- Email: [email protected]
Effects of surface modification on the reactivity of activated carbon in direct carbon fuel cells
Accepted Manuscript Effects of surface modification on the reactivity of activated carbon in direct carbon fuel cells Lijun Fan, Jun Wang, Linzhe Zhao, Nianjun Hou, Tian Gan, Xueli Yao, Ping Li, Yicheng Zhao, Yongdan Li PII:
S0013-4686(18)31714-6
DOI:
10.1016/j.electacta.2018.07.196
Reference:
EA 32393
To appear in:
Electrochimica Acta
Received Date: 2 April 2018 Revised Date:
12 June 2018
Accepted Date: 26 July 2018
Please cite this article as: L. Fan, J. Wang, L. Zhao, N. Hou, T. Gan, X. Yao, P. Li, Y. Zhao, Y. Li, Effects of surface modification on the reactivity of activated carbon in direct carbon fuel cells, Electrochimica Acta (2018), doi: 10.1016/j.electacta.2018.07.196. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Effects of Surface Modification on the Reactivity of Activated Carbon in Direct Carbon Fuel Cells Lijun Fan1,2, Jun Wang1,2, Linzhe Zhao1,2, Nianjun Hou1,2, Tian Gan1,2, Xueli Yao1,2, Ping Li1,2, Yicheng Zhao1,2*, Yongdan Li1,2,3* State Key Laboratory of Chemical Engineering (Tianjin University), Tianjin Key
RI PT
1
Laboratory of Applied Catalysis Science and Technology, School of Chemical Engineering, Tianjin University, Tianjin 300072, China
Collaborative Innovation Center of Chemical Science and Engineering (Tianjin),
SC
2
Tianjin, 300072, China
Department of Chemical and Metallurgical Engineering, Aalto University,
M AN U
3
Kemistintie 1, FI-00076 Aalto, Finland Abstract
Activated carbons (AC) pretreated with HNO3 and NaOH are investigated as the
TE D
fuel of direct carbon fuel cells. HNO3 and NaOH treatments both increase the oxygen content and decrease the graphitization degree of AC. The amount of hydroxyl groups
EP
on the surface of AC increases after the treatments with HNO3 and NaOH, and then decreases during the subsequent heating process in an inert atmosphere. On the
AC C
contrary, the carbonyl and quinone groups on the surface of the activated carbon remain stable during the heating process. The activated carbon treated with HNO3 shows the highest reactivity towards oxidation and reverse Boudouard reactions due to its lowest graphitization degree and highest oxygen content. The single cell with a
∗
Corresponding author. Email: [email protected]; Tel: +86-22-27405613; Fax: +86-22-27405243 (Y. Zhao) ∗ Corresponding author. Email: [email protected], [email protected]; Tel: +86-22-27405613; Fax: +86-22-27405243 (Y. Li) 1
ACCEPTED MANUSCRIPT 380 µm-thick yttria stabilized zirconia layer as the electrolyte and the activated carbon treated with HNO3 as the fuel exhibits the lowest polarization and the highest
RI PT
maximum power density of 128 mW cm-2 at 800 oC.
Keywords: Direct carbon fuel cell; Activated carbon; Surface modification;
AC C
EP
TE D
M AN U
SC
Electrochemical oxidation; Oxygen functional groups
2
ACCEPTED MANUSCRIPT
1. Introduction A direct carbon fuel cell (DCFC) converts the chemical energy stored in solid carbon, which could be obtained from coal and biomass, into electricity directly via [1, 2]
. The energy efficiency of traditional coal-fired power
RI PT
an electrochemical route
generation processes involving Carnot cycles is only 35-40% in general, while that of DCFCs could theoretically exceed 100% since the overall reaction (C+O2→CO2)
SC
possesses a slightly positive entropy change (1.6 J mol-1 K-1 at 600 oC) [3]. Meanwhile,
M AN U
the anodic product of DCFCs could be pure CO2, which simplifies the capture and sequestration processes of CO2 for downstream utilization[4]. Besides, the activity of solid carbon fuel keeps unity during the anodic reaction regardless of the extent of conversion, ensuring the high overall practical energy efficiency (about 80%) of
TE D
DCFCs [5].
The basic structure of a DCFC consists of three active components: a fuel electrode (anode), an oxidant electrode (cathode) and an electrolyte sandwiched
EP
between them. Based on the electrolytes used, DCFCs are categorized into three main
AC C
classes: molten hydroxide, molten carbonate and solid oxide DCFCs (SO-DCFCs). A comprehensive description of DCFCs classification could be found elsewhere
[1, 2, 6]
.
One of the key challenges for the practical application of DCFCs is their sluggish anodic kinetics, which is mainly due to the insufficient contact between the solid carbon fuel and the solid anode layer. Liquid media such as molten carbonates and metals have been utilized at the anode of DCFCs as shuttles of carbonate and oxygen ions between the carbon particles and the electrolyte layer, which enhance the 3
ACCEPTED MANUSCRIPT performance of the cells effectively [7, 8]. However, the hot caustic liquids decrease the stability of the cells simultaneously
[8]
. On the contrary, SO-DCFCs with an
all-solid-state structure without corrosive media fundamentally eliminate the
RI PT
corrosion of materials, showing a promising durability and feasibility to scale-up [9-11]. The major issue of SO-DCFCs is the difficulty in getting the solid fuel to the anode reaction zone or triple phase boundaries (TPBs) [12]. Recently, mixed ionic-electronic
[12-15]
. A tubular DCFC supported by a yttria
M AN U
area and promote the anodic kinetics
SC
conducting materials have been used as the anodes of SO-DCFCs to enlarge the TPB
stabilized zirconia (YSZ) electrolyte layer with a La0.6Sr0.4Co0.2Fe0.8O3-δ (LSCF) anode exhibited a maximum power density (Pmax) of 30 mW cm-2 at 800 oC with CO2 as the anodic purge gas [16]. The performance dropped by 40% after 11-day operation.
with solid anodes.
TE D
Nevertheless, a substantial improvement of the anodic activity is required by DCFCs
The structure and surface property of carbon influence the anodic kinetics of [6]
. Cherepy et al.[17] examined nine carbon samples in molten
EP
DCFCs significantly
AC C
carbonate, and found that crystallographic disorder degree and electrical conductivity were main factors determining the rate of discharge. Li et al.[18] pretreated activated carbon (AC) with HNO3, which increased oxygen-containing surface functional groups and improved the electrochemical reactivity of the carbon consequently. Cao et al.[19] found that both acid and base pretreatments could increase the specific surface area and microspore volume of AC. Zhang et al.[20] observed the increase of C=O and O-H groups and the disappearance of C=C groups on the surface of a 4
ACCEPTED MANUSCRIPT biomass-based AC after the treatment in HNO3. Eom et al.[21] demonstrated that the electrochemical resistance of raw coal is influenced by its surface properties such as the degree of oxidation and ash composition, which could be modified by acid
RI PT
treatments. However, the surface properties of carbon fuels were studied ex situ in most of the previous literatures, while the anodic process of DCFCs is more complex. For instance, Li et al.[22] proved that the oxygen-containing functional groups on the
SC
surface of coal decreased sharply after heat-treatment. Fuente-Cuesta et al.[23] found
M AN U
that the heat-treatment of bituminous coal led to higher structural order of carbon and lower H/C ratio, resulting in the decrease of anodic reactivity. Therefore, the alterations of the structure and surface properties of carbon fuels those affect the anodic activity of DCFCs have been not demonstrated clearly, which requires further
TE D
in-depth studies.
In this work, AC is pre-treated with HNO3 and NaOH, respectively, and the variations of its structure and surface properties before and after heat treatment are
EP
studied. The effects of surface modification of AC on the electrochemical
AC C
performances of the anode and the corresponding single cell are systematically investigated.
2. Experimental
2.1. Preparation and characterization of carbon fuels Base and acid modifications of carbon were carried out by immersing 5 g commercial AC (Aladdin, > 200 mesh) into 30 ml 2 mol L-1 NaOH and HNO3 solutions, respectively, at room temperature for 20 h. Then the samples were washed 5
ACCEPTED MANUSCRIPT with distilled water for three times. The concentrations of H+ and OH- in the acid and base solutions before and after the treatments are listed in Table S1. The washed samples were dried at 105 oC in air for 12 h, which are marked as AC-NaOH and
RI PT
AC-HNO3, respectively. The surface property of the AC samples before and after pretreatments was studied with a Fourier transform infrared (FTIR) spectrometer (Perkin-Elmer) in the
SC
region of 4000-400 cm−1. X-ray photoemission spectroscopy (XPS) was conducted
M AN U
using Escalab 250Xi system (ThermoFisher Scientific USA) with monochromatic 150 W Al-Kα radiation as the excitation source. Binding energies were calibrated to the common C1s electron binding energy at 284.8 eV.
Thermogravimetric analysis (TGA) of the AC samples were performed with an
TE D
analyzer (STA449 F3, Netzsch) in an Ar flow (50 ml min−1, STP) from room temperature to 800 oC at a heating rate of 5 oC min-1, and then the temperature was kept at 800 oC for 1 h. When the temperature cooled down to room temperature, the
EP
remaining samples were collected, and designated as AC-Ar, AC-NaOH-Ar and
AC C
AC-HNO3-Ar, respectively. Surface properties of the remaining samples were analyzed using FTIR and XPS. Raman measurements were performed with a Raman spectrometer (Renishaw inVia) with a 633 nm He-Ne laser excitation source. After the TGA in Ar, the gasification processes of the residues were also investigated with TGA in air and CO2, respectively, from room temperature to 800 oC at a heating rate of 5 oC min-1. The TGA in air is combined with an online mass spectrometer (HPR20, Hiden) to measure the amounts of CO and CO2 in the outlet gas. 6
ACCEPTED MANUSCRIPT 2.2. Single cell test Electrolyte-supported single cells were fabricated for electrochemical tests in this work. Commercial YSZ powder (Ningbo SOFCMAN Energy Technology Co., Ltd,
RI PT
China) was uniaxially pressed at 500 MPa, and then sintered at 1500 oC for 4 h in air. The YSZ electrolyte layer has a thickness of about 380 µm. Commercial Ce0.9Gd0.1O1.95 (GDC) powder (Ningbo SOFCMAN Energy Technology Co., Ltd,
SC
China) was mixed with a binder (V006, Heraeus Ltd.) to form a slurry, which was
M AN U
then screen-printed on both sides of the YSZ electrolyte pellet and sintered at 1250 oC in air for 2 h to form 7 µm-thick buffer layers to avoid the potential interfacial reaction between the YSZ electrolyte and the electrode materials at high temperatures. GDC powders were thoroughly mixed with LSCF powder at a weight ratio of 5:5 by
TE D
ball-milling as electrode materials. The LSCF powder was synthesized through an EDTA-citrate sol-gel method, and the detailed process could be found in a previous work
[24]
. The LSCF-GDC composite powder was also made into slurry,
EP
screen-printed on the GDC buffer layers, and then sintered at 1080 oC for 2 h in air.
AC C
The average thickness of the LSCF-GDC electrode layers was about 16 µm. The cross-sectional morphology of the single cell was observed with a Hitachi S-4800 scanning electron microscope (SEM). Ag paste was printed on both sides of the cell as a current collector. The geometric surface area of each electrode was 0.72 cm2. The configuration of the single cell was described in our previous work
[25]
.
