Ecological Engineering xxx(2009)xxx–xxx
Contents lists available at ScienceDirect
Ecological Engineering
j o u r n a l h o m e p a
g e:w w w.e l s e v i e r.c o m/l o c a t e/e c o l e n g
Characterising alternative recycled waste materials for use as green roof
growing media in the U.K.
Chloe J.Molineux a,?,Charles H.Fentiman b,Alan C.Gange a
a School of Biological Sciences,Royal Holloway University of London,Egham,Surrey TW200EX,United Kingdom
b Fentiman Consulting,31Nutham Lane,Southwater,West Sussex RH139GG,United Kingdom
a r t i c l e i n f o
Article history:
Received18March2009
Received in revised form8June2009
Accepted13June2009
Available online xxx
Keywords:
Extensive green roof
Substrate-based roofs
Lightweight aggregates
Material characterisation
Biodiversity
Plant growth
a b s t r a c t
We characterised four recycled materials that have been manufactured into useful substrates for use on
extensive green roofs.These were a crushed red brick(the U.K.industry standard substrate base and
therefore used as a control)and three alternative pellets made from:clay and sewage sludge(waste
clay from excavations,?y ash and sewage sludge),paper ash(from recycled newspapers)and carbonated
limestone(from quarry?nes).Investigations into optimal organic content–conifer-bark compost for plant
nutrients–and characterisations such as pH,particle size distribution,loose bulk density,particle density,
XRF and leachate analyses were performed.Greenhouse experiments showed signi?cant interactions
between the four aggregates and the amount of added organic material,meaning that organic addition
did not have the same effect on plant growth in each aggregate.The addition of organics also signi?cantly
reduced the pH of the recycled aggregates,making growing conditions for plants more favourable in
these substrates.Particle density and loose bulk density results have shown all substrates to be classed
as lightweight aggregates and leaching analysis has con?rmed that all substrates perform within legal
leachate limits for drinking water.As all the aggregates are commercially available at similar costs to the
crushed red brick control,we believe that the alternative substrates have great potential in the green roof
market and as they can be locally sourced we would also suggest that they are as good,if not better,than
the industry standard,both economically and environmentally.
?2009Elsevier B.V.All rights reserved.
1.Introduction
Green roofs are generally classi?ed into two types of sys-
tems:extensive and intensive.Intensive systems are more like roof
gardens supporting large trees and shrubs,but requiring deep sub-
strates and regular maintenance.Extensive systems are generally
substrate-based with a vegetated layer or a Sedum mat,either on
its own with a sponge membrane for moisture retention or with
a substrate base;offering between2.5and10cm deep root zones
due to restrictions by weight loading on a building’s structure.In
the U.K.,substrate-based vegetated roofs concentrate on maximis-
ing biodiversity by encouraging plant species diversity(although
there are all sorts of reasons for installing these types of roofs),
whereas Sedum mat systems generally comprise only stonecrop
plant species and are installed for clients wanting an instant‘green’
effect.The purely substrate-based green roofs are relatively cheap
to install compared to the Sedum systems,aim to recycle waste
materials(such as broken bricks)and have been shown to support
?Corresponding author.Tel.:+441784443188;fax:+441784414224.
E-mail address:C.J.Molineux@https://www.doczj.com/doc/ac6390752.html,(C.J.Molineux).
rare invertebrates and birds(Gedge and Kadas,2005).They have
predominantly been created using crushed brick or demolition
waste,including crushed concrete as their substrate,in an attempt
to mimic natural brown?eld sites found in urban environments
(Gedge,2000;Grant et al.,2003).These‘brown’or‘biodiverse’roofs
are usually constructed for this type of habitat mitigation in the
U.K.,especially in London,as the only litigation imposing construc-
tors to install green roofs comes from the conservation of a rare
bird species,the Black Redstart.This is a common species in many
parts of Europe but in the U.K.it is a rare breeding species with most
breeding sites at roof top level in large cities,especially London and
Birmingham.Replacement of old buildings poses a direct threat to
the species and as result habitat recreation is necessary to preserve
the species(https://www.doczj.com/doc/ac6390752.html,).
In the U.K.sourcing substrates raw materials is challenging
because crushed brick materials(as speci?ed by the German FLL
standards)are not always available within50km of the roof so
long distance haulage is often necessary.Other potentially suit-
able materials are available,such as crushed demolition wastes,
but these have to be processed in order to remove any nails or steel
that may harm the roof waterproo?ng membranes,adding cost.
