Attachment of Iron Oxide Nanoparticles to Carbon Nanotubes and the Consequences for Catalysis

Article abstract of DOI:10.1002/cctc.201800487

The attachment of colloidal iron-oxide nanoparticles (designated Fe-NPs) to pristine and surface-oxidized carbon nanotubes (CNTs and CNT-Ox, respectively) was investigated. The loadings of Fe-NPs (size 7 nm) on the CNT and CNT-Ox supports amounted to 3.4

Full text of DOI:10.1002/cctc.201800487

Accepted Article
Title: Attachment of Iron Oxide Nanoparticles to Carbon Nanotubes
and Consequences for Catalysis
Authors: Nynke A. Krans, Ewout C. van der Feltz, Jingxui Xie, Iulian A.
Dugulan, Jovana Zečević, and Krijn P. de Jong
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemCatChem 10.1002/cctc.201800487
Link to VoR: http://dx.doi.org/10.1002/cctc.201800487
A Journal of
www.chemcatchem.org

ChemCatChem
10.1002/cctc.201800487
COMMUNICATION
Attachment of Iron Oxide Nanoparticles to Carbon Nanotubes and
Consequences for Catalysis
[
a]
[a]
[a]
[b]
[a]
[a]
Nynke A. Krans, E.C. van der Feltz, J. Xie, A.I. Dugulan, J. Zečević and K.P. de Jong*
Inorganic Chemistry and Catalysis, Debye Institute for Nanomaterials Science, Utrecht University; Universiteitsweg 99, 3584 CG
Utrecht, The Netherlands
Attachment of colloidal iron oxide nanoparticles (designated
Fe-NP) to pristine and surface-oxidized carbon nanotubes (CNT
and CNT-Ox, respectively) was investigated. The loadings of Fe-
NP (size 7 nm) on the CNT and CNT-Ox supports amounted to
and high olefin selectivity can be obtained.[16,18] Moreover,
colloidal particles can be promoted by using an organic ligands
exchange which targets the promoters on the catalytically active
particles specifically.[15,20]
3
.4 and 2.3 wt.%, respectively, where the difference was
In FTO catalysis, choosing the support material is of
importance as well, as strongly interacting supports, such as
silica and alumina, can incorporate the iron, making reduction
and carbidization to the active iron carbide phase difficult.[
Nevertheless, there are many porous carbon materials that can
attributed to weaker Van der Waals interaction between the
colloidal Fe-NP and the surface of CNT-Ox. Fischer-Tropsch to
Olefins (FTO) synthesis was performed to investigate the
impact of support functionalization on the catalyst performance.
Weak interaction between the Fe-NP and the CNT-Ox support
facilitated particle growth and led to substantial deactivation of
Fe/CNT-Ox catalysts. Addition of promoters (Na+S) to Fe/CNT
provided remarkable activity, selectivity to lower olefins and
stability, making colloidal iron nanoparticles on pristine CNT
suitable catalysts for FTO.
21–23]
be used as catalyst supports such as activated carbon, graphitic
and templated carbon.[
24–26]
Carbon nanotube (CNT) supports
are of growing interest in research due to their chemical
inertness and mesoporosity.[
27–29]
More importantly, it was found
that functionalization of carbon supports could suppress particle
growth by anchoring particles, thereby increasing stability of
impregnated catalysts. [30–34] The attachment of colloidal particles
to support materials has so far not been researched abundantly,
however it has been found that the support material can alter the
particles behavior during catalysis.[35–38]
Commonly applied catalyst synthesis methods, such as
incipient wetness impregnation and precipitation, are often used
to obtain industrial supported catalysts.[1–3] Controlling the size,
composition and shape of metal (oxide) nanoparticles, however,
is limited with these methods. Colloidal synthesis has been
successfully applied to obtain well-defined model catalysts
suitable for fundamental studies.[4–6] For example, Fischer-
Tropsch to Olefins (FTO) is one of the catalytic reactions
particularly sensitive to particle size in which colloidal model
In the present work, the impact of carbon surface oxidation
on colloidal iron oxide nanoparticle (Fe-NP) attachment to CNTs
and their subsequent promotion with Na + S was investigated. It
was found that functionalizing the CNT support by oxidation
(CNT-Ox) impeded the colloidal particle attachment and led to
growth of the nanoparticles, which decreased catalytic activity.
