Abstract
The present work reports on the influence of nickel molar loading and impregnation solutions, nitrate salts and copper and nickel ammonium complexes, on the synthesis of Ni-Cu catalyst supported on alumina for obtaining carbon nanofibers by catalytic chemical vapor deposition of methane. These catalysts were characterized by N2 adsorption-desorption, x-ray Diffraction, Thermo-Gravimetric Analysis and Scanning Electron Microscopy for the carbon nanostructures analysis. The catalytic activity was studied through the carbon deposition under different conditions of temperature (600 °C and 650 °C) and pressure (in ultra-low pressure conditions and 0.5 bar). As a main result, catalyst using nitrate salts impregnation leads to the major carbon nanofibers yielding at the temperature of 600 °C. On the other hand, catalysts synthesized by ammonium complexes impregnation present also improved characteristics compared to conventional catalysts that allows considering this methodology as an interesting alternative in carbon nanofibers producing.
Export citation and abstract BibTeX RIS
1. Introduction
The catalytic methane decomposition by means of Catalytic Chemical Vapor Deposition (CCVD) using a catalyst has been developed these last decades because it allows generating suitable products as COx-free hydrogen and carbon nanofibers or nanotubes [1]. Moreover, CCVD is considered as a simple and economically viable process for a large-scale production compared to arc-discharge or laser methods [2, 3]. However, different types of carbon responsible of catalyst deactivation are formed during the producing of carbon nanofilaments by catalytic decomposition of hydrocarbon [4, 5]. For example, according to Menon [6], type and location of carbon atoms are more significant than the nanofibers quantity produced when considering the catalytic activity of the catalyst. Besides, the catalyst deactivation is due to carbon encapsulation (amorphous carbon) where the active surface is covered by amorphous carbon while other carbon structures, as carbon nanofilaments, obstruct the active surfaces only when the carbon yield is high [6].
The carbon nanofilaments, including carbon nanofibers (CNF) and carbon nanotubes (CNT), are filaments with diameters between 1 and 200 nm and several micrometers of length that are formed by a bunch of graphene layers [5]. These materials are composed by carbon of graphitic nature [3]. Due to their textural properties, electrical and thermal conductivities, these materials are appropriate materials for mainly catalysts [7] and energy storage [8]. Among others applications, we found supercapacitors, nanoelectromechanical switches, removal of organic solvents from water, catalytic support and filtering in biological and health sciences [9].
The morphology and structure of carbon nanofilaments mainly depend on the type of catalyst, on the characteristics of the metal particles (size, morphology and composition), on the carbon source and reaction conditions. Catalysts composed of transition metals as nickel, iron, palladium, cobalt, molybdenum are efficient for producing CNF and CNT [1, 3]. Several studies have pointed out the fact that catalysts based on nickel element are the most suitable support for Thermo-Catalytic Decomposition (TCD) of methane compare to cobalt and iron elements [10]. This is due to the fact that nickel catalyst has a high capacity of active carbon site and a lower cost [11]. Temperatures higher than 600 °C proved to be very active for Ni/Al2O3. But, higher temperatures can destabilize the catalyst so a promoter, such as copper, must be added to ensure stability and durability avoiding its deactivation. It was reported that the use of copper as promoter increases the methane decomposition activity of nickel, improves nickel dispersion over the support, and enhances thermal stability of the catalyst [12].
The procedure of catalyst synthesis has a significant role in catalytic activity during the reaction of methane decomposition [13]. The impregnation method compared to sol-gel or co-precipitation is the most effective way to allow a suitable catalytic activity [13]. Even if the impregnation solution using nitrate salts has been widely studied, we investigate in the present research the use of copper and nickel ammonium complexes as impregnation solutions. Indeed, when the molecular complex size is bigger than the pores of the support, the impregnation can be limited to the external support surface during the catalyst synthesis [14]. In addition of covering the pores of the support, it also avoids the deactivation of the channels pores [14]. The materials constituting the substrate exhibit a large variety of chemical and surface properties that have been mainly studied as catalyst for reaction [15–17]. When using complex materials in catalyst synthesis, porosity could be induced and surface area increases. In addition, a scheme of sacrifice between metals and alloys can appear resulting from their differences in thermal stabilities [18]. However, the components having a low thermal stability are selectively eliminated through a thermal treatment that formed an ordered porous pattern [19].