Before the test, 0.1 g carbon fuel was put in the anode cavity with N2 as the purge gas (5 ml min-1, STP). The cathode gas was O2 (50 ml min-1, STP). The I-V and I-P curves 7
ACCEPTED MANUSCRIPT and electrochemical impedance spectra (EIS) were recorded at 750 and 800 oC with various flow rates of N2 using an electrochemical workstation (Versastat 3, Princeton Applied Research). The EIS spectra were obtained at open circuit state in the
RI PT
frequency range of 100 kHz-50 mHz.
3. Results 3.1. Characterization of AC samples
SC
The contents of carbon and oxygen on the surface of raw AC determined by XPS
M AN U
are 88.9 and 10.1 at.%, respectively (Table 1). After AC is treated with NaOH, the content of oxygen increases to 10.7 at.% with a little decrease of that of carbon. AC-HNO3 possesses the highest oxygen content and O/C ratio of 14.9 at.% and 0.18, respectively. The results of elemental analysis of the AC samples before and after acid
TE D
and base treatments show a similar tendency (Table S2).
The FTIR spectra of the AC samples are shown in Fig. 1. The spectrum of raw AC reveals the following absorption bands: the band at 3440 cm-1 assigned to O-H
EP
stretching vibration in hydroxyl functional groups; the bands at 2922 and 2845 cm-1
AC C
due to CH2 and CH3 asymmetric and symmetric stretching vibrations in alkyl chains [26]
; the band at 1700 cm-1 attributed to C=O stretching vibration in ketone, carboxylic
acid or ester
[20]
; the band at 1600 cm-1 ascribed to C=C stretching vibration in
aromatic rings or C=O stretching vibration in carbonyls [27] and the broad band within the range of 1300–1000 cm-1 associated with various function groups containing C–O band such as ethers, phenols and hydroxyl groups
[28]
. After the NaOH treatment, the
intensity of the band at 3440 cm-1 increases significantly, indicating an increase of the 8
ACCEPTED MANUSCRIPT amount of hydroxyl groups. Meanwhile, a small new band at 1050 cm-1 appears. The modification with HNO3 leads to not only increase of bands at 3440 and 1700 cm-1 but the appearance of new bands at 1533 and 1337 cm-1, which originate from highly
RI PT
conjugated C=O and C-O stretching vibration, respectively, in carboxylic groups and carboxylate moieties [28].
The C1s core-level spectra of the AC samples are shown in Fig. S1. Four
SC
deconvoluted peaks are obtained from each curve, corresponding to graphitic carbon
M AN U
(C-C, ~284.8 eV), carbon in alcohol, phenolic or ether (C-O-R, ~286.2 eV), carbonyl or quinone groups (-C=O, ~287.3 eV) and carboxyl or ester groups (O-C=O, ~289.0 eV), respectively [29]. The area percentages of the peaks are listed in Table 2. After the treatments with NaOH and HNO3, the intensity of Peak C-C decreases remarkably,
TE D
while that of Peak C-O-R increases noticeably. The intensities of Peaks –C=O and O-C=O slightly increase after the AC is treated with NaOH. In contrast, obvious elevation of peak –C=O is observed in the result of AC-HNO3.
EP
The TGA results of the untreated and treated AC samples in Ar are illustrated in
AC C
Fig. 2. The weights of all the samples decrease gradually before the temperature reaches 400 oC, and AC-HNO3 shows the highest weight loss rate. In the temperature range of 400-800 oC, all of the samples exhibit a rapid loss of the weight. The samples are kept at 800 oC for 1 h, during which period their weights are nearly stable. The weight loss of the samples shows an order of AC-HNO3 > AC-NaOH > AC. 3.2. Characterization of AC samples after the heat treatment in Ar The contents of carbon and oxygen on the surface of the samples after the 9
ACCEPTED MANUSCRIPT Ar-TGA test are listed in Table 3. All of the samples possess a low oxygen content compared with the samples before the heat treatment. The AC-HNO3 still exhibits the highest oxygen content after the heat treatment.
RI PT
Fig. 3 shows the FTIR spectra of the AC samples after the heat treatment in Ar. The peaks at 3440 cm-1 of AC-NaOH and AC-HNO3 decrease after the heat treatment, indicating the reduction of the amount of hydroxyl groups. The three samples after the
SC
heat treatment show similar contents of hydroxyl groups. The peaks at 1700 cm-1 in
M AN U
the spectra of AC-Ar and AC-NaOH-Ar are negligible, while that in AC-HNO3-Ar is noticeable. Meanwhile, AC-HNO3-Ar exhibits a high peak at 1580 cm-1 compared with the other two samples.
The C1s core-level spectra of the samples after the heat treatment in Ar are
TE D
shown in Fig. S2, and the area percentages of the various deconvoluted peaks are listed in Table 4. Compared with the results before the heat treatment (Table 2), Peak C-C of all the samples after the heat treatment increases, while Peak C-O-R decreases.
EP
The intensity of Peak –C=O keeps almost constant, and that of AC-HNO3 exhibits the
AC C
highest intensity among the three samples. Peaks O-C=O decrease slightly after the heat treatment.
The Raman spectra of the samples after the heat treatment in Ar are shown in Fig.
4. Generally, the Raman spectra of carbonaceous materials consist of one Gaussian-shaped band (D3, ~1500 cm-1) and four Lorentzian-shaped bands, G, D1, D2 and D4, at about 1580, 1350, 1620 and 1200 cm-1, respectively
[30]
. G band
commonly originates from the stretching vibration mode (E2g symmetry) in the ideal 10
ACCEPTED MANUSCRIPT graphitic lattices, while D1 and D2 bands reflect the structural defects located at graphene layer edges and graphene surface layers, respectively
[31]
. D3 band
corresponds to amorphous carbon, and D4 band is attributed to the disordered
RI PT
graphitic lattice and other impurities. The ratios between the intensities (I) of various bands are shown in Fig. 5. ID/IG and ID3/(IG + ID2 + ID3) values of all the samples are in the order of AC < AC-NaOH < AC-HNO3, while the values of IG/IAll are in the reverse
SC
order.
M AN U
The TGA results of the Ar-treated samples in air are exhibited in Fig. 6a. The oxidation of AC-HNO3 begins at about 250 oC, approximately 100 oC lower than the oxidation temperatures of AC and AC-NaOH. A slight increase of weight in the temperature region of 350-450 oC is observed in the results of both AC and AC-NaOH,
TE D
corresponding to the chemisorption process of oxygen, which is generally more obvious in the carbon with a lower reactivity [32]. The oxidation of AC-NaOH is faster than that of AC. Fig. 6b exhibits the oxidation products of various samples. The
EP
temperatures corresponding to the peaks of CO and CO2 both decrease in the order of
AC C
AC, AC-NaOH and AC-HNO3. CO dominates in the oxidation products of AC and AC-NaOH. On the contrary, the main oxidation product of AC-HNO3 is CO2. The TGA results of the Ar-treated samples in CO2 are shown in Fig. 7. For AC
and AC-NaOH, the weight loss corresponding to the reverse Boudouard reaction (C + CO2 → 2CO) begins at about 750 oC. In comparison, the reverse Boudouard reaction of AC-HNO3 starts at about 650 oC. The rate of weight loss of AC-HNO3 is much higher than those of AC and AC-NaOH at 800 oC. 11
ACCEPTED MANUSCRIPT 3.3. Single cell performance The cross-sectional morphology of the single cell is shown in Fig. 8. The thicknesses of the dense YSZ electrolyte layer, the porous GDC buffer layer and
RI PT
LSCF-GDC electrode layer are about 380, 7 and 16 µm, respectively. Close adherence is achieved between every two neighboring layers without any cracks (Fig. 8b).
I-V and I-P curves of the single cells with 5 ml min-1 (STP) N2 as the anode
SC
purge gas and the corresponding EIS results are shown in Fig. 9. At 750 oC (Fig. 9a),
M AN U
the cells with AC and AC-NaOH as the fuels exhibit similar open circuit voltages (OCVs) of about 0.90 V and Pmax of 53 mW cm-2. Sharp decreases of the current densities are observed at about 100 mA cm-2, indicating that the cell reactions are under diffusion control. For comparison, the cell fed with AC-HNO3 obtains higher
TE D
OCV and Pmax of 1.02 V and 85 mW cm-2, respectively, and no obvious diffusion-control characteristic is observed. When the temperature increases to 800 oC (Fig. 9b), the OCVs of all the cells show no noticeable change, while the Pmax of the
EP
cells with AC, AC-NaOH and AC-HNO3 as the fuels rise to 81, 82 and 128 mW cm-2,
AC C
respectively. The performances of the cells fed with AC and AC-NaOH are still controlled by diffusion at high current densities. Fig. 9c depicts the impedance spectra of the single cells at 750 oC. The cells with
different fuels have similar total ohmic resistances, which are shown as the high-frequency interrupts of the curves on the real axis. The arcs in the curves represent complicated electrode processes. The general approach is attributing the high-frequency arc (> 50 Hz) mainly to the charge transfer at the anode, the 12
ACCEPTED MANUSCRIPT intermediate-frequency arc (50-0.5 Hz) to the adsorption/desorption process at the anode, and the low-frequency arc (< 0.5 Hz) to the mass transfer at the anode
[33-35]
.
The characteristics of the cells with various fuels are similar in the high- and
RI PT
intermediate-frequency regions. On the contrary, the low-frequency arcs of the cells fed with AC and AC-NaOH are much larger than that of the cell with AC-HNO3, implying higher concentration polarizations of the AC and AC-NaOH anodes, which
SC
is in accordance with the I-V results (Fig. 9a). The ohmic and polarization resistances
M AN U
of all the cells decrease with the increase of the temperature to 800 oC (Fig. 9d). The high- and intermediate-frequency characteristics of all of the cells are still almost the same. Meanwhile, the anode concentration polarizations of AC and AC-NaOH are much higher than that of AC-HNO3.
TE D
Fig. 10a displays the I-V and I-P characteristics of the single cell fed with AC-HNO3 at 800 oC with different N2 flow rates in the anode chamber. When the flow rate of N2 increases from 5 to 100 ml min-1 (STP), the OCV keeps almost
EP
constant and the Pmax drops from 128 to 95 mW cm-2. The corresponding EIS results
AC C
are shown in Fig. 10b. The N2 flow rate shows no obvious effect on the ohmic resistance of the cell and the charge transfer polarization of the electrodes, but exhibits significant influence on the anode concentration polarization. The concentration polarization of the anode with 5 ml min-1 N2 flow is much lower than those with 50 and 100 ml min-1 N2 flows.
4. Discussion 4.1. The effects of NaOH and HNO3 treatments 13
ACCEPTED MANUSCRIPT It has been reported that the surface properties of carbon could be modified by the treatments with acids and bases
[19, 36]
. As listed in Table 1, after the NaOH
treatment, the oxygen content on the surface of AC is slightly increased. The obvious
RI PT
increase of the intensity of the band at 3440 cm-1 in the FTIR result (Fig. 1) and the C1s core-level spectrum (Table 2) indicate a noticeable increase of the hydroxyl groups, which agrees with previous works
[19, 27]
, and a slight increase of functional
SC
groups containing C=O and COO bands on the surface of AC-NaOH. Meanwhile, the
M AN U
graphitization degree of AC is reduced after the NaOH treatment, which is proved by the decrease of Peak C-C. The HNO3 treatment results in a higher oxygen content (Table 1) and a lower graphitization degree (Peak C-C in Table 2). The more oxygen content comes not only from the increased hydroxyl groups, but from the growth of
Table 2).
