"Novel bio-based catalytic strategies for the fractionation and valorization of

lignocellulose"

Von der Fakultät für Mathematik, Informatik und Naturwissenschaften der RWTH Aachen

University zur Erlangung des akademischen Grades eines Doktors der Naturwissenschaften

genehmigte Dissertation

vorgelegt von

Diplom-Chemiker

Philipp M. Grande

aus Bergheim

Berichter: Universitätsprofessor Dr. Walter Leitner

Universitätsprofessor Dr. Marcel Liauw

Tag der mündlichen Prüfung: 05.05.2014

Diese Dissertation ist auf den Internetseiten der Hochschulbibliothek online verfügbar.

iv

This work has been performed by the scientific guidance of Univ. Prof. Dr. rer. nat. Walter

Leitner and under the supervision of Dr. rer. nat. Pablo Domínguez de María. All

experimental studies have been taken place in the research group “Lehrstuhl für Technische

Chemie und Petrolchemie” of the ITMC (Institut für Technische und Makromolekulare

Chemie) at the RWTH Aachen University (Rheinisch-Westfälisch Technische Hochschule

Aachen) within the time from February, 2010 until May, 2013.

It was performed as part of the Cluster of Excellence “Tailor-Made Fuels from Biomass”

(EXC236), which is funded by the Excellence Initiative by the German federal and state

governments to promote science and research at German universities.

Affidavit

I hereby declare, that I have written this work independently and have used only the specified

sources and resourses. The work has not been presented to any examining office, but parts of

the work have already been published in scientific journals or as conference contribution (see

“List of Patents and Publications”).

Eidesstattliche Erklärung

Hiermit erkläre ich, dass ich diese Arbeit eigenständig verfasst habe und nur die angegebenen

Quellen und Hilfsmittel verwendet habe. Die Arbeit wurde bisher keiner Prüfungsbehörde

vorgelegt, wobei Teile der Arbeit in wissenschaftlichen Zeitschriften, Patenten, sowie

Konferenzbeiträgen veröffentlicht wurden (siehe “List of Patents and Publications”).

v

ABSTRACT

In the coming decades replacing the depleting fossil resources will demand the development

of integrated biorefinery concepts. This study focuses on the use of biomass as starting

material to produce biomass-derived fuels and commodity chemicals.

Initially, a conceptual biomass pretreatment/fractionation strategy called 1-step OrganoCat

Process is shown and optimized. This biphasic process uses an aqueous reactive phase, with

oxalic acid as catalyst to hydrolyze hemicellulose, and 2-MeTHF as the second phase for the

in situ extraction of lignin. Different acids as catalyst as well as biomass sources showed how

versatile the system is. Recycling the liquid phases and increasing the ratio of substrate to

solvent and catalyst by fourfold made the process more economically feasible. A 3 L-scale

reaction was conducted, showing that scale-up of the process is possible while maintaining

the laboratory scale efficiency.

Second, the newly designed 2-step OrganoCat Process is presented. In this process

hemicellulose is hydrolyzed in seawater, catalyzed by oxalic acid. The resulting solid residue

(cellulose and lignin) is extracted by different bio-based solvents to extract lignin. While the

efficiency of the hydrolysis improved significantly, lignin extraction was not as efficient as

observed in the 1-step OrganoCat Process, making the variation with two steps more suitable

for biomass with low lignin content like algae.

Finally, different strategies for the valorization of the sugar-containing fractions of

lignocellulose are presented. The organic acid-catalyzed as well as the enzymatic hydrolysis

of crystalline cellulose are shown to work efficiently in (concentrated) seawater as solvent.

For further conversion of the resulting glucose, a chemo-enzymatic approach is presented. In

the first step commercially available immobilized glucose isomerase (IGI) is used for the

isomerization of glucose to fructose in seawater. In the second step dehydration of fructose to

HMF is conducted in a biphasic seawater-2-MeTHF system with oxalic acid as catalyst.

Consecutive operation of both steps afforded 64% HMF. Using the raw hemicellulose effluent

from the 1-step OrganoCat Process for fermentation with U. maydis is proven to afford

itaconic acid, not needing any further downstream processing. Also the iron(III) chloride-

catalyzed dehydration of xylose from hemicellulose to furfural is shown to work successfully.

Both reactions are examples for possible subsequent reaction steps in biorefineries.

vi

KURZZUSAMMENFASSUNG

Für das Ersetzen von fossilen Ressourcen werden in den kommenden Jahrzenten Verbund-

Bioraffineriekonzepte notwendig. Diese Arbeit beschäftigt sich mit der Nutzung von

Biomasse zur Herstellung biogener Kraftstoffe und Rohchemikalien.

Zunächst wurde eine konzeptuelle Strategie zur Vorbehandlung von Lignocellulose, der

1-step OrganoCat Process vorgestellt und optimiert. Dieses zweiphasige Reaktionssystem

beinhaltet eine wässrige, reaktive Phase mit Oxalsäure als Katalysator zur Hydrolyse der

Hemicellulose, und 2-MeTHF als zweite Phase zur in situ Extraktion des Lignins. Die

Vielseitigkeit des Systems wurde durch die Verwendung verschiedener Säuren als

Katalysatoren, sowie durch die Verarbeitung verschiedener Arten von Biomasse demonstriert.

Die Wirtschaftlichkeit des Prozesses wurde durch die Wiederverwendung der flüssigen

Phasen verbessert, welches eine Vervierfachung des Substrat-Lösungsmittel und des Substrat-

Katalysator-Verhältnisses ermöglicht. Die Reaktion wurde unter Beibehaltung der Effizienz

wie im Labormaßstab auf einen 3 L-Maßstab hochskaliert.

Anschließend wird der neue 2-step OrganoCat Process vorgestellt. In diesem Prozess wird

Hemicellulose mit Oxalsäure als Katalysator in Meerwasser hydrolisiert und anschließend der

feste Rückstand (Cellulose und Lignin) mit verschiedenen biogenen Lösungsmitteln extrahiert

um das Lignin zu isolieren. Während die Hemicellulose-Hydrolyse signifikant verbessert

wurde, war die Ligninextraktion weniger effizient, als beim 1-step OrganoCat Process, so

dass der neue Prozess besser für Biomasse mit geringem Ligningehalt, wie Algen geeignet ist.

Schließlich werden verschiedene Strategien für die Verwertung der Zuckerfraktionen aus

Lignocellulose vorgestellt. In (konzentriertem) Meerwasser als Lösungsmittel wird die

effektive, durch organische Säuren, oder enzymatisch katalysierte Hydrolyse von kristalliner

Cellulose gezeigt. Für die Umsetzung der Glucose wird ein chemisch-enzymatischer Ansatz

vorgestellt. Im ersten Schritt wird in Meerwasser mit handelsüblicher, immobilisierter

Glucoseisomerase (IGI) Glucose zu Fructose isomerisiert. Im zweiten Schritt wird in einem

zweiphasigen Meerwasser-2-MeTHF-System mit Oxalsäure als Katalysator Fructose zu HMF

dehydratisiert. Die konsekutive Durchführung beider Schritte resultiert in 64% HMF. Es wird

gezeigt, dass die Fermentation des unraffinierten Hemicellulose-Effluenten zu Itaconsäure mit

U. maydis möglich ist, ohne eine weitere Aufreinigung zu benötigen. Auch die Eisen(III)-

chlorid-katalysierte Dehydratisierung von Xylose aus Hemicellulose zu Furfural wird

vorgestellt. Beide Reaktionen sind Beispiele möglicher Folgeschritte in Bioraffinerien.

vii

ACKNOWLEDGEMENTS

First of all I would like to thank Prof. Dr. Walter Leitner for giving me the opportunity to

work on such a challenging and amazing topic.

I would like to thank Dr. Pablo Domínguez de María for being endlessly patient and

constantly encouraging during the whole process of learning the scientific work. Furthermore,

our countless fascinating and delighting discussions about my topic and science in general

improved my understanding of being a scientist. I would like to thank him for all his trust in

my abilities and the opportunities of handling also organizational tasks, apart from my

scientific work.

For frequent and fruitful discussions I would like to thank my colleagues and friends Henning

Kayser, Zaira Maugeri, Dr. César A. Urbina-Blanco, Christoph R. Müller, Dr. María Pérez

Sánchez and Thorsten vom Stein. For their fantastic work in the lab I would like to thank my

researching students and lab assistants Sophie Weber, Gent Mehmeti, Christian T. Bergs,

Anne-Christine Kick and Natascha Eickels. Moreover, I would like to thank the whole ITMC

group for the supporting working and social atmosphere. I would also like to acknowledge

Qingqi Yan, Tobias Klement and Bin Zhao for very interesting and fruitful cooperation as

well as Hannelore Eschmann and Julia Wurlitzer from the GC/HPLC-department and Ines

Bachmann-Remy from the NMR-department.

In addition I appreciate all the incredible support from my parents Dr. Silvana and Dr. Bernd

Th. Grande, my sister Sabrina C. and brother-in-law Jörg Grande as well as my close friends

Tamara and Sebastian Wissen, enduring all the good and bad times included during the past

years.

viii

TABLE OF CONTENTS

1.  INTRODUCTION ......................................................................................................................................... 1 

1.1. General composition of lignocellulose, classification and sources ............................................................. 2 

1.2. Biomass processing / pretreatment. State of the art. ................................................................................... 4 

1.2.1.  Organosolv process ......................................................................................................................... 5 

1.2.2.  Steam explosion process ................................................................................................................. 6 

1.2.3.  Ionic liquid pretreatment ................................................................................................................. 7 

1.3. Hydrolysis of polysaccharides .................................................................................................................... 8 

1.3.1.  Enzymatic hydrolysis ...................................................................................................................... 8 

1.3.2.  Chemical hydrolysis ........................................................................................................................ 9 

1.4. Biomass derived chemicals ...................................................................................................................... 10 

1.4.1.  2-Methyltetrahydrofuran (2-MeTHF) ........................................................................................... 10 

1.4.2.  5-(Hydroxymethyl)-2-furaldehyde (HMF) .................................................................................... 11 

1.4.3.  Furan-2-carbaldehyde (Furfural) ................................................................................................... 13 

1.4.4.  2-Methylidenebutanedioic acid (Itaconic acid) ............................................................................. 14 

2.  AIM OF THE THESIS ................................................................................................................................ 16 

3.  RESULTS & DISCUSSION ........................................................................................................................ 19 

3.1. The 1-step OrganoCat-Process ................................................................................................................. 19 

3.1.1.  A biogenic lignocellulose fractionation concept ........................................................................... 19 

3.1.2.  Assessment of selectivity in the pulping process: hemicellulose vs. cellulose .............................. 21 

3.1.3.  Influence of the reaction vessel, pressure and preliminary scale-up ............................................. 24 

3.1.4.  Influence of type of acid catalyst and type of biomass .................................................................. 28 

3.1.5.  Influence of the reaction mixture composition .............................................................................. 31 

3.1.6.  Recycling of effluents ................................................................................................................... 33 

3.1.7.  Combination of mechanical treatment with pulping ..................................................................... 34 

3.2. The 2-step OrganoCat Process ................................................................................................................. 36 

3.2.1.  A fractionation concept for low lignin content biomass ................................................................ 36 

3.2.2.  Reaction time screening of the hemicellulose hydrolysis .............................................................. 37 

3.2.3.  Screening of solvents for the lignin extraction .............................................................................. 38 

3.2.4.  Recycling of the aqueous effluent ................................................................................................. 40 

3.3. Valorization strategies for the carbohydrate fractions from lignocellulose .............................................. 41 

3.3.1.  Introduction ................................................................................................................................... 41 

3.3.2.  Salt-assisted organic acid-catalyzed hydrolysis of microcrystalline cellulose .............................. 42 

3.3.3.  Enzymatic hydrolysis of cellulose with Accellerase® 1500 .......................................................... 45 

3.3.4.  Chemo-enzymatic conversion of glucose to 5-(hydroxymethyl)-2-furaldehyde ........................... 52 

ix

3.3.5.  Synthesis of furfural from xylose using the raw 1-step OrganoCat Process effluent .................... 58 

3.3.6.  Fermentation of raw carbohydrate effluents .................................................................................. 59 

4.  SUMMARY AND CONCLUSION ................................................................................................................ 62 

5.  EXPERIMENTAL PART ............................................................................................................................ 67 

5.1. Chemicals ................................................................................................................................................. 67 

5.2. General Procedures ................................................................................................................................... 67 

5.2.1.  Standard procedure for the 1-step OrganoCat Process .................................................................. 67 

5.2.2.  Standard procedure for the 2-step OrganoCat Process .................................................................. 68 

5.2.3.  Procedure for the organic acid-catalyzed cellulose hydrolysis in seawater ................................... 69 

5.2.4.  Standard procedure for the cellulose hydrolysis with Accellerase® 1500 ..................................... 69 

5.2.5.  Procedure for the synthesis of HMF from glucose ........................................................................ 70 

5.2.6.  Procedure for the synthesis of furfural from 1-step OrganoCat Process hemicellulose effluent ... 70 

5.2.7.  Procedure for fermentation of sugars ............................................................................................ 70 

5.2.8.  Procedure for the mechanical pretreatment with a screw press ..................................................... 71 

5.3. Analysis .................................................................................................................................................... 71 

5.3.1.  High Performance Liquid Chromatography (HPLC) .................................................................... 71 

5.3.2.  Gas Chromatography (GC) ........................................................................................................... 72 

5.3.3.  Analysis with Glucose (HK) Assay Kit......................................................................................... 73 

5.3.4.  Analysis with PAHBAH method .................................................................................................. 74 

REFERENCES ........................................................................................................................................................ 75 

LIST OF PATENTS & PUBLICATIONS .................................................................................................................... 87 

1

1. INTRODUCTION

The depletion of fossil resourses is fostering the research on the utilization of biomass to

produce biofuels for transportation and energy storage, as well as chemical products for

health, care, nutrition and functional applications. In this regard, lignocellulose represents a

highly attractive alternative feedstock, being the largest natural source of organic material and

additionally not competing directly with feeding purposes. Herein, efforts comprise not only

the development and cultivation of new plants, or the use of marginal lands, but also the

exploitation of readily available biomass waste streams (e.g. paper, compost, etc.) are of great

importance. Furthermore, utilizing bio-based and worldwide available chemicals for

processing may be of utmost relevance to be completely (geopolitically) independent of fossil

resourcses and financially competitive.

To achieve this transition from fossil to renewable resources, the development of biorefinery

strategies has been put forward. The integration of low cost fuel production and high value

chemicals production, using biotechnological and chemical conversion, is important to

provide energy and economic efficiency.[1–3] Furthermore, minimalizing transportation and

storage costs by locating biorefineries close to the feedstock (e.g. combination with

agriculture or algae farms) might become another important economic factor, whereby the

valorization of rural local areas may be envisaged.[4,5]

In addition to the concerns leading to the choice of biomass for its use in future biorefineries,

the envisaged enormous consumption of water for processes might become problematic,

especially in fermentative and enzymatic reaction steps. Thus, minimal pollution of waste

water streams has to be targeted by using easily biodegradable chemicals. The sources of

potable water are becoming more and more restricted, which might lead to a competition

between the chemical industry and nutrition. To tackle this issue the utilization of seawater

instead of drinking water might also prove to be a valid alternative, especially in the field of

large-scale biomass processing. For non-coastal cities the need of depurating residual water

for its further use in biorefineries appears as a promising option as well.

Along those lines, the philosophy of “Green Chemistry” can be used as a guideline to design

chemical products and processes, reducing or even eliminating the use and generation of

hazardous substances.[6] 12 principles can be used to evaluate atom and energy efficiency,

hazards, safety, abundance of feedstock, selectivity, biodegradability and pollution

2

prevention.[7] It is expected that such large-scale biorefineries will operate following these

drivers.

1.1. General composition of lignocellulose, classification and sources

Lignocellulose is generally considered to be the inedible part of plants, not suitable for food

production.[8,9] Main components of this composite material are typically cellulose (35-50%),

hemicellulose (15-30%) and lignin (15-30%) with small amounts of other substances (10%,

e.g. proteins, oils and ashes)[10–13] (Figure 1).

Figure 1. Scheme of lignocellulose fiber and the predominant monomers in cellulose, hemicellulose and lignin.

Cellulose is the most abundant biopolymer and source of fermentable sugars on earth. It is a

polydisperse linear polymer of D-anhydroglucopyranose with -1,4-glycosidic bonds,

anhydrocellobiose being the repeating unit, and a degree of polymerization from 100 to

20000. Hydrogen bonds between the polymers lead to a highly ordered supramolecular

structure.[14] The grade of regularity in hydrogen bonds between the cellulose polymers

defines the so-called crystallinity. Cleavage of the -1,4-glycosidic bonds can be achieved

3

acid-catalyzed,[15] by enzyme catalysis (e.g. cellulases)[16] or catalyzed by bases.[17] Herein the

crystallinity is of main importance for the accessibility of the catalyst to the polymer and a

crucial factor for the hydrolysis rate.[18]

Hemicellulose, the second most abundant polysaccharide in nature, is a heteropolysaccharide

consisting mainly of 1,4-linked -D-xylopyranose units (i.e. ca. 90% in beech wood) and to a

lesser part of hexoses (e.g. glucose, mannose, galactose) and other pentoses (e.g.

arabinose).[12,19] Due to its amorphous structure the hydrolysis of hemicellulose is more

straightforward than that of cellulose.[20,21]

Lignin is an amorphous macromolecule, containing the so-called monolignols (e.g. p-

coumaryl, coniferyl and sinapyl alcohol), connected with a huge variety of linkages

(-O-4-linkage being the most dominant).[22] It shows a heterogeneous composition with

primary structure, dependent on the plant and cell-wall region.[23] In contrast to the common

hypothesis of a highly cross-linked three-dimensional structure, caused by random phenoxy

radical-radical coupling, a lamella-like or linear polystyrene-like macromolecule structure

was suggested,[24,25] containing few to no cross-links.[26] However, due to strong self-

association of lignin chains[27] under relatively harsh conditions needed for the extraction,

understanding lignin structure remains a difficult task. In terms of finding applications for the

aromatic compounds in lignocellulose, investigations of phenolic lignin compounds extracted

from different plants as well as synthesized lignin model compounds have been conducted,

showing anti-oxidant activity[28–30] due to free phenolic hydroxyl groups, scavenging

radicals.[31,32] A huge variety of different molecules can be synthesized from lignin. The

production of vanillin via selective oxidation of lignin is one already successfully applied

industrial scale process.[33,34] Also, lignin gives access to the aromatics fraction (benzene,

toluene, xylene, BTX), different macromolecules (e.g. carbon fiber, polymer alloys, etc.) and

miscellaneous monomers (e.g. substituted phenols, styrenes, biphenyls, etc.).[2]

For an economically valuable utilization of lignocellulose as feedstock the selective

fractionation of this composite material into its three main components (e.g. cellulose,

hemicellulose and lignin) is of crucial importance (Figure 2). Besides avoiding problems like

catalyst inhibition, expensive downstream processing can be kept to minimum and thus lower

costs significantly. Furthermore, the separate conversion of each fraction allows application of

appropriate conditions for each substrate and product desired. By such fractionation, an array

of commodities and biofuels can be produced (Figure 2).

4

Figure 2. Some possible (bio)chemical pathways for future biorefineries.