Furthermore,alternative lightweight substrates for green roofs,
0925-8574/$–see front matter?2009Elsevier B.V.All rights reserved.
doi:10.1016/j.ecoleng.2009.06.010
2 C.J.Molineux et al./Ecological Engineering xxx(2009)xxx–xxx
such as LECA,Lytag,pumice and lava(Emilsson and Rolf,2005) are generally manufactured overseas and are not locally available; thus varied green roof habitats for vegetation are not often pos-sible.Therefore cost-effective,recycled,sustainable alternatives to crushed brick need to be found and assessed for use in the growing U.K.market(Fentiman and Hallas,2006).
There has been very little biodiversity research conducted on green roofs in the U.K.Currently,unless there is the aforementioned compulsion to establish a green roof for black redstarts,architects and developers install green roofs for non-ecological reasons,such as aesthetical appeal,for green credentials and for economic value like thermal insulation and to reduce water run-off,as?ooding is becoming increasing problematic in the U.K.(EA,2003).For this reason they tend to use commercially available ready-made Sedum matting on very thin layers of substrate or directly onto a mois-ture mat;this generally does not allow natural plant colonisation nor offers the varied,species diverse environment that is desirable for most invertebrates that prefer deeper substrate bases(Gedge and Kadas,2005;Kadas,2007).Therefore most studies by green roof researchers seem to centre on water run-off quality and ther-mal properties provided by vegetated roofs.Water run-off quality is measured by the quantities of leachate contaminates,e.g.high phosphorus levels from too much organic fertilisation(Berndtsson et al.,2006;Emilsson et al.,2007).Studies have also been con-ducted to?nd out what effects substrate depth and roof slope have on water absorption and therefore quantities of run-off(Nicholaus et al.,2005),hydrological function(Bengtsson et al.,2005)and peak?ows(Villarreal and Bengtsson,2005).Thermal properties of green roofs(the vegetation layer)have been investigated and have revealed that the plants themselves reduce summer air tempera-tures signi?cantly(Niachou et al.,2001),thereby emphasising the importance of vegetation cover.Life cycle assessments of vegetated buildings have also been conducted,concluding that energy costs can be greatly reduced by green roofs and that they can reduce the urban heat island effect(Booth,2006;Saiz et al.,2006).These stud-ies concentrate on economic bene?ts rather than biodiversity,but are nonetheless vital if green roofs are to become part of planning and development in the U.K.and other developed countries.
In this study we discuss the nature of the substrate,which is the basis of the entire green roof system.Guidelines have been pro-duced for the green roof industry in Germany(FLL,2002),however these standards are not always compatible with the U.K.market (e.g.they do not permit the use of recycled concrete or calcareous aggregates)and in these cases British standards have been followed. Relatively little has been published on alternative green roof grow-ing media,especially from the U.K.,and we believe that in order to achieve the desired green roof,an engineered substrate must be characterised.As substrate-based green roofs in the U.K.are gener-ally for biodiversity(and Sedum roofs for economic and aesthetical appeal)it is important to determine if the alternative materials support vegetation in a similar or more successful way to the U.K. industry standard.This paper considers the following,1)can recy-cled secondary materials support vegetation like a commonly used substrate in the U.K.,and2)are these recycled substrates viable alternatives in terms of material characterisations and economical costs.
This study has taken the U.K.green roof industry standard sub-strate of crushed red brick and compared it to three other recycled aggregates–all wastes that are usually sent to land?ll–includ-ing:sewage sludge,waste clay,?y ash,paper ash and quarry?nes. The sewage sludge waste is combined with locally sourced waste clay and?y ash from Tilbury,Essex and pelletised into usable lightweight aggregate by RTAL(Tilbury),hereafter termed‘clay pel-lets’.This company also manufactures waste paper ash pellets in a similar way,using ash produced by Aylesford Newsprint Ltd.(Ayles-ford,Kent)when recycling newspapers.These‘paper ash pellets’are lightweight and can be produced to varying sizes depending on their intended purpose.Finally,Carbon8Contracting(Chatham, Kent)produce lightweight pellets from carbonating quarry waste (limestone based)by the use of waste carbon dioxide to improve structure and strength and to lower pH(Hills et al.,1999),hereafter termed‘carbon8pellets’.Each of these aggregates is combined with an organic component,producing a substrate that can viably be manufactured at similar costs to the crushed red brick.In this study all four substrates have been characterised to further understand their potential as growing media for green roofs in the U.K.and although FLL guidelines have been considered,it was not always possible to relate?ndings to those in the standards due to the calcareous nature of the materials.
