SID 5 Research Project Final Report - Defra, UK

SID 5 (Rev. 3/06) Page 1 of 25

General enquiries on this form should be made to:

Defra, Science Directorate, Management Support and Finance Team,

Telephone No. 020 7238 1612 E-mail: [email protected]

SID 5 Research Project Final Report

Note

In line with the Freedom of Information Act 2000, Defra aims to place the results of its completed research projects in the public domain wherever possible. The SID 5 (Research Project Final Report) is designed to capture the information on the results and outputs of Defra-funded research in a format that is easily publishable through the Defra website. A SID 5 must be completed for all projects.

This form is in Word format and the boxes may be expanded or reduced, as appropriate.

ACCESS TO INFORMATION

The information collected on this form will be stored electronically and may be sent to any part of Defra, or to individual researchers or organisations outside Defra for the purposes of reviewing the project. Defra may also disclose the information to any outside organisation acting as an agent authorised by Defra to process final research reports on its behalf. Defra intends to publish this form on its website, unless there are strong reasons not to, which fully comply with exemptions under the Environmental Information Regulations or the Freedom of Information Act 2000.

Defra may be required to release information, including personal data and commercial information, on request under the Environmental Information Regulations or the Freedom of Information Act 2000. However, Defra will not permit any unwarranted breach of confidentiality or act in contravention of its obligations under the Data Protection Act 1998. Defra or its appointed agents may use the name, address or other details on your form to contact you in connection with occasional customer research aimed at improving the processes through which Defra works with its contractors.

Project identification

1. Defra Project code WR0204 (WRT 236)

2. Project title

New technologies to allow beneficial reuse of silt from construction and demolition waste recycling washing plant

3. Contractor organisation(s)

Imperial College London

54. Total Defra project costs £ 201,742

(agreed fixed price)

5. Project: start date ................ 01 February 2006

end date ................. 31 January 2008


6. It is Defra’s intention to publish this form.

Please confirm your agreement to do so. ................................................................................... YES NO

(a) When preparing SID 5s contractors should bear in mind that Defra intends that they be made public. They should be written in a clear and concise manner and represent a full account of the research project which someone not closely associated with the project can follow.

Defra recognises that in a small minority of cases there may be information, such as intellectual property or commercially confidential data, used in or generated by the research project, which should not be disclosed. In these cases, such information should be detailed in a separate annex (not to be published) so that the SID 5 can be placed in the public domain. Where it is impossible to complete the Final Report without including references to any sensitive or confidential data, the information should be included and section (b) completed. NB: only in exceptional circumstances will Defra expect contractors to give a "No" answer.

In all cases, reasons for withholding information must be fully in line with exemptions under the Environmental Information Regulations or the Freedom of Information Act 2000.

(b) If you have answered NO, please explain why the Final report should not be released into public domain

Executive Summary

7. The executive summary must not exceed 2 sides in total of A4 and should be understandable to the intelligent non-scientist. It should cover the main objectives, methods and findings of the research, together with any other significant events and options for new work.

This project has investigated processing waste silt generated from aggregate washing recycling plants into secondary aggregate. The fine particles that constitute silt are filtered from water used to wash recycled aggregate. Aggregate washing plants are increasingly used because legislation and financial drivers are supporting increased use of secondary aggregates. Washing is beneficial because international aggregate standards limit the quantity of fine material (typically <63μm) that should be included in recycled aggregate. To comply with these standards both recycled and virgin aggregate are washed. The wash-water contains fine silt particles which are removed by filtration so the water can be recycled. The resulting silt filter cake is a major waste disposal issue for the industry. This project had the overall aim of manufacturing viable aggregate from waste silt produced by aggregate washing and was completed by addressing the following four objectives. Objective 1: Characterisation of silt particles produced from different UK construction and demolition waste recycling washing plants The physical characteristics of silt samples were assessed using particle size analysis, X-ray diffraction, thermal gravimetric analysis, surface area analysis and examination by microscopy. The chemical characteristics assessed were sulphate and chloride content, pH and pozzolanic activity. The silt consisted of clay and larger quartz particles. It had a water content of typically ~ 30%, which gave the material similar handling properties to a potters’ clay. It had been expected that it would contain much higher levels of un-hydrated cement from concrete crushing operations which produce recycled aggregates. However, the washing operations supplying silt to this project processed high quantities of excavated waste and soil rather than concrete, and the silts therefore had high clay content. The silts were free from chlorides and sulphates, were pH neutral and were not pozzolanic. Approximately 200 tonnes of silt is generated per day at aggregate washing plants. At the start of the project there were five operational plants, although this number increased during the project period. The five recycled aggregate washing plants visited during the project produce approximately 250,000 tonnes of silt per year. This is almost entirely sent to landfill, and is associated with significant and increasing disposal costs. In some cases, silt is disposed of on the land owned by the aggregate washing company, although this was normally only seen as a temporary solution.


Objective 2: Evaluation of the properties of cement/waste silt composites containing high silt loadings Silt was mixed with different amounts of Portland cement (PC) and organic additives such as polyvinyl alcohol (PVA). The mixtures were moulded to form cylindrical samples that could be tested for unconfined compressive strength (UCS). It was found that optimised combinations of PC and PVA gave the optimum balance of UCS, cost and processing characteristics. Increasing PC and PVA increased the UCS but also increased process costs. It was found that the UCS of fully saturated samples was reduced to approximately 30% of the dry UCS. Micro-structural analysis of these novel silt/cement/polymer composites indicated a matrix of clay and quartz particles held together by PVA, with little evidence of cement hydration products. However, PC reacted with the PVA to form strong bonds with clay particles. The optimised mix consisted of 85% silt, 1% PVA and 14% PC. This gave appropriate processing characteristics and properties suitable for use as aggregate in low-grade (bulk fill) applications. The UCS of this material was typically ~15 MPa. Compaction loads for unbound aggregate are around 0.7 MPa and low-grade concretes are made with strengths of 10 MPa. Alternative cements were also considered. These included high alumina cement which is expensive and gave marginal strength improvements. Basic oxygen converter slag cement was investigated. This gave good strengths and is made entirely from waste materials. Samples containing pulverised fuel ash (PFA) also had good properties. A range of other polymers were investigated but PVA gave the best results. Research was completed to investigate improving the wet UCS by treating samples with boric acid to crosslink the polymer. This resulted in fully saturated UCS values approximately 60% the dry UCS. Objective 3: Production at pilot plant scale, testing and micro-structural analysis of rapid setting aggregate particles manufactured by high shear mixing and extrusion of polymer modified cement/waste silt composite aggregate A specially designed and constructed extruder was used to manufacture aggregates. The extruder applied pneumatic pressure to a ram in an extrusion chamber. A sensor allowed the extruded material to be cut as it exited the die to form pellets. These were “rounded off” in a pan-pelletiser and left to cure. It was found that the optimised blend of 85% silt, 1% PVA and 14% PC could be readily extruded and formed good aggregate pellets. Unbound applications use aggregate in granular form as a fill material without the addition of a binding material. A standard ten percent fines test was completed on optimised silt aggregate. It was found that a blend of 10% silt aggregate and normal aggregate from the washing plants could be used for unbound engineering fill applications. This was in dry conditions. In wet condition the properties were potentially problematic due to deformation and crushing of silt aggregate particles, rather than excessive fines production. Bound applications are where aggregate is used in concrete or asphalt. Trials were completed using the silt aggregate in low-strength concrete. Up to 50% silt aggregate could be blended with control aggregate and this achieved the properties required for these concretes. The wet concrete UCS was approximately 80% of the dry concrete UCS and the microstructure of extruded aggregate was identical to the cylindrical samples previously examined. Objective 4: Specification of commercially viable processing plant to manufacture aggregate from waste silt, identification of exploitation routes and impact on Defra policy Using data from the previous research the equipment required for silt aggregate production was specified. The economic, environmental and social impacts were assessed to give information useful for policy decisions. It was found that there were positive environmental impacts resulting from diverting waste from landfill and saving mineral resources. There were also positive social impacts due to reduced traffic near landfill and recycling sites. The economic case was less clear. In some scenarios the process was less expensive than landfill, but in other scenarios it was more expensive. The main factor affecting economic viability was the cost of PC and PVA compared to landfill disposal costs. Initial equipment costs were not a significant factor. Carbon dioxide emissions from the silt aggregate process were much higher than landfill, creating a negative environmental impact. This was due to the significant carbon emissions associated with PC. In the opinion of the authors, the overall economic viability of the process is limited by the current low-cost of inert waste landfill, with tax at 2 £/tonne. A large increase in disposal costs would make the process commercially viable with the silt/cement/polymer mix being used as aggregate. Manufacturing aggregate and saving the cost of silt disposal is only marginally economically viable at the present time. Silt could be processed into higher value products and applications and therefore the project team are investigating a range of higher value applications for the silt-binder blend. This has a number of interesting characteristics


