Investigation of Copper-supported Ceria Nanobers for Catalytic Oxidation of Diesel Soot

In the present research, CeO 2 and CuO/CeO 2 nanobers were synthesized by electrospinning at 12 kV DC by maintaining a tip to collector distance of 10 cm. The morphology of the as-synthesized nanobers was determined by scanning electron microscopy, and their elemental composition was veried by X-ray photoelectron spectroscopy. The activity of the prepared samples for diesel soot oxidation was determined in a bench-scale xed bed tubular reactor, and euent gases were analyzed by online gas chromatography. The variations of working parameters (air ow rate, catalyst-soot ratio, catalyst-soot contact type) were evaluated in a range to optimized reaction conditions for diesel soot oxidation. The catalyst with 40% CuO/CeO 2 had the highest surface area, the smallest crystallite size, and the best activity for diesel soot oxidation (tight contact) at T f = 318 o C.


Introduction
Diesel engines have gained immense popularity as compared to gasoline engines around the world due to their better fuel e ciency, lower operating cost, higher durability, and reliability, and they are the main power source for commercial (trucks, buses, trains, ships) and off-road industrial vehicles (excavation machinery, mining equipment) (Badini et al. 1998). However, a major environmental problem associated with diesel engines is the emission of particulate matter (PM)/soot consisting of carbonaceous soot and a soluble organic fraction (SOF) of hydrocarbons from their exhaust (Abdullah and Bhatia, 2008). Due to the carcinogenic properties of aromatic compounds, photochemical hazardous products of hydrocarbons and black carbon in PM could induce serious health problems and adverse impacts on the environment, such as global warming and climate change (Mishra and Prasad, 2017). Therefore, a possible solution to reduce PM emissions from vehicles is the use of diesel particulate lters (DPFs). The introduction of regenerative catalytic DPFs can reduce soot emission by 98% (Merkel et al. 2001). Among the existing catalytic systems for soot oxidation, noble metal-based catalysts are the most promising ones; however, they are very expensive and sensitive to sulfur poisoning. The key challenge is to nd a catalyst that can decrease the combustion temperature of PM from above 600 o C to the range of diesel exhaust (150-450 o C) (Mishra and Prasad, 2014).
Due to the limited availability of noble metals more focus is on the development of non-precious catalysts i.e. CeO 2 , NiO, CuO, MoO 3 , MnO 2 , Co 3 O 4 , and Fe 3 O 4 . Among them, copper-ceria catalysts are widely accepted because of their high reducibility at low temperatures. This is due to the synergetic effect between copper and ceria which provides chemically and thermodynamically stable metal oxides (Tang et al. 2014). Wang et al. (2011) reported that nanomaterials have improved catalytic activity than conventional materials due to their higher surface-area-to-volume ratio. Pal et al. (2017) studied Nanostructured Cu-ceria in detail and suggested substantial oxidation ability, speci c surface area, and oxygen storage capacity. Moreover, 1-D ceria nanomaterials have a lower agglomeration tendency, possess high porosity, and have an exceptionally high speci c surface area. Hence, in recent years, one-dimensional (1-D) ceria nanomaterials, such as nanorods, nanotubes, nanowires, nanobelts, nanoribbons, nano bers, have been fabricated for different catalytic applications.
Nano bers can be synthesized by different methods, such as phase separation, template synthesis, and electrospinning. Among them, electrospinning is considered the most suitable method for nano ber fabrication. Electrospinning is quite versatile because it produces continuous nano bers at low cost and permits ease of production and exibility and better control over ber diameter, microstructure, and ber arrangement (Cui et al. 2008).
There is no study reported on Cu-supported ceria nano ber catalysts synthesized by electrospinning technique for diesel soot oxidation. In the present work we have designed brous structured Ceria and Cusupported ceria catalysts with potentially enhanced trapping and soot oxidation properties. Due to special morphology these nano bers increases the contact points with the soot particles compared to the layer of catalyst which is conventionally used in diesel particulate lter. In order to correlate the catalytic properties with the structure of nano bers, the textural and spectral characterization was done. This paper additionally gets picture of the impact of working parameters for example i.e. catalyst-soot ratio, catalyst-soot contact condition, and air ow rate on diesel soot oxidation. was used as the base polymer because of its large molecular weight and high solubility in polar solvents.

