Germanium and phosphorus co-doped carbon nanotubes with high electrocatalytic activity for oxygen reduction reaction

Abstract

Germanium and phosphorus co-doped carbon nanotubes (Ge–P-CNTs) were prepared by a simple and scalable approach. The morphology and structure of the Ge–P-CNTs was characterized by scanning electron microscopy, transmission electron microscopy, X-ray diffraction and X-ray photoelectron spectroscopy. The ORR electrocatalytic performances were evaluated by exchange current density, reaction pathway selectivity, on-set potential, kinetic current density and H₂O₂ yields from rotating ring-disk electrode (RRDE) measurements, indicating that the co-doped GeP₂C₂, GeP₃C and GeP₄ + PC₃ microstructures in Ge–P-CNTs are crucial to improving the ORR catalytic performance. In fact, the electrochemical performance enhancement results from the synergistic effects by the appropriate proportion of Ge and P atoms. The ORR catalytic synergistic effect has also been verified by calculating the work function based on density functional theory (DFT). Because of these outstanding features, it is expected that the Ge–P-CNTs materials will be a very suitable catalyst for fuel cells and metal–air batteries.

---

Introduction

There is a growing interest in energy sources that do not rely on fossil fuels, due to the depletable resources of these feedstocks and the environmental concerns connected to their use.¹ Electrochemical energy conversion devices, ranging from fuel cells to metal–air batteries, have been extensively studied.² The oxygen reduction reaction (ORR) is a key process in the performance of fuel cells and metal–air batteries.^3–9 It is well known that the platinum-based catalysts have been commonly used in commercial fuel cells owing to their relatively low overpotential and high current density.¹⁰ However, the widespread use of fuel cells has been limited by the high cost and inadequacy of this metal.^11,12 Intensive research efforts have been directed towards the cost reduction of fuel cells through the development of non-noble metal catalysts for the electrochemical reduction of oxygen.¹³ Recently, in alkaline medium, many metal-free cathode catalysts,^14–16 such as nitrogen-doped or phosphorus-doped carbon catalysts, with higher electrocatalytic activity, longer-term stability and better tolerance to methanol crossover towards oxygen reduction in alkaline than Pt/C catalyst have attracted enormous interest as an alternative to platinum-based catalysts.^17,18

Recently, carbon nanotubes (CNTs) have attracted a great deal of attention in different research fields because of their extraordinary electronic, thermal, and mechanical properties, and other potential applications.¹⁹ Carbon nanotubes have been reported as a promising material for fuel cell applications in acidic environment due to their large surface area, good thermal and chemical stability, and high electrical conductivity.^14,20 Furthermore, modifying CNTs by introducing heteroatom is known to improve their electrocatalytic activity for ORR.^11,21 Nitrogen-doped carbon nanotubes have been reported with excellent ORR electrocatalytic performance owing to the high electronegativity of N atoms.^22,23

Phosphorus and nitrogen are the same main group elements, having the same number of valence electrons and often showing similar chemical status. Therefore, phosphorus-doping carbon nanotubes can be also with wonderful catalytic activity for ORR. In previous studies, phosphorus-doped carbon materials²⁴ exhibit outstanding ORR activity as well as excellent selectivity and stability, which endow it great potential to be an efficient metal-free electrocatalyst, such as P-doped graphene.²⁵ Germanium (Ge), as a carbon group element, possesses the semblable outermost electron distribution with C, and its higher carrier mobility and source injection velocity.^26,27 The elemental germanium has a higher intrinsic electronic conductivity due to its smaller band gap of 0.6 eV,²⁸ so Ge is very easily doped into carbon systems and maybe improves its electrochemical properties.^29,30 Previous studies show that the Ge in nanomaterials plays a crucial role in activating and improving the kinetics of the conversion reaction.³¹

Many researchers have reported that the co-doping carbon materials have excellent electrocatalytic performance for ORR, because the co-doping can be influenced not only by the increased number of dopant heteroatoms but also by the synergistic effect arising from co-doping of heteroatoms.³² For example, B/N,³³ S/N,³⁴ or P/N³⁵ co-doped carbon materials have been used for ORR electrocatalysts. However, to our best knowledge, there are no any reports to focus on single and co-doped germanium carbon based materials for ORR catalyst.

In our studies, we have developed a simple and green approach to synthesize phosphorus and germanium co-doped carbon nanotubes (Ge–P-CNTs). The resultant Ge–P-CNTs can be used as metal-free ORR electrocatalyst for ORR. We also studied that phosphorus and germanium co-doping carbon nanotubes with various germanium contents. The unique features of the resulting Ge–P-CNTs, lead to high electrocatalytic activity, resistance to fuel crossover and long-time durability in alkaline environment.

