Industrial Pilot Scale Leaching and Recovery of Zinc from Waste-to-Energy Fly Ash using Scrubber Liquids

This thesis comprises 60 ECTS credits and is a compulsory part in the Master of Science

with a Major in Resource Recovery

– Sustainable Engineering, 120 ECTS credits

No. 2016.13.06

Industrial Pilot Scale

Leaching and Recovery of Zinc from

Waste-to-Energy Fly Ash using Scrubber Liquids

Industrial Pilot Scale Leaching and Recovery of Zinc from Waste-to-Energy Fly Ash

using Scrubber Liquids

Manuela Wagner, s144424@student.hb.se

Master thesis

Subject Category:

Technology

University of Borås

School of Engineering

SE-501 90 BORÅS

Telephone +46 033 435 4640

Examiner:

Dr. Anita Pettersson

Supervisor, name:

Dr. Karin Karlfeldt Fedje Dr. Sven Andersson

Supervisor, address: Renova AB Götaverken Miljö AB

Box 156 Anders Carlssons gata 14

401 22 Göteborg 417 55 Göteborg

Date:

2016-06-03

Keywords:

pilot plant, zinc recovery, fly ash, WtE, scrubber liquids, leaching,

precipitation, chlorides

Abstract

Previous studies from laboratory experiments and a similar process at a plant in Switzerland,

led to the pilot plant project at Renova AB, which will be described in this master thesis. In

cooperation with Götaverken Miljö AB it was investigated if fly ash, produced at the Renova

Waste-to-Energy plant in Gothenburg, could be treated with own scrubber liquids in order to

recover zinc. If successful, Renova might build this tested pilot process in to a big scale. The

pilot plant has a scale of 16 times smaller than a future big scale process.

The goal of the project is to leach zinc from fly ash and gain a fly ash residue, which is

classified as non-hazardous waste. The filtrate from the leaching campaign is treated so that

the containing zinc is recovered. The zinc cake end product shall has a quality so that it can be

sold to other industries or upgraded to high purity zinc metal.

The evaluation of the experiments showed that the pilot plant process was successful. It was

possible to leach out zinc by a maximum quote of 74%. The total recovery of zinc could be

achieved by a maximum of 72%. The final zinc cake product was achieved through a

precipitation and filtration campaign.

This thesis evaluates, which process set-ups for zinc recovery through leaching and

precipitation & filtration are the best and can be recommended for a big scale process. In

addition, it briefly analyses the zinc product quality.

Future studies will be necessary within: cost analysis of the process, zinc product quality and

an analysis of the ash residue.

Table of contents

Abstract ... iii

Table of Figures ... vi

Table of Tables ... viii

Table of Abbreviations ... ix

1.

Introduction ... 1

2.

Background ... 2

3.

Description of the Waste-to-Energy Process at Renova ... 4

4.

Material ... 6

4.1

Fly Ash ... 6

4.2

Scrubber liquids ... 7

4.2.1

Hydrochloric acid ... 7

4.2.2

Sodium sulphate solution ... 8

4.2.3

Condensate water ... 8

4.3

Other liquids ... 8

4.3.1

Sulphuric acid ... 8

4.3.2

Sodium Hydroxide ... 9

5.

Pilot Plant Principle ... 9

6.

Laboratory experiments ... 11

6.1

Laboratory Set-up... 11

6.2

Zn-Precipitation Method in Laboratory ... 13

7.

Pilot Plant: Leaching Campaign ... 14

7.1

Leaching Set-up ... 14

7.2

Leaching Method ... 17

8.

Pilot Plant: Precipitation and Filtration Campaign... 20

8.1

Precipitation and Filtration Set-up ... 21

8.2

Precipitation and Filtration Method ... 22

9.

Results and Discussion ... 27

9.1

Pilot Plant Leaching ... 27

9.2

Laboratory Precipitation and Filtration ... 30

9.3

Pilot Plant: Precipitation and Filtration ... 32

9.5

Zinc cake end product from the pilot plant ... 36

10.

Conclusion ... 39

11.

Outlook ... 40

References ... ix

Appendix 1: Calculations... xii

Appendix 2: Eurofins Result Tables

Table of Figures

Figure 1.1:

Municipal Solid Waste Treatment Methods World-Wide & Sweden

1

Figure 3.1:

Simplified WtE and Flue Gas cleaning process

5

Figure 4.1:

Fly ash from 2 different samples

6

Figure 4.2:

Average composition of the used fly ash in percent by weight

7

Figure 4.3:

HCl appearance and average composition in percent by weight

7

Figure 4.4:

Sodium Sulphate Solution average composition in percent by weight

8

Figure 5.1:

a) Leaching Principle b) Pilot Plant Pictures of Leaching Process

10

Figure 5.2:

a) Principle of Precipitation b) Pilot Plant Pictures of Precipitation Process

11

Figure 6.1:

Scheme of Laboratory Set-Up

12

Figure 6.2:

Laboratory Equipment

12

Figure 7.1:

Big bag on big bag cutter

15

Figure 7.2:

Blending tank during leaching

15

Figure 7.3:

Leaching Process Set-up

16

Figure 7.4:

Inside of the vacuum belt filer with ash residue and washing procedure

17

Figure 8.1:

Precipitation of Heavy Metal Hydroxides

20

Figure 8.2:

Precipitaion of Zinc and other metals

21

Figure 8.3

Set-up of Precipitation and Filtration Process

22

Figure 8.4:

Warming procedure of Zinc Liquid in Cipax tank

23

Figure 8.5:

Flotation procedure of Zinc Liquid in Cipax Tank

23

Figure 9.3:

Leaching result of Zinc from Fly Ash

27

Figure 9.4:

Zinc leaching efficiency over pH

28

Figure 9.5:

Leaching efficiency of Zinc and Magnesium over pH

28

Figure 9.6:

Zinc leaching efficiency over L/S

29

Figure 9.7:

Zinc leaching efficiency over Residence time

29

Figure 9.1:

Zinc precipitation and filtration rate in laboratory

31

Figure 9.2:

Magnesium precipitation and filtration rate in laboratory

31

Figure 9.8:

Precipitation and Filtration rate: Zinc

32

Figure 9.9:

Zinc Precipitation and Filtration rate over pH

33

Figure 9.13:

Zinc Precipitation and Filtration over zinc cake wash time

34

Figure 9.14:

Total zinc recovery from fly ash to zinc cake

35

Table of Tables

Table 6.1:

Overview over Laboratory variations

13

Table 6.2:

Overview over Laboratory sample preparation

14

Table 7.1:

Overview of Leaching Variations

18

Table 7.2:

Leaching Campaign with Hydrochloric Acid A, B, C, G, H, J

18

Table 7.3:

Leaching Campaign with Hydrochloric Acid and Sodium Sulphate Solution D, F

19

Table 7.4:

Leaching Campaign with Hydrochloric Acid and Condensate I

19

Table 7.5:

Leaching Campaign with Sulphuric Acid and Condensate K

19

Table 7.6:

Leaching Campaign with Condensate E

20

Table 8.1:

Overview of Precipitation and Filtration variations

26

Table 9.1:

Laboratory results: zinc and magnesium precipitation

31

Table 9.2:

Zinc win from 10 kg virgin fly ash

35

Table of Abbreviations

HCl

Hydrochloric acid

H

2

O

2

Hydrogen peroxide

L/S

Liquid to solid ratio

MSWI

Municipal Solid Waste Incineration

NaOH

Sodium hydroxide

Na

2

SO

4

Sodium sulphate solution

P&F

Precipitation and Filtration

wt%

Percent by weight

Municipal Solid Waste Treatment

Methods

1.

Introduction

Living in a “throwaway society”, the amount of solid household waste steadily increases.

Whereas Sweden and other European countries have strict legacies on how waste is handled,

most other countries world-wide still simply dump their waste in the nature or on uncontrolled

landfills (Hoornweg & Bhada-Tata 2012). As consequence, waste liquids drain into ground

water and methane gas emits in the atmosphere. Based on the data from Hoornweg &

Bhada-Tata (2012) What a Waste

– A global review of solid waste management, figure 1.1 shows

that Waste-to-Energy (WtE) is the most common waste treatment method in Sweden with

50%. Whereas, taking the average of all other countries, dumping and landfilling (19% &

52%) are far more common world-wide than WtE (4%).

*World-Wide means: Algeria, Antigua and Barbuda, Armenia, Australia, Austria, Belarus, Belgium, Belize, Bulgaria, Cambodia, Cameroon, Canada, Chile, Colombia, Costa Rica, Croatia, Cuba, Cyprus, Czech Republic, Denmark, Dominica, Greece, Grenada, Guatemala, Guyana, Haiti, Hong Kong, China, Hungary, Iceland, Ireland, Israel, Italy, Jamaica, Japan, Jordan, Korea (South), Kyrgyz Republic, Latvia, Lebanon, Lithuania, Luxembourg, Macao, China, Madagascar, Malta, Marshall Islands, Mauritius, Mexico, Monaco, Morocco, Netherlands, New Zealand, Nicaragua, Niger, Norway, Panama, Paraguay, Peru, Poland, Portugal, Romania, Singapore, Slovak Republic, Slovenia, Spain, St. Kitts and Nevis, St. Lucia, St. , St. Vincent and Grenadines, Suriname, Sweden, Switzerland, Syrian Arab Republic, Thailand, Trinidad and Tobago, Tunisia, Turkey, Uganda, United Kingdom, United States, Uruguay, Venezuela, West Bank and Gaza

Figure 1.1:

Municipal Solid Waste Treatment Methods World-Wide & Sweden

(Hoornweg & Bhada-Tata 2012)

The reason for a more sustainable waste treatment in Sweden is the high legislation standard.

Landfills need to be set up in a controlled way, fees apply to landfill waste, it is forbidden to

landfill organic waste and even higher fees apply if hazardous waste is landfilled. Bulky

wastes are therefore avoided to be landfilled as it is firstly very cost intensive and secondly

not sustainable as the number of landfills in Sweden are decreasing in the same time (Hansson

2013). A rather simple method on how to reduce the big waste volume, especially bulky

waste, is incineration. As this process produces energy, it can be considered as sustainable. To

assure a controlled incineration process without any harm to the environment the WtE plants

need to have good flue gas cleaning systems. In addition safe treatment methods of the

resulting ashes are required.

One WtE plant in Sweden is owned and run by Renova AB, located in Sävenäs in

Gothenburg: see chapter 4 for a detailed process description. In total there are 32 WtE plants

in Sweden (Johansson, 2012), which have produced over 200 000 tonnes of fly ash in the year

2012 (Flyhammar 2013). 50% of the Swedish household and industrial waste is energetically

processed for both electricity and district heating, which refers to 5 000 000 tonnes burning

material per year (Avfall Sverige, 2012).

Slag or bottom ashes from waste incineration can be used as additives in construction

materials, whereas fly ash is classified as hazardous waste due to its high concentration of

easily dissolved chlorides and heavy metals (Pettersson et al., 2013; Steenari & Zhao 2010).

As a consequence it can only be landfilled or treated outside of Sweden (Pettersson et al.

2013; Avfall Sverige, 2012). Half of the annually produced Swedish fly ash is transported to

Langøya in Norway or old German salt mines where it is treated and used as filling material

(Pettersson et al., 2013).

Renova stabilises the fly ash through the so called “Bamberg method”. The resulting product

is a “Bamberg cake” which can be landfilled as non-hazardous waste (Renova AB 2013).

Hereby, the fly ash is blended with thickened water treatment sludge by a portion of one

weight part sludge to two weight parts ash. The outcome is a pasty product, which sets fast

without any chemicals added. It is almost impermeable to water: if exposed to water only 5%

results in leakage water (Thomé-Kozmiensky, 1994). Future Swedish legislations may

however lead to the situation that this method no longer can be used. Henceforth, even though

stabilised by the Bamberg method, the fly ash would still be classified as hazardous waste.

Therefore new treatment methods need to be found in order to avoid high landfilling costs.

One interesting component in fly ash is the heavy metal zinc. Pure zinc is applied in different

industries. Even though a definite amount of virgin metal is indispensable, a certain demand

could be satisfied by metal recovery from fly ash instead of developing new mines. Studies

showed that the content of zinc in some fly ashes is comparable to the content in ores where

new zinc mines are set up (Boliden Group 2013).

As the zinc content in the fly ash from Renova is rather high, it is investigated if a recovery

process through leaching and precipitation + filtration of zinc hydroxide is applicable.

