- Email: [email protected]
Anchoring effect of Ni2+ in stabilizing reduced metallic particles for growing single-walled carbon nanotubes
Accepted Manuscript 2+ Anchoring effect of Ni in stabilizing reduced metallic particles for growing singlewalled carbon nanotubes Maoshuai He, Xiao Wang, Lili Zhang, Qianru Wu, Xiaojie Song, Alexander I. Chernov, Pavel V. Fedotov, Elena D. Obraztsova, Jani Sainio, Hua Jiang, Hongzhi Cui, Feng Ding, Esko Kauppinen PII:
S0008-6223(17)31215-0
DOI:
10.1016/j.carbon.2017.11.093
Reference:
CARBON 12623
To appear in:
Carbon
Received Date: 8 September 2017 Revised Date:
7 November 2017
Accepted Date: 29 November 2017
Please cite this article as: M. He, X. Wang, L. Zhang, Q. Wu, X. Song, A.I. Chernov, P.V. Fedotov, E.D. 2+ Obraztsova, J. Sainio, H. Jiang, H. Cui, F. Ding, E. Kauppinen, Anchoring effect of Ni in stabilizing reduced metallic particles for growing single-walled carbon nanotubes, Carbon (2017), doi: 10.1016/ j.carbon.2017.11.093. 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.
ACCEPTED MANUSCRIPT Graphical Abstract A Ni-incorporated MgO catalyst was developed for predominant synthesis of (6, 5) single walled carbon nanotubes. Density functional theory-based calculations revealed that the unreduced subsurface Ni stabilized reduced Ni atoms on the surface,
AC C
EP
TE D
M AN U
SC
RI PT
facilitating the growth of carbon nanotubes with a narrow chirality distribution.
ACCEPTED MANUSCRIPT Anchoring Effect of Ni2+ in Stabilizing Reduced Metallic Particles for Growing Single-Walled Carbon Nanotubes Maoshuai He1,2,3*, Xiao Wang2, Lili Zhang4, Qianru Wu1, Xiaojie Song1, Alexander I. Chernov5,6,
RI PT
Pavel V. Fedotov5, Elena D. Obraztsova5, Jani Sainio7, Hua Jiang7, Hongzhi Cui1, Feng Ding2*, Esko
1
SC
Kauppinen7
School of Materials Science and Engineering, Shandong University of Science and Technology,
2
Center for Multidimensional Carbon Materials, Institute for Basic Science, UNIST-gil 50,
Ulju-gun, Ulsan 44919, Korea 3
M AN U
266590 Qingdao, China
Key Laboratory of Eco-Chemical Engineering, Ministry of Education, College of Chemistry and
4
TE D
Molecular Engineering, Qingdao University of Science and Technology, Qingdao 266042, China Shenyang National Laboratory for Materials Science, Institute of Metal Research, Chinese
EP
Academy of Sciences, 72 Wenhua Road, Shenyang 110016, China A.M. Prokhorov General Physics Institute RAS, 38 Vavilov Street, 119991 Moscow, Russia
6
National Research Nuclear University MEPhI (Moscow Engineering Physics Institute), Kashirskoe
AC C
5
hwy. 31, 115409, Moscow, Russia 7
Department of Applied Physics, Aalto University School of Science, P.O. Box 15100, FI-00076
Aalto, Finland
*Corresponding author. Tel: +86 532 8605 7925. E-mail address: [email protected] (M.S. He); Tel: +82-52 217 5703. E-mail address: [email protected] (F. Ding)
ACCEPTED MANUSCRIPT Abstract The suitability of the Ni-MgO catalyst as a catalyst in chiral-selective growth of single-walled carbon nanotubes (SWNTs) by chemical vapor deposition has been assessed. It reveals that catalyst
RI PT
calcination temperature plays an important role in affecting the catalyst performances. Using CO as the carbon precursor and a chemical vapor deposition reaction temperature of 600 oC, Ni-MgO
SC
pre-calcined at 600 oC demonstrates the best performances in catalyzing the growth of SWNTs with predominant (6, 5) species. Systematic characterizations on catalysts calcinated at different
M AN U
temperatures indicate that Ni2+ ions diffuse towards the interior of MgO matrix upon annealing. DFT-based calculations reveal that the binding energy between Ni2+ and adjacent Ni(0) is larger than that between Mg2+ and Ni (0), while Ni2+ situated deep inside MgO has weak interactions with surface Ni atoms. This work highlights the importance of subsurface Ni2+ in anchoring reduced
TE D
surface Ni atom, which inhibits the aggregation of Ni particles and therefore, facilitates the growth of SWNTs with a narrow chirality distribution.
EP
Keywords: single-walled carbon nanotube, catalyst, chirality selective, reduction temperature, solid
AC C
solution, anchoring effect 1. Introduction
Because of their unique structures, single-walled carbon nanotubes (SWNTs) exhibit extraordinary electrical and optical properties, which are sensitively dependent on their chirality indices (n, m) [1]. Although SWNTs have demonstrated great potentials in nanoelectronics and optoelectronics [2], their structural inhomogeneity greatly hinders their practical applications [3, 4]. To lift the hurdle for incorporating SWNTs into nanoelectronics, it is necessary to achieve SWNTs 2
ACCEPTED MANUSCRIPT with defined properties, diameters, and even chiralities. Several post-synthesis approaches, such as density gradient ultracentrifugation [5], DNA recognition [6] and gel chromatography [7] have been devoted to separate SWNTs with single chirality in the last decade. These approaches usually suffer
RI PT
from high cost and contamination of SWNTs [8, 9]. Besides, a dispersion process, which could destroy the integrities and pristine properties of SWNTs, is generally required prior to the separation
SC
procedure.
Different from separation techniques, chemical vapor deposition (CVD) synthesis of SWNTs
M AN U
with controlled (or specific or selective) chirality is an economic way for achieving SWNTs with defined structures and properties. Most catalyst systems developed so far for chirality selective synthesis of SWNTs are listed in Table 1. All the parameters related to CVD process, such as reaction temperature, carbon source, catalyst and support, are demonstrated to affect SWNT chirality
TE D
distribution. Generally, ethanol and CO are mostly adopted as the carbon source for selective growth. As part of the produced molecular fragments from ethanol is similar to that of CO [42], it can be
EP
concluded that the use of hydrogen-free CO as the carbon source favors the growth of SWNTs with a
AC C
narrow diameter distribution. Recently, we demonstrated that CO facilitates the nucleation of SWNTs on Fe particles by perpendicular mode, i.e. SWNTs possess diameters smaller than the catalyst particles, accounting for the chirality-selective synthesis of SWNTs [43]. Besides, low temperature is usually adopted in a number of experiments for chirality-selective growth of SWNTs. The roles of low temperature could be divided into two aspects. One is that the mobility of small-sized particles is low at low reaction temperatures, facilitating the growth of small diameter SWNTs with low formation energy [44]. The other is that nanoparticles preserve their solid state at 3
ACCEPTED MANUSCRIPT low reaction temperatures. Consequently, the solid nanoparticles could serve as templates for synthesizing SWNTs with structures matching the catalyst crystal symmetry [32, 35]. Table 1. A summary of catalysts applied for chirality selective growth of SWNTs. Cat., T, P, Abs.,
RI PT
PL, ED are the abbreviation for the catalyst, temperature, pressure, absorption, photoluminescence and electron diffraction, respectively. The indicated pressure value in the table is the total pressure of the CVD reaction. Active
T (oC)
P(atm)
particle
Co
750
1-10
Fe
600
1
CoMo/MgO
Co
750
1
FeCo/Zeolite
FeCo
650
0.013
FeCu/MgO
Fe
600
1
Co-MCM-41 CoxMg1-xO
Co
(n,m)
CO
600
5.8
550
SWNT selectivity (%)
(6, 5)
Ref.
Raman
Abs
PL
ED
40
42
51
25
[1015]
CH4
(6, 5)
/
/
54
/
[14]
CH4
(7, 5)
/
26
/
/
[15]
EtOH
(6, 5)
/
/
25
/
[16]
CO
(6, 5)
/
/
60
40
[17,
CO
18] (6, 5)
/
/
39.7
/
[19, 20]
5.8
CO
(6, 5)
/
/
/
/
[21]
Co
500
1
CO
(6, 5)
/
/
67
55
[22]
Co
600
1
CO
(6, 5)
/
/
39
/
[23]
AC C
Co/SiO2
Co
EP
CoMn/MCM-41
precursor
TE D
FeRu/SiO2
Enriched
M AN U
CoMo/SiO2
Carbon
SC
Cat.
Ni/SiO2
Ni
500
1
CO
(6, 5)
/
/
41
/
[24]
Fe/MgO
Fe
700
1
CO
(6, 5)
/
/
32
/
[25]
FeMn/MgO
Fe
600
1
CO
(6, 5)
/
/
46
/
[25]
CoMo/SiO2
Co
800
17.8
CO
(6, 5)
/
/
40
/
[26]
Co/TUD-1
Co
800
5.9
CO
(9, 8)
/
/
59.1
/
[27]
CoMo/MgO
Co
800
5.9
CO
(7, 5)
/
/
39.9
/
[28]
CoPt/SiO2
CoPt
800
1
EtOH
(6, 5)
/
/
30
/
[29]
Au/Al2O3-SiO2
Au
700
0.00049
CH4/H2
(6, 5)
/
/
/
/
[30]
4
ACCEPTED MANUSCRIPT Co
600
10-6
C2H2
(6, 5)
26.8
/
/
/
[31]
Mo-sapphire
Mo2C
850
1
EtOH/H2
(12, 6)
90
/
/
/
[32]
W-sapphire
WC
850
1
EtOH/H2
(8, 4)
80
/
/
/
[32]
Mo-quartz
Mo2C
830
1
EtOH
(8, 4)
46.6
/
/
/
[33]
Fe-quartz
Fe
820-8
1
EtOH
(15, 2)
21
/
/
/
[34]
RI PT
Co-SiO2
80 W6Co7
1030
1
EtOH/H2
(12, 6)
94.4
/
/
/
[35]
CoW-SiO2
W6Co7
1050
1
EtOH/H2
(16, 0)
79.2
/
/
/
[36]
CoW-SiO2
W6Co7
1050
1
EtOH/H2O
(14, 4)
97.4
/
/
/
[37]
CoW-SiO2
W6Co6
750
0.0128
EtOH
(12, 6)
1200
1
70
/
/
/
[38]
/
M AN U
/
/
/
[39]
C Ferrocene
Fe
SC
CoW-SiO2
CH4
(9, 9),
(12, 12)
Fe
880
1
FeNi
NixFe1-x
600
1
CO/NH3
(13, 12)
/
/
/
13.7
[40]
C2H2/H2
(8, 4)
/
/
39.1
/
[41]
TE D
Ferrocene
Besides carbon feedstocks and temperature, the interplay between metal and support, which
EP
regulates the catalyst performances, can affect SWNT growth results. To facilitate the synthesis of SWNTs at low reaction temperature, it is necessary to develop catalyst systems affording metal
AC C
particles which are well stabilized at reaction conditions. Therefore, the easy reduction and the presence of anchors are necessary for forming small particles to grow SWNTs. In some bimetallic catalysts, the addition of a second metal, such as Mo [45], Ru [14], Cu [17] or Mn [25], not only facilitates the reduction of active component, but also prevents metal coalescence. In addition to bimetallic catalysts, some monometallic catalysts, like SiO2 supported Co [23] and Ni [24] catalysts, have also been demonstrated to efficiently synthesize SWNTs at low temperatures. Under reaction
5
ACCEPTED MANUSCRIPT conditions, it is postulated that part of metal ions could be reduced to generate metal clusters, which can be stabilized by the unreduced counterparts, like Co2+ and Ni2+ in the above two catalysts. Compared with SiO2-supported catalysts, MgO-supported catalysts show great promises in scaling
RI PT
up because of its low-cost and easy removal after CVD. MgO supported Co [22, 46] and Fe [25] catalysts have already been reported for chirality-selective synthesis of SWNTs. Nevertheless,
SC
monometallic Ni catalyst supported by MgO is rarely applied for controlling the chirality of SWNTs. More importantly, the anchoring mechanisms exerted by metal ions have not been well addressed
M AN U
yet.
