Evolution of Pd Species for the Conversion of Methane under

Subscriber access provided by UNIV OF LOUISIANA

Kinetics, Catalysis, and Reaction Engineering

Evolution of Pd Species for the Conversion of Methane under Operation Conditions Jianjun Chen, Yang Wu, Wei Hu, Pengfei Qu, Guochen Zhang, yi jiao, Lin Zhong, and Yao-Qiang Chen Ind. Eng. Chem. Res., Just Accepted Manuscript • DOI: 10.1021/acs.iecr.8b06226 • Publication Date (Web): 29 Mar 2019 Downloaded from http://pubs.acs.org on March 29, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

Evolution of Pd Species for the Conversion of Methane under Operation Conditions

Jianjun Chen,a Yang Wu,a Wei Hu,b, c Pengfei Qu,b Guochen Zhang,b Yi Jiao, a Lin Zhong b *, and Yaoqiang Chen a, b, d *

a

Institute of New Energy and Low-carbon Technology, Sichuan University,

Chengdu 610064, China b

College of Chemical Engineering, Sichuan University, Chengdu 610064,

China c

Institute of Atmospheric Environment, Chongqing Academy of Environmental

Science, Chongqing 401147, China d

College of Chemistry, Sichuan University, Chengdu 610064, China

Corresponding author: Lin Zhong (Email: [email protected]; Tel/Fax: +86-28-85418451) Corresponding author: Yaoqiang Chen (Email: [email protected]; Tel/Fax: +86-28-85418451)

1

ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Abstract: A Pd model catalyst was tested under dynamically operation conditions to investigate the active site for the conversion of CH4 from natural gas vehicles (NGVs) exhausts. It was demonstrated that in situ pretreatment in reaction gas significantly improved CH4 removal activity. XPS results revealed that metallic Pd phase was much more active than PdO for CH4 conversion under stoichiometric conditions. Interestingly, under rich conditions (lambda=0.98), the formation of inactive PdCx led to considerable deactivation of catalyst. Increasing O2 concentration can efficiently convert it into active Pd0 and thereby the activity was recovered to original level. By contrast, under stoichiometric conditions (lambda=1.00), CH4 conversion was stable, probably owing to the maintaining of active Pd0 phase. However, further increasing O2 amount (lambda=1.02) resulted in reduced activity, which may be due to the formation of PdO from oxidation of metallic Pd. These results implied that Pd species under dynamically working conditions was evolvable.

Keywords: Pd chemical state; CH4 conversion; Active site; Metallic Pd; Dynamic operation conditions; Evolution

2


Page 2 of 47

Page 3 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1. Introduction Nowadays, natural gas driven vehicles (NGVs) have been widely used all around the world

1, 2.

However, the unburned CH4 in NGVs exhausts is a

potent greenhouse gas, and this pollutant must be efficiently removed over supported catalysts to meet the increasingly stringent emission regulations 3-5. The after-treatment catalyst for NGVs exhaust purification varies with the working modes of the engine, such as lean burn and stoichiometric ones, operating respectively under excess of oxygen and under a stoichiometric mixture of fuel and oxygen. It is widely accepted that supported Pd-based catalysts are highly effective for CH4 oxidation

6-8.

The changes in CH4

conversion with lambda over Pd catalyst can be associated with the corresponding changes in the states of Pd and Ce, the interaction of metal and support, the size and morphology of the catalyst, and with water content in the reaction mixture

7, 9.

Particularly, water is known to be the largest

inhibitory effect in Pd-catalyzed methane combustion under lean condition (especially at temperature below ca. 420 °C) mainly owing to the formation of inactive Pd(OH)2 by reacting with PdO 10. However, the inhibitory effect seems to be insignificant under stoichiometric and rich conditions

11.

Normally, the

state of Pd on the Pd catalyst is considered to be crucial for its catalytic performance for the elimination of CH4 from NGV exhaust gas 12, 13. Nevertheless, it is still a debate about which state of Pd species responsible for the active site of CH4 conversion under lean burn conditions or 3



stoichiometric ones

7, 14, 15.

Page 4 of 47

This uncertainty may be mainly due to that Pd

chemical state exists in various forms under the operation conditions involving the variations of exhaust gas temperature and reaction gas compositions. Under lean conditions, the fluctuation of temperature around 800 °C at which PdO decomposes usually results in the change of the oxidation and coordination state of Pd species

8, 16.

S. K. Matam et al.

17

shown the change

of Pd oxidation state over Pd catalyst supported on alumina for CH4 oxidation in the excess of O2 using X-ray absorption spectroscopy (XAS). It can be clearly found that during the reactions Pd was in an entirely oxidized state below 677 °C, and that the PdO species initially decomposed to metallic Pd above 719 °C. Finally, the Pd species mainly existed as metallic Pd above 856 °C. A lot of works 18, 19 further reported that PdO phase can only be stable in air up to about 800 °C and that the stable species was metallic Pd above this temperature. Nevertheless, the exhaust temperature of lean burn NGVs engines usually does not exceed 600 ° C

20.

Below this temperature, PdO

species is more thermodynamically stable than metallic Pd ones, and PdO acts as the active site for CH4 oxidation over Pd-based catalyst

8, 21.

Interestingly, during the test of CH4 oxidation activity below 600 °C in excess of O2, A. K. Datye et al. 22 found that the light-off temperature of CH4 oxidation over Pd/Al2O3 reduced in hydrogen was lower as much as 50 ° C than that calcined in air. The results inferred that the reduced Pd/Al2O3 was more active than the calcined one for CH4 oxidation in excess of O2. However, PdO 4


Page 5 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


species was progressively formed over the reduced sample during cooling in excess of O2 (5 vol.%). Consequently, the reduced Pd species was re-oxidized, and the catalytic performance of CH4 oxidation declined significantly to a stable level. This work indicated that the effects of the O2 concentration on the chemical state of Pd species must be taken into account if identifying the active sites of CH4 conversion

23.

Under operating conditions

for stoichiometric NGVs, the air-fuel ratio was always fluctuating during sudden acceleration and deceleration, accompanied by the variation of O2 concentration in the exhaust emission 24. Therefore, even if the temperature of exhaust gas from stoichiometric NGVs (especially from heavy duty ones) was mostly below the temperature of PdO decomposition, the chemical state of Pd may be constantly changing. In this case, the mechanism of CH4 conversion becomes more complex and consequently its study is remarkably challenging 25.

Recently, extensive studies have been focused on the mechanism of CH4 conversion and the Pd chemical state over Pd-based catalysts under the conditions of dynamic air-fuel ratio

12.

Johan Nilsson et al.

26

used operando

XANES to investigate the CH4 oxidation activity in various mixture of CH4 and O2 over Pd/Al2O3. The experiments were performed by switching the flow of oxygen on and off while the flow of methane was constant at 0.1% at 360 °C. The results clearly illustrated that the Pd chemical state was mostly in the 5



Page 6 of 47

form of metallic Pd after 0.15% and 0.25% oxygen being removed from the mixture gas. At the same time, higher activity for CH4 conversion can be observed. It suggested that metallic Pd species may contribute to the enhanced activity over the Pd catalyst operating close to stoichiometric air-fuel ratio. Under conditions similar to the working ones, D. Ferris et al.