The zinc hydroxide product may then be sold directly to relevant companies or further

upgraded to high purity zinc metal.

This master thesis will take a deeper look into the overall process of the municipal solid waste

incineration (MSWI) plant Renova AB. In form of a pilot plant, a possible applicable process

is shown, on how to treat fly ash in order to minimize its toxicity and to recover zinc by using

scrubber liquids. The pilot plant is 16 times smaller than if applied to the real life process at

Renova.

2.

Background

non-hazardous waste (Pettersson et al. 2013). A method, which is already applied in a plant in

Switzerland.

The results of all studies mentioned below are used as a basis for this pilot plant: its scale,

construction and the variation of the experiments in the different campaigns. Additionally

some experiments are made in the laboratory of Renova itself.

One of the studies discussed how to wash fly ash from Renova with normal water in a

multiple step process (Steenari & Zhao 2010). The fly ash was mixed with water and blended

for a certain time. When filtrated, the filtrate liquid was stored and used to wash unwashed

ash. The result of the study showed that chlorides could be washed out by washing the ash

with normal water and reusing the wash water as first washing liquid. The fact that normal

water can wash out chlorides is used during the leaching and the precipitation and filtration

experiments (P&F) of the pilot plant.

Another study, which knowledge is used, is the master thesis from Preiss, 2013. The thesis

studied the behaviour of fly ash from Renova when mixed with HCl and, which blending is

best to leach out zinc. The most important results from the thesis applied at the pilot plant

leaching campaigns are:



pH value of 3,5 is ideal to leach zinc



liquid to solid ratio (L/S) shall be over 3



best residence time for the fly ash HCl blending is 10 – 20 minutes



zinc concentration decreases if the residence time is over 30 minutes



pH value increases over time if no more acid is added

The study also suggests a two-step leaching process, which is not applied at the pilot plant as

it is considered that a one step process will have enough zinc outcome and less costs.

The study from Pettersson, and co-workers 2013 is the main study for this pilot plant project.

A scenario is suggested in, which zinc is leached from fly ash using hydrochloric acid and

where the ash residue from the leaching campaign is washed with water in order to reduce the

chloride content. The residence time in the blending tank during the leaching process shall

last less than 30 minutes and the L/S shall be 3. The separation of the ash residue from the

zinc liquid filtrate shall be done by a vacuum belt filter. Separated from chlorides and other

heavy metals, the ash residue might then be incinerated in order to destroy dioxins. The

method of reincineration of the ash residue is already applied in a plant in Switzerland

without any negative consequences (Schlumberger, Schuster, Ringmann & Koralewska 2007

& BSH Umweltservice AG 2013). The stored liquid from the leaching process shall be set

into a neutralisation tank, in which the pH value shall be increased up to a level of 9.0

– 9.3.

This can be done by blending the liquid with sodium hydroxide. By increasing the pH value,

zinc hydroxide will precipitate and can then be separated from the liquid by using a chamber

press filter. The economical conclusion of the study is that: by applying the described process,

Renova could reduce its CO

2

emission and reduce the process costs compared to the treatment

Another possibility to leach out zinc, is to use an alkali solution. However this also implies a

lower leaching efficiency of zinc and the leaching of elements, which are not intended to be in

the end product (Pettersson et al., 2013). This possibility is therefore not considered to be

tested in the pilot plant.

3.

Description of the Waste-to-Energy Process at Renova

The process at Renova AB is a standard process similar to any other WtE plant. Figure 3.1 is

a simplified process illustration by means that it only shows the steps relevant for the pilot

plant project.

Waste is collected from the Gothenburg region by trucks and transported to the Renova WtE

plant in Sävenäs. Here the waste is collected in a big waste bunker. A crane puts a certain

amount of waste into the furnace of the boiler. While the waste material is burnt, no other fuel

needs to be added. The energy is both used for electricity and district heating production. The

bottom ash, consisting of fine particles up to larger metal scrap parts is transported to another

plant of Renova located in Tagene where the material is sorted and recycled.

The raw gas from the furnace is led into an electric filter where most of the solid particles are

captured. This is the fly ash, which is used for the pilot plant experiments.

The rest of the flue gas is led into different scrubbers for further cleaning steps.

The first step is to wash the flue gas with normal water in order to release chlorides and by

which hydrochloric acid is created. The following desulphurisation step is made by washing

the remaining flue gas with sodium hydroxide solution (NaOH). It creates the sodium

sulphate solution, which is used in the leaching campaign of the pilot plant experiments.

A future process would be to use hydrogen peroxide (H

2

O

2

) instead of NaOH in the

desulphurisation step where the outcome would be sulphuric acid instead of sodium sulphate

solution

.

The same NaOH solution, which is used for flue gas cleaning, is also used at the pilot plant

throughout the precipitation and filtration campaign.

Figure 3.1:

Simplified WtE and Flue Gas cleaning process

Waste

Bottom Ash

Furnace

Hydrochloric Acid

Sodium Sulphate Solution

Condensate Water

Raw Gas

_E-Filter

Fly Ash

H

2

O wash

NaOH wash

Simplified WtE and Flue Gas Cleaning Process at Renova

Scrubbers

Waste-to-Energy

Flue Gas Cleaning

4.

Material

All materials used during the pilot plant project are products or by-products created at the

Renova WtE plant in Gothenburg. Some of the products such as fly ash are created during the

incineration process whereas most of the liquids are produced during the flue gas cleaning

process. See the previous chapter for more information.

The element analysation of the material is based on a blending sample consisting of at least 4

single samples taken to different times throughout the experiments.

4.1

Fly Ash

Two types of ashes are created during waste incineration: bottom and fly ash. Whereas

bottom ash also contains scrap metals, fly ash mostly consists of fine powder and larger black

soot particles. Figure 4.1 shows that the colour of fly ash can vary between lighter and darker

grey and sometimes even appear slightly yellow.

Figure 4.1:

Fly ash from 2 different samples

The overall composition of fly ash from waste incineration plants is hard to predict as the

different components of the fuel “waste” are changing continuously.

Figure 4.2:

Average composition of the used fly ash in percent by weight

The fly ash samples for the leaching experiments are collected

in so called “big bags”, which

are commonly used for bulk powder storage and handling. Big bags are big plastic bags,

which have a volume of around one cubic meter and are easy to cut from the bottom. The fly

ash is collected at

the “Bamberg blender”. Therefore the sludge flow into the blender is

stopped so that only fly ash enters the blender. It is therefore possible to collect the fly ash at

the place where usually the “Bamberg cake” is the outcome.

