cleaner Production 434 (2024) 140333

lowest N2O share (8–17%) in reacted NO within the temperature win­ Data availability
dow of maximum NOx conversion, resulting in an increase in the N2
share (83–88%) compared with those of the other 0.5Pt/yZzT catalysts Data will be made available on request.
(Fig. S11). Thus, 0.5Pt/90Z10T is a promising H2-SCR catalyst,
achieving reasonable deNOx performance and producing a low N2O Acknowledgment
share with an enhanced N2 share during the H2-SCR process.
This research was supported by the Technology Innovation Program
4. Conclusions (Development of Direct Injection Hydrogen Engine Source Technology
based on Carbon-Free Hydrogen Fuel, grant number 20018473) funded
H2-SCR is a promising technique for reducing NOx emissions in lean- by the Ministry of Trade, Industry and Energy (MOTIE, Korea) and by
burn H2-ICEs owing to its excellent low-temperature catalytic perfor­ the “Development of performance optimization of combined regenera­
mance. Herein, the H2-SCR of NOx over noble-metal-based catalysts was tion PM-NOx system and DPF damage on board diagnosis technology for
investigated by varying the type and loading amount of active metal as heavy duty vehicle (2021003390004)” project of the Ministry of Envi­
well as the catalyst support, focusing enhancing the low-temperature ronment’s Technology Development Project linked to air environment
deNOx performance with satisfactory N2 transformation of reacted management commercialization.
NO. All the catalysts were coated on a monolithic ceramic substrate.
Nanoscale particles of metallic Pt and Pd were dispersed on the supports Appendix A. Supplementary data
without severe agglomeration, and no significant expansion, contrac­
tion, or collapse of the lattice in the catalyst supports occurred after Supplementary data to this article can be found online at https://doi.
impregnation with the metallic species. The xPt/TiO2 catalysts exhibited org/10.1016/j.jclepro.2023.140333.
greater than 90% NOx conversion at 135–180 ◦ C, whereas the xPd/TiO2
catalysts showed low H2-SCR activity at all reaction temperatures. Using References
0.5Pt/TiO2 as a reference catalyst, increasing H2 in the feed stream
Alghamdi, N.M., Cano, J.R., Anjum, D.H., Im, H.G., Kalamaras, C., Gascon, J., Sarathy, S.
significantly enhanced the catalytic activity from 9% to 55% at 100 ◦ C. M., 2020. Hydrogen Selective Catalytic Reduction of Nitrogen Oxide on Pt- and Pd-
However, the enhancement in NOx conversion was accompanied by Based Catalysts for Lean-Burn Automobile Applications. https://doi.org/10.4271/
increased N2O formation. Increasing the O2 concentration increased 2020-01-2173. SAE Technical Paper 2020-01-2173.
Beaumont, S.K., Alayoglu, S., Specht, C., Kruse, N., Somorjai, G.A., 2014. A nanoscale
NOx conversion and the N2 share during the light-off period between
demonstration of hydrogen atom spillover and surface diffusion across silica using
120 and 150 ◦ C. However, above 150 ◦ C, the NOx conversion decreased the kinetics of CO2 methanation catalyzed on spatially separate Pt and Co
sharply with increasing O2 concentration. Investigation of the effect of nanoparticles. Nano Lett. 14 (8), 4792–4796. https://doi.org/10.1021/nl501969k.
Börnhorst, M., Deutschmann, O., 2021. Advances and challenges of ammonia delivery by
the GHSV revealed that the deNOx performance was greatly affected by
urea-water sprays in SCR systems. Prog. Energy Combust. Sci. 87, 100949 https://
the amount of coexisting H2 in the feed. NOx conversion over the 0.5Pt/ doi.org/10.1016/j.pecs.2021.100949.
zeolite catalysts was maximized in the temperature range of 140–150 ◦ C, Cao, L., Wang, Q., Yang, J., 2020. Ultrafine Pt particles directed by polyvinylpyrrolidone
decreasing in the following order: 91% (0.5Pt/ZSM-5) > 89% (0.5Pt/ on zeolite beta as an efficient low-temperature H2-SCR catalyst. J. Environ. Chem.
Eng. 8 (1), 103631 https://doi.org/10.1016/j.jece.2019.103631.