which make it a potential material for processing into products such as low-embodied energy bricks and blocks, lightweight thermally insulating panels, lightweight aggregate, thermally insulating high thermal mass products and pumpable but fast setting high thermal mass, insulating and sound proofing filler. Potential decorative products include shaped and coloured aggregates and decorative coving and architectural mouldings. Other potential applications may include use as a catalyst support material and as an environmentally friendly lightweight packaging material. Therefore the authors believe the silt/cement/polymer material developed in this research has the potential to allow waste silt to become a valuable resource and produce environmentally-friendly low-cost materials. Further research funding is being sought to develop these options in collaboration with relevant industrial partners.

Project Report to Defra

8. As a guide this report should be no longer than 20 sides of A4. This report is to provide Defra with details of the outputs of the research project for internal purposes; to meet the terms of the contract; and to allow Defra to publish details of the outputs to meet Environmental Information Regulation or Freedom of Information obligations. This short report to Defra does not preclude contractors from also seeking to publish a full, formal scientific report/paper in an appropriate scientific or other journal/publication. Indeed, Defra actively encourages such publications as part of the contract terms. The report to Defra should include:

the scientific objectives as set out in the contract;

the extent to which the objectives set out in the contract have been met;

details of methods used and the results obtained, including statistical analysis (if appropriate);

a discussion of the results and their reliability;

the main implications of the findings;

possible future work; and

any action resulting from the research (e.g. IP, Knowledge Transfer).


New technologies to allow beneficial reuse of silt from construction and demolition waste recycling washing plant

Richard Lupo and Christopher Cheeseman

Department of Civil and Environmental Engineering Skempton Building

South Kensington Campus Imperial College London

London SW7 2AZ

1. Introduction The aim of this project was to solve the silt disposal problem at construction and demolition (C&D) recycling plants that use aggregate washing technology. It may not be economic to wash aggregate if a water treatment plant combined with the cost of silt disposal is high relative to the return from the washed aggregate. Processing silt into a product would offset these costs. This project therefore aimed to maximise recovery of high quality uncontaminated material by developing processing technology that could realise potential environmental and economic benefits. Washing recycled aggregate produces clean, high quality aggregate for which there is significant market demand. The primary aim was therefore to develop a commercially viable process to transform silt generated from C&D waste recycling plants using washing systems into viable aggregate. Specific objectives were: a) Characterisation of silt particles produced from different aggregate recycling washing plants, including assessment of their pozzolanic properties; the objective was to understand the types of silts generated by UK C&D washing plants and the factors that control variability. b) Evaluation of the properties of cement/C&D waste silt composites containing high silt loadings; the objective was to understand how silt affects the hydration and properties of different cements. The work envisaged using Portland cement, high alumina cement and novel MgO containing cement systems that harden by adsorption of CO2 from the atmosphere. c) Production at pilot plant scale, testing and micro-structural analysis of rapid setting aggregate particles manufactured by high shear mixing and extrusion of polymer modified cement/waste silt composite aggregate. The objective was to complete pilot scale trials to produce representative quantities and qualities of high silt aggregate using high shear mixer-extrusion equipment d) Specification of commercially viable processing plant that can produce aggregate from C&D waste silt, identification of routes to exploitation and impact on Defra policy on C&D waste recycling; the objective is to have identified all equipment required to process silt to aggregate in order to achieve the stated project aim. An aggregate washing plant adds value to aggregate by removing silt. A water treatment plant does not add value to the material and is a cost to plant operators. However, in many applications it would not be possible to wash aggregate without the addition of a water treatment plant as there is normally not enough land to dispose of the wash water to a lagoon. Many aggregate producers realise the benefit of using a water treatment plant as the system is fully automatic, only takes up a small area and produces a silt filter cake that can be stockpiled rather than incurring the costs of maintaining lagoons with associated health and safety issues. However silt disposal is a major issue. The project aimed to develop a commercially viable silt cake processing plant and a route to exploitation of the product developed. It should be noted that the type of novel technology used in this project could have wide applicability to other fine grained wastes arising in the UK, such as quarry dusts, combustion residues and tunnel spoil.


0

2000

4000

6000

8000

10000

12000

0 10 20 30 40 50 60 70 80

Counts

o2θ

Q Q Q

Q

0

200

400

600

800

1000

1200

14 16 18 20 22 24 26 28 30

Counts

o2θ

Clays

2. Silt characterisation 2.1 Experiments Objective 1 aimed to characterise silt produced from different aggregate recycling washing plants, including an assessment of pozzolanic properties. The industrial partner involved in the project, Duo (formerly Powerscreen), are the suppliers of aggregate washing plant at five locations in the UK. All these were visited to assess the quantities of silt produced and collect samples. These plants were in Edinburgh, Hull, Birmingham, Colchester and Welwyn Garden City. Relevant physical characteristics of all the silt samples were assessed. Water content was measured by drying at 105ºC for 24 hours. The dried samples were subject to x-ray diffraction (XRD) (Philips PW 1830 diffractometer, using Cu Kα radiation, from 1.6-70

o (2θ) with a 0.02

o step size), particle size analysis (Coulter 1500 particle size

analyser), thermal gravimetric analysis (TGA) (PL Thermal sciences, STA 1500), nitrogen absorption (Omnisorp 100 with BET surface area measurement software) and optical microscopy. The chemical characteristics were assessed. Dried samples were reacted with a barium chloride (BaCl2) solution to test for sulphate, which can lead to sulphate attack in concrete. Samples were also tested with silver nitrate (AgNO3) solution to test for the presence of chlorides, which may cause corrosion problems in steel reinforced structural concrete. For these tests, 1g of dried silt was added to 100ml of de-ionised water and shaken for 30 seconds. The liquid was then filtered and 20ml of AgNO3 was added to 20 ml of filtrate. Precipitation of white silver chloride would indicate the presence of free chloride in the sample. A further 20ml sample of filtrate was taken and 20ml of BaCl2 added. If white barium sulphate precipitated this indicates the presence of free sulphate. The pH was measured using universal indicator in a further sample of filtrate. Pozzolanicity was assessed by mixing 10g of dried sample with 2.5g of lime and adding water to achieve water to solids ratio of 0.4. 20 mm diameter x 20 mm high compression test samples were formed and left to cure for 7 days before testing for unconfined compressive strength (UCS). 2.2 Results A typical XRD data is shown in Figure 1. The major crystalline phase present was quartz. The series of relatively non-distinct peaks between 20 and 30

o 2θ have counts higher than the background at angles below 20

o 2θ, and

this is characteristic of clay. At this stage in the project the type of clays were not investigated because it was felt that the clay types would vary between sites, and the aim was to develop a generic “silt to aggregate” process that would be suitable for all sites.