1.2. Biomass processing / pretreatment. State of the art.

Biomass processing has been investigated extensively within the last century.[35] For a

complete valorization of lignocellulose a pretreatment has to fulfill several requirements. To

start with, the formation of sugars during the pretreatment or subsequent formation of sugars

by hydrolysis has to be optimized, avoiding the loss or unselective degradation of

carbohydrates. Secondly, the formation of inhibitory byproducts for the subsequent hydrolysis

and fermentation has to be kept to a minimum, and finally the overall process has to be

necessarily cost-efficient.[13] Different physical, physicochemical, chemical, and biological

processes have been developed. In this section, three of the most important processes will be

discussed due to their effectiveness concerning fractionation and improvement of subsequent

processing steps, and due to their special relevance for this PhD thesis.

5

1.2.1. Organosolv process

In the organosolv process, breaking of the internal lignin and hemicellulose bonds of

lignocellulose by hydrolysis with inorganic acids (e.g. HCl, H2SO4, H3PO4) is conducted

typically in aqueous-organic solvent miscible mixtures (Figure 3).

Figure 3. General scheme of the organosolv process.

To avoid possible inhibition of subsequent enzymatic reaction steps a straightforward

recovery of the organic solvent is crucial, and thus low boiling point solvents are preferred

(e.g. methanol, ethanol, acetone, ethylene glycol, triethylene glycol and tetrahydrofurfuryl

alcohol).[36,37] Typical process conditions are 100-250 °C and 10-45 bar,[38] whereby at

elevated temperatures (>185 °C) due to autocatalysis no addition of acids is needed. Organic

acids (e.g. oxalic, acetylsalicylic and salicylic acid) have been studied for the process as well.

Even though the organosolv process shows to be a very promising pretreatment for

lignocellulose, certain drawbacks have to be taken into account. Due to the severe processing

conditions energy consumption is considerably high, which reduces cost-efficiency. Also high

temperature and acidic conditions might lead to considerable degradation of sugars - to

furanics, levulinic acid and humins - and condensation of the lignin. Furthermore, alcohols

used as solvents may form ethers under these severe conditions. This might lead to solvent

loss among organosolv cycles, compromising the economics of the process.[39]

The separation of the three main components in a selective way is crucial for a complete

valorization of lignocellulose. It can be achieved by extensive costly washing steps with

6

organic solvents for the removal of residual lignin. However, the separation of C5 and C6

sugar fractions is often not possible, as they remain mixed in the aqueous solution.

Additionally, the inorganic acids (e.g. HCl, H2SO4) applied cannot be efficiently recycled and

may lead to environmentally unfavorable waste water streams.

1.2.2. Steam explosion process

Another commonly applied pretreatment method for lignocellulose is steam explosion. The

macro structural decomposition of chipped biomass is achieved by treating it with high-

pressure saturated steam followed by rapid decompression (Figure 4).

Figure 4. Scheme of the steam explosion process.

Typical conditions are 160-260 °C and 7-49 bar, expanding to atmospheric pressure after

several seconds to a few minutes.[13,40,41] Addition of acids (e.g. H2SO4, SO2, H2CO3) can

improve enzymatic hydrolysis of pretreated bagasse and removal of hemicellulose. A

decrease in formation of inhibitory compounds for subsequent reaction steps can also be

achieved.[42] Compared to conventional mechanical methods (e.g. mechanical comminution)

steam explosion has low energy requirement and it is very cost efficient due to the low

recycling or environmental costs.[43,44] Additionally, an overall improvement in enzymatic

hydrolysis rate of polysaccharides compared to untreated biomass is observed after steam

explosion .[45] However, significant degradation of hemicellulose sugars and transformation of

lignin is caused due to high temperatures, leading to inhibitory materials for subsequent

7

enzymatic steps.[46] Thus, washing steps of the pretreated biomass are necessary, removing

soluble lignin fragments, hemicellulose sugars and hydrolyzed parts of the cellulose.[35,47] The

steam explosion process itself is very cost-efficient. However, it lacks selectivity and

sustainability for the complete valorization of lignocellulose.

1.2.3. Ionic liquid pretreatment

Many ionic liquids (ILs) are non-flammable,[48,49] have negligible vapor pressure[50,51] and low

melting points due to large cations and anions with low symmetry and dislocated charge.[52]

They can be designed to feature high thermal stability,[53] high conductivity[54] and low

toxicity.[55,56]

ILs have been reported to dissolve and restructure the crystalline cellulose hydrogen bond

network efficiently. This leads to an amorphous-like structure of the cellulose and thus higher

depolymerization rates by increasing the number of solvent-exposed glucan chain hydrogen

bonds with water.[18,57–59] Besides application of ILs as pretreatment, in combination with

acidic metal chlorides (e.g. CuCl2, PdCl2, FeCl3) in 1-ethyl-3-methylimidazolium chloride

([EMIM]Cl) as well as with solid catalysts (e.g. Amberlyst, Zeolite) in

1-butyl-3-methylimidazolium chloride ([BMIM]Cl), direct acidic hydrolysis of cellulose has

been published recently.[60–62] Due to the fact that ionic liquids are known to dissolve lignin

quite effectively,[63] combined with the potential of dissolving cellulose, direct dissolution of

lignocellulose in ILs could be observed, mostly using [EMIM]Cl and [BMIM]Cl and leading

to significant improvement in the subsequent hydrolysis of lignocellulosic pulp.[64–70] A

general scheme of the procedure is shown in Figure 5.

Figure 5. General scheme of the ionic liquod pretreatment of lignocellulose.

8

Even though ionic liquids show promising potential as pretreatment for lignocellulosic

biomass some potential drawbacks exist (e.g. toxicity, cost, corrosiveness, recyclability and

final disposal), reducing their applicability in industrial processes.[51] Their high cost

commands efficient recycling of the ILs which is limited due to their low volatility and thus

might involve expensive extraction with organic solvents to purify recycled ILs.

To overcome the separation issue switchable ILs have been developed, where the IL is formed

by addition of an agent (e.g. CO2) and decomposed by removing this agent.[71] Some

thermally stable ionic liquids (e.g. bis[(trifluoromethyl)sulphonyl]amide) have been shown to

be distillable under reduced pressure, facilitating separation of high boiling substances from

the IL. [72–75] Another promising group of ionic solvents can be obtained by combining two

bio-based, low cost salts (hydrogen bond acceptors, e.g. choline chloride, with hydrogen bond

donors, e.g. levulinic acid, sugar-based polyols, etc.) to form deep eutectic solvents (DES),

[56,76–80] which have been reported to solubilize cellulose as well as lignin efficiently.[76,81]

1.3. Hydrolysis of polysaccharides

Depolymerization of polysaccharides is a necessary step to produce fermentable sugars,

which can subsequently be converted into desired products. While hydrolysis of

hemicellulose is straightforward due to its amorphous structure, cellulose depolymerization

remains challenging. Due to its high crystallinity the cellulose displays a low surface area for

cleavage of -1,4-glycosidic bonds, especially, for enzymatic hydrolysis. In this section, the

most important enzymatic and chemical hydrolysis of cellulose will be presented.

1.3.1. Enzymatic hydrolysis

The depolymerization of cellulose with cellulases and cellulosomes (complexed cellulase

systems) has been investigated intensely during the last decades. The recalcitrance of

lignocellulosic biomass is one of the most challenging issues to overcome.[20,82–92] As

cellulases are relatively expensive enzymes, improving volumetric productivity, utilizing

cheaper substrates to produce the enzymes, enhancing their stability concerning temperature

and pH, raising activity on solid substrates, and higher tolerance to end-product inhibition is

of crucial interest for applying them in industrial processes.[16] Most cellulases are produced

by Trichoderma sp. and Aspergillus sp.[93–95] They contain endoglucanases to expose new

9

chain ends by hydrolyzing intramolecular -1,4-glucosidic bonds, exoglucanases to release

soluble sugar oligomers (e.g. cellobiose, glucose) by cleaving cellulose chain ends and

-D-glucosidases to hydrolyze dissolved cellobiose. Primary hydrolysis, releasing soluble

oligomers from the solid substrate is the rate-limiting step, while secondary hydrolysis takes

place in the liquid phase and produces the desired glucose in a much faster manner.[16]

Inhibition of the cellulase complex by cellobiose and glucose is another major issue to tackle

within cellulose hydrolysis.[96,97] In contrast to the separate hydrolysis and fermentation

(SHF), simultaneous saccharification and fermentation (SSF) has been put forward,

converting sugars directly to less or non-inhibitory products (e.g. ethanol). This leads to

enhanced cellulose hydrolysis rates.[98–100] Moreover, for the enzymatic conversion of

lignocellulose, simultaneous saccharification and co-fermentation (SSCF) can directly

ferment cellulose- and hemicellulose-derived sugars in one reactor with the hydrolysis.[101]

However, a major drawback of both methods may be the difference in optimal temperature for

cellulases (between 45 and 50 °C) and typical fermentative organisms (between 30 and

35 °C). Thus a compromise has to be found for each combination, lowering the respective

reaction rates and stability of the enzymes.[102] Despite these issues, reduced capital costs

(lower amount of enzyme and number of vessels needed) and higher ethanol yields than in

SHF, make SSF and SCFF preferred methods in laboratory and pilot plant studies for ethanol

production.[103,104]

1.3.2. Chemical hydrolysis

In addition to enzymatic methods mentioned above, the efficient depolymerization of

cellulose has been investigated intensively using inorganic acids (like H2SO4, H3PO4 or HCl)

as catalysts. However, due to the high crystallinity of cellulose, high temperatures (>150 °C)

have to be applied, causing degradation of the produced sugars.[90,91,105–108] To enhance the

catalytically available surface area, ionic liquids have been employed due to their ability to

reorder cellulose fibers by breaking hydrogen bond bridges leading to an amorphous-like

structure.[18,57,58,109–116] Another approach along the same lines is the utilization of inorganic

salts (e.g. LiCl, ZnCl2, CaCl2) to swell or even dissolve the cellulose slurry and thus improve

the hydrolysis rate with strong inorganic acids in an analogous way to the ionic liquid

mechanism.[15,117–123] As an alternative to the inorganic acids, and due to their biogenity

organic dicarboxylic acids (e.g. oxalic, maleic and fumaric acid) have also been studied for

cellulose hydrolysis.[62,100,124–132] As a result, dicarboxylic acids were found to perform better

10

than monocarboxylic acids due to their lower pKa values.[62,128] However, the efficient

depolymerization of highly crystalline cellulose requires very high temperatures (>160 °C),

leading again to unselective sugar degradation, lowering the attractiveness of the process.[125]

To tackle these issues, our group recently published a conceptual process, combining organic

acids (e.g. oxalic acid, maleic acid) with inorganic salts (e.g. NaCl, CaCl2 typical at brine

concentrations) under mild conditions (125 °C), leading to efficient hydrolysis of

microcrystalline cellulose (e.g. AVICEL®).[133]

1.4. Biomass derived chemicals

Producing chemicals and fuels from renewable resources as an alternative to fossil resources

and exchanging hazardous solvents and reagents by more benign components has stepped into

the focus of modern research and process design.[1,2,90,134,135] A key step is the reduction of

oxygen content in sugars, derived from lignocellulose, which can be performed by removing

small, highly oxidized carbon molecules (e.g. CO2, formaldehyde and formic acid) as in

fermentative routes (e.g. producing ethanol), via hydrogenolysis, removing oxygen and

forming water with hydrogen or by dehydration of the sugars into a variety of furan-like

compounds.[136] The U.S. Department of Energy (DOE) made a top 10 list of bio-based

chemicals, updated recently.[1,137] In the following, some important bio-based chemicals will

be introduced.

1.4.1. 2-Methyltetrahydrofuran (2-MeTHF)

In order to develop benign processes the utilization of bioderived, non-hazardous chemicals is

crucial. Especially, the choice of solvents is of utter importance as they are the biggest source

of waste in chemical processes, but will be necessarily needed for several steps within

biorefineries as well.[134,135] 2-methyltetrahydrofuran (2-MeTHF) might become an alternative

to substitute less environmentally friendly solvents (e.g. CH2Cl2). 2-MeTHF can be

synthesized from biomass-derived substrates such as furfural or levulinic acid,[138–140] by a

series of reduction-dehydration steps, also leading to other interesting chemicals (e.g.

-valero-lactone, 1,4-pentanediol) as shown in Figure 6.[141]

11

Figure 6. Example of a chemical route for the synthesis of 2-MeTHF from levulinic acid.[141]

In addition to being biogenic, 2-MeTHF also shows promising properties concerning

degradability in the environment where presumably by oxidation and ring-opening, it can be

degraded abiotically by sunlight and air.[142] Also preliminary toxicological studies enable its

use for pharmaceutical processes[143] and make 2-MeTHF a viable option for industrial pilot

plants.[144] Due to its straightforward recovery from aqueous effluents by conventional

distillation,[145] 2-MeTHF is a promising solvent for many reactions. The portfolio of possible

applications substituting hazardous solvents and even improving yields and selectivity is

already broad, including Lewis-acid-mediated ring-opening reactions, organometallic

catalyzed reactions in basic conditions, carbon-carbon and carbon-heteroatom bond

formation, organocatalysis, biocatalysis and biomass processing.[146] However, a potential

drawback of 2-MeTHF is the formation of peroxides when exposed to oxygen. This can be

overcome by addition of stabilizers (e.g. butylated hydroxyl toluene), albeit more research in

that line is necessary for large-scale practical appliactions.[147]

1.4.2. 5-(Hydroxymethyl)-2-furaldehyde (HMF)

For the production of chemicals and fuels from biomass, 5-(hydroxymethyl)-furaldehyde

(HMF) is of huge importance as a platform molecule and has been reviewed regularly

throughout the past century.[136,148–152] Due to the presence of two functional groups (the

hydroxymethyl- and the aldehyde group) many different (bio)chemical conversions of HMF

are possible, encompassing the formation of esters, ethers, halides and oxidation of the

hydroxymethyl group, reactions of the furan ring, as well as reduction, condensation and

oxidation of the formyl group (Figure 7).[152]

Apart from its variety of applications, HMF shows promising properties concerning

biodegradability, as its degradation in microorganisms takes place through oxidation and/or

reduction to the furanic alcohol and acid forms.[153]

12

Figure 7. Scheme of possible products from HMF (adapted from [136]).

The synthesis of HMF can be divided into single-phase systems, biphasic systems and ionic-

liquid-based systems. A very comprehensive overview of the different processes has been

recently published by van Putten et al.[136] Most of the reactions take place in an aqueous

phase due to the ability to dissolve the sugars, whereby homogeneous acids (e.g. HCl, H2SO4,

oxalic acid) and heterogeneous acids (e.g. Amberlyst, metal oxides) as well as metal chlorides

(e.g. CrCl2, CuCl2) are used as catalysts. Due to the tendency of HMF to form humins - that

is, oligomeric condensates of sugars and HMF -, and to prevent rehydration of HMF to

levulinic and formic acid, biphasic systems have been favored, using organic solvents (e.g.

THF, butanol and methyl isobutyl ketone) for the in situ extraction of HMF, which can be

improved by addition of salts or aprotic solvents (e.g. dimethyl sulfoxide), stabilizing

HMF.[154] The mechanism of hexose triple dehydration to HMF is not completely understood

yet. Mechanistical approaches can be divided into those based on cyclic intermediates, where

starting from ketofuranose dehydration of the hemiacetal is followed by two consecutive

-dehydrations, and those based on acyclic intermediates, where a linear 1,2-enediol is

13

formed, followed by two consecutive -dehydrations, ring closure and a final elimination of

water. Additionally, the influence of the solvent (e.g. DMSO, ILs) on the mechanism needs to

be considered.[136]

1.4.3. Furan-2-carbaldehyde (Furfural)

Furfural, another promising platform chemical, can be derived by dehydration directly from

xylose or arabinose, which are main components of the hemicellulose fraction contained in

lignocellulose.[127,155,156] It offers access to a huge variety of further products, including

2-methylfuran, 2-methyltetrahydrofuran, valerate esters, 2-(ethoxymethyl)furan and

2-(ethoxymethyl)tetrahydrofuran furan as well as various C10-C15 coupling products (Figure

8).[157]

Figure 8. Scheme of possible products from furfural.

14

Typically, strong acids (e.g. H2SO4) are used as catalyst in a temperature range from 150 to

220 °C,[158,159] which due to high energy cost and low product yield results in high

investment.[157] Quantum mechanical studies suggest the protonation of C2-OH of xylose and

subsequent ring-contraction as the most favorable mechanistic pathway to furfural in

water[160].

1.4.4. 2-Methylidenebutanedioic acid (Itaconic acid)

In the quest of options to replace petroleum-based chemicals, itaconic acid is a promising

candidate to be incorporated into polymers as an alternative for acrylic or methacrylic acids.

A broad variety of applications can be targeted, including resins, lattices, fibers, detergents,

cleaners and bioactive compounds.[161,162] While the first synthesis of itaconic acid was

conducted chemically by pyrolysing citric acid, followed up by hydrolyzing the

anhydrides,[163] fermentative processes are commercially conducted due to their higher

efficiency.[161] Herein, different organisms have been studied (e.g. Aspergillus sp., Ustilago

sp., Candida sp., Rhodotorula sp.) with A. terreus being the most frequently used organism in

commercial production.[161] Even though different biochemical pathways have been discussed,

the main route is believed to proceed with citric acid and aconitic acid as intermediates which

are then enzymatically decarboxylated to itaconic acid.[164–166] (Figure 9)

15

Figure 9. Scheme of the postulated biosynthetical pathway for itaconic acid production (adapted from [161]).

Due to high cost for refined quality starting materials needed to get reproducible and

(obviously) high productivities, the development of itaconic acid producing organisms which

are able to convert lignocellulose-derived sugars as carbon source is needed.[162]

16

2. AIM OF THE THESIS

The general aim of this thesis is the conceptual development and optimization of strategies for

environmentally-friendly and cost-efficient pretreatment methods enabling the conversion of

lignocellulosic biomass into useful raw materials following the principles of “green

chemistry”. The obtained raw materials will be evaluated for their usability in subsequent

reaction steps, demonstrating their potential application in integrated biorefinery concepts.

To this end, three major topics will be investigated in detail:

1) The 1-step OrganoCat Process (see section 3.1 for details), developed in our group will

be optimized concerning selectivity of the fractionation of beech wood, separating

cellulose and hemicellulose efficiently and avoiding sugar degradation. A comparison

of different reaction vessels and scales as well as their influence on yields of the three

products (that are cellulose, hemicellulose and lignin) will be shown, also demonstrating

a preliminary scale-up of the process. The variability of the fractionation system, using

different (acid) catalysts and processing different kinds of lignocellulose will be

investigated, as well as the possibility of economical improvement by varying the

substrate-to-solvent and -catalyst ratio. The influence of mechanical pretreatment on the

1-step OrganoCat Process will be considered as well.

2) The 2-step OrganoCat Process is a variation of the 1-step OrganoCat Process. Salty

water is used as solvent for the first step to hydrolyze hemicellulose selectively and

efficiently. In a second step lignin is extracted. The potential improvement of the

hydrolysis rate in the first step by utilizing the effect of naturally occurring salts in

seawater on the acid catalyst will be investigated. Different promising candidates (e.g.

organic solvents, water-organic solvent-mixtures, deep eutectic solvents) as solvents for

lignin will be screened for the second reaction step. An improvement of the

substrate-to-solvent and –catalyst will be shown by recycling of the aqueous effluent in

the first reaction step.

