UNIVERSITY OF OULU P .O. Box 8000 F I -90014 UNIVERSITY OF OULU FINLAND

A C T A U N I V E R S I T A T I S O U L U E N S I S

University Lecturer Tuomo Glumoff

University Lecturer Santeri Palviainen

Postdoctoral research fellow Sanna Taskila

Professor Olli Vuolteenaho

University Lecturer Veli-Matti Ulvinen

Planning Director Pertti Tikkanen

Professor Jari Juga

University Lecturer Anu Soikkeli

Professor Olli Vuolteenaho

Publications Editor Kirsti Nurkkala

ISBN 978-952-62-1513-6 (Paperback)ISBN 978-952-62-1514-3 (PDF)ISSN 0355-3213 (Print)ISSN 1796-2226 (Online)

U N I V E R S I TAT I S O U L U E N S I SACTAC

TECHNICA

U N I V E R S I TAT I S O U L U E N S I SACTAC

TECHNICA

OULU 2017

C 608

Anni Keränen

WATER TREATMENTBY QUATERNIZED LIGNOCELLULOSE

UNIVERSITY OF OULU GRADUATE SCHOOL;UNIVERSITY OF OULU,FACULTY OF TECHNOLOGY

C 608

ACTA

Anni K

eränen

C608etukansi.kesken.fm Page 1 Wednesday, February 22, 2017 11:39 AM

A C T A U N I V E R S I T A T I S O U L U E N S I SC Te c h n i c a 6 0 8

ANNI KERÄNEN

WATER TREATMENT BY QUATERNIZED LIGNOCELLULOSE

Academic dissertation to be presented with the assent ofthe Doctoral Training Committee of Technology andNatural Sciences of the University of Oulu for publicdefence in Kuusamonsali (YB210), Linnanmaa, on 31March 2017, at 12 noon

UNIVERSITY OF OULU, OULU 2017

Copyright © 2017Acta Univ. Oul. C 608, 2017

Supervised byProfessor Juha TanskanenDoctor Tiina Leiviskä

Reviewed byDoctor Md. Rabiul AwualProfessor Abel Navarro

ISBN 978-952-62-1513-6 (Paperback)ISBN 978-952-62-1514-3 (PDF)

ISSN 0355-3213 (Printed)ISSN 1796-2226 (Online)

Cover DesignRaimo Ahonen

JUVENES PRINTTAMPERE 2017

OpponentProfessor Amit Bhatnagar

Keränen, Anni, Water treatment by quaternized lignocellulose University of Oulu Graduate School; University of Oulu, Faculty of TechnologyActa Univ. Oul. C 608, 2017University of Oulu, P.O. Box 8000, FI-90014 University of Oulu, Finland

Abstract

Water-related problems are increasing globally, and new, low-cost technologies are needed toresolve them. Lignocellulosic waste materials contain reactive functional groups that can be usedto provide a bio-based platform for the production of water treatment chemicals. Research on bio-based ion exchange materials in the treatment of real wastewaters is needed.

In this thesis, anion exchange materials were prepared through chemical modification(epichlorohydrin, ethylenediamine and triethylamine) using five Finnish lignocellulosic materialsas bio-based platforms. Scots pine sawdust and bark (Pinus sylvestris), Norway spruce bark (Piceaabies), birch bark (Betula pendula/pubescens) and peat were chosen due to their local availabilityand abundance. The focus was placed on NO3- removal, but uptake of heavy metals, such asnickel, was also observed and studied. Studies on maximum sorption capacity, mechanism,kinetics, and the effects of temperature, pH and co-existing anions were used to elucidate thesorption behaviour of the prepared materials in batch and column tests.

All five materials removed over 70% of NO3- at pH 3–10 (initial conc. 30 mg N/l). Quaternized

pine sawdust worked best (max. capacity 32.8 mg NO3-N/g), and also in a wide temperature range(5–70°C). Column studies on quaternized pine sawdust using mining wastewater and industrialwastewater from a chemical plant provided information about the regeneration of exhaustedmaterial and its suitability for industrial applications. Uptake of Ni, V, Co and U was observed.Column studies proved the easy regeneration and reusability of the material.

For comparison, pine sawdust was also modified using N-(3-chloro-2-hydroxypropyl)trimethylammonium chloride and utilized to remove NO3

- from groundwater and industrialwastewater. A maximum sorption capacity of 15.3 mg NO3-N/g was achieved for the syntheticsolution. Overall, this thesis provides valuable information about bio-based anion exchangematerials and their use in real waters and industrial applications.

Keywords: ion exchange, lignocellulose, nickel, nitrate, vanadium, wastewater

Keränen, Anni, Kationisoitu lignoselluloosa vesienkäsittelyssä Oulun yliopiston tutkijakoulu; Oulun yliopisto, Teknillinen tiedekuntaActa Univ. Oul. C 608, 2017Oulun yliopisto, PL 8000, 90014 Oulun yliopisto

Tiivistelmä

Edullisia ja kestäviä vedenkäsittelytekniikoita tarvitaan kasvavien vesiongelmien ratkaisemi-seen. Lignoselluloosaa, kuten sahanpurua, syntyy suuria määriä teollisuuden sivutuotteena. Senreaktiivisia funktionaalisia ryhmiä voidaan modifioida kemiallisesti ja valmistaa siten biopohjai-sia vedenkäsittelykemikaaleja. Tutkimustietoa oikeiden jätevesien puhdistuksesta biopohjaisillaioninvaihtomateriaaleilla tarvitaan lisää, jotta materiaalien käyttöä voidaan kehittää ja edistää.

Tässä väitöstyössä valmistettiin anioninvaihtomateriaaleja modifioimalla kemiallisesti viittäsuomalaista lignoselluloosamateriaalia: männyn sahanpurua ja kuorta (Pinus sylvestris), kuusenkuorta (Picea abies), koivun kuorta (Betula pendula/pubescens) ja turvetta. Menetelmässä käy-tettiin epikloorihydriiniä, etyleenidiamiinia ja trietyyliamiinia orgaanisessa liuotinfaasissa. Työs-sä keskityttiin erityisesti nitraatin poistoon sekä synteettisistä että oikeista jätevesistä. Materiaa-lien soveltuvuutta teollisiin sovelluksiin arvioitiin maksimisorptiokapasiteetin, sorptioisotermi-en, kinetiikka- ja kolonnikokeiden sekä pH:n, lämpötilan ja muiden anionien vaikutusta tutkivi-en kokeiden avulla.

Kaikki viisi kationisoitua tuotetta poistivat yli 70 % nitraatista laajalla pH-alueella (3–10).Kationisoitu männyn sahanpuru osoittautui parhaaksi materiaaliksi (32,8 mg NO3-N/g), ja se toi-mi laajalla lämpötila-alueella (5–70°C). Kolonnikokeet osoittivat sen olevan helposti regeneroi-tavissa ja uudelleenkäytettävissä. Tuotetta testattiin myös kaivos- ja kemiantehtaan jätevedenkäsittelyyn, ja kokeissa havaittiin hyviä nikkeli-, uraani-, vanadiini- ja kobolttireduktioita. Män-nyn sahanpurua modifioitiin vertailun vuoksi myös kationisella monomeerilla, N-(3-kloro-2-hydroksipropyyli)trimetyyliammoniumkloridilla. Tuotteen maksimisorptiokapasiteetiksi saatiin15,3 mg NO3-N/g ja se poisti nitraattia saastuneesta pohjavedestä. Kokonaisuudessaan väitöskir-jatyö tarjoaa uutta tietoa biopohjaisten ioninvaihtomateriaalien valmistamisesta ja niiden sovel-tuvuudesta oikeiden teollisuusjätevesien käsittelyyn.

Asiasanat: ioninvaihto, jätevesi, lignoselluloosa, nikkeli, nitraatti, vanadiini

7

Acknowledgements

This research was conducted in the Chemical Process Engineering research group

at the University of Oulu in 2012–2017. The thesis work was kindly funded by the

University of Oulu, the Maj and Tor Nessling foundation, the Anna Esteri Timola

foundation, the Tauno Tönning foundation, Maa- ja vesitekniikan tuki ry and the

Waste Management Association JHY.

This thesis was supervised by Prof. Juha Tanskanen and Dr. Tiina Leiviskä,

who provided me with support and guidance in my early academic career. I also

thank all the other co-authors, Prof. Osmo Hormi (Department of Chemistry,

University of Oulu), Prof. Bao-Yu Gao (Shandong University, China), Prof. Marina

V. Frontasyeva and Ms. Inga Zinicovscaia (Joint Institute for Nuclear Research,

Russia) for their valuable contribution to the publications included in this work.

For the pre-examination of this thesis, I would like to acknowledge the pre-

examiners Dr. Md. Rabiul Awual (Japan Atomic Energy Agency, Japan) and Prof.

Abel Navarro (Borough of Manhattan Community College, USA) for their valuable

comments to this work. I also thank Mike Jones and Sue Pearson at Pelc Southbank

Languages Ay for the proofreading of the thesis and the papers included within.

A special acknowledgment is extended to my colleagues in the Chemical

Process Engineering research group, who have been an endless source of joy and

inspiration in the course of my work. Particularly, I want to thank Marja, Juha,

Johanna and Jenna for all the fun we have had together and the discussions that

have helped me carry on with my work. With an honest heart, I can say that this

thesis would not have been finished without your priceless support. Thank you,

Jenna, for becoming my best friend in the course of gym classes and other follies

we have had! I also thank my friends, family and my lovely partner Markus for

their invaluable support and understanding for this enormous project.

Iloasia puhuttu ja juttu hupaisa oli!

Oulu, February 2017 Anni Keränen

8

9

List of symbols and abbreviations

Latin symbols

C concentration [mg/l]

m mass [g]

Q sorption capacity [mg/g]

V volume [l]

Y mass yield [%]

Subscripts

0 initial state

e equilibrium

max maximum

t at time t

Abbreviations

BET Brunauer-Emmett-Teller (surface area analysis)

CA citric acid

CA-PSD pine sawdust treated with citric acid

CAC-PSD pine sawdust treated with citric acid and CHMAC

CFA continuous flow analysis

CHMAC N-(3-chloro-2-hydroxypropyl) trimethylammonium chloride

C-PSD pine sawdust treated with CHMAC

CTAB cetyltrimethyl ammonium bromide

DMF N,N-dimethylformamide

EDA ethylenediamine

EDM a modification method involving epichlorohydrin and

dimethylamine

EDS energy dispersive X-ray spectroscopy

EPI epichlorohydrin

ETM a modification (quaternization) method involving epichlorohydrin

and triethylamine

ETM-PSD ETM-modified pine sawdust

10

FAAS flame atomic absorption spectroscopy

FESEM field emission scanning electron microscopy

FTIR Fourier transform infrared spectroscopy

GFAAS graphite furnace atomic absorption spectroscopy

GTMAC glycidyl trimethylammonium chloride

IC ion chromatography

ICP-MS inductively coupled plasma mass spectroscopy

ICP-OES inductively coupled plasma optical emission spectroscopy

LC lignocellulose, lignocellulosic

PSD pine sawdust

SHP sodium hypophosphite

TEA triethylamine

11

List of original publications

This thesis is based on the following publications, which are referred throughout

the text by their Roman numerals:

I Keränen A, Leiviskä T, Gao BY, Hormi O & Tanskanen J (2013) Preparation of novel anion exchangers from pine sawdust and bark, spruce bark, birch bark and peat for the removal of nitrate. Chemical Engineering Science 98: 59–68.

II Keränen A, Leiviskä T, Hormi O & Tanskanen J (2015) Removal of nitrate by modified pine sawdust: Effects of temperature and co-existing anions. Journal of Environmental Management 147: 46–54.

III Keränen A, Leiviskä T, Hormi O & Tanskanen J (2015) Preparation of cationized pine sawdust for nitrate removal: Optimization of reaction conditions. Journal of Environmental Management 160: 105–112.

IV Keränen A, Leiviskä T, Zinicovscaia I, Frontasyeva MV, Hormi O & Tanskanen J (2016) Quaternized pine sawdust in the treatment of mining wastewater. Environmental Technology 37(11): 1390–1397.

V Leiviskä T, Keränen A, Vainionpää N, Al Amir J, Hormi O & Tanskanen J (2015) Vanadium(V) removal from aqueous solution and real wastewater using quaternized pine sawdust. Water Science and Technology 72(3): 437–442.

The author, Anni Keränen, was the main and corresponding author of Papers I–IV.

The author conducted most of the experimental work, and was responsible for the

design of the experimental work together with the second author as well as writing

the manuscripts in Papers I–IV. The modelling in Paper III was carried out by the

author. Paper V is an extended compilation of two Bachelor’s theses published by

the University of Oulu. The present author contributed by supervising the thesis

work and preparing the quaternized pine sawdust used in the experimental work.

The co-authors in all the papers participated in the preparation of the papers by

providing valuable comments and consultation.

12

13

Table of contents

Abstract

Tiivistelmä

Acknowledgements 7 List of symbols and abbreviations 9 List of original publications 11 Table of contents 13 1 Introduction 15

1.1 Aim of the study ...................................................................................... 16 1.2 Outline of the thesis ................................................................................ 17 1.3 Definition of sorption .............................................................................. 18

2 Lignocellulose in water treatment 19 2.1 Anion removal ........................................................................................ 20

2.1.1 Weak base anion exchangers ........................................................ 20 2.1.2 Strong base anion exchangers ...................................................... 22 2.1.3 Quaternary monomers .................................................................. 24 2.1.4 Crosslinking .................................................................................. 25

2.2 Cation removal ........................................................................................ 27 3 Materials and methods 29

3.1 Chemical modification ............................................................................ 30 3.1.1 Amination using the ETM method ............................................... 31 3.1.2 Amination using CHMAC ............................................................ 33 3.1.3 Citric acid treatment ..................................................................... 35

3.2 Sorption studies ....................................................................................... 37 3.2.1 Batch shaking experiments ........................................................... 39 3.2.2 Fixed bed column studies ............................................................. 40 3.2.3 Real waters ................................................................................... 40

3.3 Analysis methods .................................................................................... 42 4 Results and discussion 45

4.1 Characterization of prepared materials ................................................... 45 4.1.1 ETM method ................................................................................. 45 4.1.2 CHMAC treatment ....................................................................... 46 4.1.3 Citric acid treatment ..................................................................... 49

4.2 Sorption studies ....................................................................................... 51 4.2.1 Effect of sorbent dose ................................................................... 51 4.2.2 Effect of pH .................................................................................. 51

14

4.2.3 Maximum sorption capacity ......................................................... 51 4.2.4 Effect of contact time ................................................................... 53 4.2.5 Effect of temperature .................................................................... 54 4.2.6 Effect of phosphate and sulphate .................................................. 55 4.2.7 Fixed bed column studies ............................................................. 58

4.3 Real waters .............................................................................................. 59 4.3.1 Nitrate-contaminated groundwater (C-PSD) ................................ 59 4.3.2 Mining wastewater (ETM-PSD) ................................................... 60 4.3.3 Ammoniacal industrial wastewater (ETM-PSD) .......................... 62

4.4 Summary ................................................................................................. 63 5 Economic feasibility 65 6 Conclusions 69 References 71 Original publications 79

15

1 Introduction

Global interest in bio-based chemicals has exploded during the past decade. A vast

amount of lignocellulosic waste is produced annually around the world, consisting

mostly of various types of crop residues from agriculture, and sawdust and wood

chips from the forestry industry. Lignocellulose (LC) provides an excellent source

of feedstock for the production of bio-based chemicals thanks to its versatile

structure and reactive functional groups, such as phenolic hydroxyl groups.

Environmental awareness and developing technologies have boosted the use of

biomass to replace fossil raw materials in the production of fuels, chemicals,

packaging and products. Among the variety of promising biochemicals are water

treatment chemicals.

Water-related conflicts are expected to become a serious problem for mankind

in the near future. Increasing use of fresh water resources, pollution, and climate

change threaten access to clean water. Droughts have become an ever more frequent

problem, not just in the Middle East and Africa, but even in the industrially

developed California (U.S. Geological Survey 2016). Pollution of water systems

poses a risk to the vulnerable aquatic ecosystems. Industrial activities such as

mining produce large volumes of wastewater, which require treatment. In Finland,

mining industry is a major pollution source of heavy metals and nutrients. Nitrate

pollution also occurs regionally due to agricultural activities.

Among the various forms of water pollution, nitrate pollution is an emerging

problem worldwide. Excessive use of fertilizers, untreated manure and leaking

septic tanks cause nitrate to leak into the groundwater and surface water, causing

eutrophication and posing a health risk to people, especially infants, through

methaemoglobinaemia (Bouwer 1989 in Chabani et al. 2006). Elevated nitrate

concentrations in groundwater have also been found in Central Europe (European

Environmental Agency 2015). A maximum concentration for nitrate in drinking

water has been set at 50 mg/l (11.3 mg/l as NO3-N), as recommended by the World

Health Organization (2011), the European Union (Council Directive 91/676/EEC

1991) and Finnish legislation (Ministry of Social Affairs and Health 2014). Other

anions, such as phosphate, sulphate and anionic metals like arsenate, chromate and

vanadate, also pose risks to the aquatic environment in terms of excess nutrient load,

salination, toxic effects and bioaccumulation. To tackle water pollution, cost-

efficient water treatment techniques must be readily available where needed.