Figure 1. Typical XRD data of a dried silt sample, with clay mineral region shown.

The highest volume fraction of material had particle sizes between 0-10µm as seen in Figure 2. Clay particles are typically less than 2µm and so this result indicates significant clay, in agreement with the XRD results. The fraction of particles in the larger size ranges indicate the presence of quartz, as also indicated by XRD analysis.


0

5

10

15

20

25

30

35

40

45

0-10

10-2

0

20-3

0

30-4

0

40-5

0

50-6

0

60-7

0

70-8

0

80-9

0

90-1

00

100-

110

110-

120

120-

130

130-

140

140-

150

150-

160

160-

170

170-

180

Size range (µm)

% V

olu

me

Figure 2. A typical trace from particle size analysis showing the distribution of particle sizes. The results of physical analysis are summarised in Table 1. The conclusion is that silt filter cakes from all plants contain high levels of clay and quartz particles. This was confirmed by optical microscopy. Factors which influence the water content of the silt include the amount of fines (particles less than 63μm) entering the process and the cycle time of the filter press. There were low-levels of organics present in the silt. Organic flocculants are used in the water treatment process and there may also be organic components in the aggregates being washed. Average production of fines is typically 200 tonnes per day, and variations between plants are due to the amount of lime added to the slurry prior to filtration. For instance, the Birmingham plant does not use any lime which results in low drainage time, and consequently a low output of both product and silt filter cake. The chemical characterisation results are summarised in Table 2. None of the samples exhibited substances considered harmful to concrete. The silts were an inert waste, as expected from the type of washing facilities supplying the materials. The silt samples had low pozzolanicity with UCS test results on lime/silt blends varying between 0.7 and 3.3MPa. This variation was probably due to the quantity of fines in the silt resulting from crushed concrete. The lowest UCS came from the Colchester silt where there is unlikely to be any residual cement in the silt. The conclusion from the physical and chemical characterisation was that these silts are inert waste containing high quantities of clay and quartz particles.


Table 1. Summary of the physical characteristics of silt from aggregate washing plants

Characteristic Site Welwyn

Garden City Edinburgh

11

Edinburgh 2

Colchester Hull Birmingham

Volume produced (tonnes/day)

240 259 N/A 150-180 280-320 64

Water content (% total wt)

33.8 31.4 32.4 36.9 36 31.3

Water content (% dry solids)

51.1 45.8 47.9 58.5 56.3 45.6

TGA % wt loss at 450

oC

2

6.6 2.3 2.4 1.0 6.8 3.1

TGA % wt loss at 690

oC

3

1.7 2.7 3.8 1.9 1.5 2.7

Surface area (m

2/g)

14.28 12.67 13.01 12.14 13.88 14.54

Particle size distribution (µm):

d10 82.9 70.62 69.67 148.5 50.8 56.65 d25 39.61 40.77 36.14 54.26 31.77 31.7 d50 20.59 17.16 14.73 27 13.43 14.15 d75 6.954 5.445 4.867 7.501 4.577 4.851 d90 2.06 1.905 1.756 2.163 1.759 1.751

Notes: 1 two samples were taken from the Edinburgh site due to reported varied input. Sample 2 was taken 2 hours after sample 1.

2 weight loss at this temperature attributed to disassociation of organic compounds

3 weight loss at this temperature attributed to loss of hydroxyl ions from clays

Table 2. Summary of the chemical characteristics of the silt samples from different sites. Site

Welwyn Garden City

Edinburgh 1

1

Edinburgh 2

Colchester Hull Birmingham

Free chloride None None None None None None Free sulphate None None None None None None pH 7 8 7 7.5 7.5 7 Pozzolanicity (MPa)

3.3 2.1 2.4 0.7 3.3 2.0

3. Properties of cement/polymer/waste silt composite materials The objective was to evaluate the properties of cement/polymer/waste silt composites containing high silt loadings. Initially Portland cement was used, but data for other cement are also reported. 175 different compositions were produced using various binders and different processing methods. The key results and the implications from these experiments are reported. 3.1 Preliminary study A preliminary study was completed to assess basic properties of the materials and to explore processing parameters. 3.1.1 Methods Silt filter cake from the Welwyn Garden City site was used, with Portland cement (PC, Lafarge Cement, grade CEM1 42,5N) and PVA (Nippon Gohsei, grade GH17S) (1). High water to cement ratio reduces the strength of cement pastes, and therefore dry sharp sand was added at a mass ratio of 23:57 sand to wet silt filter cake. This decreased the water to cement ratio and improved the processing properties of the mix. The compositions used for these trials are shown in Table 3. Batch mixing used a Hobart N50 mixer. This was sufficiently powerful to blend the binders and silt filter cake with similar consistency to clay. Each batch weighed 300g including the water content of the silt. Water was added on an empirical basis, until a consistency suitable for extrusion was achieved. Assessment of the suitability for extrusion was based on whether or not the mix could be easily forced through a 6.25mm sieve.


0

5

10

15

20

25

30

35

0 1 2 3 4% PVA (PC replacement)

14 day cure

28 day cureUC

S (

MP

a)

Once mixed, cylindrical compression test samples, 20mm diameter and approximately 20 mm high, were moulded and cured at room temperature (25ºC and 60% RH). Unconfined compressive strength (UCS) was determined using standard compression testing equipment with an Automax 5 series controller (Contest Instruments). The load rate was 300 kPa/s and samples were tested after curing for 14 and 28 days. Table 3. Compositions of mixes used for preliminary trials. 3.1.2 Results The results of UCS testing are shown in Figures 3 and 4. It is clear that at high levels of binder addition the materials are stronger, as would be expected. The low result at 2% PVA is probably due to excess water addition during mixing. At higher cement contents the water to cement ratio is higher than for lower cement ratios which contributes to increased strength. Modest polymer additions produce a marked change in properties, and replacement of 0.5% PC with PVA produced an increase in strength.

0

5

10

15

20

25

30

35

40

0 20 40 60

Cement content (%)

UC

S (

MP

a)

14 day cure

28 day cure

Figure 3. Variation of UCS with PC content

Figure 4. Variation of UCS with PVA content in silt, sand, PC and cement mixtures (80% silt/sand content, PC

content was difference between 20% and PVA content).

Wet silt (%)

Dry sand (%)

PC (%)

PVA (%)

Total water content (%)

68 27 5.0 0.0 23.9 61 24 15.0 0.0 24.8 57 23 20.0 0.0 23.9 50 20 30.0 0.0 24.8 36 14 50.0 0.0 22.2 57 23 19.5 0.5 24.8 57 23 19.0 1.0 20.3 57 23 18.0 2.0 22.0

57 23 17.0 3.0 19.8

SID 5 (Rev. 3/06) Page 10 of 25

PVA is an adhesive and therefore binds individual particles together. It absorbs some of the water from the mixture reducing the effective water to cement ratio and this will increase the strength of the material. PVA is also a rheological aid that changes the flow characteristics of the material. This allows particles to compact more closely together during processing, resulting in higher density and improved mechanical properties. 3.2 Further process development and material characterisation The preliminary study showed that interesting materials could be produced with addition of PVA. Further research was necessary to establish the range of properties that could be obtained with the PC/PVA/ silt system. The sample preparation was the same as in the preliminary study and the compositions investigated are given in Table 4. Table 4. Compositions used in the secondary study.