17

3) Different strategies for the valorization of cellulose and hemicellulose sugars will be

developed and applied to samples from differently pretreated lignocellulose and

cellulose. Application of abundantly and naturally available seawater as solvent for

chemical (acid catalyzed) and biochemical (enzymatic) depolymerization of

microcrystalline cellulose as well as in acid catalyzed dehydration of sugars (e.g.

fructose, xylose) will be investigated, also combining enzymatic isomerization and acid

catalyzed dehydration for the synthesis of HMF. The utilization of sugar effluents from

the 1-step OrganoCat Process and cellulose depolymerization for furfural production

and fermentation to itaconic acid will be shown to be successful.

19

3. RESULTS & DISCUSSION

3.1. The 1-step OrganoCat-Process

3.1.1. A biogenic lignocellulose fractionation concept

A complete lignocellulose valorization that enables the use of the three main components -

cellulose, hemicellulose and lignin -, is of utmost importance for the set-up of future

biorefineries. Starting from lignocellulosic residues, three valuable raw materials can be

achieved for further (bio)chemical processing. Desirably, lignocellulose fractionation should

be highly selective, delivering non-degraded raw materials. Moreover, produced effluents

should be directly usable without implementing expensive downstream units operations, thus

integrating such steps in a simplified pipeline. As can be inferred from sustainability,

recycling of water, organic solvents and catalysts may become a crucial part when developing

such a system, in which optimized biomass loadings are aggregated. As a result from an ideal

fractionation set-up, a largely-delignified, crystalline cellulose pulp is achieved, which may be

more prone to cellulase-catalyzed depolymerization. Furthermore, due to its amorphous

structure the hemicellulose fraction is straightforwardly hydrolyzed by acids, affording

aqueous xylose.[129] Finally, the achievement of chemically usable lignin is rare in the

literature, due to further condensations, impurities and removal of reactive sites during

pulping processes. Herein, a fractionation system that might deliver manageable and low-

degraded lignin fractions would be an asset for future biorefineries.

Focusing on all these aspects, a novel chemical pulping procedure for lignocellulosic biomass

(the so-called 1-step OrganoCat Process) was conceptually developed in our research

group[167,168] (Figure 10).

20

Figure 10. Process scheme for one-pot fractionation of lignocellulosic biomass (1-step OrganoCat Process),

Figure adapted from [167].

In a one-pot system at 125 °C with two liquid phases (1:1 water:2-MeTHF) and a biogenic

catalyst (0.2 M oxalic acid regarding the aqueous phase) lignocellulose (i.e. beech wood) is

partially depolymerized and separated into its main components cellulose, hemicellulose

sugars (mostly xylose) and lignin. Under these conditions, hemicellulose is selectively

hydrolyzed via acid catalysis leaving cellulose untouched, while lignin is partially extracted in

situ to the organic phase. The reaction is conducted at mild conditions (125 °C, 10 bar) and at

residence times of 6 h[167]. 2-Methyltetrahydrofuran (2-MeTHF) was chosen as the (biogenic)

organic solvent which, due to its low boiling point (79 °C) can be recycled into the system via

distillation. After that distillation solid lignin is obtained. Solid cellulose pulp can be filtered

off from the aqueous phase and oxalic acid can be removed via crystallization from the

aqueous phase and be reused.

21

Taking this biogenic fractionation concept as starting point, several aspects of it were in-depth

studied and improved, with a clear focus on practical lignocellulose fractionation. In the next

sections these results will be described and discussed.

3.1.2. Assessment of selectivity in the pulping process: hemicellulose vs. cellulose

The production of pure aqueous effluents of hemicellulose sugars, whereas cellulose pulp

remained suspended, was the first aspect to be optimized, taking again beech wood as

prototypical lignocellulose. Apart from obtaining different fractions - xylose and pulp - for

their further valorization in biorefineries via fermentation or chemical conversion of sugars, a

low amount of impurities in the fractions is of great importance to overcome cumbersome and

expensive product workup. To improve selectivity for the hydrolysis of amorphous

hemicellulose without depolymerizing crystalline cellulose and thus yield mostly xylose,

different temperatures and reaction times were assessed. The yields of xylose in a temperature

range from 125 to 150 °C at two different reaction times of 3 and 6 h are compared in Figure

11. While the xylose yield after 3 h reaction time improved significantly from 125 to 140 °C,

after 6 h at 125 °C and higher temperatures the theoretical maximum of hemicellulose

contained in beech wood (15-20%[12]) was reached.

22

Figure 11. Production of xylose (wt% compared to total wood weight) in the aqueous effluent after reactions in

the range of 125-150 °C at different reaction times. Conditions: Metal-based high pressure reactor, biphasic

system water/2-MeTHF (1:1, 10 mL total volume), 10 bar CO2, oxalic acid: 0.2 M in the aqueous phase, beech

wood (0.1 mm): 100 g L-1 in the aqueous phase. Xylose and glucose production were analyzed by HPLC (see

experimental part).

Quantitative hydrolysis of hemicellulose can be achieved at 3 h at 140 °C, or for longer

reaction times at lower temperatures. Significant degradation of xylose to furfural could not

be observed in any of those experiments, enabling a broad and promising framework for

lignocellulose fractionation, at least in the case of beech wood. Depending on the biomass

applied, conditions may obviously be adapted, and a compromise yield vs. reaction time set

for each specific case.

Beside xylose, hemicellulose contains other sugars as well (e.g. pentoses, hexoses) whereby

glucose is the second largest fraction in beech wood (ca. 2-5% in relation to total wood

weight).[12] As can be seen from Figure 12, at temperatures between 125 to 140 °C, the

amount of glucose is in the range of 2-3 wt% (in relation to total wood weight). Hence, it may

be assumed that at those reaction conditions glucose originates almost exclusively from

hemicellulose. Interestingly, at higher temperatures (> 150 °C) glucose yield in the aqueous

solution enhances significantly (Figure 12). These results show that cellulose

depolymerization catalyzed by organic acids starts at temperatures of 150 °C, fully consistent

23

with studies in literature for cellulose depolymerization, under similar reaction

conditions.[127,129,131,132,169–172]

Figure 12. Production of glucose (wt% compared to total wood weight) in the aqueous effluent after reaction in

the range of 125-150 °C at different reaction times. Conditions: Metal-based high pressure reactor, biphasic

system water/2-MeTHF (1:1, 10 mL total volume), 10 bar CO2, oxalic acid: 0.2 M in the aqueous phase, beech

wood (0.1 mm): 100 g L-1 in the aqueous phase. Xylose and glucose production were analyzed by HPLC (see

experimental part).

Therefore, as a conclusion from section 3.1.2, the 1-step OrganoCat Process allows

outstanding selectivity patterns, showing that, under optimized conditions, acid

depolymerization proceeds exclusively on (amorphous) hemicellulose, whereas (more

crystalline) cellulose pulp remains suspended in the aqueous solution. Conclusively, to

achieve high selectivity of the biomass fractionation in the case of beech wood, avoiding

significant hydrolysis of cellulose pulp, 6 h at 125 °C as well as 3 h at 140 °C lead to the best

deliverables. Due to the energy efficiency of the system a shorter reaction time of 3 h at only

slightly higher temperature of 140 °C seems favorable. Yet, for further decisions on actual

reaction conditions, a more holistic perspective of a whole biorefinery process should be kept,

providing a broad heat- and mass- integration analysis.[173]

24

3.1.3. Influence of the reaction vessel, pressure and preliminary scale-up

Once the selectivity xylose-glucose was properly assessed and the processing conditions were

improved, the next step was to evaluate the influence of pressure by means of different

reaction vessels in laboratory use, as well as to provide preliminary data on the potential

scale-up of the concept and its future use at operational biorefineries.

In a first array of experiments, the 1-step OrganoCat Process was set-up using different

reaction vessels (Figure 13), reaction scales and applied pressures of CO2. Figure 14

summarizes these results. All fractionation processes were conducted under the same reaction

conditions (solvents, reaction time and temperature, catalyst and wood loadings, similar to

those of Figure 10 and Figure 11).

a) b) c) d)

Figure 13. Different reaction vessels. a) GC-vial; b) Schott-bottle; c) Metal-based high pressure reactor;

d) Glass-made high pressure reactor.

25

Figure 14. Production of xylose and glucose (wt% compared to total wood weight), detected in the aqueous

effluent after reaction for 6 h at 125 °C in different reactors and at different CO2 pressures. Conditions: Biphasic

system water/2-MeTHF (1:1), oxalic acid: 0.2 M in the aqueous phase, beech wood (0.1 mm): 100 g L-1 in the

aqueous phase. Xylose and glucose production were analyzed by HPLC (see experimental part).

Except for the pretreatments performed in GC-vials, in which xylose yields were quite low,

all other reactors led to analogous results, virtually reaching full conversion of the

hemicellulose, compared to average content in beech wood.[12] The lower outcome observed

in GC-vials can be explained by the lack of tightness of such devices, causing lower pressure

and solvent leaching, partially evaporated at reaction temperature. Remarkably, metal- and

glass-made high pressure reactors showed full conversion of hemicellulose at 125 °C after

6 h. For metal-based high-pressure reactors, CO2 was added to prevent the evaporation of the

organic phase into the upper part of the reactor. As observed previously,[167] no activity of

CO2 without addition of oxalic acid was found for lignocelluose fractionation. Albeit an effect

of dissolved CO2 in the aqueous solution lowering pH is possible, it does not appear to be

significant due to comparable results observed in the glass-made high pressure reactor, where

26

no CO2 was added (Figure 14). Therefore, due to better handling and smaller dead-volume

glass-made high-pressure reactors (10 mL and 40 mL) were chosen for further fractionation

studies at bench scale.

As observed in Figure 14, by changing from 10 to 40 mL the total reaction volume, analogous

results in xylose were achieved. Envisaging the importance of scale-up for the type of

fractionation system studied, the 1-step OrganoCat Process was further conducted in a 5 L

metal-based high pressure reactor with glass inlay. These experiments were performed in

cooperation with the Max-Planck-Institute für Kohlenforschung (MPI, Mülheim/Ruhr,

Germany). For this preliminary scale-up, 150 g of beechwood (6 mm) were suspended in

1.5 L of water containing 0.1 M oxalic acid (aq.), and 1.5 L of 2-MeTHF were added. In this

reaction only half of the oxalic acid concentration was chosen as a result from studies

showing the same high activity at both 0.2 M and 0.1 M oxalic acid in the aqueous phase (see

section 3.1.5, Table 2 for details). The high pressure reactor was tightly closed and heated to

140 °C for 3 h while magnetically stirred. Figure 15a shows the setup of the reaction. The

temperature and pressure were monitored during the course of the reaction (Figure 15b).

Heating the reactor to the desired reaction temperature as well as cooling down after reaction

took ca. 2.5 h. Pressure rose to ca. 10 bar at maximum temperature. The workup procedure is

described in the experimental part (section 5.2.1).

27

a) b)

Figure 15. Reaction-setup at the Max-Planck-Institute für Kohlenforschung. a) Filled glass inlay with substrate,

catalyst and solvents; b) Closed high pressure reactor.

After filtrating and washing 53 wt% of initial amount of beech wood (80 g) were obtained as

cellulose pulp which relates nicely to the expected amount of cellulose in beech wood

(35-50 wt%).[10] Gratifyingly, the yield in xylose, which was 24.0 wt% of initial amount of

beech wood (36 g), glucose, which was 4.33 wt% of initial amount of beech wood (6.5 g) and

lignin, which was 12.6 wt% of initial amount of beech wood (19 g) appear to be in the same

range as the small scale reactions (10 mL or 40 mL scale, Figure 16). Higher yields in sugar

fractions of the 3 L-scale experiment might be explained by the higher amount of beech wood

sample used for the experiment, leading to a higher homogeneity of the material.

28

Figure 16. Production of xylose, glucose (wt% compared to total wood weight) in the aqueous effluent and

lignin (wt% compared to total wood weight) in the organic phase, after processing at different reaction scales.

Conditions: Biphasic system water/2-MeTHF (1:1), oxalic acid: 0.1 M in the aqueous phase, beech wood:

100 g L-1 in the aqueous phase, 140 °C, 3 h. Xylose and glucose production were analyzed by HPLC (see

experimental part). Lignin production was determined gravimetrically after removing 2-MeTHF via distillation.

Overall, it can be concluded that at 3 L-scale, the 1-step OrganoCat Process worked at least

as efficiently as in the smaller scales. Based on these first results, it can be envisaged that the

scale-up to a pilot plant process might be feasible. In fact, at higher scale the fractionation

appears to be even more efficient and reproducible than when smaller volumes and loadings

are charged.

3.1.4. Influence of type of acid catalyst and type of biomass

When assessing a novel lignocellulose fractionation system, a crucial aspect is its adaptability

for different reaction conditions and scalability, as reported in previous sections. Furthermore,

another important aspect is to see whether different acid catalysts could be successfully used.

The aim of these studies is the identification of a palette of catalysts that may be recovered or

further used in subsequent reaction steps. Apart from using an easily recyclable catalyst, other

efficient, more creative ways of directly utilizing the effluent of one reaction for the next step

can be taken into account. As an example, for fermentation of aqueous xylose (and other

29

sugars) different buffer systems can be used (e.g. phosphate buffer, acetate buffer, citrate

buffer). Based on those facts, different acids were tested to substitute oxalic acid in the 1-step

OrganoCat Process, which could be subsequently neutralized and later used as buffer for

fermentation of the sugars. The results of the acid screening are shown in Figure 17.

Figure 17. Production of xylose (wt% compared to total wood weight) in the aqueous effluent, detected after

reaction for 6 h at 125 °C with different acids as catalyst. Conditions: Glass-made high pressure reactor, biphasic

system water/2-MeTHF (1:1), acid 0.1 M in the aqueous phase, pH adjusted to 1.2 (determined at RT, pH at

125 °C: ca. 1.2-1.7) with HCl, beech wood (0.1 mm): 100 g L-1 in the aqueous phase. # pH adjusted to 1.8

(determined at RT, pH at 125 °C: ca. 2.0-2.3)[174,175]. Xylose production was analyzed by HPLC (see

experimental part).

Gratifyingly, as shown in Figure 17, the xylose yield remains in the same “full conversion”

range for all different acids, as long as the same pH is employed. For organic acids the actual

pH shifts at 125 °C by approximately 0.5 units higher, as investigated recently in our

group.[175,176] A pH range of 1.2-1.7 seems to be applicable for the efficient fractionation of

beech wood. As observed, the fractionation system follows an acid-driven profile, whereby

xylose production is exclusively dependent on the proton concentration. Raising pH from 1.2-

1.7 to ca. 2.0-2.3 a significant drop in hydrolysis was observed. Analogous dependency of

organic acids and hydrolysis has been observed for cellobiose in our research group as

30

well.[175] From a practical perspective, the possibility of using different (biogenic) acids for

lignocellulose processing opens clearly a broad range of opportunities for chemical

engineering, e.g. valorization of acidic waste effluents from chemical plants, combination of

acid recovery with downstream units operations, etc.

Apart from the acid versatility, another important asset for lignocellulosic pretreatment

strategies is the possibility of adapting them to diverse biomass sources. To investigate this

the same optimized parameters applied for beech wood were evaluated for other types of

biomass. (Table 1)

Table 1. Production of xylose, glucose (wt% compared to total wood weight) in the aqueous effluent and lignin

(wt% compared to total wood weight) in the organic phase, detected after reaction for 3 h at 140 °C. Conditions:

Glass-made high pressure reactor, biphasic system water/2-MeTHF (1:1, 40 mL total volume), oxalic acid:

0.1 M in the aqueous phase, lignocellulose: 100 g L-1 in the aqueous phase. Xylose and glucose production were

analyzed by HPLC (see experimental part). Lignin production was determined gravimetrically after removing

2-MeTHF via distillation.

entry raw material xylose yield

[wt%]

glucose yield

[wt%]

lignin yield

[wt%]

1 beech wood (Fagus sp.) 17 2.3 11

2 Mate tea (Ilex paraguariensis) 5.9 2.1 10

3 reed (Phragmites australis) 19 1.2 14

4 spruce (Picea sp.) 2.6 0.3 0

Mate tea (Ilex paraguariensis) is a very popular tea in South America (e.g. Argentina).

Annually only in Argentina aproximately 280.000 tons of Mate are produced, consumed, and

then discarded.[177] For the valorization of local areas, utilizing lignocellulose should not be

restricted to plants grown for this purpose, but other biomass-based waste products may

provide economic options as well. Hence, leftovers of Mate preparation (once used for food

consumption purposes) were treated with the 1-step OrganoCat Process (Table 1, entry 2).

Hemicellulose was converted completely (content in I. paraguariensis ca. 7 wt%[178]) the

amount of recovered lignin (content in I. paraguariensis ca. 25 wt%[178]) was in the same

31

range as from beech wood (ca. 50% of theoretical content, Table 1, entry 2). The solid

cellulose pulp remained suspended in the aqueous solution. Thus, exhausted leftovers of Mate

tea might be used as a source of xylose, lignin and cellulose for further local valorization.

Obviously, such conceptual approach may be envisaged and extended for other vegetal

sources as well (e.g tea leaves).

Another substrate chosen due to its high availability was reed (Phragmites australis). Due to

its efficient cultivation reed or similar plant types could be a suitable substrate for future

provision of sugars and lignin streams. Reed yields xylose as well as lignin in the same

percentual range of amount compared to the dry biomass weight as beech wood (Table 1,

entry 3).

Spruce (Picea sp.) is a commonly used wood for paper production due to its abundance in

several northern countries. Hence, utilization as lignocellulose source appears convenient.

After conducting the 1-step OrganoCat Process, the obtained amount of xylose was relatively

low (2.6 wt% xylose yield, compared to the dry biomass weight, Table 1, entry 4) and lignin

could not be recovered at all from the organic solvent fraction. Spruce, being a so called soft

wood is characterized by low hemicellulose and high lignin content.[179] Due to the fact that

the 1-step OrganoCat Process relies on depolymerization of hemicellulose to loosen the

attached lignin and extract it into the organic phase, soft woods might not be the appropriate

substrate for a pulping procedure under the mild reaction conditions employed. Applying

more severe conditions (e.g. temperature, pressure) might lead to depolymerization of

cellulose and thus extraction of lignin. In that case aspects to be carefully optimized would be

the degradation of resulting sugars and eventual formation of humins - due to harsher reaction

conditions -, that could decrease the product purity and efficiency of separation. A

compromise between efficient processing conditions and low degradation patterns appears

feasible once optimizations would be conducted.

3.1.5. Influence of the reaction mixture composition

An application-oriented lignocellulose fractionation system must be straightforward and

economically feasible. An important aspect is to provide high selectivity in fractionation

(xylose-cellulose-lignin). Another crucial step is the optimization of the biomass and catalyst

loadings, together with reduction and recycling of water and solvent effluents. Due to the

physical feasibility of stirring a suspension of wood chips in water-2-MeTHF mixtures, there

32

is a limitation for substrate loadings at ca. 100 g L-1 (referring to the concentration only in the

aqueous phase, Table 2, entry 2). Consequently, increasing the biomass loadings up to

200 g L-1 resulted in a reduction in xylose and lignin yield (Table 2, entry 3). Albeit a

compromise may be found, other strategies to increase the amount of biomass processed per

volume would be clearly desirable.

For further improvements, the amount of catalyst used for the reaction was reduced to half.

That resulted in a concentration of 0.1 M instead of 0.2 M referring to the concentration only

in the aqueous phase (Table 2, entry 1). Xylose and lignin were obtained in the same range of

amount as with higher concentration of oxalic acid. Even though oxalic acid could be recycled

via crystallization, decreasing the amount needed for the fragmentation may economically

enhance the process significantly.

