Nanostructure and reactivity of soot from biofuel 2,5-dimethylfuran pyrolysis with CO₂ additions

Abstract

This paper investigated the nanostructure and oxidation reactivity of soot generated from biofuel 2,5-dimethylfuran pyrolysis with different CO₂ additions and different temperatures in a quartz tube flow reactor. The morphology and nanostructure of soot samples were characterized by a low and a high resolution transmission electron spectroscopy (TEM and HRTEM) and an X-ray diffraction (XRD). The oxidation reactivity of these samples was explored by a thermogravimetric analyzer (TGA). Different soot samples were collected in the tail of the tube. With the increase of temperature, the soot showed a smaller mean particle diameter, a longer fringe length, and a lower fringe tortuosity, as well as a higher degree of graphization. However, the variation of soot nanostructures resulting from different CO₂ additions was not linear. Compared with 0%, 50%, and 100% CO₂ additions at one fixed temperature, the soot collected from the 10% CO₂ addition has the highest degree of graphization and crystallization. At three temperatures of 1173 K, 1223 K, and 1273 K, the mean values of fringe length distribution displayed a ranking of 10% CO₂>100% CO₂>50% CO₂ while the mean particle diameters showed the same order. Furthermore, the oxidation reactivity of different soot samples decreased in the ranking of 50% CO₂ addition>100% CO₂ addition>10% CO₂ addition, which was equal to the ranking of mean values of fringe tortuosity distribution. The result further confirmed the close relationship between soot nanostructure and oxidation reactivity.

Graphical abstract

Keywords

2 / 5-dimethylfuran pyrolysis / soot / CO₂ addition / nanostructure / reactivity

1 Introduction

World energy consumption is increasing as a result of human’s rising desire for higher living conditions and the ever-growing world population. Most of the energy is supplied by fossil fuels. However, the burning of these kinds of fuels has led to many various environment problems including global warming and climate change [1,2]. Soot particles, which were liberated from the incomplete combustion of fossil fuels, are highly carbonization materials and are regarded as the second largest contributor to global warming after carbon dioxide [3]. Furthermore, the existence of these ultrafine particles can cause great harm to human’s health [4]. All of these motivate investigators to dig out new methods to decrease soot emission.

With the development of oxy-fuel combustion, capture and storage of CO₂ become easier, thus the severe influence of greenhouse effect can be controlled to some degree [5,6]. Moreover, exhaust gas recirculation (EGR) is widely used to decrease NO₂ emission in some modern engines. EGR can lead to a low combustion temperature, which may result in an increase of soot formation. Many researchers began to conduct corresponding studies to explore the influence of CO₂ on soot emission, hoping to get some new ways to decrease emissions [7–11]. Abián et al. [7,8] investigated the effect of different CO₂ concentrations on soot and gas products from ethylene thermal decomposition and soot reactivity in conventional combustion and oxy-fuel combustion environments. They found that a 25% CO₂ addition could promote the formation of soot. However, a 78.5% CO₂ addition could result in a decrease in the yield of soot, compared with the pyrolysis of pure ethylene in N₂ atmosphere. For soot reactivity, they concluded that the presence of CO₂ only affected the soot reactivity at the highest formation temperature (1475 K) and CO₂ concentration (78.5%). In terms of combustion, Ying et al. [9] examined the nanostructure evolution and reactivity of nascent soot from n-butanol-doped inverse diffusion flames in CO₂, N₂, and He atmospheres. They discovered that CO₂ could inhibit soot formation from the inception process. In addition to experimental research of the influence of CO₂ on soot production, modeling studies of the effects of CO₂ addition on soot and gas products were widely performed [10,11]. Besides, considering the high quantities of emissions of waste and other similar materials, extensive research of the pyrolysis or combustion of coal char, waste textile, and cemented radioactive waste-forms in CO₂ atmosphere were carried out to make full use of these materials, finish the conversion of CO₂, and reduce the greenhouse effect [12–14]. Studies of the effect of CO₂ on soot emission and gas products discharge have remained hot till now.