Sand/silt

(%) Sand/silt

ratio PC (%)

PVA (%)

Total water Content (%)

90 0.4 10 0 20.6 90 0.4 9.5 0.5 20.6 90 0.4 9 1 18.5 90 0.4 8 2 17.7 85 0.4 15 0 21.9 85 0.4 14.5 0.5 21.4 85 0.4 14 1 17.8 85 0.4 13 2 17.3 80 0.4 20 0 21.2 80 0.4 19.5 0.5 22.5 80 0.4 19 1 17.4 80 0.4 18 2 17.4 85 0 14 1 23.5 85 0.1 14 1 21.3 85 0.2 14 1 19.6

3.2.1 Test methods Each batch was tested using a standard cone penetrometer with a 35mm, 30º cone. The cone was raised 10mm above the surface of the cylindrical test sample and then released. The depth of penetration gives an indication of the ability of the material to be processed by extrusion. A penetration depth of 2-5mm was found to indicate a mix suitable for extrusion. Density was measured on individual samples using Archimedes method. Water uptake was determined from the weight increase between a dry sample and the same sample after immersion in water for 24 hours and surface dried. The average results from 3 samples are reported. UCS was measured as in the preliminary study. Wet strength was measured on surface dried samples that had been submersed in water for 24 hours. In each case, the average of 3 samples is reported. Micro-structural analysis was obtained from gold coated fracture surfaces examined by scanning electron microscopy (SEM, Jeol 5610). Hydration characteristics of the system were determined using isothermal conduction calorimetry (JAF calorimeter, Wexham Developments) at 20ºC. Each sample was hand mixed for 5 minutes to give a degree of comparison similar to machine mixed samples. 3.2.2 Results Cone penetrometer data for different mixes is given in Figure 5. There is considerable variability, but in general the penetration and the ability to be extruded is controlled by the PVA content. Increased water was required for samples with low PVA content in order to produce a dough-like material suitable for sample moulding. This resulted in high cone penetration due to the lubricating effect of water between grains of silt and sand. With increased PVA content reduced water was required. The PC content did not appear to greatly affect cone penetration behaviour as hydration products would not have formed so early after sample preparation.

SID 5 (Rev. 3/06) Page 11 of 25

0

1

2

3

4

5

6

7

5 10 15 20 25

PC (%)

Cone p

enetr

ation (

mm

)

0 % PVA 0.5% PVA 1% PVA 2% PVA

0.00

5.00

10.00

15.00

20.00

25.00

30.00

5 10 15 20 25

PC (%)

UC

S (

MP

a)

2% PVA

0.5% PVA

0% PVA

1% PVA

Figure 5. Cone penetration of mixes prior to curing The data in Figure 6 indicates that increasing sand content increases cone penetration. This may be due to lower adhesion of silt particles and PVA onto sand particles causing a lower viscosity mix. Nevertheless the effect of sand to silt ratio is small compared to the effect of PVA content. No processing difficulties were encountered with any level of sand, indicating that sand may be omitted in future work.

0

0.5

1

1.5

2

2.5

3

3.5

4

0 0.1 0.2 0.3 0.4 0.5

Sand/silt ratio

Cone p

enetr

ation (

mm

)

Figure 6. Cone penetration varying against sand to silt ratio

Samples with 2% PVA and 18% PC had a dry UCS of 26 MPa. This decreased with reduction of PVA and to a lesser extent PC, as seen in Figure 7. In addition, the failure mechanism changed from brittle to ductile with increasing PVA content, reflecting the higher compliance of PVA as can be seen in Figure 8.

Figure 7. Variation in compressive strength (UCS) with PC content and PVA content

SID 5 (Rev. 3/06) Page 12 of 25

0

2

4

6

8

10

12

14

16

18

0 0.1 0.2 0.3 0.4 0.5

Sand/Silt ratio

UC

S (

MP

a)

Dry WetLinear (Dry) Linear (Wet)

Figure 8. Failure modes of compression tested samples. Samples were made with 15% PC replaced with, from left to right, 0% PVA, 0.5% PVA, 1% PVA and 2% PVA.

Figure 9 shows the effect of sand/silt ratio on dry UCS when PVA and PC concentrations were held constant. Increasing the silt content effectively increases the water content which in turn increases the water to cement (w/c) ratio. In concrete systems increased w/c ratio produces higher porosity in the cement matrix surrounding aggregate particles. However, in these trials the silt had a high surface area which suggests the distance between particles is very small. Additionally there is a relatively high w/c ratio which would result in capillary pores greater than 100 μm. It is therefore possible that there is no continuous matrix of cement surrounding each silt particle. If there is no continuous cement matrix the decrease in strength may result from stress concentration effects of the sand particles rather than the effect of w/c ratio.

Figure 9. Effect of sand/silt ratio on UCS for dry and wet samples (1% PVA, 14% PC)

Figure 10 shows wet UCS increased with increasing PVA and PC content. However the UCS was typically 30% of the corresponding dry UCS result. This suggests that there was significant water ingress during submersion which will have softened the PVA and/or provided lubrication between the silt and clay particles. There is also the possibility that increased pore pressure may have contributed to the lower UCS. Density and water uptake were inversely related and independent of PC content. Typical values were 18% water uptake and 1.73 g/cm

3 for samples with 0% PVA and 10% water uptake and 1.89 g/cm

3 for samples with 2%

PVA. The rheology of PVA enabled denser packing of silt, PC and clay particles as reported to occur in the manufacture of macro defect free (MDF) cements. Reduction in the amount of dense sand particles reduced the sample density. Similarly the water uptake increased with reducing density. Typical results for samples with 1% PVA and 14% PC and no sand were 30% water uptake and 1.60 g/cm

3.

In macro defect free (MDF) cements it is reported that PVA molecules present a pathway for moisture to ingress into the material, causing reduced wet strength (1-3). If this were the case then it would be expected that higher PVA contents would lead to a weaker material, but this was not seen in this study. However, as only a maximum of 2% PVA was added it is possible that the clay matrix may allow the permeation of water, reducing the attractive forces between clay particles. Calorimetry results are shown in Figure 11. This shows the heat of hydration in the first few hours for cement paste and cement paste containing silt and PVA (4. Cement paste has a distinct peak, indicating a high level of chemical activity. However, the combination of silt and PVA additions reduces and delays this peak. 85% silt and 2% PVA addition reduces the hydration to a very small peak after about 15 hours, suggesting inhibition of normal cement hydration reactions.

SID 5 (Rev. 3/06) Page 13 of 25

0.00

2.00

4.00

6.00

8.00

10.00

5 10 15 20 25

PC (%)

UC

S (

MP

a)

0% PVA 0.5% PVA 1% PVA 2% PVA

Figure 10. Wet strength variation with PC and PVA content Images obtained using scanning electron microscopy (SEM) of the fracture surfaces revealed a clay and silt matrix bound by finely dispersed strands of PVA. Cement hydration products were difficult to observe as seen in Figure 12a. To investigate the apparent lack of hydration products and the function of each of the system components further, images of separate systems were studied.