In addition, the amount of 2-MeTHF was reduced to assess the optimal solvent-to-water ratio.

Thus, when half of the amount of 2-MeTHF (Table 2, entry 4) was added, xylose yields in the

aqueous solutions remained complete. Albeit the lignin yield dropped slightly (from 13 to

10 wt%) the fractionation of beech wood shows to be still possible with half the amount of

2-MeTHF (water:2-MeTHF 2:1 v/v).

Table 2. Variation of reaction mixture composition. Production of xylose, glucose (wt% compared to total wood

weight) in the aqueous effluent and lignin (wt% compared to total wood weight) in the organic effluent, detected

after reaction for 3 h at 140 °C. Conditions: Glass-made high pressure reactor, biphasic system water/2-MeTHF

(10 mL total volume), oxalic acid: different concentrations in the aqueous phase, beech wood (6 mm): 100 g L-1

in the aqueous phase. Xylose and glucose production were analyzed by HPLC (see experimental part). Lignin

production was determined gravimetrically after removing 2-MeTHF via distillation.

entry catalyst loading

[mol L-1]

water:

2-MeTHF

wood loading

[g L-1]

xylose yield

[wt%]

glucose yield

[wt%]

lignin yield

[wt%]

1 0.1 1:1 100 17 1.2 14

2 0.2 1:1 100 18 1.2 13

3 0.2 1:1 200 15 2.0 9

4 0.1 1:0.5 100 18 1.2 10

33

3.1.6. Recycling of effluents

Apart from decreasing (thus optimizing) the amount of 2-MeTHF, another way to further

improve the solvent- and catalyst-substrate ratio is the recycling of the liquid phases by

adding new biomass, once the first (solid) loading has been processed. As stated above, there

is a physical limitation to the addition of solid biomass, which can be assumed to be in the

range of 100-150 g L-1 (referring to the aqueous phase for beech wood, 6 mm, Table 2). Thus,

to improve this, after conducting the reaction for 3 h at 140 °C the liquid effluents (that is, the

aqueous phase with hemicellulose sugars and oxalic acid, as well as the organic 2-MeTHF

phase with lignin) were recovered via filtration of the reaction mixture - removing solid

pulp -, and then a new beech wood loading (100 g L-1, referring to the aqueous phase) was

added (Figure 18).

Figure 18. Concept for recycling liquid phases in the 1-step OrganoCat Process.

The envisaged filtration-addition approach was performed up to four cycles (with a total wood

addition of 400 g L-1, referring to the aqueous phase). Gratifyingly, it led to a remarkable

improvement in hemicellulose sugar concentrations, producing an aqueous solution of ca.

34

60 g xylose L-1 (Figure 19). An almost linear increase in xylose concentration shows that,

applying state of the art filtration and distillation techniques, the direct recycling of the

solvents can yield much higher concentrations in product streams, resulting highly beneficial

for further reaction steps (e.g. fermentation,[180] dehydration[155]).

In addition to xylose deliverables, an increase of lignin in the organic phase was observed as

well (Figure 19). It must be noted that due to the experimental setup a loss of liquid phase

during filtrations could not be avoided. Consequently, the improvement of lignin yield, which

was determined gravimetrically, was only about twofold. However, this aspect can easily be

circumvented by optimization of the recovery process.

Figure 19. Recycling of liquid phases. Concentration of xylose, glucose in the aqueous effluent and lignin in the

organic effluent, detected after reaction for 3 h at 140 °C. Conditions (1st cycle): Glass-made high pressure

reactor, biphasic system water/2-MeTHF (1:1, 10 mL total volume), oxalic acid 0.1 M in the aqueous phase,

beech wood (6 mm): 100 g L-1 in the aqueous phase. following cycles: filtration (paper filter, MN615), addition

of new beech wood (6 mm): 100 g L-1 in the aqueous phase, 140 °C, 3 h. Xylose and glucose concentrations

were analyzed by HPLC (see experimental part); lignin production was determined gravimetrically after

removing 2-MeTHF via distillation and concentration calculated.

3.1.7. Combination of mechanical treatment with pulping

Prior to chemical pretreatment, the particle size has to be reduced by mechanical processing.

Phragmites australis was mechanically prepared at the Aachener Verfahrenstechnik (AVT,

RWTH Aachen University). One sample was processed only applying the cutting mill and

35

one was subsequently treated with the screw press. The screw press applies a shearing force,

decreasing moisture, physically changing the fibers and increasing the specific surface

area.[181]

Figure 20 shows the yields of xylose, glucose and lignin after cutting mill, chopping reed into

particles of 10 mm length, in comparison to additional screw-pressing after cutting mill.

Under optimized conditions of the 1-step OrganoCat Process xylose yield was about 18 to

20% and lignin yield about 12%. Here, no significant change was observable, comparing

cutting-mill reed to screw-press reed.

Figure 20. Impact of mechanical pretreatment on 1-step OrganoCat Process. Production of xylose, glucose

(wt% compared to total wood weight) in the aqueous effluent and lignin (wt% compared to total wood weight) in

the organic phase, detected after reaction for 3 h at 140 °C. Conditions: Glass-made high pressure reactor,

biphasic system water/2-MeTHF (1:1, 40 mL total volume), oxalic acid: 0.1 M in the aqueous phase, reed

(10 mm): 100 g L-1 in the aqueous phase. Xylose and glucose production were analyzed by HPLC (see

experimental part). Lignin production was determined gravimetrically after removing 2-MeTHF via distillation.

Further investigations at lower temperatures or shorter reaction times might show stronger

impact on separation of the lignocellulose components and thus lead to lower overall energy

costs. Likewise, the set-up of mechanical steps might contribute to homogenize the size of

biomass, and thus to enable more reproducible results during the subsequent (bio)chemical

steps. An investigation of the subsequent enzymatic hydrolysis of the resulting pulp after the

combination of cutting mill or screw press and 1-step OrganoCat Process will be discussed in

section 3.3.3.

cutting mill cutting mill + screw press

36

3.2. The 2-step OrganoCat Process

3.2.1. A fractionation concept for low lignin content biomass

For certain (bio)chemical segments contamination with organic solvents may become

unacceptable. Additionally, contamination of waste water effluents with organic solvents

represents an important ecological issue. Waste must be completely removed before recycled

water is placed back into the ecosystem. Albeit 2-MeTHF shows preliminary promising

features in biodegradability,[143] avoiding contamination completely or reducing it to a

minimum might become a desirable option in specific applications whereby sugars (e.g.

xylose) are envisaged for food industry or analogous “sensitive” markets.

To meet the requirements for those applications, the 2-step OrganoCat Process was designed

(Figure 21). In the first step, 100 g L-1 of beech wood are stirred at 140 °C for 3 h in seawater

as the sole liquid phase with 0.1 M oxalic acid as catalyst.

In a second step, after cellulose pulp filtration to achieve aqueous xylose, the obtained solid

cellulose-lignin mixture was extracted with a broad range of different organic solvents and

deep-eutectic-solvents (DES) which are known to dissolve lignin to a significant extent.[76]

Figure 21. Process scheme of the "2-Step-OrganoCat-Process".

37

For biorefineries a large consumption of water will be inevitable, due to its use as solvent in a

number of (bio)chemical conversions, and especially in enzymatic and fermentative reaction

steps. Whereas ca. 70% of the earth is covered with water, only 3% of that water is potable

water,[182] leading to an unfavorable competition between drinking water and usage for

industrial processes. To avoid this competition, utilization of water-depurated effluents from

municipal wastes for biorefineries could open up interesting opportunities. In parallel to that,

seawater is abundantly available and easily accessible, especially in coastal regions. As

described in literature seawater contains different salts (mostly NaCl, ca. 3-4 wt%),[183] which

may improve acid catalyzed cellulose hydrolysis rate.[133]

3.2.2. Reaction time screening of the hemicellulose hydrolysis

In a first step, reaction time screening of the hydrolysis of hemicellulose was assessed. As can

be seen in Figure 22, xylose yield reached a maximum already after 1 hour. Such fast results

in depolymerization are consistent with results observed in depolymerization of crystalline

cellulose in seawater and salty effluents catalyzed by organic acids (see also section

3.3.2).[133] Conducting the fractionation process at longer times (up to 3 hours) did not result

in higher content of xylose, aligned with the assumption that full hydrolysis of hemicelluloses

was reached after 1 hour already.

Figure 22. Reaction time screening of the 2-step OrganoCat Process. Production of xylose and glucose (wt%

compared to total wood weight) in the aqueous effluent. Conditions: Monophasic system (seawater, 10 mL total

volume), oxalic acid: 0.1 M, beech wood (6 mm): 100 g L-1, 140 °C. Xylose and glucose production were

analyzed by HPLC (see experimental part).

38

3.2.3. Screening of solvents for the lignin extraction

The solid cellulose pulp from the first step was separated from the aqueous phase by filtration

and washed with water until neutral pH. Then, lignin was extracted from the solid residue

(containing cellulose pulp and lignin) by adding 2-MeTHF and stirring for 3 h at room

temperature. Then, the organic solvent was removed via distillation. As observed, lignin yield

in the second step of the 2-step OrganoCat Process was not as high as in the 1-step

OrganoCat Process (Figure 23). To investigate this further, different mixtures of water and 2-

MeTHF, as well as other organic solvents were screened to see whether lignin yields could be

improved (Figure 23).

Figure 23. Screening of different solvents for lignin extraction after the first step. Conditions first step:

Monophasic system (seawater, 10 mL total volume), oxalic acid: 0.1 M, beech wood (6 mm): 100 g L-1, 140 °C,

3 h. Lignin extraction was conducted with different solvents at different reaction times and temperatures; yield

was determined gravimetrically after removing organic solvent via distillation. The amount of lignin was

determined gravimetrically and is given in wt% compared to total wood weight. *Lignin was precipitated by

addition of water to the DES, then filtered and redesolved in 2-MeTHF; organic solvent was removed via

distillation.

39

Ethanol and 1-butanol are known as good solvents for lignin from the organosolv process,[184]

as well as 2-MeTHF from the 1-step OrganoCat Process. The mixture of lactic acid with

betaine (2:1 mol:mol) forms a DES and has also been reported as a highly efficient solvent for

lignin dissolution.[76] As can be seen, all extractions led to analogous lignin yields (4-6 wt%).

Yet, similar yields to those observed with the in situ extraction from 1-step OrganoCat

Process could not be achieved in any of the experiments performed (Figure 23). This leads to

the conclusion that the set-up of in situ biphasic systems is crucial for achieving high lignin

yields. It could be hypothesized that in the absence of organic solvent lignin remains in acidic

aqueous solutions over longer reaction times, thus being prone to further irreversible

(re)polymerizations. More research in that direction is needed to validate these considerations,

which might bring a competitive advantage for the 1-step OrganoCat Process compared to

more classic organosolv treatments (typically performed in monophasic solutions under

strong acidic conditions).

In any case, the xylose production in this monophasic system remains outstanding, making it

especially appealing for biomass containing low lignin proportions (e.g. grasses), or for other

promising raw materials (e.g. algae). In fact, the improvement of hydrolytic efficiency when

using seawater shows great opportunity for processing seawater-based algae, not needing to

remove their salty-water content, and thus simplifying treatment processing. Additionally, the

lack of contamination with organic solvents in the first step shows high potential for

application of the sugars in food industry. Also in the subsequent waste (sea)water treatment

the absence of organic solvents is beneficial.

40

3.2.4. Recycling of the aqueous effluent

Based on previous results considering quantitative xylose yields, further improvements in the

solvent- and catalyst-to-biomass ratio were considered, by recycling the liquid phase in the

first step, analogous to the procedure discussed in section 3.1.5 for the 2-step OrganoCat

Process. A similar benefit for further biomass processing, like in the 1-step OrganoCat

Process (section 3.1.5), was targeted. For this purpose, after conducting the reaction for 3 h at

140 °C, the liquid effluent (that is, the 0.1 M oxalic acid solution in seawater) was recovered

via filtration of the reaction mixture – removing solid pulp –, and then new beech wood

(100 g L-1) was added. The filtration-addition approach was performed up to four cycles.

As can be seen from Figure 24, the xylose concentration increases almost linearly with every

cycle, leading to a remarkable xylose concentration of ca. 90 g L-1 after 4 cycles. This nicely

shows that also with this monophasic variant, higher xylose concentrations can be achieved

straightforwardly, significantly improving economics of this and following processing steps.

Figure 24. Recycling in 2-step OrganoCat Process. Concentration of xylose, glucose in the aqueous effluent,

detected after reaction for 3 h at 140 °C. Conditions (1st cycle): monophasic system (seawater, 10 mL total

volume), oxalic acid: 0.1 M, beech wood (6 mm): 100 g L-1, following cycles: filtration (paper filter, MN615),

addition of new beech wood (6 mm): 100 g L-1, 140 °C, 3 h. Xylose and glucose concentrations were analyzed

by HPLC (see experimental part).

41

3.3. Valorization strategies for the carbohydrate fractions from lignocellulose

3.3.1. Introduction

The valorization of all components of lignocellulose (e.g. cellulose, hemicellulose and lignin)

might become important for future biorefinery concepts. To simplify subsequent reaction

steps it is crucial to minimize downstream processing, as well as using abundantly available

and thus economic solvents and catalysts. Thus, conducting subsequent process steps after the

fractionation of lignocellulose with the not further purified effluents could demonstrate further

benefits of this process.

Based on these considerations, in this section some valorization strategies for cellulose and

xylose will be presented (Figure 25). Herein, the focus is on valorization of the carbohydrate

fractions lignin will remain out of the scope of this work. Processing of cellulose can be

achieved via enzymatic hydrolysis[104] as well as by setting up acid-catalyzed

depolymerizations.[185] Obtained sugar effluents can subsequently be converted chemically or

fermentative to synthesize so called platform molecules, such as HMF, furfural, itaconic acid,

etc. These molecules are then used to produce desired products, such as fuels, fine chemicals,

etc. (see section 1.4 for details).[141]

42

Figure 25. Flow scheme envisaged for the valorization of the 1-step OrganoCat Process product streams

3.3.2. Salt-assisted organic acid-catalyzed hydrolysis of microcrystalline cellulose

Based on findings showing that organic acids (e.g. oxalic acid, maleic acid) are able to

depolymerize amorphous cellulose[129] and the beneficial effect of adding salts, improving the

hydrolysis rate of acid-catalyzed (1-4)glycosidic bond cleavage[15] a proof-of-concept

process was designed for more crystalline celluloses.[133,167] Herein, microcrystalline cellulose

(AVICEL®) was hydrolyzed efficiently under mild conditions, combining organic acids with

NaCl.

43

Figure 26. Reaction scheme of the salt-assisted organic acid-catalyzed hydrolysis of microcrystalline cellulose.

Organic-acid catalyzed hydrolysis of cellulose in (concentrated) seawater

Abundantly available seawater was introduced as a natural source of NaCl (ca. 3-4%). Figure

27 compares hydrolysis of AVICEL® in different reaction media. Without the addition of salt

(white bars), the activity was close to zero under the mild reaction conditions applied (125 °C,

6 h). However, using seawater instead of distilled water improved hydrolysis rate significantly

(green bars). In drinking water production, seawater is commonly desalinated, leading to

waste water with high salt content,[186] which could be beneficial for the hydrolysis rate of this

reaction. To simulate this high-salt-content-seawater, seawater was concentrated fivefold

(blue bars) by distillation, reaching about the same NaCl loading as in brine. Gratifyingly, in

concentrated seawater the reaction rate is in the same range as with brine (grey bars). This

opens up a great opportunity to utilize this waste stream for sugar production out of

microcrystalline cellulose.

44

Figure 27. Impact of salt content in seawater on hydrolysis rate of organic acids in crystalline cellulose.

Conditions: AVICEL® cellulose: 20 g L-1, 0.1 M maleic acid, 125 °C, 6 h. Samples were taken after 0, 1, 4 and

6 h. Different solvents were used: distilled water, seawater, 5-fold concentrated seawater, brine. Concentrations

of reducing sugars were determined with PAHBAH method (see experimental part).

Organic-acid catalyzed hydrolysis of ionic liquid-pretreated cellulose

Another way to lower crystallinity in cellulose, enabling straightforward hydrolysis, is

dissolution and re-precipitation in ionic liquids (ILs).[187] To analyze this effect, the hydrolysis

rates of differently pretreated samples, provided by the Institut für Technische und

Makromolekulare Chemie (ITMC, RWTH Aachen University),[187,188] were compared (Figure

28). Samples of -cellulose were dissolved in different ILs and then precipitated by addition

of water. Cellulose was recovered by filtration and then dried. In comparison with only dried

-cellulose an improvement in initial reaction rate of hydrolysis with 0.1 M oxalic acid was

observed for [DMIm]Butyrate and even more after pretreatment with [EMIM]Butyrate.

Due to preliminary results, showing a positive effect in the hydrolysis rate caused by the

addition of NaCl,[133] also the impact of IL pretreatment under those conditions was

investigated. The initial reaction rate of oxalic acid with NaCl to hydrolyze untreated -

45

cellulose is already higher than without NaCl. Interestingly, after dissolution and

reprecipitation in [DMIm]Acetate the initial hydrolysis rate was doubled.

Figure 28. Effect of pretreating -cellulose with different ionic liquids. Conditions: pretreated -cellulose:

20 g L-1, 0.1 M oxalic acid, 125 °C, 6 h; # addition of 30 wt% NaCl compared to the aqueous phase. Initial

reaction rate was calculated from reducing sugar concentration, determined with PAHBAH after reaction (see

experimental part).

3.3.3. Enzymatic hydrolysis of cellulose with Accellerase® 1500

Another very efficient and selective way to hydrolyze cellulose to glucose is performed by

cellulases (see section 1.3.1). Herein, crystallinity[18] and content of associated material to

cellulose (i.e. lignin)[189] are of major importance, as well as capillary structure of cellulose

fibers, molecular arrangement and degree of water swelling of the substrate.[190] Moreover,

another important factor is product inhibition of cellulases and -glucosidases, typically

caused by increasing (soluble) concentrations of cellobiose and glucose[97]. These inhibitory

patterns can be overcome by applying simultaneous saccharification and fermentation

(SSF),[99] converting glucose in-situ through fermentation, thus removing the inhibitor (see

section 1.3.1 for details).[13]

46

The commercially available cellulase Accellerase® 1500 (kindly donated from Genencor®

International Corporation) was used to determine hydrolysis activity of different cellulose

substrates. It is actually an enzyme “cocktail”, intended specifically for lignocellulosic

biomass processing industries. The enzyme mixture is produced with a recombinant strain of

Trichoderma reesei and includes genetically improved exoglucanases, endoglucanases,

hemicellulases as well as -glucosidases at optimized proportions.

Influence of substrate and catalyst loading

In order to define the optimal conditions for hydrolysis of cellulose catalyzed by Accellerase®

1500 (see experimental part) different substrate and catalyst loadings were firstly screened.

Table 3 compares initial reaction rates at substrate loadings from 10 to 30 g L-1 and catalyst

loadings from 0.2 to 5 vol% (v/v). Best results were achieved at substrate loadings of 20 g L-1

(Table 3, entry 2). Higher cellulose loadings lead to lower reaction rates, presumably due to

substrate inhibition.[191] Likewise, the addition of more enzyme per solvent shows higher

activity but reaches maximum activity at about 1 vol% of Accellerase® 1500 (v/v) (Table 3,

entries 4-6). Hence, for further reactions 20 g L-1 AVICEL® and 1 vol% Accellerase® 1500

were chosen as standard conditions for further uses with cellulose samples.