27

investigated the effect of pretreatment on the activity of CH4 abatement over Pd/YFeO3 in a stoichiometric mixture gas of CH4-CO-NO-O2 (lambda=1). Before the reactions, the catalyst was pretreated at 900 °C for 2 h in static air (thermal ageing) and another one was pretreated at 900 °C for 2 h in above reaction gas (stoichiometric ageing). As confirmed by XANES and STEM, well-defined metallic Pd nanoparticles with 10–20 nm were observed after stoichiometric ageing while Pd species was mainly in the oxide state in the case of thermal aging. Accordingly, the activity of CH4 abatement over Pd/YFeO3 after stoichiometric aging was significantly higher than that after thermal ageing, implying that the metallic Pd nanoparticle could display activity for CH4 conversion under the given conditions. However, the pretreatments of catalysts in this work seemed to be different from those in real NGVs. Typically, the pretreatments in operating NGVs were normally performed under working NGVs exhaust gases, mainly composing of CH4, CO, NOx, O2, H2O, and CO2

28, 29.

The pretreatments on Pd-based catalyst

affected the performance of CH4 conversion significantly

30, 31.

Taking the

composition of exhaust gas and the pretreatment of catalyst into account, G. 6


Page 7 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Groppi et al.

23

further treated a commercial Pd-based catalyst in a simulated

NGVs exhaust gas at 600 °C at 5 h prior to the reactions and then tested the activity of CH4 removal under periodic lean/rich switch in the emission from NGVs. The improvement of CH4 conversion was observed after switching from lean to rich feed at small lambda oscillation amplitudes. The authors suggested that the enhancement during above oscillation conditions may resulted from the co-existence of a mixed Pd0/PdO state rather than a completely oxidized one. However, there were no further characterization studies yet to provide evidence about this opinion in the work. Our previous works

15

indicated that metallic Pd and PdO states were both active sites for

CH4 conversion under rich conditions, and that PdO species played a vital role for the reactions under lean conditions. However, this work didn’t address about which species was more active for CH4 removal under rich conditions for NGVs. Moreover, the active site for CH4 conversion under stoichiometric conditions is also not clear 23.

Our present work mainly focused on the investigation of the relationship between CH4 conversion variations and the corresponding changes in Pd state under operation conditions. A Pd model catalyst was pretreated in a simulated stoichiometric NGVs exhaust gases, and the effects of the variation of Pd state on the activity of CH4 conversion under stoichiometric conditions were studied using XPS technique. The impacts of the evolution of Pd species 7



on the activity were also investigated by periodic pulse and lambda oscillation operating experiments. It can be found that metallic Pd was much more active than PdO for CH4 conversion under stoichiometric conditions, and that the chemical state of Pd species changed dynamically with the fluctuation of the air-fuel ratio. Our work may be helpful for development of highly efficient Pd catalysts for the abatement of CH4.

2. Experimental 2.1 Catalyst Preparation A commercial γ-Al2O3 (LA-50, JINQIHUAGONG Co. Ltd) was precalcined in air at 900 °C for 8 h to result in a BET surface area of 143 m2/g. The alumina modified by CeO2 was prepared by incipient wetness using Ce(NH4)2(NO3)6 (Sigma Aldrich) as cerium precursor. The received powder was dried at 110 °C for 12 h and calcined at 550 °C for 3 h to yield catalyst support containing 3 wt.% CeO2. The Pd model catalyst was prepared by impregnating palladium(II) nitrate solution (Heraeus, 15 wt.%) on the obtained support powder. Similar as previously described, drying at 110 °C for 12 h and calcining at 550 °C for 3 h in flowing air were carried out (ramp: 4 °C/min). The loading of Pd was 1 wt.%.

The resulting Pd model catalyst powder was then balled milled with 7.00 mL deionized water/g powder for 3 h. Next, 0.03 mL 4.0 M HAc/g and 5.00 mL 8


Page 8 of 47

Page 9 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


deionized water/g were added with vigorous stirring. The concentration of the obtained slurry is about 0.30 g catalyst/ mL. The monoliths substrate (Corning, 400 cells per square inch, volume is 2.0 mL, length is 25 mm, diameter round is 11 mm and the wall thickness is 0.10 mm) was vertically immersed into the washcoat slurry for 2 min. The excess slurries were removed by a compressed air stream. Afterwards, the sample was dried at 150 °C for 30 min. After repeating this procedure, the desired Pd loading of about 1.7 g/L (50 g/ft3) was obtained. Finally, the monolithic samples were dried and calcined at 550 °C for 3 h in flowing air.

2.2 Catalyst activity test The monolithic catalysts were tested in a multiple fixed bed continuous flow reactor operated at atmospheric pressure. A K-type thermocouple was used to measure the inlet gas temperature of the catalyst. All the gases were regulated by a series of mass flow controllers. The total gas flow rate was 1680 mL/min, corresponding to a gas hourly space velocity (GHSV) = 50,400 h−1 (STP). The calculated lambda (λ) value was denoted as Eq.(1) 9. λ=(2[O2] + 2[CO2] + [CO] + [NO] + [H2O])/(2[CO2] + 2[CO] + 4[CH4] + [H2O]) (1) The standard exhaust gases from stoichiometric NGVs consisted of 1000 ppm CH4, 5000 ppm CO, 930 ppm NO, 4035 ppm O2 (λ=1), 10vol.% H2O, 10vol.% CO2 and balance N2. The testing procedures were descripted in Figure 1a. 9



Before the testing, the monolithic catalysts were pretreated at 500 °C in above reaction gas with a range of 0, 1, 3, 5, and 8 h, respectively. The pretreated catalysts were marked as Untreated catalyst, Cat-1-R, Cat-3-R, Cat-5-R, and Cat-8-R, respectively. The pretreated monolithic samples were cooled to 300 °C in flowing N2. Afterword, these catalysts were tested in the reaction gas from 300 to 500 °C at a rate of 5 °C/min since the conversion of CH4 can hardly occur below 300 °C. The outlet gases of the reactor were continually analyzed with a Fourier Transform Infrared (FT-IR) spectrometer (Thermo Fisher Scientific, Antaris IGS-6700), calibrated at 940 Torr and 165 °C. The conversions of reactant species i (Xi) were calculated using the Eq. (2): Xi = 100%  (Ci, inlet - Ci, outlet) / Ci, inlet

(2)

Where Ci, inlet and Ci, outlet represented the volumetric concentration of i species in the inlet gas and outlet gas, respectively.

2.3 Periodic pulse and lambda oscillation operating tests According to the results of above activity test, Cat-5-R with highest activity of CH4 conversion was applied to test under the conditions of both periodic oxygen pulse and oscillation lambda. The procedures of the experiments were as follows. After pretreating the monolithic sample in the reaction gas at 500 °C for 5 h, the feed was cut off at once and then the catalysts were cooled to 450 °C in flowing N2. The temperature was kept constant for several hours of time on stream and the reaction gas was turn on at the same time. 10


Page 10 of 47

Page 11 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


The catalytic activity of CH4 conversion was continually recorded at 450 °C. As presented in Figure S4a, periodic pulse experiments were conducted by performing 2 min long stoichiometric pulses (lambda=1.00) per 60 min on a stationary rich reaction gas mixture (lambda=0.98). For comparison, the activity of the CH4 conversion over the sample were also tested at lambda =1 for 60 min to evaluate the loss of the activity. Lambda changes were acquired by varying the concentration of O2 from 0.089% (lambda = 0.98) to 0.403% (lambda= 1.00). For oscillation lambda experiment (Figure S4b), the value of lambda varied from 0.98 to 1.02 at constant cycle period (120 s). Altering the oxygen concentration from 0.089% to 0.717% periodically was performed to achieve lambda oscillations by adjusting the setting value of the mass flow controller of oxygen. For keeping a constant total gas flow of 1680 mL/min during the experiments, the variation of oxygen concentration was compensated by simultaneously changing the flow of N2 component. The CH4 conversion was calculated at each setting lambda value.