Preparation of Catalysts
Ethanol (Merck) and de-ionized water were selected as the solvent and the co-solvent, respectively.
Aqueous solutions with 0 mol%, 20 mol%, 40 mol%, and 60 mol% of Cu were prepared by dissolving appropriate amounts of Cu(CH 3 COO) 2 .H 2 O in a solution of 0.4 g of cerium nitrate hexahydrate and 4 mL of de-ionized water kept in ve different vials. In another set of four vials, 10% (w/v) PVP solutions were prepared by adding an appropriate amount of PVP in 8 mL of ethanol followed by magnetic stirring. In order to prepare spinning solutions, the contents of the rst set of four vials were added separately to the vials containing PVP-ethanol solutions. The contents of the vials were mixed, and acetic acid was added drop-wise (2-3 drops) to each vial. The as-mixed solutions were then magnetically stirred for three hours at room temperature to obtain homogeneous solutions.
Each precursor solution was drawn in a plastic syringe (6 mL) equipped with a stainless-steel at-tip needle (21 gauges). The syringe was mounted vertically and attached to a syringe pump. The ow rate from the syringe was set at 1 mL/h. A metallic collector plate covered with an aluminum foil was kept directly under the needle tip to collect nano bers. The distance between the syringe tip and the collector plate was maintained at 10 cm. The syringe, the stainless-steel needle, and the solution delivery tube were washed with ethanol before casting bers from a new solution. The as-prepared nano bers were aged under ambient conditions for 24 h and then calcined in the stagnant air at 500 o C for three hours.

Preparation of Diesel Soot
The real soot was prepared by the partial combustion of locally available commercial diesel (HP) in a lamp with limited air supply and collected on the inner walls of an inverted beaker kept over the lamp. The soot was then collected from the recipient walls of the beaker and dried in an oven overnight at 120 o C to remove moisture and volatile components. The dried soot was stored in an air-tight bottle for future experiments.

Characterization of Catalysts
The textural characterization of the catalysts was carried out in a Micromeritics ASAP 2020 analyzer based on the low-temperature N 2 physisorption method. The phase compositions of the catalysts were identi ed by a powder X-ray diffraction diffractometer (XRD; Rigaku Ultima IV) under Cu-Kα1 radiation (λ

Activity measurement of the Catalysts
The performances of the as-prepared catalysts for diesel soot oxidation were evaluated in a compact xed bed tubular quartz reactor. The reactor consisted of two co-axial glass tubes with 20 mm and 50 mm diameters. A helical coil of quartz tube existing between the co-axial tubes served as a pre-heater of the air. A hole was present in the lower part of the outer tube to prevent breakage due to the expansion or contraction of air between the co-axial tubes. The pre-heated air entered the catalyst bed and owed to the inner tube. The product stream from the bottom of the reactor was cooled to ambient temperature in a condenser and then analyzed by online gas chromatography.
The reactor was mounted vertically in a split open furnace. The down air ow stream was used to avoid the distortion of the catalyst bed. The soot catalyst (catalyst bed diameter and height were 20 mm and 1.27 mm, respectively) was placed on a thin layer of quartz wool, which was supported on a perforated quartz disc inside the inner tube. A thermocouple well consisting of a 4 mm diameter tube was inserted axially from the bottom to the center of the disc for temperature measurement and control. In order to evaluate the catalytic activities of the catalysts, 110 mg of each catalyst-soot mixture was placed in the reactor and then oxidized till the total conversion of the soot at a constant heating rate of 1 o C.min -1 .
Before the reactions, each soot-catalyst mixture with a weight ratio of 1:10 was milled in an agate mortar for tight contact and with a spatula for loose contact. The detailed method of soot conversion calculations by gas chromatography is explained in the literature (Mishra and Prasad, 2015). Tight contact condition was used to explore the intrinsic activity of the catalysts because soot and catalysts were ground together to produce more contact points while loose contact mode was used to simulate real soot-catalyst contact inside catalytic converter (Bueno-López et al. 2007).