Various methods are available for the removal of nitrate, such as biological

treatment (Ilies & Mavinic 2001), reverse osmosis (Häyrynen et al. 2008),

16

electrodialysis (Hell et al. 1998), adsorption (Öztürk & Bektaş 2004) and ion

exchange (Ebrahimi & Roberts 2013). Every method has its advantages and

disadvantages, which put them in competition with each other. Ion exchange was

chosen here because of its easy operation and the reusability of the material. It is a

widely used method for drinking water production for instance in the United States

(Jensen et al. 2012). Present ion exchange resins are oil-derived products and very

expensive, which increases the investment costs. Using less expensive raw

materials for the production of ion exchange resins could bring the costs down.

Low-cost, bio-based water treatment chemicals have been widely researched

over the past few decades (Laszlo 1996, Low & Lee 1997, Gao et al. 2009,

Anirudhan et al. 2012). Lignocellulosic waste materials are available worldwide,

thus offering a local, abundant, renewable and cheap raw material for the

production of water treatment chemicals. It has been estimated that some 30 million

metric tons and over 340 million metric tons of straw are produced in Germany and

China annually, although most of it is currently burnt to produce energy (Liu 2013,

Weiser et al. 2014). In Finland, an estimated 600 000 tons of sawdust is produced

each year but is mostly burned or even dumped in a landfill (Korpinen 2010).

Recent plans in Finland have aimed at the production of bio-ethanol from sawdust

(Aamulehti 2014). Biodiesel is produced from waste tall oil by UPM (Mannonen

2015). However, the use of LC as a feedstock in biorefineries as a platform for

high-value bio-products instead of bulk chemicals (e.g. ethanol) would be

economically and ecologically justified.

1.1 Aim of the study

The aims of this work were to prepare bio-based anion exchange materials using

Finnish lignocellulosic raw materials, to develop more benign production methods

and study their application in the removal of nitrate and other contaminants from

real waters. This kind of approach is novel because the focus of most research in

this field so far has been on the treatment of synthetic solutions.

Five native Finnish LC materials—Scots pine sawdust and bark (Pinus

sylvestris), Norway spruce bark (Picea abies), birch bark (Betula pubescens/

Betula verrucosa), and peat—were chosen as raw materials because of their local

availability and abundance. The use of local waste materials was focused on

because they will play a key role in the production of bio-based chemicals

worldwide. In addition, very little information was found about the use of bio-based

anion exchange materials in the treatment of real wastewaters in general. The

17

majority of the studies on the use of quaternized LC in water treatment found in

literature deal with the basic phenomena of sorption, elucidating kinetics,

mechanisms, optimum sorption conditions, and focusing on the treatment of

synthetic solutions. Synthetic solutions are needed to produce replicable data on

basic sorption parameters, such as maximum sorption capacity, but real waters are

needed in order to find applications for these new materials. Studies on real waters

are essential if the materials are gradually to take their place aside the current

treatment technologies. This work was conducted since no research was available

on the use of bio-based anion exchange materials in the treatment of real

wastewaters or nitrate-contaminated groundwater.

1.2 Outline of the thesis

This thesis consists of five papers published in scientific journals. Paper I covers

the preparation of anion exchange materials using five different Finnish

lignocellulosic raw materials through the ETM modification method (employing

epichlorohydrin and triethylamine) and nitrate removal studies using synthetic

nitrate solutions. The prepared materials were compared with each other in terms

of sorption performance under varying conditions and the best performing material

was chosen for further studies. Paper II continued this work with studies on the

effects of temperature and co-existing anions using ETM-modified pine sawdust

(ETM-PSD).

In order to find alternative, less hazardous modification methods for

quaternization, another method employing N-(3-chloro-2-hydroxypropyl)

trimethylammonium chloride (CHMAC) and NaOH was applied to pine sawdust

in Paper III, and proved simpler than the one used in Paper I. Optimized reaction

conditions were determined using design of experiments, including an empirical

model. Sorption tests on nitrate-contaminated groundwater were performed in

addition to tests on synthetic nitrate solutions.

To gain more information about the possible applications of quaternized pine

sawdust, real wastewaters were also treated. ETM-PSD—as prepared in Paper I—

was used for the treatment of mining wastewater in batch and fixed bed ion

exchange column studies in Paper IV. Ammoniacal industrial wastewater and the

use of ETM-PSD were studied in the removal of nickel and vanadium in Paper V.

Based on the good results in nickel removal obtained in Paper IV and in

vanadium removal in Paper V, efforts were made to produce a bio-based hybrid ion

exchange material capable of removing both anions and cations. A combination of

18

two different, simple and less hazardous modification methods was used: an

anionizing citric acid treatment, and the cationizing CHMAC method, which was

utilized and optimized in Paper III. The citric acid method was seen as an

alternative, safer modification method for the production of bio-based water

treatment chemicals. However, these tests remained at a preliminary stage and the

results are referred to as “unpublished data”.

1.3 Definition of sorption

Sorption phenomena include several mechanisms, such as adsorption, absorption

and ion exchange (Naja & Volesky 2011) for instance, and they can be classified in

different ways depending on the approach.

In this thesis, sorption has been used collectively to describe all possible

sorption mechanisms, like ion exchange and adsorption, which in terms of word

choice are considered and referred to here as the prevalent mechanisms. Ion

exchange was assumed to be the main mechanism based on what was known about

the reaction routes in chemical modification. Nevertheless, the presence of other

attraction mechanisms has not been ruled out because of the complex structure of

the modified materials.

19

2 Lignocellulose in water treatment

Lignocellulose is the most abundant biopolymer in the world and it is found in all

plant material. It is composed of cellulose, hemicellulose, lignin, extractives and

inorganic material usually referred to as ash. The exact distribution of components

varies depending on the plant species. In Scots pine (Pinus sylvestris) wood, for

instance, the composition is 40–49% cellulose, 27–29% hemicelluloses, 24–28%

lignin, and 3–5% extractives, such as resins, waxes and terpenes, and less than 1%

of ash (Paper I, Politi & Sidiras 2012, Sable et al. 2012, Räisänen & Athanassiadis

2013, Korpinen 2010).

Lignocellulose is an excellent platform for the production of various types of

biochemicals. The versatile structure of LC provides both a source of C-5 and

C-6 sugars in the form of cellulose and hemicelluloses as well as the only bio-based

source of aromatic compounds in the form of lignin (Lange et al. 2013). There are

several methods for the extraction of valuable intermediate products, biofuels and

fine chemicals from LC through destructive treatments, such as pyrolysis,

gasification, and enzymatic and acid hydrolysis. An interesting approach, though

very different in terms of products, is the use of non-destructive treatment methods,

which are covered in this thesis. The focus in these methods is not in degrading the

LC structure into its sub-units but rather in the utilization of the LC as such and of

its numerous functional groups, which can be used for the direct chemical

modification of the LC structure. The functional groups include hydroxyls,

carboxyls and methoxy groups, among others, and they allow the modification of

LC surface properties through the addition of chemical substituents and sidechains

on the surface. This enables the production of novel, bio-based materials, such as

ion exchange materials, without the need for separating specific components from

the LC. This approach could also give rise to new applications for surface-modified

LC in fields other than water treatment.

Commercial ion exchange resins are widely used for water purification in

potable water production, the pharmaceutical, food and beverage industries,

precious metal recovery, desalination, and wastewater treatment. They are synthetic

oil-derived co-polymers, such as styrene-divinylbenzene, with functional groups

suitable for either anion or cation removal. LC with its many reactive functional

groups could provide a bio-based platform for the production of ion exchange

materials through the addition of ion exchange functionality bearing groups. Due

to their bio-based backbone, they could be regarded as a greener alternative in

comparison to conventional ion exchange resins. A low-cost raw material could

20

also lower production costs. The conversion of wastes into products would be a

significant step towards a circular economy.

2.1 Anion removal

Lignocellulose inherently possesses a negative surface charge (Nachtergaele 1989),

which causes anions to repel its surface. The anion removal capacity of native LC

is thus very weak (Hena et al. 2015). To facilitate anion removal, the surface of LC

must be cationized by attaching cationic groups onto it. As discussed above, this is

enabled by the reactive functional groups in the structure of LC. Amination

occurring through the use of a variety of nitrogen-bearing amine compounds is the

most commonly used cationization method for LC (Chattopadhyay 2001), due to

the fact that the electron configuration of nitrogen allows the formation of a cation.

Amination is based on the introduction of amine (ammonium) groups in the

structure of LC, and is used in this thesis to refer to the attachment of secondary,

tertiary or quaternary ammonium groups to the LC structure. In the case of

quaternary ammonium groups, the specific term quaternization is used in this thesis

as in most similar cases in the literature.

Various types of LC materials and sorbents derived therefrom through

amination reactions have been studied around the world in the removal of anions

such as nitrate and phosphate (Orlando et al. 2002a, Wang et al. 2010). The most

common amination methods are reviewed here divided into methods producing

weak or strong base anion exchangers. In addition, the use of quaternary monomers

is also reviewed. However, as amination is a versatile method with various

applications—for instance in the production of the cationized starch needed in the

paper industry or in surface modification of cotton textiles—not all the methods

presented here have been used directly in the preparation of anion exchange

materials. Sorption capacities on various anions in synthetic solutions are given for

materials applied in anion removal where available.

2.1.1 Weak base anion exchangers

The addition of tertiary ammonium groups to LC results in the formation of a weak

base anion exchanger. The nitrogen in a tertiary (or secondary) ammonium group

has a lone electron pair which is capable of binding a proton. This means that the

ion exchange properties are pH-dependent. A weak base anion exchange material

21

works best at below pH 6–7. (Nollet 2000). The molecular structures of some of

the amine compounds used in the referenced studies are presented in Fig. 1.

Fig. 1. Amine compounds used for the introduction of secondary and tertiary

ammonium groups for the production of weak base anion exchangers.

Šimkovic & Laszlo (1997) used epichlorohydrin together with ammonium

hydroxide or imidazole to produce weak base anion exchangers from sugar-cane

bagasse for the removal of two anionic dyes, and they achieved capacities of

3.3 mmol/g for Alizarin Red S and 1.1 mmol/g for Brilliant Red F3B. Orlando et

al. (2003) modified various LC materials with thionyl chloride, N,N-dimethyl-

formamide, dimethylamine and formaldehyde and obtained the highest capacity for

nitrate with modified Moringa oleifera hull (2.8 mg/g for NO3-N). Anirudhan et al.

(2012) treated cellulose with epichlorohydrin, polyethyleneimine and N,N-

methylene bisacrylamide in the presence of α,α’-azobisisobutyro nitrile for

phosphate removal (optimal pH 4.5). In their later study on the same material, they

22

reported a sorption capacity of 35.8 mg/g for NO3-N (Anirudhan & Rauf 2013).

Anirudhan & Unnithan (2007) prepared a weak base anion exchanger by treating

coconut coir pith with epichlorohydrin, dimethylamine and HCl, and reported a

sorption capacity of 13.6 mg/g for As and an optimal pH of 6–8. Orlando et al.

(2002a) used a similar method: epichlorohydrin and dimethylamine with a pyridine

catalyst in N,N-dimethylformamide to add tertiary ammonium groups to sugar-cane

bagasse (19.8 mg/g for NO3-N) and rice hull (18.5 mg/g for NO3-N). This method

known as the EDM method (employing epichlorohydrin and dimethyl or

diethylamine) has been used on various LC materials (Orlando et al. 2002b, Wang

et al. 2007, Katal et al. 2012). Katal et al. (2012) reported the optimal pH for bio-

based weak base anion exchangers to be 6–8.

2.1.2 Strong base anion exchangers

The quaternary ammonium nitrogen possesses a permanent positive charge

regardless of the pH of the surrounding solution due to the lack of a lone electron

pair (Biermann 1996). This feature gives the sorbent material the properties of a

strong base anion exchanger, which means that the ion exchange functionality is

sustained over a wide pH range.

Dizhbite et al. (1999) used epichlorohydrin, diethylamine and cetyltrimethyl

ammonium bromide (CTAB, Fig. 2) to synthesize a strong base anion exchanger

from acid hydrolysis lignins for use as sorbents for bile acids and cholesterol. The

EDM method was modified by Gao et al. (2009) in order to produce strong base

anion exchangers, i.e. by introducing quaternary ammonium groups into LC by

replacing dimethylamine with triethylamine, a tertiary amine. This method was

referred to as the ETM method (employing epichlorohydrin and trimethyl or

triethylamine). The ETM method has been a popular method among numerous

research teams for the quaternization of various LC materials and several

approaches have been reported (Xu et al. 2011, Xu et al. 2010d). Wang et al. (2010)

determined the optimum reaction conditions for the quaternization of giant reed:

2.5 ml epichlorohydrin, 1.25 ml N,N-dimethylformamide, 0.5 ml ethylenediamine

and 2.5 ml triethylamine per one gram of giant reed and they attained a nitrogen

content of 7.8%. The nitrogen content of a modified sample has been used in this

field as an indicator for the addition of quaternary ammonium groups, which

correlates well with the anion exchange capacity (Pepper et al. 1953, Gao et al.

2009). Xu et al. (2011) reported the use of three different chemical dosages in ETM

modification along with respective chemical costs.

23

Fig. 2. Amine compounds used for the introduction of quaternary ammonium groups

for the production of strong base anion exchangers.

Studies on the ETM method have long remained in laboratory scale without any

effort toward practical applications and production in a larger scale. In 2013, Xu et

al. published their studies on pilot scale modification tests. A comparison with

samples prepared in laboratory scale showed that the pilot scale products performed

slightly weaker. One of the few research papers dealing with the treatment of real

waters with bio-based anion exchange materials was published by Xu et al. (2012a).

They treated three different nitrate-containing waters (two samples of surface water

from China and effluent from a Chinese wastewater treatment plant) with ETM-

modified Arundo donax L. reed (Xu et al. 2012a). Uptakes of 70–98% for nitrate,

24

49–94% for nitrite, about 95% for phosphate, 92–99% for sulphate and 83% for As

were reported (Xu et al. 2012a).

2.1.3 Quaternary monomers

Quaternary monomers are defined here as compounds which inherently possess a

quaternary ammonium group (Fig. 2). Unlike in the ETM method, the quaternary

groups are thus attached directly to the LC structure without the need to first

synthesize them from an amine compound.

The cationic monomers presented in Fig. 2 are available commercially and are

widely used in the production of cationized products used in numerous industrial

applications, such as cationized starch and cellulose for the paper industry

(Nachtergaele 1989). N-(3-chloro-2-hydroxypropyl) trimethyl ammonium chloride

(CHMAC, Fig. 2) contains a quaternary ammonium group with methyl substituents

and a chloride counter ion. Antal et al. (1984) were among the first to use CHMAC

in the presence of NaOH to obtain a cationized product from beech sawdust (Fagus

silvatica) but the possible applications were not specified at that time. Šimkovic et

al. (1987) continued this research by treating spruce wood meal with CHMAC.

Later, Laszlo (1996), Low & Lee (1997), Marshall & Wartelle (2004),

Namasivayam & Höll (2005) and Wartelle and Marshall (2006) used CHMAC in

the treatment of aspen wood meal, rice husk, soybean hulls, bagasse, peanut shells

and corn stover, among others, in an attempt to prepare strong base anion exchange

materials for the removal of As, Cr, P, Se and anionic dyes. The following sorption

capacities have been achieved: 0.55 mmol/g for Reactive Red 180 (Laszlo 1996),

39 mg/g for As (Marshall & Wartelle 2004), 1.7 mg/g for NO3-N (Namasivayam &

Höll 2005) and 82.2 mg/g for Cr (Wartelle & Marshall 2006). The CHMAC method

has also been patented in the preparation of anion exchange materials by Yang et

al. (2015). In the patent they describe the cationization of rice straw by CHMAC

and NaOH in isopropyl alcohol for use for water treatment (Yang et al. 2015).

During the synthesis, CHMAC reacts with NaOH, forming a reactive epoxy

group-containing glycidyl trimethyl ammonium (GTMAC, Fig. 2) intermediate,

which reacts with the hydroxyl groups of LC through the reactive epoxy ring.

GTMAC is also available as a commercial product and, like CHMAC, can be

applied directly in cationization. It is worth noting that epichlorohydrin can also be

reacted with trimethylamine forming an epoxyamine intermediate equivalent to

GTMAC, which further reacts with LC (Xu et al. 2010b). Kong et al. (2015)

cationized kraft lignin using GTMAC to be used as a flocculant for anionic dye

25

removal. Sowmya & Meenakshi (2014) treated chitosan with melamine,

glutaraldehyde and GTMAC for the removal of nitrate and phosphate. They

achieved sorption capacities of 6.8 mg/g for NO3-N and 10.4 mg/g for PO4-P

(Sowmya & Meenakshi 2014). The reactions of CHMAC and GTMAC with LC do

not involve crosslinking, unlike the ETM method, leading to a less pronounced

mass gain.

Raji & Anirudhan (1998) treated rubber wood sawdust with N,N-methylene-

bisacrylamide, potassium peroxydisulphate, acrylamide and ethylenediamine, and

the prepared polyacrylamide sawdust had a sorption capacity of 12.4 mg/g for

Cr(VI). Ansari et al. (2009) coated walnut wood sawdust with two electroactive

polymers, polypyrrole (with FeCl3 as a pretreatment) and polyaniline (with

(NH4)2S2O8 as a pretreatment), for nitrite removal.

Ionic liquids are salts with ambient melting points, and they have also been

used for the modification of LC. Abbott et al. (2006) and Karachalios & Wazne

(2013) prepared quaternized cellulose and pine bark through a treatment with

chlorocholine chloride (Fig. 2) and urea. Karachalios & Wazne (2013) used the

product in nitrate removal and reported a sorption capacity of 40.8 mg/g for

NO3-N and an optimal pH of 5–8.

