Environmental Study of the

M A S T E R ' S T H E S I S

Environmental Study of the

Lignocellulose Ethanol Production at the Sekab Pilot Plant (ETEC)

Ebenezer Twumasi

Luleå University of Technology

Master Thesis, Continuation Courses

Preface

This master thesis is based on work carried out during January – June 2007 at Sekab E- technology in Örnsköldsvik, Sweden. It was financially supported by Sekab AB.

First I would like to express my gratitude to Prof. Oleg Antzutkin, Luleå University for proposing me for this thesis work. I am very grateful to the entire staff of E-technology for giving me this opportunity to do this thesis work, especially Torbjörn Lindgren. I wish to especially thank Maria Edlund, Dr. Anders Wingren, all of Sekab and Dr. Mats Lindberg, Luleå University for their supervision, support, and the many fruitful discussions that we had. I also thank deeply Carl-Axel Lalander, Sekab and the operators of the pilot plant for all the practical know-how that they taught me. A special thanks to Dr. Jörg Brücher, processum biorefinery initiative AB, for helping me with sampling techniques and equipments.

I would like to express my deepest gratitude and love to my family for their various support and encouragement through out my education. I would also extend my warmest gratitude to all the Ghanaian students and friends in Luleå. Patrick Amofah Msc.

Industrial Marketing & e-commerce, Lea Rastas Amofah, Doctorial student in the Division of Environmental Engineering, Emmanuel Essel, Doctorial student in the department of Mathematics, Frederick Ayisi Sarpong, in the Environmental Engineering and Godfred Etsey in the Chemical and Biochemical Engineering department. Your diverse support, encouragement and information are very much appreciated.

Finally, blessed be the LORD my strength and source of my knowledge. I wouldn’t have come this far without him.

Luleå, August 2007

Ebenezer Twumasi

Abstract

In order to design a future large-scale lignocellulose ethanol plant it is of crucial importance to monitor and minimize the discharge from the factory to the environment.

The Ethanol Pilot Plant in Örnsköldsvik, which is being operated near industrial conditions, is an excellent tool to study these effects. In the present study, the liquid and gas process streams have been analysed in order to close the mass balances of the studied chemical compounds. Of special interest was to close the mass balance on sulphur from H

SO

or SO

used for hydrolysis catalysts. Sulphur containing compounds have unpleasant smells and are often highly toxic to animals and human. High sulphur concentration in effluent wastewater leads to formation of high concentration of sulphide that upset the anaerobic biological organisms of wastewater. The existing gas treatment unit, wet scrubbers were also studied for their removal efficiency of TOC. The removal of TOC is very important because in air organic compounds along with oxides increase the level of ozone in the atmosphere. Wastewater samples obtained during continuous operation were investigated for COD, TOC, pH, total sulphur and TS as well as SS. The wastewaters were sampled individually from the membrane filter press, the scrubbers, the condensate from the evaporation unit and the stillage.

Mass balance estimation of sulphur indicates a loss of sulphur in the reactors, evaporators as well as the fermentation unit. The estimations further indicate that most sulphur emissions end up in the stillage.

In the gas stream after the H

O scrubber, the largest sulphur emission release (43 mg/m

) into atmosphere was registered when SO

was used as a catalyst in the hydrolysis. The use of H

SO

as a catalyst however release small amount of sulphur (4 mg/m

) into air, which is advantageous for environment. The TOC discharged into air on all occasions were in the range (9-17mg/m

). The removal efficiency of TOC is found to be a function of TOC loading, the gas flow rate and the liquid hold up. Results indicate that high liquid hold up as well as high inlet TOC concentration results in higher TOC absorption efficiency whereas a higher inlet gas flow rate decrease the absorption efficiency. The removal efficiency of the SO

scrubber shows that almost 85-97% removal efficiency could be achieved. H

O scrubber shows 2-97% removal efficiency.

Wastewater parameters investigated indicate high strength wastewater effluent as it contains high residual COD and TOC. The TOC and COD levels in the wastewater were however not dependent on the acid catalyst used in the hydrolysis process. The total effluent wastewater has a pH level that is acidic (pH 3-6) and sulphur concentration in the range (180-240 mg/l) that will be problematic for biological wastewater treatment and methane production in an anaerobic treatment plant. The stillage stream contributes to the largest COD, TOC, TS as well as sulphur emission in wastewater.

Keywords: environment, ethanol, lignocellulose, pilot plant, catalyst, sulphur, scrubber,

wastewater, stillage, COD, TOC, mass balance

CONTENTS Page No.

Preface...i

Abstract...ii

1. INTRODUCTION...1

1.1 Scope and Objectives ...2

2. LITERATURE STUDIES...3

2.1 Greenhouse effect and petroleum...3

2.1.1 Ethanol: An Alternative Fuel Source...4

2.2 Structural components of wood...5

2.2.1 Cellulose ...5

2.2.2 Hemicelluloses...6

2.2.3 Lignin...6

2.2.4 Extractives and ash ...7

2.3 Ethanol production ...7

2.3.1 Acid hydrolysis process ...7

2.3.2 Enzymatic Hydrolysis process...8

2.3.3 Fermentation ...9

2.4 Investigating parameters studied ...10

2.4.1 Organic compounds ...10

2.4.2 pH...11

2.4.3 Total Sulphur ...11

2.4.4 Total Solids (TS) and Suspended solids (SS) ...11

3. MATERIALS AND METHODS ...13

3.1 Process description of the pilot plant ...13

3.1.1 Hydrolysis Reactors ...13

3.1.2 Membrane Filter Press ...14

3.1.3 Evaporation unit...14

3.1.4 Detoxification and Fermentation ...14

3.1.5 Distillation Unit ...15

3.1.6 Wet scrubbers for gas treatment ...15

3.2 Sampling points...16

3.2.1 Hydrolysate...16

3.2.2 Wastewater sampling...18

3.2.3 Gas streams ...19

3.2.3.1 Flow and temperature measurement ...19

3.2.3.2 Sampling of Sulphur dioxide in gas stream ...19

3.2.3.3 Sampling of total organic carbon (TOC) in gas stream ...20

3.2.3.4 Sampling points of gas...21

3.2.4 Laboratory measurements...22

3.2.5 Material balance...22

3.2.6 Analytical methods ...23

3.2.6.1 Total sulphur (ICP and Schoeniger flask method)...23

3.2.6.2 TOC in water...23

3.2.6.2 COD ...23

4. RESULTS AND DISCUSSION ...24

4.1. Material balance of sulphur...24

4.1.1 Reactors...24

4.1.2 Membrane filter press. ...26

4.1.3 Evaporation Unit...27

4.1.4 Fermentation Unit ...27

4.1.5 Distillation Unit ...28

4.2. SO

scrubber...29

4.2.1 Mass balance of Sulphur and TOC ...29

4.2.2 Evaluation of SO

Scrubber Efficiency ...31

4.2.2.1 Effect of gas flow rate and TOC Concentration on removal efficiency ....31

4.2.2.2 Effect of liquid hold up on removal efficiency...32

4.3. H

O scrubber...33

4.3.1 Mass balance of Sulphur and TOC ...33

4.3.2 Evaluation of H

O Scrubber Efficiency...34

4.3.2.1 Effect of gas flow rate and inlet concentration ...34

4.3.2.2 Effect of liquid hold up on removal efficiency...34

4.4 Comparison of the efficiency of SO

and H

O scrubber ...36

4.5. Process gas discharge...37

4.5.1 TOC discharge ...37

4.5.2 Sulphur discharge...38

4.6. Wastewater characteristics ...39

5. CONCLUSIONS AND RECOMMENDATIONS...46

REFERENCES...48

APPENDIX A Scheme of the process ...50

APPENDIX B Tables...51

1. INTRODUCTION

With rapid depletion of the world reserves of petroleum, ethanol in recent years has emerged as one of the alternative liquid fuel and has generated immense activities of research in the production of ethanol and its environmental impact.

