Catalytic deoxygenation of bio-based 3-hydroxydecanoic acid to secondary alcohols and alkanes

Article abstract of DOI:10.1039/d0gc00691b

This work comprises the selective deoxygenation of bio-derivable 3-hydroxydecanoic acid to either linear alkanes or secondary alcohols in aqueous phase and H₂-atmosphere over supported metal catalysts. Among the screened catalysts, Ru-based systems were identified to be most active. By tailoring the catalyst, the product selectivity could be directed to either secondary alcohols or linear alkanes. In the absence of a Br?nsted acidic additive, 2-nonanol and 3-decanol were accessible with a yield of 79% and 6% respectively, both of which can be used in food and perfume industries as flavoring agents and fragrances. To produce alkanes, we successfully synthesized a bifunctional Ru/HZSM-5 catalyst. The acidic zeolite support facilitated the dehydration of the intermediary formed alcohols to alkenes, while the following hydrogenation occurred at the Ru centers. Thus, full 3-hydroxydecanoic acid deoxygenation to nonane and decane, which are both well-established as diesel and jet fuels, was achieved with up to 72% and 12% yield, respectively.

Full text of DOI:10.1039/d0gc00691b

Green Chemistry
View Article Online
View Journal
PAPER
Catalytic deoxygenation of bio-based
3-hydroxydecanoic acid to secondary alcohols
Cite this: DOI: 10.1039/d0gc00691b
and alkanes†
Joel B. Mensah,
Jens Artz
Adrian H. Hergesell,
*
Sebastian Brosch,
Christiane Golchert,
and Regina Palkovits
This work comprises the selective deoxygenation of bio-derivable 3-hydroxydecanoic acid to either linear
alkanes or secondary alcohols in aqueous phase and H2-atmosphere over supported metal catalysts.
Among the screened catalysts, Ru-based systems were identified to be most active. By tailoring the cata-
lyst, the product selectivity could be directed to either secondary alcohols or linear alkanes. In the
absence of a Brønsted acidic additive, 2-nonanol and 3-decanol were accessible with a yield of 79% and
6% respectively, both of which can be used in food and perfume industries as flavoring agents and
fragrances. To produce alkanes, we successfully synthesized a bifunctional Ru/HZSM-5 catalyst. The
acidic zeolite support facilitated the dehydration of the intermediary formed alcohols to alkenes, while
the following hydrogenation occurred at the Ru centers. Thus, full 3-hydroxydecanoic acid deoxygena-
tion to nonane and decane, which are both well-established as diesel and jet fuels, was achieved with up
to 72% and 12% yield, respectively.
Received 26th February 2020,
Accepted 6th May 2020
DOI: 10.1039/d0gc00691b
rsc.li/greenchem
One major challenge in bio-refinery is that the feedstocks
often comprise heterogeneous mixtures with respect to the
Introduction
Owing to global warming and the finite availability of fossil functionalities represented in their components and a high
feedstocks, biomass has evolved as a promising alternative for oxygen content.21 This critical difference to conventional oil-
the supply of energy and commodities. While renewable ener- refinery can be addressed by employing multidisciplinary
gies harvested from wind, solar, and water may supply green approaches, e.g. a combination of biotechnology, engineering,
energy, biomass additionally gives access to renewable carbon and chemistry.21,22 Lately, the development of flexible and
for the production of fuels and fine chemicals. This character- robust biotechnological methods has made great strides.23–25
istic provides an important basis for the implementation of Therefore, future trends should be directed to the further
closed carbon cycles and the crucial reduction of the global optimization of biotechnological approaches, involving
a
CO2-footprint.1–4
direct integration of subsequent thermal and/or chemo-cata-
In recent years, different bio-refinery concepts have been lytic treatment to unlock the dormant synergies: accessibility
researched, involving numerous technologies and biomass of novel biomass feedstocks, increased yields and selectivities,
feedstocks as well as bio-derived substrates for their following and an enlarged portfolio of bio-based products.23,26,27
conversion into bio-based platform molecules for chemical
Rhamnolipids (RHLs) and their precursor molecules 3-(3-
industries.5–11 Lignocellulosic biomass presents an ideal can- hydroxyalkanoyloxy)alkanoates (HAAs) present such an innova-
didate for the production of chemicals and fuels, as respective tive class of bio-based raw materials. They are produced
agricultural and agro-industrial wastes are widely available and through the microbial fermentation of glucose or xylose.28–34
inexpensive.12–17 These waste streams are rich in sugars, allow- These bio-surfactants commonly consist of two β-hydroxy fatty
ing an enzymatic or chemo-catalytic conversion into building acid-subunits which are connected through an ester bond and
blocks or precursors for carbon-based value chains.18–20
in case of RHLs also possess one or two rhamnose moieties
linked through a β-glycosidic bond.28 This recurrent hydroxy
acid backbone (with an adjustable chain length of typically 8,
10, 12 or up to 14 carbon atoms each) is an interesting starting
material for bio-based value chains.32–34 RHLs and HAAs may
find direct application in bioremediation, or as surfactants
in cosmetics and detergents.33,35 The corresponding free
Institute of Technical and Macromolecular Chemistry, Worringerweg 2,
52074 Aachen, Germany. E-mail: palkovits@itmc.rwth-aachen.de
†Electronic supplementary information (ESI) available. See DOI: 10.1039/
d0gc00691b
This journal is © The Royal Society of Chemistry 2020
Green Chem.

