D. Gupta, B. Saha
CatalysisCommunications100(2017)206–209
Scheme 1. Synthesis of 5-aminolevulinic acid from me-
O
thyllevulinate.
O
O
Step1
Br2
O
O
Br
2
Step
N3
O
O
O
Step 3
H2 Ru-CNS
H2N
OH
3
NaN3
O
O
,
O
3.2. Catalysis
polar solvent, CS2. In the latter case, we observed no ring-opening
product and GVL is formed with 100% selectivity at 35% conversion
(entry 9). A reaction in non-polar solvent, hexane, yields GVL with high
selectivity (entry 10).
A reaction using reused catalyst yields comparable GVL in hexane.
Higher activity of Ru-CNS over Ru/C in EtOH could be due to the un-
ique structure of the carbon sphere of the former catalyst as reported in
prior report [29] as well as porous and oxygen functionalities of the
carbon sphere as observed from the FTIR and HRTEM analysis. Prior
reports on the structural effect of supports for Ru and Pt catalyzed
hydrogenation of LA to GVL, [34] including the beneficial effect of
immobilized metal loaded CNS catalysts are consistent with our ob-
In addition to the high activity of the Ru-CNS catalyst for LA hy-
drogenation, we found that this catalyst is also effective for Ff and HMF
hydrogenation to the corresponding furfural alcohols. The hydrogena-
tion of Ff and HMF (2 mmol) with 20 mg Ru-CNS at 4 bar H2 and 150 °C
enables quantitative conversion of substrates with 94% and 95% iso-
lated yields of mono hydroxymethyl furan (MHMF) and bis-hydro-
xymethylfuran (BHMF), respectively, for 3 h (Scheme S2). Because of
high use of MHMF and BHMF as monomers for the synthesis of variety
of polymeric materials such as resins and artificial fibers for high-value
uses, the current global annual consumption of furfural alcohols ex-
ceeds 300,000 MT which is projected to reach about 550,000 MT by
2020. Thus, development of effective catalytic hydrogenation routes for
the conversion of furfurals to furfural alcohols has received significant
attention in recent years. In this context, high yield of furfural alcohols
using the investigated Ru-CNS catalyst is significant.
The catalytic effectiveness of Ru-CNS was tested for the hydro-
genation of compound 3 (5-azide methyllevulinate; AML), LA, HMF and
Ff. First, we used AML as a substrate which was derived from methyl
levulinate (ML) via bromination followed by a reaction with NaN3
(Scheme 1). The yields of the brominated intermediate and AML are
72% and 71%, respectively. The resultant AML (1 mmol) was hydro-
genated with Ru-CNS (20 mg) with maximum 76% yield of ALA in THF.
Overall yield of ALA from the present method is higher compared to
those obtained from microbial processes [14,30]. The yield is lower
than the Mascal's method using CMF as a starting substrate [19] in
which high amount of expensive Pd/C catalyst (60 mg for 1.5 mmol
substrate) with high Pd loading (10 wt% Pd) was used for the hydro-
genation step. Thus, the present method achieving 76% ALA yield in the
hydrogenation step using less expensive Ru metal at lower catalyst
loading (20 mg with 5.2 wt% Ru loading for 1 mmol 3) could be ben-
eficial for ALA process economics.
To test the effectiveness of the Ru-CNS catalyst further, the hydro-
genation of LA was conducted at 20 bar H2 and 140 °C (Scheme 2).
Quantitative conversion of LA with 92% isolated yield of GVL and 1,4-
pentanediol (PDO) (82% and 18% selectivity, respectively) was ob-
tained in ethanol (Table 1, entry 1). To understand the effect of solvent
on the conversion of HMF to GVL, we investigated the reactions in polar
and non-polar solvents. These results are shown in Table 1. The cor-
responding 1H NMR spectra are in the Supporting information.
PDO yield is significant in solvent with high polarity, e.g., EtOH,
indicating ring-opening occurs in this solvent [31]. Less polar solvent,
acetonitrile, yields less PDO (entries 2–3). Based on prior reports sug-
gesting homogeneous metal salts (RuCl3, PdCl2) [32,33] formed metal
nanoparticulates under the hydrogenation conditions, which effectively
catalyzed LA hydrogenation, we used homogeneous RuCl3 as a catalyst.
This reaction achieved 56% LA conversion in acetonitrile; however
selectivity to PDO is high (65%; entry 4) when compared with Ru/CNS.
A controlled experiment in THF using GVL as a starting substrate re-
veals a very little conversion (2%) of GVL to PDO for 8 h (entry 5),
which is consistent with the results of LA conversion in THF (entry 6)
and suggest that GVL is stable in this solvent. PDO yield is less in CHCl3
even after 24 h of reaction with majority of products is GVL (83%)
(entry 7). Comparison of our results using Ru/CNS in EtOH with the
literature results using commercial Ru/C in EtOH (entry 8) [21] sug-
gests that the Ru/CNS catalyst, at higher substrate loading (moles of
LA/gRu = 3.84) than the literature Ru loading (moles of LA/
gRu = 2.1), is more effective in terms of GVL yield and selectivity as
well as its ring-opening to PDO. The effect of solvent polarity on ring-
opening and GVL selectivity becomes clear when we used a very less
4. Conclusions
We reported the synthesis of a Ru-CNS catalyst under hydrothermal
condition from glucose, and its morphological and structural char-
acterization using BET, FESEM, PXRD, HRTEM, FTIR and Raman
spectroscopic techniques. The catalytic effectiveness of Ru-CNS is in-
vestigated for the synthesis of 5-aminolevulinic acid from methyllevu-
linate (ML) and for the hydrogenation of 5-hydroxymethylfurfural,
furfural and levulinic acid to the corresponding furan alcohols and γ-
valerolactone. The results show high activity of the catalyst, enabling
76% 5-aminolevulinic acid yield from 5-azidemethyllevulinate and >
90% furfural alcohols and γ-valerolactone from their respective pre-
cursors, which can be derived from biomass. In addition, we elucidated
the effect of solvent polarity on the yield and selectivity of γ-valer-
olactone and its ring-opening product, and compared the results with
commercial Ru/C catalyst.
Acknowledgements
BS acknowledges Department of Science and Technology (DST),
New Delhi for funding this research (award number SR/S1/IC-61B/
2012). BS also acknowledges the financial support from the Catalysis
Center for Energy Innovation, an Energy Frontier Research Center
funded by the U.S. Department of Energy, Office of Science, Office of
Basic Energy Sciences under Award number DE-SC0001004 for the later
Scheme 2. Ru-CNS catalyzed conversion of levulinic acid to γ-valerolactone and 1,4-
pentanediol (PDO).
208