The fly ash samples for analysation have been taken before the fly ash entered the blending

tank of the ash leaching campaign. See chapter 5 and 7 for a detailed description of the pilot

plant.

4.2

Scrubber liquids

4.2.1

Hydrochloric acid

The hydrochloric acid solution (HCl 3.5 wt%) is generated at the first flue gas cleaning

scrubber when the incoming flue gas is washed with normal water.

It has a transparent to light yellow appearance with small black particles. Figure 4.3 shows

this appearance some weeks after the sample was taken and furthermore illustrates that

around 4 % of the liquid consists of chlorides whereas the metal part is so low that it can be

neglected.

Figure 4.3:

HCl appearance and average composition in percent by weight

Cl

4 wt%

Metals

0 wt%

H

2

O +

Other

96 wt%

HCl

Cl

Metals

Cl

11%

S

5%

Zn

2%

Mg

1%

Al

3%

Ca

19%

Fe

2%

K

5%

Na

6%

Si

6%

Ti

1%

Other

39%

Average composition

of used Fly Ash in wt%

HCl has a very low pH value, which is around 0. Today at Renova it is neutralized with

CaCO

3

and Ca(OH)

2

, producing a CaCl

2

solution, which is later in the process used for the

production of gypsum.

4.2.2

Sodium sulphate solution

The Sodium sulphate solution (Na

2

SO

4

) is created during the desulphurisation cleaning step

when the flue gas is washed with sodium hydroxide (NaOH). Figure 4.4 illustrates its

composition. It is a transparent liquid, which has a low percent fraction of sodium (2%) and

sulphate (1%).

Figure 4.4:

Sodium Sulphate Solution average composition in percent by weight

After undergoing several cleaning steps the solution is nowadays mixed with the CaCl

2

solution, producing gypsum. The gypsum is dewatered in a vacuum filter, similar to the pilot

filter used in this project (Andersson 2016).

4.2.3

Condensate water

Condensate water is created when the flue gas in the condensing scrubber cools down, whilst

energy is recovered and converted for the district heating process. It has the same properties

as normal water although it has a slightly lower pH value and is warmed up to 60°C.

4.3

Other liquids

4.3.1

Sulphuric acid

The sulphuric acid solution (H

2

SO

4

) was bought special for this project. It is tested as a

substitute for HCl. Future processes at Renova might lead to an internal production of this

particular acid. The background idea is to use hydrogen peroxide (H

2

O

2

) instead of NaOH for

the desulphurisation step (Andersson et al. 2013).

4.3.2

Sodium Hydroxide

Sodium hydroxide (NaOH) is used at Renova in the internal waste water treatment. It is a

25wt% diluted, water based, transparent, solution, which is produced through

chlorine-alkali-electrolysis and bought from external producers.

5.

Pilot Plant Principle

The pilot plant set-up is based on results from previous studies and adjusted through

conclusions made within the project. The campaigns can be divided into two parts: leaching

and precipitation + filtration.

Figure 5.1

illustrates the principle of the leaching process and shows pictures from the pilot

plant leaching campaign. The fly ash is blended with hydrochloric acid or other scrubber

liquids in order to leach out zinc. The residence time and pH value of the slurry is regulated to

an intended value. The liquid is separated from the ash residue by the usage of a vacuum belt

filter. The zinc filtrate liquid is stored in so called “cipax tanks” and the ash residue in big

plastic containers.

a)

Leaching Principle

b)

Pilot Plant Leaching Campaign

Figure 5.1:

a) Leaching Principle b) Pilot Plant Pictures of Leaching Process

The second step is to blend the stored zinc liquid with sodium hydroxide in order to achieve a

precipitation of zinc hydroxide. The suspension is then filtered in a chamber press filter from,

which

the outcome is a “zinc cake” and filtration liquid. Figure 5.2 illustrates this principle

and shows pictures from the pilot plant precipitation + filtration campaign (P&F).

a)

Precipitation Principle

Big Bag with fly ash

on big bag cutter

Fly Ash is blended with

Scrubber Liquids. The

pH value of the Slurry

is pH regulated

Filtered Slurry (Ash

Residue) on vacuum

belt filter

Filtrate (Zinc Liquid)

from vacuum belt

filter stored in cipax

tank

b)

Pilot Plant P&F Campaign

Figure 5.2:

a) Principle of Precipitation b) Pilot Plant Pictures of Precipitation Process

6.

Laboratory experiments

Additional to the pilot plant, laboratory experiments are made in order to investigate the

precipitation behaviour of zinc and magnesium in the zinc liquid. The experiments are done in

the laboratory of the university in Borås.

6.1

Laboratory Set-up

Zinc liquids from the pilot plant leaching campaign are taken to Borås. Different pH values

for the zinc hydroxide precipitation are set throughout several experiments.

The zinc liquid is blended in a glass flask with the same NaOH solution that is used in the

pilot plant P&F experiments (see chapter 8). After the intended pH value of the suspension is

set and kept constant for at least 30 minutes, the suspension is filtered in a vacuum filter. The

zinc hydroxide crystals fasten in the filter whereas the filtrate liquid passes the filter and is

collected in the vacuum flask. Figure 6.1 illustrates the laboratory set-up.

NaOH is filled into

blending tank where it is

mixed with the Zinc Liquid

The pH value of the solution is

regulated so that Zinc Hydroxide

precipitates

Figure 6.1:

Scheme of Laboratory Set-Up

Samples of the produced zinc cake are then analysed with a spectrometer in order to

determine the zinc and magnesium content.

Figure 6.2

shows the same set-up in real pictures of the used equipment.

Equipment from laboratory

Figure 6.2:

Laboratory Equipment

Vacuum filter

Computer

stirring device

Laboratory Set-Up

Zinc cake

pH regulated

zinc liquid

Filtrate liquid

Suspension

pH electrode

b)

pH-adjustment

a)

Filtration of Zn(OH)

6.2

Zn-Precipitation Method in Laboratory

The pH value of the zinc liquid is adjusted with different prefill methods in the same way as it

is done in the pilot plant P&F campaign (see chapter 8). One prefill liquid is NaOH solution,

the other one is zinc liquid itself. The different prefill methods are done as the morphology of

the zinc hydroxide crystal formation is changing depending on, which prefill method is

chosen (Top & Cetinkaya 2015). Different crystal formations may lead to a different product

quality regarding different element contents in the product. The zinc cake product content is

further investigated in this report whereas the morphology isn’t. The pH values for both

prefill methods are adjusted to 7, 8, 9 and 10. The used zinc liquids are those from the

leaching campaign C and D as those are the campaigns, which seem the most probable for a

future big scale process at Renova (see chapter 7).

