Green catalysis by nanoparticulate catalysts developed for flow processing? Case study of glucose hydrogenation

Article abstract of DOI:10.1039/c4ra14559c

Heterogeneous catalysis, flow chemistry, continuous processing, green solvents, catalyst immobilization and recycling are some of the most relevant, emerging key technologies to achieve green synthesis. However, a quantification of potential effects on a case to case level is required to provide a profound answer, whether they can lead to a superior process compared to the industrial standard. To do so, holistic environmental assessment approaches are very useful tools providing insights and decision support during the process development phase. Herein, novel heterogeneous nanoparticulate ruthenium catalysts immobilized on hyperbranched polystyrene (HPS) and nitrogen-doped carbon nanotubes (NCNT) were investigated with respect to their potential environmental impacts and improvements if utilized in an industrially highly relevant process, namely glucose hydrogenation to sorbitol. The results of a comparative Life Cycle Assessment of the alternative catalytic systems under consideration of RANEY nickel as benchmark catalyst revealed that in particular Ru nanoparticles on porous HPS beads processed under flow-chemistry conditions have the potential to improve the greenness of the overall synthesis, but the concentration of glucose in the reaction mixture is in fact the most influential parameter. This journal is

Full text of DOI:10.1039/c4ra14559c

View Article Online
View Journal
RSC Advances
This article can be cited before page numbers have been issued, to do this please use: D. Gericke, D.
Ott, V. G. Matveeva, E. M. Sulman, A. Aho, D. Murzin, S. Roggan, L. Danilova, V. Hessel, P. Loeb and D.
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Accepted Manuscripts are published online shortly after
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. This Accepted Manuscript will be replaced by the edited,
formatted and paginated article as soon as this is available.
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the Ethical guidelines still
apply. In no event shall the Royal Society of Chemistry be held
responsible for any errors or omissions in this Accepted Manuscript
or any consequences arising from the use of any information it
contains.
www.rsc.org/advances

RSC Advances
Page 1 ofJ10ournal Name
Dynamic Article Links ►
View Article Online
DOI: 10.1039/C4RA14559C
Cite this: DOI: 10.1039/c0xx00000x
www.rsc.org/xxxxxx
ARTICLE TYPE
Green Catalysis by Nanoparticulate Catalysts developed for Flow
Processing? Case Study of Glucose Hydrogenation
Dörthe Gerickea, Denise Otta, Valentina G. Matveevab, Esther Sulmanb, Atte Ahoc, Dmitry Yu. Murzinc, Stefan Roggand, Lyudmila
Danilovae, Volker Hesself, Patrick Loebg, Dana Kralischa*
5
Received (in XXX, XXX) Xth XXXXXXXXX 20XX, Accepted Xth XXXXXXXXX 20XX
DOI: 10.1039/b000000x
Heterogeneous catalysis, flow chemistry, continuous processing, green solvents, catalyst immobilization
and recycling are some of the most relevant, emerging key technologies to achieve green synthesis.
10 However, a quantification of potential effects on a case to case level is required to provide a profound
answer, whether they can lead to a superior process compared to the industrial standard. To do so, holistic
environmental assessment approaches are very useful tools providing insights and decision support
already during the process development phase. Herein, novel heterogeneous nanoparticulate ruthenium
catalysts immobilized on hyperbranched polystyrene (HPS) and nitrogen-doped carbon nanotubes
15 (NCNT) were investigated with respect to their potential environmental impacts and improvements if
utilized in an industrially highly relevant process, namely glucose hydrogenation to sorbitol. The results
of a comparative Life Cycle Assessment of the alternative catalytic systems under consideration of Raney
nickel as benchmark catalyst revealed that in particular Ru nanoparticles on porous HPS beads processed
under flow-chemistry conditions have the potential to improve the greenness of the overall synthesis, but
20 the concentration of glucose in the reaction mixture is in fact the most influential parameter.
process within the last decade: i) the increasing demand for and the
severe price competition on the sorbitol market, and ii)
environmental and health concerns as well as high purification
efforts due to the leaching of Raney nickel at hydrogenation
conditions.
Therefore, the development of continuous hydrogenation processes
has been considered to a greater extend and alternative catalytically
active transition metals have been tested in hydrogenation reactions.
Even so, the improvement of the established process remains
Introduction
Sorbitol is a major specialty product and the polyol with the most
widespread use in nutrition, cosmetics, medicine and industry1 and
25 therefore represents a compound that is encountered in many 50
everyday life utilities. Large amounts are used as alternative
sweetener for diabetics, but it is also applied as humectant and
binding agent in cosmetics or food, for example in toothpaste, where
it not only prevents its dehydration but also contributes to its taste
30 and its non-cariogenic effect.2 Additionally, sorbitol is a starting 55 challenging, as all of the following requirements have to be covered:
i) selectivity and conversion need to be competitive (> 99 %); ii) the
material for the industrial synthesis of vitamin C,3 which in turn has
catalyst has to feature high stability and no leaching behaviour; iii)
various applications. The worldwide production volume exceeded
new ways of producing sorbitol have to not only be advantageous
1,000,000 t/a in 2000.1 Especially the growing demand in China4
concerning handling and health aspects, but have also to be low in
stimulates many sorbitol producers to increase their production. The
costs and environmental impact.
35 current selling price of sorbitol ranges between $ 0.95 and $ 1.40 per 60
kg (70 % syrup).5
Sorbitol is mainly produced via catalytic hydrogenation of D-glucose
– being an abundant and cheap feedstock derived from renewable
These ideas for process improvement have been followed within the
collaborative project POLYCAT. The acronym abbreviates ‘Modern
POLYmer-based CATalysts and microflow conditions as key
elements of innovations in fine chemical syntheses’. The project was
resources.1, 6 In industry, sorbitol production is typically based on
launched by the European Community's Seventh Framework
40 Raney nickel catalyzed slurry batch or sometimes fixed-bed 65
Programme. The aim is to develop an integrated, coherent and
holistic approach utilizing novel heterogeneous nanoparticulate
continuous processes.6, In a first instance, the process mostly
7
complies with the twelve principles of green chemistry8, e.g., high
catalysts in vitamin, pharmaceutical and crop protection syntheses in
atom economy, use of renewable feedstocks, use of water and
conjunction with functions of microreaction technology and ‘green’
production of degradable and innocuous materials. However, the
solvents. Thus, process design activities within POLYCAT are
45 following issues have led to a reconsideration of the manufacturing 70
accompanied by an iterative evaluation approach for most
This journal is © The Royal Society of Chemistry [year]
[journal], [year], [vol], 00–00 | 1

RSC Advances
Page 2 of 10
environmentally benign and economic viable design alternatives.9, 10 using Raney nickel catalysts due to its low price and its high
Decision making on different stages of development was supported selectivity for sorbitol.17 Leaching of nickel into the product causes
View Article Online
by applying Life Cycle Assessment approaches in order to identify 50 a limitation of the number of reuses of the catalyst and requires an
DOI: 10.1039/C4RA14559C
the best alternatives not only from a technical but also from an elaborate clean-up.1
ecological perspective. In this context, also the stage of catalyst
preparation and its influence on the greenness of redesigned or
optimized processes played a crucial role (see also Figure 1).