Table 3. Initial reaction rates with different substrate and catalyst loadings. AVICEL® cellulose, 0.1 M citrate

buffer (pH = 4.5), 50 °C, 3 h. Initial reaction rate was calculated from glucose concentrations, determined with

Glucose (HK) Assay Kit after reaction (see experimental part).

entry substrate loading

[g L-1]

Accellerase® 1500

loading* [vol%]

initial reaction rate

[g L-1 h-1]

1 10 0.2 1.6

2 20 0.2 3.7

3 30 0.2 1.7

4 20 0.6 5.7

5 20 1.0 5.0

6 20 5.0 6.3

* related to the amount of solvent

47

Multistep-hydrolysis of AVICEL® with Accellerase® 1500

Due to product inhibition during enzymatic hydrolysis of cellulose, glucose has to be

continuously removed to reach a complete cellulose depolymerization. To assess whether

commercial cellulases would hydrolyze microcrystalline cellulose completely, a sample of

AVICEL® was hydrolyzed in multiple steps. After each step the aqueous glucose solution was

removed via filtration, the residual cellulose was washed with water and new Accellerase®

1500 and solvent were added. Figure 29 shows initial reaction rates of the different cycles.

Reaction rates diminish constantly over hydrolytic cycles. This could be explained by

different grades of crystallinity of the substrate.[192,193] First, less crystalline regions (of a

microcrystalline cellulose) are hydrolyzed. Herein, depolymerizing the solid cellulose is the

rate-determining step while hydrolyzing soluble oligomers is relatively fast. In the following

cycles, only the higher crystalline regions of cellulose are left so reaction rate decreases. In

the 5th cycle, cellulases showed almost no activity on the residue of cellulose particles with

highest crystallinity. Finally, in the 6th cycle no glucose could be determined anymore,

supporting that no further hydrolysis of the cellulose particles was possible upon such highly

recalcitrant cellulosic materials.

Figure 29. Initial reaction rates observed along the multistep hydrolysis of AVICEL® catalyzed by cellulases.

Conditions: AVICEL® cellulose: 20 g L-1, 0.1 M citrate buffer (pH=4.5), 1 vol% Accellerase® 1500, 50 °C, 3 h.

Initial reaction rate was calculated from glucose concentrations, determined with Glucose (HK) Assay Kit after

reaction (see experimental part).

48

Hydrolysis of cellulose pulps obtained from the 1-step OrganoCat Process

The cellulose pulp of the 1-step OrganoCat Process was hydrolyzed with Accellerase® 1500.

Figure 30 compares produced sugar concentration of Accellerase® 1500 with different

cellulose pulps. As benchmark commercially available AVICEL® cellulose was used.

Figure 30. Kinetics of the hydrolysis with Accellerase® 1500 of AVICEL® cellulose, beech wood (6 mm) and

beech wood pulp (after 1-step OrganoCat Process). Conditions: substrate loading: 20 g L-1, 0.1 M citrate buffer

(pH=4.5), 1 vol% Accellerase® 1500, 50 °C. Reducing sugar concentrations were determined with PAHBAH

after reaction (see experimental part).

The enzymatic cocktail converts about half of the microcrystalline cellulose (compared to

theoretical full conversion of 20 g L-1 cellulose yielding ca. 22 g L-1 glucose) within 3 h. At

about 10 g L-1 product inhibition takes place, consistent with the literature[97] and practically

avoided by simultaneous saccharification and fermentation (SSF) processes.[99] Conversely,

when using non-pretreated beech wood as substrate virtually no conversion was observable.

After applying the 1-step OrganoCat Process cellulase reaction rate increased significantly.

Compared to AVICEL®, the activity is slightly lower which might be explained by residual

49

lignin in the cellulose pulp which may inhibit the cellulases. However, hydrolysis rate was

greatly improved compared to the untreated beech wood.

Furthermore, the hydrolysis rates of the differently pretreated substrates (see section 3.1.7)

after conducting the 1-step OrganoCat Process were investigated. Herein, the reaction rate of

screw-pressed reed pulp after 1-step OrganoCat Process was significantly higher than that of

cutting-mill reed pulp (Figure 31).

Figure 31. Kinetics of hydrolysis with Accellerase® 1500 of cut and screw pressed reed and only cut reed before

and after 1-step OrganoCat Process. Conditions: cellulose pulp: 20 g L-1, 0.1 M citrate buffer (pH = 4.5), 1 vol%

Accellerase® 1500, 50 °C. Glucose concentrations were determined with Glucose (HK) Assay Kit after reaction

(see experimental part).

A possible explanation for this improvement could be the enhancement of the available

surface area and the decrease in cellulose crystallinity by screw press processing, presumed to

be analogous to ball milling treatment.[190] Even though without chemical pretreatment

compared to cutting-mill reed no change in hydrolysis rate was observable, after conducting

the 1-step OrganoCat Process, the higher surface area enhanced the hydrolysis rate

50

significantly. Overall, it appears that the combination of mechanical pre-steps with further

1-step OrganoCat Process may deliver promising synergies for a more adequate biomass

processing and valorization.

Enzyme catalyzed hydrolysis of cellulose in seawater

Scientific literature offers a broad number of examples of enzymes and microorganisms (e.g.

halofilic), able to live in high concentration of salt (e.g. NaCl), even up to saturated

solutions.[180,194] Some biotechnological applications have been reported,[180] and recently also

fermentative applications producing succinic acid from glucose in seawater were

published.[183] With regard to cellulases, there are some examples of these enzymes dealing

with high saline concentrations or ionic liquids.[195–197] Yet, surprisingly, the direct use of

cellulases in (concentrated) seawater for depolymerization purposes had never been assessed

or reported in the literature.

With such state of the art in mind seawater was used instead of distilled water to prepare the

buffer for the hydrolysis of cellulose with commercial glycosidic “cocktail”

Accellerase® 1500 (see experimental part). The first step was to evaluate whether cellulase

activity in seawater was analogous to that observed in distilled-water-based standard buffer.

Figure 32 compares the kinetics of hydrolysis of AVICEL® cellulose in distilled water (white

bars) and in seawater (grey bars). Gratifyingly, no significant difference was observable when

seawater was used instead of distilled water. Therefore, some commercially available

cellulase “cocktails” might be directly used in seawater as reaction media for the provision of

fermentable sugars.

Furthermore different concentrations of seawater were compared, simulating waste stream

from e.g. desalination plants. Higher concentrations were achieved by distillation of seawater

to the desired salt concentration. At twofold concentrated seawater (blue bars), the enzymatic

“cocktail” displayed a slightly lower performance. Furthermore, with fourfold concentrated

seawater (green bars) a significant reduction of hydrolysis rate was observed, presumably due

to the higher osmolarity of the solvent. However, cellulose was still hydrolyzed efficiently,

even at this high salt concentration. Accellerase® 1500 is a cellulase “cocktail”, composed of

several glycosidases, optimized in performance and proportion. Thus further investigations

with isolated enzymes are needed to understand which components are affected by the salt. In

combination with the knowledge about halofilic cellulases this might open up a possibility for

51

genetic evolution of cellulases, to support occurring high salt concentrations in concentrated

seawater, and thus to integrate this reaction media into future biorefinery plants.

Figure 32. Effect of higher concentration of seawater on cellulose hydrolysis. Conditions: cellulose: 20 g L-1,

0.1 M citrate buffer (pH=4.5), 1 vol% Accellerase® 1500, 50 °C. Glucose concentrations were determined with

Glucose (HK) Assay Kit after reaction (see experimental part).

52

3.3.4. Chemo-enzymatic conversion of glucose to 5-(hydroxymethyl)-2-furaldehyde

In the scope of utilizing biomass-derived chemicals and fuels, several molecules have gained

importance over the past decades. In this respect, 5-(hydroxymethyl)-2-furaldehyde (HMF) is

considered a platform chemical[1,137] due to its broad number of applications for further

conversions (see also section 1.4.2).[139,141,152,198–208] It can be directly derived from glucose or

fructose, upon a triple dehydration, and a variety of syntheses have been published using

homogeneous, heterogeneous, acid or metal catalysts[202].

Effluents of biomass-derived glucose are mostly aqueous, as cellulose hydrolysis must be

obviously conducted in aqueous solutions. Hence, further water removal as work-up step

might be considered energetically not efficient for this chemical segment. Thus, most of the

reported HMF syntheses take place in aqueous media. However, straightforward production is

challenging due to the instability of HMF,[209] especially in high concentrations and in

presence of sugars, where the so-called humins (e.g. oligomeric condensates of sugars and

HMF) are formed. In addition, in water media at severe conditions there is the possibility of

HMF rehydration, leading to formic and levulinic acid formation. To overcome this, several

strategies have been investigated: a) removal of HMF from the reactive phase into a second

organic phase, such as MIBK, 2-butanol or THF, b) addition of cosolvents, such as DMSO,

stabilizing sugars and HMF, c) utilization of non-aqueous solvents, such as ionic liquids.[136]

HMF can be produced either from glucose or fructose. The synthesis using glucose as raw

material that is obviously preferred for lignocellulose-based biorefineries, typically requires

more severe conditions (e.g. high temperature, pressure, etc.).[202] However, this enhances side

reactions and thus reduces selectivity for HMF. Consequently, new approaches, avoiding high

concentration (and thus degradation) of HMF in the reaction mixture but still converting the

complete sugar substrate selectively, have to be developed.

Herein a combined chemo-enzymatic synthesis of HMF from D-glucose is presented (Figure

33). In a first step, D-glucose is enzymatically converted to D-fructose by means of a

commercially available immobilized glucose isomerase (IGI, see Experimental Part).

Isomerization takes place in seawater where the immobilized (suspended) enzyme can be

easily recycled via filtration. In a second step, dehydration of D-fructose, catalyzed by oxalic

acid dissolved in seawater leads to HMF, which is in situ extracted into a 2-MeTHF phase and

can easily be isolated via distillation, while 2-MeTHF is recycled into the system. Oxalic acid

53

in the aqueous phase can be recycled and residual sugars can be either led back into the

enzyme tank or fermented.

Figure 33. Reaction scheme of the chemo-enzymatic conversion of D-glucose to HMF.

Conversion of D-fructose to 5-(hydroxymethyl)-2-furfural

At first, the oxalic-acid-catalyzed dehydration of D-fructose to HMF in a biphasic system was

investigated. The substrate and the catalyst were dissolved in water and a second organic

phase of 2-MeTHF was added for the in situ extraction of HMF. Table 4 compares HMF

yields after synthesis in different solvents. Adding salts to the acid-catalyzed dehydration of

sugars has been reported to be beneficial by decreasing solubility of organic species in

water.[210] Consequently, addition of NaCl (20 wt%, referring to the aqueous phase) in the

synthesis of HMF led to higher yields (Table 4, entry 2) than that observed for reactions

conducted without NaCl (Table 4, entry 1). In general, results are fully consistent with

previous studies using similar NaCl-water biphasic systems for HMF production.[154] Inspired

by the results shown above, pure seawater (Table 4, entry 3) was tested as a natural source of

NaCl in water as well as two- and fourfold concentrated seawater (Table 4, entries 4-5). In

case of pure seawater, the HMF yield is slightly lower than with the addition of NaCl

54

(20 wt%, referring to the aqueous phase), as consequence of the lower salt content (3-4%

NaCl in seawater[183]). However, at twofold as well as fourfold concentration of seawater,

yields of HMF were in the same range as observed for water-NaCl mixtures (Table 4, entry

2). In all cases a brownish coloring of both phases (that are the aqueous phase and 2-MeTHF)

was observed, suggesting some by-product formation, typically observed in these HMF

syntheses [211,212]. However, no solid humins were observed in the applied reaction conditions.

This low degradation of the reagents in combination with recyclability of solvents and catalyst

as well as the possibility to further use the glucose containing seawater for fermentation

without the need for workup[183] compensates the relatively low HMF yields and presents

great opportunity for a selective and complete valorization of the sugars.

These remarkable results, implementing seawater to enhance HMF production, keeping

selectivity at considerably high level, encourage further utilization of this solvent for the

dehydration of sugars. For instance, the dehydration of algae-derived saccharides[5] to

synthesize HMF, which has been reported to be straightforward (compared to that of

glucose),[213] could be combined with desalination plants and lead to a promising approach for

an integrated HMF production.

Table 4. Production of HMF (% compared to theoretical yield) in 2-MeTHF after reaction in different solvents.

Conditions: Biphasic system water/2-MeTHF (1:1, 10 mL total volume), oxalic acid: 0.1 M in the aqueous

phase, D-fructose: 1 mmol, 140 °C, 1 h. HMF production was analyzed by HPLC (see experimental part).

entry solvent NaCl* [wt%] HMF yield [%]

1 water – 27

2 water - NaCl 20 52

3 seawater ca. 4 45

4 2x conc. seawater ca. 8 54

5 4x conc. seawater ca. 16 57

* wt% related to the aqueous phase

55

Isomerization of D-glucose to D-fructose with immobilized glucose isomerase

Whereas conversion of fructose to HMF was shown to work efficiently in the seawater-

2-MeTHF setup presented above, glucose is not converted at the same conditions as also

shown in section 3.1.2, where no significant degradation of glucose or xylose was observed.

In order to not apply harsher conditions (e.g. higher temperature, lower pH) and consequently

lose selectivity by degradation of HMF, D-glucose can be isomerized to D-fructose by the so-

called Lobry deBruyn-Alberda van Ekenstein (LBAE) transformation.[214,215] For food-

industry-related applications, glucose isomerase (GI, D-xylose ketol isomerase; EC 5.3.1.5) is

commonly applied to isomerize D-glucose into D-fructose[216]. For instance, it is used in food

industry to produce high fructose corn syrup (HFCS, via saccharification of starch),[216] and it

has also been shown to be active in ionic liquids[217] and with boron-based systems in the

synthesis of HMF[218]. However, application of the GI in seawater has not been investigated

previously and will be presented in this work. Due to its straightforward recyclability and

improved stability over soluble forms,[219] a commercially available immobilized glucose

isomerase (IGI, Sweetzyme IT Extra, 5.1.3.5; ≥300 µg-1), was chosen.

The isomerization has an equilibrium at about 50-60% fructose content,[220] which makes

multiple treatment of the aqueous phase after conversion of D-fructose to HMF necessary.

Hence, the activity of the IGI in seawater (saturated with 2-MeTHF, analogue to the

envisaged aqueous effluent of the second step, Figure 33) was compared to the activity in

phosphate buffer (Figure 34). Impressively, under these non-optimized conditions, the activity

in seawater (saturated with 2-MeTHF) was only slightly lower than in buffer and reached

equilibrium in ca. 5 h (data not shown). This is consistent with previous findings reported in

the literature, where the robustness of the enzyme is also shown by reaction in ionic liquids or

in ethanol-aqueous systems.[217,221]

56

Figure 34. Kinetics of the isomerization of D-glucose to D-fructose in phosphate buffer and in seawater

(saturated with 2-MeTHF, no addition of buffer). Conditions: D-glucose: 140 g L-1, pH 8.2, IGI: 3 wt% (referring

to D-glucose), 60 °C. Glucose concentrations in samples were determined with Glucose (HK) Assay Kit after the

reaction (see experimental part).

Following these promising results, stability and activity of the immobilized enzyme using the

raw effluent from the former step was investigated over multiple cycles. After conducting the

isomerization of 140 g L-1 D-glucose for 40 min at 60 °C with 3 wt% IGI (referring to

D-glucose), the immobilized enzyme was recovered via filtration, added to new solution of

140 g L-1 D-glucose in seawater (saturated with 2-MeTHF) and tested again (Figure 35).

Gratifyingly, the enzyme remained stable during all of the cycles, which clearly emphasizes

the robust nature of the IGI as stated above, and provides promising prognoses on the use of

these enzymes in raw salty effluents to be used in future biorefineries.

57

Figure 35. Recycling of immobilized glucose isomerase (IGI) in 2-MeTHF-saturated seawater. Conditions:

D-glucose: 140 g L-1, seawater (saturated with 2-MeTHF, no buffer added, pH 8.2), IGI: 3 wt% (referring to

D-glucose), 60 °C, 40 min. After each cycle recovery of IGI by filtration and addition to new substrate solution.

Glucose concentration was determined with Glucose (HK) Assay Kit after the reaction and enzyme activity

standardized to remaining glucose in the first cycle (see experimental part).

From D-glucose to HMF by combining the isomerization and dehydration

Since both steps (enzymatic and chemical) showed good performance in the seawater media,

the combination of them was the subsequent experiment. For this purpose, 3 wt% IGI

(referring to D-glucose) was added to a solution of 140 g L-1 D-glucose in 5 mL seawater

(saturated with 2-MeTHF). It was stirred for 2 h at 60 °C, forming ca. 90 g L-1 of D-fructose

(calculated as difference from determined D-glucose concentration of 52 g L-1). After

removing the enzyme via filtration, oxalic acid was added to achieve a 0.1 M solution. Then,

5 mL 2-MeTHF was added and the mixture was stirred for 1 h at 140 °C. Analysis of the

organic phase showed 40 g L-1 HMF, equivalent to 64% yield (relative to the theoretical HMF

yield from D-fructose).

The organic phase was removed and the pH was adjusted to 8 (with solid NaOH). To start a

second cycle D-glucose was added to achieve a total concentration of ca. 140 g L-1. 3 wt% IGI

(referring to theoretical concentration of 140 g L-1 D-glucose) was added and reaction

58

conducted for 2 h at 60 °C, leading to 60 g L-1 of D-fructose (calculated as difference from

determined D-glucose concentration of 77 g L-1).

These results show nicely that the two steps, the isomerization of D-glucose to D-fructose and

the conversion of D-fructose to HMF, can be combined straightforwardly. Since immobilized

glucose isomerase proved to be very stable under the applied conditions, the aqueous effluent

after dehydration of D-fructose can be conversely recycled back into the enzyme reactor.

Alternatively, the remnant seawater with D-glucose (ca. 50%) could be used for further

fermentative processes. Optimization of the concept may provide useful and sustainable

synergies for future biorefineries combining enzymes and bio-based acids, and using seawater

as readily available and non-potable water (and salt) source.

3.3.5. Synthesis of furfural from xylose using the raw 1-step OrganoCat Process effluent

The platform chemical furfural, which is of great importance due to its numerous

applications[139,141,151,222] can be derived from xylose, main component of hemicellulose, by

triple dehydration of the monomeric C5 sugar. For the furfural synthesis our research group

developed a concept using FeCl3 as acidic catalyst in aqueous media and addition of NaCl to

improve furfural production rate, while the product is in situ extracted into a second

2-MeTHF phase[155] (Figure 36).

Figure 36. Scheme of the iron-catalyzed xylose dehydration. Figure adapted from[155]

59

The non-purified aqueous hemicellulose sugars effluent of the 1-step OrganoCat Process,

conducted with beech wood particles (0.5-0.1 mm, Figure 16) was directly used after removal

of oxalic acid and the 2-MeTHF phase (containing lignin). Furfural was synthesized upon

xylose dehydration with FeCl3 ∙ 6 H2O as catalyst. Xylose concentration in the effluent was

ca. 30 g L-1 (determined in the aqueous phase together with small amounts of other

carbohydrates after the 1-step OrganoCat Process).