BETA) > 85% at (0.5Pt/SSZ-13). Increasing the cell density from 400 to Dong, X., Wang, B., Yip, H.L., Chan, Q.N., 2019. CO2 emission of electric and gasoline
900 cpsi for 0.5Pt/ZSM-5 positively influenced NOx conversion but also vehicles under various road conditions for China, Japan, Europe and world
promoted N2O production. To further the development of catalysts that average–prediction through year 2040. Appl. Sci. 9, 2295. https://doi.org/10.3390/
app9112295.
can simultaneously reduce NOx emissions and N2O formation, the effect Duan, K., Chen, B., Zhu, T., Liu, Z., 2015. Mn promoted Pd/TiO2–Al2O3 catalyst for the
of mixed supports of ZSM-5 and TiO2 on the H2-SCR performance was selective catalytic reduction of NOx by H2. Appl. Catal. B Environ. 176–177,
investigated. Increasing the ZSM-5 ratio significantly improved the low- 618–626. https://doi.org/10.1016/j.apcatb.2015.04.048.
Duan, K., Wang, Z., Hardacre, C., Liu, Z., Chansai, S., Stere, C., 2019. Promoting effect of
temperature NOx conversion, indicating that the mixed supports can Au on Pd/TiO2 catalyst for the selective catalytic reduction of NOx by H2. Catal.
alter the deNOx characteristics of the 0.5Pt/yZzT catalysts. The average Today 332, 69–75. https://doi.org/10.1016/j.cattod.2018.06.022.
NOx conversion over the entire reaction temperature range was highest Duan, L., Tan, P., Liu, J., Liu, Y., Chen, Y., Lou, D., Hu, Z., 2022. Emission characteristics
of a diesel engine with an electrically heated catalyst under cold start conditions.
for 0.5Pt/50Z50T (55.8%), followed by 0.5Pt/75Z25T (52.4%), 0.5Pt/
J. Clean. Prod. 162, 134965 https://doi.org/10.1016/j.jclepro.2022.134965.
90Z10T (50.5%), 0.5Pt/25Z75T (47.3%), and 0.5Pt/10Z90T (30.3%), European Parliament, 2022. EU Ban on the Sale of New Petrol and Diesel Cars from 2035
including that increasing the proportion of ZSM-5 can suppress N2O Explained. https://www.europarl.europa.eu/news/en/headlines/economy/202210
19STO44572/eu-ban-on-sale-of-new-petrol-and-diesel-cars-from-2035-explained.
formation. Compared with the other 0.5Pt/yZzT catalysts, 0.5Pt/
(Accessed 19 October 2023).
90Z10T demonstrated the lowest N2O share in the reacted NO within the Eßer, E., Müller, K., Kureti, S., 2022. Low-temperature H2-deNOx in diesel exhaust. Top.
temperature window of maximum NOx conversion, resulting in an in­ Catal. 66, 1020–1030. https://doi.org/10.1007/s11244-022-01719-x.
crease in the N2 share. Fang, S., Takagaki, A., Watanabe, M., Ishihara, T., 2020. The direct decomposition of NO
into N2 and O2 over copper doped Ba3Y4O9. Catal. Sci. Technol. 10, 2513–2522.
https://doi.org/10.1039/D0CY00194E.
CRediT authorship contribution statement Frost, R., 2023. EU 2035 petrol and diesel car ban: Germany reaches deal on synthetic
fuels. Euronews.green. https://www.euronews.com/green/2023/03/22/eu-to-ban-
petrol-and-diesel-cars-by-2035-heres-why-some-countries-are-pushing-back.
Kyungseok Lee: Conceptualization, Formal analysis, Investigation, (Accessed 19 October 2023).
Methodology, Validation, Visualization, Writing – original draft. Gholami, F., Tomas, M., Gholami, Z., Vakili, M., 2020. Technologies for the nitrogen
oxides reduction from flue gas: a review. Sci. Total Environ. 714, 136712 https://
Kyoungbok Lee: Formal analysis, Methodology, Validation. Byung­
doi.org/10.1016/j.scitotenv.2020.136712.
chul Choi: Formal analysis, Methodology, Validation. Kwangchul Oh: Gioria, E., Marchesini, F.A., Soldati, A., Giorello, A., Hueso, J.L., Gutierrez, L., 2019.