To fill this gap, we herein develop a series of MgO supported Ni catalysts prepared by impregnation and subsequent calcination. Catalyst performances are evaluated by synthesizing carbon nanotubes, the diameters of which are preliminarily characterized by Raman spectroscopy.
TE D
Photoluminescence (PL) is further adopted to determine the chirality distribution of SWNTs with small diameters. By combining catalyst characterizations and DFT calculations, the anchoring effects
EP
of Ni2+ in stabilizing reduced metallic Ni particles will be elucidated.
AC C
2. Materials and methods
2.1 Preparation of Ni-MgO catalyst Porous MgO was obtained by decomposition of magnesium carbonate hydroxide hydrate (Aldrich, 99%) at 400 oC for 1 h [17]. The BET area of MgO prepared by such a technique is 50 m2/g, as previously determined by Li et al. [47]. MgO supported Ni catalysts were prepared by impregnating 4.0 g MgO with an aqueous solution containing 1.4 g Ni(NO3)26H2O. The resulting mixtures were dried overnight at a temperature of 80 oC. Several aliquots of the dried catalysts were 6
ACCEPTED MANUSCRIPT then calcined in air for 20 h at 450 oC, 600 oC and 800 oC, respectively. The catalysts are denoted as Ni-MgO (450), Ni-MgO (600) and Ni-MgO (800) respectively. 2.2 Growth of carbon nanotubes by CVD
RI PT
CVD growth of carbon nanotubes was performed on a CCR1000 (Linkam) micro-reactor. About 20 mg catalyst powder was loaded into sample holder and heated to 600 oC under a 50 sccm Ar flow.
SC
After being stabilized at 600 oC, CO with a flow rate of 50 sccm was introduced to replace Ar and the reaction was run for 15 min. CO was finally switched off and the system was cooled down under
M AN U
the protection of Ar. All the process is performed at ambient pressure. 2.3 Characterizations of catalyst particles and carbon nanotubes
XRD measurements were performed with a Panalytical diffractometer using Cu Kα (λ= 0.15406 nm) radiation to ascertain the crystal structure of catalyst. Scans were performed over the 2θ
TE D
range from 10o to 90°. The surface concentrations of Ni in the catalysts were determined by X-ray photoelectron spectroscopy (XPS), which was carried out on a Surface Science Instruments SSX-100
EP
ESCA spectrometer using monocromated Al Kα radiation. Peak positions were calibrated using the
AC C
O 1s photoelectron peak of MgO at 530.5 eV as an internal reference [48]. Scanning electron microscopy (SEM) was carried out with a Nova Nano SEM450 to characterize the catalysts. To clarify
the
morphologies
and
compositions
of
as-prepared
catalysts,
scanning
transmission/transmission electron microscopy (STEM/TEM) observations and energy-dispersive X-ray (EDX) mapping analysis were conducted using an FEI Tecnai F20 TEM operated at 200 kV. Raman spectrometer (Jobin Yvon LabRam 300) with an excitation laser wavelength of 632.8 nm (1.96 eV) was applied to characterize the as-produced carbon nanotubes. A JEOL JSM-7500F 7
ACCEPTED MANUSCRIPT SEM was performed to characterize the morphology of the carbon nanotubes. The products were purified by hydrochloric acid with subsequent water wash. After drying in oven, the samples were sonicated in acetone and collected onto porous carbon TEM grid for structural characterizations
RI PT
(JEOL JEM-2200FS 2×CS corrected TEM). To estimate the chirality distributions, the purified SWNTs were dispersed into an aqueous solution of sodium cholate hydrate (1 wt%) by tip
SC
sonication. The dispersion was subjected to PL measurements after centrifugation at 100 000 g for 40 min (Ultracentrifuge Optima Max-E, Beckman-Coulter, MLA-80 rotor). The PL characterizations
2.4 Computational methods and details
M AN U
were carried out with a Horiba Jobin-Yvon NanoLog system.
All the first principles calculations were performed by Vienna Ab initio Simulation Package based on density functional theory. Perdew-Burke-Ernzerhof (PBE) functional within the frame of
TE D
generalized gradient approximation (GGA) was chosen to describe the electronic exchange and correlation. The interaction between the valence electrons and kernel was depicted by the projected
EP
augmented wave (PAW) method with the cutoff energy set as 400 eV. With the aim to describe the
AC C
interlayer interaction more accurate, the Grimme DFT-D2 method is included. A three-layer slab of MgO with a supercell size of 11.78×11.78×20.00 Å3 was chosen as the theoretical model. The bottom layer is fixed as the bulk and the top two layers can be relaxed. The vacuum spacing is larger than 10 Å and is enough to void avert the interference of adjacent images. As the Ni2+ ion can substitute Mg2+ ion in the MgO martix, one, two and four Ni2+ ions are chosen to replace the equivalent number of Mg2+ ions on the surface or subsurface of the slab, respectively. To investigate the anchoring effect of Ni2+ on the Ni atom, one Ni atom is placed above the slab. When optimizing 8
ACCEPTED MANUSCRIPT the geometrical structure, 4×4×1 k-points were sampled uniformly in the Brillouin zone using Monkhorst-Pack sampling method. A recommended conjugated gradient algorithm was adopted to ensure a force convergence less than 0.01 eV/A on each atom. After geometrical relaxation, the
ܧ ൌ ܧ௦ ܧே െ ܧேି௦ 3. Results and discussion
RI PT
binding energy between the Ni atom and slab is calculated for further analysis following the formula: (1)
SC
Figure 1a presents Raman spectra of carbon nanotubes grown on Ni-MgO catalysts at 600 oC
M AN U
using CO as the carbon source. In the Raman spectrum of the product grown on Ni-MgO (450), radial breathing modes (RBMs) in the low frequency range are absent, suggesting the lack of SWNTs in the product. Instead, only large diameter multi-walled carbon nanotubes were grown on the Ni-MgO (450) catalyst. As tube diameter is correlated with catalyst size, the active catalyst particles
AC C
EP
TE D
formed on Ni-MgO (450) during CVD should possess large diameters.
Figure 1. (a) Raman spectra (excitation wavelength: 632.8 nm) of carbon nanotubes grown on different Ni-MgO catalysts at 600 oC using CO as the carbon source. (b) Contour plots of normalized PL emission intensities under various excitation energies for SWNTs grown on Ni-MgO (600). (c) Normalized PL emission intensities of identified SWNTs deduced from Figure b. Different from Ni-MgO (450), Ni-MgO catalysts annealed at higher temperatures afford the synthesis of SWNTs under identical CVD conditions, as indicated by the presence of RBMs in the 9
ACCEPTED MANUSCRIPT Raman spectra of carbon nanotubes produced on Ni-MgO (600) and Ni-MgO (800). The RBMs excited by 632.8 nm laser consist of two major peaks: one is centered at ∼190 cm-1, the other is located at ∼292 cm-1. For SWNTs grown on Ni-MgO (800), the intensity of the peak centered at 190
RI PT
cm-1 is much stronger than the one at 292 cm-1, indicating most SWNTs have diameters larger than 1.0 nm. In addition, the high intensity of D mode and the relatively low intensity of G mode indicate
SC
the coexistence of much carbonaceous impurities. In contrast, Raman spectrum of SWNTs grown on Ni-MgO (600) exhibits a relatively low intensity of the D mode and a more intensive RBM peak at
M AN U
292 cm-1, implying SWNTs grown on Ni-MgO (600) have relatively small average diameters. The production of small-diameter SWNTs on Ni-MgO (600) is also verified by the appearance of intermediate frequency modes at
750 cm-1 in the Raman spectrum. The intensity of intermediate
frequency mode has been proven to increase with increasing SWNT curvature [49]. Therefore, such
TE D
characteristic intermediate frequency modes could be readily observed in sub-nanometer SWNTs [20] like what are achieved here. Overall, Ni-MgO (600) exhibits the best performances in catalyzing
EP
sub-nanometer SWNTs.
AC C
To estimate the chirality distribution of SWNTs grown on Ni-MgO (600), SWNTs dispersed in sodium cholate solution was subjected to PL characterizations. Figure 1b depicts the PL map of the SWNT dispersion where (6, 5) SWNT displays the most intense emission intensity. Besides (6, 5) SWNTs, other semiconducting SWNTs with diameters smaller than 0.85 nm, including (8, 3), (7, 5) and (8, 4) SWNTs, are the major species identified in the product. Figure 1c presents the normalized PL emission intensity of different SWNT species. Compared with SWNTs grown on SiO2 supported Ni catalysts under identical conditions [24], the average diameter of SWNTs grown on Ni-MgO 10
ACCEPTED MANUSCRIPT (600) is a bit smaller. Assuming the abundances of SWNTs are proportional to their PL emission intensities, the percentage of (6, 5) SWNT in the product is about 44%. To further confirm the morphology and structure of SWNTs, SEM and TEM characterizations
RI PT
were performed and some typical images are demonstrated in Figure 2a & 2b. Detailed TEM analysis demonstrates that the SWNTs have relatively small diameters centered at ~0.8 nm (Figure
SC
3c). The growth of smaller-diameter SWNTs on Ni-MgO (600) can be attributed to the specific metal-support interaction, which could affect the morphology of metal nanoparticles formed during
M AN U
CVD growth process. In addition, an epitaxial relationship between reduced Ni particles and MgO support [50], like in the case of Co-MgO catalyst [22] prepared by the same approach, cannot be
EP
TE D
excluded.
Figure 2. SEM image (a) and TEM image (b) of carbon nanotubes grown on Ni-MgO (600) at 600 o
images.
AC C
C using CO as the carbon source. (c) Diameter distribution of SWNTs determined from TEM
Aiming to clarify why the calcination temperature affects the catalyst performances for synthesizing SWNTs, XRD and XPS characterizations were employed on the three Ni-MgO catalysts. Figure 3a depicts the XRD patterns of Ni-MgO catalysts annealed at three different temperatures over the 2θ range from 10o to 90°. As both Ni2+ and Mg2+ possess same face-centered cubic crystal structures and similar ionic radii [51], calcination could promote the migration of Ni2+ 11
ACCEPTED MANUSCRIPT ions towards MgO interior and the formation of NiO-MgO solid solution. Such progressive diffusion of Ni ions across the MgO lattice can be well reflected by XRD features [52]. On the one hand, the lack of NiO peak signals the formation of NiO-MgO solid solution or a highly dispersed NiO form
RI PT
[52]. On the other hand, the deduced lattice parameter could reflect the forms of NiO dissolved into MgO matrix. Figure 3b presents the magnified XRD spectra in the range 2θ = 73-81o of the three
SC
Ni-MgO catalysts. All catalysts show two well-resolved XRD peaks ascribable to MgO (311) reflection and MgO (222) reflection, respectively. With the increase of the calcination temperature,
M AN U
the characteristic peaks shift to higher angles and their corresponding FWHMs decrease. On the basis of the 2θ value of the XRD peak, MgO lattice parameter (a) can be calculated from the equation:
a ൌ ඥ(ℎଶ ݇ ଶ ݈ ଶ )
ߣ 2 sin ߠ
TE D
Where h, k, l are Miller indices of planes in MgO cubic crystal and λ is the wavelength of Cu Kα (0.15406 nm). The reflection peaks for Ni-MgO (450) and Ni-MgO (600) are almost identical,
EP
suggesting the similar bulk crystal structure of MgO in both catalysts. Based on the 2θ value of 78.4o for (222) planes, the lattice parameter of MgO is determined to be 0.422 nm. Increasing the
AC C
calcination temperature of Ni-MgO to 800 oC shifts the lattice reflection to a higher 2θ value (78.6o) and therefore a smaller lattice parameter (0.421 nm). Ever-increasing incorporation of Ni2+ ions into the MgO structure accounts for the progressive 2θ upshift and the decrease in FWHM (Figure 3b) upon calcination [52].
12
SC
RI PT
ACCEPTED MANUSCRIPT
M AN U
Figure 3. (a) XRD patterns of Ni-MgO catalysts calcined at different temperatures. (b) For comparison, the overlapped spectra in the 73o-81o range are presented. (c) XPS spectra of Ni 2p in Ni-MgO catalyst series. (d) Atomic ratios of Ni to Mg in different Ni-MgO catalysts determined by XPS.
