Waste Management 32 (2012) 532541

Contents lists available at SciVerse ScienceDirect

Waste Management
journal homepage: www.elsevier.com/locate/wasman

Inuence of construction and demolition waste management on the environmental impact of buildings
Andr Coelho, Jorge de Brito
Department of Civil Engineering and Architecture, Instituto Superior Tcnico, Universidade Tcnica de Lisboa, Av. Rovisco Pais, 1049-001 Lisbon, Portugal

a r t i c l e

i n f o

a b s t r a c t
The purpose of this study is to quantify comparable environmental impacts within a Life Cycle Analysis (LCA) perspective, for buildings in which the rst (Materials) and last (End of Life) life cycle stages are adjusted to several waste/material management options. Unlike most LCAs, the approach is top-down rather than bottom-up, which usually involves large amounts of data and the use of specic software applications. This approach is considered appropriate for a limited but expedient LCA designed to compare the environmental impacts of different life cycle options. Present results, based on real buildings measurements and demolition contractor activities, show that shallow, supercial, selective demolition may not result in reduced environmental impacts. Calculations actually show an increase (generally less than 5%) in most impact categories for the Materials and End of Life stages because of extra transportation needs. However, core material separation in demolition operations and its recycling and/or reuse does bring environmental benets. A reduction of around 77% has been estimated in the climate change impact category, 57% in acidication potential and 81% in the summer smog impact (for the life cycle stages referred). 2011 Elsevier Ltd. All rights reserved.

Article history: Received 1 February 2011 Accepted 24 November 2011 Available online 17 December 2011 Keywords: Construction and demolition waste Life Cycle Analysis Waste management options

1. Introduction The work described here stems from a wider study which analyses selective and traditional demolition practices in technical and economic terms. It is also linked with construction and demolition waste (CDW) generation (Coelho and de Brito, 2010a) and distribution quantication in Portugal (Coelho and de Brito, 2010b). The work as a whole is an analysis of the technical, economic and environmental viability of a large-scale fully-developed CDW recycling plant in Portugal. It is a tentative comparative analysis of the environmental impacts of different options in CDW management at the end-of-life stage of a buildings life cycle. The purpose is to quickly quantify the environmental effects of applying deconstruction techniques and recycling options by incorporating these processes into the environmental impact of producing materials and transporting them to site (closing the cycle). The environmental impact of a buildings life cycle has been researched widely in recent years by various authors (Junnila, 2004b; Pinto, 2008; Balazs et al., 2001; Pascualino et al., 2008; Blengini, 2006, 2008; Xiaodong et al., 2010; Krogmann et al., 2009; Schuer et al., 2003). Building projects not intended for research purposes but which include LCA approaches in the design phase are very few, however (Hes, 1998, 2001a,b). One of the major
Corresponding author. Tel.: +351 218443659; fax: +351 218443071.
E-mail address: jbrito@civil.ist.utl.pt (J. de Brito). 0956-053X/$ - see front matter 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.wasman.2011.11.011

barriers to this is the reluctance to move design time lines to accommodate the extra time needed for an LCA, even though this might offer clear nancial and environmental benets (Hes, 2001a). Most of these studies have adopted a bottom-up approach using databases with embodied energy and emissions of several construction materials (Gonzlez and Navarro, 2005; Peuportier, 2001; Junnila, 2004b). These environmental impact analyses of buildings rely heavily on drawings, specications and/or data from the actual buildings. This approach is generally adequate when information on building quantities (bill of quantities), nal drawings and appropriate construction products environmental impact databases is available, bearing in mind the location of the buildings in relation to building materials suppliers and waste management operators. When these conditions are met, with high levels of transparency (clear denition of system boundaries, clear allocation methods, no mixing of data from different sources, etc.), along with coherent end-of-life and recycling considerations, present-day LCA software tools will provide reasonably accurate environmental impact results, reproducible to about 10% (Peuportier & Putzeys, 2005). Neglecting some or all of these requirements will naturally make LCA unreliable, or at least questionable. This is called the process-based LCA and it enables very specic analysis, although it requires detailed information, which makes it time consuming and costly. In fact, the longer the supply chain

A. Coelho, J. de Brito / Waste Management 32 (2012) 532541

533

the more time and money the analysis will require, particularly if local product environmental proles are not readily available (Schuer et al., 2003; Kofoworola and Gheewala, 2008). Other authors (Blengini et al., 2010), while praising the merits and advantages of the process-based LCA approach, acknowledge that it is often regarded as too complicated, too data and knowledge intensive, and rather time consuming. Moreover, the sensitivity analysis of current alternatives or future building improvements (e.g.: interventions on existing buildings or better building design) and denition of scenarios is particularly important, given the still signicant uncertainty which hinders LCA (Junnila, 2004a; Schuer et al., 2003). Not surprisingly, though, when it comes to quantifying the impacts generated at the end-of-life cycle stage and their consequence on global environmental impacts, there are divergences between process-based methods and this leads to different global impact estimations, especially in partial assessments (by life cycle phase e.g.: production, transport, end-of-life stage) (Lasvaux et al., 2009). So when it comes to LCA on buildings it is important to dene precisely what measures for end-of-life material management there are, and the implications of recycling and reuse for initial materials production. Another method of estimating life cycle environmental impacts has also been used in the last few years, especially for assessing entire industry sectors, and even for national economies (Carnegie Mellon University Green Design Institute, 2008). Its underlying theory was originally published by Leontief (1970) and later enhanced by Hendrickson et al. (1998), and it is generally called economic inputoutput life cycle assessment (EIO-LCA). Unlike the process-LCA this method tends to be applied to large-scale systems like a whole industrial sector/segment, although it can and has been applied to specic buildings (Kofoworola and Gheewala, 2008), in a hybrid approach an LCA-process with EIO-LCA. Although not applied in our study, it is referenced as a possible top-down approach which has been used on some occasions. The process-LCA bottom-up approach makes sense when it is possible to perform the life cycle inventory and environmental impact analysis using design blueprints, site data and database access for estimating the impacts of all the activities involved in the construction, operation, maintenance and end-of-life option of a building. This was not the case in this study, in which the environmental analysis was supplementary, not the main purpose. A simpler and quicker analysis was therefore necessary. The EIO-LCA was not an option, as its full application to a single building (even an averaged building, compiled from several casestudies) has no precedents and for this specic purpose has only been applied as a complement (Kofoworola and Gheewala, 2008). Furthermore there would be difculties in describing processes, linking monetary values with physical units and having to deal with the impacts of out of (local) economy imports (Carnegie Mellon University Green Design Institute, 2008). For these reasons an alternative method was sought.