2.1.4 Crosslinking

As presented in the previous section, many of the commonly used amination

methods involve the use of epichlorohydrin because of its crosslinking properties

and the ability to enhance the reactivity of LC (Wang et al. 2010), which gives it

an essential role in the quaternization reactions. Crosslinking is defined as creating

a link between the polymer chains through covalent or ionic bonds, which add the

rigidity of the structure and alter the physico-chemical properties (Steiger 1972).

Epichlorohydrin is one of the commonly used reagents for crosslinking. The above-

mentioned ETM method utilizes the crosslinking properties of epichlorohydrin

(Fig. 3), and produces a highly crosslinked, porous product with a mass gain of up

to 300–700% (Paper I). Although very hazardous for the environment and human

health, epichlorohydrin has maintained its position as a highly reactive and efficient

crosslinking agent.

Both the EDM and ETM methods involve crosslinking. Some argue that

crosslinking is required in the production of cationized LC for water treatment

purposes because it is claimed to decrease the solubility of the material (Laszlo

1996, Šimkovic 1999). Besides epichlorohydrin, it has also been proposed that

26

ethylenediamine acts as a crosslinking agent (Xu et al. 2010b) in these methods, in

addition to introducing secondary amine groups. Moreover, it is unclear whether

ethylenediamine is involved in a crosslink or in a side chain as we have proposed.

However, this question is in fact irrelevant with regard to functional amine groups

because the amine groups delivered by ethylenediamine will in any case form

secondary amine groups (see Fig. 5 on p. 32), which possess cation sorption

properties (Torres et al. 2006).

Fig. 3. Crosslink between two cellulose chains formed by epichlorohydrin (left) and

ethylenediamine (right). Adapted from Yoshida et al. (2014).

Other compounds besides epichlorohydrin are available for the crosslinking of LC.

The suitability of formaldehyde, dialdehydes, diepoxides, diisocyanates, divinyl

compounds, methylolated nitrogen-containing compounds and dihalogen-

containing compounds has been reported by Steiger (1972).

There are also benign alternatives for crosslinking. Various polycarboxylic

acids, such as citric, maleic and tartaric acid, have been shown to possess

crosslinking properties due to their polyfunctionality. Coma et al. (2003) used citric

acid to crosslink hydroxyl-propyl methyl-cellulose to produce bio-based packaging

materials. Also, Menzel et al. (2013) prepared starch films crosslinked with citric

acid. Reddy & Yang (2010) crosslinked starch films with citric acid and reported

improved water resistance (measured as film weight loss). Organic acids can also

be used for the anionization of LC to enhance cation sorption, as described below.

27

2.2 Cation removal

Cation removal and related modification methods for LC are also briefly reviewed

as a natural continuation of the topic of water treatment using LC. Due to its

negative surface charge, native LC can be used as such in the removal of aqueous

cationic compounds (Brown et al. 2000), but the sorption capacities are usually

very low—except for peat—at only a few milligrams per gram of sorbent (Asadi et

al. 2008). Chemical treatments, for example those employing formaldehyde (Taty-

Costodes et al. 2003), NaOH (Argun & Dursun 2008) and mineral acids (Argun et

al. 2007, Asadi et al. 2008), can be used to improve cation sorption capacity

through the activation of OH groups, and to prevent the elution of compounds that

would result in staining and increased COD levels in the treated water (Taty-

Costodes et al. 2003, Argun & Dursun 2008).

Despite the fact that research on the removal of anions using bio-based anion

exchange materials is rare, some studies have been published regarding the

treatment of real wastewaters and the removal of cationic heavy metals using LC,

both modified and unmodified. Asadi et al. (2008) reported the removal of Cd, Cu,

Ni, Pb and Zn from four different industrial wastewaters using rice hull and sawdust

of species of Papullus treated with heat, HCl or NaOH. They found that NaOH-

treated rice hull performed best in the treatment of plating wastewater, giving

sorption capacities of 68.4 and 56.9 mg/g for Zn in batch and fixed bed column

studies, respectively (Asadi et al. 2008). They also concluded that NaOH treated

rice hull could be reused after metal desorption with 0.1 M HCl (Asadi et al. 2008).

Metal uptake from synthetic solutions was most efficient at pH 4–7 (Asadi et al.

2008). Sorption capacities of 45 and 47 mg/g for Zn were achieved with sugarcane

bagasse and sawdust of species of Manilkara modified with ethylenediamine

tetraaceticacid dianhydride (EDTAD) (Pereira et al. 2010). Hegazi (2013) used rice

husk and fly ash in the removal of Fe, Cd, Cu, Ni and Pb from electroplating

wastewater. Velazquez-Jimenez et al. (2013) reported metal sorption capacities for

two modified agave bagasse materials: 50.1 mg/g for Pb and 20.5 mg/g for Zn

using a NaOH treatment, and 42.3 mg/g for Pb and 12.4 mg/g for Zn using a HCl

treatment. Yu et al. (2016) modified corn silk with dilute HNO3 and attained

maximum capacities of 93 mg/g for Cu, 82 mg/g for Co and 75 mg/g for Ni.

Similarly, polycarboxylic acids have been used for the preparation of bio-based

sorbents for cation removal. Polycarboxylic acids contain more than one carboxylic

group. The polyacid dehydrates at high temperature forming an acid anhydride,

which reacts with the hydroxyl groups of LC, thus increasing the number of

28

reactive hydroxyl groups (i.e. -COO-). This reaction could also potentially lead to

crosslinking two lignocellulosic OH groups together. (Wing 1996)

Wing (1996) was among the first to apply citric acid treatment in the

preparation of a bio-based sorbent from corn fibre for the removal of Cu: a capacity

of 127 mg Cu/g was obtained. Marshall et al. (1999) and Wartelle & Marshall

(2000) treated soybean hulls with citric acid after NaOH pretreatment, and achieved

Cu sorption capacities of 155 mg Cu/g and 91.5 mg Cu/g, respectively, as opposed

to the capacity of unmodified hulls (20.3 mg Cu/g). Salazar-Rabago & Leyva-

Ramos (2016) used citric, malonic and tartaric acids to modify sawdust of Pinus

durangensis and obtained sorption capacities of up to 304 mg/g for Pb as opposed

to the capacity of 19 mg Pb/g when using native sawdust.

29

3 Materials and methods

Lignocellulosic raw materials—Scots pine sawdust (PSD) and bark (Pinus

sylvestris), Norway spruce bark (Picea abies), birch bark (Betula pubescens/Betula

verrucosa) and peat—were used as bio-based platforms for the preparation of ion

exchange materials. The materials were dried at 80°C, then ground and sieved to

obtain a size fraction of 90–250 µm, which was used in the modification tests.

In Papers I–III, sodium nitrate (NaNO3, Merck, analytical grade) was used to

prepare synthetic nitrate solutions to confirm the successfulness of the chemical

modification step and for the studies on the sorption mechanism, kinetics, effect of

pH, etc. In Paper II, sodium phosphate (Na3PO4, Sigma-Aldrich) or sodium

sulphate (Na2SO4, Sigma-Aldrich), both of analytical grade, was added into the

nitrate solutions to study the effect of co-existing sulphate and phosphate. NaCl in

analytical grade and NaOH in reagent grade were used in the regeneration in the

fixed bed column tests (Papers II, IV and V). A commercial strong base anion

exchange resin, Amberjet 4200 Cl (Dow Chemical), was used in Paper I as a

reference material in the maximum sorption capacity tests. In addition to synthetic

solutions, real waters, as described below, were used in the sorption tests. Detailed

descriptions of the physico-chemical properties of the above samples are given in

the respective publications.

– NO3-contaminated groundwater, taken from a water intake plant in

Haapavesi, Finland (NO3-N 6.2 mg/l, SO42- 25 mg/l, pH 6.3, conductivity

0.25 mS/cm) (Paper III)

– mining wastewater, taken from a mine in Northern Finland (Ni 38 µg/l, NO3-

N 9.3 mg/l, SO42- 8100 mg/l, pH 8–8.2, conductivity 9.4 mS/cm) (Paper IV)

– ammoniacal industrial wastewater, taken from a synthesis gas scrubber at a

European chemical plant (V 50.4 mg/l, Ni 13.3 mg/l, SO42-

490 mg/l, NH4-N

423 mg/l, pH 5.1, conductivity 3.2 mS/cm) (Paper V)

– metal-containing industrial wastewater (Ni 42 µg/l, NO3-N 31 mg/l,

SO42- 440 mg/l, pH 8, conductivity 2.9 mS/cm) (Unpublished data)

– NO3-containing mining wastewater (NO3-N 15.3 mg/l, pH 6.8, conductivity

1.85 mS/cm) (Unpublished data)

30

3.1 Chemical modification

Two different chemical modification methods were used for the quaternization

(amination) of the surface of LC, the ETM and CHMAC methods, as described in

the following sections. A separate anionizing treatment using citric acid was trialled

on pine sawdust. In all modification tests, the sample slurries were stirred in a hot

water bath using a multi-position magnetic stirrer and under reflux to prevent the

evaporation of the solvent phase. A water-circulating thermostat (Fig. 4) was used

to maintain the required reaction temperature. All five LC materials were used in

the ETM modification tests in Paper I, whereas in Paper III only pine sawdust was

treated with CHMAC.

31

Fig. 4. Experimental apparatus used in the chemical modifications: 1) reflux condensers,

2) water-circulating thermostat, and 3) multi-position stirrer.

3.1.1 Amination using the ETM method

The ETM method was adapted in Paper I from the method used by Wang et al.

(2010) to quaternize and crosslink giant reed. The proposed reaction schemes for

the crosslinking and quaternization are presented in Figures 3 and 5.

32

Fig. 5. Proposed reaction route of the ETM quaternization method. Crosslinking is not

included in this illustration. Adapted from Xu et al. (2011).

N,N-dimethylformamide is used as a solvent. Epichlorohydrin first reacts with the

hydroxyl groups of LC, forming a reactive cellulose ether (Gao et al. 2009).

Ethylenediamine reacts with the cellulose ether and introduces secondary amino

groups and triethylamine is used to add quaternary amino groups, which act as the

functional ion exchange groups of the material (Wang et al. 2010). The chloride

O

HOH

HH

H

H

O

OHO

OHO

HOH

HH

H

HOOH

OH

n

+ ClCH2 CH

O

CH2

n

+ NH2NH2 CH2CH2

+CH3

CH3

CH3

CH2

CH2

CH2

N

n

n Cl-

Cl-

ethylenediamine

triethylamine

epichlorohydrin

R1

CH2CH

O

CH2

O

HO

HH

H

H

O

OO

OO

HO

HH

H

HOO

CH2 O

CH2CHO

CH2

R1

R1

R1

R2

R2R2

R2

NH2NH CH2CH2CHCH2

O

HO

HH

H

H

O

OO

OO

HO

HH

H

HOO

CH2 O

CH2

OH

CH2

NH2NH CH2CH2CH

OH

CH2

R3

R3

R3

R3

NHNH CH2CH2CHCH2

O

HO

HH

H

H

O

OO

OO

HO

HH

H

HOO

CH2 O

CH2

OH

CH2

NHNH CH2CH2CH

OH

CH2 CH2 CH CH2

OH CH3

CH3

CH3

CH2

CH2

CH2

N+

CH2 CH CH2

OH CH3

CH3

CH3

CH2

CH2

CH2

N+

CH2CH

O

CH2 CH2 NH2NH CH2CH2CH

OH

CH2

Cl-

CH2 NHNH CH2CH2CH

OH

CH2 CH2 CH CH2

OH CH3

CH3

CH3

CH2

CH2

CH2

N+

R1: R2:

R3:

33

anion originating from epichlorohydrin remains in the structure acting as a counter

ion and enabling the ion exchange reactions. Epichlorohydrin forms crosslinks

between the cellulose chains (Yoshida et al. 2014), and ethylenediamine might also

act as a crosslinker, as described in Figure 3.

Two grams of LC (90–250 µm) was weighed in a two-neck round-bottom flask

and 16 ml of N,N-dimethylformamide and 13 ml of epichlorohydrin was added.

The mixture was stirred at 60–70°C for 45 min. After this, 2.5 ml of ethylene-

diamine was added and the mixture was stirred for another 45 min at 60–70°C. In

the final step, 13 ml of triethylamine was added and the mixture was stirred for

120 min at 60–70°C. The prepared materials were washed with plenty of water and

dried at 105°C. The particle size increased slightly during modification to max.

0.5–1.5 mm according to visual analysis. They were not crushed because a slightly

bigger particle size is desirable for the applications. The mass yield [%] was

calculated as a ratio between the final mass and the initial sample mass using

Equation 1. (Paper I). ETM-modified pine sawdust is referred to later in the text as

ETM-PSD.

%100⋅=initial

final

m

mY , (1)

where mfinal is the final mass of the sample [g] and minitial is the initial mass of the

sample [g].

3.1.2 Amination using CHMAC

In an attempt to find a less hazardous (without the need to use epichlorohydrin) and

simpler method for quaternization, a cationic monomer, N-(3-chloro-2-hydroxy-

propyl) trimethylammonium chloride (CHMAC), was used to prepare an anion

exchange material from PSD. The CHMAC method involves less chemicals and is

therefore considered to be less hazardous. Fig. 6 describes the reaction between a

cellulose chain and CHMAC. The same reaction is assumed to take place in the

case of any hydroxyl group in LC.

One gram of PSD (90–250 µm) was first mixed with 10 ml of NaOH at room

temperature for an hour to enhance the reactivity of PSD. Then, 10 ml of aqueous

solution of CHMAC (65%) was added and the mixture was stirred under reflux at

a desired temperature for a predetermined period of time. The final product was

separated from the reaction media, washed with plenty of water and dried at 105°C.

The optimal reaction conditions were determined using design of experiments as

34

described below. The final product mass was 85–108% of the initial mass (see

Equation 1) depending on the chemical doses and synthesis conditions. (Paper III).

The products are referred to later in the text as C-PSD.

Fig. 6. Reaction between cellulose and CHMAC in the presence of NaOH. Adapted from

Ren et al. (2007)

Optimization of reaction conditions

The reaction conditions for the CHMAC treatment were optimized using design of

experiments as no information was available on the optimal reaction conditions for

the CHMAC method when applied to PSD. A central composite circumscribed

(CCC) design was used to obtain a maximum amount of information with the

minimum number of experiments. The CCC design was composed and the data

were analysed using Modde 8.0 (MKS Data Analytics Solutions, Sweden). Three

independent variables were chosen: reaction time (0.32–3.68 h), reaction

temperature (26.4–93.6 °C) and NaOH/CHMAC molar ratio (0.19–2.21). The

molar ratio was controlled by using NaOH solutions at different concentrations

(3.06–35.42% w/v). The centre point conditions were a reaction time of 2 h,

a reaction temperature of 60°C and a NaOH/CHMAC ratio of 1.2. With three

variables the number of experiments in a full factorial test is 23. The distance of the

star points from the centre point was defined in coded units as (23)1/4 = 1.682. The

design area was therefore a sphere, which makes it rotatable and ensures equal

prediction variance at every point equidistant from the centre point (Witek-

CH3CH3

N+ CH3

Cl

OHCl

-O

HOH

HH

H

H

O

OHO

OHO

HOH

HH

H

HOOH

OH

n

+ + NaOH

O

HO

HH

H

H

O

OO

OO

HO

HH

H

HOO

O

CH3CH3

N+ CH3

OH

R1

R1 n

Cl-

+ H2O +

R1

R1

R1

Cl-

CH3CH3

CH2CH CH2 N+ CH3

OH

NaCl

R1:

35

Krowiak et al. 2014). Three repetitions were conducted at each test point to ensure

repeatability except for the five repetitions at the centre point in order to provide a

reasonable estimate of the prediction error and to make the model more reliable

throughout the test region. A total of 47 experiments were conducted. Mass yield

Y, as defined by Equation 1 (p. 33), was chosen as the response. Full details about

the experimental design are found in Paper III.

3.1.3 Citric acid treatment

In our studies on the treatment of mining wastewater and ammoniacal industrial

wastewater it was discovered that ETM-PSD was also very efficient in the removal

of nickel (Papers IV and V). However, the use of such an anion exchange material,

the preparation of which requires the use of multiple hazardous chemicals may not

be an economically and ecologically feasible option for cation removal. Cation

sorption could be enhanced by milder treatment methods, as discussed above (see

section 2.2 on page 27). Therefore, an anionizing citric acid (CA) treatment was

trialled on PSD. The tests remained at a preliminary stage but further studies will

follow in the future.

Three grams of PSD was mixed with water and stirred at 60°C for 30 min.

Citric acid (Sigma-Aldrich) at doses of 0.1–1.5 g CA/g (or 0.5–7.8 mmol CA/g) of

PSD was dissolved in a small amount of water and added into the PSD slurry, which

was then stirred for another 30 min at 60°C. A sodium hypophosphite monohydrate

catalyst (SHP, NaH2PO2 · H2O, Sigma-Aldrich, 20% of CA mass) was used together

with CA in part of the experiments. Then, the slurry was dried at 105°C to evaporate

excess water, after which the mixture was reacted at 120°C for 3 h. (Unpublished

data). Citric acid is converted into a reactive acid anhydride at high temperature.

The acid anhydride then reacts with the hydroxyl groups of LC, thus increasing the

number of hydroxyl groups which enhances the cation-binding properties (Wing

1996, Low et al. 2004) (Fig. 7). The final product was cooled, washed with plenty

of water to remove all unreacted citric acid, and dried at 60°C. The products are

referred to later in the text as CA-PSD. (Unpublished data).