The co-doped carbon-based nanomaterials will open the potential application in fuel cells and metal–air batteries.

Experimental

Materials

Carboxyethyl germanium sesquioxide and triphenylphosphine were purchased from Aladdin. Ethyl alcohol and potassium hydroxide were purchased from Sinopharm Chemical Reagent Co., Ltd. Deionized (DI) water (18 MΩ) was used as a solvent in the synthesis. Multiwall carbon nanotubes (MWCNTs) were purchased from Shandong Dazhan Nano Materials Company. The Pt/C (Hispec 3000, 20% Pt on Vulcan XC-72R) electrocatalyst was purchased from Johnson Matthey.

The preparation of Ge–P-CNTs

100 mg of carbon nanotubes were dispersed ultrasonically for 20 min in 10 mL distilled water, and then 1.0, 2.0, 3.0 and 4.0 mg of carboxyethyl germanium sesquioxide were respectively also dissolved in 2 mL distilled water. The organic germanium solution was introduced into the suspension liquid of carbon nanotubes, and then ultrasonic treated for 10 min. The resulting mixtures were heated and dried in an oven at 80 °C for 24 h. The solid mixtures comprised of carbon nanotubes and organic germanium were heated at 600 °C for 1 h with a heating rate of 5 °C min⁻¹ in the atmosphere of nitrogen. The materials obtained are respectively denoted as Ge-CNTs-1, Ge-CNTs-2, Ge-CNTs-3, Ge-CNTs-4. Afterwards, the obtained solid mixtures were separately dispersed in 15 mL distilled water. 2.0 mg triphenylphosphine was dissolved in 2 mL ethyl alcohol and introduced into the suspension liquid of Ge-CNTs by ultrasonic treating for 10 min. After that, the resulting suspensions were dried in an oven at 80 °C for 24 h. The obtained solid composites were calcined by heating from 25 °C to 600 °C with a heating rate of 5 °C min⁻¹ and keeping the temperature at 600 °C for 1 h in the atmosphere of N₂. The resulting samples are denoted as Ge–P-CNTs-1, Ge–P-CNTs-2, Ge–P-CNTs-3, Ge–P-CNTs-4, respectively. For the purpose of comparison, the carbon nanotubes doped solely with P atoms (P-CNTs) were also prepared. 2.0 mg triphenylphosphine and 100 mg pure CNTs were treated by the above same process.

Electrode preparation and electrochemical test

The pretreatment of working electrode was as follows: the electrodes were polished mechanically with particle size of 0.3 micron and 0.05 micron alumina powder under an abrasive paper to obtain a mirror-like surface, respectively. And then the polished electrodes were washed with deionized water and ethanol by sonication for 15 min and dried with nitrogen. 2.0 mg of each grinded sample was dispersed in the solvent mixture of (5 wt%) Nafion (25 μL), ethanol absolute (250 μL), and de-ionized water (250 μL) by sonication, respectively. Afterwards, 10.0 μL suspension of each sample was pipetted onto the treated glassy carbon electrode of 4 mm in diameter (loading ∼ 0.303 mg cm⁻²) and dried at room temperature before measurement. Electrochemical experiments were conducted using a CHI 750E Electrochemical Analyzer/Workstation with a typical three-electrode cell. A platinum wire was used as counter electrode, glass carbon as working electrode and Ag/AgCl, KCl (3 M) electrode as reference electrode. Cyclic voltammetry (CV) experiments were performed from 0.4 V to −1.8 V with a scan rate of 50 mV s⁻¹. The linear sweep voltammograms were recorded in O₂ saturated 0.1 M KOH with a scan rate of 1 mV s⁻¹ at various rotating speeds from 225 to 2500 rpm. For the rotating ring-disk electrode (RRDE) measurements, catalyst inks were prepared by the same method as for the ORR measurements. The scanning rate of the disk electrode was 10 mV s⁻¹ and the ring potential was constant at 0.5 V. The electron transfer number (n) and the peroxide percentage (% HO₂⁻) were determined by the followed equations:where I_d is disk current, I_r is ring current and N = 0.41 is the current collection efficiency of the Pt ring.

