- Email: [email protected]
Effect of catalyst and substrate on growth characteristics of carbon nanofiber onto honeycomb monolith
ARTICLE IN PRESS
JID: JTICE
[m5G;August 21, 2015;8:40]
Journal of the Taiwan Institute of Chemical Engineers 000 (2015) 1–10
Contents lists available at ScienceDirect
Journal of the Taiwan Institute of Chemical Engineers journal homepage: www.elsevier.com/locate/jtice
Effect of catalyst and substrate on growth characteristics of carbon nanofiber onto honeycomb monolith Soraya Hosseini a, Suraya Abdul Rashid b, Ali Abbasi c, Farahnaz Eghbali Babadi b, Luqman Chuah Abdullah a, Thomas S.Y. Choong a,∗ a b c
Department of Chemical and Environmental Engineering, Universiti Putra Malaysia, 43400 Selangor, Malaysia Materials Processing and Technology Laboratory, Institute of Advanced Technology, Universiti Putra Malaysia, 43400 Serdang, Selangor, Malaysia Faculty of Chemical Engineering, Universiti Teknologi Malaysia, Skudai 81310, Johor, Malaysia
a r t i c l e
i n f o
Article history: Received 8 April 2015 Revised 13 July 2015 Accepted 15 July 2015 Available online xxx Keywords: Acid modified Carbon nanofibres Honeycomb monolith Iron Wash-coating alumina
a b s t r a c t Carbon nanofiber coated monolith with a homogeneous and consistent layer was prepared by catalytic decomposition of benzene on iron catalyst. A comparative study was carried out on carbon nanofiber growth onto bare monolith, acid modified monolith and wash-coat alumina monolith. The catalyst was prepared by dip-coating the monolith into an iron-salt solution with different concentrations (0.1–0.3 g/mL), dried, and calcined at 500 °C. It was found that the concentration of catalyst controlled Fe particles dispersion, which in turn was responsible for the catalytic activity. Lower iron concentration loaded monolith showed higher bulk density of nanofibers growth compared with higher concentration of iron solution used. The results demonstrated that after treatment with nitric acid, the surface area of cordierite monoliths could be increased to values as high as 30.6 m2 /g. Intertwined bundles of carbon nanofibers grown by this pre-treatment formed of a wide range of diameter sizes with tree like morphology. In addition, wash-coat materials such as alumina, utilized to increase the specific surface area and to distribute the catalyst on the surface of the monolith. The deposition of alumina wash-coat layer caused the iron (Fe) to appear more homogeneously distributed after drying and calcination, indicating Fe-0.2–Al2 O3 /monolith to be a superior support to grow CNFs compared to other substrates. © 2015 Published by Elsevier B.V. on behalf of Taiwan Institute of Chemical Engineers.
1. Introduction Carbon nanofibres (CNFs) are extensively applied in numerous industries due to their fascinating characteristics such as: high strength and tensile modulus, high chemical stability in acidic and basic media, high thermal and electrical conductivity, low density, high corrosion resistance, and excellent creep resistance [1,2]. CNFs form aggregates with high surface areas, high mesopore volumes and low tortuosity. These fibers are promising materials for liquid phase reactions due to their mesoporous structure without microporosity, resulting in the lack of mass transfer limitations [3]. The wider use of CNFs in powder form is hampered due to some drawbacks including high-pressure drop, plugging and flow mal-distribution for fixed bed operation as well as agglomeration and difficulty in filtration because of fines formation for slurry operation [4–6]. To circumvent these drawbacks, CNFs’ growth on macro-structured supports has been suggested. The synthesis of the nanofibers directly on such substrates can efficiently improve their ∗
Corresponding author. Tel.: +60 3 89466293; fax: +60 3 86567120. E-mail address: [email protected] (T.S.Y. Choong).
performance in a variety of applications, especially as catalyst supports, by increasing the surface area. The role of the support is not only to disperse catalysts’ active phase, but also more importantly to prevent metal nanoparticles from aggregation into larger clusters. Spreading the metal phase on a support results in the formation of small crystallite sizes on the surface which is desirable for CNF growth [7]. The growth of CNFs on metallic filters, foams, silica gel beads, carbon felts and ceramic supports has been reported. Recently monolithic materials have attracted a good deal of attention due to their applications in a wide variety of processes. In separation and purification applications, monolith is preferred since a single-piece permeable mass can separate species better than a cluster of packed particles [8,9]. The application of a catalyst in the form of a honeycomb structure with a system combining liquid, gas and solid phases is known to decrease the mass transfer limitations between the phases due to the hydraulic regime developed inside the capillary channels of the monolith [10,11]. Despite the advantages of monolithic cordierite such as high void fraction, uniform flow distribution, large geometric surface area, low pressure drop and thermal shock resistance, the use of cordierite as a catalyst support is limited due to the fact that this material has a low specific surface area
http://dx.doi.org/10.1016/j.jtice.2015.07.015 1876-1070/© 2015 Published by Elsevier B.V. on behalf of Taiwan Institute of Chemical Engineers.
Please cite this article as: S. Hosseini et al., Effect of catalyst and substrate on growth characteristics of carbon nanofiber onto honeycomb monolith, Journal of the Taiwan Institute of Chemical Engineers (2015), http://dx.doi.org/10.1016/j.jtice.2015.07.015
JID: JTICE 2
ARTICLE IN PRESS
[m5G;August 21, 2015;8:40]
S. Hosseini et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2015) 1–10
(0.02–0.5 m2 /g) and weak metal–support interactions [12]. Therefore, two different treatments including wash coating [13] and acid modification [14] have been developed to increase the specific surface area. According to the literature, CNFs can be synthesized directly on monolith as substrate either by adding a catalyst such as ferrocene in the feed or by impregnation of metal (Fe, Ni, Co) over alumina through wash coating [15,16]. Jarrah et al. reported the synthesis of CNF coated monolith by using CH4 as carbon source and Ni deposited thick alumina layer (8–17 μm) as catalyst and used the samples for enzyme immobilization [17]. Furthermore, uniform layers of carbon nanofibres appeared on Ni catalyst on alumina wash-coated monoliths via thermocatalytic decomposition of C2 H6 :H2 by GarcıaBordeje et al. [18]. Afterward, they extended their work by changing the growth conditions such as the reaction temperature, the thickness of the alumina wash-coated layer and the type of carbon source [19]. They concluded that the growth conditions and the properties of the catalyst are the parameters affecting the properties of synthesized CNF–monolith composite. Ulla et al. deposited a zeolitic layer onto cordierite monolith in order to have the sufficient number of cationic exchange sites to anchor cobalt as an active catalyst metal. By using the prepared surface for CNF formation and acetylene as carbon source, carbon nanofibers grew over the cordierite monolith [20]. The effect of oxidation treatment of CNF-coated monolith was studied by Armenise et al. [21]. The treatment with concentrated acid (HNO3 and H2 O2 ) created oxygen functionalities on the fibers. The growth of CNFs on monolithic structure using Ni–Cu alloy as catalyst and C2 H4 as carbon source has also been reported by Yan-Li et al. [22]. They concluded that the addition of Cu to Ni can inhibit the over-growth of metal particles in order to improve the growth rate of CNTs at high temperatures. The aim of the present work is to study the effect of various monolith substrates for growing CNFs and to compare the growth yield and morphology of the obtained fibers. Here, we studied the initial catalyst deposition, CNF formation and its growth. Carbon nanofibers (CNFs) were grown over structured substrate consisting of honeycomb ceramic monoliths by catalytic chemical vapor deposition (CVD). The CNFs’ synthesis mechanism involved the decomposition of benzene/H2 as carbon source and use of various metallic surfaces as catalyst. Iron was selected as catalyst layer and was coated on different pre-treated surfaces. The catalyst layer was prepared by dip-coating of the monolith into an iron-salt solution with different concentrations (0.1–0.3 g/mL). Iron was coated over three various types of monoliths including unmodified, acid modified, and alumina wash-coated and then, the coated substrates were subjected to benzene/H2 as carbon source to grow carbon nanofibers. Moreover, the effect of three various temperatures of 700, 800 and 900 °C on CNT growth and morphology was studied. 2. Chemicals and materials Cordierite monoliths (2Al2 O3 •5SiO2 2MgO) with cell density of 400 cells per square inch and the size of 25(L) × 20(D) mm2 were cut from a commercial sample (100 mm length) supplied by Beihai Haihuang Chemical Packing Co. Ltd., China, and used as substrate. The physical specification of monolith is summarized in Table 1. KOH, Fe(NO3 )3 •9H2 O, Al2 (SO4 )3 •16H2 O, sodium dodecyl sulfate (SDS) and benzene were obtained from Sigma-Aldrich, Malaysia. Purified gases of Ar (99.99%) and H2 (99.99%) were supplied by The Linde AG Company. 2.1. Catalyst preparation Three various catalyst-coated substrates were prepared to grow carbon nanofibers. In the first method, three various concentrations
Table 1 The specification of bare monolith. Specification of monolith Monolith Cross-section Circular Diameter 2.50 ± 0.02 mm Length 2.50 ± 0.02 mm Surface area 1 cm2 /g Cells Channel Square Cells 400 (cpsi) Width 1.02 ± 0.02 mm Wall thickness 0.25 ± 0.02 mm Chemical compositions 50.9 ± 1.0% SiO2 35.2 ± 1.0% Al2 O3 MgO 13.9 ± 0.5% Others <1%
of 0.1, 0.2, and 0.3 g/mL of iron solution were used to coat the unmodified monolith in order to study the effect of catalyst concentration. The iron deposited on monolith substrates acts as the catalyst for CNF growth. To increase the surface area, the monolith was pre-treated using 2 M HNO3 solution in the second procedure before being coated with iron solution. In the third procedure, the monolith substrates were wash-coated with an alumina solution and afterwards impregnated with iron solution. 2.1.1. Coating over bare monolith 10 g Fe(NO3 )3 •9H2 O was mixed with 0.2 g sodium dodecyl sulfate (SDS) and 100 mL distilled water under agitation for 30 min. The bare monolith was soaked into the solution and heated at 100 °C for 6 h. Afterward, to avoid channel blockage or iron agglomeration, the excess solution trapped inside the monolith channels was purged using pressurized air. The orange colored monolith was then soaked in 1 M, KOH solution for 4 h in order to convert iron ions to Fe(OH)3 according to the given equation.