View Article Online
Paper
Green Chemistry
β-hydroxy fatty acids, (readily obtained after hydrolysis of the may enlarge this product tree of bio-based commodities by
ester bond in HAAs) are potential new substrates for the pro- additionally giving direct access to the respective unsaturated
duction of fatty alcohols, fatty acid esters or higher alkanes fatty acids and 1,3-fatty diols, besides secondary alcohols.
with applications as fuels and lubricants.36,37
Herein, we present the first selective chemo-catalytic deoxy-
However, the considerably high costs for the production of genation of 3-HDA to secondary alcohols and linear alkanes in
bio-surfactants still limit their wide-spread use. Thus, exploit- aqueous phase over a set of supported metal catalysts.
ing the full potential of this material class in future bio-refi-
We have selected 3-HDA as a model compound as it is pre-
neries necessitates further optimization of the biotechnologi- dominant in the β-hydroxy acid backbone structures of HAAs
cal production besides the development of efficient methods and RHLs and readily obtained from the same.58 Moreover,
for their downstream processing.35 One principal method for water was chosen as a solvent as the microbial production of
the production of respective fatty alcohols from fatty acids HAAs and the subsequent hydrolysis both occur in aqueous
would be using stoichiometric or even excess amounts of a media. Thus, an aqueous reaction system is advantageous
metal hydride (e.g. LiAlH4 or NaBH4) as commonly applied in from an ecological and economical perspective as it avoids an
the reduction of ketones, carboxylic acids, and carboxylic acid additional operational step, thereby saving resources, energy,
derivatives. However, this leads to poor overall E-factors38 and costs.
resulting from the accumulation of at least stoichiometric
amounts of waste. In this regard, hydrogenation combined
with heterogeneous catalysis offers a more environmentally
Experimental
friendly alternative and has been the subject of many previous
Chemicals and materials
studies.39–48 This approach provides a remedy as it enables an
3-Hydroxydecanoic acid (≥95%) was obtained from 007
Chemicals, CDCl3 (99.9%) and D2O (99.9%) were purchased
from Deutero, ethanol (≥99.9%), HCl (≥37 wt%), heptane
(≥99.2%), and NaOH (≥99.5%) were supplied by ChemSolute,
decanoic acid (98%), ethyl heptanoate (99%) and pyridine
(99.8%, anhydrous) were acquired from Sigma Aldrich,
1-decanol (99%) and decyl decanoate (≥95.0%) were obtained
from TCI, and CeO2 from Saint Gobain.
The following catalysts (5 wt% metal loading) were commer-
cially available: Pd/C (ChemPur), Rh/C (Alfa Aesar), Ru/C
(Heraeus), Pt/C (Sigma Aldrich). All supported metal oxide cat-
alysts (5 wt% Ru) with exception of Ru/Al2O3 (Alfa Aesar) were
synthesized via wet impregnation of commercial supports:
HZSM-5 (Si : Al = 15, Clariant and Si : Al = 40, Alfa Aesar),
Silicalite-1 (SiO2 (MFI-form), Süd-Chemie) besides TiO2
(anatase, ST61120) and ZrO2 (tetragonal, SZ61152) which both
were kindly provided by Saint Gobain. For the wet impreg-
nation procedure, 1.0 g of dry support material was carefully
added to a stirred solution of the corresponding amount of
ruthenium(III) chloride hydrate (99.9% purity, 39–42% Ru,
abcr) in 20 ml of ethanol. After thorough stirring for 3 h, the
solvent was removed at 40 °C and reduced pressure. In the fol-
lowing, the catalyst was dried in a vacuum oven (60 °C) over-
night. Lastly, the material was dried at 120 °C for 2 h under
N2-flow and then reduced at 350 °C (5 °C min−1 heating ramp)
under H2-flow for 3 h.
easy separation of the products (typically in liquid phase) from
the reductant (mostly H2-gas) and the solid catalyst addition-
ally to preferably lower E-factors. Furthermore, highly selective
and versatile transformations are possible: the selective hydro-
genation of aromatic carboxylic acids and respective deriva-
tives was recently showcased by the group of Shimizu where no
dearomatization was observed over the synthesized Re/TiO2
catalyst.49,50 Moreover, the group of Breit reported a bimetallic
Pd–Re/C catalyst for the reduction of various carboxylic acids
to alcohols.51 The same work also demonstrated a selective
one-pot synthesis of alkanes starting from fatty acids by
careful adjustment of the reaction conditions.
However, so far reports on the heterogeneously catalyzed
conversion of β-hydroxy acids remain scarce and have been
mostly limited to shorter chains of three to five carbon
atoms.52–55 Recently, we have investigated the electrochemical
conversion of 3-hydroxydecanoic acid (3-HDA) into a drop-in
bio-diesel.56 This present work aims to unlock the potential of
β-hydroxy fatty acids as novel biogenic substrates via chemo-
catalysis (Scheme 1). The known routes for the valorization of
alkylic fatty acids to date mainly yield primary fatty alcohols,
fatty acid esters, alkenes, and alkanes.57 β-Hydroxy fatty acids
Autoclave experiments
In a typical reaction, 50 mg of catalyst, 250 mg of 3-HDA (or an
equimolar amount of a different substrate), and 5 ml of water
were added to a PTFE inlet and placed into an autoclave. The
autoclave was gently flushed with H2 three times to replace the
remaining air. Afterwards, an H2 pressure of 100 bar was
applied and the autoclave was placed into a preheated heating
mantle. The reaction was run at 200 °C for the desired amount
of time until it was terminated by cooling of the autoclave.
Scheme 1 Microbial HAA production and subsequent chemo-catalytic
conversion of 3-HDA into fuels and fine chemicals.
Green Chem.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Green Chemistry
Paper
Reaction work-up
an MCT detector (2 cm−1 resolution). The analyzed sample was
pressed into a two-layered wafer using a pellet press (30–40
bar, for 5 min). The wafer was 20 mm in diameter, containing
15 mg of sample and 70 mg of KBr. First, the background
signal of the empty evacuated sample cell was measured at
The aqueous phase was transferred into a glass vial and the
autoclave inlet was carefully rinsed with 5 ml heptane.
Subsequently, the aqueous and organic phases were com-
bined. Then 0.1 ml 1 M NaOH was added to transfer any
residual 3-HDA into its deprotonated form and hence make it
water-soluble to transfer it into the aqueous phase. For the
extraction, the vial was shaken vigorously and left at rest for
the phases to separate (sometimes the addition of KCl was
necessary for proper phase separation). Afterwards, both layers
were divided into two different vials and separated from the
solid catalyst using a syringe filter (0.45 μm pore size, non-
polar). The organic phase was analyzed via gas chromato-
graphy (GC) to determine the product yields.