5
Scheme 1: Reaction equation for the hydrogenation of D-glucose.
55 The advantages of using Ru as alternative catalyst are numerous: Ru
does not dissolve under the reaction conditions of the D-glucose
hydrogenation, and therefore does not cause any issues related to
metal leaching. Thus the catalyst may remain active for a longer time
and purification expenditures will be reduced. In addition, higher
60 selectivities15 and mass-specific activities were reported.1, 6 Due to
the expensive price of Ru, it is usually fixed on a support.
Figure 1: Overall evaluation approach during process design of modern
10
nanoparticulate catalysts for hydrogenation reaction.
Novel Catalytic Systems focused within POLYCAT:
Nanoparticulate Ruthenium based Catalysts
In this context, different alternatives in the development of a new
way of catalytic hydrogenation reaction using heterogeneous
nanoparticulate catalysts in flow-processing have been assessed. As
The present Life Cycle Assessment study is focused on the
development of new, nanoparticulate Ru based catalysts on
nanostructured supports catalyzing the hydrogenation of D-glucose.
A major advantage of nanoparticles in catalysis is their large surface
area that leads to a comparably high number of catalytic reactions at
70 a time and therefore increases the catalytic activity.18-22 A variety of
reasons for the better performances of nanoparticulate catalysts has
been discussed in recent literature: geometry, oxidation state,
chemical and physical environment, interparticle interactions,
nanoparticle – support interactions etc.23
Thus, the most important aspect in the development of
nanoparticulate catalysts is control over nanoparticle size and
morphology as well as prevention of their nanoparticulate
aggregation and leaching during the reaction.23, 24 Besides, low-cost,
reproducible and scalable approaches to nanocatalyst synthesis are
of great importance.25 Among the diverse catalytic nanomaterials,
supported metal nanoparticles are the most widely studied and
extensively employed in many industrial processes.25 The most
prospective supports are nanostructured polymers (e.g., dendrimers,
polymeric networks, etc.), which can serve as an organic
macroligand for growing nanoparticles and are able to provide a
solution for a number of the above-mentioned issues related to the
use of nanoparticles in catalysis.
15 mentioned above, the reaction of D-glucose converted to sorbitol is 65
already conducted in industry at large scale. Thus, it is well suited to
prove the potential greenness of modern nanoparticulate catalysts for
hydrogenation reactions in food- or cosmetic industry.
Background
20 D-Glucose Hydrogenation: State of the Art
Sorbitol can be produced from different sugars, but the synthesis
from D-glucose results in the highest purity of the product and is
therefore preferred. The most common manufacturing method is the 75
catalytic hydrogenation using metallic catalysts. There are also
25 microbiological processes known but these have not yet achieved an
economic importance.5, 11
The catalytic hydrogenation of D-glucose to give sorbitol is depicted
in Scheme 1. The reaction is conducted under elevated pressure and 80
temperature conditions using a hydrogenation catalyst. Early patents
30 starting from 1925 describe the hydrogenation with nickel or Raney
nickel catalysts in a batch process.3, 12-14An amount of 2.5 to 12 %wt
of the nickel catalyst is added to the D-glucose solution (usually 30
to 60 %wt). Hydrogen gas is added at excess pressure (> 100 bar) at 85
temperatures between 100 to 170 °C. In many production sites, the
35 procedure has remained unchanged. However, from the 1980s on,
continuous processes were tested as well.5 The continuous process is
typically conducted under similar process conditions using sublimed
or fixed-bed catalysts. If necessary, the catalyst is removed by 90
precipitation and filtration. Afterwards the product mixture has to be
40 decolorized (for example using activated charcoal) and nickel ions
have to be removed to a level that is suitable for the area of
application by using ion exchangers.
Within this study, hypercrosslinked polystyrene (HPS) and nitrogen-
doped carbon nanotubes (NCNT) were investigated in detail. Both
kinds of materials facilitate the stabilization of the metal
nanoparticles on the surface, influencing e.g. catalyst lifetime and
catalytic activities.11 In the following, both systems as well as
expected advantages justifying their application will be described in
more detail.
More recently the hydrogenation of D-glucose with ruthenium 95
catalysts at industrial scale has been described.15, 16 The process
45 conditions are comparable to the ones described above, but the used
amount of metal catalyst is typically smaller. However, to date
industrial manufacture of sorbitol is still carried out in slurry reactors100
Hyperbranched Polystyrene Supported Catalysts
HPS networks were first synthesized in the early 1980s by
crosslinking linear polystyrene chains with bifunctional crosslinking
agents like monochlorodimethylether or p-xylylen dichloride as
shown in Figure 2a.26 Originally used as sorbent material, HPS is
2
|
Journal Name, [year], [vol], 00–00
This journal is © The Royal Society of Chemistry [year]

Page 3 of 10
RSC Advances
nowadays produced commercially by several companies.27
of NCNT. Figure 3 shows a magnified image of NCNTs. The
Due to its high crosslinking degree, which can exceed 100 %, the 45 utilization of NCNT based ruthenium catalysts within the
structure contains nanosized pores. During impregnation with a hydrogenation of glucose is advantageous comparedVtioewtAhreticulesOenolinfe
DOI: 10.1039/C4RA14559C
metal salt, nanoparticles are formed inside these pores where they Ru/C, which was done before,6 as Ru/NCNT has a higher mechanical
can be kept stable. The unique feature of HPS is the ability to swell stability, porosity and therefore a higher accessible surface area.
in different solvents28 which favours the inclusion of various metal-
5
containing compounds in the HPS matrix. Because of its large
surface area, (usually specific surface area (SSA) > 1,000 m2/g), its
thermal stability (up to 300° C)29 and its well-defined interfaces HPS
10 is very suitable for the application as support material for metal
nanoparticles in catalysis.30, 31 Figure 2b symbolizes the expected
structure of the nanoparticles in the nanopores of an HPS-bead.
X-ray photoelectron spectroscopy analyses revealed that the Ru
nanoparticles have a mixed structure (their surface is partially
15 reduced to Ru(0)). Based on a kinetic study, Doluda et al. proposed
that both Ru(0) and Ru(IV) oxide on the nanoparticle surface are
responsible for catalytic hydrogenation for different reason, i.e.