The effects of air calcination on catalyst reducibility and performances could also be reflected
TE D
by the changes in both Ni surface compositions and binding energies of XPS spectrum [53]. Figure 3c compares the Ni 2p photoelectron peaks and atomic ratios of Ni/Mg in the three catalysts. In all
EP
the spectra, the binding energy of the Ni 2p3/2 core level is centered around 856.1 eV, which is much
AC C
higher than that of metallic Ni (852.8 eV), suggesting that the Ni exists as oxide state. The binding energy of the main peak reflects the nature of the ligands to which Ni is bonded because charge transfer involves transfer of an electron from an L-band to the 3d9 band [54]. Besides, the Ni 2p3/2 peaks show a symmetric profile, which is different from the multiplet-split characteristic of asymmetric peak of “free” NiO, suggesting that the Ni2+ could dissolves inside the MgO matrix. In addition, a satellite peak with a binding energy of ~862 eV is clearly seen, which is attributed to the unscreened cd8 final-state configurations [55]. Although the spectra are quite similar, Figure 3d 13
ACCEPTED MANUSCRIPT indicates that, the Ni/Mg atomic ratio detected by XPS progressively decreases, suggesting the
SC
RI PT
depletion of Ni in the sampled thickness upon calcination.
M AN U
Figure 4. (a) TEM image of Ni-MgO (600) catalyst. (b) STEM image from the same catalyst particle and EDX mapping from the crossed area indicated by the white square. TEM and SEM were also applied to characterize the morphology and element distribution of the catalysts. Figure 4a depicts a TEM image of Ni-MgO (600) catalyst. There are no clear isolated
TE D
Ni-containing nanoparticles on the MgO surface. STEM image and EDX mapping were further applied to analyze the catalyst composition (Figure 4b). It is clearly shown that the elements of Mg,
EP
O and Ni are almost overlapped, and Ni is well dispersed in the MgO support. The TEM
AC C
characterization results of Ni-MgO (450) catalyst and Ni-MgO (800) catalyst are presented in Supporting Information Figure S2&S3, respectively. In both cases, Ni elements are also well dispersed in the catalysts. Such characterization results are in agreement with SEM results, which are presented in Supporting Information Figure S4, S5 & S6. In addition, the crystalline degree of the catalysts increases with the increase of calcination temperature, as revealed by high resolution SEM (Supporting Information Figure S7). The observations also agree with the XRD characterization results. However, as the lattice constants of MgO and NiO are similar, it is difficult to distinguish 14
ACCEPTED MANUSCRIPT isolated NiO nanoparticles possibly located on MgO surface by TEM. As clarified by Arena et al.[52, 53], there are four forms of Ni2+ ions in Ni-MgO catalysts: (i) “unreacted” NiO on MgO surface, which is uninfluenced by the MgO support and could be readily
RI PT
reduced at temperatures below 400 oC; (ii) Ni2+ ions in the outermost layer of the MgO structure, having square-pyramidal coordination and can be reduced at a temperature of around 600 oC; (iii) Ni2+ ions located in the sub-surface layers of the MgO lattice, which is reducible at a temperature of
M AN U
and is reducible at temperatures higher than 1000 oC.
SC
about 750 oC; (iv) Ni2+ ions dissolved in the MgO matrix, forming a “bulk” NiO-MgO solid solution
In order to evaluate the interactions between reduced Ni atom and cations in the catalyst during CVD reaction, DFT-based calculations are performed to confirm our hypothesis. In our theoretical mode, Ni atom situated on MgO matrix where equivalent Ni2+ ions substitute for one, two and four
TE D
Mg2+ ions on the first or second layer were investigated. Without Ni2+ substitution, the Ni atom favors bonding with the O atom on the top site (shown in Figure 5a, lateral view is presented in
EP
Supporting Information Figure S1) with a calculated binding energy of 1.97 eV (Figure 5b). If some Mg2+ ions on the first layer are substituted by Ni2+, the most stable configuration for the adhered Ni
AC C
atom is bridge site between the Ni2+ and O2- instead of the top site. In addition, the adjacent Ni2+ ions can also contribute to a stronger binding to the adhered Ni atom. As shown in Figure 5b, when one Ni2+ replaces one Mg2+ on the surface, the binding energy increases from 1.97 eV to 2.41 eV compared with the case of no Ni2+ substitution. The binding energy will increase as the number of substituted Mg2+ atoms increases and ultimately reaches 3.03 eV when four Mg2+ ions are substituted (Figure 5c-e), which is 1.06 eV higher than that in the unsubstituted case. Consequently, the 15
ACCEPTED MANUSCRIPT substituted Ni2+ on the surface can act as an anchor site to stabilize the absorbed Ni atoms and impede their subsequent migration, which is propitious to nucleate small catalyst particles. When the Mg2+ ions on the second layer of MgO are substituted by the Ni2+ (Figure 5f-h), unlike the
RI PT
substitution in the first layer, the most stable site for the absorbed Ni atom is the top of the O atom, which is the same to the unsubstituted case. Besides, there is only nearly 0.10 eV increase of the binding energy and no obvious dependence on the number of substituted Mg2+ ions, which implies
SC
that the Ni2+ substitution on the second layer of MgO contributes little to stabilize the absorbed Ni
M AN U
atoms.
Under our CVD reaction conditions, “unreacted” NiO can be readily reduced and form large-diameter metallic Ni particles, responsible for the synthesis of multi-walled carbon nanotubes. Ni2+ ions in the outmost layer of MgO could also be reduced at conditions adopted in our
TE D
experiments. While subsurface Ni2+ ions and “bulk” Ni2+ ions cannot be reduced. Taking our calculation results into account, it is suggested that the subsurface Ni2+ could act as anchors to
EP
constrain the mobility of reduced surface Ni. While “bulk” Ni2+ ions, which is far from the surface,
AC C
could not stabilize the Ni clusters formed on surface.
16
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
EP
Figure 5. (a) Most stable configuration of Ni atom absorbed on MgO top layer without Ni substitution. (b) Binding energy of absorbed Ni atom on MgO. Most stable configurations of one Ni
AC C
atom absorbed on MgO top surface with one, two, four Mg2+ ions on the first layer (c-e) or second layer (f-h) substituted by equivalent Ni2+ ions, respectively. The O2-, Mg2+, Ni2+ ions are colored in red, green and blue, respectively. The absorbed Ni (0) is displayed smaller than those in the substrate for distinctness.
17
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Figure 6. Schematic illustration of catalyst evolution upon the introduction of CO at 600 oC for Ni-MgO catalysts calcined at (a) 450 oC; (b) 600 oC; (c) 800 oC. Because of sufficient subsurface Ni2+ ions in Ni-MgO (600), which can serve as anchors for constraining the mobility of reduced Ni
TE D
clusters, small diameter SWNTs with a narrow chirality distribution are achieved. ～ Therefore, the coexistence of Ni2+ ions in the outmost layer and in the subsurface of MgO is of
EP
great importance for low temperature growth of SWNTs. As for Ni-MgO (450), there exist only “unreacted” and surface NiO. Both can be readily reduced to form large particles for the growth of
AC C
large diameter multi-walled carbon nanotubes (Figure 6a).～With the increase of the calcination temperature, more Ni2+ ions migrate towards the core of MgO support, accounting for the depletion of surface Ni detected by the XPS (Figure 3d). As the XRD pattern of Ni-MgO (600) does not exhibit obvious peak shift compared with that of Ni-MgO (450) (Figure 3b), it is suggested that the depletion of surface Ni mainly contribute to the generation of subsurface Ni2+ instead of diffusion inside bulk MgO matrix. Introducing CO at CVD reaction temperature favors the formation of small Ni clusters from the reduction of outmost layer Ni2+ ions, which could be further stabilized by 18
ACCEPTED MANUSCRIPT subsurface Ni2+ ions. The stabilized metallic Ni particles are responsible for the synthesis of SWNTs with small diameters. Further considering the use of CO as the carbon source, it could facilitate the dissolution of carbon inside metal particle, resulting in the nucleation of SWNTs with a
RI PT
perpendicular mode [43] and synthesis of SWNTs with a narrow chirality distribution (Figure 6b). However, further increasing the catalyst calcination temperature to 800 oC causes the diffusion
SC
of subsurface and outmost layer Ni2+ ions. Consequently, these Ni2+ ions migrate into MgO matrix and form a bulk NiO-MgO solid solution, which explains the shift of XRD peaks (Figure 3a&3b)
M AN U
and the diminishing Ni content in XPS spectrum (Figure 3d). As a result, only trace amount of residual Ni2+ ions could be reduced to form metal Ni particles, catalyzing the growth of SWNTs. It is noted that because the amount of subsurface Ni2+ ions is diminishing, their “anchoring” effect is not as efficient as in Ni-MgO (600), particles and resulting SWNTs grown on Ni-MgO (800), have
4. Conclusion
TE D
relatively large diameters (Figure 6c).～
EP
To conclude, Ni-MgO catalysts annealed at different temperatures are applied for the synthesis
AC C
of carbon nanotubes at 600 oC using CO as the carbon feedstock. Among them, Ni-MgO (600) displays the best catalyst performance not only in growing small-diameter SWNTs, but also in achieving SWNTs with a narrow chirality distribution. By means of a comparative study, the progressive change of Ni-MgO catalysts upon calcination at different temperatures has been marked. The binding energy deduced from DFT calculations revealed that reduced Ni(0) atom can be well anchored by underlying Ni2+ ions. Therefore, the coexistence of outmost layer Ni2+ ions and subsurface Ni2+ ions is the key for forming small-diameter metallic Ni particles in reducing 19
ACCEPTED MANUSCRIPT environment at low temperatures. Such reduced Ni particles anchored by subsurface Ni2+ ions and the use of CO as carbon source are responsible for the chiral-selective growth of SWNTs. This work not only extends the development of catalyst system for chiral-selective synthesis of SWNTs, but
RI PT
also helps understand the anchoring effect of metal ions in stabilizing reduced metal particles. Acknowledgements
SC
The research leading to these results has received funding from Taishan scholarship of climbing plan (No. tspd20161006) and the National High Technology Research and Development Program of
M AN U
China (863 Program, 2015AA034404). The authors would also like to acknowledge the Natural Science Foundation of Shandong Province of China (No. ZR2016EMM10) and Scientific Research Foundation of Shandong University of Science and Technology for Recruited Talents (No.
Supplementary data
TE D
2016RCJJ001). AIC acknowledges the support by the Alexander von Humboldt Foundation.