There is a low net contribution of CO

to the atmosphere when ethanol is produced from biomass under the assumption that the outtake of biomass does not exceed the natural production, and that the CO

produced from ethanol production is absorbed by growing biomass. The history of ethanol as a fuel dates back to the early days of the automobile where for example the first Ford automobile (1888) was fuelled on ethanol.

Ethanol as an automotive fuel can be used in two ways; as low blend in cars with no engine modification or as high blend (E85) if a flexi-fuel vehicle (FFV) is used.

Ethanol can be produced from grains such as wheat, barley or corn, or from sugar cane as well as sugar beet. However it is expected that there will be a limited supply of starch and sugar in the future, therefore lignocellulosic biomass is seen as an attractive feedstock for the future supply of ethanol.

Wood is used in the Sekab E- technology pilot plant.

Conversion of wood into ethanol goes through two processes, first hydrolysis of hemicellulose and cellulose in the lignocellulosic materials to sugars followed by fermentation of the sugars to ethanol by bacteria or yeast, upon which the ethanol can be recovered and concentrated using distillation. Lignin residue is a co-product in ethanol production, and can be used for power generation.

Besides reducing air pollution and carbon dioxide build-up in atmosphere, the production of ethanol from renewable resources will create jobs, decrease the dependence of countries that do not have their own oil reserves, reduce trade deficits and improve foreign exchange balance.

Sekab E-technology is one of such research companies operating lignocellulosic ethanol pilot plant. In the ethanol production process toxic chemicals such as SO

ethanol, methanol, acetic acid, lactic acid, formaldehyde, and acetaldehyde can potentially be formed and emit into air that could be harmful to the environment and human health.

Wastewater is the main contaminated stream from ethanol production.

This stream

contains organic matter, sulphur etc. which if not properly or completely treated before

discharge can cause water contamination.

1.1 Scope and Objectives

Maria Edlund reported studies on the waste streams of the Sekab E-technology pilot plant. In her work, organic compounds in air and water were studied. The COD and BOD were studied to examine the degradability of the effluent wastewater from different stages of the process. The sulphur in the wastewater was also studied. In the gas stream individual organic substances and their origination were studied.

In the present study, the liquid and gas process streams were analysed in order to close the mass balances of the studied chemical compounds. Of special interest was to close the mass balance of sulphur from H

SO

or SO

used for hydrolysis catalysts.

The existing gas treatment unit, wet scrubbers, were also considered for its removal efficiency of total organic compounds (TOC) in gaseous stream. The removal of TOC is very important because in air organic compounds along with oxides increase the level of ozone in the atmosphere.

The combined effluent stream going to the biological wastewater treatment was investigated.

The parameters considered were suspended solids (SS), total solids (TS), total organic carbon (TOC), chemical oxygen demand (COD), pH and total sulphur.

The long-term objective of this work is to use the data acquired from the studied parameters to aid in forecast the environmental impact of up scale ethanol production plant and how to purify waste streams.

Outline of the thesis

Chapter 1: The background and the subject of the thesis are introduced. The chapter also highlight on the problem area, the purpose and objectives of the thesis.

Chapter 2: litreature studies were analysed. Greenhouse effect and petroleum were considered. Advantages and disadvantages of ethanol as fuel were also highlight. Wood as raw material for ethanol production is discussed. Structural components and building blocks of wood are clearly defined. Major pre-treatment techniques and process steps in ethanol production are described.

The investigated parameters such as total organic compounds (TOC), total sulphur (Tot-S), chemical oxygen demand (COD), pH and total solids (TS) and their impact on the environment are also described.

Chapter 3: In this chapter the material and methods used in this work is highlighted. The process description of the major process units of the pilot plant is systematically discussed.

Finally, the different analytical techniques and sampling methods applied in this work.

Chapter 4: Here the results obtained in this work are summarise and discussed.

Chapter 5: Finally, the answers to the main questions and findings from the thesis are

presented.

2. LITERATURE STUDIES

2.1 Greenhouse effect and petroleum

The general state of the earth climate change recently has been under various debates regarding its causes. One school of thought believes that variation in earth temperature could be explained by natural activity, and that there is no evidence of human activity being involved. Luckily, the earth is not hit by known natural causes including the impacts of large volcanic eruptions and collisions with comets or meteorites very often, and therefore their associated climate changes occur rarely throughout earth history.

However, other causes of climate change influence the earth on much shorter time scales with changes sometimes occurring within a single generation. Indeed, accumulation of CO

in the atmosphere owed to large consumptions of petroleum and coal has come to be recognized as a major factor of climate change and global warming in our generation.

One factor that affects temperature increase on earth is the amount of greenhouse gases present in the atmosphere. These gases allow solar radiation to pass through the transparent atmosphere to the surface of the earth. The earth and the atmosphere reflect some of it. Low- energy radiation is absorbed by the earth and re-emitted in all directions by greenhouse gas molecules. The heat energy from the sun to the earth surface over a period should be about the same as the amount of energy radiated back into space leaving the temperature of the earth’s surface roughly constant. The effect of this is to warm the surface of the earth and lower atmosphere thereby preserving the earth as a habitable planet.

Without the greenhouse effect, earth average temperature would be -18

C, rather than the present (15

C).

However, human induced greenhouse gases have increased earth surface air temperature and caused sub-surface ocean temperature to rise.

Water vapour (H

O) and carbon dioxide (CO

) are the two largest contributors to the greenhouse effect. Methane (CH

), nitrous oxide (N

O), chlorofluorocarbons (CFCs), and other greenhouse gases are present in the atmosphere only in trace amounts.

CO

concentration in the atmosphere is naturally regulated and can be described by a carbon cycle model. Plants convert CO

and water into carbohydrates and oxygen using sunlight through photosynthesis where large portion of the carbohydrates are incorporated into polysaccharides of growing biomass. When new biomass is produced the same amount of CO

is assimilated as was released during combustion of the chopped tress. On the other hand, CO

released from fossil fuels have not been part of the carbon cycle over millions of years. Humanity is liberating carbon from beneath the earth surface and putting it into the atmosphere. Figure 1 clearly illustrates the different pathway of CO

, as a cycle when biomass is used as fuel and noncyclic when fossil fuel is used.