Fe(NO3 ) + 3NaOH → Fe(OH )3 + 3NaNO3 The brown colored monolith was washed several times with distilled water to remove excessive basic solution. Initially, the monolith was dried at room temperature. During the slow drying at ambient temperature, the humidity gradient along the channels’ lengths was lower and the formed particles were more homogeneously distributed. The iron-coated monolith was then dried in an oven at 110 °C for 24 h. Consequently, the monoliths were calcined under air atmosphere at the temperature up to 500 °C for 2 h with heating rate of 10 °C/min. Three iron solution concentrations of 0.1, 0.2 and 0.3 g/mL were individually used in the coating procedure of the catalyst on the monolith surface. The samples were labeled as Fe0.1/monolith, Fe-0.2/monolith and Fe-0.3/monolith. The iron loading was determined by weighting monolith and was about 1.5, 2.8 and 4 wt% of the bare monolith for 0.1, 0.2 and 0.3 g/mL solutions, respectively. 2.1.2. Coating over acid modified monolith The bare monolith was first modified through soaking in 2 M, HNO3 (250 mL) solution for 24 h on a shaker at 30 °C. The treated monolith was then rinsed with distilled water until acquiring neutral pH. 0.3 wt% weight loss was reported by weighting modified monolith after acid treatment. The acid modified monolith was then dried at 110 °C for 24 h in an oven. Afterward, 20 g Fe(NO3 )3 •9H2 O was mixed with 0.2 g sodium dodecyl sulfate (SDS) and 100 mL distilled water under shaking for 30 min and the acid modified monolith soaked into the iron solution (0.2 g/mL). The pervious procedure was repeated to obtain iron coated acid modified monolith. The sample was labeled as Fe-0.2/acid modified monolith.
Please cite this article as: S. Hosseini et al., Effect of catalyst and substrate on growth characteristics of carbon nanofiber onto honeycomb monolith, Journal of the Taiwan Institute of Chemical Engineers (2015), http://dx.doi.org/10.1016/j.jtice.2015.07.015
ARTICLE IN PRESS
JID: JTICE
[m5G;August 21, 2015;8:40]
S. Hosseini et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2015) 1–10
3
Table 2 Metal content and BET surface areas of various catalysts. Catalyst
Fe
Al
Mg
Si
Others
Pore volume (cm3 /g)
BET surface areas (m2 /g)
Fe-0.1/monolith Fe-0.2/monolith Fe-0.3/monolith Fe-0.2/acid monolith Fe-0.2/Al2 O3 monolith
15.3 26.4 37.17 25.1 19.12
27.1 23.3 20.14 23.13 38.28
11.8 8.98 6.12 10.11 7.34
44.1 40.3 35.43 40.17 33.65
1.7 1.02 1.14 1.49 1.61
0.043 0.081 0.089 0.156 0.169
1.26 2.17 5.45 30.6 41.31
2.1.3. Coating with alumina wash-coat For the third sample, 10 g aluminum sulfate Al2 (SO4 )3 •16H2 O was dissolved in 50 mL distilled water. The bare monolith was soaked in the solution for 2 h. Afterwards, pressurized air was applied to flush along the cordierite channels to ensure homogeneous coating of the salt solution. Initially, the monolith was dried at room temperature. The drying process was continued in an oven at 110 °C for 24 h, followed by calcination at 900 °C for 3 h in a tube furnace under air atmosphere with heating rate of 10 °C/min. Decomposition of Al2 (SO4 )3 to Al2 O3 is given as: heat
Al2 (SO4 )3 −−→ Al2 O3 + 3SO3 The amount of Al2 O3 coated on the monolith was about 3.5 wt% through weight measurement. Iron was supported on the Al2 O3 -coated monolith by dip-impregnating with the iron solution (0.2 g/mL), followed by drying at 110 °C and calcination at 500 °C. the sample was marked as Fe-0.2/Al2 O3 monolith. 2.2. CVD process in a horizontal tubular reactor A tube furnace reactor was utilized to grow carbon nanofibers on the Fe-coated monolith. The furnace was comprised of a 6.2 cm diameter quartz tube and a 65 cm heating zone. The first step of the growth process involved flushing the tube with Ar gas (flow rate: 10 mL/min) to ensure the elimination of oxygen inside the reaction tube. The furnace temperature was set at 700 °C for 2 h and the gas composition was controlled using mass flow meters (Cole Parmer 65 mm). Benzene along with a 50/50 vol% mixture of H2 /Ar was fed into a vaporizer at 700 °C. The vaporized benzene mixed with H2 and Ar was continuously introduced into the tubular quartz. The growth time was set to 2 h. The hydrocarbon (benzene) undergoes a homogeneous reaction over the iron catalyst in the presence of H2 , generating carbon radicals which in turn are deposited in the form of nanofibers on the surface of the substrate. After 2 h, the furnace was cooled down to room temperature under Ar gas (10 mL/min). 2.3. Characterization methods A scanning electron microscope (SEM, Philips XL30) and a field emission scanning electron microscope (FESEM, Sirion-100, USA) were used for taking images of the formed carbon nanofibers. The porous structure of the catalysts was characterized using N2 adsorption/desorption isotherms at −197 °C using a Micromeritics Tristar apparatus after outgassing at 250 °C. TGA experiments were performed using a thermogravimetric analyzer (Netzsch STA 409) and the system was heated from room temperature to 700 °C at a rate of 10 °C/min under flowing air. Raman spectroscopic analyses were carried out using a PerkinElmer GX FT-IR/Raman spectrometer equipped with a Nd:YAG laser (λ0 = 1064 nm). The adherence of the CNF coated monoliths was evaluated with their resistance against ultrasonic vibration. The CNF coated monoliths were subjected into an ultrasonic bath (40 kHz) by immersing in acetone inside a glass vessel. At every time interval, the samples were taken out, dried and the weight of the samples both before and after the ultrasonic treatments was measured.
3. Result and discussion The growth of CNFs on monolith is of great importance in a wide variety of applications due to the numerous advantages of their geometric shape. In this work, CNFs were grown on various surfaces coated with Fe particles by thermal chemical vapor deposition (CVD). Applied conditions were optimized for CNFs synthesis on monolith substrates, in which catalyst-coated monolith heated at 700 °C and placed parallel to the flow of benzene/H2 /Ar gas mixture, resulted in the formation of CNF deposits on the as-prepared catalyst. The type of catalyst used over monolith surface affected the fibers’ growth, bulk density and diameter. In order to evaluate the effect of the substrate used in the CNFs growth, various samples were synthesized using different features. By controlling experimental parameters such as treatment of monolith surface, catalyst concentration and synthesis temperature, various morphologies of CNFs were obtained. Analyzing the data achieved from surface area studies revealed that some physical changes took place with some variation in their bulk density. It was found that the diameter of the carbon nanofibers was dependent on the pre-treatment conditions of the substrates. 3.1. Effect of catalyst concentration To elucidate the catalyst distribution on the monolith, five samples named Fe-0.1/monolith, Fe-0.2/monolith, Fe-0.3/monolith, Fe0.2/acid modified monolith and Fe-0.2/Al2 O3 monolith were used to synthesize CNFs. In order to investigate the crystalline nature and surface morphology of the catalyst, scanning electron microscopy (SEM) was carried out for monoliths loaded with Fe particles. Influence of iron precursor concentration and monolith surface was studied. A series of monolith-supported Fe catalysts were prepared by impregnation from an aqueous solution of iron nitrate at room temperature. The ICP analysis of the catalysts revealed that the metal ratios coated are very close to those predicted from the catalyst preparation (Table 2). SEM images of three samples Fe-0.1/monolith, Fe-0.2/monolith and Fe-0.3/monolith are illustrated in Fig. 1(a), (b) and (c), respectively. Fig. 1(a) demonstrates highly dispersed Fe nano-particles on the monolith and less agglomeration of Fe is observed due to the low iron concentration. On the other hand, Fe particles were significantly more uniform in the presence of surfactant (SDS) with low concentration. Increasing the iron solution concentration up to 0.2 g/mL lead to Fe particles non-homogeneously distributed on the surfaces of channels, as can be seen in Fig. 1(b). The image shows that the catalysts particles were easily aggregated into larger particles and formed amorphous phases on the substrate. SEM image of Fe-0.3/monolith (Fig. 1(c)) showed that iron was in the form of floccules without having a regular structure. The surface of the catalyst was non-smooth and porous which resulted in an increase in the surface areas of Fe loaded monolith. The iron coated monolith was mainly composed of rubble-like particles with amorphous structure. To summarize, the lower iron concentration resulted in the formation of well-dispersed Fe particles with less agglomeration on monolith surface compared to higher iron concentration. Fig. 2 represents the nitrogen adsorption–desorption isotherm for the catalyst coated monolith. Fig. 2(a) shows a type III
Please cite this article as: S. Hosseini et al., Effect of catalyst and substrate on growth characteristics of carbon nanofiber onto honeycomb monolith, Journal of the Taiwan Institute of Chemical Engineers (2015), http://dx.doi.org/10.1016/j.jtice.2015.07.015
JID: JTICE 4
ARTICLE IN PRESS
[m5G;August 21, 2015;8:40]
S. Hosseini et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2015) 1–10
Fig. 1. SEM images of iron coated monolith with different concentrations (a)Fe-0.1/monolith, (b) Fe-0.2/monolith, (c) Fe-0.3/monolith, (d) Fe-0.2/acid modified monolith, (e) Fe-0.2/Al2 O3 wash coat monolith and (f) bare monolith.
adsorption–desorption isotherm which is the characteristic of macroporous materials. The hysteresis is formed due to the presence of some degree of mesoporosity in the catalyst’s pore network. The adsorption in macropores is visible in the shown isotherm graph at high relative pressures where p/p0 → 1. This is confirmed by the sudden increase in the adsorption at quantity near a relative pressure of unity. Three samples coated with Fe particles indicated similar isotherms with different surface area. With increasing Fe concentration from 0.1 g/mL to 0.3 g/mL, BET surface area increased from 1.26 to 5.45 m2 /g, respectively. Pore size distribution (based on BJH method) of the samples is presented in Fig. 2. According to Fig. 2(a), some macrospores (<50 nm) are available in the prepared sample.
The PSD indicates that there was macroporosity within the catalyst network as the PSD graph continues to grow to the values beyond 50 nm as well. The average pore width calculated by applying BJH model to the adsorption branch of the isotherm is 53.75 nm. A type IV adsorption–desorption isotherm which is a characteristic of mesoporous materials is demonstrated in Fig. 2(b). The presence of micropores and mesopores was appeared for Fe-0.2/acid modified monolith samples. It is clear that the BET surface area and pore volume significantly increased when the monolith surface was treated with 2 M HNO3 solution which is attributed to the presence of micropores and mesopores. Fe-0.2/Al2 O3 wash-coat monolith demonstrated a 35% increase in BET surface area compared to the other samples.
Please cite this article as: S. Hosseini et al., Effect of catalyst and substrate on growth characteristics of carbon nanofiber onto honeycomb monolith, Journal of the Taiwan Institute of Chemical Engineers (2015), http://dx.doi.org/10.1016/j.jtice.2015.07.015
JID: JTICE
ARTICLE IN PRESS
[m5G;August 21, 2015;8:40]
S. Hosseini et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2015) 1–10
5
Fig. 2. Nitrogen adsorption–desorption isotherm, pore size distribution of three samples (a) Fe-0.2/monolith, (b) Fe-0.2/acid modified monolith and (c) Fe-0.2/Al2 O3 wash coat monolith.