36

Fig. 7. Proposed reaction between cellulose and citric acid: a) formation of an acid

anhydride, b) reaction of acid anhydride with cellulose, c) subsequent formation of an

acid anhydride, and d) formation of a crosslink between two cellulose chains. Adapted

from Salam et al. (2011)

Hybrid ion exchange material

Based on the excellent results on the removal of both anions and cations from

mining wastewater and ammoniacal industrial wastewater with ETM-PSD (see

sections 4.3.2 and 4.3.3), and the successful nickel removal with CA-PSD (see

37

section 4.1.3), the preparation of a hybrid ion exchange material was attempted.

Milder treatment methods were focused on, and hence, an anionizing citric acid

treatment with a subsequent cationizing treatment with CHMAC was applied to

PSD. The syntheses were carried out as described in sections 3.1.2 and 3.1.3. Pine

sawdust was first treated with citric acid (0.1–1.5 g/g or 0.5–7.8 mmol/g of PSD)

in the presence of a SHP catalyst followed by CHMAC treatment (0.01–0.04 mol/g

in the optimal reaction conditions) to enhance both anion and cation sorption

properties. The final product was washed with plenty of water, dried at 105°C, and

labelled CAC-PSD. (Unpublished data).

A combination of 0.4 g CA/g (or 2 mmol CA/g) of PSD and 0.024 mol

CHMAC/g of PSD was found in the preliminary tests to be sufficient for the

production of an efficient material in terms of both nitrate (NaNO3, uptake 39%

from the initial 20 mg N/l) and nickel removal (Ni(NO3)2, uptake 15% from the

initial 10 mg Ni/l) from synthetic single-ion solutions. This batch of CAC-PSD was

also tested briefly on a metal-containing industrial wastewater (42 µg Ni/l,

31 mg NO3-N/l). A dose of 1 g/l resulted in a nickel uptake of 15±0% and a nitrate

uptake of 24±1.1%. Higher doses were not tested at this research stage. Despite the

promising sorption results, the hybrid synthesis turned out to be very difficult to

scale up, and making a bigger batch of CAC-PSD was unsuccessful. Also, it seemed

that the CHMAC treatment neutralized the anionizing effect of CA due to the

reaction of CHMAC with the OH groups. (Unpublished data). Considering these

problems, using CA-PSD and C-PSD together, e.g. in a mixed bed column instead

of making a hybrid material, could possibly be a more reasonable approach.

Intensive work on the method and applications is still needed.

3.2 Sorption studies

To find out the best performing ETM material in terms of nitrate removal, all the

five ETM materials prepared in Paper I were tested. Basic sorption tests were

conducted using all five ETM-modified materials and synthetic solutions, starting

from the effect of sorbent dose, pH and contact time (Paper I). Further tests,

including the effects of other ions, fixed bed column tests and tests on real waters,

were performed and reported in Papers II, IV and V on ETM-PSD, which was

chosen for further studies based on Paper I. C-PSD was studied in Paper III in terms

of sorbent dose, contact time, temperature and maximum capacity using synthetic

nitrate solutions. In addition to this, nitrate-contaminated groundwater and nitrate-

38

containing mining wastewater were treated with C-PSD (Paper III, Unpublished

data).

Table 1. Summary of sorption studies. The Roman numerals refer to the original

publications.

Solution and experiment type ETM-PSD C-PSD CA-PSD CAC-PSD

NaNO3 solution (NO3)

Batch tests (sorbent in Cl- form)

Sorbent dose I1 III ud2 ud2

pH I1

Contact time I1 III

Maximum sorption capacity for NO3-N I1, II III

Co-existing PO4-P and SO4-S II

Temperature II III

Fixed bed column tests (Cl- form, NaCl reg.) I, II

NaVO3 solution (V)

Batch tests (sorbent in Cl- form)

pH V

Maximum sorption capacity for V V

Co-existing SO4-S V

Ni(NO3)2 solution (Ni)

Sorbent dose ud ud

NO3-contaminated groundwater

Sorbent dose (sorbent in Cl- form) III

Mining wastewater

Batch tests (sorbent in Cl- form)

Sorbent dose IV

Temperature IV

Contact time IV

Fixed bed column studies (Cl- form, NaCl reg.) IV

Ammoniacal industrial wastewater

Batch tests (sorbent in Cl- form)

Sorbent dose V

pH V

Fixed bed column tests (OH- form, NaOH reg.) V

Metal-containing industrial wastewater

Sorbent dose (sorbent in Cl- form) ud2

NO3-containing mining wastewater

Batch tests (sorbent in Cl- form)

Sorbent dose ud

1 All five ETM materials prepared in Paper I were also subjected to these tests. 2 Not a full test series.

ud: unpublished data

39

CA-PSD and CAC-PSD were tested in nickel removal from synthetic solution, and

CAC-PSD was tested on a metal-containing industrial wastewater (Unpublished

data). The results are presented in the section above. Table 1 summarizes all the

sorption tests on different types of modified PSD.

3.2.1 Batch shaking experiments

Batch shaking tests with synthetic solutions were used to study the sorption

efficiency of modified LC. A desired amount of modified LC was mixed with 50 ml

of solution. A concentration of 30 mg NO3-N/l in a NaNO3 solution was used by

default to test the nitrate removal efficiency of a prepared material (Papers I–III).

In Paper V, a sodium metavanadate (NaVO3) solution was used to study vanadium

removal. A 24-hour contact time was used to ensure equilibrium in all batch

shaking tests (Papers I–V). The experimental conditions were adjusted depending

on the nature of the sorption test. For the pH tests, 0.1–1 M NaOH or HCl was used

for pH adjustment (Papers I and V). For the tests on the effect of temperature, a hot

water bath (Fig. 4) or a cold room was used (Papers II and III).

Maximum sorption capacity is a useful tool for assessing the sorption

performance of a sorbent material, and therefore, in Paper I, it was determined for

nitrate in order to find out the best performing ETM material in terms of nitrate

removal. The maximum sorption capacity of ETM-PSD for vanadium from

synthetic solution was determined in Paper V. The maximum nitrate sorption

capacity of C-PSD for was determined in Paper III. Initial concentrations of 5–

400 mg NO3-N/l (Papers I and III) and 35–580 mg V/l (Paper V) were used. The

maximum capacity (Qmax) was calculated using Equation 2:

( )m

CCVQ e−

= 0max

(2)

where V is the liquid volume of sample [l], C0 is the initial concentration in solution

[mg/l], Ce is the equilibrium concentration in solution [mg/l] and m is the mass of

the sample [g]. Sorption capacities were also calculated from some the results of

the basic batch shaking tests to enable some comparison, but such results were not

referred to as maximum sorption capacities.

The effect of contact time was studied at contact times of 2 min, 5 min, 15 min,

30 min, 1 h, 2 h, 4 h and 24 h (Papers I, III and IV). In addition, kinetic modelling

was performed and reported in Paper I. The effects of phosphate and sulphate on

nitrate removal were studied by adding 10–500 mg/l PO4-P or 1–50 mg/l SO4-S

40

into the nitrate solution (30 mg NO3-N/l), as described in Paper II. The effect of

added sulphate on vanadium removal was studied by adding 15–5000 mg/l SO4-S

into the synthetic vanadium solution (22–28 mg V/l) (Paper V). Tests on real waters

were conducted in a similar manner and further details are given in Papers III–V

and section 3.2.3.

3.2.2 Fixed bed column studies

In a real industrial column process, commercial, bead-form ion exchange resins are

used in fixed bed columns through which contaminated water is passed, allowing

the ion exchange reactions to occur. Batch shaking tests are easy to conduct in

laboratory scale, which however, cannot provide all the information needed for the

accurate design of an ion exchange column process, and therefore ETM-PSD was

tested in a packed-bed column to evaluate its performance under continuous-flow

conditions, replicating the conditions of a real industrial column process. The

regeneration of the material was also investigated in column since the reuse of the

material is the cornerstone feature in the case of ion exchange.

To prepare the column for the experiments, ETM-PSD (1–2 g) was wet-packed

in a glass column (11 mm in diameter and 680 mm in length). During the ion

exchange cycles, nitrate solution (Papers I and II) or real wastewater (Papers IV

and V) was passed gravitationally through the bed (5–7 cm thick) down-flow at a

flow rate of 3–5 ml/min after which the bed was regenerated down-flow using NaCl

solution. The bed was rinsed with Milli-Q water after every ion exchange and

regeneration cycle. The solutions were collected in 40–100 ml fractions. (Papers I,

II, IV and V).

In Papers I and II, synthetic nitrate solutions (NaNO3, 10, 30 and 50 mg NO3-

N/l) were used to study the effect of initial concentration. In Paper IV, real mining

wastewater was used, and a detailed experimental procedure concerning these tests

is given in section 3.2.3. In Papers I, II and IV, the material was regenerated using

a 0.1 M NaCl solution to maintain the Cl- form of the material. In Paper V,

ammoniacal industrial wastewater was treated in a fixed bed column using ETM-

PSD in OH- form. NaOH (2 M) was thus used for regeneration (Paper V).

3.2.3 Real waters

Knowledge about sorption behaviour in the treatment of real waters is essential

because synthetic solutions alone do not alone provide enough of the information

41

that is required for the application of ion exchange resins. The majority of the

research in this field has dealt with synthetic solutions but it does not answer the

vital question: how do the materials perform in the treatment of real waters that

contain a variety of components in different proportions and concentrations? One

of the main focus areas of this work was to study the possible applications of bio-

based ion exchange materials.

Besides wastewater treatment, bio-based ion exchange materials could also

potentially be used in the production of drinking water. To demonstrate this,

nitrate-contaminated groundwater was treated in batch using C-PSD, as

described in Paper III. The nitrate concentration was slightly elevated (6.2 mg/l as

NO3-N), though below the recommended maximum limit of 11.3 mg/l as NO3-N

(50 mg/l as NO3-) in drinking water set by the Finnish legislation (442/2014). C-

PSD was also tested on a nitrate-containing mining wastewater (Unpublished

data). In our additional work, metal-containing industrial wastewater was treated

using CAC-PSD (Unpublished data).

In Paper IV, another batch of mining wastewater containing nitrate and heavy

metals was studied. The sample was taken from a stream exiting a tailings pond,

which met the requirements of the current environmental permit. Nevertheless, ever

increasing restrictions in the allowed concentrations of heavy metals and nutrients

call for more efficient treatment of mining wastewaters. The mining wastewater

was treated in batch and fixed bed column tests using ETM-PSD (for further details

see Table 6 on p. 60). The effects of sorbent dose, contact time and temperature

were studied in batch mode. Also, a column trial comprising the following

sequence was performed. A full account of the procedure is given in Paper IV.

– Three ion exchange/regeneration cycles with NaCl,

– a maintenance cycle with HCl,

– three ion exchange/regeneration cycles with NaCl,

– a maintenance cycle with HCl, and

– two ion exchange/regeneration cycles with NaCl.

This trial was used to imitate a real industrial process and to study the effects of an

acid rinse on the material. Acidic or oxidizing rinsing media are used at certain

intervals on ion exchange materials to remove accumulated and resistant impurities

and foulants that block the active sites of the material, thus hindering sorption

performance. The chemical resistance of an ion exchange material towards such

media must be determined as a vital part of the feasibility assessment. Exposure to

42

strong chemicals might deteriorate the structure and lead to the collapse of the

matrix. An additional test of the chemical resistance of ETM-PSD using HCl and

H2O2 was performed and reported in Paper V.

In Paper V, ammoniacal industrial wastewater was treated in batch and fixed

bed column tests using ETM-PSD to study the removal of anionic vanadate

compounds. Batch tests were performed using the material in Cl- form, and column

tests in OH- form (Paper V).

3.3 Analysis methods

The physico-chemical properties of PSD and the prepared anion exchange

materials were studied using various methods as listed in Table 2. Details of the

apparatus, and detailed results in the form of graphs and numerical data, are given

in the respective publications. In Paper I, besides PSD, the four other LC materials

were also subjected to the same analyses, but left out of Table 2 for clarity.

Table 2. Characterization methods for PSD and modified PSD samples. The Roman

numerals refer to the original publications.

Analysis/parameter PSD ETM-PSD C-PSD CA-PSD CAC-PSD

Elemental composition (C, H, N, O) I1 I1 III

Chemical composition (cellulose,

hemicellulose, lignin, extractives)

I1 I

Ash content I1 I

BET surface area I1 I1

Neutron activation analysis IV

EDS analysis II, V

FESEM imaging I1 I1, V III

FTIR spectra I1, ud I1 ud ud ud

1 All five unmodified and ETM materials prepared in Paper I were also subjected to these tests.

ud: unpublished data

The water analysis methods used in this thesis are listed in Table 3. The methods

are described briefly by referring either to the analysis method, apparatus or the

SFS analysis standard provided by the Finnish Standards Association. A detailed

methodological description is given in the original publications.

43

Table 3. Water analysis methods. The Roman numerals refer to the original publications.

Parameter Method/ Apparatus/ SFS standard Reference

pH Metrohm 744 pH meter I–V, ud

Total suspended solids SFS-EN ISO 872 (2005) IV

Conductivity VWR Phenomenal PC 5000H I, III, IV, V

Turbidity Hach Ratio/XR III

Alkalinity SFS-EN ISO 9963-1 (1996) III, IV

Total organic and inorganic carbon GE Sievers TOC analyser IV

Dissolved organic carbon GE Sievers TOC analyser V

NO3- Hach Lange cuvette tests for total

nitrogen (LCK138, LCK238, LCK338)

I–III, ud

Hach Lange cuvette tests for NO3-

(LCK339, LCK340)

ud

CFA III, V

IC, CFA IV

NO2- CFA III, V

CFA IV

SO42- IC II–V

PO43- CFA II

CFA III

Cl- IC III, IV

F- IC IV

NH4+ SFS 3032 (2000) III

CFA IV, V

Ni ICP-OES, ICP-MS IV

FAAS V, ud

V GFAAS V

Other metals and semimetals ICP-OES, ICP-MS III, IV

ud: unpublished data

44

45

4 Results and discussion

4.1 Characterization of prepared materials

4.1.1 ETM method

The ETM method gave products with high mass yields (300–700% of initial mass).

Nitrogen contents of 9.1–9.8% were achieved during modification. The nitrogen

content of unmodified LC materials was 0.8–1.9%. The highest increase was

observed in PSD. (Paper I). According to the theory presented by Xu et al. (2011),

the nitrogen was added mostly in the form of quaternary ammonium nitrogen, but

also presumably in the form of secondary amine nitrogen, which according to Xing

et al. (2011) originated from ethylenediamine (see Fig. 5 on p. 32). The increased

nitrogen content is assumed to describe the amount of quaternary ammonium

nitrogen and result in higher nitrate removal rates and uptake of other anions

(Pepper 1953, Gao et al. 2009).

FESEM imaging revealed the spongy, pumice-like appearance of the ETM

materials, and increased macroporosity, mostly in the range of 1–10 µm (visual

observation). Clear changes in the appearance indicated heavy crosslinking, which

led to clumping and increased particle size in the prepared materials (visual

observation). (Paper I). Hena et al. (2015) observed very similar visual changes

with ETM-modified sugarcane bagasse. It was interesting that all the other ETM

materials, apart from ETM-PSD, appeared very porous in the FESEM imaging, but

resulted in lower sorption capacities and poorer overall sorption performance

compared to ETM-PSD. Apparently, the porosity did not play such an important

role in the sorption phenomena. (Paper I). This observation was supported by the

surface area analysis.

According to the BET surface area analysis, the surface area of the untreated

LC samples was very low, only 0.5–0.97 m2/g (Paper I). The quaternization and

crosslinking did not increase the surface area much: it remained at 1.1–4.3 m2/g

(Paper I), which is of the same magnitude as the values found in the literature for

similar materials. Xu et al. (2010b) reported BET surface areas of 5.3 and 4.4 m2/g

for wheat residue and ETM-modified wheat residue (Table 4). It is noteworthy that

the surface area decreased during the modification but no explanation was given

for this. Later, Xu et al. (2011) observed similar results with the same ETM-

modified wheat residue materials (4.9–5.7 m2/g) but again, the value for

46

unmodified wheat residue was higher (6.5 m2/g) than that of the ETM-modified

samples. Xu et al. (2013) reported the same behaviour in the case of ETM-modified

wheat stalk and cotton stalk (Table 4). Tan et al. (2012) reported the BET surface

areas of both unmodified and ETM-modified wheat straw to be in the range of 4.6–

5.9 m2/g. Again, the value for the modified sample was higher than that of the

original sample (Tan et al. 2012). However, this behaviour could not be explained.

For comparison, the surface area of activated carbon materials is usually in the

range of hundreds of square metres per gram. In the case of activated carbon,

sorption is based on adsorption, which relies on the porous structure and high

surface area. However, the surface area of ETM materials did not seem to correlate

with the sorption performance, which was a clear indication of ion exchange being

the dominant mechanism (Paper I). Xu et al. (2010b, 2011) and Tan et al. (2012)

drew the same conclusions in their work. However, other sorption mechanisms,

such as adsorption or ion exchange induced by secondary amine groups, have not

been ruled out.

Table 4. BET surface areas of ETM-modified materials.

Material BET surface area [m2/g]

Reference Original Modified

Pine sawdust 0.84 1.44 Paper I

Pine bark 0.96 3.42 Paper I

Spruce bark 0.97 1.13 Paper I

Birch bark 0.97 4.33 Paper I

Peat 0.49 1.16 Paper I

Wheat residue 5.3 4.4 Xu et al. 2010b

Wheat straw 5.92 4.66 Tan et al. 2012

Wheat stalk 8.82 4.56 Xu et al. 2013

Cotton stalk 5.30 4.40 Xu et al. 2013

4.1.2 CHMAC treatment

The CHMAC treatment produced a quaternized product with similar quaternary

ammonium groups compared to the ETM materials but without crosslinking and

the addition of secondary amine groups. Comparing the visual appearances of

ETM-PSD and C-PSD, it is clear that the CHMAC treatment does not have the

same kind of drastic effect on PSD compared with the ETM method. The FESEM

micrographs showed that, unlike the ETM materials, the C-PSD materials were not

47

porous and clumpy yet slightly different in appearance compared to native PSD.