Ru(0) participates in hydrogen activation while Ru(IV) – in D-
glucose adsorption.32
20
50 Figure 3: Depiction of NCNTs by means of transmission electron
microscopy.
LCA-Modelling
In order to support the process design efforts for modern catalytic
55 pathways for glucose hydrogenation, a comparative environmental
assessment and holistic decision making procedure was followed
consistent with previous studies, see, e.g., references.38-40 The LCA
methodology 41, 42 was chosen to quantify and assess the potential
environmental impacts associated with the alternative catalytic
Figure 2: Scheme of HPS synthesis (a) and nanoparticle formation 60 systems under investigation. During iterative evaluation steps, the
in the nanopores of a HPS-bead (b).
information about environmental impact criteria is used for process
improvement. Key aspects for green process improvement are
determined and then taken into account within the next iterative step
of process design. The evaluation is refined and narrowed down in
parallel to the process development activities. At the end, the
environmental impacts of the new process design alternatives are
quantified and compared with a benchmark process.
The outcome of experimental investigations on D-glucose
hydrogenation served as database for the Life Cycle Inventory (LCI).
As ruthenium is a dependent co-product of the platinum group metal
(PGM), the extraction of the primary metal from the ore as well as
the pyrometallurgical processing and hydrometallurgical
purification in case of the secondary (recycled) metal was modelled
based on LCI data referred to the PGM group. Information on
detailed LCI gathering regarding catalyst preparations can be found
in the appendix. In case of low data availability the LCI analysis
approach introduced by Hischier et al.43 was followed, allowing the
integration of average values regarding transportation of goods45,
energy and water consumption46 and plant infrastructure. In general,
80 process waste was considered to be disposed via hazardous waste
incineration. Waste water was considered to be disposed via waste
water treatment. More information about the data quality can be
CNT / NCNT Supported Catalysts
25 The synthesis of Carbon Nanotubes (CNT) was first described in
1991 by Iijima.33 His investigations indicated outstanding 65
characteristics: The material is stronger than steel, harder than
diamond, its electrical conductivity is higher than that of copper, its
thermal conductivity is higher than that of diamond, to mention but
30 a few.33, 34 Furthermore, carbon nanotubes are an interesting support
material for metallic nanoparticulate catalysts because of their 70
unique electronic and morphological properties. There is a strong
electronic interaction between the CNT support material and the
catalytically active metal, which results in a strong adhesion and a
35 promotion of the catalytic activity. The shape of the tubes allows a
good accessibility for the substrate to the catalytically active metal 75
and good transportation properties.35, 36 Doping the CNT with N-
atoms leads to an additional functionality for the attachment of the
nanoparticles.
40 Both CNT and nitrogen-doped carbon nanotubes (NCNT) are
usually synthesized by chemical vapour deposition (CVD).37
Depending on the desired composition of the nanotubes, the
precursor can contain nitrogen atoms, which leads to the formation
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 3

RSC Advances
Page 4 of 10
found in ESI Table 1 - 4.†
50 %wt sorbitol solutions and a clean-up process (only in case of the
LCI analysis and Life Cycle Impact Assessment (LCIA) were Raney nickel catalyzed process). These processes were defined as
supported by the software tools Umberto® NXT LCA47 and the LCI integral parts of the system boundary of this LCA studyV.iew Article Online
DOI: 10.1039/C4RA14559C
database Ecoinvent v.3.0.44 LCIA was conducted applying the LCIA
methodology “ReCiPe 2008” by Goedkoop et al.48 The following
Benchmark Process
5
35 The today’s practice of industrial catalytic hydrogenation of
D-glucose to sorbitol was used as a benchmark process. It is
described in numerous literature sources. However, none of the
procedures in literature described the entire process in such a way
that it could be used to perform LCA. Therefore, process parameters
40 were selected combining the information given in several literature
sources.3, 5, 12-14, 45 For the calculation of the potential environmental
impacts of the benchmark process the following assumptions
concerning the reaction conditions are made: The reaction is carried
out in batch mode with an amount of catalyst of 2 g per kg D-glucose,
45 the D-glucose concentration is 50 %wt (5.6 M), the temperature
110oC, hydrogen pressure 40 bar, and the reaction time is 3 h. The
clean-up is performed with a polymer-based ion exchanger. The
lifetime is assumed to be three years, 2 M HCl serves as regeneration
agent.
ReCiPe environmental impact potentials were considered: Global
Warming Potential (GWP), Ozone Depletion Potential (ODP),
Terrestrial Acidification Potential (TAP), Freshwater Eutrophication
Potential (FEP), Human Toxicity Potential (HTP), Photochemical
10 Oxidant Formation Potential (POFP), Terrestrial Ecotoxicity
Potential (TETP), Natural Land Transformation Potential (NLTP),
Metal Depletion Potential (MDP) and Fossil Fuel Depletion
Potential (FDP). They were addressed at the midpoint level and
‘hierarchist’ perspective. This perspective is in accordance to the
15 most common policy principles with regard to time-frame and other
issues.49 Most frequently, the 100 year timeframe was used. The
LCIA results presented in the following are based on the functional
unit FU = 1 kg sorbitol.
The LCIA categories GWP and MDP were selected for screening
20 purposes. On the one hand, anthropogenic greenhouse gas emissions
are one of the most urgent global environmental problems and are
directly related to energy intensive process steps, thus particularly 50
awakening stakeholder’s interests. On the other hand, MDP accounts
for the consumption of the finite source of metals that cannot be
25 substituted for future generations. It is an important factor to be
considered when creating green catalytic processes.
The production processes were subdivided in five process steps: The
support preparation (HPS, NCNT), the catalyst preparation
(HPS/Ru, NCNT/Ru, Raney nickel), the D-glucose hydrogenation
30 and the vaporization of the product solution in order to give 1 kg of
Process Alternatives in Focus
The process conditions investigated for batch and continuous,
polymer supported ruthenium catalyzed hydrogenation alternatives
as well as benchmark process are summarized in tables 1 and 2. For
interested readers, more detailed information can be found in chapter
55 “Materials and Methods”.
Figure 4: Potential environmental impacts (scaled to 1 and worst case scenario, respectively) of a) preparation of the polymer-based support material
hyperbranched polystyrene (HPS) and nitrogen-doped carbon nanotubes (NCNT) with regard to GWP and MDP; b) impregnation of the supports to
give the two hydrogenation catalysts HPS/Ru and NCNT/Ru with regard to GWP and MDP; c) glucose hydrogenation (scenario: HPS Batch 1.6) in
dependence of catalyst cycles: 1, 50, ∞, exemplarily shown for GWP; d) Ru catalyst recycling and reuse (0 – 90 %) with regard to MDP (for the
example of scenario HPS Batch 1.6 M referred to benchmark RaNi Batch 5.6 M and RaNi Batch 5.6 M 90 % metal recycling).