Adsorption–desorption isotherm (IV type) which is a characteristic of mesoporous materials was observed (Fig. 2(c)). Table 2 presents the calculated values for the structural porous properties of the prepared catalysts. The SEM and FESEM images in Fig. 3(a)–(c) exhibit the morphologies of CNFs-coated monolith, revealed in the form of nanofibrous structure. The images were taken at different magnifications and using various locations inside monolith cells. It can be seen that the carbon nanofibers were grown over three samples with different iron concentration from 0.1 to 0.3 g/mL. However, the bulk density and thickness of fibers are different. The nanofibers produced by using Fe0.1/monolith had uniform diameters and smooth surfaces with high density (Fig. 3(a)). The diameters of the CNFs are typically in the range of 36–70 nm and a minority of nanofibers showed diameters as small as 25 nm. The highly dense CNFs entangled together and covered the surface of monolith, forming a web network. It was shown that by increasing iron concentration up to 0.2 g/mL, as can be seen in Fig. 3(b), the density of CNFs was decreased. The CNFs were non-homogeneous and thicker compared to those grown on Fe-0.1/monolith that can be
attributed to non-smooth catalyst surface or agglomeration of catalyst particles. The agglomeration causes a decrease in the number of surface metal particles that can be active on supported catalyst. The formed CNFs have an average diameter of around 90 nm. Fig. 3(c) shows the SEM image of grown CNFs onto Fe-0.3/monolith. Insignificant growth was observed on the catalyst surface prepared using 0.3 g/mL Fe concentration. Relatively less dense fibers along with agglomeration of fibers can be seen in the image. In summary, monolith with low iron concentration (0.1 g/mL) yielded a significantly larger dispersion and denser growth of CNFs than the other two concentrations (0.2 and 0.3 g/mL). The average diameters of the CNFs increased in the sequence of Fe-0.1/monolith < Fe-0.2/monolith < Fe0.3/monolith. Based on these results, the catalyst dispersion and Fe concentration influenced the growth of CNFs and fibers diameter. Carbon nanofibers grown by CVD of benzene at 700 °C on the iron catalyst had diameter in the range of 25–200 nm as presented in the FESEM micrographs. Comparing Fig. 3(a)–(c) reveals that the iron concentration affected the bulk density of CNFs and slightly the physical structure of the fibers.
Please cite this article as: S. Hosseini et al., Effect of catalyst and substrate on growth characteristics of carbon nanofiber onto honeycomb monolith, Journal of the Taiwan Institute of Chemical Engineers (2015), http://dx.doi.org/10.1016/j.jtice.2015.07.015
JID: JTICE 6
ARTICLE IN PRESS
[m5G;August 21, 2015;8:40]
S. Hosseini et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2015) 1–10
Fig. 3. SEM images of CNFs over catalysts (a) Fe-0.1 /monolith, (b) Fe-0.2/monolith and (c) Fe-0.3/monolith.
3.2. Effect of acid treatment Fig. 4 illustrates SEM images of CNFs on Fe-0.2/acid modified monolith. In order to achieve highly dispersed active catalysts onto the monolith, bare monolith was chemically modified by treating with nitric acid (2 M). The corrosion of the monolith increased and some pores were generated due to acid treatment. However, ironimpregnated acid modified monolith was non-smooth due to the generated pores. Fig. 1(d) shows iron coated over acid treated monolith and a rough surface with mountain shaped features was appeared. Fig. 4 shows the non-dense curly growth of the CNFs on Fe-0.2/acid modified monolith. The dispersion of the Fe particles on acid modified monolith was not uniform due to the inhomogeneous porous surface. There was a change in the fibers’ diameter and the surface of CNFs and the generated fibers was not smooth compared to unmodified monoliths. The results analysis shows that the acid modification comes with certain changes in BET surface area and pore volume. The isotherm corresponded to type III was revealed for the bare monolith that was characteristic of macroporous solids [23]. However, a different isotherm type was found for acid modified monolith
with bimodal distribution including micropores and mesopores (IV type) as shown in Fig. 2(c). The BET surface area and pore volume significantly increased after the modification with HNO3 which is attributed to the creation of micropores and mesopores. Acid modified monolith impregnated with iron-based nanoparticles showed that the BET surface area increased by about 82% from 5.45 to 30.6 m2 /g. The diameter of carbon nanofibers grown on the Fe-0.2/acid modified monolith was in the range of 25–200 nm. Comparing Fig. 3(b) and Fig. 4, it can be concluded that acid treatment provided less uniformity on the surface and CNFs are inhomogeneous. Fig. 4 shows the SEM image of the carbon nanofiber grown inside the monolith cell. It can be observed that the coating layer is highly uneven and all parts of the monolith wall are covered with CNFs. This uneven distribution of CNFs is most likely associated with a non-uniform surface and consequently irregular distribution of the Fe catalyst particles. 3.3. Effect of alumina wash coat Supporting Al2 O3 on the monolith leads to an increase in the internal surface area through providing a porous layer. A complex
Please cite this article as: S. Hosseini et al., Effect of catalyst and substrate on growth characteristics of carbon nanofiber onto honeycomb monolith, Journal of the Taiwan Institute of Chemical Engineers (2015), http://dx.doi.org/10.1016/j.jtice.2015.07.015
JID: JTICE
ARTICLE IN PRESS
[m5G;August 21, 2015;8:40]
S. Hosseini et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2015) 1–10
7
Fig. 4. SEM images of CNFs (a) over Fe-0.2/acid modified monolith, (b) CNFs over Fe-0.2/acid modified monolith (100 nm), (c) inside channel after 2 h, (d) inside channel after 30 min.
Fig. 5. SEM images of CNFs over Fe-0.2/Al2 O3 wash coat monolith.
Please cite this article as: S. Hosseini et al., Effect of catalyst and substrate on growth characteristics of carbon nanofiber onto honeycomb monolith, Journal of the Taiwan Institute of Chemical Engineers (2015), http://dx.doi.org/10.1016/j.jtice.2015.07.015
JID: JTICE 8
ARTICLE IN PRESS
[m5G;August 21, 2015;8:40]
S. Hosseini et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2015) 1–10
Fig. 6. SEM images of CNFs synthesized at reduction temperature over Fe-0.2/ monolith (a) 700 °C, (b) 700 °C (100 nm), (c) 800 °C, (d) 800 °C (100 nm), (e ) 900 °C and (f) 900 °C (100 nm).
porosity (macropores of the monolith, micro or mesopores of the alumina wash-coat) was generated on the channel structure. Iron nitrate solution dispersed on alumina wash-coat was converted into iron by calcination at high temperature. The presence of wash-coat layer resulted in more dispersion of Fe particles into the pore-network. SEM images of the channels coated with alumina wash-coats were obtained on the pieces cut from a monolith (Fig. 1(e)). The support was completely covered with alumina. As can be seen in Fig. 5(a)–(c), the CNFs grown on alumina wash-coat substrate were significantly more dense and uniform, resulting in the dispersion of the catalyst particles on the porous substrate during impregnation. The generated CNFs revealed a smooth surface and were randomly entangled together. As shown in Fig. 5(a) and (b), the locky-like morphology for generated carbon nanofibers with average diameter in the range of 20–90 nm was formed (Fig. 5(d)). 3.4. Effect of temperature The SEM images of as-synthesized CNFs at reaction temperatures of 700–900 °C are shown in Fig. 6. As can be seen, the morphologies
of CNFs are different with curly and straight fibers. For temperatures higher than 800 °C, no growth was observed which was possibly due to the deactivation of the catalyst at high temperature. Fig. 6(c) shows the typical SEM image of the ultra-long CNFs that were generated at 800 °C; the high magnification image indicates the smooth surface and straight structure of the CNFs. Some aggregated CNFs were found on the surface in the form of amorphous carbon for the sample heat-treated at 800 °C. At lower synthesis temperatures, CNFs may contain both sp2 and sp3 hybridized carbon while higher temperatures resulted in a complete hybridization of the carbon content to sp2 [24]. Therefore, CNFs generated at lower temperature are more defective in comparison to the CNFs formed at higher temperature. Fig. 6(e) and (f) indicates that very small amounts of CNFs are grown at 900 °C which can be attributed to the inactivity of iron catalyst. An obvious decrease in CNF density of the catalyst was also observed, indicating that at the high temperature of 900 °C particle sintering phenomena could happen, preventing the growth of CNFs. In addition, the CNF growth may be inhibited by deactivation of the catalyst, possibly by amorphous carbon deposition (Fig. 6(f)). The SEM results showed that the CNF have bridged between
Please cite this article as: S. Hosseini et al., Effect of catalyst and substrate on growth characteristics of carbon nanofiber onto honeycomb monolith, Journal of the Taiwan Institute of Chemical Engineers (2015), http://dx.doi.org/10.1016/j.jtice.2015.07.015
ARTICLE IN PRESS
JID: JTICE
[m5G;August 21, 2015;8:40]
S. Hosseini et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2015) 1–10
9
2 1.9 1.8 Crabon/catalyst
1.7 1.6 1.5 1.4 1.3 1.2 1.1 1 Fig. 7. Typical Raman spectra for Fe-0.2/ monolith at the three temperatures 700 °C, 800 °C and 900 °C.
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
Catalsyt loading (g) Fig. 9. The optimum amount of catalyst loaded Al2 O3 /monolith for CNFs synthesis.
120 20 min
3.5. Catalyst activity
60 min
Weight (wt%)
100
40 min
80
60
40 0
200
400 Temperature (oC)
600
800
Fig. 8. TGA analysis profile the as-synthesised CNFs at three times 20, 40 and 60 min.
catalyst particles and so, no further growth can occur on the catalyst surface. Fig. 7 shows FT-Raman spectra obtained from CNF generated at different reaction temperatures from the catalytic decomposition of benzene. Two peaks D and G were appeared at approximately 1318 and 1611/cm, respectively. The quality of CNF can be identified by using the peak intensity ratio of the G-peak (1611/cm) to D-peak (1318/cm). The ratio between the size of two G and D peaks is a useful way of comparing the quality of the CNF obtained in terms of how ordered they are. This parameter is used to show the purity of carbonic materials and a larger G/D ratio indicates a higher purity [25,26]. The Raman spectra (Fig. 7) indicate that the G/D ratio is significantly higher at reaction temperature of 800 °C, with a value of 2.46, than for the catalyst temperature of 700 °C with the value of 1.35. This shows that a higher G/D ratio for carbon deposits was obtained from higher reaction temperatures, indicating a higher quality of carbon nanofibers. It can be seen that G/D value increased with increasing the reaction temperature from 700 to 800 °C, leading to a decrease in the structural defects. A significant decrease in the G/D ratio was seen at 900 °C which was most likely due to the low activity of catalyst at this temperature, leading to the lower quality of the filamentous carbon. Thermo-gravimetric analysis (TGA) of as-synthesized CNF on Fe0.2/Al2 O3 monolith is presented in Fig. 8. The TGA was performed in air with a heating rate of 10 °C/min. The TGA result showed a slight weight loss in the range of 100–400 °C and a sharp one from 400 to 650 °C. It was reported that the oxidation temperature is lower than 450 °C for amorphous carbon and 450–600 °C for single wall carbon nanotube [27,28]. The sharp decrease in weight loss can be attributed to the oxidation of carbonaceous species such as single and multiwall carbon nanotubes [27].