The reaction conditions did not affect the appearance of the samples. (Paper III).

The optimal reaction conditions were as follows: reaction time of 1.8 h,

reaction temperature of 57°C and NaOH/CHMAC molar ratio of 1.32 based on the

optimization of mass yield (see section below). This was very much in accordance

with the optimal conditions reported earlier by Antal et al. (1984), Laszlo (1996)

and Ren et al. (2007) for the modification of beech sawdust (NaOH/CHMAC molar

ratio 1.8, 60°C, 2 h), sugarcane bagasse (NaOH/CHMAC molar ratio 1.56) and

hemicelluloses (NaOH/CHMAC molar ratio 1.2, 60°C, 2 h), respectively. The

C-PSD used in all further experiments was prepared using the optimal conditions

determined in this work.

The nitrogen content achieved in the optimal reaction conditions was lower

compared to the ETM materials, at only 2.5%. Despite the low result, the maximum

nitrate sorption capacity was 15.3 mg/g for NO3-N, which was about half of that

achieved using ETM-PSD (32.8 mg/g at a nitrogen content of 9.4%). Sorption

mechanisms other than ion exchange with the quaternary ammonium groups might

have caused this discrepancy. (Papers I–III).

It must be kept in mind that CHMAC modification is simpler, faster and could

also be considered safer than the ETM method. Therefore, the low nitrogen content

and nitrate sorption capacity should not be considered to be such poor results. This

is because producing an efficient material with less effort could be seen as

economically favourable. On the other hand, the overall hazardousness of the ETM

method and its advantages could be considered. Is the pursuit of a less hazardous

synthesis route really worthwhile? The process is very likely to be operated in a

closed circuit system, which significantly increases safety. Ultimately, it is the

sorption efficiency and capacity, and applicability in real treatment processes that

define the feasibility of the material. In order to make the choice between the ETM

and CHMAC methods, more information is needed about the column performance

and regeneration of C-PSD, which were not studied in this work.

Modelling of the CHMAC method

An empirical model was composed for the mass yield in the quaternization of PSD

using CHMAC in the presence of NaOH, as presented in Equation 3:

ACCBACBAY 68.191.691.266.548.219.136.17.104 222 −−−−+−−= , (3)

48

where Y is the mass yield [%] as defined by Equation 1, A is the reaction

temperature, B is the reaction time and C is the NaOH/CHMAC molar ratio.

Variables A, B and C are in coded units (-1…1), as presented in Paper III.

Substitution of the optimal condition parameters in Equation 3 gives a response of

105.2% as mass yield. The quadratic model describes a surface, i.e. the response

surface, which allows the determination of optimal reaction conditions for the

maximization of the yield. Nitrate uptake by C-PSD and its nitrogen content were

also modelled, but they did not produce viable and well-predicting models, hence

the modelling was focused on the mass yield. (Paper III).

Fig. 8 illustrates the observed mass yield during the CHMAC modification and

NO3- removal as functions of the nitrogen content in C-PSD. It seemed clear that

the higher the nitrogen content of the material, the better the nitrate uptake and

mass yield. Increased mass yield proves the addition of side chains onto LC. The

results in Figure 8 also support the assumption that a higher nitrogen content is

beneficial for the sorption of anions. Due to the lack of secondary amine groups, as

opposed to ETM materials, the nitrogen content represents mostly the amount of

quaternary ammonium groups, which clearly seem to contribute to the uptake of

nitrate anions.

49

Fig. 8. The observed mass yield and nitrate removal (sorbent dose 3 g/l) from 30 mg/l

NO3-N of C-PSD plotted against the nitrogen content in C-PSD. Both values are given

as the average of three repetitions with deviation. The red triangles represent the

optimal batch. Adapted from Paper III, reprinted by permission from Elsevier B.V.

A nitrate uptake of 53% was recorded for the optimal batch using a NaNO3 solution

of 30 mg N/l and a sorbent dose of 3 g/l (Fig. 8) (Paper III). C-PSD was also briefly

tested on nitrate-containing mining wastewater (initially 15.3 mg/l as NO3-N).

Sorbent doses of 3–10 g/l were tested for a contact time of 1 h at around 15°C. The

smallest doses were not effective, but uptakes of 29 and 33% were attained at doses

of 8 and 10 g/l. (Unpublished data). The suppressed nitrate sorption suggests that

the complex ionic composition of the wastewater (data not shown) had a

pronounced effect on the sorption behaviour, possibly somewhat stronger than the

effect of low temperature.

4.1.3 Citric acid treatment

The citric acid treatment did not affect the visual appearance of PSD. The FTIR

analysis showed a clear increase in the amount of carbonyl groups in CA-PSD

indicated by the signal at 1740 cm-1, which was a result of the reaction between LC

and citric acid (see Fig. 7 on p. 36). The nickel sorption performance of CA-PSD

was tested by using a Ni(NO3)2 solution (10 mg Ni/l) at a sorbent dose of 1 g/l with

single repetitions. Nickel uptakes of 59–73% and mass yields of 111–130% were

50

observed at CA doses of 0.3–1.5 g/g (or 1.6–7.8 mmol/g) with the SHP catalyst

(Fig. 9). The catalytic effect of SHP remained unclear because the tests were not

repeated. Moreover, it was beyond the scope of this study. (Unpublished data).

Fig. 9. Effect of CA dose on mass yield and nickel uptake of CA-PSD (Unpublished data).

The curing temperature, 120°C, was chosen based on the literature. Some research

teams have reported optimal curing temperatures of 120°C (Marshall et al. 1999)

and 140°C (Low et al. 2004) for CA modification. On the other hand, Menzel et al.

(2013) stated that the curing temperature did not in fact play a significant role in

the case of starch films. Similar conclusions have been made by Yang (1991). In

addition to the introduction of OH groups, citric acid is also claimed to create ester

bond-containing crosslinks between the cellulose chains (Wing 1996), as presented

in Fig. 7 (p. 36). This would require the reaction of two hydroxyl groups in different

parts of the LC structure or two different particles with the same CA molecule.

Naturally, the formation of a crosslink between two LC particles cannot be regarded

as an axiomatic fact. It is difficult to confirm whether crosslinking actually takes

place. Many have identified crosslinks through the detection of ester bonds using

FTIR spectroscopy (Coma et al. 2003, Reddy & Yang 2010), which is based on the

detection of ester bonds and ester carbonyl groups (-C=O). The signals of both

51

carboxylic carbonyls and ester carbonyls appear in the same wave number range of

1600–1750 cm-1, and therefore distinguishing between them might be difficult

because of overlapping signals and slight changes in carbonyl frequencies due to

resonance effects etc. (Smith 1999).

4.2 Sorption studies

4.2.1 Effect of sorbent dose

All five ETM materials performed well in the nitrate sorption tests: doses of

2–6 g/l resulted in nitrate uptakes of 70–89% from synthetic solution. The materials

were compared to a commercial strong base anion exchange resin, Amberjet

4200 Cl in Cl- form, which gave uptakes of 89–98% at doses of 2–6 g/l. However,

compared to various bio-based sorbent materials found in the literature, prepared

through the EDM or ETM methods, the present ETM materials performed better in

terms of nitrate sorption capacity. The highest nitrate uptake was achieved with

ETM-modified pine sawdust (ETM-PSD). (Paper I).

4.2.2 Effect of pH

The sorption performance of all ETM materials was tested at a pH range of 2–10

using synthetic nitrate solutions. Uptakes of 81–86% were observed at pH 3–10,

which also supported the hypothesis of a strong base anion exchanger. (Paper I).

Xu et al. (2010b, 2011) found the optimal pH to be 4–10 for phosphate and nitrate

removal. At pH 2, the high concentration of Cl- due to pH adjustment with HCl was

likely to cause the nitrate uptakes to drop to 54–67% with all materials. Again,

ETM-PSD performed best. (Paper I). A wide operating pH range would enable the

treatment of various types of waters.

4.2.3 Maximum sorption capacity

The maximum nitrate sorption capacity was determined for all five ETM materials

mentioned in Paper I. ETM-PSD performed best giving a nitrate sorption capacity

of 30.1 mg N/g. The capacities of the other ETM materials were in the range of

24.2–26.5 mg N/g. (Paper I). In Paper III, the nitrate sorption capacity of C-PSD

was determined to be 15.3±1.4 mg N/g (as an average of capacity values at 215–

52

350 mg N/l). The nitrate sorption capacity of C-PSD was about half of that of ETM-

PSD which, however, can be considered a good result despite the relatively low

nitrogen content and considering the simple synthesis procedure.

The maximum sorption capacity was also determined for vanadium. Fig. 10

compares the sorption capacities of ETM-PSD and C-PSD for nitrate and the

capacity of ETM-PSD for vanadium from synthetic solutions (Papers I, III and V).

ETM-PSD showed excellent vanadium sorption performance. A maximum sorption

capacity of 130 mg V/g was attained for a synthetic NaVO3 solution and

103 mg V/g for ammoniacal industrial wastewater (Paper V). In our earlier study

on the same wastewater, a vanadium sorption capacity of 48.9 mg V/g had been

achieved in batch tests at pH 7 using the commercial Amberjet 4200 Cl anion

exchange resin but this was not the maximum capacity (Keränen et al. 2015). Sirviö

et al. (2016) reported a maximum sorption capacity of 100.9 mg V/g from synthetic

solution using bisphosphonate nanocellulose. A sorption capacity of 24.9 mg V/g

using ZnCl2 activated carbon was reported by Namasivayam & Sangeetha (2006).

Compared to these literature values, ETM-PSD showed strong potential as a bio-

based sorbent for vanadium removal, even from complex industrial wastewaters.

53

Fig. 10. Nitrate sorption capacities of ETM-PSD for NO3-N and V and C-PSD for NO3-N

using synthetic solutions. Sorbent doses: 4 g/l (ETM-PSD + NO3-N, one repetition), 3 g/l

(C-PSD + NO3-N, three repetitions) and 1.5 g/l (ETM-PSD + V, two repetitions). (Papers I,

III and V)

4.2.4 Effect of contact time

From a process design point of view, it is important to know the effect of contact

time on the sorption phenomenon. In the case of a fixed bed ion exchange column

process, the contact time between the material and the water is a corner stone in the

design of the process, having a direct influence on the total treatment capacity.

Sufficient reaction time between the solution and the sorbent is required to enable

the diffusion of nitrate onto the surface of the sorbent and the reaction with the

functional groups. The binding mechanisms of ETM-modified anion exchange

materials may include adsorption and chelation as well as ion exchange, due to

secondary amine groups for instance, which might affect the length of the contact

time.

During the tests (initial conc. 30 mg NO3-N/l), it became evident that nitrate

sorption from synthetic solution was very fast with all five ETM materials:

0 100 200 300 400 500 6000

5

10

15

20

25

30

35

C-PSD + NO3-N

ETM-PSD + NO3-N

ETM-PSD + V

Initial concentration in solution [mg/l]

NO

3-N

sor

ptio

n ca

paci

ty [

mg/

g]

0

20

40

60

80

100

120

140

V s

orpt

ion

capa

city

[m

g/g]

54

equilibrium was attained in 5–10 min at room temperature (uptakes of 75–86%)

and over 70% removal were observed with all materials after only 2 min. (Paper I).

Xu et al. (2010a, 2012a) reported minimum contact times of 20 min for ETM-

modified Arundo donax L. Due to the short contact time, the flow rate in the ion

exchange column could be increased, which therefore would increase the overall

treatment capacity. A short reaction time is also beneficial in the case of a mixed

tank process.

Very similar behaviour was observed with C-PSD although the uptake

remained significantly lower than with the ETM materials: 64±1.8% of nitrate was

removed in 10 minutes. This could be due to the different structure of C-PSD

compared to ETM-PSD (lack of crosslinking and lower nitrogen content). (Paper

III).

4.2.5 Effect of temperature

Water temperature is an important factor when a sorbent material is applied in water

treatment. It is known that sorption is hindered by increasing temperature, due to

the exothermic nature of the phenomenon and increased Brownian motion of the

ions (Worch 2012). In the mining industry in the Arctic hemisphere, the wastewater

effluents are very cold, particularly in wintertime. On the other hand, effluents from

other industrial activities can be very warm and allowing the effluent to cool down

before treatment would require extra effort and increase costs. Fulfilling these

requirements is the key to creating an efficient sorbent material.

Temperatures of 5–40°C were studied at NO3-N concentrations of 10–500 mg/l.

Additional sorption tests were carried out at 50, 60 and 70°C at a constant NO3-N

concentration of 250 mg/l to provide comparison to higher temperatures. ETM-

PSD proved to work well in this wide temperature range. However, the sorption

capacities did decrease at higher temperatures (Fig. 11). (Paper II). This

observation is in accordance with the general sorption theory and the same

behaviour has been detected with various sorbent materials (Anirudhan & Unnithan

2007, Xu et al. 2010c, Karachalios & Wazne 2013, Hena et al. 2015). The deviation

between the repetitions increased with increasing temperature. The results suggest

that ETM-PSD would work very well at low temperatures, which supports

applications in the mining industry.

The effect of temperature in the treatment of mining wastewater was also

studied, and the results are discussed in section 4.3.2.

55

Fig. 11. a) Effect of temperature on the sorption capacity of nitrate at 5–40°C and

b) comparison of sorption capacities at 5–70°C at 250 mg NO3-N/l. The error bars

represent the deviation of three repetitions. (Paper II). Reprinted by permission from

Elsevier B.V.

4.2.6 Effect of phosphate and sulphate

As wastewaters are complex mixtures of various ions, the presence of competing

ions must also be studied. Sorbent materials usually have different affinities for

ions of different valence and atomic mass (DeSilva 1999). In Paper II, the effects

of sulphate and phosphate on nitrate sorption by ETM-PSD were studied as they

are common contaminants in wastewaters. The nitrate uptake was about 71% with

no added sulphate or phosphate (initially 30 mg N/l), but it decreased gradually to

16% in the presence of 100 mg SO4-S/l and to 26% in the presence of 50 mg

PO4-P/l (Table 5). Energy dispersive X-ray analysis (EDS) confirmed the presence

of phosphate and sulphate on the surface of ETM-PSD. (Paper II)

The initial pH values of the sulphate and phosphate solutions were 6.1–6.4 and

7.5–12, respectively. In this pH range, phosphate exists as bivalent HPO42- and

56

trivalent PO43-, with the former being predominant (Eliaz & Sridhar 2008). The

presence of these co-existing species, namely H2PO4-, HPO4

2- and PO43- at the

initial solution pH of 7.5–12.2, affects the sorption of nitrate. Due to the protonation

of phosphate, the pH increases with increasing phosphate concentration. HPO42- is

likely to have sorbed more readily than nitrate, thus lowering the nitrate uptake at

concentrations of 10 mg P/l and beyond (Table 5). When comparing SO4-S and

PO4-P doses of 50 mg/l, which are practically equal in moles (0.5 mol), one can

observe that the presence of phosphate had a greater suppressive effect on nitrate

uptake, and phosphate was sorbed better than sulphate. This suggests that ETM-

PSD would have a higher affinity for phosphate than sulphate, which might be due

to the pH-dependent variations in phosphate speciation. SO4-S concentrations of

300 mg/l and over inhibited nitrate sorption completely although this would

probably have happened with an equal phosphate dose as well. (Paper II).

Table 5. Effects of phosphate and sulphate on nitrate removal from synthetic solution

using ETM-PSD. Values are given as the average of three repetitions with standard

deviations. Sorbent dose 3 g/l. (Paper II). Reprinted by permission from Elsevier B.V.

Ninitial [mg/l] Sinitial [mg/l] Nfinal [mg/l] N uptake [%] Sfinal [mg/l] S uptake [%]

30 0 8.4 ± 0.3 71.4 ± 0.8 - -

30 10 6.9 ± 0.7 76.7 ± 2.3 - -

30 30 9.0 ± 0.1 70.6 ± 0.3 0.23 ± 0.1 99.3 ± 0.2

30 50 12.0 ± 0.4 60.0 ± 1.4 0.30 ± 0.0 99.4 ± 0.0

30 100 25.3 ± 1.2 15.9 ± 3.8 14.6 ± 0.2 83.8 ± 0.2

30 300 30.7 ± 0.5 0 186 ± 6.8 31.2 ± 2.5

30 500 31.6 ± 0.2 0 367 ± 0.6 26.5 ± 0.1

0 500 - - 395 ± 25.5 18.4 ± 5.3

Ninitial [mg/l] Pinitial [mg/l] Nfinal [mg/l] N uptake [%] Pfinal [mg/l] P uptake [%]

30 0 7.8 ± 0.3 73.4 ± 1.0 - -

30 1 8.6 ± 0.4 71.7 ± 1.2 - -

30 5 8.8 ± 0.5 71.7 ± 1.7 - -

30 10 10.6 ± 0.4 65.6 ± 1.3 - -

30 15 11.5 ± 1.2 63.4 ± 4.0 0.32 ± 0.04 97.7 ± 0.3

30 30 15.3 ± 0.3 49.2 ± 0.8 0.76 ± 0.02 97.5 ± 0.1

30 50 22.3 ± 1.0 25.8 ± 3.1 3.6 ± 0.06 92.7 ± 0.1

0 50 - - 1.6 ± 0.1 96.4 ± 0.2

Dashes represent no measurement.

