Fischer-Tropsch synthesis: Deuterium labeled ethanol tracer studies on iron catalysts

Article abstract of DOI:10.1016/j.apcata.2010.06.039

The addition of D₃- and D₅-ethanol during Fischer-Tropsch synthesis with a promoted iron catalyst gives results that are consistent with ethanol initiating chain growth. However, there is considerable H/D exchange in the alcohol prior to initiating chain growth so that the D/molecule ratios in the alkane and alkene products are in the range 1.3-1.5. The nearly constant D/molecule ratio in the alkane and alkene products in the C₅-C₁₄ carbon number range is consistent with ethanol initiating chain growth.

Full text of DOI:10.1016/j.apcata.2010.06.039

Applied Catalysis A: General 385 (2010) 46–51
Contents lists available at ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Fischer–Tropsch synthesis: Deuterium labeled ethanol tracer studies on iron
catalysts
Muthu K. Gnanamania, Robert A. Keogha, Wilson D. Shafera, Buchang Shib, Burtron H. Davisa,∗
a Center for Applied Energy Research, University of Kentucky, 2540 Research Park Dr., Lexington, KY 40511, United States
b Department of Chemistry, Eastern Kentucky University, Richmond, KY 40475, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
The addition of D3- and D5-ethanol during Fischer–Tropsch synthesis with a promoted iron catalyst
gives results that are consistent with ethanol initiating chain growth. However, there is considerable H/D
exchange in the alcohol prior to initiating chain growth so that the D/molecule ratios in the alkane and
alkene products are in the range 1.3–1.5. The nearly constant D/molecule ratio in the alkane and alkene
products in the C5–C14 carbon number range is consistent with ethanol initiating chain growth.
© 2010 Elsevier B.V. All rights reserved.
Received 5 February 2010
Received in revised form 16 June 2010
Accepted 19 June 2010
Available online 25 June 2010
1. Introduction
olefin to paraffin ratio of the products and observed only a slight
degree of chain incorporation upon addition of 1-olefin or ethanol
The Fischer-Tropsch synthesis (FT) can be briefly defined as the
catalytic conversion of carbon monoxide and hydrogen to hydro-
carbon products. It has long been known that smaller 1-alkene and
alcohol molecules may be able to act as a chain initiator under nor-
mal FT synthesis conditions [1]. The results from the addition of a
C2 monomer during FT has been reviewed by a number of workers
[2–4], who advanced a number of different surface species such as
in major amounts. Therefore, it is not only the quantity of the
labeled compound that is co-fed along with a syngas during FT
synthesis but also the synthesis conditions that may determine the
rate of incorporation of the added compound into the FT synthesis
products.
In the 1960s, Kryukov et al. [11] used ␣-deuteroethanol as a
tracer to study the mechanism of the FT reaction. The authors
concluded that intermediate surface compounds obtained from
the alcohol take part in the formation of hydrocarbons and that
they retain at least a part of the deuterium of the added alcohol.
Brundage and Chuang [12] obtained a two-unequal-hump response
for d1- and d2-propionaldehyde formation from D2 and C2D4 pulses
during ethylene hydroformylation using Mn-Rh/SiO2 at 513 K and
0.1 MPa. The authors demonstrated using FT-IR/mass spectrometry
that the different deuterium incorporation pathways into the pro-
pionaldehyde was possibly due to H/D exchange on adsorbed acyl
and acyl hydrogenation. The earlier work with 14C-labeled ethanol
clearly established the ability of ethanol to initiate chain growth
(e.g., 5,9) and that rupture of the C–C bond did not occur to a mea-
surable extent. However, that work could not provide any measure
of the lability of the C–H bond during synthesis.