Furan family fuels including 2,5-Dimethylfunan (DMF) have aroused more and more attention because of the development of the second generation synthesize biofuel technology. Through the large-scale industrial synthesis production of DMF from starch, cellulose, glucose and fructose, the problem of global energy shortage and environmental pollution could be relieved to some extent [15,16]. An increasing number of researchers have investigated the pyrolysis and combustion characteristics of pure DMF, DMF/gasoline mixture and DMF/diesel blends [17–25]. Studies of the effects of DMF as an addition fuel on soot nanostructure and reactivity were performed to check whether DMF could become a kind of alternative fuel to reduce the dependence on fossil fuels and decrease emissions [17–22]. Jiang et al. [17] explored the nanostructure and oxidation reactivity of the nascent soot produced in n-heptane/2,5-dimethylfuran (DMF) inverse diffusion flames (IDF) with/without the influence of magnetic fields. The results showed that the additions of DMF-doped could enhance soot production in IDF. With DMF blends, typical core-shell structures with well-organized fringes were presented. Jia et al. [18] studied the nanostructure and oxidation reactivity of soot particles formed from DMF/n-heptane non-swirling and swirling flames. The typical core-shell structure was also discovered from the soot collected in the 50% n-heptane/50% DMF flame. But from pure n-heptane and 80% n-heptane/20% DMF flames, younger soot with partly graphitic and partly amorphous structures was found. Gogoi et al. [19] explored the effects of 2,5-dimethylfuran addition to diesel on soot nanostructures and reactivity. The soot emission was found to decrease with increasing amount of DMF in the DMF/diesel blends while the oxidation reactivity of soot particles increased. Zhang et al. [20,21] studied the combustion and emissions of 2,5-dimethylfuran addition in a diesel engine. Ma et al. [22] investigated the combustion characteristics and emissions of gasoline-DMF blends in a gasoline internal combustion engine. Furthermore, researches of the combustion and pyrolysis of pure DMF were also widely conducted [23–25]. Cheng et al. [23] carried out the experimental and kinetic modeling study of 2,5-dimethylfuran pyrolysis at different pressures. The main pathways in the decomposition of DMF were drawn and the growth of aromatics was determined through the rate of production and sensitivity analyses. The results demonstrated that the formation of aromatics in the pyrolysis of DMF was promoted compared with the pyrolysis of cyclohexane and methylcyclohexane under very close conditions. This discovery hinted the potentially high sooting tendency of DMF. Alexandrino et al. [24] studied the influence of temperature and concentration of 2,5-Dimethylfuran on its sooting tendency. With the increase of temperature or inlet 2,5-DMF concentration, the soot yield was greater. Due to their tendency to form cyclopentadienyl radicals, which played an important role as precursors in the growth processes to larger and larger PAH, without going through benzene as intermediate, 2,5-DMF was thought to have a high sooting tendency. However, the research highlights the different gas products of different reaction processes. However, as for the relationship between nanostructure and reactivity of soot particles collected from different reaction condition, very little literature could be found. In recent years, many researchers performed their own investigations by paying great attention to soot nanostructure and oxidation reactivity in order to further understand soot formation process and invent efficient technologies for soot reduction [26–31]. Liu et al. [26] studied the nanostructure and reactivity of carbon particles (soot) from co-pyrolysis of biodiesel surrogate methyl octanoate blended with n-butanol at temperatures from 1023 K to 1223 K in a quartz tube flow reactor. It was found that with the rise of n-butanol addition, the soot showed a higher graphitization degree and a lower oxidation reactivity. Ying et al. [27,28] discovered that the soot which presented the lowest degree of crystallization with the shortest fringe length and largest fringe tortuosity could have the highest oxidation reactivity. Paladpokkrong et al. [29] studied the nanostructure and reactivity of soot particles in ethylene/DMC normal and inverse diffusion flames. It was found that soot particles obtained with DMC addition had more disordered layers and a higher oxidation reactivity. Luo et al. [30] carried out an experimental study of laboratory jet impinging ethylene diffusion flames focusing on the characteristics of soot forming on the impinging plate. The results also uncovered the relationship between the soot structure and oxidation reactivity that the soot with the lower degree of crystallization was easier to oxidize. Yehliu et al. [31] found that soot nanostructure disorder correlates with a faster oxidation rate through quantitative analysis of HRTEM images and the results from a thermogramatric analyzer. Pyrolysis is one of the pretty important steps in the combustion process. Studies focusing on the influence of various fuels and temperatures under different pyrolysis conditions on soot production were conducted [32–35]. However, the nanostructure and reactivity of the soot collected from DMF pyrolysis under CO₂ atmosphere are rare.

Therefore, the aim of this paper was to investigate the nanostructure and reactivity of soot collected from pure DMF pyrolysis of different CO₂ additions at three temperatures (1073 K, 1123 K, and 1173 K) in the laboratory scale, hoping to provide more data and information for the practical use of the new biomass fuel. The ratio of CO₂ in the reactant gas in different pyrolysis conditions was kept to be 0%, 10%, 50%, and 100% separately. A low transmission electron microscope (TEM) and a high transmission electron microscope (HRTEM) were employed to obtain the morphology and nanostructure of the soot samples. An X-ray diffractometer (XRD) was used to measure the graphization degree of the soot samples. A thermogravimetric analyzer (TGA) was applied to obtain the oxidation reactivity of the soot samples. This paper could provide information of the influences of CO₂ on soot characteristics through correlating soot nanostructure with its oxidation reactivity.