Catalyst plays a key role in the production of carbon nanofibers and some research work has been devoted to the development of efficient catalysts. Generally, the catalytic activity of iron catalysts decreases with deposition of a large amount of amorphous carbons on the catalysts during the synthesis. Therefore, it is required to develop catalysts with a long life as well as high activity; the optimum amount of catalyst should be determined. In the present study, benzene decomposition was carried out over monolith-supported iron catalyst (Fe/Al2 O3 monolith) under different conditions. The catalytic activity has directly affected the carbon nanofiber yield. The carbon/catalyst value is defined as the number of benzene molecules decomposed per one Fe atom in each Fe coated monolith. The ratio of carbon/Fe versus the loaded catalyst was used to investigate the catalyst activity. Fig. 9 shows the carbon yields when Fe/Al2 O3 monolith was applied as the catalyst to generate carbon nanofiber at 700 °C. The C/Fe value was determined via measuring the amount of deposited carbons and the weight of the catalyst for benzene decomposition. The iron catalyst was loaded from 0.1 to 0.8 g into alumina wash-coat and the generated CNFs were weighted for each run during 2 h. The C/Fe value increased significantly as the loading amounts of iron increased from 0.1 to 0.4 g and the value reached the maximum (C/Fe = 1.9) at 0.4 g loading of iron metal. When the loading amounts exceeded 0.4 g, the C/Fe values decreased sharply as shown in Fig. 9. The deactivation was found to be faster on higher amount Fe crystals. This clearly indicates an optimum Fe loaded (0.4 g) for CNF growth at the conditions used in the present work. The results clearly indicate that the amount of iron has very significant effects on the CNF growth due to promote coking rate and increasing the deactivation rate. Carbon yields were calculated by weighing the sample after the reduction process and after the carbon nanotube synthesis as following:
CNFs(wt%) =
(M f − Mi ) Mi
∗ 100
where Mf is the total mass obtained at the end of the process, and Mi is mass of the used catalyst [29]. 4. Conclusion The concentration of catalyst’s active phase particles dispersed on the monolith was found to be of key importance in determining the bulk density as well as the shape and diameter of the produced CNF. The morphology and diameter of CNF are directly linked to the concentration and the size of active catalyst particles. Various pretreatment procedures were applied to the substrate surface and their effects on generated carbon nanofibers were investigated. It was found that the surface of the monolith resulted from each pretreatment process had a direct influence on the density and type
Please cite this article as: S. Hosseini et al., Effect of catalyst and substrate on growth characteristics of carbon nanofiber onto honeycomb monolith, Journal of the Taiwan Institute of Chemical Engineers (2015), http://dx.doi.org/10.1016/j.jtice.2015.07.015
JID: JTICE 10
ARTICLE IN PRESS
[m5G;August 21, 2015;8:40]
S. Hosseini et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2015) 1–10
of filamentous carbon. The spatial density seems to increase by using alumina wash-coating and highly dense and entangled CNFs are formed due to the intermediate layer using Al2 O3 with high roughness. In this study, we have found that not only morphology but also diameter of grown carbon nanofibers was affected by temperature. Acknowledgment This work was financially supported by the Ministry of Higher Education, Government of Malaysia and Universiti Putra Malaysia (via grant number GP-IBT/2013/9416900). References [1] Wang G, Wang J, Wang H, Bai J. Preparation and evaluation of molybdenum modified Fe/MgO catalysts for the production of single-walled carbon nanotubes and hydrogen-rich gas by ethanol decomposition. J Environ Chem Eng 2014;2:1588– 95. [2] Peng B, Jiang Sh, Zhang Y, Zhang J. Enrichment of metallic carbon nanotubes by electric field-assisted chemical vapor deposition. Carbon 2011;49:2555–60. [3] Velasquez M, Batiot-Dupeyrat C, Gallego J, Santamari A. Chemical and morphological characterization of multi-walled-carbon nanotubes synthesized by carbon deposition from an ethanol–glycerol blend. Diam Relat Mater 2014;50:38–48. [4] Ruiz Cebollada P, Marties Latorre L, Monzon Bescos A, Garcia Bordeje E. Preparation of stainless steel microreactors coated with carbon nanofiber layer: impact of hydrocarbon and temperature. Catal Today 2009;147:S87–93. [5] Secordel X, Yoboue A, Berrier E, Cristol S, Lancelot Ch, Capron M, Paul J. Supported oxorhenate catalysts prepared by thermal spreading of metal Re0 for methanol conversion to methylal. J Solid State Chem 2011;184:2806–11. [6] Nawaz K, Schmidt Sh J, Jacobi AM. Effect of catalyst and substrate on the moisture diffusivity of silica-aerogel-coated metal foams. Int J Heat Mass Transf 2014;73:634–44. [7] Sekulic DP. Wetting and spreading of liquid metals through open microgrooves and surface alterations. Heat Transf Eng 2011;32:648–57. [8] Svec F, Huber CG. Monolithic materials, promises, challenges, achievements. Anal Chem 2006;78:2100–7. [9] Cong HP, Ren XC, Wang P, Yu SH. Macroscopic multifunctional graphene-based hydrogels and aerogels by a metal ion induced self-assembly process. ACS Nano 2012;6:2693–703. [10] Masoudi Soltani S, Hosseini Soraya, Malekbala Mohamad Rasool. A review on monolithic honeycomb structures and fabrication techniques. J Appl Sci Res 2013;9:2548–60. [11] Hosseini S, Choong TSY, Hamid M. Adsorption of a cationic dye from aqueous solution on mesoporous carbon-based honeycomb monolith. Desalin Water Treat 2012;49:326–36. [12] Marco Y, García-Bordeje E, Franch C, Palomares AE, Yuranova T, Minsker LK. Bromate catalytic reduction in continuous mode using metal catalysts supported on monoliths coated with carbon nanofibers. Chem Eng J 2013;230:605–11.
[13] Villegas L, Masset F, Guilhaume N. Wet impregnation of alumina-washcoated monoliths: effect of the drying procedure on Ni distribution and on autothermal reforming activity. Appl Catal A: Gen 2007;320:43–55. [14] Cheah W, Hosseini S, Khan MA, Chuah TG, Choong TSY. Acid modified carbon coated monolith for methyl orange adsorption. Chem Eng J 2013;215–216:747– 54. [15] Liu Q, Liu Z, Su J. Al2 O3 -coated cordierite honeycomb supported CuO catalyst for selective catalytic reduction of NO by NH3 : surface properties and reaction mechanism. Catal Today 2010;158:370–6. [16] Li W, Wang H, Ren Z, Wang G, Bai J. Co-production of hydrogen and multi-wall carbon nanotubes from ethanol decomposition over Fe/Al2 O3 catalysts. Appl Catal B: Environ 2008;84:433–9. [17] Jarrah N, van Ommen JG, Lefferts L. Development of monolith with a carbonnanofiber-washcoat as a structured catalyst support in liquid phase. Catal Today 2003;79–80:29–33. [18] Garcia-Bordeje E, Kvande I, Chen D, Ronning M. Carbon nanofibers uniformly grown on gamma-alumina washcoated cordierite monoliths. Adv Mater 2006;18:1589–91. [19] Garcıa-Bordeje E, Kvande I, Chen D, Ronning M. Synthesis of composite materials of carbon nanofibers and ceramic monoliths with uniform and tuneable nanofibre layer thickness. Carbon 2007;45:1828–38. [20] Ulla MA, Valera A, Ubieto T, Latorre N, Romeo E, Milt VG, Monzon A. Carbon nanofiber growth onto a cordierite monolith coated with Co-mordenite. Catal Today 2008;133–135:7–12. [21] Armenise S, Roldan L, Marco Y, Monzon A, García-Bordeje E. Elucidation of catalyst support effect for NH3 decomposition using Ru nanoparticles on nitrogenfunctionalized carbon nanofiber monoliths. J Phys Chem 2012;116:26385–95. [22] Yan-Li W, Xu-Jian W, Liang Z, Wen-Ming Q, Xiao-Yi L, Li-Cheng L. A high strength carbon nanofiber/honeycomb cordierite composite produced by chemical vapor deposition. New Carbon Mater 2012;27:153–6. [23] Nijhuis TA, Beers AEW, Vergunst T, Hoek I, Kapteijn F, Moulijn JA. Preparation of monolithic catalysts. Catal Rev – Sci Eng 2001;43:345–80. [24] Ding Q, Song X, Yao X, Qi X, Au C, Zhong W, Du Y. Large-scale and controllable synthesis of metal-free nitrogen-doped carbon nanofibers and nanocoils over watersoluble Na2 CO3 . Nanoscale Res Lett 2013;545:1–8. [25] Qian X, Liu H, Huang C, Chen S, Zhang L, Li Y, Wang J, Lia Y. Self-catalyzed growth of large-area nanofilms of two-dimensional carbon. Sci Rep 2015;5:7756. [26] Bahrami B, Khodadadi A, Mortazavi Y, Esmaielib M. Short time synthesis of high quality carbon nanotubes with high rates by CVD of methane on continuously emerged iron nanoparticles. Appl Surf Sci 2011;257:9710–16. [27] Li F, Wang Y, Wang D, Wei F. Characterization of single-wall carbon nanotubes by N2 adsorption. Carbon 2004;42:2375–83. [28] Kitiyanan B, Alvarez WE, Harwell JH, Resasco DE. Controlled production of singlewall carbon nanotubes by catalytic decomposition of CO on bimetallic Co–Mo catalysts. Chem Phys Lett 2000;317:497. [29] Louis B, Gulino G, Vieira R, Amadou J, Dintzer T, Galvagno S, Centi G, Ledoux MJ, Pham-Huu C. High yield synthesis of multi-walled carbon nanotubes by catalytic decomposition of ethane over iron supported on alumina catalyst. Catal Today 2005;102–103:23–8.