2.4 Characterizations To identify the effects of pretreatment in the reaction gas on the texture of catalyst, N2 adsorption-desorption (Quantachrome, Autosorb SI) experiments operated at -196 °C was performed to evaluate the total pore volume and the Brunauer-Emmett-Teller

(BET)

surface

area

according

Barrett-Joyner-Halenda (BJH) method and BET model, respectively. 11


to

the


Page 12 of 47

To investigate the effects of the pretreatment on activity of CH4 conversion, the catalysts were also tested after pretreatment in N2. The procedures of the pretreatment were displayed in Figure S3.The catalyst was also tested from 300 to 500 °C at a rate of 5 °C/min just as the above.

The X-ray photoelectron spectroscopy (XPS) was recorded on a Kratos XSAM 800 spectrometer equipped with an Mg Kα radiation (1253.6 eV). The experiment was performed at 13 kV and 20 mA. The XP-signal positions were calibrated by the C 1s peak position at 284.6 eV. One effective approach to rapidly quench the Pd catalyst for preserving Pd state was to purge the Pd catalyst with N2 at the operating temperature, and then cool quickly to room temperature in N2 without exposure to air

7, 32.

The procedures of preparation of the samples for XPS measurements were shown in Figure 3. The feed was cut off immediately once the pretreating process finished. The catalysts were then rapidly cooled to room temperature in flowing N2. Afterward, the samples were stored in a N2-filled container. Finally, these catalysts were quickly introduced into the XPS vacuum chamber for the testing and the XPS spectra were recorded.

Transmission electron microscopy (TEM) measurements were used to investigate the state of Pd over the as synthesized catalyst and pretreated 12


Page 13 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


catalyst. The experiments were performed on a Tecnai G2 F20 S-TWIN TEM (FEI Company, USA, 200 kV accelerating voltage). The powder catalysts were diluted in ethanol and dispersed by ultrasonic prior to the measurements. The samples were finally prepared by depositing a drop of the solution on Cu grids coated by a carbon film and dry in air.

To detect the deposited carbon on the catalyst under fuel-rich conditions, thermo-gravimetric (TG) measurement was carried out on a NETZSCH TG 209F1 type thermos-gravimetric. The sample was heating from 25 to 600 °C with a rate of 10 °Cmin-1 in 20/80 mLmin-1 of the oxygen/argon flow. The outlet gas from the reactor was continuously monitored using mass spectrometry (MS) following the m/z 44 (CO2).

3. Results and discussions 3.1 Monolithic Pd catalyst for CH4 conversion It is well known that the catalyst for the after-treatment system of industrial NGVs must be pretreated under the real NGVs exhaust gases before working 33, 34.

However, the impacts of pretreatment on the catalytic performance are

still not very clear. In this work, we evaluated the activity of CH4 conversion over the Pd model catalyst after pretreatment in reaction gas for 0, 1, 3, 5, and 8 h, respectively. The procedures of the pretreatment were descripted in Figure 1a. The results were presented in Figure 1b and Figure S1. For all the 13



Page 14 of 47

catalysts, CO conversions were almost 100% in the testing temperature from 300 to 420 °C and the conversions decreased slightly above ca. 420 °C with a value of 96% at 500 ° C (Figure S1a). NO conversion nearly followed the behavior of CH4 abatement during the measurement (Figure S1b). These observations were in agreement with previous results

35, 36.

However,

compare to the Untreated catalyst, the CH4 light-off temperature (temperature of 50% CH4 conversion) of the pretreated catalysts (Cat-1-R, Cat-3-R, Cat-5-R, and Cat-8-R) shifted by about 20-50 ° C to lower temperatures (Figure 1b). The results suggested that the pretreated samples were more active than the un-pretreated ones. Typically, the CH4 light-off temperature over the pretreated catalysts monotonously decreased from 444 to 395 °C as the pretreatment times increased from 0 to 5 h, but it slightly increased to 398 ° C with the pretreatment times further increasing to 8 h. As a result, Cat-5-R presented the highest catalytic activity for CH4 conversion under stoichiometric conditions among these samples. The results implied that the pretreatment of catalyst had a strong effect on the activity of CH4 removal.

Note that the pretreated catalysts were endured in long process during the pretreatment in the reaction gas (Figure 1), which might affect the texture of catalyst and hence the catalytic activity of CH4 elimination. Taking that into account, the N2 adsorption-desorption measurement was performed over the catalyst pretreated in the reaction gas. All the catalysts presented a similar 14


Page 15 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


pore size of maximum distribution at around 5.4 nm (Table S1 and Figure S2) and shown the same type of isotherms IV

37.

Besides, the pore volume

remained almost constant during the pretreatment. The results suggested that the pretreatment in reaction gas did not destroy the pore texture of the catalysts. It can be also seen from Table S1 that the BET surface areas decreased slightly from 142 to 135 m2/g as the pretreatment times raised from 0 to 8 h. Nevertheless, with a view of the distinct differences of CH4 abatement activity after the pretreatment, slight reduction of BET surface areas appeared insignificant to the differences. As a result, the textural properties of these pretreated catalysts would play a minor role in the activity of CH4 conversion.

15



(a)

(b)

Figure 1. (a) Description of the pretreatment procedures under reaction gas; (b) CH4 light-off curves as temperature was raised from 300 to 500 °C in simulated stoichiometric NGVs exhausts for Pd model catalyst (50 g/ft3) after the above pretreatment. Conditions: 1000 ppm CH4 + 5000 ppm CO + 930 ppm NO + 4035 ppm O2 + 10vol. %CO2 + 10vol. %H2O balanced with N2 (lambda=1); Gas hourly space velocity (GHSV) = 50,400 h−1. 16


Page 16 of 47

Page 17 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


The chemical state of Pd on the Pd-based catalysts was known to be crucial to its catalytic performance for the conversion of CH4 from NGV exhaust gases

4, 7, 8, 15, 22, 23, 27.

Different conditions of pretreatment may result

in various chemical states of Pd species and thereby had different activity of CH4 conversion. To further investigate the effects of pretreatment on the activity of CH4 removal, the catalysts were allowed to treat at 500 ° C in flowing N2 for 0, 1, 3, 5, and 8 h, respectively. The procedures of the pretreatment in N2 were displayed in Figure S3a. The flow rate of N2 gas was the same as that of the reaction gas and the N2 feed was maintained during cooling process. The activity of CH4 elimination was tested and the results were shown in Figure S3b. It can be found that the curves of CH4 conversion for the catalysts pretreated in N2 appeared to be similar to that of the Untreated catalyst. The small differences in the CH4 light-off temperature for these curves might result from the coating deviation of the samples. During the pretreatment in flowing N2, the Pd species almost maintained its oxidized state, and the activity of CH4 conversion was nearly constant. As mentioned above, the CO conversions of the catalysts pretreated in reaction gas were lowered to about 96% at 500 °C and there was still ca. 200 ppm CO in the gas feed (Figure S1a). It indicated that the stream may have a moderate reduction property. Accordingly, the chemical states of Pd species over the catalysts pretreated in reaction gas may be in both metallic Pd and PdO. Oxide Pd species can catalyze the reaction of CH4 conversion in the 17



conditions of stoichiometric NGV exhaust gas

4, 15, 23.