Over the past two hundred and fifty years, the concentration of CO

has risen from about 270ppm to about 370ppm. This rise has happened in time of industrial revolution largely due to the combustion of fossil fuels.

Sadly, this year the concentration has blown past 380ppm. So human activities, for the last 440,000 years at least, have forced the CO

concentration to oscillate between 180 and 280 ppm, and now it has been pushed up to

380ppm.

About 2.7 percent annual growth in developing countries is expected and will go

beyond the industrialised countries near 2018. This means that the CO

emissions will

increase due to developing countries dependency on fossil fuels. For instance, it is expected

that between 2001 and 2025, the world CO

emissions will be increased by 1.9 percent

annually in developing countries like China and India where there is economic development

with fossil energy dependency.

During earlier periods of the Earth’s history, fossilisation of biological material created the deposits of coal, oil and natural gas, which take millions of years to generate.

The fuel we are using today were made more than three million years ago before the time of dinosaurs but is being consumed in a fraction of that time. Humanity has already used a large amount of the oil reserves beneath the earth surface and it is expected that crude oil production will start declining from 2010.

A few countries in the word have massive deposits of petroleum, while for example United States does not have reserves that are sufficient to meet their domestic needs. This increases their dependence on other countries for a very vital need of humanity: energy. This is detrimental for economic security considering the fact that oil price keeps fluctuating and the uncertainty about the supply of crude oil mainly from the Middle East. World energy bodies including World Energy Council, the International Energy Agency and the US Energy Information Administration have projected an increase of some 60% in global primary energy consumption from 2002 to 2030 to reach a level of 16.5 billion tonnes of oil equivalents in 2030.

The current challenge is to respond effectively to the risks of global climate change whilst continuing to meet the high-energy demands of mature economies and the rapidly increasing energy demands of developing economies.

2.1.1 Ethanol: An Alternative Fuel Source

The high octane number of ethanol makes its blends achieve the same octane boosting or anti-knock effect as petroleum derived aromatics like benzene.

Aside high octane number, ethanol has a high evaporation heat and high flammability temperature that influences the engine performance positively and increases the compression ratio.

The blend E85 consisting of 15vol% unleaded gasoline and 85vol% ethanol has a prevalent usage (above - 21

C) as alternative fuel because of its advantage over pure ethanol which has a high risk of cold starting problem.

The low energy content of ethanol (21.2 GJ per m

) compared to gasoline (31.4 GJ per m

) means that ethanol-fuelled cars must be refilled more frequently than gasoline-fuelled ones.

The ethanol-fuelled car can run only 60% of the distance of gasoline fuelled ones.

2.2 Structural components of wood

Bio-ethanol can be produced either from grains such as wheat, barley or corn, or from sugar cane as well as sugar beet. However, it is expected that there will be a limited supply of starch and sugar in the future, therefore lignocellulosic biomass is seen as an attractive feedstock for the future supply of ethanol. It has been estimated that, the global annual generation of dry biomass from the photosynthetic process is about 2.2

10

tonnes, which is equivalent to about ten times the energy used globally per year.

In Sweden for example, the current forest harvest as well as forest residue correspond to 150 TWh per year, of which about 1,8 TWh amounts to the equivalent of ethanol and biogas produced annually.

This makes wood an attractive raw material in Sweden. Sekab E-technology operates a pilot plant using wood, e.g. spruce, as raw material and wood will therefore be considered in more detail in this section.

Wood is composed of a series of tubular fibres or cells cemented together. Each fibre wall is composed of various quantities of three polymers: cellulose, hemicelluloses, and lignin.

In addition wood contains low molecular organic and inorganic substance, extractives and ash, representing only a minor proportion. Hardwoods contain several cell types consisting of supporting tissue, including both libriform cells and fibre tracheids, and storage tissue of ray parenchyma cells. The wood substance in softwoods is composed of two different cells:

tracheids (90-95%) and ray cells (5-10%).

2.2.1 Cellulose

Wood in its dry state contains 40-55% cellulose

. Cellulose is a polydisperse polymer with a degree of polymerization (DP) that ranges from 3500 to 36,000 and comprises more than one-third of all vegetable matter. Cellulose is a homopolysaccharide composed of

- glucopyranose units, which are linked together by ß- (1→4)-glycosidic bonds (Figure. 2).

Two glucose molecules linked together by ß- (1→4)-glycosidic bond is called cellobiose, which is the smallest repeating unit in cellulose. Celluloses are largely linear polymers and have a strong tendency to form intra and intermolecular hydrogen bonds. They have a high tensile strength and are not soluble in water because of the presence of strong intermolecular hydrogen bonds and sometimes the presence of a small amount of crosslinking. Bundles of cellulose molecules are thus aggregated together in the form of microfibrils, in which highly ordered (crystalline) regions alternate with less ordered (amorphous) regions.

High molecular weight native cellulose, which is insoluble in 17.5% aqueous sodium hydroxide solution, is called α-cellulose. The fraction that is soluble in 17.5% sodium hydroxide solution but insoluble in 8% solution is called β- cellulose, and that which is soluble in 8%

sodium hydroxide solution is called γ- cellulose.

Figure 2: The structure of cellulose.

2.2.2 Hemicelluloses

The amount of hemicelluloses in dry wood is usually between 20 and 30%. The composition and structure of the hemicelluloses in the softwoods differ in characteristic from those in the hardwoods. Hemicelluloses are heteropolysaccharides with the degree of polymerization of only 200. The principal hemicelluloses in softwoods (about 20%) are Galactoglucomannans (see figure 3). The backbone is a linear or slightly branched chain of ß- (1→4)-linked D- mannopyranose and D-glucopyranose units.

In addition to galactoglucomannans, softwoods hemicellulosic structures also contain an arabinoglucuronoxylan (5-10%) and arabinogalactan thus introducing the 5-carbon monosacharides arabinose and xylose

. The major hemicellulose in hardwood is xylan. The backbone of xylan consists of ß- (1→4)- linked xylopyranose units. (Figure 4). Hardwood also contains glucomannan with a backbone of ß- (1→4)-linked D-mannopyranose and D-glucopyranose units. Unlike softwood xylan, hardwood xylan does not contain arabinose units.

Figure 3: The structure of major softwood hemicellulose, glucomannan.

Figure 4: The structure of major hardwood hemicellulose, xylan.

2.2.3 Lignin

In plant cells, lignin binds together the components of the cell wall and is responsible for mechanical strength and protects the cellulose structure from degradation and against microbial attack.

Lignins are polymers of p-hydroxyphenylpropane units. It makes up about one-quarter to one-third of the dry mass of wood. It has several unusual properties for being a biopolymer, such as having a network structure and lacking a defined primary structure. The building blocks are guaiacyl, syringyl and p-hydroxyphenylpropane . Hemicellulose components are bound to lignin mainly through arabinose, xylose, and galactose moieties. The Lignin content is different in softwoods and hardwoods. Normal hardwood contains 20-30% while softwood contains 26-32% of lignin. In softwoods lignin, the main precursors is trans-coniferyl alcohol while in hardwoods trans-sinapyl alcohol and trans-p-coumaryl alcohol are also lignin precursors.