Textural characterization of catalysts by N 2 -sorption
The textural properties of the ceria nano bers (BET surface area, total pore volume, and average pore diameter) are summarized in Table 1. It is noticeable that the ceria nano bers had a high speci c surface area of 80-102 m 2 .g -1 and an average pore diameter of 108-130 Å. In comparison to pure ceria, the surface area of the Cu-supported Ceria nano bers increased gradually when 10 wt%, 20 wt%, and 40 wt% of copper oxide were added. However, a drop in the surface area was observed at the highest copper oxide loading content of 60 wt%. The incorporation of 10-40wt% of CuO favored the increase in porosity, causing an increase in the surface area from 80 m 2 .g -1 to 102 m 2 .g -1 . In contrast, the incorporation of 60 wt% CuO decreased the porosity and widened the pore distribution, causing the blocking of pores and a drastic decrease of the surface area to 93 m 2 .g -1 .

XRD Analysis of the Catalysts
The XRD patterns of the CeO 2 and CuO/CeO 2 nano bers with different copper loadings (20-60 mol%) are illustrated in Fig. 1. The detected characteristic peaks correspond to the crystalline planes of pure cubic cerium oxide (ceria uorite structure; JCPDS 34-0394). The small difference in the widths of these ceria peaks indicates different ceria domain sizes. Moreover, in the Cu-supported catalysts, the re ections of CuO (copper oxide; JCPDS: 45-0937) were prominent at 33.8º and 56.67 o . The most intense re ection of CeO 2 was detected at 2θ = 28.83 o . The average crystallite size of CeO 2 (14 nm) and CuO/CeO 2 (9-12 nm) were calculated by the Debye-Scherer formula ( Table 1). The intensity of the CuO peaks increased with the rising copper loading in the catalysts; however, no signi cant difference was observed between the pure and Cu-loaded ceria nano ber catalysts. It is already reported that Cu impregnation marginally in uences the lattice parameters of ceria. When the uorite-like lattice of ceria was doped with divalent Cu2þ cations, the small ionic radii of both Cu and Cu2þ caused a decrease in lattice parameters; however, the increased vacancies and lattice distortion led to an increase in lattice parameters (Si et al. 2012).

FTIR Characterization
The FTIR spectra of the CeO 2 and CuO/CeO 2 nano bers are displayed in Fig. 2. Broad bands at 3700- to Ce-O bonds. In the present experiment, the bands around 3440 cm -1 , 1630 cm -1 , 1385 cm -1 , and several stretches below 1200 cm -1 con rm that the samples contained copper acetate and cerium nitrate (used as precursor).

SEM Characterization
The SEM analysis con rmed the formation of cylindrical nano bers with varying diameters. Figure 3 exhibits the SEM images of the pure and Cu-supported ceria nano bers with varying copper loadings (0-60 mol%). The average diameters of the ceria nano bers containing 20 mol%, 40 mol%, and 60 mol% of Cu were measured as 117 nm, 108 nm, and 115 nm, respectively. Tang  Cu x+ Ce due to oxidative/reductive conditions. Therefore, the XPS analysis of Cu must be considered in conjunction with other characterization methods to elucidate the nature of Cu ions in these catalysts (Gibbons et al. 2014). The rise in the relative strength can be ascribed to the incorporation of CuO into the ceria lattice during oxidation at high temperatures.