However, the metallic

Pd may also convert CH4 pollutant efficiently in the present conditions, in consistent with the previous works 12, 17. For verifying the roles of reduced and oxide Pd species, Cat-5-R and Untreated catalyst in Figure 1a were allowed to cool in flowing air, and those were named as Cat-5-R (air) and Untreated catalyst (air), respectively. Similarly, Cat-5-R (N2) and Untreated catalyst (N2) represented the samples being cooled in N2. As shown in Figure 2, the activity of CH4 conversion over Untreated catalyst (air) was nearly the same as that of Untreated catalyst (N2). Since the as-prepared catalyst was calcined in air at 550 °C for 3 h, the initial Pd species should be in the form of PdO (Pd2+). Heating the catalyst in air and further cooling in N2 (or air) would not change the Pd chemical state. It may be concluded that the Pd species retained initially oxidized state over the Untreated catalyst after these two cooling steps. However, the CH4 light-off temperature of Cat-5-R (N2) was about 27 ° C lower than that of Cat-5-R (air), indicating that Cat-5-R cooled in N2 became more active than that cooled in air. The decreased activity over Cat-5-R (air) could be attributed to the gradually oxidized of the initial reduced Pd species during cooling in air. The results were in good agreement with the recent studies on Pd/Al2O3 where PdO species was progressively formed over the reduced Pd/Al2O3 in the conditions of excess O2 leading to considerably decreased activity of CH4 conversion

22.

Similar observations of declined

activity for CH4 elimination over reduced Pd catalyst after O2 pulsing were 18


Page 18 of 47

Page 19 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


found on other literature

26, 31.

Therefore, reduced Pd species may be

essential for CH4 conversion in the conditions of stoichiometric NGV exhaust gas.

C B A D

Figure 2. CH4 conversions as a function of temperature over the Pd model catalyst after different cooling conditions. (A) The Pd catalyst was cooled in N2 after pretreatment in reaction gas for 5 h. (B) The Pd catalyst was cooled in air after pretreatment in reaction gas for 5 h. (C) The untreated catalyst in Figure 1 was cooled in N2. (D) The untreated catalyst in Figure 1 was cooled in air.

19



Page 20 of 47

3.2 Pd chemical state for the activity of CH4 conversion To unravel the chemical state of Pd species responsible for these distinct differences in catalytic activity of CH4 conversion, XPS as an effective technique was applied. Figure 3 presented the procedures of preparation of the samples for XPS tests. Deconvolution of Pd 3d spectra from the catalysts

pretreated

in

reaction

gas

was

performed

by

fitting

a

Gaussian-Lorentzian function. The results were shown in Figure 4. For Untreated catalyst, the peaks at ca.336.4 eV and 341.7 eV were assigned to PdO 3d5/2 and PdO 3d3/2, respectively. It was found that the chemical state of surface Pd over Untreated catalyst existed entirely in PdO. For the pretreated catalysts (Cat-1-R, Cat-3-R, Cat-5-R, and Cat-8-R), the peaks ascribing to PdO became much more weakly. Additionally, two new peaks around 335.0 eV and 340.3 eV were present over the pretreated samples, and the peaks can be attributed to metallic Pd species. Since the peak area of metallic Pd species was much more than that of PdO one, the former was the primary Pd species over the pretreated catalysts.

Figure 3. Schematic preparation of the samples for XPS measurements. 20


Page 21 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 4. The X-ray photoelectron Pd 3d spectra of the Pd model catalyst after various pretreatment under reaction gas.

21



For the purpose of further investigating the state of Pd over the catalysts, HRTEM experiments were employed. Figure 5 showed the HRTEM results of Untreated catalyst (a and b) and Cat-5-R (c and d). The state of Pd nanoparticles was analyzed by d-spacing measurements. PdO species was presented over Untreated catalyst, and the PdO nanoparticles were crystalline with the spacing of 0.265 nm and 0.215 nm corresponding to PdO(1 0 1) and PdO(1 1 0), respectively. In contrast, metallic Pd was found over Cat-5-R. The observed metallic Pd nanoparticles primarily exposed Pd(1 1 1) and Pd(2 0

0). Figure 5. TEM images of Untreated catalyst (a and b) and Cat-5-R (c and d). 22


Page 22 of 47

Page 23 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


In order to identify the relative amount of metallic Pd and PdO, the quantitative

analysis

accounted

for

a

nonlinear

Shirley

background

substraction using the Casa XPS software. Figure 6 presented the relationship between the ratio of Pd/(Pd + PdO) and the CH4 light-off temperature at various pretreated time in reaction gas. It can be seen that the ratio of Pd/(Pd + PdO) sharply increased from 0 to 89% as the pretreatment times increased from 0 to 5 h, but it slightly decreased to 85% with the pretreatment times further increasing to 8 h. Interestingly, the CH4 light-off temperature monotonously declined from 444 to 395 °C as the ratio of Pd/(Pd + PdO) inclined from 0 to 89%, and then that slightly increased to 398 °C as this ratio decreased to 85%. It suggested that the CH4 light-off temperature was closely associated with the ratio of Pd/(Pd + PdO), and that the higher ratio of Pd/(Pd + PdO) was beneficial for the improvement of CH4 conversion activity under stoichiometric conditions. Much more works reported that PdO was the most active phase over Pd-based catalyst for CH4 conversion under lean conditions

38-40.

However, there is still no answer about which state of Pd

species acting as more active site for CH4 conversion under stoichiometric conditions of NGVs

4, 23, 27, 41, 42.

The results indicated that PdO and metallic

Pd species can both act as active sites but the metallic Pd phase may play a vital role in the elimination of CH4 from stoichiometric NGV exhaust gases.

23



Figure 6. CH4 light-off temperature and the ratio of Pd/(Pd+PdO) as a function of the pretreatment time in reaction gas.

Alternative CO/O2 treated experiments were used to further verify the roles and relative activity of metallic Pd and PdO species for CH4 conversion. The samples for the experiments were prepared according to the process as shown in Figure 7a. The results were displayed in Figure 7b, Figure 7c, and Figure 7d. Firstly, the Pd model catalyst was reduced in 1% CO at 500 °C for 1 h during the step I, and then this sample was tested in the reaction gas (-light-off test) after cooled in N2. It can be seen that the change of pretreatments resulted in remarkably different catalytic activities of CH4 conversion depending on the pretreated gas. As shown in Figure 7b, the CH4 light-off temperature over the reduced Pd catalyst lowered to 361 ° C. This 24


Page 24 of 47

Page 25 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


value was even 34 ° C lower than that of Cat-5-R, which had the highest catalytic activity among the catalysts pretreated in reaction gas. Secondly, the catalyst after the test I was then oxidized by 2% O2 at 500 °C for 1 h during the step II. However, the CH4 light-off temperature in -light-off test shifted strikingly by about 90 °C to higher temperatures. The results suggested that the reduced catalyst was remarkably more active than the oxidized sample once again. In order to identify the repeatability of above observations, the catalyst was continuously treated in CO/O2 and then tested in the reaction gas. As shown in Figure 7a, the procedures of step III and V were the same as that of step I, and correspondingly the procedures of step IV and VI were similar to that of step II. Interestingly, the CH4 light-off temperature for -light-off test was 95 ° C lower than that for -light-off test. A similar trend was also observed over the following -light-off test and -light-off test. The results verified the repeatability of the considerably different activities between the reduced and oxidized Pd catalyst. The sample after reduced in CO for 1 h during step III and the one after oxidized in O2 for 1 h during step IV were measured by XPS technique to acquire the chemical state of Pd species. As shown in Figure 8, the Pd chemical state in the end of step III and step IV was completely metallic Pd and PdO, respectively. It confirmed that these two Pd species were the active phases in the conversion of CH4 under stoichiometric conditions. Considering that the activity of CH4 conversion over this reduced Pd catalyst (100 % metallic Pd) was relatively higher than those of Cat-5-R 25



Page 26 of 47

(89% metallic Pd) and Untreated catalyst (100 % PdO), respectively, metallic Pd species was suggested to be much more active than PdO phase in stoichiometric NGVs exhaust gas conditions.