2 Experimental and analysis method

2.1 Experimental device

The DMF pyrolysis experiments were conducted at different reaction temperatures and different mixing fractions of CO₂ in the Ar atmosphere. The concrete physical properties and related information of the substances used in the experiments were listed in Table 1 while different pyrolysis conditions were shown in Table 2. The total flow rate of 1000 mL/min (STP) was fed into the flow reactor. The amount of the carrier Ar used to carry the fuel gas acquired in the evaporator was kept to be 2% of the total gas in all the pyrolysis conditions. The amount of the fuel gas was also kept to be 2%. The ratio of CO₂ in the remaining 96% reactant gas (960 mL/min) varied in different pyrolysis conditions. For each CO₂ addition, pyrolysis reactions took place at three different temperatures, 1173 K, 1223 K, and 1273 K. Considering the different conversions of DMF at different temperatures [23,24] and the aim of the present work to get the soot characterization, these three temperatures were selected.

A flow reactor which had a quartz tube of 50 mm in internal diameter and 700 mm in length was used to run the pyrolysis reaction. The detailed experimental system was displayed in Fig. 1. The flow reactor was placed inside an electric furnace to offer sufficient power to maintain the required temperature. To make sure that the inlet of the flow reactor was totally in gas phase, the whole pipelines were wrapped in heating band. Temperature controllers were used to maintain the temperature of the pipe at 120°C. DMF pumped by a Harvard PHD2000 springe pump was fed into the flow reactor by an individual line using Ar as the carrier gas. The Ar and CO₂ flow rates were controlled by mass flow controllers. The concrete parameters of these devices were shown in Electronic Supplementary Materials.

The outlet of the reactor had the soot collection system in the tail of the quartz tube including a filtration membrane and some quartz fiber cotton in each pyrolysis conditions. The membrane had a diameter of 50 mm and an aperture of 0.45 mm to collect soot. The temperature of the filtration membrane at the outlet of the tube was pretty low, which was about 350 K during the running of the flow reactor. The quartz fiber cotton with a pore light of lower than 5 mm was used to prevent uncollected soot from blocking the back pipeline. The mass of each membrane was weighed by using a high precision balance before putting into the quartz tube. The same operation was done after getting the membrane out from the quartz tube when the pyrolysis reaction finished. The soot amount could be acquired by the difference. The experiment of each condition ran for 1 h. There is a screen on the flow reactor, which showed the temperature of the point where the S type thermocouple was placed in the flow reactor. By operating the buttons, the reactant temperature like 1173 K was set. The temperature profiles along the center of the quartz tube were measured by moving a K-type thermocouple from the beginning position, as shown in Fig. 2.

2.2 Soot characterization methods

To get the impact of different temperatures and CO₂ additions on soot morphology and nanostructure, an FTI Tecnai G2 F30 S-Twin transmission electron microscope operated at an accelerating voltage of 300 kV with a 0.17 HR STEM resolution and a 0.10 nm line resolution was used. The samples were ultrasonically shocked in ethanol for 120 min in order to make soot samples disperse evenly. Then one or two drops of the suspension liquid were dropped on the carbon film coated lacy grid (200-mesh). Nanostructure characteristics including fringe length and fringe tortuosity were accessed by the processing of HRTEM images using a homemade MATLAB software [9,18,27,28,30] with the algorithms adopted from Refs. [36,37]. The detailed flowchart of the image processing procedure used to process HRTEM images can be found in Ref. [27]. The fringe length could reveal the physical extent of the carbon layer planes. The fringe tortuosity which is defined as the ratio of the real fringe length to the straight-line distance within carbon atom could show the degree of the fringe curvature.

The XRD can evaluate the structure and morphology of the atoms and molecules in the tested material A D8 Advanced X-ray diffractometer (Bruker, Karlsruhe, Germany) with Cu-Ka radiation (l = 0.15418 nm, 45 kV, 25 mA) was applied to get the graphitic degree of soot. The scan range was 10°–70° with a scan step size of 0.05 and a scan speed of 0.2 s/step.

To study the oxidation reactivity of the soot samples under different pyrolysis conditions, an STA 449 F3 Jupiter thermogravimetric analyzer from NETZSCH with the recording software was employed. Each soot samples were weighed 10±0.5 mg on the high precision balance and then spread uniformly in the quartz crucible. First, the soot sample was heated up from 30°C to 300°C at a rate of 10°C /min in the Ar atmosphere (100 mL/min). Then, the constant temperature (300°C) was maintained for 60 min to remove the volatile compounds. After that, the soot was continually heated from 300°C to 500°C at the same heating rate as before. Finally, the reactant gas was changed to an oxygen mixture flow, which contained 78% of Ar and 22% of O₂. The total flow rate was held at 100 mL/min. The oxidation reactivity of soot was evaluated through the isothermal oxidation at 500°C. The constant temperature of 500°C was kept for 120 min. The mass loss of each sample in the oxidation period was normalized. Repeated experiments were conducted to make the TGA results reliable.