Lateral view of the optimized configurations when Ni(0) atom is placed on top of MgO with (or
EP
without) Ni2+ substitution; Detailed TEM and SEM characterization results of the Ni-MgO catalysts
AC C
calcinated at different temperatures. References
[1] M.S. Dresselhaus, G. Dresselhaus, R. Saito, A. Jorio, Raman spectroscopy of carbon nanotubes, Phys. Rep. 409 (2005) 47-99. [2] S.M. Bachilo, M.S. Strano, C. Kittrell, R.H. Hauge, R.E. Smalley, R.B. Weisman, Structure-assigned optical spectra of single-walled carbon nanotubes, Science 298 (2002) 2361-2366. [3] M.C. Hersam, Progress towards monodisperse single-walled carbon nanotubes, Nat. Nanotechnol. 3 (2008) 387-394. [4] R.H. Baughman, A.A. Zakhidov, W.A. de Heer, Carbon nanotubes--the route toward applications, Science 297 (2002) 7877-92. [5] M.S. Arnold, A.A. Green, J.F. Hulvat, S.I. Stupp, M.C. Hersam, Sorting carbon nanotubes by electronic structure using density differentiation, Nat. Nanotechnol. 1 (2006) 60-65. 20
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
[6] X. Tu, S. Manohar, A. Jagota, M. Zheng, DNA sequence motifs for structure-specific recognition and separation of carbon nanotubes, Nature 460 (2009) 250-253. [7] H. Liu, D. Nishide, T. Tanaka, H. Kataura, Large-scale single-chirality separation of single-wall carbon nanotubes by simple gel chromatography, Nat. Commun. 2 (2011) 309. [8] T. Lei, I. Pochorovski, Z. Bao, Separation of semiconducting carbon nanotubes for flexible and stretchable electronics using polymer removable method, Acc. Chem. Res. 50 (2017) 1096-1104. [9] C. Luo, D. Wan, J. Jia, D. Li, C. Pan, L. Liao, A rational design for the separation of metallic and semiconducting single-walled carbon nanotubes using a magnetic field, Nanoscale 8 (2016) 13017-13024. [10] D.E. Resasco, W.E. Alvarez, F. Pompeo, L. Balzano, J.E. Herrera, B. Kitiyanan, et al., A scalable process for production of single-walled carbon nanotubes (swnts) by catalytic disproportionation of co on a solid catalyst, J. Nanopart. Res. 4 (2002) 131-136. [11] A. Jorio, A. Santos, H. Ribeiro, C. Fantini, M. Souza, J. Vieira, et al., Quantifying carbon-nanotube species with resonance raman scattering, Phys. Rev. B 72 (2005) 075207. [12] Y. Tian, H. Jiang, I.V. Anoshkin, L.J. Kauppinen, K. Mustonen, A.G. Nasibulin, et al., A reference material of single-walled carbon nanotubes: Quantitative chirality assessment using optical absorption spectroscopy, RSC Advances 5 (2015) 102974-102980. [13] S.M. Bachilo, L. Balzano, J.E. Herrera, F. Pompeo, D.E. Resasco, R.B. Weisman, Narrow (n,m)-distribution of single-walled carbon nanotubes grown using a solid supported catalyst, J. Am. Chem. Soc. 125 (2003) 11186-11187. [14] X. Li, X. Tu, S. Zaric, K. Welsher, W.S. Seo, W. Zhao, et al., Selective synthesis combined with chemical separation of single-walled carbon nanotubes for chirality selection, J. Am. Chem. Soc 129 (2007) 15770-15771. [15] G. Lolli, L. Zhang, L. Balzano, N. Sakulchaicharoen, Y. Tan, D.E. Resasco, Tailoring (n,m) structure of single-walled carbon nanotubes by modifying reaction conditions and the nature of the support of como catalysts, J. Phys. Chem. B 110 (2006) 2108-2115. [16] Y. Miyauchi, Fluorescence spectroscopy of single-walled carbon nanotubes synthesized from alcohol, Chem. Phys. Lett. 387 (2004) 198-203. [17] M. He, A.I. Chernov, P.V. Fedotov, E.D. Obraztsova, J. Sainio, E. Rikkinen, et al., Predominant (6,5) single-walled carbon nanotube growth on a copper-promoted iron catalyst, J. Am. Chem. Soc. 132 (2010) 13994-13996. [18] M. He, B. Liu, A.I. Chernov, E.D. Obraztsova, I. Kauppi, H. Jiang, et al., Growth mechanism of single-walled carbon nanotubes on iron–copper catalyst and chirality studies by electron diffraction, Chem. Mater. 24 (2012) 1796-1801. [19] C.Z. Loebick, S. Derrouiche, N. Marinkovic, C. Wang, F. Hennrich, M.M. Kappes, et al., Effect of manganese addition to the co-mcm-41 catalyst in the selective synthesis of single wall carbon nanotubes, J. Phys. Chem. C 113 (2009) 21611-21620. [20] C. Zoican Loebick, R. Podila, J. Reppert, J. Chudow, F. Ren, G.L. Haller, et al., Selective synthesis of subnanometer diameter semiconducting single-walled carbon nanotubes, J. Am. Chem. Soc. 132 (2010) 11125-31. [21] N. Li, X. Wang, F. Ren, G.L. Haller, L.D. Pfefferle, Diameter tuning of single-walled carbon nanotubes with reaction temperature using a co monometallic catalyst, J. Phys. Chem. C 113 (2009) 21
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
10070-10078. [22] M. He, H. Jiang, B. Liu, P.V. Fedotov, A.I. Chernov, E.D. Obraztsova, et al., Chiral-selective growth of single-walled carbon nanotubes on lattice-mismatched epitaxial cobalt nanoparticles, Sci. Rep. 3 (2013) 1460. [23] M. He, A.I. Chernov, P.V. Fedotov, E.D. Obraztsova, E. Rikkinen, Z. Zhu, et al., Selective growth of swnts on partially reduced monometallic cobalt catalyst, Chem. Commun. 47 (2011) 1219-1221. [24] M. He, A.I. Chernov, E.D. Obraztsova, J. Sainio, E. Rikkinen, H. Jiang, et al., Low temperature growth of swnts on a nickel catalyst by thermal chemical vapor deposition, Nano Res. 4 (2011) 334-342. [25] M. He, P.V. Fedotov, A. Chernov, E.D. Obraztsova, H. Jiang, N. Wei, et al., Chiral-selective growth of single-walled carbon nanotubes on fe-based catalysts using co as carbon source, Carbon 108 (2016) 521-528. [26] B. Wang, L. Wei, L. Yao, L.J. Li, Y. Yang, Y. Chen, Pressure-induced single-walled carbon nanotube (n,m) selectivity on co-mo catalysts, J. Phys. Chem. C 111 (2007) 14612-14616. [27] H. Wang, B. Wang, X.Y. Quek, L. Wei, J.W. Zhao, L.J. Li, et al., Selective synthesis of (9,8) single walled carbon nanotubes on cobalt incorporated tud-1 catalysts, J. Am. Chem. Soc. 132 (2010) 16747-16749. [28] B. Wang, Y. Yang, L.-J. Li, Y. Chen, Effect of different catalyst supports on the (n,m) selective growth of single-walled carbon nanotube from co–mo catalyst, J. Mater. Sci. 44 (2009) 3285-3295. [29] B. Liu, W. Ren, S. Li, C. Liu, H.M. Cheng, High temperature selective growth of single-walled carbon nanotubes with a narrow chirality distribution from a copt bimetallic catalyst, Chem. Commun. 48 (2012) 2409-2411. [30] Z. Ghorannevis, T. Kato, T. Kaneko, R. Hatakeyama, Narrow-chirality distributed single-walled carbon nanotube growth from nonmagnetic catalyst, J. Am. Chem. Soc. 132 (2010) 9570-9572. [31] M. Fouquet, B.C. Bayer, S. Esconjauregui, R. Blume, J.H. Warner, S. Hofmann, et al., Highly chiral-selective growth of single-walled carbon nanotubes with a simple monometallic Co catalyst, Phys. Rev. B 85 (2012) 235411. [32] S. Zhang, L. Kang, X. Wang, L. Tong, L. Yang, Z. Wang, et al., Arrays of horizontal carbon nanotubes of controlled chirality grown using designed catalysts, Nature 543 (2017) 234-238. [33] L. Kang, S. Deng, S. Zhang, Q. Li, J. Zhang, Selective growth of subnanometer diameter single-walled carbon nanotube arrays in hydrogen-free cvd, J. Am. Chem. Soc. 138 (2016) 12723-12726. [34] Q. Zhao, Z. Xu, Y. Hu, F. Ding, J. Zhang, Chemical vapor deposition synthesis of near-zigzag single-walled carbon nanotubes with stable tube-catalyst interface, Sci. Adv. 2 (2016) e1501729. [35] F. Yang, X. Wang, D. Zhang, J. Yang, D. Luo, Z. Xu, et al., Chirality-specific growth of single-walled carbon nanotubes on solid alloy catalysts, Nature 510 (2014) 522-524. [36] F. Yang, X. Wang, D. Zhang, K. Qi, J. Yang, Z. Xu, et al., Growing zigzag (16, 0) carbon nanotubes with structure-defined catalysts, J. Am. Chem. Soc. 137 (2015) 8688-8691. [37] F. Yang, X. Wang, J. Si, X. Zhao, K. Qi, C. Jin, et al., Water-assisted preparation of high-purity semiconducting (14, 4) carbon nanotubes, ACS Nano 11 (2016) 186-193. [38] H. An, A. Kumamoto, H. Takezaki, S. Ohyama, Y. Qian, T. Inoue, et al., Chirality specific and 22
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
spatially uniform synthesis of single-walled carbon nanotubes from a sputtered co–w bimetallic catalyst, Nanoscale 8 (2016) 14523-14529. [39] R.M. Sundaram, K.K. Koziol, A.H. Windle, Continuous direct spinning of fibers of single‐ walled carbon nanotubes with metallic chirality, Adv. Mater. 23 (2011) 5064-5068. [40] Z. Zhu, H. Jiang, T. Susi, A.G. Nasibulin, E.I. Kauppinen, The use of nh3 to promote the production of large-diameter single-walled carbon nanotubes with a narrow (n,m) distribution, J. Am. Chem. Soc. 133 (2011) 1224-1227. [41] W.-H. Chiang, R. Mohan Sankaran, Linking catalyst composition to chirality distributions of as-grown single-walled carbon nanotubes by tuning nixfe1−x nanoparticles, Nat. Mater. 8 (2009) 882-886. [42] S. Zhang, Y. Hu, J. Wu, D. Liu, L. Kang, Q. Zhao, et al., Selective scission of c-o and c-c bonds in ethanol using bimetal catalysts for the preferential growth of semiconducting swnt arrays, J. Am. Chem. Soc. 137 (2015) 1012-1015. [43] M. He, Y. Magnin, H. Amara, H. Jiang, H. Cui, Fr, et al., Linking growth mode to lengths of single-walled carbon nanotubes, Carbon 113 (2017) 231-236. [44] Q. Wang, M.F. Ng, S.W. Yang, Y. Yang, Y. Chen, The mechanism of single-walled carbon nanotube growth and chirality selection induced by carbon atom and dimer addition, ACS Nano 4 (2010) 939-946. [45] W.E. Alvarez, B. Kitiyanan, A. Borgna, D.E. Resasco, Synergism of Co and Mo in the catalytic production of single-wall carbon nanotubes by decomposition of CO, Carbon 39 (2001) 547-558. [46] M. He, H. Jiang, I. Kauppi, P.V. Fedotov, A.I. Chernov, E.D. Obraztsova, et al., Insights into chirality distributions of single-walled carbon nanotubes grown on different coxmg1−xo solid solutions, J. Mater. Chem. A 2 (2014) 5883-5889. [47] Q.W. Li, H. Yan, Y. Cheng, J. Zhang, Z.F. Liu, A scalable CVD synthesis of high-purity single-walled carbon nanotubes with porous mgo as support material, J. Mater. Chem. 12 (2002) 1179-1183. [48] S. Ardizzone, C.L. Bianchi, M. Fadoni, B. Vercelli, Magnesium salts and oxide: An XPS overview, Appl. Surf. Sci. 119 (1997) 253-259. [49] R. Podila, J. Reppert, N. Li, C.Z. Loebick, S.J. Stuart, L.D. Pfefferle, et al., Curvature-induced symmetry lowering and anomalous dispersion of the g-band in carbon nanotubes, Carbon 49 (2011) 869-876. [50] S. Sao-Joao, S. Giorgio, C. Mottet, J. Goniakowski, C.R. Henry, Interface structure of Ni nanoparticles on MgO (1 0 0): A combined HRTEM and molecular dynamic study, Surf. Sci. 600 (2006) L86-L90. [51] M. Jafarbegloo, A. Tarlani, A.W. Mesbah, Nio–mgo solid solution prepared by sol–gel method as precursor for Ni/MgO methane dry reforming catalyst: Effect of calcination temperature on catalytic performance, Catal. Lett. 146 (2016) 1-11. [52] F. Arena, F. Frusteri, A. Parmaliana, L. Plyasova, A. Shmakov, Effect of calcination on the structure of ni/mgo catalyst: An x-ray diffraction study, J. Chem. Soc., Faraday Trans. 92 (1996) 469-471. [53] F. Arena, A. Licciardello, A. Parmaliana, The role of Ni2+ diffusion on the reducibility of nio/mgo system: A combined trp-xps study, Catal. Lett. 6 (1990) 139-149. 23
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
[54] H.W. Nesbitt, D. Legrand, G.M. Bancroft, Interpretation of ni2p xps spectra of ni conductors and ni insulators, Phys. Chem. Miner. 27 (2000) 357-366. [55] A.P. Grosvenor, M.C. Biesinger, R.S.C. Smart, N.S. Mcintyre, New interpretations of xps spectra of nickel metal and oxides, Surf. Sci. 600 (2006) 1771-1779.