(a)

(b)

(c)

Figure 7. (a) Description of the procedures for alternative CO/O2 treated experiment, and (b, c, and d) the CH4 light-off tests for the Pd model catalyst (50 g/ft3) after the above pretreatment.

26


(d)

Page 27 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 8. The X-ray photoelectron Pd 3d spectra of the Pd model catalyst after the reduction by CO in Step III (a) and the oxidation by O2 in Step IV (b), respectively.

3.3 The evolution of Pd species under operation conditions The real catalyst normally worked in the fluctuation of air-fuel ratio deviated from the stoichiometric value under working conditions (lambda=0.98-1.02) Our previous work

15

43.

studied the variation of Pd species and the

corresponding catalytic performance of CH4 removal at the right side of the stoichiometric value (lambda=1.003-1.02). However, for the left side of the stoichiometric value (lambda=0.98-1.00), the effects of the potential change in Pd species on CH4 conversion are not yet clear. To reveal the actual Pd 27



Page 28 of 47

species under operation conditions, the experiments about periodic stoichiometric

pulse

(lambda=1.00)

under

stationary

rich

conditions

(lambda=0.98) were conducted. The catalyst was firstly pretreated at 500 °C for 5 h in the reaction gas and then cooled from 500 to 450 °C in flowing N2 to maintain the metallic Pd species. The catalyst was tested at 450 ° C in lambda=0.98 from 0 to 60 min, following by evaluating at the same temperature in lambda=1.00 from 60 to 62 min. As shown in Figure. 9a, the activity of CH4 conversion can be nearly retained at the beginning of the testing under stationary rich conditions (lambda=0.98). It may be due to that the oxygen-storing capacity (OSC) of ceria is capable of dumping lambda-oscillations to some extent

44,

thus permitting the catalyst to operate

close to stoichiometric conditions to achieve high activity of CH4 removal

45.

Note that the CH4 conversion decreased gradually from initial 99% to 74% within the first 60 min. It indicated that the active state in CH4 conversion converted progressively to one less active or inactive state with relatively low activity of CH4 removal. Additional O2 was then added to the reaction system for increasing the value of lambda to 1.00 between 60 and 62 min. Strikingly, the CH4 conversion was immediately recovered to the initial level (about 98%) during the O2 pulse, which may be due to the present of the active phase over the Pd model catalyst. To identify these observations, the catalyst was further tested in the periodic stoichiometric pulse in lambda=0.98 gas feed for two circles as displayed in Figure 9a. Actually, the CH4 conversion once again 28


Page 29 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


declined from 99% to 71% at lambda=0.98 between 62 and 122 min, and that was rapidly recovered to 98% at lambda=1.00 in 122-124 min after O2 pulsing.

(a)

(b)

Figure 9. (a) Effect of periodic stoichiometric pulse (lambda=1.00) on CH4 conversion under stationary rich (lambda=0.98) reaction gas mixture (Toven =450 °C); (b) CH4 conversion under stationary stoichiometric (lambda=1.00) reaction gas mixture (Toven =450 °C).

Besides, a similar trend of CH4 abatement was found on the following 124-184 min test. As a result, the catalyst shown a poor stability of CH4 conversion under stationary rich conditions (lambda=0.98). For comparision, the stability of the catalyst in stoichiometric reaction gas was also examined. As presented in Figure 9b, the catalyst after the pretreatment was tested at 450 ° C in lambda=1.00 for 1 h. Note that the CH4 conversion decreased slightly from 98% to 94.5% and the activity can be almost maintained during 29



testing in lambda=1.00. The results seemed to be different from the previous work

23

that reported the gradually reduced conversion of CH4 over Pd-mainly

commercial catalyst from initial 52 to 43% after testing under stationary stoichiometric conditions for 30 min. However, it was not clear about what were the cooling conditions of the Pd-mainly catalyst after the pretreatment. If the pretreated Pd-mainly catalyst was cooling in lean conditions, the CH4 conversion was supposed to decline considerably owing to the formation of PdO species according to the results of Figure 2. Additionally, the slight decrease of CH4 conversion activity might be associated with the behaviors of self-oscillations during CH4 oxidation and that might involve the gradual formation and removal of PdCx

30.

Therefore, the results suggested that the

activity of CH4 conversion under lambda=1.00 was more stable than that under lambda=0.98. It may be due to that the active Pd0 phase was mainly retained under stationary stoichiometric conditions.

To explore the reason responsible for the considerable deterioration of CH4 conversion activity under stationary rich conditions, TG-MS and XPS techniques were applied. The sample for these tests was prepared as follows. The reaction was terminated at once when the CH4 conversion decreased to 71% as shown in the green point of Figure 9a. The catalyst was collected after cooled to room temperature in flowing N2. The results of TG-MS were shown in Figure 10. During heating in 20%O2, a strong peak of CO2 at ca. 30


Page 30 of 47

Page 31 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


331 ° C and a weak peak of CO2 at ca. 393 ° C were observed. These two peaks were ascribed to C(ads) and Cgraph, respectively

46.

It indicated the

existence of coke over the catalyst. This was in well consistent with the report of M. Stockenhuber et al

47.

They studied the oxidation of CH4 in an excess of

air and found the formation of carbon at very low temperature (ca. 180 ° C) during the reaction. Besides, the deposition of carbon can be initially removed after heating in air at about 320 ° C by detecting CO2 in the outlet of the reactor.

Figure 10. The TG-MS following the m/z 44 (CO2) of the Pd model catalyst under stationary rich conditions (lambda=0.98) for 1 h as shown in the green point of Figure 9;

31



Page 32 of 47

The oxygen-storing ceria in the catalyst can also favor the removal of coking deposition smoothly

45.

Accordingly, the recovery of CH4 conversion activity

over the Pd model catalyst after O2 pulse suggested that the deposited carbon can be easily removed by heating in 20%O2 at 450 °C. Therefore, the remarkable reduction of CH4 elimination activity under stationary rich conditions could be attributed to the cover of the active Pd0 species by coke. By studying the conversion of CH4 during net-reducing conditions, J. Nilsson et al. 31 regarded the accumulation of carbonaceous species on the surface of the Pd catalyst as a Pd-carbide species. A similar phenomenon was also found in the work of M. Stockenhuber et al.

47,

who attributed this species to

one form of Pd-carbide species. With regarding to the formation of PdCx, it was supposed that the dissociation of CH4 may occur at 450 ° C in the rich condition, resulting in the formation of H2 and carbon cluster

46.