The objective of the present work is to study the differences related to the characteristics of catalysts synthesized using the two impregnation solutions, nitrate salts called NixN where Ni denotes the nickel, x the atomic percent of Ni and N the nitrate salts and ammonium complexes referred NixC where C denotes the ammonium complexes. In the two situations, a small amount of copper (5 at%) is added as metal. By means of Thermo-Catalytic Decomposition of methane, the performance in carbon nanofibers deposition is evaluated using x-ray Diffraction (XRD), Thermo-Gravimetric Analysis (TGA) and Scanning Electron Microscopy (SEM) for analysis the catalyst and the structure of the carbon nanofibers.
2. Experiments and analysis
2.1. Catalyst preparation
Six Nickel-Copper catalysts over alumina were prepared at three different molar loadings: 50%Ni-5%Cu, 40%Ni-5%Cu and 30%Ni-5%Cu, using two different impregnation solutions: nitrate salts or ammonium complexes. The following notation NixNT or NixCT will be adopted, respectively, for each experiment as detailed below:
- •Ni describes the nickel over alumina for catalysts preparation.
- •x indicates the nickel molar percentage: 30%, 40% and 50%,
- •N or C inform nitrate salts or ammonium complexes impregnations, respectively.
- •T indicates the temperature of impregnation (600 and 650 °C).
Table 1 collects the composition of the different catalyst types.
Table 1. Designation and chemical composition of the catalysts.
| Catalyst designation | Mol relations (Ni:Cu/Al2O3) | Ni–Cu in wt% |
|---|---|---|
| Ni50Cu | 50:5/45 | 37.43 % Ni–4.05 % Cu |
| Ni40Cu | 40:5/55 | 28.28 % Ni–3.84 % Cu |
| Ni30Cu | 30:5/65 | 20.23 Ni–3.65 % Cu |
For the nitrate salts impregnation (NixN), three catalysts at different molar loadings (Ni30N, Ni40N and Ni50N) were prepared by successive impregnations at the same conditions of agitation, i.e. at 70 °C during 3 h, for which Cu (NO3)2.3 H2O P.A. and Ni(NO3)2.6 H2O P.A. salts were dissolved and diluted in distilled water. The first impregnation was carried out with the sample of Cu (NO3)2.3 H2O, and then dried at 105 °C for 5 h, milled and calcined at 350 °C for 3 h. Then, this sample is cooled inside the oven (closed oven) during one day up to room temperature, in order to perform the second impregnation with Ni(NO3)2.6 H2O. The resultant suspension was reserved for 24 h at room temperature, dried during 6 h at 105 °C, milled and calcined at 450 °C for 4 h.
For the ammonium complexes impregnation (NixC), impregnation solutions of copper and nickel ammonium complexes were used as precursors for the three catalyst preparations at different molar loadings (Ni30C, Ni40C and Ni50C). The complexes were obtained by means of a metal dissolution mixture taking into account the appropriated amount of ammonia for obtaining a stoichiometric quantity. The formation of [Ni(NH3)4]2+ and [Cu(NH3)4]2+ complexes with an addition of 25% of ammonia is expected. This mixture was reserved for 24 h and diluted in distilled water in order to obtain a homogeneous solution. The first impregnation was carried out with the solution of copper ammonium complex under constant agitation during 3 h at room temperature. Afterwards, the second impregnation consists of adding the solution of nickel ammonium complex under constant agitation for 3 h with a magnetic stirrer. The resulting suspension was reserved for 24 h at room temperature. The water and ammonia in excess were separated using a Büchi R-100 rotary evaporator during 6 h at 105 °C up to evacuate 90% of liquid. The sample was then dried at 105 °C for 4 h or the time required to reach a constant weight. Finally, the sample was then milled and calcined at 350 °C during 2 h followed by 450 °C during 5 h.
2.2. Carbon nanofiber deposition
Carbon nanofibers deposition was carried out from catalytic decomposition of methane in a quartz reactor using a SK2C-5TPB horizontal furnace. The following parameters: temperature, catalyst nature (according to impregnation solution), molar loading of catalyst, growth time and pressure were considered in the experiments. Previously, the experiments with Ni50N and Ni50C catalysts were performed for evaluating the carbon nanofibers deposition, as a function of temperature (600 °C and 650 °C), time (3 and 5 h) and for determining the optimum conditions for producing the higher quantity of carbon deposited. In addition, the experiments were performed under barometric pressures at in ultra-low pressure conditions (ULPC) and 0.5 bar. Prior to the carbon deposition experiments, the catalysts were activated with H2 with 100 ml min−1 of flow rate under N2 atmosphere using 150 ml min−1 of flow rate during 2 h at 500 °C for obtaining a metal catalyst (Ni-Cu/Al2O3). When the metal catalyst was obtained, the growing tests related to the carbon deposition were performed with a flow rate of 83 ml min−1 of CH4 and 55 ml min−1 of N2.
