landfill 2011 seminar, 18 - 20 october, durban, kwazulu-natal

Proceedings of the LANDFILL 2011 Seminar, “Waste Management Facilities – The New Order” 18 – 20 October 2011 Proceedings ISBN Number: 978-1-920017-52-1 Durban, South Africa Produced by: Document Transformation Technologies cc

Keynote Address: Towards Appropriate Contaminant Containment Standards K Legge ............................................................................................................................................. Not available

Composition of Waste Landfilled at Metropolitan Landfills and the Impact of the New Draft Waste Regulations C C Wise, R C Emery and J A Coetzee ............................................................................................................... 1

An Overview of ENER-G Systems Experience with Landfill Gas Utilisation Projects in South Africa D Cornish ............................................................................................................................................................. 11

Climatic Influences on Landfill Gas Extraction, with a Unique Comparison of Typical Performance between Landfills Located in the Highveld and KZN Coastal Regions B J Barratt, B L Jewaskiewitz and B L Mills ........................................................................................................ 16

Buffer Zones: The Long Term Interface L Moodley, R Winn and J Parkin ........................................................................................................................ 24

Asphalt Liners in Landfill Construction T Egloffstein and G Burkhardt ............................................................................................................................ 34

Landfill Gas Extraction Schemes - A Contractor's Viewpoint J Butt ................................................................................................................................................................... 43

Drainage Sand Bulk Sampling a New Lined Residue Disposal Facility I Nyirenda and C van Renssen .......................................................................................................................... 55

Geosynthetic Clay Liners (GCLs) – The Effects of Peel Strength on Performance C MacKenzie and N du Toit ................................................................................................................................ 64

Addressing an Integrated Approach to the use of Alternative Technologies as an Economically Viable Alternative to Landfill C Whyte .............................................................................................................................................................. 73

Denitrification of Treated Landfill Leachate using MBT Waste as a Carbon Source: The Bisasar Rd Test Cells Z Hussain, N Sawyerr and C Trois .................................................................................................... Not available

Why Spend Money on Independent CQA Services during the Installation of Geosynthetic Lining Systems? B L Mills and B L Jewaskiewitz ........................................................................................................................... 83

Economic Comparison of Landfilling with and without Anaerobic Pre-Treatment G Burkhardt and T Egloffstein ............................................................................................................................ 94

Landfill Leachate Treatability L J Strachan, H D Robinson and L R Gravelet-Blondin ................................................................................... 104

Behaviour of Baling-Wrapping MSW Landfill Application in Spain J M Baldasano ................................................................................................................................................... 114

GCLs: The Buffelsdraai Landfill Experience G J Payne .......................................................................................................................................... Not available

Design and Construction of a Deep Upstream Groundwater Cut-off Trench J Shamrock, J Glendinning, J Mzisa and H Janse van Rensburg .................................................... Abstract only

eThekwini Municipality: Electron Road Waste Management Facility Presentation C Wise ............................................................................................................................................... Not available

Mechanical Biological Treatment (MBT) Presentation R Lombard ......................................................................................................................................... Not available

Slope Stability of Liners and Covers Workshop R Thiel ................................................................................................................................................ Presentation

A Sustainable Mineral Barrier Option M Naismith, E Timmermans and H Mulleneers ............................................................................................... 170

09:3

0

10:3

010

:30

11

:05

11:0

5

11:4

011

:40

12

:15

12:1

5

12:5

013

:50

14

:30

14:3

0

15:1

015

:10

15

:50

16:1

0

16:5

016

:50

17

:30

Tea

Keynote Address: Towards Appropriate Contaminant Containment Standards

Kelvin Legge

Composition of Waste Landfilled at Metropolitan Landfills and the Impact o

the New Draft Waste Regulations (CC Wise, RC Emery and

JA Coetzee)

An Overview of ENER-G Systems Experience with Landfill Gas Utilisation

Projects in South Africa (D Cornish)

Climatic Influences on Landfill Gas Extraction, with a Unique Comparison

of Typical Performance Between Landfills Located in the Highveld and

KZN Coastal Regions (BJ Barratt, BL Jewaskiewitz and

BLMills)

Buffer Zones: The Long Term Interface(L Moodley, R Winn and J Parkin)

Lunch Sponsored By Wilson & Pass

Asphalt Liners in Landfill Construction(Dr T Egloffstein and G Burkhardt)

Landfill Gas Extraction Schemes - A Contractor's Viewpoint

(J Butt)

Drainage Sand Bulk Sampling a New Lined Residue Disposal Facility

(I Nyirenda and C van Renssen)

Tea

Geosynthetic Clay Liners (GCLs) – TheEffects of Peel Strength on

Performance (C MacKenzie and N du Toit)

Addressing an Integrated Approach to the use of Alternative Technologies asan Economically Viable Alternative to

Landfill (C Whyte)

Cocktails Drinks Sponsored By Barloworld

La

nd

fill

Inte

res

t G

rou

p S

em

ina

r

18

to

20

Oc

tob

er

20

11

Fin

al P

rog

ram

me

12:5

0

13:5

015

:50

16

:10

18:0

0 +

Se

ss

ion

1

Ses

sio

n C

hai

rman

: Jo

hn

Par

kin

Se

ss

ion

3

S

essi

on

Ch

airm

an:

Geo

ff P

urn

ell

Se

ss

ion

5

Ses

sio

nC

hai

rman

:C

oli

nR

edm

an12

:30

Se

ss

ion

6

Ses

sio

nC

hai

rman

:

08:0

0

09:0

0

Se

ss

ion

4

S

essi

on

Ch

airm

an:

Registration

10:1

0

Se

ss

ion

2

Ses

sio

n C

hai

rman

: L

og

an M

oo

dle

y

DAY 1 : Tuesday 18 October 2011

09:0

0

09:3

0

09:3

0

10:1

010

:30

11

:10

11:1

0

11:5

011

:50

12

:30

13:3

0

14:1

014

:10

14

:50

14:5

0

15:3

0

2011

-09-

13:

Fin

al

Ses

sio

n C

hai

rman

: C

oli

n R

edm

an

12:3

0

13:3

0S

essi

on

Ch

airm

an:

Ses

sio

n C

hai

rman

: 10

:10

10

:30

08:1

0

08:5

008

:50

09:3

0

Tea

Economic Comparison of Landfilling with and without Anaerobic Pre-Treatment

(G Burkhardt and Dr T Egloffstein)

Denitrification of treated landfill leachate using MBT waste as a carbon source: The Bisasar Rd Test Cells

(Z Hussain, N Sawyerr and Dr C Trois)

Why Spend Money on Independent CQA Services During the Installation of Geosynthetic Lining Systems?

(BL Mills and BL Jewaskiewitz)

DAY 2 : Wednesday 19 October 2011

10:3

0

12

:30

eThekwini Municipality: Electron Road Waste Management Facility Presentation

(C Wise)

Mechanical Biological Treatment (MBT) Presentation (R Lombard)

16:0

0

Lunch Sponsored By Aquatan

Design and Construction of a Deep Upstream Groundwater Cutoff Trench

(J Shamrock, J Glendinning, J Mzisa and H Janse van Rensburg)

GCLs: The Buffelsdraai Landfill Experience (GJ Payne)

Landfill Leachate Treatability (LJ Strachan, HD Robinson and LR Gravelet-

Blondin)

Behaviour of Baling-Wrapping MSW Landfill Applicationin Spain

(Dr José M Baldasano)

3 : Thursday ctober 2011

10:1

0

10:3

0 a

12:3

0

13:3

0

ch d By

emele td

08:1

5

10:1

0

Slo

pe

Sta

bili

ty o

f L

ine

rs a

nd

S

lop

e S

tab

ility

of

Lin

ers

an

d

Se

ss

ion

7

S

essi

on

Ch

airm

an:

Pet

er D

avie

s

Se

ss

ion

8

Ses

sio

n C

hai

rman

: P

eter

Dav

ies

DAY 320 Oc

Tea

LuncSponsoreBosch Ste

(Pty) Lt

py

Co

vers

Wo

rks

ho

p

Ric

k T

hie

l

py

Co

vers

Wo

rks

ho

p

Ric

k T

hie

l


COMPOSITION OF WASTE LANDFILLED AT METROPOLITAN LANDFILLS AND

THE IMPACT OF THE NEW DRAFT WASTE REGULATIONS

Wise C C*, Emery R C* and Coetzee J A* * Jeffares & Green (Pty) Ltd. P.O.Box 38561, Pinelands, 7430, Tel: +27 21 532 0940; E-mail: [email protected]

SUMMARY The Draft National Standard for Disposal of Waste to Landfill and the new Domestic Waste Collection Standards aim to divert certain waste fractions from landfill. In order to effectively plan for this diversion, an understanding of the composition of the waste being landfilled is required. Landfill-based waste characterization information has been sourced from Metropolitan municipalities for Johannesburg, Cape Town and Durban and is compared. The percentages of annual landfill airspace consumed by the recyclables, greens and builders rubble are then calculated. The analysis shows that of these portions, greens make up the largest portion of the landfill airspace consumed followed by recyclables (based on volume). The new targets set by the draft waste regulations are then evaluated to estimate the potential landfill airspace savings that would be achieved if these targets were met. 1 INTRODUCTION As it becomes more difficult to find suitable sites and to obtain authorisations for new landfill sites within Metropolitan areas, there is an increasing need to divert waste from landfills in order to extend the life of the existing landfills. Further to this, the new Draft National Standard for Disposal of Waste to Landfill (Government Notice 432 of 2011) and National Domestic Waste Collection Standards (DEA, 2011) require and/or encourage certain waste streams to be diverted from landfill. However, in order to plan for effective waste diversion from landfill, one has to understand the composition of the waste that is actually being landfilled. This paper analyses various studies into the characterisation of waste to landfill in Metropolitan areas in order to establish a baseline from which to plan effective waste diversion. It will also analyse the waste characterisations in light on the new Draft National Standard for Disposal of Waste to Landfill (Government Notice 432 of 2011) in an attempt to determine what affect these regulations will have on landfill airspace savings.

1

2 EXISTING WASTE CHARACTERISATION STUDIES A number of waste characterisation studies have been undertaken within metropolitan areas in the last 10 to 15 years. Most of these have been focused on establishing the amount of recyclables available in waste streams from various income groups in order to determine the viability of initiating recycling programmes. As such these studies take representative samples from various areas (generally domestic waste with some commercial waste) and divide it into its various waste fractions. These include Ingerop (1999), KV3 et al. (2004), Gibb (2008), Ball (2003), Ball (2004), Haultec (1998) and GreenEng (2010). Each of these studies have been useful for their intended purpose, but are generator- based and are thus focused on the breakdown within the waste streams and not the actual quantity that’s is landfilled. There are fewer waste characterisations that attempt to analyse the composition of the waste actually landfilled. This is understandable given the complexity and difficulty of such a task. However, four studies have been identified that are specifically landfill-based characterisation. These are:

1. Ball (2001): Landfill Salvager Survey. Report for the Greater Johannesburg Metropolitan Council.

2. PNDA et. al. (2007): Strategic Road Map Phase 2 for Pikitup. 3. Akhile Consortium (2010): Section 78(3) Analysis for the Solid Waste Department, City

of Cape Town, Status Quo Report. 4. GreenEng / Jeffares & Green (2010): Analysis of the Bisasar Road Landfill weighbridge

data for the sizing of Electron Road Transfer Station.

Of these four, the three most recent studies PNDA et al (2007), Akhile (2010) and GreenEng (2010) were analysed in more detail as they were considered more recent. A summary of the purpose, assumptions and limitations of each of these studies follows. 2.1 Pikitup Strategic Road Map, Phase 2 (2007) This study was focused primarily on an analysis of the waste currently going to landfill with the aim of identifying and quantifying specific streams for diversion. It was a comprehensive study that included a summary of the composition of waste landfilled to Municipal landfills in Johannesburg. The characterisation of the household waste was based on KV3 et al. (2004), while the remaining quantities were derived from weighbridge data. Some assumptions were made with regard to the composition of non-domestic streams (such as illegal dumping and street cleaning), but these quantities are relatively small and the assumptions would not significantly affect the general conclusions drawn. The data used was for the 2005/2006 financial year. 2.2 Section 78(3) review for the City of Cape Town, Status Quo report This study was also focused on the waste being landfilled at Municipal sites in an attempt to quantify the waste fraction that could be diverted for beneficiation. The characterisation of the household waste was based on Gibb (2008), while the other fractions were derived from weighbridge data. It should be noted that the definition of recyclables in the household waste characterisation study included all plastic and paper even if it was contaminated. This study also looked at hazardous waste, which was quantified based on an analysis of the weighbridge data (for commercial and industrial hazardous waste) and Gibb (2008) for household hazardous waste. The quantities were determined for the 2008/2009 financial year. 2.3 Bisasar Road Landfill Waste Characterisation, Durban Solid Waste This study was undertaken by GreenEng for Jeffares & Green (Pty) Ltd in order to accurately determine the size of the new Electron Road Transfer station. It looked at weighbridge data from 2006 until 2010 and established an average for each of the major waste streams. It was

2

however not the intention of the study to break down the household waste stream further (i.e into recyclables, garden etc.). 3 SUMMARY OF RESULTS FROM EXISTING LANDFILL-BASED

CHARACTERISATION STUDIES The results from the three different landfill-based waste characterisations are summarised and presented here. 3.1 Pikitup Strategic Road Map, Phase 2 (2007) The Strategic Road Map study categorised the waste in the landfill, based on the collection mechanism. The following categories were thus used; Collection Vehicles, Street Cleaning, Illegal Dumping, Drop offs, Informal Settlements, Bulk, Builders Rubble and Greens. However, each of these categories was divided into various fractions such as recyclables, garden refuse, putrescibles etc. This data was thus collated to provide an estimated waste characterisation of the waste landfilled for the 2005/2006 financial year, as shown in Table 1. The report at the time allowed for 15% greens in the household waste. However, since then greens are not allowed to be disposed of in wheelie bins. The data was thus adjusted to only allow for 2% greens in household waste, assuming some residents will “hide” greens in their wheelie bins.

Table 1: Estimated waste characterisation (by mass) for waste landfilled in Municipal landfill sites in Johannesburg for the 2005/2006 financial year (adapted from PDNA et al. 2007)

Annual Mass Landfilled

(t/yr x 1000)

Recyclables 206.1 13.4%

Putrecibles 126.1 8.2%

Greens and Garden 166.0 10.8%

Fines/residue 225.3 14.7%

Misc. & other 351.3 22.9%

Compacted bulk 37.2 2.4%

Builders' rubble 420.8 27.5%

Total 1533.0 This data was compared with the Pikitup IWMP which showed a greens percentage for 2009/2010 as 9% of the total waste landfilled, which compares well with the data in Table 1 as it allows for some greens diversion that has been taking place.

3.2 City of Cape Town Section 78(3) Review (2010) This study characterised the waste landfilled on Municipal Landfill sites according to the new Waste Act classification system. There was therefore a defined split between hazardous, inert and general waste. General waste was further split between domestic and commercial/industrial. A summary of the estimated waste characterisation as reported in that study for the 2008/2009 financial year is provided in Table 2. It must be noted that this study indicated that this data should be used with caution as the basic underlying assumptions were applicable for the purposes of that study and may not apply for other purposes.

3

Table 2: Estimated waste characterisation (by mass) for waste landfilled in Municipal landfill sites in Cape Town for the 2008/2009 financial year (Akhile et al. 2010)


(t/yr x 1000)

General 1,444 92%

Household (incl. informal settlements) 571 36%

Recyclables 339 22%

Food 84 5%

Other 148 9%

Street and area cleaning (excl. Builders rubble) 94 6%

Trade (commercial & industrial, excl hazardous) 301 19%

Inert Waste 479 30%

Builders rubble 349 22%

Greens directly to landfill as free waste 31 2%

Household greens (incl. informal) 99 6%

Hazardous 132 8%

Household 41 3%

Industrial 66 4%

Business & Commercial 25 2%

Total 1,577 3.3 Bisasar Road Landfill Waste Characterisation (2010) This study was based only on the weighbridge data and was therefore a course estimate than the previous two studies. Nonetheless, some of the comparisons are useful for further analysis. The categories that were used in the characterisation were based on the classification at the weighbridge, e.g. Durban Solid Waste vehicles, builders’ rubble, mixed loads etc. The average for the 4 years, 2007 until 2010 are presented in Table 3. Table 3: Average waste characterisation (by mass) for waste landfilled in Bisasar landfill site in

Durban for the period from 2007 until 2010 (GreenEng et al. 2010)


(t/yr x 1000)

DSW Loads 417.8 38.0%

General Solid Waste 69.7 6.3%

Garden Refuse 36.6 3.3%

Builder's Rubble 107.6 9.8%

Mixed Loads 13.4 1.2%

Sand & Cover Mat. 418.8 38.1%

Tyres 1.0 0.1%

Light Refuse 0.2 0.0%

Other 33.7 3.1%

Total 1098.8

4

4 COMPARISON As can be noted, the categories used for the waste characterisation were different in each of the studies, which made a direct comparison difficult. In particular, the general waste fraction for the Durban data was not further categorised into recyclables and putrescibles. Thus, only the Johannesburg and Cape Town data could be compared by grouping the various waste fractions into 5 broad categories, namely; recyclables, non-green organics (putrescibles, food and other organics excluding greens), builders rubble (which included sand), greens (garden refuse and other greens from parks etc.) and other. This was considered sufficient for the purposes of this study as they represent the major waste fractions that would be diverted from landfill. The comparison of the waste categorisation for Johannesburg and Cape Town are presented in Table 4.

Table 4: Average waste characterisation (by mass) for waste landfilled in Johannesburg and Cape Town

Cape Town (2008/2009)

Johannesburg (2005/2006)

Annual Mass Landfilled (t/yr x 1000)

Annual Mass Landfilled (t/yr x 1000)

Recyclables 339.2 21.5% 206.1 13.4%

Non-green organics 152.3 9.7% 126.1 8.2%

Greens and Garden 129.9 8.2% 166.0 10.8%

Builders' rubble 348.8 22.1% 420.8 27.5%

Other 606.5 38.5% 613.9 40.0%

TOTAL 1576.7 1532.9 The differences between the data for each of the waste fractions is discussed briefly.

4.1 Recyclables Although it may appear that Cape Town generates more recyclables, the reason for the higher apparent figure is that the methodology used in the household waste characterisation study (Gibb 2008) included portions of the fines, which appeared to be paper or plastic, in the paper and plastic categories. As such the amount of recyclables tended to be over-estimated. 4.2 Non- green Organics The percentage of non-green organics are relatively similar with Cape Town at 9.7% and Johannesburg at 8.2%. It must however be noted that the Cape Town figure excludes organic (non-hazardous) special wastes (approx. 37, 000 tons/year) that is landfilled at the Vissershok H:h site. That same waste fraction in Johannesburg would be landfilled at one of the private sites and was thus taken out of the analysis for comparison. It should be noted that this fraction however has a high diversion and beneficiation value. 4.3 Greens and Garden Waste It would appear that Johannesburg (10.8%) landfills slightly more green waste than Cape Town (8.2%). This could however be attributed to the Johannesburg data being slightly older (2005/2006) and thus does not make allowance for the diversion measures that Pikitup have subsequently initiated. For example, Cape Town’s 2008/2009 figures show that they are diverting approximately 67,000 tons/year of green waste by means of a chipping contract, representing a diversion of 35% of the estimate mass of green generated.

5

4.4 Builders Rubble The mass of builders rubble tends to vary quite significantly from year to year as the building industry fluctuates. Thus a comparison over different financial years is not ideal. However, the difference between mass of builders rubble landfilled in Cape Town (22.1%) and Johannesburg (27.5%) is considered to be relatively close given that the figures were recorded two years apart.

5 AIRSPACE CONSUMED BY WASTE CATEGORIES Although most waste characterisation is normally undertaken by means of mass, the volume that the waste consumes in the landfill is more relevant when it comes to determining the life of the landfill. The waste characterisations, as shown in Table 4 were converted to estimated landfill volumes by means of the average landfill densities shown in Table 5, obtained from de Wit (2009).

Table 5: Average landfill compaction densities for various waste fractions (de Wit, 2009)

Density

kg/m3

Glass 830

Plastic 326

Builders Rubble 1502

Paper 606

Cans 200

Greens 200

Wet waste (food etc.) 900 The landfilled density of all other wastes was assumed to be 800 kg/m3. Using the household recyclables breakdown from Gibb (2008) and KV3 (2004), the weighted average density of recyclables on landfill was calculated as 512 kg/m3 for Cape Town and 540 kg/m3 for Johannesburg. The percentage of each waste fraction landfilled by volume could thus be calculated and is shown along with the percentage by mass for Cape Town and Johannesburg in Figures 1 and 2.

Figure 1: Estimated waste characterisation (by mass and volume) for waste landfilled in Municipal landfill sites in Cape Town for the 2008/2009 financial year

6

Figure 2: Estimated waste characterisation (by mass and volume) for waste landfilled in Municipal landfill sites in Johannesburg for the 2005/2006 financial year

5.1 Greens The most striking result is that greens and garden refuse, although relatively small with regards to mass, make up a significant portion by volume. As such it is evident that, even though the figures for Johannesburg be an over estimate for the current situation (due to the data preceding some waste diversion initiatives), there is no doubt that the diversion of greens would not make a significant impact with respect to mass, but would make a far-reaching impact on extending the lifespan of a landfill. 5.2 Builders Rubble Builders’ rubble, although making up a large portion with respect to mass, has less on an impact on the landfill airspace consumed. Certainly there is significant merit in diverting builders’ rubble from landfill sites, but that will not have as significant an impact on the life of the landfill as would greens. 5.3 Recyclables Recyclables are generally considered a prized waste stream for diversion, primarily because of the value attached to that waste. This data also shows that recyclables can contribute to a relatively large portion of the annual landfill airspace consumed. 5.4 Tyres (supplementary analysis to graphs in Figure 1 and 2) Waste tyres has for many years been considered a significant problem on landfill sites. The Bisasar Road data (see Table 3) recorded the mass of waste tyre separately and was thus used to determine the percentage of the annual airspace that is consumed by tyres. The bulk density of waste tyres has been previously reported to be between 350 kg/m3 and 500 kg/m3 (Wallingford, 2005). For the purposes of this study it was assumed the no waste collects between the tyres on a landfill and thus a conservative landfill density of 350 kg/m3 was used. In 2007, 1,777 tons of tyres were disposed of on the Bisasar Road landfill. Using the landfill densities in Table 5, the percentage by volume of the total airspace consumed by tyres was 0.4%. Furthermore, the South African Tyre Recycling Process Company (SATRP) reports that the larger Cape Town area produces about 12,000 tons of waste tyres per year, of which 3,000 tons per year are recycled. Assuming the remaining 9,000 tons are landfilled, this figure would constitute 1.0% of the City’s landfill airspace. Thus, the data appears to indicate that waste tyres do not consume as much landfill airspace as is commonly thought.

7

6 NEW DRAFT WASTE REGULATIONS AND STANDARDS 6.1 Greens The Draft National Standard for Disposal of Waste to Landfill (Clause 5(1)) has set targets for the diversion of a number of particular waste streams. The majority of these are hazardous waste, with the exception of tyres and greens. The effect of these targets on future landfill airspace was calculated based on the results of this estimated waste characterisation. A proposed target has been set to divert 25% (from baseline) of garden waste in 5 years and 50% in 10 years. Using the data presented in Figures 1 and 2, the impact that this would have on landfill airspace for Cape Town and Johannesburg is shown in Table 6. Table 6: Estimated landfill airspace savings (by volume) if Draft National Standard for Disposal

of Waste to Landfill targets are achieved for diversion of Green waste

Landfill Airspace Savings

Greens Diversion Rate Cape Town Johannesburg

25% (5 year target) 6.6% 8.6%

50% (10 year target) 13.1% 17.3%

It can thus be estimated that if the 10 year target for greens is achieved (50% diversion) then landfill lifespans could potentially be extended by between 13% and 17%.

6.2 Recyclables The Domestic Waste Collection Standard (DEA, 2011) states (Clause 4.1) that “All domestic waste must be sorted at source (i.e. the -households) in all Metropolitan and secondary cities”. There is thus strong pressure for Metropolitan municipalities to initiate a system to divert recyclables from landfill. Although many studies calculate the financial viability of at-source separation and recycling initiatives based on the income from sales of recyclables, a very important consideration is the financial benefit of landfill airspace savings. The landfill airspace savings for Johannesburg and Cape Town were estimated assuming that either 25% or 50% of the recyclables are diverted from landfill. This is shown in Table 7. Table 7: Estimated landfill airspace savings (by volume) for 25% and 50% recyclables diversion

Landfill Airspace Savings

Recyclables Diversion Rate Cape Town Johannesburg

25% 6.7% 4.0%

50% 13.4% 7.0% It can thus be estimated that a diversion of 50% of the recyclables (which is relatively difficult to achieve) would result in extending landfill lifespans by between 7% and 13%. 6.3 Summary Based on the estimates presented here, a 50% diversion of green waste has almost twice as much impact on extending the life of a landfill than if 50% of recyclables can be diverted. One has to also consider how easy it would be to divert greens as opposed to recyclables. The breakdown of the green waste sources for Cape Town and Johannesburg is given in Table 8.

8

Table 8: Estimate of sources of green waste (by mass) for Cape Town (8,2% of all waste landfilled, see Table 4) and Johannesburg (10,8% of all waste landfilled, see Table 4)

Cape Town (2008/2009)

Johannesburg (2005/2006)


(t/yr x 1000) Annual Mass Landfilled

(t/yr x 1000)

Domestic Collections 99.1 76% 12.7 8%

Drop offs 67.7 (already diverted) 82.8 50%

Directly 26.3 20% 66.0 40%

Other 4.5 3% 4.5 3%

Total 129.9 166.0 Diverting green waste from domestic collections tends to more difficult as one has to provide a suitable attractive alternative method of disposal of garden waste to prevent residents putting greens in wheelie bins. However, the greens delivered to the drop-offs and landfills are much easier to divert as it is generally separated and this should be the initial target for municipalities. If one were to look at the Johannesburg scenario, a 50% diversion of green waste could easily be achieved by diverting the greens at drop-offs and direct deliveries to the landfill. It would however be more difficult for Cape Town to achieve a further 50% diversion rate without having to target the wheelie bins. It is thus important that DEA define the baselines from which the targets are to be achieved, i.e. is the baseline greens currently landfilled or greens generated? 7 CONCLUSIONS The following conclusions can be drawn from this short study (note that this paper does not take into account the financial implications where a full cost determination would be necessary, in meeting the regulatory requirements or targets for diversion): 1. A consistent method of classifying waste at landfills is required and the implementation of the

New Waste Classification will greatly assist in this regard. Furthermore, a clear definition of what constitutes each of the waste fractions (e.g. recyclables, greens etc.) is also required;

2. Although greens don’t make up a large portion of the waste landfilled by mass, this fraction consumes a very significant portion of the landfill airspace (by volume). Thus the diversion of greens should be considered a priority;

3. Builders’ rubble, although generally easier to divert from landfill, does not offer the same landfill airspace savings as greens (it should also be noted that builders rubble is often used for daily cover material at landfill sites);

4. Recyclables consume enough of the landfill airspace for it to be worth diverting. The emphasis on source separation of recyclables in the National Domestic Waste Collection Standards is therefore justified in terms of extending the life of landfills;

5. If the targets set by the Draft National Standard for Disposal of Waste to Landfill for the diversion of greens are met, the lifespan of landfills can potentially be increased by between 13% and 17%;

6. Diversion of recyclables tends to result in less landfill airspace savings compared with the diversion of a similar percentage of greens;

7. Greens delivered to landfills and drop-offs for Johannesburg tend to make up more than 50% of the greens generated and should be the target for achieving the targets set in the Draft National Standard for Disposal of Waste to Landfill;

8. Clarity is required as to whether the baseline for measuring diversion of greens is based on the amount of greens landfilled or amount of greens generated.

9

REFERENCES Akhile Consortium (2010): Section 78(3) Analysis for the Solid Waste Department, City of Cape Town, Status Quo Report. Ball J. and Associates (2001). Landfill Salvager Survey. Report for the Greater Johannesburg Metropolitan Council. Ball J. & Associates (2004) Integrated Waste Management Strategy for the Rustenberg Local Municipality: Waste Stream Analysis (WSA). Report for Rustenberg Local Municipality. Department of Water Affairs (2011) National Environmental Management: Waste Act, 2008 (act no. 59 of 2008) National Domestic Waste Collection Standards (Government Gazette No. 33935, 21 January 2011). Department of Water Affairs (2011) National Environmental Management: Waste Act, 2008 (act no. 59 of 2008) Draft National Standard for Disposal of Waste to Landfill, Notice 432 of 2011. De Wit MP (2009) Costing the Integrated Waste Management Bylaw with specific reference to airspace savings. GreenEng (2010) Advanced Integrated Solid Waste Management System for uMgungundlovu District Municipality: Waste Characterisation Study Report GreenEng / Jeffares & Green (2010): Analysis of the Bisasar Road Landfill weighbridge data for the sizing of Electron Road Transfer Station, Report for Durban Solid Waste. Haultec (1998) Waste Stream Analysis Report for Durban Metro HR Wallingford (2005) Sustainable Re-use of Tyres in Port, Coastal and River Engineering, Report SR 669. Ingerop (1999) Swartklip Refuse Transfer Station: Waste Characterisation Study, Report for the Cape Metropolitan Council PNDA et. al. (2007): Strategic Road Map Phase 2 for Pikitup. http://www.pikitup.co.za/ upload/files/StratRoadMap2of3.pdf

10


AN OVERVIEW OF ENER-G SYSTEMS EXPERIENCE WITH LANDFILL GAS UTILISATION PROJECTS IN SOUTH

AFRICA

David Cornish General Manager of ENER-G Systems, 205 Northway, Durban North, Durban SUMMARY The attached paper has been written with the purpose of sharing a broad outline of the experiences that ENER-G Systems has had over the last six years in the development and running of landfill gas utilisation projects registered under the Clean Development Mechanism allowing them to produce carbon credits. This paper also outlines the processes involved in generating and selling the electrical power from the projects. Introduction The development of landfill gas utilisation projects over the past five years in South Africa was initially viewed as low hanging fruit. Although this may have been true in theory it has proven to be far from what has been experienced in practise. There have however been a number of isolated success stories such as the Durban Solid Waste Bisasar project, the Chloorkop landfill gas CDM project and the Alton landfill gas CDM to electricity project. ENER-G Systems has over the past six years been involved in the development of over ten landfill gas utilisation projects and through this experience it has become evident that this industry is not for the faint hearted or the impatient. Initially the waste industry viewed the development of landfill gas utilisation projects as synonymous with the destruction of methane gas for the generation of carbon credits under the Kyoto protocol with the added benefit of being able to produce some electricity at the same time, the sale of electricity was subject to you being able to secure a buyer for the power. This is however no longer as attractive as it was once thought to be with projects taking on average over 24 months to obtain registration as CDM projects and the majority of registered projects only realising half of their forecasted CER generation. ENER-G Systems has from inception viewed landfill gas projects as primarily energy projects and viewed the generation of carbon credits as a secondary benefit that assisted in making the project viable. It is now far more evident that the waste industry will become synonymous with the production of energy in various forms from the production of process heat to electricity. ENER-G Systems has over the past five years been focused on the potential to develop landfill gas projects as energy projects and as such has been engaging with the Department of Energy and the National Energy Regulator and has been an active participant in the establishment of support mechanisms for renewable electricity generation projects. This is becoming a greater focus as waste to energy plants become more and more a part of the waste management strategy. As the uncertainty around the longevity of the Kyoto protocol and its replacement mechanism increases the importance of support for landfill gas to electricity projects.

11

CDM Registration and CER Issuance Alton is a small landfill site that was closed in 2003 and belongs to the uMhalthuze Municipality. In 2006 the municipality issued a request for proposal to develop the site with the view to making use of the landfill gas for power generation and to have the project registered as a CDM project and generate power. ENER-G Systems was successful in securing the right to develop the project on a build own and operate basis and the following is a brief outline of the projects development to date. In 2007 the Environmental Impact assessment was conducted and the Record of Decision was obtained under the old ECA regulations. This was the first hurdle that then allowed the project to proceed with its CDM registration process. The CDM registration process took place from 2008 through to the first quarter of 2009 and the project was registered as a CDM project on the 18th August 2009. The project was constructed from June 2009 through to commissioning in November 2009 and has been running ever since.

Picture 1 of the Alton landfill site generator

Picture 2 of the Alton landfill site CDM flare.

12

This then marks the date from which ENER-G Systems was able to start accruing CERs (Certified Emission Reductions) from the destruction of landfill gas from the site through the collection of the data from the flare. Once the project has accrued sufficient CERs (typically one year of operations and depends on the number of CERs a particular project produces) then it is able to have the accrued CERs verified and then issued. This is not a straight forward process as the verification process has very little to look at, touch or count in order to verify the CERs. This is because the CERs have been produced by burning the methane content in the landfill gas and this has floated off into the atmosphere. The verification then turns to the data that has been collected as the source clarification for the verification of the CERs and the verification process then sets out to audit this process using the PDD (Project Design Document) as the template against which to measure every action and parameter. The verification process is very difficult to predict and manage as the proverbial goal posts of what is expected from the project in terms of data and how this data is managed and presented and what is viewed as the critical elements or not changes from Verifier to Verifier. Having completed two verifications on the Enviroserv Chloorkop CDM project is has become apparent that the process of ensuring that the system that monitors the quality and traceability of the data provides a great deal of confidence when it comes to verification of the data used to calculate the CERs produced from a project. This along with a thorough understanding of what is detailed in the projects PDD is essential to the successful completion of a verification of a project’s CERs. Verifiers often focus on items that are beyond the scope of the project and this often hampers the issuance of CER’s for project and it is left up to the project participants to understand the verifier’s roles as well what they do in order to drive the process through to successful issuance. Alton Electricity Sales The sale of power from the Alton project was less straight forward than anticipated as the local municipality was not in a position to purchase the power at a premium from the project due to restrictions imposed by the Municipal Finance Management Act so a private purchaser needed to be found. ENER-G Systems engaged BHP Billiton’s Hillside plant which is directly next door to the Alton landfill site with the view to selling the power from the project into the Hillside smelter. It took from June 2007 through to July 2009 to negotiate and conclude the power sale agreement with BHP that also included amending their power supply agreement with Eskom to allow BHP Billiton to buy power from ENER-G Systems and this was no small task as BHP is a key Eskom account who use over 1150MW of power. This then allowed ENER-G Systems to fully commercialise the project and sell both the power and the CERs from the project. This was followed by a two year application process to obtain a power generation and distribution license from the National Energy Regulator of South Africa (NERSA). Currently in South Africa any Independent Power Producer (IPP) is required to obtain a license from NERSA to generate and sell power regardless of the size of the project. Developing and registering a landfill gas utilisation project as a CDM project is only a small component of a successful landfill gas utilisation project. The challenges of monitoring and getting carbon credits verified is often fraught with inconceivable challenges that are often not anticipated and this poses huge challenges to the development of landfill CDM projects. There is also the ongoing challenge of being able to sell the power under the REFIT (Renewable Feed In Tariff) program that is set to be implemented in mid 2011. Electricity Sales from Landfill Gas to Electricity Projects The sale of electricity from any renewable electricity project in South Africa has been fraught with both regulatory and legislative challenges over the last five years not to mention the price barrier to entering into the market and competing with Eskom. Renewable electricity generation projects and in particular landfill gas is not able to compete with Eskom Megaflex rates as the cost and scale of economy of landfill power generation is not able to produce power at Eskom current rates. This is

13

also due to Eskom’s historic low cost to produce electricity were the market is only now starting to realise the real cost of electricity. However when comparing the price that a landfill gas projects need to make them viable and that of what Municipalities are paying for power from Eskom, it is evident that some form of support mechanism is required to stimulate investment in this sector. In order to overcome this issue the National Energy Regulator of South Africa (NERSA) introduced a REFIT (Renewable Feed In Tariff) program that is set to be implemented in mid 2011. This mechanism was established to aggregate the premium cost of power from renewable electricity generation projects over that total generation in South Africa. The process of aggregating the higher cost of renewable electricity from other technologies over the base load generation plants will result in the smallest impact on the overall cost electricity for the end user. This is why NERSA have adopted the REFIT mechanism to support renewable electricity project in South Africa. Landfill gas is one of seven qualified technologies that are in line to receive support under the REFIT program and has been allocated a tariff of 90c/KWh for the first round of REFIT. The challenges are however not limited to securing an allocation under REFIT as the process of concluding the power purchase agreement now involves a number of additional stakeholders and agreement as depicted below in Diagram 1.

Diagram 1 Depiction of the sale of power under REFIT.

The Future of the Waste Industry It has become increasingly evident through the implementation of these projects that there is a natural progression of the waste industry into the energy industry as waste streams are increasingly viewed as alternate and in most cases as renewable sources of energy,. These will increasingly be viewed as a mechanism for achieving environmental compliance through emission reductions and or avoidance and will provide valuable avenues for complying with the Polokwane declaration for waste minimisation. This will provide the waste industry with many new challenges

14

and opportunities as waste services become ever more integrated and the above three factors become increasingly important for the sector. For landfill sites to remain competitive in the ever increasing cost sensitive environment that has a growing focus on sustainability and environmental compliance it cannot leave energy from waste out of the equation for an effective and efficient waste management program. Conclusion With the growing focus on reducing, reusing and recycling waste the option of using waste as an energy source by both municipalities and private waste operators is growing rapidly. This provides a fresh set of challenges for the waste sector with the incorporation of electricity generation as part of a holistic successful waste management strategy. This along with the opportunity to reduce the effects that the waste sector has on green house gas emissions is increasing the complexity of landfill gas projects as well as waste to energy projects. It is evident from more developed waste sectors like Europe that all of these aspects need to considered together and need to work in ever increasing interdependency in order for the industry to adapt to a rapidly changing financial and legislative environment.

15


CLIMATIC INFLUENCES ON LANDFILL GAS EXTRACTION, WITH A UNIQUE

COMPARISON OF TYPICAL PERFORMANCE BETWEEN LANDFILLS

LOCATED IN THE HIGHVELD AND KZN COASTAL REGIONS

BARRATT B J, JEWASKIEWITZ B L and MILLS B L

Envitech Solutions (Pty) Ltd, PO Box 1677, Hillcrest, 3650 e-mail: [email protected], [email protected] or [email protected] ABSTRACT It is well documented that there are many factors influencing landfill gas generation and recovery. Whilst waste composition, density and age, and landfilling practice are variable contributing elements, seasonal climatic variables such as ambient temperature, rainfall, wind and atmospheric pressure play a significant role in landfill gas generation and recovery. The Mariannhill and Weltevreden landfills are GLB+ and GLB- sites located near Durban and Brakpan respectively, and the landfill gas extraction system operating data obtained from each of these sites was used for comparative purposes. The climates at each of these sites are termed KZN Coastal and Highveld respectively. Climatic, or atmospheric, conditions can affect the temperature, pressure, moisture content and oxygen levels within a landfill. Ambient temperature and wind can affect the biological activity within surface horizons of landfills, while atmospheric pressure variations can temporarily influence passive gas emissions by causing pressure gradients. Precipitation, however, is considered to have the greatest influence on landfill gas production by raising moisture contents closer to the optimum for anaerobic decomposition. This study has provided evidence that gas extraction from landfills in both the KZN Coastal and Highveld regions is significantly influenced by seasonal climatic variations. The findings suggest that a combination of atmospheric conditions lead to a distinct seasonal difference in landfill gas generation and recovery at each site, with both sites exhibiting typically higher gas recovery during the relatively warmer, wetter summer months than during the colder, drier winter months. Assuming that the landfills under consideration are representative of other typical landfills located in their respective climatic regions, it was determined that seasonal differences in landfill gas recovery of some 10% and 20% could be expected for a landfill located in the KZN Coastal region and the Highveld region respectively.

