Direct Transformation of Syngas to Aromatics over Na-Zn-Fe5C2 and Hierarchical HZSM-5 Tandem Catalysts

Article abstract of DOI:10.1016/j.chempr.2017.06.017

Direct synthesis of aromatics from syngas is a great challenge because of severe operating conditions and low yield of aromatics. Making this process more competitive than the MTA (methanol to aromatics) process will require high energy efficiency and low CO₂ emission. A combination of Na-Zn-Fe₅C₂ and hierarchical HZSM-5 with uniform mesopores dramatically changed the product distribution of Fischer-Tropsch synthesis, leading to 51% aromatic selectivity under the stable stage with CO conversion >85%. C₁₂₊ heavy hydrocarbons almost disappeared, and the catalyst showed good stability. The hierarchical zeolitic structure and Br?nsted acidity of HZSM-5 could be precisely tuned by controlling the alkali treatment conditions and the degree of ion exchange. The appropriate density and strength of the Br?nsted acid sites and the hierarchical pore structure of HZSM-5 endowed the catalyst with an unprecedented aromatic yield. This work shows a broad area for development for syngas conversion.

Full text of DOI:10.1016/j.chempr.2017.06.017

Article
Direct Transformation of Syngas to Aromatics
over Na-Zn-Fe5C2 and Hierarchical HZSM-5
Tandem Catalysts
Bo Zhao, Peng Zhai, Pengfei
Wang, ..., Qianwen Zhang,
Weibin Fan, Ding Ma
fanwb@sxicc.ac.cn (W.F.)
dma@pku.edu.cn (D.M.)
HIGHLIGHTS
An unprecedented aromatic yield
via direct syngas conversion is
obtained
The aromatic selectivity is
dependent on the acidity and
porous structure of HZSM-5
The pivotal role of the density and
strength of acid sites is discussed
Aromatics, the basic chemicals, could be produced from syngas (CO/H2) directly
over a physical mixture of a highly effective FTS catalyst, Na-Zn-Fe5C2, and
HZSM-5 zeolite. The Na-Zn-Fe5C2 component was highly selective toward olefins;
the appropriate density and strength of the Brønsted acid sites and the hierarchical
pore structure of the HZSM-5 component converted the formed olefins and
endowed the tandem catalyst with an unprecedented aromatic yield.
Zhao et al., Chem 3, 1–11
August 10, 2017 ª 2017 Published by Elsevier
Inc.
http://dx.doi.org/10.1016/j.chempr.2017.06.017

Please cite this article in press as: Zhao et al., Direct Transformation of Syngas to Aromatics over Na-Zn-Fe5C2 and Hierarchical HZSM-5 Tandem
Catalysts, Chem (2017), http://dx.doi.org/10.1016/j.chempr.2017.06.017
Article
Direct Transformation of Syngas
to Aromatics over Na-Zn-Fe5C2
and Hierarchical HZSM-5 Tandem Catalysts
Bo Zhao,1,6 Peng Zhai,1,6 Pengfei Wang,2,3 Jiaqi Li,1 Teng Li,1 Mi Peng,1 Ming Zhao,4 Gang Hu,5
1,7,
Yong Yang,2,3 Yong-Wang Li,2,3 Qianwen Zhang,4 Weibin Fan,2, and Ding Ma
*
*
SUMMARY
The Bigger Picture
The depletion of petroleum
Direct synthesis of aromatics from syngas is a great challenge because of severe
operating conditions and low yield of aromatics. Making this process more
competitive than the MTA (methanol to aromatics) process will require high
energy efficiency and low CO2 emission. A combination of Na-Zn-Fe5C2 and
hierarchical HZSM-5 with uniform mesopores dramatically changed the product
distribution of Fischer-Tropsch synthesis, leading to 51% aromatic selectivity
under the stable stage with CO conversion >85%. C12+ heavy hydrocarbons
almost disappeared, and the catalyst showed good stability. The hierarchical
zeolitic structure and Brønsted acidity of HZSM-5 could be precisely tuned by
controlling the alkali treatment conditions and the degree of ion exchange.
The appropriate density and strength of the Brønsted acid sites and the hierar-
chical pore structure of HZSM-5 endowed the catalyst with an unprecedented
aromatic yield. This work shows a broad area for development for syngas con-
version.
reserves threatens economic
development and regional
stability nowadays. Building an
alternative route for fuel and
chemical production has become
a focused research topic in recent
years. Although conversion of
syngas to fuels by the Fischer-
Tropsch synthesis (FTS) process
has been industrialized in various
countries, expanding its products
to value-added chemicals is still a
great challenge. Aromatics are
very important feedstock; they are
used as blending mixtures for
gasoline with a high octane
INTRODUCTION
Transformation of syngas (CO/H2) derived from shale gas, biomass, and coal has
been developed as a promising alternative to oil for the preparation of liquid fuels
and commodity chemicals.1–5 The Fischer-Tropsch synthesis (FTS) process is well
known. The product selectivity of FTS is limited by the law of polymerization, i.e.,
Anderson-Schulz-Flory distribution, which restricts industrial application as the
result of a very wide product distribution. Recently, different novel catalyst design
strategies have been developed to confine the products into a narrow range by
altering the electronic property of metal active sites and the steric structure of
the catalyst.6–12 With S and Na as promoters, high selectivity to low olefins was
obtained over a supported Fe catalyst.7 In addition, the introduction of acidic
zeolite in an FTS catalyst initiates consecutive reactions, such as cracking and isom-
erization of primary products, leading to excellent selectivity toward high-quality
gasoline or diesel.8–10 Very recently, it was reported that coupling of high-temper-
ature methanol synthesis catalyst with SAPO (silico-alumino-phosphate) molecular
sieves produced more than 80% of C2–C4 olefins in hydrocarbons, although high
temperature (673 K) facilitated the water gas shift reaction and generated a large
amount of CO2.11,12
number in some countries and are
the most important molecular
platform for the polymer industry.
With the fast development of the
polymer industry, the gap
between supply and demand calls
for a new strategy for the synthesis
of aromatics. We have established
an intriguing process that directly
converts syngas to aromatics on
the basis of a physical mixture of a
highly effective FTS catalyst and
zeolite, which shows enormous
potential in comparison with the
traditional multiple-step pathway.
Aromatics, mainly produced from petroleum refinement processes, are used as a
fuel additive in some countries and are the most important platform molecules for
the polymer industry. With the fast development of the economy, the gap between
Chem 3, 1–11, August 10, 2017 ª 2017 Published by Elsevier Inc.
1