The ratio used was 1.5: 1 (CH4: N2) [13], the proposed flow rates are within that reported by Suelves et al [20]. Finally, all the tests were performed by using 0.2 g of catalyst. Table 2 collects all the information related to the different tests performed at different loadings of catalyst, natures of impregnation solutions and temperatures. The total amount of solid carbon was measured by using the following equation [21].
Table 2. Details of the experiments of carbon nanofibers deposition, growth time, temperature and pressure.
| Ni50-Catalyst | Ni (At%) | Time (h) | T (°C) | Pressure (Bar) |
|---|---|---|---|---|
| N | 30 | 5 | 600 and 650 | ULPC |
| 40 | ||||
| 50 | 3 and 5 | |||
| 50 | 600 | 0.5 | ||
| C | 30 | 5 | 600 and 650 | ULPC |
| 40 | ||||
| 50 | 3 and 5 | |||
| 50 | 600 | 0.5 |
In equation (1), wtot and wcat represents the total and initial weight of the catalyst after and before reaction, respectively. This yield was also expressed in the form of gdep/gNi.
2.3. Physical and structural catalysts and carbon nanofibers characterization
These different catalysts and carbon nanofibers were characterized using:
- •Specific Surface Area (SBET) technique is applied in order to compare the variation of surface area when catalysts are synthesized using different impregnates solutions and nickel loadings. The specific surface area for each catalyst was determined by means of Brunauer, Emmet and Teller (BET) method and the pore volume by Barrett, Joyner and Halenda (BHJ) method, using a Micromeritics Accelerated Surface Area and Porosimetry System (ASAP 2020, Micromeritics) with N2 adsorption at −196 °C.
- •X-ray Diffraction (XRD) is used for identifying the crystalline phases. For the different powders and carbon nanofibers analysis, the x-ray Diffractograms were obtained using a Bruker (D8 Avance Davinci) diffractometer with CuKα (λ = 0.1542 nm) radiation operated at 40 kV and 40 mA. The 2θ Bragg's angles range was from 10° to 80° at a scan rate of 1°/min.
- •Thermo-Gravimetric Analysis (TGA) is applied to determine the changes in mass. In addition, thermal stability at 800 °C was carried out for all catalysts using TA Instruments SDT – Q60, at a heating rate of 10 °C min−1 under N2 atmosphere. Aluminum crucibles and 11 mg of sample were required for this experiment. TGA of carbon nanofibers was carried out using a SETERAM Thermal Gravimetric Analyzer from room temperature up to 800 °C. Nitrogen was used as a carrier gas with a constant flow rate during the analysis.
- •Scanning Electron Microscopy is employed for characterizing the morphology of the nanofibers. The surfaces of the different catalysts were analyzed with a ZEISS 1430 Scanning Electron Microscope (SEM) and a Zeiss Ultra55/EDS Bruker SEM (with a higher resolution) was used to observe the shape, the diameters and the lengths of the carbon nanofibers. Prior to these experiments, the nanofibers are deposited on a planar substrate for SEM observation according to the process described hereafter. First, 20 ug of nanofibers were mixed with 200 ul of Hewamethyldisilazane (HDMS) solution acting as an adhesion promoter. Then, the mixture is deposited on a polished mirror silicon substrate using a Karl Suss® RC8 spin coater with 500 rpm speed condition.
3. Results and discussion
3.1. Catalysts characterization
3.1.1. Specific surface area (BET)
The specific surface area (SBET), the pore volume (VP) and the diameter (DP) of nickel-copper catalysts synthesized by nitrate salts (Ni50N, Ni40N and Ni30N) and ammonium complexes impregnation (Ni50C, Ni40C and Ni30C) are collected in table 3.
Table 3. Textural properties of catalysts impregnated with nitrate salts (Ni30N, Ni40N and Ni50N) and with ammonium complexes (Ni30C, Ni40C and Ni50C).
| Sample | SBET (m2 g−1) | VP (cm3 g−1) | DP (Å) | |
|---|---|---|---|---|
| Alumina support | 175.5 | 0.183 | 41.3 | |
| Nitrate salts | Ni30N | 181.5 | 0.222 | 47.2 |
| Ni40N | 169.5 | 0.189 | 49.7 | |
| Ni50N | 155.2 | 0.207 | 51.4 | |
| Ammonium complexes | Ni30C | 315.0 | 0.150 | 26.9 |
| Ni40C | 290.3 | 0.181 | 37.1 | |
| Ni50C | 166.0 | 0.100 | 47.9 |
As indicated on table 3, the specific surface area is reduced for the two types of catalysts when increasing the nickel loading, this result being already observed by Ashour [22]. This author explains that when increasing the amount of Ni and Cu metals over the alumina support of a catalyst, the BET surface area decreases and, subsequently, that the pore diameter increases. Therefore, it could be pointed out that metal loading is related to the pore size which can explain the lowest values obtained for the surface area for the two Ni50N and Ni50C catalysts.