3 Results and discussion

3.1 Morphology analysis by TEM

Figure 3 illustrates representative TEM images of various soot particles collected from different pyrolysis conditions. For each sample, three to four locations were chosen to magnify and get the TEM pictures. It was found that the morphology was similar for the same sample. For the selection of the representative images, the pictures which owned the single-layer structure were chosen because this kind of TEM pictures could make it convenient to select the locations to get the HRTEM pictures. To facilitate the understanding, the horizontal axis referred to the temperature while the vertical axis referred to the fraction of CO₂ addition. The number 1, 2, and 3 symbolized the temperature of 1173 K, 1223 K, and 1273 K, respectively while the alphabet A, B, C, and D represented the fraction of CO₂ addition of 0%, 10%, 50%, and 100%, respectively.

At a fixed addition of CO₂, the sizes of soot particles tended to be smaller with the increase of temperature. The soot collected at 1173 K with a 0% CO₂ addition presented a film-like material, which may develop from the condensation of heavy PAHs. It was difficult to identify individual particles. This liquid-like deposition with an irregular shape was the same as that in previous researches [38–40]. Some distinct particles could be noticed at 1223 K. The particles were tinier at 1273 K compared with those collected at 1223 K and 1173 K. As a result, soot could break into small particles with temperature increasing.

At 1173 K, the soot collected with 10%, 50%, and 100% CO₂ additions showed particle-like protrusions. Especially with the 10% CO₂ addition and the 100% CO₂ addition, the monomers or approximate spherical particles could be observed transparently. Compared to the soot collected with the 10% and 100% CO₂ additions at 1173 K, the soot collected with the 50% CO₂ addition at 1173 K showed a higher amount of liquid-like material. At 1223 K, some individual particles could be found under the coverage of a film-like material. Besides, the coverage area of the liquid-like material of the soot collected with 10% and 100% CO₂ additions was smaller than that of the soot collected with 0% and 50% CO₂ additions at 1223 K. For higher temperatures, the soot displayed chain-like or tufted aggregates. The different replacement ratios of AR by CO₂ at the same temperature may cause different morphology of soot samples because of the thermal effects, dilution effects, and chemical effects [7,10]. Considering the similar molecular weight and resemble density of AR and CO₂ under the standard condition, the dilution effects of CO₂ may be little. Moreover, on the one hand, the existence of O may take part in the chemical reaction and decrease the reactions related to soot formation. Furthermore, the soot particles may be oxidized by the O atom. On the other hand, the specific heat capacity of CO₂ is larger than that of AR, resulting in the enhancement of the thermal effect, which may promote soot production to some degree. The competitions of the two kinds of effects may lead to different results.

Some TEM pictures of soot particles were processed by Image J to acquire the quantitative information of the average sizes of particles. The mean diameters of the soot particles were exhibited in Fig. 4, the representative particle size distribution of the soot under different pyrolysis conditions were presented in Electronic Supplementary Material, while some parameters of the morphology of the agglomerate in the TEM pictures [41–43] such as the aggregate projected maximum length (L), the radius of gyration (R_g), the aggregate projected area (A_a), and the number of particles per agglomerate (N) were tabulated in Electronic Supplementary Material. It was a little difficult to measure the diameters of the particles because the particles were bundled together and it was hard to distinguish the clear boundary. Therefore, the mean diameter of the particles may have been overestimated, as mentioned in Ref. [27]. It could be found that with the increase of temperature, the mean particles size decreased. At 1273 K, soot particles may have stopped growth. During this period, the crack between particles took the lead, which resulted in the decrease of the mean diameter of the particles. At one fixed temperature, the variation of the mean diameters of the soot particles was not linear with the increase of the ratio of CO₂ addition. It followed the order of 10% CO₂>0% CO₂>100% CO₂>50% CO₂. When the 50% CO₂ was introduced, the collected particles showed the smallest average diameter of the four values from the four CO₂ addition ratios. On the one hand, the chemical effect of CO₂ may benefit the oxidation of soot particles and limit the growth of soot particles, which takes the lead. On the other hand, the thermal effect of CO₂, which may absorb the heat and benefit the production of soot, was not as remarkable as the pyrolysis condition when the 100% CO₂ was added. The average particle diameters of soot particles collected from 0%, 10%, and 100% CO₂ additions were similar. When the 10% CO₂ was added, the O atom may take part in the oxidation of the soot particles. The thermal effect of CO₂ may play a more important role when the 100% CO₂ was injected.

3.2 Nanostructure analysis by HRTEM

The images of soot particles with a higher magnification were obtained to get the information about soot nanostructures in different pyrolysis conditions, as displayed in Fig. 5. Figure 6 illustrates the extracted skeleton images from Fig. 5. In Fig. 6, the soot collected at 1173 K had highly curved, short, and disorganized lamellae compared to those produced at 1223 K and 1273 K. Apparent planar lamellae collected at 1223 K could be seen. A typical core-shell structure could be found in soot at 1273 K, as exhibited in A3, B3, C3, and D3. The outer shell consisted of extended, disconnected, and planar crystallites which were placed in a concentric way and were mostly parallel to each other. The orientations of fringes influenced the crystalline and amorphous structure of soot particles. The parallel fringes stood for the graphitization part, while the curved fringes symbolized the existence of PAHs. The long and parallel graphene layers represented the crystallite structure of mature soot [18,44]. The above features revealed that soot nanostructures greatly differ at different pyrolysis temperatures. Generally, the higher degree of crystallization and graphitization corresponds to the lower reactivity [17,18,26–28]. It could be deduced that the soot collected at 1273 K may have a lower reactivity than that at 1173 K and 1223 K.