4
|
Journal Name, [year], [vol], 00–00
This journal is © The Royal Society of Chemistry [year]

Page 5 of 10
RSC Advances
Table 1: Conditions of the glucose hydrogenation process alternatives under
investigation in laboratory scale used for the LCA.
becomes obvious that for both production processes, GWP is mainly
influenced by the consumption of the feedstocks styrene (in case of
HPS Batch 1.6 M (75 %) and acetonitrile (iDnOcIa:s1e0.1o0f3VN9ie/CwCN4ArRTtiAcl1Be4Oa5nt5cl9inhCe
35 0.1 M (35 %)). Further, both materials are produced by applying a
catalyst. In both cases the impact of these catalysts on GWP is
marginal, however, their supply dominates the category MDP: The
tin salt used for the HPS production26 causes 97 % of the overall
MDP. For the NCNT-production the catalyst made from numerous
40 metal salts35 causes 48 % of the total MDP (figure 4a). In figure 4b
the impacts of the impregnation step, giving the final hydrogenation
catalysts HPS/Ru and NCNT/Ru, is visualized. Obviously, the
largest share of the potential impacts results from the consumption
of the Ru salt used for the precursor solution. This clear result can
45 be observed for numerous other LCIA categories as well and
emphasizes the importance of a thoughtful consumption of this metal
and/or metal recovery from waste streams. To give an example, the
overall MDP would drop by 30 %, if the ruthenium loss during the
preparation of the polymer-based catalytic HPS/Ru (3.0 %wt. Ru)
50 system is decreased to 0 % compared to currently 47 %. For the
NCNT-based system (3.6 %wt.), a reduction from 32 % Ru loss to
0 % results in a decrease of the MDP to 45 % of the initial value.
Figure 4c, which puts the production of the catalyst in the context of
its application in D-glucose hydrogenation and downstream chains,
55 points out the strong dependency of the environmental impact
potentials on the catalyst lifetime. The GWP for example could be
reduced by 53 % when reusing the catalyst endlessly instead of
exchanging it after every run. To give a better impression, a catalyst
reuse number of 50 can be taken as a rough value clearly
60 outweighing environmental burdens resulting from the catalyst
manufacturing.
Batch Processes
Continuous Processes
HPS Batch
1.6 M
NCNT
Batch
0.1 M
3.6
HPS
Conti
0.5 M
2.85
NCNT Conti
0.2 M
Scenario
3.0
2.2
Catalyst
[%wt Ru]
0.56
1.39
85
0.02
0.60
Catalyst/ glucose
ratio [g/kg]
218
2,535
376
TOF
[mol sorbitol/
mol Ru*h]
98
99
96
97
99.3
99
91
98
Conversion [%]
Selectivity [%]
* at T > 100°C
5 Table 2: Theoretical scenarios – industrial application of glucose
hydrogenation process alternatives used for the LCA.
Batch Processes
Continuous
Processes
Benchmark
HPS
Batch
5.6 M
3.0
NCNT
Batch
5.6 M
3.6
HPS
Conti
5.6 M
2.85
NCNT RaNi Batch
Scenario
Conti
5.6 M
2.2
5.6 M
-
Catalyst
[%wt Ru]
0.56
218
1.39
85
0.02
0.53
376
1.98
132
Catalyst/ glucose
ratio [g/kg]
2,535
TOF
[mol sorbitol/
mol cat.*h]
97
95
99
80
98
Conversion [%]
Besides the option of catalyst reuse, the use of secondary materials
instead of primary raw materials is an important key issue for
sustainability. In figure 4d, the influence of the application of
secondary ruthenium or Raney nickel on the resulting MDP of the
overall D-glucose hydrogenation process is exemplarily shown by
means of the scenario HPS Batch 1.6 M compared to the benchmark
RaNi Batch 5.6. The MDP of recycled catalytic material is lower by
two orders of magnitude improving the environmental performance
to a great extent. It was calculated that the share of recycled metal
compared to the overall metal consumption used for the
impregnation of the support has to be around 88 % in order to reduce
the impact of this scenario even down to the level of the RaNi Batch
5.6 process calculated assuming primary metals. It becomes
impressively obvious that long-term stability and reusability of the
newly developed catalysts are essential requirements for the
greenness of the overall process.
* at T > 100°C
First of all, the evaluation studies were focused on lab scale process 65
10 options (batch and continuous mode) using the novel developed
catalyst systems to identify bottlenecks and improvement potentials
to be considered for a later pilot or industrial scale implementation.
The conditions described were chosen according to experimental
investigations (firstly, without any optimization, i.e. base cases, see 70
15 table 1). After these optimization studies, the environmental
performances of future batch and continuous processes were
compared to the industrial benchmark process using Raney nickel
(see table 2). On the one hand, these scenarios indicate a potential
up-scale of the most promising, technically feasible experimental 75
20 scenarios. On the other hand, these options describe potential,
hypothetical ones to investigate parameters guaranteeing
competitiveness to the industrial benchmark process.
View for the Whole: Key Criteria for Green D-Glucose
Hydrogenation Process in Batch
Results and Discussion
80 Next, the investigations were extended by further LCIA impact
categories to guarantee a more holistic decision making towards
green process design options. Thus, fingerprint analyses of the key
environmental impacts of the existing batch procedures for
D-glucose hydrogenation were performed (Figures 5 and 6). Figure
5 shows that in case of the HPS Batch 1.6 M scenario three main
aspects of the reaction dominate the potential environmental impacts.
The environmental impacts in a number of categories are mainly
caused by the supply of the feedstock D-glucose. It causes for
example 87 % of the TETP. In the categories MDP and TAP the main
Environmental Impacts along the Manufacturing Process
25 Chains and Utilization of Nanostructured Catalytic Systems
Before evaluating the performance of the catalysts in the
hydrogenation reaction, the Life Cycle Impact potentials of the
alternative batch catalytic systems themselves were investigated in 85
detail. Figure 4a depicts the results of the investigation of the
30 environmental impacts of the support materials HPS and NCNT by
means of two representative LCIA categories, GWP and MDP. It
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 5

RSC Advances
Page 6 of 10
contribution (60 %) is caused by the supply of the catalyst. The 30 the environmental impact of the catalyst used should be minimized
vaporization step causes significant impacts in several categories (up by means of reduced metal load and/or the application of recycled
View Article Online
DOI: 10.1039/C4RA14559C
to 37 % in the category HTP).
material; iii) high conversion rates and selectivities are also
important for a positive environmental performance, since the
agricultural supply of the raw material D-glucose has a relevant
35 share on the overall environmental impact.