80 °C. The sample was then heated to 250 °C (at 5 °C min−1
)
for 1 h under vacuum to remove any residues. Afterwards, the
background signal of the sample cell was recorded at 80 °C. In
the following, pyridine adsorption was induced by exposing
the evacuated sample to pyridine vapor for 2 min. The removal
of physisorbed pyridine was achieved by treating the sample at
200 °C (at 5 °C min−1) for 1 h. Afterwards, the measurement of
chemisorbed pyridine was performed at 80 °C.
Gas chromatography
Results and discussion
15 mg of ethyl heptanoate were added as an internal standard
to ca. 1 g of the sample prior to the analysis performed on a
Thermo Scientific Chromatograph Trace GC Ultra equipped
with a CP-WAX-52 (60 m × 250 μm × 0.25 μm) column. N2 was
used as carrier gas with a flow rate of ca. 1.5 ml min−1 along
with an inlet temperature of 250 °C and a split flow of 50 ml
min−1. A temperature program of 70 °C (5 min) and 200 °C
(15 min) with a heating ramp of 8 °C min−1 was used for the
analysis of a 1 µl sample.
Catalyst screening for secondary alcohol production
As different metals supported on carbon were previously
reported to catalyze the reduction of carboxylic acids and their
derivatives, the same served as the basis for the herein investi-
gated deoxygenation of 3-HDA.57 The data depicted in Fig. 1
clearly shows that the addition of a catalyst enhances the reac-
tion outcome as the blank test in the absence of any catalyst
showed a considerably low yield of 4% nonane and 3%
decanoic acid.
The addition of Pd/C resulted in an increase of the total
yield to 14% and a change in selectivity leading to alcohols
only (7% 3-decanol, 4% 1-decanol, 3% 2-nonanol). Pt/C and
Rh/C both showed a very similar reaction outcome in terms of
total yields (27% and 30%) with 2-nonanol observed as the
main product. While the application of Pt/C resulted in a
slightly higher yield for 2-nonanol (25%) and the formation of
low amounts of decanoic acid (2%), Rh/C gave rise to 18%
2-nonanol besides alkane by-products (8% nonane and 1%
decane) and small amounts of 3-decanol (3%). Amongst the
Gas chromatography–mass spectrometry
The samples were prepared analogously to regular GC
samples. Measurements were carried out on
a Varian
CP-3800 gas chromatograph equipped with a CP-WAX-52
column (60 m × 250 μm × 0.25 μm) and a Varian 1200L
Quadrupole MS/MS fitted with an electron ionization (EI)
source. The volume of injected sample was 1 μl (diluted in
methanol 1 : 10) while the GC signal was recorded with a flame
ionization detector (FID). The EI voltage was set at 70 eV with a
transfer capillary temperature of 250 °C and an ion source
temperature of 250 °C.
Temperature programmed desorption of ammonia
Temperature programmed desorption of ammonia (NH3-TPD)
measurements were performed using a Micromeritics ASAP
2920 device and a method previously reported by Liu et al.59
Approximately 90 mg of sample were dried in situ under a He
flow of 10 ml min−1 using a temperature ramp of 5 °C min−1
up to 600 °C prior to the analysis. The sample was then cooled
to 100 °C, thereafter NH3 with a flow of 25.3 ml min−1 was fed
to the sample. The sample was then heated to 600 °C (5 °C
min−1 heating rate) to start the NH3-desorption. The amount
of desorbed NH3 was quantified with a thermal conductivity
detector (TCD). A cold trap was employed to prevent any water
from passing through the TCD.
IR-spectroscopy of adsorbed pyridine
IR-measurements of adsorbed pyridine were performed on a
Fig. 1 Metal screening for the conversion of 3-HDA. Conditions: 25 mg
Vertex 70 spectrometer equipped with a transmission cell and cat. (5 wt% metal), 250 mg 3-HDA, 5 ml water, 100 bar H2, 200 °C, 1 h.
This journal is © The Royal Society of Chemistry 2020
Green Chem.

View Article Online
Paper
Green Chemistry
Fig. 2 Screening of different supports for Ru in the conversion of
3-HDA. Conditions: 50 mg cat. (5 wt%), 250 mg 3-HDA, 5 ml water, 100
bar H2, 200 °C, 1 h.
product distributions, with higher yields of C9-compounds
(compared to C10) and 2-nonanol as the main product and
small amounts of alkanes (4–9%).
Commonly, the metal dispersion and the support acidity
play a decisive role in the support influence. The aforemen-
tioned study by our group also comprised a further characteriz-
ation of Ru supported on alumina, titania, and zirconia.62
Slight differences were found for the metal dispersion, while
more pronounced variations were revealed for the support
acidity. However, no consistent correlation with the herein
observed activities was deducible, which necessitates further
studies. Nevertheless, the conducted screening established
Ru/C as the most active catalyst serving as a promising starting
point for the selective production of secondary alcohols, i.e.
79% of 2-nonanol.
Scheme 2 Proposed reaction mechanism (a) for the deoxygenation of
3-HDA into alcohols over Ru/C and (b) further Brønsted acid-catalyzed
alcohol dehydration followed by hydrogenation of the resulting alkenes
over Ru.
investigated catalysts, Ru/C possessed the best performance
leading to the selective formation of secondary alcohols i.e. a
yield of 44% 2-nonanol and 8% 3-decanol. Moreover, small
amounts of alkanes (3%) were observed as by-products.
While Pt, Pd, and Rh are well known in the literature for
their hydrogenation properties, all three showed significantly
lower alcohol yields compared to Ru.60,61 In a recent study by
our group, a similar activity trend was found for the reduction
Proposed mechanism
of N-(2-hydroxyethyl)succinimide over Ru, Rh, Pt, and Pd sup- Based on the works by Chen et al.63 and Vorotnikov et al.,67
a
ported on carbon.62 The significantly higher activity of Ru was mechanism for the conversion of 3-HDA over Ru/C into
attributed to its capability to simultaneously cleave hydrogen 2-nonanol and 3-decanol is proposed in Scheme 2a, aiming to
and activate the carbonyl function, which is also required in explain the formation of the mainly formed secondary alco-
the conversion of 3-HDA (see proposed mechanism in hols. At first, the hydroxy fatty acid 1 is reduced into the
Scheme 2a).63,64 In this regard, Pt and Pd have been reported hydroxy acyl species 2. Starting from the activated acyl 2, one
to demand the use of significantly higher temperatures in the pathway (top) leads to the activation and reduction of the car-
conversion of fatty acids, with high product yields only bonyl function (3 and 4), which results in the formation of the
obtained at temperatures above 300 °C.65,66 Hence, Ru was diol 5. Subsequently, after further reduction and hydrodeoxy-
chosen as active metal component for further optimization genation of the latter, 3-decanol is finally obtained as one of
and the examination of different metal oxide supports (Fig. 2).