Conceptualization, Formal analysis, Funding acquisition, Investigation, Green Synthesis of a Cu/SiO2 Catalyst for Efficient H2-SCR of NO. Appl. Sci. 9 (19),
Methodology, Project administration, Supervision, Writing – review & 4075. https://doi.org/10.3390/app9194075.
Granger, P., Dhainaut, F., Pietrzik, S., Malfoy, P., Mamede, A.S., Leclercq, L.,
editing. Leclercq, G., 2006. An overview: comparative kinetic behaviour of Pt, Rh and Pd in
the NO + CO and NO + H2 reactions. Top. Catal. 39, 65–76. https://doi.org/
10.1007/s11244-006-0039-0.
Declaration of competing interest Guan, Y., Liu, Y., Lv, Q., Wang, B., Che, D., 2021. Review on the selective catalytic
reduction of NOx with H2 by using novel catalysts. J. Environ. Chem. Eng. 9 (6),
106770 https://doi.org/10.1016/j.jece.2021.106770.
The authors declare that they have no known competing financial Hajjari, M., Tabatabaei, M., Aghbashlo, M., Ghanavati, H., 2017. A review on the
interests or personal relationships that could have appeared to influence prospects of sustainable biodiesel production: a global scenario with an emphasis on
the work reported in this paper.

16
K. Lee et al. Journal of Cleaner Production 434 (2024) 140333

waste-oil biodiesel utilization. Renew. Sustain. Energy Rev. 72, 445–464. https:// Park, S.M., Jang, H., Kim, E.S., Han, H., Seo, G., 2012. Incorporation of zirconia onto
doi.org/10.1016/j.rser.2017.01.034. silica for improved Pt/SiO2 catalysts for the selective reduction of NO by H2. Appl.
Heck, R.M., Gulati, S., Farrauto, R.J., 2001. The application of monoliths for gas phase Catal. A-Gen. 427–428, 155–164. https://doi.org/10.1016/j.apcata.2012.03.045.
catalytic reactions. Chem. Eng. J. 82 (1–3), 149–156. https://doi.org/10.1016/ Park, S.M., Kim, M., Kim, E.S., Han, H., Seo, G., 2011. H2-SCR of NO on Pt–MnOx
S1385-8947(00)00365-X. catalysts: Reaction path via NH3 formation. Appl. Catal. A-Gen. 395 (1–2), 120–128.
Hong, Z., Wang, Z., Chen, D., Sun, Q., Li, X., 2018. Hollow ZSM-5 encapsulated Pt https://doi.org/10.1016/j.apcata.2011.01.033.
nanoparticles for selective catalytic reduction of NO by hydrogen. Appl. Surf. Sci. Patel, V.K., Sharma, S., 2021. Effect of oxide supports on palladium based catalysts for
440, 1037–1046. https://doi.org/10.1016/j.apsusc.2018.01.239. NO reduction by H2-. SCR. Catal. Today 375, 591–600. https://doi.org/10.1016/j.
Hong, Z., Sun, X., Wang, Z., Zhao, G., Li, X., Zhu, Z., 2019. Pt/SSZ-13 as an efficient cattod.2020.04.006.
catalyst for the selective catalytic reduction of NOx with H2. Catal. Sci. Technol. 9, Peng, Z., Li, Z., Liu, Y., Yan, S., Tong, J., Wnag, D., Ye, Y., Li, S., 2017. Supported Pd
3994–4001. https://doi.org/10.1039/C9CY00898E. nanoclusters with enhanced hydrogen spillover for NOx removal via H2-SCR: the
Hosseini, S., Moghaddas, H., Masoudi Soltani, S., Kheawhom, S., 2020. Technological elimination of “volcano-type” behaviour. Chem. Commun. 53, 5958–5961. https://
applications of honeycomb monoliths in environmental processes: a review. Process doi.org/10.1039/C7CC02235B.
Saf. Environ. Protect. 133, 286–300. https://doi.org/10.1016/j.psep.2019.11.020. Price, L.A., Ridley, C.J., Bull, C.L., Wells, S.A., Sartbaeva, A., 2021. Determining the
Hosseini, S.H., Tsolakis, A., Alagumalai, A., Mahian, O., Lam, S.S., Pan, J., Peng, W., structure of zeolite frameworks at high pressures. CrystEngComm 23, 5615–5623.
