Energy and Buildings 40 (2008) 837848 www.elsevier.

com/locate/enbuild

A life-cycle energy analysis of building materials in the Negev desert

N. Huberman *, D. Pearlmutter
Ben-Gurion University of the Negev, Jacob Blaustein Institute for Desert Research, Albert Katz International School for Desert Studies, Sede Boqer Campus 84990, Israel Received 22 May 2007; accepted 19 June 2007

Abstract Environmental quality has become increasingly affected by the built environmentas ultimately, buildings are responsible for the bulk of energy consumption and resultant atmospheric emissions in many countries. In recognizing this trend, research into building energy-efciency has focused mainly on the energy required for a buildings ongoing use, while the energy embodied in its production is often overlooked. Such an approach has led in recent years to strategies which improve a buildings thermal performance, but which rely on high embodied-energy (EE) materials and products. Although assessment methods and databases have developed in recent years, the actual EE intensity for a given material may be highly dependent on local technologies and transportation distances. The objective of this study is to identify building materials which may optimize a buildings energy requirements over its entire life cycle, by analyzing both embodied and operational energy consumption in a climatically responsive building in the Negev desert region of southern Israelcomparing its actual material composition with a number of possible alternatives. It was found that the embodied energy of the building accounts for some 60% of the overall life-cycle energy consumption, which could be reduced signicantly by using alternative wall inll materials. The cumulative energy saved over a 50-year life cycle by this material substitution is on the order of 20%. While the studied wall systems (mass, insulation and nish materials) represent a signicant portion of the initial EE of the building, the concrete structure (columns, beams, oor and ceiling slabs) on average constitutes about 50% of the buildings pre-use phase energy. # 2007 Elsevier B.V. All rights reserved.
Keywords: Building materials; Energy-efciency; Life-cycle analysis; Embodied energy

1. Introduction World energy demand is projected to increase by up to 71% between 2003 and 2030 [1]. At present the vast majority of this energy consumption is based on fossil fuels, and despite notable advances in renewable energy technology, it is questionable whether such a demand trajectory can be met in an environmentally sustainable manner [2]. It has been proposed, then, that the only way to avoid a drastic reduction in accepted standards of living is to achieve an order-of-magnitude improvement in energy-efciency, dened as the ratio between energy services provided and energy consumed [3]. 1.1. Energy in Israel As in other industrialized countries, energy consumption and CO2 emissions in Israel have increased steadily over the
* Corresponding author. Tel.: +972 8 6596875; fax: +972 8 6596881. E-mail address: norah@bgu.ac.il (N. Huberman). 0378-7788/$ see front matter # 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.enbuild.2007.06.002

past decades. The country obtains nearly all of its energy from imported fossil fuels [4], though it is unique in mandating the use of solar energy for water heating in all new residential buildings. Since the 1970s Israels electrical power generation has been based primarily on coal [5] and the country also has sizeable deposits of oil shale [4]. Rapid population growth has resulted in overcrowding in the center of the country, causing a spill-over of construction to peripheral areas such as the Negev desert. The Negev comprises 65% of Israels land area, but accommodates less than 8% of its population. Construction in the Negev typically requires longer transportation distances from Israels commercial and industrial centers, increasing energy requirements for physical development. The harshness of the desert climate also affects energy consumption, due to the heavy heating and cooling loads in residential and commercial buildings. By and large, planning and design follow practices that are standard in the countrys more temperate regions, and particular adaptation to local conditions is the exception rather than the rule [6]. The distribution of Israels energy use among different sectors of the economy is representative of industrialized

838

N. Huberman, D. Pearlmutter / Energy and Buildings 40 (2008) 837848

countries, where buildings account for a large fraction of the overall consumption: in the U.S., the combined residential and commercial building sectors account for approximately 40% of the total [7]. These sectors, however, only include the energy consumed in buildings during the period of their active usage. The share of energy used by buildings increases signicantly when the energy used in their production is included as well. 1.2. Energy-efciency in the life cycle of buildings Any comprehensive assessment of architectural energy consumption must in fact consider the entire life cycle of the building, which can be divided into three phases: pre-use phase (embodied energy, EE), use phase (operational energy, OE) and post-use phase (demolition or possible recycling and reuse). The intensity of energy consumption in the rst of these phases for the production of buildings and their components has increased dramatically with industrialization. In contrast to traditional building practices based on locally available raw materials and human energy, modern methods have allowed vast quantities of fuel energy to be harnessed in the manufacture of standardized, quality-controlled building products. The hightemperature processes used to produce steel, aluminum, cement, glass and expanded foam insulation are prime examples. Industrial technologies have also led to sharp increases in operational energy consumption, most notably with the advent and proliferation of air-conditioning. Efforts in recent decades to moderate the use of non-renewable energy for heating and cooling have led to signicant savings through climatically responsive design approaches, including technological innovations for improving the thermal efciency of the building envelope [8,9]. At the same time, however, technologies yielding solutions such as super-insulated walls and windows have contributed to operational energy-efciency through the exploitation of high embodied-energy materials. Therefore, strategies which reduce a buildings energy needs for maintaining thermal comfort do not necessarily lower energy demand in the production phase, or in the overall life cycle. While reducing operational energy consumption has been a goal of designers for many years, embodied energy has received much less attention. There are several reasons for this, among them the lack of a clear assessment methodology and the data required to implement it, as well as a common assumption that the initial energy needed for production of a building is minor compared to its long-term operational needs. Some studies have indicated that this is indeed the case, citing gures in the range of 80% running energy to 20% embodied energy [10]. It is clear, however, that as operational energy use becomes lower, the role of embodied energy in minimizing overall consumption becomes increasingly prominent [1013]. In recent years the methodologies for embodied-energy assessment have improved, as have the reliability and availability of data. One recent report [9] indicated that the embodied energy in an ofce building may be as much as 67 times its annual operational energy, though most studies show more modest ratios. Depending on the expected lifetime of the building and its energy-efciency level for operation, the

