Life cycle assessment of a micromorph photovoltaic system

Energy 36 (2011) 4297e4306

Contents lists available at ScienceDirect

Energy journal homepage: www.elsevier.com/locate/energy

Life cycle assessment of a micromorph photovoltaic system Mirko Bravi a, Maria Laura Parisi b, Enzo Tiezzi a, b,1, Riccardo Basosi a, b, c, * a

Polo Universitario Colle di Val d’Elsa, Via Matteotti 15, 53100 Siena, Italy Department of Chemistry, University of Siena, Via A. Moro 2, 53100 Siena, Italy c Centre for the Study of Complex Systems, Via Roma 56, 53100 Siena, Italy b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 28 June 2010 Received in revised form 5 April 2011 Accepted 6 April 2011 Available online 5 May 2011

In this paper the results from a in-depth life cycle analysis of production and use of a novel grid-connected photovoltaic micromorph system are presented and compared to other thin film and traditional crystalline silicon photovoltaic technologies. Among the new thin film technologies, the micromorph tandem junction appears to be one of the most promising devices from the industrial point of view. The analysis was based on actual production data given to the authors directly from the PRAMAC Swiss Company and it is consistent with the recommendations provided by the ISO norms and updates. The gross energy requirement, green house gas emissions and energy pay-back time have been calculated for the electric energy output virtually generated by the studied system in a lifetime period of 20 years. A comparative framework is also provided, wherein results obtained for the case study are compared with data from literature previously obtained for the best commercially available competing photovoltaic technologies. Results clearly show a significant decrease in gross energy requirement, in green house gas emissions and also a shorter energy pay-back time for the micromorph technology. Ó 2011 Elsevier Ltd. All rights reserved.

Keywords: Photovoltaic Thin film Life cycle assessment

1. Introduction Since 2003, the production of Photovoltaic panels has experienced a rapid development in Europe (from 2006 to 2008 the sector has grown in EU approximately by 200%) due to specific energy policies and major programs of public support for photovoltaic technologies. As was shown by a recent European Photovoltaic Industry Association study [1], the development of this sector seems to move toward a greater promotion of applications such as BAPV and BIPV and the employment of technologies able to introduce significant and efficient innovations [2]. In this context, new TF technologies can play an important role as they show to have already reached a good maturity. TF technologies provide single junction devices that use less material while maintaining the efficiencies of devices based on mono- and multi-Si PV technologies. TF solar cells use a-Si, CIGS, CdTe or Poly-Si deposited on lowcost substrates. These technologies are successful because CdTe,

* Corresponding author. Department of Chemistry, University of Siena, Via A. Moro 2, 53100 Siena, Italy. Tel.: þ39 0577234240; fax: þ39 0577234239. E-mail addresses: [email protected] (M. Bravi), [email protected] (M.L. Parisi), [email protected] (R. Basosi). 1 Prof. Tiezzi passed away prematurely while this paper was under submission. Authors wish to underline that this work is dedicated to his memory and his pioneer work on environmental sustainability. 0360-5442/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.energy.2011.04.012

CIGS and a-Si absorb the solar spectrum much more efficiently than crystalline or multi-Si. Moreover TF solar cells use only 1e10 mm of active material and exhibit lifetimes comparable to crystalline Si based technologies. Although the predominant PV technology is still crystalline silicon (with a market share of 83% in 2008), TF technology has reached a 13% share of PV sector in the last year, with an increase by almost 120% over the previous year. The growth of the TF technology market share has been facilitated by the increasing shortage of silicon [3] as well as by the acquisition of “turnkey” production plants. Another favorable aspect of the TF technology is the complete integration of module production. Unlike the crystalline Si technology manufacturing process, which is characterized by the presence of several operations on different levels of the production chain, the TF technology production chain, can lead to the manufacture of the final module in a unique process. Short term estimates [4] agree that the industry of TF will show a market growth rate significantly and structurally higher than the traditional PV technologies. The mono- and poly-Si technologies, already established in terms of available performance and production costs, can get marginal improvements in the efficiency and cost savings related to scale and system economies. It would appear that the TF technologies have the greatest potential for a substantial reduction in prices as well as for efficiency improvement in the short-tomedium term [5].

4298

M. Bravi et al. / Energy 36 (2011) 4297e4306

Nomenclature a-Si BAPV BIPV BOS CIS CdTe CIGS EPBT GER GHG LCA MCPH Micro-Si Mono-Si Multi-Si PECVD Poly-Si PV TF

amorphous silicon building applied photovoltaic building integrated photovoltaic balance of system copper-indium-selenide, CuInSe cadmium telluride, CdTe/CdS copper-indium-gallium-selenide, CuIn(Ga)Se2 energy pay-back time gross energy requirement green house gas life cycle assessment micromorph micro-crystalline silicon (mc-Si) mono-crystalline silicon multi-crystalline silicon plasma enhanced chemical vapor deposition poly-crystalline silicon photovoltaic thin film

The main rationale for the TF technology is the utilization of cheap materials (glass, metal, plastic) with smaller amount of highpriced semiconductors. In order to obtain higher absorption of sunlight and to better exploit sunlight diffusion, it is also possible to create junctions in series stacking multiple layers of different semiconductor materials, thus obtaining optimal results for different wavelength ranges of the electromagnetic spectrum. TF PV technology saw the first a-Si commercial product in 1980. In the ‘90s, the proposal was to employ a thin micro-Si film as a promising absorber material for solar cells. Micro-Si is advantageous material for the production of solar cells since it is more stable compared to a-Si when exposed to sunlight. To overcome the conversion efficiency limits typical of single junction cells, one strategy that has proven to be very valuable in obtaining higher performances is that of using multi-junction solar cells (two or three p-i-n junctions

connected in series) of micro-Si and a-Si devices [6e8]. Currently, the multi-junction TF device that appears to be the most promising from the industrial point of view is the tandem junction called Micromorph. In particular, it combines an a-Si layer with a micro-Si one (a-Si/mc-Si). This structure produces devices that better exploit the solar spectrum and contributes to higher and more stable efficiencies. At the same time, it preserves the economic and technological advantages of simple a-Si [9e11]. Other TF PV technologies currently available as commercial products are the CIS, the CIGS and CdTe. In general these have better performance with lower costs when compared to a-Si devices. However they have the intrinsic disadvantage of involving rare, expensive and in some cases, potentially toxic materials [12e14].