To 2.5 mL of this raw solution, FeCl3 6 H2O (0.24 M) was added together with NaCl

(30 wt%, related to the aqueous phase). Then 2.5 mL of 2-MeTHF were added and the

reaction mixture was stirred for 2 h at 140 °C. Remarkably ca. 7 g L-1 of furfural were

observed in the organic phase, indicating a rate of formation of ca. 3.5 g L-1 h-1 and

corresponding to an initial yield of 37%, still under these largely non-optimized process

conditions.

Figure 37. Conversion of xylose into furfural mediated by FeCl3 6 H2O, using aqueous non-purified xylose

effluent obtained from the 1-step OrganoCat Process. Reaction conditions: ca. 30 g L-1 xylose, 0.12 M

FeCl3 6 H2O, 30 wt% NaCl with respect to aqueous phase, 2.5 mL 2-MeTHF, 2.5 mL of aqueous phase

produced in 1-step OrganoCat Process, 140 °C, 2 h.

3.3.6. Fermentation of raw carbohydrate effluents

Itaconic acid may be used as a building block for polymerization (e.g. plastics, adhesives,

elastomers and coatings),[161,162] as well as for the synthesis of potential biofuels or biogenic

60

solvents (e.g. 3-MeTHF) (see also section 1.4.4).[141] It can be synthesized via chemical

multistep reactions[223] or via biotransformations.[161] To keep downstream processing of sugar

effluents at a minimum, research with the goal of demonstrating compatibility of raw

effluents with subsequent processes is of high interest. In cooperation with

Bioverfahrenstechnik (AVT, RWTH Aachen University) different effluents of the 1-step

OrganoCat Process (section 3.1) and enzymatically hydrolyzed cellulose in seawater (section

3.3.3) were fermented by U. maydis to itaconic acid[180] (Figure 38).

Figure 38. Fermentation of raw sugar effluents with U. maydis.

Due to the fact that the organism preferes glucose over xylose as substrate, especially in the

initial growing phase, the sole fermentation of the hemicellulose effluent (13 g L-1 xylose in

reaction medium after preparation of the Tabuchi medium) yielded very low amounts of

itaconic acid. However, by adding 30 g L-1 glucose to that hemicellulose effluent (10 g L-1

xylose in reaction medium) a final itaconic acid concentration of 4.1 g L-1 was found after

36 h, consuming both xylose and glucose. In a reference experiment using a solution of

61

30 g L-1 xylose and 90 g L-1 glucose a concentration of 13.1 g L-1 itaconic acid was achieved.

The product concentration is much lower in the case of the hemicellulose sugars due to the

overall lower concentrations.[180] However, fermentation of the hemicellulose effluent of the

1-step OrganoCat Process was successfully achieved, even though higher sugar

concentrations in the reaction media as well as genetically improving acceptance of xylose as

substrate for the organism could further improve production.

The effluent of enzymatically hydrolyzed cellulose in seawater was produced by stirring

100 g L-1 amorphous cellulose (Sigmacell®101) with 1 vol% Accellerase® 1500 at 50 °C for

seven days in 0.1 M citric buffer (pH 4.5) prepared with seawater. The raw glucose solution

(ca. 60 g L-1 glucose, determined with HPLC) was used to prepare the Tabuchi medium

(without further addition of glucose) for fermentation. Remarkably, a slight increase of the

overall production of itaconic acid, compared to a reference cultivation using regular Tabuchi

medium with 60 g L-1 glucose, was observed. After 96 h, a final itaconic acid concentration of

6.4 g L-1 was obtained, demonstrating excellent results for the fermentation of this raw

sample.

62

4. SUMMARY AND CONCLUSION

Figure 39. Summary scheme.

63

In this thesis development and optimization of conversion strategies for lignocellulosic

biomass was targeted. It was desired to yield value-added raw materials following the

principles of “green chemistry”. Evaluating usability of the effluents originating from single

reaction steps in subsequent reactions was conducted to demonstrate a potential application in

integrated biorefinery concepts.

The 1-step OrganoCat Process [224] (a biphasic, biogenic lignocellulose fractionation strategy)

was in-depth studied in terms of selectivity and optimized parameters, identifying the best

conditions for an energy efficient, quantitative depolymerization of hemicellulose, leaving

cellulose untouched (3 hours at 140 °C). As a result, three non-degraded raw materials are

obtained from beech wood (xylose, cellulose and lignin). Moreover, a glass-made high

pressure reactor setup was shown to display the best results in laboratory-scale, regarding

yields of product streams and handling at 10 mL as well as 40 mL reaction scale (referring to

total solvent volume). Furthermore, the reaction was conducted at a 3 L scale (referring to

total solvent volume), proceeding as efficiently as in the smaller lab-scales, showing that a

scale-up for practical purposes can be realized straightforwardly. Additionally, the versatility

of catalysts used in the process was successfully demonstrated, showing full conversion of

hemicellulose, independently of the kind of acid, as well as the adaptability to a broad range

of different biomass sources with moderate-to-high hemicellulosic content, whereas

softwoods (e.g. spruce) still have some room for improvement due to their low hemicellulose

content, which might be tackled by applying harsher conditions (e.g. higher temperature,

pressure or addition of salt), thus extracting lignin at the cost of some cellulose degradation

extent. Furthermore, to improve solvent-to-substrate and -catalyst ratio, pretreatments at high

substrate loadings of 200 g L-1 were shown, as well as the reduction of organic solvent within

the process, yet leading to slightly lower yields. This issue was successfully overcome by

recycling the liquid effluent and adding new lignocellulose in multiple steps, leading to

remarkably enhanced hemicellulose sugar concentrations (ca. 60 g L-1 after four pretreatment

cycles) as well as recovered increased amounts of lignin. Finally, the combination of a

mechanical screw press pretreatment with chemical pulping resulted in comparable results

considering the product streams. Herein, reduction of reaction time and temperature, and thus

working at less optimized conditions might show higher impact of the mechanical

pretreatment on the fractionation of lignocellulose.

64

The 2-step OrganoCat Process, a variation conducted in two steps has been put forward as

well. First hemicellulose was hydrolyzed selectively in natural and abundantly available

seawater as reaction medium with bioderived and recyclable oxalic acid as catalyst. In a

second step lignin was extracted from the solid residue in different organic solvents, water-

organic solvent mixtures and deep eutectic solvent (DES). The utilization of salty water

proved to enhance the hydrolysis rate in the first step significantly (full hydrolysis of the

hemicellulose fraction after 1 hour at 140 °C). The lignin yield obtained was lower than with

the original 1-step OrganoCat Process (ca. 3-6 wt% compared to ca. 12 wt% in the 1-step

OrganoCat Process, in relation to total wood weight), independent on the extraction solvent

used, which might be explained by lignin being exposed to the acidic medium longer than in

the biphasic system, and thus being prone to further irreversible (re)polymerizations. Finally,

the straightforward recycling of the reactive phase in the first step led to significantly

improved substrate-to-solvent and substrate-to-catalyst ratios (90 g L-1 xylose after total

addition of 400 g L-1 beech wood to 0.1 M oxalic acid in seawater within 4 cycles). Those

promising results display great opportunities for lignocellulose containing low amounts of

lignin (e.g. grasses, algae, etc.), efficiently and selectively hydrolyzing hemicellulose and

leaving cellulose for subsequent hydrolysis. Furthermore, sugar effluents not being

contaminated with organic solvents might favor application in nutritional and pharmaceutical

processes.

After showing two promising concepts for the fractionation of lignocellulose various

strategies for the valorization of the effluents cellulose and hemicellulose sugars were

presented, focusing on the application of integrated biorefinery concepts. For the salt-assisted

organic-acid catalyzed depolymerization of microcrystalline cellulose[133] seawater can be

used as natural source of salt, improving the hydrolysis rate and substituting more exhaustable

potable water reservoirs. As another method to reduce crystallinity of cellulose, pretreatment

(e.g. dissolution and reprecipitation) with ionic liquids (e.g. [DMIm]Acetate) was shown to

significantly improve the hydrolysis rate of the organic-acid as well as the salt-assisted

organic-acid catalyzed depolymerization of -cellulose.

As an example for the enzymatic hydrolysis of cellulose, the commercially available cellulase

cocktail Accellerase® 1500 was characterized and shown to efficiently depolymerize

microcrystalline cellulose (Avicel®) at optimum conditions (e.g. 20 g L-1 substrate loading,

0.1 vol% enzyme compared to solvent, 50 °C). Furthermore, hydrolysis of Avicel® cellulose

65

was performed removing the liquid product mixture and adding new enzyme solution in

multiple cycles, thus avoiding product inhibition of the catalyst. As a result, particles of high

crystallinity were obtained which could not be hydrolyzed anymore, and might find

application as defined-size cellulose particles, when controlling crystallinity appropriately. To

further evaluate the effect of the different pretreatments hydrolysis of cellulose samples from

the 1-step OrganoCat Process and the combination with mechanical pretreatment was

compared to the untreated biomass. In case of beech wood, after the 1-step OrganoCat

Process the enzymatic activity arose significantly compared to the non-pretreated sample,

which virtually showed no activity at all. Compared to Avicel® cellulose the activity was

slightly lower, presumably due to inhibition by residual lignin. However, based on those good

results, evolving cellulase and adapting it to the specific kind of biomass, hydrolysis

effectivity could be further improved and optimized. When combined with mechanical screw

press pretreatment of reed hydrolysis rate of the cellulose pulp was shown to be improved

greatly. The effect of the screw press on the 1-step OrganoCat Process is not yet well

understood and has to be investigated further. However, considerations involving an extended

surface area and a decrease in cellulose crystallinity, analogous to ball milling treatment,[190]

are probable and might deliver promising synergies of mechanical pretreatment and the 1-step

OrganoCat Process. Furthermore, application of commercial cellulase (e.g.

Accellerase® 1500) using seawater to prepare the buffer was successfully conducted, showing

the same high activity as when distilled water was used. Also in higher concentrated seawater

decent activity was observed.

To synthesize HMF from glucose a chemo-enzymatic concept conducted in seawater was put

forward, using IGI for the isomerization of glucose to fructose, followed by the oxalic acid

catalyzed triple-dehydration of fructose to HMF in a biphasic system. Herein, seawater was

used as source of salt, boosting the activity of the dehydration reaction while keeping

selectivity at a considerably high level. IGI proved to be very robust, efficiently working in

2-MeTHF saturated seawater over several cycles, and thus enabling the combination of both

reaction steps for multiple cycles, adding new glucose for the isomerization and yielding

HMF in the dehydration. Those examples indicate great opportunity for integrated processes

at coastal areas, using salt-water based biomass (e.g. algae) and converting it directly in their

natural medium seawater. Additionally, utilization of the waste stream of drinking water

production (e.g. concentrated seawater) might further enhance cost-effectiveness of both

processes and make them more benign and integrated.

66

Finally, the direct utilization of the hemicellulose effluent from the 1-step OrganoCat Process

for subsequent conversion into platform chemicals was shown. In the iron-catalyzed

dehydration of xylose to furfural in a bio-based biphasic system[155] the crude effluent

afforded 3.5 g L-1 h-1 under non-optimized conditions, which could be straightforward

improved by higher xylose concentration in the effluent. Also the fermentation with U.

maydis showed first promising results in the production of itaconic acid from glucose,

obtained from enzymatic hydrolysis of cellulose in seawater, and from xylose, derived from

beech wood in the 1-step OrganoCat Process.[180] Avoiding an expensive downstream

processing unit for the sugars, both chemical and biochemical approaches reveal promising

success in the utilization of the crude effluents coming from the pretreatment units.[155,180]

67

5. EXPERIMENTAL PART

5.1. Chemicals

Avicel® (PH-101), -cellulose, Amorphous cellulose (phosphoric-acid-swollen-cellulose,

PASC), fructose, glucose, xylose, oxalic acid, phosphoric acid, acetic acid, citric acid, maleic

acid, hydrochloric acid, lactic acid, betaine, sodium chloride, magnesium sulfate, iron(III)

chloride hexahydrate, calcium chloride, sodium hydroxide, dipotassium hydrogen phosphate,

potassium dihydrogen phosphate, trisodium citrate, glucose isomerase Sweetzyme IT Extra

(5.1.3.5; ≥300 mg-1), Glucose (HK) Assay Kit (GAHK20), reagents for PAHBAH method,

1-butanol, ethanol, and 2-MeTHF were purchased from Sigma–Aldrich. Beech wood, reed

and spruce were obtained from local suppliers and dried until constant weight. Mate tea leaves

were obtained in supermarket. After tea consumption, leaves were dried for several weeks and

used for fractionation systems. Seawater was collected in Bandol (France), and stored at 4 °C

until use. Accellerase® 1500 was kindly donated by Genencor International Corporation. It is

an enzyme cocktail containing several enzymes with endoglucanase activity

(2200-2800 CMC U g-1; CMC: carboxymethylcellulose), and -glucosidase activity

(525-775 pNPG U g-1; pNPG: p-nitrophenyl glucoside), with an optimum pH of 4.5.

5.2. General Procedures

5.2.1. Standard procedure for the 1-step OrganoCat Process

Glass-made high pressure reactor

In a glass-made high pressure reactor (Ace pressure tube, 15 mL or 80 mL) 2-MeTHF (5 mL

or 20 mL) and the same volume of aqueous oxalic acid solution with the indicated

concentration (0.1-0.2 M, referring to the aqueous phase) were introduced. 100 g L-1

(referring to the aqueous phase, e.g. 500 mg or 2 g) lignocellulose and a stirring bar were

added and the reactor was closed tight. The suspension was stirred at room temperature for

20 min and then heated to the indicated reaction temperature for the indicated time.

After reaching desired reaction time the reactor was cooled down to room temperature and

opened. The two phases were separated, 2-MeTHF was removed over reduced pressure and

lignin was weighed and analyzed via NMR. The aqueous phase with solid residue was

68

centrifuged at 4000 rpm for 5 min. Then, 2 mL of the aqueous phase were filtered with a

syringe filter (Multoclear-13 PTFE, 0.45 µm) and analyzed in HPLC. The solid cellulose pulp

was filtered off with a filter paper (MN 615), washed neutral with distilled water and then

dried.

Standard procedure for the metal-based high pressure reactor

In a stainless steel high pressure reactor (10 mL or 20 mL) 2-MeTHF (5 mL or 10 mL) and

the same volume of aqueous oxalic acid solution with the indicated concentration (0.1-0.2 M,

referring to the aqueous phase) were introduced. 100 g L-1 (referring to the aqueous phase,

e.g. 500 mg or 1 g) lignocellulose and a stirring bar were added. The indicated amount of CO2

was pressed on and the reactor was closed tight. The suspension was stirred for 20 min at

room temperature and then heated to the indicated reaction temperature for the indicated time.

After reaching desired reaction time the reactor was cooled down to room temperature and the

high pressure reactor was vented carefully to release CO2. The workup procedure was

analogous to the reaction in the glass-made high pressure reactor.

Procedure for the 5L metal-based high pressure reactor

In a glass inlay 100 g L-1 beech wood lignocellulose (150 g) was suspended in water with

0.1 M oxalic acid (1.5 L). As the organic phase, 2-MeTHF (1.5 L) was added, the glass inlay

was inserted into the metal-based high-pressure reactor and the reactor was closed. The

reactor was heated to 140 °C for 3 h. After cooling of the reactor, the organic phase was

separated by decantation. The aqueous phase was filtered to isolate the cellulose-pulp, and

xylose and glucose concentrations were determined in the aqueous phase. The solid residue

was washed with distilled water until neutral pH, dried until constant weight, extracted with

2-MeTHF (1.5 L) at room temperature for 24 h and then filtrated and the solid residue dried

again until constant weight. The organic phases were collected, and 2-MeTHF was evaporated

to obtain the lignin fraction.

5.2.2. Standard procedure for the 2-step OrganoCat Process

In a glass-made high pressure reactor (Ace pressure tube, 15 mL or 80 mL) 100 g L-1

lignocellulose was suspended in seawater with oxalic acid (0.1 M). The temperature was set at

140 °C for 3 h reaction time. After cooling off the reactor, the aqueous phase was filtered to

isolate the cellulose-pulp, and xylose and glucose concentrations were determined in the

69

aqueous phase via HPLC. The solid residue was washed with distilled water until neutral pH

and dried until constant weight.

Dried pulp obtained was suspended (final concentration 100 g L-1) in the respective solvent

(2-MeTHF / distilled water, 1:1 v/v; 2-MeTHF / seawater, 1:1 v/v; 2-MeTHF / 0.1 M oxalic

acid, 1:1 v/v; 2-MeTHF; EtOH; 1-BuOH; DES, lactic acid / betaine, 2:1 mol/mol), and stirred

at the indicated temperature for the indicated time. After cooling and filtration the organic

phase was decanted (in case of mixtures with water), and the organic solvent was removed via

distillation to obtain the lignin fraction. In case of the DES after filtration lignin was

precipitated by addition of water, filtrated and redissolved in 2-MeTHF. The organic solvent

was then removed via distillation to obtain the lignin fraction.

5.2.3. Procedure for the organic acid-catalyzed cellulose hydrolysis in seawater

Slurries of Avicel® cellulose (20 g L-1) were suspended in distilled water, brine, seawater and

5-fold concentrated seawater, in which 0.1 M maleic acid or oxalic acid were aggregated.

Reactions were magnetically stirred at 125 °C for 6 h in a refluxing system (106 °C as

effective temperature). At 0, 1, 4 and 6 h samples of 0.3 mL were taken and analyzed with the

PAHBAH method.

5.2.4. Standard procedure for the cellulose hydrolysis with Accellerase® 1500

Preparation of citrate-buffer in distilled water

44 mmol citric acid and 52 mmol of trisodium citrate were added to 900 mL of distilled water,

leading to a pH of 4.5 (0.1 M buffer). The buffer solution was stored at 4 °C until use.

Preparation of citrate-buffer in seawater

12 mmol of citric acid and 14 mmol of trisodium citrate were added to 250 mL of seawater,

leading to a pH of 3.8. Trisodium citrate was further added (ca. 1 mmol) until pH was

equilibrated at 4.5 (0.1 M buffer). The buffer solution was stored at 4 °C until use.

Hydrolysis procedure

Cellulose samples (15-40 g L-1) were suspended in citrate buffer in a round flask with a

stirring bar. The suspension was heated to 50 °C and the indicated amount of

Accellerase® 1500 was introduced. 0.3 mL samples were taken at indicated times, heated to

100 °C for 5 min and then analyzed with the Glucose (HK) Assay Kit.

70

5.2.5. Procedure for the synthesis of HMF from glucose

Preparation of phosphate buffer

In 1 L distilled water 7.4897 g K2HPO4, 0.9526 g KH2PO4, and 0.3000 mg MgSO4 were

dissolved under stirring at room temperature. The solution was stored at 4 °C until use.

Standard procedure for the isomerization of D-glucose to D-fructose with IGI

Solutions of glucose (140 g L-1) were prepared in phosphate buffer and in 2-MeTHF-saturated

seawater. In a closed glass reactor 5 mL reaction phase and 3 wt% (referring to glucose) of

immobilized glucose isomerase (IGI) were added. The reaction mixture was shaken in an

Eppendorf Thermomixer Comfort at 600 rpm and 60 °C for the indicated time. After cooling

the reactor was opened, the liquid phase was decanted and analyzed with Glucose (HK) Assay

Kit.