24
S0008-6223(17)31215-0
DOI:
10.1016/j.carbon.2017.11.093
Reference:
CARBON 12623
To appear in:
Carbon
Received Date: 8 September 2017 Revised Date:
7 November 2017
Accepted Date: 29 November 2017
Please cite this article as: M. He, X. Wang, L. Zhang, Q. Wu, X. Song, A.I. Chernov, P.V. Fedotov, E.D. 2+ Obraztsova, J. Sainio, H. Jiang, H. Cui, F. Ding, E. Kauppinen, Anchoring effect of Ni in stabilizing reduced metallic particles for growing single-walled carbon nanotubes, Carbon (2017), doi: 10.1016/ j.carbon.2017.11.093. 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.
ACCEPTED MANUSCRIPT Graphical Abstract A Ni-incorporated MgO catalyst was developed for predominant synthesis of (6, 5) single walled carbon nanotubes. Density functional theory-based calculations revealed that the unreduced subsurface Ni stabilized reduced Ni atoms on the surface,
AC C
EP
TE D
M AN U
SC
RI PT
facilitating the growth of carbon nanotubes with a narrow chirality distribution.
ACCEPTED MANUSCRIPT Anchoring Effect of Ni2+ in Stabilizing Reduced Metallic Particles for Growing Single-Walled Carbon Nanotubes Maoshuai He1,2,3*, Xiao Wang2, Lili Zhang4, Qianru Wu1, Xiaojie Song1, Alexander I. Chernov5,6,
RI PT
Pavel V. Fedotov5, Elena D. Obraztsova5, Jani Sainio7, Hua Jiang7, Hongzhi Cui1, Feng Ding2*, Esko
1
SC
Kauppinen7
School of Materials Science and Engineering, Shandong University of Science and Technology,
2
Center for Multidimensional Carbon Materials, Institute for Basic Science, UNIST-gil 50,
Ulju-gun, Ulsan 44919, Korea 3
M AN U
266590 Qingdao, China
Key Laboratory of Eco-Chemical Engineering, Ministry of Education, College of Chemistry and
4
TE D
Molecular Engineering, Qingdao University of Science and Technology, Qingdao 266042, China Shenyang National Laboratory for Materials Science, Institute of Metal Research, Chinese
EP
Academy of Sciences, 72 Wenhua Road, Shenyang 110016, China A.M. Prokhorov General Physics Institute RAS, 38 Vavilov Street, 119991 Moscow, Russia
6
National Research Nuclear University MEPhI (Moscow Engineering Physics Institute), Kashirskoe
AC C
5
hwy. 31, 115409, Moscow, Russia 7
Department of Applied Physics, Aalto University School of Science, P.O. Box 15100, FI-00076
Aalto, Finland
*Corresponding author. Tel: +86 532 8605 7925. E-mail address: [email protected] (M.S. He); Tel: +82-52 217 5703. E-mail address: [email protected] (F. Ding)
ACCEPTED MANUSCRIPT Abstract The suitability of the Ni-MgO catalyst as a catalyst in chiral-selective growth of single-walled carbon nanotubes (SWNTs) by chemical vapor deposition has been assessed. It reveals that catalyst
RI PT
calcination temperature plays an important role in affecting the catalyst performances. Using CO as the carbon precursor and a chemical vapor deposition reaction temperature of 600 oC, Ni-MgO
SC
pre-calcined at 600 oC demonstrates the best performances in catalyzing the growth of SWNTs with predominant (6, 5) species. Systematic characterizations on catalysts calcinated at different
M AN U
temperatures indicate that Ni2+ ions diffuse towards the interior of MgO matrix upon annealing. DFT-based calculations reveal that the binding energy between Ni2+ and adjacent Ni(0) is larger than that between Mg2+ and Ni (0), while Ni2+ situated deep inside MgO has weak interactions with surface Ni atoms. This work highlights the importance of subsurface Ni2+ in anchoring reduced
TE D
surface Ni atom, which inhibits the aggregation of Ni particles and therefore, facilitates the growth of SWNTs with a narrow chirality distribution.
EP
Keywords: single-walled carbon nanotube, catalyst, chirality selective, reduction temperature, solid
AC C
solution, anchoring effect 1. Introduction
Because of their unique structures, single-walled carbon nanotubes (SWNTs) exhibit extraordinary electrical and optical properties, which are sensitively dependent on their chirality indices (n, m) [1]. Although SWNTs have demonstrated great potentials in nanoelectronics and optoelectronics [2], their structural inhomogeneity greatly hinders their practical applications [3, 4]. To lift the hurdle for incorporating SWNTs into nanoelectronics, it is necessary to achieve SWNTs 2
ACCEPTED MANUSCRIPT with defined properties, diameters, and even chiralities. Several post-synthesis approaches, such as density gradient ultracentrifugation [5], DNA recognition [6] and gel chromatography [7] have been devoted to separate SWNTs with single chirality in the last decade. These approaches usually suffer
RI PT
from high cost and contamination of SWNTs [8, 9]. Besides, a dispersion process, which could destroy the integrities and pristine properties of SWNTs, is generally required prior to the separation
SC
procedure.
Different from separation techniques, chemical vapor deposition (CVD) synthesis of SWNTs
M AN U
with controlled (or specific or selective) chirality is an economic way for achieving SWNTs with defined structures and properties. Most catalyst systems developed so far for chirality selective synthesis of SWNTs are listed in Table 1. All the parameters related to CVD process, such as reaction temperature, carbon source, catalyst and support, are demonstrated to affect SWNT chirality
TE D
distribution. Generally, ethanol and CO are mostly adopted as the carbon source for selective growth. As part of the produced molecular fragments from ethanol is similar to that of CO [42], it can be
EP
concluded that the use of hydrogen-free CO as the carbon source favors the growth of SWNTs with a
AC C
narrow diameter distribution. Recently, we demonstrated that CO facilitates the nucleation of SWNTs on Fe particles by perpendicular mode, i.e. SWNTs possess diameters smaller than the catalyst particles, accounting for the chirality-selective synthesis of SWNTs [43]. Besides, low temperature is usually adopted in a number of experiments for chirality-selective growth of SWNTs. The roles of low temperature could be divided into two aspects. One is that the mobility of small-sized particles is low at low reaction temperatures, facilitating the growth of small diameter SWNTs with low formation energy [44]. The other is that nanoparticles preserve their solid state at 3
ACCEPTED MANUSCRIPT low reaction temperatures. Consequently, the solid nanoparticles could serve as templates for synthesizing SWNTs with structures matching the catalyst crystal symmetry [32, 35]. Table 1. A summary of catalysts applied for chirality selective growth of SWNTs. Cat., T, P, Abs.,
RI PT
PL, ED are the abbreviation for the catalyst, temperature, pressure, absorption, photoluminescence and electron diffraction, respectively. The indicated pressure value in the table is the total pressure of the CVD reaction. Active
T (oC)
P(atm)
particle
Co
750
1-10
Fe
600
1
CoMo/MgO
Co
750
1
FeCo/Zeolite
FeCo
650
0.013
FeCu/MgO
Fe
600
1
Co-MCM-41 CoxMg1-xO
Co
(n,m)
CO
600
5.8
550
SWNT selectivity (%)
(6, 5)
Ref.
Raman
Abs
PL
ED
40
42
51
25
[1015]
CH4
(6, 5)
/
/
54
/
[14]
CH4
(7, 5)
/
26
/
/
[15]
EtOH
(6, 5)
/
/
25
/
[16]
CO
(6, 5)
/
/
60
40
[17,
CO
18] (6, 5)
/
/
39.7
/
[19, 20]
5.8
CO
(6, 5)
/
/
/
/
[21]
Co
500
1
CO
(6, 5)
/
/
67
55
[22]
Co
600
1
CO
(6, 5)
/
/
39
/
[23]
AC C
Co/SiO2
Co
EP
CoMn/MCM-41
precursor
TE D
FeRu/SiO2
Enriched
M AN U
CoMo/SiO2
Carbon
SC
Cat.
Ni/SiO2
Ni
500
1
CO
(6, 5)
/
/
41
/
[24]
Fe/MgO
Fe
700
1
CO
(6, 5)
/
/
32
/
[25]
FeMn/MgO
Fe
600
1
CO
(6, 5)
/
/
46
/
[25]
CoMo/SiO2
Co
800
17.8
CO
(6, 5)
/
/
40
/
[26]
Co/TUD-1
Co
800
5.9
CO
(9, 8)
/
/
59.1
/
[27]
CoMo/MgO
Co
800
5.9
CO
(7, 5)
/
/
39.9
/
[28]
CoPt/SiO2
CoPt
800
1
EtOH
(6, 5)
/
/
30
/
[29]
Au/Al2O3-SiO2
Au
700
0.00049
CH4/H2
(6, 5)
/
/
/
/
[30]
4
ACCEPTED MANUSCRIPT Co
600
10-6
C2H2
(6, 5)
26.8
/
/
/
[31]
Mo-sapphire
Mo2C
850
1
EtOH/H2
(12, 6)
90
/
/
/
[32]
W-sapphire
WC
850
1
EtOH/H2
(8, 4)
80
/
/
/
[32]
Mo-quartz
Mo2C
830
1
EtOH
(8, 4)
46.6
/
/
/
[33]
Fe-quartz
Fe
820-8
1
EtOH
(15, 2)
21
/
/
/
[34]
RI PT
Co-SiO2
80 W6Co7
1030
1
EtOH/H2
(12, 6)
94.4
/
/
/
[35]
CoW-SiO2
W6Co7
1050
1
EtOH/H2
(16, 0)
79.2
/
/
/
[36]
CoW-SiO2
W6Co7
1050
1
EtOH/H2O
(14, 4)
97.4
/
/
/
[37]
CoW-SiO2
W6Co6
750
0.0128
EtOH
(12, 6)
1200
1
70
/
/
/
[38]
/
M AN U
/
/
/
[39]
C Ferrocene
Fe
SC
CoW-SiO2
CH4
(9, 9),
(12, 12)
Fe
880
1
FeNi
NixFe1-x
600
1
CO/NH3
(13, 12)
/
/
/
13.7
[40]
C2H2/H2
(8, 4)
/
/
39.1
/
[41]
TE D
Ferrocene
Besides carbon feedstocks and temperature, the interplay between metal and support, which
EP
regulates the catalyst performances, can affect SWNT growth results. To facilitate the synthesis of SWNTs at low reaction temperature, it is necessary to develop catalyst systems affording metal
AC C
particles which are well stabilized at reaction conditions. Therefore, the easy reduction and the presence of anchors are necessary for forming small particles to grow SWNTs. In some bimetallic catalysts, the addition of a second metal, such as Mo [45], Ru [14], Cu [17] or Mn [25], not only facilitates the reduction of active component, but also prevents metal coalescence. In addition to bimetallic catalysts, some monometallic catalysts, like SiO2 supported Co [23] and Ni [24] catalysts, have also been demonstrated to efficiently synthesize SWNTs at low temperatures. Under reaction
5
ACCEPTED MANUSCRIPT conditions, it is postulated that part of metal ions could be reduced to generate metal clusters, which can be stabilized by the unreduced counterparts, like Co2+ and Ni2+ in the above two catalysts. Compared with SiO2-supported catalysts, MgO-supported catalysts show great promises in scaling
RI PT
up because of its low-cost and easy removal after CVD. MgO supported Co [22, 46] and Fe [25] catalysts have already been reported for chirality-selective synthesis of SWNTs. Nevertheless,
SC
monometallic Ni catalyst supported by MgO is rarely applied for controlling the chirality of SWNTs. More importantly, the anchoring mechanisms exerted by metal ions have not been well addressed
M AN U
yet.
To fill this gap, we herein develop a series of MgO supported Ni catalysts prepared by impregnation and subsequent calcination. Catalyst performances are evaluated by synthesizing carbon nanotubes, the diameters of which are preliminarily characterized by Raman spectroscopy.