2. Literature review Among the studies that focus on single buildings LCA are Schuer et al. (2003), Kofoworola and Gheewala (2008), Dewulf et al. (2009) and Asif et al. (2007). In the last work only the materials life cycle stage was analysed, and only from an embodied energy perspective, though others analyse several buildings, allowing comparison and consideration of their differences from an environmental impact point of view (Junnila, 2004a; Peuportier, 2001). Other studies have emphasised the performance of low energy buildings, focusing on life cycle energy (Thormark, 2001) or using large arrays of indicators (Blengini et al., 2010). These studies are highly detailed and site specic, which is an advantage when the

precise determination of environmental impacts is important. Generalization of these results is not advisable, however, although some general remarks are given (Blengini et al., 2010). In (Thormark, 2001), in particular, transportation calculations have been over simplied and may not reect regional transportation needs, while for recycling strategies it assumes that all materials taken from the building, when demolished in the future, are reinserted in the building construction products chain. This ignores the possibility of down-cycling, which is the fate of most of the recycled construction material mass. Some focus has also been directed at the end-of-life stage in construction projects, such as in Blengini and Garbarino (2010), which analyses the process of recycling demolished construction materials quite deeply especially concrete/ceramic aggregates and steel but deliberately excludes other building life-cycle stages. This study, although making use of a combined top-down/ bottom-up approach, in choosing a meaningful suite of indicators, is clearly bottom-up as far as the LCA calculation process is concerned. It is also dedicated to the aggregate industry processes, involving natural or recycled aggregates, in order to reach some conclusions on what a sustainable supply mix (of aggregates for the construction industry) should be. In this domain, studies like that of Weil et al. (2006) also specifically analyse the aggregate production industry, with particular reference to civil and structural engineering works. The authors apply an LCA to the concrete production industry in Germany, in which a certain amount of recycled aggregates, calculated from a mass-ow analysis, are integrated in the industrial process and compared to their present form. This has been done in view of stricter environmental threshold values for materials which will be in contact with the soil. Although relevant and useful, it is mainly an up-stream analysis in relation to the building construction cycle that does not take into account, for instance, the end-of-life operations and their possible variations. Embodied impacts have also been studied in some detail, from a life cycle perspective, by Chen et al. (2000). Their study focuses solely on embodied energy, and considers only one end-of-life scenario in which steel and aluminum are entirely recycled. This scenario, although linked to the fact that, in the study, steel and aluminum are the most energy intensive materials used in the buildings (two high rise housing blocks in Hong Kong), is unlikely to be recreated in practice, since a 100% replacement level is considered. Also, transportation energy requirements are determined in a very simplied manner, since distances, especially those out of the island, are calculated considering that all materials coming from a neighbor country come from its geometric centre. A less traditional approach to life cycle embodied impacts, following the same line of research, was performed by Dewulf et al. (2009). Here the concept of Exergy was used to quantify the potential to perform work, embodied in construction materials applied in a single dwelling. The Cumulative Exergy Consumption impact parameter is employed to quantify and compare three scenarios in end-of-life materials management, including a reference scenario where all materials are disposed of and two other scenarios combining reuse, recycling and incineration (combined heat and power) techniques.

3. Methodology This method resembles a process-based LCA, but one constructed from a top-down point of view. This approach was based on relevant published data, in particular (Junnila, 2004a; Blengini, 2006).

534

A. Coelho, J. de Brito / Waste Management 32 (2012) 532541

This option has naturally meant that only a scenario comparison has been possible, and not an absolute environmental impact value ranking. Calculations have therefore been made in the original environmental impact parameter units (kg CO2eq/m2, kg SO2eq/m2 and so on) but the resulting values are not used strictly to obtain environmental impact gures, in those same units, for buildings in other contexts. However, considering all data retrieved from real buildings and demolition operations, the purpose is to give a representative image of Portugal as a whole, as far as End-of-Life and Materials life cycle stages of existing buildings over 30 years old, chosen using the criteria established in (Coelho and de Brito, 2010a) are concerned. In Junnila (2004a) a detailed environmental LCA was performed for all life cycle stages in the buildings under analysis, although scenarios involving deconstruction were not included. Among the resulting conclusions were: Similar results are to be expected in housing and services buildings, from a life cycle environmental impact point of view. Different services buildings, designed by different teams, built by different contractors and occupied by different organizations, show a similar pattern for each life cycle stage proportion within the buildings global environmental impact. The end-of-life stage impact is generally below 3%, which is small compared with the global impact in all the analysed categories. But the differences found in the inclusion of recycled materials can lead to signicant changes in the summer smog and heavy metals categories (38% and 20% respectively). Furthermore, Junnilas work used case studies chosen in a rationale of replication to obtain similar results for buildings with signicantly different characteristics. This means that average values (here taken from Junnilas case studies) can be used to estimate the environmental impact distribution over the life cycle of a generalized housing or ofce building to compare scenarios which may inuence the environmental impacts of one or more life cycle stages. For this comparison, some LCAs use weighting methodologies to reach single score indexes (Blengini et al., 2010; Mithraratne and Vale, 2004), which may help decision-makers to read and easily interpret LCA results. However, the comparison presented in this study does not attribute different relative importance factors to the impact categories, which were chosen according to Junnila (2004a). The impact categories are: climate change (kg CO2eq/m2), acidication (kg SO2eq/m2); summer smog (kg C2H4eq/m2); nitrication (kg PO4eq/m2); and heavy metals (kg Pbeq/m2). They result from the emission of a wide array of substances, all of which are known to be important pollutants. Natural resources extraction and waste generation have not been considered explicitly, but it naturally follows that waste management practices involving the recycling and/or reuse of materials/components lead to less extraction of raw materials and less waste generation, since these practices limit new raw materials use. Importance factors were not used in relation to impact categories because of the subjectivity and controversy implied (Craighill & Powell, 1999).This aspect is aggravated by the many variables at stake which may give rise to the higher or lower relative importance of certain factors over others: the ecological sensitivity of certain regions, human concerns about that region, stakeholders interests/opinions, time constraints, economic reasons, among others. By not applying these factors this study attributes equal importance to all impact categories under consideration.