The experimental transformation from batch to continuous
processing applying both catalytic systems enabled the reduction of
the D-glucose / catalyst ratio to 4 % (HPS) and 43 % (NCNT) of the
batch ratio. However, the process changes led to a decrease of the
40 conversion from 95 % to 89 % in case of the NCNT-based catalytic
system. Under laboratory scale conditions, low concentrated
aqueous substrate solutions and primary ruthenium for catalyst
preparation were used (Table 1). In order to estimate the effect of an
increase of the concentration of the feed solution to 50 %wt, likely
45 in industrial practice, some theoretical scenarios were calculated as
well. Additionally, a theoretical recycling scenario was investigated.
A ratio of secondary ruthenium being 3/4 of the overall ruthenium
demand was considered for the example of the NCNT-based
catalytic system (Figure 7).
5
Figure 5: Fingerprint analysis for the base case (HPS Batch 1.6 M) of the
HPS-based glucose hydrogenation.
50
Figure 6: Fingerprint analysis for the NCNT base case (NCNT Batch 0.1 M).
10 Transfer from Batch to Continuous Processing: Prognosis and
Actions to be taken
Taking the above key criteria into account, hypothetical scenarios
were defined (see Table 2) allowing a first glance on their
environmental competitiveness to the industrial benchmark process,
15 see Figure 7.
The LCIA fingerprint of the NCNT Batch scenario (Figure 6)
revealed the following: i) almost all impact categories are dominated 55
by the vaporization step (up to 80 % in the category FEP), which is
a result of the low glucose concentration employed; ii) The share of
20 the impact of the D-glucose hydrogenation is higher than in the HPS-
based process (up to 17 % in the category FDP); iii) The dominating
contribution of the catalyst in the categories MDP and TAP (50 %) 60
is comparable to the HPS-based process.
By means of these insights, the key criteria for process improvements
25 were defined as follows: i) energy intensive processes such as the
heating and vaporizing of the solvent water contribute largely to the
potential environmental impacts in many categories. The 65
development of a catalyst that works at a high concentration, such
as 30 to 50 %wt. as described in references 12-15, 45, is favourable; ii)
Figure 7: Assessment of alternative catalytic process scenarios compared to
benchmark regarding GWP and MDP.
In both experimentally investigated continuous process alternatives
the amount of catalyst per kg of D-glucose was reduced, as an
exchange of the catalyst was not required as often as in the batch
process. In addition, the amount of ruthenium on the support was
decreased. Therefore the MDP is decreased by around 42 % in case
of the HPS Conti 0.5 M scenario.
If furthermore a common concentration for the industrial D-glucose
hydrogenation is assumed as done in the theoretical scenarios (Table
2), HPS Batch 5.6 M and HPS Conti 5.6 M, the GWP would be in a
competitive range compared to the industrial benchmark process
RaNi Batch 5.6 M. As the HPS batch process has already been
conducted at a higher substrate concentration, chances are good, that
the theoretical scenario HPS Conti 5.6 M which ranges in the same
magnitude of impact in both categories as the Raney nickel
6
|
Journal Name, [year], [vol], 00–00
This journal is © The Royal Society of Chemistry [year]

Page 7 of 10
RSC Advances
benchmark process does, can be practically put into realization.
20
The same consideration was made for the NCNT-based system: The The increase of the concentration to an industrial level as done in the
performance of the catalyst of the original batch process was theoretical scenarios NCNT Batch 5.6 M and NCNTViCewonArttiic5le.O6nMline
DOI: 10.1039/C4RA14559C
improved in the continuous process by decreasing the amount of will decrease the environmental burdens further, so that the GWP
catalyst used and decreasing the amount of metal impregnated on the can be reduced to 1/10 of the original value and has then a
support. This way, the MDP was decreased by about 44 %. As the 25 comparable environmental impact to the impact of the Raney nickel
concentration in the first batch experiments was lower, the increase process. Unfortunately, the MDP would be still larger than for the
of the concentration also contributes to the improvement of the GWP RaNi Batch 5.6 benchmark process.
5
by around 1/4.
Interestingly, the environmental impact caused by the additional Ni
separation step required in case of the conventional process were
30 found less influential than expected.
10
Finally, the learnings gained by means of the exemplarily evaluation
of the impact categories GWP and MDP were proven by the holistic
assessment of ten of the categories considered by the ReCiPe
midpoint LCIA. The results are shown in Figure 8. They underline
35 the positive effect of the transfer of sorbitol production from batch to
continuous processing. They also support the findings resulting from
the analysis of GWP and MDP that HPS/Ru may become a superior
catalytic system for hydrogenation reactions compared to Raney
nickel.
40 Conclusion and Outlook
Within this comparative LCA study, the utilization of novel
heterogeneous, nanoparticulate catalytic systems for hydrogenation
reactions were critically evaluated against the state of the art by
means of a reaction of high industrial relevance. With both newly
45 developed systems, catalyst leaching into the product sorbitol can be
totally avoided, whereas activity and selectivity can be kept high
over time.
The results confirmed that the polymer-based nanoparticulate
catalyst HPS/Ru will be able to compete with the industrial
benchmark catalyst Raney nickel in the hydrogenation of D-glucose,
if the reaction can be performed also under industrial reaction
conditions in a continuously running process utilizing highly
concentrated D-glucose solution. In contrast, the industrial scale use
of NCNT/Ru for sorbitol production was forecasted to be less
55 competitive, being due to lower substrate concentrations, conversion
and higher catalyst/glucose ratio. However, this outcome may
depend on the current state of development of both systems under
laboratory scale conditions and can vary after further optimisation
and scale up.
Figure 8: Comparison of environmental impact potentials of glucose
hydrogenation process alternatives referred to benchmark RaNi Batch 5.6 M 50
- Constraint: high concentrated reaction mixtures.
15
60 According to the findings, a strategy was developed to compete with
the classical sugar hydrogenation procedure. The overall strategy in
order to decrease the environmental burdens caused by the D-glucose
hydrogenation catalyzed by NCNT/Ru or HPS/Ru is depicted in
Figure 9 in an order of priority in which they should be pursued.
65 Although the process per se seemed to be green at a first glance,
several hot spots occurred during the development process
accompanied by a detailed evaluation of all process chains which
emphasizes the need for a more critical and holistic evaluation of
catalytic processes. Thus, as an example, although the catalyst is
70 used in marginal amounts, its environmental impact is indeed
significant. Further, water – often implied as a green solvent – maybe
no longer green if energy-intensive separation steps becomes
necessary.