Characterization

X-ray diffraction (XRD, DX2700, X-ray diffractometer with Cu Kα radiation of wavelength λ = 0.1541 nm at 40 kV and 40 mA), scanning electron microscopy (SEM, JEOL-JEM-6100F) and transmission electron microscopy (TEM, JEOL-JEM-1200) were used to characterize the morphology and structure of the samples. X-ray photoelectron spectroscopy (XPS) measurements were performed on ESCALab-250 electron spectrometer from Thermo Scientific Corporation with monochromatic 150 W Al Kα radiations. Pass energy for the narrow scan is 30 eV and the base pressure was about 6.5 × 10⁻¹⁰ mbar. Elemental mapping was conducted using EDAX detector attached on JEM-2010. Raman spectra were tested by Raman spectroscopy (Renishaw in Via plus, UK).

Calculation for work function of Ge–P-CNTs

In the present work, we have taken a (8, 0) zigzag nanotube having length twice of its lattice constant for the workfunction of Ge–P-CNTs with different doping-geometries. It has total 64 C atoms and is placed in a 15 Å × 15 Å × 8.572 Å tetragonal supercell. The equilibrium configurations were determined by relaxation of all atoms in the supercell. All calculations were performed with Vienna Ab initio Simulation Package (VASP 5.2) at the spin-polarized DFT level.^36–39 The projector-augmented wave (PAW) approach was adopted and the plane-wave cut-off was 500 eV.^40,41 The exchange and correlation energies were also calculated using the Perdew–Burke–Ernzerhof form of the spin-polarized generalized gradient approximation (GGA-PBE).^42,43 Theoretical calculation details were shown in ESI.†

Results and discussion

The representative SEM and TEM images of the Ge–P-CNTs were shown in the Fig. 1a–d. These results clearly revealed that the diameters of the nanotubes are in the range of 10–20 nm, whereas the wall thickness is only about 2.5 nm. From the element mapping (Fig. 1e and f), the product contains carbon, oxygen, germanium and phosphorus elements, indicating germanium and phosphorus have been doped into carbon nanotubes. We can also see the elements content of germanium and phosphorus is very low relative to carbon and oxygen.

Fig. 1 (a and b) SEM and (c and d) TEM images of Ge–P-CNTs-2; (e and f) SEM images of Ge–P-CNTs-2 with corresponding elemental mapping images.

The XRD spectra (Fig. 2a) for the P and Ge co-doped carbon nanotubes with different Ge content are basically consistent with the typical spectra of pure CNTs, indicating that the content of phosphorus and germanium for these materials is too low to make difference. Further structural information about the doped CNTs was observed from Raman spectra (Fig. 2b). The D band situated around 1343 cm⁻¹ arises from the disordered carbon atoms, and the G band located around 1574 cm⁻¹ results from sp²-hybridized graphitic carbon atoms. Obviously, the I_D/I_G ratio of Ge-CNTs-2 and P-CNTs is higher than pristine CNTs, which is due to the introduction of defects by Ge-doping and P-doping. Interestingly, the I_D/I_G ratio of Ge–P-CNTs-2 is slightly lower than the pristine CNTs, indicating larger crystalline graphitic domains in Ge–P-CNTs-2.³⁴

Fig. 2 (a) XRD spectra for CNTs and Ge–P-CNTs with different Ge content; (b) Raman spectra for CNTs, P-CNTs, Ge-CNTs-2 and Ge–P-CNTs-2.

Fig. 3a shows XPS survey spectra for Ge–P-CNTs-2. The dominant carbon C 1s at about 285 eV, P 2p peaks at about 132 eV, Ge 3d at about 31 eV, and O 1s at about 533 eV indicate the successful incorporation of heteroatoms such as phosphorus, germanium, and oxygen with the carbon nanotubes. Fig. 3b shows the deconvoluted XPS P 2p spectrum of the Ge–P-CNTs-2 has three peaks at 130.1 (87.83 atom%), 133.3 (3.57 atom%) and 133.5 eV (8.60 atom%). The three peaks can be assigned to the P–C, P–O and P–Ge bonding respectively, which suggest that phosphorus atoms are successfully incorporated into the carbon nanotubes through a thermal annealing process.^25,44 The XPS Ge 3d spectrum (Fig. 3c) of Ge–P-CNTs-2 is observed at 29.9, 30.9, 31.4 and 32.0 eV. The intense Ge 3d peak at about 30.9 eV (56.56 atom%) is due to the formation of C–Ge bonding, indicating the incorporation of Ge elements into carbon nanotubes. The peaks at 31.4 and 32.0 eV may be assigned to GeO and GeP_x.^31,44 The peak at 29.9 eV is due to pure Ge and other Ge chemical status. Fig. 3d displays the XPS C 1s spectrum of the Ge–P-CNTs-2 has peaks at 284.9, 285.6, 288.2, 288.3, 290.1 and 291.5 eV. The main peak at 284.9 eV (74.95 atom%) is assigned to the C–C bonds, which is agreement with the carbon-based structure of Ge–P-CNTs. The peak at 285.6 eV is ascribed to the C–Ge bonds or C atoms no longer in the regular tubular structure. The peak at 288.2 eV is attributed to the C–P from phosphorus-doping.⁴⁵ The peak at 288.3 eV is also attributed to the C=O groups or C–O carbon atoms. Two peaks at 290.1 and 291.5 eV are assigned to carbonates and π–π* transitions. Accordingly, the O 1s XPS spectra for Ge–P-CNTs-2 is shown in Fig. S1,† two strong peaks at 531.9 and 533.2 eV are assigned to oxygen atoms in amorphous hydrogenated carbon and C–O–P.⁴⁶ The peak at 532.2 eV is attributed to P–O bonding, which is possibly due to physisorbed oxygen.⁴⁷ Two peaks at 531.2 and 532.3 eV is due to GeO, GeO₂ respectively.⁴⁸