0

0.5

1

1.5

2

2.5

3

0 5 10 15 20 25Time (Hrs)

Heat

rate

(W

/Kg)

Paste10% silt + 1% PVA20% silt + 2% PVA85% silt + 2% PVA

Figure 11. Conduction calorimetry data showing the combined effect of PVA and silt on cement hydration. The

cement paste had a w/c ratio of 0.38 for all calorimeter runs and all silt additions were in the wet state. Figure 12b shows that the PC-PVA system produces cement hydration products covered by a fine film of PVA. This suggests that initial hydration occurs in the presence of PVA, but that further hydration is prevented by this film. A strong bond may exist between PC particles and PVA. The PC-silt system consisted of randomly orientated clusters of clay particles but PC or PC hydration products were not evident, suggesting a fine dispersal of cement particles (Figure 12c). Alternatively, strong PC-silt bonds may be present underneath a layer of clay particles with this clay layer fracturing in preference to the PC-silt bonds.

SID 5 (Rev. 3/06) Page 14 of 25

Figure 12. Fracture surface of a) PVA, PC, silt and sand in the ratio 1:14:61:24, 28 day cure, b) PVA, PC and sand in the ratio 1:14:24 plus water to hydrate, 4 day cure, c) PC, silt and sand in the ratio 14:61:24, 4 day cure,

d) PVA, silt and sand in the ratio 1:61:24, 4 day cure.

The silt-PVA system, shown in Figure 12d, consisted of a clay matrix with aligned particles. A PVA film was not visible, suggesting that either the PVA was finely dispersed throughout the matrix or that strong PVA-clay bonds were present below the fractured surface. 3.2.3 Discussion on results of secondary study The dry strengths obtained appear suitable for non-structural applications. For example, the compaction load applied when reinstating an excavation is reported as 0.7 MPa and footings must be designed to withstand a load of 0.2 MPa. Non-structural grades of concrete have strength of 10 MPa which indicates that aggregate manufactured from this material is suitable. The wet strength, however, is approximately 30% of the corresponding dry strength. In some cases this still appears suitable for non-structural applications but this behaviour is not ideal. From the SEM observations it is suggested that the microstructure of PVA-PC-silt consists of clay and silt particle matrix. Within this matrix are regions of clay particles that have been aligned by the adhesive forces of the PVA. There is no evidence of a continuous matrix of PC hydration products. An increase in PVA concentration increases the number of PVA-clay and PVA-PC bonds resulting in increased strength. The PVA-PC bonds may be particularly strong due to a chemical reaction between the two components, as indicated by the calorimetry results. In addition, PVA acts as a rheological aid which allows denser packing of the particles which in turn leads to higher strength. PVA results in increased alignment of clay particles and consequently an increase in density. This has the effect of decreasing the water uptake and would explain why PVA has a greater effect on water uptake and density than PC. The denser packing effect appears to reduce any negative effect of water ingress which may occur along PVA molecules. Sand content has little effect on properties apart from producing a non-permeable particle which decreases water ingress. The wet strength also increases with both increasing PVA and PC due to the increased number of bonds. However, the ingress of water causes the PVA to soften and attractive forces between silt particles to weaken, resulting in wet strengths of approximately 30% of the dry UCS values. 3.3 Effect of different cements A series of experiments were completed using a range of alternative cements. High alumina cement (HAC) was used extensively in MDF cement technology (1). It is approximately 4 times as expensive as PC, but it is useful in certain applications that require fast setting. Magnesium oxide (MgO) cements require lower temperatures to manufacture and are associated with lower carbon dioxide emissions than PC. In addition, MgO based systems are being developed at Imperial College that are reported to absorb carbon dioxide from the atmosphere during

a b)

c d)

SID 5 (Rev. 3/06) Page 15 of 25

0

5

10

15

20

25

30

0 0.5 1 1.5 2 2.5PFA/PC ratio

UC

S (

MP

a)

70 % Silt

80% Silt

90% Silt

curing (5, 6). Cement manufactured from waste basic oxygen converter (BOC) slag and waste gypsum was also investigated as this is claimed to be significantly more sustainable than PC (7). The samples were manufactured and tested in the same way as previous experiments except the curing conditions were modified for the MgO cements. Results are presented in Table 5. The effect of adding pulverised fuel ash (PFA) was also investigated. Figure 13 shows that replacing approximately 50% of the PC with PFA resulted in only a marginal decrease in UCS. The results show that other cements, including low carbon cements and PFA have potential to be used in the PVA silt system. Table 5. Summary of experiments with other cements and results

Parameter

Cement type and quantity (%)

20 HAC 5 HAC 18 HAC 5 MgO

1,

13 PC 18 MgO

1

5 MgO, 13 PC

18 BOC

Silt (%) 57 67 57 57 57 57 57 Sand (%) 23 27 23 23 23 23 23 PVA (%) 0 0 2 2 2 2 2

8 day UCS (MPa) - - - 15.6 13 - - 14 day UCS (MPa) 9.2 1.5 20.2 - - 17.9 - 28 day UCS (MPa) 10.0 1.5 - - - 19.9 27.2

Note.

1 Cured in a carbonating atmosphere of 5% CO2, 26ºC and 90% relative humidity

Figure 13. Effect on strength (UCS) of pulverised fuel ash (PFA) addition to PC-silt-PVA materials. All samples contained 1% PVA.

3.4 Effect of processing conditions Mixing PVA with the silt prior to PC addition appeared to give greater UCS than mixing all components together simultaneously. Samples with 61% silt, 24% sand, 1% PVA and 14% PC had 9 day UCS of 18.6 MPa. A further trial was conducted with samples containing 85% silt, 14% PC and 1% PVA bulk mixed. Samples cured in air had a UCS of 15 MPa, whilst those cured under water for 7 days and then dried had a UCS of 20 MPa. 3.5 Wet strength The most successful method to improve wet strength involved curing in boric acid (Optibor, Borax Europe) to cross-link the PVA. Samples were prepared as previously but were cured in air before being placed in Optibor solution. The results of these trials are shown in Figure 14. Optibor could not be included in the mix as this caused immediate crumb formation due to instantaneous cross-linking of the PVA. Samples placed in Optibor solution before curing in air cracked due to preferential surface shrinkage. Optimisation of the PVA cross-linking process using Optibor solution could form the basis for patentable IPR resulting from this research.

SID 5 (Rev. 3/06) Page 16 of 25

0

2

4

6

8

10

12

Control 3 days optibor 7 days optibor

Treatment

UC

S (

MP

a)

Dry

Wet

Figure 14. Effect of boric acid (Optibor) treatment during curing on the wet strength of PC-silt-PVA samples.

4. Aggregate production The objective of this part of the research was the production at pilot plant scale, testing and micro-structural analysis of rapid setting aggregate particles manufactured by high shear mixing and extrusion of polymer modified cement/waste silt composite aggregate. 4.1 Equipment A Hobart A200N mixer fitted with a “K” type paddle was used for preparing batches for extrusion. The batches were extruded using a purpose built extruder as shown in Figure 15. The extrusion chamber was 75mm in diameter and the extruder plunger was driven by a 100mm pneumatic cylinder with 6.5 bar air pressure. The cylinder was equipped with an adjustable throttle which controlled the air flow into the cylinder and hence the speed at which the extruder plunger advanced in the chamber. At the end of the chamber was a die with 31, 7mm diameter holes. The die was arranged such that when the extrudate had exited approximately 7mm from the die it triggered a light sensor, energising a wire cutter which swept along the face of the die to cut the extrudate into pellets. An Eirich 400mm diameter pan pelletiser was used to “round off” the square edges of the cut pellets. The spoiler blade was removed as it tended to trap material between it and the pan causing formation of high quantities of small and distorted pellets. 4.2 Methods Based on earlier experiments with cylindrical samples a composition of 85% silt, 1% PVA and 14% PC was used to produce appropriate aggregate properties at an affordable cost. 1 Kg batches were processed throughout the aggregate testing trials. The silt was weighed into the mixer and broken up by the blade. The PVA was added and mixed for approximately 5 minutes until the batch had a pasty texture. Approximately 50ml of water was added to ensure the material had appropriate properties to be extruded. PC was added and the mix blended for 5 minutes until a dough-like texture was produced. The correct dough consistency is important because if it is too stiff it cannot be extruded and if it was too soft the extruded pellets would tend to adhere to each other during subsequent pelletising.