2.Materials and methods
https://www.doczj.com/doc/ac6390752.html,anic content
Before characterisation and trial experiments could begin, investigations were undertaken to establish the optimal amount of organics that should be added to the substrates,as a source of nutrients.Nutrients are required for healthy plant growth and as substrate-based green roofs should require very little mainte-nance,the right organic content in a substrate is vital if further fertilisation of the system is to be avoided(Emilsson et al.,2007). Commercially available top dressing compost was chosen,contain-ing50:50conifer-bark compost and medium clay soil hereafter termed‘organics’.Seventy-two pots(8cm height×6cm width) were set up in a greenhouse containing nine replicates of each aggregate(crushed red brick,clay pellets,paper ash pellets and carbon8pellets)with15%(by volume)organics and nine replicates with25%organics.The greenhouse temperature ranged from12to 27?C over the duration of the experiment and watering was given to all pots in equal amounts approximately every2–3days.The pots were sown with ten Plantago lanceolata(Ribwort plantain)seeds. https://www.doczj.com/doc/ac6390752.html,nceolata,a commonly used bait plant or phytometer(Bartelt-Ryser et al.,2005),was chosen to represent a wide range of plant species and it is often used in ecological experiments as it can with-stand a wide range of pH values,is found pan-globally and can survive in all types of habitats;even harsh environments(Grime et al.,1988).After initial germination,seedlings were removed to leave three healthy individuals per pot,most of these seedling sur-vived but in a few cases the replicate number was reduced due to mortality.Plant heights and total shoot biomass were measured for each pot after two months of growth.
2.2.Aggregate characterisation
2.2.1.pH values
The pH was determined for each aggregate and then each sub-strate(aggregate plus organic component).The?rst measurement was for the four aggregates where each had nine replicates.Thirty grams of material was soaked in75ml of distilled water for24h then three readings,using a HANNA HI4521pH meter,were taken for each sample to get an accurate mean for each of the replicates (as there can sometimes be small variations between readings).The second measurement was taken in the same way for the four sub-strates two months later,after plants were grown and subsequently harvested from the materials used in the greenhouse experiment (described in Section2.1).
2.2.2.Particle size distribution
The particle size distribution within batches of different mate-rials(aggregate with no organics added)was determined using BS
C.J.Molineux et al./Ecological Engineering xxx(2009)xxx–xxx3
EN933-2:1996.Representative batches of each aggregate(approx-imately1kg from25kg samples)were separated through a range of sieves sizes(mm):<0.063,0.125,0.25,0.5,1,2,4,6.3,10,14and 20.After each sieve was passed it was weighed to determine the amount of material in the batch to be of that particular size.
2.2.
3.Loose bulk density and void spaces
Characterising the bulk density and void spaces using BS EN 1097-3:1998shows how each material will naturally compact down on a roof and provides information on volume and weight.A container(1.4l capacity)was?lled with each aggregate and indi-vidually weighed and re-weighed three times to give a mean loose bulk density per aggregate.The materials were not pressed down in any way,but levelled by hand to ensure no material was above the container edge.
2.2.4.Particle density and water-holding capacity
Unlike loose bulk density,which considers the amount of mate-rial that will?t into a container including air spaces,particle density (BS EN13055-1:2002)can be used to accurately calculate mass to volume ratios.Firstly,containers with approximately600ml capac-ities were?lled with distilled water and a glass slide was placed on top making them airtight.These were then weighed.Once emp-tied and dried they were then?lled with300g of each aggregate (oven dried to60?C for24h)and soaked in400ml distilled water for24h to ensure all materials were fully saturated.Distilled water was then?lled up to600ml,the glass slides were placed on top making them airtight and then they were re-weighed.Finally,the 300g samples of each aggregate were surface dried and re-weighed. Results include densities when oven dry and fully saturated with water(also giving%water content)and can be used with loose bulk density to calculate%void space within a sample.For all aggregates, representative samples were taken from large batches(over25kg).
2.2.5.X-ray?uorescence analysis
As the chosen substrates are from waste sources it is important to?nd out what elements,potentially harmful to the environment, are contained within the manufactured aggregates.Therefore,the four aggregates were subjected to standard XRF testing following XRF Scienti?c Ltd.Preparation guidelines,to determine chemi-cal oxide composition for selected elements(Al,Ca,Cr,Cu,Fe,K, Mg,Mn,Na,Ni,P,Rb,S,Si,Sr,Ti,Zn and Zr).The results,from Philips PW1400XRF,contain elemental oxide percentages,which for all aggregates should fall between the standard error ranges of 98–102%.
2.3.Leachate analysis
During rainfall it is common for materials to leach elements contained within them,including heavy metals possibly damaging to the environment.Therefore leachate analysis using inductively coupled plasma optical emissions spectrometry(ICP-OES)was con-ducted.The manufactured aggregates(with no compost added) were analysed using BS EN12457-3:2002.For the leaching experi-ment,20g of each substrate(replicated three times)were weighed into plastic nitric acid washed bottles and combined with40ml of distilled water.These were placed onto a rotating turntable and left for6h.After this time,leachates were collected and?ltered through Whatman0.45?m CA w/GMF disposable housing with?lters.The materials were then drained and the bottles were topped up with 200ml of water(absorbed water by the materials was taken into account).These were then put back onto the turntable and left for a further18h.Leachates were
collected and?ltered as before and all samples were then run through a Perkin Elmer Instrument,Otima 4300DV for analysis,in particular for Al,Ba,Ca,Cd,Cr,Cu,Fe,K,Mg,Fig.1.(a)Height of Plantago lanceolata when grown with varying compost quanti-ties in four different aggregates.Bars represent means±one standard error.(b)Leaf biomass of Plantago lanceolata when grown with varying compost quantities in four different aggregates.Bars represent means±one standard error.