TE D
the functional groups with C=O band such as carbonyl and quinone groups (Fig. 1 and
4.2. The effects of heat treatment in Ar
EP
The TGA results shown in Fig. 2 reflect the release of the volatile components in
AC C
the AC fuels during the heating process of the single cell. Meanwhile, the volatiles released during the TGA tests analyzed with a mass spectrometer are shown in Fig. S3. The gradual weight loss below 400 oC is attributed to the release of absorbed water and some of the carboxylic groups, while the rapid loss of the weight in the temperature range of 400-800 oC is corresponding to the losing of some of the oxygen-containing functional groups, such as ether and phenols
[37]
. The weight loss
of AC-HNO3 is the largest when the temperature reaches 800 oC probably due to its 14
ACCEPTED MANUSCRIPT highest oxygen content (Table 1). After the heat treatment in Ar at 800 oC for 1 h, the volatiles in the AC samples release almost completely. Most of the volatiles in the AC samples are oxygen-containing functional groups.
RI PT
As a result, the oxygen content on the surface of the samples decreases remarkably after the heat treatment in Ar (Table 3). The amounts of carbon in C-O-R groups in AC-NaOH and AC-HNO3 decrease significantly after the heat treatment (Table 4).
SC
The FTIR spectra of all the samples after the heating process display similar peak
M AN U
intensities at 3440 cm-1 (Fig. 3), implying the loss of hydroxyl groups in AC-NaOH and AC-HNO3. Meanwhile, the peaks between 1700-1500 cm-1 of all the samples become weaker after the heat treatment probably due to the loss of carboxyl and ester groups (Peak O-C=O in Table 4). On the contrary, the constant intensity of Peak C=O
TE D
implies that the carbonyl and quinone groups remain stable during the heat treatment, which are consistent with the report that the decomposition of these groups in an inert atmosphere requires a higher temperature than those of hydroxyl and carboxyl groups . The amount of C=O band in AC-HNO3-Ar is much higher than those in AC-Ar
EP
[38]
AC C
and AC-NaOH-Ar (Table 4), which is also proved by the highest peaks of AC-HNO3-Ar at 1700 and 1580 cm-1 in the FTIR spectra (Fig. 3). It is widely believed that the complete oxidation of carbon at anodes of DCFCs involves the adsorption of two oxygen ions on the surface carbon in series accompanied with the release of electrons
[17]
. A high surface C=O content may facilitate the oxidation of
carbon and thus improve performance of cells. In general, heat treatment in an inert atmosphere would lead to an increase of the 15
ACCEPTED MANUSCRIPT graphitization degree of carbonaceous species
[23]
. In this work, the increase of Peak
C-C of the samples after the heat treatment (Table 4) indicates an increased graphitization degree probably due to the removal of the carbon in the
RI PT
oxygen-containing functional groups. For the samples after the heat treatment, AC possesses the highest graphitization degree, while AC-HNO3 exhibits the lowest degree. The graphitization degrees of the samples after the heat treatment are also
SC
evaluated based on the Raman spectra. G and D3 bands in Fig. 4 originate from the
M AN U
graphitic and amorphous structures of carbon, respectively. The ratios between the intensities of various bands are usually applied to evaluate the disorder degree of carbonaceous materials qualitatively. Typically, higher ID1/IG and lower IG/IAll values imply a higher disorder degree of the structure of carbon, and ID3/(IG + ID2 + ID3) is [39, 40]
. As shown in Fig. 5,
TE D
closely related to the content of amorphous carbon
compared with untreated AC, AC-NaOH and AC-HNO3 exhibit high ID1/IG and ID3/(IG + ID2 + ID3) values and low IG/IAll values, implying a lower graphitization degree of
EP
AC-NaOH and AC-HNO3. Among all the samples, AC-HNO3 possesses the highest
AC C
disorder degree of the structure, which is consistent with the XPS results. The TGA results of the Ar-treated samples in air (Fig. 6) reflect the reactivity of
the oxidation of the samples. The lowest oxidation temperature of AC-HNO3 indicates the highest reactivity of AC-HNO3 for oxidation attributed to its lowest graphitization degree and highest oxygen content. The high rate of oxidation of AC-NaOH compared with that of AC implies a higher reactivity of AC-NaOH. The temperatures corresponding to the peaks of the products in the TPO-MS results (Fig. 6b) are 16
ACCEPTED MANUSCRIPT another aspect reflecting the oxidation reactivity of the samples. AC-HNO3 with the highest oxidation reactivity exhibits the lowest temperatures corresponding to the peaks of CO and CO2. Meanwhile, the high contents of CO in the oxidation products
RI PT
of AC and AC-NaOH indicate an incomplete oxidation process, which would lead to a lower fuel efficiency of the DCFC. On the contrary, CO2 is the main oxidation product of AC-HNO3, implying the complete oxidation of the fuel due to its high
SC
oxidation reactivity. Similar to the TGA results in air, AC-HNO3 exhibits the lowest
M AN U
initial temperature and the highest rate of weight loss in the TGA results in CO2, which demonstrate that it also possesses the highest reactivity for the reverse Boudouard reaction. 4.3. Single cell performance
TE D
The reactivity of the carbon fuel in the anode has important effects on the OCV and the output power density of the single cell. Though the differences in the activation polarization among the anodes with various fuels are indistinguishable from
EP
the complex EIS results at 750 oC (Fig. 9c), the low OCVs of the cells fed with AC
AC C
and AC-NaOH (Fig. 9a) indicate their low anode activity. Since the reverse Boudouard reactions of AC and AC-NaOH just begin at about 750 oC (Fig. 7), the direct oxidation of the carbon fuels should be the main reaction at these anodes. The cell fed with AC-HNO3 possesses the highest OCV and Pmax partly due to the highest oxidation reactivity of the fuel (Fig. 6). Meanwhile, the CO generated by the reverse Boudouard reaction could be also involved in the anodic reaction, resulting in the lowest concentration polarization resistance. When the temperature rises to 800 oC, 17
ACCEPTED MANUSCRIPT the reverse Boudouard reaction is involved in the anode reactions of the cells with AC and AC-NaOH as the fuels, bringing about higher limited current densities. Meanwhile, the cell fed with AC-HNO3 still exhibits the lowest polarization resistance
RI PT
and the highest Pmax of 128 mW cm-2, which is superior to most of the electrolyte-supported DCFCs with similar configurations reported previously (Table S1).
SC
The CO generated through the reverse Boudouard reaction could be diluted by
M AN U
the inert purge gas at the anode, leading to a lower performance of the cell. The decrease of the Pmax of the cell with the increase of the flow rate of N2 from 5 to 50 ml min-1 in the anode chamber (Fig. 10a) is mainly due to the insufficient supply of CO to the active sites at the anode, which is proved by the obvious diffusion limitation at
TE D
higher current densities accompanied with a much larger low-frequency arc (Fig. 10b). A similar behavior has been observed in SO-DCFCs with demineralized coal as the fuel and high purity N2 as purge gas.[34, 37] However, no obvious further increase of the
EP
concentration polarization is observed when the flow rate of N2 rises to 100 ml min-1,
AC C
and it is generally believed that the contribution of the electrochemical oxidation of CO to the performance of the cell is negligible when the flow rate of the anode purge gas reaches 100 ml min-1 [12, 41]. Therefore, it is reasonable to attribute the Pmax of 95 mW cm-2 of the cell with the 100 ml min-1 N2 flow to the direct electrochemical oxidation of the AC-HNO3 at the anode.
5. Conclusions In this work, the structure and surface properties of AC are modified by HNO3 18
ACCEPTED MANUSCRIPT and NaOH treatments. The content of oxygen on the surface of AC increases considerably after the treatment with HNO3. Compared with untreated AC, AC-NaOH and AC-HNO3 both possess a low graphitization degree and a high amount of C-O-R
RI PT
functional groups. Meanwhile, AC-HNO3 has more carbonyl and quinone groups than other samples. After the heating process in Ar atmosphere, the amount of C-O-R groups decreases significantly, leading to a reduction of oxygen content on the carbon
SC
surface. However, the C=O groups on the surface of AC-HNO3 are more stable. The
M AN U
lowest graphitization degree and highest oxygen content of AC-HNO3 result in its highest reactivity for oxidation and reverse Boudouard reactions. The single cell fed with AC-HNO3 exhibits the highest Pmax of 128 mW cm-2 and lowest polarization resistance at 800 oC. The increase of the N2 flow rate at the anode impedes the reverse
Acknowledgments
TE D
Boudouard reaction, leading to a decrease of the performance of the cell.
The financial support of NSF of China under contract numbers 51402210 and the
EP
support of Tianjin Municipal Science and Technology Commission under contract
AC C
numbers 15JCQNJC06500 are gratefully acknowledged. The work has been also supported by the Program of Introducing Talents to the University Disciplines under file number B06006, and the Program for Changjiang Scholars and Innovative Research Teams in Universities under file number IRT 0641. References [1] C. Jiang, J. Ma, G. Corre, S.L. Jain, J.T.S. Irvine, Challenges in developing direct carbon fuel cells, Chem. Soc Rev. 46 (2017) 2889-2912. [2] T.M. Gur, Critical review of carbon conversion in "carbon fuel cells", Chem. Rev. 113 (2013) 6179-6206.
19
ACCEPTED MANUSCRIPT [3] J.F. Cooper, J.R. Selman, Analysis of the carbon anode in direct carbon conversion fuel cells, Int. J. Hydrogen Energy 37 (2012) 19319-19328. [4] W. Hao, X. He, Y. Mi, Achieving high performance in intermediate temperature direct carbon fuel cells with renewable carbon as a fuel source, Appl. Energy 135 (2014) 174-181. [5] A.C. Rady, S. Giddey, A. Kulkarni, S.P.S. Badwal, S. Bhattacharya, Catalytic gasification of carbon in a direct carbon fuel cell, Fuel 180 (2016) 270-277. [6] Y. Zhao, C. Xia, L. Jia, Z. Wang, H. Li, J. Yu, Y. Li, Recent progress on solid oxide fuel cell:
RI PT
Lowering temperature and utilizing non-hydrogen fuels, Int. J. Hydrogen Energy 38 (2013) 16498-16517.
[7] A. Jayakumar, R.K. Ungas, S. Roy, A. Javadekar, D.J. Buttrey, J.M. Vohs, R.J. Gorte, A direct carbon fuel cell with a molten antimony anode, Energy Environ. Sci. 4 (2011) 4133-4137.
[8] B. Yu, Y. Zhao, Y. Li, A SnO2-samarium doped ceria additional anode layer in a direct carbon fuel
SC
cell, J. Power Sources 306 (2016) 387-393.
[9] S. Nurnberger, R. Bussar, P. Desclaux, B. Franke, M. Rzepka, U. Stimming, Direct carbon conversion in a SOFC-system with a non-porous anode, Energy Environ. Sci. 3 (2010) 150-153. Solid State Ionics 271 (2015) 121-127.