57

The effect of sulphate on vanadium sorption from synthetic solution using ETM-

PSD was studied in Paper V. According to Zeng & Cheng (2009), vanadium is

present in valences of -3, -4 and -6 in the initial pH of the solutions (around 7, final

pH 6–6.8) (Fig. 12). Added sulphate did not affect the uptake of vanadium: uptakes

of over 97% were observed at concentrations of up to 5000 mg SO4-S/l with an

initial vanadium concentration of 22–28 mg/l at pH 7, and over 99% without added

sulphate. Sulphate uptake was not determined. The higher charge of the vanadate

species is likely to cause the sorbent to favour vanadium over sulphate. (Paper V).

Fig. 12. The distribution of vanadium species as a function of pH. Total VO2+

concentration is 14.33 mM. The dashed line represents the sum of the distribution

percentages of the decavanadates. (Zeng & Cheng 2009). Reprinted by permission from

Elsevier B.V.

It is assumed that the sorption affinity of ions is related not only to the valence of

the ion but also to the atomic number (mass): the higher the atomic number, the

higher the affinity (Wang et al. 2007). Very few studies concerning ETM materials

and competitive sorption were found. Xu et al. (2012a) reported using ETM-

modified Arundo donax L. for the simultaneous removal of NO3-, NO2

-, PO43- and

SO42- (all ions present in the same solution) and observed a sorption affinity order

of SO42- > PO4

3- = NO3- > NO2

-. Despite the difference in both atomic numbers and

58

valence, they reported phosphate and nitrate to have equal affinity (Xu et al. 2012a).

Wang et al. (2007) reported competitive sorption of NO3-, NO2

-, H2PO4- and SO4

2-

(all ions present in the same solution) using EDM-modified wheat straw (weak base

anion exchanger), and detected a preferential affinity order of SO42- > H2PO4

- >

NO3- > NO2

-. Unlike the observation of Xu et al. (2012a), these results suggested

the atomic weight had a strong influence on the sorption: phosphorus as a heavier

atom had a higher affinity than nitrogen. Nevertheless, a strong base anion

exchanger like ETM-PSD might behave differently, as demonstrated by Xu et al.

(2012a) and additional affinity studies are needed.

4.2.7 Fixed bed column studies

ETM-PSD was tested in a fixed bed column using both synthetic NaNO3 solutions

at NO3-N concentrations of 10, 30 and 50 mg/l (Papers I and II) and mining

wastewater (Paper IV). In addition, ammoniacal industrial wastewater was also

treated in column using ETM-PSD (Paper V). The column results concerning real

waters are presented in section 4.3.

The column studies on synthetic NaNO3 solutions showed the concentration-

dependence of the sorption phenomenon (Fig. 13). The results were in agreement

with the results reported by Xing et al. (2011) who used ETM-modified wheat straw.

The higher the nitrate concentration, the faster the breakthrough. This behaviour is

closely connected to the sorption kinetics: in batch mode, a lower nitrate

concentration requires a short residence time, whereas, in the case of a higher

concentration achieving the same sorption efficiency, a longer contact time is

needed. In the case of a fixed bed column, the sorbent dose should be increased.

On the other hand, the driving force of diffusion is greater in the case of higher

concentrations. The sorption performance was maintained well over five ion

exchange cycles. The pH of the nitrate solutions increased slightly during some of

the sorption cycles but the overall increase was negligible (initial pH 5.8–6.6, final

pH 5.5–6.9). (Papers I and II).

The particle size of ETM-PSD was not uniform (based on visual observation)

and rather small compared to commercial resins. This could cause channelling and

an unnecessarily large pressure drop in the column. A column process might not be

the optimal form of application for quaternized LC as such, and therefore other

options should be investigated, such as a mixed tank with subsequent separation by

settling or filtration. Alternatively, agglomeration could also be considered to

increase the particle size.

59

Fig. 13. Breakthrough of NO3-N in a fixed bed column packed with ETM-PSD at varying

initial concentrations in NaNO3 solutions. Adapted from Papers I and II, reprinted by

permission from Elsevier B.V.

4.3 Real waters

4.3.1 Nitrate-contaminated groundwater (C-PSD)

In Paper III, C-PSD was tested in the treatment of nitrate-contaminated ground-

water. The effect of the sorbent dose (0.5–6 g/l) was tested. An adequate uptake of

48% was attained at a dose of 2 g/l, and the uptake increased quite steadily,

reaching a maximum removal of 71% at a dose of 6 g/l (initially 6.2 mg N/l).

Practically complete removal of sulphate was also observed (97%) at a dose of 3 g/l

(initially 25 mg/l) while the nitrate uptake was 59%. The pH was slightly elevated

from the initial 6.3 to about 7, which can be regarded as advantageous in terms of

drinking water production as alkalinization is usually used to raise the pH and

increase the buffering ability of the water. Owing to the ion exchange reactions

chloride was released in the treated water resulting in an increased chloride

60

concentration of 43 mg/l, which remains below the threshold value of 250 mg/l set

by EU legislation (Council Directive 98/83/EC 1998). C-PSD could therefore also

have potential in the production of drinking water. (Paper III).

4.3.2 Mining wastewater (ETM-PSD)

In Paper IV, ETM-PSD was used in the treatment of mining wastewater with a focus

on nitrate removal. The physico-chemical properties of the mining wastewater and

treated samples are presented in Table 6 and a full account including the

concentrations of a variety of metals is given in Paper IV.

Table 6. Properties of mining wastewater before and after treatment with ETM-PSD in

batch and column tests. Adapted from Paper IV, reprinted by permission from Elsevier

B.V.

Parameter Unit Value

Untreated

wastewater

Batch,

resin dose 1 g/l

Column, cycle 1,

fraction 2–3

pH - 8–8.2 8.35 7.8

Conductivity mS/cm 9.4 - -

Cl- mg/l 28 113.3 31

SO42- mg/l 8100 8800 8900

PO4-P mg/l 0.025 0.013 0.013

NH4-N mg/l 17.0 16.3 16.3

NO3-N mg/l 9.3 9.5 9.9

NO2-N µg/l < 0.005 500 520

Co µg/l 5.9 < 3 < 1

Ni µg/l 38.4 < 5 3

U µg/l 2.9 - < 0.2

V µg/l < 5 < 5 < 3

Dashes represent no measurement.

As can be seen in Table 6, the conductivity of the mining wastewater was very high

but within the normal range given for industrial wastewaters (Fondriest

Environmental, Inc. 2014). The wastewater contained neither solids nor colour.

Effect of sorbent dose

After the first batch sorption tests, it became evident that ETM-PSD could not

remove nitrate from this particular wastewater. A negligible nitrate uptake of 9%

61

was attained. Neither was any uptake of sulphate observed even though successful

sulphate removal from groundwater was observed with C-PSD in Paper III. The

reason for this might lie in the difference in magnitude of the initial sulphate levels

(only 25 mg/l in groundwater) and concentrations of other ions. Surprisingly,

however, nickel uptakes of about 85% were observed at doses of 0.5–2 g/l.

Although the initial nickel concentration in the mining wastewater was not

comparable with that of the ammoniacal industrial wastewater (13.3 mg/l)

observed in Paper V—which probably enabled the high removal rates—the result

implies that ETM-PSD has clear potential in the treatment of dilute wastewaters,

such as run-off waters. Dilute concentrations are often especially difficult to treat

with conventional techniques such as chemical precipitation. The threshold value

for nickel in surface water has been set at 20 μg/l by the European Union Directive

(Council Directive 2008/105/EC) whereas it was initially 38.4 μg/l in wastewater.

(Paper IV). The further experimental work on this mining wastewater focused on

nickel removal.

Effect of temperature

In the Arctic hemisphere, mining wastewaters are usually very cold, only about 5–

10°C. To elucidate the sorption behaviour of ETM-PSD at low temperature, mining

wastewater was used for sorption tests at 5°C in addition to room temperature

(23°C), testing different contact times simultaneously. The nickel sorption pattern

was very similar at both temperatures but higher and faster uptakes were observed

at 23°C. In 60 minutes, 31% of nickel was removed at 5°C while a 50% uptake was

recorded at 23°C. An approximate 86–90% uptake was achieved in 24 hours at both

temperatures. (Paper IV).

Fixed bed column studies

In Paper IV, mining wastewater was also tested in column with an extended test

series simulating a real ion exchange process. The test included eight ion exchange

cycles (1500 ml each) and eight regeneration cycles, and two maintenance cycles

with HCl after regeneration cycle 3 and 6. The focus was on nickel removal, and

uptakes of 84–100% were recorded. Uptake of uranium (38–93%) and cobalt

(87–93%) were also detected in the first sorption cycle. Fractions in the other

sorption cycles were not analysed in terms of U and Co. Neutron activation analysis

(NAA) revealed a variety of other metals on the surface of exhausted ETM-PSD

62

(Paper IV). The material’s affinity for cations could be explained by the secondary

amine groups originating from ethylenediamine, which are likely to possess cation

binding properties due to the lone electron pair (Torres et al. 2006). However, the

sorption mechanism has not been studied in depth because it was not in the scope

of this study. According to Krestou & Panias (2004), uranium appears as an anion

at pH 7 and above, which could explain the results (the initial pH of the mining

wastewater was 8). A weak nitrate uptake (7–29%) was detected at the beginning

of each of the eight ion exchange cycles but no nitrate was sorbed during the rest

of the cycles (Paper IV).

No decrease in sorption performance with regard to nickel or nitrate was

observed during the eight sorption cycles. The maintenance cycles with HCl after

regeneration cycles 3 and 6 had no suppressing effect on the nickel sorption

performance of ETM-PSD. The high affinity for nickel suggested the nickel

selectivity of ETM-PSD, yet the initial concentration was higher than that of U and

Co. Nickel was detected in the HCl maintenance solutions, and in low

concentrations in the NaCl regeneration solutions, together with nitrate. Naturally,

in a real-life situation the need for maintenance would be less frequent compared

to this simulation, and therefore, it might be too early to make conclusions on the

desorption efficiency of NaCl and HCl, even though the maintenance cycles

seemed to be more efficient in restoring the sorption efficiency of the material.

Nevertheless, this trial produced valuable data that can be used for the design of

further experiments.

4.3.3 Ammoniacal industrial wastewater (ETM-PSD)

In Paper V, ammoniacal industrial wastewater from a chemical plant containing

high amounts of Ni and V was treated with ETM-PSD. Over 99% of vanadium was

removed in batch tests at all pH values (pH 5.1, 7 and 9). As in the case of mining

wastewater, excellent nickel uptake was observed (92% at pH 7 and 97% at pH 9),

and somewhat poorer uptake at the unadjusted pH of 5.1 (56%). Exposing ETM-

PSD to HCl or H2O2 did not inhibit vanadium sorption from the synthetic solution,

but the H2O2 treatment caused the material to break down, resulting in poor settling

and turbidity. (Paper V).

In the first column tests it was discovered that vanadium could not be desorbed

from ETM-PSD in Cl- form using NaCl solution. Hence, prior to the column

experiment, ETM-PSD was treated (activated) with 2 M NaOH to convert the

material from Cl- form into OH- form. Vanadium removal was efficient in column

63

and the material was successfully regenerated using 2 M NaOH. Nickel could not

be desorbed from the material using NaOH. (Paper V). In our previous study on the

same wastewater, nickel fouling was observed when using the commercial anion

exchange resin, Amberjet 4200 Cl, in column (Keränen et al. 2015a). An acid rinse

(1 M H2SO4) was used to successfully desorb nickel (Keränen et al. 2015a) but this

method was not employed in this study. Based on these observations, ETM-PSD

again showed good potential for the treatment of real wastewaters.

4.4 Summary

The above results are an excellent example of the fundamental problem in the case

of synthetic solutions and real wastewaters. As observed in Paper I, the sorption of

nitrate onto ETM-PSD from a synthetic solution was very fast. However, in the

studies of Paper IV, the very same material showed no affinity towards nitrate in

the case of a real mining wastewater although it removed nickel successfully. In

Paper V, excellent vanadium uptake was detected both from synthetic solution,

even in the presence of 5000 mg/l SO4-S, and ammoniacal industrial wastewater

using ETM-PSD. These results show how essential tests on real wastewaters are. It

is true that research on model compounds is the only way to simulate and model

the phenomena in real, non-ideal systems as accurately as possible and to produce

data comparable to other studies. The importance of the research on synthetic

solutions cannot be denied. However, in the light of these results regarding real

waters, it is obvious that model compounds alone simply cannot provide detailed

information on the behaviour of real systems that would help in the design of a real

treatment process.

Table 7 presents a summary of the sorption capacities of ETM-PSD and

C-PSD for NO3-N, PO4-P, SO4-S, Ni and V in synthetic solutions and real waters.

Most of the capacity values have been calculated from the batch test results

(e.g. the test series on the effect of sulphate and phosphate) to enable comparison

with the maximum capacity values. In terms of nitrate, ETM-PSD performed better.

Conversely, C-PSD was not tested on other ions, mining wastewater or industrial

wastewater, which would have provided more detailed information on the

performance of the material. Hence, the table does provide an adequate comparison

between ETM-PSD and C-PSD but rather a summary of capacities achieved using

ETM-PSD. Extended tests on C-PSD are definitely needed to enable a thorough

assessment of its applicability in water treatment.

64

Table 7. Comparison of sorption capacities [mg/g] of ETM-PSD and C-PSD on various

ions (on elemental basis) and solutions. The Roman numerals in parentheses refer to

the original publications.

Solution ETM-PSD C-PSD

NaNO3 solution, NO3-N 30.11 (I), 32.81 (II) 15.31 (III)

NaNO3 (30 mg N/l) + Na3PO4 (50 mg P/l) solution, PO4-P 15.4 (II) -

Na3PO4 (50 mg P/l) solution, PO4-P 14.4 (II) -

NaNO3 (30 mg N/l) + Na2SO4 (500 mg S/l) solution, SO4-S 43.9 (II) -

Na2SO4 solution, SO4-S 40.9 (II) -

NO3-contaminated groundwater, NO3-N (6.3 mg N/l) - 6.8 (III)

Ammoniacal industrial wastewater

Ni (13.3 mg Ni/l) 2.3 (V) -

V (50.4 mg V/l) 1031 (V) -

NaVO3 solution, V 1301 (V) -

1 Maximum capacity.

65

5 Economic feasibility

The cost of water treatment chemicals has a significant impact on the overall

economics of the treatment method. An estimate of the total production costs would

be required to assess the feasibility of modified LC in water treatment more

accurately. At such an early stage of research, i.e. without proper knowledge about

the scale-up of the modification, it is very difficult to create a fully comprehensive

assessment of production costs. This evaluation thus aims at creating an

approximate cost estimate, and comparing the production costs and material

properties with various bio-based anion exchange materials found in the literature,

and with the present commercial ion exchange resins. This approach, however, does

not take into account the profit margin of commercial resins, which is included in

the market price, but it provides a reasonable estimate.

On account of their inherent nature of being waste, LC raw materials, as they

are defined in this thesis, are often regarded free of costs. In simple terms, total

costs can be divided into costs derived from collection, transport, pre-treatment,

chemicals and electricity. The instrumentation and the expenses derived thereof

have been left out of consideration to simplify the calculations.

Perhaps the most essential consideration in cost analysis is the comparison

between the two quaternization methods presented and applied in this thesis,

namely the ETM and CHMAC methods. In fact, the cost assessment could be

carried out by focusing merely on the chemical costs of these two methods, since

the expenses of raw material acquisition (assuming that the same raw material, e.g.

sawdust, is used), i.e. transport and pre-treatment, can be considered to be equal.

Some estimates regarding the chemical costs of quaternized LC were found in the

literature and this information was used as a reference for the materials prepared in

this work. Table 8 presents estimates of the chemical costs of ETM-PSD and

C-PSD with comparison to materials prepared through the same modification

methods, except for the case of CHMAC modified sugarcane bagasse described by

Laszlo (1996), where epichlorohydrin was used as a crosslinking agent. In addition

to chemical costs, chemical doses, nitrogen contents and sorption capacities are

compared.

66

Ta

ble

8. C

om

pa

riso

n o

f e

sti

mate

d c

hem

ica

l co

sts

, c

he

mic

al d

os

es

, n

itro

ge

n c

on

ten

ts a

nd

so

rpti

on

ca

pa

cit

ies

of

va

rio

us

ET

M-

an

d

CH

MA

C-m

od

ifie

d L

C m

ate

ria

ls. C

hem

ica

l co

sts

[U

SD

/kg

of

fin

al p

rod

uct]

ha

ve

be

en

es

tim

ate

d u

sin

g t

he c

urr

en

t p

ric

es

giv

en

in

th

e

foo

tno

te. T

he R

om

an

nu

me

rals

refe

r to

th

e o

rig

ina

l p

ub

lic

ati

on

s.

Raw

mate

ria

l D

ose

[m

l/g o

f LC

]

Yie

ld

[%]

Cost

[US

D/k

g]

N c

onte

nt

(orig

. N

) [%

]

Qm

ax

for

NO

3-N

or

PO

4-P

[m

g/g

] R

efe

rence

E

PI

DM

F

ED

A

TE

A

NaO

H

CH

MA

C

To

tal

Pin

e s

aw

du

st

6.5

8

.0

1.2

5

6.5

-

- 2

2.3

55

0

4.5

8

9.4

(0

.8)

30

.1;

32

.8

I, I

I

Wh

ea

t st

raw

5

.0

5.0

0

.75

5

.0

- -

15

.75

na

-

14

.5 (

0.3

5)

7.5

1

Xu e

t al.