SID 5 (Rev. 3/06) Page 17 of 25

Apply incremental load

Load plate

Plunger

Container

Aggregate

Measure displacement

Figure 15. The extruder used for aggregate manufacture.

It was found that when 1 kg of material was insufficient mix to extrude the whole sample. This was because the material compacting to a higher density. The material nearest the die extruded but when compacted, higher density material approached the die the extruder did not have sufficient power. To overcome this, the dough was split into 4 equal sub-samples that were individually extruded. This produced material that could be extruded and pellets that retained shape when cut. The pellets were loaded into the pan pelletiser and the speed set to slow. The pan angle was adjusted so that the pellets were subject to a tumbling action and pelletising continued for 10 minutes. The pelletised aggregate were left to harden in the laboratory for 28 days before testing. 4.3 Testing The British Standard 10% fines test (BS 812-111:1990) was selected as the most appropriate. The 10% fines test results in the load required to produce10% fine material as a result of crushing. This load is known as the ten per cent fines value (TPV). The apparatus had the dimensions suitable for the 10 to 6.3 mm aggregate as describe in the standard. This consisted of a 78 mm diameter container which was filled with aggregate to a height of 65 mm. Testing was carried out using a Mayes 750 kN universal testing machine. The loading rate was set the achieve 50 kN in 600 seconds. Each sample was subject to either 10 kN or 50 kN loads and the load and compaction data were continuously monitored. Compaction data would be useful for applications where the aggregate was being used as unbound granular fill. The testing arrangement is shown schematically in Figure 16.

Figure 16. Schematic of 10% fines aggregate testing arrangement.

The silt aggregate was compared with the aggregate generated from the washing plants which acted as control data. This aggregate had historically been sold but no TPVs were available. Therefore there was no TPV target

SID 5 (Rev. 3/06) Page 18 of 25

0

5

10

15

20

25

30

0 10 20 30 40 50 60

Load (kN)

Dis

pla

cem

ent

(mm

)

0SA10D

0SA50D

10SA10D

10SA50D

100SA10D

100SA50D

value. Blends of silt aggregate and control aggregate were tested dry and in wet conditions after soaking in water for 24 hours. The mixes tested are given in Table 6.

Table 6. Aggregate mixes used in the testing trials

Sample ID Silt aggregate (wt %)

Control aggregate (wt %)

Load (kN) Wet/dry

0SA50D 0 100 50 Dry 0SA10D 0 100 10 Dry

10SA50D 10 90 50 Dry 10SA10D 10 90 10 Dry 100SA50D 100 0 50 Dry 100SA10D 100 0 10 Dry 0SA50W 0 100 50 Wet 0SA10W 0 100 10 Wet 10SA50W 10 90 50 Wet 10SA10W 10 90 10 Wet 100SA50W 100 0 50 Wet 100SA10W 100 0 10 Wet

Each sample was then left to dry in the laboratory for 24 hours, sieved and the portion below 1.7mm weighed. The percentage fines for each load was determined and the TPV estimated by interpolation. 4.4 Results

The curves for the 10kN and 50kN follow the same loading pattern, as seen in Figure 17. It can be seen that increasing the proportion of silt aggregate results in higher displacement and hence the stiffness of the material is lower than the control. The TPV values for the blends are shown in Table 7.

Figure 17. Load-displacement curves obtained during 10% fines aggregate testing.

It can be seen that as the proportion of silt aggregate increases, the TPV decreases. This is to be expected given the results from the loading curves. As the particles are crushed, the depth of the particles decreases and the fines occupy void space in the compacted sample.

Table 7. Estimated TPV values for aggregate blends

Blends Fines at 10 kN (%)

Fines at 50 kN (%)

Estimated TPV value (kN)

0SA10D, 0SA50D 0.5 11.3 45 10SA10D, 10SA50D 1.8 12.8 40

100SA10D, 100SA50D 9.0 24.5 13 Figure 18 shows that the 10 kN and 50 kN loading patterns were identical. For all blends the displacement was higher than the corresponding sample when wet, indicating that the samples had crushed more in the wet state than in when dry. Furthermore, blends with higher silt aggregate content showed higher displacement than blends with high control aggregate. Both these observations indicate a weaker silt aggregate.

SID 5 (Rev. 3/06) Page 19 of 25

0

5

10

15

20

25

30

35

40

0 10 20 30 40 50 60

Load (kN)

Dis

pla

cem

ent

(mm

)

0SA10W

0SA50W

10SA10W

10SA50W

100SA10W

100SA50W

Figure 18. Displacement load curve for wet samples obtained from 10% fines testing.

The TPV data shown in Table 8 indicates higher TPV values occur for blends with higher silt aggregate content. This is an anomalous result that is explained by the observation that when crushed, silt aggregate particles did not fall apart to generate fines. Instead, the aggregate deformed and reconsolidated into other particles that were too large to pass through the 1.7mm sieve. This consolidating effect can be seen in Figure 19 a) and b). Table 8. TPVs of blends of silt and control aggregate

Blends Fines at 10 kN (%)

Fines at 50 kN (%)

Estimated TPV value (kN)

0SA10W, 0SA50W 1.2 10.5 48 10SA10W, 10SA50W 3.6 11.8 41

100SA10W, 100SA50W 15.6 10.0 50 The tests above were repeated with differing PC contents and for samples containing PFA. Testing of the TPV in wet condition resulted in similar deformation and consolidation of the particles. Aggregate made using 20% PC blended with 20% silt aggregate showed similar results to the aggregate made with 14% PC. However, the 100% silt aggregate blend had over twice the TPV, indicating that a small increase of PC could make a disproportionate improvement in strength. a) b)

Figure 19. Consolidation effect of wet silt aggregate particles subjected to load a) without control aggregate, b) with control aggregate.

Samples containing PFA did not perform as well as in the cylindrical sample testing. A reduction in strength of approximately 17% was observed when the composition was changed to 10% PC and 10% PFA. However, the reduction in strength, as measured by TPV, was approximately 67%. The differences in production and testing may account for the different strengths observed. Cylindrical samples were subject to higher pressures during forming and this may have resulted in greater contact between the binders and silt particles resulting in higher strength. Alternatively the geometry of the testing arrangement may have had an effect.

SID 5 (Rev. 3/06) Page 20 of 25

4.4.1 Strength of aggregate in concrete Trials were carried out using the silt aggregate/control aggregate blends in concrete. The mix was designed to generate a C 20 concrete which would have a minimum strength of 20 MPa. The concrete contained cement and sand in the mass ratio 0.6 and water was added to maintain workability suitable for moulding into 100 mm cube samples. The proportions used are shown in Table 9. Each cube was allowed to cure for 2 days in the mould and then released and placed in the curing tank. Unless otherwise stated, all UCS measurements were carried out on wet cubes. The table shows the effects of substituting control aggregate for silt aggregate that had cured for 14 days. There is a steady decline in properties, but at 50% substitution the concrete still meets the 20 MPa concrete design criteria. Table 9. UCS of concrete containing silt aggregate.