Longer and straighter fringes could be observed with 10% and 100% CO₂ additions compared to 0% and 50% CO₂ additions at 1173 K or 1223 K. However, the difference in soot nanostructure which resulted from different CO₂ additions was not so great by visual observations. At 1173 K, the soot collected with the 0% CO₂ addition showed amorphous structures with no crystallinity. All the fringes were short and disordered. With CO₂ addition, more organized and parallel fringes could be found. At 1223 K, the soot owned short radii of curved fringes and completely closed or partially closed outer shells could be noted, which implied the existence of fullerenic structure no matter CO₂ was added or not. However, the fullerenic structure was clearer when CO₂ was added. At 1273 K, the soot showed a representative core-shell structure with CO₂ additions.

Soot fringe length and tortuosity were extracted from the skeleton images for quantitative analysis, as shown in Figs. 7 and 8. The quantitative average values of all calculating parameters were given in Tables 3 and 4. For each sample, three skeleton pictures were calculated. Finally, the average value of the three mean values was considered to be the average value in Tables 3 and 4.

As shown in Fig. 7, the percent of fringe length, which was greater than 2 nm, was larger with the increase of the pyrolysis temperature. Besides, the average value of fringe length distribution exhibited an order of 1273 K>1223 K>1173 K. It could be deduced that as the temperature increased, the physical length of the carbon layer increased. Figure 8 showed that the fringe tortuosity distribution tended to be narrower with the increase of pyrolysis temperature. In addition, the average values of fringe tortuosity distribution exhibited an order of 1173 K>1223 K>1273 K. The larger fringe tortuosity of soot particles represented the higher curvature of the carbon layer. It could be discovered that with the rise in temperature, the carbon layer tended to be more tortuous. Moreover, the inner disorder degree of soot particles could change with the alteration of the fringe tortuosity of soot particles. As discovered in previous studies [9,17,18,26,27,30], the fringe length and tortuosity of soot particles were closely related to its chemical reactivity. The soot with the largest average fringe length value and the smallest mean fringe tortuosity value may have the lowest reactivity. It could be deduced that the soot collected at 1273 K may have a lower oxidation reactivity than the soot collected at 1173 K and 1223 K. There was no visual distinction in the histograms for fringe tortuosity since the four soot samples collected with different CO₂ additions had the similar broaden width range at one fixed temperature. The mean values of fringe length distribution had a ranking of 10% CO₂>100% CO₂>50% CO₂ at one fixed temperature. The average value of fringe length distribution of the 0% CO₂ addition was the smallest at 1173 K and 1223 K and the second smallest at 1273 K. The average values of the fringe tortuosity distributions were in the opposite order. Because of the competitions of the three kinds of effects of CO₂-thermal effects, dilution effects and chemical effects, the results may differ. The nonlinear relationship of CO₂ addition was also found by Abián et al. [7]. The results agreed with the results acquired from the HRTEM analysis and the TEM analysis.

3.3 XRD analysis

The XRD patterns of soot samples obtained under different pyrolysis conditions were illustrated in Fig. 9. The 002 band (2q ≈ 25°) stood for the existence of crystalline graphitic carbon, from which the graphitization degree of soot samples could be obtained. Therefore, the concrete values about the diffraction angles of the soot particles were summarized from the XRD spectra in Fig. 9, as exhibited in Table 5. For a constant CO₂ addition, the diffraction peaks transformed and approached to the 002 band (2q ≈ 25°) with the increase of temperature, revealing an enhancement in the graphitic character and the ordering of soot structure and thus the lower reactivity. For a constant temperature, the peak diffraction angle of the soot with the 10% CO₂ addition was nearest to the 002 band, which was followed by the soot with the 100% CO₂ addition, leading to a higher graphitization degree than the soot with the 0% CO₂ addition and the 50% CO₂ addition. The peak diffraction angle of the soot with the 0% CO₂ addition was farthest to the 002 band except at 1273 K, representing the lowest degree of graphitization. However, at 1273 K, the peak diffraction angle of the soot with the 0% CO₂ addition was quite close to that of soot with the 50% CO₂ addition. The results were consistent with the morphology analysis of TEM, the nanostructure analysis of HRTEM and fringe parameters including fringe length and tortuosity.

3.4 Isothermal oxidation analysis

The oxidation reactivity is one of the most important properties of soot particles, which reflects the reactivity in the oxidation process and is affected by the aggregation morphology and nanostructure of soot samples.

Figure 10 showed the normalized mass loss curves during the oxidation process. The horizontal coordinate stood for the reaction time from the beginning of the oxidation and the longitude coordinate represented the percentage of the remaining soot mass.