On the other hand, and according to IUPAC classification, the isotherms of all the catalysts belong to type IV are shown in figure 1 [23].
Figure 1. Adsorption–desorption isotherms of catalysts: (a) Catalysts impregnated with nitrate salts and (b) Catalysts impregnated with ammonium complexes.
Download figure:
Standard image High-resolution imageThese figures exhibit the hysteresis cycle which is characteristic of mesoporous solids. By comparing the catalysts according impregnation solutions, the catalysts synthesized by nitrate salts present the lower specific surface area. However, the corresponding pore diameter and pore volume are higher than the catalysts impregnated with nickel and copper ammonium complexes. According to Avraham et al [18] and Li et al [19] by means of the use of complex materials in catalysts, it is possible to induce porosity and to increase the surface area. This is due to the fact that in this case complexes are crystalline materials that generate homogeneous nucleation [13]. This is confirmed by the presence of complexes in NixC catalysts which result from the metallic particles deposition. These results also show that the specific surface area of support has increased, for instance, the value of SBET for the particles of support is 175.5 m2.g−1, while for Ni30C and Ni40C catalysts are increased with values of 315.0 m2.g−1 and 290.3 m2.g−1, respectively. Therefore, we suggest that for supported metal catalysts synthesized by impregnation, the metals deposited as crystalline particles (complexes) are highly dispersed on particles of the support as mentioned by Xu et al [24]. It is well known that in catalyst, the dispersion plays a major role for both catalytic activity and selectivity [24]. Consequently, by comparing the two impregnation methods using nitrate salts and ammonium complexes, the last one is revealed itself more advantageous. The lower surface area of the samples prepared by impregnation using nitrate salts may be due to the sintering of smaller particles to bigger ones [13]. According to Sebastián del Río [3], catalysts prepared by impregnation present size distributions of wide particles. Therefore, the growth control of nanoparticles is complex and it is also combined with the structure and surface chemistry of the support.
3.1.2. X-ray diffraction (XRD)
The XRD spectrum of catalysts synthesized by nitrate salts and ammonium complexes of copper and nickel are shown in figure 2. Regardless of the impregnation solutions used in the catalysts preparation, the observed peaks are characteristics of NiO at 37.3°, 43.4°, 62.9° and 75.5° [25–27] and 79.0° [27] and those observed at 43.3° and 62.8° to the (Ni-Cu)O [28]. The highest intensity of NiO peaks dropped as Ni amount decreases. The presence of NiO in new catalysts is due to successive steps of calcination and the decomposition of nickel nitrate [10] and nickel ammonium complexes. Besides, NiAl2O4 structures formed in a few quantity is also observed with characteristic peaks located at 37.0°, 66.2° and 45.5° [29].
Figure 2. Diffractograms of catalysts synthesized by impregnation using nitrate salts and ammonium complexes.
Download figure:
Standard image High-resolution imageThe FWHM of the Ni reflections shown in figure 2 are highly variable. NixC catalysts have shown the broadest Ni reflections indicating that their respective Ni domain sizes are small in these samples [30].
As expected, it is highly dependent on the method of catalyst preparation used: on one side, NixC synthesized by complexes seems to enhance the formation of NiO with small domain sizes, while on the other side, impregnation by nitrates (NixN) promote the formation of relatively large NiO crystals. Due the high difference in Ni domain sizes, it would be expected to observe substantial differences in performance of fresh catalysts [31].
3.1.3. Thermo-gravimetric analysis (TGA)
Thermo-Gravimetric Analysis under nitrogen atmosphere of non-reduced catalysts is shown in figure 3. It can be observed in figure 3(a) that the weight loss of catalysts varies accordingly to the type of impregnation solution used in catalyst preparation, as well as of nickel loading. This behavior was also observed by Al-Fatesh et al [13] whom concluded that this behavior affects the carbon deposition. Likewise, the authors observed that there is a maximum weight loss of catalyst at a higher temperature which explains the maximum carbon deposition obtained despite the preparation method [13]. It is already noticeable that the weight loss is more important for the Ni30C compared to the Ni50C, the same tendency is observed for the NixN catalysts with a smallest variation. In figure 3(b), the position of the temperature peak for the NixC is located between 41.6 °C and 50 °C with a most important decreases of dTG parameter for Ni50C. For the NixN, two peaks are observed between 83.3 °C and 141.6 °C, we can note that Ni50N also presents a lower dTG parameter, while for the Ni40N and Ni30N the order is reversed.