2.2.4 Extractives and ash

Extractives in wood usually represent a minor proportion of about 4-10% or as much as 20%

of the dry weight of normal wood species depending on the climate of growth. Extractives consists of organic compounds including fats, waxes, alkaloids, proteins, phenolics, simple sugars, pectin, mucilage, gums, resins, terpenes, glycosides, saponins and essential oils. They contribute to wood properties such as odour, colour and decay resistance.

Ash is the inorganic residue remaining after total combustion of the wood matrix. The residues are approximately 25% water-soluble. In wood, the principal elemental components of ash are Ca and K with lesser amounts of Mg, Na, Mn and Fe. CO

, PO

, [SiO

]

, C

O

, and SO

are likely anions.

2.3 Ethanol production

Production of ethanol from starch is a well-known industrialised technique that has been employed since ancient days, purposely for producing beverages. However, the starch low sugar yield per hectare compared with prevalent forms of sugar in cellulose and hemicellulose from wood makes starchy materials less interesting for large-scale production

[2,3,4]

. Cellulosic ethanol production prevents the danger that food cropping (such as grain or sugarcane) will turn into more lucrative fuel-cropping. The supply of raw material is also more abundant than for corn-based ethanol production. Conversion of wood into ethanol goes through two processes, first hydrolysis of hemicellulose and cellulose in the lignocellulosic materials to sugars followed by fermentation of the sugars to ethanol by bacteria or yeast, upon which the ethanol can be recovered and concentrated using distillation. Lignin residue is a co-product in ethanol production, and can be used for power generation.

There are three major hydrolysis processes to produce sugars from lignocellulosic material:

dilute acid, concentrated acid and enzymatic hydrolysis.

2.3.1 Acid hydrolysis process

In a hydrolysis process, the hemi cellulose and cellulose polymer react with water and liberate sugar.

(C

H

O

)

+ (n) H

O → n C

H

O

(1)

Hydrolysis of polysaccharides is catalysed by acid. Figure 5 shows the reaction pathway of acid hydrolysis .The reaction between the cellulose and acid catalyst is a heterogeneous reaction in which a proton is transferred to the glycosidic oxygen linking two sugar units (I).

The conjugate acid (II) dissociates to form the cyclic carbonium ion (III) which reacts with

water to produce free sugar and a proton.

Figure 5: Mechanism of acid-catalyzed hydrolysis

When concentrated acids such as H

SO

and HCl are used, high yields of carbohydrate are recovered in orders of over 90% of both hemicellulose and cellulose sugars, but recovering the large quantities of acids used is still not economically valuable. In addition, concentrated acids are hazardous and corrosive and require reactors and pipes that are resistant to corrosion.

Dilute acid hydrolysis on the other hand operates with acid concentration in the range of 9-18mmol/l and achieves a sugar yield of 70-98% based on the hemicelluloses but only 50% based on the glucan, all at lab scale.

The process has a fast rate of reaction that facilitates continuous process. The hydrolysis takes place in two stages. In the first stage, the process operates at moderate temperatures ranged from 170-190

C and low sulphuric acid concentration (approximately 9 mmol/l), where hemicelluloses are readily hydrolyzed to 5 and 6-carbon sugars.

Cellulose hydrolysis takes place in the second stage at high temperatures ranged from 200

C to 230

C. In this stage, there are some unwanted reactions that occur at long residence time, namely the formation of furfural, formic acid, hydroxymethylfurfural (HMF) and levulinic acid. Aside lowering the yield of fermentable monosaccharides, these compounds may also be toxic to the microorganisms in the fermentation stage.

2.3.2 Enzymatic Hydrolysis process

In enzymatic hydrolysis, biological catalysts (enzymes) are employed to break down

cellulose into sugar. Acid pre-treatment of the lignocellulosic material is normally performed

to increase the susceptibility of cellulosic substrates to enzyme attack. Cellulase enzymes

under mild conditions (pH 4.8 and Temperature 45-50 oC) normally carry out enzymatic

hydrolysis. Cellulase can be produced from both bacteria and fungi, and is usually a mixture

of several enzymes that often contain carbohydrate-binding affinity to facilitate the

interaction between the enzyme and the substrate surface. During hydrolysis, the cellulases

catalyse three major phases. First through binding to the surface of crystalline cellulose,

endoglucanase (EG, endo-1, 4-

- glucanohydrolase) attacks low crystallinity regions in the

cellulose fibre creating free chain-ends. Exoglucanase or cellobiohydrolase (CBH, 1, 4- ß-

-

glucan cellobiohydrolase) further degrades the molecule by removing cellobiose unites from

the free chain-ends. In the final stage, ß-glucosidase hydrolyzes cellobiose to produce

glucose. In addition, a number of auxiliary enzymes such as glucuronidase, acetylesterase,

xylanase, ß-xylosidase, galactomannanase and glucomannanase attack hemicellulosic

structures. Combination of enzymatic conversion of cellulose and hemicellulose with

minimizing product (sugar) inhibition.

This process is referred to as Simultaneous Saccharification and Fermentation (SSF).

2.3.3 Fermentation

Fermentation can be defined as respiration in an anaerobic environment with no external electron acceptor. Production of ethanol from sugar derived from starch and biomass has been commercially dominated by the yeast Saccharomyces cerevisiae. The yeast uses the sugar as substrate and adenosine tri-phosphate (ATP) as an intra cellular energy transporter.

Reaction 2 is the fermentation reaction of glucose (C

H

O

) carried out by yeast to produce ethanol (C

H

OH) along with production of carbon dioxide.

C

H

O

→ 2C

H

OH + 2CO

+ 2 ATP (Energy Released: 118 kJ/mol) (2)

The reaction takes the glycolysis pathway in which glucose breaks down into two molecules of pyruvate. (See figure 6) The ionized carboxyl group (COO

) is removed from the pyruvate to generate a molecule of carbon dioxide, which is released by the yeast into its surroundings.

The resulting molecule, acetaldehyde accepts hydrogen from NADH. This hydrogen and the H

ion released in the earlier stage of glycolysis are added to the acetaldehyde producing ethanol.

Figure 6: Fermentation of ethanol

Ideally, conversion of both glucose and xylose to ethanol is 0.51 gram ethanol per gram sugar but because the yeast needs some energy (sugar) for cell metabolism and growth, conversions higher than 0.45g EtOH/g sugar have not been achieved.

2.4 Investigating parameters studied

Investigation of substances that might be expected in the various effluent streams and their impact on the environment is of special interest when considering the ethanol pilot plant process. This will help to minimize the pollutants in the effluent streams by using efficient treatment system or employed process technique that is environmentally friendly. Typical wastewater environmental parameters that were considered include wastewater flow, Total Organic Carbon (TOC), Total Sulphur (Tot-S), Chemical Oxygen Demand (COD), total solids (TS) and pH. Air emissions were also monitored for TOC and Tot-S. Some of these parameters and their reasons for monitoring are highlighted below.