This shows that functionalization of support is not always
beneficial, but can have detrimental effect when colloidal
particles are used.
systems can be beneficial.[
catalysts and is becoming increasingly important as an
alternative to traditional cracking of crude oil fractions for lower
7,8]
]
FTO involves iron based
2
-C
4 2
) and aromatics production from syngas (CO/H
).[9–
olefins (C
13]
.
In addition, colloidal particles have shown to be relatively
After synthesis, the Fe-NP size and shape were investigated
by transmission electron microscopy (TEM) which showed that 7
nm spherical Fe-NP with a narrow size distribution were
obtained (Figure 1). Subsequently, the impact of oxidation on
the acidity and pore volume of the CNT supports was
stable and obtained comparable catalytic properties to
conventional catalysts, making them interesting model systems for
fundamental studies on FTO.[14,15]
To increase the activity and selectivity of an FTO catalyst
2
investigated by N -physisorption in combination with acid-base
addition of promoters is of vital importance, as unpromoted iron
catalysts usually show high selectivity towards methane.[
promoting iron particles with Na and S promoters, low methane
titration (Table S1). CNT-Ox displayed significantly higher acidity,
with the number of acidic groups being 1.7 nm-2 indicating
successful oxidation. The Fe-NP were then anchored to both
CNT and CNT-Ox which hereafter are named Fe/CNT and
Fe/CNT-Ox. After attachment of the Fe-NP, sodium and sulfur
promoters were added via a ligand exchange method[15,20] and
particle size and distribution were studied with TEM (Figures 2
and S1), these samples are hereafter referred to as FeP/CNT
and FeP/CNT-Ox (P denoting the Na and S promoters). The iron
particles were well distributed on the supports, with no
indications of aggregation or particle growth. In contrast to Fe-
NP on CNT, the particles were less abundant on CNT-Ox
16–19]
By
[
a]
N.A. Krans, E.C. van der Feltz, J. Xie, J. Zečević and K.P. de Jong
Inorganic Chemistry and Catalysis
Debye institute for Nanomaterials Science
Universiteitsweg 99, 3584 CG, Utrecht (The Netherlands)
E-mail: n.a.krans@uu.nl
E-mail: k.p.dejong@uu.nl
Dr. A.I. Dugulan
Fundamental Aspects of Materials and Energy Group
Delft University of Technology
Mekelweg 15, 2629 JB, Delft, The Netherlands
E-mail: a.i.dugulan@tudelft.nl
*
b]
[
(Figure 2b and d) which implied a weaker interaction between
CNT-Ox and the nanoparticles. This was also inferred from
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201800487
COMMUNICATION
Figure 1. TEM image of well dispersed, sterically stabilized by the oleic acid
and oleylamine ligands, 7 nm iron oxide colloidal nanocrystals. The inset
histogram shows the narrow size distribution.
Figure 2. TEM images of fresh catalysts Fe/CNT (A), Fe/CNT-Ox (B),
FeP/CNT (C) and FeP/CNT-Ox (D).
particle growth during TEM imaging that led to a broader particle
size distribution in comparison with the other three samples
(
Figure S1). By XRD analysis (Figure S2) it was observed that
Table 1. Iron, sodium and sulfur weight loading on the different supports
measured by ICP-AES.
all catalysts contained mixed γ-Fe and Fe phases, which
O
2 3
3 4
O
[
14]
is in agreement with our previous findings. Furthermore, ICP-
AES (Table 2) showed that Fe loading in both Fe/CNT and
FeP/CNT samples was 3.4 wt.% while a significantly lower Fe
loading of 2.3 and 2.0 wt.% was detected in Fe/CNT-Ox and
FeP/CNT-Ox respectively. This is in agreement with the TEM
results in Figure 2 revealing a lower number of iron particles
attached to CNT-Ox supports, indicating weaker attachment.