M AN U
[10] M. Dudek, On the utilization of coal samples in direct carbon solid oxide fuel cell technology, [11] M.T. Mehran, T.-H. Lim, S.-B. Lee, J.-W. Lee, S.-J. Park, R.-H. Song, Long-term performance degradation study of solid oxide carbon fuel cells integrated with a steam gasifier, Energy 113 (2016) 1051-1061.
[12] A. Kulkarni, F.T. Ciacchi, S. Giddey, C. Munnings, S.P.S. Badwal, J.A. Kimpton, D. Fini, Mixed ionic electronic conducting perovskite anode for direct carbon fuel cells, Int. J. Hydrogen Energy 37 (2012) 19092-19102.
TE D
[13] Y. Leng, S. Chan, Q. Liu, Development of LSCF–GDC composite cathodes for low-temperature solid oxide fuel cells with thin film GDC electrolyte, Int. J. Hydrogen Energy 33 (2008) 3808-3817. [14] X. Wu, X. Zhou, Y. Tian, X. Kong, J. Zhang, W. Zuo, X. Ye, Stability and electrochemical performance of lanthanum ferrite-based composite SOFC anodes in hydrogen and carbon monoxide, Electrochim. Acta 208 (2016) 164-173.
EP
[15] J. Xiao, D. Han, F. Yu, L. Zhang, J. Liu, Z. Zhan, Y. Zhang, P. Dong, Characterization of symmetrical SrFe0.75Mo0.25O3−δ electrodes in direct carbon solid oxide fuel cells, J. Alloy. Compd. 688 (2016) 939-945.
AC C
[16] S. Giddey, A. Kulkarni, C. Munnings, S.P.S. Badwal, Performance evaluation of a tubular direct carbon fuel cell operating in a packed bed of carbon, Energy 68 (2014) 538-547. [17] N.J. Cherepy, R. Krueger, K.J. Fiet, A.F. Jankowski, J.F. Cooper, Direct Conversion of Carbon Fuels in a Molten Carbonate Fuel Cell, J. Electrochem Soc. 152 (2005) A80-A87. [18] X. Li, Z.H. Zhu, J.L. Chen, R. De Marco, A. Dicks, J. Bradley, G.Q. Lu, Surface modification of carbon fuels for direct carbon fuel cells, J. Power Sources 186 (2009) 1-9. [19] D.X. Cao, G.L. Wang, C.Q. Wang, J. Wang, T.H. Lu, Enhancement of electrooxidation activity of activated carbon for direct carbon fuel cell, Int. J. Hydrogen Energy 35 (2010) 1778-1782. [20] J.B. Zhang, Z.P. Zhong, D.K. Shen, J.X. Zhao, H.Y. Zhang, M. Yang, W.L. Li, Preparation of Bamboo-Based Activated Carbon and Its Application in Direct Carbon Fuel Cells, Energy & Fuels 25 (2011) 2187-2193. [21] S. Eom, S. Ahn, K. Kang, G. Choi, Correlations between electrochemical resistances and surface properties of acid-treated fuel in coal fuel cells, Energy 140 (2017) 885-892.
20
ACCEPTED MANUSCRIPT [22] X. Li, Z.H. Zhu, R. De Marco, J. Bradley, A. Dicks, Modification of Coal as a Fuel for the Direct Carbon Fuel Cell, J. Phys. Chem. A 114 (2010) 3855-3862. [23] A. Fuente-Cuesta, C. Jiang, A. Arenillas, J.T.S. Irvine, Role of coal characteristics in the electrochemical behaviour of hybrid direct carbon fuel cells, Energy Environ. Sci. 9 (2016) 2868-2880. [24] P. Zeng, R. Ran, Z. Chen, H. Gu, Z. Shao, J. Dacosta, S. Liu, Significant effects of sintering temperature on the performance of La0.6Sr0.4Co0.2Fe0.8O3−δ oxygen selective membranes, J. Membrane Science 302 (2007) 171-179.
RI PT
[25] L. Jia, Y. Tian, Q. Liu, C. Xia, J. Yu, Z. Wang, Y. Zhao, Y. Li, A direct carbon fuel cell with (molten carbonate)/(doped ceria) composite electrolyte, J. Power Sources 195 (2010) 5581-5586.
[26] Y. Jiao, J. Zhao, W. An, L. Zhang, Y. Sha, G. Yang, Z. Shao, Z. Zhu, S.-D. Li, Structurally modified coal char as a fuel for solid oxide-based carbon fuel cells with improved performance, J. Power Sources 288 (2015) 106-114. Adsorptive Properties, Langmuir (2004) 2233-2242.
SC
[27] J.P. Chen, Acid/Base-Treated Activated Carbons Characterization of Functional Groups and Metal [28] A.-N.A. El-Hendawy, Influence of HNO3 oxidation on the structure and adsorptive properties of corncob-based activated carbon, Carbon 41 (2003) 713-722.
M AN U
[29] J.-H. Zhou, Z.-J. Sui, J. Zhu, P. Li, D. Chen, Y.-C. Dai, W.-K. Yuan, Characterization of surface oxygen complexes on carbon nanofibers by TPD, XPS and FT-IR, Carbon 45 (2007) 785-796. [30] A. Sadezky, H. Muckenhuber, H. Grothe, R. Niessner, U. Pöschl, Raman microspectroscopy of soot and related carbonaceous materials: Spectral analysis and structural information, Carbon 43 (2005) 1731-1742.
[31] A.M. Rao, J. Chen, E. Richter, U. Schlecht, P.C. Eklund, R.C. Haddon, U.D. Venkateswaran, Y.K. Kwon, D. Tomanek, Effect of van der Waals interactions on the Raman modes in single walled carbon
TE D
nanotubes, Physical review letters 86 (2001) 3895-3898.
[32] S. Kizgut, S. Yilmaz, Characterization and non-isothermal decomposition kinetics of some Turkish bituminous coals by thermal analysis, Fuel Process. Technol. 85 (2003) 103-111. [33] A.C. Rady, S. Giddey, A. Kulkarni, S.P.S. Badwal, S. Bhattacharya, Direct Carbon Fuel Cell Operation on Brown Coal with a Ni-GDC-YSZ Anode, Electrochim. Acta 178 (2015) 721-731.
EP
[34] A.C. Rady, S. Giddey, A. Kulkarni, S.P.S. Badwal, S. Bhattacharya, Degradation Mechanism in a Direct Carbon Fuel Cell Operated with Demineralised Brown Coal, Electrochim. Acta 143 (2014) 278-290.
AC C
[35] S. Giddey, A. Kulkarni, C. Munnings, S.P.S. Badwal, Composite anodes for improved performance of a direct carbon fuel cell, J. Power Sources 284 (2015) 122-129. [36] X. Wang, N. Gao, Q. Zhou, H. Dong, H. Yu, Y. Feng, Acidic and alkaline pretreatments of activated carbon and their effects on the performance of air-cathodes in microbial fuel cells, Bioresour Technol. 144 (2013) 632-636. [37] A.C. Rady, S. Giddey, A. Kulkarni, S.P.S. Badwal, S. Bhattacharya, B.P. Ladewig, Direct carbon fuel cell operation on brown coal, Appl. Energy 120 (2014) 56-64. [38] J.L. Figueiredo, M.F.R. Pereira, M.M.A. Freitas, J.J.M. Órfão, Modification of the surface chemistry of activated carbons, Carbon 37 (1999) 1379-1389. [39] P. Li, Y. Zhao, B. Yu, J. Li, Y. Li, Improve electrical conductivity of reduced La2Ni0.9Fe0.1O4+δ as the anode of a solid oxide fuel cell by carbon deposition, Int. J. Hydrogen Energy 40 (2015) 9783-9789. [40] C. Sheng, Char structure characterised by Raman spectroscopy and its correlations with
21
ACCEPTED MANUSCRIPT combustion reactivity, Fuel 86 (2007) 2316-2324. [41] H. Jang, Y. Park, J. Lee, Meticulous insight on the state of fuel in a solid oxide carbon fuel cell,
AC C
EP
TE D
M AN U
SC
RI PT
Chem. Eng. J. 308 (2017) 974-979.
22
ACCEPTED MANUSCRIPT
Table 1. XPS elemental analysis results of the raw and treated AC. C (at.%)
O (at.%)
O/C
AC
88.9
10.1
0.11
AC-NaOH
87.5
10.7
0.12
AC-HNO3
84.5
14.9
RI PT
Sample
AC C
EP
TE D
M AN U
SC
0.18
23
ACCEPTED MANUSCRIPT
Table 2. The area percentages of the deconvoluted peaks in the C1s spectra of raw and treated AC samples. The area percentages of the peaks of carbon (%) Sample C-O-R
-C=O
AC
81
9
4
6
AC-NaOH
76
12
5
7
AC-HNO3
72
13
8
7
SC M AN U TE D EP AC C 24
O-C=O
RI PT
C-C
ACCEPTED MANUSCRIPT
Table 3. XPS elemental analysis results of the samples after the heat treatment in Ar. C (at.%)
O (at.%)
O/C
AC-Ar
94.9
5.0
0.05
AC-NaOH-Ar
94.3
5.1
0.05
AC-HNO3-Ar
91.8
7.9
RI PT
Sample
AC C
EP
TE D
M AN U
SC
0.09
25
ACCEPTED MANUSCRIPT
Table 4. The area percentages of the deconvoluted peaks in the C1s spectra of raw and treated AC samples after the heat treatment in Ar. The area percentages of the peaks of carbon (%) Sample C-O-R
-C=O
AC-Ar
83
8
4
5
AC-NaOH-Ar
80
9
5
6
AC-HNO3-Ar
76
9
9
6
SC M AN U TE D EP AC C 26
O-C=O
RI PT
C-C
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
Fig. 1. FTIR spectra of the raw and treated AC.
27
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
Fig. 2. TGA curves of AC samples in Ar atmosphere.
28
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
Fig. 3. FTIR spectra of the raw AC and treated AC after the heat treatment in Ar.
29
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Fig. 4. Raman spectra of (a) AC, (b) AC-NaOH and (c) AC-HNO3 after the heat treatment in Ar. 30
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
Fig. 5. Band area ratios of AC samples after the heat treatment in Ar.
31
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Fig. 6. (a) TGA curves in air and (b) TPO-MS results of AC samples after the heat treatment in Ar.
32
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
Fig. 7. TGA curves in CO2 of AC samples after the heat treatment in Ar.
33
RI PT
ACCEPTED MANUSCRIPT
SC
Fig. 8. SEM cross-sectional micrographs of (a) the single cell and (b) the
AC C
EP
TE D
M AN U
electrolyte-electrode interface.
34
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
35
SC
RI PT
ACCEPTED MANUSCRIPT
Fig. 9. I-V and I-P curves of single cells with various carbon fuels and 5 ml min-1 N2
M AN U
as the anode purge gas at (a) 750 oC and (b) 800 oC and the corresponding EIS results
AC C
EP
TE D
at (c) 750 oC and (d) 800 oC.
36
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
EP
Fig. 10. (a) I-V and I-P curves and (b) the corresponding EIS of the single cell with
AC C
AC-HNO3 as the fuel at 800 oC with different flow rates of the anode purge gas.