The formed C

species was suggested to further interact with Pd component to generate the PdCx, in line with the work by O. Balmes

48

who reported that carbon species

can dissolve into the Pd nanoparticles to form one carbide phase. As a result, the probable formation of PdCx resulted in the decreased concentration of the active Pd0 species and thereby the reduced activity of CH4 removal. Considering the immediate recovery of activity after O2 pulse, it indicated that the PdCx species can be easily converted into active metallic Pd state after increasing O2 concentration. The effects of lambda oscillation on the activity of CH4 conversion were 32


Page 33 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


further examined. The catalyst was pretreated in the reaction gas at 500 °C for 5 h and then cooled from 500 to 450 °C in flowing N2 to retain the metallic Pd species. The activity of CH4 elimination was recorded as the lambda value increased from 0.98 to 1.02 and consequently decreased from 1.02 to 0.98 at constant cycle period of 120 s. As displayed in Figure 11, starting at lambda = 0.98 with 30 s when the deposit of coke rarely occurred, nearly 98% conversion of CH4 can be observed. Afterward, further increasing O2 amount (lambda=1.00) had a little effect on the activity. The results suggested that the active phase for CH4 conversion may exist mainly as metallic Pd species when switching from lambda =0.98 to 1.00. However, further increasing O2 concentration towards slight lean conditions (lambda=1.02), CH4 conversion decreased rapidly from 96% to 65% with 30 s owing to the sharply converting the active phase for CH4 elimination to the less active or inactive one.

33



Page 34 of 47

Figure 11. Effect of lambda oscillation on CH4 conversion (Toven=450 °C). The value of lambda oscillation varied from 0.98 to 1.02 at constant cycle period of 120 s.

Under the 30 s lean conditions, oxygen initially was adsorbed on the partial of surface metallic Pd and some PdO could be formed

49.

Nevertheless, the

surface of the remaining metallic Pd species may be saturated by the oxygen because of “oxygen poisoning” effect

50.

In this situation, the metallic Pd

species covering by oxygen were not beneficial for the adsorption and activation of CH4 molecules

51, 52,

leading to a relatively low activity for CH4

conversion. However, under this lean conditions, PdO served as the most active phase for CH4 elimination. According to the studies of Nilsson

26, 31,

the

activity of CH4 conversion was able to gradually increase after prolonging the 34


Page 35 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


reaction time when the oxidation of Pd bulk was built. Following the increase of O2 content (lambda=1.02), decreasing the O2 concentration (lambda=1.00) resulted in a slight increment of CH4 conversion with a stable level about 73%. Within the 30 s of lambda=1.00, the primary state of Pd during this period may be still the oxygen saturated Pd surface despite the metallic Pd species covering by oxygen was partially reduced. Under this circumstance, the activity of CH4 abatement would be not high. Additionally, the decreased activity under lean condition (λ=1.02) is partial due to the inhibitory effect of water associated with the formation of Pd-OH species

10.

Interestingly, further

decreasing O2 concentration (lambda=0.98) nearly recovered the CH4 conversion to the initial level with 99%. It was mainly due to that the active Pd0 species can be smoothly presented during this test. A similar behavior was observed in the following two circles of periodic lambda oscillation tests. The results indicated that the Pd species can dynamically change with the variation of lambda under conditions of NGVs exhaust gases.

The pulse experiments depended significantly on the OSC, not only CH4 conversion but also CO. The changes in the conversions of CH4 and CO with lambda and the CO/NO crossover point were displayed in Figure. S5. At sub-stoichiometric regions (λ =0.98), the conversions of CH4 and CO were high 97%) at the beginning of the experiment (0-30 s). Even though the presence of water can decrease the OSC of ceria 35


44, 53,

ceria still had the


ability to dump the lambda-oscillation to some extent, and thereby favor the CH4 and CO conversions. Besides, during this rich conditions, the CO conversion can also be promoted by the reaction of CO with ceria (CO + Ce2O4  CO2 + Ce2O3)

44.

Under stoichiometric conditions (30-60 s), ceria

may play a role in maintaining the active Pd species, which allowed CH4 and CO to be converted smoothly. Further increasing the value of lambda to 1.02 (60-90 s), the CO conversion was almost 100%, which was benefited from the excess of oxygen. However, the CH4 conversion presented a significant decrease. As mentioned above, the main surface Pd species during this 30 s lean conditions may be metallic Pd, despite that some PdO have been formed by the reaction of Pd0 with O2. Therefore, the CH4 conversion in this situation would be reduced. The green heart-shaped point in Figure. S5 indicated the CO–NO crossover point. Ce may exist mainly in CeIV (CeO2) at the right of CO-NO cross-over point (lean conditions). In the circumstances, Pd may get occupied easily by CO for further conversion instead of CH4. It was confirmed by the observations of the increase in CO conversion and the decrease in CH4 conversion at the right of CO-NO cross-over point.

According to the above results, the evolution of the Pd species under operation conditions of stoichiometric NGVs was proposed. Under working conditions, sudden slowing down and acceleration at any time usually accompanied by the change of O2 concentration in NGVs exhaust gas. In this 36


Page 36 of 47

Page 37 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


case, the fluctuation of the ratio of air-fuel had an essential role on the chemical state of the Pd species. Under rich conditions (lambda=0.98), metallic Pd can acted as the active site for the conversion of CH4 over Pd catalyst, but this active phase suffered from the cover by deposit carbon to form the inactive PdCx. However, the PdCx species can smoothly transform to the active Pd0 phase after increasing O2 amount to stoichiometric one (lambda=1.00). Further increasing O2 concentration towards slight lean conditions (lambda=1.02), PdO was widely accepted to the most active species in the elimination of CH4 from NGVs exhaust gases 7, 8, 12, 14, 15.

4. Conclusion In summary, pretreating the Pd model catalyst in reaction gas can remarkably enhance the conversion of CH4 from a stoichiometric NGVs exhaust gas. The alternative CO/O2 treated experiments combined with the XPS analysis indicated that metallic Pd phase was much more active than PdO for the elimination of CH4 under stoichiometric conditions (lambda=1.00). However, under stationary rich conditions (lambda=0.98), inactive PdCx was gradually formed, resulting in considerable reduced catalytic activity. Fortunately, the PdCx species can be efficiently converted to the active Pd0 phase after pulsing O2 (lambda=1.00). In lean conditions (lambda=1.02), Pd species progressively existed in PdO acting as the most active phase for CH4 conversion

7, 14.

The findings of the evolution of Pd species would help in the 37



development and operation of highly efficient Pd catalysts for NGVs emission control.

5. Acknowledgments This work was supported by the National Key R&D Program of China (2016YFC0204902), the National Natural Science Foundation of China (21673146), and the Graduate Student’s Research and Innovation Fund of Sichuan University (2018YJSY093). We also thank Shanling Wang of The Analytical & Testing Center (ATC) of Sichuan University for performing TEM characterization.

Supporting Information CO and NO conversions, catalyst texture, periodic oscillation operating tests, and the changes in the conversions of CH4 and CO with lambda and the CO/NO crossover point were included; other Figures as mentioned in the text.

References (1) Petrov, A. W.; Ferri, D.; Krumeich, F.; Nachtegaal, M.; van Bokhoven, J. A.; Kröcher, O. Stable complete methane oxidation over palladium based zeolite catalysts. Nat. Commun. 2018, 9, (1), 2545. (2) Johnson, T.; Joshi, A. Review of Vehicle Engine Efficiency and Emissions.