Mn,Na,Ni,Pb,Sr,Ti and Zn.Results are given as elements present, as the analysis will account for speciation.
2.4.Data analysis and statistics
Differences in plant growth between the organic treatments and substrates were examined with two-factor ANOVA after testing for normality and homogeneity of variances.One-factor ANOVA was used to examine differences in pH values between the substrates and for selected leachates.Means were separated with a Tukey’s HSD post hoc test(Fowler et al.,1998).All analyses were conducted using the statistical package UNISTAT?.
3.Results
https://www.doczj.com/doc/ac6390752.html,anic content
We found that an increased organic component produced larger plants(F1,56=14.429,P<0.001)(Fig.1a)with more shoot biomass (F1,14=36.279,P<0.001)(Fig.1b)in each of the four aggregate types. There were also signi?cant differences(F3,56=9.150,P<0.001) found between the aggregates for plant heights.Taller plants were observed in the crushed red brick compared to the clay pellets and carbon8pellets;and the paper ash pellets produced taller plants than the carbon8pellets(Fig.1a).For the shoot biomass data there was signi?cantly more biomass(F3,14=28.643,P<0.001) from plants grown in the crushed red brick compared to those in the paper ash pellets;and from plants in the carbon8pellets com-pared to those in the paper ash pellets(Fig.1b).However,there was also a signi?cant interaction between organic content and aggre-gate type for both plant height(F3,56=2.840,P<0.05)and shoot
4
C.J.Molineux et al./Ecological Engineering xxx (2009)xxx–xxx
Table 1
Material characterisations of the four aggregates intended for use as green roof growing media.
Aggregate Water Content,%Particle Density (Dry),Mg/m 3Particle Density (Wet),Mg/m 3Apparent Density,Mg/m 3Loose Bulk Density,Mg/m 3Void Space,%Red Brick 20.70 1.59 1.92 2.360.8347.45Clay Pellets
17.70 1.49 1.75 2.010.8344.50Paper Ash Pellets 5.70 1.82 1.92 2.030.8951.11Carbon8Pellets
26.00
1.85
2.50
2.50
0.91
64.00
Fig.2.pH of the four aggregates in their raw state,combined with 15%compost and combined with 25%compost.Bars represent means ±one standard error.
biomass (F 3,14=8.370,P <0.01).For the plant height data,the higher organic component did not produce taller plants in the clay pellets,yet did in the other aggregate types;similarly,the shoot biomass was not signi?cantly increased in the clay pellets with increasing organic content.
3.2.Aggregate characterisation
3.2.1.pH Values
The pH varied considerably between the four aggregates (F 3,96=2310.897,P <0.001).The lowest mean values were for clay pellets (pH 8.53)showing mild alkalinity,followed by paper ash pellets (pH 9.33)and red brick (pH 9.71).The highest mean values were for carbon8pellets (pH 11.80)showing strong alkalinity.There was a decrease in pH for all materials after the substrates were com-bined with organics and plants were grown in them (Fig.2)and for carbon8pellets,pH was reduced by an average of 2.71units.This was signi?cant in all but the paper ash pellets after 15%organics was added.Therefore there was a signi?cant interaction (F 6,96=209.262,P <0.001)between the organic treatment (15%or 25%)and aggre-
gate type because otherwise,the addition of organics would have reduced the pH values in every case.
3.2.2.Particle size distribution
Particle size distribution testing clearly showed variations in size proportions between the different materials (Fig.3).The crushed red brick had higher concentrations of smaller particles ranging from 5to 9mm,whereas the paper ash pellets were larger,mostly ranging from 7to 11mm.The clay pellets distribution was between that of the red brick and paper ash,(mostly 6–9mm)and the car-bon8pellets had the widest range spanning from 5to 11mm.3.2.3.Loose bulk density,particle density and water-holding capacity
The various methods of material characterisation (Table 1)have shown that all the materials can be classi?ed as lightweight aggre-gates,which is very important for green roof substrates,as all fall within the set limits of particle density ≤2.00Mg/m 3and loose bulk density (≤1.20kg/m 3)(BS EN 13055-1).Moreover,water-holding capacity varied between aggregates,ranging from 5.7%to 26%(Table 1).
3.2.