Our previous study added D2O as a tracer to elucidate the
FT mechanism for a cobalt catalyst and these show that there
is no measurable H/D exchange in alkanes and the exchange in
alkenes appears to be limited to the vinyl positions [13]. In the
present study, partially deuterated ethanol (i.e., CD3CD2OH and
CD3CH2OH) was used as a tracer to investigate the iron-catalyzed
Fischer–Tropsch mechanism at different synthesis conditions. An
attempt has been made to identify the surface intermediates
formed from the added D-ethanol through an analysis of FT hydro-
carbon products.
vinyl (CH2 CH (ad)) and vinylidene (CH2
C (ad)). In the pioneer-
ing work of Emmett and co-workers [5,6], 14C labeled ethanol (1.5%)
was added to the syngas feed to an iron catalyst operated at 508 K
and 1 atm, using a plug-flow reactor. Based on the observed con-
stant radioactivity (14C) of the hydrocarbon products, the authors
concluded that ethanol, or an adsorption complex derived from the
ethanol, could act as a chain initiator. Schulz et al. [7] probed the
mechanism of the FT reaction for a cobalt-thoria-kieselguhr cata-
lyst that was operated at 458 K and atmospheric pressure using 14
C
alkenes and found that nearly 30% of the ethene and propene added
were incorporated into the FT products. Co-feeding 14C-labeled
ethene with the syngas over cobalt catalysts demonstrated that
the alkene added at concentrations lower than 2 mol% can act as
an initiator and also acts as a propagator, when the ethene con-
centration exceeds 5 mol% [8]. Tau et al. [9] reported that 1- and
2-propanol incorporate into the products of the FT synthesis and
claimed that the surface species generated from 1- and 2-propanol
are distinct and these initiate chain growth to form different prod-
ucts. Hanlon and Satterfield [10] have reported that the addition
of ethanol or ethene to the feed to an iron catalyst increased the
∗
Corresponding author.
E-mail address: davis@caer.uky.edu (B.H. Davis).
0926-860X/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.apcata.2010.06.039

M.K. Gnanamani et al. / Applied Catalysis A: General 385 (2010) 46–51
47
Table 1
Effect of d5-ethanol co-feeding on olefin–paraffin ratio of hydrocarbon products of
Fe-FT synthesis at 543 K and 0.8 MPa using a syngas (H2/CO: 0.7).
Runs
Sample no.
Olefin/Paraffin ratio
C2 /C2
C3 /C3
C4 /C4
3
4
1.59
1.62
5.80
5.70
6.50
6.60
Without
C2D5OH
5
6
1.98
1.98
7.01
6.20
7.41
7.02
With
C2D5OH
7
8
9
1.64
1.60
1.62
5.80
5.52
5.50
6.54
6.53
6.56
Without
C2D5OH
2. Experimental
The Fischer–Tropsch runs were conducted using a 1-L CSTR.
The experimental apparatus and materials used are described else-
where [14]. The iron catalyst has a composition of 4.4 wt% Si, 56 wt%
Fe, 0.5 wt% K (i.e., a low ␣ catalyst). The ethanol-D5 (CD3CD2OH) and
ethanol-D3 (CD3CH2OH) were purchased from Cambridge Isotope
Laboratory, Inc., with an isotopic purity of 98%. For each run, 20 g
of crushed catalyst was mixed with 310 g of C30 oil and pretreated
for 24 h using a syngas (H2/CO = 0.7) at 543 K and 0.10 MPa. The ini-
tial conditions for the FT reaction were 0.8 MPa and 543 K using a
syngas feed at 60 slph with a percentage composition of 41.2 H2
and 58.8 CO. About 1.50 mol% of ethanol-D5 or ethanol-D3 (5 g of
labeled ethanol was mixed with 17 g of n-heptane) was introduced
into the reaction system during a period of 300 min using a Mil-
tonRoy mini-piston pump. The D-ethanol was introduced into the
reaction system along with a syngas stream after reaching a steady
CO conversion (∼40%) at a feed rate of 0.926 slph. In order to see the
effect of pressure on the D-distribution of hydrocarbon products, a
separate FT reaction was performed at the same reaction conditions
as mentioned above except that the pressure was set at 2.2 MPa. The
gaseous products were analyzed by an online-gas chromatograph
(GC) during and following D-ethanol addition. Hot (373 K) and cold
(273 K) traps were emptied before the addition of deuteroethanol.