Substitution of control aggregate (%)

Water (%)

Control Aggregate (%)

Silt Aggregate (%)

21 day UCS (MPa)

0 8 52 0 44 20 10 41 8 30 50 12 26 19 22 100 16 0 40 16

As the research aimed to develop an industrial process suitable for the type of site where the silt is generated, it was necessary to know how long the aggregate should cure, before being suitable for use in concrete. It was found that a minimum of 7 days curing at room temperature was necessary. In order to investigate the effects of water saturation on concrete made with silt aggregate, a concrete with 50% silt aggregate substitution was produced. The aggregate had cured for 14 days and the concrete was cured for 14 days. The results show that the wet strength of the concrete was approximately 80% of the dry strength. This is far better than for the aggregate alone. However, for comparison, the wet strength of normal concrete would be expected to be around 90% of the dry strength. 4.4.2 Microstructure The microstructure was examined using SEM as above and found to be identical to the microstructure in cylindrical samples. 5. Whole life costing An objective of the project was to specify commercially viable processing plant to produce aggregate from waste silt, identify routes to exploitation and assess impact on Defra policy on aggregate recycling. A quantitative sustainability assessment was therefore completed to provide information for informed policy decisions in this area. Experimental work showed that potentially appropriate methods existed to improve the wet strength of slit derived aggregate. The sustainability assessment has been completed on the assumption that the process would manufacture aggregate with adequate properties. 5.1 Method of assessment In order to ensure that all relevant impacts were included in the assessment and that different options were considered in the same context, the system boundary shown in Figure 20 was used. The system starts with 40,000 tonnes of wet silt and describes two options. The silt aggregate production option involves importing binders and processing to produce 29,800 tonnes of dry aggregate. For the landfill option, the 40,000 tonnes of silt are sent to landfill, but even more excavated material must be imported and processed to generate the 29,800 tonnes of aggregate, which generates even more silt. The impacts of both processes have been considered.

SID 5 (Rev. 3/06) Page 21 of 25

Figure 20. A schematic of the system boundary used in the sustainability assessment. The data collected was incorporated into a spreadsheet model. This was used to include a range of values for certain factors and to provide results in terms of future scenarios. 5.2 Factors and assumptions 5.2.1 Economic Experiments found that 1% PVA and 14% PC resulted in a potentially viable aggregate. The grade of PVA used costs £2.50/kg for the quantities likely to be used in the process. Experiments with PVA of the same grade but with larger particle size were quoted at £1.50 /kg. The cost of PC is ~ £112/tonne but project partners report construction companies purchasing PC at ~£60/tonne. The water content used was 30%. This is relevant because the silt aggregate dries to produce a product with a lower weight than the mixed components. The cost of a mixer with sufficient power, based on 90 watts/kg delivered to material, was quoted at £125,000 for a Winkworth RT2500 mixer/extruder. The mixing capacity is 2500 litres which equates to 5000 kg batches which should be mixed within 15 minutes. An 8 hour day for 250 days a year equates to processing 40,000 tonnes of material per year. Trials indicated that an extruder with a chamber pressure of > 6 MPa but < 50 MPa was sufficient to extrude the material. This could be achieved with a gear pump and the cost has been assumed to be £11,000. The extruded pellets would then be processed through a pan pelletiser. Industrial experience indicates that a pelletiser sufficiently large to process 40,000 tonnes costs ~ £135,000. It was assumed that one person would be employed at a cost of £30,000. Maintenance costs were estimated at 7.5% of capital investment, increasing every year by a further 2.5%. The expected lifetime of the capital investment was 5 years, so maintenance at the end of this period was 17.5%. Depreciation was calculated linearly, decreasing capital investment to zero over 5 years. Power for the plant was estimated at 2% of operating costs. The economic assessment was carried out using net present value (NPV) analysis at a discount rate of 3.5%. The selling price of silt derived aggregate from the process was assumed to be equivalent to the current price of aggregate generated from the site (9 - 15 £/tonne).

5.2.2 Landfill option Total disposal costs, including haulage, landfill tax and landfill gate fees, were reported by industrial partners to range from 20 - 24 £/tonne waste. Processing of excavated waste to produce 29,800 tonnes of aggregate costs 4.14 £/tonne of aggregate and generated more silt at a rate of 20% of input material. This was offset by a gate fee ranging from 1 - 4 £/tonne. The discount factor for NPV calculations was 3.5% as in the silt aggregate option. It was noted that some aggregate washing plant operators dispose of silt on site. However at some time in the future this will probably not be possible. The options would then be to dispose of silt at other landfill sites incurring costs, or cease operating with proportionate loss of revenue. 5.2.3 Environmental In the absence of accurate data, the use of 1 tonne of PVA was assumed to deplete fossil fuel reserves by 1 tonne. Fuel for materials transport (silt, PVA, PC, excavated wastes - quantities given in economic data) was calculated using the following formula: Diesel usage x distance travelled x total quantity of x density of diesel Lorry weight material transported

Source of excavated

material

Input PVA, PC

Existing process

Silt agg process

Landfill

29,800 tonnes output at

weighbridge

Key Existing = Silt agg =

Silt

Silt

SID 5 (Rev. 3/06) Page 22 of 25

A 25 tonne lorry was assumed for these operations, with diesel use 0.415 litres/km and the distance travelled was 48-80 km for each round trip. The fuel required to recycle aggregate in the existing process was calculated from literature sources of CO2 emissions associated with recycled aggregate and ranged from 1.25-12.7 kg CO2/tonne aggregate. Using a standard rate of 2.63 kg CO2/litre of diesel and a diesel density of 0.83 kg/litre, fuel use can be calculated.

5.2.4 Mineral depletion The quantity of PC used in the aggregate contributed to mineral depletion in the process. For the landfill option, aggregate will need to be supplied from other sources. This will be from additional excavated soils from development sites. The silt content of these soils has not been included in order to maintain comparability with PC use.

5.2.5 Carbon dioxide emissions Transport of all materials and wastes involves combustion of diesel. The rate used was 2.63 kg CO2/ltr of diesel. This was also used to calculate CO2 emissions from the diesel generators used to power site equipment. CO2 emissions are also associated with PC, resulting from the energy to heat and grind the constituent components and from the chemical conversion of CaCO3 to CaO which releases CO2. 0.87 kg CO2/kg PC has been used in this assessment. It has been suggested that CO2 from the atmosphere will react with hydrated cement to reform CaCO3, reducing the impact of PC. The quantity of CaO in cement means that 0.51 kg CO2/kgPC would be theoretically absorbed by PC containing mixtures. Due to unavailability of data describing CO2 emissions associated with PVA production, a figure of 2 kg CO2/kg PVA was used based on emissions associate with other polymer production.