Please cite this article in press as: Zhao et al., Direct Transformation of Syngas to Aromatics over Na-Zn-Fe5C2 and Hierarchical HZSM-5 Tandem
Catalysts, Chem (2017), http://dx.doi.org/10.1016/j.chempr.2017.06.017
supply and demand calls for a new strategy for the synthesis of aromatics. Starting
from syngas, typically, a two-step process is needed, i.e., we need to transform syn-
gas to methanol on a CuZn catalyst (493–573 K) first, and then convert methanol to
aromatics over a zeolite catalyst (MTA, >673 K).13 However, this two-step process is
highly cost and energy effective. If we could combine the two processes into one,
i.e., direct conversion of syngas into aromatics (STA), this would not only greatly
decrease the equipment and operation costs but would also have an advantage in
product separation and recycling because the reactants are in the gas phase and
the main products are liquid.
(n + 6)CO + (2n + 9)H2 / Cn+6H2n+6 + (n+6)H2O
(n R 0)
Because of the mismatch between the operating temperature of the CuZn catalyst
and that of aromatization catalysts,14 many efforts have been devoted to coupling
a high-temperature methanol synthesis catalyst with an MTA catalyst.15,16 Pioneer-
ing work by Chang et al.17 indicated that 70% aromatics selectivity could be reached
over a Zn-Cr/HZSM-5 catalyst, demonstrating the effectiveness of this concept.
However, the consecutive reactions required severe operating conditions and
showed low reactivity. On the other hand, some researchers combined an Fe-based
FTS catalyst with zeolite to directly synthesize aromatics from syngas. Guan et al.18
reported 40.4% aromatics selectivity in C5+ products on an Fe/MnO-GaZSM-5 cata-
lyst, but rapid deactivation occurred with time on stream. Recently, Yan et al.19
investigated the influence of reaction temperature, pressure, and space velocity
on the production of aromatics on an Fe-Pd/HZSM-5 catalyst, yet it is difficult to
obtain high aromatic selectivity as well as high CO conversion on these catalysts.20,21
As the combination of high CO conversion and high aromatic selectivity would
confer an immediate advantage in industrial applications of the STA process, the
development of more efficient catalyst is highly desired.
Here, we have constructed a tandem catalyst system by combining Na-Zn-Fe5C2
catalysts (with high CO conversion and high selectivity toward alkenes)22 with
hierarchical HZSM-5 for the direct conversion of syngas to aromatics. By physically
mixing the highly selective Na-Zn-Fe5C2 catalyst with the mesoporous HZSM-5
zeolite, we successfully obtained high aromatic selectivity with high CO conversion.
The total selectivity of undesired C1 products, including CH4 and CO2, has been
successfully suppressed to less than 35%, and the yield of aromatics reached as
high as 32.6% in a single run.
1Beijing National Laboratory for Molecular
Sciences, College of Chemistry and Molecular
Engineering, Peking University, Beijing 100871,
China
RESULTS
Preliminary Catalytic Performance for STA
2State Key Laboratory of Coal Conversion,
Institute of Coal Chemistry, Chinese Academy of
Sciences, 27 South Taoyuan Road, Taiyuan,
Shanxi 030001, China
Table 1 shows the catalytic activity and product distribution of syngas conversion
at 613 K and 20 bar. Fe-based FTS processes generally operate between 523 and
623 K. The main products are alkanes containing 30.8% CH4 over a simple SiO2-
supported iron catalyst (Table 1, entry 1). At such a temperature range, aromatiza-
tion of alkanes through dehydrogenation/oligomerization/cyclization over a solid
acid catalyst is not possible because it needs a high reaction temperature, e.g.,
823 K.23 Therefore, a combination of Fe and HZSM-5 (SiO2/Al2O3 = 30) (Table 1,
entry 2) generated C2+ paraffin as the main product, with aromatic selectivity of
only 10.9% at 613 K. By comparing with the catalytic results of Fe/SiO2, it can be
deduced that aromatics are formed at the expense of olefins on acidic HZSM-5.
3Synfuels China, Beijing 100195, China
4Faculty of Materials and Chemical Engineering,
Beijing Institute of Petrochemical Technology,
Beijing 100029, China
5Israel Chemicals Limited, Shanghai 200021,
China
6These authors contributed equally
7Lead Contact
*Correspondence: fanwb@sxicc.ac.cn (W.F.),
dma@pku.edu.cn (D.M.)
Recently, a high-performance catalyst, Na-Zn-Fe5C2 (termed as FeZnNa), was devel-
oped by the co-precipitation method (Figures S1–S3).22 The molar weight ratio of
http://dx.doi.org/10.1016/j.chempr.2017.06.017
2
Chem 3, 1–11, August 10, 2017