Support
Material
Fe wt.%
Na wt.%
S wt.%
Atomic
ratio Na/S
Fe/CNT
3.4
0.04
0.16
0.05
0.73
0.07
1.11
<DL
0.06
<DL
0.08
0.04
0.07
FeP/CNT
Fe/CNT-Ox
FeP/CNT-Ox
CNT
3.4
2
2.3
The difference in Fe loading observed with TEM as well as
with ICP-AES is tentatively ascribed to a different interaction
between the iron Fe-NP and the supports. This interaction
difference is linked to the Van der Waals forces between
particles and support that relate to the Hamaker constants,
2.0
12
<DL
<DL
CNT-Ox
which is larger for graphitic CNT than for graphene-oxide like
CNT-Ox.[39,40] Additionally, the apparent particle growth during
[a]. After correcting for blank CNTs with an average amount of 0.06 wt.% Na
and 0 wt.% S. DL = Detection Limit
TEM imaging, shown in the histograms in Figure S1, of iron Fe-
NP in FeP/CNT-Ox compared to Fe/CNT-Ox, is explained by
particle growth induced by the electron beam due to this weaker
interaction. The more direct interaction of the Fe-NP surface
towards CNT-Ox once the ligands were exchanged with smaller
significantly higher in CNT-Ox support compared to pristine CNT
Table 1).
(
2
Na S molecules, caused FeP/CNT-Ox to be more stable during
electron microscopy measurements.
The impact of the observed interactions between the iron
particles and promoters with the support on catalyst activity,
selectivity and stability for FTO was studied. As known from
previous literature it was found that adding Na + S promoters
ICP confirmed that ligand exchange was successful as both
promoted catalysts exhibited an increase in Na and S (Table 1).
The Na/S atomic ratio in FeP/CNT was 2, equal to the precursor
decreases the hydrogen coverage on the iron surface and
2
Na S, but the amount of sodium was much higher in FeP/CNT-
[41,42]
thereby the methane selectivity.
As expected, the promoted
Ox than in FeP/CNT catalyst. This is attributed to ion exchange
of Na ions with the surface acidic groups present on the CNT-Ox,
suggesting that most of the Na was present on the support. To
verify this hypothesis, we exposed both CNT and CNT-Ox to the
same amount of S and Na promoters used for the catalyst
preparation. As expected, the amount of sodium was
catalysts showed higher iron time yield (FTY) compared to the
unpromoted catalysts (Figure 3), as well as a decrease in the
CH and an increase in the C -C = selectivity (Table 2). However,
4 2 4
after a 100 h on stream, FeP/CNT-Ox displayed lower iron-
normalized activity than FeP/CNT, most possibly due to particle
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201800487
COMMUNICATION
Table 2. FTO catalysis activity and selectivity. Catalytic activity and selectivity of pristine and oxidized support materials. (10 bar, 340 ºC, H
00 h) FTY of the promoted samples clearly show in an increase due to the Na + S promoters, as well a decrease in methane and increase in lower olefin and
5+ selectivity. The CH , C -C = and C5+ selectivities are given CO excluded.
2
/CO = 2, TOS =
1
C
4
2
4
2
FTY
C5+ selectivity
(% C)
Support Material
CH
4
selectivity (% C)
C
2
-C
4
= selectivity (% C)
2
CO selectivity (% C)
-4
(10 mol/gFe/s)
Fe/CNT
3.5
0.5
6.0
4.5
40
26
10
28
Fe/CNT-Ox
FeP/CNT
N.D.
11
8
N.D.
48
N.D.
34
N.D.