Tabatabaei, M., Aghbashlo, M., 2023. Use of hydrogen in dual-fuel diesel engines. https://doi.org/10.1039/D1CE00142F.
Prog. Energy Combust. Sci. 98, 101100 https://doi.org/10.1016/j. Reitz, R.D., Ogawa, H., Payri, R., Fansler, T., Kokjohn, S., Moriyoshi, Y., Agarwal, A.,
pecs.2023.101100. Arcoumanis, D., Assanis, D., Bae, C., Boulouchos, K., Canakci, M., Curran, S.,
Hu, Z., Yang, R.T., 2019. 110th anniversary: recent progress and future challenges in Denbratt, I., Gavaises, M., Guenthner, M., Hasse, C., Huang, Z., Ishiyama, T.,
selective catalytic reduction of NO by H2 in the presence of O2. Ind. Eng. Chem. Res. Johansson, B., Johnson, T., Kalghatgi, G., Koike, M., Kong, S., Leipertz, A., Miles, P.,
58 (24), 10140–10153. https://doi.org/10.1021/acs.iecr.9b01843. Novella, R., Onorati, A., Richter, M., Shuai, S., Siebers, D., Su, W., Trujillo, M.,
Hu, Z., Yong, X., Li, D., Yang, R.T., 2020. Synergism between palladium and nickel on Uchida, N., Vaglieco, B.M., Wagner, R., Zhao, H., 2020. IJER editorial: the future of
Pd-Ni/TiO2 for H2-SCR: a transient DRIFTS study. J. Catal. 381, 204–214. https:// the internal combustion engine. Int. J. Engine Res. 21 (1), 3–10. https://doi.org/
doi.org/10.1016/j.jcat.2019.11.006. 10.1177/146808741987799.
Hu, Z., Zhang, T., Li, D., Yang, R.T., 2021. Understanding the promotional effect of 3d Rouleau, L., Duffour, F., Walter, B., Kumar, R., Nowak, L., 2021. Experimental and
transition metals (Fe, Co, Cu) on Pd/TiO2 for H2-SCR. Catal. Sci. Technol. 11, numerical investigation on hydrogen internal combustion engine. In: SAE Technical
886–894. https://doi.org/10.1039/D0CY01840F. Paper 2021-24-0060. https://doi.org/10.4271/2021-24-0060.
Huang, Y., Yu, Z., Guo, M., Liu, H., Liu, X., Han, J., Cui, S., Liu, B., Zhao, Y., Wei, J., Satsuma, A., Hashimoto, M., Shibata, J., Yoshida, H., Hattori, T., 2003. Nitrous oxide free
Liu, B., Chen, S., 2023. Coated monolithic catalysts for better selective catalytic pathway for selective reduction of NO by hydrogen over supported Pt catalysts.
reduction: concerns about structural integrity, catalytic activity and anti-poisoning Chem. Commun. 14, 1698–1699. https://doi.org/10.1039/B304806N.
performance. Catal. Commun. 178, 106667 https://doi.org/10.1016/j. Savva, Z., Petallidou, K.C., Damaskinos, C.M., Olympiou, G.G., Stathopoulos, V.N.,
catcom.2023.106667. Efstathiou, A.M., 2021. H2-SCR of NOx on low-SSA CeO2-supported Pd: the effect of
IEA, 2021. Key World Energy Statistics 2021. https://www.iea.org/reports/key-world-e Pd particle size. Appl. Catal. A-Gen. 615, 118062 https://doi.org/10.1016/j.
nergy-statistics-2021. (Accessed 19 October 2023). apcata.2021.118062.
Kim, G.J., Shin, J.H., Kim, S.B., Hong, S.C., 2023. The role of Pt valence state and La Schott, F.J.P., Balle, P., Adler, J., Kureti, S., 2009. Reduction of NOx by H2 on Pt/WO3/
doping on titanium supported Pt-La/TiO2 catalyst for selective catalytic reduction ZrO2 catalysts in oxygen-rich exhaust. Appl. Catal. B-Environ. 87 (1–2), 18–29.
with H2. Appl. Surf. Sci. 608, 155040 https://doi.org/10.1016/j. https://doi.org/10.1016/j.apcatb.2008.08.021.
apsusc.2022.155040. Sharma, S., Ghoshal, S.K., 2015. Hydrogen the future transportation fuel: from
Koch, D.T., Eßer, E., Kureti, S., Sousa, A., 2020. H2-deNOx catalyst for H2 combustion production to applications. Renew. Sustain. Energy Rev. 43, 1151–1158. https://doi.
engines. MTZ worldwide 81, 30–35. https://doi.org/10.1007/s38313-020-0229-3. org/10.1016/j.rser.2014.11.093.