A snapshot of the global impacts calculated in (Junnila, 2004a) for each building is presented in Table 1. Averaging the life cycle impact proportions for the buildings referred to in Table 1 (also taken directly from Junnilas work) will provide the above generalized/reference buildings life cycle stage average impacts, as used here, as both percentage proportions and environmental impact units (Table 2). As an example, the climate change impact factor of 271 kg CO2eq/m2 for the generalized building Materials stage was obtained as the result of a 7.3% share of the 3700 kg CO2eq/m2 global average climate change impact presented in Table 1. All other Table 2 gures were obtained in the same way. The average life-cycle shares are calculated from (Junnila, 2004a), from the buildings A, B and C individual shares. Only the Materials and End-of-Life cycle stages were considered in the analysis, mainly to eliminate differences in impacts that may occur in the operational phase of buildings located in Finland and Portugal. These differences in impacts may appear for several reasons, such as climatic differences, building features and living habits. However, results including the complete life cycle (even though with basic data sourced from Finnish LCA case study) are also shown so as to give an idea of the possible impact of the proposed Materials and End-of-Life scenarios when all life cycle stages are accounted for (Table 12). Having characterized the generalized reference building in terms of environmental impacts, scenarios can now be dened. Five cases were considered, depending on end-of-life management activities: traditional demolition with landll or recycling and deconstruction, the material quantities landlled, and different options for incorporating recycled or reused materials in new construction. These scenarios are summarized in Table 3. Demolition/end-of-life environmental consequences are mainly conditioned by transportation, even though the impacts of direct demolition activities and materials transportation have both been considered in the specied scenarios. In fact, according to Blengini (2006), demolition environmental consequences only account for 7% of total demolition/end-of-life stage activities, the rest arises from transportation. Even though Blenginis gure applies only to energy consumption, it is here assumed valid for other impact categories since it has been demonstrated that around 80% of all emissions into the atmosphere derive from energy production (International Energy Agency, 2005). Different demolition procedures, from traditional demolition to total piece-by-piece deconstruction, will naturally lead to different energy consumption and waste generation patterns from the activities themselves, but given the relatively small importance of this sub-stage in life cycle environmental impacts, the differences were ignored. The demolition activity impacts for scenario 1 were therefore considered constant for all scenarios. Transportation impacts, with much higher relative importance, were specically considered for each scenario, depending on recycling and reuse percentages. Fifteen demolition cases were analysed to establish a relationship between the percentage of materials sent for recycling and transportation distance, with information supplied by a specialized contractor. This data is given by a sample of this contractors normal activity, which involves selective demolition, demolition, onsite separation and transport. Demolished buildings comprise houses, schools, ofce buildings and commercial sites. Manual labour is generally used for the selective demolition of accessible covering elements such as wood oors, doors, windows, cabinets and ceilings. The core dense structural materials, i.e. bricks, stone and concrete, plus various attached materials such as structural wood frames, xed sanitary

A. Coelho, J. de Brito / Waste Management 32 (2012) 532541 Table 1 Global environmental impacts obtained by (Junnila, 2004a), by impact category and by building analyzed. Impact category Climate change Acidication Summer smog Nitrication Heavy metals Unit kg kg kg kg kg CO2eq/m SO2eq/m2 C2H4eq/m2 PO4eq/m2 Pbeq/m2
2

535

Building A 4700 15.1 2.1 1.6 0.0021

Building B 3100 8.5 1.6 1 0.001

Building C 3300 9.8 2.3 1.3 0.001

Average 3700 11.1 2.00 1.30 0.0014

Notes: (1) All life cycle impact phases are included, from Materials (extraction, production and transport) to End-of-Life (demolition and waste management operations); (2) Life span considered: 50 years; (3) Main climate conditions: Finland.

Table 2 Environmental impacts, by category and life cycle stage, for the average reference building. Impact category Unit Life cycle stages reference case generalized building Materials Quantity Climate change Acidication Summer smog Nitrication Heavy metals kg kg kg kg kg CO2eq/m2 SO2eq/m2 C2H4eq/m2 PO4eq/m2 Pbeq/m2 271 1.26 0.51 0.13 0.00057 % 7.3 11.3 25.3 10.0 42.0 Demolition/end-of-life Quantity 37.0 0.37 0.05 0.06 0.00003 % 1.0 3.3 2.7 4.7 2.3 Total Quantity 308 1.63 0.56 0.19 0.0006 % 8.3 14.7 28.0 14.7 44.3

Note: Percentages of total life-cycle environmental impact, for each impact category.

Table 3 End-of-life and materials life cycle stages scenario. Scenario 1 Short description Complete demolition of building and transport of materials to landll Selective demolition of non-structural elements (soft stripping), followed by complete demolition of all other materials Notes/observations/assumptions 1. Environmental impacts are considered equal to those of the generalized reference building.a However, no extra transportation distance was added (due to recycling), since this scenario does not involve recycling 1. Elements removed during soft stripping: mechanically xed coverings, water supply and waste water drainage pipes, electrical cables, doors, windows. Destination: recycling 2. Elements destroyed during traditional demolition: masonry walls, structural support elements (concrete, steel and wood). Destination: landll. 3. Demolition (not transport) environmental impacts are considered equal to those incurred by the generalized reference building 4. Environmental impacts due to transportation of materials for recycling are calculated from scenario 1 basis (using the transportation distance enlargement factor) 5. Extra environmental impacts due to recycling operations are accounted for as a percentage of the original (scenario 1) impacts with the materials supply (as shown in Table 9) 6. Input data from housing buildings measured in (Coelho and de Brito 2010a) 1. Demolished structural elements are sent to recycling, whenever possible (depends on each real case situation) 2. Demolition (not transport) environmental impacts are considered equal to those incurred for the generalized reference building 3. Total recycled aggregate quantity included in new construction is limited to 10% (only general lls) of the total CDW generated from deconstruction and demolition of the building parts 4. Input data from services buildings measured in (Coelho and de Brito 2010a), along with 12 other real cases of demolition operations performed by a contractor 1. Hazardous materials are landlled (in a controlled way) 2. Recovered masonry and concrete are recycled and used to produce new masonry bricks and concrete 3. All other materials are recycled and used in new construction materials 4. The same input data as in scenario 2 is used 1. Hazardous materials are landlled (in a controlled way) 2. Recovery for reuse purposes do not cause signicant environmental impacts (extra impacts due to recovering activities are ignored), when compared to recycling impacts 3. Extra transportation needed for material reuse refers exclusively to these materials (not mixed with those being recycled), assuming an average distance equal to where the landll is located 4. Of all recovered materials aimed for reuse, only 50% is actually reusable (the rest is recycled) 5. The same input data as in scenario 2 is used

Deconstruction of non-structural elements, followed by complete demolition of all other materials

Full deconstruction, sending all materials to recycling

Full deconstruction, sending all materials which cannot be reused to recycling

a In the generalized reference building no allocation of emissions was assumed for the future products that are made from recycled building materials. This means that no reduction in the environmental impacts of the materials life cycle stage was considered. This derives directly from (Junnila, 2004a).