2.4.1 Organic Compounds

Organic compounds in air and water are of environmental importance for several reasons.

In air, organic compounds play a significant role in the chemical reactions that form ozone.

Ozone is formed when the emissions of organic compounds and nitrogen oxides react in the presence of sunlight.

Health problems such as damages to lung tissue, reduce lung function, and adversely sensitizer the lungs to other irritants, are all related to formation of ozone in the troposphere.

Whilst beneficial in the stratosphere, ozone formation in the troposphere depletes the good ozone in the stratosphere. The depletion of ozone can cause increased amount of UV radiation to reach the Earth that can lead to more cases of skin cancer, cataracts, and impaired immune system.

There are several possible environmental consequences associated with organic matter in water. One feature of aqueous organic matter is that oxygen and other oxidizing agents in water can oxidize the organic material. Therefore, when organic material is released into a water body, the bulk organic matter degrades, consuming oxygen and leaving the system in an oxygen-deprived state (Anoxic conditions). This leads to a low pE environment and changes the entire chemistry of the aquatic system. The anoxic condition makes respiration hard to maintain, promotes dissolution of metals like mercury in water and creates stress on many aquatic organisms including fish.

Organic matter may also undergo other reactions leading to formation of toxic products in the aquatic environment. For instance, inorganic tin undergoes alkylation which is biological processed in aquatic environment to form monomethyl tin (CH

Sn

) and dimethyl tin ((CH

)

Sn

). The organotin products species are more toxic to aquatic biota than are the original inorganic tin compounds. Toxicity becomes grater as the number of organic groups increases in the series R

Sn

for n=1 to 3. The toxicity is also inversely related to the length of R and is at a maximum where R is a methyl or ethyl group.

Organic carbon in wastewater is composed of a variety of organic compounds in various

oxidation states. The COD method can be used to determine the amount of oxygen that is

needed to oxidise the organic matter. The TOC analysis on the other hand gives direct

expression of total organic content than COD, but it does not provide the same kind of

information as of COD. Because TOC is independent of the oxidation state of the organic

matter and does not measure inorganic or organically bound matter that can contribute to

COD, TOC measurement cannot take the place of COD. However if a relation can be

established between TOC and COD, then TOC can be used as an estimate for COD under

the same conditions for a specific source of wastewater.

2.4.2 pH

The hydrogen–ion concentration of wastewater is an important parameter, because it affects both treatment and the environment. The concentration range appropriate for most biological life is quite narrow and critical. Wastewater with an adverse concentration of hydrogen ion is difficult to treat by biological means, and the concentration in the natural waters will change if the pH of the effluent wastewater is not altered before it is discharged.

2.4.3 Total Sulphur

During several processes, sulphur containing compounds that are damaging to animals as well as humans are released to the environment. Sulphur in its elemental form is not toxic, but many sulphur derivatives are, such as sulphur dioxide (SO

) and hydrogen sulphide (H

S). These compounds have unpleasant smells and are highly toxic. In addition, acidic rain occurs when SO

and nitrogen oxides undergo chemical transformation in the atmosphere, and are absorbed by water droplets in clouds. The droplets then fall to earth as rain, snow, mist, dry dust, hail or sleet and increase the acidity of the soil, and affect the chemical balance of lakes and streams.

The sulphate ion occurs naturally in most waters and is present in wastewater as well.

Sulphate is reduced biologically under anaerobic conditions to sulphide that combined with hydrogen to form hydrogen sulphide. The H

S can be oxidized biologically to sulphuric acid that is corrosive to sewer pipes. In addition, H

S is a poisonous gas that smells like rotten egg.

During anaerobic wastewater treatment, organic matter decomposes into methane (CH

) and CO

. If the organic matter contains sulphur, unwanted sulphide is produced and that upsets the biological process.

At Domsjö, the anaerobic treatment step can tolerate approximately 100mg/L of sulphur and still have acceptable methane production.

To stabilize an organic waste efficiently in the anaerobic treatment, the nonmethanogenic and methanogenic bacteria must be in a state of dynamic equilibrium. To maintain such a condition, the system should be free of dissolved oxygen and ion concentration of inhibitory compounds such as heavy metals and sulphides. For effective anaerobic treatment, the sulphur in the wastewater should be eliminated.

2.4.4 Total Solids (TS) and Suspended Solids (SS)

Total solids include both total suspended solids and total dissolved solids (the portion that passes through a filter) left in a container after evaporation and drying of a water sample.

Suspended solids (SS) are the portion of total solids (TS) retained by a filter, i.e. they are particles that are able to settle in stagnant water. SS consists of small fibre fragments, fines, non-settled biological sludge (agglomeration of microorganisms) and other small particles down to about 1μm in size.

High concentrations of SS of wastewater can cause many problems for stream health and aquatic life if not treated before discharge to water bodies.

High TS can block light from reaching underwater vegetation. Photosynthesis slows down as the amount of light passing through the water is reduced. Reduced rates of photosynthesis cause less amounts of oxygen to be released into the water by plants. The plants will stop producing oxygen and die as the light is completely blocked from bottom dwelling plants.

Low dissolved oxygen can lead to fish kills. In addition, high TS can cause an increase in

surface water temperature, because the suspended particles absorb heat from sunlight and

will cause the dissolved oxygen levels to fall even further since warmer waters can hold less

dissolved oxygen. Suspended sediments can clog fish gills, reduce growth rates and decrease

resistance to disease, and prevent egg and larval development. High TS can cause problems

for industrial use, because the solids may clog or scour pipes and machinery.

The level of

TS in this work will give an idea about the physical unit operation (sedimentation or flotation

unit etc.), to be used in future wastewater treatment plant to produce a clarified effluent and

concentrated sludge that can easily be handled and treated.

3. MATERIALS AND METHODS

3.1 Process description of the pilot plant

The ethanol pilot plant in Örnsköldsvik has a capacity to produce around 300-400 litres of ethanol per day, using approximately 2 tonnes (dry weight) of wood chips (spruce). The plant operation is based on the technology of two stage dilute acid hydrolysis that could be followed by enzymatic hydrolysis. In dilute acid hydrolysis, sulphuric acid or sulphur dioxide is used as a catalyst. In this work, samples were analysed when dilute acid hydrolysis was in use and will be discussed in details in this section followed by major unit operations in the plant. A flow sheet of the process is presented in appendix A. The main product streams as well as wastewater and gas streams are illustrated in different line style.

The wood chips are stored in containers and blown up by a fan via a cyclone to the rooftop of the plant. Chips from the cyclone are then fed to the first step of the process, the pre- steaming.

In the pre-steaming process, steam is used to pre-heat the raw material and liberate trapped air in the chips. Likely organic substances and particles from the chips and excess steam are found in the gas streams. A pre-steaming scrubber is connected to this unit to remove the particles and the substances in the gas stream. Un-trapped chemicals in the gas stream are led to a H

O scrubber for further treatment.