KEYWORDS

Landfill gas, seasonal climatic variations, atmospheric conditions, landfill gas extraction, Weltevreden, Mariannhill, Ekurhuleni Metropolitan Municipality, eThekwini Metropolitan Municipality, Durban Solid Waste

16

INTRODUCTION

It is well documented that there are many factors influencing landfill gas generation and recovery. Whilst waste composition, density and age and landfilling practice are variable and significant contributing elements, climatic variables such as ambient temperature, rainfall, wind and atmospheric pressure also play a significant role in landfill gas generation and recovery. Landfill gas is generated by the anaerobic decomposition of organic waste, thus it stands to reason that the climatic variables that promote anaerobic decomposition should result in increased production and recovery of landfill gas. The Mariannhill landfill is a GLB+ site located near Durban, operated by Durban Solid Waste, within the eThekwini Metropolitan Municipality. Whilst active gas extraction has been taking place for some time at this landfill, a Clean Development Mechanism (CDM) landfill gas-to-electricity project was registered and finally commissioned in December 2006. Both flaring and electricity generation contribute to landfill gas combustion as part of this project. For the purpose of this study, the climate at the Mariannhill landfill is termed KZN Coastal. The Weltevreden landfill is a GLB- site located near Brakpan, within the Ekurhuleni Metropolitan Municipality (EMM). This site is one of four landfills that are part of the EMM Landfill Gas Extraction and Combustion CDM Project, which was officially commissioned in May 2008. Currently, only flaring is carried out on this project. For the purpose of this study, the climate at the Weltevreden landfill is termed Highveld. Operating data obtained from each of the landfill gas extraction systems, captured in accordance with CDM requirements, was used for this study. Measurements of gas flows and methane concentration, in particular, have been used in this study for assessing landfill gas recovery (quantity and quality). Data from weather stations at each of these sites has also been obtained for the periods since commissioning of the projects for the purpose of this study. CLIMATIC INFLUENCES ON LANDFILL GAS EXTRACTION Climatic (or atmospheric) conditions can affect the temperature, pressure, moisture content and oxygen levels within a landfill. Lower temperatures may reduce biological activity and higher temperatures may increase biological activity in landfills, however, this is generally limited to the surface horizons as the heat generated by biological activity deeper within the landfill usually counteracts any outside temperature differences (U.S. Army Corps of Engineers, 2008). Atmospheric pressure can temporarily affect passive landfill gas emissions as pressure gradients develop due to the difference in the rate at which pressures in a landfill equalize compared to outside barometric pressures (Gebert et al., 2006). Gebert et al. (2006) found that passive landfill gas emissions could be directly related to atmospheric pressure changes such as diurnal variations, passing of pressure highs and lows as well as changes resulting from auto-oscillation of air, resulting in temporary increases in landfill gas emissions following a decrease in barometric pressure and vice versa. In terms of landfill gas production and recovery, however, the effects of atmospheric pressure changes do not appear to be well researched. Rainfall infiltration affects landfill gas generation through the addition of moisture to the waste decomposition process. This is usually a positive effect as the natural moisture content of municipal solid waste is typically less than the optimum for gas production. However, rainfall can also carry with it dissolved oxygen, which negatively affects the anaerobic conditions within the landfill (U.S. Army Corps of Engineers, 2008). Similarly, wind can carry oxygen into the landfill but as with temperature, this typically affects only the surface horizons. Wreford (1995) found that precipitation is a predominant factor in methane production by enhancing anaerobic conditions, mixing organic matter and nutrients through downward percolation and stimulating bacterial growth, while also diluting metabolic inhibitors. All these influences were found to lead to increased methane production, which was highest after a time lag

17

of about fourteen (14) days. Interestingly, it was also found by Wreford (1995) that the methane to carbon dioxide ratio in the landfill gas showed a strong relationship to cumulative precipitation seven (7) days prior to sampling, especially if heavy rainfall was experienced. Wreford (1995) further found that barometric pressure fluctuations did not appear to have any significant effect on landfill gas production.

WELTEVREDEN LANDFILL: CLIMATE DATA AND GAS RECOVERY Data retrieved from the weather station at the Weltevreden landfill site, dating back to November 2008, comprised monthly rainfall totals, average monthly wind speed and direction, average monthly evaporation and monthly high and low temperatures. Climate data for Johannesburg comprising average maximum and minimum monthly temperatures and average monthly rainfall based on information from the thirty (30) year period 1961-1990 was also retrieved from the South African Weather Service (SAWS) for the purpose of this study. The gas extraction and collection network at the Weltevreden landfill comprises eighteen (18) vertical wells of between 13m and 15m depth and six (6) horizontal wells of between approximately 50m and 150m in length. Although the extraction and flaring of landfill gas has been taking place since May 2008, the gas extraction and collection network (well field) has not been stable for much of this time. Several horizontal wells were commissioned in February 2009 with several more in April 2009. The horizontal wells were closed in October/November 2009 for maintenance and repairs, and were only re-opened in June/July 2010. These changes to the gas extraction and collection network render it very difficult to establish any long-term trends in gas recovery. However, for the period from August 2010 to June 2011, when the gas extraction and recovery network was stable, a clear increase in the average monthly gas flows was observed, from approximately 370Nm3/h in October 2010 to almost 460Nm3/h in April 2011, with a decrease to approximately 420Nm3/h in May 2011. A similar trend in the average methane concentration was observed, increasing from 48% v/v in September 2010 to 55% v/v in April 2011, and decreasing to 52% v/v by May 2011. Analyzing both the weather data from the site and the SAWS averages for Johannesburg, the seasonal differences in rainfall, maximum and minimum temperatures and wind speed are fairly pronounced. The average monthly rainfall is less than 10mm for the winter months of June to August but greater than 100mm for the summer months of November to February. Average daily temperatures range from 4o to 19o during the winter months of May to August and from 13o to 26o during the summer months of December to March. In addition, significantly greater wind speeds are experienced during the months of June through November, peaking in September. Taking all these factors into consideration, one cannot attribute the above-mentioned trends in landfill gas recovery to any particular climatic influence without an in-depth assessment of changes in landfill gas production and recovery over much smaller time scales. However, it certainly appears that the combination of seasonal climatic influences at the Weltevreden site have an influence on landfill gas (and methane) generation and recovery. This is clearly reflected by an effective increase in landfill gas recovery of approximately 90Nm3/h for the warmer, wetter summer months as opposed to the colder, drier and windier winter months. This increase equates to approximately twenty percent (20%) of the maximum April 2011 landfill gas flows recorded. Figure 1, below, illustrates the increase in landfill gas recovery and methane concentration for the above-mentioned period, while also indicating the average monthly rainfall and average daily minimum and maximum temperatures for the same period.

18

0

20

40

60

80

100

120

140

0

100

200

300

400

500

600

Sep-10 Nov-10 Jan-11 Mar-11 May-11

Tem

p. (

oC

), C

H4

(% v

/v)

and

Rai

nfal

l (m

m)

Tota

l G

as F

low

(N

m3/

h)WELTEVREDEN LANDFILL SITE

Average total flow for month (Nm3/h)

Average daily max. temp for month (deg C)

Average daily min. temp for month (deg C)

Average CH4 for month (% v/v)

Average monthly rainfall for Johannesburg (mm)

Figure 1: Landfill gas generation and climate data for the Weltevreden landfill site

MARIANNHILL LANDFILL: CLIMATE DATA AND GAS RECOVERY Data retrieved from the weather station at the Mariannhill landfill, dating back to December 2008, comprised wind speed, air temperature, relative humidity, barometric pressure and rainfall recorded at 10 minute intervals. However, the data was severely fragmented and incomplete, with large periods of no records. The only months for which meaningful data could be extracted were January to March 2009 and September 2010 to January 2011. Rainfall data for the latter period, however, was not correctly recorded due to a blockage in the rain gauge. Climate data for Durban comprising average maximum and minimum monthly temperatures and average monthly rainfall based on information from the thirty (30) year period 1961-1990 was also retrieved from the South African Weather Service (SAWS) for the purpose of this study.

19

The gas extraction and collection network at the Mariannhill landfill comprises seventeen (17) vertical wells of between 8m and 19m depth, nine (9) risers of which only five (5) are operational, and six (6) horizontal wells. Although the extraction and flaring of landfill gas and electricity generation has been taking place since December 2006, like Weltevreden, the gas extraction and collection (well field) network has not been stable for much of this time. Four (4) new vertical wells were commissioned in July 2007 and three (3) horizontal wells were commissioned in July 2009 with another two (2) horizontal wells being commissioned in November 2009. An additional horizontal well was commissioned in May/June 2010. These changes to the gas extraction and collection network render it very difficult to establish any long-term trends in gas production and recovery. Nevertheless, for the period from June 2010 to March 2011, when the gas extraction and collection network was stable, the average monthly gas flows increased from approximately 530Nm3/h in June 2010 to some 590Nm3/h in March 2011, followed by a slight decrease to around 540m3/h in April 2011. The average methane concentration varied between 50% v/v and 55% v/v during this period, but no particular trend was observed. Analyzing both the useable weather data from the site and the SAWS averages for Durban, the seasonal differences in rainfall, maximum and minimum temperatures and wind speed are reasonably apparent. Average monthly rainfall is between 30mm and 60mm during the winter months of May to August but exceeds 100mm for the summer months of November through to March. Average daily temperatures range from 11o to 25o during the winter months of May to August and from 20o to 28o during the summer months of December to March. Wind speeds are also variable, with slightly less windy conditions typically experienced during the months of May to July. As with the Weltevreden landfill, one cannot attribute the above-mentioned trends to any particular climatic influence without an in-depth assessment of changes in landfill gas recovery over much smaller time scales. However, it certainly appears that the combination of seasonal climatic influences at the Mariannhill landfill have a similar influence on landfill gas generation and recovery. This is reflected by the effective increase in landfill gas recovery of approximately 60Nm3/h for the warmer, wetter summer months as opposed to the colder, drier winter months. This increase corresponds to approximately ten percent (10%) of the maximum March 2011 landfill gas flows recorded. Figure 2, below, illustrates the increase in landfill gas recovery for the above-mentioned period, while also indicating the average monthly rainfall and average daily minimum and maximum temperatures for the same period.

20

0

20

40

60

80

100

120

140

160

0

100

200

300

400

500

600

Jun-10 Aug-10 Oct-10 Dec-10 Feb-11

Tem

p. (

oC

), C

H4

(% v

/v)

and

Rai

nfal

l (m

m)

Tota

l G

as F

low

(N

m3/

h)

MARIANNHILL LANDFILL SITE

Average total flow for month (Nm3/h)

Average daily max. temp for month (deg C)

Average daily min. temp for month (deg C)

Average CH4 for month (% v/v)

Average monthly rainfall for Durban (mm)

Figure 2: Landfill gas generation and climate data for the Mariannhill landfill site

PERFORMANCE COMPARISON

The weather data obtained illustrates that the differences between the KZN Coastal and Highveld climates are significant:

The difference between the average daily summer and winter minimum temperatures is some 10o (11o in winter and 21o in summer) for the KZN Coastal region and some 11o (4o in winter and 15o in summer) for the Highveld region;

The difference between the average daily summer and winter maximum temperatures is some 5o (23o in winter and 28o in summer) for the KZN Coastal region and some 10o (16o in winter and 26o in summer) for the Highveld region;

In summer, the difference between the average daily minimum and maximum temperatures is some 7o for the KZN Coastal region and some 11o for the Highveld region;

In winter, the difference between the average daily minimum and maximum temperatures is some 12o for both the KZN Coastal region and the Highveld region;

21

Both the KZN Coastal and Highveld regions receive, on average, in excess of 100mm during the summer months, however, in winter the KZN Coastal region receives an average of 30-60mm per month compared to an average of 4-10mm per month for the Highveld region.

Based on the above, it would be expected that the higher winter rainfall and less severe temperature differences between the seasons would have a moderating effect on landfill gas generation and recovery in the KZN Coastal region compared to the Highveld region. The observations of this study appear to confirm this, indicating that although the landfills in both regions exhibit definite seasonal differences in gas recovery, the difference is some 10% for a landfill located within the KZN Coastal area and some 20% for a landfill located within the Highveld area, assuming that the landfills under consideration are representative of other typical landfills located in their respective climatic regions.

CONCLUSIONS Landfills are not homogeneous and it is extremely difficult to compare two landfills in different regions as there are many variables that contribute towards landfill gas generation and recovery. To compound the matter, landfill gas extraction systems are constantly evolving, with new wells being installed and commissioned as time progresses. In addition, the level of gas recovery from wells located in relatively older parts of the landfill tend to reduce with time, so without continuous gas recovery data for each well, it is difficult to find an extended period of time during which the entire gas extraction and collection system is stable. Nevertheless, the influences of climatic conditions and variations on landfills have been studied at many landfill sites in different parts of the world, and the influences of different atmospheric conditions have been well documented. This study has aimed to determine whether seasonal climatic variations can be expected to affect landfill gas recovery at landfills located in the KZN Coastal and Highveld regions of South Africa respectively, and whether any relative seasonal differences in gas recovery can be attributed to the specific climatic regions themselves. The Weltevreden landfill is a GLB- site located near Brakpan, while the Mariannhill landfill is a GLB+ site located near Durban. Both sites are of similar size, with similar extents of gas extraction and collection systems. However, the climatic regions in which the sites are located are significantly different. A review of the recorded landfill gas extraction operating data obtained from each site has indicated that both sites have been affected by the expansion of the gas extraction systems, as well as variations due to maintenance issues. Yet a period of relative stability of several months was found for each site, which enabled reasonable assessments of seasonal climatic influence to be made and comparisons to be drawn. This study provides an indication that gas recovery from landfills in both the KZN Coastal and Highveld regions is influenced by seasonal climatic variations. The findings suggest that a combination of atmospheric conditions lead to a distinct seasonal difference in landfill gas recovery at each site, with both sites exhibiting typically higher gas recovery rates during the relatively warmer, wetter summer months than during the colder, drier months. It was found that a seasonal difference in landfill gas recovery of some 10% could be expected for a landfill located in the KZN Coastal region, whereas a seasonal difference in landfill gas recovery of some 20% could be expected for a landfill located in the Highveld region, assuming that the landfills under consideration are representative of other typical landfills located in their respective climatic regions. It is noted that the findings of this study are sufficiently encouraging to warrant a more detailed study, with assessment on a real-time basis using data recorded at much smaller time intervals rather than on a monthly average basis.

22

ACKNOWLEDGEMENTS

The authors would sincerely like to thank the Ekurhuleni Metropolitan Municipality and Durban Solid Waste for the use of the gas extraction operating data and for supplying data from their weather stations for this study. REFERENCES

Gebert, J. and Groengroeft, A. (2006) Passive landfill gas emission - Influence of atmospheric pressure and implications for the operation of methane-oxidising biofilters. Waste Management 26, pp 245-251.

Solid Waste Association of North America. (2002) Landfill Gas Operation & Maintenance. Manual of Practice.

U.S. Army Corps of Engineers. (2008) Engineering and Design - Landfill Off-Gas Collection and Treatment Systems. EM1110-1-4016. Department of the Army.

Wreford K. (1995) An Analysis of the Factors Affecting Landfill Gas Composition and Production and Leachate Characteristics at the Vancouver Landfill Site at Burns Bog. A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science. The University of British Columbia, Vancouver, Canada.

23


BUFFER ZONES: THE LONG TERM INTERFACE

Moodley L*, Winn R** and Parkin J*** *Design & Contracts Engineer. eThekwini Municipality, Department of Cleansing and Solid Waste, DSW. P.O. Box 1038, Durban, 4000. [email protected] ** Conservancy Manager. Richard Winn, Environmental Consultant, Landscaper and Nurseryman. P.O. Box 5505, Durban, 4000. [email protected] ***Deputy Head: Plant & Engineering. eThekwini Municipality, Department of Cleansing and Solid Waste, DSW. P.O. Box 1038, Durban, 4000. [email protected]

SUMMARY

The Plant and Engineering Section of the Department of Cleansing and Solid Waste, DSW, had decided to place emphasis in changing the mindset from “hard engineering to green engineering” on the buffer zone areas. Coupled with this, was the fact that modern landfills have a larger buffer zone area which increases the management and maintenance associated with it. Particular focus was placed on setting up a buffer zone model that would be self sustaining in terms of biodiversity control, forestation of living firebreaks and upgrading the green lung for the eThekwini Municipality. This paper presents the buffer zone management tools and environmental management concepts implemented at the Mariannhill and Buffelsdraai Landfill Sites to achieve sustainable buffer zones with minimal effort over time. The paper further highlights the linking of the buffer zone management into the sites’ EMP and creating the basis for securing special long term processes e.g. introduction of red data species, indigenous reforestation, Clean Development Mechanisms (CDM) and climate change mitigation etc. KEYWORDS Buffer Zone Management, Environmental Management, Landfill Rehabilitation, Green Engineering, Sustainability, Reforestation, PRUNIT (Plant Rescue Unit), Biodiversity. INTRODUCTION The Minimum Requirements for Waste Disposal by Landfill was introduced in 1994 by the Department of Water Affairs and Forestry (DWAF) and more recently this responsibility has been handed to the Department of Environmental Affairs and Tourism (DEAT) in order to regulate waste disposal practices and minimise the impacts on the receiving environment. Landfill operators issued with operating permits are continually tasked with ensuring permit compliance to the regulator. Moreover the compliance to the Record of Decision (ROD) and Environmental Management Plan (EMP) for landfill sites further adds to the “compliance stockpile”. As such, the operational management and engineering of nowadays landfills referred to as “waste management facilities” become increasingly challenging. One key area of scrutiny by any regulator, neighbouring community, environmental compliance auditor etc tends to zone on the site’s buffer zone. There can be much debate on the above mentioned environmental regulations and permit requirements but the main objective is ensuring a protected environment. Moreover, emphasis is placed on pollution prevention, environmental degradation prevention, preservation of natural assets and protection to ecological sustainable development (Torr, 2009). The Department of Cleansing and Solid Waste, DSW, which is accountable for the waste management within the

24

eThekwini Metropolitan Area (EMA) is renowned both locally and internationally for an innovative approach to landfill management where waste disposal operations are integrated with landfill rehabilitation during the operational life of the landfill site. DSW uniquely rehabilitates its landfill sites by rescuing indigenous vegetation prior to landfilling from the landfill footprint and buffer zone areas. As a result, an indigenous staging nursery was constructed to stock all indigenous plant sand seedlings not only prior to landfilling but also prior to any earthworks onsite. The rescuing of vegetation approach had led to the creation of the Plant Rescue Unit (PRUNIT) which has shown to be cost effective and environmentally acceptable (Parkin et al., 2006). As a result, DSW is well done the line in mastering the rehabilitation component of the landfill footprint and is now making strides into the long term management of the landfill sites buffer zones. There has been limited research to date on the proper management of South African landfill buffer zones and therefore limited experiences on the tools and conceptual models that can be used to ensure good practice and sustainable long term buffer zones. The interest of this aspect has initiated the Plant and Engineering team of the Department of Cleansing and Solid Waste, DSW, to change the mindset of how to manage and apply naturalistic engineering concepts at the two new aged landfills namely, the Mariannhill Landfill Site and the Buffelsdraai Landfill Site, Kwa-Zulu Natal. This was identified as a vital component in linking the interface between the landfill footprint and the landfill’s buffer zone to avoid future probable pollution into the receiving environment. This paper presents some practical examples of buffer zone management and highlights the start of a long term study of the biodiversity, ecological science, climate change, and reforestation. Therefore comments on preliminary progress achieved through the application of unique management tools and environmental management concepts are highlighted. SITE DESCRIPTION The Mariannhill Landfill Site The Mariannhill Landfill Site is situated outside Pinetown, some 20 kilometres west of Durban and is a permitted GLB+ landfill where landfill engineering methods have successfully combined to realize South Africa’s first landfill conservancy (Conservancy status acquired in August 2002). The Mariannhill Landfill Site, opened in July 1997, was located to text book standards, being well hidden from the public view by the natural topography and well established vegetation (Strachan et al, 2002). The total area of the site covers some 49.5 ha of which 18.5ha is landfill footprint and the remaining 31ha incorporating the buffer zone. The landfill presently receives approximately 750tons/day of MSW (Municipal Solid Waste) which serves for the majority of the western areas of the eThekwini Metropolitan Area (EMA). Based on airspace calculations predicting 4.4 million cubic metres capacity, the site will be operational beyond the year 2024. The Buffelsdraai Landfill Site The Buffelsdraai Landfill Site is situated in Verulam, approximately 50km north of Durban and is also permitted as a GLB+ landfill. The site was a successor to the closed La Mercy Landfill and was commissioned in May 2006 and is positioned out of the public eye on a sugar cane farm. The total area of the site is some 887ha of which 100ha is the actual landfill footprint and the remaining 787ha consists of the buffer zone. The landfill currently receives some 400tons/day of MSW and services predominately the northern areas of the EMA. This daily tonnage is expected to increase to some 3500tons/day post closure of the city’s central landfill i.e. The Bisasar Road Landfill Site by early 2014. Airspace calculations reveal a total of 45 million cubic metres capacity which equates to an operational life of some 70 years.

25

THE BUFFER ZONE MANAGEMENT CONCEPT Landfill engineering nowadays concentrate significantly on the design, operation and aftercare of the landfill footprint and to some extent neglects the required management and due responsibility of the landfill’s surrounding buffer zone. Most landfill operators have a negative perception with regards to the duration of responsibility and size attached the footprint and the buffer zones. Figure 1 below illustrates the conceptual mindset of the interface between the landfill footprint and buffer zone of any landfill as well as duration of responsibility for each area. Figure 1: The conceptual mindset of the interface between the landfill footprint and buffer zone As depicted in Figure 1 above, the required duration of responsibility by any permitted landfill operator is given by the operational life (x) years of the landfill plus a 30 years after care period. In general a minimum value of x is usually in the order of 20 years and as a result the corresponding minimum duration of responsibility is some 50 years and varies up 70 years which is a lifetime commitment. It is further noted that x is also a function of the available airspace but more importantly with the drive seen in the industry for waste minimisation, recycling etc, would directly increase the duration of responsibility. The reality is that there is some 50~70 years responsibility and the mindset change adopted by the Department of Cleansing & Solid Waste – DSW was to accept the long term commitment and apply simple management tools and environmental asset concepts to turn what as previously seen as negatives into positives for a self sustaining system in the long term. The Legislative Framework South Africa has compiled a strong legal framework to ensure that all development is biophysically, socially and economically sustainable (Diederichs and Van Nierick, 2009). A legislative management tool in achieving this is the national Environmental Impact Assessment (EIA) Regulations which provide different activities that are known to have a negative effect on the receiving environment if not managed properly. The role of the EIA is to access if the activity can avoid against such detrimental impacts. The EIA is also used to satisfy the constitutional rights as all South African citizens have a right to live in protected environment which is free from harm to human health (DWAF, 2005). The process evaluates both positive and negative impacts which are presented to environmental decision makers regarding the developments sustainability and resulting acceptability or not. The establishment of the Mariannhill and Buffelsdraai Landfill Sites having been subjected to the full scrutiny of the EIA process ensured some 10 years “interface” with all registered interested and affected parties and the natural progress thereafter with the EIA approval followed by the issue of the Record of Decision (ROD) by the Department of Agriculture & Environmental Affairs (DAEA) and lastly the operating permit for the landfill (now referred to as waste management facilities). From this, an Environmental Management Plan (EMP) was compiled using the generic eThekwini’s EMP with added peculiarities which were unique to the system and situation. Generally the EIA will investigate the science of the EMP and final compilation will be done by the landfill site’s Monitoring Committee. It must be noted that this process is not a fresh/new process as in general the MC is represented by the same interested and affected parties that were registered with the

Buffer Zone Responsibility = x + 30years

Landfill Footprint Responsibility = x + 30years

Where: x = operational life of landfill site

Apply Management Tools & Environmental Asset Concepts to Achieve Long Term Sustainability

26

initial EIA process. At the first meeting of the MC, the EMP was expanded and the baseline Terms of Reference (TOR) for the management of the sites was compiled. The most important part of this process is that the EMP and TOR are living documents and can be added to improve on the sites management which generally includes for all processes that the buffer zone could help in cementing the above mentioned compliance issues. The Plant and Engineering team that represented the operator on the MC created the vision which was later realised by all on the representatives to move into a conservancy committee. The understanding is that there could be pressure from external factors to influence the decisions of the buffer zone and emphasis must be placed on attracting the “correct culture” of partners that would work together in achieving long term management goals with adding value. The buffer zone must be deemed as an environmental asset that attracts positive working relations and possible partners. This process had initially started with good intentions and been able to move with the changing times (PRUnit, Rehabilitation, Clean Development Mechanisms, climate change etc).The sooner the management starts, the quicker the onset of the interface. Waste Management License holders for landfills should ideally own the land within the buffer zone and if not already owned, seek to purchase that land in the near future, or have binding agreements where by the adjoining landowner will not use the land for sensitive land uses within an agreed timeframe. It is from the eThekwini experience, that should the buffer zone not be owned by the license holder then there is too large a risk that any management/investment in this area will have long term sustainability and benefits cannot be enjoyed. As a result, both the buffer zones of the Mariannhill and Buffelsdraai Landfill Sites are owned by the eThekwini Municipality which has ensured controlling interest for long term benefits. MANAGEMENT TOOLS AND INTERPRETATION Landfill engineering should factor in selected environmental management tools which are fundamental for environmental acceptance. Such tools may incorporate the use of simple cost effect yet durable tools/systems that form some component of the landfill. The following management tools were identified by the Department of Cleansing & Solid Waste - DSW:

Rescue and Relocation (Flora & Fauna): All flora and fauna found on the footprint would naturally occur in the buffer zone should be relocated directly to its new long term home in the buffer zone. Emphasis must be placed on moving everything once if possible to guarantee the lowest carbon footprint with minimal disturbance ecologically.

Rehabilitation Nursery Process: Should plants/flora not be able to be moved once, then a nursery process must be utilized. Specialised hardened off species can thereafter be utilized for rehabilitation of the buffer zone.

Endemic Indigenous: By definition, all seed sources must be collected with a 50 kilometer zone for international auditing purposes. The DSW approach only uses seed source from its own landfill sites and therefore has always been compliant – As a result there has been no need to change any nursery processes to fulfill the international standard.

Screening for Aesthetic and Visual: Most ROD’s have such a requirement and therefore this rehabilitation/screening process automatically assists in fulfilling the ROD.

Vegetation for Odour Movement: All landfills have associated odour problems i.e. migration of landfill gas would normally move down the valleys and these drainage lines are rehabilitated as a priority in all biodiversity management processes. Drainage lines are 32m on both sides (NEMA) and therefore erosion control, leachate movement, gas migration and wind scatter can be mitigated by this rehab approach. Scented species can be utilized for natural odour management e.g. heteropexis Natalensis (Natal Lavender).

Wind breaks for Wind Scatter: Final screening from footprint moving outwards into the buffer zone area. It is a long term guarantee that the footprint does not interfere with the buffer zone but does not preclude internal scatter fences.

27

Forestation for Living Fire-Breaks: The mindset is to grow a green firebreak in the form of a forest which cannot burn – long term sustainability (developed within 10 years) rather than cutting as conventionally done which is continually for the duration of the landfill commitment. As a result this application becomes sustainable with minimal efforts that are economically viable.

Closure Rehabilitation: Should the landfill footprint be surrounded by a viable endemic indigenous seed source, would imply that the long term footprint rehabilitation will be enhanced and cost effective. This would ensure enhancement of the buffer zone (best practice in the buffer zone) which would work towards achieving best practice after closure of the footprint.

ENVIRONMENTAL ASSET CONCEPT All environmental processes associated with the buffer zone must be deemed as an asset to start out with the correct mindset. The following environmental asset concepts were identified and implemented by the Department of Cleansing & Solid Waste - DSW:

Rescue and Relocate: All fauna within the footprint and buffer zone must be deemed as an asset in creating management tools for the asset. This is seen as an economical viable solution as the only costs attached are labour and time and no external capital costs for the fauna. An example of implementation at the Mariannhill Landfill Site is the Boma fire protection

i.e. requirement from eThekwini Fire Department which was created using relocation plants/trees from the footprint.

All the initial ROD screening for the Osidiswini Hospital was rescued from the Buffelsdraai Landfill entrance road.

Seed Source: All endemic indigenous species seed sources must be collected irrespective if

the plant/tree cannot be relocated for future use and for fulfilling the requirement off the 50 km zone. Some seeds are propagated directed into the PRUnit process whilst others are stored for future use therefore avoiding purchasing costs and increasing the biodiversity potential. Seed sources collected from both the Mariannhill and Buffelsdraai Landfill Sites are not

only planted on the sites but also used in rehabilitating inherited “old dumps”. The initial seed source and grown on trees for the Reforestation Tree-Preneur process

at Buffelsdraai Landfill Site in collaboration with the eThekwini Environmental Planning and Climate Protection Branch, were provided through the PRUnit process. It should be noted that this process provided the opportunity to “kick-start” this process i.e. some 20 000 trees were utilized which saved an 18month lead up time.

Topsoil: Where possible this should be utilized for instant rehabilitation – forest base to

forest, grassland to grassland. If this is not possible, then it can be used in the nursery mix for potting etc and never be stored in excess of 2m in height so as not to destroy its existing seed source within the stockpile. Mariannhill Landfill Conservancy (MLC) red data species grassland was created simply

by spreading grassland topsoil some 300mm thick and then doing alien plant control until a high quality grassland was re-established.

Buffelsdraai Landfill’s buffer zone planting and the sites road verges were done using rescued topsoil.

Rocks: Rocks salvaged from civil works and blasting can be utilized for road edge protection and creating habitats on the buffer zone interface with the footprint for long term management tools for beneficial wildlife species. Too lessen the potential of rodent increase in this area, snake population should have somewhere to thrive and the habitats created from the rock stockpiles will guarantee a balance in the ecosystem. Further uses could be for landscaping of site entrances, protection of manholes and erosion protection.

28

MLC has a natural rock stockpile between the community and the landfill to lessen rodent migration.

Electron Road Transfer Station Facility has already had its rocks designated for landscaping set aside at Mariannhill Landfill from blasted rock of Cell 4 Phase 3.

The rock in the buffer zone disturbed by the sugar cane farmer has been reinstated to create the same original habitat for rodent control and other habitats. This was placed by the local community under the DSW team’s guidance.

Groundwater: By enhancing the habitat around the riverrine sections of the buffer zone, there

will be an automatic enhancement of the water quality i.e. through filtration, absorption and transpiration. This is not only relevant to the landfill in the footprint in the long term but also for upgrading the original buffer zone use which probably would have had contaminants from “old school” agriculture. Buffelsdraai Landfill had been heavily farmed for sugarcane in excess of 60 years and

therefore the residues of the fertilizer, herbicides, insecticides and ripening agents all had to be dealt with as they slowly leach into the sites groundwater system. A minimum requirement is the 64m drainage line enhancement and temporary process of plugged all old sugarcane from footprint area into drainage lines had an immediate polishing solution with natural reeds for any initial run off before leachate system was in place.

MLC: An upstream man made silt trap and reedbead was created in 2001/2002 for initial desilting and polishing and in 2011 a set of gabion baskets/organic filter have been placed downstream.

Leachate: Treated leachate from the site can be utilized for rehabilitation in the footprint but

the buffer zone interface is always available for initially for rehabilitation and therefore the fire breaks can be irrigated using this long before final foot rehabilitation. MLC: Treated leachate has been used in rehabilitation the buffer zone below the

leachate treatment plant and for Boma firebreak. Buffelsdraai Landfill: Treated leachate has been used for dust suppression and for

watering the screening plants along the road in the buffer and for irrigation on local community boundaries.

Landfill Gas: By nature any landfill gas is heavier than air and as a result migrates down

valley lines making the buffer zone drainage lines habitat upgrade even more important. Included in these plantings must be scented plants to assist in odour control and these plants/trees will break up odour plumes before reaching neighbouring communities. MLC: Retaining and upgrading forests in the closest community areas on the sites

eastern boundary. This is mutually beneficial in also assisting with wind scatter on the landfill. Due to the close proximity of the community the site has an active odour control management system which will probably not be needed at Buffelsdraai as a result of a larger buffer zone.

Buffelsdraai Landfill: The reforestration project has focused initially on riverrine areas and therefore minimizes the potential for gas migration.

Green Waste: Green waste treatment (mechanically or biologically) will provide end uses

within the landfill such as mulching final side slopes, soil amelioration for better rehabilitation and compost for the nursery process. The nursery has provided plants/trees for the buffer zone whilst the mulch is utilized for the understory enhancement of the forest which hasl positively impacted on the biodiversity process. MLC: PRUnit nursery has utilized compost for its bagging process and the mulch was

used for the creation of PEAT in the upper wetland. This enhances water retention, polishing ability and improves plant growth in a short space of time making the wetland able to fulfill its objective quicker.

Buffelsdraai Landfill: All eradicated alien plants are left insitu within the buffer zone to enhance plant growth whilst all seed heads are removed.

29

SECURING SPECIAL LONG TERM PROCESSES / RELATIONSHIPS If best practice is applied to the buffer zone and its interface with the landfill footprint, then opportunities are created for securing long term relationships with various stakeholders of good standing. Biocontrol agents released on the site which are tested (7 years) insect’s specific to the degradation of an alien plant in an area which is supplied by the Department of Agriculture only to areas with a long term management plan for the site. These processes demand long term commitment from the site owner/operator who will gain major benefit if best practice is implemented in relationship to the buffer zone. Table 1 below illustrates the biocontrol release at the MLC. This process commenced in 2003/2004 and has shown initial succession and conclusive results will be forthcoming – normally a 10~12 year process. In the interim the MLC has entomological tours when there are local conferences. Table 1: Biocontrol Release at the Mariannhill Landfill Conservancy, (ARC, 2011)

Weed Degree

of Control*

Biocontrol Agents

Year of Release

Estab- Main Feeding

Guild

Damage to weed*

lished RELEASED IN MARIANNHILL CONSERVANC

Y IN 2004: ? Calycomyza eupatorivora

(Agromyzidae) 2003 Yes Leaf miner Moderate

Science in Progress

Chromolaena odorata

Pareuchaetes aurata aurata

1990 No Leaf

chewer - (triffid weed/ paraffienbos) (Arctiidae)

Pareuchaetes

insulata 1998 Yes

Leaf chewer

Considerable (Arctiidae)

Pareuchaetes pseudoinsulat

a 2001 No

Leaf chewer - (Arctiidae)

Red Data Species/Specialised Habitats: Once habitats are be created that the scientific

authorities accept as viable for release of endangered species, and then higher authorities such the International Union of conservation (IUCN) will favour the project as all red data species come with a full international protocol. MLC has the black headed dwarf chameleon (Bradypodion melanocephalum) has been

relocated to the grassland surrounding the upper wetland in the buffer zone. The science before and after has been completed and the relocation has been deemed to be highly successful as the population has increased from the original 15 to 47 at the last count. Table 2 below shows the initial release of the 15 black headed dwarf chameleon at the MLC.

30

Table 2: Black Headed Dwarf Chameleon Release at the Mariannhill Landfill Conservancy

No. Species Male Female Adult Sub-adult WP1 S WP2 E Location

1 Black-headed X 29°52.863' 30°57.200' on Chromolaena

2 Black-headed X X 29°52.863' 30°57.200' on Chromolaena

3 Black-headed X X 29°52.856' 30°57.198' Turpentine grass





8 Black-headed X X 29°52.863' 30°57.198' on creeper

9 Black-headed X 29°52.863' 30°57.198' on creeper







29°58.749' 30°57.951' Mariannhill Conservancy

Black-Headed Dwarf Chameleon rescue/relocation - Hilltop Housing development

Observation date: 11, 18, July 2007 - Elevation: 75 m ASL

Weather conditions on the above date: warm, cloudy, moderate East, drizzle

Release co-ordinate By enticing red data relocation onto the site, one would find a new set of allies to concur future pressures on the buffer zone such as land invasion or an EIA for a pipeline (refer to Petronet pipeline on buffer zone – IUCN and KZN Ezemvelo Wildlife assisted in preventing this process from moving forward.

Reforestation/Tree-Preneur: Is a process to add value to the management process and satisfy the needs of the ROD, EMP and TOR for the site. The process is basically PRUnit being implemented via the surrounding local community and therefore major job creation can be realized without been included in the landfill footprint which demands high skills and safety levels. The project also addresses social upliftment issues through upgrading and maintaining the sites buffer zones with the potential to expedite the rehabilitation for probable “nature reserve status”.

This also leads to the direction of Carbon Sinks, Clean Development Mechanisms (CDM) and Climate Change processes for the future and the DSW teams are always striving to develop a “green lung” from the landfill footprint and buffer zone interface.

CONCLUSIONS There have been limited South African experiences on the tools and conceptual models that can be used on landfills to ensure good practice and sustainable long term buffer zones. The practical examples and preliminary data presented in this paper underline the unique yet simple management tools and environmental management concepts that can be successfully integrated into any landfill facility. The progress made thus far at the Mariannhill and Buffelsdraai Landfills Site has shown that landfill operators/owners need not be apprehensive of management required for landfill buffer zone but instead adopt a change in mindset by creating a beneficial link between the landfills footprint and buffer zone. The much talked about “interface” is merely ensuring the buffer zone can be an asset for the landfill footprint whereby the rehabilitation process would start from buffer zone moving internally towards the footprint i.e. self seeding potential, processes through reforestation, habitat systems etc will ensure restoration of the footprint to natural

31

environment will minimal efforts in the long term. It is recommended that the implementation of environmental asset concepts highlighted in this paper should only be done provided the buffer zones owned by the landfill operator as this guarantees control and vested interest for long term sustainability. Further detailed information on any particular aspect related to this paper can be reviewed on the following website: www.landfillconservancies.com

Plate 1: Release of Black Headed Dwarf Chameleon into the MLC

Plate 2: Rescue & Relocation of a Cussonia Spicata (Indigenous Tree) from footprint to buffer zone at Buffelsdraai Landfill Site

Plate 3: Community Reforestation Project of Buffelsdraai Landfills Buffer Zone

Plate 4: Rhino Award received at the MLC from Ezemvelo KZN Wildlife (2009/2010)

ACKNOWLEDGMENTS This project is viable with the cooperative assistance between the Department of Cleansing & Solid Waste Department – DSW and Richard Winn Environmental Consultant, Landscaper and Nurseryman. Credit must also be given to the teams “un-sung heroes” associated to the sites success i.e. operational teams, resident engineers, monitoring committees etc and the author commends Richard Winn and his specialist teams for their “greening vision”.