536

A. Coelho, J. de Brito / Waste Management 32 (2012) 532541

elements, water and drainage pipes and roof shingles are demolished in a more traditional fashion, using mechanical equipment (Coelho and de Brito, 2011). Transportation distances were calculated for each case and material managed, given its destination. These destinations are naturally selected by the contractor to be as close as possible to the worksite, for optimization and cost savings; for our contractor and for jobs in the Lisbon area, the distance will generally be less than 50 km). However, and especially when sending materials to be recycled, local operators are not always available (e.g.: recycling operators for wood, glass and some oil contaminated materials), which can imply transportation distances over 5 times more than the rest of the materials average transportation distance. A sample table is shown (Table 4) with the quantities, destinations and transported distance related to a certain demolition job. A total transportation distance was derived from this, to be compared with a reference (hypothetical) situation where all materials would be sent to landll. In the calculations, transported volumes are calculated from loose material densities, listed in Table 6. Transportation distance in the latter case depends heavily on the selected waste operator, and so its choice has been based on these criteria: 1. The waste operator must be located within 100 km of the site, or be the nearest waste operator otherwise. 2. The waste operator must be a generalist (i.e. accepting several types of CDW). 3. After points 1 and 2 are considered, the waste operator chosen will be the one to which, in reality, a greater distance was covered (measured by the sum of all distances listed to each operator which check criteria 1 and 2). Criterion item 1 is used to give preference to close range general operator locations, which are generally (but not always) within 100 km. Criterion item 2 excludes specialized operators (e.g.: glass or wood only recyclers), which cannot receive loads of co-mingled CDW materials. The nal criterion item selects the operator which in practice, for each job and after the selection implied by items 1 and 2, has been preferred by the contractor as far as total transport distance is concerned. Table 5 was built up by applying these criteria and the method used to compile Table 4 to all remaining real demolition cases. An average transportation magnication factor (Mt) could then be derived, along with an averaged percentage of recycled material that could be used in new construction. Mt is calculated by dividing the total transportation distance for each real example (with its recycling options) by the distance calculated for all materials being sent to landll. Plotting the relationship Recycling % vs. Mt results in Fig. 1, from which a quadratic function is tted, with a good approximation factor (R2 = 0.86). From this function it is now possible to estimate Mt, for each scenario (using each recycling ratio as indicated in Table 7), which will affect their environmental impacts considerably. Demolition/end-of-life environmental impacts are then quantied based on the generalized building impacts, shown in Table 2. Table 8 presents these values, for each impact category and for each scenario, using each Mt value as a proportion factor, derived from the equation presented in Fig. 1. As these scenario impacts all derive from the generalized building impacts described above, which implies a 27% recycling rate (and a complementary 73% landll rate), then obtaining transportation impacts for scenario 1 involves dividing the generalized building transportation only impacts by Mt (scenario 1 implies 0% of materials sent to recycling), while for the other scenarios it will mean multiplying by Mt.

Besides the direct impact of demolition activities and material transportation at the end-of-life stage, other environmental effects must be considered which result from the recycling processes themselves; these will depend on the amount sent for recycling, in each scenario. As for reuse, certainly some impacts are intrinsic to the activity (apart from transportation), e.g.: cutting wood, cleaning masonry bricks, storing material. This aspect has been ignored in this analysis since the effects are generally weaker than those inherent to recycling activities (Craighill & Powell, 1999; Gao et al., 2001). These references deal with energy consumption and the broad environmental impact of reusing scenarios. However, the results reported in Craighill & Powell (1999) consider transportation and comprise an average of several case studies, mostly related to mineral extraction and road maintenance. In spite of this, the reuse scenario in that study leads to an average reduction of almost 70% in environmental impacts compared with the scenario involving 50% landll and 50% recycling. In (International Energy Agency, 2005) reusing activities have been assumed to add no energy consumption, in a direct comparison with recycling activities. Data from Blengini (2008) was used to quantify the extra impacts caused by recycling. According to this study, certain net environmental benet reductions may be expected from recycling operations despite some additional impacts resulting from the processes used to recycle aggregate and steel products. An average of three environmental impact indicators was calculated for the extra impacts from recycling, resulting in 17% for aggregates and 39% for steel products. These extra impact percentages mean that, whatever the environmental impact reduction brought about by recycling, the nal impact of the considered life cycle stages including recycling will be 17% and 39% higher for aggregate and steel mass respectively. Knowing the quantities of materials used for the measured buildings/demolition examples in the listed scenarios (Table 3), particularly the aggregate and metals generated, these percentages are applied to give the overall extra environmental impact percentages presented in Table 9. The direct reduction of environmental impacts in the materials life cycle stage equals the percentage of recycled (plus reuse, if any) material in a certain scenario, as summarized in Table 7. In other words, a simple subtraction takes the rst into account by deducting the equivalent environmental impacts of the new materials initially planned for each scenario, but considering that some material sent for recycling may not be used again in building construction (factor u as described in the equations below). The resulting gure is then affected by the extra recycling operations impact percentages described above. For scenario 3, however, even though most of the materials are sent to landll, it was assumed that when aggregates can be recycled only 10% are used in new construction. This was to account for some general bulk lls, which are relatively common practice today. As for reuse (only in scenario 5), only part can actually be reused and the rest is sent for recycling. This part is considered to be 50%, based on published data (Guy, 2000, 2005; Southworth, 2009). Environmental impact values for the materials life cycle stage were therefore nally calculated using the following expressions:

Scenario 2 :