Ru/CeO2 showed a significantly different reaction outcome
the targeted secondary alcohols.
The second path (bottom) comprises the activation of the
compared to the remaining supported Ru-catalysts with β-C–H bond of the surface species 2, forming the acyl inter-
decanoic acid as the main product (16%), a considerably lower mediates 8 and 9. In the following, Ru catalyzes the decarbony-
2-nonanol yield of 9% and no alkane formation. This was not lation of the terminal C–C-bond to produce CO besides the
the case for Ru/TiO2, Ru/Al2O3, Ru/ZrO2, and Ru/C, for which respective Cn−1 alcohol 10. The hydrogenation of both
no decanoic acid formation was verified. Moreover, similar adsorbed species finally leads to 2-nonanol – the main product
outcomes were observed for these catalysts with regard to their of the Ru/C-catalyzed hydrogenation of 3-HDA – besides
Green Chem.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Green Chemistry
Paper
methane. The latter, together with CO and CO2, was also veri-
fied in a qualitative analysis of the gas phase (Scheme S2,
ESI†). The further reduction of 2-nonanol and 3-decanol over
Ru/C into nonane and decane was only observed in minor
amounts and is thus, in the interest of clarity, not shown in
Scheme 2a.
Selective alkane production
The second part of this work targeted the selective full deoxy-
genation of 3-HDA to alkanes e.g. for the application as renew-
able fuels. Since both nonane and decane were found in small
quantities in the final product mixture of the deoxygenation
over Ru/C, a prolongation of the reaction time was tested.
However, no significant enhancement of the alkane yields was
verified after increasing the reaction time from 1 to 3 h
(Fig. S1, ESI†). As the deoxygenation of 3-HDA over Ru/C
seems to run into a dead-end at the alcohol stage, we explored
a second strategy to shift the product selectivity from partially
deoxygenated alcohols towards the corresponding fully deoxy-
genated alkanes.
Fig. 3 Shifting the selectivity in 3-HDA conversion from alcohols to
alkanes via addition of Brønsted acids. Conditions: 50 mg 5 wt% Ru/C
(2 wt% H2SO4 or 50 mg HZSM-5 as co-cat.) or 50 mg
5 wt% Ru/HZSM-5, 250 mg 3-HDA, 5 ml H2O, 100 bar H2, 200 °C, 1 h.
The contemplated tool for this aim was a one-pot dehydra-
tion-hydrogenation-sequence comprising the Brønsted acid-
catalyzed dehydration of the present alcohols to alkenes and
the subsequent reduction to alkanes over the Ru-species
(Scheme 2b). The general applicability was first tested with the
2-nonanol intermediate as a substrate. Whitmore et al. have
shown the Brønsted acid-catalyzed transformation of 2-octanol
to octene isomers.68 To check for the formation of alkenes via
dehydration, 2-nonanol was reacted over Brønsted-acidic
HZSM-5 (15) under N2-atmosphere. In the absence of H2, the
expected nonene isomers could be evidenced through GC-MS,
proving a successful dehydration step. This was further corro-
borated as the conversion of 2-nonanol over Ru/C under H2-
pressure only yielded very low amounts of nonane (4%), while
application of Ru/C and HZSM-5 (15) under H2-atmosphere
lead to the selective production of nonane (Scheme S1, ESI†).
This highlights the possibility to overcome the bottleneck of
alcohol formation for an effective full deoxygenation of 3-HDA
to alkanes through a catalyzed dehydration step as envisioned
in Scheme 2b. For a quick assessment of the feasibility in a
one-pot synthesis of alkanes starting from 3-HDA, a soluble
Brønsted acid was added to Ru/C as a co-catalyst. Indeed, the
positive impact of the addition of 2 wt% H2SO4 was clearly
visible which manifested in the complete shift of the product
selectivity to alkanes (Fig. 3). Noteworthy, the total product
elaborate separation and recovery of the catalyst from the reac-
tion solution. This issue could be overcome by replacing
soluble H2SO4 with a solid Brønsted acid, enabling an easier
separation e.g. through filtration. In this regard, the addition
of HZSM-5 (15) as a co-catalyst proved to be effective, again
leading to a selectivity shift towards fully deoxygenated
nonane and decane (52% and 26% yield) (Fig. 3).
Furthermore, the considerably lower total yield for the H2SO4–
Ru/C tandem could be enhanced by using HZSM-5 (15)–Ru/C,
which resulted in a less pronounced decrease of the initial pH
value and thus an expected higher solubility of the substrate
(Table S2, ESI†).
Finally, these findings and similar previously reported con-
cepts lead us to combine the dehydration properties of
Brønsted acidic HZSM-5 (15) and the hydrogenation capability
of Ru particles in a single catalyst.69,70 This was successfully
implemented in the preparation and testing of Ru/HZSM-5
(15) for the targeted selective production of alkanes. The appli-
cation of the bifunctional catalyst led to a high total alkane
yield of 84% (72% nonane and 12% decane) after only 1 h and
thus presents a suitable method for the full deoxygenation of
3-HDA.