TE D
Photoluminescence (PL) is further adopted to determine the chirality distribution of SWNTs with small diameters. By combining catalyst characterizations and DFT calculations, the anchoring effects
EP
of Ni2+ in stabilizing reduced metallic Ni particles will be elucidated.
AC C
2. Materials and methods
2.1 Preparation of Ni-MgO catalyst Porous MgO was obtained by decomposition of magnesium carbonate hydroxide hydrate (Aldrich, 99%) at 400 oC for 1 h [17]. The BET area of MgO prepared by such a technique is 50 m2/g, as previously determined by Li et al. [47]. MgO supported Ni catalysts were prepared by impregnating 4.0 g MgO with an aqueous solution containing 1.4 g Ni(NO3)26H2O. The resulting mixtures were dried overnight at a temperature of 80 oC. Several aliquots of the dried catalysts were 6
ACCEPTED MANUSCRIPT then calcined in air for 20 h at 450 oC, 600 oC and 800 oC, respectively. The catalysts are denoted as Ni-MgO (450), Ni-MgO (600) and Ni-MgO (800) respectively. 2.2 Growth of carbon nanotubes by CVD
RI PT
CVD growth of carbon nanotubes was performed on a CCR1000 (Linkam) micro-reactor. About 20 mg catalyst powder was loaded into sample holder and heated to 600 oC under a 50 sccm Ar flow.
SC
After being stabilized at 600 oC, CO with a flow rate of 50 sccm was introduced to replace Ar and the reaction was run for 15 min. CO was finally switched off and the system was cooled down under
M AN U
the protection of Ar. All the process is performed at ambient pressure. 2.3 Characterizations of catalyst particles and carbon nanotubes
XRD measurements were performed with a Panalytical diffractometer using Cu Kα (λ= 0.15406 nm) radiation to ascertain the crystal structure of catalyst. Scans were performed over the 2θ
TE D
range from 10o to 90°. The surface concentrations of Ni in the catalysts were determined by X-ray photoelectron spectroscopy (XPS), which was carried out on a Surface Science Instruments SSX-100
EP
ESCA spectrometer using monocromated Al Kα radiation. Peak positions were calibrated using the
AC C
O 1s photoelectron peak of MgO at 530.5 eV as an internal reference [48]. Scanning electron microscopy (SEM) was carried out with a Nova Nano SEM450 to characterize the catalysts. To clarify
the
morphologies
and
compositions
of
as-prepared
catalysts,
scanning
transmission/transmission electron microscopy (STEM/TEM) observations and energy-dispersive X-ray (EDX) mapping analysis were conducted using an FEI Tecnai F20 TEM operated at 200 kV. Raman spectrometer (Jobin Yvon LabRam 300) with an excitation laser wavelength of 632.8 nm (1.96 eV) was applied to characterize the as-produced carbon nanotubes. A JEOL JSM-7500F 7
ACCEPTED MANUSCRIPT SEM was performed to characterize the morphology of the carbon nanotubes. The products were purified by hydrochloric acid with subsequent water wash. After drying in oven, the samples were sonicated in acetone and collected onto porous carbon TEM grid for structural characterizations
RI PT
(JEOL JEM-2200FS 2×CS corrected TEM). To estimate the chirality distributions, the purified SWNTs were dispersed into an aqueous solution of sodium cholate hydrate (1 wt%) by tip
SC
sonication. The dispersion was subjected to PL measurements after centrifugation at 100 000 g for 40 min (Ultracentrifuge Optima Max-E, Beckman-Coulter, MLA-80 rotor). The PL characterizations
2.4 Computational methods and details
M AN U
were carried out with a Horiba Jobin-Yvon NanoLog system.
All the first principles calculations were performed by Vienna Ab initio Simulation Package based on density functional theory. Perdew-Burke-Ernzerhof (PBE) functional within the frame of
TE D
generalized gradient approximation (GGA) was chosen to describe the electronic exchange and correlation. The interaction between the valence electrons and kernel was depicted by the projected
EP
augmented wave (PAW) method with the cutoff energy set as 400 eV. With the aim to describe the
AC C
interlayer interaction more accurate, the Grimme DFT-D2 method is included. A three-layer slab of MgO with a supercell size of 11.78×11.78×20.00 Å3 was chosen as the theoretical model. The bottom layer is fixed as the bulk and the top two layers can be relaxed. The vacuum spacing is larger than 10 Å and is enough to void avert the interference of adjacent images. As the Ni2+ ion can substitute Mg2+ ion in the MgO martix, one, two and four Ni2+ ions are chosen to replace the equivalent number of Mg2+ ions on the surface or subsurface of the slab, respectively. To investigate the anchoring effect of Ni2+ on the Ni atom, one Ni atom is placed above the slab. When optimizing 8
ACCEPTED MANUSCRIPT the geometrical structure, 4×4×1 k-points were sampled uniformly in the Brillouin zone using Monkhorst-Pack sampling method. A recommended conjugated gradient algorithm was adopted to ensure a force convergence less than 0.01 eV/A on each atom. After geometrical relaxation, the
ܧ ൌ ܧ௦ ܧே െ ܧேି௦ 3. Results and discussion
RI PT
binding energy between the Ni atom and slab is calculated for further analysis following the formula: (1)
SC
Figure 1a presents Raman spectra of carbon nanotubes grown on Ni-MgO catalysts at 600 oC
M AN U
using CO as the carbon source. In the Raman spectrum of the product grown on Ni-MgO (450), radial breathing modes (RBMs) in the low frequency range are absent, suggesting the lack of SWNTs in the product. Instead, only large diameter multi-walled carbon nanotubes were grown on the Ni-MgO (450) catalyst. As tube diameter is correlated with catalyst size, the active catalyst particles
AC C
EP
TE D
formed on Ni-MgO (450) during CVD should possess large diameters.
Figure 1. (a) Raman spectra (excitation wavelength: 632.8 nm) of carbon nanotubes grown on different Ni-MgO catalysts at 600 oC using CO as the carbon source. (b) Contour plots of normalized PL emission intensities under various excitation energies for SWNTs grown on Ni-MgO (600). (c) Normalized PL emission intensities of identified SWNTs deduced from Figure b. Different from Ni-MgO (450), Ni-MgO catalysts annealed at higher temperatures afford the synthesis of SWNTs under identical CVD conditions, as indicated by the presence of RBMs in the 9
ACCEPTED MANUSCRIPT Raman spectra of carbon nanotubes produced on Ni-MgO (600) and Ni-MgO (800). The RBMs excited by 632.8 nm laser consist of two major peaks: one is centered at ∼190 cm-1, the other is located at ∼292 cm-1. For SWNTs grown on Ni-MgO (800), the intensity of the peak centered at 190
RI PT
cm-1 is much stronger than the one at 292 cm-1, indicating most SWNTs have diameters larger than 1.0 nm. In addition, the high intensity of D mode and the relatively low intensity of G mode indicate
SC
the coexistence of much carbonaceous impurities. In contrast, Raman spectrum of SWNTs grown on Ni-MgO (600) exhibits a relatively low intensity of the D mode and a more intensive RBM peak at
M AN U
292 cm-1, implying SWNTs grown on Ni-MgO (600) have relatively small average diameters. The production of small-diameter SWNTs on Ni-MgO (600) is also verified by the appearance of intermediate frequency modes at
750 cm-1 in the Raman spectrum. The intensity of intermediate
frequency mode has been proven to increase with increasing SWNT curvature [49]. Therefore, such
TE D
characteristic intermediate frequency modes could be readily observed in sub-nanometer SWNTs [20] like what are achieved here. Overall, Ni-MgO (600) exhibits the best performances in catalyzing
EP
sub-nanometer SWNTs.
AC C
To estimate the chirality distribution of SWNTs grown on Ni-MgO (600), SWNTs dispersed in sodium cholate solution was subjected to PL characterizations. Figure 1b depicts the PL map of the SWNT dispersion where (6, 5) SWNT displays the most intense emission intensity. Besides (6, 5) SWNTs, other semiconducting SWNTs with diameters smaller than 0.85 nm, including (8, 3), (7, 5) and (8, 4) SWNTs, are the major species identified in the product. Figure 1c presents the normalized PL emission intensity of different SWNT species. Compared with SWNTs grown on SiO2 supported Ni catalysts under identical conditions [24], the average diameter of SWNTs grown on Ni-MgO 10
ACCEPTED MANUSCRIPT (600) is a bit smaller. Assuming the abundances of SWNTs are proportional to their PL emission intensities, the percentage of (6, 5) SWNT in the product is about 44%. To further confirm the morphology and structure of SWNTs, SEM and TEM characterizations
RI PT
were performed and some typical images are demonstrated in Figure 2a & 2b. Detailed TEM analysis demonstrates that the SWNTs have relatively small diameters centered at ~0.8 nm (Figure
SC
3c). The growth of smaller-diameter SWNTs on Ni-MgO (600) can be attributed to the specific metal-support interaction, which could affect the morphology of metal nanoparticles formed during
M AN U
CVD growth process. In addition, an epitaxial relationship between reduced Ni particles and MgO support [50], like in the case of Co-MgO catalyst [22] prepared by the same approach, cannot be
EP
TE D
excluded.
Figure 2. SEM image (a) and TEM image (b) of carbon nanotubes grown on Ni-MgO (600) at 600 o
images.
AC C
C using CO as the carbon source. (c) Diameter distribution of SWNTs determined from TEM
Aiming to clarify why the calcination temperature affects the catalyst performances for synthesizing SWNTs, XRD and XPS characterizations were employed on the three Ni-MgO catalysts. Figure 3a depicts the XRD patterns of Ni-MgO catalysts annealed at three different temperatures over the 2θ range from 10o to 90°. As both Ni2+ and Mg2+ possess same face-centered cubic crystal structures and similar ionic radii [51], calcination could promote the migration of Ni2+ 11
ACCEPTED MANUSCRIPT ions towards MgO interior and the formation of NiO-MgO solid solution. Such progressive diffusion of Ni ions across the MgO lattice can be well reflected by XRD features [52]. On the one hand, the lack of NiO peak signals the formation of NiO-MgO solid solution or a highly dispersed NiO form
RI PT
[52]. On the other hand, the deduced lattice parameter could reflect the forms of NiO dissolved into MgO matrix. Figure 3b presents the magnified XRD spectra in the range 2θ = 73-81o of the three
SC
Ni-MgO catalysts. All catalysts show two well-resolved XRD peaks ascribable to MgO (311) reflection and MgO (222) reflection, respectively. With the increase of the calcination temperature,
M AN U
the characteristic peaks shift to higher angles and their corresponding FWHMs decrease. On the basis of the 2θ value of the XRD peak, MgO lattice parameter (a) can be calculated from the equation:
a ൌ ඥ(ℎଶ ݇ ଶ ݈ ଶ )
ߣ 2 sin ߠ
TE D
Where h, k, l are Miller indices of planes in MgO cubic crystal and λ is the wavelength of Cu Kα (0.15406 nm). The reflection peaks for Ni-MgO (450) and Ni-MgO (600) are almost identical,
EP
suggesting the similar bulk crystal structure of MgO in both catalysts. Based on the 2θ value of 78.4o for (222) planes, the lattice parameter of MgO is determined to be 0.422 nm. Increasing the
AC C
calcination temperature of Ni-MgO to 800 oC shifts the lattice reflection to a higher 2θ value (78.6o) and therefore a smaller lattice parameter (0.421 nm). Ever-increasing incorporation of Ni2+ ions into the MgO structure accounts for the progressive 2θ upshift and the decrease in FWHM (Figure 3b) upon calcination [52].
12
SC
RI PT
ACCEPTED MANUSCRIPT
M AN U
Figure 3. (a) XRD patterns of Ni-MgO catalysts calcined at different temperatures. (b) For comparison, the overlapped spectra in the 73o-81o range are presented. (c) XPS spectra of Ni 2p in Ni-MgO catalyst series. (d) Atomic ratios of Ni to Mg in different Ni-MgO catalysts determined by XPS.