At the end of the addition, both oil and wax products were collected
from the cold and hot traps, respectively, and further separated
from the aqueous phase. The aqueous phase was analyzed by GC
using a Q-Porapack column at 373 K. The detailed experimental
procedures are described elsewhere [14]. Quantitative analysis of
ethanol in an aqueous solution was determined using a standard
calibration procedure and the D-distribution in ethanol was quan-
tified using GC–MS. It is known that fragments of deuteroethanol
could interfere with a parent peak of another D isotopomer of
D-ethanol [15], a correction factor was introduced based on the
GC–MS pattern obtained using standard D5-ethanol or D3-ethanol
compounds. The D-distribution of various hydrocarbons present
in the oil and wax samples were analyzed using GC–MS. Compli-
cations are involved in analyzing the C1–C4 gaseous hydrocarbon
products due to exchange and scrambling of deuterium among FT
products during the FT reaction. The D isotopomers for different
carbon numbers were calculated after correcting 13C for their par-
ent ion peak (m+ ion); the details are described elsewhere [16].
Fig. 1. Effect of D5-ethanol co-feeding on C2 olefin-to-paraffin ratio.
alyst with added ethanol, which causes an increase in the C2, C3,
and C4 olefin/paraffin ratios. Nearly constant CO conversion and a
linear increasing trend in the selectivity to methane were observed
upon D5-ethanol addition and this suggests that the adsorbed D5-
ethanol is not involved in C–C bond breaking. These results are
consistent with the conclusions made from other studies where the
small amounts of added ethanol did not significantly affect the CO
conversion and methane selectivity [10]. A considerable increase in
selectivity to ethylene with D5-ethanol addition (Table 2) suggests
that this is due to the dehydration of D5-ethanol.
It is well-known from ethanol and ethylene co-feeding studies
during FT synthesis using iron catalysts that the rate of incorpo-
ration of ethanol is much higher than the rate of incorporation of
ethylene [10]. In the present system, about 55.4% of the added D5-
ethanol was recovered from the aqueous phase; thus, up to 44.6% of
D5-ethanol could be converted into various hydrocarbons by enter-
ing into the growing hydrocarbon chain. It is hard to predict the
exact amount of D5-ethanol that is incorporated into the grow-
ing hydrocarbon chains as evidenced by the lack of increase of C3
hydrocarbon selectivity (Table 1) with added D5-ethanol. Kokes et
al. [17] applied 14C tracer experiments to define the mechanism for
a doubly promoted iron catalyst used at varying pressures and tem-
peratures and found that the percentage incorporation of ethanol
falls from 18% to 7%, and then from 7% to 2.2% as the pressure
increased from 0.1 to 0.75 to 2.1 MPa, respectively. The authors
compared their results to Kummer’s work and concluded that less
incorporation occurred over a doubly promoted catalyst contain-
ing potassium and magnesium than over a singly promoted catalyst
without potassium. In our system, a singly promoted Fe catalyst at
low potassium content and at low pressure (0.8 MPa) may favor
more D5-ethanol incorporation.
Table 2
Effect of d5-ethanol co-feeding on hydrocarbon product selectivity of Fe-FT synthesis
at 543 K and 0.8 MPa using a syngas (H2/CO: 0.7).