Please cite this article as: S. Hosseini et al., Effect of catalyst and substrate on growth characteristics of carbon nanofiber onto honeycomb monolith, Journal of the Taiwan Institute of Chemical Engineers (2015), http://dx.doi.org/10.1016/j.jtice.2015.07.015
JID: JTICE
[m5G;August 21, 2015;8:40]
Journal of the Taiwan Institute of Chemical Engineers 000 (2015) 1–10
Contents lists available at ScienceDirect
Journal of the Taiwan Institute of Chemical Engineers journal homepage: www.elsevier.com/locate/jtice
Effect of catalyst and substrate on growth characteristics of carbon nanofiber onto honeycomb monolith Soraya Hosseini a, Suraya Abdul Rashid b, Ali Abbasi c, Farahnaz Eghbali Babadi b, Luqman Chuah Abdullah a, Thomas S.Y. Choong a,∗ a b c
Department of Chemical and Environmental Engineering, Universiti Putra Malaysia, 43400 Selangor, Malaysia Materials Processing and Technology Laboratory, Institute of Advanced Technology, Universiti Putra Malaysia, 43400 Serdang, Selangor, Malaysia Faculty of Chemical Engineering, Universiti Teknologi Malaysia, Skudai 81310, Johor, Malaysia
a r t i c l e
i n f o
Article history: Received 8 April 2015 Revised 13 July 2015 Accepted 15 July 2015 Available online xxx Keywords: Acid modified Carbon nanofibres Honeycomb monolith Iron Wash-coating alumina
a b s t r a c t Carbon nanofiber coated monolith with a homogeneous and consistent layer was prepared by catalytic decomposition of benzene on iron catalyst. A comparative study was carried out on carbon nanofiber growth onto bare monolith, acid modified monolith and wash-coat alumina monolith. The catalyst was prepared by dip-coating the monolith into an iron-salt solution with different concentrations (0.1–0.3 g/mL), dried, and calcined at 500 °C. It was found that the concentration of catalyst controlled Fe particles dispersion, which in turn was responsible for the catalytic activity. Lower iron concentration loaded monolith showed higher bulk density of nanofibers growth compared with higher concentration of iron solution used. The results demonstrated that after treatment with nitric acid, the surface area of cordierite monoliths could be increased to values as high as 30.6 m2 /g. Intertwined bundles of carbon nanofibers grown by this pre-treatment formed of a wide range of diameter sizes with tree like morphology. In addition, wash-coat materials such as alumina, utilized to increase the specific surface area and to distribute the catalyst on the surface of the monolith. The deposition of alumina wash-coat layer caused the iron (Fe) to appear more homogeneously distributed after drying and calcination, indicating Fe-0.2–Al2 O3 /monolith to be a superior support to grow CNFs compared to other substrates. © 2015 Published by Elsevier B.V. on behalf of Taiwan Institute of Chemical Engineers.
1. Introduction Carbon nanofibres (CNFs) are extensively applied in numerous industries due to their fascinating characteristics such as: high strength and tensile modulus, high chemical stability in acidic and basic media, high thermal and electrical conductivity, low density, high corrosion resistance, and excellent creep resistance [1,2]. CNFs form aggregates with high surface areas, high mesopore volumes and low tortuosity. These fibers are promising materials for liquid phase reactions due to their mesoporous structure without microporosity, resulting in the lack of mass transfer limitations [3]. The wider use of CNFs in powder form is hampered due to some drawbacks including high-pressure drop, plugging and flow mal-distribution for fixed bed operation as well as agglomeration and difficulty in filtration because of fines formation for slurry operation [4–6]. To circumvent these drawbacks, CNFs’ growth on macro-structured supports has been suggested. The synthesis of the nanofibers directly on such substrates can efficiently improve their ∗
Corresponding author. Tel.: +60 3 89466293; fax: +60 3 86567120. E-mail address: [email protected] (T.S.Y. Choong).
performance in a variety of applications, especially as catalyst supports, by increasing the surface area. The role of the support is not only to disperse catalysts’ active phase, but also more importantly to prevent metal nanoparticles from aggregation into larger clusters. Spreading the metal phase on a support results in the formation of small crystallite sizes on the surface which is desirable for CNF growth [7]. The growth of CNFs on metallic filters, foams, silica gel beads, carbon felts and ceramic supports has been reported. Recently monolithic materials have attracted a good deal of attention due to their applications in a wide variety of processes. In separation and purification applications, monolith is preferred since a single-piece permeable mass can separate species better than a cluster of packed particles [8,9]. The application of a catalyst in the form of a honeycomb structure with a system combining liquid, gas and solid phases is known to decrease the mass transfer limitations between the phases due to the hydraulic regime developed inside the capillary channels of the monolith [10,11]. Despite the advantages of monolithic cordierite such as high void fraction, uniform flow distribution, large geometric surface area, low pressure drop and thermal shock resistance, the use of cordierite as a catalyst support is limited due to the fact that this material has a low specific surface area
http://dx.doi.org/10.1016/j.jtice.2015.07.015 1876-1070/© 2015 Published by Elsevier B.V. on behalf of Taiwan Institute of Chemical Engineers.
Please cite this article as: S. Hosseini et al., Effect of catalyst and substrate on growth characteristics of carbon nanofiber onto honeycomb monolith, Journal of the Taiwan Institute of Chemical Engineers (2015), http://dx.doi.org/10.1016/j.jtice.2015.07.015
JID: JTICE 2
ARTICLE IN PRESS
[m5G;August 21, 2015;8:40]
S. Hosseini et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2015) 1–10
(0.02–0.5 m2 /g) and weak metal–support interactions [12]. Therefore, two different treatments including wash coating [13] and acid modification [14] have been developed to increase the specific surface area. According to the literature, CNFs can be synthesized directly on monolith as substrate either by adding a catalyst such as ferrocene in the feed or by impregnation of metal (Fe, Ni, Co) over alumina through wash coating [15,16]. Jarrah et al. reported the synthesis of CNF coated monolith by using CH4 as carbon source and Ni deposited thick alumina layer (8–17 μm) as catalyst and used the samples for enzyme immobilization [17]. Furthermore, uniform layers of carbon nanofibres appeared on Ni catalyst on alumina wash-coated monoliths via thermocatalytic decomposition of C2 H6 :H2 by GarcıaBordeje et al. [18]. Afterward, they extended their work by changing the growth conditions such as the reaction temperature, the thickness of the alumina wash-coated layer and the type of carbon source [19]. They concluded that the growth conditions and the properties of the catalyst are the parameters affecting the properties of synthesized CNF–monolith composite. Ulla et al. deposited a zeolitic layer onto cordierite monolith in order to have the sufficient number of cationic exchange sites to anchor cobalt as an active catalyst metal. By using the prepared surface for CNF formation and acetylene as carbon source, carbon nanofibers grew over the cordierite monolith [20]. The effect of oxidation treatment of CNF-coated monolith was studied by Armenise et al. [21]. The treatment with concentrated acid (HNO3 and H2 O2 ) created oxygen functionalities on the fibers. The growth of CNFs on monolithic structure using Ni–Cu alloy as catalyst and C2 H4 as carbon source has also been reported by Yan-Li et al. [22]. They concluded that the addition of Cu to Ni can inhibit the over-growth of metal particles in order to improve the growth rate of CNTs at high temperatures. The aim of the present work is to study the effect of various monolith substrates for growing CNFs and to compare the growth yield and morphology of the obtained fibers. Here, we studied the initial catalyst deposition, CNF formation and its growth. Carbon nanofibers (CNFs) were grown over structured substrate consisting of honeycomb ceramic monoliths by catalytic chemical vapor deposition (CVD). The CNFs’ synthesis mechanism involved the decomposition of benzene/H2 as carbon source and use of various metallic surfaces as catalyst. Iron was selected as catalyst layer and was coated on different pre-treated surfaces. The catalyst layer was prepared by dip-coating of the monolith into an iron-salt solution with different concentrations (0.1–0.3 g/mL). Iron was coated over three various types of monoliths including unmodified, acid modified, and alumina wash-coated and then, the coated substrates were subjected to benzene/H2 as carbon source to grow carbon nanofibers. Moreover, the effect of three various temperatures of 700, 800 and 900 °C on CNT growth and morphology was studied. 2. Chemicals and materials Cordierite monoliths (2Al2 O3 •5SiO2 2MgO) with cell density of 400 cells per square inch and the size of 25(L) × 20(D) mm2 were cut from a commercial sample (100 mm length) supplied by Beihai Haihuang Chemical Packing Co. Ltd., China, and used as substrate. The physical specification of monolith is summarized in Table 1. KOH, Fe(NO3 )3 •9H2 O, Al2 (SO4 )3 •16H2 O, sodium dodecyl sulfate (SDS) and benzene were obtained from Sigma-Aldrich, Malaysia. Purified gases of Ar (99.99%) and H2 (99.99%) were supplied by The Linde AG Company. 2.1. Catalyst preparation Three various catalyst-coated substrates were prepared to grow carbon nanofibers. In the first method, three various concentrations
Table 1 The specification of bare monolith. Specification of monolith Monolith Cross-section Circular Diameter 2.50 ± 0.02 mm Length 2.50 ± 0.02 mm Surface area 1 cm2 /g Cells Channel Square Cells 400 (cpsi) Width 1.02 ± 0.02 mm Wall thickness 0.25 ± 0.02 mm Chemical compositions 50.9 ± 1.0% SiO2 35.2 ± 1.0% Al2 O3 MgO 13.9 ± 0.5% Others <1%
of 0.1, 0.2, and 0.3 g/mL of iron solution were used to coat the unmodified monolith in order to study the effect of catalyst concentration. The iron deposited on monolith substrates acts as the catalyst for CNF growth. To increase the surface area, the monolith was pre-treated using 2 M HNO3 solution in the second procedure before being coated with iron solution. In the third procedure, the monolith substrates were wash-coated with an alumina solution and afterwards impregnated with iron solution. 2.1.1. Coating over bare monolith 10 g Fe(NO3 )3 •9H2 O was mixed with 0.2 g sodium dodecyl sulfate (SDS) and 100 mL distilled water under agitation for 30 min. The bare monolith was soaked into the solution and heated at 100 °C for 6 h. Afterward, to avoid channel blockage or iron agglomeration, the excess solution trapped inside the monolith channels was purged using pressurized air. The orange colored monolith was then soaked in 1 M, KOH solution for 4 h in order to convert iron ions to Fe(OH)3 according to the given equation.