32

REFERENCES Diederichs, N and Van Nierick, M, (2009). Municipal Project Managers Handbook: Managing the Environmental Impact Assessment Processes. Environment Department Library, eThekwini Municipality, Durban, South Africa. Department of Water Affairs and Forestry (2005). Waste Management Series. Minimum Requirements for the Handling, Classification and Disposal of Hazardous Waste. Draft 3rd edn. Pretoria Parkin, J, Strachan, L, Bowers A, Wright M and Winn, R (2006). Extreme Landfill Engineering: Developing and Managing South Africa’s busiest and largest landfill facilities. Wastecon 2006. Conference Proceedings, Cape Town, South Africa, 2006. Strachan L, Rolando A and Wright M (2002). Rescue, Reinstate and Remediate – Landfill Engineering Methods that Conserve the Receiving Environment. Proceedings Wastecon 2002, International Congress, IWMSA, pp 443-451 Torr LC, (2009). Application of Diary Waste Water to Agricultural Land: An Environmental Perspective. Master of Arts, University of Stellenbosch, Department of Geology, Geography and Environmental Studies, Stellenbosch, South Africa. ARC-Plant Protection Research Institute (Website: http://www.arc.agric.za/)

33


ASPHALT LINERS IN LANDFILL CONSTRUCTION

Thomas Egloffstein and Gerd Burkhardt ICP Ingenieurgesellschaft Prof. Czurda und Partner mbH, Karlsruhe, Germany – [email protected]

1 GENERAL

Asphalt pavements are common in road construction. But asphalt concrete and bitumen have also been used in hydraulic engineering for a long time very successfully to serve sealing purposes, e.g. as impervious cores or surface seals of dams and for reservoir linings. Similarly, in areas where clayey soils are not available, asphalt liners have been used successfully for bottom barrier systems of domestic waste landfills. In Switzerland and in Germany asphalt is considered an established alternative liner material in landfill engineering, and pertinent regulations, based on experience and on performance evaluations including investigations on exhumed landfill asphalt liners (Schellenberg 1996), have been introduced. Starting in the mid nineties of the past century, asphalt bottom liners were also constructed at landfills in countries outside Europe, e.g. landfill “Deir el Balah” in Gaza-Palestine (Burkhardt 1996/1997) and landfill “Al Multaqaa” in Oman (Burkhardt 2004) or the landfills at Sousse and Kairouan in Tunisia. As compared to compacted clay liners and to geomembranes, asphalt liners have the following advantages:

Asphalt is a well known construction material. Its constituents, bitumen and mineral aggregate are available in most countries of the world, and so is the equipment for processing, placing and compacting asphalt. The craftsmanship for handling asphalt construction is well understood.

Asphalt liners are robust. The material is ductile and can follow differential settlements without rupture.

Asphalt liners of adequate quality are perfect seals with respect to advective flow of leachate, comparable to geomembranes in sealing efficiency.

Asphalt liners can be constructed in hot and dry climates, where placement and compaction of clay liners is very demanding because of rapid desiccation.

Asphalt liners can be constructed in cold climates as well, provided they are placed upon a frost proof coarse grained well compacted base course. Freezing problems which are typical of unprotected compacted clay liners do not exist.

The disadvantages of asphalt liners are: Composite liners of asphalt and mineral sealing layers as promoted by the German guide

lines, are cumbersome and expensive.

Asphalt has no sorption capacity for contaminants. If there is a defect, leachate will migrate to the ground directly.

Asphalt changes its properties by ageing to some extent.

Asphalt is not resistant against all substances which might occur in a solid waste landfill. Asphalt bottom liners should not be executed for hazardous waste landfills, if organic solvents, oil, diesel fuel, benzene or similar organic liquids have to be anticipated among the deposited waste material.

34

2 PROPERTIES OF ASPHALT LINERS

Asphalt is a mixed in plant construction material composed of specified gradations of crushed rock, sand, mineral filler and bitumen. Depending on the function it serves, the dosage of the components, the type of bitumen used, details of the mixing process and compaction requirements may vary. Special rules with this respect have to be followed in the production of landfill asphalt liners. In spite of thorough compaction after placement, a certain amount of air voids remain trapped within the bitumen matrix. But since these voids are not interconnected, asphalt can be considered practically impervious to advective flow of water or leachate at the base of a solid waste landfill. Provided the recommendations given subsequently for dosage, processing, placement and compaction are followed, the relationship between void ratio and water pressure shown in Figure 1 is applicable to asphalt liners at the base of landfills.

Figure 1: Permeability of sealing asphalt (0/11) in relation to void ratio and water pressure

(Vater 1996) Bitumen that constitutes the impervious matrix of asphalt is not sufficiently resistant to a number of chemicals. Table 1 lists the most common liquids which might be of concern in connection with solid waste landfill applications of asphalt. Asphalt bottom liners shall not be constructed in cases where large amounts or high concentrations of substances can occur which are aggressive to bitumen, which dissolve bitumen or cause rapid ageing of bitumen. 3 REQUIREMENTS ON ASPHALT LINERS

In Germany asphalt liners in landfills („landfill asphalt“) usually had to meet the demands of the German “General Approval“ (“Allgemeine bauaufsichtlichen Zulassung” No Z-67.11-1) of the German Institute for Technical Construction (Deutsches Institut für Bautechnik) as of 23.07.1996 (which is currently not in force). The mixture for asphalt layers shall be produced by dry mixing of crushed aggregates, quarry sand and natural sand (any or all of it) of the required grading with bitumen. The layers shall be classified as follows:

35

Table 1: Resistance of bitumen to chemical substances (Deutsche Shell 1990)

Table 2: Requirements on asphalt sealings for landfills according to TL Min-Stb 91, Germany. a) First asphalt layer (asphalt liner 1 or asphalt bearing layer)

Grain size distribution 0/16

Bitumen content (DIN 1995-1)

≥ 5.2 % by weight

Bitumen B65 or B80

Thickness 8 to 12 cm

Void ratio (air voids) Less than 5 % b) Second asphalt layer (asphalt liner 2 or asphalt sealing layer)

Grain size distribution 0/11

Bitumen content (DIN 1995-1)

≥ 6.5 to 7.5 % by weight

Bitumen B65 or B80

Thickness Minimum 6 cm

Void ratio (air voids) Less than 3 %

4 REQUIREMENTS ON MATERIALS

4.1 Aggregates

All types of aggregate and sand, except for natural sand, shall be crushed from natural rock or crushed river gravel. Crushing shall consist of at least two stages. The aggregates shall meet the following requirements: They shall not contain grains and sand, except for natural sand, which is a product of blasting. They shall not contain more than 0.25 % of clay particles and lumps. River gravel shall be crushed in such a manner as will ensure that at least 80 %, by weight, of the aggregate particles retained on sieve No. 4 shall be crushed particles. Crushed particle means a particle containing at least one face broken mechanically.

36

The sand equivalent of material passing a No. 4 Sieve shall be at least 50% after passing a drying drum. Grading of the aggregates shall permit their proportioning and mixing according to grading curve lying within the gradation range as specified below. Natural sand, if used to improve the grading, shall not exceed 10 % of the total weight of aggregates. Natural sand shall be free of salts, clay particles or lumps, organic matter or other foreign substances. Testing of aggregates for quality: The samples for determining the quality of aggregates shall be taken daily throughout the production period. The samples shall be taken from the crusher conveyers, from the stockpiles or from the hot bins in the asphalt plant after drying.

Figure 2: Grain size distribution - aggregates of Asphalt Liner No 1 (0/16) and No. 2 (0/11) – according to German guidelines (TL Min-Stb 91)

4.2 Bitumen for Asphalt Concrete

The bitumen material requires prior approval by an experienced Engineer. The samples shall be submitted for approval at least 30 days before the use of the asphalt is to begin and also during the progress of the work according to the Experts instruction. The bitumen shall be delivered in containers directly from the place of production.

37

4.3 Composition of the Mixture

The characteristics of this finished mix to be prepared and tested shall be as follows:

Table 3: Composition of mixture (acc. to DIN 1995-1) Characteristics examined Asphalt Liner

No. 1 Asphalt Liner

No. 2 Voids in mix (%) max. 5 max. 3 Bitumen (% by weight of aggregates and bitumen)

5.2

6.5 - 7.5

The exact quantity of bitumen shall be determined as follows: The grading curves of the aggregates according to one of the ranges given above should be tested by batches of the mix containing 5.5% bitumen and 6.0% bitumen. The next tests should be made with a bitumen content of 5,7 % and 6,2 % and so on. For a mix according to the proposed grade line a Marshall test shall be made in the laboratory and the optimum asphalt content according to DIN ISO or AASHTO procedures (or similar) shall be determined. The permissible deviation from the prescribed bitumen content shall not exceed ± 0.3% (± three-tenths of a percent).

5 PRODUCTION PROCESS

5.1 General

Asphalt concrete shall be supplied to the site from a production plant located in the vicinity of the construction site or at such reasonable distance from the site as will permit supply of asphalt at the required temperature. The output of the plant should be sufficient to the demand. Direct communication between the work site and the production plant shall be maintained by means of a telephone line or radio to ensure steady supply and reliable control over production.

5.2 Screening and Feeding

After drying and heating the aggregates shall be screened and separated into at least three fractions in the internal storage bins. Each aggregate shall be separately conveyed and introduced from the storage bins to the feeding system. Introduction of material for drying shall be done in suitable proportions and at a rate, which permits heating and drying as required. 5.3 Weighing and Mixing

The aggregates shall be weighed to an accuracy of ± 5 kg and shall be conveyed to the mixer in the required quantities. The bitumen shall be made flowing from the measuring device to the mixer and spread throughout. The mixing shall continue until a homogeneous mixture is obtained but shall last for not less than 50 seconds after introduction of all the ingredients into the mixer. The temperature of the bitumen during mixing shall be between 150 and 160 °C. The temperature of the aggregates in the mixer shall be 160 to 170 °C during addition of the bitumen. Mixes whose temperatures do not fall within this range shall be rejected. The moisture content of the aggregates after drying in the plant shall not exceed 0.3%. The moisture content shall be determined as follows: A sample of 5 kg taken from the internal bins (after the sifting plant) shall be placed in a closed dry container. The container shall be weighed with the aggregates and placed in the oven, which shall be kept at a temperature of 120 °C for 24 hours with the cover of the container. The container shall then be removed from the oven and weighed again. The difference between the obtained weights before and after the drying process is

38

taken to determine the moisture content.

6 PLACING OF ASPHALT CONCRETE LAYERS

6.1 Preparations of Areas

Before spreading asphalt the area shall be swept clean of all foreign matter, spilt material or dust. After the surface of the area has dried it shall be sprayed with bitumen emulsion or sprayings with an approved mechanical sprayer. The quantity applied shall be checked by tests. The quantity applied by spraying shall not be less than that specified and shall not exceed the quantity by more than 50%. Each spraying shall be done one day before placing the overlying layer. Before spraying the area will be swept with brooms so that it is free of all foreign matter or dumped material and also free of dust and other dirt, which may have accumulated. In the event that a layer or the surface on which it is placed is contaminated with diesel oil or kerosene such layer will be removed. The entire area shall be sprayed uniformly with the aid of mechanical sprayers and shall remain exposed to the atmosphere for a period of time to be determined by the Engineer before covering with the next layer. Pools of bitumen shall be covered with fine sand serving as an absorbent, which will subsequently be removed. The Contractor shall prevent people and mechanical equipment from moving over areas, which have been sprayed. 6.2 Transporting of the Asphalt Concrete

The ready asphalt concrete shall be hauled from the mixing plant to the area to be paved in trucks with clean smooth metal load boxes. On cold and wet days the boxes shall be covered in order to protect the asphalt from rain and cooling-off. To prevent sticking of asphalt to the box walls the latter may be coated with a small amount of a suitable compound or soaked lime. The boxes shall be cleaned out at the end of each working day. The transport distance shall not be large, as the installation temperature of asphalt must not be lower than minimum 130-140°C.

6.3 Spreading with Finishers

Spreading of the various asphalt layers shall be done with road finishers on which sensors and smoothers are installed which advance on steel cables tensioned and prepared beforehand on pegs in accordance with the thickness of the required layer. The steel cables shall be of an approved type. Only where spreading by machine is impossible (e. g. at the slopes of an embankment dam) manual spreading shall be permitted, provided that written approval is obtained and further provided that the spreading is not done with rakes but with hot blades by a method which will prevent segregation. The surface of the layer before and after the compaction shall be smooth and free of cracks, tear holes and perforations. The surface of the layer shall conform after compaction to the required gradient and shape. Uninterrupted supply of asphalt concrete to the finisher has to be ensured in order to prevent material held in the finisher and in the worm screw from cooling and to enable the material spread behind the finisher to be compacted before it cools. 6.4 Joints

Longitudinal joints in any layer shall be staggered against those in the underlying layer by at least 60 cm. Transverse joints shall be staggered by at least 60 cm. Joints shall be made by cutting strips at least 5 cm wide and throughout the thickness of the layer at the shoulder of the plot and the material removed. If not stated otherwise the cut area shall be heated by an approved heating machine attached to the finisher to a temperature of 110 - 130 °C.

39

6.5 Rolling and Compaction

A three-wheeled roller, a tandem roller and/or heavy pneumatic-tire rollers shall do compaction immediately after spraying and at a temperature of 130 °C to 140 °C. Any layer having a lower temperature has to be rejected. No repairs of the layer surface once it has been rolled shall be permitted. Any damage to a layer by removal of another upper layer shall be repaired.

7 EXAMPLES

7.1 Landfills with Asphalt Sealings in Germany

In Germany and Switzerland a lot of asphalt liners in landfill construction were built. Some examples were given in the following.

Figure 3: Asphalt sealing following the “General Approval“ (“Allgemeine bauaufsichtlichen Zulassung” No Z-67.11-1) of the German Institute for Technical Construction (Deutsches Institut

für Bautechnik – dai 1996)

Figure 4: Asphalt bottom liner of a landfill of an iron works in Germany

30 cm drainage layer

6 cm asphalt sealing layer (asphalt liner no 2)


12 cm Hydraulically bonded base course

40 cm bearing layer consisting of blast furnace slag and up to 20 % asphalt rubble

Natural underground

Bitumen membrane as addiditional impervious layer




8 cm asphalt bearing layer (asphalt liner no 1)

40 cm compacted clay liner (CCL)

Natural underground

40

Figure 5: Photo of asphalt liner construction of a landfill of an iron works in Germany

7.2 Landfill Al Multaqaa, Oman

Since middle of the nineties last century asphalt sealings were built in other countries too. An example from Oman is given in this chapter. The landfill was designed by ICP in 2001/2002.

Figure 6: Bottom liner of landfill Al Multaqaa, Oman

8 REFERENCES

Asphalt Institute (1994): Mix Design Methods for Asphalt Concrete and other Hot-Mix Types, Manual Series No. 2 (MS-2) Sixth Edition, Asphalt Institute, Lexington, Kentucky, 1994 Burkhardt, G. (2004): Asphalt Liners for Sanitary Landfills – Landfill of Muscat Municipality, Oman, Lecture at the First Chinese/German-Geo-Environmental Forum, Wuhan, China, 2004 dai (Deutsches Asphaltinstitut 1996): Asphalt für Deponieabdichtungen, Allgemeine Bauaufsichtliche Zulassung Nr. Z-67.11-1 as of 23.07.1996



8 cm asphalt bearing layer (asphalt liner no 1)

30 cm compacted base course

Compacted natural underground

41

Deutsche Shell Aktiengesellschaft (1990): Empfehlungen für die Ausführung von Asphaltarbeiten im Wasserbau (EAAW 83), 4. Auflage, 1990 DVWK (Deutscher Verband für Wasserwirtschaft und Kulturbau: Merkblätter zur Wasserwirtschaft, 237/1996, Deponieabdichtungen in Asphaltbauweise DIN 1995-1 (1989): Deutsches Institut für Normung: Bitumen und Steinkohlenteerpeche – Anforderungen an die Bindemittel – Straßenbaubitumen Müller, W., August, H.,Jakob, R.,Tatzky-Gerth and Vater E-J. (1995): Die Wirkunsgweise der Kombinationsdichtung – Immersionsversuche zur Schadstoffmigration in Deponieabdichtungssystemen, in Burkhardt, G. und Egloffstein, Th. (Hrsg.): Asphaltdichtungen im Deponiebau, expert Verlag, Renningen-Malmsheim, 1996 TL Min-Stb. 91 (1991): Technische Prüfvorschrift für Mineralstoffe im Straßenbau, Forschungsgesellschaft für Straßen- und Verkehrswesen, FGSV) Vater, E.-J. (1995): Eigenschaften von Asphaltbeton für die Anwendung in Deponieabdichtungen – Stand der Beratungen im DIBt, in Burkhardt, G. und Egloffstein, Th. (Hrsg.): Asphaltdichtungen im Deponiebau, expert Verlag, Renningen-Malmsheim, 1996

42


LANDFILL GAS EXTRACTION PROJECTS – A CONTRACTOR’S VIEWPOINT

Justin Butt PrEng, BSc (Eng), Dip. Proj. Mgt, MSAICE, AArb. SUMMARY: Whilst much has been written regarding the theory and design of landfill gas projects, little can be found regarding the construction of such schemes. This paper aims to highlight some of the issues facing a contractor, from the weather, to installing the gas wells, sub-contractors and other challenges. It is hoped that this may assist consultants and operators in the planning and running of landfills in order to access the gas sooner for the client.

1. INTRODUCTION A lot has been written and studied about the theory of landfill gas (LFG), and how starch plus water, in the absence of oxygen, will yield methane plus carbon dioxide, and how this may be used to generate electricity or carbon credits through flaring. The labyrinthine process of getting a LFG project certified with the UNFCCC, and the length of time it takes to get payment for your first carbon credits is well known. A Google search will quickly yield the principles of the process and will offer step by step solutions to applying for carbon credits. But none of this theory is worth much without the practicalities of physically building the scheme, getting the gas out the waste and to the flare or engines. This paper aims to shed some light on the construction phase of a LFG extraction project and how landfill operators and consultants can assist the construction through their daily operations and designs. It is hoped that this will assist consultants in the designing and planning of such projects thereby speeding up construction and the path to attaining carbon credits and electricity sales. 2. BASIS This paper is based on personal experiences Fountain Civil Engineering has had in constructing many of these projects at various landfills around South Africa. Sites we have been instrumental in include Bisasar Road, Mariannhill, La Mercy, Rietfontein, Simmer & Jack, Weltevreden, Rooikraal, Second Creek, Robinson Deep and Marie Louise. 3. MATTERS AFFECTING CONSTRUCTION This section of the paper sets out the physical challenges and issues that we have been faced with during construction of LFG projects. 3.1 Program and Weather The weather can be one of a LFG project’s biggest enemies. Due to the requirement for oxygen and gas exclusion, clay materials with a high P.I. are usually specified for significant sections of the work. The problem with these materials is that even a small rainfall can render the material useless for days at a time. To compound the situation, the 4:00pm storms in Johannesburg every day in summer or the continuous rain for days at a time in Durban result in the material not getting a chance to dry, sometimes for weeks at a time. In December 2007 our project at Bisasar Road in

43

Durban was put on hold for almost 5 weeks due to the summer rains. Even after the rains had ended and with the hot Durban summer, most of the site was still inaccessible for days at a time; and just as we would be getting ready to start work again, the next batch of rain would hit and the site would be saturated all over again. Figure 1 below shows a gas wellhead at Bisasar Road submerged in about 500mm of water. This illustrates the challenge that weather can provide when working with clay-type materials. Despite digging trenches to allow the water to run away, and the hot summer days, the material remains saturated and unworkable for days after the rains have abated.

Fig. 1: Gas well at Bisasar Road still submerged days after the rains have abated. Whilst it may be difficult to plan just when a project will start, due to the long delay in designing, and getting funding for the project, especially when the client is a municipality, a LFG project constructed during the drier months will have a much greater chance of early completion than one that begins at the same time as the rainy season. 3.2 Pipework 3.2.1 Pipe Material Arguably the biggest component of the construction phase of a LFG project is the installation of the gas carrier pipe network. HDPE is the preferred medium as the material is strong but flexible and has a high resistance to the corrosive properties of landfill gas. One of the major issues with the use of HDPE, however, is that because the material is oil-based, the price of the material is strongly linked to the fuel price. This can have repercussions later in the project, when the pipe supplier may have quoted or may have even delivered materials at one price, and then cannot supply the same materials for the same price 2 months later. Therefore it may be worth investigating the options of treating HDPE material as a rise and fall item, linked to a base rate at tender stage.

44

A second impact of the material being oil-based is that strikes in the petroleum industry affect the HDPE market. This was the case during the recent strikes in July 2011 when the petroleum industry went on strike for 3 weeks. HDPE pipe supply was severely hampered to our Johannesburg sites at Simmer & Jack and Rooikraal, the result of which was a 3 week set back in our (already rather tight) program. 3.2.2 Pipe Welding The two most common methods of connecting pipes on LFG schemes are butt welding and electrofusion (e/f) couplings. If carried out correctly, butt welding is by far the preferred method from a contractor’s viewpoint. Not only is it quicker and easier to arrange when welding long lengths of pipe, but it also has a lower failure rate than e/f couplings. The failure rate on e/f couplings has been found to be as high as 25%, especially on larger diameter couplings that require a more powerful generator than is usually available on site. On our 2011 contract at Robinson Deep, oxygen levels as high as 5.5% were being recorded at the flare station. Upon (much) investigation, it was found that four of the 500mm diameter e/f couplings were leaking. These had to be cut out and re-welded (now with 2 e/f’s in the place of one to make up for the extra length of pipe in the cut out section) which had obvious time and financial implications – at over R7500 per coupling plus the additional time and labour to find and replace the faulty couplings. Figure 2 below shows an example of two of these problematic 500mm e/f couplings at Robinson Deep.

Fig. 2: An example of the leaking 500mm diameter electrofusion welds at Robinson Deep Butt welding, on the other hand, can be much better monitored, and if done according to the correct pre-determined specifications is virtually 100% failsafe. There are very sophisticated welding machines available these days to ensure the welding is carried out correctly and eliminate any human error as far as possible. It is possible to monitor each weld individually with a data logger that can then be downloaded to a computer. In this way each and every weld can be carefully monitored for any non-conformance. An example of a tracked welding machine with a computerised data logger, used at Bisasar Road, is shown in figure 3 below.

45

Fig. 3: Tracked HDPE pipe welding machine with data logger (Bisasar Road) Unfortunately not all pipe welding operations are as sophisticated as the tracked machine shown above. Some of our sub-contractors in the past have made use of builders’ rubble and bits of wood as pipe supports while welding, as illustrated in figure 4 below.

Fig 4: Sub-standard pipe welding ‘supports’ Compromised welding techniques, as illustrated above, lead to sub-standard welds which may result in a weaker weld, but will more likely lead to an unacceptable internal bead inside the pipe. This is a particular problem when welding two 12m pipes together, as it is nearly impossible to guarantee that the weld beads are acceptable. Figure 5 below shows an example of an unacceptable internal bead created as a result of sub-standard welding practice.

46

Fig 5: Unacceptable internal bead created as a result of sub-standard butt welding 3.3 Well Installation Naturally the first step in any successful LFG project is obtaining the gas. To this end we have installed both vertical and horizontal gas wells at numerous landfills around the country. Both systems have their merits and demerits and will be discussed in turn. 3.3.1 Vertical well installation To date we have installed over 160 vertical wells at 5 different landfills, using 3 different types of wellhead design. The first step in the process of installing a vertical well is to auger the well to the required depth. We have used both a continuous flight auger (CFA) and a kelly bar auger and have found the CFA to be much better suited to the challenges of drilling into a waste body. A CFA pushes the waste to the surface continuously, whilst a kelly bar arrangement only brings up approximately 1m of compacted waste at a time, and then has to be re-inserted into the hole to drill a further 1m, and so on. Further when drilling into wet waste (generally from about 10m onwards) the waste tends to fall off the kelly bar cutting head before being brought to the surface, and so progress can be extremely slow. At Robinson Deep we started drilling with a kelly bar auger, but excessive leachate levels high in the waste body resulted in us having to remove that rig and continuing with a CFA machine which was able to complete the job. This naturally meant two establishments for the drilling teams – a cost which we had to carry. Examples of CFA and Kelly bar augers are shown in figures 6 and 7 below.

47

Fig.6: CFA Piling rig (Bisasar) Fig. 7: Kelly bar auger (Rietfontein) Once the well has been augered out, and the perforated gas recovery pipe inserted into the well the wellhead needs to be installed in order to monitor and control the well. At Bisasar Road and Second Creek we installed “HofstetterTM” though-flow type wellheads. The design of these wellheads is such that they need to be installed over the recovery pipe, inside the 450mm diameter augered hole some 3m – 4m below the surface of the landfill. Without any light for guidance (torches are not allowed as there is a danger of the flammable gas interacting with the bulb which could be fatal) this can be a bit like trying to thread a needle with boxing gloves! Once safely installed, however, the carrier pipes are then easily laid, with no more than 750mm excavation between adjacent wells, due to the fact that each well in fact acts as a knock out for condensate in the system, and the carrier pipes connect to the body of the wellhead below the surface of the ground. This type of wellhead and carrier pipe layout is illustrated in figure 8 below.

Fig. 8: “HofstetterTM” type wellhead showing shallow excavation of carrier pipes (Bisasar)

48

At Rietfontein, and Simmer & Jack we installed a different type of wellhead designed by Jones & Wagener. This wellhead has the advantage of being installed inside the recovery pipe at the surface and is therefore simpler to install since one is working at close quarters and can see clearly what they are doing. The drawback to installing this type of wellhead, however, is that because the gas outlet is above ground, condensate cannot be knocked out back into the well. The result is that all carrier pipes need to be laid at 3% - 4% minimum grades away from the first well, with the consequence that the excavation becomes very deep – up to 5m or more – especially at the crest of the landfill. Because of the unpredictable nature of compacted waste excavating is a slow process, even with a large excavator digging shallow trenches, so these deep trenches are very slow going, not to mention the inherent dangers of deep excavations. An example of these wellheads is shown in figure 9 below.

Fig. 9: Jones & Wagener designed wellhead with gas offtake above ground (Rietfontein) At Robinson Deep and Marie Louise a third type of vertical was installed. This well, designed by EnerG Systems, was simply a 90 degree elbow welded to the top of the recovery pipe, approximately 500mm below the surface. This wellhead is by far the simplest, but due to the simple design, each well is individually controlled at a central manifold and thus separate carrier pipes need to be laid for every well, often resulting in twin pipes laid for long distances in a common trench. An example of this type of wellhead is shown in figure 10 below.

49

Fig. 10: EnerG Systems wellhead – a simple 90 degree elbow 500mm below the surface

3.3.2 Horizontal well installation Since 2008 we have installed over 80 horizontal wells at 6 landfill sites. The design of the wells has not changed significantly from one site to the next, but the one thing that does influence construction is the gradient of the surrounding area. Due to their design, horizontal wells are usually over 100m long and must be constructed ideally at a fall of between 3% – 5%. At this gradient it is easy to see that, unless the landfill itself slopes accordingly the end of the horizontal well gets very deep, and putting staff into the well to ensure adequate stone bedding and pipe installation is extremely dangerous. For this reason wherever possible landfill operators who wish to install horizontal wells should attempt to slope the surface of the landfill at a 3% grade in the direction of the horizontal well. When carried out successfully, it is possible to install numerous horizontal wells quickly and easily, with the result that the client gets access to the gas sooner, and the contractor does not get in the way of the landfill operator whilst trying to finish very deep wells. This can be seen at Bisasar Road in figure 11 below, where 5 horizontal wells have been completed (the first 2 being covered already by the next terrace of waste) with a 6th currently underway, and a 7th soon to be started, all within less than 2 weeks. The result is that the operator can landfill without being hindered in any way by the contractor.

50

Completed wells Next well set out Well under construction

Fig. 11: Seven horizontal wells successfully constructed (Bisasar) 3.4 Control Chambers Control of the gas flow from individual wells is critical to the successful running of any LFG scheme. To this end we have installed numerous control chambers, known as manifolds, on LFG sites around the country. At Bisasar and Mariannhill these chambers took the form of plastered brick chambers with the individual pipes being built into the wall of the chamber. These chambers are easy to add additional sections to and are strong enough to withstand the rigours of wayward plant operators who have been known to ride into them. The result was that, while the wall may have been damaged, the pipework was unharmed, and the system ran without faltering. An example of one of these brick manifold chambers is shown in figure 12 below.

Fig. 12: Brick chamber and manifold (Bisasar)

51

We have also installed modular type HDPE manifolds that are built as a unit. These manifolds have the advantage of being constructed in a factory, and then simply being dropped into place on site, however one doubts that they would be able to withstand a collision with a dozer. An example of a modular HDPE manifold is shown below in figure 13.

Fig. 13: Modular HDPE manifold box built in a factory (Robinson Deep) 3.5 Sub-contractors Due to the many varied requirements in a LFG project, there are always numerous sub-contractors for the main contractor to deal with. The drilling of the vertical wells, welding of the HDPE pipe, electrical connections in the flare compound, engine installation and maintenance, SCADA and other IT issues to do with engines and flaring required for verification and numerous other tasks associated with any LFG project are usually sub-contracted out to experts in their respective fields. This carries its own challenges for the main contractor – having to rely on others’ equipment which may be sub-standard, or the availability of competent staff to carry out specific tasks etc. To this end, Fountain Civil Engineering has purchased a CFA piling rig to carry out all well drilling and two HDPE welding machines capable of welding pipes up to 250mm in diameter in order to reduce our reliance on sub-contractors wherever possible. With these acquisitions we are able to construct the entire scheme from gas well to flare inlet without the input of any subcontractors on site. 3.6 Other challenges facing a LFG contractor We have overcome numerous other challenges on LFG schemes, but will cover just two in this section – high leachate levels, and vandalism. 3.6.1 High leachate levels High leachate levels in the landfill can cause problems for any contractor on site. At Robinson Deep the leachate was unexpectedly high, within 1m of the surface. This resulted in severe drilling problems and the need to change drilling and well installation methods. This high level of leachate can be seen in figure 14 below.

52

Fig.14: High leachate levels are a contractor’s enemy (Robinson Deep) 3.6.2 Vandalism Due to the high number of scavengers on any landfill site vandalism can be a cause of concern for the LFG contractor. At Rietfontein all our vertical wellheads were vandalised in order to break the 63mm galvanised steel flanges off for scrap metal value. Other items, such as plant batteries, generators, tools and even the site office fridge have all been stolen at one point or another. Figure 15 below shows a damaged wellhead at Rietfontein, vandalised to get at something as small as a 63mm galvanised flange.

Fig. 15: Wellhead vandalised to get 63mm galvanised flange for scrap metal value

53

4. CONCLUSIONS No LFG project would be possible, and no carbon credits obtained or kilowatts generated if it were not for the contractor building the scheme. There are ways that consultants and landfill operators can design and plan that will enable the contractor to complete the project sooner and thereby allow the client to get access to the revenue-producing gas sooner. These include

planning for the construction phase to be in the dry season wherever possible specifying butt welding rather than electrofusion couplings for HDPE pipes identifying the best type of wellhead for the situation, taking into account the associated

challenges of installing the corresponding carrier pipes landfilling and capping at a 3% grade to suit horizontal well installation investigating the possibility of treating HDPE materials as a rise and fall item, linked to

a base rate at tender stage. Other challenges that we have overcome whilst in the construction phase of many LFG projects include high leachate levels, vandalism that is rife on most landfills, and dealing with numerous sub-contractors for the many varied tasks on a project – something that we at Fountain Civil Engineering have tried to mitigate by purchasing our own well drilling and HDPE pipe welding machinery.

54


DRAINAGE SAND BULK SAMPLING A NEW LINED RESIDUE DISPOSAL

FACILITY

Irene Nyirenda and Chris van Renssen Golder Associates Africa (Pty) Ltd, PO Box 600, Halfway House, 1685, South Africa e-mail: [email protected]; [email protected] SUMMARY An H:H landfill was designed and developed for a manganese company by Wates, Meiring and Barnard and Golder Associates Africa (Pty) Ltd (GAA). The site was commissioned in January 2006. During construction, the Contractor submitted a grading envelope for a filter sand that was evaluated and accepted by the GAA Engineer.. Subsequent to completion of construction, large stones were found in the sand drainage layer, and concern was expressed that the presence of these stones could result in damage to and subsequent leakage of leachate through the geomembrane liner below the sand drainage layer. 1. INTRODUCTION An H:H landfill site situated on the western outskirts of the Nelspruit town, approximately 6km from the plant and 10km from Nelspruit CBD was designed, permitted and developed in June 2003 for a manganese company by Wates, Meiring and Barnard and Golder Associates Africa (Pty) Ltd (GAA). Its construction completed in December 2005 to receive manganese waste in the form of a filter cake (Larox) generated by the plant. Site is classified as H:H according to DWAF’s: “Minimum Requirements for Waste Disposal by Landfill”. A daily volume of approximately 230 tons of residue is generated continuously at an in-situ (compacted) density of 1 600 kg/m3. 2. DESIGN CRITERIA The facility was one of the first hazardous waste facilities in South Africa that adopted the risk based approach for the liner design. The first phase of the landfill has a footprint of 5.5 Ha, consisting of four cells with a life span of eight years. The design philosophy for the construction of the disposal facility was based on the need to utilise the potential airspace to achieve the following objectives:

Compliance with the Minimum Requirements for Waste disposal by Landfill Dispose of the volume of residue to satisfy the life of project requirements Provide a facility that is simple and economical to operate Effectively eliminate or minimise risk of surface and ground water pollution

55

3. LINER SYSTEM OF THE FACILITY The lining system installed for the hazardous landfill is shown below.

The lining system described from top to bottom:

Leachate Collection System: The leachate collection system consists of a 300mm thick free draining sand layer. Perforated collection pipes (110 dia.) are placed at a horizontal spacing of approximately 50 m. The perforated pipes are covered with 19mm stone which is wrapped with a nonwoven needle punched geotextile.

Primary Composite Liner: The primary liner consists of a 2 mm High Density Polyethylene (HDPE) geomembrane over a 4000 g/m2 geosynthetic clay liner (GCL).

Leakage Collection Layer: The leakage collection layer consists of a 750 µm HDPE cuspated drain. The cusps are placed pointing down and filled with sand. Perforated collection pipes are placed at a horizontal spacing of 50m.

Secondary Composite Liner: The secondary liner consists of a 1.5 mm HDPE geomembrane over a 4000 g/m2 GCL.

In-situ material: The in-situ material has been ripped and recompacted to a depth of 150mm and a density of 95% Mod AASHTO

Both the primary and secondary liners are composite in nature consisting of a High Density Polyethylene (HDPE) geomembrane and a 4 000 g/m2 geosynthetic clay liner (GCL). The GCL layers were assumed to replace the compacted natural clay liner specified in the Department of Water Affairs & Forestry (DWAF) Minimum Requirements for waste disposal by landfill. The liner system serves to:

Separate the residue material from the groundwater system Direct leachate to a collection point Intercept leachate leaks through the upper barrier and direct it to a leakage collection

point 4. COMPATIBILITY OF LAROX WASTE MATERIAL AND FILTER SAND The filter design requirement that was used during the design stage was as follows:

1. D15 of filter to be less than 5 times D85 of soil to be retained 2. D15 of filter to be between 4 and 20 times D15 of soil to be retained 3. D85 of filter to be less than 5 times D15 of filter

Figure 1: Liner System of the H:H Landfill Facility

56

The first two requirements ensure that particles will not move from the retained material into the filter, and also that the filter will be significantly more permeable than the material retained. The third requirement ensures that the particle size distribution of the filter material will not change as water seeps through it. The 300mm free draining sand layer was specified based on waste/filter compatibility outlined above. 5. PROBLEM STATEMENT During the construction of the disposal facility, the construction quality assurance was done by a local consultant appointed by the facility owner who was not affiliated to GAA. The contractor submitted a grading envelope of the filter sand to be used in the 300mm leachate collection layer. The GAA design Engineer evaluated the grading envelope of the filter sand as per filter design requirements and accepted its use. The grading envelope was introduced onto the construction drawings and issued to the Contractor. Subsequent to completion of construction, approximately a few months into operating the landfill, GAA and the owner of the disposal facility found large stones in the sand drainage layer, and concern was expressed that the presence of these stones could result in damage to and subsequent leakage of leachate through the geomembrane liner below the sand drainage layer. GAA as the design Engineers of the landfill facility were requested by the owner of the disposal facility to facilitate an investigation into the possible damage the stones found could have on the overall performance of the facility, as well as the immediate potential damage to the liner. 6. SELECTED METHODOLOGY TO RESOLVE THE DISPUTE At the time of the dispute, the first two cells of the disposal facility were filled with larox waste to approximately 2.5 metres in height, for this reason very little investigation could be done over these areas. In order to resolve the dispute, a programme of how the sampling of the filter sand placed in Cells 3 & 4 will be achieved had to be determined and defined; keeping in mind that approximately 3.5 ha of the site was covered in 300mm sand layer. For this reason, it was decided that statistical representation methods will be adopted. It was also important to identify an independent party, who would evaluate the processes, methodology and findings in which the dispute was to be resolved should the matter proceed to arbitration. An independent statistical specialist, who was familiar with statistical quality assurance techniques in road building, was approached by GAA to assist in trying to resolve the conflicting views of the different parties on the acceptability of the sand delivered to site as a drainage/filter layer.

57

7. SCOPE AND OBJECTIVES The objectives of the investigation were to:

Confirm whether the filter sand that was placed on the liner is within the specification of the grading envelope that was evaluated and approved by the GAA design Engineer.

Verify the presence of oversize stones/pebbles within the filter sand drainage layer and if so confirm the sizes of these oversize stones.

To determine whether the stone in the filter sand has a substantial negative influence on the filtering capacity of the filter sand.

Establish if the placing of the sand containing river pebbles and stones allegedly damaged the liner by means of visual inspection and by means of electric leak detection testing.

8. CHOSEN SAMPLING TECHNIQUE As it is practically impossible to evaluate each and every square metre of the drainage layer, statistical techniques were employed to evaluate the quality of limited number of samples which are “assumed” to represent the total population (i.e. the total area of the filter layer). In order to get a true picture of the quality it was essential that every square metre of each cell has got an equal chance of being selected for evaluation. This was done by means of random sampling techniques, in other words selecting the sampling positions without knowing what the quality is like at that particular position. This also means that there should not be any “visual” differences before selection starts. In this case the liner of the section covered with the filter sand should be evaluated separately from the liner not covered by the filter sand. Even though sampling positions are normally determined by random numbers there is a slight possibility that these positions may be bunched together. In order to prevent this unsatisfactory situation use was made of stratified random sampling techniques. This involved dividing the investigation section into “n” equal portions where “n” is sample size selected. The sampling position in each of these “n” equal-sized sections is then determined in a random manner. The randomly determined sampling positions for the actual investigation are shown in Figure 2 below.