3.1.1 Hydrolysis Reactors

Figure 7: The two-stage hydrolysis process. a : separation of unhydrolyzed cellulose and a liquid phase containing sugar. b : solid phase consisting of lignin and unhydrolyzed cellulose. c : solid phase consisting of lignin, unhydrolyzed cellulose and a liquid phase containing sugar and by-products.

Hydrolysis is performed in two stages to maximize sugar yield from the hemicellulose and

cellulose fractions. In the first stage, a horizontal plug-flow reactor (R1) is fed with the

steamed wood chips and in the reactor the chips are impregnated with a dilute sulphuric acid

solution. When SO

is used, the chips are impregnated before fed to the reactor. The

temperature in the reactor is in the range of 170-190˚C resulting in the hydrolysis of most of

the hemicellulose. After the first reactor, the solubilized sugar in the liquid phase is recovered and transported to a storage tank while solid phase consisting of lignin and unhydrolyzed cellulose is forwarded to the second reactor (R2). Before entering the second reactor the lignocellulose is reimpregnated with acid. In the second stage, high temperature is applied, in the range of 190-210˚C that is sufficient to hydrolyze the resistant cellulose fraction. The whole hydrolysis process is characterized by being conducted in continuous mode. The liquid hydrolysates recovered from each stage are transported to one storage tank before channelled to the membrane filter press. Gases generated from the two reactors are led to the SO

scrubber.

3.1.2 Membrane Filter Press

The slurry from the two reactors contains a solid phase consisting of lignin, unhydrolyzed cellulose and a liquid phase containing sugar and by-products which are forwarded to the membrane filter press. In the membrane filter press, a pressure driven flow of slurry through a membrane filter is used to separate lignin residue from the hydrolysate. After recovering the hydrolysate, water (i.e. wash water) is pumped through the lignin residue retained in the membrane to washout the residual hydrolysate in the lignin. The wash water displaces the residual hydrolysate, which joins the hydrolysate stream to the storage tank. As the washed water exit the press, the valve to the environmental tank (Waste water tank) automatically opens. The rest of the wash water containing only small amounts of sugar is squeezed out from the lignin and channelled to the Wastewater tank. The wet lignins (filter cake) retained

in the filter press are later burnt in a boiler at Domsjö fabriker.

3.1.3 Evaporation unit

This step intends to evaporate water and to concentrate the sugar solution to about 15%. In this step, the hydrolysate is fed to a backward-fed double-effect evaporator. During the process, steam is supplied to the second effect and the fresh hydrolysate enters the first effect. The hydrolysate of 5% concentration enters the first effect that operates at a low temperature and pressure. The 7% concentrate product leaves the first effect and continues to high temperature and pressure second effect. From the second effect 15% concentrate of hydrolysate is recovered and channelled to the detoxification unit. The condensates (water) from both effects are channelled to the wastewater tank.

3.1.4 Fermentation

There are four fermentors employed in the plant and they can be operated separately or

combined in batches or in a continues process. Ammonium nitrate and phosphoric acid are

added as nutrients to the fermentative microorganism and to keep pH up at 5.0-5.5 around

30˚C. After the fermentation, yeast is separated from the fermentated hydrolysate by a

centrifugal pump and recycled to the fermentation tank. The CIP (cleaning in place) system

is linked to the recycle loop to clean pipes and tanks, kill bacteria and yeast culture. The

system uses sodium hydroxide, sulfo-amino acid and steam as cleaning agents.

3.1.5 Distillation Unit

The fermentation products (mash) are fed to a distillation column in which the ethanol is

concentrated to 90-95wt%. The process involves boiling the liquid mixture around 100-103˚C in which volatile ethanol exit as vapour while water remains at the bottom of the

column. The vapour is condensed and recovered as liquid ethanol at the top of the column.

The remaining stillage at the bottom of the column is transported to the environmental tank.

3.1.6 Wet scrubbers for gas treatment

The plant uses three wet scrubbers to remove air pollutants before discharging the process gases to the atmosphere. The pre-steaming scrubber, SO

scrubber and H

O scrubber absorb gas pollutants from different process sources. This is illustrated in figure 8. The pre-steaming scrubber treats the gas stream from the pre-steaming process that contains mostly organic compounds in lower concentrations. The gas stream going into the SO

scrubber originates from the two hydrolysis reactors. The SO

scrubber absorbs high concentration organic compounds and sulphur in the gas from the reactors. The outlet gas from the SO

scrubber and the pre-steaming scrubber contains a lower concentration of organic compounds and sulphur, and these gas streams enter the H

O scrubber. In addition, gases originating from the CIP, detoxification tank, evaporation and fermentation are also led to the H

O scrubber.

The gas stream flows up the chamber (counter current to the liquid). The scrubbing liquid is sprayed from the top of the scrubber and comes in contact with the gas stream to be cleaned on its way down. The gases pass through a filler layer, which increases the contact area between the gas particles and the water droplets. The washed gas stream leaves at the top of the scrubber with low concentration of pollutants while the droplet water leaves the scrubber at the bottom with high concentration of pollutants to the wastewater tank. The process is continues in which fresh water is constantly pump to the top of the scrubber. NaOH is used as absorbing solution in SO

scrubber and to maintain pH at around 6. There is no chemical absorbent and pH regulation in the H

O scrubber. The pre-steaming scrubber has no pH regulation, neither filler layers nor circulation of water.

Figure 8: The gas treatment system.

3.2 Sampling points

Figure 9 shows sampling points for hydrolysate. Wastewater and gas stream sampling points have been indicated on the process flow sheet at appendix 1.

3.2.1 Hydrolysate

Hydrolysates from different process steps were sampled at different points. The samples represent only a snapshot of the hydrolysate during the operation period considered. Sample from R1 was taken from a pump transporting hydrolysate to storage tank F7103. Samples from this point have a TS% of 8-15 depending on the process condition. Attached to F7104 is a pipe with valve, where thick slurry hydrolysate from the 2nd reactor is sampled by opening the valve. The TS% of the slurry lies in the range of 27-45. Combined hydrolysate from R1 and R2 goes to F7103. A sample is taken by opening a valve on the pipe connected to F7103.

This is sampled before the filtration step. During the filtration step, the hydrolysate from the filter press goes to the storage tank F7114. Sampling is carried out by opening a valve on the pipe connecting the filter press and F7114. The lignin residue from the filter plate in the press is sampled. For filter plates selected, the lignin residue at the top and bottom are taken as separate samples. From the evaporation unit, sampling is carried out by opening a valve on the pipe linked to the pump connected to the second effect evaporator. Sample from the fermentation tank is taken by opening a valve on a pipe connected to the pump that is linked to the fermentation tank. After yeast separation the storage tank F7301 is sampled and its represent sample (Mash) before Distillation. The stillage from the distillation column is sampled by opening valve on the recirculation loop of the distillation column.

stillage Sampling points

Liquid phase containing sugar

Solid phase consisting of lignin and unhydrolyzed cellulose

Dist

F7301 Fe

Evap m F7114 F7104

F7103 R1

Lignin cake

Figure 9: Flowsheet of storage tanks and different sampling points.