2 IC i

1 IC i

PC 2 uC2 =100 1 r 100

! AC 2 1 100

2 In which:IC environmental impact value, for category i, in scei C1 nario 2; Ii environmental impact value, for category i, in scenario 2 C2 1; P C r material percentage sent for recycling, in scenario 2; u weighted average, over waste weight, of possible recycled material

Table 4 Total transportation distance, with and without recycling options (one of 15 real examples). EWL Code Name Weight kg Destination % of total, in weight Waste operator Site distance, km Volume m3 Number of journeys Transportation distance, km Usable in new construction products, after recycling? Integration form

With recycling options 170107 170201 170202

Mixture of concrete Wood Glass

23020 97220 3680

Landll Recycling Recycling

10.74 45.38 1.72

Citri Ecociclo Vidrologic

48 210 234

16 324 11

1 17 1

48 3570 234

Yes Yes

170405

Iron and steel

10380

Recycling

4.84

Palmela recycling centre Quimitcnica

46

12

46

Yes

Plywood, Viroc, etc. Recovered glass, used in several construction products Iron products

A. Coelho, J. de Brito / Waste Management 32 (2012) 532541

170407 170603

170604 170903 200101 200301 200121

Mixed metals other insulation materials containing dangerous substances Glass wool CDW with dangerous substances Paper and cardboard Mixed municipal waste Fluorescent tubes and other mercury-containing waste Discarded electrical and electronic equipment Wood Name

14440 7940

Recycling Landll

6.74 3.71

37

16 611

1 32

46 1184

Yes

Metal products

8520 23440 3200 17520 130

Landll Landll Recycling Landll Recycling

3.98 10.94 1.49 8.18 0.06

Citri Quimitcnica A.S Simes Citri Ambicare

48 37 27 48 48

655 156 32 117 0

34 9 2 7 1

1632 333 54 336 48 Yes Yes Miscellaneous packages Recovered metal and glass, used in several construction products

200136 200138 EWL Code

920 3840 Weight kg

Recycling Recycling Destination

0.43 1.79 Waste operator

Ambitrena Ecociclo Site distance, km

48 210 Average loose waste density, kg/m3 404

6 13 Volume m3

1 1 Number of journeys 28

48 210 Transportation distance, km

No Yes Plywood, Viroc, etc.

Without recycling options 170903

Total transportation distance, with recycling options, km Total transportation distance, without recycling options, km

Construction and demolition wastes containing dangerous substances 7789

214.250

Landll

Citri

48

531

1344

1344

Notes: The number of journeys is calculated using standard 19.3 m3 sized trucks; Volumes are calculated, from weights, using loose material density values shown in Table 6.

537

538

A. Coelho, J. de Brito / Waste Management 32 (2012) 532541

Table 5 Transportation distances, transportation magnication factors and percentage of recycled material available for incorporation into new construction (15 real demolition examples). Case/ site Generated CDW, kg Materials destination, % Landll 1 50005 2 179200 3 214250 4 496992 5 1535710 6 2138652 7 3411760 8 4101440 9 4135354 10 4483375 11 4554410 12 4934126 13 7213303 14 18435606 15 61174652 Simple average Weighted average (on CDW generated kg) Standard deviation 7.0 16.1 37.5 73.5 4.8 98.0 99.6 98.6 99.6 99.2 98.5 98.1 10.9 97.3 99.9 69.25 92.1 Recycling 93.0 83.9 62.5 26.5 95.2 2.0 0.4 1.4 0.4 0.8 1.5 1.9 89.1 2.7 0.1 30.75 7.9 Total transportation distance, with recycling options, km Landll 153 552 3533 560 930 2337 6194 5854 8647 5434 5048 72083 33694 264144 70272 31962 2145929 Recycling 2041 5131 4256 1567 7200 147 720 1886 1148 495.6 2524 1766 31601 4774 1001 4417 3662 Total 2194 5683 7789 2127 8130 2484 6914 7740 9795 5930 7572 73849 65295 268918 71273 36379 88261 Total transportation distance, without recycling options, km Landll 408 1225 1344 735 1729 1886 5312 3861 3225 3977 4176 63455 9390 125712 67146 19572 58926 Average transportation magnication factor (Mt) 5.4 4.6 5.8 2.9 4.7 1.3 1.3 2.0 3.0 1.5 1.8 1.2 7.0 2.1 1.1 3.05 1.8 Percentage of recycled material available to be incorporated in new construction, % 100 100 98.8 100 100 100 100 100 100 100 100 100 31.6 100 100 95.4 95.8

35.09

Table 6 Loose densities for several CDW materials. Material Wood Gypsum board Plaster (gypsum) Brick Mixed CDW (all materials) Mixed CDW (only aggregate) Paper and cardboard Plastics Glass Metals Plant waste Municipal Solid Waste Loose waste density (kg/m3) 300 350 1000 1400 1400 1464 100 13 347 900 150 150 Source Construction Materials Recycler, Lisle, Ill. At: http://wasteage.com/mag/waste_news_briefs/

At: http://www.sustainability.vic.gov.au/resources/ documents/ Waste_Volume_to_Weight_Conversion_Table.pdf

Note: Loose densities correspond to average volume densities of non-compacted waste piles, for each material.

use in new construction; AC 2 recycling operations averaged extra environmental impact, in percentage. Eq. (1) applies to scenario 3, using a different percentage for recycling materials and so a different value for Mt. As for scenario 4, since all materials are considered to be incorporated into new construction, the uC factor equals 100% and can be eliminated, and so Eq. (1) becomes:
4 C1 IC 1 i Ii 4 PC r 100

4. Results and conclusions The results for each scenario are presented in Table 10. The totals for each scenario make it possible to determine percentile variations from scenario 1, representing a traditional approach of landlling all materials. Table 11 summarizes these variations, considering only the Materials and End of Life impact stages. For general perspective purposes only, Table 12 shows the percentage variations with regard to scenario 1, when all the life cycle stages are accounted for. Table 10 shows that the impacts share assigned to the Materials phase is always over 75%, for any analysed impact category, in scenarios 1 through 3. This is because, as replacement levels of input materials is low in these scenarios <9% the production and transportation of construction elements made of virgin materials still prevails. The impact share assigned to the Materials phase is always below 35%, for any analysed impact parameter, in scenarios 4 and 5. Actually, with selective demolition, not only soft-stripping, and