Figure 3. Thermo-gravimetric analysis of non-reduced catalysts under N2 atmosphere: (a) Mass variation rate and (b) dTG as a function of temperature.
Download figure:
Standard image High-resolution image3.2. Deposition of carbon nanofiber
3.2.1. Effect of growth temperature
Figure 4 shows that growth temperature has a significant influence on the carbon nanofiber producing. The catalysts synthesized by nitrate salts impregnation have a greater deposition of carbon nanofiber at 600 °C than that at 650 °C, which can indicate that the catalyst deactivation effect is more pronounced than new filaments nucleation and growth at higher reaction temperature [32]. Therefore, this explains the lower carbon deposition at high temperature [32].
Figure 4. Evolution of carbon growth rate for Ni50N and Ni50C catalysts versus time showing the influence of the growth temperature.
Download figure:
Standard image High-resolution imageThe deactivation rate is faster at high reaction temperatures and, as a result, a reduction in the carbon content is observed. When deactivation takes place, the carbon formation rate curves show an initial period of rapid growth until a maximum is reached, this is clearly observed in Ni50C-650 °C at 200 min, whereas for Ni50N-650 °C it occurs at shorter times that were not shown in the curves; this period is followed by a decrease in the carbon formation rate until a residual constant value is reached being for Ni50C is greater than Ni50N at 650 °C while at 600 °C there are no difference. The differences in decay initiation time, indicate that the catalyst by complexes has greater stability during longer times.
On the other hand, the catalysts synthesized by ammonium complexes impregnation have a maximum carbon deposition at 650 °C, which means those catalysts are more stable at high temperature. This result was expected because the Thermo-Catalytic Decomposition (TCD) of methane is an endothermic reaction and consequently, the balance conversion of methane rises when increasing temperature [33].
For a better assessment of all catalysts, Mc was calculated (using equation (1)) in terms of the quantity of nickel due to the carbon formation rate was highly subjected to the quantity of Ni active sites. Nevertheless, a maximum carbon yield of 744.84 gC/gNi was observed after 5 h of reaction at 600 °C with Ni50N catalyst. On the other hand, the highest Mc value for NixC catalyst was 584.82 gC/gNi with Ni50C. At 650 °C the carbon yield starts to decrease until the complete deactivation of catalyst. Hernadi et al [34] have reported the carbon production up to 1.6 gC/gcat for the cracking of an array of organic reactants over Co and Fe supported silica and zeolite. Our carbon yields compare favorably with those generated from methane over supported Ni-alumina (244 gC/gNi) and from over Ni/Cu/Al (585 gc/gNi) [Li Y, Chen J, Qin Y, Chang L [35].
Table 4 shows that Ni50N catalyst produced the greatest reductions in Mc yield at 650 °C, for which the evolution of carbon growth rate for Ni50N and its comparison with Ni50C have been studied, the results are shown in figure 4.
Table 4. Experimentally observed the behavior of accumulated carbon (AC) with total methane conversion for two types of Ni/Y zeolite used catalyst.
| Run N°. | Sample | T (ºC) | Wtot (g) | Mc (%) | Mc (gdep/gNi) |
|---|---|---|---|---|---|
| With nitrate salts NixN | |||||
| 1 | Ni30N | 600 | 1.8989 | 947.08 | 159.03 |
| 650 | 1.2604 | 629.26 | 105.67 | ||
| 2 | Ni40N | 600 | 3.5215 | 1758.99 | 448.97 |
| 650 | 2.2297 | 1110.41 | 283.42 | ||
| 3 | Ni50N | 600 | 3.8862 | 1943.10 | 744.84 |
| 650 | 1.9631 | 978.61 | 375.13 | ||
| With ammonium complexes NixC | |||||
| 4 | Ni30C | 600 | 0.3213 | 160.41 | 24.71 |
| 650 | 0.3491 | 174.46 | 26.88 | ||
| 5 | Ni40C | 600 | 2.4959 | 1247.95 | 316.38 |
| 650 | 1.9461 | 973.05 | 246.68 | ||
| 6 | Ni50C | 600 | 3.6783 | 1837.31 | 584.82 |
| 650 | 2.8350 | 1416.08 | 450.74 | ||
3.2.2. Ni loading effect in catalyst and type of impregnated solution
Carbon nanofibers (CNF) deposition based on catalysts synthesized by impregnation using both nitrate salts (Ni50N, Ni40N and Ni30N) and ammonium complexes (Ni50C, Ni40C and Ni30C) at 600 °C and 650 °C during 5 h is shown in figure 5; the final result is the average of five replicated samples. Ni50 loading produces the maximum quantity of carbon deposited followed by Ni40 and Ni30 in the following sequence Ni50 > Ni40 > Ni30, this behavior is found to be the same for the two types of impregnated solutions. The physicochemical characteristics (TGA, SBET, others) of these catalysts used depend on their composition. The technique XRD used showed that the prereduced active catalysts NixN and NixC were formed by particles of CuO and NiO (probably Ni–Cu alloy). The mode of synthesis of the catalisys also had a slight influence in the SBET. The study of the carbon growth from the catalytic decomposition of methane indicates on the one hand good activity of the catalysts NixN and on the other hand good stability over time of the NixC catalyst compared with the homologue.