38
FeP/CNT-Ox
51
37
38
Figure 3. Iron-time yield (FTY) for CNT-supported iron catalysts, Fe/CNT
closed squares) and FeP/CNT (open squares) are shown in black, while
(
Fe/CNT-Ox (closed circles) and FeP/CNT-Ox (open circles) are shown in red.
Figure 4. TEM images of spent catalysts Fe/CNT (A), Fe/CNT-Ox (B),
FeP/CNT (C) and FeP/CNT-Ox (D).
growth during catalysis. Moreover, Fe/CNT-Ox showed no
activity after the first 20 hours, leading to contrasting results
compared to literature. Although the literature points to the
beneficial effect of functionalized groups on CNTs for
impregnated iron catalyst,[30] it is clear that when colloidal iron
Fe-NP are used the effect is adverse. Spent catalysts were
analyzed by TEM as shown in Figure 4. Due to conversion of the
iron oxide Fe-NP into the active carbide phase, core-shell
particles had formed and were observed in all catalysts however
Both promoted catalysts were fully carbidized as indicated
by Mössbauer, which matched the TEM results showing 100%
of core-shell particles. More importantly, CNT-Ox supported
catalysts showed noticeable particle growth (Figure 4b and
Figure S4b and d), most probably as a consequence of the
weaker attachment of the Fe-NP to the oxidic surface which led
to sintering during the FTO reaction. These results are
comparable to previous results from our group for impregnated
catalysts on an ordered mesoporous carbon support, where poor
catalytic activity was attributed to growth and encapsulation of
the particles by carbon.[ Finally, these findings indicated that
attaching the colloidal particles to an oxidized carbon nanotube
support inferred deactivation during the catalytic reaction due to
the poor interaction of the Fe-NP to the surface.
[
17]
mostly in the promoted ones.
From quantitative analysis of
34]
TEM images it appears that 35% of such core-shell particles are
present in Fe/CNT, while Fe/CNT-Ox had 75%. This nicely
matched the Mössbauer spectroscopy results (Tables S2 and
Figure S3) that the same amount of iron oxide Fe-NP had
converted into ε-Fe2.2
C
and χ-Fe
5
C
2
phases, which are
In conclusion, colloidal iron oxide nanoparticles (Fe-NP)
were attached to two supports, i.e. pristine CNTs and oxidized
CNTs (CNT-Ox), to investigate the impact on the attachment
and promoter addition on the catalyst activity, selectivity and
stability for FTO catalysis. It was found that the oxidation of the
support led to lower weight loading of Fe-NP because of a
weaker interaction between the CNT-Ox and the particles.
Addition of Na + S promoters resulted in a highly active catalyst
considered active in FTO.[ Interestingly it was observed that
only the core-shell particles had grown, suggesting that
carbidization coincides with particle growth. Even though
Fe/CNT-Ox was inactive after only 20 h on stream this catalyst
seemed to have more activated particles than Fe/CNT. Further
investigations will be necessary to explain the cause of such low
inactivity.
43]
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201800487
COMMUNICATION
for pristine CNTs, while particle growth impeded the FeP/CNT-
Ox activity. Our hypothesis is that the lower activity of CNT-Ox
supported catalysts is due to insufficient attachment of iron Fe-
NP to the support, causing particle growth under FTO conditions.
Finally, it can be concluded that colloidal Fe-NP interact
differently on surface-oxidized carbon supports related to
different Van der Waals forces, which can have a big impact on
the catalyst performance.
Dugulan, K. P. de Jong, Science. 2012, 335, 835–838.
[17]
J. Xie, H. M. Torres Galvis, A. C. J. Koeken, A. Kirilin, A. I. Dugulan,
M. Ruitenbeek, K. P. De Jong, ACS Catal. 2016, 6, 4017–4024.
M. Oschatz, N. Krans, J. Xie, K. P. De Jong, J. Energy Chem. 2016,
[18]
[19]
[20]
[21]
[22]
25, 985–993.