Figure 2: Site layout showing randomly determined sampling positions

58

9. SAMPLING METHODOLOGY USED FOR THE INVESTIGATION Six panel positions in each cell were selected by stratified random sampling techniques without anyone knowing what the condition was like at any of these positions. By using random sampling techniques where every section of each sampled cell has an equal chance of being selected for evaluation, one could conclude that the combined lot of six panels gives one an overall indication of the quality of the sand filter layer and the liner quality in that particular cell. Additional panel on the roadway was investigated to verify if presences of stones with the induced traffic load could cause damage to the liner. The size of each sample area was 2 m wide by 5 m long. The sand recovered from the sample strip was first sieved to screen out all aggregate particles larger than 12.5 mm. Plastic shovels were used in order to ensure that liner was not damaged by the spades. The fraction of stones larger than 12.5 mm was separated from the rest of the filter sand layer. The volume of the stone fraction was recorded and a sample taken to the laboratory to determine its grading. The filter sand sample was then riffled to reduce the sample sent to the laboratory for grading analysis.

Figure 3: Sieving and collection of oversize stones All parties involved in the resolution of conflict were present during the actual sampling exercise ensuring that there was no selection bias.

Figure 4: Sealed bag of oversize stones ready to be taken to the laboratory

59

10. ELECTRIC LEAK TESTING ON 2mm HDPE GEOMEMBRANE The electric leak detection test was conducted in accordance with ASTM 6747 and ASTM D7002-03 Standards (Water Lance Method of Leak Detection). This method uses electrical current or electrical potential to detect and locate potential leak paths. A potential leak path, for the purposes of the investigation was defined as any unintended opening, perforation, breach, slit, tear, puncture, and crack or seam breach. Scratches, gouges, dents, or other aberrations that do not completely penetrate the geomembrane were not considered. The exposed liner area was swept with brooms to remove all loose particles prior to the visual inspection and electric leak detection test. The principle of the electric leak detection test is to place a voltage between the soil under the geomembrane and a stream of water projected onto the liner surface, and this locates areas where electrical current flows through discontinuities in the liner. The lower conductive layer is usually the soil and the upper conductive layer being water. In this case, the lower conductive layer was the GCL immediately beneath the primary geomembrane liner. A cathode ground was established at the anchor trench of the cell in the primary GCL and an anode was placed in a water stream of a lance. Water was supplied by gravity from a tank truck parked at a higher elevation than the lined area. For this technique to be effective, the leaking water must come into contact with the electrical conducting medium to which the ground electrode of the 24 volts dc supply was connected. Since the geomembrane is not a perfect electrical insulator, a steady background signal is audible. As the water flows through a leak path, there is an increase in the signal. A detector informs the operator (via acoustical and visual signals) of the presence of an infiltration and thus a perforation in the geomembrane. Leak paths as small as 1 mm in size are located by an audio signal or by measuring a current of magnitude related to the size of the leak.

Figure 5: Testing the dipole for a signal prior to commencement of liner testing An initial 1mm hole was punctured in the liner as a test hole. After this defect was picked up by the detector, it was then run on the rest of the test panels to locate any visual or hidden defects on the liner. Extreme care was taken around the presence of wrinkles and waves, steep slopes and lack of contact between the liner and the conductive soil at bottom of slopes. Conductive paths such as metal pipe penetrations or batten strips on concrete along the cell benches were carefully isolated.

60

11. QUALITY CONTROL PROTOCOL Quality Assurance Protocol was initiated at every stage of the investigation. The contractor and GAA had full time supervision. The independent soil laboratory team, in agreement with contractor and GAA, established the following:

Ratio of oversize pebbles to excavated sand. An estimate of total volume retained (Excavated volume less oversize pebbles volume)

vs. total volume excavated; The depth of the panel was measured at six different points along the length of the

panel, and another 3 points across the width of the panel. An average from the 18 measurements was then determined and recorded.

Percentage sieved vs. total placed sand. Visual condition of liner prior to testing and number of defects recorded per test strip

Daily notes and photographs were maintained and records of all visual observations (surface characteristics) during the investigation were kept. These were signed off on a daily basis by all parties involved. 12. BACKGROUND TO THE EVALUATION OF THE GRADINGS OF THE FILTER

SAND Until fairly recently the only way Civil Engineers could ensure that the desired functional properties were achieved with a granular material was by specifying a grading curve. As it is practically impossible to ensure a fixed grading, a certain amount of variation is allowed for by specifying and upper and lower grading curve, creating a grading envelope, within which all delivered gradings are “expected” to fall to be acceptable. The COMPACT models were developed by CJ Semmelink (2004) from the results of investigations which covered the evaluation of the effect of the properties of untreated granular materials on their compactability. The models have since successfully been used to solve compaction problems all over the world covering natural and stabilized road building materials ranging from fill to crushed stone bases, asphalt mixes, concrete mixes and the determination of the void content of a toxic waste fill. The material property which has by far the greatest effect on the compactabiliy (i.e. space occupied by solid particles) is the grading (i.e. particle size distribution) of the material after compaction CJ Semmelink (2004). These models generally express the density in terms of space occupied by particle solids (i.e. % Solid Density) (%SD). The percentage void space in the “compacted” material is equal to 100 minus the %SD. In the case of a filter layer this is the property that ensures the functionality of the layer as filter. Unfortunately the Author could find any reference in literature as to what this percentage should be for a filter layer. The grading evaluation of filter sand retrieved from these positions showed that the filter sand void content, which is the property which allows the sand to act as a filter, is pretty uniform and seems satisfactory in terms of the sand grading envelope’s void content for the upper lower grading limits. The fact that some the grading curves are slightly outside the grading limits for certain sieve sizes is not critical as the grading is only used as an indicator of the density of the sand in terms of the space occupied by solid particles (percentage solid density) (%SD) (as determined with the COMPACT prediction models) and indirectly of the porosity of the layer (i.e.100–SD (%)).

61

13. INTERPRETATION OF DATA

All the excavated panels had a significant amount of oversize material. The stone sizes ranged from 13 mm – 75 mm. The presence of few stones larger than 75mm that were found in panel C4 West 3 raised serious concerns.

The volume of oversize stones found in all thirteen panels represented 0.70% of the total volume of sand excavated.

During the liner testing and visual observations, the twelve panels plus the roadway panel all showed no visual damage and no leak defects were detected. A few fold/wrinkle damages were observed on four of the test panels, but this was not the result of the presence of oversize stones. In this instance the presence of oversize stones within the sand layer did not have an adverse effect on the state and condition of the liner.

It was observed that the areas of exposed liner along the eastern edges of cells 2, 3 and 4 had been damaged, both in terms of holes and some deep scratches. These areas were tested by the leak detection method, and all identified leaks were repaired and subsequently tested.

However some of the scratches were at least 0.5 mm deep and there was concern over the long-term performance of the liner, as the scratches could initiate environmental stress cracking.

14. CONCLUSIONS Based on the objectives of the investigation the following conclusions were made:

No visual damage or leak defects were recorded on any of the sampled sections on the covered area including the road which has been exposed to traffic for several years. The presence of oversize stones showed no detrimental effect/damage on the liner in this instance.

Although significant defect/holes were recorded in the open strip areas, these were

outside the scope of the exercise, and were not as a result of the presence of oversized material.

The predicted void contents of the combined gradings of the sand and stone were only a

fraction of one per cent less than the predicted void contents of the sand only for all the sampled sections. It was therefore concluded that the stone had an insignificant effect on the filtering capacity of the filter layer.

Although the gradings of the filter sand as well as the gradings of the combined material

are not totally within the grading envelope, the predicted void contents compare favourably with the predicted void contents of the grading envelope. It also ties in with the range of the inherent porosities for properly designed granular filters. It was therefore concluded that the material will perform effectively as a filter layer and does not need to be replaced. The grading specification was only a means of ensuring that the required performance criteria of the material will be achieved in practice.

The liner section from which the sand was previously removed was damaged in places.

This was most likely caused by the use of metal spades during the removal of the filter layer. It is recommended that this section be covered with filter sand as soon as the damage has been repaired.

62

15. REFERENCES ASTM 6747: Standard Guide for Selection of Techniques for Electrical Detection of Potential Leak Paths in Geomembranes, November 2002 ASTM D7002-3: Standards Practice for Leak Location on Exposed Geomembranes using the Water Puddle System, 2003 GAA Technical Memorandum: Kingston Vale Drainage Sand Bulk Sampling Method Statement, July 2008 Golder Associates Africa: Drainage Sand Bulk Sampling Kingston Vale Residue Management Facility, Report No: 3138-8204-198, October 2008 Peter Legg, September 2008 “Kingston Vale Residue Management Facility Investigations of Drainage Systems Memorandum” Semmelink CJ. “Statistical Sampling Method Selected in the Evaluation of the filter sand quality and possible damage caused by stone found in the filter sand at the Kingston Vale Residue Management Facility”, September 2008.

63


GEOSYNTHETIC CLAY LINERS (GCLs) – THE EFFECTS OF PEEL STRENGTH ON

PERFORMANCE Clint Mackenzie and Neville du Toit Kaytech Engineered Fabrics, PO Box 1658, Dassenberg, 7350, South Africa SUMMARY Reinforced Geosynthetic Clay Liners are used in applications where the shear strength of the GCL is critical. The Peel Strength of the GCL can be related to its shear strength and is measured by using test methods like ASTM D6496 – 04a. Inter-laboratory Studies have established the repeatability (within laboratory) and reproducibility (inter-laboratory) of the Peel Strength results. ABSTRACT The test method that is currently used internationally by manufacturers and accepted by designers and end-users, is ASTM D6496 – 04a (Reapproved 2009) (Standard Test Method for Determining the Average Bonding Peel Strength between the Top and Bottom layers of a Geosynthetic Clay Liner). This standard test method makes allowance for variations in the recorded test results. The variation in the recorded peel strength test results during testing by either the manufacturer’s laboratory, as well as the results obtained by external third party laboratories, will be discussed. In this design, not only should the peel strength be of prime consideration, but additional manufacturing processes and testing should also be taken into account. The discussion will, in addition to the peel strength tests, also touch on the effects of the needling together of the two geotextiles through the bentonite powder layer to form the composite product; the hydrated internal shear strength of the GCL and its importance on embankment slopes; the raw materials that are used and their possible influences, including the nonwoven cover geotextile; the woven carrier geotextile and the natural sodium bentonite, and finally the effect of ‘Thermal Locking’ of the nonwoven geotextile filaments to the woven geotextile. Manufacturing Quality Assurance (MQA), is considered, which also plays a very important role in obtaining consistent and accurate test results. The values for the test results used in this discussion are based on within-laboratory and between-laboratory studies carried out by the manufacturer and an independent third party external laboratory. 1. INTRODUCTION The average bonding peel strength of GCLs (ASTM D6496-04a) formulates the calculation of the average force per unit of width required to separate the geotextiles on either side of the clay component of a GCL. “Several researchers (Heerten et al., 1995; Richardson, 1997; Olsta and Crosson, 1999: Mackay and von Maubeuge, 1999) have correlated GCL peel strength with internal hydrated shear strength. Olsta and Crosson (1999) performed peel testing on needle-punched GCLs and reported that a GCL with a peak peel strength of 65N (100mm wide sample) should result in an internal shear strength of approximately 24KPa, which is sufficient for most low normal

64

load applications.” (Hurst & Rowe. 2006). Internal Hydrated shear testing ASTM D6243 is a very costly and time consuming test and thus not geared for production control on a large scale at high frequencies. “ASTM D6496-04a can therefore very successfully be used as an index test that provides a reliable production control measure at more frequent intervals.” (Hurst & Rowe 2006). That in turn provides an indication of the hydrated internal shear strength. Essentially the difference between the two failures modes in GCLs can be explained as follows: When sliding at the interface between the top or bottom of the GCL and the adjacent material occurs, this is termed ‘interface shear’. However, when sliding occurs through the contained bentonite between the geotextiles in the composite, then this is termed ‘internal shear’. To the authors’ knowledge, “no full scale field failures related to the internal shear strength of reinforced GCLs have been reported.” (Zornberg. 2009). The significance of this is that peel strength has to be adequately controlled by the manufacturer in order to ensure that the required internal hydrated shear strength is achieved. The focus of this paper is on the average bonded peel strength of needle-punched reinforced GCLs. We made an in depth study of the ASTM D6496–04a test. Included in the study is the consideration of the inter-laboratory test program, (ILS – Inter-laboratory Studies), the interpretation of repeatability and reproducibility as well as their limits. The research on the title subject of this paper revealed that numerous studies on “The effects of peel strengths on Performance” have been carried out. Thus there is the need to focus on obtaining an in depth understanding of the peel strength test and the interpretation of Peel Strength results. 2. DISCUSSION

This paper focuses on needle-punched; Thermal Lock®

GCLs. Needle punching reinforces the otherwise weak layer of bentonite clay. Unreinforced bentonite has a weak shear resistance and is susceptible to shear failure, even on gentle slopes. The needle punching process consistently reinforces the bentonite layer with thousands of high tenacity fibres that resist and transfer the shearing stresses into the encapsulating geotextile. The nonwoven cover geotextile is thus

effectively bonded to the woven carrier geotextile and the Thermal Lock®

consolidates this bond also providing a rough frictional, interface surface. The GCL used in the study presented in this

paper is Envirofix®

X800. A detailed cross section of Envirofix®

is shown in Plate 1 below.

Plate 1

65

“The need to consider ‘precision’ arises because tests performed on presumably identical materials in presumably identical circumstances; do not in general, yield identical results. This is attributed to unavoidable random errors inherent in every measurement procedure; the factors that influence the outcome of a measurement cannot be completely controlled. In the practical interpretation of measurement data, this variability has to be taken into account. For instance, the difference between a test result and some specified value may be within the scope of unavoidable random errors, in which case a real deviation from such a specified value has not been established. Similarly, comparing test results from two batches of material will not indicate a fundamental quality difference if the difference between them can be attributed to the inherent variation in the measurement procedure.” (ISO 5725-6. 1994). The assessment of “Precision” is carried out in practice by calculating repeatability and reproducibility. In terms of the ASTM D6496-04a repeatability is represented as a limit, CVSr%, and is defined as within-laboratory variability of results. For a 95% confidence limit this repeatability limit is multiplied by a factor. Between-laboratory reproducibility is represented as a limit, CVSR%, and is defined as the variance in results between-laboratories. For a 95% confidence limit this reproducibility limit is multiplied by a factor. “The best available estimate of the true value of the characteristic under study is m the overall mean of all the results. A single reported value, y will not in general, be equal to m. The overall error y-m contains two parts, e and B, where e is the deviation of y from the mean of a large number of results. B is the deviation between the conceptual mean and the overall mean.” (ISO5725-2.1994). See diagram 1 below.

Diagram 1

2.1 INTERPRETATION OF PRECISION AND BIAS The following extract is taken from the ASTM D6496-04a standard test method:

66

12. Precision and Bias 12.1 Inter-Laboratory Test Program - an inter-laboratory study of this test method was run in 2002/2003. Four different geosynthetic clay liner samples were distributed to ten laboratories. Three sets of test results were generated for each sample by each of the laboratories. 12.2 Test Results - the precision information is given in Table 1. The results are presented for the average of the coefficient of variation, CV %, for the four samples. 12.3 Bias - the procedure in this test method for measuring the bonding peel strength of the geosynthetic clay liner has no bias because the values of the bonding peel strength can be defined in the terms of this test method.

TABLE 1 Test Results Statistic ILS Average Within laboratory repeatability limit, CVSr% 7 Between laboratory reproducibility limit, CVSR% 11 95 % confidence limit within laboratory repeatability, CVr% 18 95 % confidence limit between laboratory reproducibility, CVR% 32 The following interpretation of Table 1 above was done based on: ASTM Standardization News May/June.2009 DataPoints What Are Repeatability and Reproducibility? Part 2: The E11 Viewpoint1 By Neil Ullman http://www.astm.org/SNEWS/MJ_2009/datapoints_mj09.html

Using a specified value of minimum average roll value, (MARV), peel strength of 360N/m, taken from the specification GRI-GCL3 REV2 26 July 2010, http://www.geosynthetic-institute.org/grispecs/gcl3.pdf . If an Inter-Laboratory Study using ASTM D6496-04a yields a result of 400N/m the following is the interpretation of Table 1 on Precision and Bias from ASTM D6496-04a:

1. A result is the average obtained after testing 5 off 100mm x 200mm specimens as per

ASTM D6496-04a. 2. If we had an inter-laboratory average result of 400N/m for Peel Strength from various

laboratories, and we had a single operator in one laboratory run many tests on that material, then 95 percent of the results would fall within a range of approximately ±55N/m (or about 1.96 times the repeatability standard deviation Coefficient of Variation in ASTM D6496 Table 1 ILS CVSr=7% i.e. a total range of about 110N/m).

3. But if only two results were obtained at random, (in one laboratory; CM & NDT), then 95

percent of the time the difference between those two results should not be more than 72N/m (the repeatability limit in ASTM D6496 Table 1 ILS CVr=18%).

4. Similarly, if many laboratories obtained a single result then 95 percent of the single results

would fall in a range of about 144N/m (± the repeatability limit in ASTM D6496 Table 1 ILS CVr=18%), but pairs of results would rarely have a difference of greater than 128N/m (the reproducibility limit in ASTM D6496 Table 1 ILS CVR=32%).

2.2 WITHIN-LABORATORY STUDY In order to investigate the repeatability values laid out in the ASTM D6496-04a, an internal study of

results within a single laboratory was carried out on Envirofix®

X800. A Within-Laboratory Study Procedure was used (annexure 1), for the testing in order to ensure quality and validity of the data.

67

The data was gathered from two samples immediately adjacent to one another. Thirty individual specimens were carefully cut, marked and data recorded from each sample and the results are represented in Table 2 below. The data in the table from the manufacturer’s production laboratory yields an overall repeatability variance, (CVSr%), of 4%. This value is within the repeatability variance as reported in the ASTM D6496-04a ILS. Similarly at 95% confidence level this repeatability variance is within the variance as reported in the ASTM D6496-04a ILS.

Table 2

Kaytech Laboratory Repeatability Variance

Kaytech RU1 Peel

strength result (N/m)

Kaytech RU2 Peel

strength result (N/m) Average

Variance between averages

Test 1 Result 837 834






Average 808 799 804 37

Variance, Sr2 1064 727 896

CVSr% 4%

ASTM ILS CVSr% 7%

CVSr% (95% confidence) 10%

ASTM ILS CVSr% (95% confidence) 18%

2.3 INTER-LABORATORY STUDY In order to investigate the reproducibility values laid out in the ASTM D6496-04a, an inter-laboratory study of results between two laboratories was carried out on Envirofix® X800. An Inter-Laboratory Study Procedure was used (annexure 2), for the testing in order to ensure quality and validity of the data. The data was gathered from two samples immediately adjacent to one another. Thirty individual specimens were carefully cut, marked and data recorded from each sample and the results are represented in Table 3 below. The data in the table from the Inter-Laboratory Study yields an overall reproducibility variance, (CVSR%), of 10%. This value is within the reproducibility variance as reported in the ASTM D6496-04a ILS. Similarly at 95% confidence level this reproducibility variance is within the variance as reported in the ASTM D6496-04a ILS.

68

Table 3 Between Laboratory, Kaytech and TRI, Reproducibility Variance Kaytech

Peel strength result (N/m)

TRI Peel strength result (N/m) Average

Variance between Averages







Average 525 583 554 1710

Variance, Sr2 957 2045 1501

Between Lab Variance, SL2 1459

Reproducibility Variance, SR2 2961

CVSR% 10%

ASTM ILS CVSR% 11%

CVSR% (95% confidence) 28%

ASTM ILS CVSR% (95% confidence) 32%

3. CONCLUSION No known failures, outside of test laboratories, related to the internal shear strength of reinforced GCLs have been reported. Average bonded peel strengths of GCLs can be directly correlated to internal hydrated shear strength hence the necessity to understand and control the peel strength of reinforced GCLs.

This paper focuses on the understanding and interpretation of ASTM D6496-04a test method. When manufacturing and designing GCLs the understanding of the reproducibility and repeatability within test methods is a factor that must be considered. The comparison of results between a manufacturer and an external independent laboratory will have a variation. An allowance has been made in the ASTM test method in order to deal with this variation. The Authors thus concur with “comparing test results from two batches of material will not indicate a fundamental quality difference if the difference between them can be attributed to the inherent variation in the measurement procedure.” (ISO 5725-6. 1994). The within and inter-laboratory studies carried out in this paper confirm that the Table 1 values published in the ASTM D6496-04a test method are indicative of the studies carried out in this paper. 4. REFERENCES ASTM D6496 – 04a (Reapproved 2009) – Standard Test Method for Determining Average Bonding Peel Strength between Top and Bottom Layers of Needle-punched Geosynthetic Clay Liners

69

ASTM D6243-09 Test Method to Determine the Internal and Interface Shear Resistance of Geosynthetic Clay Liners by the Direct Shear Method ISO 5725-6 Accuracy (trueness and precision) of measurement methods and results - Part 6 Use in practice of accuracy values ISO 5725-2:1994 Accuracy (trueness and precision) of measurement methods and results – Part 2: Basic method for the determination of repeatability and reproducibility of a standard measurement method Geosynthetic Institute – GRI-GCL3 – (2010) – Test Methods, Required Properties, and Testing Frequencies of Geosynthetic Clay Liners (GCLs) Jorge G Zornberg – Advances in the use of Geosynthetics for Waste Containment Freilich B.J, Li C and Zornberg J.G. – Effective Shear Strength of Fibre-Reinforced Clays Kent P von Maubeuge, Henning Ehrenberg (1999) - Comparison of Peel and Shear Tensile Test Methods for Needle-punched Geosynthetic Clay Liners Jorge G Zornberg, M.ASCE, John S McCartney, S.M.ASCE and Robert H. Swan Jr – Analysis of a Large Database of GCL Internal Shear Strength Results P.J.Fox and T.D. Stark (2004) – State-of-the-Art Report: GCL Shear Strength and its Measurement H.T.Eid, T.D.Stark and C.K.Doerfler – Effect of Shear Displacement Rate on Internal Shear Strength of a Reinforced Geosynthetic Clay Liner George R. Koerner, Ph.D., P.E., CQA Associate Director, Geosynthetic Institute – Further comments on the concept of MARV. P. Hurst, R.K. Rowe – Average Bonding Peel Strength of Geosynthetic Clay Liners after Short-term Exposure to Water and Jet Fuel A-1.

70

Annexure 1 MEASUREMENT PRODUCT VARIABILITY TESTING PROCEDURE FOR DESTRUCTIVE TESTING WITHIN-LABORATORY

1.) Organize the test specimens and keep them in order during testing. The specimens must be tested “left-to-right” across the roll width, do not test the specimens randomly.

2.) Cut twelve (12) test specimen groups ordered “left-to-right” across the sample roll. Each

test specimen group must have the number of test specimens dictated by the test standard (5 specimens per group).

3.) Test specimen groups “RU 1” under single operator, single day, single apparatus

conditions. Record temperature, apparatus, grip pressure, grip faces, any other “adjustable” test attributes and the technician. Do NOT vary any test parameters during testing, such as grip faces, grip pressures, specimen cutting punch/knife size etc.

4.) Test specimen groups “RU 2” on same day with same operator and the same technician

under the same conditions for “RU 1”.

5.) Asign specimen groups RU 1 and RU 2 for ASTM D 6496-04a. Test all Groups at Kaytech Production Laboratory.

TEST SPECIMEN LAYOUT

Machine Direction

Destructive testing includes all mechanical tests and any test where the test specimen is ruptured or in any way altered by the testing procedure.

RU 1 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

RU 2 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

LEFT RIGHT

71

Annexure 2 MEASUREMENT PRODUCT VARIABILITY TESTING PROCEDURE FOR DESTRUCTIVE TESTING WITHIN-LABORATORY

1.) Organize the test specimens and keep them in order during testing. The specimens must

be tested “left-to-right” across the roll width, do not test the specimens randomly. 2.) Cut twelve (12) test specimen groups ordered “left-to-right” across the sample roll. Each

test specimen group must have the number of test specimens dictated by the test standard (5 specimens per group).

3.) Test specimen groups “RU 1” under single operator, single day, single apparatus

conditions. Record temperature, apparatus, grip pressure, grip faces, any other “adjustable” test attributes and the technician. Do NOT vary any test parameters during testing, such as grip faces, grip pressures, specimen cutting punch/knife size etc.

4.) Test specimen groups “RU 2” on a different day with different operator, different technician

in another Independent laboratory. Record temperature, apparatus, grip pressure, grip faces, any other “adjustable” test attributes and the technician. Do NOT vary any test parameters during testing, such as grip faces, grip pressures, specimen cutting punch/knife size etc.

5.) Asign specimen groups “RU 1” and “RU 2” for ASTM D 6496-04a. Test Groups “RU 1” at

Kaytech and send groups “RU 2” to Independent accredited laboratory, (TRI), for testing.

TEST SPECIMEN LAYOUT

Machine Direction Destructive testing includes all mechanical tests and any test where the test specimen is ruptured or in any way altered by the testing procedure

RU 1 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

RU 2 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

LEFT RIGHT

72


ADDRESSING AN INTEGRATED APPROACH TO THE USE OF ALTERNATIVE

TECHNOLOGIES AS AN ECONOMICALLY VIABLE ALTERNATIVE TO LANDFILL

Christopher Whyte B.Sc.(Hons); Dip.S.Bus.Man.; M.Sc. (GI, London); Pr,Sci.Nat

Managing Director of USE-IT: eThekwini Waste Materials Recovery Industry Development Cluster Office 44, The Village Centre, 59/61 Old Main Road, Hillcrest, 3650 Tel: (031)7652349, Email [email protected] Summary:

Landfill space is being rapidly depleted in South Africa and the creation of new landfill sites is becoming increasingly problematic. The adoption of globally tried and tested applications in landfill diversion through waste beneficiation methodologies can not only extend the life of landfill, but can also result in the development of a new, vibrant and sustainable economy.

Abstract

There is a dichotomy in attitudes towards waste management – that of “landfill is best” and the proponents of “zero waste”. Somewhere in between these two extremes is an entire plethora of sound ideas and philosophies suggesting a balance between the two by taking an integrated approach to waste management through the appropriate application of alternative technologies that beneficiate waste.

As a developing country with very specific socio-economic conditions to deal with, South Africa needs to look at lessons learned from the global industry’s current benchmarks in maximizing landfill reduction. We need to take a critical look at our own limitations in terms of economic viability factors when dealing with waste and contextualize similar limitations in terms of foreign direct investment and public/private sector involvement. There are a number of globally tried and tested alternative technologies that are simply not viable under local economic conditions. On the other hand there are as many local and international technologies that could be ideal for local application and provide the solutions we need to reduce landfill volumes.

Taking an integrated approach requires a better understanding of the different components of the waste stream. We can show how we can create specific interventions in the waste stream that will result in significant waste diversion whilst creating economically sustainable business opportunities in a manner that fits the national directives of job creation and greening the economy. Projects are currently being implemented locally through the Non-Profit Organisation called USE-IT: the eThekwini Waste Materials Recovery Industry Cluster, where the focus is developing a waste beneficiation value-chain. This initiative is addressing opportunities available in recycling and waste beneficiation using local knowledge and international best practice to implement realistic projects under local conditions. The concept of Zero Waste in South Africa will not be achieved in the short to medium-term and there will be a need for landfills for some time to come – in the interim we need to take small steps towards landfill diversion in a manner befitting a vision towards sustainable economic development.

73

Introduction The Polokwane Declaration of September 2001 (DEAT, 2001) was signed by representatives from the business community, civil society and government and recognized the “need for urgent action to reduce, re-use and recycle waste in order to protect the environment”. The main goals of this declaration (DEAT, 2001) were to:

Reduce waste generation by 50% by 2012 Reduce waste disposal by 25% by 2012, and Develop a plan for zero waste by 2022

To date our achievements of these goals have been somewhat lacking. The reality is that since the signing of this declaration, the rates of waste generation have increased substantially. As a result of population and economic growth, the levels of waste generation are expected to increase at 2-3% per year (Fiehn & Ball, 2005). It is more likely that combined with the issues of globalization changing our consumer habits and the accelerated rates of urbanization (Urban Studies, 2004) that the country is dealing with, that this figure is more likely to be in the region of 4-5%. At these rates we have essentially increased the amount of waste going to landfill by up to 50% since the signing of the Polokwane Declaration ten years ago. To get exact volumes of solid waste disposal in South Africa is also difficult, with estimates ranging from 15 million tons per annum (Mt/a) (DEAT, 2006), to 24.1 Mt/a (Purnell, 2009) up to 42 Mt/a (Wuite & McKenzie, 2004). The variances in estimates relate largely to the poor state of record-keeping at most of our landfill sites (PPDC, 2004). The rate of urbanization is going to complicate matters. In 1975 the percentage of population living in urban areas was 48% (Urban Studies, 2004), increasing to 57.6% in 2001 and expected to reach 67.2% in 2015. The current rate of urbanization is estimated to be 1.2% per annum with the current status of total population living in urban areas at 62% in 2010 (Index Mundi, 2011). Waste generation per capita differs across the socio-economic profile with low, middle and high income groups generating 0.41, 0.74 and 1.29 kg/capita/day respectively (Fiehn & Ball, 2005). The General Household Survey (Stats SA, 2007) indicates that 39% of households do not receive regular waste collection. With a current population of 50.59 million (Stats SA, 2011), if we multiplied the median waste generation factor above we would be looking at in the region of 35 Mt/a, so our total waste going to landfill considering the percentage of households not receiving waste clearance services, is more likely to be in the region of the 24.1 Mt/a as indicated by Purnell (2009). The critical issue we face if taking the projection of Wuite & McKenzie (2004), is that at the current rate of disposal, by 2009 waste generated will exceed landfill capacity in five of the nine provinces by up to 67%. Thus, even though landfills will still play a critical role in waste management for South African municipalities, we have already exceeded the critical threshold at which we need to implement alternative waste management procedures that divert and process waste to save what little landfill space we have left.

This is by no means a new issue and both the private and public sector have been critically aware of the looming crisis for at least a decade (DEAT 2000 & 2001), and yet we have still done little to rectify the situation. We need to urgently change our mindset to look at regarding waste as a resource rather than its current form as a liability in landfill. The international community has multiple benchmarks in sustainable landfill diversion that we as a country can strive to achieve. Waste beneficiation technologies take many forms from the large-scale mass incineration platforms employed in many countries to generate electricity, to waste-specific interventions that extract greater value out of the waste stream and create many more jobs. In order to unpack the opportunities we need to understand the waste stream. Specific waste information at landfills is somewhat limited as we tend to group waste into very general categories. Nonetheless, there are generic characteristics of the waste stream generated locally that we can use as a starting point.

The Waste Stream Characteristics of the waste stream tend to vary according to the socio-economic status of the areas contributing to a landfill (DEAT, undated). Globalization, urbanization and the trend towards a more consumerist society are tending to level out these socio-economic variations on type of waste, but not volume of waste. However, it is difficult to get fixed contributions of different waste

74

streams without doing detailed waste audits at each of the landfill sites. For the purposes of this paper, one can work to average estimated urban waste compositions indicated below:

Waste Fraction Percentage

Green waste 12-15

Builder’s rubble 10-15

Soil waste and soil cover material 15-20

Paper and cardboard 7-13

Plastics 5-9

Glass 3-7

Metals 2-5

Sawdust and wood waste 5-7

Textiles 1-4

White waste and electronic waste 1-2

Tyres 1-2

Other 10-30 In the last category we tend to clump mixed wastes that are largely organic (food and organic wastes from domestic sources), but more difficult to process than the clean green waste emanating from gardens, markets and municipal clearing functions.

Understanding some of the more general landfill waste stream characteristics, we can identify immediate opportunities for diversion that can help us to achieve the diversion requirements of the Polokwane Declaration. Green waste and builder’s rubble are both generally pre-sorted before they get to landfill and would be easy to divert and process separately – this alone could achieve a 22-30% reduction. Soil waste and soil cover material also comprise a large percentage of the waste stream, but are currently used as cover in compliance with the permit requirements of landfill to cover the waste. However, if we implement diversion schemes for the other waste streams, the requirement for cover material would also decrease accordingly and the costs of importing cover material would similarly decrease. Extraction of the other recyclable waste streams (paper & cardboard, glass, plastics, metals, e-waste and tyres) would require slightly more complex interventions such as the establishment of Material Recovery Facilities (MRF’s), but the ultimate would be to implement stringent kerbside recovery systems over time to reduce the contamination of this waste before it reaches landfill.

The other aspect is to look at is landfill airspace versus percentage contribution to mass. If we look at an average weight composition of plastics at 7%, this can comprise more than 18% by volume due to the low density of the material. This is exacerbated by the fact that the plastics in a landfill take hundreds of years to break down. As such we should be specifically targeting plastics for recycling. Glass and metals are some of the easiest wastes to recycle and already have well established markets for offtake. Paper is a little more problematic to deal with at landfill as contamination renders a proportion unrecyclable, but this can still be used with green waste in composting or incineration. Considering all of the above, we can show that diversion from landfill into recycling and waste beneficiation can deal with the bulk of waste going to landfill. The next step is to identify economically viable solutions that are able to use this waste that are practical for South African applications and can contribute meaningfully to job creation. At least 50% of waste going to landfill can be diverted, and this can increase to more than 80% if we employ more complex and expensive systems of energy extraction.

75

Solutions For those involved in landfill operations, they should by now be fully aware of the waste hierarchy which classifies waste management strategies according to their desirability as illustrated in Figure 1. The most favoured options being the prevention, minimisation and reuse of products, followed by recycling and then energy recovery, followed by the least favourable option of disposal via landfills.

Figure 1: The Waste Hierarchy (source: www.wikipedia.org)

There is little doubt that there will be a need for landfills for some time to come in South Africa considering that our more developed neighbours such as the USA and the EU have only reached 33.4% and 41% of the total waste stream recycled respectively (Friends of the Earth, 2010). Nonetheless, South Africa needs to strive towards increasing its diversion objectives to catch up with international best practice. Considering that landfill operations are already a liability for municipalities to deal with, we need to ensure that solutions implemented are economically viable, sustainable and environmentally acceptable.

There are a number of globally and locally tried and tested products and processes that can be applied to specific sectors of the waste stream. Local regulation, certification and bureaucracy tend to stifle access to waste streams and implementation of innovative solutions, and this does require some attention to unlock potential benefits from waste diversion and beneficiation. In order to understand some of the opportunities available, several specific focus areas will be addressed, inter alia:

Bio-Organic fertilisers and accelerated composting Recycled composite manufacture Builder’s Rubble & Compressed Earth Blocks Tyres and Rubber Electronic Waste Plastics Glass

Each of these are being addressed by a Non-Profit organisation established in eThekwini Municipality under a Memorandum of Agreement with the city’s Economic Development Unit, called USE-IT: the eThekwini Waste Material Recovery Industry Development Cluster. The focus of this NGO is to establish a waste materials recovery value chain for the city, as well as addressing similar opportunities for the Province of KwaZulu-Natal.

Bio-Organic Fertilizers and Accelerated Composting:

One of the largest waste streams we have in any of our urban centres is organic waste which can comprise more than 40% of the total MSW (DEAT, Undated). Organic waste can be converted to composts for a variety of applications from landfill cover to urban gardening. There are often issues with use of this MSW-derived compost because of potential contaminants, but the reality is that a large percentage of this can be considered “clean-greens” such as garden waste and market

76

waste. In many municipalities this comprises between 12 and 15% of the total daily waste stream. The other problem faced is that compost is generally a low-value product that needs local markets as transport costs make its distribution unviable. The opportunity is to further beneficiate this product with specific microbial and nutrient additives to make a Bio-Organic Fertilizer (BOF). A BOF can have at least 5 times the economic value of compost (up to 12 times the value as a bagged and exported commodity), and the processing of this is accelerated and controlled to eliminate leachate, odour and vector issues so the process can be done in an urban setting. BOF is an organic replacement for chemical fertilizers saving emissions and contributing to a more sustainable future. Chemical fertilizers cost in excess of R4000 per ton, whereas BOF can be manufactured locally and distributed for under R1000 per ton. Considering that South Africa is a water-poor region, the additional benefit of crops grown using BOF is that higher water-retention capacity can be achieved and virtually no nutrients are leached into our streams as nutrients are locked into the soil matrix and only bio-available. Large scale composting and BOF operations from this waste stream can generate thousands of jobs in South Africa and contribute positively to both the economy and the environment. There are also opportunities for Carbon Credits from composting operations as diversion and processing save landfill space, avoid methane production and replace energy-intensive and high-emission chemical fertilizers. BOF is generally produced by accelerated composting using the addition of effective micro-organisms and then enhanced with additives that can include manure, chicken litter, sewage sludge, natural lime, bio char and volcanic ash as bulking agents with nutrient value. Using a controlled process of thermophyllic aerobic bacterial decomposition ensures that the final product contains no seeds or pathogens and is suitable for both domestic and large-scale agricultural markets.

Recycled Composite Manufacture: One of the fastest growing trends in international “green” manufacturing is the use of different recycled products back into environmentally-friendly products. Using blends of recycled commodities helps to create a market offtake for recycled products that drives recycling. One of the biggest composite manufacturing opportunities is Wood-Plastic Composites (WPC’s) where a blend of plastics and wood waste are used to create products that replace standard timber applications with the benefit of being stronger, longer-wearing and resistant to mould, borers, rot and other issues that affect timber. Main international applications are decking, balustrades, cladding and edgings. Other composites include combinations of plastic, rubber, glass and wood. Roof tiles and manhole covers can be made from plastics and glass, rubber composites can be used to make roof tiles, and different plastic composites can be used as building components or automotive components. These are all high-investment green technologies that create many jobs. USE-IT is currently facilitating one small project that would use 370 tons a month of plastics and 600 tons per month of glass (0.5% of the monthly waste stream in eThekwini) that can create 390 direct jobs. Total estimated potential for composites manufacture for eThekwini alone would be in excess of 1000 jobs and the creation of a half a billion Rand Industry.

Builder’s Rubble and Compressed Earth Blocks (CEB’s) USE-IT is currently piloting a project on the manufacture of CEB’s for use in low-cost housing. The products currently being manufactured are a blend of available soils and a 30% blend of crushed builder’s rubble which are achieving a 9.4Mpa average compression rating (SABS approved for double-storey structures). The building products are less than half the cost of concrete blocks, but are stronger and more thermally efficient. In addition, the building process is much simpler and faster saving up to half the costs on the actual building process as well. The aim of this project is to provide a viable alternative to providing cost-effective, thermally-efficient high-quality homes for the low-cost housing shortage in the country. South Africa landfills enough builders’ rubble each year to produce in excess of 160,000 40m2 houses per year using a 30% blend of crushed rubble. USE-IT is currently completing the Agrément Certification for the CEB’s, and once done will address the commercialisation of the project for eThekwini and other municipalities in the country. The potential for job creation for this component of waste beneficiation, product manufacture and construction is in excess of 5000 jobs.