Fig. 3 (a) XPS survey spectra for Ge–P-CNTs-2; (b) P 2p XPS spectra for Ge–P-CNTs-2; (c) Ge 3d XPS spectra for Ge–P-CNTs-2; (d) C 1s XPS spectra for Ge–P-CNTs-2.

Fig. 4a shows the CV curves for electrochemical activities of CNTs, P-CNTs, Ge-CNTs-2, and Ge–P-CNTs-2 electrodes in N₂- and O₂-saturated 0.1 M KOH solution, referenced to the Ag/AgCl electrode. No apparent peak can be found in CVs obtained in the N₂-saturated solution (dash line), indicating no reduction reaction occurs. However, the CVs obtained in the O₂-saturated solution (solid line) show a well-defined characteristic ORR peaks centered at −0.24 V for Ge–P-CNTs-2 is rather positive than those at −0.32, −0.31 V, and −0.31 V of single doped Ge-CNTs-2, P-CNTs, and pure CNTs. To gain further insight about the ORRs with CNTs, P-CNTs, Ge-CNTs-2 and Ge–P-CNTs-2, the linear sweep voltametry (LSV) measurements on a rotating-disk electrode (RDE) were also performed in 0.1 M KOH electrolyte saturated with O₂ at a rotation rate of 1600 rpm. As shown in Fig. 4b. It can be seen that the Ge–P-CNTs-2 electrodes have the higher current density and the more positive onset potential than those of CNTs, Ge-CNTs-2, and P-CNTs electrodes. Clearly, the onset potential for Ge–P-CNTs-2 starts at −0.10 V, while for the Ge-CNTs-2, P-CNTs, and CNTs at −0.18 V, −0.17 V, and −0.18 V, respectively. This result is in agreement with the CV observations, which suggests that the ORR catalytic activity of Ge–P-CNTs-2 is better than those of solely doped P-CNTs, Ge-CNTs-2 electrodes. We also found that P-CNTs exhibited better ORR electro-catalytic activity than CNTs. It is noting that Ge-CNTs has similar onset potential to CNTs, while it could approach lower current densities than pure CNTs at the same potential.

Fig. 4 (a) CV curves for CNTs, P-CNTs, Ge-CNTs-2 and Ge–P-CNTs-2 electrodes in N₂ and O₂-saturated 0.1 M solution of KOH; (b) LSV curves for various materials in O₂-saturated 0.1 M solution of KOH with a rotation rate of 1600 rpm.

To gain further insight into the role of Ge–P-CNTs-2 during the ORR electrochemical process, we studied the reaction kinetics by rotating-disk voltammetry. The rotating disk electrode (RDE) measurements for Ge–P-CNTs with different Ge content in O₂-saturated 0.1 M KOH as the electrolyte shows that the current density was enhanced by an increase in the rotation rate (from 225 to 2500 rpm; Fig. 5a and S3,† respectively). Fig. S2† shows the LSV curves for the Ge–P-CNTs electrodes with different Ge content in O₂-saturated 0.1 M KOH solution at a rotation rate of 1600 rpm. We can clearly see the Ge–P-CNTs-2 has the more positive onset potential at −0.10 V than those of Ge–P-CNTs-1 (−0.12 V), Ge–P-CNTs-3 (−0.12 V) and Ge–P-CNTs-4 (−0.12 V) electrodes. The corresponding Koutecky–Levich (K–L) plots at different electrode potentials were obtained from the LSV curves at various rotating speeds, exhibiting good linearity (see in Fig. 5b). The slopes are basically consistent over the potential range from −0.475 to −0.55 V, which indicates that the electron transfer numbers for oxygen reduction at various electrode potentials are similar. The transferred electron numbers (n) per oxygen molecule in the ORR process at the Ge–P-CNTs/GCE can be calculated from the Koutecky–Levich (K–L) equation:

J⁻¹ = J_K⁻¹ + B⁻¹ω^−1/2 (3)

B = 0.2nFC₀(D₀)^2/3ν^−1/6 (4)

J_K = nFkC₀ (5)

where J is the measured current density, J_K is the kinetic current density, J_L is the diffusion-limiting current density, ω is the angular velocity of the disk (ω = 2πN, N is the linear rotation speed). n is the overall number of electrons transferred in oxygen reduction, F is the Faraday constant (F = 96 485 C mol⁻¹), C₀ is the bulk concentration of O₂, (C₀ = 1.2 × 10⁻³ mol L⁻¹), ν is the kinematic viscosity of the electrolyte (ν = 0.01 cm² s⁻¹). D₀ is the diffusion coefficient of O₂, in 0.1 M KOH (1.9 × 10⁻⁵ cm² s⁻¹). According to eqn (3) and (4), the number of electrons transferred n and J_K can be obtained from the slope and intercept of the K–L plots, respectively. Fig. 5c shows the n and J_K of each sample were calculated at −0.475 V. Obviously, the electron-transfer numbers (n) of Ge–P-CNTs-2 is 3.8 at −0.475 V, which is superior to other Ge–P-CNTs with different Ge content. This result suggests that a four-electron-transfer is main route for ORR on Ge–P-CNTs-2 electrode. Furthermore, the n-values of Ge–P-CNTs with different Ge content are also calculated at other potentials (−0.30 V, −0.40 V, −0.50 V, −0.70 V, −0.10 V). As shown in Table S2,† we can see that the n-values of Ge–P-CNTs-2 are higher than those of other Ge–P-CNTs at these potentials. The J_K value (21.3 mA cm⁻²) of Ge–P-CNTs-2 at −0.475 V exceeds those of Ge–P-CNTs-1 and Ge–P-CNTs-3. The above results further confirm that the suitable for Ge-doping amount is important factor for enhancing the electrocatalytic activity for oxygen reduction. Excellent ORR activity of the Ge–P-CNTs-2 catalyst was also gleaned from the much smaller Tafel slope of 70 mV per decade at low over-potentials (Fig. 5d) than that measured with Pt/C (87 mV per decade) in 0.1 M KOH.

Fig. 5 (a) LSV curves for the Ge–P-CNTs-2 at different rotating speeds; (b) K–L plots of Ge–P-CNTs-2 at various potential; (c) kinetic limiting current of different samples as well as the corresponding electron-transfer numbers obtained from K–L plots at −0.475 V; (d) Tafel plots obtained from the RDE measurements on Pt/C and Ge–P-CNTs-2 at 1600 rpm.

To confirm the ORR catalytic pathways of the materials, we performed rotating ring-disk electrode (RRDE) measurements monitor the formation of peroxide species (HO₂⁻) during the ORR process (Fig. 6a). As shown in Fig. 6b, the measured HO₂⁻ yields are below ∼10% for Ge–P-CNTs-2, over the potential range of −1.0 to −0.45 V, corresponding an electron transfer number of ∼3.8. It is worth noting that the calculated HO₂⁻ yields are above ∼20% and the electron transfer numbers are 3.3–3.5 in the potential range of −0.4 to −0.3 V. The electron transfer numbers (n) calculated from the rotating ring-disk electrode (RRDE) measurements are similar as those derived from K–L plots, reconfirming the nearly 4e oxygen reduction pathway with a small amount of 2e oxygen reduction.

Fig. 6 (a) Rotating ring-disk electrode voltammograms recorded with Ge–P-CNTs-2 (loading ∼ 0.303 mg cm⁻²) in O₂-saturated 0.1 M KOH at 1600 rpm. Disk current (I_d) (red line) is shown on the lower half and ring current (I_r) (blue line) is shown on the upper half of the graph. The disk potential was scanned at 10 mV s⁻¹ and the ring potential was constant at 0.5 V versus Ag/AgCl. (b) Percentage of peroxide (red line) and the electron transfer number (n) (black line) of Ge–P-CNTs-2 at various potentials, based on the corresponding RRDE data in (a).

The above electrochemical results indicate that the Ge, P co-doping in CNTs improves their electrochemical catalytic activity for ORR due to a synergistic effect between Ge and P atoms. How to understand the interesting synergistic effect?