Please cite this article in press as: Zhao et al., Direct Transformation of Syngas to Aromatics over Na-Zn-Fe5C2 and Hierarchical HZSM-5 Tandem
Catalysts, Chem (2017), http://dx.doi.org/10.1016/j.chempr.2017.06.017
Table 1. Comparison of the CO Conversion and Product Distribution of Different Catalysts for Syngas Conversion
Entry Catalyst
Al (mmol/g)a B Site
(mmol/g)b
XCO (%) CO2
Selectivity (%)
Selectivity in Hydrocarbons (%)
c
CH4 C2–C4 Olefin C2–C4 Paraffin C5+
Aromatics
1
2
3
4
5
6
7
8
9
Fe/SiO2
–
–
38.6
28.1
74.9
88.3
84.2
89.2
86.6
88.5
88.8
11.3
30.8 13.6
29.5 1.0
10.9 25.5
11.3 1.4
17.0
29.1
4.4
38.6
–
Fe/HZSM-5
1.06
–
1.02
–
13.7
27.3
27.7
26.5
26.9
27.2
27.8
27.5
29.5 10.9
58.1 1.1
28.9 19.8
22.7 30.9
21.8 40.5
52.6 2.1
38.2 26.9
13.2 50.6
FeZnNa
FeZnNa@HZSM-5
FeZnNa@0.1-HZSM-5
FeZnNa@0.6-HZSM-5
FeZnNa@0.6-NaZSM-5
FeZnNa@2.0-HZSM-5
1.06
1.04
1.36
1.36
1.69
1.02
0.46
0.35
0
38.6
35.6
26.8
15.8
21.9
25.6
9.9
9.8
0.7
1.0
10.1 20.4
0.36
0.20
9.8
9.6
2.9
1.0
FeZnNa@0.6-HZSM-5-a 1.36
Reaction conditions: 120 mg catalyst (20 mg FeZnNa + 100 mg ZSM-5, physical mixture), CO/H2/CO2/Ar = 24:64:8:4, 613 K, 20 bar, total flow rate = 20 mL/min,
time on stream = 40 hr. Catalyst was treated in 5%H2/N2 at 623 K for 2 hr before the reaction.
aAluminum content per gram of zeolite, determined by ICP.
bBrønsted sites per gram of zeolite, determined by 1H MAS NMR.
cC5+ products except for aromatics.
Zn and Fe was 1:1, with 0.8 wt % sodium content caused by residual precipitate.
Seventy-eight percent of the hydrocarbon product was olefins, and undesired
C1 (CO2, CH4) products were successfully suppressed with high CO conversion
(Table 1, entry 3). It was found that Fe5C2 was the active phase in this catalyst, and
the sodium acted as an electronic promoter that affected the elementary reaction
on the surface to produce more alkenes. The thermodynamic calculation shows
that, different from alkanes, aromatization of alkenes is feasible below 623 K and
long-chain alkenes are apt to cyclization without oligomerization (Figure S4).24
Thus, the alkene-favorable Na-Zn-Fe5C2 catalyst would be an ideal platform for suc-
cessive aromatization reactions.
After physically mixing Na-Zn-Fe5C2 with HZSM-5, the aromatic selectivity increased
sharply from 1.1% to 19.8% (Table 1, entry 4), whereas C2–C4 olefin selectivity
decreased dramatically from 25.5% to 1.4%, and the olefin to paraffin ratio of hydro-
carbons decreased greatly from 5.8 to 0.036. Simultaneously, CO conversion
increased from 74.9% to 88.3% in comparison with pure Na-Zn-Fe5C2 catalyst.
This can be attributed to the tension force by consecutive consumption of interme-
diates, which in turn favors the conversion of the initial reactants. The sharp increase
in C2–C4 paraffin selectivity from 4.4% to 38.6% is due to the occurrence of dealky-
lation of polyalkylbenzene and the hydrogen transfer reaction within the reaction
network.25 Hydrogen generated during the aromatization process is transferred
into unsaturated hydrocarbons, leading to the formation of additional paraffin.
Modified HZSM-5 by Base Treatment
Although HZSM-5 can catalyze different types of acid-based reactions, such as dehy-
dration of alcohols, cracking of hydrocarbons, oligomerization and cyclization of ole-
fins, and dehydrogenation of paraffin,26 different types of reactions require different
acidity, porosity, and reaction conditions. As the acid properties of zeolites depend
primarily on the location, the geometry, and the coordination of aluminum in its
framework, zeolite acidity can be engineered to improve its catalytic performance
in specific reactions. In the current study, removal of excessive acid sites and creation
of additional pore structures seem to be a promising way to further increase the
selectivity toward aromatics, because aromatization reactions usually require mild
Chem 3, 1–11, August 10, 2017
3