100 ml zinc liquid and as much NaOH solution as needed are mixed until the intended pH

value is achieved. The stirring device is set to a slow blending rotation for 30 minutes before

the suspension is filtered.

The suspension is filled into the filter. It is done by filling in 25 ml at a time, pouring in more

suspension before cracks in the zinc cake are formed. The filter time is measured until all the

suspension is filled into the filter and the first cracks are formed. The filter time measurement

indicates better or worse filtration. Table 6.1 shows the variations and filter time of the

laboratory experiments.

Table 6.1:

Overview over Laboratory variations

The filter cake and the filtrate is analysed by a spectrometer. The water content of the zinc

cake is analysed by drying a sample in the oven over 2 days at a temperature of 100°C.

Experiment

Zinc liquid

pH

NaOH prefill

Time of filtration [min]

The zinc cake samples are prepared by dissolving a small amount in nitric acid, which is then

diluted with distilled water. Table 6.2 shows the dilution.

Table 6.2:

Overview over Laboratory sample preparation

After the liquid samples from the pipe have been taken, they are mixed with 250 ml of

distilled water in a glass flask. A sample from the glass flask is then taken to analysation. For

the filtrate: 0.4 ml filtrate sample is taken and mixed with 100 ml distilled water in a glass

flask. It is then analysed with a spectrometer.

7.

Pilot Plant: Leaching Campaign

Based on the results from previous studies a liquid to solid ratio of a minimum of 3 and a pH

value of 3.5 is mainly studied throughout the leaching experiments (see chapter 2). A

residence time, lower than 20 minutes is set in the blending tank.

7.1

Leaching Set-up

The regulation of the level in the tank and the pH value is stirred manually. The numbers

shown in the tables below refer to the average value when the intended circumstances remain

constant throughout at least 20 minutes and samples are taken.

Figure 7.3

shows the set-up of the leaching process. The tank is first filled with scrubber

liquid until a level of 60 litres. The big bag with fly ash is set onto the big bag cutter (also see

figure 7.1

). Two knives cut a hole into the bottom of the bag. The fly ash falls into transport

screws, which move the fly ash into the blending tank. The ash flow is turned on once the

liquid in the blending tank reaches the 60 litres mark. The blending device is turned on and

Zinc Cake

Experiment

Total suspension

[ml]

sample

cake [g]

nitric acid

[ml]

dist. Water in

pipe [ml]

both flows fill up the tank to an intended level (usually 90

– 110 litres; also see figure 7.2).

The produced slurry in the blending tank is pH regulated (usually to around pH 3.5). The pH

value is measured by a pH electrode, which is located at the inside of the tank. A pump

transports the slurry from the blending tank to the vacuum belt filter once the intended slurry

level is reached. The pump efficiency is set to a level so that the inflow stream is equal the

outflow stream. The slurry from the tank reaches the vacuum belt filter from the top. The belt

of the filter is moving from one end to another in several steps. Once the slurry reaches the

belt, a vacuum from the bottom sucks away the zinc liquid filtrate. It is pumped into a cipax

tank and stored for the P&F campaign. The ash residue on the belt filter is washed twice. First

with the collected washing water from the second washing step and then with condensate

water from the scrubber (see figure 7.3). Once the ash residue is transported to the end of the

filter it is collected in a plastic container.

Figure 7.1:

Big bag on big bag cutter

Figure 7.2:

Blending tank during leaching

Big Bag with

Fly Ash

Big Bag

cutter

Fly Ash

Leaching Set-Up

Figure 7.3:

Leaching Process Set-up

The blending device in the blending tank is a small propeller in the middle of the tank. As the

propeller is only attached on the top of the tank a liquid level of a minimum of 60 litres is

needed. A lower liquid level might cause a high vibration of the propeller, which may lead to

a damage. The primary liquid fill up also minimizes the risk of agglomeration of fly ash in the

tank, which otherwise could lead to a process stop or a damage of the propeller. A scale was

manually written on the outside of the tank in order to control the liquid level in the tank.

Throughout the experiments it happened that the agglomeration in the hoses from the

blending tank to the filter is so high that a pump stop occurs. If that occurs the pump direction

is changed every 5 seconds so that the back and forth movement of the liquid resolves the

blockage in the hose.

The pH electrode in the blending tank is calibrated every week.

Depending on the set-up, the belt of the vacuum belt filter moves every 10 to 15 seconds. In

total it has around 12 movement steps from when the slurry enters until the ash residue is

collected. The filter is divided into different zones (also see figure 7.4). The first zone, which

covers two movement steps, collects the zinc liquid filtrate from the slurry. The washing

process of the ash residue covers around three movement steps. During the remaining

movement steps the liquid in the ash residue is vacuumed, collected in a different zone from

the filtrate and led to the drain.

Big Bag with

Fly Ash

Knives

Dosing Screws

Blending Tank

Income Scrubber

Liquids

Slurry

Vacuum Belt Filter

Zinc Liquid to Cipax

Figure 7.4:

Inside of the vacuum belt filer with ash residue and washing procedure

7.2

Leaching Method

Eleven different leaching variations are made in how the fly ash is blended with scrubber

liquids in order to leach out zinc. The campaigns are called A to K. They are mainly focused

on mixtures with fly ash and hydrochloric acid as this will most likely be applied at Renova

AB in a big scale process. Table 7.1

gives an overview over the variations of the leaching

campaign. The different colours in the table indicate the different scrubber liquids. All

parameters for the blending are:



different scrubber liquids:

Hydrochloric acid

Sodium sulphate solution

Condensate

Sulphuric acid



pH value, which was tried to be kept between 3 and 4 when acid liquids were used:

pH 3.2

– 3.6



Liquid to solid ratio (L/S); indicates how much liquid is added to ash  L/S 3 means

that there is 1 part ash to 3 parts liquid:

L/S 3

– 6



residence time in the blending tank:

Residence time 8

– 16 min

Campaign

Hydrochloric

Acid

Sodium

Sulphate

Solution

Condensate

Sulphuric

Acid

pH

L/S

Residence

time

[min]

A

x

-

-

-

3,6

3,1

13

B

x

-

-

-

3,6

3,3

8

C

x

-

-

-

3,3

3,5

10

G

x

-

-

-

3,3

3,6

11

H

x

-

-

-

3,2

4,4

16

J

x

-

-

-

3,4

3,3

13

D

x

x

-

-

3,5

6,2

8

F

x

x

-

-

3,6

5,5

9

I

x

-

x

-

3,5

5,8

9

K

-

-

x

x

3,6

3,9

10

E

-

-

x

-

10,2

3,9

9

Table 7.1:

Overview of Leaching Variations

The leaching in campaign A, B, C, G, H and J is made with hydrochloric acid and fly ash

only. An overview therefore can be seen in Table 7.2. The purpose of campaign A is to get to

know the equipment and develop a feeling for the manual stirring by using the real material.

An ash flow of 100 kg/h and a hydrochloric flow of 306 l/h is found in order to keep the pH

value at 3.6 at a residence time of 11 minutes. In campaign B the maximum capacity of the

vacuum belt filter is tested by using a higher ash and hydrochloric acid flow. The residence

time hereby is kept rather low at 7 minutes. The campaigns C, G and J are set up like

campaign A although ashes from different big bags are used. This is made to obtain more data

and henceforth achieve more reliable results. Campaign H has the purpose to test a higher

residence time.

Ash (kg/h)

HCl [l/h]

Slurry on

filter [l/h]

L/S

Level in

Tank [l]

Residence

time [min]

pH

A

100

300

516

3,1

98

11

3,6

B

150

500

816

3,3

96

7

3,6

C

100

350

450

3,5

82

9

3,3

G

100

364

575

3,6

93

10

3,3

H

100

437

647

4,4

155

14

3,2

J

80

325

493

3,3

97

12

3,4

Table 7.2:

Leaching Campaign with Hydrochloric Acid A, B, C, G, H, J

hydrochloric acid might be used for the leaching of zinc, a big part of the sodium sulphate

solution might be overproduced. It is hence tested how the leaching efficiency of zinc from

fly ash is affected if sodium sulphate solution is also brought into the process. Table 7.3

illustrates that the campaigns are set up in a rather similar way. The pH value is set to 3.5 and

3.6 with a residence time of 8 minutes in both experiments. The main variation is made within

the liquid ratio. In campaign D a higher ratio of sodium sulphate solution is used whereas

campaign F has a higher ratio of hydrochloric acid.

Ash

(kg/h)

HCl

[l/h]

Sodium

Sulphate

Solution [l/h]

Slurry

on filter

[l/h]

L/S

Level in

Tank [l]

Residence

time

[min]

pH

D

75

211

250

619

6,2

81

8

3,5

F

100

356

194

760

5,5

106

8

3,6

Table 7.3:

Leaching Campaign with Hydrochloric Acid and Sodium Sulphate Solution D, F

After evaluating the first results it is considered that a high L/S ratio is profitable for a high

zinc leaching efficiency and a good process stirring. Campaign I investigates if it is really due

to the higher L/S ratio and not due to the sodium sulphate solution. A blending as in campaign

D is therefore set up by using condensate water instead of sodium sulphate solution. The pH

value is kept at 3.5 and the residence time at 9 minutes (table 7.4).

Ash

(kg/h)

HCl

[l/h]

Condensate

[l/h]

Slurry

on filter

[l/h]

L/S

Level in

Tank [l]

Residence

time [min]

pH

I

75

216

223

596

5,8

86

9

3,5

Table 7.4:

Leaching Campaign with Hydrochloric Acid and Condensate I

As already mentioned, a future process at Renova could lead to a production of sulphuric

acid. It is therefore tested in campaign K if sulphuric acid can replace hydrochloric acid in the

zinc leaching campaign. The used sulphuric acid solution has a higher concentration as the

hydrochloric acid solution from the scrubber and is therefore blended with condensate water.

The L/S ratio, residence time and pH value is approximately the same as when hydrochloric

acid is used. Table 7.5 illustrates the data of campaign K.

Ash

(kg/h)

Condensate

[l/h]

Sulphuric

acid (l/h)

Slurry

on filter

[l/h]

L/S

Level in

Tank [l]

Residence

time

[min]

pH

K

100

320

71

601

3,9

93

9

3,6

Table 7.5:

Leaching Campaign with Sulphuric Acid and Condensate K

Campaign E is made in order to evaluate the chloride washout in the fly ash. Therefore ash is

blended with condensate water only. As the pH value is far too high (pH 10.2) to leach out

zinc it is not expected to obtain a good zinc liquid filtrate. Table 7.6 shows the details of

campaign E.

Ash

(kg/h)

Condensate

[l/h]

Slurry on

filter [l/h]

L/S

Level in

Tank [l]

Residence

time [min]

pH

E

125

490

753

3,9

97

8

10,2

Table 7.6:

Leaching Campaign with Condensate E

The calculations of the different values in the tables can be found in Appendix 1. During the

campaign experiments itself, the numbers were estimated in a simpler way so that faster

adjustments of the settings could be made.

The ash residue is delivered back into the furnace after being washed on the vacuum belt

filter. In theory the chlorides are washed out (Steenari & Zhao 2010), dioxins get destroyed

through re-incineration and instead of recreating fly ash the ash residue will terminate as

bottom ash (Vehlow et al, 1991). Results and further description of the back delivery of the

ash residue i

nto the furnace isn’t discussed in this thesis but in the final report of the

“Energiforsk” project: Hållbar behandling av avfallsflygaska genom lakning, termisk

behandling och zinkåtervinning

.

8.

Pilot Plant: Precipitation and Filtration Campaign

The precipitation of zinc hydroxide from the previously stored zinc liquid shall be achieved

by regulating the pH value of the liquid to 9.25 (EPA 1980). As figure 8.1 shows, this is the

value where the solubility of zinc hydroxide is the lowest and therefore the precipitation the

highest.