Y. Cheng, J. Lin, K. Xu, H. Wang, X. Yao, Y. Pei, S. Yan, M. Qiao, B.
Zong, ACS Catal. 2016, 6, 389–399.
A. Nag, M. V. Kovalenko, J. S. Lee, W. Liu, B. Spokoyny, D. V.
Talapin, J. Am. Chem. Soc. 2011, 133, 10612–10620.
C. H. Zhang, H. J. Wan, Y. Yang, H. W. Xiang, Y. W. Li, Catal.
Commun. 2006, 7, 733–738.
Experimental Section
J. Y. Park, Y. J. Lee, P. K. Khanna, K. W. Jun, J. W. Bae, Y. H. Kim,
J. Mol. Catal. A Chem. 2010, 323, 84–90.
Details of the fabrication of the iron oxide nanoparticles, the attachment
of the Fe-NPs to a CNT, oxidation of the CNT support by liquid-phase
oxidation and the addition of Na
found in articles previously published by our group.
2
S promoters by ligand exchange can be
[23]
K. R. P. M. Rao, F. E. Huggins, V. Mahajan, G. P. Huffman, V. U. S.
Rao, B. L. Bhatt, D. B. Bukur, B. H. Davis, R. J. O’Brien, Top. Catal.
[14,15,27]
Synthesis
methods and characterization can be found in the supporting information.
1995, 2, 71–78.
[
[
[
[
24]
25]
26]
27]
F. Rodríguez-Reinoso, Carbon. 1998, 36, 159–175.
Y. Yang, K. Chiang, N. Burke, Catal. Today 2011, 178, 197–205.
J. Zhu, A. Holmen, D. Chen, ChemCatChem 2013, 5, 378–401.
T. O. Eschemann, W. S. Lamme, R. L. Manchester, T. E.
Parmentier, A. Cognigni, M. Rønning, K. P. De Jong, J. Catal. 2015,
Acknowledgements
We acknowledge the European Research Council, EU FP7 ERC
Advanced Grant no. 338846. H. C. de Waard (UU) is
acknowledged for ICP-AES measurements. Willem Kegel and
Alfons van Blaaderen are acknowledged for useful discussions
3
28, 130–138.
P. Serp, M. Corrias, P. Kalck, Appl. Catal. A Gen. 2003, 253, 337–
58.
W. Chen, Z. Fan, X. Pan, X. Bao, J. Am. Chem. Soc. 2008, 130,
414–9419.
H. J. Schulte, B. Graf, W. Xia, M. Muhler, ChemCatChem 2012, 4,
50–355.
[
[
[
28]
29]
30]
3
Keywords: Nanoparticles • Fischer-Tropsch • Attachment •
9
Promoters • Carbon Supports
3
[
1]
P. Munnik, P. E. De Jongh, K. P. De Jong, Chem. Rev. 2015, 115,
687–6718.
[
[
31]
32]
J. L. Figueiredo, J. Mater. Chem. A 2013, 1, 9351–9364.
M. S. Shafeeyan, W. M. A. W. Daud, A. Houshmand, A. Shamiri, J.
Anal. Appl. Pyrolysis 2010, 89, 143–151.
6
[
[
2]
3]
F. C. Meunier, ACS Nano 2008, 2, 2441–2444.
J. Ma, Z. Zhu, B. Chen, M. Yang, H. Zhou, C. Li, F. Yu, J. Chen, J.
Mater. Chem. A 2013, 1, 4662.
[
[
33]
34]
C. Alegre, M. E. Gálvez, E. Baquedano, E. Pastor, R. Moliner, M. J.
Lázaro, J. Catal. 1989, 115, 98–106.
[
[
[
[
4]
5]
6]
7]
Y. Yin, A. P. Alivisatos, Nature 2005, 437, 664–670.
G. A. Somorjai, F. Tao, J. Y. Park, Top. Catal. 2008, 47, 1–14.
C.-J. Jia, F. Schüth, Phys. Chem. Chem. Phys. 2011, 13, 2457.