Figure 5. Carbon nanofibers deposition (g) during 5 h and a growth temperature of 600 °C and 650 °C with NixN and NixC catalysts.
Download figure:
Standard image High-resolution imageFurthermore, if the carbon deposition is compared at 600 °C, according the two types of impregnated solutions, Ni50N catalyst presents a higher deposition of 3.89 g than Ni50C having a 3.68 g. This behavior was observed in catalysts synthesized by impregnation compared to co-precipitation and sol-gel method as well, impregnation method being more effective in CNF deposition even if their surface area is lower than the ones synthesized by co-precipitation and sol-gel [13]. This also occurs for NixN and NixC.catalysts. According to previous works [36], it is possible to obtain higher amount of carbon when the surface of the catalyst is coated by metallic particles that increase the catalytic activity of these sites when the metal is deposited through impregnation method.
3.2.3. Growth pressure effect
Figure 6 shows carbon nanofibers deposition with Ni50N and Ni50C catalysts at 600 °C, for 3 and 5 h with the two pressures of in ultra-low pressure conditions and 0.5 bar.
Figure 6. CNF deposition (g) at 600 °C with Ni50N and Ni50C at in ULPC (indicated as 0 bar in the graph) and 0.5 bar of growth pressure for 3 and 5 h.
Download figure:
Standard image High-resolution imageWhen a pressure of 0.5 bar is applied, the CNF deposition decreases for the two types of impregnation solutions. As an example, Monzón et al [32] investigated the influence of methane concentration applying pressures of 0.1, 0.075, 0.05 and 0.025 bar in carbon nanofiber manufacturing. These authors have observed that for greater pressures of methane, the formation rate of polymeric species increases which causes both the encapsulation and the deactivation of the surface of metallic crystallites of the catalysts then leading to a decrease of the carbon nanofiber producing.
3.3. Carbon nanofiber characterization
3.3.1. Thermo-gravimetric analysis (TGA)
Figure 7 shows the TGA profiles of carbon nanofibers obtained at 600 °C and 650 °C under N2 atmosphere which allows the evaluation of the thermal stability. Figures 7(a) and (b) reveal that thermal stability increases following the sequence Ni30 < Ni40 < Ni50. Concerning the growth temperature, the graphitic structures have a better stability at 600 °C for all the catalysts impregnated with nitrate salts. On the other hand, catalysts impregnated with ammonium complexes (Ni50C and Ni40C) produce fibers with a lower weight loss at 650 °C than at 600 °C. In few works, it has been reported that a high graphitic character of carbon nanofiber is due to the lowest quantity of structural defects and poor porosity [37, 38]. Additionally, according to Díaz et al [38], carbon nanofibers with the lowest graphitic character get oxidized at lowest temperatures.
Figure 7. TGA profile of carbon nanofibers under N2 atmosphere for fibers obtained for impregnated catalysts with (a) nitrate salts and (b) ammonium complexes.
Download figure:
Standard image High-resolution imageAs is shown figure 8, the oxidation temperatures vary following the Ni30 < Ni40 < Ni50 sequence and the fibers produced with Ni30 catalysts oxidized at the lowest temperature. For the NixN600, at 50% of weight loss the range of temperature is between 500 °C and 564 °C, while for the NixC650 the variation of temperature at the same weight loss is located between 512 °C and 564 °C. It is interesting to note that for a composition of 50% of nickel the temperature level has no significant influence at 50% of weight loss. Concerning the first highest peak observed in the figure 8(b), we note that for a highest concentration of Ni, the first peak rises at a maximum temperature of 563 °C for the Ni50C650 and 552 °C for Ni50N600. As it has been demonstrated here, the residual weight was higher in the materials obtained using the Ni30C and Ni30N catalysts due to the quantity of metal catalyst.