5.2.6 Waste disposal These values were based on the original 40,000 tonnes of wet silt, plus the extra silt generated from extra excavated soil processing. 5.2.7 Social issues Two social impacts have been considered in this analysis, job creation and road congestion. For this analysis it has been assumed that extra jobs generated from the silt aggregate process, would be offset against job losses from the reduction of haulage of material, resulting in no net creation. Congestion on the roads has been quantified in terms of lorry loads. 5.3 Results Various factors have a range of values which can be used to assess the impact on sustainability. A high value for one factor will have an adverse effect on the relative sustainability of one option and this improves the relative sustainability of other options. For this reason two assessments have been completed. The first is where all the impacts have been given values favourable to silt aggregate production. The second assessment has used values favourable to the landfill option. The values used for the assessments are shown in Table 10 and the results of the assessments are shown in Table 11. 5.4 Policy implications From the assessment the social benefits of adopting silt recycling are clear. Waste minimisation and reduction of landfill reliance are also evident. However, the CO2 emissions due to haulage of silt and extra excavated waste in the landfill option is far less than the CO2 emissions associated with PC use in the silt aggregate. The economic case is also marginal if the silt aggregate route is adopted. Table 10. Values used for each of the assessments

Value description Value favourable for silt aggregate option

Value favourable for landfill option

PVA price (£/tonne) 1500 2500 PC price (£/tonne) 60 111.90

Disposal to landfill costs (£/tonne) 24 20 Gate fees to washing site (£/tonne) 1 4 Aggregate selling price (£/tonne) 15 9

Haulage (round trip) (km) 80 48 Embodied energy of recycled aggregate

(kg CO2/tonnes aggregate) 12.7 1.3

PC CO2 absorption (kg CO2/kg PC) 0.51 0

SID 5 (Rev. 3/06) Page 23 of 25

To make the silt aggregate process more sustainable these issues need to be addressed. The economic sustainability could be addressed in a number of ways. Further research could be carried out using cheaper binders. Alternatively the costs of disposal could increase. Using the model, the disposal costs would have to increase to 39 £/tonne in order for silt aggregate to become favourable. Disposal costs are a combination of haulage, landfill tax and gate fee. There is evidence to suggest that haulage prices are increasing but landfill tax is 2 £/tonne. This would have to increase to approximately 19 £/tonne for the process to be economically viable. Table 11. Comparison of sustainability of silt aggregate and landfill options

Impact Criteria

Favourable to silt aggregate option

Favourable to landfill option

Silt aggregate option

Landfill option

Silt aggregate option

Landfill option

Economic 5 yr NPV (£) -3,095,850 -3,587,061 -7,083,782 -3,032,838

Environmental Fossil fuel depletion

(tonnes/yr) 888 172 885 43

Environmental Minerals extraction (tonnes/yr)

5,600 29,800 5,600 29,800

Environmental CO2 emissions

(tonnes/yr) 2,911 544 5,758 137

Environmental Waste disposal

(tonnes/yr) 0 47,450 0 47,450

Social Lorry journeys

(No./yr) 240 3,388 240 3,388

Another alternative would be to manufacture higher value products. Bricks or lightweight aggregate require similar properties and could be manufactured using similar processing. The price of these products are significantly higher than aggregate, and above the 40 £/tonne necessary to make the process viable. A further advantage of making these products is that they are currently manufactured using high temperature sintering. This may mean that CO2 emissions associated with PC are offset by the environmental benefits. Environmental impacts could also be improved by using more sustainable binders with lower carbon footprint. Experimental work has shown that there is potential to use BOC slag cement. Waste pozzolans such as PFA also have potential benefits. An additional benefit in using binders derived from waste is that the materials are cheaper and a gate fee could be charged to offset other costs. No attempt has been made to determine if the reduced social impacts justify the extra CO2 emissions or whether the reduced wastes disposed offset the marginal economic case. This is because there are no standard methods for doing this. The assessments have indicated the areas that need to be addressed if the silt aggregate process were more sustainable than the current situation. Extent to which the objectives have been met Silt generated in the UK by aggregate washing plants has been fully characterised. Evaluation of the silt-cement-polymer composites has been completed as detailed in section 3. Additionally MgO and high alumina cement (HAC) have been evaluated as reported in section 3.2.5. Pilot scale extrusion facilities were installed at Imperial College London in order to produce sufficient quantities of aggregate for evaluation as reported in section 4 of the report. During the project it became clear that a high shear mixer-extruder would not be available as originally expected. However alternative mixers were found which were suitable for the experimental work (section 4.1) and the contract was amended to reflect this change. Specification of viable processing equipment and specific policy issues have been addressed in section 5 of the report. Reliability of results Results of each objective are discussed in the relevant sections. The methods of obtaining these results have been detailed in the report. All equipment used to obtain the results was operated by appropriately trained personnel. All the results have been presented to, and subject to scrutiny during regular steering committee meetings with project team members, industrial collaborators and the Defra monitor present (8). In addition the project successfully satisfied a UKAS audit team that the research complied with the Joint Code of Practise for Research. Main implications of the research The main implications of the research are that it is possible to manufacture a silt/cement/PVA composite product with strength appropriate for use in construction products. The composites do suffer from lower wet strength but

SID 5 (Rev. 3/06) Page 24 of 25

there are ways of improving this. The sustainability of this material and manufacturing process needs to be improved by using low carbon cements and producing a higher value product. Possible future work Future work should investigate improving wet strength, measuring long term properties of silt/PVA/cement composites and production of higher value products such as bricks or lightweight aggregate using low carbon cements. Actions resulting from the research (e.g. IP, Knowledge Transfer) The early results of the work have been released in the public domain and as such IPR have not been pursued. However, the project has been advised that, should a means of improving the wet strength be developed, then this would be a patentable invention.

SID 5 (Rev. 3/06) Page 25 of 25

References to published material

9. This section should be used to record links (hypertext links where possible) or references to other published material generated by, or relating to this project.

1. Lewis J.A., Boyer M.A., Effects of an organotitanate cross-linking additive on the processing and properties of macro-defect-free cement, Advanced Cement Based Materials, Volume 2, 1, (1995) 2-7. 2. McN. Alford N., Groves G.W., Double D.D., Physical properties of high strength cement pastes, Cement and Concrete Research, Volume 12, 3, (1982) 349-358. 3. Donatello S., Tyrer M., Cheeseman C.R., Recent developments in macro-defect-free (MDF) cements, Construction and Building Materials, Volume 23, 5, (2009) 1761-1767. 4. Snelson D.G., Wild S., O'Farrell M., Heat of hydration of Portland Cement–Metakaolin–Fly ash (PC–MK–PFA) blends, Cement and Concrete Research, Volume 38, 6, (2008) 832-840. 5. Vandeperre L.J., Liska M., Al-Tabbaa A., Microstructures of reactive magnesia cement blends, Cement and Concrete Composites, Volume 30, 8, (2008) 706-714. 6. Vlasopoulos N. and Cheeseman C.R., Use of Magnesium Oxide-Cement binders for the production of blocks with lightweight aggregates, Sustainable construction materials and technologies, Chun, Claisse and Ganjian (eds) Taylor & Francis Group, London, 2007, 287-294. 7. Claisse P.A., Ganjian E., Sadeghi-Pouya H., Site trilas of concrete with a very low carbon footprint, Sustainable construction materials and technologies, Chun, Claisse and Ganjian (eds) Taylor & Francis Group, London, 2007, 11-18. 8. Lupo R., Tyrer M., Cheeseman C.R. and Donatello S., Manufactured aggregate from waste materials, Sustainable construction materials and technologies, Chun, Claisse and Ganjian (eds) Taylor & Francis Group, London, 2007, 763-767. Other output arising from the research: R. Lupo, C.R. Cheeseman, The silt route. Minerals, quarrying and recycling, Official Journal of the Minerals Engineering Society, Vol 36, No 01, 2007, UK R. Lupo. From Silt to stone, CIWM, April 2007, The Chartered Institution of Waste Management, UK T. Vounaki. Beneficial reuse of silt from construction and demolition waste recycling washing plant. MSc Project dissertation (2007) Imperial College London, UK C. Unluer. An experimental evaluation of recycled aggregate concrete from waste silt. MSc project dissertation (2007) Imperial College London, UK

Documents

SID 5 Research Project Final Report - Defra, UK