Runs
Sample no. Selectivity, %C
CH4 C2
8.21 4.68 2.93 10.46 1.57 9.08 1.39
3. Results and discussion
C2
C3
C3
C4
C4
3
4
Without
C2D5OH
3.1. d5-Ethanol
7.87 4.45 2.74 9.80 1.72 8.45 1.28
5
6
8.38 5.78 2.92 10.70 1.77 8.40 1.08
8.21 5.63 2.88 10.51 1.69 8.22 1.17
There is a significant increase of olefin-to-paraffin ratio for C2,
C3, and C4 with D5-EtOH addition as shown in Table 1 and Fig. 1;
this suggests that deuteroethanol can compete with CO and hydro-
gen for adsorption sites on the catalyst surface. Kummer et al. [5]
also noted this decrease in the hydrogenation activity of the cat-
With C2D5OH
7
8
9
8.63 4.88 2.97 10.61 1.83 9.14 1.40
8.45 4.68 2.92 10.39 1.88 8.84 1.35
8.40 4.66 2.87 10.28 1.93 8.75 1.34
With out
C2D5OH

48
M.K. Gnanamani et al. / Applied Catalysis A: General 385 (2010) 46–51
Table 3
ysis of oil and wax samples, and is shown in Tables 3 and 4. The
Distribution of deuterium in paraffins for co-feeding D5-ethanol during Fe-FT syn-
thesis at 543 K and 0.8 MPa using a syngas (H2/CO: 0.7).
D/molecule for alkanes was nearly constant (∼1.4) up to C10 and
may increase slightly with increasing carbon number after C10
.
Paraffin
mol%
D/molecule*
However, the D/molecule ratio for 1-olefin exhibits an increasing
trend starting from 1.2 for C5 and increases to 1.5 for C12. The D0
isotopomer for each compound comes from two sources: a normal
FT synthesis during the tracer period and from the accumulated
product. The predominance of the D1 isotopomer in all hydrocar-
bons suggests that considerable H/D exchange occurred during the
interaction of D5-ethanol on the catalyst surface and this would
create a pool of deuterium which may eventually be introduced
into hydrocarbon products.
Carbon no. D0
D1
D2
7.21
7.15
10.07
11.33
12.92
12.41
16.53
D3
D4
D5
5
6
8
67.62
22.93
28.3
32.72
34.44
35.91
34.8
1.64
1.39
3.06
5.02
4.01
4.75
5.15
0.60
0.68
0.47
0.25
1.45
1.88
2.21
0.00
0.00
0.00
0.00
0.00
0.00
0.00
1.38
1.31
1.38
1.43
1.45
1.49
1.60
62.31
53.68
48.77
45.04
45.46
44.8
9
10
11
12
31.18
*
The D/molecule were calculated without including d0.
If a C2 species derived from D5-ethanol acts as an initiator, the
D/molecule must be a constant for all compounds. It appears that
ethanol competes with the reactants (CO and H2) as well as prod-
ucts (H2O, alkenes, etc.) for adsorption sites. The amount of ethanol
added is small compared to the added CO; thus, the added ethanol
does not impact the total conversion of CO to a measurable extent.
The added ethanol does impact the secondary hydrogenation reac-
tion so that the alkene/alkane ratio increases and the amount of
methane decreases.
The data obtained in this study display a constant D/molecule
for alkanes indicating that the added D5-ethanol could initiate
the carbon chain growth, but only after extensive H/D exchange
on the catalyst surface. The D-content per molecule for the 1-
olefin increased linearly from 1.2 to 1.5 with increasing carbon
number in the C5–C12 range, suggesting that 1-olefin readsorp-
tion that was followed by some H/D exchange could cause
this observation. It is known that iron catalysts readily effect
hydrogenation–dehydrogenation of alcohols and so it is com-
mon to relate the participation of alcohol in H/D exchange
by reverse dehydrogenation of alcohol. Scheme 1 shows that
more than one step likely exists where H/D exchange could
take place at the ␣ and ␤ positions of adsorbed d5-ethanol.
Thus, the intermediate compound (1), (2) will readily exchange
deuterium with adsorbed hydrogen on the catalyst surface lead-
Table 4
Distribution of deuterium in 1-olefins from co-feeding D5-ethanol during Fe-FT
synthesis at 543 K and 0.8 MPa using a syngas (H2/CO: 0.7).