Flow calculations

The volumetric flow rate of the hydrolysate out from the reactor system was determined by using the Delta V software. The software monitors and controls the process parameters. The software also registers the process parameters and it is therefore possible to know the history of the process. The flow is calculated by first generating the percentage level or volume of hydrolysate in the tank of interest at 10s or 10min time interval. A plot of total volume against the time interval gives a graph from which the slope is calculated as a flow rate (m

/h). In calculating the flow rate of the two reactors, a matlab program was used to plot the graphs and generate flow rate (m

/h). The flow rate of different sampling points and measurement technique applied is present in table 1.

Table 1: Sampling points and flow measurement techniques applied.

Sampling points

Direct display on Delta V

Calculated from generated values

from Delta V

Calculated from generated values from Delta V using matlab

Measured by hand

Raw material X

Reactor columns

X

F7103 X

F7114 X

F7223 X

F7301 X

F7302 X

Ethanol stream X

Evaporator X Fermentor X Condensate

water X

Scrubbers wastewater

X Filterwash

water

X

3.2.2 Wastewater sampling

Pre-steaming scrubber water

Sampling is carried out by closing a valve on the pipe that connects the scrubber and the wastewater tank. On top of this valve is another valve, sample is then taken by opening the top valve. The Volumetric flow-rate of the water is calculated by determining the time to collect 12L volume of water that flows out. For each flow rate determination, three measurements are performed and an average is used.

H

O and SO

scrubber water

Wastewater from H

O scrubber is sampled by switching off the fresh water going into the scrubber for at least one hour and vice versa if sample is taken from SO

scrubber. Another valve on the pipe that connect wastewater tank is open to get wastewater sample. This is possible since the pipe is horizontal and the opened valve is vertical. In measuring the flow- rate, the same technique as used in the pre-steaming scrubber flow measurement is applied.

Wash water from filter Press

This sample is taken when the valve of the pipe connecting the hydrolysate stream from the press to F7114 automatically switch to the wastewater tank. Figure 10 show the switch pattern of the filter press. The volume of the wash water to the wastewater tank is calculated by considering the time it takes for all the wash water and squeeze water to go to the wastewater tank when the valve switched. The flow rate (l/h) of the washed water and squeeze can be determined directly from Delta V. With the known washed water used, the filter cake mass, and average dry mass of lignin (50%) for each batch of press, the volume of waste water can be determined by multiplying the flow rate and the time it takes for wash water and squeeze water to exit the filter press plus amount of water in the filter cake (i.e.

50% × filter cake mass). For 24 hour operation the volumetric flow of wastewater from the filter press is calculated by multiplying the volume of wastewater effluent by the number of press procedures per day (i.e. 3 for 24hour). The resulted value is then divided by 24 hours.

Figure 10: The flow rate pattern of wastewater from the filter press.

Collective condensate (water) sample from the evaporation units

The combined condensate from the evaporation system is sampled by opening a valve connecting to the wastewater tank. The flow rate was measured using the same procedure as of scrubber water.

3.2.3 Gas streams

The procedure used in measuring the flow and temperature of the gas stream is first

explained, followed by how sampling is carried out for each substance considered (i.e. sulphur and total organic carbon (TOC)) and finally the description of each sampling

point. Sampling was carried out as manual single measurements for short period.

3.2.3.1 Flow and temperature measurement

The flow velocity as well as the temperature of the gas in the pipe is measured with a measuring sensor. The sensor consists of a measuring probe with a wheel, connected to hand held instrument that displays velocity (m/s) and temperature (˚C). During measurement, the probe is inserted at 90 degrees angle to the pipe in the direction of the gas flow. The presence of steam in the gas stream affects the speed of the wheel with water precipitate. To overcome this, only the initially displayed value was considered. The velocity is independent of the density, pressure, temperature and chemical substances in the gas. The volumetric flow rate (m

/s) of the gas is calculated by multiplying the velocity (m/s) obtained by the area (m

) of the pipe assuming equal velocity over the whole cross section.

3.2.3.2 Sampling of Sulphur dioxide in gas stream

The sulphur in the gas stream is sampled using 10g Na

CO

/l solution in two wash flasks arranged in series. The idea is to capture the sulphur in the first or second flask in which the Na

CO

and the sulphur undergo the follow reaction.

2CO

(aq)

+ H

SO

⇔ 2HCO

+ SO

The resulting sulphite in the solution is then analysed for sulphur using ICP or Schoeniger flasks method at MoRe-Research.

During sampling, exhaust gas stream from the pipe passed through two washed bottles, placed in series containing 250ml Na

CO

solution in each. A pump is connected to the end of the second bottle to withdraw gas from the pipe through the bottles. See arrangement in figure 11. The volume of gas that flows through the bottles is registered by the pump meter. The volume of gas allowed depends on the concentration of sulphur at the sampling point. After sampling, the connected tubes are rinsed with deionised water into the flasks.

Figure 11: Arrangement of wash flasks for sulphur sampling

The concentration (C

mg/l) obtained from the analysis is sulphur in the solvent. This was recalculated to concentration (C

mg/m

) of sulphur in the gas phase. The volume of Na

CO

solution (V

in litres) used is multiplied by the concentration (C

mg/l) to give the mass (m mg) of sulphur in the solution. The concentration (C

mg/m

) of sulphur is then calculated by dividing mass m by the volume of gas pump through the bottles (V

).

3.2.3.3 Sampling of total organic carbon (TOC) in gas stream

Total organic carbons in the gas streams were measured by Sick total hydrocarbon analyser model 3006. See figure 12 for the equipment. The analyser operates according to the principle of comparison. The concentration of sampled gas is compared with the known concentration of the calibration gas. Before sampling, the instrument is calibrated by first allowing a test gas (Nitrogen) with a concentration value of zero (zero gas) to pass through the instrument and the zero point is set to zero. In a second stage, a calibration gas 1% vol.

propane in air is fed to the instrument, and the sensitivity of the instrument is then set to this value. After the calibration, the tube of the instrument is inserted into the pipe in the direction opposite to the flow of gas stream. The physically measured quantity is converted into an electric signal by means of a flame ionisation detector (FID). During the operation of the analyser, hydrogen is constantly consumed as combustion gas. Carbon monoxide and carbon dioxide are not registered. The measured value is registered as ppm hydrocarbons relative to propane (C

H

). By multiplication with factor 1.608 this value can be converted to mg

“organic” C per standard cubic meters (mgC/m

). The results of the analysis give the concentration of gas at standard temperature and pressure. To determine the concentration of TOC at the conditions in the pipe, the ideal gas law equation was used.

Ideal gas law: PV=nRT where P = standard pressure at 1atm V = volume of gas

T = temperature in Kelvin R = the gas constant

n = number of moles of gas

The concentration (C

mg/m

) of TOC at the conditions in pipe was calculated by C

=

2 1

V C .