77

There are other uses for crushed builder’s rubble if centralised crushing and screening facilities are implemented as different size fractions can be diverted to crusher-run for roads maintenance and construction, CEB manufacture, compacted foundation stone for construction and as pipeline bedding material.

Tyres and Rubber

Eleven million waste tyres are produced in South Africa each year (SA Tyre Recycling Process Company at www.rubbersa.com) and with new legislation governing the collection and recycling of these tyres (DEAT) there is an opportunity for processing and beneficiation. Although used tyres are a small fraction of the total waste stream, they are a problematic waste to deal with. Cutting, granulating and de-beading (removing metal) tyres creates a product that has a value in excess of $200 per ton on the international market – but the concept would be to take beneficiation and remanufacturing to the next step. The opportunity is to create downstream value-add applications that use the granulated rubber in new composite manufacturing opportunities. In eThekwini alone it is estimated that tyre processing could attract R35 million investment and create 100 jobs, but remanufacturing into new products could attract three times this investment.

Electronic Waste Electronic Waste or e-waste is also a small component of the waste stream, but potentially one of the most hazardous. USE-IT has partnered with the Electronic Waste Association of South Africa (EWASA) to implement a take-back scheme for e-waste and a closed-loop recycling initiative for the greater eThekwini area. The idea initially would be to create a local recycling and refurbishment centre that could create in excess of 60 jobs. Thereafter, the intention is to establish waste beneficiation applications for all components of this waste stream – metals, plastics, printed circuit boards, CRT monitors and printer cartridges that could justify hundreds of new jobs and greater investment. The current vision is to create eThekwini as an e-waste hub for Africa.

Plastics Plastics are hugely problematic at landfill as whilst they only make up about 7% of the waste stream, they comprise 18% of the volume going to landfill. The applications for recycled plastics are numerous, from primary washing, granulating and pelletizing, to downstream remanufacturing of new environmentally-friendly plastic and composite components. The initial requirement would be to extract the plastics from the waste source and then to sort it into the different polymers.

The primary issue with plastic waste is that there is a high level of contamination and also unprocessed plastic waste has a low value. As with other recycling initiatives, there are only small margins to be made in recycling (collecting) plastic unless it is taken at least part of the way through the beneficiation cycle. An option currently being mooted is to pool resources from Material Recovery Facilities, curb side recycling, community collectors and Community Buy-Back Centres to increase volumes and then channel these through a central processing centre (sorting, washing, granulating, pelletizing) before distribution to existing plastics manufacturers or development of new composite manufacturing plants.

In eThekwini Municipality it is estimated that in the region of 350 tons per day of plastics are landfilled. The best scenario would be to try and target this waste stream before it gets mixed with the general waste – creating hundreds of informal jobs in the collection of waste (which would need to be backed up by a guaranteed offtake market through buyback centres at a standardised best-industry-price). Materials recovery at landfills could divert what gets through the primary collection. All waste could then be processed centrally where profit-sharing options would be redirected to the collectors and operators. Unusable and contaminated plastics could be processed as Refuse Derived Fuel (RDF) for cement kilns, or even more preferably diverted through a catalytic depolymerisation system that would convert the waste to a raw diesel that would power the collection trucks and backup generators for the processing plant. The plastics waste in the

78

eThekwini area has the potential to create more than 500 jobs and a three hundred million Rand Industry all on its own. To achieve this requires an integrated approach unlocking all opportunities in the value chain.

Glass Glass recycling is widely publicised as the most practically and successfully recycled commodity in South Africa – the reality is that in areas remote from the main glass smelters like eThekwini, glass recycling is not very successful and this is largely due to the low value by weight and the transport requirements to its only market in Johannesburg. Glass comprises only about 5% of the waste stream, but this implies that eThekwini landfills in the region of 250 tons of glass every day. The focus on glass recycling needs to be concentrated on creating more efficient collection and distribution systems both at source and from the landfills. Thereafter, the economic imperatives need to be designed around creating local beneficiation opportunities that avoid the transportation issues to distant markets and focus on local job creation. There are a number of beneficiation opportunities for glass and these include the remanufacture of new glass products, application in the ceramics and paint industries, sand blasting, mix as a composite material in roof tile manufacture, water filtration, asphalt surface hardening and bead production for local crafts, to mention a few. A similar initiative is planned for glass in integrating approaches to successful recycling and local beneficiation. An integrated approach to glass recycling in eThekwini can create at least 450 jobs.

Energy from Waste There are a multitude of technologies available for converting waste to energy (W2E), but as a country we need to make a conscious decision to steer clear of mass incineration technologies. This by no means negates the application of W2E technologies, but we need to observe the waste hierarchy in putting recycling before energy recovery and be mindful of the fact that these need not be mutually exclusive. There are extensive proven applications internationally extolling the virtues of taking a holistic approach to waste management where one first maximises the returns from resource recovery before extracting energy from the remaining waste fraction. Recycling sustains many more jobs than landfilling or incineration and the value extracted is far higher. There are even international examples of W2E applications which have created an increase in recycling figures (Solid Waste & Recycling, 2011). This is largely due to the process of implementing a Materials Recovery Facility as a primary stage of waste processing before incineration to extract inert materials. Generally once this is in place, it makes sense to extract other recyclable components at the same time. This makes economic sense from two different perspectives – the first is the greater rate of return on recycled products, and the second is savings on capital invested to process total tonnage. At a rate of between US$110,000 and $140,000 per daily ton processed throughput (e.g., a 100 ton per day system costs USD$11-14 million), it makes sense to remove low-energy, inert and high-value products from the waste stream first. Technological innovations in the past decade have realised a number of more cost-effective, higher-efficiency, waste-specific and lower emission W2E applications that render this a viable and effective application as part of a suite of integrated interventions in waste management. Systems and processes are constantly evolving and there are new systems entering the market that are promising lower-efficiency W2E systems, but with the advantage of far lower capital expenditure and the same particulate and emission control systems. Although due diligence still needs to be completed on some of these units, the capital costs could be in the region of 7-10% of conventional W2E technologies. Specific systems for specific waste streams need to be optimally determined using a critical a balance of cost, efficiency and applicability. Catalytic depolymerisation, for example, could be more applicable than gasification or pyrolysis for a feedstock richer in plastics and rubber, whereas carbonisation and heat-conversion systems may be more applicable where biomass dominates the waste stream. Alternatively, anaerobic digestion or fermentation systems may be more suited to waste high in food scraps, sewage sludge, green waste and abattoir waste.

79

Essentially there are no globally accepted ‘black-box’ systems in W2E and solutions need to be optimally designed around the feedstock (and available capital). Jobs With the South African Government’s current push to generate jobs in the country, recycling and waste beneficiation should be receiving priority attention. The current Jobs Fund established by the government aims to generate 150,000 jobs with a R9 Billion fund (www.jobsfund.org.za managed by the Development Bank of South Africa). A simple calculation shows that this equates to R60,000 per job – a difficult task. Whilst the jobs created per million Rand expenditure are in the region of 17-20 for more labour intensive industries such as agriculture and mining, the jobs created per million Rand investment into industry and manufacturing are in the region of 1-3 jobs (The KZN Investment Strategy, in press). If we look at some of our global initiatives for benchmarking, recycling and waste beneficiation are key development and job-creation opportunities. In the United States of America (USA) in 2007 some 254 Mt/a of municipal waste were produced, of which 85Mt/a or 33.4% were recovered for recycling and composting (US EPA, 2008). Nationally, the recycling sector was estimated to generate US$236 Billion in turnover and directly employ over 1.1 million people at 56,000 public and private facilities in 2001 (US EPA, 2002). That equates to one job created per 77.3 tons of waste or 129 jobs per 10,000 tons diverted from landfill and processed annually. Furthermore, the Friends of the Earth (2010) estimates that if we extend the direct jobs to ‘indirect’ and ‘induced’ jobs, this recycling industry is responsible for 3.9 million jobs in the USA. In other words in addition to the economic activity of the recycling and reuse industry itself, further economic activity such as supply companies, accounting, legal, building and transport rely on this industry for business – referred to as ‘indirect’ employment. In addition, the direct and ‘indirect’ employees support another round of economic activity when they spend their wages in the economy – called ‘induced’ employment (Friends of the Earth, 2010). In the USA, for every direct job created across the sector, a further 1.2 ‘indirect’ jobs and 1.3 ‘induced’ jobs were created in the wider economy (Friends of the Earth, 2010). Similar numbers can be derived from the Friends of the Earth (2010) report for the United Kingdom (UK) and the European Union (EU), although a higher level of automation creates less jobs at one per 159 tons for the UK and one per 204 tons for the EU, but these still exceed the 36 jobs per 10,000 tons recycled stated by the Institute for Local Self-Reliance (1997) as industry has become more sophisticated in beneficiation opportunities for waste.

If we took these figures from the USA and extrapolated them to an achievable 50% of the total municipal solid waste in South Africa reported by Purnell (2009) at 24.1 million tons, then we have the potential to create almost 156,000 jobs. If we include the multiplier effect of the ‘induced’ and ‘indirect’ jobs, this number could reach 545,000 jobs. Add to this the potential savings to municipalities in landfill avoidance at a conservative cost of R150 per ton to landfill, and this would save the country an additional R1.8 Billion every year. To take the extrapolation further, if we estimated a very conservative rate of return for recycled products entering back into the market at R300 per ton (basic compost at R150 per ton, glass at R540 per ton and plastics at R3000 per ton as examples), we would be creating an industry turnover worth more than R3.6 billion a year. More difficult to quantify would be the environmental benefits that would accrue by reducing our landfill by half, but reductions in emissions and leachate from landfill should not be overlooked. What is evident from this simple extrapolation is that we may be able to meet the job-creation targets of the national Jobs Fund in the waste sector alone by diverting and beneficiating waste.

80

Conclusions and Recommendations

Landfill as a strategic waste management option is going to be needed for some time to come, but as a country we can ill afford not to explore, invest in and implement a variety of waste diversion strategies and waste beneficiation technologies to unlock the value in the resources we currently bury. There are a number of internationally tested and proven waste beneficiation technologies that we can implement now and we can use this opportunity to reach the job-creation objectives of the South African Government in this sector alone. Not only jobs, but the interventions can allow us to go a long way to addressing service delivery backlogs and strengthen the agriculture, water and environmental health sectors at the same time. There have been too many years of talking about the issues and not actually implementing the solutions we know are readily available. The new Waste Act allows us the opportunity to make this difference now with the Minister able to define ‘priority wastes’ as per section 14 (1) b wherein ‘the imposition of specific waste management measures in respect of the waste may improve reduction, re-use, recycling and recovery rates or reduce health and environmental impacts’ (Republic of South Africa Government Gazette, 2009).

Implementation of waste beneficiation in preference to landfill has multiple effects: saving landfill airspace and costs, creating jobs, diversifying and improving the economy, reducing reliance on imports, improving agriculture, saving water, reducing poisonous leachates and emissions, implementing clean development mechanisms, contributing to a reduction in carbon emissions and global warming, underpinning the services and support sectors and creating a healthier and happier environment for all South Africans to live in. It’s time to make the change.

References

Department of Environmental Affairs and Tourism (DEAT) (Undated). Working with Waste: Guideline on Recycling of Solid Waste. 94pp Department of Environmental Affairs and Tourism (DEAT) (2000) White Paper on Integrated Pollution and Waste Management for South Africa: A policy on pollution prevention, waste minimisation, impact management and remediation. Pretoria. ISBN 0-621-3002-8. Department of Environmental Affairs and Tourism (DEAT) (2001) Polokwane Declaration on Waste Management, Polokwane, Northern Province, 26-28 September, 2001, Available online at www.environment.gov.za/ProjProg/WasteMgmt/Polokwane_declare.htm Department of Environmental Affairs and Tourism (2006). Implementation Plan for Transfer of the Waste Permitting Function. Department of Environmental Affairs and Tourism: Pretoria Fiehn, H and Ball, J. (2005). Background research paper: Waste. South Africa Environment Outlook. National State of the Environment Project. Department of Environmental Affairs and Tourism: Pretoria Friends of the Earth, 2010. Friends of the Earth Report. More Jobs, Less Waste: Potential for Job Creation through Higher Rates of Recycling in the UK and EU. September 2010, 54pp. Available online at www.foeeurope.org Index Mundi (2011). South Africa Demographics profile 2011. Accessed online 14th July 2011 at http://www.indexmundi.com/south_africa/demographics_profile.html. Institute for Local Self-Reliance, Washington, DC, 1997. Available online at www.ilsr.org/recycling KZN Investment Strategy Promotion and Facilitation Imperatives: Investment Strategy and Implementation Plan (in press). Trade & Investment KZN (TIKZN). 44pp. PPDC (2004). A Report for the assessment of waste disposal sites in the province of KwaZulu-Natal. A study commissioned by the Provincial Planning and Development Commission, prepared by SiVEST (Pty) Ltd. 51pp

81

Purnell, G. (2009). National Waste Quantification and Waste Information System. Research Paper commissioned for the National Waste Management Strategy. 47pp. Accessed online 11th July 2011 at http://www.wastepolicy.co.za/sites/default/files/Waste_quantification_WIS.pdf Republic of South Africa, Government Gazette (2009). No. 59 of 2008: National Environmental Management: Waste Act, 2008. Vol 525, Cape Town. 47pp Solid Waste & Recycling, 2011. Solid Waste & Recycling online news, May 23, 2011. Accessed online 23rd June 2011 at http://www.solidwastemag.com/news/us-waste-to-energy-facilities-have-high-recycling-rates-studies/1000440391/?link_source=aypr_S Statistics South Africa (Stats SA) (2007). General Household Survey 2007. Statistical release P0318. Available online at: www.statssa.gov.za. Statistics South Africa (Stats SA) (2011). Mid-Year Population Estimates 2011. Statistical release P0302. Available online www.statssa.gov.za/. United States Environmental Protection Agency (2002). Recycling is working in the United States, January 2002. United States Environmental Protection Agency (2008). Municipal Solid Waste in the United States, 2007 Facts and Figures. Urban Studies (2004). The Human Development Index: Where South Africa Stands, Urban Studies. Accessed online 22nd July 2011 at http://www.urbanstudies.co.za/mar-mayjun04.html Wuite, M. & McKenzie, J. (2004). Defusing the Landfill Timebomb. In: Delivery. The magazine for Local Government, Spring 2004. P 57-58.

82


WHY SPEND MONEY ON INDEPENDENT CQA SERVICES DURING THE

INSTALLATION OF GEOSYNTHETIC LINING SYSTEMS?

Mills B L and Jewaskiewitz B L Envitech Solutions (Pty) Ltd, PO Box 1677, Hillcrest, 3650. Tel: +27-31-7641555 Fax: +27-31-7641555 Mobile: +27 79 4925992. E-mail: [email protected] / [email protected] ABSTRACT The general purpose of a geosynthetic lining system is to ensure minimal or negligible leakage from one entity (e.g. a waste body or lagoon) to an adjacent entity (eg. natural soil or an engineered structure). However, the extensive time, money and other resources expended on the design and development of an effective geosynthetic lining system may all be wasted if the system’s integrity is compromised during the initial construction phase, even before use of the facility has commenced. Research has shown that 97% of all liner defects are directly related to the construction phase of a project, including liner installation and the placement of drainage and protection layers respectively. The use of an on-site Construction Quality Assurance (CQA) organisation can greatly assist in, at the least, reducing, if not completely alleviating, construction and installation related flaws, defects and damages to the geosynthetic barrier system. The importance of on-site, independent, CQA services during the installation of geosynthetic lining systems has been realized since geosynthetic liners became common use in the 1980’s. In some countries in Europe and in North America it is a regulatory requirement to have a third-party CQA officer on-site during the installation of any geosynthetic lining system. The CQA organisation should be independent from the installer, construction company, consultant and even the client. Koerner (1993) describes Construction Quality Assurance or CQA (as opposed to CQC, MQC and MQA) as: “A planned system of activities that provides the owner and permitting agency assurance that the facility was constructed as specified in the design. CQA includes inspections, verifications, audits, and the evaluation of materials and workmanship necessary to determine and document the quality of the constructed facility. CQA refers to measures taken by the CQA organisation to assess if the installer or contractor is in compliance with the plans and specifications for a project.” CQA (and MQA) is performed independently from Construction Quality Control (CQC) and Manufacturer Quality Control (MQC). The essence of on-site, independent, CQA services is to identify possible non-conformances before and during the installation phase and to initiate and monitor all remedial work. Even though a CQA officer has little authority over work progress on site, it is his/her duty to provide sufficient motivation to the contractor to rectify any non-conformances, failure of which could result in non-approval of the project as a whole. This paper will discuss some of the benefits of on-site CQA, consider CQA costs in relation to total project costs and finally illustrate how easily things can (and do) go wrong during the liner

83

installation phase, greatly compromising the integrity of the barrier system. It will conclude that the use of independent CQA services provides the necessary reassurance that every measure had been taken to ensure the installation of a geosynthetic lining system of the highest quality and workmanship.

KEYWORDS

Construction Quality Assurance (CQA), Geosynthetic lining systems, landfill liners, Geosynthetic Clay Liner (GCL), Geomembrane, HDPE, CQA Services, MQA, MQC, CQC

CONTENT

The following subjects will be discussed:

Background and new waste disposal legislation

What is CQA?

What CQA is not

Benefits of having third party independent on-site CQA services during liner construction

Money Talks: The economics of CQA

CQA Hall of Shame (photographs)

BACKGROUND AND NEW WASTE DISPOSAL LEGISLATION

With the advancement in liner technology and scarcity of natural sources, the replacement of natural liner systems with geosynthetic liner systems has become the preferred design model for new waste disposal sites. However, a need has been identified pertaining to the quality of the geosynthetic liner installation, in particular during the installation and construction phase. In North America and many European countries, the use of third party Construction Quality Assurance (CQA) services during barrier system installation is now a regulatory requirement. This is understandable, as research has shown that 97% (Nosko, 1996) of all liner defects are directly related to the construction phase of a project, including liner installation, and in particular the subsequent placement of drainage and protection layers. In 2008 the new National Environmental Management: Waste Act no. 59 of 2008 (NEMWA) was promulgated, and later the National Waste Management Strategy (NWMS) as an implementation aid for the Waste Act. On studying both these documents, one realises the intention to greatly reduce the negative effects of waste and waste related products on the environment and community health. In addition, it is clear that legislative non-compliance will not be tolerated, with penalties of up to R10million and/or jail time. In addition to this, according to the new Draft National Standard for Disposal of Waste to Landfill (Government Gazette, 1 July 2011) the Landfill Classification and Containment Barrier Designs require all landfills (exclusive of inert waste landfills) to have, as the minimum, a clay liner, protective silty sand/ geotextile/ HDPE GSM layer and a geotextile liner as a barrier between the waste body and the in-situ material. The Standard allows for natural layers to be replaced with synthetic liner systems. Section 3 (2) (g) of the Standard also states that ‘Construction Quality Assurance is required on site’. It is therefore clear that all new waste disposal sites in South Africa will require a composite geosynthetic lining system, with the general purpose of the liner being to ensure minimal or negligible leakage from one entity (eg. a waste body or lagoon) to an adjacent entity (eg. natural soil or an engineered structure).

84

WHAT IS CQA? The US EPA Technical Guidance Document: Quality Assurance and Quality Control for Waste Containment Facilities, Koerner (1993), describes Construction Quality Assurance or CQA (as opposed to CQC, MQC and MQA) as: “A planned system of activities that provides the owner and permitting agency assurance that the facility was constructed as specified in the design. CQA includes inspections, verifications, audits, and the evaluation of materials and workmanship necessary to determine and document the quality of the constructed facility. CQA refers to measures taken by the CQA organization to assess if the installer or contractor is in compliance with the plans and specifications for a project.” CQA (and MQA) is performed independently from Construction Quality Control (CQC) and Manufacturer Quality Control (MQC). In essence, the aim of CQA is ensuring that the installation of a geosynthetic liner system is carried out to the highest level of quality and workmanship, thereby preventing possible future remedial work resulting in high financial and resource losses. Some of the most important CQA tasks on GSL installation projects include, but are not limited to: Compilation of a Quality Assurance Plan – This essential document includes a detailed

description of all CQA activities that will take place during construction to manage the installed quality of the facility. It is tailored to each specific facility to be constructed and becomes completely integrated into the project plans and specifications. Any differences between the CQA Plan and project specifications should be resolved before any construction work commences.

Documentation of installation related activities including: o Daily Summary Reports; o Inspection and Testing Reports; o Non-conformance and remediation Reports; o Daily as-built HDPE panel installation layouts; o Photographic record/diary of liner installation; o Final report documentation and certification.

Inspection of: o delivered lining materials; o anchor trenches and sub-grade preparation; o backfill / drainage material; o deployed GCL, geocomposite, geotextile and geosynthetic membrane panels for defects,

physical damage and correct overlapping. Ensuring:

o that relevant MQC documentation is supplied for the delivered material on site; o conformance testing is done if required, and test results are received and scrutinized; o destructive test samples are collected from field seams, tested on-site and results recorded,

and keeping archive samples for the Engineer or Employer; o All site collected samples for archiving are collected and distributed to relevant parties.

Observing: o deployment of lining materials; o seaming pre-weld performance and destructive test results; o and inspecting field seaming of geosynthetic membrane panels; o and verifying non-destructive air pressure testing, vacuum box testing and high-voltage

spark testing of field seams, extrusion welds, patches and repairs; o drainage material back-filling operations; o Weather conditions and recommending work stoppages accordingly.

Open communication through attending and chairing site meetings eg. pre-construction and progress meetings.

Providing advice to the civil contractor with respect to the anchor trench construction. Identification and noting of non-conformances, as well as monitoring rectification work. Checking and confirming the contractors field installation reports.

85

Although most CQA work is focused on project documentation, on-site independent CQA personnel will identify possible non-conformances before and during the installation phase and will initiate and monitor the remedial work. The physical on-site presence of an independent CQA officer should not be under-estimated.

WHAT CQA IS NOT The role of a CQA officer on-site can often be misunderstood. His/her role is not: An authoritative decision maker – often the Employer, design engineers, or the site contractors

look to the CQA officer (as an independent third party) for final decision making – in particular during disputes. Although the CQA officer’s opinion and experience can be invaluable during these times, he/she is not in an authoritative position to override another contracted party’s decision. In extreme cases, should the CQA officer be of the opinion that a particular decision will in no uncertain terms negatively affect the function of the liner system, he/she may decide to approach the Employer directly.

The design engineer – a CQA officer may be able to verify design calculations, provide alternative design options and practical deployment solutions, however, the design engineer is still responsible for the final project design and technical calculations.

Quantity Surveyor – a CQA officer’s role is to verify the quality of an installation, and in no way verifies quantities of installed material for claim/invoicing purposes.

Part of the Installation Team – although a CQA officer must always work very closely with the GSL installer, his/her position must remain independent.

The CQA organization is also not in a position to accept any liability for either the efficacy of the design or the quality of the delivered product to the Employer. These responsibilities rest with the Engineer and Contractor/Installer respectively, and should be covered by the relevant contractual conditions and guarantees to be furnished in terms of the project specification. This should be clearly understood by all parties concerned. BENEFITS OF HAVING THIRD PARTY INDEPENDENT ON-SITE CQA SERVICES DURING LINER CONSTRUCTION The primary benefit of independent CQA services during the installation of a geosynthetic liner system is to provide the Employer with the confidence and assurance that the end product will prevent contaminants from leaking into the adjacent natural soil and/or groundwater with detrimental effects on the environment and community health. Although it is widely accepted that no containment system is 100% leak-free, the use of independent CQA services would ensure a barrier system of the best possible quality and durability. In a paper by B. Forget, A.L. Rollin and T. Jacquelin entitled ‘Lessons learned from 10 years of leak detection surveys on geomembranes’, based on a study of 89 projects totalling 2 652 000 m2, it was concluded that: ‘the average leak density on exposed geomembranes (many types and thicknesses) that were installed under a rigorous CQA program is approximately 4 leaks per hectare. Conversely, the statistics show a sharp climb, to 22 leaks per hectare, in the absence of such a CQA program. The situation was found to be similar with covered geomembranes: a negligible leak density (0.5 leaks/ha) was found on geomembranes installed under a strict CQA program and a prior water puddle leak detection survey on the exposed geomembrane, climbing sharply to a density of 16 leaks/ha in the absence of both a CQA program and the water puddle leak detection survey.’ Essentially, this equates to a leak increase of 5,5 times for exposed geomembranes and an astronomical 32 times for covered geomembranes, if a stringent CQA program is not followed.

86

In addition to the quality assurance duties and responsibilities of the CQA officer including inspection, evaluation, auditing and verification undertaken before, during and after the geomembrane installation process, there are added benefits to having an independent third-party CQA officer on-site. Some additional benefits include, but are not limited to: Site ‘policeman’ – just as the presence of a traffic officer with a speed camera deters motorists

from disregarding the speed limit, the mere physical presence of a CQA officer on site ‘encourages’ the installation team to, at the very least, comply with the project specifications.

Employer ‘representative’ – although most large installation projects will have a dedicated

Employer appointed representative providing liaison between the Employer and project manager, attending meetings and performing site visits, this person often lacks practical installation expertise, and are often not based on the project site. The CQA officer can act as the Employer’s ‘eyes and ears’ on site, with the project goal as the main focus, regardless of other parties involved.

Unofficial dispute resolution – although without contractual authority, the CQA officer is often

approached for an opinion during dispute resolution, and during discussions surrounding practical design and/or installation issues. This can be at any point before or during the installation process, although it is often the last resort before formal dispute resolution procedures are initiated. Informal consultation with the CQA officer can prevent endless delays, thereby saving time and money, and ensuring the project is completed on time whilst achieving the highest possible level of quality.

Independent Project Administration – the completion of ‘paperwork’ is a necessary and vital

aspect of any construction project and many may agree that construction supervisors, in particular, are not well known for efficiently maintaining paperwork of the required standard. Many regard it as ‘a waste of time’, that is, until a dispute or subsequent legal action arises. This may occur during construction, or many years after completion of the project. As a consequence, the project documentation becomes of vital importance, with legal battles being won or lost purely based on this. With the meticulous, systematic and daily documentation of site conditions, project progress, problems encountered etc., the CQA documentation produced during the execution of the project can be used at any stage to provide relevant and factual installation data by any party involved.

Installation Planner – an effective CQA officer will anticipate any foreseeable installation

problems and assist with co-ordination between the earthworks contractor and the installation specialists.

License approval – a licensing agency may wish to view CQA documentation during the site

approval process in order to validate installation materials and workmanship. MONEY TALKS: THE ECONOMICS OF CQA Where independent CQA is not mandatory, either in terms of legislation or the project specification, the short term financial cost of such services will unfortunately take precedence over the environmental risk and future cost benefits of preventing or limiting leakages. The following table – based on actual data for completed projects in the Middle East – illustrates the cost of CQA services in relation to total project costs.

87

Table 1: Cost of CQA Services as a % of Total Project Cost

Installation area

ELL * survey done?

m2 (approx.)

% of Total Project Cost

% of Total Project

Cost

% of GSL Installer

Cost

% of Total Project

Cost

% of GSL Installer

Cost

% of Total Project Cost

1 Al-Ghabawi Landfill Cell 1, Amman, Jordan 180 000 10.0% 1.9% 19.0% Yes 2.0% 20.0% 3.9%2 Al-Ghabawi Landfill Cell 2, Amman, Jordan 180 000 10.0% 1.8% 18.0% Yes 2.0% 20.0% 3.8%3 Al-Ghabawi Landfill Cell 3, Amman, Jordan 180 000 10.0% 2.0% 20.0% Yes 1.4% 14.3% 3.4%4 KEMAPCO Package 5, Aqaba, Jordan 565 000 5.8% 1.0% 17.1% Yes 0.7% 11.4% 1.7%5 KEMAPCO Pond A, Aqaba, Jordan 80 000 5.0% 1.0% 20.0% Yes 1.7% 33.3% 2.7%6 Pearl GTL Waste Pond, Ras Laffan, Qatar 30 000 8.0% 1.9% 23.8%7 NDIA Engineered Landfill, Mesaieed, Qatar 750 000 0.9% 0.2% 21.0%

* 3rd Party Electric Leak Location

Total cost of ELL survey

GSL instal-lation cost to Client

Project DescriptionCost of CQA

services as % of Project Cost

Combined Cost of

CQA & ELL

Third Party CQA & Leak Location Survey Services

Based on the above table, the cost of on-site independent CQA services (provided by an international organization) amounts to approximately 20% of the GSL installation costs. However, when compared to the total project cost, the cost of employing the services of an independent CQA organisation equates to a nominal figure of up to 2%, or up to 4% if an Electric leak Location Survey is included. It can therefore be deduced that the overall economic and environmental benefit achieved through the use of independent CQA services far outweighs the financial costs of these services. Without effective independent CQA, the Employer will be exposed to a high risk of significant potential future costs, delays and work/process stoppages – in particular if the system is in operation – and especially if treatment and remediation of contaminated soil and/or groundwater is required. CQA HALL OF SHAME The photographs below were taken at various project sites, and provide an illustration of typical damages/problems observed through the execution of independent CQA services. All of these were rectified immediately under the watchful eye of the CQA officer on site.

Figure 1: Shipping damage to GCL Figure 2: Manufacturing flaw (GCL)

88

Figure 3: Heavy equipment driving on GCL Figure 4: GCL hydrated during rainfall

Figure 5: HDPE damage (bullet hole!) Figure 6: HDPE equipment handling damage

Figure 7: HDPE stone damage (substrate) Figure 8: Backfilling equipment damage

89

Figure 9: Pinhole leak detection through use of HV spark leak detector

Figure 10: HDPE double fusion weld burn

Figure 11: HDPE thermal contraction Figure 12: HDPE welding flaw due to wrinkling

Figure 13: Stone removal (substrate) Figure 14: Double fusion weld failure

90

Figure 15: HDPE manufacture defect Figure 16: HDPE folding upon backfilling

Figure 17: Construction debris within sand material (rebar!)

Figure 18: Use of a plough for stone backfill removal in sand backfill material above HDPE liner

Figure 19: Boulders within backfill material Figure 20: Stones and debris removed from

backfill

91

Figure 21: Specified ‘sand’ backfill on top and below HDPE liners - rejected by CQA officer

Figure 22: Visible stones within sand backfill

Figure 23 and Figure 24: Damage to geomembrane due to equipment placing backfill material CONCLUSION Even though an independent CQA officer has no contractual authority over work progress on site, it is his/her duty to provide sufficient motivation to the contractor, engineer and Employer, if necessary, to rectify any non-conformances, failure of which could result in non-approval and functional failure of the project as a whole. CQA services provided on-site during the installation of a geosynthetic barrier system should provide the Employer with the confidence and assurance that the best possible quality and durability of the completed lining system will be achieved. The financial cost of on-site independent CQA services is nominal in comparison with the potential environmental and economic costs of failure. The use of CQA would serve to significantly reduce these risks to an acceptable level. REFERENCES Draft National Standard for Disposal of Waste to Landfill, Government Gazette Notice 432 of 2011, 1 July 2011 EPAl600/R·93/182: Technical Guidance Document: QUALITY ASSURANCE AND QUALITY CONTROL FOR WASTE CONTAINMENT FACILITIES by David E. Daniel and Robert M. Koerner; September 1993

92

Forget B., Rollin A.L. and Jacquelin T., “Lessons learned from 10 years of Leak detection surveys on Geomembranes” Nosko, V., Andrezal, T., Gregor, T., and Ganier, P., 1996, “SENSOR Damage Detection System (DDS) – The Unique Geomembrane Testing Method”, Geosynthetics: applications, design and construction, de Groot, M.B., den Hoedt, G., and Termaat, R.J., Editors, Balkema, Proceedings of the First European Geosynthetics Conference EuroGeo1, Maastrict, Netherlands, September 1996, pp. 743-748.

93


ECONOMIC COMPARISON OF LANDFILLING WITH AND WITHOUT

ANAEROBIC PRE-TREATMENT

Gerd Burkhardt and Thomas Egloffstein ICP Ingenieurgesellschaft Prof. Czurda und Partner mbH, Karlsruhe, Germany, [email protected] 1. Introduction The introduction of proper and environmentally compatible waste management is supported financially with the aim of fundamentally improving the infrastructure in many developing and emerging countries by numerous development banks, such as, for example, the German KfW bank group, the EBRD or EIB (development banks of the EU) or by the World Bank. Until a few years ago, fundamental waste management measures such as collection and transport of the waste as well as proper landfilling were sponsored here by the FRG through the KfW bank group and the Gesellschaft für Internationale Zusammenarbeit (giz). These measures can in general be borne economically by the populations in these states, whereas higher quality waste disposal systems including technically more sophisticated facilities, such as e.g. a mechanical-biological treatment plant (MBA), composting or a fermentation plant, seem too expensive. The basis used for this assessment is the principle that the costs of waste disposal should not be more than approx. 1 % of the average income of the population. If it can be seen that waste management measures are too expensive and cannot be borne by the population in the long run, such projects are usually not sponsored either, since this sponsoring would not be sustainable. The new guidelines of the environmental and development policies in the FRG go beyond simple waste disposal, however. Where possible, recycling and climate-friendly technologies should be given attention. This certainly seems possible in emerging countries such as Turkey, for example, if the economic conditions are favourable. The KfW has published the following figures for the projects it has sponsored in Turkey (see tables 1 and 2). The figures given for the overall costs of fermentation of 60 – 90 €/Mg seem too high here. Presumably these figures are based on the relatively expensive, high tech facilities in Germany. On the one hand the authors also know of definitely lower figures, on the other hand the costs for a waste management concept have to be seen as a whole and not only in terms of the individual type of plant. Cost advantages (when landfilling or during transport) might result through the comparatively expensive treatment of waste at other sites.

94

Table 1: Costs of waste management plants in Turkey, sponsored by the KfW development bank, in comparison with figures from Germany (Schenk and Pfaff-Simoneit 2008)

Investment costs

Dynamic prime costs

Costs per household(in % per income)

Country Region /

City Inhabitant

s million

€ € per capita

€ / Mg Average income

Income of the

poorest 20 %

Turkey

Denizli 330,000 11.0 33.0 24.0 0.60% 1.10%

Samsun 380,000 17.6 46.0 26.0 0.60% 1.10%

Erzurum 330,000 14.2 43.0 32.0 1.00% 2.00%

Trazon / R. 601,000 27.1 45.0 36.0 0.80% 1.50%

Germany 150 - 180 0.30% 1.00%

Table 2: Cost comparison of different waste treatment systems - all figures in €/Mg waste (Schenk and Pfaff-Simoneit 2008)

Waste Treatment / Disposal Sanitary Landfill

Biological Treatment

(Stabilisation)

Fermen-tation

Incineration

Total costs 12 – 20 25 – 60 60 – 90 120 - 180 Benefit by electric power generation 4 – 6 - 8 – 14 18 - 35 Benefit of CDM 4 – 5 8 – 10 10 – 11 12 - 14 Net costs (without utilisation of heat) 4 – 8 17 – 50 45 – 65 90 - 130

It was for this reason that the consortium of companies IGIP / L.e.e. / ICP decided to pool the experiences of the firms and to carry out a cost comparison for a specific example. As a basis, a project was chosen where the costs had already been calculated for a landfill (currently under construction) as well as for an MBA (aerobic treatment of the organic components of waste). The firm L.e.e. has in addition planned a fermentation plant of a relevant size which is already in operation, so that confirmed figures for the construction of such a facility are also available. Using this sample project we intend to investigate to what extent the costs differ for disposal systems with a deposit of non-treated waste and the costs of deposit after anaerobic treatment. A project in Tunisia was chosen for this, as this project was one of those being sponsored by the KfW and the costs calculation for the construction had already been confirmed by an invitation for tenders. Moreover, Tunisia is a relatively highly developed country in North Africa, where without doubt higher quality waste treatment plans could be put into operation. In order to make the cost calculations comparable, an Excel tool was developed at the same time, which automatically calculates the specific dynamic prime costs or draws up a business plan suitable for bank approval from the usual costs calculations of the planners (investment costs, re-investment costs and operating costs). Due to the structure of this tool even sensitivity analyses may also be carried out very easily.

2. Description of the waste management situation in Tunisia In the project for which the cost comparison calculations were carried out, we are dealing with the waste management concept for Greater Tunis with about 2 million inhabitants. Besides an

95

optimized collecting and re-loading of the waste, in particular a second regulated landfill was to be planned to reduce strain on the landfill Jebel Chékir, constructed in 2000. Tunisia may be described (not only) from a waste management point of view as the most advanced country in North Africa. One after the other, modern waste management systems have been and are still being developed all over the country. Here, first the most populated cities and governorates along the coast were supplied with modern waste collection, re-loading points and a regulated landfill. The first of these landfills thus developed was Jebel Chékir, mentioned earlier, which was sponsored by the World Bank. Up to now 10 regular landfills for 9 governorates have been developed (see table 3 and figure 1). The lines in table 3 with a light grey background are those governorates already equipped with a modern landfill; the first 4 governorates form the greater Tunis area. Thus in the foreseeable future about 7 million inhabitants in Tunisia (about 68%) will be connected with a regular landfill. The planning of three further landfills in Medjerdatal (Governorate Béja, Jendouba, Le Kef and Siliana) is under way at present and further projects are planned for the near future. The nine new landfills (without the landfill Jebel Chékir) were surveyed last year and correspond completely both in construction as well as in operation to European standards, even if due to lack of experience there were initially (and at some sites still are) problems with the disposal of leachate and at present still no degasification is installed. This is planned, however, at all landfills and will be installed soon. It is intended that the second planned landfill „Kabouti“ for the greater Tunis area should take in exactly half of the waste generated there. The volume of waste in the first year of operation (estimated at the beginning of 2012) will be approx. 350,000 Mg and because of the rapidly increasing population and specific waste amount reach about 600,000 Mg in the year 2030. The overall volume is planned for maximally about 10 million Mg or resp. up to 23 operating years. The deposit surface of the landfill is approx. 22.5 hectares. All details in the following text are based on a feasibility study which was sponsored by the KfW (IGIP / IU / ICP 2008). In Table 4 the most important figures concerning Tunisia are compared with those of Turkey and Germany. The grey marked Governorates are linked to a sanitary landfill.