1-olefin
mol%
D/molecule*
Carbon no. D0
D1
D2
4.13
6.16
9.74
11.42
11.87
15.90
17.65
D3
D4
D5
5
6
8
69.60
25.36
28.35
33.06
34.87
35.15
34.81
33.61
0.74
1.13
2.00
2.48
3.59
4.97
4.30
0.11
0.20
0.14
0.70
0.15
2.14
0.68
0.00
0.00
0.00
0.00
0.00
0.00
0.00
1.19
1.25
1.35
1.37
1.38
1.55
1.50
64.08
56.80
50.36
49.25
42.06
43.76
9
10
11
12
*
The D/molecule were calculated without including d0.
The detailed relative amounts of isotopomers in alkanes and 1-
olefins produced in the D5-ethanol tracer run are summarized in
Tables 3 and 4, respectively. The important characteristics of these
data are: (1) more than 60% of the alkanes and 1-olefins are D0
isotopomer, and D1 is the most abundant D isotopomer of all the
FT products formed irrespective of carbon number; (2) the relative
mol% of D1, D2, D3 and D4 isotopomer increased with increasing
carbon number; and (3) the total D content per molecule for both
alkanes and 1-olefins was between 1.2 and 1.6 in the carbon range
ing to
a spectrum of D-ethanols (D0, D1, D2, D3, D4 and
D5). As shown in Fig. 2, the recovered ethanol contains D0,
D1, D2, D3, and D5 isotopomers. After the rapid elimination-
addition reactions, the ethanol that participates in the chain
growth is actually the partially deuterated ethanol (C2H5DO,
C2H4D2O, C2H3D3O), and not the added D5-ethanol. GC–MS anal-
ysis of ethanol after the FT run shows the presence of the
following D isotopomers (mole, %): C2H6O = 22.8, C2H5DO = 14.3,
C5–C12
.
If the added D5-ethanol is only involved in initiating the hydro-
carbon chain one would expect the maximum number of deuterium
atoms that an alkane or 1-olefin could have is 5; however, our
experimental results show that there is no evidence for the pres-
ence of D5 isotopomer in any of the hydrocarbons. The D/molecule
for each hydrocarbon can be calculated based on the GC–MS anal-
Scheme 1. Hydrogenation–dehydrogenation of d5-ethanol on Fe catalysts.

M.K. Gnanamani et al. / Applied Catalysis A: General 385 (2010) 46–51
49
Fig. 3. Effect of methane production rate during D5-ethanol addition at 2.2 MPa and
543 K using a syngas ratio of H2/CO = 0.7.
Fig. 2. The D-distribution of ethanol in aqueous phase during Fe-FT synthesis D5-
ethanol co-feeding at 543 K and 0.8 MPa using a syngas (H2/CO: 0.7).
Table 5
Effect of D5-ethanol co-feeding on olefin–paraffin ratio of products from D5-ethanol
co-feeding during Fe-FT synthesis at 543 K and 2.2 MPa using a syngas (H2/CO: 0.7).
C2H4D2O = 9.8, C2H3D3O = 7.1, C2H2D4O = 0, C2D5HO = 42.9. It
reveals that hydrogenation–dehydrogenation takes place readily
during adsorption of D5-ethanol (CD3CD2OH) and this leads to deu-
terium exchanging with hydrogen of the reaction mixture on the
catalyst surface.