Figures 10(a)–10(c) illustrated the influences of different CO₂ additions on oxidation reactivity at three constant temperatures. At 1173 K, 1223 K, and 1273 K, the oxidation reactivity of different soot samples decreased in the ranking of 50% CO₂ addition>100% CO₂ addition>10% CO₂ addition. The oxidation reactivity showed a nonlinear variation. With the 50% CO₂ addition, the soot samples had the highest oxidation reactivity. Due to the co-existence of the chemical effects, the dilution effects, and the thermal effects of CO₂, the competition of the effects could influence the oxidation reactivity of soot particles. The results were consistent with the HRTEM analysis and XRD results. As exhibited in the previous HRTEM analysis, the mean values of fringe length distribution displayed a ranking of 10% CO₂>100% CO₂>50% CO₂ at one fixed temperature. The soot collected with the 10% CO₂ addition had the longest fringe length and smallest fringe tortuosity compared with the soot collected with 0%, 50%, and 100% CO₂ additions at the fixed temperature. Moreover, the soot collected with the 10% CO₂ addition had the largest value of the peak diffraction angles compared to the soot collected with 0%, 50%, 100% CO₂ additions at the fixed temperature, as shown in the XRD analysis. It implicated that the soot with the smallest fringe length and largest fringe tortuosity had the lowest graphization degree and the highest oxidation reactivity. Not enough soot could be collected for the TGA analysis with the 0% CO₂ addition at 1173 K. At a lower temperature, the conversion of DMF was low. With the increase of temperature, the soot amount will increase. Therefore, at 1223 K and 1273 K, more than 10 mg soot was collected to conduct the test. At 1173 K with the 0% CO₂ addition, less than 10 mg soot was collected. However, when CO₂ was introduced, the soot amount increased. CO₂ can affect the soot formation process by influencing the concentration of the OH radical poor via the reaction CO₂ + H ⇋ CO+ OH. The OH radicals formed can deplete molecular hydrogen, which may enhance the PAH formation and then increase the soot amount. The soot sample collected with the 0% CO₂ addition at 1223 K had a higher oxidation reactivity than that with the 10%, 50%, and 100% CO₂ addition at the same temperature.

Figures 10(d)–10(f) exhibited the effect of different temperatures on soot oxidation reactivity at three constant CO₂ additions. The curve of the soot formed at 1173 K dropped fastest while that of the soot formed at 1273 K decreased slowest. The oxidation reactivity followed the order of 1173 K>1223 K>1273 K, meaning that as the temperature decreased, the oxidation reactivity increased. When the temperature was higher, the nanostructure of the soot was more organized and the degree of soot graphitization was higher, resulting in the lower reactivity. These result agreed well with the HRTEM analysis and XRD results. A high correlation was found between the soot oxidation reactivity and nanostructure.

4 Conclusions

This paper concentrated on the nanostructure and reactivity of soot samples from DMF pyrolysis at different CO₂ addition and different temperatures. CO₂ could influence soot formation, nanostructure, and oxidation rate through thermal function, dilution effect, and complicated chemical reactions. The dominating conclusions obtained were summarized as follows.

The variations of soot nanostructure and reactivity resulted from the different additions of CO₂ were not linear. The soot with the 10% CO₂ addition had more ordered nanostructure, more mature structures, and a higher graphization than the soot with the 0%, 50%, and 100% CO₂ addition at 1173 K, 1223 K and 1273 K. The mean values of fringe length distribution displayed a ranking of 10% CO₂>100% CO₂>50% CO₂. Besides, the mean values of fringe tortuosity distribution displayed a ranking of 10% CO₂<100% CO₂<50% CO₂. Moreover, the oxidation reactivity of different soot samples decreased in the ranking of 50% CO₂ addition>100% CO₂ addition>10% CO₂ addition.

The soot yield at 1273 K had a typical onion-like structure, where the outer shell was formed by crystallites with oriented, extended, and parallel lamellae. The degree of graphization drastically increased with the rise of the pyrolysis temperature. The oxidation reactivity followed the order of 1173 K>1223 K>1273 K, meaning that as the temperature decreased, the oxidation reactivity increased.

A high correlation was found between nanostructure and oxidation reactivity of carbon materials. A longer fringe length, a lower fringe tortuosity, and a higher graphization lead to a lower reactivity.