embodied energy typically represents between 10% and 60% of the total energy used during the lifetime of the building [2,14,15]. The choice of a given building material can have multiple effects on a buildings energy consumption over the different phases of its life cycle, and as suggested previously, these effects can be contradictorysince properties such as high insulation value may yield relative savings in operational energy together with higher embodied-energy costs. The balance of these factors is especially signicant since a buildings external structure and envelope (roof, oor, walls and windows) tend to account for the greatest portion of its EE [16]. Often a range of different materials can be found to fulll the same function in a building, and since their energy-efciency may vary signicantly, savings can be achieved through substitution. In some cases these savings arise from the use of renewable energy in the production process, and in others from the reuse or recycling of existing products. Materials which incorporate industrial and consumer wastes (such as y-ash concrete, recycled plastic lumber, etc.) can reduce both the depletion of natural resources and the pollution generated by disposal. These environmentally friendly materials are becoming more widespread [9], though it is crucial that their benet be gauged within the larger life-cycle context. To obtain a comprehensive picture of a products whole-life environmental costs, a number of guidelines and draft standards have been developed in recent years. The process whereby the component and overall environmental ows in a system are quantied and evaluated is known as life-cycle assessment (LCA) [9]. It treats the life cycle of any product as a series of stagesfrom cradle (raw material extraction and harvesting), through manufacturing, packaging, transportation and use, to grave (disposal). While energy-related building regulations have begun to proliferate, life-cycle environmental assessments are still voluntary in almost all countries [17]. LCA studies generally consist of four phases, as set out in ISO Standard 14040 [18]: goal and scope denition, life-cycle inventory, impact assessment and interpretation. These four steps of the LCA methodology can be applied specically for life-cycle energy analysis (LCEA), which uses energy as the only measure of environmental impact. This does not replace the broader LCA environmental assessment method, but facilitates decision-making concerning energyefciency as an indicator of a buildings overall resource efciency [11,19]. 1.3. Previous studies Several recent studies have attempted to evaluate the environmental impacts of buildings in an integrated fashion over the entire life cycle. Adalberth [20] suggested an organized LCEA methodology, and showed an example of its application [21] in three prefabricated single-unit dwellings in Sweden. The LCEA methodology was applied in a variety of cases for evaluating energy ows in residential buildings, such

N. Huberman, D. Pearlmutter / Energy and Buildings 40 (2008) 837848

839

as the green house in Melbourne, Australia [11]. A detailed study [22] analyzed the life-cycle energy consumption of a residential building in Michigan, including energy costs, different impact categories and the inuence of future energy cost scenarios. In the context of LCEA, a number of studies have evaluated the potential for reducing the pre-use energy in buildings through material substitution. An Indian study presented a comparison of EE requirements for three residential buildings using different structural systems [23], and a Canadian research [24] examined the EE and greenhouse gas emissions associated with the on-site construction of a number of structural systems. A study by Pierquet et al. [25] analyzed both the thermal performance and embodied energy of wall systems in a cold climate in the U.S. Substantial net savings in the pre-use phase have been found for buildings constructed in India with local materials, due to reductions in transportation distances [26]. Two other Indian studies presented a comparison of energy embodied in building materials, one for both common and alternative materials and building systems [27] and the other for a wide range of wall elements [28]. Meanwhile, the accuracy of methods to assess EE data as part of LCA studies is still questioned. Pullen [29] compared diverse EE values from different origins and evaluated the inuence of the methodology used on the results obtained, concluding that LCEA practitioners should be aware of the imprecision of EE data. In response, Treloar et al. [30] presented methods for assessing and improving the reliability of EE calculations for making decisions in the context of LCEA. A broad overview of previous LCEA studies was recently compiled [31], in which the importance of ongoing thermal efciency in life-cycle energy use was re-emphasized. In addition, different LCA applications have been proposed to address a broader list of environmental impacts and improve accuracy, even analyzing re-use [32,33] and recycling methods in the context of LCA [34]. Tools for improved LCA in buildings have also begun to proliferate, as summarized by Erlandsson [35]. The validity of using single point indicators such as energy to assess overall environmental impact has also been scrutinized. Grant [36] posited that single indicators do not represent all environmental aspects, but are more tangible than complex models and therefore easier to understand. A report by Peuportier et al. [37] noted that complete LCA is more appropriate for evaluating environmental impacts than for prescribing particular materials, though it was subsequently concluded that LCA can also be useful for determining suitable technologies [38,39]. Decisions concerning the relative importance of different indicators also reects current environmental pressures, and LCA weighting has even been characterized as a political rather than a scientic act, subject to changing political agendas in the future [40]. The ongoing challenge of evaluating environmental impacts generated by buildings is addressed in the present study, which aims to ll several perceived voids. First is a lack of building LCA studies conducted in Israel, particularly for desert

buildings. While most of the literature on LCEA uses existing EE coefcients from different countries when local values are not found, this study introduces the peculiarities of the Negev desert of southern Israel into the calculation of local EE values, which are then compared with values from different sources. In order to evaluate the net savings of different building systems during the whole life cycle of a climatically responsive building in the Negev desert, the obtained EE values of building materials are then applied in diverse congurations to calculate the EE of a building as part of the complete LCEA study. 2. Methods 2.1. LCEAgoal and scope denition This study evaluates the inuence of different building material congurations on the energy-efciency of a desert building in Israel. For that purpose, the total energy budget of a particular building is assessed by applying the LCEA methodology. Following the methodological stages of the LCA detailed above, the heart of the LCEA process consists of a life cycle inventory, in which the essential quantication of energy ows is conducted, and an impact assessment for converting energy consumption into observable impacts such as CO2 emissions [11]. The LCEA consists of an analysis of the principal energy ows during the pre-use and use phases of the buildings life cycle, for the purpose of identifying material congurations with minimal cumulative energy consumption over these phases. (Energy consumption in the post-use phase, as well as recurring embodied energy in the use phase, are excluded from the analysis due to the high level of uncertainty and variability involved in their estimation.) Emphasis is placed on comparing traditional materials with more commonly used industrial materials, and gauging their relative inuence on life-cycle energy consumption. An additional aim of the study is to quantify the payback period of a given alternative, or to quantify the time required for an operational energy advantage (or disadvantage) to counterbalance an initial investment (or savings) in EE. 2.1.1. Studied system An existing building served as a base case for the analysis. The building is located at the Sede-Boqer campus of BenGurion University, in the arid Negev region of Israel at 30.88N latitude, at approximately 480 m above sea level (Fig. 1). The climate of the region is characterized by sharp daily and seasonal thermal uctuations, dry air and clear skies with intense solar radiation. Summers are hot and dry, with a mean daily maximum temperature of 32 8C, while nights are cool (daily minimum of 17 8C). Global radiation is intense, averaging 7.7 kWh/m2 per day during June and July. Winter days are typically sunny but cool, with a mean daily maximum temperature of 14.9 8C and a nightly minimum of 3.8 8C in January. Prevailing winds are northwesterly and consistently strong during the late afternoon and evening [6,41].