Feed
Olefin-to-paraffin ratio
C2 /C2
C3 /C3
C4 /C4
0.462
0.458
2.614
2.555
3.408
3.415
Without
C2D5OH
It is also possible that intermediate surface species could be
an ethene ␲-complex derived from the dehydration of added
deuteroethanol (Scheme 2). After the metal bonds to one carbon
in ethene to produce (4), it then undergoes a hydrogen atom shift
to produce (5). Finally, the carbon-metal double bond is formed (6)
which can initiate chain growth. The intermediates (4) and (5) can
also undergo H/D exchange with surface adsorbed hydrogen atoms
and this leads to a total deuterium atoms are less than 4 in the C2
unit. Furthermore, the co-feeding experiments [18] have revealed
that ethanol and propanol are to a considerable extent converted to
ethene and ethane or to propene and propane, respectively. It has
been reported [17] that 14C labeled ethene addition also initiates
chain growth but that the molar radioactivity in the products is sig-
nificantly smaller than observed in ethanol co-feeding runs. These
results suggest that it is more difficult to form the chemisorbed
surface structure (3) in Scheme 2 in the case of ethene addition
than for ethene produced from added ethanol. Another possibil-
ity is that the ethanol incorporation mechanism may follow more
than one pathway. At present, we do not have any direct experi-
mental evidence to show the specific C2 intermediate derived from
the added deuteroethanol which is incorporated into the growing
hydrocarbon chain. However, the fact that under the same reac-
tion conditions ethanol initiates chain growth about 50-100 times
greater than ethene suggests that most of the H/D exchange occurs
With C2D5OH
0.802
5.232
4.689
0.408
0.400
2.267
2.181
3.257
3.189
Without
C2D5OH
prior to loss of the oxygen–carbon bond.
3.2. Effect of pressure on d5-ethanol incorporation
Quantitatively, the picture for D5-ethanol co-feeding at high
pressure (2.2 MPa) is not significantly different from the results
obtained at lower pressure (0.8 MPa). As shown in Fig. 3 and Table 5
there is a depression of methane formation and an increase of
olefin-to-paraffin ratio of C2, C3 and C4 hydrocarbons at higher
pressure which could suggest that some hydrogenation sites were
occupied by the added deuteroethanol. Hanlon and Satterfield [10]
also observed a similar decrease in methane selectivity with the
addition of ethanol during FT synthesis using a fused magnetite
catalyst containing potassium at 521 K and 0.78 MPa. Also, Snel
and Espinoza [8] observed a substantial decrease in the rate of
methane formation that depended on the co-feed concentration
as well as the catalyst used. In contrast, a large number of reports
indicate little or no influence of ethanol co-feeding on the rate of
methane formation. The authors observing suppression of methane
Scheme 2. The possible pathway for the formation of C2 species.

50
M.K. Gnanamani et al. / Applied Catalysis A: General 385 (2010) 46–51
Table 8
Distribution of deuterium in 1-olefins while co-feeding D5-ethanol during Fe-FT
synthesis at 543 K and 2.2 MPa using a syngas (H2/CO: 0.7).
1-olefin
mol%
D0
D/molecule*
Carbon no.
D1
D2
D3
D4
D5
5
6
8
72.27
70.64
66.44
64.99
69.32
68.02
72.56
22.25
22.92
24.39
22.74
20.00
20.94
17.03
4.44
5.10
6.80
9.29
6.69
7.88
7.23
0.80
1.05
2.15
2.61
2.60
2.64
2.38
0.20
0.19
0.22
0.36
1.40
0.52
0.61
0.00
0.00
0.00
0.00
0.00
0.00
0.00
1.30
1.30
1.35
1.45
1.52
1.46
1.53
9
10
11
12
*
The D/molecule were calculated without including d0.
more than 80% and 70% of the alkanes and 1-olefins, respectively,
are D0 isotopomer; (2) the relative mol% of D1 isotopomer of the
1-olefin is almost two times higher than the corresponding alkane
and; (3) the D/molecule remains close to a constant value of 1.35
(C5–C12) for alkanes, perhaps showing an increasing trend from
1.30 to 1.5 for 1-olefins (C5–C12). Fig. 5 shows the D-distribution of
ethanol which was recovered from the reactor during D5-ethanol
addition and it reveals that H/D exchange occurs at the catalytic
sites before being involved in the chain growth. As described earlier,
the C2 intermediate species derived from the adsorbed D5-ethanol
are responsible for the observed D-distribution in the hydrocar-
bon products. A noticeable difference can be seen for D0 and D1
distributions between low and high CO conversion. As CO conver-
sion increases the relative mole % of deutero compounds (D1, D2,
and D3) decreased irrespective of carbon number due to the fact
that most of the catalyst surface will be covered with intermedi-
ates which are derived from the normal FT synthesis. This reveals
that the possibility for H/D exchange is lower at high pressure and
could be a dominating process at low CO conversion. Indeed, the
overall D/molecule remains more or less the same at 1.35, which
is slightly lower than the D/molecule (1.45) obtained at low CO
conversion. It is likely that there may be a considerable difference
between the interaction of added D5-ethanol and the catalyst sur-
face that impacts the D-distribution among hydrocarbon products.