The effects of air calcination on catalyst reducibility and performances could also be reflected
TE D
by the changes in both Ni surface compositions and binding energies of XPS spectrum [53]. Figure 3c compares the Ni 2p photoelectron peaks and atomic ratios of Ni/Mg in the three catalysts. In all
EP
the spectra, the binding energy of the Ni 2p3/2 core level is centered around 856.1 eV, which is much
AC C
higher than that of metallic Ni (852.8 eV), suggesting that the Ni exists as oxide state. The binding energy of the main peak reflects the nature of the ligands to which Ni is bonded because charge transfer involves transfer of an electron from an L-band to the 3d9 band [54]. Besides, the Ni 2p3/2 peaks show a symmetric profile, which is different from the multiplet-split characteristic of asymmetric peak of “free” NiO, suggesting that the Ni2+ could dissolves inside the MgO matrix. In addition, a satellite peak with a binding energy of ~862 eV is clearly seen, which is attributed to the unscreened cd8 final-state configurations [55]. Although the spectra are quite similar, Figure 3d 13
ACCEPTED MANUSCRIPT indicates that, the Ni/Mg atomic ratio detected by XPS progressively decreases, suggesting the
SC
RI PT
depletion of Ni in the sampled thickness upon calcination.
M AN U
Figure 4. (a) TEM image of Ni-MgO (600) catalyst. (b) STEM image from the same catalyst particle and EDX mapping from the crossed area indicated by the white square. TEM and SEM were also applied to characterize the morphology and element distribution of the catalysts. Figure 4a depicts a TEM image of Ni-MgO (600) catalyst. There are no clear isolated
TE D
Ni-containing nanoparticles on the MgO surface. STEM image and EDX mapping were further applied to analyze the catalyst composition (Figure 4b). It is clearly shown that the elements of Mg,
EP
O and Ni are almost overlapped, and Ni is well dispersed in the MgO support. The TEM
AC C
characterization results of Ni-MgO (450) catalyst and Ni-MgO (800) catalyst are presented in Supporting Information Figure S2&S3, respectively. In both cases, Ni elements are also well dispersed in the catalysts. Such characterization results are in agreement with SEM results, which are presented in Supporting Information Figure S4, S5 & S6. In addition, the crystalline degree of the catalysts increases with the increase of calcination temperature, as revealed by high resolution SEM (Supporting Information Figure S7). The observations also agree with the XRD characterization results. However, as the lattice constants of MgO and NiO are similar, it is difficult to distinguish 14
ACCEPTED MANUSCRIPT isolated NiO nanoparticles possibly located on MgO surface by TEM. As clarified by Arena et al.[52, 53], there are four forms of Ni2+ ions in Ni-MgO catalysts: (i) “unreacted” NiO on MgO surface, which is uninfluenced by the MgO support and could be readily
RI PT
reduced at temperatures below 400 oC; (ii) Ni2+ ions in the outermost layer of the MgO structure, having square-pyramidal coordination and can be reduced at a temperature of around 600 oC; (iii) Ni2+ ions located in the sub-surface layers of the MgO lattice, which is reducible at a temperature of
M AN U
and is reducible at temperatures higher than 1000 oC.
SC
about 750 oC; (iv) Ni2+ ions dissolved in the MgO matrix, forming a “bulk” NiO-MgO solid solution
In order to evaluate the interactions between reduced Ni atom and cations in the catalyst during CVD reaction, DFT-based calculations are performed to confirm our hypothesis. In our theoretical mode, Ni atom situated on MgO matrix where equivalent Ni2+ ions substitute for one, two and four
TE D
Mg2+ ions on the first or second layer were investigated. Without Ni2+ substitution, the Ni atom favors bonding with the O atom on the top site (shown in Figure 5a, lateral view is presented in
EP
Supporting Information Figure S1) with a calculated binding energy of 1.97 eV (Figure 5b). If some Mg2+ ions on the first layer are substituted by Ni2+, the most stable configuration for the adhered Ni
AC C
atom is bridge site between the Ni2+ and O2- instead of the top site. In addition, the adjacent Ni2+ ions can also contribute to a stronger binding to the adhered Ni atom. As shown in Figure 5b, when one Ni2+ replaces one Mg2+ on the surface, the binding energy increases from 1.97 eV to 2.41 eV compared with the case of no Ni2+ substitution. The binding energy will increase as the number of substituted Mg2+ atoms increases and ultimately reaches 3.03 eV when four Mg2+ ions are substituted (Figure 5c-e), which is 1.06 eV higher than that in the unsubstituted case. Consequently, the 15
ACCEPTED MANUSCRIPT substituted Ni2+ on the surface can act as an anchor site to stabilize the absorbed Ni atoms and impede their subsequent migration, which is propitious to nucleate small catalyst particles. When the Mg2+ ions on the second layer of MgO are substituted by the Ni2+ (Figure 5f-h), unlike the
RI PT
substitution in the first layer, the most stable site for the absorbed Ni atom is the top of the O atom, which is the same to the unsubstituted case. Besides, there is only nearly 0.10 eV increase of the binding energy and no obvious dependence on the number of substituted Mg2+ ions, which implies
SC
that the Ni2+ substitution on the second layer of MgO contributes little to stabilize the absorbed Ni
M AN U
atoms.
Under our CVD reaction conditions, “unreacted” NiO can be readily reduced and form large-diameter metallic Ni particles, responsible for the synthesis of multi-walled carbon nanotubes. Ni2+ ions in the outmost layer of MgO could also be reduced at conditions adopted in our
TE D
experiments. While subsurface Ni2+ ions and “bulk” Ni2+ ions cannot be reduced. Taking our calculation results into account, it is suggested that the subsurface Ni2+ could act as anchors to
EP
constrain the mobility of reduced surface Ni. While “bulk” Ni2+ ions, which is far from the surface,
AC C
could not stabilize the Ni clusters formed on surface.
16
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
EP
Figure 5. (a) Most stable configuration of Ni atom absorbed on MgO top layer without Ni substitution. (b) Binding energy of absorbed Ni atom on MgO. Most stable configurations of one Ni
AC C
atom absorbed on MgO top surface with one, two, four Mg2+ ions on the first layer (c-e) or second layer (f-h) substituted by equivalent Ni2+ ions, respectively. The O2-, Mg2+, Ni2+ ions are colored in red, green and blue, respectively. The absorbed Ni (0) is displayed smaller than those in the substrate for distinctness.
17
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Figure 6. Schematic illustration of catalyst evolution upon the introduction of CO at 600 oC for Ni-MgO catalysts calcined at (a) 450 oC; (b) 600 oC; (c) 800 oC. Because of sufficient subsurface Ni2+ ions in Ni-MgO (600), which can serve as anchors for constraining the mobility of reduced Ni
TE D
clusters, small diameter SWNTs with a narrow chirality distribution are achieved. ～ Therefore, the coexistence of Ni2+ ions in the outmost layer and in the subsurface of MgO is of
EP
great importance for low temperature growth of SWNTs. As for Ni-MgO (450), there exist only “unreacted” and surface NiO. Both can be readily reduced to form large particles for the growth of
AC C
large diameter multi-walled carbon nanotubes (Figure 6a).～With the increase of the calcination temperature, more Ni2+ ions migrate towards the core of MgO support, accounting for the depletion of surface Ni detected by the XPS (Figure 3d). As the XRD pattern of Ni-MgO (600) does not exhibit obvious peak shift compared with that of Ni-MgO (450) (Figure 3b), it is suggested that the depletion of surface Ni mainly contribute to the generation of subsurface Ni2+ instead of diffusion inside bulk MgO matrix. Introducing CO at CVD reaction temperature favors the formation of small Ni clusters from the reduction of outmost layer Ni2+ ions, which could be further stabilized by 18
ACCEPTED MANUSCRIPT subsurface Ni2+ ions. The stabilized metallic Ni particles are responsible for the synthesis of SWNTs with small diameters. Further considering the use of CO as the carbon source, it could facilitate the dissolution of carbon inside metal particle, resulting in the nucleation of SWNTs with a
RI PT
perpendicular mode [43] and synthesis of SWNTs with a narrow chirality distribution (Figure 6b). However, further increasing the catalyst calcination temperature to 800 oC causes the diffusion
SC
of subsurface and outmost layer Ni2+ ions. Consequently, these Ni2+ ions migrate into MgO matrix and form a bulk NiO-MgO solid solution, which explains the shift of XRD peaks (Figure 3a&3b)
M AN U
and the diminishing Ni content in XPS spectrum (Figure 3d). As a result, only trace amount of residual Ni2+ ions could be reduced to form metal Ni particles, catalyzing the growth of SWNTs. It is noted that because the amount of subsurface Ni2+ ions is diminishing, their “anchoring” effect is not as efficient as in Ni-MgO (600), particles and resulting SWNTs grown on Ni-MgO (800), have
4. Conclusion
TE D
relatively large diameters (Figure 6c).～
EP
To conclude, Ni-MgO catalysts annealed at different temperatures are applied for the synthesis
AC C
of carbon nanotubes at 600 oC using CO as the carbon feedstock. Among them, Ni-MgO (600) displays the best catalyst performance not only in growing small-diameter SWNTs, but also in achieving SWNTs with a narrow chirality distribution. By means of a comparative study, the progressive change of Ni-MgO catalysts upon calcination at different temperatures has been marked. The binding energy deduced from DFT calculations revealed that reduced Ni(0) atom can be well anchored by underlying Ni2+ ions. Therefore, the coexistence of outmost layer Ni2+ ions and subsurface Ni2+ ions is the key for forming small-diameter metallic Ni particles in reducing 19
ACCEPTED MANUSCRIPT environment at low temperatures. Such reduced Ni particles anchored by subsurface Ni2+ ions and the use of CO as carbon source are responsible for the chiral-selective growth of SWNTs. This work not only extends the development of catalyst system for chiral-selective synthesis of SWNTs, but
RI PT
also helps understand the anchoring effect of metal ions in stabilizing reduced metal particles. Acknowledgements
SC
The research leading to these results has received funding from Taishan scholarship of climbing plan (No. tspd20161006) and the National High Technology Research and Development Program of
M AN U
China (863 Program, 2015AA034404). The authors would also like to acknowledge the Natural Science Foundation of Shandong Province of China (No. ZR2016EMM10) and Scientific Research Foundation of Shandong University of Science and Technology for Recruited Talents (No.
Supplementary data
TE D
2016RCJJ001). AIC acknowledges the support by the Alexander von Humboldt Foundation.