840

N. Huberman, D. Pearlmutter / Energy and Buildings 40 (2008) 837848

Fig. 2. The case study building, as shown in elevation from the south (emphasizing the selected half-block, with four one-storey apartments) and the rst oor plan.

Fig. 1. Location of the case study site (Sede-Boqer, at 30.88N latitude), in the Negev region of southern Israel. The locations of raw materials are shown relative to a 50-km radius of the city of Beer-Sheva, the assumed manufacturing site.

For the building selected as a base case, the functional unit is the service provided by four student apartments of 28 m2 each, over 50 years (the buildings assumed life span). 2.1.2. Building materials The actual building used as the base case for the analysis was built with reinforced concrete, cast in place for external walls as well as for oor and ceiling slabs. These concrete walls (and roof) are covered with extruded polystyrene (XPS), a rigid foam with closed cells that is produced in a continuous extrusion process and marketed locally as Rondopan. This external wall insulation layer has a thermal resistance value (R) of 1.82 km2/ W (approximately double that required by Israel Standard 1045), and is protected by a stone veneer approximately 5 cm thick as the exterior nish material. Exterior insulation on the roof provides a resistance of R = 3.0. For all of the alternative (hypothetical) envelope congurations, a reinforced concrete skeleton was taken as a constant including roof and oor slabs, as well as necessary vertical structural elements estimated to cover 33% of the original wall area. The remaining concrete (67% of wall area) was then substituted by different inll masonry block types (using a wall thickness of 20 cm for external walls and 10 cm for internal partitions). The total wall R-value (1.96 km2/W) and nish layers were kept constant (identical to the base case) in all congurations, by adjusting the insulation thickness depending on the insulating value of the particular mass material. Windows, oors and roof systems were also maintained constant for all congurations. Lightweight construction was not considered since, in addition to being uncommon for residential buildings in

The analysis intentionally focuses on a modern building, in which a conscious attempt was made in the design stages to minimize operational energy costs, thus amplifying the relative importance of its embodied-production energy. To evaluate the actual energy-savings and the potential for additional reductions in the cumulative energy consumption over the life span of the building, different envelope congurations were evaluated by hypothetically substituting particular building materials. The building selected is part of a student dormitory complex (Fig. 2), and was designed with a number of passive heating and cooling featuresincluding south facing windows to capture solar radiation in winter, cross ventilation for summer nocturnal cooling, double-glazing and insulated shutters, and massive insulated walls. The complex includes 24 individual apartment block buildings, and the building type used as a model for the case study consists of eight single-storey apartments arranged symmetrically over two storeys. All calculations were based on a half block, ensuring a realistic representation of internal and external walls. The four apartments in this half-block module comprise 112 m2 of oor area, with approximately 21 m2 of south-facing glazed openings and 14 m2 facing north. All openings are treated with operable insulated aluminum roller shutters.

N. Huberman, D. Pearlmutter / Energy and Buildings 40 (2008) 837848

841

the region, it has been found to be climatically inappropriate for the Negev [42]. When considering different masonry blocks to be compared, two principal groups of materials were considered. The rst group includes standard materials which are commonly available in the Israeli market and in general usage for residential construction. The rst of these is the standard hollow concrete block (HCB), with dimensions of 40 cm 20 cm 20 cm, with four cavities in two rows. These blocks are ordinarily manufactured using 1012% Portland cement. Another common inll material that was chosen for comparison is the autoclaved aerated concrete (AAC) block, marketed in Israel as Ytong or Ashkalit and which is lighter in weight (400500 kg/m3) and higher in thermal resistance than ordinary concrete blocks. The process of their manufacture is characterized by the high rates of EE necessary for manufacturing of lime and Portland cement, for the grinding of these materials and of sand, as well as for autoclaving. The second group of materials considered includes alternative masonry block types which are not currently manufactured and marketed on a large scale, but that could potentially fulll the same function as the standard block types. These materials have been found in laboratory testing to meet the requirements of the Israeli construction industry for compressive strength, durability and water absorption [43]. The following alternative materials were chosen for the analysis: stabilized soil blocks (SSB) and y-ash blocks (FAB). SSB: In many desert regions, adobe bricks can be produced on-site from local soil and dried in the sun, with a very low investment of non-renewable energy. However, the loess soils in the Negev have a high percentage of montmorillonite clay, which experiences large volumetric changes with varying moisture content that lead to cracking of the material. The durability of blocks made with this type of clay may be improved, and their compressive strength nearly doubled, with the addition of 4% lime and 2% cement to the mixture [44]. The strength of montmorillonite-based materials may further increased by using mechanical compaction [45]. While in Israel this material has only been used in individual initiatives, the industrialization of earth block manufacture in some locations, such as parts of the U.S. and Australia (where labor is relatively expensive), has made the material economically competitive [26]. FAB: Research carried out in the Negev has shown that durable masonry blocks can be produced entirely without cement which is highly energy-intensive in its production by replacing it with two types of locally produced y-ash (based respectively on low-calcium coal and oil shale). Both types of ash are industrial waste materials, produced as by-products in power-generation plants, and since a large portion (estimated at close to 1 million tonnes per year) of this waste is not currently recycled, it constitutes a potentially large-scale source of pollution [46]. Given that the two-types of y-ash are available waste materials, the products EE was assumed to include only transportation (in various stages of the process) and nal block manufacture. The composition of mixed y-ash blocks (FABm) used for the present analysis was 50% sand, 35% oil shale yash and 15% coal y-ash, with 20% mixing water.