Recently, it has been found that the ORR activity of the doped carbon is strongly related with the nanoscale work function, which is the required minimum energy for transferring an electron from the surface of carbon to the vacuum.^49–53 The work function of sp² carbon could be reduced by doping or co-doping, promoting the electron donation from electrocatalysts to O₂ and ORR intermediates. On this basis, the Ge/P mono- and co-doping effects on ORR were studied with DFT by calculating the work function of the carbon nanotubes with different Ge/P co-doping configurations. Fig. 7 shows the calculating mold and calculated results. The work function is 4.46 eV for the pristine sp² carbon, which is closed to the previous studies.⁵⁴ So its experimental results of ORR activity is low (CNTs in Fig. 4). After P mono-doping, the work function is slightly increased to 4.48 eV (PC₃ in Fig. 7), indicating the similar electron donation capacity in ORR process. This is confirmed by the similar ORR activity of P-CNTs with the pristine CNTs (Fig. 4). For Ge mono-doping, Ge dopant is inclined to remain in the carbon di-vacancy defect, forming a steady 4 coordinated state (GeC₄ in Fig. 7).⁴⁶ The work function of GeC₃ is 4.72 eV, even greater than the pristine sp² carbon (4.46 eV). This is consistent with the lower ORR activity of the Ge mono-doped CNTs (Ge-CNTs in Fig. 4). By gradually replacing adjacent carbon atoms of germanium with phosphorus, four Ge/P co-doping configurations were obtained (GePC₃, GeP₂C₂, GeP₃C, and GeP₄ + PC₃ in Fig. 7). The work functions gradually decrease by increasing the number of P atoms coordinated to Ge, which are 4.77 eV (GePC₃), 4.39 eV (GeP₂C₂), 4.27 eV (GeP₃C) and 4.23 eV (GeP₄ + PC₃). The theoretical results indicate that the synergistic effect of appropriate Ge–P-CNTs could effectively decrease the work function of the sp² carbon, and thus increase the ORR activity.

Fig. 7 The computed work functions for carbon nanotube and different Ge/P mono- or co-doping configurations. Atoms are show by balls of different color and labeled accordingly.

As we all known, for a new metal-free ORR electrocatalysts, a high anti-methanol performance and durability are very important for the application in fuel cell and metal–air batteries. The chronoamperometric method was used to further discuss the electrochemical selectivity of Ge–P-CNTs-2 and Pt/C, and the results are shown in Fig. 8a. Clearly, when equal methanol is added into 0.1 M KOH solution respectively, the Ge–P-CNTs-2 has less obvious change of the current track of ORR than the commercial Pt/C. This confirms that Ge–P-CNTs-2 has better tolerance to methanol electro-oxidation effect than Pt/C, and may have potential application in alkaline direct methanol fuel cells. The durability of Ge–P-CNTs-2 and Pt/C was also evaluated by chronoamperometric response (Fig. 8b). A 5.5 h test of Ge–P-CNTs-2 showed only a slight performance attenuation of 16.0%. In contrast, the Pt/C catalyst suffered an activity loss of nearly 26.0%. This result suggests that the durability of Ge–P-CNTs-2 is superior to Pt/C catalyst, which can be attributed to the strongly bonded heteroatoms in Ge–P-CNTs and the improved chemical and mechanical stability of the carbon nanotubes relative to that of the carbon-black-based Pt/C.

Fig. 8 (a) Methanol crossover tolerance test of Ge–P-CNTs-2 and Pt/C conducted by chronoamperometric response in 0.1 M KOH aqueous electrolyte; (b) current–time (i–t) chronoamperometric response of Ge–P-CNTs-2 and Pt/C electrodes at −0.27 V in 0.1 M KOH at 1600 rpm.

Conclusions

In summary, we have demonstrated for the first time the one-step fabrication of Ge and P co-doped carbon nanotubes (Ge–P-CNTs) catalysts with excellent ORR performance, which is better than that of carbon nanotubes catalysts doped solely with Ge atoms (Ge-CNTs) or P atoms (P-CNTs). The ORR catalytic synergistic effect of P and Ge elements in CNTs has been verified by density functional theory calculations, showing that the Ge/P co-doping could effectively reduce the work function of sp² carbon, thus facilitate electron donation during ORR process. Also, compared to the fuel-sensitive and vulnerable Pt/C, the Ge–P-CNTs electrocatalyst is much more tolerant to fuel crossover and displays long-term durability in alkaline media. As a low-cost fuel-cell catalyst for ORR, the Ge–P-CNTs catalysts has potential in other applications such as lithium–air batteries, photocatalysis, and heterocatalysis.

Acknowledgements

The work was financially supported by National Natural Science Foundation of China (No. 51503109, No. 21501105 and No. 51473081), ARC Discovery Project (Grant No. 130104759) for financial support.