4.X-ray ?uorescence analysis
XRF analysis (Table 2)identi?ed 18oxides common to all mate-rials.These ranged in abundance (measured as %by weight)and showed what each aggregate was made from.In particular,alumina content was highest in clay
pellets (19.9%)and lowest in carbon8pellets (8.6%),calcium oxide was highest in Carbon8pellets (40.4%)and lowest in clay pellets (3.3%)and potassium and sodium oxides were evenly ranged between the materials (from 0.5%to 2.9%and 0.2%to 0.5%,respectively).3.3.Leachate analysis
Statistical analysis of leachates concentrated on four elements;namely aluminium,calcium,potassium,and sodium,because
Fig.3.Particle size distribution,where cumulative weight has been calculated,for 1kg sample for each material.
C.J.Molineux et al./Ecological Engineering xxx(2009)xxx–xxx5
Table2
XRF analysis showing elemental composition,%of oxides,and the loss on ignition
(LOI)for the four aggregates.
Element(oxide)Red Brick Clay Pellets Paper Ash
Pellets
Carbon8Pellets
Al2O317.0219.9111.558.69
CaO10.37 3.3936.9940.47
CrO0.040.010.020.02
Cu2O0.010.010.020.01
Fe2O3 5.13 4.89 1.37 3.03
K2O 2.98 2.460.50 1.18
MgO 1.55 1.64 2.89 2.33
MnO0.040.070.050.11
Na2O0.540.560.220.46
NiO0.000.010.010.01
P2O50.21 1.170.290.38
Rb2O0.030.010.010.01
SiO254.1356.8825.7142.83
SO0.600.020.200.42
SrO0.040.090.080.09
TiO20.97 1.870.350.35
ZnO0.020.020.020.01
ZrO20.020.030.040.01
LOI(%) 3.10 3.7024.1023.10
Table3
Heavy metal leachates,mg/l from the four aggregates when analysed by ICP-OES.
Heavy metal leachate (mg/l)Red Brick Clay Pellets Paper Ash
Pellets
Carbon8Pellets
Aluminium0.2a0.3a 3.5b0.9a
Barium0.00.00.30.0 Cadmium0.00.00.00.0
Calcium543.5a37.0b443.8a176.8b Chromium0.00.00.00.1
Copper0.00.00.00.0
Iron0.00.10.10.0
Lead0.00.00.00.0 Magnesium 2.48.10.10.0 Manganese0.00.00.00.0
Nickel0.00.00.00.0 Potassium11.6ab 5.8a24.7b11.3a Sodium 1.4a 4.3a34.5b 2.8a Strontium0.20.30.00.4
Titanium0.010.819.00.3
Zinc0.00.10.10.0
All values are means from three replicates per aggregate.Aluminium,calcium,potas-sium and sodium analysed with ANOVA and differences separated with Tukey-test. Values not sharing the same letter indicate a signi?cant difference(P<0.05).
these were most prominent in our samples and were common to all materials(Table3).There were signi?cantly higher con-centrations of aluminium(F3,20=15.004,P<0.001)and sodium (F3,20=14.943,P<0.001)in the paper ash pellets compared to all other materials.There were also signi?cantly higher levels of cal-cium leachates(F3,20=49.648,P<0.001)found in crushed red brick and paper ash pellets compared to the clay and carbon8pellets. Finally there were signi?cantly higher concentrations of potassium (F3,20=5.759,P<0.001)found in the paper ash pellets compared to clay and carbon8pellets,but levels were similar to those in the red brick leachates.
4.Discussion
The greenhouse experiment revealed signi?cant effects of organic content as well as substrate type and a signi?cant interac-tion between the two variables.This shows that if novel materials from recycled sources are to be considered for green roof substrates, then careful consideration of the amount of organics to include is highly important.Interestingly,plant growth in the crushed red brick was no different(in terms of plant heights)to the paper ash pellets and for plant shoot biomass data;the red brick was com-parable to the clay pellets.Plant heights however,may not be as important as biomass for determining plant success on a green roof, as wind speeds and less competition for sunlight(due to sparse veg-etation cover and little shading)may be a limiting factor on how tall plants will grow.It has been suggested that ideal extensive green roof growing media should include between10%and25%organic matter with90–75%substrate(Beattie and Berghage,2004).Our ‘organic matter’was a blend of compost and soil(50:50)added to the aggregates in our greenhouse experiments as15%and25% by volume,thus the amount of real organic matter actually added was half these amounts,7.5%and12.5%respectively,which falls within the limits outlined by FLL standards(FLL,2002).From these results we have decided to add12.5%conifer-bark compost(or25%‘organics’by volume)to all of our aggregates for all future experi-ments.This should be the optimal amount to sustain healthy plant growth,not add too much weight to the substrates and help reduce substrate shrinkage,which occurs when too much organic mat-ter/compost is added(Snodgrass and Snodgrass,2006).Clearly the suggested ratio of organic:inorganic content of a growing medium (Beattie and Berghage,2004)has a large range and will affect plant growth performance and biomass depending on the type of sub-strate used,as shown in our growth experiment,where results indicate that certain substrates perform as well as the red brick and could be used as alternative materials on extensive green roofs. Further work,testing these substrate mixes with25%organics on a green roof situation,are underway and it will be interesting to see if these observations are true for other plant species.We would suggest that different substrates would support slightly different plant communities due to the interactions occurring between the substrates themselves and the organic component(on a chemical level).