The energy needed for production of FAB can be reduced even further more by substituting local soil for sand, which in fact must be transported from outside the Negev region. While this conguration may reduce transportation energy, the extent of the reduction is dependent on the location of the manufacturing plant in which the nal product is produced and its distance from the source of soil. This type of soil/y-ash block (FABs), based on a composition of 70% loess soil and 30% oil shale y-ash (with 15% mixing water), has also been tested for compressive strength and durability with promising results. 2.2. LCEAinventory methodology The LCEA inventory involves the actual quantication of energy inputs to the system in the different life-cycle phases. All energy values are expressed in primary energy terms, using the common unit of GJ (109 joules) in order to allow comparisons and additions between them. The particular methodology utilized to calculate energy ows in each phase is as follows: 2.2.1. Pre-use phase Energy ows in the pre-use phase were quantied so as to account for all direct energy inputs, whereas only a part of the indirect energy was included. The level of analysis was thus limited to IFIAS level II, which is intended to capture most (on the order of 90%) of the energy inputs to the system [47]. Perunit embodied-energy values were derived for individual building components (including major nish materials as well as bulk and insulation materials in the envelope) and then multiplied by their quantities within the building as designed. While ranges of various raw material EE values were obtained from published studies [27,28,4850], the embodied energy of major components was calculated for the local situation by combining the average of available data for raw materials (e.g. cement) with actual manufacturing processes (e.g. for the production of concrete) and transportation energy requirements according to resource locations within the region. In most cases EE values are based on the process analysis method, which considers hierarchically the actual processes responsible for producing the materialfrom the level of raw material extraction, to building materials and element production, to construction of the entire building [47,51]. In cases for which raw material data were unavailable (such as expanded polystyrene), the EE of the nal product was obtained directly from literature. For transportation energy, it was assumed that all nal product manufacturing took place in the city of Beer-Sheva, within a 50 km radius of which most of the resources are located (see Fig. 1). A common energy intensity factor of 1.57 MJ/(tonnes km) was adopted for all transportation of materials, based on typical fuel consumption and related energy costs of trucking [52]. Energy required for on-site construction of the building itself was estimated as a percentage of the overall material embodied energy. The gure of 8% of initial EE was adopted as an intermediate rate, based on different approaches found in the literature [11,32,53].

842

N. Huberman, D. Pearlmutter / Energy and Buildings 40 (2008) 837848

2.2.2. Use phase Energy required for the provision of thermal comfort was quantied on a yearly basis for heating and cooling seasons and assumed to be constant over the 50-year life span of the building. An active thermal simulation employing Quick II software (TEMMI, Ltd.) was performed to quantify the operational energy requirements of the building system for heating and cooling. Loads were calculated by establishing a seasonal comfort temperature set point (20 8C for winter days and 24 8C for summer days, with 50% humidity and inltration rates typical for local construction), and quantifying the thermal energy required to maintain this interior temperature. A ventilation rate of six air changes per hour (ACH) was introduced during summer night time hours. To quantify the yearly energy requirement, typical hot (July) and cold (January) daily cycles were simulated and the resulting daily energy requirements were multiplied by statistical factors representing the length of the respective cooling and heating seasons (120 days for summer and 100 for winter). Thermal loads were converted to delivered energy using the appropriate efciency factors (COP = 2.9 for electrical air-conditioning in summer, and 0.7 for gas-red heating in winter). Physical properties of the various building materials (density, conductivity and specic heat) were input to the QUICK II software based on values from the software database, supplemented by local values [54] when appropriate. It should be noted that numerous elements and factors were acknowledged and considered outside the scope of the analysis. These include the EE of detailed features such as furniture, appliances, infrastructure, and landscaping, as well as usephase energy for lighting, cooking, water heating, etc. As mentioned previously, the analysis did not include upstream indirect EE, recurring EE or post-use energy, and does not address actual economic costs or aesthetic and social image factorsany of which could be crucial for decision-making in an actual design process. Aside from energy consumption itself, global warming was the only impact category studied in this LCA, as expressed by
Table 1 Embodied-energy values of building materials in the Negev. Sources: [4446,59] Material Concrete Reinforced concrete Hollow concrete block Autoclaved aerated concrete block Stabilized soil block Fly-ash block (mixed) Fly-ash block (soil) Stone c Expanded polystyrene c Glass c Reinforcing steel c Aluminumc
a b c

energy-related carbon dioxide emissions. A direct translation of primary energy values to CO2 emissions was made by applying a conversion factor [47]. 2.3. LCEAimpact assessment Results obtained in the LCEA inventory from the different phases were summed over the assumed life span of the building for further evaluation. These primary energy values were then translated into quantities of CO2, a prime indicator of environmental impact due to its role in global warming [36,40,55]. Two different analyses were performed in order to evaluate the inventory results. The rst one was the calculation of the cumulative energy expenditure, measuring the energy life-cycle impacts of a given conguration by adding its overall embodied energy and its total operational energy over a 50-year life span (both values given in terms of equivalent primary energy), yielding energy consumption totals. The second one was the payback period which in a general sense represents a breakeven point between various congurations, after which a usephase advantage difference outweighs an opposite difference in the pre-use phase. Since the energy results are shown here as a total primary energy value without listing the details of fuels used, the conversion factor adopted for CO2 emissions quantication represents an average value of different energy sources and their related emissions. However, based on the published coefcients and the differences in the type of energy used in the analyzed phases it was decided to use different approximate values for each phase (100 kg of CO2 produced per GJ of EE, and 50 kg/GJ for OE). 3. Results 3.1. Pre-use phase: embodied energy In Table 1 the derived embodied-energy values are shown for the various building products considered in the analysis. These

EE (MJ kg1) 1.15 2.60 1.08 3.27 0.49 0.23 0.21 0.79 116 18 35 211

EE (MJ m3) 2,852 6,230 1,216 1,536 938 184 179 1,890 2,710 46,766 273,180 570,420

EE (MJ m2 element) 466548 10591246 243 338 138 38 36 95 54271 374 593698 6845

EE (MJ m2 oor area) 1486a, 1024b 3378a, 2328b 241b 335b 136b 38 b 36 b 140 376a, 216336b 96 1891a, 1303b 326

Base case. Alternative cases. Calculated as average of published values.