96

Table 3: Basic Data on the Governorates of Tunisia (Institut National de la Statistique, Tunisia 2009)

No. Governorate Area Inhabitants (2009)

Inhabitants per km2 (2009)

km2 Capita Capita/km2 1 Tunis 346 994 900 2 875.4 2 Tunis Ariana 482 485 700 1 007.7 3 Tunis Ben Arous 761 567 500 745.7 4 Manouba 1 137 364 600 320.7 5 Nabeul 2 788 743 500 266.7 6 Zaghouan 2 768 169 100 61.1 7 Bizerte 3 685 542 400 147.2 8 Béja 3 558 304 600 85.6 9 Jendouba 3 102 421 200 135.8

10 Le Kef 4 965 256 100 51.6 11 Siliana 4 631 233 100 50.3 12 Kairouan 6 712 555 900 82.8 13 Kasserine 8 066 428 300 53.1 14 Sidi Bou Sid 6 994 408 800 58.5 15 Sousse 2 621 602 300 229.8 16 Monastir 1 019 504 700 495.3 17 Mahdia 2 966 392 900 132.5 18 Sfax 7 545 917 000 121.5 19 Gafsa 8 990 334 900 37.3 20 Tozeur 4 719 102 300 21.7 21 Kebili 22 084 149 100 6.8 22 Gabès 7 175 357 400 49.8 23 Médenine 8 588 451 200 52.5 24 Tataouine 38 889 145 000 3.7

Total 154 5911 10 432 500 67.5

Table 4: Comparison of Economic Data of Tunisia, South Africa, Germany and other countries (IMF 2010, Wikipedia 2011, World Bank 2008)

1 According to the source, different figures are given for the area. Figures concerning population density thus vary accordingly.

97

3. Brief description of the sample project The figures 1 and 2 show the simplified site plan of the landfill Kabouti as well as the basis sealing system used, where a combination sealing was chosen.

Figure 1: Layout of sanitary landfill „Kabouti“ (IGIP / IU / ICP 2008)

Figure 2: Bottom liner system of landfill „Kabouti“ (variant 1)

Bottom liner system of sanitary

landfill Kabouti (variant 1)

Waste

Drainage layer

Geotextile (filter layer)

Sand protection layer

PE-HD geomembrane 2.5 mm

Compacted clay liner (2 layers) k ≤ 1 x 10-9 m/s

Ground level

98

The investment costs of this landfill can be seen in Table 5.

Table 5: Investment costs of the landfill Kabouti (landfill sections 1 to 3)

Landfill section 1 and infrastructure

Landfill section 2 Landfill section 3

Investment costs2 (without equipment)

about 14 million € about 3.1 million € about 5 million €

Investment costs equipment

about 2.2 million € Re-investment after period of

amortization

Re-investment after period of amortization

The specific dynamic production costs for the landfill described are calculated on the basis of the investment costs shown in Table 5, the re-investment for equipment etc. over the operating period and the operating costs at around about 12.57 €/Mg. A fixed rate of interest of 3% was assumed for the calculation. Proceeds from landfill gas use (production of electrical energy) were not taken into consideration here. If the costs for collection, re-loading and transport are also added, this results in total dynamic production costs of 36.88 €/Mg for 50% of the waste amount dealt with in the investigation /4/ for the settlement area of Tunis. If revenue from the conversion of landfill gas into electricity is taken into account, then the above costs are reduced to 10.38 €/Mg for the landfill or resp. 34.69 €/Mg for the overall disposal. The amount of gas extracted and the energy obtained is estimated quite conservatively, since it has been shown that the prognoses for this in the past were usually calculated much too optimistically and could not be realized in practice. 4. Comparison of the systems 4.1 Comparison of the landfills in the variants considered In the following the fundamentals of the planning for the two landfills in the variants under investigation are briefly described. Variant 1 The landfill Kabouti mentioned before was planned as a crude waste landfill taking account of the EU landfill directive (1999). In the first landfill section almost all the infrastructure of the landfill was developed, e.g.: access road and internal road, all buildings, weighing machine, leachate storage basin, surface drainage with retention. This planning thus corresponds to the basic variant 1 of the following comparison. In the case of the waste quantities described above, the costs were calculated over the whole period of the landfill operation and on the basis of the detailed planning of the landfill as well as on costs calculation carried out and meanwhile verified through the invitation for tenders and allocating of contracts for the construction. Likewise taken into consideration were the costs incurred in the after-care phase, assuming that the landfill was equipped with leak-proof surface sealing after filling and that thus the leachate amount is reduced to zero in the after-care phase. These costs were used as a basis for comparison for the second scenario developed – fermentation of waste and the subsequent deposit on a thus reduced-size landfill. Variant 2 For the alternative variant it was simply assumed that the fermentation plant was designed for the medium amount of waste, so that in the real case it would be under-dimensioned from start to finish. This assumption is certainly greatly simplified, but the planning with a modular structure 2 Without surface sealing (capping system)

99

would have involved too much effort for this costs comparison. For variant 2 it is assumed that the organic waste and recyclable waste will be removed from the overall waste stream and that only the remainder will reach the landfill. Due to the pre-treatment of the waste, the amounts to be deposited are reduced considerably. On average only about 156,000 Mg per annum are to be deposed of over the period under consideration. The landfill in the comparison variant 2 can, for the same disposal period, thus be developed to a considerably lesser extent. The following table shows the main differences between the landfills under consideration. In addition to the differences described in Table 6, a reduced amount of equipment and personnel is required for the residue landfill in Variant 2. A simple surface sealing was planned for both landfill variants, which for both was calculated at 35/€/m2. The dynamic prime costs of both variants can be seen in Table 6. Here it should be noted that the costs of both variants refer respectively to the total waste quantity occurring in the period under consideration.

Table 6: Comparison between the crude waste landfill in Variant 1 and the residual waste landfill in Variant 2

Landfill Variant 1 Landfill Variant 2 Operating time 23 years 23 years After-care period 30 years 30 years Area 22,5 ha 14,0 ha Bottom lining system Composite liner Single liner Degasification necessary yes no Average quantity of leachate about 90.000 m3/a about 45.000 m3/a Leachate treatment 3 step treatment plant 1 step treatment plant Investment costs3 about 22.1 million € about 10 million € Dynamic prime costs without benefits of landfill gas utilization4

12.57 €/Mg

5.63 €/Mg

4.2 Costs of waste pre-treatment It is not only the costs of depositing waste at the landfill which are to be compared, however. In Variant 2, the costs for waste treatment have to be added, of course. These are made up of the costs for the pre-treatment of the waste before fermentation and the costs for the fermentation plant. The pre-treatment of the waste is essential, as collection is of mixed waste, so that first the recycling waste and also the contaminants have to be sorted out. The pre-treatment consists of sieving (drum screen) as well as hand sorting on a conveyer belt. Further processing of the waste (e.g. using a pulper) was added to the costs of fermentation. Here, the scenario being considered was based on the actual composition of the waste. The investment costs for pre-treatment amount to 23 million Euros (construction costs and costs for equipment), those for the fermentation plant to a total of 43 million Euros. As in the case of the landfill, the plant operating times were estimated at 23 years, so that the systems could be kept comparable in the time axis too. The calculation shows that the dynamic prime costs for Variant 2 are - without taking the revenues into account - in the area of costs specified by the KfW (see Tab. 2). Thus the fermentation alone including the accompanying pre-treatment of the waste incurs costs of about 22 €/Mg (in terms of

3 Without equipment and surface sealing 4 In terms of the total amount of waste occurring

100

the total amount of waste involved). The specific costs according to the amount of the waste treated are of course higher, as waste had already been sorted out in the pre-treatment phase (either as recycling materials or as contaminants which had to be deposited. The costs area of 60 – 90 €/Mg depicted in Table 2 for fermentation seems despite this very high, but nevertheless still lies in the range of the costs calculated, if one relates this only to the waste amount treated. The actual costs of the treatment plants depend quite considerably on the proceeds acquired from recycling materials and resp. the energy produced. To demonstrate this, the dynamic prime costs were calculated respectively both for the sorting as well as for the fermentation, depending on the amount of the proceeds. For this see figures 3 and 4.

Figure 3: Costs for pre-treatment / sorting depending on the amount of achievable proceeds for the recyclable materials in €/Mg

Figures 3 and 4 demonstrate that when the proceeds for the materials obtained or energy produced are higher, then the treatment costs for pre-treatment of waste are distinctly reduced. It is also evident, however, that taking into account the proceeds for recyclable materials on the world market today and in the near future, the costs for sorting cannot be reduced as desired. This pre-treatment is, however, necessary in any case in a mixed waste collection before higher quality measures such as fermentation can be carried out. This can even provide profit through the production of energy if the payment for the electricity fed into the mains supply is approx. 12 €-cent per kWh.

Figure 4: Costs for the fermentation depending on the amount of achievable proceeds for the electrical energy produced

101

4.3 Cost comparison of the two systems

In Table 7 a clear overview of the costs for the systems under consideration (variants 1a to 1b, also 2a to 2d) is given. First the costs for both variants were calculated, without taking into account payment for recycled materials acquired or energy produced (variants 1a and 2a). In the next step, the costs when estimating realistic profits (30 € / Mg recycled materials; 0.11 €/kWh electrical energy) were calculated (variants 1b, 2b and 2c). Since in the fermentation process a fertilizer or compost is also produced, a further variant 2c was considered for the sale of these products (for 10 €/Mg). It becomes evident, however, that in many countries the problematic sale of fertilizer/compost does not contribute much to reducing costs. Finally, yet another estimate was calculated, based on the approach that the proceeds from recycled materials as well the energy produced develop differently compared to all other costs of living. For this an annual increase in proceeds of 1% was estimated (variant 1c and 2d). The transport costs were not calculated for variant 2 because they are subject to location. To simplify matters the costs of Variant 1 were adopted but not shown in table 7. The comparison of systems thus reveals that higher quality waste disposal systems with a fermentation plant compared to the simple disposal of the mixed waste on a landfill are economically also clearly justifiable. If profits are estimated from CO2 trading - although this has not yet been done in the investigation above – then financial advantages even result for the Variant 2.

Table 7: Costs comparison between the depositing of waste without previous anaerobic treatment (Variant 1) and depositing after previous anaerobic treatment (Variant 2)

3,00 %

2009 ‐

2005 ‐ 2009 ‐

2010 ‐

2032 ‐

30 €/Mg

0,11 €/kWh

10 €/Mg

1 %

Variant 1a Variant 1b Variant 1c Variant 2a Variant 2b Variant 2c Variant 2d

without

landfill gas

utilization

with landfill gas

utilization and

revenues for

electric power

generation

with landfill

gas utilization

and increase of

revenues

Without any

revenues

With revenues

for electric

power

generation and

recycables

With revenues

for electric

power

generation and

recycables and

fertilizer

Like variant 2c

but with an

increase of

revenues

Mechanical pre‐treatment 0,00 0,00 0,00 11,67 10,82 10,82 10,82

Anaerobic degestion 0,00 0,00 0,00 21,57 2,13 1,01 ‐1,91

Landfilling 12,57 10,38 9,94 5,63 5,63 5,63 5,63

Sum 12,57 10,38 9,94 38,87 18,57 17,45 14,54

Basic data

Revenues by recycables

Revenues by electric power generation

Revenues by fertilizers / compost

Increase of revenues per year

Interest

First year

Design phase

End of disposal at Kabouti (Variant 1):

Commencement of w landfill construction

Dynamic prime costs without collection, transfer and transportation

All figures in €/Mglandfill without pre‐treatment

landfill with pre‐treatment (biogas generation by anaerobic

degestion)

Variant 2Variant 1

5. Conclusions In the considerations above we are dealing only with hypothetical models of course. The models are based on real figures, however, and also on the construction of a landfill and a fermentation plant. The fermentation plant is a very good size as it should treat all the remaining waste after pre-treatment (i.e. without recycled materials and without contaminants about 250,000 Mg/a). This size of course results in relatively low specific treatment costs. The specific costs for fermentation will be somewhat higher for smaller facilities. It is assumed that all waste to be deposited will be pre-treated, so that there is a considerable reduction in costs for the landfill.

102

It is desirable that after a phase in which first pilot plants with a performance of less than 100,000 Mg/a in countries such as South Africa, Tunisia or Turkey are built and successfully operated, then the waste quantities of whole regional authorities should be treated before they are deposited on landfills. In addition to purely monetary considerations, the other advantages of such a high quality waste disposal - such as environmental, ecological and social aspects - should also to be taken into account. The increasing scarcity of raw materials as well as the rise in energy prices will in future be a factor in making higher quality waste disposal systems of interest when compared to just depositing the waste without treatment. Probably even financial advantages will be possible. An important pre-requisite for this in particular is reliable legislation, for example for the feeding of electrical energy into the national grid(s) of a country. This contributes quite considerably to a long term certainty in planning. The corresponding legislation in Germany can here serve as an example. The costs for varying waste disposal systems can be drawn up relatively quickly using the Excel tool developed and can be subjected to a sensitivity analysis. 6. References

Schenk, B. und Pfaff-Simoneit (2008): “KfW’s Support for Turkey’s Evolving Waste Management System”, Lecture on the Conference “Towards a Clean Environment in Turkey – Challenges & Financing Options”, Nov. 4th and 5th 2008 in Ankara, Turkey World Bank (2008): World Development Indicators, http://data.worldbank.org Wikipedia (2010): www.wikipedia.com Institut National de la Statistique (2009) : Statistical data on Tunisia Consortium IGIP / IU / ICP (2008): Feasibility study on the waste management for Greater Tunis, elaborated for ANGeD (Agence National de la Gestion des Déchets), Tunis funded by Kreditanstalt füe Wiederaufbau, Frankfurt (KfW) European Union (1999): Directive 1999/31/EG as of April 26th 1999

103


LANDFILL LEACHATE TREATABILITY Strachan L J*, Robinson H D** and Gravelet-Blondin L R*** * Director, GreenEng (Pty) Ltd, P O Box 641, Westville, 3630, Durban, Kwazulu-Natal, Tel: +27(0)31 266 8083, Fax: +27(0)86 600 5186, email: [email protected]; ** Technical Director for SKM-Enviros (UK), Shrewsbury, Shropshire, UK; Enviros House, Shrewsbury Business Park, Shrewsbury, Shropshire, SY2 6LG; Tel: +44 (0)1743 284 877; Fax: +44 (0)1743 245 558; Mobile: +44 (0)7714 757 835; e-mail: [email protected] *** Water Quality & Waste Water Specialist, Water Lily Consulting, email: [email protected] ABSTRACT

“He makes the deep to boil like a pot.” (Job 41:31)

In line with the theme ‘Waste Management Facilities – The New Order’, this paper presents a case of ‘what still lies below’ highlighting the environmental burden and risk even a ‘New Order waste management facility’ or modern landfill represents. Many landfills produce high strength effluents or leachates which originate primarily from the percolation of rainfall ingress through the landfilled wastes, as well as from liquids “squeezed” from the waste body – particularly in the case of the many ‘co-disposal’ landfills where liquid wastes and sludges flow through the landfilled waste matrix. There are numerous landfill sites scattered across South Africa in particular in the higher rainfall regions, which produce high strength leachate emissions. The management methods which have been typically engineered are usually:

discharge to a retention or evaporation pond; or discharge to a nearby sewer-line.

The discharge (or often termed ‘spill-over’) of raw leachate to the natural environment is omnipresent. In more recent times as the water shortages in SA are starting to take a grip, the commonplace discharge of landfill leachate to the nearby sewer system is no longer always viable. ‘New Water’ waste water treatment practices are on the increase in SA where the municipal sewerage system is ‘harvested’ for available water to be treated for reuse by industry or, in extreme circumstances (and where the ‘yuck-factor’ can be overcome!) for potable water supply. The treatability of landfill leachate is discussed in this paper and ‘leachate treatability trials’ are offered particular detail including one where the flow of leachate from a co-disposal low-hazardous waste landfill site was restricted by the legislators requiring an economical leachate treatment and disposal method to be found. Unfortunately the particular case study site could not be named in this paper, but the treatment trial results nonetheless provided. The principal reason for a treatability trial is to provide a treated effluent, at low cost, of the full-scale plant equivalence which can be assessed for the method of disposal or further treatment. There are very few cases where biological processes for a landfill leachate cannot be applied, however, the process adopted must ensure that the bacterial environment be able to accept the biological loading at any stage of the process. A sequencing batch reactor (SBR) process is strongly advantageous over most other biological systems, for example a Membrane Bio-Reactor (MBR), since the influent leachate at any one stage is never toxic or too extreme to the biological system. Keywords: Landfill Leachate, Biological Loading, Biological Nitrification and Denitrification, Treatability Trials, Sequencing Batch Reactor (SBR), Landfill Emissions, New Water

104

INTRODUCTION The co-disposal landfill site considered as a case study in this paper is a permitted H:h landfill (pre-new regulations of 2011) engineered in accordance with the Minimum Requirements for Landfill which is currently regulated by the Department of Environmental Affairs (DEA) of the national government of South Africa. The site is nearing the end of its planned disposal lifespan and contains some 2,5 to 3,0 million m3 of disposed wastes. The landfill produces volumes of landfill leachate (which drains from the engineered drainage systems) of up to some 100m3/day which is largely transported off site for disposal by tanker trucks. To achieve discharge standards to the municipal sewer and/or sea outfall, the leachate was pre-treated with hydrogen peroxide (H2O2) prior to tankering off site. An on-site landfill leachate treatment solution was required to achieve an economical disposal to the municipal sewer line. The undertaking of a leachate treatability trial would inform the possibilities for, and design of, an appropriate leachate treatment plant at the site. It was considered valuable to conduct pilot-scale treatability trials on representative leachate from the site at an early stage, in order to assess whether leachates contained any components which might be likely to constrain or compromise the successful biological treatment of the leachate in an on-site plant. Such toxicity can prove extremely difficult to predict or exclude, based solely on analytical results. A successful biological trial will give confidence both to the landfill site owner and to regulatory authorities, that there are no fundamental reasons why a full scale plant could not operate consistently and reliably, to achieve appropriate standards for discharge of treated leachate into, either the local sewer system, or possibly into an adjacent watercourse after further polishing treatment.

Many landfills produce high strength effluents or leachates which originate primarily from the percolation of rainfall ingress through the landfilled wastes, from a high moisture content “squeezed” out from the waste body and/or alternatively, specifically in the case of this site, from the co-disposal of liquid wastes and sludges into the landfill. Currently there are three known leachate treatment plants in South Africa employing primary biological treatment using SBR’s namely the Mariannhill, Buffelsdraai (both located in the Durban Municipal region) and Vissershok (located in the City of Cape Town Municipality) leachate treatment plants sites which treat some 50m3/day, 120m3/day and some 300m3/day respectively.

As previously stated, the principal reason for a leachate treatability trial is to provide a treated effluent, at low cost, of the full-scale plant equivalence which can be assessed for the method of disposal or further treatment. This paper explains the applicability and necessary design principles of biological treatment processes for landfill leachate. Biological treatment processes can include nitrification, partial-denitrification, full denitrification, dissolved-air floatation (DAF), engineered reedbeds and wetland systems. If required, physical post treatment methods such as ultra-filtration, reverse-osmosis, etc, may also be applied to offer a treated leachate quality suitable for discharge to the natural environment or natural watercourse. This paper concludes that the understanding of the natural or biological treatability of any landfill leachate should be understood by any landfill operator and/or owner. This can be achieved at relatively low cost through treatability trials.

OBJECTIVES OF LEACHATE TREATABILITY TRIALS Objectives of leachate treatability trials are generally as follows:

(i) Leachate Treatability: To confirm whether there are any fundamental features of leachate quality from the landfill under consideration, which may be likely to impair successful treatment by standard aerobic biological processes in an extended aeration Sequencing Batch Reactor (SBR) system. Of particular concern to the success of the trials may be low concentrations of chromium in the leachate, which might possibly provide some level of inhibition to one or both of the stages of the biological nitrification process. In practice, a treatability trial is the only way in which the possible significance of such toxicity can be evaluated.

105

(ii) Treatment Design Parameters: To obtain leachate-specific treatment data. The trial would also determine whether, and to what extent, provision must be made for the addition of extra alkalinity, or any other chemicals, to the process of treating the leachate, and would also assess the actual quantities of sludge likely to be generated during the treatment process.

(iii) Effluent Quality: To demonstrate the degree of treatment performance that could be anticipated from a well-designed and operated full-scale biological treatment plant, by providing reliable estimates of typical effluent quality, based on actual simulation of the treatment processes.

(iv) To further characterise leachate and effluent quality from the landfill, particularly with respect to more detailed analysis including leachate toxicity, and levels of residual “hard” COD likely to be present in final effluent.

(v) To minimise the hydraulic retention time (HRT) period that is required to achieve successful treatment of the leachate under consideration, and so optimise the operation and performance of a full-scale treatment plant.

GENERAL COMPOSITION OF LEACHATE The raw leachate of the landfill under consideration in this paper, to offer an appropriate case study, is typical of strong methanogenic leachates from many large landfill sites in many countries of the world (Ref: Robinson, 1995, 2003, 2007). Main contaminants requiring treatment include degradable organic contaminants which contribute to COD values from 10,000 to 15,000mg/l, concentrations of ammoniacal-N of up to some 1,300mg/l (many unreliable results are present in the data set for ammoniacal-N), as well as other inorganic parameters. Aerobic biological treatment processes will not, on their own, be able to achieve compliance with all required consent limits for all parameters, but would nevertheless undoubtedly provide great improvements in quality, at reasonable cost. The leachate collected from the main leachate storage tank at the landfill site is considered to be representative of that which will continue to require treatment in a full-scale plant, based on previous assessment of historical leachate quality data. The quality of the leachate blend was typically as follows (mg/l except pH-value):

Parameter: Influent

concentration:

COD 8800

BOD20 4000

BOD5 2800

ammoniacal-N 1250

alkalinity (as CaCO3) 12000

chloride 6250

pH-value ~8.5

This is generally a methanogenic leachate, typical of many derived from similar landfill sites throughout the World, where the key treatment to be provided is full nitrification of relatively high concentrations of ammoniacal-N, by biological processes.

106

ACHIEVEMENT OF TREATMENT OBJECTIVES There were no immediately obviously toxic components contained in the raw leachate in the case study – although both cyanides (typically 6-10mg/l) and phenols (often >250mg/l) can both exhibit some toxicity, the nature of the SBR process proposed for treatment would overcome such problems, as each compound is very degradable. Apparent presence of mineral oil (50 percent of samples containing in excess of 50mg/l of oil, although improved sampling protocols would be required for accurate assessment of overall oil content), may well require an oil interception process in advance of the biological reactor, since treatment is only likely to remove limited quantities of oil. One contaminant which can often be important in the inhibition of biological processes of nitrification is chromium. Inhibition is typically noted when concentrations of chromium exceed 1.0mg/l, and occasionally below this, depending on the proportion of hexavalent chromium. The analytical separation of the different states of chromium is notoriously difficult, but in the leachate samples obtained, hexavalent chromium (exceeding one sample in which the reported result for hexavalent chromium exceeds that for total chromium), accounts for on average less than 3 per cent of total chromium. A laboratory result was 1.06mg/l, mean values for total chromium were 0.92mg/l, and treatability trials will be valuable in determining whether this results in significant inhibition of nitrification processes, or if acclimatisation of bacteria can overcome this readily. Other metals are unlikely to cause any significant problems to treatment processes. Reported levels of sulphides in the leachate are typically from 100 to 250mg/l. These are very high concentrations indeed, and if accurate, are well in excess of the limit for discharge of 1.0mg/l. Nevertheless, the nature of the SBR process may well enable sulphides to be oxidised, with minimal emission of hydrogen sulphide during the treatment process. There is always the back-up option of pre-treatment with hydrogen peroxide, to convert sulphide to elemental sulphur – a process which uses very low quantities of peroxide solution, since the reaction is very specific. Typically, to fully treat 250mg/l of sulphide, a concentration of only 300mg/l of H2O2 would be required. The treatability trials will examine the extent to which hydrogen sulphide is likely to compromise the use of a simple aerobic biological process.

There are some determinands which cannot be reduced to below presently required discharge limits, by biological treatment alone. These include conductivity (very consistent at 50,000µS/cm), chlorides (typically 10,000mg/l, compared to a discharge limit of 1,000mg/l), fluorides (very variable, from 1 to 50mg/l, compared to a discharge limit of 5mg/l), sulphates (typically 500 to 1500mg/l, with a discharge limit which appears to have been reduced from 250mg/l to 21.5mg/l since mid-2010), and total dissolved solids of 30,000mg/l, compared to a discharge limit of 500mg/l. The only way in which such high concentrations of dissolved inorganic solids can be reduced would involve either evaporation (expensive in energy terms, unless waste heat is available from landfill gas combustion), or reverse osmosis, and in either instance such post-treatment would be far more readily achievable with a biologically pre-treated leachate. Many of the issues discussed above will be resolved by a successful biological treatability trial. THE TREATABILITY TRIALS PILOT PLANT

Design and Construction of the Leachate Treatment Pilot Plant Unit A pilot-scale leachate treatability unit was constructed and used for the trials by GreenEng, in a similar manner to those which have been used successfully in more than one hundred previous studies, over many years (e.g.: Robinson and Luo, 1991; Strachan & Robinson et al, 2000).

The pilot-scale leachate treatability unit was constructed as shown in Figures 1 and 2 below, and consisted of a 240-litre capacity polyethylene wheelie bin, modified as follows:

A 20mm ID pipe was inserted to act as a small bell-mouth overflow weir, at a measured level within the tank, providing a liquid capacity (operational volume) of 160 litres;

The discharge end of this pipe was suitably connected via a miniature electrical solenoid valve to an effluent storage container;

107

The solenoid valve was normally closed, but when supplied with mains current opened, allowing effluent to flow across the overflow weir into the effluent container;

An air supply was fed through a control valve into a specialised micro-porous tubular element, to act as an air diffuser in the base of the wheeled bin;

Air flow rates within a typical range from 20 to 40 litres per minute through the diffuser could be achieved, and these rates caused both oxygenation and adequate turbulent mixing of solids (mixed liquor suspended solids);

A variable diaphragm chemical dosing pump (range of possible pumping rates from 0.1 to 8 litres per hour) was programmed precisely to dose leachate from an individual 100-litre capacity stock tank into the aeration unit as required;

The unit was expected to operate at ambient temperatures of about 25 to 30ºC, considered likely to be representative of a full-scale plant in the area of the landfill site;

Operation of the treatability unit was controlled by an electrical panel that comprised several programmable time switches, based on quartz clocks, which were set to control the operation of the air compressor, the leachate feed dosing pump, and the solenoid valve. The leachate feed dosing pump can be adjusted by means of both the pumping rate and the time during which it is switched on (for example, 1 minute every 5 minutes) to deliver evenly the required volume of leachate feed each day into the reactor, within the selected dosing period. Since the volume of the aeration reactor was 160 litres, desired hydraulic retention periods could be achieved by control of leachate feed rates. For example, 16 litres of leachate feed per day would give a mean hydraulic retention time (HRT) of 10 days. Other required HRTs could be readily achieved in a similar manner.

Figure 1: Images of the leachate treatment pilot plant established by GreenEng at the University of Kwazulu-Natal for leachate treatability trials. Lindsay Strachan (left) and Howard Robinson (right) is shown in the photo on the left. A treatability trial should precede a landfill leachate treatment plant

design to assess any inhibition of the biological processes. A trial may last for some 8 to 12 weeks.

108

Commissioning of the Pilot Plant Treatment Unit

The treatability trials were carried out using a mixture of 100 litres of activated sludge that was obtained from the eThekwini Municipality’s Northern Wastewater Treatment Works, in Durban, and 50 litres of mixed liquor obtained from the DSW’s Buffelsdraai Leachate Treatment Plant of eThekwini Municipality. Relatively rapid establishment of successful treatment was anticipated for the new leachate feed because such activated sludge had previously been used successfully as a biological seed for treatability trials on leachates from other landfills in the Durban area (Bisasar Road, Mariannhill and Buffelsdraai treatment trials). Provided that no significant toxic inhibition of treatment processes was encountered, the sludge was deemed well acclimatised to treatment of a mixed domestic/industrial wastewater

Figure 2: Basic design of the experimental unit used to achieve treatment of landfill leachates. Operation of the Pilot Plant Treatment Unit

The operation of the pilot plant treatment unit to achieve biological nitrification is shown in detail in Figure 3 below and is described as follows:

i. Compressed air was fed into the aeration unit from 14h00 to 10h00, and during this period leachate was pumped (dosed) into the unit evenly to (ultimately) achieve a desired hydraulic retention time (HRT) and rate of treatment.

ii. After each period of leachate dosing and aeration, at 10h00, when the liquid level in the aeration unit had risen, the compressor was turned off automatically and sludge was allowed to settle in a quiescent phase until 13h00. At this time the solenoid valve was energised and opened for a 30 minute period, during which time clarified effluent discharged by gravity over the overflow weir into the collection vessel/drum (as illustrated in Figure 6 previously). The overflow was fitted with a 2 mm mesh cover, to prevent occasional large floating particles and/or biological flocs from blocking the solenoid valve.

iii. After the solenoid valve had closed again at 13h30, a further quiescent phase period with no aeration was allowed until 14h00. Aeration turned on at 14h00, sludge was re-suspended, and leachate additions and the treatment processes over a next 24 hour treatment cycle repeated. The inside of the unit was brushed each day, during the aeration

109

0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 10.00 11.00 12.00 13.00 16.00 17.00 18.00 19.00 20.00 21.00 22.00 23.00 24.00

(1)

(2)

(3)

(4)

(5)

(1) Aeration on from 14h00 to 10h00 (20 hours per day).

(2)

(3) Solenoid valve energised (open for effluent release) from 13h00 to 13h30.

Aeration (20 hrs/day)

8.00 9.00 15.0014.00

Treatment Processes Daily Cycle

Time

Settlement (2 hrs/day)

(10.00 - 13.00)

(14.00 - 10.00)

Leachate Dosing (18 hrs/day)

(14.30 - 08.30)

Notes:

quiescent period (no activity)

(13.30 - 14.00)

Leachate dosing on within period (1)above, on 3 occasions each per hour, at variable rates over 5 minutes each, to achieve desired daily amounts treated.

Decant (Soleniod Valve)

(13.00 - 13.30)

phase, to ensure that sludge remained in suspension, and did not adhere to the polyethylene walls of the wheelie bin.

iv. After 14h00 each day effluent could then be collected, the volume that had been treated could be measured, and samples submitted for immediate analysis as required. Effluent samples were tested immediately each day (generally), (using test strips) for ammoniacal-N, nitrate-N and nitrite-N. Temperature and pH-value were usually determined in the aeration reactors once a day, at around 13h30.

Figure 3: Detailed schematic of the operation of the treatability trial pilot plant unit

v. Detailed laboratory determination of a wide range of parameters was undertaken regularly on effluent samples, and suspended solids (SS) and volatile suspended solids (VSS) were also regularly determined in the mixed liquor within the aerated reactor.

vi. Alkalinity could be added as sodium bicarbonate, as necessary, (for convenience by a single daily addition of a measured weight directly into the aerobic reactors as needed), in order to maintain pH values within the range 7.5 to 8.5, and provide optimum conditions for nitrification to take place. However, as speculated previously in this report, it was highly probable that very little pH control adjustments may be required owed to the high alkalinity concentration of the raw leachate.

vii. Phosphoric acid was added as a nutrient to the reactor, as a single slug dose of 20ml of concentrated H3PO4 on day 2, and then no further additions were made. Thereafter, 20ml of phosphoric acid was to be added once every 3 weeks.

viii. For routine analysis of effluent samples, these were passed thought a coarse GF/D grade filter circle, after decanting, specifically to overcome any day-to-day variations in sludge settlement which might result from operation of the small overflow arrangement on the pilot-scale treatment unit, and could obscure real changes in effluent BOD or COD results.

110

0

200

400

600

800

1000

1200

1400

1600

1800

2000

0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200

Influent ammoniacal nitrogen (mg/l)

Eff

luen

t C

OD

val

ue

(mg

/l)

Full Scale Plants

Pilot Scale Trials

Linear (Pilot Scale Trials)

Linear (Full Scale Plants)

TREATMENT EFFICIENCY Treatment efficiency of the pilot plant very rapidly stabilised at final levels, with >99.98% removal of ammoniacal-N, and apparently complete removal of all biodegradable COD, indicated by the very stable values of residual COD in the final effluent, representing overall COD removal rates of about 70%. Nevertheless, the final effluent COD values in excess of 1250mg/l were somewhat greater than might be anticipated for treatment of similar leachates from domestic waste landfills. In extensive experience of treating leachates using biological processes, at both pilot-scale and full-scale (Robinson et al, 1995; Strachan et al, 2000), it has been found that levels of residual and intractable “hard” COD in treated leachates (treated effluents) are not generally linked to the levels of COD in leachates being treated, but rather are much more closely related to concentrations of ammoniacal-N in raw leachates. This may well be due to both being the product of the same anaerobic processes of degradation, taking place within landfill sites.

Figure 4 below provides correlations between concentrations of ammoniacal-N in leachates being treated, and COD values in final effluents, for a large number of full-scale SBR plants and pilot-scale trials on leachates from both domestic and industrial waste landfill sites (Robinson et. al., 2004). A simple interpolation on the figure, for a leachate with influent ammoniacal-N in the order of about 1250 mg/l, indicates that treatment of similar leachates has generally resulted in effluents containing COD values of about 1100 mg/l. While this is probably indicative of the nature of the raw leachate, it does not imply that it necessarily contains any toxic or inhibitory substances. Indeed, the fact that the relatively sensitive nitrification process was not apparently inhibited, makes it unlikely that any unusual organic compounds are present in the treated leachate – and more likely that the COD simply represents higher concentrations of relatively inert compounds such as humic or fulvic acids.

Figure 4: Correlation between concentrations of ammoniacal-N in leachates, and residual “hard” COD in treated effluents, for full-scale treatment plants and detailed pilot-scale studies on a range of landfill leachates (all results in mg/l) [after Robinson et. al., 2004]

Results for nitrate-N in treated leachate at the end of the trial, of 1271mg/l, were within 2% of the concentration of 1253mg/l of ammoniacal-N in raw leachate, and demonstrated that removal was likely to be entirely by full biological nitrification. It was not necessary to add any external alkalinity to maintain pH-values during the treatment trials, and pH-values remained stable at optimum values between 8.0 and 8.5. Overall reduction in alkalinity was 8737mg/l (as CaCO3). In theory (refer to Section 3.1) the nitrification of 1253mg/l of

111

ammoniacal-N would consume about 7.14 times this value as alkalinity, that is 8946mg/l (as CaCO3), and this value is very close to the alkalinity reduction actually observed. Measured specific rates of treatment of ammoniacal-N, based on typical treatment rates during the 25 days from Day 51 to Day 76 (16.49 litres/day), on a concentration of ammoniacal-N of 1253mg/l, and on typical concentrations of MLSS of 4500 mg/l, and of MLVSS of 3400 mg/l were as follows: 106 mg N / g MLSS / day, or

140 mg N / g MLVSS / day. These values, although unexpectedly high, and indicative of a relatively uninhibited nitrification process, do not necessarily represent maximum rates of treatment since it is possible that the plant might have treated leachate at slightly greater rates. Nonetheless, leachate treatment plants treating similar leachates have shown to be able to operate at rates similar to these. Therefore, these figures obtained in the present trials will form a sound basis for the design of a full-scale leachate treatment plant for the proposed landfill site. CONCLUSIONS The biological treatability trails have been completed successfully on a high strength methanogenic landfill leachate from a co-disposal landfill site (previously registered as a H:h landfill), during an 82 day trial period. The trials were commissioned using non-acclimatised biological seed sludge from the eThekwini Municipality’s Northern Waste Water Treatment Works in Durban. Nearly 993 (992.7) litres of leachate was treated at an overall mean rate of 12.11 litres per day, representing an overall mean hydraulic retention time (HRT) of about 13.2 days, and equivalent to about 6.2 bed volumes of the SBR reactor, providing complete flushing out of any contaminants that had initially originated in the seed sludge materials. During the 25 day period from Day 51 to Day 76, when treatment had reached stable conditions, 412.2 litres of leachate was treated at a mean rate of 16.5 litres per day – a mean HRT of just below ten days (9.70 days). The trials were carried out very precisely over a total of 82 days, during which complete, reliable and consistent treatment of leachate was maintained, with full biological nitrification of ammoniacal-N at all times, and removal of essentially all degradable COD. Sludge settled well at all times during the trials, providing a well-clarified effluent, which was a dark reddish-brown in colour. Engineered reedbeds could assist somewhat with colour removal at full scale. Samples of treated leachate at the end of the trial period were submitted for toxicity testing at the laboratories of Alcontrol in Chester, in the UK. Although the Microtox testing was unsuccessful because of the colour of the sample, testing of Respiration and Nitrification Inhibition was able to demonstrate that toxicity is of insignificant concern for final discharge of treated leachate to the municipal sewer line. Landfill leachate treatability trials have confirmed that an appropriately designed SBR leachate treatment plant at the landfill site considered should work reliably and consistently, and be able to treat leachates to standards which will be able to meet proposed consents for discharge to the available sewer line. No toxic inhibition whatsoever has been detected to the treatment process, and net rates of sludge production have been demonstrated to be low. Detailed design of a full-scale treatment plant, and discussions with regulators to determine suitable discharge consent standards, can both now be progressed confidently without delay.

112

REFERENCES Robinson H.D. and Maris P.J. (1979) Leachate from Domestic Waste : Generation, Composition and Treatment. A Review. Water Research Centre Technical Report TR108, 38 pp, March 1979. Robinson H.D. & Grantham G. (1988) The Treatment of Leachates in On-Site Aerated Lagoon Plants: Experience in Britain and Ireland. Water Research, 22 (6), June 1988, pp. 733-747. Robinson H.D. (1990) Leachate Treatment – The Practice. Paper presented to the Conference, “Landfill Problems and Solutions, held at Brunel University, Uxbridge, on 15-18 May 1990. Proceedings, 14pp. Robinson H.D. and Luo M.M.H. (1991) Characterisation and Treatment of Leachates from Hong Kong Landfill Sites. Paper presented to IWEM ’90, ‘Design and Construction of Works for Water and Environmental Management’, held at the Scottish Exhibition and Conference Centre, Glasgow, 4-6 September 1990, paper 15, 20pp, and published in the Journal of the Institution of Water and Environmental Management, 5, (3), June 1991, 326-335. Robinson H.D. (1992) Aerated Lagoons. Landfilling of Waste: Leachate, Edited by Christensen, Cossu and Stegmann, Elsevier Applied Science, pp. 203-210. Robinson H.D., Chen C.K., Formby R.W. & Carville M.S. (1995). Treatment of Leachates from Hong Kong Landfills with full Nitrification and Denitrification. Sardinia ’95 Fifth International Landfill Symposium, CISA, Vol I pp. 511-534. Robinson H.D., Farrow S., Last S., & Jones D. (2003) Remediation of Leachate Problems at Arpley Landfill Site, Warrington, Cheshire, UK. CIWM Scientific & Technical Review, December 2003, pp. 18-26. Robinson H.D., Novella P., Strachan L. & Last S. (2004) The Use of SBR Systems for Treatment of Leachates at South African Landfills. Waste 2004 Integrated Waste Management and Pollution Control: Policy and Practice, Research and Solutions. Conference Proceedings. 2004, pp. 536-549. Robinson H.D. (2007) The composition of leachates from very large landfills: An international review. Paper presented to Torbay 2006, The Annual Conference and Exhibition of CIWM “Changing the Face of Waste Management”, Paignton, June 2006. Published in Communications in Waste and Resource Management, June 2007, 8 (1), pp. 19-32. Strachan L.J. (1999). An Investigation into Biological Treatment Systems for Landfill Leachate. Masters Degree Research Thesis (Research works carried out under supervision of Professor Raffaello Cossu of University of Cagliari (now Padova), Italy), University of Natal, Durban, South Africa. Strachan L.J., Trois C., Robinson H.D. & Olufsen J.S. (2000) Appropriate Treatment of Landfill Leachates with Full Nitrification and Denitrification. WISA 2000 Conference, Sun City, South Africa, Water Institute of South Africa. Strachan L.J., Robinson H.D., Last S.D., Payne G. & Wright M.R. (2007). Development of leachate treatment at a large new tropical landfill site. Proceedings Sardinia 2007, Eleventh International Landfill Symposium, CISA, Cagliari, Sardinia, Italy.