Standard procedure for the synthesis of HMF from fructose

Solutions of D-fructose (72 g L-1) were prepared in indicated aqueous solvents, in which

oxalic acid (0.2 M) was aggregated. In a closed glass reactor a mixture of 2.5 mL reaction

phase and 2.5 mL 2-MeTHF phase was magnetically stirred for 1 h at 140 °C. After cooling

the reactor was opened, the organic phase was decanted and analyzed via HPLC.

5.2.6. Procedure for the synthesis of furfural from 1-step OrganoCat Process hemicellulose

effluent

The aqueous effluent from the 1-step OrganoCat Process (see section 0 for details) contained

ca. 30 g L-1 xylose (determined in the aqueous phase together with small amounts of other

carbohydrates after 1-step OrganoCat Process). Oxalic acid was removed by adding 2 mmol

CaCl2 to 20 mL of the effluent, stirring for 1 h at room temperature, and filtrating. To 2.5 mL

of this solution, FeCl3 6 H2O (0.12 M, related to the total volume) was added together with

NaCl (30 wt%, related to the aqueous phase). Then 2.5 mL 2-MeTHF was added and the

reaction mixture was stirred for 2 h at 140 °C. After cooling the reactor was opened, the

organic phase was decanted and analyzed via GC.

5.2.7. Procedure for fermentation of sugars

The experiments were conducted by Tobias Klement at the Bioverfahrenstechnik (AVT, Prof.

Dr. J. Büchs, RWTH Aachen University), following the protocol of Klement et al.[180]

71

5.2.8. Procedure for the mechanical pretreatment with a screw press

The experiments were conducted by Qingqi Yan at the Aachener Verfahrenstechnik (AVT,

Prof. Dr. M. Modigell, RWTH Aachen University), following the protocol of Yan et al.[181]

5.3. Analysis

5.3.1. High Performance Liquid Chromatography (HPLC)

HPLC measurements of glucose and xylose were carried out on a Jasco HPLC equipped with

a SUGARSH1011 column with a 0.01 wt% aqueous acid solution as the eluent. The flow rate

was set to 0.6 mL min-1 and samples of 30 µL were injected. Amounts of xylose and glucose

present in the hemicellulose fraction were determined based on calibration curves built using

commercially available authentic substrates.

Figure 40. Calibration curves of xylose and glucose for HPLC measurement. Retention times: xylose 13.2 min,

glucose 12.2 min.

HPLC measurements of HMF were carried out on a Jasco HPLC equipped with a

Kromasil 100 C18 column with a 10 wt% aqueous methanol solution as the eluent. The flow

rate was set to 0.8 mL min-1 and samples of 50 µL were injected. The amount of HMF present

in 2-MeTHF was determined based on calibration curves built using commercially available

authentic substrates.

72

Figure 41. Calibration curve of HMF for HPLC measurement. Retention time HMF 9.8 min.

5.3.2. Gas Chromatography (GC)

GC measurements to quantify the concentration of furfural in the 2-MeTHF phase were

conducted with a 50 m OV1-IVA column, nitrogen as carrier gas, a split ratio of 33:1, and a

flame-ionization detector. The initial temperature was 50 °C, raised at 8 °C min-1 to 250 °C.

Quantification was done using n-decane and n-dodecane as internal standards.

73

5.3.3. Analysis with Glucose (HK) Assay Kit

10 µL of the sample were diluted with 990 µL distilled water. After shaking for 1 min 50 µL

of the diluted sample were transferred with an Eppendorf pipette into a microtiter plate well.

Then 200 µL of Glucose (HK) Assay Kit were added and absorption at 340 nm was measured

with a UV/VIS spectrometer (BioTek PowerWave HT) until absorption was constant.

Figure 42. Calibration curve for the Glucose (HK) Assay Kit.

74

5.3.4. Analysis with PAHBAH method

5 g p-hydroxy benzoic acid hydrazide (PAHBAH) were added to 30 mL distilled water and

slurried. Then 5 mL of concentrated HCl were added and the volume filled to 100 mL with

distilled water (solution A). In a second flask 12.45 g trisodium citrate were dissolved in

500 mL distilled water, 1.10 g CaCl2 and 20.00 g NaOH were added. Then the volume was

adjusted to 1 L with distilled water (solution B). 10 mL of solution A were added to 90 mL of

solution B, stirred and stored on ice freshly for every analysis (PAHBAH-reagent).[225]

50 µL sample were mixed with 100 µL PAHBAH-reagent, heated to 100 °C for 10 min and

then absorption at 410 nm was measured with a UV/VIS spectrometer (BioTek PowerWave

HT).

Figure 43. Calibration curve for the PAHBAH-method.

75

REFERENCES

[1] T. Werpy, G. Petersen, A. Aden, J. Bozell, Top Value Added Chemicals From Biomass.

Volume I-Results of Screening for Potential Candidates From Sugars and Synthesis

Gas, 2004.

[2] J. Bozell, J. Holladay, D. Johnson, J. White, Top Value Added Chemicals from

Biomass-Volume II: Results of Screening for Potential Candidates from Biorefinery

Lignin., 2007.

[3] J. J. Bozell, Clean Soil Air Water 2008, 36, 641–647.

[4] J. Leboreiro, A. K. Hilaly, Bioresour. Technol. 2011, 102, 2712–2723.

[5] P. M. Foley, E. S. Beach, J. B. Zimmerman, Green Chem. 2011, 13, 1399–1405.

[6] P. T. Anastas, Chem.rev. 2007, 107, 2167–2168.

[7] P. Anastas, N. Eghbali, Chem. Soc. Rev. 2010, 39, 301–312.

[8] K. Sanderson, Nature 2011, 474, S12–S14.

[9] B. Kamm, M. Kamm, M. Schmidt, T. Hirth, M. Schulze, in Biorefineries-Industrial

Process. Prod. Status Quo Futur. Dir. (Eds.: B. Kamm, P.R. Gruber, M. Kamm),

Wiley-VCH Verlag GmbH, Weinheim, Germany, 2008.

[10] C. E. Wyman, B. Yang, Calif. Agr. 2009, 64, 185–190.

[11] C. Somerville, H. Youngs, C. Taylor, S. C. Davis, S. P. Long, Science (80-. ). 2010,

329, 790–792.

[12] A. Wisogel, S. Tyson, D. Johnson, in Handb. Bioethanol Prod. Util. (Ed.: Ch.E.

Wyman), Taylor & Francis, Washington D.C., 1996, pp. 105–116.

[13] Y. Sun, J. Cheng, Bioresour. Technol. 2002, 83, 1–11.

[14] V. L. Finkenstadt, R. P. Millane, Macromolecules 1998, 31, 7776–7783.

[15] R. Rinaldi, F. Schüth, ChemSusChem 2009, 2, 1096–1107.

[16] Y.-H. Percival Zhang, M. E. Himmel, J. R. Mielenz, Biotechnol. Adv. 2006, 24, 452–

481.

[17] F. Hu, A. Ragauskas, Bioenerg. Res. 2012, 5, 1043–1066.

[18] S. P. S. Chundawat, G. Bellesia, N. Uppugundla, L. da Costa Sousa, D. Gao, A. M.

Cheh, U. P. Agarwal, C. M. Bianchetti, G. N. Phillips, P. Langan, et al., J. Am. Chem.

Soc. 2011, 133, 11163–11174.

76

[19] B. C. Saha, J. Ind. Microbiol. Biot. 2003, 30, 279–291.

[20] C. E. Wyman, Annu. Rev. Energ. Env. 1999, 24, 189–226.

[21] A. Ebringerová, Z. Hromádková, T. Heinze, Adv. Polym. Sci. 2005, 186, 1–67.

[22] J. Zakzeski, P. C. A. Bruijnincx, A. L. Jongerius, B. M. Weckhuysen, Chem.rev. 2010,

110, 3552–3599.

[23] L. B. Davin, N. G. Lewis, Curr. Opin. Biotech. 2005, 16, 407–415.

[24] K. Ruel, F. Barnoud, D. A. I. Goring, Wood Sci. Technol. 1978, 12, 287–291.

[25] T. Hatakeyama, T. Yoshida, H. Hatakeyama, Tenth Int. Symp. Wood Pulping Chem.

1999, 1, 478–481.

[26] M. Dolk, F. Pla, J. F. Yan, J. L. McCarthy, Macromolecules 1986, 19, 1464–1470.

[27] W. J. Connors, S. Sarkanen, J. L. McCarthy, Holzforschung 1980, 34, 80–85.

[28] J. M. Cruz, H. Domı́nguez, J. C. Parajó, Food Chem. 2005, 90, 503–511.

[29] C. Pouteau, P. Dole, B. Cathala, L. Averous, N. Boquillon, Polym. Degrad. Stabil.

2003, 81, 9–18.

[30] L. R. C. Barclay, F. Xi, J. Q. Norris, J. Wood Chem.Technol. 1997, 17, 73–90.

[31] C. Rice-Evans, N. Miller, G. Paganga, Trends iPlant Sci. 1997, 2, 152–159.

[32] X. Pan, J. F. Kadla, K. Ehara, N. Gilkes, J. N. Saddler, J. Agric. Food. Chem. 2006, 54,

5806–5813.

[33] M. B. Hocking, J. Chem. Educ. 1997, 74, 1055–1059.

[34] S. Ramachandra Rao, G. Ravishankar, J. Sci. Food Agr. 2000, 80, 289–304.

[35] J. McMillan, in Pretreat. Lignocellul. Biomass Enzym. Convers. Biomass Fuels Prod.

ACS Symp. Ser. (Eds.: M.E. Himmel, J.O. Baker, R.P. Overend), American Chemical

Society, Washington, DC, 1994, pp. 292–324.

[36] H. L. Chum, D. K. Johnson, S. Black, J. Baker, K. Grohmann, K. V Sarkanen, K.

Wallace, H. a Schroeder, Biotechnol. Bioeng. 1988, 31, 643–649.

[37] R. W. Thring, E. Chornet, R. P. Overend, Biomass 1990, 23, 289–305.

[38] X. Zhao, K. Cheng, D. Liu, Appl. Microbiol. Biot. 2009, 82, 815–827.

[39] M. Mascal, E. B. Nikitin, ChemSusChem 2010, 3, 1349–1351.

[40] J. Li, G. Henriksson, G. Gellerstedt, Bioresour. Technol. 2007, 98, 3061–3068.

77

[41] S. M. Ewanick, R. Bura, J. N. Saddler, Biotechnol. Bioeng. 2007, 98, 737–746.

[42] P. J. Morjanoff, P. P. Gray, Biotechnol. Bioeng. 1987, 29, 733–741.

[43] M. T. Holtzapple, a E. Humphrey, J. D. Taylor, Biotechnol. Bioeng. 1989, 33, 207–

210.

[44] T. A. Clark, K. L. Mackie, J. Wood Chem.Technol. 1987, 7, 373–403.

[45] W. R. Grous, A. O. Converse, H. E. Grethlein, Enzym. Microb. Tech. 1986, 8, 274–

280.

[46] K. L. Mackie, H. H. Brownell, K. L. West, J. N. Saddler, J. Wood Chem.Technol.

1985, 5, 405–425.

[47] M. Mes-Hartree, B. E. Dale, W. K. Craig, Appl. Microbiol. … 1988, 29, 462–468.

[48] D. M. Fox, J. W. Gilman, A. B. Morgan, J. R. Shields, P. H. Maupin, R. E. Lyon, H. C.

De Long, P. C. Trulove, Ind. Eng. Chem. Res. 2008, 47, 6327–6332.

[49] M. Smiglak, W. M. Reichert, J. D. Holbrey, J. S. Wilkes, L. Sun, J. S. Thrasher, K.

Kirichenko, S. Singh, A. R. Katritzky, R. D. Rogers, Chem. Commun. 2006, 2554–

2556.

[50] P. Wasserscheid, T. Welton, Eds., Ionic Liquids in Synthesis, Wiley-VCH Verlag

GmbH & Co. KGaA, Weinheim, Germany, 2007.

[51] S. Tan, D. MacFarlane, Top. Curr. Chem. 2010, 290, 311–339.

[52] P. Wasserscheid, W. Keim, Angew. Chem. Int. Ed. 2000, 39, 3772–3789.

[53] A. Bösmann, P. S. Schulz, P. Wasserscheid, Monatshefte für Chemie - Chem. Mon.

2007, 138, 1159–1161.

[54] P. Bonhôte, A.-P. Dias, N. Papageorgiou, K. Kalyanasundaram, M. Grätzel, Inorg.

Chem. 1996, 35, 1168–1178.

[55] E. B. Carter, S. L. Culver, P. a Fox, R. D. Goode, I. Ntai, M. D. Tickell, R. K. Traylor,

N. W. Hoffman, J. H. Davis, Chem. Commun. 2004, 630–631.

[56] P. Nockemann, B. Thijs, K. Driesen, C. R. Janssen, K. Van Hecke, L. Van Meervelt, S.

Kossmann, B. Kirchner, K. Binnemans, J. Phys. Chem. B 2007, 111, 5254–5263.

[57] A. P. Dadi, C. a Schall, S. Varanasi, Appl. Biochem. Biotech. 2007, 137-140, 407–421.

[58] R. P. Swatloski, S. K. Spear, J. D. Holbrey, R. D. Rogers, J. Am. Chem. Soc. 2002,

124, 4974–4975.

[59] C. Graenacher, Cellulose Solution, 1934, US1943176.

78

[60] Y. Su, H. M. Brown, G. Li, X. Zhou, J. E. Amonette, J. L. Fulton, D. M. Camaioni, Z.

C. Zhang, Appl. Catal. A-Gen. 2011, 391, 436–442.

[61] R. Rinaldi, R. Palkovits, F. Schüth, Angew. Chem. Int. Ed. 2008, 47, 8047–8050.

[62] L. Vanoye, M. Fanselow, J. D. Holbrey, M. P. Atkins, K. R. Seddon, Green Chem.

2009, 11, 390–396.

[63] Y. Pu, N. Jiang, A. J. Ragauskas, J. Wood Chem.Technol. 2007, 27, 23–33.

[64] I. Kilpeläinen, H. Xie, A. King, M. Granstrom, S. Heikkinen, D. S. Argyropoulos, J.

Agric. Food. Chem. 2007, 55, 9142–9148.

[65] D. a. Fort, R. C. Remsing, R. P. Swatloski, P. Moyna, G. Moyna, R. D. Rogers, Green

Chem. 2007, 9, 63–69.

[66] J. Viell, W. Marquardt, Chem. Ing. Tech. 2009, 81, 1213–1213.

[67] J. Viell, W. Marquardt, Holzforschung 2011, 65, 519–525.

[68] S. H. Lee, T. V Doherty, R. J. Linhardt, J. S. Dordick, Biotechnol. Bioeng. 2009, 102,

1368–1376.

[69] S. S. Y. Tan, D. R. MacFarlane, J. Upfal, L. a. Edye, W. O. S. Doherty, A. F. Patti, J.

M. Pringle, J. L. Scott, Green Chem. 2009, 11, 339–345.

[70] J. B. Binder, R. T. Raines, P. Natl. Acad. Sci. USA 2010, 107, 4516–4521.

[71] P. G. Jessop, D. J. Heldebrant, X. Li, C. A. Eckert, C. L. Liotta, Nature 2005, 436,

1102.

[72] M. J. Earle, J. M. S. S. Esperança, M. a Gilea, J. N. C. Lopes, L. P. N. Rebelo, J. W.

Magee, K. R. Seddon, J. a Widegren, Nature 2006, 439, 831–834.

[73] P. Wasserscheid, Nature 2006, 439, 797.

[74] A. E. Rosamilia, C. R. Strauss, J. L. Scott, Pure. Appl. Chem. 2007, 79, 1869–1877.

[75] U. P. Kreher, A. E. Rosamilia, C. L. Raston, J. L. Scott, C. R. Strauss, Molecules 2004,

9, 387–393.

[76] M. Francisco, A. van den Bruinhorst, M. C. Kroon, Green Chem. 2012, 14, 2153–2157.

[77] Z. Maugeri, P. Domínguez de María, RSC Adv. 2012, 2, 421–425.

[78] P. Domínguez de María, Z. Maugeri, Curr. Opin. Chem. Biol. 2011, 15, 220–225.

[79] Z. Maugeri, W. Leitner, P. Domínguez de María, Tetrahedron Lett. 2012, 53, 6968–

6971.

79

[80] C. Ruß, B. König, Green Chem. 2012, 14, 2969–2982.

[81] R. M. Hertel, A. S. Bommarius, M. J. Realff, Y. Kang, Deep Eutectic Solvent System

Comprising Betaine Monohydrate and Hydrogen Bond Donor, 2012, WO2012145522.

[82] M. M. Chang, T. Y. C. Chou, G. T. Tsao, Adv. Biochem. Eng. Biot. 1981, 20, 15–42.

[83] A. L. Demain, M. Newcomb, J. H. D. Wu, Microbiol. Mol. Biol. R. 2005, 69, 124–154.

[84] L. T. Fan, Y.-H. Lee, M. M. Gharpuray, Adv. Biochem. Eng. Biot. 1982, 23, 157–187.

[85] H. E. Grethlein, Biotechnol. Adv. 1984, 2, 43–62.

[86] K. W. Lin, M. R. Ladisch, D. M. Schaeffer, C. H. Noller, V. Lechtenberg, G. T. Tsao,

AIChE Symp. Ser. 1981, 203, 102–106.

[87] J. D. McMillan, ACS Sym. Ser. 1994, 566, 234–292.

[88] M. A. Millett, A. J. Baker, L. D. Satter, Biotechnol. Bioeng. Symp. 1976, 6, 125–153.

[89] N. Moreira, Sci. News. Online 2005, 168, 209–224.

[90] N. Mosier, C. Wyman, B. Dale, R. Elander, Y. Y. Lee, M. Holtzapple, M. Ladisch,

Bioresour. Technol. 2005, 96, 673–86.

[91] J. Weil, P. Westgate, K. Kohlmann, M. R. Ladisch, Enzym. Microb. Tech. 1994, 16,

1002–1004.

[92] C. E. Wyman, B. E. Dale, R. T. Elander, M. Holtzapple, M. R. Ladisch, Y. Y. Lee,

Bioresour. Technol. 2005, 96, 1959–1966.

[93] J. R. Cherry, A. L. Fidantsef, Curr. Opin. Biotech. 2003, 14, 438–443.

[94] H. Esterbauer, W. Steiner, I. Labudova, A. Hermann, M. Hayn, Bioresour. Technol.

1991, 36, 51–65.

[95] O. Kirk, T. V. Borchert, C. C. Fuglsang, Curr. Opin. Biotech. 2002, 13, 345–351.

[96] M. Holtzapple, M. Cognata, Y. Shu, C. Hendrickson, Biotechnol. Bioeng. 1990, 36,

275–287.

[97] A. V Gusakov, A. P. Sinitsyn, Biotechnol. Bioeng. 1992, 40, 663–671.

[98] P. Ghosh, N. B. Pamment, W. R. B. Martin, Enzym. Microb. Tech. 1982, 4, 425–430.

[99] J. D. Wright, C. E. Wyman, K. Grohmann, Appl. Biochem. Biotech. 1988, 18, 75–90.

[100] J.-W. Lee, R. C. L. B. Rodrigues, T. W. Jeffries, Bioresour. Technol. 2009, 100, 6307–

6311.

80

[101] L. C. Teixeira, J. C. Linden, H. A. Schroeder, Appl. Biochem. Biotech. 2000, 84-86,

111–128.

[102] C. Tengborg, M. Galbe, G. Zacchi, Biotechnol. Progr. 2001, 17, 110–117.