Please cite this article in press as: Zhao et al., Direct Transformation of Syngas to Aromatics over Na-Zn-Fe5C2 and Hierarchical HZSM-5 Tandem
Catalysts, Chem (2017), http://dx.doi.org/10.1016/j.chempr.2017.06.017
Brønsted acidity27,28 and a hierarchical pore structure, which ensures fast product
diffusion and desorption.29,30
We treated pristine HZSM-5 with NaOH solutions of different concentrations, which
not only adjusts its acidity but also creates mesostructured porosity.31,32 The treated
zeolites were termed x-HZSM-5, where x represents the NaOH concentration. For
example, 0.1-HZSM-5 means HZSM-5 treated with 0.1 M NaOH solution at 353 K
for 1 hr and then exchanged with 1 M NH4NO3 solution three times (all the samples
were exchanged three times unless specified otherwise). Figure 1A shows that all the
treated HZSM-5 samples show diffraction patterns characteristic of MFI structure,
but the diffraction lines of the samples treated with R0.6 M NaOH solution were
obviously broadened and decreased in intensity. This phenomenon suggests that
NaOH treatment partially destroyed HZSM-5 structure by desilication,33 whereas
the main MFI architecture was preserved.
The N2 adsorption/desorption isotherms clearly show that HZSM-5 has a typical
microporous structure. In contrast, mesoporous structures were formed in NaOH-
treated HZSM-5, as revealed by its H4-type N2 sorption isotherms (Figures 1B
and 1C). Upon NaOH treatment, an obvious increase in the mesopore volume was
observed despite the fact that the surface area and micropore volume remained
similar (except for 2.0-HZSM-5; its surface area and micropore volume declined
significantly as a result of severe destruction of the ZSM-5 structure) (Table S1).
The mesopore volume and diameter determined from the desorption branch by
the Barrett-Joyner-Halenda (BJH) method increased with increasing NaOH con-
centration up to 1.0 M, indicating the presence of large numbers of mesopores in
the sample. Transmission electron microscopy shows that 2–3 mm pristine HZSM-5
crystals with smooth surfaces became sponge-like because plenty of mesopores
ranging from 4 to 10 nm were created as a result of the NaOH treatment (Figures
1D and 1E). These results indicate that the hierarchical pore structure in ZSM-5 crys-
tals was formed through a desilication process (Table S2).
Given that acid site and aluminum in HZSM-5 play a critical role in the aromatization
process, the distribution and reconstruction of aluminum species and acid sites in
parent HZSM-5 and those treated with NaOH were tracked with NH3-tempera-
ture-programmed desorption (TPD), pyridine-infrared, 27Al magic angle spinning
(MAS) and multiple quantum (MQ)-MAS nuclear magnetic resonance (NMR), and
1H MAS NMR techniques (Figures 2 and S5–S7). Treatment with 0.1 M NaOH solu-
tion partially removed framework aluminum (AlF, at 55 ppm, the main origin of
Brønsted acid sites) and led to the formation of species with distorted framework
Al (AlDF, with relatively weak Brønsted acidity), and six-coordinated (Al6NF, at about
0 ppm) Al species.34 With the increase of NaOH concentration (0.6 M), more AlF
species were expelled from the zeolite framework; five-coordinated (Al5NF, around
30 ppm) and ex-framework or tri-coordinated Al species (a very broad hump, so-
called invisible Al) were clearly resolved on 27Al MAS and 27Al MQ MAS NMR spectra
(Figures 2A, 2D, and 2E).35,36 As strong Brønsted acid sites (BAS) are associated with
AlF, the removal of AlF from the zeolitic lattice will decrease the amount of BAS in the
catalyst. The NH3-TPD profile indicates that the ammonia desorption associated
with strong acid sites (>573 K) decreased with the degree of NaOH treatment
(Figure 2C). Direct evidence about BAS is provided by the 1H MAS NMR profile (Fig-
ure 2B). It shows that the peaks characteristic of Brønsted acid sites (4.0 and 6.1 ppm)
decreased sharply after NaOH treatment (0.1 M), suggesting the removal of large
amounts of BAS in comparison with parent HZSM-5 (0.46 versus 1.02 mmol/g).
When the concentration of NaOH solution increased to 0.6 M, the concentration
4
Chem 3, 1–11, August 10, 2017

Please cite this article in press as: Zhao et al., Direct Transformation of Syngas to Aromatics over Na-Zn-Fe5C2 and Hierarchical HZSM-5 Tandem
Catalysts, Chem (2017), http://dx.doi.org/10.1016/j.chempr.2017.06.017
Figure 1. The Structure of the Modified HZSM-5
(A) X-ray diffraction patterns of modified HZSM-5.
(B) N2 adsorption-desorption isotherms for modified HZSM-5.
(C) Pore-size distributions of HZSM-5 derived from the N2 adsorption-desorption isotherms by the
BJH method. The number before HZSM-5 represents the NaOH concentration of the solution that
treated HZSM-5.
(D) TEM micrograph of the HZSM-5. The scale bar represents 50 nm.
(E) TEM micrograph of 0.6-HZSM-5. The scale bar represents 50 nm.
Chem 3, 1–11, August 10, 2017
5

Please cite this article in press as: Zhao et al., Direct Transformation of Syngas to Aromatics over Na-Zn-Fe5C2 and Hierarchical HZSM-5 Tandem
Catalysts, Chem (2017), http://dx.doi.org/10.1016/j.chempr.2017.06.017
Figure 2. The Acidity and Framework Structure of Modified ZSM-5
(A) 27Al MAS NMR spectra.
(B) 1H MAS NMR spectra.
(C) NH3-TPD profiles for HZSM-5 prepared with different concentrations of NaOH or different ion-
exchange times.
(D) 27Al MQ MAS NMR spectra of HZSM-5.
(E) 27Al MQ MAS NMR spectra of 0.6-HZSM-5.
of BAS decreased further to 0.35 mmol/g, and its strength also weakened
(Figure 2C).
STA on FeZnNa and Modified ZSM-5 Tandem Catalysts
After Brønsted acidity decreased with 0.1 M NaOH treatment, the aromatic
selectivity increased greatly to 30.9% on FeZnNa@0.1-HZSM-5 with a decrease in
selectivity toward methane and C2–C4 paraffins (Table 1, entry 5). With the further
reduction of strong Brønsted acidity sites, the selectivity toward aromatics reached
40.5% on FeZnNa@0.6-HZSM-5 (Table 1, entry 6; Figures S8–S10), which is twice
that of FeZnNa@HZSM-5 catalyst. Although BAS is the active site for aromatization,
as supported by the very low aromatic selectivity obtained over the FeZnNa@0.6-
NaZSM-5 catalyst (which has not been exchanged with NH4NO3 solution and does
not have BAS; Table 1, entry 7), an excessive amount of strong BAS could facilitate
the cracking of long-chain hydrocarbons and generation of coking species,37,38
which decrease the catalytic stability and selectivity toward aromatics. The acidity
of BAS increased with 2.0 M NaOH-treated HZSM-5 by desilication, leading to the
decline in aromatic selectivity (Table 1, entry 8).
Alongside acidity, the hierarchical pore structure created during NaOH treatment
can also promote the catalytic performance of the FeZnNa@zeolite catalyst. Such
a pore structure can serve as highly effective micro-reactors for aromatization of
6
Chem 3, 1–11, August 10, 2017