Soot Oxidation over Ceria nano bers
A soot oxidation reaction is a typical heterogeneous catalytic reaction in which solid particles act as reactants. The catalytic activities of the nano bers were measured at the temperatures at which 0%, 50%, and 100% of weight loss were detected (denoted as T o , T 50 , and T f , respectively). The soot conversion percentages for different catalysts under the loose and tight contacts are presented in Fig. 5. The T o , T 50 , and T f values of the catalysts are reported in Table 2. The T o , T 50 , and T f values of the 40% Cu-CeO 2 catalyst were measured as 258°C, 305°C, and 370°C, which are lower than those of the pure and Cusupported CeO 2 catalysts (Table 2 and  catalyst was found as 267 o C, which is much lower than those of other catalysts (above 400 o C). Therefore, the 40% Cu-CeO 2 catalyst possessed an obvious advantage with a much lower T 50 value under the tight contact. The as-prepared Cu-supported ceria nano bers with different Cu percentages exhibited different activities for soot combustion under the loose and tight contacts. The oxidation of diesel soot is directly dependant on three factors. The rst one is the reducibility of the transition metal cations in order to produce oxygen vacancies. These vacancies are essential to allow the soot oxidation. The second factor is the amount of -oxygen available to initiate the reaction of soot oxidation into CO 2 . The third factor is crystallite size and surface area.The 40% Cu-CeO 2 catalyst has high speci c surface area and consecutively increased activity of the catalyst. The observation is obvious as the catalysis is a surface phenomenon, higher the surface contacts (tight contact) higher the activity. Other side is that oxygen vacancy order to disorder transition occurs when temperature increases which can decrease the mobility of oxygen species in the bulk and result in the decreasing soot oxidation over 60% Cu-CeO 2 . In order to verify the reproducibility of the experimental data, each experiment was performed at least twice for soot oxidation, and the data were found to be reproducible with ± 1% of deviation.   The catalyst-soot ratio and gas hourly space velocity (GHSV) for soot oxidation on the 40% Cu-CeO 2 catalyst were further optimized. The activity orders at different catalyst-soot ratios were found in the following descending sequence 10:1 > 15:1 > 20:1 > 5:1 ( Fig. 6 and Table 3 (a)). The increasing GHSV from 7500 h − 1 to 22500 h − 1 caused a decrease in the characteristic temperatures (T i , T 50 , T f ); however, the further rise in the ow rate to 30000 h − 1 increased the characteristic temperatures (Fig. 7). This phenomenon explains two counter effects of the increasing GHSV on soot oxidation. The increasing air ow rate generated more oxidants (O 2 ) and improved the combustion of soot particles. Further, the increase in the air ow rate shortened the contact time of the reactants and negatively affected soot oxidation. These two counter effects resulted in the optimum GHSV value of 22500 h − 1 at the lowest characteristic temperatures of T i = 201 o C, T 50 = 246 o C, and T f = 318 o C.

Conclusions
Four catalysts, pure CeO 2 and Cu-supported CeO 2 nano bers containing 20 mol%, 40 mol%, and 60 mol% Cu were successfully fabricated by electrospinning. The average diameter of the pure CeO 2 nano bers was larger than that of the Cu-loaded CeO 2 ones due to the reduction of the gel solution viscosity after the addition of copper acetate monohydrate. Moreover, the average diameter of the nano bers decreased by about 20-30% after calcination. The optimum catalyst composition for diesel soot oxidation after calcination in the air was 40% Cu-CeO 2 (T f = 318 o C and 370 o C for tight and loose contacts, respectively).
The high catalytic activity of the 40% Cu-CeO 2 nano bers could be elucidated from the following aspects: rst, the nanoscale size of both the catalyst and soot particulates ensured a high contact e ciency between them, and second, the synergistic effect of Cu and CeO 2 increased the oxidation ability of the catalyst. Figure 2 FTIR spectra of nano bers pure and Cu loaded.    Characteristic light off temperature for soot oxidation over 40% Cu-CeO2 (with different Catalyst/soot ratio) Figure 7 Effect of Gas Hourly space Velocity on soot oxidation, by 40% Cu-CeO2, (Catalyst/Soot: 10/1)

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. graphicalabstract.pptx