Fe(NO3 ) + 3NaOH → Fe(OH )3 + 3NaNO3 The brown colored monolith was washed several times with distilled water to remove excessive basic solution. Initially, the monolith was dried at room temperature. During the slow drying at ambient temperature, the humidity gradient along the channels’ lengths was lower and the formed particles were more homogeneously distributed. The iron-coated monolith was then dried in an oven at 110 °C for 24 h. Consequently, the monoliths were calcined under air atmosphere at the temperature up to 500 °C for 2 h with heating rate of 10 °C/min. Three iron solution concentrations of 0.1, 0.2 and 0.3 g/mL were individually used in the coating procedure of the catalyst on the monolith surface. The samples were labeled as Fe0.1/monolith, Fe-0.2/monolith and Fe-0.3/monolith. The iron loading was determined by weighting monolith and was about 1.5, 2.8 and 4 wt% of the bare monolith for 0.1, 0.2 and 0.3 g/mL solutions, respectively. 2.1.2. Coating over acid modified monolith The bare monolith was first modified through soaking in 2 M, HNO3 (250 mL) solution for 24 h on a shaker at 30 °C. The treated monolith was then rinsed with distilled water until acquiring neutral pH. 0.3 wt% weight loss was reported by weighting modified monolith after acid treatment. The acid modified monolith was then dried at 110 °C for 24 h in an oven. Afterward, 20 g Fe(NO3 )3 •9H2 O was mixed with 0.2 g sodium dodecyl sulfate (SDS) and 100 mL distilled water under shaking for 30 min and the acid modified monolith soaked into the iron solution (0.2 g/mL). The pervious procedure was repeated to obtain iron coated acid modified monolith. The sample was labeled as Fe-0.2/acid modified monolith.
Please cite this article as: S. Hosseini et al., Effect of catalyst and substrate on growth characteristics of carbon nanofiber onto honeycomb monolith, Journal of the Taiwan Institute of Chemical Engineers (2015), http://dx.doi.org/10.1016/j.jtice.2015.07.015
ARTICLE IN PRESS
JID: JTICE
[m5G;August 21, 2015;8:40]
S. Hosseini et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2015) 1–10
3
Table 2 Metal content and BET surface areas of various catalysts. Catalyst
Fe
Al
Mg
Si
Others
Pore volume (cm3 /g)
BET surface areas (m2 /g)
Fe-0.1/monolith Fe-0.2/monolith Fe-0.3/monolith Fe-0.2/acid monolith Fe-0.2/Al2 O3 monolith
15.3 26.4 37.17 25.1 19.12
27.1 23.3 20.14 23.13 38.28
11.8 8.98 6.12 10.11 7.34
44.1 40.3 35.43 40.17 33.65
1.7 1.02 1.14 1.49 1.61
0.043 0.081 0.089 0.156 0.169
1.26 2.17 5.45 30.6 41.31
2.1.3. Coating with alumina wash-coat For the third sample, 10 g aluminum sulfate Al2 (SO4 )3 •16H2 O was dissolved in 50 mL distilled water. The bare monolith was soaked in the solution for 2 h. Afterwards, pressurized air was applied to flush along the cordierite channels to ensure homogeneous coating of the salt solution. Initially, the monolith was dried at room temperature. The drying process was continued in an oven at 110 °C for 24 h, followed by calcination at 900 °C for 3 h in a tube furnace under air atmosphere with heating rate of 10 °C/min. Decomposition of Al2 (SO4 )3 to Al2 O3 is given as: heat
Al2 (SO4 )3 −−→ Al2 O3 + 3SO3 The amount of Al2 O3 coated on the monolith was about 3.5 wt% through weight measurement. Iron was supported on the Al2 O3 -coated monolith by dip-impregnating with the iron solution (0.2 g/mL), followed by drying at 110 °C and calcination at 500 °C. the sample was marked as Fe-0.2/Al2 O3 monolith. 2.2. CVD process in a horizontal tubular reactor A tube furnace reactor was utilized to grow carbon nanofibers on the Fe-coated monolith. The furnace was comprised of a 6.2 cm diameter quartz tube and a 65 cm heating zone. The first step of the growth process involved flushing the tube with Ar gas (flow rate: 10 mL/min) to ensure the elimination of oxygen inside the reaction tube. The furnace temperature was set at 700 °C for 2 h and the gas composition was controlled using mass flow meters (Cole Parmer 65 mm). Benzene along with a 50/50 vol% mixture of H2 /Ar was fed into a vaporizer at 700 °C. The vaporized benzene mixed with H2 and Ar was continuously introduced into the tubular quartz. The growth time was set to 2 h. The hydrocarbon (benzene) undergoes a homogeneous reaction over the iron catalyst in the presence of H2 , generating carbon radicals which in turn are deposited in the form of nanofibers on the surface of the substrate. After 2 h, the furnace was cooled down to room temperature under Ar gas (10 mL/min). 2.3. Characterization methods A scanning electron microscope (SEM, Philips XL30) and a field emission scanning electron microscope (FESEM, Sirion-100, USA) were used for taking images of the formed carbon nanofibers. The porous structure of the catalysts was characterized using N2 adsorption/desorption isotherms at −197 °C using a Micromeritics Tristar apparatus after outgassing at 250 °C. TGA experiments were performed using a thermogravimetric analyzer (Netzsch STA 409) and the system was heated from room temperature to 700 °C at a rate of 10 °C/min under flowing air. Raman spectroscopic analyses were carried out using a PerkinElmer GX FT-IR/Raman spectrometer equipped with a Nd:YAG laser (λ0 = 1064 nm). The adherence of the CNF coated monoliths was evaluated with their resistance against ultrasonic vibration. The CNF coated monoliths were subjected into an ultrasonic bath (40 kHz) by immersing in acetone inside a glass vessel. At every time interval, the samples were taken out, dried and the weight of the samples both before and after the ultrasonic treatments was measured.
3. Result and discussion The growth of CNFs on monolith is of great importance in a wide variety of applications due to the numerous advantages of their geometric shape. In this work, CNFs were grown on various surfaces coated with Fe particles by thermal chemical vapor deposition (CVD). Applied conditions were optimized for CNFs synthesis on monolith substrates, in which catalyst-coated monolith heated at 700 °C and placed parallel to the flow of benzene/H2 /Ar gas mixture, resulted in the formation of CNF deposits on the as-prepared catalyst. The type of catalyst used over monolith surface affected the fibers’ growth, bulk density and diameter. In order to evaluate the effect of the substrate used in the CNFs growth, various samples were synthesized using different features. By controlling experimental parameters such as treatment of monolith surface, catalyst concentration and synthesis temperature, various morphologies of CNFs were obtained. Analyzing the data achieved from surface area studies revealed that some physical changes took place with some variation in their bulk density. It was found that the diameter of the carbon nanofibers was dependent on the pre-treatment conditions of the substrates. 3.1. Effect of catalyst concentration To elucidate the catalyst distribution on the monolith, five samples named Fe-0.1/monolith, Fe-0.2/monolith, Fe-0.3/monolith, Fe0.2/acid modified monolith and Fe-0.2/Al2 O3 monolith were used to synthesize CNFs. In order to investigate the crystalline nature and surface morphology of the catalyst, scanning electron microscopy (SEM) was carried out for monoliths loaded with Fe particles. Influence of iron precursor concentration and monolith surface was studied. A series of monolith-supported Fe catalysts were prepared by impregnation from an aqueous solution of iron nitrate at room temperature. The ICP analysis of the catalysts revealed that the metal ratios coated are very close to those predicted from the catalyst preparation (Table 2). SEM images of three samples Fe-0.1/monolith, Fe-0.2/monolith and Fe-0.3/monolith are illustrated in Fig. 1(a), (b) and (c), respectively. Fig. 1(a) demonstrates highly dispersed Fe nano-particles on the monolith and less agglomeration of Fe is observed due to the low iron concentration. On the other hand, Fe particles were significantly more uniform in the presence of surfactant (SDS) with low concentration. Increasing the iron solution concentration up to 0.2 g/mL lead to Fe particles non-homogeneously distributed on the surfaces of channels, as can be seen in Fig. 1(b). The image shows that the catalysts particles were easily aggregated into larger particles and formed amorphous phases on the substrate. SEM image of Fe-0.3/monolith (Fig. 1(c)) showed that iron was in the form of floccules without having a regular structure. The surface of the catalyst was non-smooth and porous which resulted in an increase in the surface areas of Fe loaded monolith. The iron coated monolith was mainly composed of rubble-like particles with amorphous structure. To summarize, the lower iron concentration resulted in the formation of well-dispersed Fe particles with less agglomeration on monolith surface compared to higher iron concentration. Fig. 2 represents the nitrogen adsorption–desorption isotherm for the catalyst coated monolith. Fig. 2(a) shows a type III
Please cite this article as: S. Hosseini et al., Effect of catalyst and substrate on growth characteristics of carbon nanofiber onto honeycomb monolith, Journal of the Taiwan Institute of Chemical Engineers (2015), http://dx.doi.org/10.1016/j.jtice.2015.07.015
JID: JTICE 4
ARTICLE IN PRESS
[m5G;August 21, 2015;8:40]
S. Hosseini et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2015) 1–10
Fig. 1. SEM images of iron coated monolith with different concentrations (a)Fe-0.1/monolith, (b) Fe-0.2/monolith, (c) Fe-0.3/monolith, (d) Fe-0.2/acid modified monolith, (e) Fe-0.2/Al2 O3 wash coat monolith and (f) bare monolith.
adsorption–desorption isotherm which is the characteristic of macroporous materials. The hysteresis is formed due to the presence of some degree of mesoporosity in the catalyst’s pore network. The adsorption in macropores is visible in the shown isotherm graph at high relative pressures where p/p0 → 1. This is confirmed by the sudden increase in the adsorption at quantity near a relative pressure of unity. Three samples coated with Fe particles indicated similar isotherms with different surface area. With increasing Fe concentration from 0.1 g/mL to 0.3 g/mL, BET surface area increased from 1.26 to 5.45 m2 /g, respectively. Pore size distribution (based on BJH method) of the samples is presented in Fig. 2. According to Fig. 2(a), some macrospores (<50 nm) are available in the prepared sample.
The PSD indicates that there was macroporosity within the catalyst network as the PSD graph continues to grow to the values beyond 50 nm as well. The average pore width calculated by applying BJH model to the adsorption branch of the isotherm is 53.75 nm. A type IV adsorption–desorption isotherm which is a characteristic of mesoporous materials is demonstrated in Fig. 2(b). The presence of micropores and mesopores was appeared for Fe-0.2/acid modified monolith samples. It is clear that the BET surface area and pore volume significantly increased when the monolith surface was treated with 2 M HNO3 solution which is attributed to the presence of micropores and mesopores. Fe-0.2/Al2 O3 wash-coat monolith demonstrated a 35% increase in BET surface area compared to the other samples.
Please cite this article as: S. Hosseini et al., Effect of catalyst and substrate on growth characteristics of carbon nanofiber onto honeycomb monolith, Journal of the Taiwan Institute of Chemical Engineers (2015), http://dx.doi.org/10.1016/j.jtice.2015.07.015
JID: JTICE
ARTICLE IN PRESS
[m5G;August 21, 2015;8:40]
S. Hosseini et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2015) 1–10
5
Fig. 2. Nitrogen adsorption–desorption isotherm, pore size distribution of three samples (a) Fe-0.2/monolith, (b) Fe-0.2/acid modified monolith and (c) Fe-0.2/Al2 O3 wash coat monolith.