Lateral view of the optimized configurations when Ni(0) atom is placed on top of MgO with (or
EP
without) Ni2+ substitution; Detailed TEM and SEM characterization results of the Ni-MgO catalysts
AC C
calcinated at different temperatures. References
[1] M.S. Dresselhaus, G. Dresselhaus, R. Saito, A. Jorio, Raman spectroscopy of carbon nanotubes, Phys. Rep. 409 (2005) 47-99. [2] S.M. Bachilo, M.S. Strano, C. Kittrell, R.H. Hauge, R.E. Smalley, R.B. Weisman, Structure-assigned optical spectra of single-walled carbon nanotubes, Science 298 (2002) 2361-2366. [3] M.C. Hersam, Progress towards monodisperse single-walled carbon nanotubes, Nat. Nanotechnol. 3 (2008) 387-394. [4] R.H. Baughman, A.A. Zakhidov, W.A. de Heer, Carbon nanotubes--the route toward applications, Science 297 (2002) 7877-92. [5] M.S. Arnold, A.A. Green, J.F. Hulvat, S.I. Stupp, M.C. Hersam, Sorting carbon nanotubes by electronic structure using density differentiation, Nat. Nanotechnol. 1 (2006) 60-65. 20
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
[6] X. Tu, S. Manohar, A. Jagota, M. Zheng, DNA sequence motifs for structure-specific recognition and separation of carbon nanotubes, Nature 460 (2009) 250-253. [7] H. Liu, D. Nishide, T. Tanaka, H. Kataura, Large-scale single-chirality separation of single-wall carbon nanotubes by simple gel chromatography, Nat. Commun. 2 (2011) 309. [8] T. Lei, I. Pochorovski, Z. Bao, Separation of semiconducting carbon nanotubes for flexible and stretchable electronics using polymer removable method, Acc. Chem. Res. 50 (2017) 1096-1104. [9] C. Luo, D. Wan, J. Jia, D. Li, C. Pan, L. Liao, A rational design for the separation of metallic and semiconducting single-walled carbon nanotubes using a magnetic field, Nanoscale 8 (2016) 13017-13024. [10] D.E. Resasco, W.E. Alvarez, F. Pompeo, L. Balzano, J.E. Herrera, B. Kitiyanan, et al., A scalable process for production of single-walled carbon nanotubes (swnts) by catalytic disproportionation of co on a solid catalyst, J. Nanopart. Res. 4 (2002) 131-136. [11] A. Jorio, A. Santos, H. Ribeiro, C. Fantini, M. Souza, J. Vieira, et al., Quantifying carbon-nanotube species with resonance raman scattering, Phys. Rev. B 72 (2005) 075207. [12] Y. Tian, H. Jiang, I.V. Anoshkin, L.J. Kauppinen, K. Mustonen, A.G. Nasibulin, et al., A reference material of single-walled carbon nanotubes: Quantitative chirality assessment using optical absorption spectroscopy, RSC Advances 5 (2015) 102974-102980. [13] S.M. Bachilo, L. Balzano, J.E. Herrera, F. Pompeo, D.E. Resasco, R.B. Weisman, Narrow (n,m)-distribution of single-walled carbon nanotubes grown using a solid supported catalyst, J. Am. Chem. Soc. 125 (2003) 11186-11187. [14] X. Li, X. Tu, S. Zaric, K. Welsher, W.S. Seo, W. Zhao, et al., Selective synthesis combined with chemical separation of single-walled carbon nanotubes for chirality selection, J. Am. Chem. Soc 129 (2007) 15770-15771. [15] G. Lolli, L. Zhang, L. Balzano, N. Sakulchaicharoen, Y. Tan, D.E. Resasco, Tailoring (n,m) structure of single-walled carbon nanotubes by modifying reaction conditions and the nature of the support of como catalysts, J. Phys. Chem. B 110 (2006) 2108-2115. [16] Y. Miyauchi, Fluorescence spectroscopy of single-walled carbon nanotubes synthesized from alcohol, Chem. Phys. Lett. 387 (2004) 198-203. [17] M. He, A.I. Chernov, P.V. Fedotov, E.D. Obraztsova, J. Sainio, E. Rikkinen, et al., Predominant (6,5) single-walled carbon nanotube growth on a copper-promoted iron catalyst, J. Am. Chem. Soc. 132 (2010) 13994-13996. [18] M. He, B. Liu, A.I. Chernov, E.D. Obraztsova, I. Kauppi, H. Jiang, et al., Growth mechanism of single-walled carbon nanotubes on iron–copper catalyst and chirality studies by electron diffraction, Chem. Mater. 24 (2012) 1796-1801. [19] C.Z. Loebick, S. Derrouiche, N. Marinkovic, C. Wang, F. Hennrich, M.M. Kappes, et al., Effect of manganese addition to the co-mcm-41 catalyst in the selective synthesis of single wall carbon nanotubes, J. Phys. Chem. C 113 (2009) 21611-21620. [20] C. Zoican Loebick, R. Podila, J. Reppert, J. Chudow, F. Ren, G.L. Haller, et al., Selective synthesis of subnanometer diameter semiconducting single-walled carbon nanotubes, J. Am. Chem. Soc. 132 (2010) 11125-31. [21] N. Li, X. Wang, F. Ren, G.L. Haller, L.D. Pfefferle, Diameter tuning of single-walled carbon nanotubes with reaction temperature using a co monometallic catalyst, J. Phys. Chem. C 113 (2009) 21
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
10070-10078. [22] M. He, H. Jiang, B. Liu, P.V. Fedotov, A.I. Chernov, E.D. Obraztsova, et al., Chiral-selective growth of single-walled carbon nanotubes on lattice-mismatched epitaxial cobalt nanoparticles, Sci. Rep. 3 (2013) 1460. [23] M. He, A.I. Chernov, P.V. Fedotov, E.D. Obraztsova, E. Rikkinen, Z. Zhu, et al., Selective growth of swnts on partially reduced monometallic cobalt catalyst, Chem. Commun. 47 (2011) 1219-1221. [24] M. He, A.I. Chernov, E.D. Obraztsova, J. Sainio, E. Rikkinen, H. Jiang, et al., Low temperature growth of swnts on a nickel catalyst by thermal chemical vapor deposition, Nano Res. 4 (2011) 334-342. [25] M. He, P.V. Fedotov, A. Chernov, E.D. Obraztsova, H. Jiang, N. Wei, et al., Chiral-selective growth of single-walled carbon nanotubes on fe-based catalysts using co as carbon source, Carbon 108 (2016) 521-528. [26] B. Wang, L. Wei, L. Yao, L.J. Li, Y. Yang, Y. Chen, Pressure-induced single-walled carbon nanotube (n,m) selectivity on co-mo catalysts, J. Phys. Chem. C 111 (2007) 14612-14616. [27] H. Wang, B. Wang, X.Y. Quek, L. Wei, J.W. Zhao, L.J. Li, et al., Selective synthesis of (9,8) single walled carbon nanotubes on cobalt incorporated tud-1 catalysts, J. Am. Chem. Soc. 132 (2010) 16747-16749. [28] B. Wang, Y. Yang, L.-J. Li, Y. Chen, Effect of different catalyst supports on the (n,m) selective growth of single-walled carbon nanotube from co–mo catalyst, J. Mater. Sci. 44 (2009) 3285-3295. [29] B. Liu, W. Ren, S. Li, C. Liu, H.M. Cheng, High temperature selective growth of single-walled carbon nanotubes with a narrow chirality distribution from a copt bimetallic catalyst, Chem. Commun. 48 (2012) 2409-2411. [30] Z. Ghorannevis, T. Kato, T. Kaneko, R. Hatakeyama, Narrow-chirality distributed single-walled carbon nanotube growth from nonmagnetic catalyst, J. Am. Chem. Soc. 132 (2010) 9570-9572. [31] M. Fouquet, B.C. Bayer, S. Esconjauregui, R. Blume, J.H. Warner, S. Hofmann, et al., Highly chiral-selective growth of single-walled carbon nanotubes with a simple monometallic Co catalyst, Phys. Rev. B 85 (2012) 235411. [32] S. Zhang, L. Kang, X. Wang, L. Tong, L. Yang, Z. Wang, et al., Arrays of horizontal carbon nanotubes of controlled chirality grown using designed catalysts, Nature 543 (2017) 234-238. [33] L. Kang, S. Deng, S. Zhang, Q. Li, J. Zhang, Selective growth of subnanometer diameter single-walled carbon nanotube arrays in hydrogen-free cvd, J. Am. Chem. Soc. 138 (2016) 12723-12726. [34] Q. Zhao, Z. Xu, Y. Hu, F. Ding, J. Zhang, Chemical vapor deposition synthesis of near-zigzag single-walled carbon nanotubes with stable tube-catalyst interface, Sci. Adv. 2 (2016) e1501729. [35] F. Yang, X. Wang, D. Zhang, J. Yang, D. Luo, Z. Xu, et al., Chirality-specific growth of single-walled carbon nanotubes on solid alloy catalysts, Nature 510 (2014) 522-524. [36] F. Yang, X. Wang, D. Zhang, K. Qi, J. Yang, Z. Xu, et al., Growing zigzag (16, 0) carbon nanotubes with structure-defined catalysts, J. Am. Chem. Soc. 137 (2015) 8688-8691. [37] F. Yang, X. Wang, J. Si, X. Zhao, K. Qi, C. Jin, et al., Water-assisted preparation of high-purity semiconducting (14, 4) carbon nanotubes, ACS Nano 11 (2016) 186-193. [38] H. An, A. Kumamoto, H. Takezaki, S. Ohyama, Y. Qian, T. Inoue, et al., Chirality specific and 22
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
spatially uniform synthesis of single-walled carbon nanotubes from a sputtered co–w bimetallic catalyst, Nanoscale 8 (2016) 14523-14529. [39] R.M. Sundaram, K.K. Koziol, A.H. Windle, Continuous direct spinning of fibers of single‐ walled carbon nanotubes with metallic chirality, Adv. Mater. 23 (2011) 5064-5068. [40] Z. Zhu, H. Jiang, T. Susi, A.G. Nasibulin, E.I. Kauppinen, The use of nh3 to promote the production of large-diameter single-walled carbon nanotubes with a narrow (n,m) distribution, J. Am. Chem. Soc. 133 (2011) 1224-1227. [41] W.-H. Chiang, R. Mohan Sankaran, Linking catalyst composition to chirality distributions of as-grown single-walled carbon nanotubes by tuning nixfe1−x nanoparticles, Nat. Mater. 8 (2009) 882-886. [42] S. Zhang, Y. Hu, J. Wu, D. Liu, L. Kang, Q. Zhao, et al., Selective scission of c-o and c-c bonds in ethanol using bimetal catalysts for the preferential growth of semiconducting swnt arrays, J. Am. Chem. Soc. 137 (2015) 1012-1015. [43] M. He, Y. Magnin, H. Amara, H. Jiang, H. Cui, Fr, et al., Linking growth mode to lengths of single-walled carbon nanotubes, Carbon 113 (2017) 231-236. [44] Q. Wang, M.F. Ng, S.W. Yang, Y. Yang, Y. Chen, The mechanism of single-walled carbon nanotube growth and chirality selection induced by carbon atom and dimer addition, ACS Nano 4 (2010) 939-946. [45] W.E. Alvarez, B. Kitiyanan, A. Borgna, D.E. Resasco, Synergism of Co and Mo in the catalytic production of single-wall carbon nanotubes by decomposition of CO, Carbon 39 (2001) 547-558. [46] M. He, H. Jiang, I. Kauppi, P.V. Fedotov, A.I. Chernov, E.D. Obraztsova, et al., Insights into chirality distributions of single-walled carbon nanotubes grown on different coxmg1−xo solid solutions, J. Mater. Chem. A 2 (2014) 5883-5889. [47] Q.W. Li, H. Yan, Y. Cheng, J. Zhang, Z.F. Liu, A scalable CVD synthesis of high-purity single-walled carbon nanotubes with porous mgo as support material, J. Mater. Chem. 12 (2002) 1179-1183. [48] S. Ardizzone, C.L. Bianchi, M. Fadoni, B. Vercelli, Magnesium salts and oxide: An XPS overview, Appl. Surf. Sci. 119 (1997) 253-259. [49] R. Podila, J. Reppert, N. Li, C.Z. Loebick, S.J. Stuart, L.D. Pfefferle, et al., Curvature-induced symmetry lowering and anomalous dispersion of the g-band in carbon nanotubes, Carbon 49 (2011) 869-876. [50] S. Sao-Joao, S. Giorgio, C. Mottet, J. Goniakowski, C.R. Henry, Interface structure of Ni nanoparticles on MgO (1 0 0): A combined HRTEM and molecular dynamic study, Surf. Sci. 600 (2006) L86-L90. [51] M. Jafarbegloo, A. Tarlani, A.W. Mesbah, Nio–mgo solid solution prepared by sol–gel method as precursor for Ni/MgO methane dry reforming catalyst: Effect of calcination temperature on catalytic performance, Catal. Lett. 146 (2016) 1-11. [52] F. Arena, F. Frusteri, A. Parmaliana, L. Plyasova, A. Shmakov, Effect of calcination on the structure of ni/mgo catalyst: An x-ray diffraction study, J. Chem. Soc., Faraday Trans. 92 (1996) 469-471. [53] F. Arena, A. Licciardello, A. Parmaliana, The role of Ni2+ diffusion on the reducibility of nio/mgo system: A combined trp-xps study, Catal. Lett. 6 (1990) 139-149. 23
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
[54] H.W. Nesbitt, D. Legrand, G.M. Bancroft, Interpretation of ni2p xps spectra of ni conductors and ni insulators, Phys. Chem. Miner. 27 (2000) 357-366. [55] A.P. Grosvenor, M.C. Biesinger, R.S.C. Smart, N.S. Mcintyre, New interpretations of xps spectra of nickel metal and oxides, Surf. Sci. 600 (2006) 1771-1779.
24