SAE Technical Paper 2018-01-0329 2018. 38


Page 38 of 47

Page 39 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(3) Cargnello, M.; Delgado Jaen, J. J.; Hernandez Garrido, J. C.; Bakhmutsky, K.; Montini, T.; Calvino Gamez, J. J.; Gorte, R. J.; Fornasiero, P. Exceptional activity for methane combustion over modular Pd@CeO2 subunits on functionalized Al2O3. Science 2012, 337, (6095), 713-7. (4) Ferri, D.; Elsener, M.; Kröcher, O. Methane oxidation over a honeycomb Pd-only three-way catalyst under static and periodic operation. Appl. Catal., B 2017, 220, 67-77. (5) Winkler, A.; Dimopoulos, P.; Hauert, R.; Bach, C.; Aguirre, M. Catalytic activity and aging phenomena of three-way catalysts in a compressed natural gas/gasoline powered passenger car. Appl. Catal., B 2008, 84, (1-2), 162-169. (6) Huang, W.; Goodman, E. D.; Losch, P.; Cargnello, M. Deconvoluting Transient Water Effects on the Activity of Pd Methane Combustion Catalysts.

Ind. Eng. Chem. Res. 2018, 57, (31), 10261-10268. (7) Ciuparu, D.; Lyubovsky, M. R.; Altman, E.; Pfefferle, L. D.; Datye, A. Catalytic Combustion of Methane over Palladium-Based Catalysts. Catal. Rev. 2002, 44, (4), 593-649. (8) Gélin, P.; Primet, M. Complete oxidation of methane at low temperature over noble metal based catalysts: a review. Appl. Catal., B 2002, 39, (1), 1-37. (9)Beulertz, G.; Votsmeier, M.; Moos, R. Effect of propene, propane, and methane on conversion and oxidation state of three-way catalysts: a microwave cavity perturbation study. Appl. Catal., B 2015, 165, 369-377. 39



(10) Gholami, R.; Alyani, M.; Smith, K. Deactivation of Pd Catalysts by Water during Low Temperature Methane Oxidation Relevant to Natural Gas Vehicle Converters. Catalysts 2015, 5, (2), 561-594. (11) Salaün, M.; Kouakou, A.; Da Costa, S.; Da Costa, P. Synthetic gas bench study of a natural gas vehicle commercial catalyst in monolithic form: On the effect of gas composition. Appl. Catal., B 2009, 88, (3-4), 386-397. (12) Granger, P. Challenges and breakthroughs in post-combustion catalysis: how to match future stringent regulations. Catal. Sci. Technol. 2017, 7, 5195-5211. (13) Xie, S.; Liu, Y.; Deng, J.; Zhao, X.; Yang, J.; Zhang, K.; Han, Z.; Dai, H. Three-dimensionally ordered macroporous CeO2 -supported Pd@Co nanoparticles: Highly active catalysts for methane oxidation. J. Catal. 2016,

342, 17-26. (14) Monai, M.; Montini, T.; Gorte, R. J.; Fornasiero, P. Catalytic Oxidation of Methane: Pd and Beyond. Eur. J. Inorg. Chem. 2018, 2884-2893. (15) Huang, F.; Chen, J.; Hu, W.; Li, G.; Wu, Y.; Yuan, S.; Zhong, L.; Chen, Y. Pd or PdO: Catalytic active site of methane oxidation operated close to stoichiometric air-to-fuel for natural gas vehicles. Appl. Catal., B 2017, 219, 73-81. (16) Chen, J.; Arandiyan, H.; Gao, X.; Li, J. Recent Advances in Catalysts for Methane Combustion. Catalysis Surveys from Asia 2015, 19, (3), 140-171. (17) Matam, S. K.; Aguirre, M. H.; Weidenkaff, A.; Ferri, D. Revisiting the 40


Page 40 of 47

Page 41 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Problem of Active Sites for Methane Combustion on Pd-Al2O3 by Operando XANES in a Lab-Scale Fixed-Bed Reactor. J. Phys. Chem. C 2010, 114, 9439-9443. (18) Colussi, S.; Trovarelli, A.; Vesselli, E.; Baraldi, A.; Comelli, G.; Groppi, G.; Llorca, J. Structure and morphology of Pd/Al2O3 and Pd/CeO2/Al2O3 combustion catalysts in Pd–PdO transformation hysteresis. Appl. Catal., A 2010, 390, (1-2), 1-10. (19) Zou, X.; Rui, Z.; Song, S.; Ji, H. Enhanced methane combustion performance over NiAl2O4-interface-promoted Pd/γ-Al2O3. J. Catal. 2016, 338, 192-201. (20) Kim, J.; Kim, E.; Han, J.; Han, H. S. Pt/Pd Bimetallic Catalyst with Improved Activity and Durability for Lean-Burn CNG Engines. SAE

International Journal of Fuels and Lubricants 2013, 6, (3), 651-656. (21) Willis, J. J.; Gallo, A.; Sokaras, D.; Aljama, H.; Nowak, S. H.; Goodman, E. D.; Wu, L.; Tassone, C. J.; Jaramillo, T. F.; Abild-Pedersen, F.; Cargnello, M. Systematic Structure-Property Relationship Studies in Palladium-Catalyzed Methane Complete Combustion. ACS Catal. 2017, 7, 7810-7821. (22) Xiong, H.; Wiebenga, M. H.; Carrillo, C.; Gaudet, J. R.; Pham, H. N.; Kunwar, D.; Oh, S. H.; Qi, G.; Kim, C. H.; Datye, A. K. Design considerations for low-temperature hydrocarbon oxidation reactions on Pd based catalysts.

Appl. Catal., B 2018, 236, 436-444. 41



(23) Bounechada, D.; Groppi, G.; Forzatti, P.; Kallinen, K.; Kinnunen, T. Effect of periodic lean/rich switch on methane conversion over a Ce–Zr promoted Pd-Rh/Al2O3 catalyst in the exhausts of natural gas vehicles. Appl. Catal., B 2012, 119-120, 91-99. (24) Chi-Bum In, S.-H. K., Chang-Dae Kim, and Won-Suk Cho. Catalyst Technology Satisfying Low Emission of Natural Gas Vehicle. SAE Technical

Paper 970744 1997. (25) Kalz, K. F.; Kraehnert, R.; Dvoyashkin, M.; Dittmeyer, R.; Glaser, R.; Krewer, U.; Reuter, K.; Grunwaldt, J. D. Future Challenges in Heterogeneous Catalysis: Understanding Catalysts under Dynamic Reaction Conditions.

ChemCatChem 2017, 9, (1), 17-29. (26) Nilsson, J.; Carlsson, P.-A.; Martin, N. M.; Adams, E. C.; Agostini, G.; Grönbeck, H.; Skoglundh, M. Methane oxidation over Pd/Al2O3 under rich/lean cycling followed by operando XAFS and modulation excitation spectroscopy. J.