37
ACCEPTED MANUSCRIPT
Research Highlights ► Activated carbon is treated with HNO3 and NaOH as the fuel of DCFCs ► AC-HNO3 shows the lowest graphitization degree and highest surface oxygen
RI PT
content ► AC-HNO3 shows the highest reactivity for oxidation and reverse Boudouard reactions
AC C
EP
TE D
M AN U
SC
► The cell supported by a YSZ layer exhibits the Pmax of 128 mW cm-2 at 800 oC
S0013-4686(18)31714-6
DOI:
10.1016/j.electacta.2018.07.196
Reference:
EA 32393
To appear in:
Electrochimica Acta
Received Date: 2 April 2018 Revised Date:
12 June 2018
Accepted Date: 26 July 2018
Please cite this article as: L. Fan, J. Wang, L. Zhao, N. Hou, T. Gan, X. Yao, P. Li, Y. Zhao, Y. Li, Effects of surface modification on the reactivity of activated carbon in direct carbon fuel cells, Electrochimica Acta (2018), doi: 10.1016/j.electacta.2018.07.196. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Effects of Surface Modification on the Reactivity of Activated Carbon in Direct Carbon Fuel Cells Lijun Fan1,2, Jun Wang1,2, Linzhe Zhao1,2, Nianjun Hou1,2, Tian Gan1,2, Xueli Yao1,2, Ping Li1,2, Yicheng Zhao1,2*, Yongdan Li1,2,3* State Key Laboratory of Chemical Engineering (Tianjin University), Tianjin Key
RI PT
1
Laboratory of Applied Catalysis Science and Technology, School of Chemical Engineering, Tianjin University, Tianjin 300072, China
Collaborative Innovation Center of Chemical Science and Engineering (Tianjin),
SC
2
Tianjin, 300072, China
Department of Chemical and Metallurgical Engineering, Aalto University,
M AN U
3
Kemistintie 1, FI-00076 Aalto, Finland Abstract
Activated carbons (AC) pretreated with HNO3 and NaOH are investigated as the
TE D
fuel of direct carbon fuel cells. HNO3 and NaOH treatments both increase the oxygen content and decrease the graphitization degree of AC. The amount of hydroxyl groups
EP
on the surface of AC increases after the treatments with HNO3 and NaOH, and then decreases during the subsequent heating process in an inert atmosphere. On the
AC C
contrary, the carbonyl and quinone groups on the surface of the activated carbon remain stable during the heating process. The activated carbon treated with HNO3 shows the highest reactivity towards oxidation and reverse Boudouard reactions due to its lowest graphitization degree and highest oxygen content. The single cell with a
∗
Corresponding author. Email: [email protected]; Tel: +86-22-27405613; Fax: +86-22-27405243 (Y. Zhao) ∗ Corresponding author. Email: [email protected], [email protected]; Tel: +86-22-27405613; Fax: +86-22-27405243 (Y. Li) 1
ACCEPTED MANUSCRIPT 380 µm-thick yttria stabilized zirconia layer as the electrolyte and the activated carbon treated with HNO3 as the fuel exhibits the lowest polarization and the highest
RI PT
maximum power density of 128 mW cm-2 at 800 oC.
Keywords: Direct carbon fuel cell; Activated carbon; Surface modification;
AC C
EP
TE D
M AN U
SC
Electrochemical oxidation; Oxygen functional groups
2
ACCEPTED MANUSCRIPT
1. Introduction A direct carbon fuel cell (DCFC) converts the chemical energy stored in solid carbon, which could be obtained from coal and biomass, into electricity directly via [1, 2]
. The energy efficiency of traditional coal-fired power
RI PT
an electrochemical route
generation processes involving Carnot cycles is only 35-40% in general, while that of DCFCs could theoretically exceed 100% since the overall reaction (C+O2→CO2)
SC
possesses a slightly positive entropy change (1.6 J mol-1 K-1 at 600 oC) [3]. Meanwhile,
M AN U
the anodic product of DCFCs could be pure CO2, which simplifies the capture and sequestration processes of CO2 for downstream utilization[4]. Besides, the activity of solid carbon fuel keeps unity during the anodic reaction regardless of the extent of conversion, ensuring the high overall practical energy efficiency (about 80%) of
TE D
DCFCs [5].
The basic structure of a DCFC consists of three active components: a fuel electrode (anode), an oxidant electrode (cathode) and an electrolyte sandwiched
EP
between them. Based on the electrolytes used, DCFCs are categorized into three main
AC C
classes: molten hydroxide, molten carbonate and solid oxide DCFCs (SO-DCFCs). A comprehensive description of DCFCs classification could be found elsewhere
[1, 2, 6]
.
One of the key challenges for the practical application of DCFCs is their sluggish anodic kinetics, which is mainly due to the insufficient contact between the solid carbon fuel and the solid anode layer. Liquid media such as molten carbonates and metals have been utilized at the anode of DCFCs as shuttles of carbonate and oxygen ions between the carbon particles and the electrolyte layer, which enhance the 3
ACCEPTED MANUSCRIPT performance of the cells effectively [7, 8]. However, the hot caustic liquids decrease the stability of the cells simultaneously
[8]
. On the contrary, SO-DCFCs with an
all-solid-state structure without corrosive media fundamentally eliminate the
RI PT
corrosion of materials, showing a promising durability and feasibility to scale-up [9-11]. The major issue of SO-DCFCs is the difficulty in getting the solid fuel to the anode reaction zone or triple phase boundaries (TPBs) [12]. Recently, mixed ionic-electronic
[12-15]
. A tubular DCFC supported by a yttria
M AN U
area and promote the anodic kinetics
SC
conducting materials have been used as the anodes of SO-DCFCs to enlarge the TPB
stabilized zirconia (YSZ) electrolyte layer with a La0.6Sr0.4Co0.2Fe0.8O3-δ (LSCF) anode exhibited a maximum power density (Pmax) of 30 mW cm-2 at 800 oC with CO2 as the anodic purge gas [16]. The performance dropped by 40% after 11-day operation.
with solid anodes.
TE D
Nevertheless, a substantial improvement of the anodic activity is required by DCFCs
The structure and surface property of carbon influence the anodic kinetics of [6]
. Cherepy et al.[17] examined nine carbon samples in molten
EP
DCFCs significantly
AC C
carbonate, and found that crystallographic disorder degree and electrical conductivity were main factors determining the rate of discharge. Li et al.[18] pretreated activated carbon (AC) with HNO3, which increased oxygen-containing surface functional groups and improved the electrochemical reactivity of the carbon consequently. Cao et al.[19] found that both acid and base pretreatments could increase the specific surface area and microspore volume of AC. Zhang et al.[20] observed the increase of C=O and O-H groups and the disappearance of C=C groups on the surface of a 4
ACCEPTED MANUSCRIPT biomass-based AC after the treatment in HNO3. Eom et al.[21] demonstrated that the electrochemical resistance of raw coal is influenced by its surface properties such as the degree of oxidation and ash composition, which could be modified by acid
RI PT
treatments. However, the surface properties of carbon fuels were studied ex situ in most of the previous literatures, while the anodic process of DCFCs is more complex. For instance, Li et al.[22] proved that the oxygen-containing functional groups on the
SC
surface of coal decreased sharply after heat-treatment. Fuente-Cuesta et al.[23] found
M AN U
that the heat-treatment of bituminous coal led to higher structural order of carbon and lower H/C ratio, resulting in the decrease of anodic reactivity. Therefore, the alterations of the structure and surface properties of carbon fuels those affect the anodic activity of DCFCs have been not demonstrated clearly, which requires further
TE D
in-depth studies.
In this work, AC is pre-treated with HNO3 and NaOH, respectively, and the variations of its structure and surface properties before and after heat treatment are
EP
studied. The effects of surface modification of AC on the electrochemical
AC C
performances of the anode and the corresponding single cell are systematically investigated.
2. Experimental
2.1. Preparation and characterization of carbon fuels Base and acid modifications of carbon were carried out by immersing 5 g commercial AC (Aladdin, > 200 mesh) into 30 ml 2 mol L-1 NaOH and HNO3 solutions, respectively, at room temperature for 20 h. Then the samples were washed 5
ACCEPTED MANUSCRIPT with distilled water for three times. The concentrations of H+ and OH- in the acid and base solutions before and after the treatments are listed in Table S1. The washed samples were dried at 105 oC in air for 12 h, which are marked as AC-NaOH and
RI PT
AC-HNO3, respectively. The surface property of the AC samples before and after pretreatments was studied with a Fourier transform infrared (FTIR) spectrometer (Perkin-Elmer) in the
SC
region of 4000-400 cm−1. X-ray photoemission spectroscopy (XPS) was conducted
M AN U
using Escalab 250Xi system (ThermoFisher Scientific USA) with monochromatic 150 W Al-Kα radiation as the excitation source. Binding energies were calibrated to the common C1s electron binding energy at 284.8 eV.
Thermogravimetric analysis (TGA) of the AC samples were performed with an
TE D
analyzer (STA449 F3, Netzsch) in an Ar flow (50 ml min−1, STP) from room temperature to 800 oC at a heating rate of 5 oC min-1, and then the temperature was kept at 800 oC for 1 h. When the temperature cooled down to room temperature, the
EP
remaining samples were collected, and designated as AC-Ar, AC-NaOH-Ar and
AC C
AC-HNO3-Ar, respectively. Surface properties of the remaining samples were analyzed using FTIR and XPS. Raman measurements were performed with a Raman spectrometer (Renishaw inVia) with a 633 nm He-Ne laser excitation source. After the TGA in Ar, the gasification processes of the residues were also investigated with TGA in air and CO2, respectively, from room temperature to 800 oC at a heating rate of 5 oC min-1. The TGA in air is combined with an online mass spectrometer (HPR20, Hiden) to measure the amounts of CO and CO2 in the outlet gas. 6
ACCEPTED MANUSCRIPT 2.2. Single cell test Electrolyte-supported single cells were fabricated for electrochemical tests in this work. Commercial YSZ powder (Ningbo SOFCMAN Energy Technology Co., Ltd,
RI PT
China) was uniaxially pressed at 500 MPa, and then sintered at 1500 oC for 4 h in air. The YSZ electrolyte layer has a thickness of about 380 µm. Commercial Ce0.9Gd0.1O1.95 (GDC) powder (Ningbo SOFCMAN Energy Technology Co., Ltd,
SC
China) was mixed with a binder (V006, Heraeus Ltd.) to form a slurry, which was
M AN U
then screen-printed on both sides of the YSZ electrolyte pellet and sintered at 1250 oC in air for 2 h to form 7 µm-thick buffer layers to avoid the potential interfacial reaction between the YSZ electrolyte and the electrode materials at high temperatures. GDC powders were thoroughly mixed with LSCF powder at a weight ratio of 5:5 by
TE D
ball-milling as electrode materials. The LSCF powder was synthesized through an EDTA-citrate sol-gel method, and the detailed process could be found in a previous work
[24]
. The LSCF-GDC composite powder was also made into slurry,
EP
screen-printed on the GDC buffer layers, and then sintered at 1080 oC for 2 h in air.
AC C
The average thickness of the LSCF-GDC electrode layers was about 16 µm. The cross-sectional morphology of the single cell was observed with a Hitachi S-4800 scanning electron microscope (SEM). Ag paste was printed on both sides of the cell as a current collector. The geometric surface area of each electrode was 0.72 cm2. The configuration of the single cell was described in our previous work
[25]
.