! 1

! AC 2 100

Finally, for scenario 5, which includes a certain amount of reused materials, the equation is simply updated as follows:
5 IC i 1 IC i 5 PC 5 pC r 1 r 100

! AC 2 1 100

5 In which pC r equals the reused material percentage, for scenario 5.

A. Coelho, J. de Brito / Waste Management 32 (2012) 532541

539

Average transportation magnification factor (Mt)

6 5 4 3 2 1 0

0.55 0.078 0.55 0.078 0.24 0.034 0.19 0.027 0.12 0.017 0.52 0.075 0.52 0.075 0.21 0.030 0.16 0.023 0.10 0.014 0.34 0.050 0.03 0.004
2

54.5

Scenario 5

Scenario 4

Scenario 3

Scenario 2

20

40

60

80

100

Recycling %
Total Fig. 1. Relationship between the recycling percentage and Mt. Scenario 1 0.020 1.05E-05 Nitrication Heavy metals 0.061 3.19E-05 0.004 2.26E-06 0.056 2.96E-05 0.016 8.20E-06 0.026 0.00001 0.034 1.80E-05 0.085 4.47E-05 0.085 4.472E05

Scenario 1 2 3 4 5 Note: Only scenario 5 includes reuse.

Recycling (+ reuse), % 0 1.9 7.9 96.3 96.3 Scenario 4 51.8

Scenario 3

sending more than 95% of the materials for recycling/reuse, significant impact reductions are possible at the Materials stage. Impacts resulting from this stage can even be lower than those coming from End-of-Life, assuming that these recycled materials are duly reused in the construction of new buildings. Table 10 also shows that the highest total impact reductions occur in the Heavy Metals (88%), Summer Smog (81%) and Climate Change (77%) categories, when shifting from scenario 1 up to 5. These are the impact categories where the highest relative contributions of the Materials stage happen. So, when the latter undergoes a sharp reduction due to inclusion of recycled/reused materials in new construction, it affects mostly the highest contributors (in a proportional manner). This global environmental impact comparison (based on a top down LCA methodology) of building demolition scenarios with different waste management approaches and recycled/reused materials incorporation considerations in the materials life cycle stage enables the following conclusions to be drawn: Soft-stripping, or the removal of non structural elements for recycling, followed by a traditional demolition of all other materials and their removal to a landll, will generally not imply any environmental impact reduction in the impact categories studied. It may even slightly aggravate the environmental impact (scenario 2 and some impact categories in scenario 3, although generally below 5%) because of longer transportation distances and because of some extra recycling activity impacts. This is largely to do with the means of transportation mostly used in the construction/demolition industry: by road, with diesel trucks. The use of (net energy saving) biofuels, running electric vehicles and/or greater use of rail transport could thus tip this environmental balance to the side of savings. Medium to low material reuse percentage will not translate into appreciable environmental impact reductions compared with the equivalent recycled quantities (scenario 5 compared to scenario 4). This is mainly because what really brings down environmental

Scenario 2

Transportation only

Scenario 1

Generalized building

Table 8 Demolition/end-of-life environmental impacts for all categories and scenarios.

Generalized building demolition only

Generalized building Total Landll 73%; Recycling 27%

2.62

34.4

9.5

15.9

20.9

51.8

Table 7 Recycling and reuse weight percentages for the scenarios under consideration.

Scenario 5

12.1

kg CO2eq/m2

Units

Climate change Acidication Summer smog

Impact category

kg SO2eq/m kg C2H4eq/ m2 kg PO4eq/m2 kg Pbeq/m2

0.37 0.053

37

0.030 1.60E-05

18.5

0.039 2.03E-05

y = -0,00054x 2 + 0,09360x + 1,45338 R = 0,86588

23.5

0.089 4.69E-05

54.4

0.089 4.69E-05

540

A. Coelho, J. de Brito / Waste Management 32 (2012) 532541 Table 11 Global environmental impact percentage differences between scenarios 2 through 5 and scenario 1 (Materials and Demolition/end-of-life stages). Impact category Scenario 2 Climate change Acidication Summer smog Nitrication Heavy metals 0.57 3.02 0.06 5.45 0.78 3 2.34 2.49 3.35 7.28 5.01 4 76.7 56.7 80.9 36.8 87.8 5 76.9 56.9 81.1 37.0 88.0

Table 9 Extra environmental impact percentages due to recycling, for the selected scenarios. Scenario Aggregates sent to recycling/total waste mass, % 0.66 4.8 93.7 53.9 Metals sent to recycling/total waste mass, % 0.10 2.05 0.64 0.63 Extra environmental impact weighted average, % 0.15 1.62 16.3 9.48

2 3 4 5

impact is replacing raw materials with recycled and/or reused elements, with very little being due to recycling or reuse. This is true even when considering that recycling entails more embodied impacts than reuse. The average reuse percentage calculated in the present study is 25%, which is considered low but technically attainable, considering the typical buildings analysed (in the Portuguese context). Greater impact reductions through reuse will require much higher percentages of materials to be reused, which could, in extremely favourable conditions, reach an average of 50% as a theoretical maximum reusable material weight. Other studies have considered reuse percentages of 17% (Gao et al., 2001) and as much as 100% (Craighill & Powell, 1999), resulting in a 11% lower energy consumption in the rst case and a 68% average environmental impact reduction in the latter (compared to an all-landll standard options, concerning End-of-Life and Materials life cycle stages). To reach the 50% level, however, for typical Portuguese buildings over 30 years old, special care must be taken with deconstruction. The careful extraction of wooden elements, masonry bricks and metal elements, for instance, cannot be done by standard construction or demolition crews; experienced deconstruction workers are needed who know exactly how to take the valuable pieces apart and handle them with reuse in mind. Given

Table 12 Global environmental impact percentage differences between scenarios 2 through 5 and scenario 1 (all life cycle stages). Impact category Scenario 2 Climate change Acidication Summer smog Nitrication Heavy metals 0.045 0.39 0.016 0.65 0.34 3 0.19 0.32 0.91 0.86 2.20 4 6.08 7.30 22.0 4.36 38.5 5 6.10 7.33 22.0 4.38 38.6

this reality it is currently better to direct policy efforts to increase recycling and reinsert materials into the construction industry cycle than to make the extra effort to encourage materials reuse, except obvious examples such as clean, straight wooden girders, or well-maintained sanitary equipment. It is particularly important to try and reintroduce recycled or reused materials into new construction, thereby helping to close the materials cycle. This will help to reduce the summer smog and heavy metals impact categories, since these account for