References

[1] Bose B K. Global warming: energy, environmental pollution, and the impact of power electronics. IEEE Industrial Electronics Magazine, 2010, 4(1): 6–17

[2] Steinberg M. Fossil fuel decarbonization technology for mitigating global warming. International Journal of Hydrogen Energy, 1999, 24(8): 771–777

[3] Santos F, Fraser M P, Bird J A. Atmospheric black carbon deposition and characterization of biomass burning tracers in a northern temperate forest. Atmospheric Environment, 2014, 95: 383–390

[4] Jaramillo I C, Gaddam C K, Vander Wal R L, Huang C H, Levinthal J D, Lighty J A S. Soot oxidation kinetics under pressurized conditions. Combustion and Flame, 2014, 161(11): 2951–2965

[5] Stanger R, Wall T, Spörl R, Paneru M, Grathwohl S, Weidmann M, Scheffknecht G, McDonald D, Myöhänen K, Ritvanen J, Rahiala S, Hyppänen T, Mletzko J, Kather A, Santos S. Oxyfuel combustion for CO₂, capture in power plants. International Journal of Greenhouse Gas Control, 2015, 40: 55–125

[6] Buhre B J P, Elliott L K, Sheng C D, Gupta R P, Wall T F. Oxy-fuel combustion technology for coal-fired power generation. Progress in Energy and Combustion Science, 2005, 31(4): 283–307

[7] Abián M, Millera A, Bilbao R, Alzueta M U. Experimental study on the effect of different CO₂, concentrations on soot and gas products from ethylene thermal decomposition. Fuel, 2012, 91(1): 307–312

[8] Abián M, Jensen A D, Glarborg P, Alzueta M U. Soot reactivity in conventional combustion and oxy-fuel combustion environments. Energy & Fuels, 2012, 26(8): 5337–5344

[9] Ying Y, Liu D. Nanostructure evolution and reactivity of nascent soot from inverse diffusion flames in CO₂, N₂, and He atmospheres. Carbon, 2018, 139: 172–180

[10] Liu D. Chemical effects of carbon dioxide addition on dimethyl ether and ethanol flames: a comparative study. Energy & Fuels, 2015, 29(5): 3385–3393

[11] Zhang Y, Wang L, Liu P, Guan B, Ni H, Huang Z, Lin H. Experimental and kinetic study of the effects of CO₂, and H₂O addition on PAH formation in laminar premixed C₂H₄/O₂/Ar flames. Combustion and Flame, 2018, 192: 439–451

[12] Zhou Y, Jin X, Jin Q. Numerical investigation on separate physicochemical effects of carbon dioxide on coal char combustion in O₂/CO₂ environments. Combustion and Flame, 2016, 167: 52–59

[13] Wen C, Wu Y, Chen X, Jiang G, Liu D. The pyrolysis and gasification performances of waste textile under carbon dioxide atmosphere. Journal of Thermal Analysis and Calorimetry, 2017, 128(1): 581–591

[14] Hartmann T, Paviet-Hartmann P, Rubin J B, Fitzsimmons M R, Sickafus K E. The effect of supercritical carbon dioxide treatment on the leachability and structure of cemented radioactive waste-forms. Waste Management (New York, N.Y.), 1999, 19(5): 355–361

[15] Qian Y, Zhu L, Wang Y, Lu X. Recent progress in the development of biofuel 2,5-dimethylfuran. Renewable & Sustainable Energy Reviews, 2015, 41: 633–646

[16] Chidambaram M, Bell A T. A two-step approach for the catalytic conversion of glucose to 2,5-dimethylfuran in ionic liquids. Green Chemistry, 2010, 12(7): 1253–1262

[17] Jiang B, Wang P, Ying Y, Luo M, Liu D. Nanoscale characteristics and reactivity of nascent soot from n-heptane/2,5-dimethylfuran inverse diffusion flames with/without magnetic fields. Energies, 2018, 11(7): 1698

[18] Jia P, Ying Y, Luo M, Jiang B, Liu D. Effects of swirling combustion on soot characteristics in 2,5-dimethylfuran/ n-heptane diffusion flames. Applied Thermal Engineering, 2018, 139: 11–24

[19] Gogoi B, Raj A, Alrefaai M M, Stephen S, Anjana T, Pillai V, Bojanampati S. Effects of 2,5-dimethylfuran addition to diesel on soot nanostructures and reactivity. Fuel, 2015, 159: 766–775

[20] Zhang Q, Chen G, Zheng Z, Liu H, Xu J, Yao M. Combustion and emissions of 2,5-dimethylfuran addition on a diesel engine with low temperature combustion. Fuel, 2013, 103: 730–735

[21] Zhang Q, Yao M, Zheng Z. Study on combustion and emission characteristics fueled with diesel, diesel-butanol and diesel-DMF blends. Chinese Internal Combustion Engine Engineering, 2014, 4: 45–50

[22] Ma Z, Shen H, Xu C. Experiment of combustion characteristics and emissions of gasoline-DMF blends. Journal of Zhejiang University, 2013, 47: 1965–1969

[23] Cheng Z, Xing L, Zeng M, Dong W, Zhang F, Qi F, Li Y. Experimental and kinetic modeling study of 2,5-dimethylfuran pyrolysis at various pressures. Combustion and Flame, 2014, 161(10): 2496–2511

[24] Alexandrino K, Salvo P, Millera Á, Bilbao R, Alzueta M U.Influence of the temperature and 2,5-dimethylfuran concentration on its sooting tendency. Combustion Science and Technology, 2016, 188(4-5): 651–666