N. Huberman, D. Pearlmutter / Energy and Buildings 40 (2008) 837848

843

coefcients are listed in terms of energy per unit mass (MJ/kg) of the given material, as well as per unit volume (MJ/m3) to account for varying material density. Calculated material EE values are also shown per unit area of the given vertical or horizontal element to account for its thickness, and per unit oor area to account for actual material quantities in the case study building (in both its actual and alternative congurations). It can be seen that the volumetric EE coefcients of secondary materials like glass and aluminum are higher by an order of magnitude than those of mass materials like reinforced concrete. When the relative volume of these materials is taken into consideration, however, it is clear that the latter account for the bulk of the buildings total embodied energy (see also Fig. 4). The EE coefcients calculated for these mass materials are geographically specic to the Negev site in Israel, but were found to fall within the range of published values from various studies in other countries. This can be seen in Fig. 3, which shows the Negev data in relation to the maximum values found in the published range (from an Australian study [50]) and to the minimum values (from India [28]). Fig. 4 shows a breakdown of embodied energy by material, in the entire building as designed and with the hypothetical substitution of non-structural concrete by various masonry inll materials. In each of these alternatives the insulation thickness is adjusted to maintain a constant overall wall resistance (Rvalue), and the nish material is held constant. These substitutions result in substantial reductions in overall EE, as all of the different block types are at least 50% less energyintensive in their production as reinforced concrete. These reductions are in the range of 3040% for the entire building, and the lowest values are obtained with alternative (soil and y-ash) blocks. It is interesting to note that even though the congurations with reinforced concrete and hollow concrete blocks are both based on cement (with its high embodied energy per unit volume), the building constructed with concrete blocks consumes more than 25% less total initial energy than the base case (a reduction of over 150 GJ). This is due to the large quantity of cement in the solid cast-concrete relative to the

Fig. 4. Comparison of embodied energy by building conguration (not including on-site construction), with constant total wall R-value.

hollow block, as well as to the additional reinforcing steel, whose high EE coefcient of 275 GJ/m3 [48] makes it particularly energy-intensive. The alternative conguration incorporating AAC (Ytong) blocks reaches a slightly lower EE total than the building with ordinary concrete blocks, despite the relatively high EE coefcient of the AAC itself. This is because its higher thermal resistance allows an equivalent wall R-value to be achieved with thinner layer of EXP insulation, which is far more EEintensive per unit volume than either type of block. The EE results expressed in Fig. 4 are in total GJ for the entire building system of a given oor area. In order to evaluate the scale of these values with respect to data from other studies, however, they may be better expressed in EE per unit oor area, and in this case the values reported here range from 3.28 to 4.91 GJ/m2. The literature on initial EE of entire buildings has produced a wide array of results, depending on the methodology used, the country analyzed, the system boundaries, the construction technology, types of transportation, etc. but many of these identify ranges similar to the results found here. Some examples include 310 GJ/m2 [27], 412 GJ/m2 [12], and up to 11 GJ/m2 [56,57]. 3.2. Use phase: operational energy As can be seen in Fig. 5, the system with autoclaved aerated concrete (and considerably less supplemental insulation) had the highest OE requirements for both heating and cooling. As all the wall congurations in this step are equivalent in their overall thermal resistance, it is clear that the higher energy requirement for this conguration is due to the relatively low heat capacity of the lightweight AAC block, and its relative inability to store available energy (particularly absorbed solar energy in winter, as well as excess heat on summer days) and to reduce loads during the critical hours. Among the different congurations, the lowest OE consumption is seen for those congurations which combine a highly resistant external insulation layer with an internal mass material of high density (and high heat storage capacity)such as reinforced concrete (RC) and stabilized soil blocks (SSB).

Fig. 3. Comparison of EE coefcients of mass materials, as calculated for the Negev and as published in studies from Australia [50] and India [28].

844

N. Huberman, D. Pearlmutter / Energy and Buildings 40 (2008) 837848

Fig. 5. Comparison of annual operational energy by building conguration, with constant total wall R-value.

3.3. Life-cycle energy/impact In Fig. 6, the cumulative energy consumption over an assumed 50-year life span is shown for building congurations with different wall mass materials. Values at year zero thus represent the embodied energy of the given conguration (including on-site construction as well as material EE), and values at year 50 represent the total life-cycle energy requirement including both production and operational primary energy consumption. The most prominent difference seen between these congurations is the relatively high life-cycle energy consumption of the reinforced concrete base case building, which exceeds that of a concrete block building by over 150 GJ and of any other option by at least 200 GJand the source of this difference is the concretes excess energy in production. The

savings yielded by these options relative to the base case amount to between 27 and 33% in terms of embodied energy, and after 50 years of OE consumption up to 20% in terms of cumulative life-cycle energy. Thus the scale of energy required for producing the building and its materials, and the potential for reducing it through simple material substitution, are in this case signicant in life cycle terms. The consumption of the AAC conguration has a higher cumulative energy requirement than any of the other masonry block options. While its embodied energy is slightly lower than the conguration with regular concrete block, its higher operational needs become signicant over a 50-year period. As emphasized previously, this difference is a direct expression of the walls thermal mass deciency in a desert climate, since its total thermal resistance is equivalent in this case to all other options. At the same time, this lack of thermal mass does not increase the cumulative consumption of the AAC building (within its 50-year life span) to the level of the base case whose massive concrete walls are thermally advantageous, but are produced with a large initial investment of embodied energy. Another interesting result is the lifetime consumption of the stabilized soil block (SSB) conguration, whose enhanced thermal performance leads to the lowest cumulative energy total of any conguration. This further demonstrates the signicance, here in life cycle terms, of utilizing both internal mass and external insulation in desert buildings. An informative way of gauging the differences between congurations is to quantify the embodied-energy savings of a given option relative to the base case, in terms of the equivalent number of years worth of operational energy. Under the circumstance of the present analysis, the production energy saved by substituting alternative materials in place of poured concrete is equivalent to the buildings heating and cooling requirements over a period of 2330 years, depending on particular type of masonry block. Although the SSB congurations initial EE is higher than either of the options using yash, the SSB buildings lower operational requirements lead to a payback after less than 20 years, thus, constituting the preferable mass material within the analyzed congurations. Using the conversion factors detailed in Section 2.3 above, the life-cycle carbon emissions may be estimated for this best case (SSB) relative to the base case conguration. The base case was found to have a lifetime emission total of 76 tonnes of CO2, while the stabilized soil conguration totals 58 tonnes (Table 2). Therefore the substitution of stabilized soil block for the non-structural concrete in the actual building is responsible for an estimated reduction of 24% in total CO2 emissions during the life of the building. 4. Discussion

Fig. 6. Cumulative energy consumption by building conguration over a 50year life-cycle. Values at year zero represent pre-use embodied energy (including on-site construction)which for the base case building exceeds that of the concrete block alternative by an amount equivalent to 25 years of operational energy.