Therefore, the probability that D5-alcohol participates in FT synthe-
sis decreases with increasing pressure, in agreement with previous
studies [10]. Similar results were obtained for a CD3CH2OH tracer
run. As shown in Fig. 6, the D/molecule exhibits a constant value of
1.4, indicating that the C2 species derived from the ethanol acted
as an initiator for the FT synthesis.
Fig. 4. Anderson–Schulz–Flory product distribution plot for low ␣-Fe catalyst with
and without D5-ethanol co-feed at 2.2 MPa, 543 K and H2/CO = 0.7.
Table 6
Effect of D5-ethanol co-feeding on hydrocarbon product selectivity during Fe-FT
synthesis at 543 K and 2.2 MPa using a syngas (H2/CO: 0.7).
Runs
Sample no. Selectivity, %C
CH4 C2
C2
C3
C3
C4
C4
3
4
7.76 2.83 6.11 8.79 3.36 7.48 2.20
7.86 2.59 5.65 8.63 3.38 7.50 2.20
Without
C2D5OH
With C2D5OH
5
6.36 3.58 4.46 9.72 1.86 7.56 1.61
6
7
8.06 2.71 6.64 8.79 3.88 7.42 2.28
7.78 2.82 7.03 8.76 4.02 8.03 2.52
Without
C2D5OH
formation attributed it to scavenging of surface hydrogen and C1
intermediates. In our study, the rate of methane formation remains
unaffected by D5-ethanol addition at low pressure (0.8 MPa) but a
noticeable change is observed at higher pressure (2.2 MPa) and at
high CO conversion suggesting that added D5-ethanol forms sur-
face intermediates which could scavenge a surface hydrogen to
form ethane but this was not found to be the case. Also, it cannot
be attributed to the reaction of adsorbed ethanol with adsorbed
methane precursors, since such a reaction should yield a higher C3
formation. Fig. 4 shows only a slight variation of the ASF distribu-
tion with added D5-ethanol. As seen from Table 6, the selectivity
to propene increased to about the same extent as the propane
decreases.
The relative amounts of isotopomers in alkanes and 1-olefins
produced in the D5-ethanol tracer run at high pressure (2.2 MPa)
are summarized in Tables 7 and 8, respectively. Compared to
the data obtained at low pressure (0.8 MPa), D5-ethanol co-fed at
high pressure (2.2 MPa) exhibited a distinguished D-distribution in
hydrocarbon products (alkanes and 1-olefins). The following data
were obtained at higher CO conversion (high pressure, 2.2 MPa): (1)
Table 7
Distribution of deuterium in paraffins while co-feeding D5-ethanol during Fe-FT
synthesis at 543 K and 2.2 MPa using a syngas (H2/CO: 0.7).
Paraffin
mol%
D0
D/molecule*
Carbon no.
D1
D2
D3
D4
D5
5
6
8
79.41
81.64
84.73
88.65
91.44
88.64
90.45
16.11
13.86
10.85
8.17
6.29
8.08
3.33
3.25
3.15
2.56
1.76
2.73
2.69
0.83
0.73
1.05
0.44
0.32
0.47
0.76
0.16
0.24
0.23
0.18
0.20
0.06
0.16
0.00
0.00
0.00
0.00
0.00
0.00
0.00
1.30
1.36
1.39
1.35
1.35
1.35
1.51
9
10
11
12
5.89
Fig. 5. The D-distribution of ethanol in aqueous phase during Fe-FT synthesis with
D5-ethanol co-feeding at 543 K, 2.2 MPa and a syngas of H2/CO = 0.7.