Table 10 Environmental impacts, by category and life cycle stage, for all scenarios. Impact category Life cycle stages scenario 1 Climate change Acidication Summer smog Nitrication Heavy metals Life cycle stages scenario 2 Climate change Acidication Summer smog Nitrication Heavy metals Life cycle stages scenario 3 Climate change Acidication Summer smog Nitrication Heavy metals Life cycle stages scenario 4 Climate change Acidication Summer smog Nitrication Heavy metals Life cycle stages scenario 5 Climate change Acidication Summer smog Nitrication Heavy metals Unit kg kg kg kg kg kg kg kg kg kg kg kg kg kg kg kg kg kg kg kg kg kg kg kg kg CO2eq/m2 SO2eq/m2 C2H4eq/m2 PO4eq/m Pbeq/m2 CO2eq/m2 SO2eq/m2 C2H4eq/m2 PO4eq/m Pbeq/m2 CO2eq/m2 SO2eq/m2 C2H4eq/m2 PO4eq/m Pbeq/m2 CO2eq/m2 SO2eq/m2 C2H4eq/m2 PO4eq/m Pbeq/m2 CO2eq/m2 SO2eq/m2 C2H4eq/m2 PO4eq/m Pbeq/m2 Materials 271 1.26 0.51 0.13 5.74E-04 267 1.24 0.50 0.13 5.64E-04 252 1.17 0.47 0.12 5.34E-04 11.6 0.054 0.022 0.006 2.45E-05 10.9 0.051 0.020 0.005 2.30E-05 Demolition/end-of-life 12.1 0.12 0.017 0.02 1.05E-05 18.5 0.19 0.027 0.030 1.60E-05 24.3 0.24 0.035 0.040 2.09E-05 54.4 0.55 0.078 0.089 4.69E-05 54.5 0.55 0.078 0.089 4.69E-05 Total 283 1.38 0.52 0.15 5.84E-04 285 1.43 0.52 0.16 5.80E-04 277 1.42 0.51 0.16 5.55E-04 66.0 0.60 0.10 0.09 7.14E-05 65.3 0.60 0.10 0.09 7.00E-05

Note: The environmental benets of recycling and reuse have been accounted for in the Materials phase, as prevented impacts.

A. Coelho, J. de Brito / Waste Management 32 (2012) 532541

541

considerable parts of the life cycle impact from the materials stage (25.3% and 42% respectively, for scenario 1), even though all other categories will also see their impacts reduced. Even though climatic differences, building features and living habits may differ between Portugal and Finland, which is where the basic data for this study was collected, the potential environmental impact reduction scenarios studied with global life cycle impacts are much lower than those observed in Table 11. Table 12 shows around 6% and 7% reductions in climate change and acidication related impacts for scenarios 4 and 5, but a gure of almost 40% is reached for the Heavy Metals category. From a complete life cycle perspective, and in order to get any obvious environmental impact reduction, especially in the climate change and acidication categories, recycling will have to be raised to above 90%, and efforts must be made to incorporate the resulting materials into new construction. Recycling percentages of this order of magnitude can be achieved and have actually been exceeded in recent real demolitions (Bjerregaard et al., 2007; Kohler, 2007a,b; Kohler & Davies, 2007). Acknowledgements Thanks are due to the FCT (Foundation for Science and Technology) for the postdoctoral research grant awarded to the rst author and to the ICIST IST research centre. Special thanks are due to Brito Cardoso and Vera Teixeira, from Ambisider, Recuperaes Ambientais, SA. References
Asif, M., Muneer, T., Kelley, R., 2007. Life cycle assessment: a case study of a dwelling home in Scotland. Building and Environment 42 (3), 13911394. Balazs, S., Antonini, E., Tarantitni, M., 2001. Application of Life Cycle Assessment (LCA) methodology for valorization of building demolition materials and products. Society of Photo-Optical Instrumentation Engineers (Eds.), SPIE the International Society for Optical Engineering 4193, 382390. Bjerregaard, M., Shaw, H., Kohler, N. 2007. Recycling demolition arising at Barts and the London Hospital. Waste & Resources Action Programme, Project Code WAS006-002: <http://www.wrap.org.uk/downloads/Barts_and_The_London_ Report_ full_case_study.c49324d6.4823.pdf> (August 23, 2010). Blengini, G.A., 2006. Life Cycle Assessment tools for sustainable development: case studies for the mining and construction industries in Italy and Portugal. Doctoral Thesis in Mining Engineering, Instituto Superior Tcnico, Technical University of Lisbon, Portugal. Blengini, G.A., 2008. Life cycle of buildings, demolition and recycling potential: a case study in Turin, Italy. Building and Environment 44 (2), 319330. Blengini, G.A., Di Carlo, Tiziana, 2010. The changing role of life cycle phases, subsystems and materials in the LCA of low energy buildings. Energy and Buildings 42 (6), 869880. Blengini, Andrea, Garbarino, Elena, 2010. Resources and waste management in Turin (Italy): the role of recycled aggregates in the sustainable supply mix. Journal of Clearer Production 18 (1011), 10211030. Carnegie Mellon University Green Design Institute (2008). Economic InputOutput Life Cycle Assessment (EIO-LCA), US 1997 Industry Benchmark model: <http:// www.eiolca.net> (September 2, 2010). Chen, T.Y., Burnett, J., Chau, C.K., 2000. Analysis of embodied energy use in the residential building of Hong Kong. Energy 4 (26), 323340. Coelho, A., de Brito, J., 2010a. Generation of CDW in Portugal. Waste Management & Research 29 (7), 739750. Coelho, A., de Brito, J., 2010b. Distribution of materials in Construction and Demolition Waste in Portugal. Waste Management & Research 29 (8), 843853. Coelho, A., de Brito, J., 2011. Economic analysis of conventional versus selective demolition a case study. Resources, Conservation & Recycling 55 (3), 382392. Craighill, A., Powell, J., 1999. A Lifecycle Assessment and Evaluation of Construction and Demolition Waste. Centre for Social and Economic Research on the Global Environment, CSERGE Working Paper WM 99-03, University of East Anglia, UK. Dewulf, J., Van der Vorst, G., Versele, N., Jansens, A., Van Langenhove, H., 2009. Quantication of the impact of the end-of-life scenario on the overall resource consumption for a dwelling house. Resources, Conservation and Recycling 53 (4), 231236. Gao, W., Ariyama, T., Ojima, T., Meier, A., 2001. Energy impacts of recycling disassembly material in residential buildings. Energy and Buildings 33 (6), 553 562. Gonzlez, M.J., Navarro, J.G., 2005. Assessment of the decrease of CO2 emissions in the construction eld through the selection of materials: practical case study of three houses of low environmental impact. Building and Environment 41 (7), 902909.