Figure 9: Measures to decrease the environmental burdens caused by the
glucose hydrogenation catalyzed by NCNT/Ru and HPS/Ru. They are put in
an order of priority in which they should be pursued.
At the current stage of development, a combination of the LCA study
75 with a full Life Cycle Costing analysis was not realizable, since too
many uncertainties and inconsistencies in the data especially
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 7

RSC Advances
Page 8 of 10
concerning material costs remained. However, first cost estimations In the following, the experimental procedures of catalyst preparation
revealed some hints concerning hot-spots and improvement and D-glucose hydrogenation, being the basis of LCI analysis, are
View Article Online
DOI: 10.1039/C4RA14559C
potentials:
given.
The costs for the HPS catalyst preparation are mainly caused by the
high costs of ruthenium salt and labour costs. Nevertheless, the 60
prices of the HPS-based catalyst and the Raney-Nickel benchmark
are in a comparable range. The same insight could be gained for the
glucose hydrogenation process under application of the HPS based
catalyst. Here, we also found the costs for labour to exceed the other
Preparation of Nanoparticulate Catalytic Systems on
Nanostructured Supports
5
HPS/Ru. The Ru-containing HPS-based catalysts (designed as
HPS/Ru) were synthesized according to the procedure described
elsewhere.46 In a typical laboratory-scale experiment, 3 g of the
support material HPS (purchased from Purolite Int., United
Kingdom, as Macronet MN100) was washed with water and dried
under vacuum (moisture < 2 %). The procedure for the production of
the HPS material published in detail in Tsyurupa et al..26 Then the
support was impregnated with 8 mL of ruthenium catalyst precursor
10 cost aspects by far, yet. This will surely substantially decrease in
parallel to a process scale-up. In case of the NCNT-based catalyst, 65
the NCNT support itself currently cause a major price difference
when comparing only the three catalysts. Due to this and due to the
comparatively high use and heating of water, the cost estimation of
15 NCNT-based glucose hydrogenation showed that this process needs
some more process optimization steps in order to be competitive.
All in all, the results indicated that a switch from batch to continuous
processing utilizing novel nanoparticulate HPS/Ru catalysts can
pave the way to a more efficient sorbitol production, but further
20 investigations under pilot-scale conditions need to reveal whether the
new catalyst alternatives stand the reaction conditions of the 75
industrial established process. NCNT/Ru based catalysts will even
need more development time and efforts to become a green and cost
efficient catalyst for sugar hydrogenation.
solution, which was prepared in
a separate vessel from
70 tetrahydrofurane (6 mL), methanol (1 mL), water (1 mL) and
Ru(OH)Cl3 (0.39 g). This solution was mixed for 20 min, before it
was added to the HPS material. Impregnation was carried out by
continuous stirring during 10 min. After the impregnation the
catalyst was dried at 70 °C and boiled with 0.1 M sodium hydroxide
with the addition of hydrogen peroxide (30 %) for about 2 h. Such a
treatment induced the formation of RuO2 nanoparticles inside the
pores of the HPS matrix. Afterwards, the catalysts containing RuO2
nanoparticles were separated from this solution by filtration, washed
with water and dried at 70 oC. Before testing the obtained catalysts
80 in D-glucose hydrogenation they were reduced in hydrogen flow at
25
o
300 C and ambient pressure during 2 h. The treatment resulted in
Abbreviations
nanoparticles with mixed valence structure Ru(0) and Ru(IV).46 Two
types of HPS-based catalysts were synthesized: i) powdered HPS
with a support particle size of less than 63 m (for batch process),
85 and ii) as-received HPS with the size of granules varying from < 0.3
mm (1 %) up to < 0.8 mm (90 %) (in case of continuous
hydrogenation). The content of Ru in the catalysts is 3.0 %wt (batch
process experiments) and 2.85 %wt (continuous experiments)
according to the results of elemental analysis. Transmission electron
90 microscopy pictures of the Ru/HPS catalysts can be found in Doluda
et al..32
CNT carbon nanotubes
CVD chemical vapour deposition
FDP fossil fuel depletion potential
30 FEP freshwater eutrophication potential
GWP global warming potential
HPS hypercrosslinked polystyrene
HPLC high pressure liquid chromatography
HTP human toxicity potential
35 LCA life cycle assessment
LCI
life cycle inventory
NCNT/Ru. A precursor solution was prepared from 0.610 g
RuCl3*xH2O (41 %wt Ru) and 250 mL deionized water, to which
4.75 g NCNT were added. The latter had been prepared as reported
95 elsewhere36 and consists of agglomerates containing particles sizes
between 100 µm and 800 µm and a maximum in the particle size
distribution at 355 µm. 150 mL of a 30 % aqueous solution of H2O2
were then added drop wise and under stirring keeping T ≤ 60 °C.
After complete addition, the mixture was heated to 80 °C for further
100 3 h. The overall procedure took 5 h. The impregnated catalyst was
filtered , washed, and dried in an oven at T = 100 °C The amount of
water for the washing was on average 50 mL per 1 g of NCNT/Ru.
The loading of Ru in the final catalyst used in batch experiments is
3.6 %wt. For all continuous D-glucose hydrogenation experiments,
a similar catalyst containing 2.2 %wt Ru was employed. The size of
the nanoparticles were found to be in the range of 1.38 ± 0.35 nm.
More information about particle sizes and transmission electron
microscopy pictures have been reported in Aho et al..47, 48
LCIA life cycle impact assessment
MDP metal depletion potential
NCNT nitrogen-doped carbon nanotubes
40 NLTP natural land transformation potential
ODP ozone depletion potential
POFP photochemical oxidant formation potential
RaNi Raney nickel
SSA Specific surface area
45 TAP terrestrial acidification potential
TETP terrestrial ecotoxicity potential
Acknowledgement
50 The authors gratefully acknowledge the financial support provided105
by the European Community's 7th Framework Programme for
Research and Technological Development under grant agreement
no.: CP-IP 246095-2 POLYCAT.