Influence of support acidity
yield (56%) decreased significantly with H2SO4 as an additive, Lastly, the effect of the support acidity on the reaction
which might be rooted in the hindered substrate solubility outcome was investigated in more detail by varying the Si : Al
caused by the presence of a Brønsted acid. A decreased solubi- ratio. For this purpose, two further catalysts with MFI-struc-
lity and hampered conversion of 3-HDA is expected as the tured supports were synthesized, namely Ru/HZSM-5 (40) and
initial pH value of 2.4 was significantly lower compared to a Ru/Silicalite (only Si, i.e. non-acidic ZSM-5). The respective
pH value of 4.1 for Ru/C without the additive (a more detailed supports were probed via NH3-TPD and pyridine-IR-spec-
description is provided in the ESI and Table S2†).
troscopy to characterize the acidity and correlate it to the cata-
After the general feasibility of the dehydration-hydrogen- lytic performance.
ation-sequence was successfully demonstrated, further optim-
The number and relative strength of the acid sites were
ization of the reaction system was conducted. One general assessed with the help of NH3-TPD (Fig. 4). All three materials
drawback of many homogeneous systems is an often more exhibit two broader desorption regimes in the low-temperature
This journal is © The Royal Society of Chemistry 2020
Green Chem.

View Article Online
Paper
Green Chemistry
Fig. 4 NH3-TPD measurement of MFI-structured silicalite, HZSM-5 (15)
and (40).
Fig. 5 Alkane yield over concentration of total acid sites in the support.
Conditions: 50 mg catalyst (5 wt% Ru), 250 mg 3-HDA, 5 ml water, 100
bar H2, 200 °C, 1 h.
region (170–280 °C) and high-temperature region (above
370 °C), indicating the presence of at least two different types
of acid sites.71 The peaks in the low-temperature regime are
assigned to NH3-desorption from weaker acid sites, while
stronger acid sites are found in the high-temperature regime
corresponding to the desorption of strongly bound NH3.72 To
differentiate between the contributions of the weak and strong
acid centers, the TPD profiles were deconvoluted applying a
Gaussian fit (Fig. S2, ESI†). Thus, after integration of the
deconvoluted peaks, the number of acid sites was accessible
for each acid type.
In accordance with the literature, an increase in acid site
concentration is observed with decreasing Si : Al ratio i.e.
higher Al content (Table 1).73 The incorporation of Al has a
dramatic effect on the resulting concentration of acid centers
as it increases from 0.062 mmol g−1 for the solely Si-based sili-
calite support by one order of magnitude to 0.512 and
0.703 mmol g−1 for Si : Al ratios of 40 and 15 respectively.
Moreover, a shift of all peak maxima to higher temperatures
with more Al was verified as a general trend – pointing at a
higher acidity. Overall, the TPD-analyses indicate the expected
trend for the support acidity: HZSM-5 (15) > HZSM-5 (40) >
Silicalite; that is a decreasing acidity with lower Al amounts.
Finally, the corresponding bifunctional Ru-catalysts were
applied to the deoxygenation of 3-HDA for the production of
alkanes (Fig. S3†). The testing revealed an influence of the
Si : Al ratio, and thus acidity, on the total alkane yield: a con-
siderably lower alkane yield of 8% was obtained for the less
acidic Ru/Silicalite, while alkane yields of 55% and 84% are
obtained over the more acidic HZSM-5 (40) and (15) supports.
To further investigate the relationship between the structure of
the support and alkane yield, the latter was examined as a
function of total acid sites (determined via NH3-TPD). Fig. 5
depicts a clear impact of the support acidity on the reaction
outcome, suggesting a direct correlation between the total con-
centration of acid sites and the final alkane yield.
IR-spectroscopy of adsorbed pyridine enabled a further
qualitative (differentiation between Brønsted and Lewis acid
sites) and semi-quantitative (relative amount of respective acid
sites) analysis of the support acidity (Fig. S4, ESI†). Due to the
low total acidity of the silicalite support, only the
HZSM-5 materials were examined for which both a co-existence
of Lewis and Brønsted acidic sites was verified. The band at
around 1455 cm−1 corresponds to the interaction of pyridine
with Brønsted acid centers, Lewis acid sites are evidenced by
the signal at around 1545 cm−1, while the band at 1490 cm−1
relates to both Brønsted and Lewis acidic sites interacting with
pyridine.74
In the following, we employed an empirical correlation pre-
viously reported by Zhu et al. to combine the quantitative NH3-
TPD result (total acidity atotal in mmol g−1) and the estimated
Lewis- to Brønsted acid ratio RL/B determined through pyri-
dine-IR spectroscopy (see ESI† for a detailed description).75
The analysis enabled further insights into the distribution of
Brønsted and Lewis acidity in both HZSM-5 materials. The
data presented in Table 1 confirm the expected trend of an
increasing number of Brønsted acid sites with more Al, as the
Brønsted acidity in zeolites originates from the replacement of
Si with a formal valency of four by Al with a respective value of
three. The resulting negative excess charge is balanced by
protons (attached to the bridging O-atoms between Si and Al)
leading to a chemically stable framework and the formation of
Table 1 Characterization of support acidity via NH3-TPD and pyridine-
IR
Acid sites [mmol g−1
]
Catalyst
support
Si : Al
(mol mol−1
)
Weak Strong Lewis Brønsted Total
a
Silicalite
∞
0.03
0.354 0.158
0.454 0.249
0.032
n.d.b n.d.b
0.093 0.419
0.033 0.670
0.062
0.512
0.703
HZSM-5 (40) 40
HZSM-5 (15) 15
a Does not contain any Al but only Si. b Not determined due to low
total acidity.
Green Chem.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Green Chemistry
Paper
Scheme 3 Proposed reaction network for the conversion of 3-HDA to alcohols and alkanes in the presence of a Ru-catalysts and/with Brønsted
acidity.
Brønsted acidic properties.76 Thus, in accordance with pre- path for decanoic acid and its derivatives was examined more
vious literature reports, the catalytic activity in the crucial closely with experiments using the different intermediates as
alcohol dehydration step is mainly attributed to Brønsted acid starting materials (Scheme S3, ESI†).
sites.75,77
The removal of the hydroxy-group in 3-HDA leads to
decanoic acid which is then further reduced to 1-decanol
over intermediary formed decanal. Low levels of 1-decanol
were observed to convert into decane or react with remain-
ing decanoic acid to yield decyl decanoate. However, a fast
hydrolysis of the ester into the alcohol and the acid is
expected, due to the aqueous environment and elevated
temperature. Additionally, the ester can be transformed
into 1-decanol and 1-nonene via decarbonylation.57
The latter alkene is believed to be mainly formed through
decarbonylation of decanal; however, 1-nonene subsequently
is readily hydrogenated to nonane over Ru (Scheme S3,
ESI†).