Guy, B., 2000. Building deconstruction: reuse and recycling of building materials. Alachua County solid wastes management innovative recycling project program, U.S.A: <http://www.lifecyclebuilding.org/les/Six%20House%20 Building%20Deconstruction.pdf> (February 26, 2010). Guy, B., 2005. The optimization of building deconstruction for department of defense facilities: Ft. McClellan deconstruction project. University of South Carolina, U.S.A.: <http://www.lifecyclebuilding.org/les/The %20Optimization %20of%20Building%20Deconstruction.pdf> (January 23, 2010). Hendrickson, C.T., Horvath, A., Lave, L., 1998. Economic inputoutput models for environmental Life-Cycle assessment. ACS Environmental Science and Technology 4, 184A191A. Hes, D., 1998. The Olympic stadium analysed using the DPWS. Centre for Design at RMIT: <http://buildlca.rmit.edu.au/menu9.html> (August 26, 2010). Hes, D., 2001. ABS Ofces, Belconnen ACT. Centre for Design at RMIT: <http:// buildlca.rmit.edu.au/menu9.html> (August 26, 2010). Hes, D., 2001. 60L Green Building. Centre for Design at RMIT: <http:// buildlca.rmit.edu.au/menu9.html> (August 26, 2010). International Energy Agency, 2005. IEA Annex 31. Energy related environmental impacts of buildings: <http://www.iisbe.org/annex31/index.html> (January 23, 2011). Junnila, S., 2004. Life Cycle Assessment of environmentally signicant aspects of an ofce building. Nordic Journal of Surveying and Real Estate Research 2 (Special Series), 8197. Junnila, S., 2004. The environmental impact of an ofce building throughout its life cycle, Doctoral dissertation, Research report 2, Helsinki University of Technology Construction Economics and Management, Espoo, Finland. Extended overview: <http://lib.tkk./Diss/2004/isbn9512272857/isbn951227 2857.pdf> (January 23, 2011). Kofoworola, O.F., Gheewala, S.H., 2008. Environmental Life Cycle Assessment of a commercial ofce building in Thailand. International Journal of Life Cycle Assessment 13 (6), 498511. Kohler, N., 2007. Recycling demolition arising at Cardiff city centre. Waste & Resources Action Programme, Project Code WAS006-002: <http:// www.wrap.org.uk/downloads/Cardiff_City_Centre_full_case_study.559cacaa. 4825.pdf> (August 23, 2010). Kohler, N., 2007. Recycling demolition arising at Gairbraid Avenue in Glasgow. Waste & Resources Action Programme, Project Code WAS006-002: <http:// www.wrap.org.uk/downloads/Glasgow_Housing_Association_full_case_study. 8a6a9713.4827.pdf> (August 23, 2010). Kohler, N., Davies, S., 2007. Recycling demolition arising at the Bryan Donkin site. Waste & Resources Action Programme, Project Code WAS006-002: <http://www.wrap.org.uk/downloads/Bryan_Donkin_Report_full_case_study. 78e94892.4824.pdf> (August 23, 2010). Krogmann, U., Minderman, N., Senick, J., Andrews, C., 2009. Life-cycle assessment of a green educational building: a case-study. Proceedings of the Life Cycle Assessment IX Toward the global life cycle economy, 29 September-2 October paper 166. Boston, USA. Lasvaux, S., Peuportier, B., Chevalier, J., 2009. Modeling the recycling processes in the LCA of buildings. Proceedings of the 1st International Exergy, Life Cycle Assessment and Sustainability Workshop and Symposium (ELCAS), Nisyros Island, Greece, chapter V, article 12. Leontief, W., 1970. Environmental repercussions and economic structure input output approach. Review of Economics and Statistics 52 (3), 262271. Mithraratne, Nalanie, Vale, Branda, 2004. Life cycle analysis model for New Zealand houses. Building and Environment 39 (4), 483492. Pascualino, J.C., Ortiz, O.O., Castells, F., 2008. Life Cycle Assessment as a tool for material selection and waste management within the building sector. In: Kenny, P., Brophy, V., Owen Lewis, J. (Eds.), PLEA Dublin 2008. Proceedings of the PLEA 200825th Conference on Passive and Low Energy Architecture, 2224 October, paper 556. Dublin, Ireland. Peuportier, B., 2001. Life Cycle Assessment applied to the comparative evaluation of single family houses in the French context. Energy and Buildings 33 (5), 443450. Peuportier, B., Putzeys, K., 2005. Inter-comparison and benchmarking of LCA-based environmental assessment and design tools. Practical Recommendations for Sustainable Construction (PRESCO), Work Package 2 Final Report, February 2005: <http://www.etn-presco.net/generalinfo/index.html> (August 30, 2010). Pinto, A., 2008. Life cycle assessment applied to the environmental and energy analysis of buildings. PhD Thesis in Mechanical Engineering, Instituto Superior Tcnico, Technical University of Lisbon, Portugal (in Portuguese). Schuer, C., Keoleian, G.A., Reppe, P., 2003. Life cycle energy and environmental performance of a new university building: modelling challenges and design implications. Energy and Buildings 35 (10), 10491064. Southworth, M.J., 2009. Building 802 Deconstruction project deconstruction results report. Alameda County Waste Management Authority, Oakland, U.S.A.: <http://www.stopwaste.org/docs/port-of-oakland-deconstruction-pilot.pdf> (February 19, 2010). Thormark, C., 2001. A low energy building in a life cycle its embodied energy, need for operation and recycling potential. Building and Environment 37 (4), 429 435. Weil, M., Jeske, U., Schebek, Liselotte., 2006. Closed-loop recycling of construction and demolition waste in Germany in view of stricter environmental threshold values. Waste Management & Research 24 (3), 197206. Xiaodong, L., Zhu, Y., Zhihui, Z., 2010. An LCA-based environmental impact assessment model for construction processes. Building and Environment 45 (3), 766775.