Adsorption–desorption isotherm (IV type) which is a characteristic of mesoporous materials was observed (Fig. 2(c)). Table 2 presents the calculated values for the structural porous properties of the prepared catalysts. The SEM and FESEM images in Fig. 3(a)–(c) exhibit the morphologies of CNFs-coated monolith, revealed in the form of nanofibrous structure. The images were taken at different magnifications and using various locations inside monolith cells. It can be seen that the carbon nanofibers were grown over three samples with different iron concentration from 0.1 to 0.3 g/mL. However, the bulk density and thickness of fibers are different. The nanofibers produced by using Fe0.1/monolith had uniform diameters and smooth surfaces with high density (Fig. 3(a)). The diameters of the CNFs are typically in the range of 36–70 nm and a minority of nanofibers showed diameters as small as 25 nm. The highly dense CNFs entangled together and covered the surface of monolith, forming a web network. It was shown that by increasing iron concentration up to 0.2 g/mL, as can be seen in Fig. 3(b), the density of CNFs was decreased. The CNFs were non-homogeneous and thicker compared to those grown on Fe-0.1/monolith that can be
attributed to non-smooth catalyst surface or agglomeration of catalyst particles. The agglomeration causes a decrease in the number of surface metal particles that can be active on supported catalyst. The formed CNFs have an average diameter of around 90 nm. Fig. 3(c) shows the SEM image of grown CNFs onto Fe-0.3/monolith. Insignificant growth was observed on the catalyst surface prepared using 0.3 g/mL Fe concentration. Relatively less dense fibers along with agglomeration of fibers can be seen in the image. In summary, monolith with low iron concentration (0.1 g/mL) yielded a significantly larger dispersion and denser growth of CNFs than the other two concentrations (0.2 and 0.3 g/mL). The average diameters of the CNFs increased in the sequence of Fe-0.1/monolith < Fe-0.2/monolith < Fe0.3/monolith. Based on these results, the catalyst dispersion and Fe concentration influenced the growth of CNFs and fibers diameter. Carbon nanofibers grown by CVD of benzene at 700 °C on the iron catalyst had diameter in the range of 25–200 nm as presented in the FESEM micrographs. Comparing Fig. 3(a)–(c) reveals that the iron concentration affected the bulk density of CNFs and slightly the physical structure of the fibers.
Please cite this article as: S. Hosseini et al., Effect of catalyst and substrate on growth characteristics of carbon nanofiber onto honeycomb monolith, Journal of the Taiwan Institute of Chemical Engineers (2015), http://dx.doi.org/10.1016/j.jtice.2015.07.015
JID: JTICE 6
ARTICLE IN PRESS
[m5G;August 21, 2015;8:40]
S. Hosseini et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2015) 1–10
Fig. 3. SEM images of CNFs over catalysts (a) Fe-0.1 /monolith, (b) Fe-0.2/monolith and (c) Fe-0.3/monolith.
3.2. Effect of acid treatment Fig. 4 illustrates SEM images of CNFs on Fe-0.2/acid modified monolith. In order to achieve highly dispersed active catalysts onto the monolith, bare monolith was chemically modified by treating with nitric acid (2 M). The corrosion of the monolith increased and some pores were generated due to acid treatment. However, ironimpregnated acid modified monolith was non-smooth due to the generated pores. Fig. 1(d) shows iron coated over acid treated monolith and a rough surface with mountain shaped features was appeared. Fig. 4 shows the non-dense curly growth of the CNFs on Fe-0.2/acid modified monolith. The dispersion of the Fe particles on acid modified monolith was not uniform due to the inhomogeneous porous surface. There was a change in the fibers’ diameter and the surface of CNFs and the generated fibers was not smooth compared to unmodified monoliths. The results analysis shows that the acid modification comes with certain changes in BET surface area and pore volume. The isotherm corresponded to type III was revealed for the bare monolith that was characteristic of macroporous solids [23]. However, a different isotherm type was found for acid modified monolith
with bimodal distribution including micropores and mesopores (IV type) as shown in Fig. 2(c). The BET surface area and pore volume significantly increased after the modification with HNO3 which is attributed to the creation of micropores and mesopores. Acid modified monolith impregnated with iron-based nanoparticles showed that the BET surface area increased by about 82% from 5.45 to 30.6 m2 /g. The diameter of carbon nanofibers grown on the Fe-0.2/acid modified monolith was in the range of 25–200 nm. Comparing Fig. 3(b) and Fig. 4, it can be concluded that acid treatment provided less uniformity on the surface and CNFs are inhomogeneous. Fig. 4 shows the SEM image of the carbon nanofiber grown inside the monolith cell. It can be observed that the coating layer is highly uneven and all parts of the monolith wall are covered with CNFs. This uneven distribution of CNFs is most likely associated with a non-uniform surface and consequently irregular distribution of the Fe catalyst particles. 3.3. Effect of alumina wash coat Supporting Al2 O3 on the monolith leads to an increase in the internal surface area through providing a porous layer. A complex
Please cite this article as: S. Hosseini et al., Effect of catalyst and substrate on growth characteristics of carbon nanofiber onto honeycomb monolith, Journal of the Taiwan Institute of Chemical Engineers (2015), http://dx.doi.org/10.1016/j.jtice.2015.07.015
JID: JTICE
ARTICLE IN PRESS
[m5G;August 21, 2015;8:40]
S. Hosseini et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2015) 1–10
7
Fig. 4. SEM images of CNFs (a) over Fe-0.2/acid modified monolith, (b) CNFs over Fe-0.2/acid modified monolith (100 nm), (c) inside channel after 2 h, (d) inside channel after 30 min.
Fig. 5. SEM images of CNFs over Fe-0.2/Al2 O3 wash coat monolith.
Please cite this article as: S. Hosseini et al., Effect of catalyst and substrate on growth characteristics of carbon nanofiber onto honeycomb monolith, Journal of the Taiwan Institute of Chemical Engineers (2015), http://dx.doi.org/10.1016/j.jtice.2015.07.015
JID: JTICE 8
ARTICLE IN PRESS
[m5G;August 21, 2015;8:40]
S. Hosseini et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2015) 1–10
Fig. 6. SEM images of CNFs synthesized at reduction temperature over Fe-0.2/ monolith (a) 700 °C, (b) 700 °C (100 nm), (c) 800 °C, (d) 800 °C (100 nm), (e ) 900 °C and (f) 900 °C (100 nm).
porosity (macropores of the monolith, micro or mesopores of the alumina wash-coat) was generated on the channel structure. Iron nitrate solution dispersed on alumina wash-coat was converted into iron by calcination at high temperature. The presence of wash-coat layer resulted in more dispersion of Fe particles into the pore-network. SEM images of the channels coated with alumina wash-coats were obtained on the pieces cut from a monolith (Fig. 1(e)). The support was completely covered with alumina. As can be seen in Fig. 5(a)–(c), the CNFs grown on alumina wash-coat substrate were significantly more dense and uniform, resulting in the dispersion of the catalyst particles on the porous substrate during impregnation. The generated CNFs revealed a smooth surface and were randomly entangled together. As shown in Fig. 5(a) and (b), the locky-like morphology for generated carbon nanofibers with average diameter in the range of 20–90 nm was formed (Fig. 5(d)). 3.4. Effect of temperature The SEM images of as-synthesized CNFs at reaction temperatures of 700–900 °C are shown in Fig. 6. As can be seen, the morphologies
of CNFs are different with curly and straight fibers. For temperatures higher than 800 °C, no growth was observed which was possibly due to the deactivation of the catalyst at high temperature. Fig. 6(c) shows the typical SEM image of the ultra-long CNFs that were generated at 800 °C; the high magnification image indicates the smooth surface and straight structure of the CNFs. Some aggregated CNFs were found on the surface in the form of amorphous carbon for the sample heat-treated at 800 °C. At lower synthesis temperatures, CNFs may contain both sp2 and sp3 hybridized carbon while higher temperatures resulted in a complete hybridization of the carbon content to sp2 [24]. Therefore, CNFs generated at lower temperature are more defective in comparison to the CNFs formed at higher temperature. Fig. 6(e) and (f) indicates that very small amounts of CNFs are grown at 900 °C which can be attributed to the inactivity of iron catalyst. An obvious decrease in CNF density of the catalyst was also observed, indicating that at the high temperature of 900 °C particle sintering phenomena could happen, preventing the growth of CNFs. In addition, the CNF growth may be inhibited by deactivation of the catalyst, possibly by amorphous carbon deposition (Fig. 6(f)). The SEM results showed that the CNF have bridged between
Please cite this article as: S. Hosseini et al., Effect of catalyst and substrate on growth characteristics of carbon nanofiber onto honeycomb monolith, Journal of the Taiwan Institute of Chemical Engineers (2015), http://dx.doi.org/10.1016/j.jtice.2015.07.015
ARTICLE IN PRESS
JID: JTICE
[m5G;August 21, 2015;8:40]
S. Hosseini et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2015) 1–10
9
2 1.9 1.8 Crabon/catalyst
1.7 1.6 1.5 1.4 1.3 1.2 1.1 1 Fig. 7. Typical Raman spectra for Fe-0.2/ monolith at the three temperatures 700 °C, 800 °C and 900 °C.
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
Catalsyt loading (g) Fig. 9. The optimum amount of catalyst loaded Al2 O3 /monolith for CNFs synthesis.
120 20 min
3.5. Catalyst activity
60 min
Weight (wt%)
100
40 min
80
60
40 0
200
400 Temperature (oC)
600
800
Fig. 8. TGA analysis profile the as-synthesised CNFs at three times 20, 40 and 60 min.
catalyst particles and so, no further growth can occur on the catalyst surface. Fig. 7 shows FT-Raman spectra obtained from CNF generated at different reaction temperatures from the catalytic decomposition of benzene. Two peaks D and G were appeared at approximately 1318 and 1611/cm, respectively. The quality of CNF can be identified by using the peak intensity ratio of the G-peak (1611/cm) to D-peak (1318/cm). The ratio between the size of two G and D peaks is a useful way of comparing the quality of the CNF obtained in terms of how ordered they are. This parameter is used to show the purity of carbonic materials and a larger G/D ratio indicates a higher purity [25,26]. The Raman spectra (Fig. 7) indicate that the G/D ratio is significantly higher at reaction temperature of 800 °C, with a value of 2.46, than for the catalyst temperature of 700 °C with the value of 1.35. This shows that a higher G/D ratio for carbon deposits was obtained from higher reaction temperatures, indicating a higher quality of carbon nanofibers. It can be seen that G/D value increased with increasing the reaction temperature from 700 to 800 °C, leading to a decrease in the structural defects. A significant decrease in the G/D ratio was seen at 900 °C which was most likely due to the low activity of catalyst at this temperature, leading to the lower quality of the filamentous carbon. Thermo-gravimetric analysis (TGA) of as-synthesized CNF on Fe0.2/Al2 O3 monolith is presented in Fig. 8. The TGA was performed in air with a heating rate of 10 °C/min. The TGA result showed a slight weight loss in the range of 100–400 °C and a sharp one from 400 to 650 °C. It was reported that the oxidation temperature is lower than 450 °C for amorphous carbon and 450–600 °C for single wall carbon nanotube [27,28]. The sharp decrease in weight loss can be attributed to the oxidation of carbonaceous species such as single and multiwall carbon nanotubes [27].