The pH analysis of the four materials used in this study has revealed that all of the growing media are alkaline in nature,even the industry standard red brick is higher than those limits set out by FLL guidelines(pH6.0–8.5).Both the clay pellets and paper ash pel-lets were lower in pH than the crushed red brick before any organics were added and for all cases the carbon8pellets remained the high-est;this may mean that once the carbon8pellets are on a green roof, plant growth may be limited to certain species that can tolerate highly alkaline conditions.Carbon8Contracting are improving their substrate by changing the mixture contents and ratios and are also trialling large-scale production and ways to improve carbonation (Gunning,https://www.doczj.com/doc/ac6390752.html,m.);hopefully this will mean they manufac-ture pellets with lower pH values that may be more suitable as a growing media.The pH was however,signi?cantly reduced with the addition of organics(slightly acidic in nature)and with the growth of https://www.doczj.com/doc/ac6390752.html,nceolata plants for all aggregates.Intriguingly,the amount of organics added made little difference to pH,15%gave a similar reduction as25%in all but the paper ash pellets.As this material produced the only pH not to be reduced by15%organics,there must be a signi?cant interaction between the aggregate type and the amount of organics added.
The material characterisation data have shown that all the substrates differ in terms of particle size distribution.This may be important for plant species richness and abundance once these substrates are contained on new green roofs.A heteroge-neous substrate will be more effective at supporting a diverse plant community(Wilson,1999;Brenneisen,2003)compared to a homogenous one,due to the way in which the materials compact down,trap organic matter in the void spaces and hold water(Steila and Pond,1989),both within the pores of the substrate itself and in the microstructure of a particle’s surface.The particle densities and loose bulk densities show water-holding capacity as well as the
6 C.J.Molineux et al./Ecological Engineering xxx(2009)xxx–xxx
void spaces available.The clay pellets and carbon8pellets have sim-ilar water-holding capacities to the red brick(approximately20%) and whilst the paper ash pellets hold far less water(only around 5.7%),which may inhibit successful plant growth,they do have a higher percentage of void spaces(51%).These spaces could provide an increased opportunity for trapping organic and particulate mat-ter as well as air spaces for increased aeration;all important abiotic factors for successful plant growth(Steila and Pond,1989).Water logging on green roofs is possible due to reduced storm-water run-off(Emilsson,2008)therefore aeration is particularly important (FLL,2002)to prevent plant root rotting.
Further material characterisations include X-ray?uorescence (XRF)analysis showing chemical analysis of the four aggregates. The crushed red brick and clay pellets contained substantial amounts of alumina(elemental analysis between17%and20%)and all contained high amounts of silica(elemental analysis between 26%and56%)indicating that these materials contain aluminium silicates(found in clay minerals)and silicon oxide(or quartz). As expected,calcium oxide content was high(40%)in the car-bon8pellets,due to limestone and calcite,which forms during the carbonation process(calcium carbonate plus carbon dioxide). Elemental oxide analysis of the Earth’s crust show similar values, as found for our materials,for iron,magnesium,potassium and sodium(Steila and Pond,1989)and soil from forests and pastures in Brazil(analysed using XRF)have also shown similar composition percentages,for aluminium,manganese,phosphorus,silicon and titanium oxides(Herpin et al.,2002).Therefore,all four materials seem promising as potential green roof growing media.
Leachates found for all materials were below the legal limit set for drinking water by the world health organisation(WHO, 1998)and the U.S.environmental protection agency.Aluminium levels were signi?cantly higher in leachates from the paper ash pellets(3.5mg/l)compared to all other materials,however these levels are still lower than those thought to be toxic to plants(Rowe and Abdel-Magid,1995).Calcium concentrations were higher from the crushed red brick and paper ash pellets compared to the clay and carbon8pellets and concentrations of potassium and sodium were also higher from paper ash pellets compare to all others.
A high level of calcium in growing media is not thought to be directly toxic to plants(Rowe and Abdel-Magid,1995)but it will cause an increase in substrate pH and have an indirectly negative affect on plant performance,as many plants cannot tolerate high pH levels.Both calcium and potassium are major soil cations and essential macronutrients for plants(Troeh and Thompson,2005) therefore the release of these metals from the aggregates should increase availability to the plant roots and improve plant growth and health.