D-Glucose Hydrogenation
Appendix: Detailed description of the experimental
110 Hydrogenation of D-Glucose using HPS/Ru.
55 procedures
Batch Process. Hydrogenation of D-glucose to sorbitol was carried
8
|
Journal Name, [year], [vol], 00–00
This journal is © The Royal Society of Chemistry [year]

Page 9 of 10
RSC Advances
out in a house made isothermal stainless steel batch reactor (120 mL) 60
installed in a shaker according to a procedure described elsewhere.46
Notes
a Friedrich Schiller University of Jena, Institute of PharmacyV, iDewepAartrictmleeOnntline
The following reaction conditions were applied: temperature 140 °С,
hydrogen pressure 40 bar, initial concentration of D-glucose 1.6 M,
catalyst loading 0.4 g. Samples of the reaction mixture were analysed
via High Pressure Liquid Chromatography (HPLC) complex 65
UltiMate 3000 equipped with the comparison refractometer detector
Waters 410.
of Pharmaceutical Technology, Otto-Schott-Str. 41, D-07745 Jena,
DOI: 10.1039/C4RA14559C
Germany. Fax: +49 3641 949942; Phone: +49 3641 949951; E-mail:
dana.kralisch@uni-jena.de
5
bTver Technical University, A.Nikitina str. 22, 170026, Tver, Russia.
Tel./Fax: +7(4822)449317; E-mail: sulman@online.tver.ru
cProcess Chemistry Centre, Åbo Akademi University, Biskopsgatan 8,
20500, Turku/Åbo, Finland. Fax: +359 2215 4479; Phone:+358 2215
4985; E-mail:dmurzin@abo.fi
Continuous Process. Hydrogenation was carried out in a tubular
dBayer Technology Services GmbH, D-51368 Leverkusen, Germany. Fax:
+49-214-30-81118; Phone: +49-214-30-48420; E-mail:
stefan.roggan@bayer.com
10 mini-fixed-bed reactor setup with an inner volume of 150 mL 70
equipped with a tank containing the solution of D-glucose, tanks with
hydrogen and nitrogen, flowmeter, temperature and pressure
controllers, plate heat-exchanger and thermostat. The catalyst
dClariant (RUS), LLC, Andropov ave. 18/6, 115432, Moscow, Russia. Fax:
+7(495)7972598; Phone: +7(495)7972599; E-mail:
loading was 2 g. The reaction was carried out under the following 75 Lyudmila.Danilova@clariant.com
eLaboratory of Chemical Reactor Engineering/Micro Flow Chemistry and
Process Technology, Department of Chemical Engineering and Chemistry,
Eindhoven University of Technology, P.O. Box 513, 5600 MB Eindhoven,
The Netherland; Phone: +40-247 2973; E-mail: v.hessel@tue.nl
15 conditions: D-glucose concentration 0.5 M, temperature 110 oC,
hydrogen pressure 40 bar, glucose solution flow rate 19 mL/min.
Samples of the reaction mixture were analysed as described for the
batch process.
80 sfFraunhofer ICT-IMM Continuous Chemical Engineering Department,
Carl-Zeiss-Str. 18-20, 55129 Mainz, Germany. Fax: +49 6131 990-205;
Phone: +49 6131 990-377; E-Mail: Patrick.Loeb@imm.fraunhofer.de
20 Hydrogenation of D-Glucose using NCNT/Ru.
Batch Process. Hydrogenation of D-glucose over NCNT/Ru was
investigated in a Parr 4561 autoclave (300 mL). The autoclave was
equipped with a gas entrainment impeller, baffles, heating jacket and
a cooling coil, sampling line, pressure, temperature and stirring rate
25 controllers. The D-glucose solution was pre-heated and saturated
with hydrogen in a separate chamber. The catalyst sample was put in
the reactor which was flushed with nitrogen and hydrogen before
† Electronic Supplementary Information (ESI) available: See
85 DOI: 10.1039/b000000x/
References
1.
B. Kusserow, S. Schimpf and P. Claus, Advanced Synthesis &
Catalysis, 2003, 345, 289-299.
heating. When the reactor reached the desired temperature of 120°C 2.
the D-glucose solution was fed to the reactor and the total pressure 90
H.-D. Belitz, W. Grosch and P. Schieberle, Lehrbuch der
Lebensmittelchemie, Springer, Berlin, 2008.
30 was increased to 20 bars of hydrogen. Samples (1-2 mL) were
periodically withdrawn through a 0.5 μm sinter during the
experiments. A 0.1 mol/L 120 mL D-glucose (Fluka, ≥ 98 % purity)
solution was used. The stirring rate was 1000 rpm. The experiments
were carried out between 120 and 180 minutes and the amount of 95
35 catalyst was between 0.1 and 0.2 g.
The concentrations of D-glucose and sorbitol were determined by
HPLC (HITACHI Chromaster HPLC) equipped with a refractive
index detector. A Biorad HPX-87C carbohydrate column was used,
the mobile phase was 1.2 mM CaSO4. The temperature of the column
40 was 70 °C and the flow rate of the mobile phase 0.5 mL/min, the100
detector was at 40 °C. Calibrations were made for D-glucose and
sorbitol.
3.
J. Karolyi, Industrial & Engineering Chemistry Process Design
and Development, 1968, 7, 107-110.
4.
G.
Parmar,
Global
Sorbitol
Market,
http://www.academia.edu/2590245/Global_Sorbitol_Market_-
_Isosorbide_Propylene_Glycol_Glycerol_and_Other_Downstrea
m_Opportunities_Applications_Toothpaste_Vitamin_C_Sweete
ner_Etc._Size_Share_Growth_Trends_And_Forecast_2012_-
_2018, Accessed 10, 2013.
5.
6.
H. Schiweck, A. Bär, R. Vogel, E. Schwarz, M. Kunz, C.
Dusautois, A. Clement, C. Lefranc, B. Lüssem, M. Moser and S.
Peters, in Ullmann's Encyclopedia of Industrial Chemistry,
Wiley-VCH Verlag GmbH & Co. KGaA, 2000.
P. Gallezot, N. Nicolaus, G. Fleche, P. Fuertes and A. Perrard,
Journal of Catalysis, 1998, 180, 51–55.
Continuous Process. Continuous hydrogenation of D-glucose over
NCNT/Ru was investigated in a continuously operating fixed bed
45 reactor set-up having six reactors in parallel, operated in co-current105 7.
trickle flow. The inner diameter of the reactor tubes was 12.5 mm
and the length of the electrically heated part was 12 cm. K-type
thermocouples were used to control the temperature. Knauer
Smartline 100 pumps were used to feed the D-glucose solution
T. Turek and C. Eisenbeis, Abschlussbericht zum Projekt
„Hydrierung
von
D-Glucose
Institut
zu
D-Sorbitol
an
Monolithkatalysatoren“,
für
Chemische
Verfahrenstechnik,
TU
Clausthal;
Max-Buchner-
Forschungsstiftung, Clausthal-Zellerfeld, 2009.
110 8.
P. Anastas and J. Warner, Green Chemistry: Theory and Practice,
Oxford University Press, 1998.
50 (Fluka, > 98%). The reaction conditions used for the continuous set-
up are the following: temperature 130 °C, glucose solution flow rate
2.0 mL/min, hydrogen flow rate 50 mL/min at a pressure of 20 bar.