Figure 8. (a) TGA profile of carbon nanofibers under O2 atmosphere and (b) mass variation rate as a function of temperature.
Download figure:
Standard image High-resolution imageTable 5 collecting the initial degradation temperature shows that the onset of degradation varies according to the nickel loading. Especially when the nickel loading increases, the degradation temperature rises. It can also be noted that the fibers obtained by means of complexes have a degradation which arises earlier.
Table 5. Temperature of start Degradation of catalysts impregnated with nitrate salts (Ni30N, Ni40N and Ni50N) and with ammonium complexes (Ni30C, Ni40C and Ni50C).
| Sample | Initial degradation temperature (°C) | |
|---|---|---|
| Nitrate Salts | Ni30N600 | 418 |
| Ni40N600 | 438 | |
| Ni50N600 | 474 | |
| Ammonium complexes | Ni30C650 | 408 |
| Ni40C650 | 427 | |
| Ni50C650 | 446 |
3.3.2. X-ray Diffraction (XRD) analysis
XRD spectrum of carbon deposited on different catalysts at several nickel loadings and temperatures of 600 °C and 650 °C present no significant differences, as shown in figure 9.
Figure 9. X-ray Diffractograms of the carbon nanofibers synthesized by catalysts impregnated with nitrate salts (a) and ammonium complexes (b).
Download figure:
Standard image High-resolution imageXRD patterns of all samples show peaks at 26.2° (002) and 44.4° (101) corresponding to 2 H—Graphite (dominant phase) and a peak at 52.3° (200) that belongs to Ni (metallic phase) [39, 40]. The diffraction peaks of graphitic carbon (ICDS NO: 01-0640) are observed at 2θ values of 26.3°, 44.8° and 54.2°, respectively [10]. For the samples of Ni30C650 and Ni30C600 catalysts, high intensity peaks are observed at 44.4° and 52.3° which could be result from the presence of metallic phase of nickel (ICDS NO: 04-0850) from the deactivated catalyst [10]. The conditions for these samples produce the lowest quantity of carbon nanofibers, although the weight of the catalyst used was the same. Therefore, it is possible that these carbon nanofibers contain a major amount of nickel. As shown in figure 9, when the reaction temperature increased from 600 °C up to 650 °C, the intensity of the peaks at 44.8° and 52.3° increases thus suggesting a progressive graphitization of CNF [10].
3.3.3. Scanning electron microscopy (SEM) observations
The catalysts prepared with impregnation using nitrate salts (Ni30N, Ni40N, Ni50N) and ammonium complexes (Ni30C, Ni40C, Ni50C) were observed by ZEISS1430 SEM (figure10). These observations reveal a rough surface with high porosity mainly for Ni30C and Ni30N catalysts where Ni-Cu metal is immersed into the pores. Ni50 catalysts, with the higher load, present a less rough surface with close, rounded and smooth conglomerates. When the Ni loading increases, the conglomerate of catalysts is more disconnected having bigger and irregular sizes. These results are related to the SBET of the catalysts shown in table 3. Moliner et al [40] concluded that a high SBET provided a long lifetime for the catalyst.
Figure 10. SEM images of catalysts prepared by impregnation using nitrates salts (a) to (c) and ammonium complexes (d) to (f).
Download figure:
Standard image High-resolution imageSEM observations of Carbon Nanofibers (CNF) is presented in figures 11 to 12 for the nitrate salts NixN and NixC obtained at 600 °C and 650 °C during 5 h, respectively. For all the situations, the general shape of the CNF is comparable to clusters which were more or less agglomerated together, these clusters seem to be more isolated for the ammonium complexes but agglomerated with smaller ones. From a general point of view, carbon nanofibers formed from the two NixN and NixC catalysts present smooth and regular shapes. These nanofilaments were often twisted but they exhibited a large range of diameters. This can be attributed to the decrease of the ammonium complexes density as also observed by Calafat and Sánchez [41] with 40Fe/ZrO2 catalyst. Furthermore, several differences can be observed between nitrate salts NixN and ammonium complexes NixC impregnated at 600 °C and 650 °C. Concerning the influence of the temperature and regarding figures 11 and 12, it is noticeable that an increase of the temperature leads to a global decreasing of the CNF diameters which are collected in table 6.