In the U.K.there are relatively few sources of suitable crushed red brick for the manufacture of green roof substrates and conse-quently most of that used in the U.K.is sourced from one plant in Cambridgeshire,U.K.and transported around the country to where it is needed.There are also other markets for this material meaning that its demand and therefore price,is relatively high(Shirem-inerals,https://www.doczj.com/doc/ac6390752.html,m.).For London based roofs,haulage costs are signi?cant and also the environmental impacts,in terms of carbon footprints,can be high.The alternative materials used in this study are not only from recycled sources but are also local to the London area and therefore haulage costs;both economically and environ-mentally,are kept to a minimum.The demand for these secondary materials is also lower than that for brick as other markets for their use are limited,thus the cost of these aggregates is generally lower than that of crushed brick(Shireminerals,https://www.doczj.com/doc/ac6390752.html,m.).For this reason the alternative materials used in this study have the poten-tial of being delivered to the London area more economically than the industry standard—crushed red brick.Clearly this also depends on the commercial availability of the alternative materials tested, but this study has established the principle that locally sourced recycled materials can provide economically viable alternatives to crushed red brick and whenever possible should be assessed for suitability.The hope for future green roof substrates is that they are manufactured regionally with suitable local secondary materials so that these bene?ts can also be seen in a wide variety of locations. Once research has determined the ability of these alternative sub-strates to support plant growth effectively,life cycle assessments (LCAs)should be carried out;unfortunately this was not within the scope of this study.
5.Conclusion
This study has shown that by selecting alternative recycled materials for use as engineered green roof substrates,similar prop-erties and characteristics of the already used crushed red brick, that conforms to FLL standards,can be achieved.If alternative sub-strates can be provided and even combined together on green roofs,enhanced plant species diversity and healthy plant growth would be possible.This would not only give extensive green roofs improved biodiversity,but would also mean that waste materials like sewage sludge,paper ash and quarry?nes could?nd a sec-ondary use and so be recycled rather than being sent to land?ll sites as they are at present.Clearly other recycled materials could also be assessed for possible use and we would recommend that this type of research be carried out as well as more advanced economical and environmental impact evaluations in the form of LCAs.
Our results have shown that the alternative substrates perform as well if not better,than the widely used crushed red brick as grow-ing media,in terms of plant growth and material characterisations. They are also similar in price to the red brick substrate(Shirem-inerals,https://www.doczj.com/doc/ac6390752.html,m.)and are already commercially available.In conclusion we suggest that these materials are viable alternatives, and due to reduced haulage and transportation costs,are more envi-ronmentally sound than the crushed red brick for green roofs in London.
Acknowledgements
We are grateful to the Natural Environment Research Council (NERC)and CASE partner Carbon8Contracting,for funding this study.We would like to thank Peter Gunning at University of Greenwich for his assistance in material characterisation,to Chris Hallas of Shire Minerals(Southern)for donating large quantities of aggregate/substrate and to Dusty Gedge for his help and advice throughout the project.
References
Bartelt-Ryser,J.,Joshi,J.,Schmid,B.,Brandl,H.,Balser,T.,2005.Soil feedbacks of plant diversity on soil microbial communities and subsequent plant growth.
Persp.Plant Ecol.Evol.Syst.7,27–49.
Beattie,D.,Berghage,R.,2004.Green roof media characteristics:the basics.In: Greening Rooftops for Sustainable Communities,Proc.North American Green Roof Conf.Portland,Toronto:The Cardinal Group.
Bengtsson,L.,Grahn,L.,Olsson,J.,2005.Hydrological function of a thin extensive green roof in Southern Sweden.Nordic Hydrol.36,259–268.
Berndtsson,J.C.,Emilsson,T.,Bengtsson,L.,2006.The in?uence of extensive vege-tated roofs on runoff water quality.Sci.Total Environ.355,48–63.
Booth,B.,2006.Green roofs are cool!Environ.Sci.Technol.,4046–4047. Brenneisen,S.,2003.?kologisches Ausgleichspotenzial von Extensiven Dachbegrü-nungen-Bedeutung des Ersatz-?kotops für den Arten-und Naturschutz und die Stadtentwicklungsplanung,Universit?t Basel.Dissertation.
BS EN933-2,1996.Tests for geometrical properties of aggregates.Part2:Determi-nation of particle size distribution-test sieves,nominal size of apertures.ISBN0 580257703.
BS EN1097-3,1998.Tests for mechanical and physical properties of aggregates.Part 3:Determination of loose bulk density and voids.ISBN0580300862.