Before the test, 0.1 g carbon fuel was put in the anode cavity with N2 as the purge gas (5 ml min-1, STP). The cathode gas was O2 (50 ml min-1, STP). The I-V and I-P curves 7
ACCEPTED MANUSCRIPT and electrochemical impedance spectra (EIS) were recorded at 750 and 800 oC with various flow rates of N2 using an electrochemical workstation (Versastat 3, Princeton Applied Research). The EIS spectra were obtained at open circuit state in the
RI PT
frequency range of 100 kHz-50 mHz.
3. Results 3.1. Characterization of AC samples
SC
The contents of carbon and oxygen on the surface of raw AC determined by XPS
M AN U
are 88.9 and 10.1 at.%, respectively (Table 1). After AC is treated with NaOH, the content of oxygen increases to 10.7 at.% with a little decrease of that of carbon. AC-HNO3 possesses the highest oxygen content and O/C ratio of 14.9 at.% and 0.18, respectively. The results of elemental analysis of the AC samples before and after acid
TE D
and base treatments show a similar tendency (Table S2).
The FTIR spectra of the AC samples are shown in Fig. 1. The spectrum of raw AC reveals the following absorption bands: the band at 3440 cm-1 assigned to O-H
EP
stretching vibration in hydroxyl functional groups; the bands at 2922 and 2845 cm-1
AC C
due to CH2 and CH3 asymmetric and symmetric stretching vibrations in alkyl chains [26]
; the band at 1700 cm-1 attributed to C=O stretching vibration in ketone, carboxylic
acid or ester
[20]
; the band at 1600 cm-1 ascribed to C=C stretching vibration in
aromatic rings or C=O stretching vibration in carbonyls [27] and the broad band within the range of 1300–1000 cm-1 associated with various function groups containing C–O band such as ethers, phenols and hydroxyl groups
[28]
. After the NaOH treatment, the
intensity of the band at 3440 cm-1 increases significantly, indicating an increase of the 8
ACCEPTED MANUSCRIPT amount of hydroxyl groups. Meanwhile, a small new band at 1050 cm-1 appears. The modification with HNO3 leads to not only increase of bands at 3440 and 1700 cm-1 but the appearance of new bands at 1533 and 1337 cm-1, which originate from highly
RI PT
conjugated C=O and C-O stretching vibration, respectively, in carboxylic groups and carboxylate moieties [28].
The C1s core-level spectra of the AC samples are shown in Fig. S1. Four
SC
deconvoluted peaks are obtained from each curve, corresponding to graphitic carbon
M AN U
(C-C, ~284.8 eV), carbon in alcohol, phenolic or ether (C-O-R, ~286.2 eV), carbonyl or quinone groups (-C=O, ~287.3 eV) and carboxyl or ester groups (O-C=O, ~289.0 eV), respectively [29]. The area percentages of the peaks are listed in Table 2. After the treatments with NaOH and HNO3, the intensity of Peak C-C decreases remarkably,
TE D
while that of Peak C-O-R increases noticeably. The intensities of Peaks –C=O and O-C=O slightly increase after the AC is treated with NaOH. In contrast, obvious elevation of peak –C=O is observed in the result of AC-HNO3.
EP
The TGA results of the untreated and treated AC samples in Ar are illustrated in
AC C
Fig. 2. The weights of all the samples decrease gradually before the temperature reaches 400 oC, and AC-HNO3 shows the highest weight loss rate. In the temperature range of 400-800 oC, all of the samples exhibit a rapid loss of the weight. The samples are kept at 800 oC for 1 h, during which period their weights are nearly stable. The weight loss of the samples shows an order of AC-HNO3 > AC-NaOH > AC. 3.2. Characterization of AC samples after the heat treatment in Ar The contents of carbon and oxygen on the surface of the samples after the 9
ACCEPTED MANUSCRIPT Ar-TGA test are listed in Table 3. All of the samples possess a low oxygen content compared with the samples before the heat treatment. The AC-HNO3 still exhibits the highest oxygen content after the heat treatment.
RI PT
Fig. 3 shows the FTIR spectra of the AC samples after the heat treatment in Ar. The peaks at 3440 cm-1 of AC-NaOH and AC-HNO3 decrease after the heat treatment, indicating the reduction of the amount of hydroxyl groups. The three samples after the
SC
heat treatment show similar contents of hydroxyl groups. The peaks at 1700 cm-1 in
M AN U
the spectra of AC-Ar and AC-NaOH-Ar are negligible, while that in AC-HNO3-Ar is noticeable. Meanwhile, AC-HNO3-Ar exhibits a high peak at 1580 cm-1 compared with the other two samples.
The C1s core-level spectra of the samples after the heat treatment in Ar are
TE D
shown in Fig. S2, and the area percentages of the various deconvoluted peaks are listed in Table 4. Compared with the results before the heat treatment (Table 2), Peak C-C of all the samples after the heat treatment increases, while Peak C-O-R decreases.
EP
The intensity of Peak –C=O keeps almost constant, and that of AC-HNO3 exhibits the
AC C
highest intensity among the three samples. Peaks O-C=O decrease slightly after the heat treatment.
The Raman spectra of the samples after the heat treatment in Ar are shown in Fig.
4. Generally, the Raman spectra of carbonaceous materials consist of one Gaussian-shaped band (D3, ~1500 cm-1) and four Lorentzian-shaped bands, G, D1, D2 and D4, at about 1580, 1350, 1620 and 1200 cm-1, respectively
[30]
. G band
commonly originates from the stretching vibration mode (E2g symmetry) in the ideal 10
ACCEPTED MANUSCRIPT graphitic lattices, while D1 and D2 bands reflect the structural defects located at graphene layer edges and graphene surface layers, respectively
[31]
. D3 band
corresponds to amorphous carbon, and D4 band is attributed to the disordered
RI PT
graphitic lattice and other impurities. The ratios between the intensities (I) of various bands are shown in Fig. 5. ID/IG and ID3/(IG + ID2 + ID3) values of all the samples are in the order of AC < AC-NaOH < AC-HNO3, while the values of IG/IAll are in the reverse
SC
order.
M AN U
The TGA results of the Ar-treated samples in air are exhibited in Fig. 6a. The oxidation of AC-HNO3 begins at about 250 oC, approximately 100 oC lower than the oxidation temperatures of AC and AC-NaOH. A slight increase of weight in the temperature region of 350-450 oC is observed in the results of both AC and AC-NaOH,
TE D
corresponding to the chemisorption process of oxygen, which is generally more obvious in the carbon with a lower reactivity [32]. The oxidation of AC-NaOH is faster than that of AC. Fig. 6b exhibits the oxidation products of various samples. The
EP
temperatures corresponding to the peaks of CO and CO2 both decrease in the order of
AC C
AC, AC-NaOH and AC-HNO3. CO dominates in the oxidation products of AC and AC-NaOH. On the contrary, the main oxidation product of AC-HNO3 is CO2. The TGA results of the Ar-treated samples in CO2 are shown in Fig. 7. For AC
and AC-NaOH, the weight loss corresponding to the reverse Boudouard reaction (C + CO2 → 2CO) begins at about 750 oC. In comparison, the reverse Boudouard reaction of AC-HNO3 starts at about 650 oC. The rate of weight loss of AC-HNO3 is much higher than those of AC and AC-NaOH at 800 oC. 11
ACCEPTED MANUSCRIPT 3.3. Single cell performance The cross-sectional morphology of the single cell is shown in Fig. 8. The thicknesses of the dense YSZ electrolyte layer, the porous GDC buffer layer and
RI PT
LSCF-GDC electrode layer are about 380, 7 and 16 µm, respectively. Close adherence is achieved between every two neighboring layers without any cracks (Fig. 8b).
I-V and I-P curves of the single cells with 5 ml min-1 (STP) N2 as the anode
SC
purge gas and the corresponding EIS results are shown in Fig. 9. At 750 oC (Fig. 9a),
M AN U
the cells with AC and AC-NaOH as the fuels exhibit similar open circuit voltages (OCVs) of about 0.90 V and Pmax of 53 mW cm-2. Sharp decreases of the current densities are observed at about 100 mA cm-2, indicating that the cell reactions are under diffusion control. For comparison, the cell fed with AC-HNO3 obtains higher
TE D
OCV and Pmax of 1.02 V and 85 mW cm-2, respectively, and no obvious diffusion-control characteristic is observed. When the temperature increases to 800 oC (Fig. 9b), the OCVs of all the cells show no noticeable change, while the Pmax of the
EP
cells with AC, AC-NaOH and AC-HNO3 as the fuels rise to 81, 82 and 128 mW cm-2,
AC C
respectively. The performances of the cells fed with AC and AC-NaOH are still controlled by diffusion at high current densities. Fig. 9c depicts the impedance spectra of the single cells at 750 oC. The cells with
different fuels have similar total ohmic resistances, which are shown as the high-frequency interrupts of the curves on the real axis. The arcs in the curves represent complicated electrode processes. The general approach is attributing the high-frequency arc (> 50 Hz) mainly to the charge transfer at the anode, the 12
ACCEPTED MANUSCRIPT intermediate-frequency arc (50-0.5 Hz) to the adsorption/desorption process at the anode, and the low-frequency arc (< 0.5 Hz) to the mass transfer at the anode
[33-35]
.
The characteristics of the cells with various fuels are similar in the high- and
RI PT
intermediate-frequency regions. On the contrary, the low-frequency arcs of the cells fed with AC and AC-NaOH are much larger than that of the cell with AC-HNO3, implying higher concentration polarizations of the AC and AC-NaOH anodes, which
SC
is in accordance with the I-V results (Fig. 9a). The ohmic and polarization resistances
M AN U
of all the cells decrease with the increase of the temperature to 800 oC (Fig. 9d). The high- and intermediate-frequency characteristics of all of the cells are still almost the same. Meanwhile, the anode concentration polarizations of AC and AC-NaOH are much higher than that of AC-HNO3.
TE D
Fig. 10a displays the I-V and I-P characteristics of the single cell fed with AC-HNO3 at 800 oC with different N2 flow rates in the anode chamber. When the flow rate of N2 increases from 5 to 100 ml min-1 (STP), the OCV keeps almost
EP
constant and the Pmax drops from 128 to 95 mW cm-2. The corresponding EIS results
AC C
are shown in Fig. 10b. The N2 flow rate shows no obvious effect on the ohmic resistance of the cell and the charge transfer polarization of the electrodes, but exhibits significant influence on the anode concentration polarization. The concentration polarization of the anode with 5 ml min-1 N2 flow is much lower than those with 50 and 100 ml min-1 N2 flows.
4. Discussion 4.1. The effects of NaOH and HNO3 treatments 13
ACCEPTED MANUSCRIPT It has been reported that the surface properties of carbon could be modified by the treatments with acids and bases
[19, 36]
. As listed in Table 1, after the NaOH
treatment, the oxygen content on the surface of AC is slightly increased. The obvious
RI PT
increase of the intensity of the band at 3440 cm-1 in the FTIR result (Fig. 1) and the C1s core-level spectrum (Table 2) indicate a noticeable increase of the hydroxyl groups, which agrees with previous works
[19, 27]
, and a slight increase of functional
SC
groups containing C=O and COO bands on the surface of AC-NaOH. Meanwhile, the
M AN U
graphitization degree of AC is reduced after the NaOH treatment, which is proved by the decrease of Peak C-C. The HNO3 treatment results in a higher oxygen content (Table 1) and a lower graphitization degree (Peak C-C in Table 2). The more oxygen content comes not only from the increased hydroxyl groups, but from the growth of
Table 2).