113


BEHAVIOUR OF BALING-WRAPPING MSW LANDFILL. APPLICATION IN SPAIN

Dr. Jose M. Baldasano ETSEIB, Technical University of Catalonia (UPC) Avd. Diagonal 647, 08028 Barcelona, Spain. [email protected] SUMMARY Baling-wrapping has been used in MSW landfills from the 70's and now it can be found in many countries worldwide. The landfills using plastic wrapped bales presented clear advantages over conventional landfills. The monitoring results from different case studies conducted at operational level in Spain are presented and analysed. The bale characterization revealed that the biodegradable fraction presented a progressive evolution to compost characteristics. 1. INTRODUCTION Baling-wrapping is a procedure for handling different types of materials such as municipal solid waste (MSW). Baling has been practiced extensively by farmers (ensilage storage) and the recycling industry in developing and developed countries. Nowadays, landfill of MSW using baling-wrapping technology can be found in many countries such as Germany, Italy, Spain, Portugal, Sweden, Korea and Lebanon, etc. The baling procedure of MSW is performed in two different ways, leading to cylindrical or rectangular bales. This technology has been used in MSW landfills since the early 70's (Hogland et al, 1999), but clearly takes place in the 90's, accumulating 40 years of experience. Baldasano et al. (2003) and Nammari (2006) reviewed the performance of MSW landfilling by baling-wrapping technology versus the conventional system. The most commonly used plastic material is low density polyethylene (LDPE). Landfills using plastic wrapped bales (LPBs) appear to present less environmental problems and clearly operational advantages than conventional landfills (CLs). In fact, LPBs is a complex media, wrapped bales are different to CLs, with very limited water infiltration due to plastic wrapping, they can behave anaerobically or aerobically; the generated gases pass through the plastic by diffusion or via small holes or by a combination of both. Unwrapped bales behaviour is similar to CLs, they can be considered as highly compacted landfills with infiltration. To understand the behaviour of bales it is a prerequisite to revisit waste decomposition in landfills, aerobic landfills, dry tombs, bioreactor landfills to look into the generation and percolation of leachate through the waste. Regarding the biological behaviour of the wrapped bales, the most extensive studies have been carried out by Tamaddon et al. (1995), DEKRA (1996), Robles-Martínez and Gourdon (1999, 2000), Hogland et al. (2001) and Nammari et al. (2003). Robles-Martínez and Gourdon (1999) showed that over several weeks of incubation, the microbial activity inside the bales was inhibited in the acidogenic phase due to accumulation of volatile fatty acids which acidified medium. Robles-Martínez and Gourdon (2000) reported on assays with only one bale over a significantly longer period (27–34 months). These studies were carried out exclusively on cylindrical bales. Hogland et al. (2001) studied the effects of composition, structural integrity variation and temperature on gas emission from eight bales (six cylindrical and two rectangular) to observe whether the bales exhibited any tendencies for self-ignition, and performed burning tests. In general, the experiments

114

show that this system considerably reduces gaseous and leachates in the short and half-term compared with CLs. However, the previous studies do not focus on the behaviour of the bales under real landfilling conditions. This work analyses and discusses the results of chemical, physical and biological processes and the environmental performance of MSW compacted and plastic-wrapped into air-tight bales with LDPE. The main case study considered is a household waste wrapped-baled landfill with three sites during five years (2000-2004) in Leon area (Northwest part of Spain) with 396300 tons. Two other case studies are also taken into account: a) a field-scale experiment of 37242 tons during four years (2004-2007) in the Garraf MSW landfill (Barcelona, Catalonia, Spain) which was carried out in order to assess the behaviour of wrapped-baled pre-sorted refuse from household waste; and b) a field-scale experiment of 26339 tons during five years (2005-2009) in the Vallensana (Barcelona, Catalonia, Spain) of wrapped-baled pre-sorted refuse from household waste. Figure 1 shows the location of the three case studies..

Figure 1 Location of the three case studies (red) using baling-wrapping in Spain (Europe). 2. LEON AREA SITE: LOCATION AND QUANTITIES The city of Leon and other near municipalities (located in northwestern Spain, see Figure 1) began to apply cylindrical baling-wrapping to their MSW in a temporal strategy for municipal waste disposal. Three different sites of Leon area were used in different time periods over 2000-2004 years.

Site 1. Trobajo of Cerecedo: adjacent to the transfer station, near Leon city, March 2000 - June 2000, accumulate 7000 tons, the land used were pre-conditioning.

Site 2. Area near Bernesga Ferral (municipality of San Andrés del Rabanedo); between September 2000 to January 2001, about 26300 tons were deposited, in the area were made measures of conditioning.

115

Site 3. Zone of Santa Maria del Paramo, April 2001 to September 2004. Here 363,000 tons was deposited. This area was operated with a system for collecting rainwater and leachate, and storage water ponds.

The MSW composition results in the Leon area in 2005 obtained by handpicking are shown in Table 1. The content of the biodegradable fraction (also known as organic matter) has a high value (46%), which is the fraction that can really cause pollution problems, while the remaining fractions are essentially inert (54%).

Table 1. MSW composition (%) in the León area, April 2005

MSW fraction Composition (%) Biodegradable 46,3 Paper and cardboard 20,5 Plastics 13,2 Brick 1,6 Ferrous 2,5 Non ferrous 0,3 Glass 4,2 Wood 1,7 Textiles 2,2 Footwear 0,7 Tyres 0,0 Ceramics and RDC 3,4 Other materials 3,4

3. METHODOLOGY The purpose of this methodology is to assess the composition of the waste inside of the bales, degree of evolution and quantify the residual level of biodegradable material, made in summer 2005. The following methodology was used: 1. Visits to the three sites: The visits consist of conducting field observations and taking

photographs accordingly. 2. Selection and removal of bales in each site depending on their age and situation both in the

interior of the bales lodged and the outer area. For this purpose we used a boom truck, they took a total of 18 bales: Trobajo of Cerecedo: three bales from the first half of year 2000. Bernesga Ferral: three bales from the second half of year 2000. Santa Maria del Paramo: a) three bales in the first half of year 2002; b) three bales in the

first half of year 2003; c) three bales in the first half of year 2004; and d) three bales in the third quarter of year 2004.

3. Weighing, characterizing and sampling of each of the 18 selected bales, which took place in

the recycling and composting San Román de la Vega treatment plant. The samplings of each of the bales were arranged for transportation to the laboratory, and further characterization and analysis. We determined the gases concentration (CO2, CH4, O2, CO and NH3) inside of the bales through a Dräger X-am 7000 (Keison) tube. Sorting the contents of the following fractions: a) glass and metal; b) paper and paperboard (including food packaging "tetrabrick"); c) plastics; d) fermentable material (food waste and plant debris); e) textiles; and f) other. Each of the fractions was weighed and sampled for

116

subsequent laboratory analysis, and the following determined: a) humidity; b) loss at 450 °C; c) ashes to 975 °C; and d) heavy metals: As, Cd, total Cr, Hg, Zn, Pb, Cu and Ni.

4. In each of the three sites the available waters were sampled, according to the following conditions: Trobajo of Cerecedo: we collected a sample of the little water that surrounded the bales. Ferral Bernesga: three water samples were collected, one of the water between the bales

and two flows water conductions to the leachate ponds. Santa Maria del Paramo: we collected five samples, three of them from different storage

ponds, one from the water conduction pipeline from the landfill and another from one external natural pond next to this site.

The parameters analysed for each water sample are: pH, conductivity, COD, TOC, AOX, chloride, sulphate, total nitrogen, ammonia, As, Cd, total Cr, Hg, Zn, Pb, Cu, and Ni.

5. Weather data analysis from the area, which aims to know the weather conditions which have been suffer the bales, correspond to the meteorological station of the Spanish Meteorological Office: 2661, León “Virgen del Camino”.

4. ANALYSIS AND DISCUSSION 4.1 Visual Inspection of the Bales From visual inspection performed in the older bales, the fermentable fractions showed that stabilized material presented lower moisture content and looked like compost. In recent bales from 2004, it was evident that the fermentable material showed little degradation, since we could easy distinguish the materials type concerned (leaves, eggshells, etc.). They had more moisture and were more attached to different material, therefore their separation was much more complicated in some of these recent bales. In addition, the presence of gases was detected. This clearly suggests a process of stabilization of the fermentable material as age advances. The behaviour of paper and cardboard were very similar to the fermentable fraction materials. In the oldest bales the presence of these materials was not easily distinguished, indicating that they had gone through a certain process of biodegradation. However, in the bales from the second half of 2004 the paper remained well preserved, and anyone could even read what was written. Textiles had no clear change in their constitution between the different bales analysed. In the case of plastics, it was observed that they lost some of their properties with the age of the bales: plastic bags broke more easily, were less flexible and discoloured. By contrast, in recent bales, plastics were virtually intact even while maintaining the brilliance of their colours. In the case of glass, there was no visual difference related to age of the bales, this material remained the same, but this did not happen with metals, which were more oxidized and brittle with age in the package. The fraction of "other" consisted mainly of construction debris, and no age-related changes were observed in the bales. The above results indicate that fermentable fractions, paper and cardboard and certain plastics have a progressive degradation over time. A stabilization process of the fermentable fraction.is clear

117

4.2 Percentage of Different Fractions A temporal pattern in the proportion of different fractions was not observed. The fermentable fractions are the majority in most of the bales tested, closely followed by plastics, and thirdly a fraction formed by paper and cardboard, glass and metals. Figure 2 shows the proportion of the different fractions found in each of the analysed bales.

Figure 2. Waste fractions (%) inside the bales. 4.3 Mechanical Strength of Bales From the viewpoint of the mechanical strength of the bales, a good behaviour is noted, independent of its initial handling at the time of being generated, transported and deposited. During this study, they were taken by the crane, put on the truck, unloaded for weighing, and finally transported to the area to be opened. After this set of manipulations, they continued to maintain their physical structure. The opening process to be submitted for analysis must be powered in order to spread the material. However, it should be mentioned that this is probably due to the grid system used in the compression process with these cylindrical bales. The wrapping-plastic eventually undergoes an aging properties process, easily detectable. Consequently, in general terms, it can be said that the bales showed a significant strength for both handling and maintaining their functionality. Nevertheless, the plastic degradation with the time was clear on direct exposure to the external weather conditions without any kind of protection or cover. 4.4 Gases in the Bales We proceeded to measure a single point made in each bale prior to opening, so that readings were taken with appropriate caution (Figure 3).

118

In all cases, the levels of CO2 did not exceed 10% by volume, except for a bale in 2004 and two in 2004, implying an evolution beyond what is reported in the DEKRA (1996) report, where bales were monitored for 9 months and found that CO2 levels are stabilized by between 20 and 25% after a few days. This behaviour is inversely related to the amount of O2, since in all cases the amount detected was greater than 10% (10-20%), except the bales where CO2 levels exceed this percentage, in these three cases oxygen is less than 10%. The behaviour of O2 does not fit the situation described in the report by DEKRA (1996), which found that in the early days a quick aerobic fermentation occurs which consumes almost all the oxygen, leaving less than 5 %. This indicates an aerobic fermentation process may be due to the entry of O2 into the bales because of the 10 bales that have O2 levels above 15%, 6 were open or had holes. Robles-Martínez and Gourdon (2000), attributes this to the deteriorating state of the cover of LDPE from old bales. However, the DEKRA (1996) report examined the bales and the bales pierced completely wrapped and did not identify significant deference behaviour. The degree of compaction prevents a significant inflow of O2 into the bale; of 8 bales with O2 levels below 15%, 6 of them have good compaction. It is noteworthy that the degree of compaction of the bales is also affected by the wear of the bearings of compaction machinery and not just lack of experience in the process or the type of materials being compacted. There was an absence of CH4 for almost all the bales, except for 2 (2004), the presence of CH4 in these two bales coincided with very low O2 levels (<2%), indicating that an anaerobic fermentation was taking place. Bales containing CH4 showed increases in CO concentration.

Figure 3. Values of CO2, CH4, O2 (% by volume) CO and NH3 (ppm) in the bales. Ammonia (NH3) presented low levels (~16 ppm) in all the bales, with the lowest values in older bales. There are no previous data known so far on the behaviour of NH3 within the bales. However, it is a compound that occurs in both aerobic fermentation and in anaerobic fermentation. It is difficult to define the prevalence of a particular fermentation process, given the heterogeneity of the materials; it is clear there is competition between an aerobic fermentation, probably dominant, and an anaerobic fermentation. But, it is well-defined process of stabilization of the fermentable fraction.

119

4.5 Water Analysis This section presents the analysis of the measurements performed in the different water ponds (Figure 4). The water analysed presented very similar characteristics. The values obtained were compared to Spanish limits to discharge into waterways (red columns) and into wastewater treatment networks (yellow columns). None of the samples had pH problems, they had a slightly alkaline pH, but within the normal range. The total nitrogen and NH3 presented important values, but without reaching the values measured in a typical leachate from a traditional landfill (> 2500 mg/l). The conductivity values clearly showed a significant content of dissolved salts. But again had lower values than those measured in traditional landfill leachates (> 20000 mS/cm). The contents of chlorides and sulphates could be described as significant. The results of COD and TOC, in some cases had high values, responding to the content of biodegradable material of the bales that shot on the layer in contact with soil, water has flooded into contact with bales. But again, their values were lower than measured in leachate from a traditional landfill (> 10000 mg/l). The AOX values indicated that the content of organic material associated with halogen was minimal and very limited. AOX values were not very high, but prevent the direct discharge into waterways. None of the samples showed problems of heavy metal pollution. Their concentrations were very low. A clear result, without to be clean water, was that it did not cause a pollution level comparable to a typical leachate from a traditional landfill. In general, it would be admitted to a network of wastewater treatment. Another important result is related to the external pond adjacent to the road in the Santa Maria del Paramo site. All parameters showed good water quality values, except for the unexpected high Hg concentration. This value was compared with the analysis of the ponds of the area and the channel, confirming that its origin was not from the bales.

120

Figure 4. Results of the water sample analysis, and comparison with different discharge limits.

4.6 Bales Analysis Table 2 shows the results of bales characterization. Furthermore, it is included a sample of land located adjacent to the Santa Maria del Paramo site. The humidity values were between 38 - 62%, with lower values in older bales (except one), and more homogeneous in the Santa Maria del Paramo site. Analyses of weight loss at 475 °C and 975 ºC, showed consistent values and the dispersion pattern was not homogeneous. There were no problems with the heavy metals concentrations and it was noticeable that there was no presence of As and Hg. The content of heavy metals was lower in the most recent bales.

Table 1. MSW composition (%) in the León area, April 2005

121

4.6 Other Aspects: Odours - Rodents Odours: In all three bales sites significant typical odours were not detected from waste fermentation. In the process of bales characterization, the presence of odour was important, and clearly present. Rodents: No rodents were detected during the visits. In recent bales, during the opening process, the presence of insects was detected. This does not imply that there may be rodents at some of the sites. 5. CONCLUSIONS The case study of a household waste wrapped-baled landfill located in Leon area (Northwest part of Spain) with 396300 tons (2000-2004) is presented. Different types of assays have been performed: (1) bale characterization in terms of waste fractions, weight and humidity in different time periods after storage; (2) measures of gas composition, evolution of the concentration of O2, CO2 and CH4; and (3) leachate composition analysis under real landfilling conditions. The bale characterisation revealed that the biodegradable fraction presented a progressive stabilization over time with compost characteristics. This was confirmed by the gas measurements. During the first months a competition among aerobic and anaerobic processes took place, while during the last months, an aerobic and stabilisation situation prevailed. The leachates characteristics were very different from the classic leachate measured in a traditional sanitary landfill; they could be accepted in a wastewater treatment plant. The obtained results confirmed that the baling-wrapping process halts the short and half-term biological activity and consequently the emission of gases and leachates. It also facilitated the handling of the refuse, and considerably reduced the main environmental impacts of a landfill. These results confirm those obtained in two pilot tests conducted in Barcelona area (Catalonia, Spain): a) field-scale experiment of 37242 tons during four years (2004-2007) in the Garraf MSW landfill and b) field-scale experiment of 26339 tons during five years (2005-2009) in the Vallensana quarry. In both cases: wrapped-baled pre-sorted refuse from household waste were deposited. The LPBs landfill presented clear advantages over CLs. LPBs eliminate the factors of visual impact of a CLs and the dispersion of light wastes by the wind. It also results in a significant reduction of the covering material, with a greater utilisation of the volume of the storage area. Subsidence problems should diminish, due the greater density and consistency of the bales. The classic vectors of pollution of CLs are practically avoided, since it is more difficult for birds, rats and dogs to gain access to the waste. The same applies to spontaneous combustion. The integration of this system in a transfer station allows the flow of vehicles to be reduced. The use of bales also simplifies the operations of transport within the controlled disposal area. It also reduces the needs for heavy machinery inside storage areas, as the bales are handled by means of a telescopic fork lift truck. The reduced land requirement of LPBs makes them also preferable from the point of view of land availability, which is increasingly scarce. 6. REFERENCES Baldasano, J.M. & Gassó, S. & Pérez, C. (2003) Environmental performance review and cost analysis of MSW landfilling by baling-wrapping technology versus conventional system. Waste Management, 23, pp 795–806. DEKRA (1996), Informe final sobre el ensayo piloto de almacenaje intermedio de basura según el método RPP. DEKRA Umwelt GmbH, Munich (Germany). Hogland, W. & Marques, M. & Grover, V. (1999) In Kalmar Eco-Tech'99, Ecological technology and management; Hogland, W., Ed.; University of Kalmar: Kalmar, Sweden, pp 203-213.

122

Hogland, W. & Marques, M. & Nimmermark, S. & Nammari, D. & Moutavtchi, V. (2001) Seasonal and Long-term Storage of Waste Fuels with Baling Technique (Preliminary Version). University of Kalmar: Kalmar, Sweden. Nammari, D.R. & Hogland, W. & Moutavtchi, V. & Marques, M. & Nimmermark, S. (2003) Physical and chemical processes in baled waste fuel with emphasis on gaseous emissions. Waste Management Research, 21, pp 309–317. Nammari, D.S. (2006) Seasonal and long-term storage of baled municipal solid waste. Ph.D. thesis memory. University of Lund, pp 71. Robles-Martínez, F. & Gourdon, R. (1999) Effect of baling on the behaviour of domestic wastes: laboratory study on the role of pH in biodegradation. Bioresource Technology, 69, pp 15–22. Robles-Martínez, F. & Gourdon, R. (2000) Long-term behaviour of baled household waste. Bioresource Technology, 72, pp 125–130. Tamaddon, F. & Hogland, W. & Kjellberg, J. (1995) Storage of Waste-Fuel by Baling Technique. Report 3188. Waste Management and Recovery Division, Department of Water Resources Engineering, Lund Institute of Technology, Lund University, Sweden.

123


DESIGN AND CONSTRUCTION OF A DEEP UPSTREAM GROUNDWATER

CUT-OFF TRENCH

J Shamrock*, J Glendinning*, J Msiza* and H Janse van Rensburg** *Jones & Wagener Consulting Civil Engineers, [email protected], [email protected] and [email protected] **Aquisim Consulting, [email protected] ABSTRACT This paper will highlight a case study of the design and construction of a deep upstream groundwater cut-off trench at a hazardous waste landfill in Gauteng. The cut-off trench was required to divert significant subsoil water flow in a perched groundwater table around an existing waste disposal site. Based on knowledge of the geotechnical and geohydrological characteristics of the site it was necessary to construct the cut-off to an average depth of 10m below natural ground. There were three main phases to this project and each of these will be discussed: Groundwater modelling. The setup, calibration and main findings of the geohydrological model developed for the site will be presented. The model predictions vs real flows measured in the installed section will also be discussed. Cut-off design. The design of the cut-off using known geotechnical and geohydrological information about the site will be presented. The main challenges to solve with such a deep cut-off will be highlighted, and the eventual solution and construction of a trial section as a proof of concept will be explained. The safety aspects of deep restricted excavations and how these were dealt with will also be highlighted. Cut-off construction. The construction of the trial section, and the lessons learned for the eventual full scale cut-off system, will be presented.

124

1

Slope Stability of Liners and

Landfill Interest Group Seminar, Oct 2011Short Course on:

Covers

Presented by: Richard Thiel, PE, GEPresident, Thiel Engineering, Oregon House, CA

richard@rthiel [email protected]

Slope stability

Deep-seated failure

V f ilVeneer failure

Stability Presentation Outline

• PART 1 – What is Shear Strength?• PART 2 – What is Slope Stability?• PART 3 – How Do you Measure Shear

Strength?• PART 4 – How Do you Evaluate the Data?• PART 5 – Slope Stability of Bottom Liners• PART 6 – Slope Stability of Cover Systems• PART 7 – What About Seismic Issues?

PART 1

WHAT IS SHEAR STRENGTH?

125

2

Land and Sea – Extremes of Shear Strength

DefinitionsPeak soil shear strength The maximum resistance of a soil to shearing stresses, usually

ith f t ti l l ff ti twith reference to a particular normal effective stress.

Residual soil shear strength The resistance of a soil to shearing stresses that does not change with increased deformation.

Normal stressNormal stress The compressive stress perpendicular to the direction of shearing.

Shear strength has a nickname called “friction” PART 2

WHAT IS SLOPE STABILITY (ANALYSIS)?

126

3

The results of the analysis are usually reported in terms of a “Factor of Safety”

(FS)(FS)

forceShearstrengthShearFS =

f

Slope Stability of Cover Veneer Layers is Similar to a Block System

Geosynthetic tension, T

Toe buttressingφ = internal friction angle

Shear forces caused by: weight of soil layer

Interface shear strengthδ = interface friction anglea = adhesion

φ = internal friction anglec = cohesion

Courtesy of JP Giroud

For soil slopes, it is the same concept but the “block” of soil slips out in a circular

geometryFor bottom liner slopes, the critical slip plane is now along the liner system, and the “block” of waste slips on the liner

MSW1x1

x21

H (γ, φ)

the block of waste slips on the liner.

L

SubgradeLiner (φ)

127

4

Key Items for Stability Analysis

• Geometry• Geometry• Material Properties (mainly shear strengths

and weights)• Pore Pressures (affects normal force)• Seismic Parameters (or construction equip )• Seismic Parameters (or construction equip.)• Analysis Method (e.g. Limit Equilibrium)• Acceptability Criteria (e.g. Factor of Safety)

PART 3

HOW DO YOU MEASURE SHEAR STRENGTH?

Lab Shear Tests

• Direct Shear (most common withDirect Shear (most common with geosynthetics)

• Triaxial Test (performed on cylinders - soil only)

• Torsional Shear (great for large strains, but small specimen and anisotrop notsmall specimen, and anisotropy, not especially common)

• Tilt Table (limited to low normal loads)

Going back to our sliding block on an inclined plane…

128

5

Most common method for geosynthetics is “large-scale” direct-shear device

TxGM vs GCL @ 10,000 psf

Specifying Shear Testing

• ASTM D5321 – Standard Method for Determining the Coefficient of Soil and gGeosynthetic, or Geosynthetic and Geosynthetic, Friction by the Direct Shear Method

• Requires a lot of direction from the engineer.

Direct Shear 1- Basic Setup

• Shear rate: often fixed at 1 mm/min forShear rate: often fixed at 1 mm/min for GS/GS; soil dependent for clays. GCL’s where bentonite is involved with shear should be at d / (50 * t50)

• Interface Condition: Sprayed wet or as-preparedprepared

• Normal Loads: Very important to select representative of project.

129

6

Direct Shear 2 – Hydration and Consolidation

• Hydrate yes or no If yes HOW? What• Hydrate – yes or no. If yes, HOW? What normal load during hydration? How long? Monitor swell?

• Consolidate – How long? Monitor deflection?

Consolidation of GCL/geomembraneInterface shear test ASTM D6243; Normal load = 345 kPa

Direct Shear 3 – Specify Shear Plane and Adjacent Materials

• Select Failure Plane: Interface internal orSelect Failure Plane: Interface, internal, or floating?

• Natural Soil Interface : Specify moisture and density

• Geosynthetic Interface :Clamped or floating condition?condition?

• Upper and Lower Textiles of GCL:Clamped or floating?

Direct Shear 3 – cont.

• Substrate :Soil plywood or concrete (if• Substrate :Soil, plywood, or concrete (if soil, specify compaction)

• Box constraint : is the top box allowed to move for dilation or not?

• Individual set-ups may have details that are p yquite different from each other.

130

7

Take-home message for Part 3:

“Specifying and testing for shear strength is not for the

uninitiated.”

PART 4

HOW DO EVALUATE SHEAR STRENGTH DATA?

Shear force vs deformation curve

Plot normal vs shear stresses

131

8

Mohr-Coulomb Envelope

Generically in addition to a “friction angle”, there also may be a y-intercept we call “cohesion” (c), or sometimes “ dh i ” ( )“adhesion” (a).

Normal load range

Linear strength envelope if cohesion is ignored

Shear strength “left on the table” if cohesion is ignored

Importance of normal load range relative to shear strength parmaters

@ 2.5" Displacement14000

16000

Actual test results showing curved envelope for textured geomembrane vs hydr GCL

6000

8000

10000

12000

HE

AR

ST

RE

SS

(ps

f)

0

2000

4000

0 2000 4000 6000 8000 10000 12000 14000 16000 18000

NORMAL STRESS (psf)

S

862 kPa

Example Safe Shear Strength Results Extrapolations

3000

3500

) Best-fit straight

1000

1500

2000

2500

Sh

ear

Str

ess

(p

sf)

gline through data

Data point (typ)

Extrapolate backwards

Extrapolate forwards horizontally

0

500

0 2000 4000 6000 8000 10000

Normal Stress (psf)

Extrapolate backwards through origin

478 kPa

132

9

3000

3500

f)

Best-fit straight line through data giving results of c = 520 psf, and friction angle = 17 4°

True-life example where actual friction angle is less than that specified, but results are passing

1000

1500

2000

2500

hea

r S

tres

s (p

sf

and friction angle = 17.4

Specification line for friction

Data point (typ)

0

500

0 2000 4000 6000 8000 10000

Normal Stress (psf)

Sh Specification line for friction

angle = 20 degrees and c = 0

478 kPa

3500 Best-fit straight line through

Example where one of the shear points fails the spec, but review of results may indicate passing

1000

1500

2000

2500

3000

ar

Str

ess

(psf

)

g gdata giving results of c = 520 psf, and friction angle = 17.4°

0

500

1000

0 2000 4000 6000 8000 10000

Normal Stress (psf)

Sh

ea

Specification line for friction angle = 22 degrees and c = 0

478 kPa

The effect of dimpling can increase shear strength by up to 30%

(Breitenbach, GFR Jan 2004) DISPLACEM ENT

vs. SHEAR STRESS

Normal

Stress

Test

Point

6000

Point # 5, w/ upper

Textured GM vs. GCL Displacement CurvesDivide by ≈20 for kPa

psf

1000

2500

5000

10000

10000

M OISTURE DATA:

Stress

psi

Point

17.4

34.72.

1.

#####

3.

4.

5.

(GCL)

69.4

6.9

69.4

2000

3000

4000

5000

SH

EA

R S

TR

ES

S (

psf)

Point # 4, w/ upper textile anchored

pptextile not anchored

3) 97

97.0

1) 131

25.6%

NA pcf

Initial Water Content:

Final Water Content:(%)

4) 76

2) 117

116.6

Initial Dry Density:

(GCL)

0

1000

2000

0.0 0.5 1.0 1.5 2.0 2.5 3.0

HORIZONTAL DISPLACEMENT (inches)75 mm

133

10

Shear Secant

Stress Friction

PEAK STRENGTH

Normal

Stress

Test

Point

10000

Textured GM vs. GCL: Peak (#831F)Divide by ≈20 for kPa

psf psf Angle

1000 650 332500 1140 255000 2610 28

10000 4010 2210000 4130 22

##### #####

Adhesion: 350 psf

psi

2.

5.

3.

4.

6.9

17.4

69.4

#####

34.7

69.4

1.

4000

6000

8000

SH

EA

R S

TR

ES

S (

psf)

Mohr-Coulomb parameters

Mohr-Coulomb envelope

Friction Angle: 21 degrees

NOTE: GRAPH NOT TO SCALE

Coefficient of Friction:

0.38

0

2000

0 2000 4000 6000 8000 10000 12000

NORMAL STRESS (psf)

Note concept of “secant”

478 kPa

Shear Secant Normal

STRENGTH ENVELOPE

Test(at 3.0 in. displacement)

10000

Textured GM vs. GCL: “Post-peak”(#831F)Divide by ≈20 for kPa

Stress Friction

psf psf Angle

1000 460 252500 750 17

5000 1620 18

10000 3680 20

10000 3940 22

##### #####

Adhesion: 0 psf

#####

Point

psi

4.

2.6.9

5.

1.

3.

Stress

34.7

69.4

69.4

17.4

4000

6000

8000

SH

EA

R S

TR

ES

S (

psf)

Friction Angle: 20 degrees

NOTE: GRAPH NOT TO SCALE

0.37Coefficient of Friction:

0

2000

0 2000 4000 6000 8000 10000 12000

NORMAL STRESS (psf)

S

478 kPa

GCL peel (ASTM D4632 = 271 N)

Results for geomembrane vs “strong” GCL

Adhesion (kPa) 16.6 12.1

766 kPa

Same project but GCL fails internally (note peel)

GCL peel (ASTM D4632 = 182 N)

766 kPa

134

11

GM/GCL test with internal GCL failure at high normal load

1915 kPa

Interface Shear Strength of TxGm vs NWNP GCL for Two Levels of Texturing

9000

Effect of different types of geomembrane texturing

2000300040005000600070008000

Shea

r St

ress

(psf

)

Peak - asperity 30 milsPost-peak - asperity 30 milsPeak asperity 23 mils

010002000

0 5000 10000 15000 20000

Normal Stress (psf)

Peak - asperity 23 milsPost-peak - asperity 23 mils

Multiple interfaces require a composite strength curve for design

For example:For example:• Some interface may control at lower normal

loads• Internal shear strength of a GCL may

control at higher normal loads.

The controlling slip surface is the one with the lowest peak strength.

PART 5

SLOPE STABILITY CONSIDERATIONS FOR

BOTTOM LINER SYSTEMS

135

12

Typical geometry for bottom liner slope stability analysis showing two common

failure modes.

MSW1x1

x21

H (γ, φ)

L

SubgradeLiner (φ)

16

18

20

6

8

10

12

14

FoundationLow FillHigh Fill

0

2

4

6

HEAP SLOPE FAILURES

1st lift Most Important for Heaps• 1st lift imposes greatest incremental change in stress

A fill h i h i i l h i• As fill height increases, incremental change in stress decreases

• Large incremental change in stress may create unstable conditions until consolidation has occurred

• Wedge analysis shows 1st lift as most unstable

Small slide in edge of heap triggered by waste dump above

Courtesy of Mark Smith

136

13

Failure on dynamic heap caused by oversaturation from broken pipe on top


Toe failure - fine-grained ore with high water level


Kettleman Hills failure Kettleman Hills failure

137

14

Aerial of Kettleman Hills after failure

Liner System

(Base liner)

2-D Slope Stability Analyses

138

15

Finite Element StudyFilz, Esterhuizen, & Duncan (Oct. 2001)

ASCE Journal of Geotec. & Geoenv. Engrg.

FINITE ELEMENT STUDY FROMFilz, Esterhuizen, & Duncan (Oct. 2001) ASCE

Journal of Geotec. & Geoenv. Engrg.

INITIALINITIAL PROGRESSION OF FAILURE

FINITE ELEMENT STUDY FROMFilz, Esterhuizen, & Duncan (Oct. 2001) ASCE

Journal of Geotec. & Geoenv. Engrg.

AT FAILURE

Lesson Learned from Kettleman:

Provide fill-sequencing plan.

139

16

Liner Failure in Liner Failure in Midwest. Midwest.

End where GS pulled out End where GS pulled out of anchor trenchof anchor trench

Exposed clay linerExposed clay liner

Courtesy of Craig Benson

Torn Edge of Torn Edge of GeomembraneGeomembrane

AnchorAnchortrenchtrench

Exposed Exposed clay linerclay liner

Torn edge of geotextileTorn edge of geotextile

Courtesy of Craig Benson

(In-)Famous Rumpke Failure

The Rumpke failure couldfailure could just as well have involved a GCL interface

140

17

Just before failure…

Courtesy of Bob Koerner

Just after failure.

Courtesy of Bob Koerner

Effect of liquid level on slope stability São João Landfill –August/2007

141

18

Main factors:- Steep slope 1(V):1,5(H); Large area without cover during the rain period ( ~100mm ); lack of internal drainage.

Bogota: Flow slide due to aggressive leachate injection

WHICH SHEAR STRENGTH TO USE?WHICH FS TO USE?

??

?

Limit Equilibrium Analyses• Spencer’s Method• Bishop’s Method• Morgenstern and Price• Janbu• NAVFAC Block and Wedge method• Etc.What do they all have in common?FS is assumed to be same at any location. (This is an

invalid assumption regarding shear stress distribution, as we learned from the FEA.)

142

19

Potential for Progressive Failure

• NON-UNIFORM SHEAR STRESS DISTRIBUTION

• PORE PRESSURES REDUCING EFFECTIVE SHEAR• PORE PRESSURES REDUCING EFFECTIVE SHEAR STRENGTH

• SEISMIC LOADING

• CONSTRUCTION DEFORMATIONS

• WASTE AND FOUNDATION SETTLEMENT

• AGING AND CREEP

• 25% OR MORE UNBONDED FROM THE START

Points of Discussion• Many GS interfaces are strain-softening

(brittle) – including internal geocomposite• Scenarios exist that can lead to exceedance

of peak strength• Those 2 factors can lead to progressive

failure• No standard of practice evolved in united• No standard-of-practice evolved in united

states regarding selection of FS vs. shear strength

Recommendations for Practice• Conduct material-specific tests• Attempt to position critical plane above

i bprimary geomembrane• Consider evaluating slope stability using

residual with FS at least > unity (use residual strengths of interfaces that have the lowest peak strengths over each normal l d )load range)

• Think about unexpected pore pressure scenarios

Trend: Probabilistic vs. Deterministic• Classic “Factor of Safety” is deterministic• Some industry trend to look at probability of

failure, or the inverse which is reliability, y• See Duncan, April 2000 ASCE J. of Geotech &

Geoenv. Engrg; also GRI XVI (Dec. 2002)• Using both approaches is the most intelligent – the

idea is than in some instances a project with FS=1.3 can have less probability of failure than

th j t ith FS 1 5another project with FS=1.5.• A simple spreadsheet can be set up to be a

powerful tool to help understand the sensitivity and reliability of certain variables

143

20

Actual probability analysis from project for bottom liner stability

Variables: Soft Soils, F1: Waste Strengths, F2:Lowest conceivable value (LCV) = 19.16 kPa 20 degreesHighest conceivable value (HCV) = 119.7 kPa 40 degreesOne std dev (σ) of shear strength = (HCV-LCV)/6 = 16.76 3.3Most likely value (MLV) of shear strength = 47.89 kPa 30 degreesg

MLV-σ = 31 kPa 26.7 degrees

MLV+σ = 64.66 kPa 33.3 degreesConsolidation Strength Gain; New LCV = 41.67 kPa

VF = coefficient of variation of the factor of safety.

Soft Soils Waste Strengths

log-normal

CONDITION FMLV FS-σ FS+σ ΔF1 FS-σ FS+σ ΔF2 σF VF

normal relia-bility index

βLN

Relia-bility

Proba-bility of failure

Pf

3:1 + benches 1.54 1.26 1.81 0.55 1.49 1.6 0.11 0.284 0.184 2.2693 98.84% 1.162%

Fill to 1st Bench 2.49 2.08 2.91 0.83 0 0.415 0.167 5.4286 100.00% 0.000%3:1+Strength Gain 1.54 1.44 1.81 0.37 1.49 1.6 0.11 0.198 0.129 3.3050 99.95% 0.047%3:1+single bench 1.5 1.2 1.73 0.53 1.49 1.6 0.11 0.274 0.183 2.1444 98.40% 1.600%

Key Stability Issues for Big Leach Pads

• Deeper fills challenge the limits of testing• Variability in ore grade/material segregationVariability in ore grade/material segregation• Ore degradation due to dissolutioning• Pile saturation from irrigation and precip., valley

configuration, and ore degradation• Interlift liners are common• Largest pads are often in most seismically activeLargest pads are often in most seismically active

areas (e.g. northern Chile and southern Peru)• Static liquefaction from saturated low-perm ore

All this sounds good, but…

• Most important factor is to perform good• Most important factor is to perform good basic analysis

• Use appropriate shear strengths• Naïve slope stability engineering shown in

following 2 examples, all done by reputable g p , y p“experienced” engineering firms.

Case History No. 1

144

21

Case History No. 1 Case History No. 1 Revised Analysis

Case History No. 2 Case History No. 2

145

22

Revised Slope Stability Analysis- Case History No. 2

Take heed:

• Slope stability should be done by qualified• Slope stability should be done by qualified practitioner

• Slope stability is the single most immediate and largest liability for the designer (and gravely impacts the other stakeholders as well)

Final note on slope stability for “wet” landfills

• Abundant side slope seeps and flooding of• Abundant side slope seeps and flooding of gas wells indicative of saturated waste

• Clogged LCS may result in leachate mounding

• Slope stability can be at riskp y

PART 6

DESIGN OF VENEER COVER SYSTEMS

146

23

Geosynthetic tension, T

SLOPE STABILITY MECHANISMS

Cause of Instability:

Interface shear strength

Toe buttressingφ = internal friction anglec = cohesion

Cause of Instability: weight of soil layer

gδ = interface friction anglea = adhesion

AND DEFINITION OF STABILITY PARAMETERSCourtesy of J.P. Giroud

Seepage Forces

PRECIPITATION

GEOMEMBRANE

WATER

Water thickness: partial or total

Courtesy of J.P. Giroud

Slope Stability EquationsINFINITE SLOPE WITHOUT WATER

tanFS

δ

INFINITE SLOPE WITH WATER

tan tanγ δ δ

tanFS =

β

ABOVE BELOW

tan tan

tan tan

0.50 to 0.55 0.5

A BbA B

sat

b

sat

= = FSFSγ δ δ

β βγγ

γ= ≈ VERY SIGNIFICANT

Courtesy of J.P. Giroud

Schematic of Head Buildup in the Drainage Layer (after Thiel & Stewart, 1993, Geo ‘93, Vancouver BC)

Water percolating through topsoil into drainage layer

L

Pi d/ hi age layer (K )2Topsoil (K )1

Final Cover

DGeomembrane

Pipe and/or trench drainage outlet (TYP)

D = maximum depth of flow

Drainage

β

147

24

Drainage Design For Side Slopes in Landfill Caps

• Qin = kveg*L*1• Q = k *i*A = (k *t)*i = θ*i• Qout = kg i A = (kg t) i = θ i

Cover Soil (Saturated)Gradient = 1

Qin

*vegreq

K Li

θ =

L

Qout

Geocomposite

inreq i

Infinite slope eqn with seepage

( )[ ]( )[ ]( )[ ] βγγγ

βφγγγγsinhhhh

costanhhhhhcFSSAT2w2w211

wwSAT2w2w211

+−+−+−++

=

Thiel and Stewart, 1993

Typical outlet design Natural slope veneer failure

Infinite slope failure in Seattle, WA, USA. Sheets of soil slide off the hillside.