Where V

is volume of the gas in the pipe and was calculated by the formula V

=

2 2

P T nR ×

where T

and P

(assumed as 1atm) are the conditions in the pipe where the sample was taken. nR is calculated using the conditions where the analysis was done. i.e.

nR=

1 1 1

T V

P where T

is 273.15 K and P

as well as V

are assumed as 1atm and 1 m

respectively.

Figure 12: Sick total hydrocarbon analyzer.

3.2.3.4 Sampling points of gas

H

O scrubber outlet gas

This is the gas leaving the plant after the H

O scrubber. It was sampled first because it has the lowest concentration of substances.

Pre-steaming scrubber outlet gas

The design of the scrubber makes it possible to sample only outlet gas from the scrubber. The outlet gas measured here help to give an idea about the total organic compounds and sulphur generated from the pre-steaming.

Inlet gas of H

O scrubber

Gas streams into the scrubber system originate from many process stages. See figure 8. In this work, gas samples taken at this point represent gas stream mainly from the pre-steaming scrubber and the SO

scrubber.

Inlet gas of SO

scrubber

The gas sampled here originates from the two hydrolysis process reactors. The gas is warm

and contains steam that makes flow measurement difficult.

3.2.4 Laboratory measurements

The pH in this work was measured with pH electrode model WTW InoLab pH 720. Two different suspended solids (SS) were calculated. That is SS (percentage weight of solid) for hydrolysate and SS (mg/l) for wastewater. Total solid (TS) of hydrolysate was determined by using Sartorius model TE 6100. For TS of wastewater, a well-mixed sample is evaporated in a weighed dish and dried to constant weight in an oven at 105°C. The increase in weight over that of the empty dish represents the total solids. In determination of SS (percentage weight of solid), measured volume of sample (V ml) is filtered through a filter with know mass (A mg) by vacuum suction. Less volume was used when the filter gets clogged too quickly and more if the sample filters through very fast. The filter is rinsed with 20ml distilled water and then dried for two hours at 105˚C. The filter after drying is allowed to reach room temperature and weighed (B mg). The SS (percentage weight of solid) is then calculated as SS% =

⎠

⎜ ⎞

⎝

⎛ −

V A

B . In measuring the SS (mg/l) of wastewater, three filters from the pack to be used were chosen, one from top, middle and bottom. The three filters are then weighed together as mass (a mg). 200ml distilled water is passed through the three filters which are then dried in oven for two hours at 105˚C. The filters after drying are allowed to reach room temperature and weighed as (b mg). The difference between the two masses divided by 3 gives a calibration mass of the filters in the pack. i.e. mg = ^{( )}

3 a

b − = c . The SS (mg/l) is then calculated as X= ( − + ) _{× 1000}

V c A B

3.2.5 Material balance

In the present study, the amounts of sulphur in some streams (i.e. gas, wastewater and hydrolysate stream) were measured and material balances were used to determine the flow in the other. Material balance based on the application of the law of conservation of mass is applied. Essentially, if there is no accumulation within the system, then all the materials that go into the system must come out either in the product, by-products, waste streams or as releases. In calculating, sulphur concentrations were multiplied by the flow rate to obtain sulphur value per unit time. i.e. R s = C s * F

Where Rs

rate of sulphur in mg/hour

C

concentration of sulphur in the sample mg/litre F

flow of sample in litre/hour

In considering mass balance around the press, volume change (∆V in litres) was used instead of the flow F. This gives mass of sulphur (M kg). The mass balance (M e

) should satisfy the equation ∑ R

=∑ R

Where

R

=

input rate of sulphur in g/hour R

= output rate of sulphur in g/hour

Gas-liquid absorption idea was used in calculating the material balance around the scrubbers.

Mass balance on individual scrubber was considered as well as the mass balance over the two scrubbers as one unit. The gas removal efficiency of the scrubber η is given by

η = 1- Q

C

Where Q is volumetric gas flow-rate (m

/s) in the pipe and C

is

3.2.6 Analytical methods

Different analytical methods were used to analyse samples for sulphur, TOC and COD concentrations. These analyses were performed by accredited laboratories in Örnsköldsvik, Sweden. TOC and COD analysis were performed by Akzo laboratory while total sulphur analysis was performed by MoRe-Research.

3.2.6.1 Total sulphur (ICP and Schoeniger flask method)

Two different analytical methods were used in analysis of sulphur depending on the volality of sulphur in the solution. The two techniques are inductively coupled plasma (ICP) and Schoeniger flask method. Jobin Yvon ICP model 2000 was used to perform the analysis. It is based on inductive coupled plasma as a method of producing ions with a mass spectrometer as a method of separating and detecting the ions. A sample mixture is introduced into the analysis by the use of a nebulizer. As droplets of nebulized sample enters the plasma chamber, it evaporates and any solid that were dissolved in the liquid vaporize and then break down into atoms. The plasma chamber consists of radio-frequency electric current induction coil that produced a spark with argon gas. At the high temperature prevailing in the chamber, atoms of many chemical elements are ionized, each forming a singly charged ion. The atoms from the plasma are extracted via a series of cones into a mass spectrometer. The ions are separated on the basis of their mass-to-charge ratio and a detector receives an ion signal proportional to the concentration.

Schoeniger flask method consists of Erlenmeyer flask with a platinum basket. A few ml of dilute NaOH solution is placed in the empty combustion flask, which is then purged with a moderate stream of oxygen to displace air. A few milligrams of sample are weighed onto a small piece of ashless filter, which is folded around the sample and place in the Platinum basket. While the flask is being swept with oxygen, the stopper assembly containing the charged platinum basket is lit. The stopper is quickly and smoothly insert into the flask. After the combustion is complete, the flask is rotated to wet the glass wall with absorbing solution and to allow the ash of the sample to drift down into the solution. The solution is then analysed by Ion chromatography.

3.2.6.2 TOC in water

The TOC in this work was analyzed with O.I analytical total organic analyzer model 1020A.

The analyzer operates on the technique of combustion to oxidize and detect the organic carbon. A liquid sample is introduced to the analyzer via loop sampling. The sample is initially introduced to an inorganic carbon (IC) removal stage, where acid is added to the sample. At this point, the IC is converted into carbon dioxide that is purged from the sample and then detected by an infrared detector calibrated to directly display the mass of carbon dioxide measured. The inorganic carbon-free sample remaining is then oxidized and the carbon dioxide generated from the oxidation process is directly proportional to the TOC in the sample .

3.2.6.2 COD

Potassium dichromate solution with mercury sulphite and 90% sulphuric acid is used for COD determination in this work. The measurement is carried out by mixing certain volume of water sample with the known amount of dichromate. The mixture is heated in thermostat at148˚C for two hours. Under this condition, the organic substance found in the water sample undergoes oxidation while potassium dichromate is reduced to Cr

. The amount of Cr

determined by a photometer is used as an indirect measure of the organic contents of the

water sample and the results is expressed in mg O

/l.