[25] Somers K P, Simmie J M, Gillespie F, Conroy C, Black G, Metcalfe W K, Battin-Leclerc F, Dirrenberger P, Herbinet O, Glaude P A, Dagaut P, Togbé C, Yasunaga K, Fernandes R X, Lee C, Tripathi R, Curran H J. A comprehensive experimental and detailed chemical kinetic modelling study of 2,5-dimethylfuran pyrolysis and oxidation. Combustion and Flame, 2013, 160(11): 2291–2318

[26] Liu D, Wang W, Ying Y, Luo M. Nanostructure and reactivity of carbon particles from co-pyrolysis of biodiesel surrogate methyl octanoate blended with n-butanol. Fullerenes, Nanotubes, and Carbon Nanostructures, 2018, 26(5): 278–290

[27] Ying Y, Liu D. Effects of butanol isomers additions on soot nanostructure and reactivity in normal and inverse ethylene diffusion flames. Fuel, 2017, 205: 109–129

[28] Ying Y, Liu D. Effects of flame configuration and soot aging on soot nanostructure and reactivity in n-butanol-doped ethylene diffusion flames. Energy & Fuels, 2017, 32: 1–84

[29] Paladpokkrong C, Liu D, Ying Y, Wang W, Zhang R. Soot reduction by addition of dimethyl carbonate in normal and inverse ethylene diffusion flames: nanostructural evidence. Journal of Environmental Sciences (China), 2018, 72: 107–117

[30] Luo M, Ying Y, Liu D. Soot in flame-wall interactions: views from nanostructure and reactivity. Fuel, 2018, 212: 117–131

[31] Yehliu K, Vander Wal R L, Armas O, Boehman A L. Impact of fuel formulation on the nanostructure and reactivity of diesel soot. Combustion and Flame, 2012, 159(12): 3597–3606

[32] Esarte C, Abián M, Millera Á, Bilbao R, Alzueta M U. Gas and soot products formed in the pyrolysis of acetylene mixed with methanol, ethanol, isopropanol or n-butanol. Energy, 2012, 43(1): 37–46

[33] Sánchez N E, Millera Á, Bilbao R, Alzueta M U. Polycyclic aromatic hydrocarbons (PAH), soot and light gases formed in the pyrolysis of acetylene at different temperatures: effect of fuel concentration. Journal of Analytical and Applied Pyrolysis, 2013, 103: 126–133

[34] Ruiz M P, Callejas A, Millera A, Alzueta M U, Bilbao R. Soot formation from C₂H₂, and C₂H₄, pyrolysis at different temperatures. Journal of Analytical and Applied Pyrolysis, 2007, 79(1-2): 244–251

[35] Ruiz M P, de Villoria R G, Millera A, Alzueta M U, Bilbao R. Influence of the temperature on the properties of the soot formed from C₂H₂ pyrolysis. Chemical Engineering Journal, 2007, 127(1-3): 1–9

[36] Yehliu K, Vander Wal R L, Boehman A L. Development of an HRTEM image analysis method to quantify carbon nanostructure. Combustion and Flame, 2011, 158(9): 1837–1851

[37] Yehliu K, Vander Wal R L, Boehman A L. A comparison of soot nanostructure obtained using two high resolution transmission electron microscopy image analysis algorithms. Carbon, 2011, 49(13): 4256–4268

[38] Wang W, Liu D, Ying Y, Liu G, Wu Y. On the response of nascent soot nanostructure and oxidative reactivity to photoflash exposure. Energies, 2017, 10(7): 961

[39] Velásquez M, Mondragón F, Santamaría A. Chemical characterization of soot precursors and soot particles produced in hexane and diesel surrogates using an inverse diffusion flame burner. Fuel, 2013, 104: 681–690

[40] Blevins L G, Fletcher R A, Benner B A Jr, Steel E B, Mulholland G W.The existence of young soot in the exhaust of inverse diffusion flames. Proceedings of the Combustion Institute, 2002, 29(2): 2325–2333

[41] Bogarra M, Herreros J M, Tsolakis A, York A P E, Millington P J, Martos F J. Impact of exhaust gas fuel reforming and exhaust gas recirculation on particulate matter morphology in gasoline direct injection engine. Journal of Aerosol Science, 2017, 103: 1–14

[42] Brasil A M, Farias T L, Carvalho M G. A recipe for image characterization of fractal-like aggregates. Journal of Aerosol Science, 1999, 30(10): 1379–1389

[43] Ying Y Y, Liu D. Effects of water addition on soot properties in ethylene inverse diffusion flames. Fuel, 2019, 247: 187–197

[44] Xiao H, Hou B, Zeng P, Jiang A, Hou X, Liu J. Combustion and emission characteristics of diesel engine fueled with 2,5-dimethylfuran and diesel blends. Fuel, 2017, 192: 53–59

RIGHTS & PERMISSIONS

Higher Education Press and Springer-Verlag GmbH Germany, part of Springer Nature