[103] N. Mosier, C. Wyman, B. Dale, R. Elander, Y. Y. Lee, M. Holtzapple, M. Ladisch,

Bioresour. Technol. 2005, 96, 673–686.

[104] M. J. Taherzadeh, K. Karimi, Bioresources 2007, 2, 707–738.

[105] R. Rinaldi, F. Schüth, ChemSusChem 2009, 2, 1096–1107.

[106] R. P. Chandra, R. Bura, W. E. Mabee, A. Berlin, X. Pan, J. N. Saddler, Adv. Biochem.

Eng. Biot. 2007, 108, 67–93.

[107] D. Klemm, B. Heublein, H.-P. Fink, A. Bohn, Angew. Chem. Int. Ed. 2005, 44, 3358–

3393.

[108] J. Tollefson, Nature 2008, 451, 880–883.

[109] J. B. Binder, J. J. Blank, A. V. Cefali, R. T. Raines, ChemSusChem 2010, 3, 1268–

1272.

[110] F. Schüth, R. Rinaldi, Method for the Depolymerization of Cellulose, 2009, WO

2009/115075 A1.

[111] T. Balensiefer, J. Brodersen, G. D’Andola, K. Massonne, S. Freyer, V. Stegmann,

Method for Producing Glucose by Hydrolysis of Cellulose That Can Be Pretreated with

an Ionic Liquid Containing a Polyatomic Anion., 2008, WO2008090155.

[112] M. Klemens, G. D’Andola, V. Stegmann, W. Mormann, M. Wezstein, W. Leng,

Method for Breaking down Cellulose in Solution, 2009, US2009062524.

[113] R. Rinaldi, R. Palkovits, F. Schüth, Angew. Chem. Int. Ed. 2008, 47, 8047–8050.

[114] P. Domínguez de María, A. Martinsson, Analyst 2009, 134, 493–496.

[115] A. Pinkert, K. N. Marsh, S. Pang, M. P. Staiger, Chem.rev. 2009, 109, 6712–6728.

[116] C. Graenacher, Cellulose Solution, 1934, US1943176.

[117] T. Heinze, A. Koschella, Polim. Cienc. Tecnol. 2005, 15, 84–90.

[118] P. L. Ragg, P. R. Fields, P. B. Tinker, Phil. Trans. R. Soc. Lond. A 1987, 321, 537–547.

[119] L. F. Chen, C.-M. Yang, Quantitative Hydrolysis of Cellulose to Glucose Using Zinc

Chloride, 1984, US4452640.

[120] S. A. Barker, P. J. Somers, A. J. Beardsmore, B. L. F. Rodgers, Solubilisation and

Hydrolysis of Cellulose-Containing Materials, 1983, EP 0096497.

81

[121] R. A. Penque, Method of Hydrolyzing Cellulose to Monosaccharides, 1977,

US4018620.

[122] E. C. Sherrard, W. H. Gauger, Ind. Eng. Chem. 1923, 15, 63–64.

[123] H. G. Deming, J. Am. Chem. Soc. 1911, 33, 1515–1525.

[124] U. Wongsiriwan, Y. Noda, C. Song, P. Prasassarakich, Y. Yeboah, Energ. Fuel. 2010,

24, 3232–3238.

[125] A. M. J. Kootstra, H. H. Beeftink, E. L. Scott, J. P. Sanders, Biotechnol. Biofuels 2009,

2, 31.

[126] J. P. M. Sanders, P. H. M. De Bot, A. M. J. Kootstra, Method for Treating Vegetable

Material with Acid as Well as Products Obtained with This Method, 2009,

WO2009145617.

[127] A. M. J. Kootstra, N. S. Mosier, E. L. Scott, H. H. Beeftink, J. P. M. Sanders, Biochem.

Eng. J. 2009, 43, 92–97.

[128] J.-W. Lee, R. C. L. B. Rodrigues, H. J. Kim, I.-G. Choi, T. W. Jeffries, Bioresour.

Technol. 2010, 101, 4379–4385.

[129] Y. Lu, N. S. Mosier, Biotechnol. Progr. 2007, 23, 116–123.

[130] N. S. Mosier, J. J. Wilker, M. R. Ladisch, Biotechnol. Bioeng. 2004, 86, 756–764.

[131] N. S. Mosier, C. M. Ladisch, M. R. Ladisch, Biotechnol. Bioeng. 2002, 79, 610–618.

[132] N. S. Mosier, A. Sarikaya, C. M. Ladisch, M. R. Ladisch, Biotechnol. Progr. 2001, 17,

474–480.

[133] T. vom Stein, P. M. Grande, F. Sibilla, U. Commandeur, R. Fischer, W. Leitner, P.

Domínguez de María, Green Chem. 2010, 12, 1844–1849.

[134] C. Jiménez-González, P. Poechlauer, Q. B. Broxterman, B.-S. Yang, D. am Ende, J.

Baird, C. Bertsch, R. E. Hannah, P. Dell’Orco, H. Noorman, et al., Org. Process Res.

Dev. 2011, 15, 900–911.

[135] D. J. C. Constable, C. Jimenez-Gonzalez, R. K. Henderson, Org. Process Res. Dev.

2007, 11, 133–137.

[136] R.-J. van Putten, J. C. van der Waal, E. de Jong, C. B. Rasrendra, H. J. Heeres, J. G. de

Vries, Chem.rev. 2013, 113, 1499–1597.

[137] J. J. Bozell, G. R. Petersen, Green Chem. 2010, 12, 539–554.

[138] G. W. Huber, S. Iborra, A. Corma, Chem.rev. 2006, 106, 4044–4098.

[139] A. Corma, S. Iborra, A. Velty, Chem.rev. 2007, 107, 2411–2502.

82

[140] R. Palkovits, Angew. Chem. Int. Ed. 2010, 49, 4336–4338.

[141] F. M. A. Geilen, B. Engendahl, A. Harwardt, W. Marquardt, J. Klankermayer, W.

Leitner, Angew. Chem. Int. Ed. 2010, 49, 5510–5514.

[142] Sigma-Aldrich, 2-methyltetrahydrofuran Mater. Saf. Data Sheet [MSDS] 2012,

Retrived from http://www.sigmaaldrich.com.

[143] V. Antonucci, J. Coleman, J. B. Ferry, N. Johnson, M. Mathe, J. P. Scott, J. Xu, Org.

Process Res. Dev. 2011, 15, 939–941.

[144] R. K. Henderson, C. Jiménez-González, D. J. C. Constable, S. R. Alston, G. G. a.

Inglis, G. Fisher, J. Sherwood, S. P. Binks, A. D. Curzons, Green Chem. 2011, 13,

854–862.

[145] B. Comanita, S. Chemicals, Spec. Chem. Mag. 2006.

[146] V. Pace, P. Hoyos, L. Castoldi, P. Domínguez de María, A. R. Alcántara,

ChemSusChem 2012, 5, 1369–1379.

[147] D. F. Aycock, Org. Process Res. Dev. 2007, 11, 156–159.

[148] F. H. Newth, Adv. Carbohyd. Chem. 1951, 6, 83–106.

[149] C. J. Moye, Rev. Pure Appl. Chem. 1964, 14, 161–170.

[150] M. S. Feather, J. F. Harris, Adv. Carbohyd. Chem. Bi. 1973, 28, 161–224.

[151] B. F. M. Kuster, Starch - Stärke 1990, 42, 314–321.

[152] J. Lewkowski, ARKIVOC 2001, i, 17–54.

[153] N. Wierckx, F. Koopman, H. J. Ruijssenaars, J. H. de Winde, Appl. Microbiol. Biot.

2011, 92, 1095–1105.

[154] D. M. Alonso, J. Q. Bond, J. A. Dumesic, Green Chem. 2010, 12, 1493–1513.

[155] T. vom Stein, P. M. Grande, W. Leitner, P. Domínguez de María, ChemSusChem 2011,

4, 1592–1594.

[156] F. Carvalheiro, G. Garrote, J. C. Parajó, H. Pereira, F. M. Gírio, Biotechnol. Progr.

2005, 21, 233–243.

[157] J.-P. Lange, E. van der Heide, J. van Buijtenen, R. Price, ChemSusChem 2012, 5, 150–

166.

[158] A. P. Dunlop, Ind. Eng. Chem. 1948, 40, 204–209.

[159] D. L. Williams, A. P. Dunlop, Ind. Eng. Chem. 1948, 40, 239–241.

83

[160] M. R. Nimlos, X. Qian, M. Davis, M. E. Himmel, D. K. Johnson, J. Phys. Chem. A

2006, 110, 11824–11838.

[161] T. Willke, K.-D. Vorlop, Appl. Microbiol. Biot. 2001, 56, 289–295.

[162] M. Okabe, D. Lies, S. Kanamasa, E. Y. Park, Appl. Microbiol. Biot. 2009, 84, 597–

606.

[163] S. Baup, Ann. Pharm. 1836, 19, 29–38.

[164] R. Bentley, C. Thiessen, J. Biol. Chem. 1957, 226, 673–687.

[165] N. Winskillm, Microbiology 1983, 129, 2877–2883.

[166] P. Bonnarme, B. Gillet, a M. Sepulchre, C. Role, J. C. Beloeil, C. Ducrocq, J.

Bacteriol. 1995, 177, 3573–3578.

[167] T. vom Stein, Organic-Acid-Catalysed Selective Fractionation of Lignocellulose,

Diplomarbeit, 2010.

[168] P. Domínguez de María, T. vom Stein, P. M. Grande, F. Sibilla, W. Leitner, Integrated

Process for the Selective Fractionation and Separation of Lignocellulose in Its Main

Components., 2011, EP 11154705.5.

[169] A. M. J. Kootstra, H. H. Beeftink, E. L. Scott, J. P. M. Sanders, Biochem. Eng. J. 2009,

46, 126–131.

[170] A. M. J. Kootstra, H. H. Beeftink, E. L. Scott, J. P. Sanders, Biotechnol. Biofuels 2009,

2, 31.

[171] N. S. Mosier, J. J. Wilker, M. R. Ladisch, Biotechnol. Bioeng. 2004, 86, 756–764.

[172] N. Li, G. a. Tompsett, T. Zhang, J. Shi, C. E. Wyman, G. W. Huber, Green Chem.

2011, 13, 91–101.

[173] J. A. Vaselenak, I. E. Grossmann, A. W. Westerberg, Ind. Eng. Chem. Process Des.

Dev. 1986, 25, 357–366.

[174] P. Dominguez de Maria, H. Kayser, F. Rodríguez-Ropero, W. Leitner, M. Fioroni, RSC

Adv. 2013, DOI 10.1039/c3ra41307a.

[175] H. Kayser, Mechanistic Studies on Organic Acid Catalyzed Hydrolysis of Glycosidic

Bonds, Diplomarbeit, 2010.

[176] E. L. Shock, Am. J. Sci. 1995, 295, 496–580.

[177] C. I. Heck, E. G. de Mejia, J. Food Sci. 2007, 72, R138–R151.

84

[178] C. M. Pagliosa, K. N. de Simas, R. D. M. C. Amboni, A. N. N. Murakami, C. L. O.

Petkowicz, J. de D. Medeiros, A. C. Rodrigues, E. R. Amante, Ind. Crop. Prod. 2010,

32, 428–433.

[179] T. E. Timell, Wood Sci. Technol. 1967, 1, 45–70.

[180] T. Klement, S. Milker, G. Jäger, P. M. Grande, P. Domínguez de María, J. Büchs,

Microb. Cell Fact. 2012, 11, 43.

[181] Q. Yan, M. Modigell, Chem. Eng. Trans. 2012, 29, 601–606.

[182] M. Pidwirny, “Introduction to the Oceans”. Fundamentals of Physical Geography,

2006.

[183] C. S. K. Lin, R. Luque, J. H. Clark, C. Webb, C. Du, Energ. Environ. Sci. 2011, 4,

1471–1479.

[184] A. Johansson, O. Aaltonen, P. Ylinen, Biomass 1987, 13, 45–65.

[185] M. Taherzadeh, K. Karimi, Bioresources 2007, 2, 472–499.

[186] B. Van der Bruggen, C. Vandecasteele, Desalination 2002, 143, 207–218.

[187] B. Zhao, L. Greiner, W. Leitner, RSC Adv. 2012, 2, 2476–2479.

[188] B. Zhao, Syntheses and Applications of Ionic Liquids as Solvents and Reactants, PhD

Thesis, 2012.

[189] A. Berlin, M. Balakshin, N. Gilkes, J. Kadla, V. Maximenko, S. Kubo, J. Saddler, J.

Biotechnol. 2006, 125, 198–209.

[190] L. T. Fan, Y. Lee, D. H. Beardmore, Biotechnol. Bioeng. 1980, 22, 177–199.

[191] P. Väljamäe, G. Pettersson, G. Johansson, Eur. J. Biochem. 2001, 268, 4520–4526.

[192] M. Hall, P. Bansal, J. H. Lee, M. J. Realff, A. S. Bommarius, FEBS J. 2010, 277,

1571–1582.

[193] P. Bansal, B. J. Vowell, M. Hall, M. J. Realff, J. H. Lee, A. S. Bommarius, Bioresour.

Technol. 2012, 107, 243–250.

[194] A. Oren, J. Ind. Microbiol. Biot. 2002, 28, 56–63.

[195] S. Grant, D. Y. Sorokin, W. D. Grant, B. E. Jones, S. Heaphy, Extremophiles 2004, 8,

421–429.

[196] C. Liang, Y. Xue, M. Fioroni, F. Rodríguez-Ropero, C. Zhou, U. Schwaneberg, Y. Ma,

Appl. Microbiol. Biot. 2011, 89, 315–326.

85

[197] J. Pottkämper, P. Barthen, N. Ilmberger, U. Schwaneberg, A. Schenk, M. Schulte, N.

Ignatiev, W. R. Streit, Green Chem. 2009, 11, 957–965.

[198] D. M. Alonso, J. Q. Bond, J. A. Dumesic, Green Chem. 2010, 12, 1493–1513.

[199] P. Domínguez de María, ChemSusChem 2011, 4, 327–329.

[200] G. W. Huber, J. N. Chheda, C. J. Barrett, J. a Dumesic, Science (80-. ). 2005, 308,

1446–1450.

[201] B. Kamm, P. R. Gruber, M. Kamm, Biorefineries — Industrial Processes and

Products. Status Quo and Future Directions, Wiley-VCH, Weinheim, 2010.

[202] A. A. Rosatella, S. P. Simeonov, R. F. M. Frade, C. A. M. Afonso, Green Chem. 2011,

13, 754–793.

[203] X. Tong, Y. Ma, Y. Li, Appl. Catal. A-Gen. 2010, 385, 1–13.

[204] F. M. A. Geilen, T. vom Stein, B. Engendahl, S. Winterle, M. A. Liauw, J.

Klankermayer, W. Leitner, Angew. Chem. Int. Ed. 2011, 50, 6831–6834.

[205] W. Xu, Q. Xia, Y. Zhang, Y. Guo, Y. Wang, G. Lu, ChemSusChem 2011, 4, 1758–

1761.

[206] J. Julis, M. Hölscher, W. Leitner, Green Chem. 2010, 12, 1634–1639.

[207] S. Lima, M. M. Antunes, M. Pillinger, A. a. Valente, ChemCatChem. 2011, 3, 1686–

1706.

[208] M. E. Zakrzewska, E. Bogel-Łukasik, R. Bogel-Łukasik, Chem.rev. 2011, 111, 397–

417.

[209] A. D. Patel, J. C. Serrano-Ruiz, J. A. Dumesic, R. P. Anex, Chem. Eng. J. 2010, 160,

311–321.

[210] Y. Román-Leshkov, C. J. Barrett, Z. Y. Liu, J. a Dumesic, Nature 2007, 447, 982–985.

[211] H. C. Silberman, J. Org. Chem. 1961, 26, 1967–1969.

[212] T. J. Christian, M. Manley-Harris, R. J. Field, B. a Parker, J. Agric. Food. Chem. 2000,

48, 1823–1837.

[213] B. Kim, J. Jeong, S. Shin, D. Lee, S. Kim, H.-J. Yoon, J. K. Cho, ChemSusChem 2010,

3, 1273–1275.

[214] J. Speck, Adv. Carbohyd. Chem. 1958, 13, 63–103.

[215] S. Angyal, Glycoscience 2001, 215, 1–14.

[216] S. H. Bhosale, M. B. Rao, V. V Deshpande, Microbiol. Rev. 1996, 60, 280–300.

86

[217] T. Ståhlberg, J. M. Woodley, A. Riisager, Catal. Sci. Technol. 2012, 2, 291.

[218] R. Huang, W. Qi, R. Su, Z. He, Chem. Commun. 2010, 46, 1115–1117.

[219] R. Dicosimo, J. McAuliffe, A. J. Poulose, G. Bohlmann, Chem. Soc. Rev. 2013, DOI:

10.1039/c3cs35506c.

[220] Y. B. Tewari, Appl. Biochem. Biotech. 1990, 23, 187–203.

[221] K. Visuri, A. M. Klibanov, Biotechnol. Bioeng. 1987, 30, 917–920.

[222] A. S. Mamman, J.-M. Lee, Y.-C. Kim, I. T. Hwang, N.-J. Park, Y. K. Hwang, J.-S.

Chang, J.-S. Hwang, Biofuels, Bioprod. Bioref. 2008, 2, 438–454.

[223] G. P. Chiusoli, US3025320: Process for Preparing Itaconic Acid, and 2, 3-Butadienoic

Acid, 1962.

[224] T. vom Stein, P. M. Grande, H. Kayser, F. Sibilla, W. Leitner, P. Domínguez de María,

Green Chem. 2011, 13, 1772–1777.

[225] M. Lever, Anal. Biochem. 1972, 47, 273–279.

87

LIST OF PATENTS & PUBLICATIONS

[1] P. Domínguez de María, T. vom Stein, P.M. Grande, F. Sibilla, W. Leitner,

EP 11154705.5, 2011, Integrated process for the selective fractionation and

separation of lignocellulose in its main components.

[2] T. Klement, S. Milker, G. Jäger, P.M. Grande, P. Domínguez de María, J. Büchs,

Microb. Cell Fact. 2012, 11, 43, Biomass pretreatment affects Ustilago maydis in

producing itaconic acid.

[3] P.M. Grande, C. Bergs, P. Domínguez de María, ChemSusChem 2012, 5, 1203-1206,

Chemo-enzymatic conversion of glucose into 5-hydroxymethylfurfural in seawater.

[4] P.M. Grande, P. Domínguez de María, Bioresource Technol. 2012, 104, 799-802,

Enzymatic hydrolysis of microcrystalline cellulose in concentrated seawater.

[5] T. vom Stein, P.M. Grande, W. Leitner, P. Domínguez de María, ChemSusChem 2011,

4, 1592-1594, Iron-catalyzed furfural production in biobased biphasic systems: from

pure sugars to direct use of crude xylose effluents as feedstock.

[6] T. vom Stein, P.M. Grande, H. Kayser, F. Sibilla, W. Leitner, P. Domínguez de María,

Green Chem. 2011, 13, 1772-1777, From biomass to feedstock: one-step fractionation

of lignocellulose components by the selective organic acid-catalyzed depolymerization

of hemicellulose in a biphasic system.

[7] T. vom Stein, P. Grande, F. Sibilla, U. Commandeur, R. Fischer, W. Leitner,

P. Domínguez de María, Green Chem. 2010, 12, 1844-1849, Salt-assisted

organic-acid-catalyzed depolymerization of cellulose.