Figure 8.1:

Precipitation of Heavy Metal Hydroxides

Figure 8.2

underlines why a pH of 9.25 is chosen. It shows that the preciptiation of zinc can

also be achieved at higher pH values towards pH 10. Though is also the precipitation of

cadmium, nickel and lead higher towards this value. As the precipitation and filtration

campaign concentrates on the precipitation of zinc, the pH value is chosen, which lays below

a high precipitation rate of these elements: 9.25.

Figure 8.2:

Precipitaion of Zinc and other metals

(Hoffland Env. 2012)

Further investigation of the precipitation behaviour with the used material were made in the

laboratory of the University of Borås (see the results in chapter 9.1).

8.1

Precipitation and Filtration Set-up

The same blending tank and pH electrode as in the leaching experiments are used for the P&F

experiments. Both are cleaned before the start of the experiments and the pH electrode is

newly calibrated. The regulation of the pH value in the blending tank is stirred manually in

the same way as it was done throughout the leaching experiments.

Figure 8.3

illustrates the set-up of the precipitation and filtration experiment. The zinc liquid

is filled into the blending tank to a level of approximately 190 litres. The pH value is

pressure and pressurised air. The out coming wash water filtrate is pumped to the drain. The

chamber press filter is then opened and the zinc cake collected.

Precipitation and Filtration Set-Up

Figure 8.3

Set-up of Precipitation and Filtration Process

The calibration of the pH electrode is done regularly.

The formation of zinc hydroxide crystals is ensured by setting the blending device to a low

mixing rate for a certain time (minimum 30 minutes).

The suspension is pumped from the blending tank into the chamber press filter through a

hose, which is put from the top opening of the blending tank, reaching down to the tank

bottom. The flow rate of the suspension into the chamber press filter is stirred by manual

pump regulation with no direct flow rate measurement. If more than 100 litres of suspension

are pumped into the filter, the zinc hydroxide

crystals don’t agglomerate anymore as the

capacity of the filter is exceeded. Instead, the suspension instantly exits the filter.

8.2

Precipitation and Filtration Method

Different parameters are set throughout the P&F experiments. After the first results of the

experiments are evaluated further variations are made based on this data.

The parameters of the P&F experiments are:



Zinc Liquid

Different zinc liquids originating from the different leaching campaigns are tested. In

order to know, which zinc liquid is used, the P&F campaigns are marked with the

NaOH 25%

Zinc Liquid

Blending Tank

Suspension

Wash Water

1. Filtrate to internal

water cleaning

2. Wash Water Filtrate

to drain

same letter as used in the leaching campaign (A-K; e.g. FA refers to the zinc liquid

from the leaching campaign A).



_{Warmed Zinc Liquid}

The zinc liquid is warmed to approximately 40°C directly in the cipax tank. A hose

with warm condensate water is therefore put into the cipax. The condensate water is

not mixed with the zinc liquid but led through a hose into the drain. Figure 8.4

illustrates this warming process.

Figure 8.4:

Warming procedure of Zinc Liquid in Cipax tank



Flotation

The purpose of flotation is to mix the liquid in the cipax tank. The fine solid particles

(fly ash that follows the zinc liquid filtrate at the leaching campaign), which have

sunken to the bottom during the storing time are bloated back into the liquid.

Therefore pressurised air is led into the cipax by a hose and an attached staff at its end.

Figure 8.5

illustrates this procedure.

Figure 8.5:

Flotation procedure of Zinc Liquid in Cipax Tank

Cipax with Zinc Liquid

Warm

Condensate

Condensate

to drain

Pressurised Air



Prefill of Zinc Liquid or Sodium Hydroxide Solution

The regulation of the pH value is done by two variations. The first one is to fill up the

blending tank with zinc liquid and then adjust the pH value by adding sodium

hydroxide. The other variation is to prefill the tank with the sodium hydroxide solution

and then regulate the pH value by adding the zinc liquid. The difference between those

variations may affect the formation behaviour of the zinc hydroxide crystals. If the

zinc liquid is prefilled, the crystals experience an alkali environment around pH 6. At

the variation where the sodium hydroxide solution is prefilled the crystals experience a

high pH value environment of around pH 12 before the value is lowered to the

intended pH 9.25. Different elements can therefore be built into the crystal structure,

which may lead to a changed zinc cake composition (Separation Science and

Technology, 2009).



_{pH Value}

The variation of the pH value may as well affect the composition of the zinc cake.

Depending on its value other components (e.g. magnesium) of the zinc liquid also

precipitate more or less (Materials science forum, 2014). A variation is therefore made

in order to find the perfect zinc cake composition.



Blending Time

The blending time affects the formation of the zinc hydroxide crystals. In total all

experiments are blended for at least 30 minutes. Some of them even have a blending

time of more than one hour.



Precoating

Precoating means that a bit of ash residue from the leaching campaign is blended into

the regulated pH suspension. The idea is that the zinc cake gets more porous and is

therefore better to wash as the occurrence of big cracks minimizes.



Flocculent

A flocculent shall bind the formed crystals more together in order to ease the filtration.

However the experiments showed that the addition of the flocculent (Magnafloc) had

no effect, which leads to the decision to use no flocculent at the P&F experiments.

Therefore the flocculent is only added in the first experiments.



Amount of suspension into filter

The amount of suspension into the filter is tried to be kept at 100 litre. The same

suspension volume ensures an accurate comparison of the zinc cake product.



Zinc cake build up time

A pump regulates the flow rate of the suspension into the filter and therefore the

rapidness of the cake build up (~10-30 min). A slower cake build up might lead to a

better washing rate and therefore a better product.



Type of wash water

Two different wash waters are used: normal tab water and warm condensate water

from the scrubber (~40°C). Both liquids might get used at a big scale process.



Amount of wash water and time of washing

Different wash settings may lead to different properties of the zinc cake product (e.g.

chloride content). A lower flowrate might minimize big cracks in the zinc cake and

therefore avoid a one-sided washing.



Washing method

Two different wash settings are available at the used chamber press filter. One way is

to wash the cake in the filter from the centre to the outside. The other method is to do

a cross wash where the wash water enters the cake from each corner. A variation of

these washing methods is done by either using only the centre wash or applying both

methods. If both methods are applied it is firstly washed from the centre and then

changed to the cross wash (half cross wash).



Repulp

The repulp campaign is tested in order to reduce the chloride content in the zinc cake

product. A zinc cake product is put into a big bucket and blended with water. This mix

is manually blended for around 20 minutes. The suspension is then pumped into the

filter and filtrated again.