J. P. Den Breejen, P. B. Radstake, G. L. Bezemer, J. H. Bitter, V.
Frøseth, A. Holmen, K. P. De Jong, J. Am. Chem. Soc. 2009, 131,
M. Oschatz, J. P. Hofmann, T. W. van Deelen, W. S. Lamme, N. A.
Krans, E. J. M. Hensen, K. P. De Jong, ChemCatChem 2017, 4,
620–628.
[
[
[
35]
36]
37]
B. H. Juárez, M. Meyns, A. Chanaewa, Y. Cai, C. Klinke, H. Weller,
J. Am. Chem. Soc. 2008, 130, 15282–15284.
7197–7203.
D. Park, S. M. Kim, S. H. Kim, J. Y. Yun, J. Y. Park, Appl. Catal. A
Gen. 2014, 480, 25–33.
[
8]
H. M. Torres Galvis, J. H. Bitter, T. Davidian, M. Ruitenbeek, A. I.
Dugulan, K. P. De Jong, J. Am. Chem. Soc. 2012, 134, 16207–
B. Ritz, H. Heller, A. Myalitsin, A. Kornowski, F. J. Martin-Martinez,
S. Melchor, J. A. Dobado, B. H. Juárez, H. Weller, C. Klinke, ACS
Nano 2010, 4, 2438–2444.
16215.
[
[
[
9]
I. Wender, Fuel Process. Technol. 1996, 48, 189–297.
10]
11]
M. Boudart, M. A. McDonald, J. Phys. Chem. 1984, 88, 2185–2195.
E. De Smit, B. M. Weckhuysen, Chem. Soc. Rev. 2008, 37, 2758–
[
[
[
38]
39]
40]
K. An, S. Alayoglu, T. Ewers, G. A. Somorjai, J. Colloid Interface Sci.
2012, 373, 1–13.
2
781.
J. L. Weber, I. Dugulan, P. de Jongh, K. P. de Jong, ChemCatChem
017, DOI 10.1002/cctc.201701667.
J. L. Li, J. Chun, N. S. Wingreen, R. Car, I. A. Aksay, D. A. Saville,
Phys. Rev. B - Condens. Matter Mater. Phys. 2005, 71, 1–6.
M. J. McAllister, J. L. Li, D. H. Adamson, H. C. Schniepp, A. A.
Abdala, J. Liu, M. Herrera-Alonso, D. L. Milius, R. Car, R. K.
Prud’homme, et al., Chem. Mater. 2007, 19, 4396–4404.
J. Xie, J. Yang, A. I. Dugulan, A. Holmen, D. Chen, K. P. De Jong,
M. J. Louwerse, ACS Catal. 2016, 6, 3147–3157.
[
12]
2
[
[
13]
14]
H. M. Torres Galvis, K. P. De Jong, ACS Catal. 2013, 3, 2130–2149.
M. Casavola, J. Hermannsdörfer, N. De Jonge, A. I. Dugulan, K. P.
De Jong, Adv. Funct. Mater. 2015, 25, 5309–5319.
[
[
41]
42]
[
[
15]
16]
M. Casavola, J. Xie, J. D. Meeldijk, N. A. Krans, A. Goryachev, J. P.
Hofmann, A. I. Dugulan, K. P. De Jong, ACS Catal. 2017, 7, 5121–
H. M. Torres Galvis, A. C. J. Koeken, J. H. Bitter, T. Davidian, M.
Ruitenbeek, A. I. Dugulan, K. P. De Jong, Catal. Today 2013, 215,
5128.
H. M. Torres Galvis, J. H. Bitter, C. B. Khare, M. Ruitenbeek, A. I.
95–102.
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201800487
COMMUNICATION
[
43]
J. W. Niemantsverdriet, A. M. Van der Kraan, W. L. Van Dijk, H. S.
Van der Baan, J. Phys. Chem. 1980, 84, 3363–3370.
This article is protected by copyright. All rights reserved.