Underlying this analysis was the notion that in a modern, climatically responsive desert building constructed with industry-standard methods and materials, the energy required for the buildings production could be just as signicant as the energy required to maintain thermal comfort in it over its entire

N. Huberman, D. Pearlmutter / Energy and Buildings 40 (2008) 837848 Table 2 Comparison of energy and related CO2 emissions by building conguration EE Energy (GJ) Reinforced concrete Autoclaved aerated concrete block Hollow concrete block Fly-ash block (soil) Fly-ash block (mixed) Stabilized soil block 615 445 445 410 411 427 CO2 (tones) 61 44 45 41 41 43 OE Energy (GJ) 309 375 336 358 358 314 CO2 (tones) 15 19 17 18 18 16 Cumulative energy Energy (GJ) 923 820 782 768 769 741

845

CO2 (tonnes) 77 63 61 59 59 58

useful life. This was indeed found to be the case, with embodied energy accounting for some 60% of the buildings overall lifecycle consumptiona relation which is consistent with previous statements that production accounts for 4060% of the total energy use in low-energy houses [14]. It was further found that this embodied energy could be reduced signicantly through design decisions involving simple material substitutions in its wall construction. When the scale of these reductions is evaluated in life cycle terms, it is seen that they may be equivalent to decades worth of operational energy expenditures, and that they represent considerable proportional savings in total consumption even after the accumulation of such expenditures over a 50-year life span. The greatest reductions, as expected, are obtained from alternative materials using recycled waste and other local resources, which have inherent benets in terms of the energy intensity of their manufacture and transportation. What was less expected is the large scale of savings that can be obtained from standard materials i.e. conventional hollow concrete blocks when compared to a structure built with full poured-concrete walls. An overview of energy consumption by life-cycle phase for different mass material congurations is shown in Fig. 7, as normalized for the oor area of the building. This summary shows that the base case building has a cumulative lifetime energy consumption equaling just over 8 GJ/m2 of oor area, and that this total may be reduced to 6.57 GJ/m2 by substituting alternative wall materials.

Fig. 7. Energy consumption by mass material and life-cycle phase, per unit oor area (over a 50-year life-cycle).

Other studies that have analyzed the life-cycle energy use in residential buildings for a 50-year period have typically shown cumulative values of at least 15 GJ/m2 [11,14,57], with great differences in maximum values depending on the methodologies, system boundaries, geographical location, materials and building design. The relatively low values in this study may be attributed to a combination of low EE coefcients (relative to values used in the Australian studies cited above) and low operating requirements, due to the buildings thermally efcient design. Also the analysis does not include operational energy that is not connected to heating and cooling requirements (appliances, recurrent embodied energy, etc.), nor does it account for the post-use phasewhich could noticeably increase the nal values when comparing with other studies. One of the distinctions revealed by the analysis is the specic dependence of the results on local circumstances in the desert climate. For example, when lightweight AAC blocks (which are generally assumed to be energy-efcient) are used in a wall section whose overall thermal resistance is no higher than in the other options, the resulting cumulative consumption is higher than that of any other block walldue to the relative lack of thermal mass as a basic quality of the material. The importance of thermal mass was also demonstrated by the life-cycle energy advantage found for stabilized soil blocks, which have a higher heat capacity than those made from y-ash. It was found in a sensitivity analysis that the importance of this thermal mass would be further increased if the buildings detailed design allowed the potential for solar gain to be better realized, modifying the life-cycle energy consumption values of the congurations and the relationship between them. In this and other ways, the results of the analysis are in fact dependent on the specic design details of the building examined, and on the subjective criteria adopted. In this sense the study is not an optimization exercise per se, but rather a comparative analysis using a selected set of realistic alternatives. The results of the analysis also highlight the importance of several other methodological issues. Since LCA software tools are still of limited usefulness due to the extent of available databases, site-dependence, etc., the identication of meaningful relationships necessitated a detailed local analysis of material energy properties for each building phase studied. In the absence of EE coefcients for Israel, and particularly for the Negev, this required an in-depth process analysis (in this case was conducted up to IFIAS level II), accounting for energy ows from the stage of raw material extraction all the way to the

846

N. Huberman, D. Pearlmutter / Energy and Buildings 40 (2008) 837848

construction site. It is again stressed that there are numerous types of upstream indirect energy inputs which are beyond the scope of this analysis. Life-cycle energy results demonstrated the importance of evaluating EE values of basic building materials when used in realistic quantities in an entire building, rather than simply comparing their embodied-energy intensity per unit volume or weight. One illustration of this is the comparison of hollow concrete block and reinforced concrete (RC) as wall inll materials. Per unit wall volume, solid poured concrete has high concentration of cement and also contains reinforcing steel leading to a material EE coefcient which is 500% higher than hollow concrete block. When placed in the whole building, however, the total embodied energy of the RC conguration is only 37% higher than the building with HCB. Another example is the conguration with wall blocks based on y-ash: while the ash itself is considered a zero production-energy industrial byproduct (which if not reused represents a potential pollutant), it is part of a building whose overall EE is nevertheless considerable. It should also be emphasized that the actual life span of a building is dependent on the durability of its materials and construction. In the present study, the materials selected and the options considered were limited to those which could be assumed to have a useful life span of 50 years, without signicant energy expenditures for maintenance or renovation (recurrent EE). In addition, all the building systems include a reinforced concrete structure in order to meet the requirements of local standards (for seismic resistance, etc.). However, this assumption of a 50-year useful life span for all of the material congurations is based on limited data, and it is not possible to quantify the actual amounts of energy which would be required for their maintenance over this period. This is especially true for alternative materials such as stabilized soil brick, which has been shown to meet durability requirements in laboratory testing [44,45,58], but is not commonly found in the existing local building stock. While the studied wall systems (mass, insulation and nish materials) represent a signicant portion of the initial EE of the building, the concrete structure (columns, beams, oor and ceiling slabs) on average constitutes about 50% of the buildings pre-use phase energy. This proportion could diminish to some extent if recurrent EE were included in the analysis (given that walls generally need maintenance and the structure is assumed to last for the duration of the buildings life), but there can be no doubt that the energy-efciency of structural systems is an issue which has not been sufciently addressed in the LCA literature. It is also a crucial determinant which, if improved upon signicantly, could lead to vast reductions in building energy consumption and environmental impact (for instance by considering roof forms such as vaults and domes which, through their structural efciency, can make use of alternative materials whose strength is insufcient for at slabs). Alternative materials for insulation represent still another possibility for life-cycle energy reduction. Operational energy needs, at least for heating and cooling, are quite commonly calculated by applying existing tools and assuming currently realistic temperature set points, patterns of