Table 8.1

shows the parameters for each P&F campaign. The different colours imply, which

zinc liquid was used. Grey stands for zinc liquid that was produced by only using

hydrochloric acid, green for hydrochloric acid + sodium sulphate solution, blue for

Washing

Campaign

Warmed Flotation

Zinc

Liquid

prefill

NaOH

prefill

Precoating

pH

Flocculent

1h

blending

Suspension

[l]

Water

Condensate

water

Cake build

up [min]

Cake

Washing

[min]

Cake

build up

[l]

Cake

Washing

[l]

half

cross

wash

F1B

x

-

x

-

-

9,0

x

-

100

-

x

30

12

90

80

-

F2B

x

-

x

-

-

9,0

x

-

90

-

x

21

14

80

30

-

F3A

x

-

x

-

-

9,1

x

-

85

-

x

12

8

60

50

-

F4A*

x

-

x

-

-

9,2

x

-

60

-

x

16

5

60

40

-

F5A

x

x

x

-

-

10,5

x

-

72

-

x

10

8

70

50

-

F6C

x

x

x

-

-

9,0

-

-

98

-

x

9

8

90

80

-

F7C*

x

x

x

-

-

9,2

-

-

98

x

-

14

24

110

80

x

F12C

x

x

-

x

x

10,5

-

x

87

x

-

26

18

70

110

-

F13C*

x

x

-

x

x

8,1

-

x

95

x

-

22

20

80

110

-

F11D

-

-

x

-

-

9,1

x

-

110

x

-

13

0

130

0

-

F14D

x

x

-

x

-

9,3

-

x

90

x

-

17

24

80

110

-

F15D*

x

x

-

x

x

9,3

-

x

75

x

-

22

23

80

110

-

F16D

x

x

-

x

-

6,9

-

x

82

x

-

22

36

70

110

-

F17D*

x

x

-

x

-

7,9

-

x

83

x

-

19

35

80

120

-

F10I

-

-

x

-

-

9,3

-

-

120

x

-

9

6

110

70

x

F8K

x

-

x

-

-

9,0

-

-

98

x

-

12

10

90

70

x

F9K*

x

-

x

-

-

9,0

-

-

79

x

-

10

12

80

70

x

F18K

x

x

-

x

-

9,6

-

x

85

x

-

36

37

70

110

-

F19K*

x

x

-

x

-

7,8

-

x

99

x

-

37

37

90

110

-

Repulp**

-

-

-

-

-

8,5

-

-

95

x

-

-

-

80

0

-

*Mixture is based on suspension of previous campaign e.g.: F3A was the base for the suspension of F4A

**Zinc cake from campaign F11D was used

9.

Results and Discussion

All results and conclusions are funded on: the samples from the pilot plant, which are analysed by

an external laboratory, the experiments in the laboratory in the University of Borås and the

conclusions during the pilot plant process.

The samples from the pilot plant experiments are aggregate samples. They consists of single

samples, which are collected to different times of the experiments and then mixed together before

sent in to the external laboratory. These aggregate samples consist of a minimum of four different

single samples. The purpose of this procedure is to prevent single outrages. The rule of thumb is:

the more single samples form the aggregate sample, the more realistic are the results.

9.1

Pilot Plant Leaching

The average zinc leaching efficiency of all leaching experiments is 60% (figure 9.3). It refers

to the zinc content in the fly ash and the zinc concentration in the zinc liquid. The average of

60% doesn’t include the results from campaign E as the leaching of zinc is not its main

priority. However, all results are shown in figure 9.3, including campaign E. The best

leaching result of 74% is achieved in campaign D where the fly as is mixed with HCl and

sodium sulphate solution. 47% is the lowest leaching result except from E (25%) and

occurred at campaign I where HCl was mixed with condensate water. Campaign B, G, H and

K have leaching results above average whereas A, C, J and F show lower results. As predicted

E has a much lower leaching result with only 25 %.

Figure 9.3:

Leaching result of Zinc from Fly Ash

The pH value is kept between 3.2 and 3.7. As figure 9.4 illustrates, all tested pH values led to

a zinc leaching efficiency over 45%. The best pH value from previous laboratory

investigation is at 3.5 (Pettersson et al., 2013). However, in the pilot plant the best and the

worst leaching efficiency was achieved at this value. The results don’t show if different pH

values shall be applied to different liquid blends. It can be assumed that the blending with

HCl and condensate water leads to the worst leaching of zinc (campaign I 45%). However as

this blending was only tested once the result can also be an individual case. Future studies in

0%

20%

40%

60%

80%

100%

D

H

K

G

B

C

F

A

J

I

E

Leaching: Zinc

Linear (Total)

Linear (Average)

*

this field might be of interest. In total it can be said that all pH settings showed a satisfying

zinc leaching efficiency.

Figure 9.4:

Zinc leaching efficiency over pH

A valuable end product has a high purity of zinc. In order to obtain a high purity product, a

leaching efficiency is preferred that gives a low concentration of other metals. Figure 9.5

compares the leaching efficiency of zinc and magnesium. The magnesium content is of

importance as relevant customers demand a low content (Boliden: 2735 mg/kg Zinc).

Relevant customers are companies, to, which it might be able to sell the zinc product. It can

be said that if the leaching efficiency of zinc is higher, the leaching efficiency of magnesium

is higher as well. The linear line for both metals shows that the magnesium leaching

efficiency at higher pH values seems to minimize whereas the leaching efficiency of zinc

keeps rather constant.

For the big scale process a higher pH value of the tested range (pH 3.7) is therefore

recommended.

Figure 9.5:

Leaching efficiency of Zinc and Magnesium over pH

Previous laboratory studies showed that the best leaching efficiency of zinc was achieved at a

pH value of 4.2 (Preiss 2013). It might therefore even be possible to set the future leaching

process to 4.2. As the pH value of 4.2 wasn’t tested throughout the leaching experiment it

cannot be confirmed that the high zinc leaching efficiency also applies to a big scale process.

A higher pH value blending will be positive as less acid solution is necessary to achieve this

value.

0%

20%

40%

60%

80%

100%

3,2

3,3

3,4

3,5

3,6

3,7

pH

Leaching efficiency:

pH

Leaching rate Zinc

Leaching rate Magnesium

Linear (Leaching rate Zinc)