*
The D/molecule were calculated without including d0.

M.K. Gnanamani et al. / Applied Catalysis A: General 385 (2010) 46–51
51
topomers than for D3-ethanol. The tracer runs (high and low CO
conversions) yielded a nearly constant D/molecule for alkanes and
1-olefins suggesting that the C2 species derived from the partially
deuterated ethanol acted as an initiator for the FT synthesis. None of
the compounds were found to contain the D5 isotopomer, indicat-
ing extensive H/D exchange of adsorbed D5-ethanol with hydrogen
on the surface of the catalyst before initiating a growing hydrocar-
bon chain.
References
[1] H. Schulz, M. Claeys, Appl. Catal. A: Gen. 186 (1999) 71–90.
[2] M.E. Dry, J.C. Hoogendoorn, Catal. Rev. Sci. Eng. 23 (1981) 265–278.
[3] M.L. Turner, H.C. Long, A. Shenton, P.K. Byers, P.M. Maitlis, Chem. -Eur. J. 1 (1995)
549–556.
[4] M.L. Turner, N. Marsih, B.E. Mann, R. Quyoum, H.C. Long, P.M. Maitlis, J. Am.
Chem. Soc. 124 (2002) 10456–10472.
[5] J.T. Kummer, H.H. Podgurski, W.B. Spencer, P.H. Emmett, J. Am. Chem. Soc. 73
(2) (1951) 564–569.
[6] J.T. Kummer, P.H. Emmett, J. Am. Chem. Soc. 75 (1953) 5177–5182.
[7] H. Schulz, B.R. Rao, M. Elstner, Erdol Kohle 22 (1970) 651–655.
[8] R. Snel, R.L. Espinoza, J. Mol. Catal. 43 (1987) 237–247.
[9] L.-M. Tau, H.A. Dabbagh, B.H. Davis, Energy Fuels 4 (1990) 94–99.
[10] R.T. Hanlon, C.N. Satterfield, Energy Fuels 2 (2) (1988) 196–204.
[11] Y.B. Kryukov, A.N. Bashkirov, F.A. Fridman, L.G. Liberov, R.M. Smirnova, A.A.
Pegov, Neftekhimiya 5 (1965) 62–67.
Fig. 6. Variation of total deuterium content with carbon number for paraffins
formed during Fe-FT synthesis with D3-ethanol co-feeding at 543 K, 2.2 MPa and
using a syngas of H2/CO = 0.7.
[12] M.A. Brundage, S.S.C. Chuang, J. Catal. 174 (1998) 164–176.
[13] B. Shi, J. Li, R.A. Keogh, B.H. Davis, Appl. Catal. A: Gen. 229 (2002) 283–289.
[14] L.M. Tau, R. Robinson, R.D. Ross, B.H. Davis, J. Catal. 105 (1987) 335–341.
[15] L.E. Friedman, J. Turkevich, J. Am. Chem. Soc. 74 (1952) 1666–1668.
[16] B. Shi, B.H. Davis, J. Catal. 157 (1995) 626–630;
B. Shi, B.H. Davis, J. Chromatogr. A 731 (1996) 351–354.
[17] R.J. Kokes, W.K. Hall, P.H. Emmett, J. Am. Chem. Soc. 79 (1957)
2989–2996.
4. Conclusions
Partially deuterated D5- and D3-ethanol were used as a tracer to
study the Fe-FT synthesis mechanism. A significant amount of deu-
terium was found in the hydrocarbon products (alkane, 1-olefin)
irrespective of carbon number. Addition of D5-ethanol during the
FT synthesis produces more D1, D2, D3 and to some extent D4 iso-
[18] R. Snel, A.L. Espinoza, J. Mol. Catal. 54 (1989) 213–223.