Notes and references

1. N. Daems, X. Sheng, I. F. J. Vankelecom and P. P. Pescarmona, J. Mater. Chem. A, 2014, 2, 4085–4110.
2. Y. M. Tan, C. F. Xu, G. X. Chen, X. L. Fang, N. F. Zheng and Q. J. Xie, Adv. Funct. Mater., 2012, 22, 4584–4591.
3. R. Cao, J.-S. Lee, M. Liu and J. Cho, Adv. Energy Mater., 2012, 2, 816–829.
4. L. Dai, D. W. Chang, J. B. Baek and W. Lu, Small, 2012, 8, 1130–1166.
5. Y. C. Jin, F. Y. Chen, Y. M. Lei and X. Q. Wu, ChemCatChem, 2015, 7, 2377–2383.
6. Q. Li, R. Cao, J. Cho and G. Wu, Adv. Energy Mater., 2014, 4, 1301415.
7. H. W. Park, D. U. Lee, M. G. Park, R. Ahmed, M. H. Seo, L. F. Nazar and Z. Chen, ChemSusChem, 2015, 8, 1058–1065.
8. G. L. Tian, M. Q. Zhao, D. Yu, X. Y. Kong, J. Q. Huang, Q. Zhang and F. Wei, Small, 2014, 10, 2251–2259.
9. J. Yin, L. Wang, P. Yu, L. Zhao, C. Tian, B. Jiang, D. Zhao, W. Zhou and H. Fu, ChemElectroChem, 2015, 10, 1002.
10. C. Arbizzani, S. Righi, F. Soavi and M. Mastragostino, Int. J. Hydrogen Energy, 2011, 36, 5038–5046.
11. T. C. Nagaiah, S. Kundu, M. Bron, M. Muhler and W. Schuhmann, Electrochem. Commun., 2010, 12, 338–341.
12. W. Zhao, P. Yuan, X. L. She, Y. Z. Xia, S. Komarneni, K. Xi, Y. K. Che, X. D. Yao and D. J. Yang, J. Mater. Chem. A, 2015, 3, 14188–14194.
13. K. Waki, R. A. Wong, H. S. Oktaviano, T. Fujio, T. Nagai, K. Kimoto and K. Yamada, Energy Environ. Sci., 2014, 7, 1950–1958.
14. T. Cao, D. Wang, J. Zhang, C. Cao and Y. Li, Chemistry, 2015, 21, 14022–14029.
15. L. Liu, Y. P. Zhu, M. Su and Z. Y. Yuan, ChemCatChem, 2015, 7, 2765–2787.
16. C. Xu, Y. Su, D. Liu and X. He, Phys. Chem. Chem. Phys., 2015, 17, 25440–25448.
17. Z. W. Liu, Q. Q. Shi, F. Peng, H. J. Wang, H. Yu, J. C. Li and X. Y. Wei, Catal. Commun., 2012, 22, 34–38.
18. X. She, D. Yang, D. Jing, F. Yuan, W. Yang, L. Guo and Y. Che, Nanoscale, 2014, 6, 11057–11061.
19. Z. Sui, Q. Meng, X. Zhang, R. Ma and B. Cao, J. Mater. Chem., 2012, 22, 8767.
20. X. J. Huang, Y. N. Luan, P. F. Yao, J. S. Xie, L. Yu, Z. Y. Wu and P. Chen, RSC Adv., 2015, 5, 76599–76606.
21. L. Chen, X. Cui, Y. Wang, M. Wang, F. Cui, C. Wei, W. Huang, Z. Hua, L. Zhang and J. Shi, Chem.–Asian J., 2014, 9, 2915–2920.
22. Y. W. Ma, L. Y. Sun, W. Huang, L. R. Zhang, J. Zhao, Q. L. Fan and W. Huang, J. Phys. Chem. C, 2011, 115, 24592–24597.
23. D. Li, C. Lv, L. Liu, Y. Xia, X. She, S. Guo and D. Yang, ACS Cent. Sci., 2015, 1, 261–269.
24. S. Zhao, C. Li, H. Huang, Y. Liu and Z. Kang, New J. Chem., 2015, 10, 1039.
25. C. Zhang, N. Mahmood, H. Yin, F. Liu and Y. Hou, Adv. Mater., 2013, 25, 4932–4937.
26. S.-i. Takagi, VLSI Technol., Dig., 2003, 14, 115.
27. C. W. L. Minjoo, L. Lee, Z. Cheng, D. A. Antoniadis and E. A. Fitzgerald, Appl. Phys. Lett., 2001, 79, 3344.
28. X. H. Liu, S. Huang, S. T. Picraux, J. Li, T. Zhu and J. Y. Huang, Nano Lett., 2011, 11, 3991–3997.