Lee, K., Choi, B., 2021. HC-SCR system combining Ag/Al2O3 and Pd/Al2O3 catalysts with Shelef, M., Jones, J.H., Kummer, J.T., Otto, K., Weaver, E.E., 1971. Selective catalytic
resistance to hydrothermal aging for simultaneous removal of NO, HC, and CO. reaction of hydrogen with nitric oxide in the presence of oxygen. Environ. Sci.
J. Ind. Eng. Chem. 102, 51–68. https://doi.org/10.1016/j.jiec.2021.06.030. Technol. 5 (9), 790–798. https://doi.org/10.1021/es60056a003.
Lee, K., Kosaka, H., Sato, S., Yokoi, T., Choi, B., 2019. Effect of Cu content and zeolite Sterlepper, S., Fischer, M., Claßen, J., Huth, V., Pischinger, S., 2021. Concepts for
framework of n-C4H10-SCR catalysts on de-NOx performances. Chem. Eng. Sci. 203, hydrogen internal combustion engines and their implications on the exhaust gas
28–42. https://doi.org/10.1016/j.ces.2019.03.028. aftertreatment system. Energies 14 (23), 8166. https://doi.org/10.3390/
Lee, K., Choi, B., Kim, C., Lee, C., Oh, K., 2021. De-NOx characteristics of HC-SCR system en14238166.
employing combined Ag/Al2O3 and CuSn/ZSM-5 catalyst. J. Ind. Eng. Chem. 93, Wang, X., Wang, X., Yu, H., Wang, X., 2019a. The functions of Pt located at different
461–475. https://doi.org/10.1016/j.jiec.2020.10.026. positions of HZSM-5 in H2-. SCR. Chem. Eng. J. 355, 470–477. https://doi.org/
Li, X., Zhang, X., Xu, Y., Liu, Y., Wang, X., 2015. Influence of support properties on H2 10.1016/j.cej.2018.07.207.
selective catalytic reduction activities and N2 selectivities of Pt catalysts. Chin. J. Wang, Z., Zhao, K., Xiao, B., Gao, P., He, D., Cai, T., Yuan, J., 2019b. Fabrication of
Catal. 36 (2), 197–203. https://doi.org/10.1016/S1872-2067(14)60197-2. monolithic catalysts: comparison of the traditional and the novel green methods.
Liu, Z., Lu, Y., Yuan, L., Ma, L., Zheng, L., Zhang, J., Hu, T., 2016. Selective catalytic Catalysts 9 (12), 981. https://doi.org/10.3390/catal9120981.
reduction of NOx with H2 over WO3 promoted Pt/TiO2 catalyst. Appl. Catal. B Zhang, Y., Zeng, H., Jia, B., Liu, Z., 2021. Selective catalytic reduction of NOx by H2 over
Environ. 188, 189–197. https://doi.org/10.1016/j.apcatb.2016.02.008. a novel Pd/FeTi catalyst. Catal. Today 360, 213–219. https://doi.org/10.1016/j.
Liu, Z., Wu, J., Hardacre, C., 2018. Research progress in the selective catalytic reduction cattod.2020.05.042.
of NOx by H2 in the presence of O2. Catal. Surv. Asia 22, 146–155. https://doi.org/ Zhang, Y., Zeng, H., Jia, B., Wang, Z., Liu, Z., 2019. Selective catalytic reduction of NOx
10.1007/s10563-018-9248-3. by H2 over Pd/TiO2 catalyst. Chin. J. Catal. 40 (6), 849–855. https://doi.org/
Liu, Y., Tursun, M., Yu, H., Wang, X., 2019. Surface property and activity of Pt/Nb2O5- 10.1016/S1872-2067(19)63297-3.