usage, etc. However it is clear that the assumption of constant operational energy use over the buildings lifetime is contingent on many factors, and is subject to future changes in energy use which could result from prices, inhabitant behavior, regulations, environmental concerns, etc. The usefulness of life-cycle energy assessments could be enhanced through predictive models which extrapolate from current trends and account for these factors, though such predictions would inevitably be limited by the assumptions made. Certainly any attempt to translate life-cycle energy consumption into monetary costs would necessitate the use of varying scenarios depicting the future trajectory of energy prices. 5. Conclusions Stated briey, the following conclusions may be drawn from the analysis concerning energy-efcient material selection in desert buildings:  The embodied energy of common building materials in the Negev is close to the average of values found in recent literature from a range of different countries.  Use of the alternative building materials analyzed here can reduce the initial production energy required for a reinforced concrete building by 3040%, or the equivalent of 2530 years of operational energy. The energy saved cumulatively over a 50-year life cycle by this material substitution is on the order of 1520%.  Well-insulated stabilized soil blocks appear to be an energetically cost-effective option for desert buildings, with minimal long-term investment of non-renewable energy. Flyash blocks also provide signicant life-cycle energy benets, making productive re-use of an industrial waste product.  Hollow concrete blocks are only marginally more energyintensive than these alternative materials, due to the small amounts of cement used and sufcient thermal mass.  Without additional insulation, the deciency of thermal mass in AAC becomes more apparent and makes its life-cycle energy prospects less desirable. These ndings offer evidence that alternative materials using waste or local resources can be utilized in the creation of more sustainable desert architecture, minimizing energy demand while enhancing the quality of the built environment. Acknowledgements The authors are grateful to the Albert Katz International School for Desert Studies of the J. Blaustein Institute for Desert Research, Ben-Gurion University of the Negev, for making this study possible. References
[1] EIA, International Energy Outlook 2006, Energy Information Administration (EIA), #:DOE/EIA-0484 (2006), Washington, DC, 2006, pp. 15. [2] C.A. Langston, G.K.C. Ding, Sustainable Practices in the Built Environment, Butterworth Heinemann, 2001.

N. Huberman, D. Pearlmutter / Energy and Buildings 40 (2008) 837848 [3] S.H. Wade, Measuring Changes in Energy Efciency, Energy Information Administration (EIA), 2002, pp. 117. [4] EIA, Israel Country Analysis Brief, Energy Information Administration (EIA), USA, 2003, pp. 111. [5] H.A. Foner, T.L. Robl, Coal use and y ash disposal in Israel, Energeia 8 (5) (1997) 13. [6] Y. Etzion, D. Pearlmutter, E. Erell, I.A. Meir, Adaptive architecture: integrating low-energy technologies for climate control in the desert, Automation in Construction 6 (1997) 417425. [7] IEA, Annex 31, Energy-Related Environmental Impact of Buildings, International Energy Agency (IEA), 2001. [8] B. Der-Petrossian, Environment-friendly construction practices, United Nations Center for Human SettlementsUNCHS (Habitat), HS/596/00E, Vienna, 2000, pp. 143169. [9] C. Scheuer, G.A. Keolian, Evaluation of LEEDTM using life cycle assessment methods, in: National Institute of Standards and Technology (NIST), U.S. Department of Commerce, Michigan, 2002. [10] RAIA, Towards a National Framework for Energy EfciencyIssues and Challenges, The Royal Australian Institute of Architects (RAIA), Australia, 2004, pp. 136. [11] R. Fay, G. Treloar, Life-cycle energy analysisa measure of the environmental impact of buildings, Environment Design Guide GEN 22 (1998) 17. [12] R. Cole, P.C. Kernan, Life-cycle energy use in ofce buildings, Building and Environment 31 (4) (1996) 307317. [13] T. Mumma, Reducing the embodied energy of buildings, Home Energy Magazine 12 (1) (1995). [14] C. Thormark, A low energy building in a life cycleits embodied energy, energy need for operation and recycling potential, Building and Environment 37 (2002) 429435. [15] G. Milne, C. Reardon, Materials Use: 3.1-Embodied Energy, Commonwealth of Australia, 2004. Available from: http://www.greenhouse.gov.au/ yourhome/technical/fs31. [16] C. Atkinson, S. Hobbs, J. West, S. Edwards, Life cycle embodied energy and carbon dioxide emissions in buildings, Industry and Environment 2 (1996) 2931. [17] G. Treloar, P.E.D. Love, O. Faniran, B.D. Ilozor, Introduction, International Journal of Environmental Technology and Management 2 (13) (2002) 17. [18] ISO 14040, Environmental ManagementLife cycle AssessmentPrinciples and Framework, International Organization for Standardization, Geneva, Switzerland, 1997. [19] B. Der-Petrossian, E. Johansson, Construction and environmentimproving energy efciency, Building IssuesReports 10 (2) (2000) 121. [20] K. Adalberth, Energy use during the life cycle of buildings: a method, Building and Environment 32 (4) (1997) 317320. [21] K. Adalberth, Energy use during the life cycle of single-unit dwellings: examples, Building and Environment 32 (4) (1997) 321329. [22] S. Blanchard, P. Reppe, Life cycle analysis of a residential home in Michigan, in: School of Natural Resources and Environment, University of Michigan, Michigan, 1998, p. 60. [23] A. Debnath, S.V. Singh, Y.P. Singh, Comparative assessment of energy requirements for different types of residential buildings in India, Energy and Buildings 23 (1995) 141146. [24] R. Cole, Energy and greenhouse gas emissions associated with the construction of alternative structural systems, Building and Environment 34 (3) (1999) 335348. [25] P. Pierquet, J.L. Bowyer, P. Huelman, Thermal performance and embodied energy of cold climate wall systems, Forest Products Journal 48 (6) (1998) 5360. [26] J.C. Morel, A. Mesbah, M. Oggero, P. Walker, Building houses with local materials: means to drastically reduce the environmental impact of construction, Building and Environment 36 (2001) 11191126. [27] B.V. Venkatarama Reddy, K.S. Jagadish, Embodied energy of common and alternative building materials and technologies, Energy and Buildings 35 (2) (2003) 129137. [28] P.S. Chani, Najamuddin, S.K. Kaushik, Comparative analysis of embodied energy rates for walling elements in India, Architectural Engineering 84 (2003) 4750.