TE D
the functional groups with C=O band such as carbonyl and quinone groups (Fig. 1 and
4.2. The effects of heat treatment in Ar
EP
The TGA results shown in Fig. 2 reflect the release of the volatile components in
AC C
the AC fuels during the heating process of the single cell. Meanwhile, the volatiles released during the TGA tests analyzed with a mass spectrometer are shown in Fig. S3. The gradual weight loss below 400 oC is attributed to the release of absorbed water and some of the carboxylic groups, while the rapid loss of the weight in the temperature range of 400-800 oC is corresponding to the losing of some of the oxygen-containing functional groups, such as ether and phenols
[37]
. The weight loss
of AC-HNO3 is the largest when the temperature reaches 800 oC probably due to its 14
ACCEPTED MANUSCRIPT highest oxygen content (Table 1). After the heat treatment in Ar at 800 oC for 1 h, the volatiles in the AC samples release almost completely. Most of the volatiles in the AC samples are oxygen-containing functional groups.
RI PT
As a result, the oxygen content on the surface of the samples decreases remarkably after the heat treatment in Ar (Table 3). The amounts of carbon in C-O-R groups in AC-NaOH and AC-HNO3 decrease significantly after the heat treatment (Table 4).
SC
The FTIR spectra of all the samples after the heating process display similar peak
M AN U
intensities at 3440 cm-1 (Fig. 3), implying the loss of hydroxyl groups in AC-NaOH and AC-HNO3. Meanwhile, the peaks between 1700-1500 cm-1 of all the samples become weaker after the heat treatment probably due to the loss of carboxyl and ester groups (Peak O-C=O in Table 4). On the contrary, the constant intensity of Peak C=O
TE D
implies that the carbonyl and quinone groups remain stable during the heat treatment, which are consistent with the report that the decomposition of these groups in an inert atmosphere requires a higher temperature than those of hydroxyl and carboxyl groups . The amount of C=O band in AC-HNO3-Ar is much higher than those in AC-Ar
EP
[38]
AC C
and AC-NaOH-Ar (Table 4), which is also proved by the highest peaks of AC-HNO3-Ar at 1700 and 1580 cm-1 in the FTIR spectra (Fig. 3). It is widely believed that the complete oxidation of carbon at anodes of DCFCs involves the adsorption of two oxygen ions on the surface carbon in series accompanied with the release of electrons
[17]
. A high surface C=O content may facilitate the oxidation of
carbon and thus improve performance of cells. In general, heat treatment in an inert atmosphere would lead to an increase of the 15
ACCEPTED MANUSCRIPT graphitization degree of carbonaceous species
[23]
. In this work, the increase of Peak
C-C of the samples after the heat treatment (Table 4) indicates an increased graphitization degree probably due to the removal of the carbon in the
RI PT
oxygen-containing functional groups. For the samples after the heat treatment, AC possesses the highest graphitization degree, while AC-HNO3 exhibits the lowest degree. The graphitization degrees of the samples after the heat treatment are also
SC
evaluated based on the Raman spectra. G and D3 bands in Fig. 4 originate from the
M AN U
graphitic and amorphous structures of carbon, respectively. The ratios between the intensities of various bands are usually applied to evaluate the disorder degree of carbonaceous materials qualitatively. Typically, higher ID1/IG and lower IG/IAll values imply a higher disorder degree of the structure of carbon, and ID3/(IG + ID2 + ID3) is [39, 40]
. As shown in Fig. 5,
TE D
closely related to the content of amorphous carbon
compared with untreated AC, AC-NaOH and AC-HNO3 exhibit high ID1/IG and ID3/(IG + ID2 + ID3) values and low IG/IAll values, implying a lower graphitization degree of
EP
AC-NaOH and AC-HNO3. Among all the samples, AC-HNO3 possesses the highest
AC C
disorder degree of the structure, which is consistent with the XPS results. The TGA results of the Ar-treated samples in air (Fig. 6) reflect the reactivity of
the oxidation of the samples. The lowest oxidation temperature of AC-HNO3 indicates the highest reactivity of AC-HNO3 for oxidation attributed to its lowest graphitization degree and highest oxygen content. The high rate of oxidation of AC-NaOH compared with that of AC implies a higher reactivity of AC-NaOH. The temperatures corresponding to the peaks of the products in the TPO-MS results (Fig. 6b) are 16
ACCEPTED MANUSCRIPT another aspect reflecting the oxidation reactivity of the samples. AC-HNO3 with the highest oxidation reactivity exhibits the lowest temperatures corresponding to the peaks of CO and CO2. Meanwhile, the high contents of CO in the oxidation products
RI PT
of AC and AC-NaOH indicate an incomplete oxidation process, which would lead to a lower fuel efficiency of the DCFC. On the contrary, CO2 is the main oxidation product of AC-HNO3, implying the complete oxidation of the fuel due to its high
SC
oxidation reactivity. Similar to the TGA results in air, AC-HNO3 exhibits the lowest
M AN U
initial temperature and the highest rate of weight loss in the TGA results in CO2, which demonstrate that it also possesses the highest reactivity for the reverse Boudouard reaction. 4.3. Single cell performance
TE D
The reactivity of the carbon fuel in the anode has important effects on the OCV and the output power density of the single cell. Though the differences in the activation polarization among the anodes with various fuels are indistinguishable from
EP
the complex EIS results at 750 oC (Fig. 9c), the low OCVs of the cells fed with AC
AC C
and AC-NaOH (Fig. 9a) indicate their low anode activity. Since the reverse Boudouard reactions of AC and AC-NaOH just begin at about 750 oC (Fig. 7), the direct oxidation of the carbon fuels should be the main reaction at these anodes. The cell fed with AC-HNO3 possesses the highest OCV and Pmax partly due to the highest oxidation reactivity of the fuel (Fig. 6). Meanwhile, the CO generated by the reverse Boudouard reaction could be also involved in the anodic reaction, resulting in the lowest concentration polarization resistance. When the temperature rises to 800 oC, 17
ACCEPTED MANUSCRIPT the reverse Boudouard reaction is involved in the anode reactions of the cells with AC and AC-NaOH as the fuels, bringing about higher limited current densities. Meanwhile, the cell fed with AC-HNO3 still exhibits the lowest polarization resistance
RI PT
and the highest Pmax of 128 mW cm-2, which is superior to most of the electrolyte-supported DCFCs with similar configurations reported previously (Table S1).
SC
The CO generated through the reverse Boudouard reaction could be diluted by
M AN U
the inert purge gas at the anode, leading to a lower performance of the cell. The decrease of the Pmax of the cell with the increase of the flow rate of N2 from 5 to 50 ml min-1 in the anode chamber (Fig. 10a) is mainly due to the insufficient supply of CO to the active sites at the anode, which is proved by the obvious diffusion limitation at
TE D
higher current densities accompanied with a much larger low-frequency arc (Fig. 10b). A similar behavior has been observed in SO-DCFCs with demineralized coal as the fuel and high purity N2 as purge gas.[34, 37] However, no obvious further increase of the
EP
concentration polarization is observed when the flow rate of N2 rises to 100 ml min-1,
AC C
and it is generally believed that the contribution of the electrochemical oxidation of CO to the performance of the cell is negligible when the flow rate of the anode purge gas reaches 100 ml min-1 [12, 41]. Therefore, it is reasonable to attribute the Pmax of 95 mW cm-2 of the cell with the 100 ml min-1 N2 flow to the direct electrochemical oxidation of the AC-HNO3 at the anode.
5. Conclusions In this work, the structure and surface properties of AC are modified by HNO3 18
ACCEPTED MANUSCRIPT and NaOH treatments. The content of oxygen on the surface of AC increases considerably after the treatment with HNO3. Compared with untreated AC, AC-NaOH and AC-HNO3 both possess a low graphitization degree and a high amount of C-O-R
RI PT
functional groups. Meanwhile, AC-HNO3 has more carbonyl and quinone groups than other samples. After the heating process in Ar atmosphere, the amount of C-O-R groups decreases significantly, leading to a reduction of oxygen content on the carbon
SC
surface. However, the C=O groups on the surface of AC-HNO3 are more stable. The
M AN U
lowest graphitization degree and highest oxygen content of AC-HNO3 result in its highest reactivity for oxidation and reverse Boudouard reactions. The single cell fed with AC-HNO3 exhibits the highest Pmax of 128 mW cm-2 and lowest polarization resistance at 800 oC. The increase of the N2 flow rate at the anode impedes the reverse
Acknowledgments
TE D
Boudouard reaction, leading to a decrease of the performance of the cell.
The financial support of NSF of China under contract numbers 51402210 and the
EP
support of Tianjin Municipal Science and Technology Commission under contract
AC C
numbers 15JCQNJC06500 are gratefully acknowledged. The work has been also supported by the Program of Introducing Talents to the University Disciplines under file number B06006, and the Program for Changjiang Scholars and Innovative Research Teams in Universities under file number IRT 0641. References [1] C. Jiang, J. Ma, G. Corre, S.L. Jain, J.T.S. Irvine, Challenges in developing direct carbon fuel cells, Chem. Soc Rev. 46 (2017) 2889-2912. [2] T.M. Gur, Critical review of carbon conversion in "carbon fuel cells", Chem. Rev. 113 (2013) 6179-6206.
19
ACCEPTED MANUSCRIPT [3] J.F. Cooper, J.R. Selman, Analysis of the carbon anode in direct carbon conversion fuel cells, Int. J. Hydrogen Energy 37 (2012) 19319-19328. [4] W. Hao, X. He, Y. Mi, Achieving high performance in intermediate temperature direct carbon fuel cells with renewable carbon as a fuel source, Appl. Energy 135 (2014) 174-181. [5] A.C. Rady, S. Giddey, A. Kulkarni, S.P.S. Badwal, S. Bhattacharya, Catalytic gasification of carbon in a direct carbon fuel cell, Fuel 180 (2016) 270-277. [6] Y. Zhao, C. Xia, L. Jia, Z. Wang, H. Li, J. Yu, Y. Li, Recent progress on solid oxide fuel cell:
RI PT
Lowering temperature and utilizing non-hydrogen fuels, Int. J. Hydrogen Energy 38 (2013) 16498-16517.
[7] A. Jayakumar, R.K. Ungas, S. Roy, A. Javadekar, D.J. Buttrey, J.M. Vohs, R.J. Gorte, A direct carbon fuel cell with a molten antimony anode, Energy Environ. Sci. 4 (2011) 4133-4137.
[8] B. Yu, Y. Zhao, Y. Li, A SnO2-samarium doped ceria additional anode layer in a direct carbon fuel
SC
cell, J. Power Sources 306 (2016) 387-393.
[9] S. Nurnberger, R. Bussar, P. Desclaux, B. Franke, M. Rzepka, U. Stimming, Direct carbon conversion in a SOFC-system with a non-porous anode, Energy Environ. Sci. 3 (2010) 150-153. Solid State Ionics 271 (2015) 121-127.
M AN U
[10] M. Dudek, On the utilization of coal samples in direct carbon solid oxide fuel cell technology, [11] M.T. Mehran, T.-H. Lim, S.-B. Lee, J.-W. Lee, S.-J. Park, R.-H. Song, Long-term performance degradation study of solid oxide carbon fuel cells integrated with a steam gasifier, Energy 113 (2016) 1051-1061.