ZrO2 for selective catalytic reduction of NO by H2. Mol. Catal. 464, 22–28. https:// Zhang, C., Gao, Y., Yan, Q., Wang, Q., 2020. Fundamental investigation on layered
doi.org/10.1016/j.mcat.2018.12.015. double hydroxides derived mixed metal oxides for selective catalytic reduction of
Liu, L., Xu, K., Su, S., He, L., Qing, M., Chi, H., Liu, T., Hu, S., Wang, Y., Xiang, J., 2020a. NOx by H2. Catal. Today Off. 355, 450–457. https://doi.org/10.1016/j.
Efficient Sm modified Mn/TiO2 catalysts for selective catalytic reduction of NO with cattod.2019.07.006.
NH3 at low temperature. Appl. Catal. A-Gen. 592, 117413 https://doi.org/10.1016/ Xu, C., Sun, W., Cao, L., Li, T., Cai, X., Yang, J., 2017. Highly efficient Pd-doped
j.apcata.2020.117413. aluminate spinel catalysts with different divalent cations for the selective catalytic
Liu, Z., Jia, B., Zhang, Y., Haneda, M., 2020b. Engineering the metal–support interaction reduction of NO with H2 at low temperature. Chem. Eng. J. 308, 980–987. https://
on Pt/TiO2 catalyst to boost the H2-SCR of NOx. Ind. Eng. Chem. Res. 59 (31), doi.org/10.1016/j.cej.2016.09.119.
13916–13922. https://doi.org/10.1021/acs.iecr.0c01792. Zhang, Y., Chen, J., Liu, Z., 2022. Selective catalytic reduction of NOx by hydrogen over
Messou, D., Bernardin, V., Meunier, F., Ordoño, M.B., Urakawa, A., Machado, B.F., PtIr/TiO2 catalyst. Catal. Today 402, 115–121. https://doi.org/10.1016/j.
Collière, V., Philippe, R., Serp, P., Le Berre, C., 2021. Origin of the synergistic effect cattod.2022.03.025.
between TiO2 crystalline phases in the Ni/TiO2-catalyzed CO2 methanation reaction. Zhang, L., Shan, Y., Yan, Z., Liu, Z., Yu, Y., He, H., 2024. Efficient Pt/KFI zeolite catalysts
J. Catal. 398, 14–28. https://doi.org/10.1016/j.jcat.2021.04.004. for the selective catalytic reduction of NOx by hydrogen. J. Environ. Sci. 138,
Moon, S., Park, D.C., Lee, E., You, Y., Heo, I., Kim, Y.J., Kim, D.H., 2023. Excellent 102–111. https://doi.org/10.1016/j.jes.2023.03.004.
activity and selectivity of Pd/ZSM-5 catalyst in the selective catalytic reduction of Zhang, Y., Xu, S., Li, J., He, E., Liu, Z., 2023a. Unraveling the promotional effect of Co on
NOx by H2. Environ. Res. 227, 115707 https://doi.org/10.1016/j. the Pd/TiO2 catalyst for H2-SCR of NOx in the presence of oxygen. J. Phys. Chem. C
envres.2023.115707. 127 (15), 7248–7256. https://doi.org/10.1021/acs.jpcc.3c01239.
Onorati, A., Payri, R., Vaglieco, B., Agarwal, A., Bae, C., Bruneaux, G., Canakci, M., Zhang, Y., Xu, S., Wang, Y., Liu, Z., 2023b. Selective catalytic reduction of NOx by H2
Gavaises, M., Günthner, M., Hasse, C., Kokjohn, S., Kong, S.-C., Moriyoshi, Y., over Na modified Pd/TiO2 catalyst. Int. J. Hydrogen Energy 48 (43), 16267–16278.
Novella, R., Pesyridis, A., Reitz, R., Ryan, T., Wagner, R., Zhao, H., 2022. The role of https://doi.org/10.1016/j.ijhydene.2023.01.099.
hydrogen for future internal combustion engines. Int. J. Engine Res. 23 (4), 529–540. Zhao, X., Zhang, X., Xu, Y., Liu, Y., Wang, X., Yu, Q., 2015. The effect of H2O on the H2-
https://doi.org/10.1177/14680874221081947. SCR of NOx over Pt/HZSM-5. J. Mol. Catal. A-Chem. 400, 147–153. https://doi.org/
10.1016/j.molcata.2015.02.013.

17