29. K. H. Seng, M. H. Park, Z. P. Guo, H. K. Liu and J. Cho, Angew. Chem., Int. Ed., 2012, 51, 5657–5661.
30. C. H. Kim, H. S. Im, Y. J. Cho, C. S. Jung, D. M. Jang, Y. Myung, H. S. Kim, S. H. Back, Y. R. Lim, C.-W. Lee, J. Park, M. S. Song and W.-I. Cho, J. Phys. Chem. C, 2012, 116, 26190–26196.
31. K. H. Seng, M. H. Park, Z. P. Guo, H. K. Liu and J. Cho, Nano Lett., 2013, 13, 1230–1236.
32. Z. T. Cui, S. G. Wang, Y. H. Zhang and M. H. Cao, J. Power Sources, 2014, 259, 138–144.
33. Y. Zheng, Y. Jiao, L. Ge, M. Jaroniec and S. Z. Qiao, Angew. Chem., Int. Ed., 2013, 52, 3110–3116.
34. J. Liang, Y. Jiao, M. Jaroniec and S. Z. Qiao, Angew. Chem., Int. Ed., 2012, 51, 11496–11500.
35. H. Jiang, Y. Zhu, Q. Feng, Y. Su, X. Yang and C. Li, Chemistry, 2014, 20, 3106–3112.
36. G. Kresse and J. Hafner, Phys. Rev. B: Condens. Matter Mater. Phys., 1994, 49, 14251–14269.
37. G. Kresse and J. Furthmuller, Comput. Mater. Sci., 1996, 6, 15–50.
38. G. Kresse and J. Furthmuller, Phys. Rev. B: Condens. Matter Mater. Phys., 1996, 54, 11169–11186.
39. G. Kresse and J. Hafner, Phys. Rev. B: Condens. Matter Mater. Phys., 1993, 47, 558.
40. P. E. Blochl, Phys. Rev. B: Condens. Matter Mater. Phys., 1994, 50, 17953–17979.
41. G. Kresse and D. Joubert, Phys. Rev. B: Condens. Matter Mater. Phys., 1999, 59, 1758–1775.
42. J. P. Perdew, K. Burke and M. Ernzerhof, Phys. Rev. Lett., 1996, 77, 3865–3868.
43. J. P. Perdew, K. Burke and M. Ernzerhof, Phys. Rev. Lett., 1997, 78, 1396–1401.
44. R. Binions, C. J. Carmalt and I. P. Parkin, Polyhedron, 2003, 22, 1683–1688.
45. P. Song, X. J. Bo, A. Nsabimana and L. P. Guo, Int. J. Hydrogen Energy, 2014, 39, 15464–15473.
46. K. S. Prasad, R. Pallela, D. M. Kim and Y. B. Shim, Part. Part. Syst. Charact., 2013, 30, 557–564.
47. Z. W. Liu, F. Peng, H. J. Wang, H. Yu, W. X. Zheng and J. Yang, Angew. Chem., Int. Ed., 2011, 50, 3257–3261.
48. T. O. K. Prabhakaran, Surf. Sci., 1995, 325, 263–271.
49. I. Kruusenberg, S. Ratso, M. Vikkisk, P. Kanninen, T. Kallio, A. M. Kannan and K. Tammeveski, J. Power Sources, 2015, 281, 94–102.
50. K. Vignarooban, J. Lin, A. Arvay, S. Kolli, I. Kruusenberg, K. Tammeveski, L. Munukutla and A. M. Kannan, Chin. J. Catal., 2015, 36, 458–472.
51. Z. Zhou, X. Gao, J. Yan and D. Song, Carbon, 2006, 44, 939–947.
52. Y. Zhao, L. Yang, S. Chen, X. Wang, Y. Ma, Q. Wu, Y. Jiang, W. Qian and Z. Hu, J. Am. Chem. Soc., 2013, 135, 1201–1204.
53. L. Yang, S. Jiang, Y. Zhao, L. Zhu, S. Chen, X. Wang, Q. Wu, J. Ma, Y. Ma and Z. Hu, Angew. Chem., Int. Ed., 2011, 50, 7132–7135.
54. W. S. Su, T. C. Leung and C. T. Chan, Phys. Rev. B: Condens. Matter Mater. Phys., 2007, 76, 235413.

---

Footnotes

† Electronic supplementary information (ESI) available: XPS spectra and electrochemical properties of the Ge–P-CNTs electrodes. See DOI: 10.1039/c5ra26675k

‡ Q. Li and F. Yuan contributed equally to this work.

---