2010a

Wh

ea

t st

raw

1

.25

1

.25

0

.4

1.0

-

- 3

.9

n

a

- n

a

na

T

an e

t al.

2012

Gia

nt

reed

2

.5

1.2

5

0.5

2

.5

- -

6.7

5

2

42

3

.58

7

.8 (

0.9

) n

a

Wa

ng e

t al.

2010

Su

ga

rca

ne

ba

ga

sse

2.0

2

.5

0.4

2

.0

- -

6.9

na

-

na

1

5.3

1

Hena e

t al.

2015

Su

ga

rca

ne

ba

ga

sse

10

.0

12

.0

2.0

1

0.0

-

- 3

4.0

34

7

11

.14

6

.5 (

0.9

) 2

0.4

M

ao e

t al.

2016

Pin

e s

aw

du

st

- -

- -

10

.02

10

.0

20

.0

1

05

1

4.3

6

2.5

(0

.8)

15

.3

III

So

ybe

an

hu

lls

- -

- -

5.0

3

5.0

10.0

65

11.8

3

na

19.5

1

Ma

rsh

all

& W

art

elle

20

04

Co

rn s

tove

r -

- -

- 5

.03

5.0

10.0

71

10.8

6

2.8

(0.8

5)

na

Wart

elle

& M

ars

ha

ll 2006

Su

ga

rca

ne

ba

ga

sse

0.2

4

- -

- 1

.25

3

1.0

2.4

9

130

2.4

6

2.4

(0.1

) na

Lasz

lo 1

99

6

Su

ga

rca

ne

ba

ga

sse

- -

- -

1.2

53

1.0

2.2

5

82

1.9

2

(0.1

) na

Lasz

lo 1

99

6

Am

be

rje

t 4

20

0 C

l -

- -

- -

- -

-

7.5

5

na

2

3.2

4

I

Am

be

rlite

IR

A 9

00

-

- -

- -

- -

-

na

n

a

21

.0;

16

.8

Orl

an

do

et a

l. 2

00

2a,b

Wh

atm

an

QA

52

-

- -

- -

- -

-

na

1

5.0

(n

a)

na

W

art

elle

& M

ars

ha

ll 2

00

6

1 P

O4-P

capa

city

; 2 C

onc.

21.2

% (

w/v

); 3

Con

c. 5

N;

4 N

ot a m

ax.

capa

city

. na: N

ot ava

ilable

. O

rig. N

: N

conte

nt

in u

nm

odifi

ed s

am

ple

. P

rice

s [U

SD

/kg]:

EP

I 1

.60

, D

MF

0.6

0,

ED

A 1

.10

, T

EA

1.5

0,

CH

MA

C (

65

%)

1.2

0,

Na

OH

(2

1.2

%)

0.1

0.

So

me

of

the

co

sts

cou

ld n

ot

be

de

term

ine

d d

ue

to

mis

sing y

ield

data

.

67

The dosage reported by Wang et al. (2010), 6.75 ml in total, was used as a

guideline in the design of the first ETM modifications. However, it was not long

before the chemical doses had to be increased dramatically to achieve adequate

mixing conditions (a total of 22.3 ml/g of PSD was used in this work). However,

the high mass yield caused the chemical costs per gram of final product to fall. The

chemical doses used by Tan et al. (2012), Wang et al. (2010) and Hena et al. (2010a)

are very low (3.9–6.9 ml/g LC), which is likely to decrease the chemical costs

significantly depending on the mass yield. It might be that PSD is more absorbent

than reed or wheat residue despite using the same size fraction. Unfortunately, data

on mass yield were not reported in many of the referenced studies, and therefore,

the chemical costs could not be determined. For instance, Xu et al. (2010a) used a

total of 15.75 ml reagents for the treatment of wheat straw. The chemical doses and

their effect on the quaternization reactions and mixing were not optimized, and this

makes a full comparison between the expenses impossible.

Wartelle & Marshall (2006) have estimated the production costs of CHMAC-

modified corn stover, and assumed that chemical costs accounted for the majority

of the total costs. They used 5 ml of both NaOH (5 N) and CHMAC (65%) per

gram of corn stover (Wartelle & Marshall 2006). They concluded that, by recycling

93.1% of CHMAC, the chemical costs could be reduced from 25 USD/kg down to

7.40 USD/kg (Wartelle & Marshall 2006). These costs were re-estimated as

10.86 USD/kg (without recycling) and 4.41 USD/kg (with recycling) using current

chemical prices. The recycling of CHMAC was not considered in this thesis, but it

is a potential means of reducing the production costs. It was observed in the course

of this work that the chemical doses could be reduced from 10 ml NaOH and 10 ml

CHMAC to 6 ml of each. Therefore, the chemical costs of C-PSD could be reduced

from 14.36 USD/kg down to 8.62 USD/kg according to our unpublished

observations. Laszlo (1996) estimated the production costs of CHMAC modified

sugarcane bagasse crosslinked with epichlorohydrin to be 2 USD/kg of resin.

Table 8 also compares the chemical doses, yields and sorption capacities of

various ETM materials. In this work, high chemical doses resulted in high nitrogen

contents in the final modified PSD product (9.4%) (Paper I). Considering the results

obtained previously (Wang et al. 2010, Xu et al. 2010a), successful quaternization

can be achieved through smaller chemical doses without losing the quaternary

ammonium groups and anion exchange properties, which means that the chemical

costs of ETM-PSD need not be as high as suggested. For instance, the DMF dose

could be decreased because of its role as a mere inert solvent as long as a well-

mixing slurry is maintained.

68

The ecological aspects regarding modification can also be taken into

consideration. Given that the CHMAC method involves only two chemicals, of

which only one is classified as posing a possible carcinogenic threat, whereas all

four of the reagents used in the ETM method are considered immediately hazardous

or carcinogenic, the CHMAC method was regarded as the safer alternative.

However, the production process is likely to be operated in a closed circuit system,

which increases the safety aspect of the ETM method.

Due to the addition of modification reagents onto the LC structure during ETM

synthesis, a large proportion of the ETM material is composed of fossil, oil-based

components. However, the bio-based skeletal structure of the ETM products

increases the share of bio-based components compared to conventional ion

exchange resins. Therefore, the products are seen as an important step towards the

development of greener chemicals. At such an early stage of development, the high

proportion of fossil components is inevitable, but the development of bio-based

bulk chemicals and modification methods could add to the share of biocomponents.

Epichlorohydrin is already produced from bio-based glycerol, a side-product of

bio-diesel production through transesterification (Sheldon 2014), which is,

however, likely to increase the production costs of an ETM material, but is a

necessary step towards greener, bio-based chemistry and syntheses.

At this stage, the economic feasibility of the bio-based anion exchange

materials looks weak but that is partly due to the poor comparability of the data.

There might be differences between the absorption properties of wheat straw and

sawdust, which could explain the extremely low chemical doses used by Xu et al.

(2010a, 2011), Wang et al. (2010) and Xing et al. (2011). Based on the present

analysis, it is clear that the chemical consumption in the ETM and CHMAC

methods should be optimized in order to create a thorough estimate of the chemical

costs.

69

6 Conclusions

This work showed the potential of Finnish lignocellulosic waste materials as bio-

based platforms for the preparation of ion exchange materials, and their application

in the treatment of real waters. Chemical modification methods were employed to

produce anion exchange materials through quaternization of LC: the ETM method

involving epichlorohydrin, ethylenediamine, triethylamine and N,N-dimethyl-

formamide, and the CHMAC method involving N-(3-chloro-2-hydroxypropyl)

trimethylammonium chloride and NaOH. In addition to this, cation sorption was

studied by treating PSD with citric acid to anionize the surface and nickel removal

studies. By combining the citric acid and CHMAC treatments, the preparation of a

hybrid ion exchange material was also attempted. New information about the

treatment of real waters using bio-based sorbents was also presented which is

essential for the development of new, sustainable water treatment chemicals.

The removal of NO3 was very rapid with ETM-PSD in a wide temperature and

pH range using synthetic solutions. Real wastewaters were tested as well, both in

batch and fixed bed column tests, to gain information about the sorption in complex,

multi-ion systems. Fixed bed column studies proved the ease of regeneration with

NaCl and NaOH, and the sorption efficiency was maintained over multiple ion

exchange cycles, which confirmed ion exchange to be the dominant sorption

mechanism. Uptake of Ni and V was observed from real wastewaters, which

implied the suitability of ETM-PSD for the treatment of real waters. Whether the

exact form of application should be a fixed bed column or a stirred tank is yet to be

investigated. The sorption efficiency of ETM-PSD varied significantly, depending

on the solution and its composition. Therefore, it must be kept in mind that the

suitability of a sorbent material for the removal of a specific pollutant must be

tested individually for each wastewater, because the ionic composition of

wastewaters varies depending on their source and related process conditions.

However, this is true for all sorbents, not only for novel bio-sorbents but also for

commercial materials.

C-PSD also showed potential for nitrate removal from real groundwater and

mining wastewater. More information is still needed about the removal of other

ions and their combined effects, and about the column behaviour. More research on

real waters is also necessary to form a comprehensive evaluation of the

applicability of C-PSD, but this remains to be studied in future work.

Both the ETM and CHMAC methods are based on the addition of oil-based

chemical substituents on LC. However, increasing the proportion of bio-based

70

components in water treatment chemicals by using a bio-polymer like LC as a

platform is a necessary step towards the development of greener products and

chemistry and would help reduce the dependency on fossil raw materials. The use

of the entire LC structure without the need to separate certain compounds, like

lignin or cellulose, to be used as a platform is advantageous because it simplifies

the overall production process thereby reducing costs.

For the commercialization of both the ETM and CHMAC methods, the

chemical doses should be optimized to avoid unnecessary formation of waste. In

later stages of research and development, the successful scale-up of the synthesis

reactions will also become the cornerstone of the whole process and its feasibility.

Thorough scale-up trials are therefore required. In order to create a comprehensive

assessment of the ETM and CHMAC methods and to compare the products, the

column behaviour of C-PSD should be studied. In addition, the effect of particle

size on the column behaviour should be investigated.

71

References

Aamulehti (2014) Suomen maassa suuressa joko valmiina tai suunnitteilla 29 puupohjaista raaka-ainetta käyttävää biojalostamoa. Aamulehti, May 23, 2014. In Finnish. URI: http://aamulehdenblogit.ning.com/profiles/blogs/suomessa-29-puupohjaista-raaka-ainetta-k-ytt-v-biojalostamoa. Cited 2016-05-12.

Abbott AP, Bell TJ, Handa S & Stoddart B (2006) Cationic functionalisation of cellulose using a choline based ionic liquid analogue. Green Chem 8(9): 784–786.

Anirudhan T & Rauf TA (2013) Adsorption performance of amine functionalized cellulose grafted epichlorohydrin for the removal of nitrate from aqueous solutions. J Ind Eng Chem 19(5): 1659–1667.

Anirudhan TS, Rauf TA & Rejeena SR (2012) Removal and recovery of phosphate ions from aqueous solutions by amine functionalized epichlorohydrin-grafted cellulose. Desalination 285: 277–284.

Anirudhan TS & Unnithan MR (2007) Arsenic(V) removal from aqueous solutions using an anion exchanger derived from coconut coir pith and its recovery. Chemosphere 66(1): 60–66.

Ansari R, Khoshbakht Fahim N & Fallah Delavar A (2009) Removal of nitrite ions from aqueous solutions using conducting electroactive polymers. The Open Process Chemistry Journal 2: 1–5.

Antal M, Ebringerová A & Šimkovic I (1984) New aspects in cationization of lignocellulose materials. I. Preparation of lignocellulose materials containing quarternary ammonium groups. J Appl Polym Sci 29(2): 637–642.

Argun ME & Dursun S (2008) Cadmium removal using activated pine bark. J Int Environ Appl Sci 3(1): 37–42.

Argun ME, Dursun S, Ozdemir C & Karataş M (2007) Heavy metal adsorption by modified oak sawdust: Thermodynamics and kinetics. J Hazard Mater 141(1): 77–85.

Asadi F, Shariatmadari H & Mirghaffari N (2008) Modification of rice hull and sawdust sorptive characteristics for remove heavy metals from synthetic solutions and wastewater. J Hazard Mater 154(1–3): 451–458.

Biermann CJ (1996) 22 - Papermaking Chemistry. In Biermann CJ (ed) Handbook of pulping and papermaking (second edition). Academic Press, San Diego, USA. pp. 438–462.

Brown PA, Gill SA & Allen SJ (2000) Metal removal from wastewater using peat. Water Res 34(16): 3907–3916.

Chabani M, Amrane A & Bensmaili A (2006) Kinetic modelling of the adsorption of nitrates by ion exchange resin. Chem Eng J 125(2): 111–117.

Chattopadhyay D (2001) Cationization of cotton for low-salt or salt-free dyeing. Indian J Fibre Text Res 26(1/2): 108–115.

Coma V, Sebti I, Pardon P, Pichavant F & Deschamps A (2003) Film properties from crosslinking of cellulosic derivatives with a polyfunctional carboxylic acid. Carbohydr Polym 51(3): 265–271.

72

Council Directive 91/676/EEC of 12 December 1991 concerning the protection of waters against pollution caused by nitrates from agricultural sources.

Council Directive 98/83/EC of 3 November 1998 on the quality of water intended for human consumption.

Council Directive 2008/105/EC of 16 December 2008 on environmental quality standards in the field of water policy, amending and subsequently repealing Council Directives 82/176/EEC, 83/513/EEC, 84/156/EEC, 84/491/EEC, 86/280/EEC and amending Directive 2000/60/EC of the European Parliament and of the Council.

DeSilva FJ (1999) Essentials of ion exchange. 25th Annual WQA Conference, 17 Mar 1999. Dizhbite T, Zakis G, Kizima A, Lazareva E, Rossinskaya G, Jurkjane V, Telysheva G &

Viesturs U (1999) Lignin — a useful bioresource for the production of sorption-active materials. Bioresour Technol 67(3): 221–228.

Ebrahimi S & Roberts DJ (2013) Sustainable nitrate-contaminated water treatment using multi cycle ion-exchange/bioregeneration of nitrate selective resin. J Hazard Mater 262: 539–544.

Eliaz N & Sridhar T (2008) Electrocrystallization of hydroxyapatite and its dependence on solution conditions. Cryst Growth Des 8(11): 3965–3977.

European Environmental Agency (2015) Nitrates in groundwater. URI: http://www.eea.europa.eu/data-and-maps/indicators/nitrate-in-groundwater-1. Cited 2016-02-29.

Fondriest Environmental, Inc. (2014) Conductivity, salinity and total dissolved solids. Fundamentals of Environmental Measurements. 3 Mar 2014. URI:

http://www.fondriest.com/environmental-measurements/parameters/water-quality/ conductivity-salinity-tds/. Cited 2016-06-22.

Gao B, Xu X, Wang Y, Yue Q & Xu X (2009) Preparation and characteristics of quaternary amino anion exchanger from wheat residue. J Hazard Mater 165(1–3): 461–468.

Häyrynen K, Langwaldt J, Pongrácz E, Väisänen V, Mänttäri M & Keiski RL (2008) Separation of nutrients from mine water by reverse osmosis for subsequent biological treatment. Miner Eng 21(1): 2–9.

Hegazi HA (2013) Removal of heavy metals from wastewater using agricultural and industrial wastes as adsorbents. HBRC Journal 9(3): 276–282.

Hell F, Lahnsteiner J, Frischherz H & Baumgartner G (1998) Experience with full-scale electrodialysis for nitrate and hardness removal. Desalination 117(1–3): 173–180.

Hena S, Atikah S & Ahmad H (2015) Removal of phosphate ion from water using chemically modified biomass of sugarcane bagasse. Int J Eng Sci 4(1): 51–62.

Ilies P & Mavinic DS (2001) The effect of decreased ambient temperature on the biological nitrification and denitrification of a high ammonia landfill leachate. Water Res 35(8): 2065–2072.

Jensen VB, Darby JL, Seidel C & Gorman C (2012) Technical report 6: Drinking water treatment for nitrate - With a focus on Tulare Lake basin and Salinas Valley ground-water.

Karachalios A & Wazne M (2013) Nitrate removal from water by quaternized pine bark using choline based ionic liquid analogue. J Chem Technol Biotechnol 88(4): 664–671.

73

Katal R, Baei MS, Rahmati HT & Esfandian H (2012) Kinetic, isotherm and thermodynamic study of nitrate adsorption from aqueous solution using modified rice husk. J Ind Eng Chem 18(1): 295–302.

Keränen A, Leiviskä T, Salakka A & Tanskanen J (2015) Removal of nickel and vanadium from ammoniacal industrial wastewater by ion exchange and adsorption on activated carbon. Desalin Water Treat 53(10): 2645–2654.

Kong F, Parhiala K, Wang S & Fatehi P (2015) Preparation of cationic softwood kraft lignin and its application in dye removal. Eur Polym J 67: 335–345.

Korpinen R (2010) On the potential utilisation of sawdust and wood chip screenings. Doctoral thesis. Åbo Akademi, Finland.

Krestou A & Panias D. Uranium (VI) speciation diagrams in the UO22+/CO3

2-/H2O system at 25°C. Eur J Miner Process Environ Protect 2004(4): 113–129.

Lange H, Decina S & Crestini C (2013) Oxidative upgrade of lignin – Recent routes reviewed. Eur Polym J 49(6): 1151–1173.

Laszlo JA (1996) Preparing an ion exchange resin from sugarcane bagasse to remove reactive dye from wastewater. Text Chem Color 28(5): 13–17.

Leiviskä T, Keränen A, Vainionpää N, Al Amir J, Hormi O & Tanskanen J (2015) Vanadium(V) removal from aqueous solution and real wastewater using quaternized pine sawdust. Water Sci Technol 72(3): 437–442.