Please cite this article in press as: Zhao et al., Direct Transformation of Syngas to Aromatics over Na-Zn-Fe5C2 and Hierarchical HZSM-5 Tandem
Catalysts, Chem (2017), http://dx.doi.org/10.1016/j.chempr.2017.06.017
alkene produced over FeZnNa catalyst by greatly facilitating the diffusion of reac-
tants and products, resulting in improved accessibility and transformation efficiency.
The reaction behavior of both C2H4 and C3H6 was investigated to study the aroma-
tization properties of different ZSM-5 zeolites (Table S3; Figures S11 and S12).
Na-type ZSM-5 cannot catalyze the transformation of light olefins, which confirmed
the deduction above. Although oligomerization seems to occur on both HZSM-5 and
0.6-HZSM-5, cyclization and dehydrogenation occur much more easily on the latter
catalyst. More importantly, the mesoporous zeolite shows high catalytic stability in
a diluted C2H4 atmosphere, whereas HZSM-5 almost died after 35 hr on stream.
Rapid deactivation was also observed on untreated HZSM-5 when n-C6H12 was
used as feed (Table S4; Figures S13 and S14).
Besides desilication and the creation of the hierarchical pore structure by NaOH
solution treatment, the catalytic performance of treated HZSM-5 can be finely tuned
by control of the type and concentration of cations attached to AlF of ZSM-5. If ZSM-5
is fully exchanged with proton, e.g., ion-exchanged with NH4NO3 three times, one
AlF site has one associated proton, thus, equivalent to one BAS. We can finely tune
the zeolitic acidity of zeolite by controlling the degree of proton-sodium exchange.
For example, after treating HZSM-5 with 0.6 M NaOH, we can exchange it with 1 M
NH4NO3 solution just once to get 0.6-ZSM-5-a. Compared with 0.6-HZSM-5
(exchanged with NH4NO3 three times), it has the same pore structure and almost
identical aluminum species distribution but different amounts of protons associated
with AlF species (0.20 versus 0.35 mmol BAS/g zeolite) (Figure 2B; Table S5). In
addition, the strength of BAS was also concomitantly modulated, with more strong
BAS removed (Figure 2C). Such a sample exhibited promising aromatics selectivity,
being as high as 50.6% when coupling with FeZnNa (Table 1, entry 9; Figures S15
and S16), whereas the conversion of CO remains almost intact, consequently result-
ing in an aromatic yield of more than 32% in a single run. This is the highest aromatic
yield reported in a single-run reaction in the FTS process. A comparison of the hydro-
carbon selectivity of FeZnNa and FeZnNa@0.6-ZSM-5-a indicates that almost all ole-
fins participate in the aromatization process (Figure 3).
DISCUSSION
The hierarchical pore structure and suitable BAS density and strength of
FeZnNa@0.6-ZSM-5-a are pivotal for its high selectivity toward aromatics in FTS.
From the results of the olefin conversion experiments, a simplified scheme for this
process is shown as Scheme S1, where olefins play an irreplaceable role as the
central hub. The dramatic decrease in the selectivity of heavy hydrocarbons (C12+
)
suggests that these molecules have been cracked into olefins and subsequently
transformed into aromatics and light paraffins on acid sites. Significantly, the selec-
tivity toward CH4 and CO2 is less than 35%, ensuring that most carbon resources
transformed into aromatics (dominated by C6–9 aromatics) and light hydrocarbons.
The high CO conversion and high product selectivity are pivotal for the separation
and practical application of this direct STA process.
EXPERIMENTAL PROCEDURES
Synthesis of FeZnNa Catalyst
FeZnNa catalyst was prepared by a co-precipitation method. Equal molar weights of
ferric nitrate and zinc nitrate were dissolved in water together for a mixed solution.
An aqueous solution of sodium carbonate and mixed solution was added to a
flask dropwise under stirring, and the pH value was controlled at 8.5–9.0 for the
dropping process. After aging for 5 hr, the precipitate was washed with deionized
Chem 3, 1–11, August 10, 2017
7

Please cite this article in press as: Zhao et al., Direct Transformation of Syngas to Aromatics over Na-Zn-Fe5C2 and Hierarchical HZSM-5 Tandem
Catalysts, Chem (2017), http://dx.doi.org/10.1016/j.chempr.2017.06.017
Figure 3. Comparison of Product Distribution over FeZnNa and FeZnNa@0.6-ZSM-5-a with All
the Carbon-Containing Products Counted
Reaction conditions: 20 mg FeZnNa, 20 mg FeZnNa + 100 mg 0.6-ZSM-5-a, CO/H2/CO2/
Ar = 24:64:8:4, 613 K, 20 bar, total flow rate = 20 mL/min, time on stream = 40 hr. Catalyst was
treated in 5% H2/N2 at 623 K for 2 hr before the reaction.
water (5 3 100 mL) and dried at 393 K for 10 hr and then calcined in a muffle
furnace at 623 K for 4 hr. The content of sodium was 0.8 wt % derived from induc-
tively coupled plasma (ICP) atomic emission spectrometry analysis (conducted
with a PROFILE SPEC).
Synthesis of Modified ZSM-5
H-ZSM-5 with a SiO2/Al2O3 ratio of 30 was purchased from Nankai University
Catalyst. HZSM-5 (4.0 g) was treated in 100 mL of aqueous solutions of NaOH at
different concentrations for 1 hr at 353 K. The filter cake was washed with deionized
water, dried at 393 K for 10 hr, and then calcined at 623 K for 5 hr. The powder
obtained was denoted as NaZSM-5. NaZSM-5 (1.0 g) was exchanged in 100 mL
of an aqueous solution of 1.0 mol/L NH4NO3. The filter cake was dried at 393 K
for 10 hr and then calcined at 623 K for 5 hr to get proton-formed zeolite. Unless
otherwise indicated, the zeolite used in the reaction was exchanged three times.
Synthesis of Reference Catalysts
Fe/SiO2 and Fe/HZSM-5 were prepared by the incipient impregnation method using
ferric nitrate as the precursor. After impregnation, the sample was dried at 393 K for
10 hr, followed by calcination in air at 773 K for 5 hr. The loading of Fe in each catalyst
was fixed at 10.0 wt %.
8
Chem 3, 1–11, August 10, 2017