[12] A. Kulkarni, F.T. Ciacchi, S. Giddey, C. Munnings, S.P.S. Badwal, J.A. Kimpton, D. Fini, Mixed ionic electronic conducting perovskite anode for direct carbon fuel cells, Int. J. Hydrogen Energy 37 (2012) 19092-19102.
TE D
[13] Y. Leng, S. Chan, Q. Liu, Development of LSCF–GDC composite cathodes for low-temperature solid oxide fuel cells with thin film GDC electrolyte, Int. J. Hydrogen Energy 33 (2008) 3808-3817. [14] X. Wu, X. Zhou, Y. Tian, X. Kong, J. Zhang, W. Zuo, X. Ye, Stability and electrochemical performance of lanthanum ferrite-based composite SOFC anodes in hydrogen and carbon monoxide, Electrochim. Acta 208 (2016) 164-173.
EP
[15] J. Xiao, D. Han, F. Yu, L. Zhang, J. Liu, Z. Zhan, Y. Zhang, P. Dong, Characterization of symmetrical SrFe0.75Mo0.25O3−δ electrodes in direct carbon solid oxide fuel cells, J. Alloy. Compd. 688 (2016) 939-945.
AC C
[16] S. Giddey, A. Kulkarni, C. Munnings, S.P.S. Badwal, Performance evaluation of a tubular direct carbon fuel cell operating in a packed bed of carbon, Energy 68 (2014) 538-547. [17] N.J. Cherepy, R. Krueger, K.J. Fiet, A.F. Jankowski, J.F. Cooper, Direct Conversion of Carbon Fuels in a Molten Carbonate Fuel Cell, J. Electrochem Soc. 152 (2005) A80-A87. [18] X. Li, Z.H. Zhu, J.L. Chen, R. De Marco, A. Dicks, J. Bradley, G.Q. Lu, Surface modification of carbon fuels for direct carbon fuel cells, J. Power Sources 186 (2009) 1-9. [19] D.X. Cao, G.L. Wang, C.Q. Wang, J. Wang, T.H. Lu, Enhancement of electrooxidation activity of activated carbon for direct carbon fuel cell, Int. J. Hydrogen Energy 35 (2010) 1778-1782. [20] J.B. Zhang, Z.P. Zhong, D.K. Shen, J.X. Zhao, H.Y. Zhang, M. Yang, W.L. Li, Preparation of Bamboo-Based Activated Carbon and Its Application in Direct Carbon Fuel Cells, Energy & Fuels 25 (2011) 2187-2193. [21] S. Eom, S. Ahn, K. Kang, G. Choi, Correlations between electrochemical resistances and surface properties of acid-treated fuel in coal fuel cells, Energy 140 (2017) 885-892.
20
ACCEPTED MANUSCRIPT [22] X. Li, Z.H. Zhu, R. De Marco, J. Bradley, A. Dicks, Modification of Coal as a Fuel for the Direct Carbon Fuel Cell, J. Phys. Chem. A 114 (2010) 3855-3862. [23] A. Fuente-Cuesta, C. Jiang, A. Arenillas, J.T.S. Irvine, Role of coal characteristics in the electrochemical behaviour of hybrid direct carbon fuel cells, Energy Environ. Sci. 9 (2016) 2868-2880. [24] P. Zeng, R. Ran, Z. Chen, H. Gu, Z. Shao, J. Dacosta, S. Liu, Significant effects of sintering temperature on the performance of La0.6Sr0.4Co0.2Fe0.8O3−δ oxygen selective membranes, J. Membrane Science 302 (2007) 171-179.
RI PT
[25] L. Jia, Y. Tian, Q. Liu, C. Xia, J. Yu, Z. Wang, Y. Zhao, Y. Li, A direct carbon fuel cell with (molten carbonate)/(doped ceria) composite electrolyte, J. Power Sources 195 (2010) 5581-5586.
[26] Y. Jiao, J. Zhao, W. An, L. Zhang, Y. Sha, G. Yang, Z. Shao, Z. Zhu, S.-D. Li, Structurally modified coal char as a fuel for solid oxide-based carbon fuel cells with improved performance, J. Power Sources 288 (2015) 106-114. Adsorptive Properties, Langmuir (2004) 2233-2242.
SC
[27] J.P. Chen, Acid/Base-Treated Activated Carbons Characterization of Functional Groups and Metal [28] A.-N.A. El-Hendawy, Influence of HNO3 oxidation on the structure and adsorptive properties of corncob-based activated carbon, Carbon 41 (2003) 713-722.
M AN U
[29] J.-H. Zhou, Z.-J. Sui, J. Zhu, P. Li, D. Chen, Y.-C. Dai, W.-K. Yuan, Characterization of surface oxygen complexes on carbon nanofibers by TPD, XPS and FT-IR, Carbon 45 (2007) 785-796. [30] A. Sadezky, H. Muckenhuber, H. Grothe, R. Niessner, U. Pöschl, Raman microspectroscopy of soot and related carbonaceous materials: Spectral analysis and structural information, Carbon 43 (2005) 1731-1742.
[31] A.M. Rao, J. Chen, E. Richter, U. Schlecht, P.C. Eklund, R.C. Haddon, U.D. Venkateswaran, Y.K. Kwon, D. Tomanek, Effect of van der Waals interactions on the Raman modes in single walled carbon
TE D
nanotubes, Physical review letters 86 (2001) 3895-3898.
[32] S. Kizgut, S. Yilmaz, Characterization and non-isothermal decomposition kinetics of some Turkish bituminous coals by thermal analysis, Fuel Process. Technol. 85 (2003) 103-111. [33] A.C. Rady, S. Giddey, A. Kulkarni, S.P.S. Badwal, S. Bhattacharya, Direct Carbon Fuel Cell Operation on Brown Coal with a Ni-GDC-YSZ Anode, Electrochim. Acta 178 (2015) 721-731.
EP
[34] A.C. Rady, S. Giddey, A. Kulkarni, S.P.S. Badwal, S. Bhattacharya, Degradation Mechanism in a Direct Carbon Fuel Cell Operated with Demineralised Brown Coal, Electrochim. Acta 143 (2014) 278-290.
AC C
[35] S. Giddey, A. Kulkarni, C. Munnings, S.P.S. Badwal, Composite anodes for improved performance of a direct carbon fuel cell, J. Power Sources 284 (2015) 122-129. [36] X. Wang, N. Gao, Q. Zhou, H. Dong, H. Yu, Y. Feng, Acidic and alkaline pretreatments of activated carbon and their effects on the performance of air-cathodes in microbial fuel cells, Bioresour Technol. 144 (2013) 632-636. [37] A.C. Rady, S. Giddey, A. Kulkarni, S.P.S. Badwal, S. Bhattacharya, B.P. Ladewig, Direct carbon fuel cell operation on brown coal, Appl. Energy 120 (2014) 56-64. [38] J.L. Figueiredo, M.F.R. Pereira, M.M.A. Freitas, J.J.M. Órfão, Modification of the surface chemistry of activated carbons, Carbon 37 (1999) 1379-1389. [39] P. Li, Y. Zhao, B. Yu, J. Li, Y. Li, Improve electrical conductivity of reduced La2Ni0.9Fe0.1O4+δ as the anode of a solid oxide fuel cell by carbon deposition, Int. J. Hydrogen Energy 40 (2015) 9783-9789. [40] C. Sheng, Char structure characterised by Raman spectroscopy and its correlations with
21
ACCEPTED MANUSCRIPT combustion reactivity, Fuel 86 (2007) 2316-2324. [41] H. Jang, Y. Park, J. Lee, Meticulous insight on the state of fuel in a solid oxide carbon fuel cell,
AC C
EP
TE D
M AN U
SC
RI PT
Chem. Eng. J. 308 (2017) 974-979.
22
ACCEPTED MANUSCRIPT
Table 1. XPS elemental analysis results of the raw and treated AC. C (at.%)
O (at.%)
O/C
AC
88.9
10.1
0.11
AC-NaOH
87.5
10.7
0.12
AC-HNO3
84.5
14.9
RI PT
Sample
AC C
EP
TE D
M AN U
SC
0.18
23
ACCEPTED MANUSCRIPT
Table 2. The area percentages of the deconvoluted peaks in the C1s spectra of raw and treated AC samples. The area percentages of the peaks of carbon (%) Sample C-O-R
-C=O
AC
81
9
4
6
AC-NaOH
76
12
5
7
AC-HNO3
72
13
8
7
SC M AN U TE D EP AC C 24
O-C=O
RI PT
C-C
ACCEPTED MANUSCRIPT
Table 3. XPS elemental analysis results of the samples after the heat treatment in Ar. C (at.%)
O (at.%)
O/C
AC-Ar
94.9
5.0
0.05
AC-NaOH-Ar
94.3
5.1
0.05
AC-HNO3-Ar
91.8
7.9
RI PT
Sample
AC C
EP
TE D
M AN U
SC
0.09
25
ACCEPTED MANUSCRIPT
Table 4. The area percentages of the deconvoluted peaks in the C1s spectra of raw and treated AC samples after the heat treatment in Ar. The area percentages of the peaks of carbon (%) Sample C-O-R
-C=O
AC-Ar
83
8
4
5
AC-NaOH-Ar
80
9
5
6
AC-HNO3-Ar
76
9
9
6
SC M AN U TE D EP AC C 26
O-C=O
RI PT
C-C
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
Fig. 1. FTIR spectra of the raw and treated AC.
27
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
Fig. 2. TGA curves of AC samples in Ar atmosphere.
28
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
Fig. 3. FTIR spectra of the raw AC and treated AC after the heat treatment in Ar.
29
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Fig. 4. Raman spectra of (a) AC, (b) AC-NaOH and (c) AC-HNO3 after the heat treatment in Ar. 30
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
Fig. 5. Band area ratios of AC samples after the heat treatment in Ar.
31
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Fig. 6. (a) TGA curves in air and (b) TPO-MS results of AC samples after the heat treatment in Ar.
32
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
Fig. 7. TGA curves in CO2 of AC samples after the heat treatment in Ar.
33
RI PT
ACCEPTED MANUSCRIPT
SC
Fig. 8. SEM cross-sectional micrographs of (a) the single cell and (b) the
AC C
EP
TE D
M AN U
electrolyte-electrode interface.
34
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
35
SC
RI PT
ACCEPTED MANUSCRIPT
Fig. 9. I-V and I-P curves of single cells with various carbon fuels and 5 ml min-1 N2
M AN U
as the anode purge gas at (a) 750 oC and (b) 800 oC and the corresponding EIS results
AC C
EP
TE D
at (c) 750 oC and (d) 800 oC.
36
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
EP
Fig. 10. (a) I-V and I-P curves and (b) the corresponding EIS of the single cell with
AC C
AC-HNO3 as the fuel at 800 oC with different flow rates of the anode purge gas.
37
ACCEPTED MANUSCRIPT
Research Highlights ► Activated carbon is treated with HNO3 and NaOH as the fuel of DCFCs ► AC-HNO3 shows the lowest graphitization degree and highest surface oxygen
RI PT
content ► AC-HNO3 shows the highest reactivity for oxidation and reverse Boudouard reactions
AC C
EP
TE D
M AN U
SC
► The cell supported by a YSZ layer exhibits the Pmax of 128 mW cm-2 at 800 oC