Catal. 2017, 356, 237-245. (27) Lu, Y.; Keav, S.; Marchionni, V.; Chiarello, G. L.; Pappacena, A.; Di Michiel, M.; Newton, M. A.; Weidenkaff, A.; Ferri, D. Ageing induced improvement of methane oxidation activity of Pd/YFeO3. Catal. Sci. Technol. 2014, 4, (9), 2919. (28) Raj, A. Methane Emission Control. Johnson Matthey Technol. Rev. 2016,

60, (4), 228-235. (29) Klingstedt, F.; Karhu, H.; Neyestanaki, A. K.; Lindfors, L. E.; Salmi, T.; 42


Page 42 of 47

Page 43 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Väyrynen, J. Barium Promoted Palladium Catalysts for the Emission Control of Natural Gas Driven Vehicles and Biofuel Combustion Systems. J. Catal. 2002, 206, (2), 248-262. (30) Bychkov, V. Y.; Tulenin, Y. P.; Slinko, M. M.; Khudorozhkov, A. K.; Bukhtiyarov, V. I.; Sokolov, S.; Korchak, V. N. Self-oscillations during methane oxidation over Pd/Al2O3 : Variations of Pd oxidation state and their effect on Pd catalytic activity. Appl. Catal., A 2016, 522, 40-44. (31) Nilsson, J.; Carlsson, P.-A.; Martin, N. M.; Velin, P.; Meira, D. M.; Grönbeck, H.; Skoglundh, M. Oxygen step-response experiments for methane oxidation over Pd/Al2O3 : An in situ XAFS study. Catal. Commun. 2018, 109, 24-27. (32) Datye, A. K.; Bravo, J.; Nelson, T. R.; Atanasova, P.; Lyubovsky, M.; Pfefferle, L. Catalyst microstructure and methane oxidation reactivity during the Pd↔PdO transformation on alumina supports. Appl. Catal., A 2000, 198, (1), 179-196. (33) Johnson, T.; Joshi, A. Review of Vehicle Engine Efficiency and Emissions. SAE Technical Paper 2017-01-0907 2017. (34) Maunula, T.; Kallinen, K.; Savimäki, A.; Wolff, T. Durability Evaluations and Rapid Ageing Methods in Commercial Emission Catalyst Development for Diesel, Natural Gas and Gasoline Applications. Top. Catal. 2016, 59, (10-12), 1049-1053. (35) Ohtsuka, H. Pt-Rh/CeO2-Al2O3 for Controlling Emissions from Natural 43



Gas Engines: Three-Way Catalytic Activity at Low Temperatures and Effects of SO2 Aging. Emission Control Science and Technology 2014, 1, (1), 108. (36) Hu, W.; Li, G. X.; Chen, J. J.; Huang, F. J.; Wu, Y.; Yuan, S. D.; Zhong, L.; Chen, Y. Q. Enhanced catalytic performance of a PdO catalyst prepared via a two-step method of in situ reduction-oxidation. Chem Commun 2017, 53, (45), 6160-6163. (37) SING, K. S. W. Reporting physisorption data for gas/solid systems with special reference to the determination of surface area and porosity. Pure Appl.

Chem. 1985, 57, (4), 603. (38) Lou, Y.; Ma, J.; Hu, W.; Dai, Q.; Wang, L.; Zhan, W.; Guo, Y.; Cao, X.-M.; Guo, Y.; Hu, P.; Lu, G. Low-Temperature Methane Combustion over Pd/H-ZSM-5: Active Pd Sites with Specific Electronic Properties Modulated by Acidic Sites of H-ZSM-5. ACS Catal. 2016, 6, (12), 8127-8139. (39) Willis, J. J.; Goodman, E. D.; Wu, L.; Riscoe, A. R.; Martins, P.; Tassone, C. J.; Cargnello, M. Systematic Identification of Promoters for Methane Oxidation Catalysts Using Size- and Composition-Controlled Pd-Based Bimetallic Nanocrystals. J Am Chem Soc 2017, 139, (34), 11989-11997. (40) Colussi, S.; Gayen, A.; Farnesi Camellone, M.; Boaro, M.; Llorca, J.; Fabris, S.; Trovarelli, A. Nanofaceted Pd-O Sites in Pd-Ce Surface Superstructures: Enhanced Activity in Catalytic Combustion of Methane.

Angew. Chem. Int. Ed. 2009, 48, (45), 8481-8484. (41) Lu, Y.; Michalow, K. A.; Matam, S. K.; Winkler, A.; Maegli, A. E.; Yoon, S.; 44


Page 44 of 47

Page 45 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Heel, A.; Weidenkaff, A.; Ferri, D. Methane abatement under stoichiometric conditions on perovskite-supported palladium catalysts prepared by flame spray synthesis. Appl. Catal., B 2014, 144, 631-643. (42) Klingstedt, F.; Neyestanaki, A. K.; Byggningsbacka, R.; Lindfors, L.-E.; Lundén, M.; Petersson, M.; Tengström, P.; Ollonqvist, T.; Väyrynen, J. Palladium based catalysts for exhaust aftertreatment of natural gas powered vehicles and biofuel combustion. Appl. Catal., A 2001, 209, (1–2), 301-316. (43) Bounechada, D.; Groppi, G.; Forzatti, P.; Kallinen, K.; Kinnunen, T. Enhanced Methane Conversion Under Periodic Operation Over a Pd/Rh Based TWC in the Exhausts from NGVs. Top. Catal. 2013, 56, (1-8), 372-377. (44) Möller, R.; Votsmeier, M.; Onder, C.; Guzzella, L.; Gieshoff, J. Is oxygen storage in three-way catalysts an equilibrium controlled process? Appl. Catal.,

B 2009, 91, (1-2), 30-38. (45) Montini, T.; Melchionna, M.; Monai, M.; Fornasiero, P. Fundamentals and Catalytic Applications of CeO2-Based Materials. Chem Rev 2016, 116, (10), 5987-6041. (46) Holger Dropsch, M. B. Interaction of methane with supported Pd catalysts studied by adsorption microcalorimetry and TPD/TPSR techniques. Appl.

Catal., A 1997, 165, 159-169. (47) Setiawan, A.; Kennedy, E. M.; Dlugogorski, B. Z.; Adesina, A. A.; Tkachenko, O.; Stockenhuber, M. Evidence of the Formation of Surface Palladium Carbide during the Catalytic Combustion of Lean Methane/Air 45



Mixtures. Energy Technology 2014, 2, (3), 243-249. (48) Balmes, O.; Resta, A.; Wermeille, D.; Felici, R.; Messing, M. E.; Deppert, K.; Liu, Z.; Grass, M. E.; Bluhm, H.; Rijn, R. v.; Frenken, J. W. M.; Westerström, R.; Blomberg, S.; Gustafson, J.; Andersen, J. N.; Lundgren, E. Reversible formation of a PdCx phase in Pd nanoparticles upon CO and O2 exposure. Phys. Chem. Chem. Phys. 2012, 14, (14), 4796. (49) Carlsson, P.-A.; Fridell, E.; Skoglundh, M. Methane oxidation over Pt/Al2O3 and Pd/Al2O3 catalysts under transient conditions. Catal. Lett. 2007,

115, (1-2), 1-7. (50) Se H. Oh; Patricia J. Mitchell; Siewert, R. M. Methane Oxidation over Alumina-Supported Noble Metal Catalysts with and without Cerium Additives.

J. Catal. 1991, 132, 287-301. (51) McCarty, J. G. Kinetics of PdO combustion catalysis. Catal. Today 1995,

26, (3), 283-293. (52) Chin, Y. H.; Buda, C.; Neurock, M.; Iglesia, E. Consequences of metal-oxide interconversion for C-H bond activation during CH4 reactions on Pd catalysts. J Am Chem Soc 2013, 135, (41), 15425-42. (53) Padeste, C.; Cant, N. W.; Trimm, D. L. The influence of water on the reduction and reoxidation of ceria. Catalysis Letters 1993, 18, (3), 305-316.

46


Page 46 of 47

Page 47 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Table of Contents graphic:

47


Evolution of Pd Species for the Conversion of Methane under

Recommend Documents