Please cite this article in press as: Zhao et al., Direct Transformation of Syngas to Aromatics over Na-Zn-Fe5C2 and Hierarchical HZSM-5 Tandem
Catalysts, Chem (2017), http://dx.doi.org/10.1016/j.chempr.2017.06.017
MAS NMR and MQ MAS
1H and 27Al MAS NMR experiments were performed on a Bruker Avance III 600
spectrometer at resonance frequencies of 600.5, 156.5, and 119.3 MHz. Before
1
each H MAS NMR experiment, about 0.3 g of the sample was packed into a glass
tube, which was later connected to a vacuum line. The temperature was gradually
increased at a rate of 2 K/min to 693 K and kept at a pressure below 10À2 Pa for
20 hr to dehydrate the sample. After the samples were cooled to room tempera-
ture, the glass tubes were flame sealed. Then the samples were transferred into
a 4 mm ZrO2 rotor under a dry nitrogen atmosphere in a glove box. 1H MAS
NMR spectra were acquired with a Hahn-echo sequence for removal of the back-
ground signal arising from the probe; the p/2 and p pulse lengths were 4.2 and
8.4 ms, respectively, at a spinning rate of 13 kHz and a recycle delay of 5 s. The
27Al MAS NMR spectra were acquired on a 4 mm probe with a spinning rate of
13 kHz, a p/18 pulse length of 0.3 ms, a recycle delay of 1 s, and with 1 M
Al(NO3)3 solution as a reference. The chemical shifts of 1H were referenced to
trimethylsilane.
MQ MAS spectra were recorded with a sequence of three pulses. The pulse lengths
were p1 = 6.8 ms, p2 = 2 ms, and p3 = 29 ms. The data were processed and sheared
after Fourier transformation with Bruker Topspin.
Catalytic Tests
The reaction was carried out in a three-channel fixed-bed reactor. All the catalyst was
sieved through a 40–60 mesh. FeZnNa and ZSM-5 were physically mixed and put
into a stainless steel pipe with a quartz tube. Before the reaction, all catalyst was pre-
treated in a 5% H2 stream at 623 K for 5 hr and cooled down to 553 K. A syngas (24%
CO, 8% CO2, 64% H2, and 4% Ar) stream was pumped at a flow rate of 20 mL/min.
The pressure in the reactor was tuned to 20 bar and the temperature of the catalyst
bed was maintained at 613 K. Olefin conversion experiment conditions: zeolite
100 mg, 613 K, 20 bar, C2H4/Ar/N2 = 20:5:75, C3H6/Ar/N2 = 7.5:2.5:90, C3H8/Ar/
N2 = 10:2:88, total flow rate 20 mL/min. Feed of n-C6H12 0.015 mL/min by Lab Alli-
ance Series II pump, flow rate of N2 20 mL/min, 1.0 bar.
The products were separated by a condensing collector. Gaseous products
were detected online by an Agilent GC 7890A system. A Hayesep-Q plot and a
MolSieve 5A plot were packed on a thermal conductivity detector for the separa-
tion of CO2, Ar, CH4, and CO. An HP PLOT Al2O3 was packed on a flame ionization
detector (FID) for the separation of hydrocarbons (C1–C7). CO conversion was
calculated by the internal standard method with Ar. Liquid products were detected
on an Agilent GC 7820 system with an HP-5 capillary column and an FID. The
detailed aromatic product distribution was determined by Agilent GC-MS
(7890A GC+ 5975 MS).
SUPPLEMENTAL INFORMATION
Supplemental Information includes Supplemental Experimental Procedures, 16 fig-
ures, 5 tables, and 1 scheme and can be found with this article online at http://dx.
doi.org/10.1016/j.chempr.2017.06.017.
AUTHOR CONTRIBUTIONS
D.M. and W.F. conceived and designed this work; B.Z. and P.Z. designed the exper-
iments, catalyst preparation, and catalytic reactions; P.W. did the NMR measure-
ments; J.L., T.L., M.P., and M.Z. did part of the catalytic evaluation as well as other
Chem 3, 1–11, August 10, 2017
9