148

25

Cover Slope Failure ‘Block’ Failure of Final Cover

Why did it fail? Given these parameters:

• Slope Angle = 8.5 degrees, slope length = 94 mδ 12 d b t t d th• δ = 12 degrees between geonet and smooth

HDPE liner)• Cover Profile: 15 cm top soil, 45 cm silty sands

(k=5*10-4 cm/sec), single bonded geocomposite,and a smooth HDPE geomembrane.

• Geocomposite Transmissivity = 8*10-4 m3/sec-mGeocomposite Transmissivity 8 10 m /sec m(4 gal/min/ft)

• A HELP analysis indicated that the top soil andsand did not saturateThe predicted drainage F.S. = 4.5

Why did it fail ?

• Inspection of the failed cover clearly indicatedthat the cover had saturated.The actual drainage F.S. = 0.26 !(unit gradient assumption – saturated slope)

The slope stability F.S. = 0.7 !Cl l h fi l ill b blClearly the final cover will become unstablewhen the vegetative soil layer saturates.

149

26

Landfill cover failure during constructionSurface Erosion at Top of Slope

Veneer Failure on Balance of Slope Why did it fail ? Given these parameters:3:1 slope, β = 18.4°Slope length L = 400 ft = 122 mGradient I = 0.33 Permeability of the cover soil K=5*10-3 cm/secSaturated unit weight of cover soil βsat = 112 pcfTransmissivity of the composite T = 3.5* 10-4

m2/secGeocomposite / Texture geomembrane frictionGeocomposite / Texture geomembrane friction angle φ =27°

150

27

Why did it fail ?• Qin = 0.006 m3/s; Qout = 0.00011 m3/s

Drainage FS = Qin / Qout = 0 02Drainage FS = Qin / Qout = 0.02 • Slope stability FS:• - FS (dry) = (tan27/tan18.3) = 1.5

– FS (sat) = (112-62.4)/112 * (tan27/tan18.3) = 0 7 != 0.7 !

Seepage induced failure is predictable !

Courtesy of Mauricio Abramento

CONCLUSION

IMPORTANCE OF DRAINAGE

FOR STABILITY OF

LANDFILL COVERSLANDFILL COVERS

Material Failure – Example on 2:1 Slope

151

28

Internal failure of geocomposite underdrain while placing clay liner (UK landfill) Why did it fail?

• Significant unbonded areas• Significant unbonded areas.• 2(H):1(V) slope (27 degrees)• Internal friction angle on unbonded areas 24

degrees• Equipment loading can be up to 30% of• Equipment loading can be up to 30% of

equipment weight in localized area• Progressive failure

152

29

TWO GEOCOMPOSITE PANELS

Plan View of Situation

UNBONDED MFGRGDEFECT ("BLISTER")

DOZER PUSHING

UNBONDEDSEAM AREA

SOIL UPSLOPE

From Thiel and Narejo, GRI-18, “Geofrontiers”, Austin, TX, Jan 2005

[ ] PPhPhC

FS a

30sin)(tancos])[(

++×+×+

=βγ

φβγ

)2)(2(2 hyhxWP

++=

Soi

lde

pth

h

[ ] PPh 3.0sin)( ++× βγ

CRITICALINTERFACE

N

F WW * a/g (equip

accel.)

Stress DrivingStress Resisting

=FS

y = 0.1111x

100

125

150

(kPa

)

= Current

= Koerner, 2001

100

150

y = 0.11x, which equates to:

0 11 kPa per gm/cm of peel;

50

75

Shea

r-ad

hesi

on S

treng

th (

ear-a

dhes

ion

(kPa

)

50

0.11 kPa per gm/cm of peel;

or 417 psf per ppi peel

0

25

0 200 400 600 800 1000 1200Ply-adhesion Strength (grams/cm)

(Chiado & Walker, 1993)

Ply-adhesion (gm/cm)

She

200 400 600 800 1000

ADHESION BETWEEN GEOTEXTILE AND CORE

PEEL TEST

153

30

CHART FOR PEEL STRENGTH OF 180 gm/cm (1 ppi) (Chart assumes 0.3m wide unbonded seam that results

in unbonded area below dozer track = 1 m2)

2

2.5

0.5

1

1.5

2

FA

CT

OR

OF

S

AF

ET

Y

perfect bondingoutside seams1 sq meter blister

2 sq meter blister

0

0.5

0% 10% 20% 30% 40% 50% 60% 70%

SLOPE ANGLE

3 sq meter blister

Considerations for Internal Shear Strength of Geocomposites (interface between geonet and

laminated geotextile)

• You start out with a significant unbondedYou start out with a significant unbonded area due to edges and, and manufacturing start/stop operations.

• Non-uniformity in manufacturing.• Assume at least 25% unbonded when it is

installedinstalled.• High potential for progressive unbonding on

slopes due to equipment loads.

Lessons Learned

• Progressive failure can occur withProgressive failure can occur with equipment loads acting over unbonded areas (specify limits on unbonded areas due to manufacturing)

• Specify minimum 1 ppi (180 g/cm, or 175 N/m) peel MARV175 N/m) peel MARV Case History – Veneer Reinforcement

Failure

154

31

Problem was geogrid directly on liner; should have been “embedded” in soil Lessons Learned on Veneer

Reinforcement

• Understand limitations of reinforcement• Understand limitations of reinforcement –embeddment is very important

• Construction QC is critical

155

32

Veneer Failure GT/GMCase History: Cincinnati Test Plots

3H: 1V Plots

2H: 1V Plots

Slide 3 Weeks after Construction Slides at Geomembrane-GCL Interface

Woven GT facing up

156

33

Lab Test Results from Cincinnati Test Program (all were single-point tests at a normal load of 17 kPa = 355 psf) Lessons Learned from Cincinnati

• GCLs hydrate on subgrade use minimum• GCLs hydrate on subgrade – use minimum 6-oz non-woven geotextile carrier against geomembrane

• Long term reinforced GCLs on 3:1 slopes should be no problem for covers

Recent Cover Failure10 years after Cincinnati the same mistake is being made

Case History – Failure Due to Landfill Gas Pressures

157

34

How to design for gas uplift?Infinite slope equation with qgas forces

Thiel, 1999

158

35

Schematic of gas-relief system LFG Collection Blanket Design

⎥⎤

⎢⎡ 2DQ

u gasgasγ

maxu allowable gas pressure=

Thiel, 1999

⎥⎦

⎢⎣

=8maxu

reqθ

LFG Collection Blanket DesignDesign Example - Slide 1

60 ft=WASTEH3

3

2

60 ft

50 lb/ft10 m12.8 N/m

6"H O 1490 N/

=

=

=

ρ

γ

WASTE

WASTE

LFG

H

D

226"H O 1490 N/m= =allowableu

The allowable gas pressure is selected to prevent geomembrane uplift

LFG Collection Blanket DesignDesign Example - Slide 2

= ρLFG LFG WASTE WASTEq r H

( )( )

3 2

6 3 2

2

6 27 2

0.1 50 60 300 ft /ft /year

2.9 10 m /m /s

8

2.9 10 12.8 10 3.1 10 m /s

−

−−

= × × =

= ×

⎡ ⎤= ⎢ ⎥

⎣ ⎦

× ⎡ ⎤= = ×⎢ ⎥

γθ

θ

LFG

LFG LFGLFGreq

allowable

LFG

q

q Du

There is no factor of safety in the above calculations. It should be included into Wreqθ

( )( )7 6 2

3.1 10 m /s1490 8

10 3.1 10 3.1 10 m /s− −

⎢ ⎥⎣ ⎦

= × = ×

θ

θ

LFGreq

Wreq

159

36

Sand and pipe gas underdrain system

Underdrain GC with pipes Combination Gas and Seep Collection Layer

(0.45 m)

(45 m)

160

37

Seeps where there is no GC

Geocomposite underdrainGeocomposite underdrain installed

No geocomposite

Seeps

Case History: Cross section of liner system

41

New slide

30 cm GRAVEL LAYER

PVC GEOMEMBRANE

1

W-NW GCL WITH NW SIDE UP

SUBGRADE

LCRS Placement on 4:1 Slope View of failed slope; PVC ripped at crest and slid down with gravel; GCL is intact and not

stressed

161

38

Initial Shear Strength Test Result 35° Friction (Secant) – results do not simulate failure

5 kPa

5 kPa

Close-up view of torn PVC and underlying GCL – condensation noted on

underside of PVC

Shear Strength Test Result with Sprayed Interface 15.9° Friction Peak; 13.5° Friction

Post-Peak (4:1 slope = 14.04°)

5 kPa

5 kPa

Lessons (re)learned:

• Shear strength testing needs to be• Shear strength testing needs to be performed at the normal loads reflecting the design conditions. This may require low normal loads for construction, mid-range normal loads for operations, and high

l l d f l i di inormal loads for ultimate conditions.

162

39

Lessons (re)learned:

• Virtually all geomembranes installed willVirtually all geomembranes installed will form condensation when placed over earthen subgrades (or GCLs on earthen subgrade), even in – this is a real issue.

• Conclusion: Spray all relevant geomembrane interfaces with water during test assembltest assembly.

PART 7

SEISMIC ISSUES

Earthquake Review - Plate Tectonics

Location and position of continents and oceans over the past 237-millions-year period, on the basis of plate movements.

After 171 million years of drift (66 million years ago).

163

40

Seismic Stability• The Problem: earthquakes

add vertical and horizontal forces to slopesforces to slopes.

• The horizontal forces may be significant destabilizing forces for a slope.

The waste is characterized as a sliding block with gsome coefficient of friction on the base, that moves when subjected to a critical acceleration.

This analysis revolves around determining the critical (or yield) acceleration, ky, that causes movement to begin.

Factors that influence seismic slope stability include:•duration of earthquake (seconds) (longer duration ⇒more displacement)

•magnitude of earthquake (acceleration usually a•magnitude of earthquake (acceleration, usually a percent of the acceleration due to gravity (g)) (larger magnitude ⇒more displacement)

•frequency content of earthquake

•natural period of soil profile and/or waste profile (influences resonance). When the predominant frequency of shaking is similar to that of the soil at the q y gsite and/or the waste mass at the site, resonance is likely to occur. Large displacements occur.

•strength of the soil

•slope geometry

Landfills have proven very robust against earthquakes

Northridge Earthquake Mw=6.7

Liquefaction Potential of Heap Leaches

Irrigation and rain causeIrrigation and rain cause saturation

In this case earthquakeIn this case earthquake triggered liquefaction (southern Peru, 2001,

M = 8.4)

164

41

Recent 2010 Concepcion Subduction EQ

• Mw = 8 8 with 7 2 aftershock• Mw = 8.8 with 7.2 aftershock• No heap failures reported• One unlined up-stream-method tailings dam

failed and killed two people

Seismic Concerns

• Slippage a k a slope stability failure• Slippage, a.k.a. slope-stability failure. Could be in foundation, along base liner, or cover veneer.

• Liquefaction (foundation or within waste)• Other (utility rupture, settlement)( y p , )

Seismic Stability(1)

Procedure to determine the critical acceleration ky that causes movement to begin

1 Estimate the ma im m hori ontal acceleration1 Estimate the maximum horizontal acceleration (MHA) in a rock below the facility

(1) Method of Bray et al., 1998

2 Calculate the seismic coefficient for base and cover sliding (“base” means between the base of the waste and the liner; “cover” means for the cover veneer sliding off the top of a completed landfill).

• MHArock is the maximum horizontal acceleration in the bedrock underneath the landfill (from seismicity

)

0.75 rockMHAkg

=

map)• This approach not applicable to deep, soft clay

foundations, and those susceptible to liquefaction.

165

42

Step 3. Calculate the yield acceleration, ky, for the bottom liner or the cover system. Generally you would use equations or a slope stability program to calculate the acceleration that results in FS = 1.0.

Step 4. If ky > k, expect seismic-induced di l t l th 150 (f thdisplacements less than 150 mm (for the base) and 300 mm (for the cover).If ky < k (or if 150 mm and 300 mm movements are unacceptable), proceed with the “simplified displacement

d ” t ti t thprocedure”, to estimate the displacements more closely.

Chart-Based Method Reference• J.D. Bray, E.M. Rathje, A.J. Augello, and S.M.

Merry (1998), "Simplified Seismic Design Procedure for Geosynthetic Lined Solid WasteProcedure for Geosynthetic-Lined, Solid-Waste Landfills“, Geosynthetics International 1998, Vol. 5, Nos. 1-2, Pages 203-235

• More recent: J.D. Bray and Thaleia Travasarou, “Simplified Procedure for Estimating Earthquake Induced Deviatoric Slope Displacements”,Induced Deviatoric Slope Displacements , Journal of Geotechnical and Geonvironmental Engineering, ASCE, V. 133(4), pp. 381-392, April 2007

Simplified Procedure to Estimate Base and Cover Displacements (Bray and Travasarou, ASCE JGGE, April 2007)

The following is used to estimate the base and cover displacements. There are three steps in this procedure.

166

43

Step 1. Estimate the dynamic properties of the fill

A Estimate yield acceleration (k ) usingA. Estimate yield acceleration (ky) using standard slope-stability analysis

B. Estimate the average shear-wave velocity of the sliding mass using literature estimates (or field measurements)

C Estimate the f ndamental period of theC. Estimate the fundamental period of the sliding mass.

Tools for properties of sliding massAvg shear wave velocity (Kavazanjian et al. (1996))

Fundamental period (Bray, 2007)References

• Kavazanjian et al. (1996). "In Situ Shear Wave Velocity of Solid Waste from Surface Wave Measurements " Proceedings of the SecondMeasurements. Proceedings of the Second International Congress on Environmental Geotechnics

• Bray, J.D. (2007) “Chapter 14: Simplified Seismic Slope Displacement Procedures,” Earthquake Geotechnical Engineering, 4th Inter. Conf. onGeotechnical Engineering, 4th Inter. Conf. on Earthquake Geotechnical Engineering - Invited Lectures, in Geotechnical, Geological, and Earthquake Engineering Series

167

44

Step 2. Estimate the earthquake characteristics (from hazard analysis)

A. Estimate earthquake moment magnitude q gMw

B. Estimate the spectral acceleration at a period equal to 1.5 times the fundamental period of the sliding mass

Example: Fundamental period Ts of sliding mass estimated to be 0.3s. Find spectral acceleration at degraded period = 1.5Ts

Site-specific spectral acceleration curve from design earthquakeT design earthquakeTs

Th t f th i i

Step 3. Calculate displacement

The amount of the non-zero seismic displacement D (cm):

ln(D) = −1.10 − 2.83*ln(ky) − 0 333*[ln(ky)]2 + 0 566*ln(ky)*ln(Sa) +0.333*[ln(ky)]2 + 0.566*ln(ky)*ln(Sa) + 3.04*ln(Sa) − 0.244*[ln(Sa)]2 + 1.5*Ts + 0.278*(M − 7) + ε

Where:ky=yield coefficient; Ts=initial fundamental period of the sliding mass in seconds; Sa=5% d d l ti t l l ti f th it ’damped elastic spectral acceleration of the site’s design ground motion at a period of 1.5Ts in the unit of g; M=earthquake’s moment magnitude; and ε= normally distributed random variable with zero mean and standard deviation of 0.66. To eliminate the bias in the model, when Ts < 0.05s, the first term should be replaced with -0.22 (instead of -1.10).

168

45

Example calculation results showing sensitivity of estimated displacement to yield acceleration

1000

)

Median

10

100

Med

ian

Dis

pla

cem

ent

(cm

84% Percentile

16th Percentile

1

0.00 0.10 0.20 0.30 0.40

M

Yield Coefficient

Landfill Interest Group Seminar , Oct 2011

Thank you.A tiAny questions

regarding slope stability?

169


A SUSTAINABLE MINERAL BARRIER OPTION

Dipl.-Geol. M. Naismith1, Ing. E. Timmermans2 and Dr. H. Mulleneers2

1Trisoplast Mineral Liners International BV, German Branch, Kruppallee 21, 24146 Kiel, Germany; PH +49 431 6686944; FAX 49 431 6686945; email: [email protected]. 2&3Trisoplast Mineral Liners International BV, Head Office, Oude Weistraat 17, 5334 LK Velddriel, the Netherlands; PH +31 418 636030; FAX +31 418 633790 email: [email protected] SUMMARY New concepts have been developed to improve the properties of different lining systems in order to enable them to better cope with any occurring detrimental effects. On Trisoplast®, as one of these modified barriers, intensive independent testing has proven that the performance of the mineral material can be significantly improved by making use of a special bentonite-polymer technique. 1. INTRODUCTION The main aspects that commonly have an adverse affect on the functionality of mineral liners are discussed, whereas the pros and cons of geomembranes are not evaluated in this article. Since the requirements for sealing systems in landfills and contaminated land applications are generally higher compared to other applications a lot of research is based on these challenging conditions. The ready-mixed material is installed (see Figure 1) as a robust layer of generally between 6 cm and 10 cm compacted thickness and should be covered shortly after by a geotextile or geomembrane in combination with a mineral covering layer which provides the necessary confining pressure. Water from the environment surrounding the layer causes the bentonite clay to swell and to further develop a network of chemical bonds with the dissolved polymer thus creating a dense hydro gel structure (see Figure 2).

+

+

+

+

+

++

++

+

+

+

+

+

+

+

++

+++

++

++

++

+

+

+

+

-

-

-

-

-

++

+

+

+

+

+

+

Legend

+

-

Mineral filler

Pore water

Air

Bentonite

Polymer Carbon

van der Waals Forces

Cationic Binding

Anionic Binding

Hydrogen Bridge-Binding

Specific Binding

Monovalent Cations

Bivalent Cations++ +

++

++

- - -

- - -

Figure 1. Overview Installation Figure 2. Schematic Principle After intensive independent, mainly government sponsored research and testing, the unique polymer-enhanced mineral liner initially became the preferred mineral barrier for landfill applications in the Netherlands. As a result of its performance combined with further ongoing research it is now used in landfill, contaminated land, industrial, infrastructure and construction as

170

well as landscaping projects within Europe and an increasing number of other countries. The main reason for its widespread application is that its use raises the level of soil and groundwater protection significantly higher than the level required by the individual national regulations. Ongoing investigations have proven that the clay gel offers improved properties compared to traditionally used mineral barriers. 2. CATION EXCHANGE, DESICCATION & PLANT ROOT PENETRATION Intensive field and lysimeter investigations especially in the 90ies showed that traditional clay seals (CCLs) may fail within several years even under rather humid climatic conditions due to desiccation, penetration and water uptake by plant roots as well as shrinkage, if they are not protected by a thick layer of topsoil with high storage capacities for plant-available water or by other protective measures, e.g. geomembranes (Melchior, 2001). Additionally the low tensile strength of thick clays and BES (sand bentonite seals) can lead to crack formation due to stresses resulting from differential settlements. Desiccation of mineral liners underneath landfills resulting from moisture movement under the influence of temperature gradients have also been modelled (Döll, P., 1996) and reported (Philip et al., 2002).

GCL (geosynthetic clay liners), too, are sensitive to desiccation, plant root penetration, and shrinkage. Field data on the performance of GCL is given in Melchior, 2002. Furthermore cation exchange leads to a rapid decrease of the swelling capacity of sodium bentonites unless the overlying cover soil is exceptionally sodium rich. This not only leads to a general increase in permeability but can, in combination with desiccation and cracking, be the cause of a dramatic loss of the sealing function of the mineral seal as the remaining swelling capacity of the bentonite is not sufficient to re-seal the gaps and cracks that were initially caused. Evaluation of field data has shown that covering a GCL with a soil layer of 750-1,000 mm thickness or with a geomembrane overlain by soil does not ensure protection against ion exchange or large increases in hydraulic conductivity (Meer, S. and Benson, C., 2007).

The polymer within the modified mineral liner interacts with the bentonite and sand forming a spider-web like gel structure. As a result of these bonds the total cation exchange capacity (CEC) of the clay gel is reduced (Hoeks et al., 1991) and thus the risk of cation exchange. Investigations have shown that cation exchange was significantly reduced when compared to BES and GCLs, making it more stable and more analogous to the relatively stable clay mineral Illite (Boels 2001, Guyonnet 2008). Permeability tests with a wide range of salt solutions are currently under investigation to determine the performance of the mineral liner under extreme conditions. The tests indicate that if high salt concentrations which are strongly Ca² dominated are already present in the filler material, a negative influence can also be noticed. The initial gel formation thus is obviously negatively influenced by the reduced swelling capacity of the clay. As a result it is advised to limit the electric conductivity value of the sand to 500 µS/cm, whenever the Ca²+ to Na+ ratio is uncertain or strongly dominated by Ca²+. Desiccation is a result of water evaporating from within the soil and depends on several factors (heat conductivity, soil temperature, air convection etc.). Ductile behaving soils reduce their volume in relation to the moisture escaping, due to shrinkage during the drying process. The process continues until the shrinkage limit, as defined in soil mechanics, is reached. Until then occurring shrinkage is almost directly proportional to the moisture content. Further reduction in volume leads to characteristic shrinkage cracks caused by tensile stresses as a result of the menisci bending into the void spaces, which pull the soil particles together. The volume reduction leads to tensile forces between these relatively dry-resistant areas. If the tensile strength of the soil is exceeded, cracks start forming.

171

The tensile strength of different soil types varies significantly. The tensile strength σz is related by Coulomb’s break conditions to the shear angle φ and the cohesion c respectively (σz = 2c * tan (45 – φ/2). Behrens & Egloffstein (2002) showed that the tensile strength at break (slow loading) of mineral liners according to Coulomb can be determined related to the moisture content. Figure 3 shows the results of the modified mineral liner and an average clay. The modified material shows significant advantages compared to conventional clay based mineral barriers when assuming that the prevention of crack formation as a result of drying out can be related to the tensile strength of the material (Behrens et al., 2002). This remarkable behaviour is confirmed by laboratory experiments in a purpose built apparatus to simulate field conditions. The modified mineral liner and a natural clay barrier were subjected to drying and wetting cycles with calcium rich water and their behaviour was compared. Whilst the clay sample showed desiccation cracks and quick responses to changing conditions after the first drying cycle, the enhanced sample remained homogeneous and undisturbed even after a second desiccation period with soil water tensions of well above 1,000 hPa (Melchior et al., 2001). Whilst further tests with other clays and various types of GCL under the same conditions all showed strong influences and increased permeability the modified mineral sample remained fully functional even after 6 year duration of extreme conditions. Further laboratory tests showed a similar behaviour of the mineral materials when subjected to plant root penetration in combination with desiccation. Whereas the clay barrier had dried out with open gaps, the enhanced mineral samples maintained their integrity and the permeability remained unaffected (Melchior et al., 2001).

0

25

50

75

100

125

150

175

5 10 15 20

moisture content %

tens

ile s

tren

gth

kN

/m²

CLAY

TRISOPLAST®

Figure 3. Tensile strength of modified barrier and clay (Behrens et al., 2002)

3. SLOPE STABILITY Long term slope stability can be problematic when designing with clay based mineral barriers. The problem occurs due to the clay’s generally very low friction angle. This can be a particular problem with the very low shear strength of a pure bentonite layer. The problem is addressed within GCL by mobilising the tensile strength of fibres or threads in needle punched or stitch-bonded GCL respectively. Their durability under aggressive environmental conditions, e.g. as found in a landfill or contaminated land application is critically important for long term slope stability. A subject of present discussion and research work is the selection of the correct design value as GCL show a significant difference between the peak and the residual shear strength. The modified mineral barrier makes use of the advantages of both non-cohesive and cohesive behaviour because its sandy structure gives high shear angles, with numerous contacts of the coarse grained particles, combined with high cohesion values resulting from the clay fraction. Thus significantly steeper slopes can be designed, without further reinforcement.

172

The independent German working group AK Trisoplast, 2002 recommends using the shear characteristics shown in Table 1 as typical values for the construction phase (moisture content on the dry side of the proctor curve) and the long term evaluation (wet side of the proctor curve up to full saturation). These values include a safety margin and offer a guideline for preliminary design planning with modified mineral liners. Behrens & Neumann, 2002 report significantly higher cohesion values gained from laboratory measurements. As a result, projects with slope angles of up to 1:1.5 (34° inclination) and more than 60 m length have successfully been realised.

Table 1. Shear characteristics of the modified mineral liner (recommended for slope stability calculations for preliminary design planning for landfill covers according to AK Trisoplast, 2002

Construction phase Long term shear parameters State of failure State of sliding State of failure State of sliding ’ = 35° ’r = 30° ’ = 30° ’r = 30° c’ = 20 kN/m2 c’ = 10 kN/m2 c’ = 10 kN/m2 c’ = 10 kN/m2

4. DIFFERENTIAL SETTLEMENT Poor or soft subgrades in general and to a greater extent the heterogeneity of the waste body in landfill applications lead to differential settlement that causes extra stresses and deformation within lining systems. Due to the low tensile strength of clays and BES this can easily lead to the formation of cracks resulting in an increase in permeability. In order to demonstrate the deformation capability of the enhanced mineral liner, a special testing device was developed by Boels (see Figure 4) which allows the measurement of the permeability after subsequent deformation steps. Tests were performed on a 2.5 cm thick layer of saturated as well as unsaturated samples. It is remarkable that even at 10% biaxial deformation, which locally reached even higher values, the permeability was not significantly affected (Boels et al., 1999). TD Umwelttechnik GmbH & Co. KG determined the stretching limits for the modified mineral liner by direct tensile tests. Neglecting the general stretching (normally less than 0.01 % and insignificant when comparing it to the stretching of about 1% to 3% at the outer surface) the acceptable radius (R) as shown in Figure 5 can be determined in accordance with the deformation requirements of the Deutsche Gesellschaft für Geotechnik e.V. (DGGT), 1997:

R = rFε*3

d*2, (d = barrier thickness, rFε = stretching at surface incl. safety factor of 2).

The above mentioned permeability results by Boels et al., 1999 show that in practice even smaller radii (< 1 m) are possible. The reasons for this derive from the positive effects caused by the high swelling and healing ability of the bentonite-polymer component in combination with the loads applied by the covering materials.

173

Figure 4. Trisoplast layer after deformation in opened strain testing device

Figure 5. Acceptable minimum bending radius R in m for the 0.1 m thick modified mineral layer

5. DURABILITY OF THE POLYMER Before its market introduction in the Netherlands the chemical and physical durability of the newly developed mineral barrier was examined intensively. Various tests showed that the material was not or only negligibly affected by any of the tested conditions. Based on the investigation results it was concluded that the polymer in the clay-polymer gel will not be attacked or degraded to a measurable level. Based on these results the clay-polymer gel was evaluated and approved in the Netherlands as a resistant system for basal lining and capping of landfills. Additional experiments were carried out by Dr. Reinhard Wienberg – Umwelttechnisches Labor in Hamburg, Germany. Wienberg labelled the polymer using radioactive 14C isotopes and subjected the samples to a variety of microbiologically and chemically active environments in closed reactors and studied the degradation. The radioactive labelling technique enables to measure degradation of the polymer in a fairly short time even at very low levels. Over two years of quantifying testing on 14C-labelled polymers showed that the actual polymer is marginally or totally non-degradable even under extreme microbiological conditions. Wienberg concluded that microbiological degradation of the polymer offers no risk for the application of the modified mineral barrier as sealing material (Wienberg, 2003). 6. FIELD INVESTIGATION & TESTING ON EXCAVATED SAMPLES Samples from excavations of the enhanced mineral barrier four to ten years after installation were carried out in order to gain field data regarding the performance of the mineral barrier under different conditions. The permeability values measured on samples of excavated locations in the Netherlands are listed in Table 2. Most chosen locations were using a single mineral lining system since potential aging effects would be more likely to occur. In order to examine the durability of the sealing layers, the moisture content and permeability of every sealing layer was determined after a visual inspection.

174

Table 2. Permeability values of excavated Trisoplast in the Netherlands

Project Covering Layers Age in years

Kvalue in 10-11 m/s

Landfill, Rotterdam Partly geomembrane, 0.6 m soil 6 1.3 -2,6

Tank Park, Rotterdam

Geotextile, 0.6 m sand/gravel 5 1.6

Landfill, Almere Geotextile, 0.2 m sand, 0.8 m soil 5 1.5 – 4

Landfill, Soesterberg

Geomembrane, drainage mat, 1.0 m soil

5 2.1

Highway, Badhoevedorp

Road construction, 1.3 m covering soil 9 1.4

Landfill, Mook Drainage mat, 0.8 m covering soil 4 2.1

Tank Park, ‘s-Hertogenbosch

Geotextile and 0.4 m sand/gravel 6 2.0 – 3.0

Remediation Diemerzeedijk

Geotextile and 1.2 m covering soil 10 ~0.5

The excavated sites showed no signs of desiccation, cracks or other structural changes within the modified mineral barrier. The samples were still moist and ductile (see Figure 6). Electron microscopy used on samples gained from some of the sites demonstrated that even at microscopic level the composition of the layer was homogeneous and unaffected (see Figure 7). The picture shows the polymer forming a gel with the bentonite which fills the pores and binds to the individual sand grains thus forming an almost impenetrable, elastic layer.

Figure 6. Excavated mineral layer Figure 7. Microscopic picture 7. FUNCTIONAL LIFETIME & AFTERCARE As ‘eternal’ aftercare of closed landfill sites plays an increasingly important role, especially with respect of sustainability, the expected functional lifetime of a sealing system has to be considered when determining the funds that have to be reserved for overhaul and/or repair. The functional lifetime of a mineral barrier is defined as the elapsed time until the performance of the material drops below the required minimum performance. There are a number of factors as described in the former paragraphs that lead to negative changes within the mineral barrier’s performance. A method for determining the functional lifetime as a function of the surrounding environment controlled by the chemical composition of the soil layers adjacent to the liner and the liquid passing through is described in Boels et al. 2003.

175

In the past ten years approx. 80% of the landfill basal liners and cappings in the Netherlands were constructed using the enhanced mineral barrier due to its quality and competitiveness. Despite the sometimes higher initial investments (installation costs) compared to other barrier constructions permitted according to the regulations (e.g. clay and sand-bentonite) it turned out to be financially more attractive to use the modified mineral lining system because of its longer functional lifetime resulting in lower funds required for the aftercare of closed landfills. Unlike the Netherlands, the majority of countries do not have similar regulations for fund reservations ensuring sufficient aftercare costs to be taken into consideration. Unfortunately this often results in short term solutions purely based on the lowest installation costs and resulting in significantly higher costs in the long run as well as failing to achieve a sustainable situation. Thus not only the actual costs and environmental risks are simply passed on to the next generation but in most cases the costs for remediation and repair as well as the environmental risks are significantly higher than if carried out correctly in the first place. 8. EQUIVALENCY TO THICKER CLAY LINERS An alternative mineral liner to be approved is generally compared with the 0.5 m to 1 m thick reference clay barrier. Its acceptance should be based on its overall performance taking the effects resulting from the surrounding environment into consideration. Beside its hydraulic properties, mechanical properties such as slope stability, crack-free deformability or resistance to desiccation as well its resistance to cation exchange need to be evaluated. The retardation capability of contaminants can be one of the main requirements for certain applications. ‘Ranking the performances of different liners in a construction after the EC-guidelines on the basis of the emission attenuation, the following sequence was found: Trisoplast > clay (clay fraction 40%) > sand-bentonite > clay (clay fraction 10%) > GCL’ (Boels et al. 2003). Elaborate modelling research after measuring diffusion and adsorption properties with radioactive labelling also confirmed that, with regard to its pollutant retention capacity, the modified mineral barrier layer (0.09 m) – particularly in combination with a naturally existing mineral layer (0.41 m) – is equivalent to the natural reference mineral sealing layer) as specified in the EU Landfill Directive (Wienberg, R, 2005). 9. CONCLUSIONS Mineral Barriers are widely used for sealing applications due to their natural durability, robustness and their resulting long functional lifetime. With Trisoplast the combination of its natural mineral components with a specially developed polymer has lead to enhanced properties in a number of relevant aspects for various kinds of applications. These improved performances have not only been demonstrated by various independent and renowned testing laboratories and institutes on lab samples but also on excavated field samples after several years of use. The resulting long functional lifetime is an important contribution towards the general requirement of realising sustainable designs. 10. REFERENCES AK Trisoplast (2002) "Empfehlungen zur Herstellung von Abdichtungen aus Trisoplast" (Version: 17.07.2002). Appendix 1 of 4 AK Trisoplast (2002) "Gemeinsame Stellungnahme der im Arbeitskreis Trisoplast vertretenen Landesbehörden vom 12.08.2002" Behrens, W. & Egloffstein, T, (2002) "Zur Einbaudicke von Trisoplast bei temporären und endgültigen Abdichtungssystemen. " Müll & Abfall, Vol. 11

176

Behrens, W. & M. Neumann (2002) "Untersuchungsergebnisse zu einigen mechanischen Eigenschaften von Trisoplast". Müll & Abfall, Vol. 2 Boels, D, H. te Beest, H. Zweers & P. Groenendijk (2003) "Investigation of the functional lifetime of Trisoplast in relation to chemical compositions of pore water solutions in barriers." Wageningen, Alterra, Green World Research. Alterra-rapport 528. 50 pp. 16 figs.; 12 tables; 25 refs Boels, D. & K. van der Wal (1999) "Trisoplast: New developments in soil protection." In: Christensen, T. H., Cossu, R & R. Stegmann (ed.): Sardinia 1999. Proc. of the seventh Int. Landfill Symposium in Cagliari, Italia, Vol. I, p. 77 – 84 Boels, D. (2001) "Comparing performance of Trisoplast with different mineral liner materials." In Christensen, T. H., Cossu, R & R. Stegmann (ed.): Sardinia 2001. Proc. of the eighth Int. Landfill Symposium in Cagliari, Italia, Vol. III, p. 45 – 54 Boels, D. and G.J. Veerman, (1996) "Permeability of Trisoplast for different liquids." The Netherlands, Wageningen, DLO-Winand Staring Centre. Rapport 487, pp. 24 (Dutch) Boels, D. and K. van der Wal (1999) Trisoplast : "New developments in soil protection." In Proceedings Sardinia 99, Seventh International Waste Management and Landfill Symposium. S. Margharita di Pula, Cagliari, Italy; 4-8 October 1999 Boels, D. en D. Schreiber (1999) "Effects of bi-axial strain on the permeability of mineral liner materials." Wageningen, DLO-Staring Centrum. 21 pp. .5 fig.; 4 tab. (Dutch) Deutsche Gesellschaft für Geotechnik e.V. (DGGT) (1997) "GDA-Empfehlungen Geotechnik der Deponien und Altlasten." 3.Auflage, Ernst@Sohn Verlag, Berlin Döll, P. (1996): "Modelling of moisture movements under the influence of temperature gradients: Desiccation of mineral liners below landfills." PH.D Thesis, Technical University of Berlin, Bodenökologie und Bodengenese Guyonnet, D., D. Cazaux, H. Vigier-Gailhanou, B. Chevrier, M. Gamet (2008) "Trisolix: Compatibility testing of Trisoplast®." BRGM/RP-56850-FR. McNeal, B.L. and N.T. Coleman (1966) "Effect of solution composition on hydraulic conductivity." Soil Science Society of America Proceedings Vol. 30: 308-312 Meer, R., Benson, C, (2007) "Hydraulic conductivity of Geosynthetic Caly Liners Exhumed from Landfill Final Covers" Melchior, S. (2001) "Performance and design of cappings for contaminated sites and landfills." In Sarsby, R.W. & T. Meggyes (Hrsg.): The exploitation of natural resources and the consequences. Thomas Telford Publishing, London, S. 95-106 Melchior, S. (2002) "Field studies and excavations of geosynthetic clay barriers in landfill covers." In Zanzinger, H., R. M. Koerner & E. Gartung (eds.): Clay Geosynthetic Barriers, A.A. Balkema Publ., Lisse, Abingdon, Exton (PA), Tokyo, p. 321- 330 Melchior, S., B. Steinert & O. Flöter (2001) "A comparison of traditional clay barriers and the polymer-modified material Trisoplast in landfill covers." In Christensen, T. H., Cossu, R & R. Stegmann (ed.): Sardinia 2001. Proc. of the eighth Int. Landfill Symposium in Cagliari, Italia, Vol. III, p. 55-64 Melchior, S., Steinert, B., Boels, D. (2002) "Aufgrabungen von Oberflächenabdichtungen mit Trisoplast – Zwischenergebnisse." In Ramke et al. (2002): Austrocknungsverhalten mineralischer Abdichtungsschichten in Deponieoberflächensystemen. Status-Workshop der Deutschen Gesellschaft für Geotechnik am 31.1. bis 1.2.2002 in Höxter, 317-329

177

Philip, L.K., Shimell, H., Hewitt, P.J., Ellard, H.T. (2002) "A field-based test cell examining clay desiccation in landfill liners", Quarterly Journal of Engineering Geology and Hydrogeology, 35, 345 – 354, Geological Society of London Quirk, J.P. and R.K. Schofield (1955) "The effect of electrolyte concentration on soil permeability." Journal of Soil Science, Vol. 6,2: 163-178 Rowell, D. L. (1963) "Effect of electrolyte concentration on the swelling of oriented aggregates of montmorillonite." Soil Science, 96: 368-375 Wienberg, R. (2003) "Bericht über die Untersuchung zur Beständigkeit von Deponieabdichtungen aus Trisoplast gegenüber mikrobieller Beeinflussung" Wienberg, R. (2005) "Berechnungen zur Schadstoffrückhaltung von Trisoplast" Wienberg, R. (2003) "Laboruntersuchungen am Deponiedichtungsmaterial Trisoplast zur Bestimmung der die Schadstoffrückhaltung bestimmenden Parameter Sorption und Diffusion"

178

landfill 2011 seminar, 18 - 20 october, durban, kwazulu-natal

Documents