Proposed reaction network
Lastly, a reaction network for the Ru-catalyzed hydroprocessing
of 3-HDA is proposed (Scheme 3). Following the postulated
mechanism (Scheme 2a) and previous literature reports,63,67
the conversion of the acid mainly occurs through the initial
reduction to the corresponding carbonyl compound, here
3-hydroxydecanal. This intermediate plays a crucial role for the
number of carbon atoms in the final product. Due to the short
lifetime of the intermediate and its fast conversion, it could
not be isolated. However, the occurrence of aldehyde species
over metal catalysts in carboxylic acid reduction was previously
reported in the literature.57,63
The formation of the herein obtained C9-products orig-
inates from the cleavage of the terminal C–C-bond in 3-hydro-
xydecanal leading to CO and 2-nonanol. The former is further
Conclusions
reduced to methane, while the latter is dehydrated to To the best of our knowledge, this is the first report on the
2-nonene, which is promoted by Brønsted acid sites. Finally, chemo-catalytic deoxygenation of a β-hydroxy fatty acid to alco-
the alkene converts into nonane via hydrogenation over the hols and alkanes. Herein, we provide an efficient green proto-
metal site. The preservation of the entire C10-structure is col for the deoxygenation of 3-HDA, the dominant recurrent
achieved through another path involving the reduction of motive in biogenic HAAs and RHLs, over supported Ru-cata-
3-hydroxydecanal into 1,3-decanediol and the subsequent lysts in aqueous phase. The use of water as solvent saves
reduction of the terminal hydroxy-moiety into 3-decanol. additional unit operations and energy requirements as the
Lastly, the alcohol is transformed into decane via the dehydra- microbial production of the HAA precursor is performed in
tion-hydrogenation-sequence (Scheme 2b).
aqueous medium, while the application of H2 as reducing
Furthermore, trace amounts of 1-decanol, decanoic acid, agent avoids the accumulation of stoichiometric waste
and decyl decanoate were also present. The proposed reaction amounts e.g. as with LiAlH4.
This journal is © The Royal Society of Chemistry 2020
Green Chem.

View Article Online
Paper
The use of Ru/C gave rise to the selective and fast pro-
Green Chemistry
Oliveira Vigier and F. Jérôme, ChemSusChem, 2018, 11,
2673–2676.
6 B. Kamm, Pure Appl. Chem., 2014, 86, 821–831.
7 A. A. Kiss, J.-P. Lange, B. Schuur, D. W. F. Brilman,
A. G. J. van der Ham and S. R. A. Kersten, Biomass
Bioenergy, 2016, 95, 296–309.
8 K. Li and Y. Sun, Chem. – Eur. J., 2018, 24, 18258–18270.
9 B. You, N. Jiang, X. Liu and Y. Sun, Angew. Chem., Int. Ed.,
2016, 55, 9913–9917.
duction of secondary alcohols (79% 2-nonanol and 6%
3-decanol). A study of the support acidity by NH3-TPD and pyr-
idine-IR measurements revealed its crucial role in obtaining
high alkane yields. The incorporation of Brønsted acidity into
the catalyst support resulted in a bifunctional Ru/HZSM-5 (15)
catalyst which proved to be highly effective for tailoring the
product selectivity to the corresponding alkanes (72% nonane
and 12% decane yield).
Finally, this work proposes a reaction network and mecha- 10 A. J. Ragauskas, G. T. Beckham, M. J. Biddy, R. Chandra,
nism based on the observed intermediates and their testing as
substrate. This provides valuable first insights into the conver-
sion of functionalized fatty acids and the development of tai-
lored catalysts for their selective downstream processing.
F. Chen, M. F. Davis, B. H. Davison, R. A. Dixon, P. Gilna,
M. Keller, P. Langan, A. K. Naskar, J. N. Saddler,
T. J. Tschaplinski, G. A. Tuskan and C. E. Wyman, Science,
2014, 344, 1246843–1246843.
Overall, this study demonstrates the potential of bio-accessible 11 A. Hommes, H. J. Heeres and J. Yue, ChemCatChem, 2019,
β-hydroxy fatty acids for bio-based value chains, thereby 11, 4671–4708.
expanding the scope of renewable substrates for fuel and fine 12 S. I. Mussatto and J. A. Teixeira, in Current Research,
chemical markets.
Technology and Education Topics in Applied Microbiology and
Microbial Biotechnology, Formatex, 2010, pp. 897–907.
13 J. B. Binder and R. T. Raines, J. Am. Chem. Soc., 2009, 131,
1979–1985.
14 I. Delidovich, K. Leonhard and R. Palkovits, Energy Environ.
Sci., 2014, 7, 2803–2830.
Conflicts of interest
The authors declare no conflicts of interest.
15 J.-P. Lange, R. Price, P. M. Ayoub, J. Louis, L. Petrus,
L. Clarke and H. Gosselink, Angew. Chem., Int. Ed., 2010,
49, 4479–4483.
16 T. Shahzadi, S. Mehmood, M. Irshad, Z. Anwar, A. Afroz,
N. Zeeshan, U. Rashid and K. Sughra, Adv. Biosci.
Biotechnol., 2014, 5, 246–251, 6 pp.
Acknowledgements
We gratefully acknowledge funding by the Federal Ministry of
Food and Agriculture (Bundesministerium für Ernährung und
Landwirtschaft) under project number 22403415 (project 17 J. Zakzeski, P. C. A. Bruijnincx, A. L. Jongerius and
executing body: Fachagentur Nachwachsende Rohstoffe). B. M. Weckhuysen, Chem. Rev., 2010, 110, 3552–3599.
J.B.M. and R.P. thank the European Regional Development 18 M. E. Himmel, S.-Y. Ding, D. K. Johnson, W. S. Adney,
Fund (ERDF) and the Federal State of North Rhine-Westphalia,
Germany for financial support under the operational program
M. R. Nimlos, J. W. Brady and T. D. Foust, Science, 2007,
315, 804–807.