Please cite this article in press as: Zhao et al., Direct Transformation of Syngas to Aromatics over Na-Zn-Fe5C2 and Hierarchical HZSM-5 Tandem
Catalysts, Chem (2017), http://dx.doi.org/10.1016/j.chempr.2017.06.017
characterization measurements; D.M. and P.Z. wrote the paper; and all authors
participated in the analysis of the experimental data and discussions of the results,
as well as preparation of the paper.
ACKNOWLEDGMENTS
This work was financially supported by the Natural Science Foundation of China
(91645115, 21473003, 21222306, 21473229, and 91545121) and 973 Project
(2017YFB0602200 and 2013CB933100).
Received: February 28, 2017
Revised: March 27, 2017
Accepted: June 26, 2017
Published: August 3, 2017
REFERENCES AND NOTES
1. Khodakov, A.Y., Chu, W., and Fongarland, P.
(2007). Advances in the development of novel
cobalt Fischer-Tropsch catalysts for synthesis
of long-chain hydrocarbons and clean fuels.
Chem. Rev. 107, 1692–1744.
mesoporous zeolite-Y-supported cobalt
nanoparticles. Angew. Chem. Int. Ed. 54, 4553–
4556.
19. Yan, Q.G., Lu, Y.W., Wan, C.X., Han, J.,
Rodriguez, J., Yin, J.J., and Yu, F. (2014).
Synthesis of aromatic-rich gasoline-range
hydrocarbons from biomass-derived syngas
over a Pd-promoted Fe/HZSM-5 catalyst.
Energy Fuels 28, 2027–2034.
10. Bao, J., He, J., Zhang, Y., Yoneyama, Y., and
Tsubaki, N. (2008). A core/shell catalyst
produces a spatially confined effect and shape
selectivity in a consecutive reaction. Angew.
Chem. Int. Ed. 47, 353–356.
2. Schweicher, J., Bundhoo, A., and Kruse, N.
(2012). Hydrocarbon chain lengthening in
catalytic CO hydrogenation: evidence for a
CO-insertion mechanism. J. Am. Chem. Soc.
134, 16135–16138.
20. Nimz, M., Lietz, G., Volter, J., Lazar, K., and
Guczi, L. (1988). Direct conversion of syngas to
aromatics on FePd/SiO2 catalyst. Catal. Lett. 1,
93–98.
11. Jiao, F., Li, J.J., Pan, X.L., Xiao, J.P., Li, H.B., Ma,
H., Wei, M.M., Pan, Y., Zhou, Z.Y., Li, M.R., et al.
(2016). Selective conversion of syngas to light
olefins. Science 351, 1065–1068.
3. Xiang, Y.Z., Barbosa, R., and Kruse, N. (2014).
Higher alcohols through CO hydrogenation
over CoCu catalysts: influence of precursor
activation. ACS Catal. 4, 2792–2800.
21. Wijayapala, R., Yu, F., Pittman, C.U., and Mlsna,
T.E. (2014). K-promoted Mo/Co- and
Mo/Ni-catalyzed Fischer-Tropsch synthesis of
aromatic hydrocarbons with and without a Cu
water gas shift catalyst. Appl. Catal. A. Gen.
480, 93–99.
12. Cheng, K., Gu, B., Liu, X.L., Kang, J.C., Zhang,
Q.H., and Wang, Y. (2016). Direct and highly
selective conversion of synthesis gas into lower
olefins: design of a bifunctional catalyst
combining methanol synthesis and carbon-
carbon coupling. Angew. Chem. Int. Ed. 55,
4725–4728.
4. Sartipi, S., Parashar, K., Makkee, M., Gascon, J.,
and Kapteijn, F. (2013). Breaking the Fischer-
Tropsch synthesis selectivity: direct conversion
of syngas to gasoline over hierarchical
Co/H-ZSM-5 catalysts. Catal. Sci. Technol. 3,
572–575.
22. Zhai, P., Xu, C., Gao, R., Liu, X., Li, M.Z., Li, W.Z.,
Fu, X.P., Jia, C.J., Xie, J.L., Zhao, M., et al.
(2016). Highly tunable selectivity for syngas-
derived alkenes over zinc and sodium-
modulated Fe5C2 catalyst. Angew. Chem. Int.
Ed. 55, 9902–9907.
13. Olsbye, U., Svelle, S., Bjorgen, M., Beato, P.,
Janssens, T.V.W., Joensen, F., Bordiga, S., and
Lillerud, K.P. (2012). Conversion of methanol to
hydrocarbons: how zeolite cavity and pore size
controls product selectivity. Angew. Chem. Int.
Ed. 51, 5810–5831.
5. Calderone, V.R., Shiju, N.R., Curulla-Ferre, D.,
Chambrey, S., Khodakov, A., Rose, A.,
Thiessen, J., Jess, A., and Rothenberg, G.
(2013). De novo design of nanostructured iron-
cobalt Fischer-Tropsch catalysts. Angew.
Chem. Int. Ed. 52, 4397–4401.
23. Meriaudeau, P., and Naccache, C. (1997).
Dehydrocyclization of alkanes over zeolite-
supported metal catalysts: monofunctional or
bifunctional route. Catal. Rev. 39, 5–48.
14. Chang, C.D., Lang, W.H., Silvestri, A.J., and
Smith, R.L. (1978). Conversion of synthesis
gas to hydrocarbon mixtures. US patent
US4096163A, filed October 8, 1976, and
published June 20, 1978.
6. Sun, Z.K., Sun, B., Qiao, M.H., Wei, J., Yue, Q.,
Wang, C., Deng, Y.H., Kaliaguine, S., and Zhao,
D.Y. (2012). A general chelate-assisted
co-assembly to metallic nanoparticles-
incorporated ordered mesoporous carbon
catalysts for Fischer-Tropsch synthesis. J. Am.
Chem. Soc. 134, 17653–17660.
24. Scurrell, M.S. (1987). Prospects for the direct
conversion of light alkanes to petrochemical
feedstocks and liquid fuels - a review. Appl.
Catal. 32, 1–22.
15. Simard, F., Sedran, U.A., Sepulveda, J., Figoli,
N.S., and Delasa, H.I. (1995). ZnO-
25. Kim, Y.T., Chada, J.P., Xu, Z.R., Pagan-Torres,
Y.J., Rosenfeld, D.C., Winniford, W.L., Schmidt,
E., and Huber, G.W. (2015). Low-temperature
oligomerization of 1-butene with H-ferrierite.
J. Catal. 323, 33–44.
Cr2O3+ZSM-5 catalyst with very-low Zn/Cr ratio
for the transformation of synthesis gas to
hydrocarbons. Appl. Catal. A. Gen. 125, 81–98.
7. Galvis, H.M., Bitter, J.H., Khare, C.B.,
Ruitenbeek, M., Dugulan, A.I., and de Jong,
K.P. (2012). Supported iron nanoparticles as
catalysts for sustainable production of lower
olefins. Science 335, 835–838.
16. Dagle, R.A., Lizarazo-Adarme, J.A., Dagle, V.L.,
Gray, M.J., White, J.F., King, D.L., and Palo,
D.R. (2014). Syngas conversion to gasoline-
range hydrocarbons over Pd/ZnO/Al2O3 and
ZSM-5 composite catalyst system. Fuel
Process. Technol. 123, 65–74.
26. Rahimi, N., and Karimzadeh, R. (2011). Catalytic
cracking of hydrocarbons over modified ZSM-5
zeolites to produce light olefins: a review. Appl.
Catal. A. Gen. 398, 1–17.
8. Kang, J.C., Cheng, K., Zhang, L., Zhang, Q.H.,
Ding, J.S., Hua, W.Q., Lou, Y.C., Zhai, Q.G., and
Wang, Y. (2011). Mesoporous zeolite-
supported ruthenium nanoparticles as highly
selective Fischer-Tropsch catalysts for the
production of C5-C11 isoparaffins. Angew.
Chem. Int. Ed. 50, 5200–5203.
27. Ma, D., Deng, F., Fu, R.Q., Dan, X.W., and Bao,
X.H. (2001). MAS MMR studies on the
dealumination of zeolite MCM-22. J. Phys.
Chem. B 105, 1770–1779.
17. Chang, C.D., Lang, W.H., and Silvestri, A.J.
(1979). Synthesis gas conversion to aromatic-
hydrocarbons. J. Catal. 56, 268–273.
28. Ma, D., Lu, Y., Su, L.L., Xu, Z.S., Tian, Z.J., Xu, Y.,
Lin, L.W., and Bao, X.H. (2002). Remarkable
improvement on the methane aromatization
reaction: a highly selective and coking-resistant
catalyst. J. Phys. Chem. B 106, 8524–8530.
9. Peng, X.B., Cheng, K., Kang, J.C., Gu, B., Yu, X.,
Zhang, Q.H., and Wang, Y. (2015). Impact of
hydrogenolysis on the selectivity of the Fischer-
Tropsch synthesis: diesel fuel production over
18. Guan, N., Liu, Y., and Zhang, M. (1996).
Development of catalysts for the production of
aromatics from syngas. Catal. Today 30,
207–213.
10
Chem 3, 1–11, August 10, 2017