C.J.Molineux et al./Ecological Engineering xxx(2009)xxx–xxx7
BS EN12457-3,2002.Characterisation of waste.Leaching.Part3:Compliance test for leaching of granular waste materials and sludges.Two stage batch test at a liquid to solid ratio of2l/kg and8l/kg for materials with a high solid content and with a particle size below4mm(without or with size reduction).ISBN0 580405818.
BS EN13055-1,2002.Lightweight aggregates.Part1:Lightweight aggregates for concrete,mortar and grout.ISBN:0580397769.
Emilsson,T.,Rolf,K.,https://www.doczj.com/doc/ac6390752.html,parison of establishment methods for extensive green roofs in southern Sweden.Urban For.Urban Green.3,103–111. Emilsson,T.,Czemiel Berndtsson,J.,Mattsson,J.E.,Rolf,K.,2007.Effect of using conventional and controlled release fertilizer on nutrient runoff from various vegetated roof systems.Ecol.Eng.29,260–271.
Emilsson,T.,2008.Vegetation development on extensive vegetated green roofs: in?uence of substrate composition,establishment method and species mix.Ecol.
Eng.33,265–277.
Fentiman,C.H.,Hallas,C.,2006.A?ne waste of a roof.Mats.World,24–26.
FLL,2002.Guidelines for the Planning,Execution and Upkeep of Green-Roof Sites (English version).Forschungsgesellschaft Landschaftsentwicklung Landschafts-bau e.V.,Troisdorf,Germany.
Fowler,J.,Cohen,L.,Jarvis,P.,1998.Practical Statistics for Field Biology,second ed.
John Wiley&Sons.
Gedge,D.,2000.Roof space—a place for brown?eld biodiversity?ECOS22,69–74. Gedge, D.,Kadas,G.,2005.Green roofs and biodiversity.Biologist52,161–169.
Grant,G.,Engleback,L.,Nicholson,B.,2003.Green Roofs:their existing status and potential for conserving biodiversity in urban areas.Eng.Nat.498,1–61. Grime,J.P.,Hodgson,J.G.,Hunt,R.,https://www.doczj.com/doc/ac6390752.html,parative plant ecology.In:A functional Approach to Common British Species.Unwin Hyman,London,UK.
Herpin,U.,Cerri,C.C.,Concei,M.,Carvalho,S.,Markert,B.,Enzweiler,J.,Friese, K.,Breulmann,G.,2002.Biogeochemical dynamics following land use change from forest to pasture in a humid tropical area(Rond?nia,Brazil):a multi-
element approach by means of XRF-spectroscopy.Sci.Total Environ.286,97–109.
Hills,C.D.,Sweeney,R.E.H.,Buenfeld,N.R.,1999.Micro-structural study of car-bonated cement-solidi?ed synthetic heavy metal waste.Waste Manage.19, 325–331.
Kadas,G.J.,2007.Can green roofs provide habitat for invertebrates in an urban environment?Unpub.Ph.D thesis,Royal Holloway Univ.London.
Nicholaus,D.,Van Woert,N.D.,Rowe,D.B.,Andressen,J.A.,Rugh,C.L.,Fernandes,R.T., Xiao,L.,2005.Green roof stormwater retention:effects of roof surface,slope and media depth.J.Environ.Qual.34,1036–1044.
Niachou,A.,Papakonstantinou,K.,Santamouris,M.,Tsangrassoulis,A.,Mihalakakou,
G.,2001.Analysis of the green roof thermal properties and investigation of its
energy performance.Energy Build.33,719–729.
Rowe,D.R.,Abdel-Magid,I.M.,1995.Handbook of Wastewater Reclamation and Reuse.CRC Press.
Saiz,S.,Kennedy,C.,Bass,B.,Pressnail,K.,https://www.doczj.com/doc/ac6390752.html,parative life cycle assessment of standard and green roofs.Environ.Sci.Technol.40,4312–4316. Snodgrass,E.,Snodgrass,L.,2006.Green Roof Plants,A Planting and Resource Guide.
Timber Press.
Steila,D.,Pond,T.E.,1989.The Geography of Soils:Formation,Distribution,and Management,second ed.Rowman&Little?eld Publishers.
Troeh,F.R.,Thompson,L.M.,2005.Soils and Soil Fertility,sixth ed.Blackwell Pub-lishing.
Villarreal,E.L.,Bengtsson,L.,2005.Response of a Sedum green-roof to individual rain events.Ecol.Eng.25,1–7.
WHO,1998.Guidelines for Drinking Water Quality,second ed.World Health Organ-isation,Geneva.
Wilson,S.D.,1999.Heterogeneity,diversity and scale in plant communities.In: Hutchings,M.J.,John,E.A.,Stewart,A.J.A.(Eds.),The Ecological Consequences of Environmental Heterogeneity,vol.4.BES.CUP,pp.53–65.