‘Regional Competitiveness and Employment’ project 19 M. Arshadi and A. Sellstedt, in Introduction to Chemicals
‘Sustainable Chemical Synthesis’. We thank Prof. Pieter
Bruijnincx for the opportunity to measure TPD, Dr Wirawan
from Biomass, John Wiley & Sons, Chichester, UK, 2008,
pp. 143–178.
Ciptonugroho for performing TPD-analyses, and Elke Biener, 20 Z. Anwar, M. Gulfraz and M. Irshad, J. Radiat. Res. Appl.
Hannelore Eschmann, Heike Fickers-Boltz, and Wolfgang Sci., 2014, 7, 163–173.
Falter for performing GC(-MS) measurements. We are grateful 21 E. de Jong and G. Jungmeier, Ind. Biorefin. White
to Matthias Fischer, Dr F. Joschka Holzhäuser, Dr M. Ghith Biotechnol., 2015, 3–33.
Al-Shaal, Marie M. Döring, and the engine room 38B 320 for 22 O. Y. Abdelaziz, K. Ravi, F. Mittermeier, S. Meier,
their support to the success of this project.
A. Riisager, G. Lidén and C. P. Hulteberg, ACS Sustainable
Chem. Eng., 2019, 7, 11640–11652.
23 J. Becker and C. Wittmann, Angew. Chem., Int. Ed., 2015,
54, 3328–3350.
24 R. Hatti-Kaul, in Industrial Biotechnology, Wiley-VCH Verlag
References
1 C. Chatterjee, F. Pong and A. Sen, Green Chem., 2015, 17,
40–71.
GmbH
279–321.
& Co. KGaA, Weinheim, Germany, 2010, pp.
2 P. Gallezot, Chem. Soc. Rev., 2012, 41, 1538.
3 E. I. Gürbüz, S. G. Wettstein and J. A. Dumesic,
ChemSusChem, 2012, 5, 383–387.
4 V. Quaschning, Understanding Renewable Energy Systems,
Routledge, London, 2016.
25 R. A. Sheldon and D. Brady, ChemSusChem, 2019, 12, 2859–
2881.
26 F. Harnisch and C. Urban, Angew. Chem., Int. Ed., 2018, 57,
10016–10023.
27 C. Plass, A. Hinzmann, M. Terhorst, W. Brauer, K. Oike,
H. Yavuzer, Y. Asano, A. J. Vorholt, T. Betke and H. Gröger,
ACS Catal., 2019, 9, 5198–5203.
5 P. N. Amaniampong, J. Clément, D. Gigmes, C. Ortiz
Mellet, J. M. García Fernández, Y. Blériot, G. Chatel, K. De
Green Chem.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Green Chemistry
Paper
28 A. Wittgens, B. Santiago-Schuebel, M. Henkel, T. Tiso, 49 T. Toyao, S. M. A. Hakim Siddiki, A. S. Touchy, W. Onodera,
L. M. Blank, R. Hausmann, D. Hofmann, S. Wilhelm,
K.-E. Jaeger and F. Rosenau, Appl. Microbiol. Biotechnol.,
2018, 102, 1229–1239.
K. Kon, Y. Morita, T. Kamachi, K. Yoshizawa and
K. Shimizu, Chem. – Eur. J., 2017, 23, 1001–1006.
50 T. Toyao, S. M. A. Hakim Siddiki, Y. Morita, T. Kamachi,
A. S. Touchy, W. Onodera, K. Kon, S. Furukawa, H. Ariga,
K. Asakura, K. Yoshizawa and K. Shimizu, Chem. – Eur. J.,
2017, 23, 14848–14859.
29 A. Wittgens, T. Tiso, T. T. Arndt, P. Wenk, J. Hemmerich,
C. Müller, F. Rosenau and L. M. Blank, Microb. Cell Fact.,
2011, 10, 80.
30 A. Wittgens, F. Kovacic, M. M. Müller, M. Gerlitzki, 51 J. Ullrich and B. Breit, ACS Catal., 2018, 8, 785–789.
B. Santiago-Schübel, D. Hofmann, T. Tiso, L. M. Blank, 52 D. Dugar and B. Neltner, US20160145227A1, 2016.
M. Henkel, R. Hausmann, C. Syldatk, S. Wilhelm and 53 L. Craciun, G. P. Benn, J. Dewing, G. W. Schriver, W. J. Peer,
F. Rosenau, Appl. Microbiol. Biotechnol., 2017, 101, 2865–
2878.
31 L. Blank, T. Tiso and A. M. Germer, WO2017006252A1,
2017.
32 A. Germer, T. Tiso, C. Müller, B. Behrens, C. Vosse,
K. Scholz, M. Froning, H. Hayen and L. M. Blank, Appl.
Environ. Microbiol., 2020, 86, DOI: 10.1128/AEM.02317-19.
33 J. D. Desai and I. M. Banat, Microbiol. Mol. Biol. Rev., 1997,
61, 47–64.
34 S. Lang and D. Wullbrandt, Appl. Microbiol. Biotechnol.,
1999, 51, 22–32.
35 T. Tiso, S. Thies, M. Müller, L. Tsvetanova, L. Carraresi,
S. Bröring, K.-E. Jaeger and L. M. Blank, in Consequences of
B. Siebenhaar and U. Siegrist, US7538247B2, 2009.
54 M. A. Lilga, J. F. White, J. E. Holladay, A. H. Zacher,
D. S. Muzatko and R. J. Orth, US7687661B2, 2010.
55 A. Zacher, J. Holladay, M. Lilga, J. White, D. Muzatko,
R. Orth, P. Tsobanakis, X. Meng and T. Abraham,
US20070219390A1, 2007.
56 J. Meyers, J. B. Mensah, F. J. Holzhäuser, A. Omari,
C. C. Blesken, T. Tiso, S. Palkovits, L. M. Blank,
S. Pischinger and R. Palkovits, Energy Environ. Sci., 2019,
12, 2406–2411.