Please cite this article in press as: Zhao et al., Direct Transformation of Syngas to Aromatics over Na-Zn-Fe5C2 and Hierarchical HZSM-5 Tandem
Catalysts, Chem (2017), http://dx.doi.org/10.1016/j.chempr.2017.06.017
29. Verboekend, D., and Perez-Ramirez, J. (2011).
Design of hierarchical zeolite catalysts by
desilication. Catal. Sci. Technol. 1, 879–890.
bifunctional solids for the one-step synthesis of
liquid fuels from syngas: a review. Catal. Sci.
Technol. 4, 893–907.
multinuclear MAS NMR study. J. Phys. Chem. B
105, 1786–1793.
36. Ma, D., Han, X.W., Xie, S.J., Bao, X.H., Hu,
H.B., and Au-Yeung, S.C. (2002). An
investigation of the roles of surface aluminum
and acid sites in the zeolite MCM-22. Chem.
Eur. J. 8, 162–170.
30. Groen, J.C., Jansen, J.C., Moulijn, J.A., and
Perez-Ramirez, J. (2004). Optimal aluminum-
assisted mesoporosity development in MFI
zeolites by desilication. J. Phys. Chem. B 108,
13062–13065.
33. Chen, H., Wang, Q.F., Zhang, X.W., and
Wang, L. (2014). Hydroconversion of
jatropha oil to alternative fuel over
hierarchical ZSM-5. Ind. Eng. Chem. Res. 53,
19916–19924.
37. Lin, X.Y., Fan, Y., Shi, G., Liu, H.Y., and Bao, X.J.
(2007). Coking and deactivation behavior of
HZSM-5 zeolite-based FCC gasoline hydro-
upgrading catalyst. Energy Fuels 21, 2517–
2524.
31. Niu, X.J., Gao, J., Miao, Q., Dong, M., Wang,
G.F., Fan, W.B., Qin, Z.F., and Wang, J.G.
(2014). Influence of preparation method on the
performance of Zn-containing HZSM-5
catalysts in methanol-to-aromatics.
Microporous Mesoporous Mater. 197,
252–261.
34. van Bokhoven, J.A., Koningsberger, D.C.,
Kunkeler, P., van Bekkum, H., and Kentgens,
A.P. (2000). Stepwise dealumination of zeolite
beta at specific T-sites observed with 27Al MAS
and 27Al MQ MAS NMR. J. Am. Chem. Soc.
122, 12842–12847.
38. Sahoo, S.K., Viswanadham, N., Ray, N., Gupta,
J.K., and Singh, I.D. (2001). Studies on acidity,
activity and coke deactivation of ZSM-5 during
n-heptane aromatization. Appl. Catal. A. Gen.
205, 1–10.
35. Ma, D., Shu, Y.Y., Han, X.W., Liu, X.M., Xu, Y.D.,
and Bao, X.H. (2001). Mo/HMCM-22 catalysts
for methane dehydroaromatization: a
32. Sartipi, S., Makkee, M., Kapteijn, F., and
Gascon, J. (2014). Catalysis engineering of
Chem 3, 1–11, August 10, 2017
11