847

[29] S. Pullen, Estimating the embodied energy of timber building products, Journal of the Institute of Wood Science 15 (3) (2000). [30] G. Treloar, P.E.D. Love, O.O. Faniran, Improving the reliability of embodied energy methods for project life-cycle decision making, Logistic Information Management 14 (5) (2001) 303317. [31] I. Sartori, A.G. Hestnes, Energy use in the life cycle of conventional and low-energy buildings: a review article, Energy and Buildings 39 (3) (2007) 249257. [32] P. Crowther, Design for disassembly to recover embodied energy, in: Proceedings of the 16th International Conference on Passive and Low Energy Architecture (PLEA), Sustaining the Future: EnergyEcology Architecture, Brisbane, Australia, (1999), pp. 95100. [33] C. Thormark, Environmental analysis of a building with reused building materials, International Journal of Low Energy and Sustainable Buildings 1 (2000). [34] C. Thormark, The effect of material choice on the total energy need and recycling potential of a building, Building and Environment 41 (8) (2006) 10191026. [35] M. Erlandsson, M. Borg, Generic LCA methodology applicable for buildings, constructions and operation services today practice and development needs, Building and Environment 38 (2003) 919938. [36] T. Grant, The development and use of single point indicators, in: Proceedings of the Pathways to Eco Efciency, Second National Conference on Life Cycle Assessment, Melbourne, Australia, 2000. [37] B.L.P. Peuportier, N. Kohler, C. Boonstra, European Project REGENER Life Cycle Analysis of Buildings in: Proceedings of the Second International Conference on Buildings and the Environment, CSTB, Paris, France, vol. 1, 1997, pp. 3340. [38] B.L.P. Peuportier, Life cycle assessment applied to the comparative evaluation of single family houses in the French context, Energy and Buildings 33 (5) (2001) 443450. [39] C. Boyle, Sustainable buildings in New Zealand, in: Proceedings of the IPENZ Conference, 2004. [40] N. Svensson, L. Roth, M. Eklund, A. Martensson, Environmental relevance and use of energy indicators in environmental management and research, Journal of Cleaner Production 14 (2006) 134145. [41] A. Bitan, S. Rubin, Climatic Atlas of Israel for Physical Planning and Design, Israel Meteorological Service and Ministry of Energy and Infrastructure, 1991. [42] D. Pearlmutter, I.A. Meir, Assessing the climatic implications of lightweight housing in a peripheral arid region, Building and Environment 30 (3) (1995) 441451. [43] D. Pearlmutter, C. Freidin, N. Huberman, Alternative materials for desert buildings: a comparative life cycle energy analysis, Building Research & Information 35 (2) (2007) 144155. [44] G. Minke, Earth Construction Handbook, WIT Press, UK, 2000. [45] Y. Etzion, M. Saller, L. Conroy, The use of acrylic polymers and vibration for the production of stabilised soil bricks, Current Practices in Geotechnical Engineering 4 (1987). [46] C. Freidin, Cementless buildings units based on oil shale and coal y ash binder, Construction and Building Materials 13 (1999) 363369. [47] W.P.S. Dias, S.P. Pooliyada, Quality based energy contents and carbon coefcients for building materials: a systems approach, Energy 29 (4) (2004) 561580. [48] A. Alcorn, G. Baird, Embodied energy analysis of New Zealand building materialsmethods and results, in: Proceedings of the Embodied EnergyThe Current State of Play, Deakin University, Geelong, Australia, 1996, pp. 6171. [49] G. Baird, A. Alcorn, P. Haslam, The energy embodied in building materialsupdated New Zealand coefcients and their signicance, IPENZ Transactions Civil Engineering Section 24 (1/CE) (1997) 4654. [50] B. Lawson, LCA and embodied energy; some contentious issues, in: Proceedings of the Embodied EnergyThe Current State of Play, Deakin University, Geelong, Australia, 1996, pp. 7376. [51] R. Emmanuel, Estimating the environmental suitability of wall materials: preliminary results from Sri Lanka, Building and Environment 39 (2004) 12531261.

848

N. Huberman, D. Pearlmutter / Energy and Buildings 40 (2008) 837848 [56] G. Treloar, R. Fay, Building materials selection: greenhouse strategies, Environment Design Guide DES 35 (2000) 19. [57] R. Fay, G. Treloar, Factors inuencing the lifetime energy of Australian suburban housing, in: Proceedings of the 16th International Conference on Passive and Low Energy Architecture (PLEA), Sustaining the Future: EnergyEcologyArchitecture, Brisbane, Australia, (1999), pp. 257262. [58] B.V. Venkatarama Reddy, Sustainable building technologies, Current Science 87 (7) (2004) 899907. [59] L. Struble, J. Godfrey, How sustainable is concrete? in: Proceedings of the International Workshop on Sustainable Development and Concrete Technology, Beijing, 2004.

[52] G.D. Salomonsson, M.D. Ambrose, Product comparison methods, in: Proceedings of the Embodied Energy: The Current State of Play, Deakin University, Geelong, Australia, 1996, pp. 2331. [53] C. Scheuer, G.A. Keolian, P. Reppe, Life cycle energy and environmental performance of a new university building: modeling challenges and design implications, Energy and Buildings 35 (10) (2003) 10491064. [54] Israeli StandardSI 1045, Thermal Insulation of Buildings: Residential Buildings, The Standards Institution of Israel, Tel Aviv, 2003. [55] R. Fay, G. Treloar, U. Iyer-Raniga, Life-cycle analysis of buildings: a case study, Building Research and Information 28 (1) (2000) 3141.