Figure 11. Evolution of CNF with different ammonium complexes at 600 °C during 5 h: (a)–(b) Ni50N, (c)–(d) Ni40N and (e)–(f) Ni30N and at 650 °C during 5 h: (g)–(h) Ni50N, (i)–(j) Ni40N and (k)–(l) Ni30N.
Download figure:
Standard image High-resolution image
Figure 12. Evolution of CNF with different ammonium complexes at 600 °C during 5 h: (a)–(b) Ni50C, (c)–(d) Ni40C and (e)–(f) Ni30C and at 650 °C during 5 h: (g)–(h) Ni50C, (i)–(j) Ni40C and (k)–(l) Ni30C.
Download figure:
Standard image High-resolution image
Table 6. Average of nanofilaments of carbon for the different conditions of impregnation.
| CNF diameters | 600 °C | 650 °C |
|---|---|---|
| Ni50N | 50 to 140 nm | 60 to 120 nm |
| Ni40N | 50 to 140 nm | 60 to 120 nm |
| Ni30N | 40 to 95 nm | 37 to 50 nm |
| Ni50C | 40 nm | 25 to 40 nm |
| Ni40C | 50 nm | 37 to 85 nm |
| Ni30C | 85 nm | 40 to 80 nm |
The NixN and NixC catalysts lead to the formation of CNF, figures 11 and 12. The ratio of carbon weight to grams in the catalysts used at the end of the test, as a parameter to measure the efficiency of the catalyst, is also shown in table 4. Taking into account the important differences in the SBET in the new catalysts, it would be expected to obtain substantial differences in the CNF deposition parameter between the different catalysts tested. However, the fixed-time activity tests did not show very high differences in the evolution of CNF and, additionally, all remained active at the end of the test.
We also note that the CNF shape is more dislocated for the NixN thus explaining the great variation of the nanofilaments diameters. In figures 11(a) to (d) and figures 11(g) to (j), we can observe several bright spots located at the top of the growing carbon fibers. Suelves et al [20] attributed this phenomenon to the presence of Ni particles. They also mentioned a reduction of the bright spots when the temperature increases, as it is observed in this work. These authors noted that carbon nanofilaments were thinner for higher temperatures thus explaining the difference observed between 600 °C and 650 °C.
Moreover, when comparing the ammonium complexes (NixC) (figure 12) and nitrate salts (NixN) (figure 11) impregnated at the same temperature, the carbon nanofilaments diameters are more regular (constant). Concerning the diameter values, they are smaller for NixC than NixN at 600 °C and 650 °C even if we note an increase of the size for the Ni30C at 650 °C. This phenomenon can be attributed to the presence of the bright spots in the composition as we can see on figure 12(k), which does not exist for Ni50C and Ni40C.
On figure 11(b), as already mentioned by Reshetenko et al [42], we also observe a 'mother' particle divided in thinner filaments. For these authors, several filaments can grow from only one particle to form a so-called 'octopus' structure, and the diameters of the growing filaments are logically lower than that of the 'mother' particle. In this case, the graphite layers in the filaments are stacked perpendicular to their axes. In our study, we can note that the thinner fibers are not perpendicular to the mother one and consequently the filaments can grow in many directions. For Avdeeva et al [43], 'octopus' carbon nanofibers were obtained when the samples contain more than 9 at% of Cu in the catalysts, because a particle shape being close to octahedral.
4. Conclusions
Catalysts prepared by impregnation using ammonium complexes show relevant characteristics that renders interesting such types of impregnation compared to conventional catalysts used in carbon nanofibers production. For example, the influence of nickel loading on the synthesis of carbon nanofibers was investigated and reveals that an increase of the nickel loading leads to a greater carbon deposition. However, the influence of impregnation solutions used in catalyst preparation shows that the performance of catalytic activity between catalysts based on nitrate salts and ammonium complex impregnation solutions was not improved significantly.
The results obtained under different conditions of thermo-catalytic decomposition (TCD) of methane show globally a greater carbon deposition at 600 °C compared to that obtained at 650 °C. On the other hand, Ni50C catalyst produces the maximum carbon deposition at 650 °C suggesting its major stability at higher temperatures. All catalysts synthesized, irrespective of the impregnation solution used in their preparation, facilitate the formation of very long carbon filaments having few nanometers of diameter and some micrometers of length. It is noticeable that the graphitic character of carbon deposited is highly dependent on the nickel loading.
Acknowledgments
The authors are grateful to FONDECYT-CONCYTEC and to the Catholic University of Saint Mary (UCSM) by co-financing this research and to the University Federal of Paraiba and University of Lille for their help and technical support. This work was partly supported by the French RENATECH network.