57 R. W. Gosselink, S. A. W. Hollak, S. W. Chang, J. Van
Haveren, K. P. De Jong, J. H. Bitter and D. S. Van Es,
ChemSusChem, 2013, 6, 1576–1594.
Microbial Interactions with Hydrocarbons, Oils, and Lipids: 58 S. Miao, S. S. Dashtbozorg, N. V. Callow and L. K. Ju,
Production of Fuels and Chemicals, ed. S. Y. Lee, Springer
International Publishing, Cham, 2017, pp. 587–622.
36 D. J. Anneken, S. Both, R. Christoph, G. Fieg,
U. Steinberner and A. Westfechtel, in Ullmann’s
J. Agric. Food Chem., 2015, 63, 3367–3376.
59 F. Liu, C. Okolie, R. M. Ravenelle, J. C. Crittenden,
C. Sievers, P. C. A. Bruijnincx and B. M. Weckhuysen, Appl.
Catal., A, 2018, 551, 13–22.
Encyclopedia of Industrial Chemistry, Wiley-VCH, Weinheim, 60 G. Webb, in Comprehensive Chemical Kinetics, ed.
Germany, 2006.
C. H. Bamford and C. F. H. Tipper, Elsevier, 1978, vol. 20,
37 H. R. Beller, T. S. Lee and L. Katz, Nat. Prod. Rep., 2015, 32,
1508–1526.
pp. 1–121.
61 U. K. Singh and M. A. Vannice, Appl. Catal., A, 2001, 213, 1–
38 R. A. Sheldon, Green Chem., 2007, 9, 1273–1283.
24.
39 T. Toyao, K. W. Ting, S. M. A. Hakim Siddiki, A. S. Touchy, 62 M. O. Haus, Y. Louven and R. Palkovits, Green Chem., 2019,
W. Onodera, Z. Maeno, H. Ariga-Miwa, Y. Kanda, 21, 6268–6276.
K. Asakura and K. Shimizu, Catal. Sci. Technol., 2019, 9, 63 L. Chen, Y. Zhu, H. Zheng, C. Zhang, B. Zhang and Y. Li,
5413–5424. J. Mol. Catal. A: Chem., 2011, 351, 217–227.
40 K. Kon, W. Onodera, S. Takakusagi and K. I. Shimizu, 64 J. Shangguan, M. V. Olarte and Y.-H. (Cathy) Chin, J. Catal.,
Catal. Sci. Technol., 2014, 4, 3705–3712. 2016, 340, 107–121.
41 Y. Takeda, Y. Nakagawa and K. Tomishige, Catal. Sci. 65 J. Fu, C. Yang, J. Wu, J. Zhuang, Z. Hou and X. Lu, Fuel,
Technol., 2012, 2, 2221–2223. 2015, 139, 678–683.
42 Y. Takeda, M. Tamura, Y. Nakagawa, K. Okumura and 66 I. Kubičková, M. Snåre, K. Eränen, P. Mäki-Arvela and
K. Tomishige, Catal. Sci. Technol., 2016, 6, 5668–5683. D. Y. Murzin, Catal. Today, 2005, 106, 197–200.
43 J.-W. Jiang, C.-C. Tu, C.-H. Chen and Y.-C. Lin, 67 V. Vorotnikov, T. R. Eaton, A. E. Settle, K. Orton,
ChemCatChem, 2018, 10, 5449–5458.
44 X. Gao, D. Tong, H. Zhong, B. Jin, F. Jin and H. Zhang, RSC
Adv., 2016, 6, 27623–27626.
E. C. Wegener, C. Yang, J. T. Miller, G. T. Beckham and
D. R. Vardon, ACS Catal., 2019, 9, 11350–11359.
68 F. C. Whitmore and J. M. Herndon, J. Am. Chem. Soc., 1933,
55, 3428–3430.
45 B. Rozmysłowicz, A. Kirilin, A. Aho, H. Manyar,
C. Hardacre, J. Wärnå, T. Salmi and D. Y. Murzin, J. Catal., 69 W. Luo, W. Cao, P. C. A. Bruijnincx, L. Lin, A. Wang and
2015, 328, 197–207. T. Zhang, Green Chem., 2019, 21, 3744–3768.
46 Z. Luo, Q. Bing, J. Kong, J. Liu and C. Zhao, Catal. Sci. 70 K. B. Baharudin, M. Arumugam, J. Hunns, A. F. Lee,
Technol., 2018, 8, 1322–1332.
47 C. Delhomme, D. Weuster-Botz and F. E. Kühn, Green
Chem., 2009, 11, 13–26.
48 W. Rachmady and M. A. Vannice, J. Catal., 2002, 207,
317–330.
E. Mayes, Y. H. Taufiq-Yap, K. Wilson and D. Derawi, Catal.
Sci. Technol., 2019, 9, 6673–6680.
71 L. Rodríguez-González, F. Hermes, M. Bertmer,
E. Rodríguez-Castellón, A. Jiménez-López and U. Simon,
Appl. Catal., A, 2007, 328, 174–182.
This journal is © The Royal Society of Chemistry 2020
Green Chem.

View Article Online
Paper
Green Chemistry
72 N. Katada, H. Igi and J.-H. Kim, J. Phys. Chem. B, 1997, 101, 75 H. Zhu, A. Ramanathan, J.-F. Wu and B. Subramaniam,
5969–5977. ACS Catal., 2018, 8, 4848–4859.
73 J. A. Rabo and G. J. Gajda, in Guidelines for Mastering 76 R. A. van Santen, G. J. Kramer and W. P. J. H. Jacobs,
Properties of Molecular Sieves, ed. D. Barthomeuf,
E. G. Derouane and W. Holderlich, Springer, 1990,
pp. 273–297.
in Elementary Reaction Steps in Heterogeneous Catalysis,
ed. R. W. Joyner and R. A. van Santen, Springer Netherlands,
Dordrecht, 1993, pp. 113–131.
74 F. Jin and Y. Li, Catal. Today, 2009, 145, 101– 77 C. D. Baertsch, K. T. Komala, Y.-H. Chua and E. Iglesia,
107.
J. Catal., 2002, 205, 44–57.
Green Chem.
This journal is © The Royal Society of Chemistry 2020