The glucose concentration was 0.2 mol/L and the amount of catalyst
was 0.05-0.50 g. The particle size of the catalysts used in the
9.
D. Ott, D. Kralisch, I. Denčic, Y. Laribi, P. D. Perrichon, V.
Hessel, C. Berguerand, L. Kiwi-Minsker and P. Loeb,
ChemSusChem, doi: 10.1002/cssc.201402313.
55 continuous reactors were between 125 and 250 µm. Samples were115 10.
periodically withdrawn from the reactors and analysed with HPLC
as described in the previous section.
I. Denčić, D. Ott, D. Kralisch, T. Noël, J. Meuldijk, M. de Croon,
V. Hessel, Y. Laribi and P. Perrichon, Organic Process Research
& Development, doi: 10.1021/op5000573.
Further synthesis conditions for both, batch and continuous
11.
German Patent, DE 000003936757 A1, 1989.
hydrogenation are summarized in Tables 1 and 2.
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 9

RSC Advances
Page 10 of 10
12.
13.
M. Abdel Akher, J. Ghali, M. S. Raouf and M. Roushdi, Starch - 39.
D. Kralisch, D. Ott, S. Kressirer, C. Staffel, I. Sell, U. Krtschil and
Stärke, 1974, 26, 307-312.
P. Loeb, Green Processing and Synthesis, 2013, 2, 465-478.
M. Abdel Akher, J. Ghali, M. S. Raouf and M. Roushdi, Starch - 40.
D. Kralisch, C. Staffel, D. Ott, S. Bensaid, GV. ieSwaArartciccloe,OnPl.ine
DOI: 10.1039/C4RA14559C
Stärke, 1974, 26, 352-355.
60
Bellantoni and P. Loeb, Green Chemistry, 2013, 15, 463-477.
EN ISO 14040, International Organization for Standardization,
Geneva, 2006.
5 14.
M. W. Kearsley and G. G. Birch, Starch - Stärke, 1977, 29, 425- 41.
429.
15.
16.
17.
German Patent, DE 102004058811, 2006.
German Patent, DE 10128203 A1, 2002.
42.
EN ISO 14044, International Organization for Standardization,
Geneva, 2006.
T. Kilpiö, A. Aho, D. Murzin and T. Salmi, Industrial & 65 43.
Engineering Chemistry Research, 2013, 52, 7690-7703.
J. Wisniak, M. Hershkowitz and S. Stein, Product R&D, 1974, 13, 44.
232-236.
R. Hischier, S. Hellweg, C. Capello and A. Primas, The
International Journal of Life Cycle Assessment, 2005, 10, 59-67.
Swiss Centre for Life Cycle Inventories, Dübendorf, 3.0 edn.,
2013.
10
18.
19.
J. Wisniak and R. Simon, Industrial & Engineering Chemistry 45.
German Patent, DE 544666, 1932.
Product Research and Development, 1979, 18, 50-57.
D. Astruc, F. Lu and J. R. Aranzaes, Angewandte Chemie
International Edition, 2005, 44, 7852-7872.
70 46.
V. N. Sapunov, M. Y. Grigoryev, E. M. Sulman, M. B. Konyaeva
and V. G. Matveeva, The Journal of Physical Chemistry A, 2013,
117, 4073-4083.
15 20.
21.
22.
G. Schmid, Chemical Reviews, 1992, 92, 1709-1727.
J. H. Fendler, Nanoparticles and Nanostructured Films, Wiley-
VCH, Weiheim, 1998.
47.
A. Aho, S. Roggan, O. A. Simakova, T. Salmi and D. Y. Murzin,
Catalysis Today, doi: 10.1016/j.cattod.2013.12.031.
A. Aho, S. Roggan, K. Eranen, T. Salmi and D. Y. Murzin,
Catalysis Science & Technology, doi: 10.1039/c4cy01088d.
75 48.
20 23.
24.
B. R. Cuenya, Thin Solid Films, 2010, 518, 3127-3150.
J. M. Campelo, D. Luna, R. Luque, J. M. Marinas and A. A.
Romero, ChemSusChem, 2009, 2, 18-45.
25.
A. Wielckowski, C. Vayenas and E. Savinova, Catalytic and
Electrocatalysis at Nanoparticle Surfaces, Marcel Decker, Inc.,
New York, 2003.
25
26.
M. P. Tsyurupa and V. A. Davankov, Journal of Polymer Science:
Polymer Chemistry Edition, 1980, 18, 1399-1406.
M. P. Tsyurupa and V. A. Davankov, Reactive and Functional
Polymers, 2006, 66, 768-779.
27.
30 28.
29.
V. A. Davankov and M. P. Tsyurupa, Reactive Polymers, 1990,
13, 27-42.
A. V. Bykov, M. A. Rubin, M. G. Sulman and E. M. Sulman,
Catalysis in Industry, 2014, 6, 143-149.
30.
E. M. Sulman, P. M. Valetsky, M. G. Sulman, L. M. Bronstein,
A. I. Sidorov, V. Y. Doluda and V. G. Matveeva, Chemical
Engineering and Processing: Process Intensification, 2011, 50,
1041-1053.
35
31.
F. Cárdenas-Lizana, C. Berguerand, I. Yuranov and L. Kiwi-
Minsker, Journal of Catalysis, 2013, 301, 103-111.
V. Y. Doluda, J. Wärnå, A. Aho, A. V. Bykov, A. I. Sidorov, E.
M. Sulman, L. M. Bronstein, T. Salmi and D. Y. Murzin,
Industrial & Engineering Chemistry Research, 2013, 52, 14066-
14080.
40 32.
33.
S. Iijima, Nature, 1991, 354, 56-58.
45 34.
M. Kumar and Y. Ando, Journal of Nanoscience and
Nanotechnology, 2010, 10, 3739-3758.
35.
36.
German Patent, DE 102007062421 A1, 2009.
L. Mleczko and G. Lolli, Angewandte Chemie, 2013, 125, 9540-
9557.
50 37.
O. G. Griffiths, J. P. O'Byrne, L. Torrente-Murciano, M. D. Jones,
D. Mattia and M. C. McManus, Journal of Cleaner Production,
2013, 42, 180-189.
38.
D. Kralisch, I. Streckmann, D. Ott, U. Krtschil, E. Santacesaria,
M. Di Serio, V. Russo, L. De Carlo, W. Linhart, E. Christian, B.
Cortese, M. H. J. M. de Croon and V. Hessel, ChemSusChem,
2012, 5, 300-311.
55
10
|
Journal Name, [year], [vol], 00–00
This journal is © The Royal Society of Chemistry [year]