Catalyst plays a key role in the production of carbon nanofibers and some research work has been devoted to the development of efficient catalysts. Generally, the catalytic activity of iron catalysts decreases with deposition of a large amount of amorphous carbons on the catalysts during the synthesis. Therefore, it is required to develop catalysts with a long life as well as high activity; the optimum amount of catalyst should be determined. In the present study, benzene decomposition was carried out over monolith-supported iron catalyst (Fe/Al2 O3 monolith) under different conditions. The catalytic activity has directly affected the carbon nanofiber yield. The carbon/catalyst value is defined as the number of benzene molecules decomposed per one Fe atom in each Fe coated monolith. The ratio of carbon/Fe versus the loaded catalyst was used to investigate the catalyst activity. Fig. 9 shows the carbon yields when Fe/Al2 O3 monolith was applied as the catalyst to generate carbon nanofiber at 700 °C. The C/Fe value was determined via measuring the amount of deposited carbons and the weight of the catalyst for benzene decomposition. The iron catalyst was loaded from 0.1 to 0.8 g into alumina wash-coat and the generated CNFs were weighted for each run during 2 h. The C/Fe value increased significantly as the loading amounts of iron increased from 0.1 to 0.4 g and the value reached the maximum (C/Fe = 1.9) at 0.4 g loading of iron metal. When the loading amounts exceeded 0.4 g, the C/Fe values decreased sharply as shown in Fig. 9. The deactivation was found to be faster on higher amount Fe crystals. This clearly indicates an optimum Fe loaded (0.4 g) for CNF growth at the conditions used in the present work. The results clearly indicate that the amount of iron has very significant effects on the CNF growth due to promote coking rate and increasing the deactivation rate. Carbon yields were calculated by weighing the sample after the reduction process and after the carbon nanotube synthesis as following:
CNFs(wt%) =
(M f − Mi ) Mi
∗ 100
where Mf is the total mass obtained at the end of the process, and Mi is mass of the used catalyst [29]. 4. Conclusion The concentration of catalyst’s active phase particles dispersed on the monolith was found to be of key importance in determining the bulk density as well as the shape and diameter of the produced CNF. The morphology and diameter of CNF are directly linked to the concentration and the size of active catalyst particles. Various pretreatment procedures were applied to the substrate surface and their effects on generated carbon nanofibers were investigated. It was found that the surface of the monolith resulted from each pretreatment process had a direct influence on the density and type
Please cite this article as: S. Hosseini et al., Effect of catalyst and substrate on growth characteristics of carbon nanofiber onto honeycomb monolith, Journal of the Taiwan Institute of Chemical Engineers (2015), http://dx.doi.org/10.1016/j.jtice.2015.07.015
JID: JTICE 10
ARTICLE IN PRESS
[m5G;August 21, 2015;8:40]
S. Hosseini et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2015) 1–10
of filamentous carbon. The spatial density seems to increase by using alumina wash-coating and highly dense and entangled CNFs are formed due to the intermediate layer using Al2 O3 with high roughness. In this study, we have found that not only morphology but also diameter of grown carbon nanofibers was affected by temperature. Acknowledgment This work was financially supported by the Ministry of Higher Education, Government of Malaysia and Universiti Putra Malaysia (via grant number GP-IBT/2013/9416900). References [1] Wang G, Wang J, Wang H, Bai J. Preparation and evaluation of molybdenum modified Fe/MgO catalysts for the production of single-walled carbon nanotubes and hydrogen-rich gas by ethanol decomposition. J Environ Chem Eng 2014;2:1588– 95. [2] Peng B, Jiang Sh, Zhang Y, Zhang J. Enrichment of metallic carbon nanotubes by electric field-assisted chemical vapor deposition. Carbon 2011;49:2555–60. [3] Velasquez M, Batiot-Dupeyrat C, Gallego J, Santamari A. Chemical and morphological characterization of multi-walled-carbon nanotubes synthesized by carbon deposition from an ethanol–glycerol blend. Diam Relat Mater 2014;50:38–48. [4] Ruiz Cebollada P, Marties Latorre L, Monzon Bescos A, Garcia Bordeje E. Preparation of stainless steel microreactors coated with carbon nanofiber layer: impact of hydrocarbon and temperature. Catal Today 2009;147:S87–93. [5] Secordel X, Yoboue A, Berrier E, Cristol S, Lancelot Ch, Capron M, Paul J. Supported oxorhenate catalysts prepared by thermal spreading of metal Re0 for methanol conversion to methylal. J Solid State Chem 2011;184:2806–11. [6] Nawaz K, Schmidt Sh J, Jacobi AM. Effect of catalyst and substrate on the moisture diffusivity of silica-aerogel-coated metal foams. Int J Heat Mass Transf 2014;73:634–44. [7] Sekulic DP. Wetting and spreading of liquid metals through open microgrooves and surface alterations. Heat Transf Eng 2011;32:648–57. [8] Svec F, Huber CG. Monolithic materials, promises, challenges, achievements. Anal Chem 2006;78:2100–7. [9] Cong HP, Ren XC, Wang P, Yu SH. Macroscopic multifunctional graphene-based hydrogels and aerogels by a metal ion induced self-assembly process. ACS Nano 2012;6:2693–703. [10] Masoudi Soltani S, Hosseini Soraya, Malekbala Mohamad Rasool. A review on monolithic honeycomb structures and fabrication techniques. J Appl Sci Res 2013;9:2548–60. [11] Hosseini S, Choong TSY, Hamid M. Adsorption of a cationic dye from aqueous solution on mesoporous carbon-based honeycomb monolith. Desalin Water Treat 2012;49:326–36. [12] Marco Y, García-Bordeje E, Franch C, Palomares AE, Yuranova T, Minsker LK. Bromate catalytic reduction in continuous mode using metal catalysts supported on monoliths coated with carbon nanofibers. Chem Eng J 2013;230:605–11.
[13] Villegas L, Masset F, Guilhaume N. Wet impregnation of alumina-washcoated monoliths: effect of the drying procedure on Ni distribution and on autothermal reforming activity. Appl Catal A: Gen 2007;320:43–55. [14] Cheah W, Hosseini S, Khan MA, Chuah TG, Choong TSY. Acid modified carbon coated monolith for methyl orange adsorption. Chem Eng J 2013;215–216:747– 54. [15] Liu Q, Liu Z, Su J. Al2 O3 -coated cordierite honeycomb supported CuO catalyst for selective catalytic reduction of NO by NH3 : surface properties and reaction mechanism. Catal Today 2010;158:370–6. [16] Li W, Wang H, Ren Z, Wang G, Bai J. Co-production of hydrogen and multi-wall carbon nanotubes from ethanol decomposition over Fe/Al2 O3 catalysts. Appl Catal B: Environ 2008;84:433–9. [17] Jarrah N, van Ommen JG, Lefferts L. Development of monolith with a carbonnanofiber-washcoat as a structured catalyst support in liquid phase. Catal Today 2003;79–80:29–33. [18] Garcia-Bordeje E, Kvande I, Chen D, Ronning M. Carbon nanofibers uniformly grown on gamma-alumina washcoated cordierite monoliths. Adv Mater 2006;18:1589–91. [19] Garcıa-Bordeje E, Kvande I, Chen D, Ronning M. Synthesis of composite materials of carbon nanofibers and ceramic monoliths with uniform and tuneable nanofibre layer thickness. Carbon 2007;45:1828–38. [20] Ulla MA, Valera A, Ubieto T, Latorre N, Romeo E, Milt VG, Monzon A. Carbon nanofiber growth onto a cordierite monolith coated with Co-mordenite. Catal Today 2008;133–135:7–12. [21] Armenise S, Roldan L, Marco Y, Monzon A, García-Bordeje E. Elucidation of catalyst support effect for NH3 decomposition using Ru nanoparticles on nitrogenfunctionalized carbon nanofiber monoliths. J Phys Chem 2012;116:26385–95. [22] Yan-Li W, Xu-Jian W, Liang Z, Wen-Ming Q, Xiao-Yi L, Li-Cheng L. A high strength carbon nanofiber/honeycomb cordierite composite produced by chemical vapor deposition. New Carbon Mater 2012;27:153–6. [23] Nijhuis TA, Beers AEW, Vergunst T, Hoek I, Kapteijn F, Moulijn JA. Preparation of monolithic catalysts. Catal Rev – Sci Eng 2001;43:345–80. [24] Ding Q, Song X, Yao X, Qi X, Au C, Zhong W, Du Y. Large-scale and controllable synthesis of metal-free nitrogen-doped carbon nanofibers and nanocoils over watersoluble Na2 CO3 . Nanoscale Res Lett 2013;545:1–8. [25] Qian X, Liu H, Huang C, Chen S, Zhang L, Li Y, Wang J, Lia Y. Self-catalyzed growth of large-area nanofilms of two-dimensional carbon. Sci Rep 2015;5:7756. [26] Bahrami B, Khodadadi A, Mortazavi Y, Esmaielib M. Short time synthesis of high quality carbon nanotubes with high rates by CVD of methane on continuously emerged iron nanoparticles. Appl Surf Sci 2011;257:9710–16. [27] Li F, Wang Y, Wang D, Wei F. Characterization of single-wall carbon nanotubes by N2 adsorption. Carbon 2004;42:2375–83. [28] Kitiyanan B, Alvarez WE, Harwell JH, Resasco DE. Controlled production of singlewall carbon nanotubes by catalytic decomposition of CO on bimetallic Co–Mo catalysts. Chem Phys Lett 2000;317:497. [29] Louis B, Gulino G, Vieira R, Amadou J, Dintzer T, Galvagno S, Centi G, Ledoux MJ, Pham-Huu C. High yield synthesis of multi-walled carbon nanotubes by catalytic decomposition of ethane over iron supported on alumina catalyst. Catal Today 2005;102–103:23–8.
Please cite this article as: S. Hosseini et al., Effect of catalyst and substrate on growth characteristics of carbon nanofiber onto honeycomb monolith, Journal of the Taiwan Institute of Chemical Engineers (2015), http://dx.doi.org/10.1016/j.jtice.2015.07.015