Liu C (2013) Can straw provide China's energy needs? Scientific American, 23 Feb 2013. URI: http://www.eea.europa.eu/data-and-maps/indicators/nitrate-in-groundwater-1. Cited 2016-03-01.

Low KS & Lee CK (1997) Quaternized rice husk as sorbent for reactive dyes. Bioresour Technol 61(2): 121–125.

Low KS, Lee CK & Mak SM (2004) Sorption of copper and lead by citric acid modified wood. Wood Sci Technol 38(8): 629–640.

Mannonen S (2015) Innovaatiot edellä – UPM valmistaa mäntyöljystä vähäpäästöistä uusiutuvaa dieseliä. Biotalous, May 18, 2015. In Finnish. URI: http://www.biotalous.fi/ innovaatiot-edella-upm-valmistaa-mantyoljysta-vahapaastoista-uusiutuvaa-dieselia/. Cited 2017-02-15.

Mao Y, Zhang M, Yu D, Wang W, Song Z, Zhao X & Yue Q (2016) Improving the properties of an anion exchanger based on sugarcane bagasse by applying pretreatment methods. Desalin Water Treat 57(38): 17944–17954.

Marshall WE & Wartelle LH (2004) An anion exchange resin from soybean hulls. J Chem Technol Biotechnol 79(11): 1286–1292.

Marshall WE, Wartelle LH, Boler DE, Johns MM & Toles CA (1999) Enhanced metal adsorption by soybean hulls modified with citric acid. Bioresour Technol 69(3): 263–268.

Menzel C, Olsson E, Plivelic TS, Andersson R, Johansson C, Kuktaite R, Järnström L & Koch K (2013) Molecular structure of citric acid cross-linked starch films. Carbohydr Polym 96(1): 270–276.

74

Ministry of Social Affairs and Health (2014) Sosiaali- ja terveysministeriön asetus talousveden laatuvaatimuksista ja valvontatutkimuksista annetun sosiaali- ja terveys-ministeriön asetuksen muuttamisesta (442/2014). In Finnish.

Nachtergaele W (1989) The benefits of cationic starches for the paper industry. Starch - Stärke 41(1): 27–31.

Naja G & Volesky B (2011) The mechanism of metal cation and anion biosorption. In: Kotrba P, Mackova M & Macek T (eds) Microbial Biosorption of Metals. Germany, Springer-Verlag: 19–58.

Namasivayam C & Höll WH (2005) Quaternized biomass as an anion exchanger for the removal of nitrate and other anions from water. J Chem Technol Biotechnol 80(2): 164–168.

Namasivayam C & Sangeetha D (2006) Removal and recovery of vanadium(V) by adsorption onto ZnCl2 activated carbon: Kinetics and isotherms. Adsorption 12:103–117.

Nollet LLM (ed) (2000) Handbook of water analysis. New York, USA, Marcel Dekker. Orlando US, Baes AU, Nishijima W & Okada M (2002a) A new procedure to produce

lignocellulosic anion exchangers from agricultural waste materials. Bioresour Technol 83(3): 195–198.

Orlando US, Baes AU, Nishijima W & Okada M (2002b) Preparation of agricultural residue anion exchangers and its nitrate maximum adsorption capacity. Chemosphere 48(10): 1041–1046.

Orlando US, Okuda T, Baes AU, Nishijima W & Okada M (2003) Chemical properties of anion-exchangers prepared from waste natural materials. React Funct Polym 55(3): 311–318.

Öztürk N & Bektaş TE (2004) Nitrate removal from aqueous solution by adsorption onto various materials. J Hazard Mater 112(1–2): 155–162.

Pepper KW, Paisley HM & Young MA (1953) 833. Properties of ion-exchange resins in relation to their structure. Part VI. Anion-exchange resins derived from styrene-divinyl-benzene copolymers. J Chem Soc: 4097–4105.

Pereira FV, Gurgel LVA & Gil LF (2010) Removal of Zn2+ from aqueous single metal solutions and electroplating wastewater with wood sawdust and sugarcane bagasse modified with EDTA dianhydride (EDTAD). J Hazard Mater 176(1–3): 856–863.

Politi D & Sidiras D (2012) Wastewater treatment for dyes and heavy metals using modified pine sawdust as adsorbent. Procedia Eng 42: 1969–1982.

Räisänen T & Athanassiadis D (2013) Basic chemical composition of the biomass components of pine, spruce and birch. URI: http://www.biofuelregion.se/UserFiles/file/ Forest%20Refine/1_2_IS_2013-01-31_Basic_chemical_composition.pdf. Cited 2016-05-12.

Raji C & Anirudhan T (1998) Batch Cr(VI) removal by polyacrylamide-grafted sawdust: kinetics and thermodynamics. Water Res 32(12): 3772–3780.

Reddy N & Yang Y (2010) Citric acid cross-linking of starch films. Food Chem 118(3): 702–711.

75

Ren JL, Sun RC, Liu CF, Lin L & He BH (2007) Synthesis and characterization of novel cationic SCB hemicelluloses with a low degree of substitution. Carbohydr Polym 67(3): 347–357.

Sable I, Grinfelds U, Jansons A, Vikele L, Irbe I, Verovkins A & Treimanis A (2012) Properties of wood and pulp fibers from logdepole pine (Pinus contorta) as compared to Scots pine (Pinus sylvestris). Bioresources 7(2): 1771–1783.

Salam A, Pawlak JJ, Venditti RA & E-tahlawy K (2011) Incorporation of carboxyl groups into xylan for improved absorbency. Cellulose 18: 1033–1041.

Salazar-Rabago J & Leyva-Ramos R (2016) Novel biosorbent with high adsorption capacity prepared by chemical modification of white pine (Pinus durangensis) sawdust. Adsorption of Pb(II) from aqueous solutions. J Environ Manage 169: 303–312.

SFS-EN ISO 872 (2005) Water quality. Determination of suspended solids. Method by filtration through glass fibre filters. Helsinki. Finnish Standards Association.

SFS-EN ISO 9963-1 (1996) Water quality. Determination of alkalinity. Part 1: Determination of total and composite alkalinity. In Finnish. Helsinki. Finnish Standards Association.

SFS 3032 (2000) Veden ammonioumtypen määritys. Helsinki. Finnish Standards Association.

Sheldon RA (2014) Green and sustainable manufacture of chemicals from biomass: state of the art. Green Chem 16(3): 950–963.

Šimkovic I, Mlynár J, Alföldi J & Antal M (1987) New aspects in cationization of lignocellulose materials. VII. Modification of spruce wood meal with quarternary ammonium groups. J Appl Polym Sci 33(8): 2651–2656.

Šimkovic I (1999) Preparation of anion exchangers from beech sawdust and wheat straw. Ind Crops Prod 10(3): 167–173.

Šimkovic I & Laszlo JA (1997) Preparation of ion exchangers from bagasse by crosslinking with epichlorohydrin-NH4OH or epichlorohydrin-imidazole. J Appl Polym Sci 64(13): 2561–2566.

Sirviö JA, Hasa T, Leiviskä T, Liimatainen H & Hormi O (2016) Bisphosphonate nanocellulose in the removal of vanadium(V) from water. Cellulose 23(1): 689–697.

Smith BC (1999) Infrared spectral interpretation: A systematic approach. CRC Press, Boca Raton, USA. ISBN 0-8493-2463-7.

Sowmya A & Meenakshi S (2014) A novel quaternized chitosan–melamine–glutaraldehyde resin for the removal of nitrate and phosphate anions. Int J Biol Macromol 64: 224–232.

Steiger FH (1972) Personal Products Company, assignee. Absorbent products from wet cross-linked pulpboard and methods of making the same. Patent US3,658,613.

Tan X, Gao B, Xu X, Wang Y, Ling J, Yue Q & Li Q (2012) Perchlorate uptake by wheat straw based adsorbent from aqueous solution and its subsequent biological regeneration. Chem Eng J 211–212: 37–45.

Taty-Costodes VC, Fauduet H, Porte C & Delacroix A (2003) Removal of Cd(II) and Pb(II) ions, from aqueous solutions, by adsorption onto sawdust of Pinus sylvestris. J Hazard Mater 105(1–3): 121–142.

76

Torres JD, Faria EA & Prado AGS (2006) Thermodynamic studies of the interaction at the solid/liquid interface between metal ions and cellulose modified with ethylenediamine. J Hazard Mater 129(1–3): 239–243.

U.S. Geological Survey (2016) The California drought. URI: http://ca.water.usgs.gov/ data/drought/. Cited 2016-03-11.

Velazquez-Jimenez LH, Pavlick A & Rangel-Mendez JR (2013) Chemical characterization of raw and treated agave bagasse and its potential as adsorbent of metal cations from water. Ind Crops Prod 43: 200–206.

Wang WY, Yue QY, Xu X, Gao BY, Zhang J, Li Q & Xu JT (2010) Optimized conditions in preparation of giant reed quaternary amino anion exchanger for phosphate removal. Chem Eng J 157(1): 161–167.

Wang Y, Gao BY, Yue WW & Yue QY (2007) Preparation and utilization of wheat straw anionic sorbent for the removal of nitrate from aqueous solution. J Environ Sci 19(11): 1305–1310.

Wartelle L & Marshall W (2000) Citric acid modified agricultural by-products as copper ion adsorbents. Adv Environ Res 4(1): 1–7.

Wartelle LH & Marshall WE (2006) Quaternized agricultural by-products as anion exchange resins. J Environ Manage 78(2): 157–162.

Weiser C, Zeller V, Reinicke F, Wagner B, Majer S, Vetter A & Thraen D (2014) Integrated assessment of sustainable cereal straw potential and different straw-based energy applications in Germany. Appl Energy 114: 749–762.

Wing RE (1996) Corn fiber citrate: preparation and ion-exchange properties. Ind Crops Prod 5(4): 301–305.

Witek-Krowiak A, Chojnacka K, Podstawczyk D, Dawiec A & Pokomeda K (2014) Application of response surface methodology and artificial neural network methods in modelling and optimization of biosorption process. Bioresour Technol 160: 150–160.

Worch E (2012) Adsorption technology in water treatment: fundamentals, processes, and modeling. Berlin/Boston, Walter de Gruyter.

World Health Organization (2011) Guidelines for drinking-water quality. Vol. 1, fourth edition.

Xing X, Gao BY, Zhong QQ, Yue QY & Li Q (2011) Sorption of nitrate onto amine-crosslinked wheat straw: Characteristics, column sorption and desorption properties. J Hazard Mater 186(1): 206–211.

Xu X, Gao B, Wang W, Yue Q, Wang Y & Ni S (2010a) Effect of modifying agents on the preparation and properties of the new adsorbents from wheat straw. Bioresour Technol 101(5): 1477–1481.

Xu X, Gao B, Zhao Y, Chen S, Tan X, Yue Q, Lin J & Wang Y (2012a) Nitrate removal from aqueous solution by Arundo donax L. reed based anion exchange resin. J Hazard Mater 203–204: 86–92.

Xu X, Gao BY, Yue QY, Zhong QQ & Li Q (2011) Preparation of new types of anion exchange resins from agricultural by-products and their utilization in the removal of various toxic anions from solutions. Chem Eng J 167(1): 104–111.

77

Xu X, Gao BY, Yue QY & Zhong QQ (2010b) Preparation of agricultural by-product based anion exchanger and its utilization for nitrate and phosphate removal. Bioresour Technol 101(22): 8558–8564.

Xu X, Gao BY, Yue QY, Zhong QQ & Zhan X (2010c) Preparation, characterization of wheat residue based anion exchangers and its utilization for the phosphate removal from aqueous solution. Carbohydr Polym 82(4): 1212–1218.

Xu X, Gao BY, Yue QY & Zhong QQ (2010d) Preparation and utilization of wheat straw bearing amine groups for the sorption of acid and reactive dyes from aqueous solutions. J Hazard Mater 182(1–3): 1–9.

Yang CQ (1991) FT-IR spectroscopy study of the ester crosslinking mechanism of cotton cellulose. Text Res J 61(8): 433–440.

Yang H, Zhang W, Yang Z & Li A (2015) Nanjing University, assignee. Straw modified into strong cation type absorbent and preparation method thereof. Patent CN102600806 (B).

Yoshida N, Ishida K, Sasaki S & Yamaoka I (2014) Tokyo JNC Corporation, assignee. Metal salt of crosslinked cellulose derivative. Patent US8,779,119.

Yu H, Pang J, Ai T & Liu L (2016) Biosorption of Cu2+, Co2+ and Ni2+ from aqueous solution by modified corn silk: Equilibrium, kinetics, and thermodynamic studies. J Taiwan Inst Chem Eng 62: 21–30.

Zeng L & Cheng CY (2009) A literature review of the recovery of molybdenum and vanadium from spent hydrodesulphurisation catalysts. Part II: Separation and purification. Hydrometallurgy 98(1–2): 10–20.

78

79

Original publications

I Keränen A, Leiviskä T, Gao BY, Hormi O & Tanskanen J (2013) Preparation of novel anion exchangers from pine sawdust and bark, spruce bark, birch bark and peat for the removal of nitrate. Chemical Engineering Science 98: 59–68.

II Keränen A, Leiviskä T, Hormi O & Tanskanen J (2015) Removal of nitrate by modified pine sawdust: Effects of temperature and co-existing anions. Journal of Environmental Management 147: 46–54.

III Keränen A, Leiviskä T, Hormi O & Tanskanen J (2015) Preparation of cationized pine sawdust for nitrate removal: Optimization of reaction conditions. Journal of Environmental Management 160: 105–112.

IV Keränen A, Leiviskä T, Zinicovscaia I, Frontasyeva MV, Hormi O & Tanskanen J (2016) Quaternized pine sawdust in the treatment of mining wastewater. Environmental Technology 37(11):1390–1397.

V Leiviskä T, Keränen A, Vainionpää N, Al Amir J, Hormi O & Tanskanen J (2015) Vanadium(V) removal from aqueous solution and real wastewater using quaternized pine sawdust. Water Science and Technology 72(3): 437–442.

Reprinted with permission from Elsevier B.V. (Papers I–III), Taylor and Francis

(Paper IV) and IWA Publishing (Paper V).

Original publications are not included in the electronic version of the dissertation.

80

A C T A U N I V E R S I T A T I S O U L U E N S I S

Book orders:Granum: Virtual book storehttp://granum.uta.fi/granum/

S E R I E S C T E C H N I C A

593. Liyanage, Madhusanka (2016) Enhancing security and scalability of Virtual PrivateLAN Services

594. Darif, Bouchra (2016) Synthesis and characterization of catalysts used for thecatalytic oxidation of sulfur-containing volatile organic compounds : focus onsulfur-induced deactivation

595. Juholin, Piia (2016) Hybrid membrane processes in industrial water treatment :separation and recovery of inorganic compounds

596. Augustine, Bobins (2016) Efficiency and stability studies for organic bulkheterojunction solar cells

597. Ylioinas, Juha (2016) Towards optimal local binary patterns in texture and facedescription

598. Mohammadighavam, Shahram (2017) Hydrological and hydraulic design ofpeatland drainage and water treatment systems for optimal control of diffusepollution

599. Louis, Jean-Nicolas (2016) Dynamic environmental indicators for smart homes :assessing the role of home energy management systems in achieving decarbonisationgoals in the residential sector

600. Mustamo, Pirkko (2017) Greenhouse gas fluxes from drained peat soils : acomparison of different land use types and hydrological site characteristics

601. Upola, Heikki (2017) Disintegration of packaging material : an experimental studyof approaches to lower energy consumption

602. Eskelinen, Riku (2017) Runoff generation and load estimation in drained peatlandareas

603. Kokkoniemi, Joonas (2017) Nanoscale sensor networks : the THz band as acommunication channel

604. Luoto, Petri (2017) Co-primary multi-operator resource sharing for small cellnetworks

605. Yrjölä, Seppo (2017) Analysis of technology and business antecedents forspectrum sharing in mobile broadband networks

606. Suikkanen, Essi (2017) Detection algorithms and ASIC designs for MIMO–OFDMdownlink receivers

607. Niemelä, Ville (2017) Evaluations and analysis of IR-UWB receivers for personalmedical communications

C608etukansi.kesken.fm Page 2 Wednesday, February 22, 2017 11:39 AM

UNIVERSITY OF OULU P .O. Box 8000 F I -90014 UNIVERSITY OF OULU FINLAND

A C T A U N I V E R S I T A T I S O U L U E N S I S

University Lecturer Tuomo Glumoff

University Lecturer Santeri Palviainen

Postdoctoral research fellow Sanna Taskila

Professor Olli Vuolteenaho

University Lecturer Veli-Matti Ulvinen

Planning Director Pertti Tikkanen

Professor Jari Juga

University Lecturer Anu Soikkeli

Professor Olli Vuolteenaho

Publications Editor Kirsti Nurkkala

ISBN 978-952-62-1513-6 (Paperback)ISBN 978-952-62-1514-3 (PDF)ISSN 0355-3213 (Print)ISSN 1796-2226 (Online)

U N I V E R S I TAT I S O U L U E N S I SACTAC

TECHNICA

U N I V E R S I TAT I S O U L U E N S I SACTAC

TECHNICA

OULU 2017

C 608

Anni Keränen

WATER TREATMENTBY QUATERNIZED LIGNOCELLULOSE

UNIVERSITY OF OULU GRADUATE SCHOOL;UNIVERSITY OF OULU,FACULTY OF TECHNOLOGY

C 608

ACTA

Anni K

eränenC608etukansi.kesken.fm Page 1 Wednesday, February 22, 2017 11:39 AM