2. The LCA study The analysis performed in this study consists of a thorough Life Cycle Assessment of the production and use of TF MCPH technology in accordance to the relevant recommendations by the International Organization for Standardization. The LCA methodology allows one to assess the potential environmental impacts associated with a product or a service during its whole life cycle. The four phases of LCA, defined by ISO 14040 and 14044 standards [15,16], are goal and scope definition, inventory analysis, impact assessment and interpretation. The goal and scope definition defines the purpose, extent and intended audience of the study. Moreover it contains a description of the system as well as geographical and time boundaries (cut-off criteria). The basis for the comparison of different systems is the functional unit of a product or service delivered. Inventory analysis consists of collection and analysis of data (such as consumed resources, electric and thermal energy use, air, water and soil emissions, by-products, etc.) associated with each process in the life cycle in order to build the life cycle inventory (LCI). When processes have more than one output, it is necessary to determine how to allocate the environmental burdens. Impact assessment evaluates the outcome of the inventory with respect to their environmental relevance. The purpose is to determine the relative importance of each of the inventory items and to

Fig. 1. Structure of a typical MCPH PV module and its light absorption range.

M. Bravi et al. / Energy 36 (2011) 4297e4306

4299

Fig. 2. Flow-chart of manufacturing process of PRAMAC MCPH module with main input flows per 1 kWp. Process Units are: GPF: Glass Preparation Front; GLM: Glass Loading and Marking Station; GCL: Glass Cleaning System; TCO1200: Transport Conductive Oxide Deposition System for Front and Back Contact; LSS1200: System for Laser Scribing; KAI1200: System with Plasma Box; LIA: Lain Automation; CTD: Contacted Tested Device; ELA: Encapsulation Lamination Assembly; QAC: Quality Assurance Component.

aggregate them to a small set of indicators, or even to a single indicator, in order to identify the processes which contribute most to the overall impact. Two mandatory elements are considered: classification (assignment of inventory data to different impact categories) and characterization (calculation of category indicator results for each impact category using characterization factors). A series of optional elements, such as normalization and weighting may be considered in order to find a significant final score, even if the weighting of environmental problems is difficult to estimate. Finally, an interpretation is necessary to evaluate the study for recommendations and conclusions. All calculations were performed with the SimaPro Software version 7.1 [17]. In order to compare results from this study with Table 1 Total lifetime electricity generation per kWp of MCPH system. L (yr)

I (kWh/m2/yr)

E

1 1700 8.74% 2 1700 8.65% 3 1700 8.57% 4 1700 8.48% 5 1700 8.40% 6 1700 8.31% 7 1700 8.23% 8 1700 8.15% 9 1700 8.06% 10 1700 7.98% 11 1700 7.90% 12 1700 7.83% 13 1700 7.75% 14 1700 7.67% 15 1700 7.59% 16 1700 7.52% 17 1700 7.44% 18 1700 7.37% 19 1700 7.29% 20 1700 7.22% Total lifetime electricity generation

PR

S (m2/kWp)

G (kWh)

75.00% 75.00% 75.00% 75.00% 75.00% 75.00% 75.00% 75.00% 75.00% 75.00% 75.00% 75.00% 75.00% 75.00% 75.00% 75.00% 75.00% 75.00% 75.00% 75.00%

11.44 11.44 11.44 11.44 11.44 11.44 11.44 11.44 11.44 11.44 11.44 11.44 11.44 11.44 11.44 11.44 11.44 11.44 11.44 11.44

1275 1262 1249 1237 1225 1212 1200 1188 1176 1165 1153 1141 1130 1119 1107 1096 1085 1075 1064 1053 23214

those published previously in literature, the specific indicators used in this study are: the gross energy requirement (GER) measured in MJ of primary energy per kWh of installed electric power; the green house gas (GHG), measured in g of CO2 equivalents per kWh of electricity generation; the energy pay-back time (EPBT), measured in years and computed as the ratio of the GER to the avoided primary energy requirement for the production of the same amount of electricity delivered by the system (assuming the average conversion efficiency of the chosen electric mix). Lastly, in order to carry out specific comparative analyses for the MCPH

Table 2 Inventory data of main input flows to the MCPH module and the BOS for 1 kWp. Module Electricity Compressed Dry Air Water Supply Solar Glass, low-iron Gas Supply (N2, SiH4, H2, PH3/H2, TMB/H2, CH4, CO2, NF3, Ar, DEZ, B2H6/H2) Wrapping BOS Material Steel Aluminum Copper Wrapping Paper Framework Stainless Steel (screws) Aluminum Rubber sheath Cables and contact boxes Copper Rubber Plastic Electricity

[kWh/kWp] [l/kWp] [m3/kWp] [kg/kWp] [kg/kWp]

369.00 80883.00 0.27 192.00 49.93

[kg/kWp]

1.20

[kg/kWp] [kg/kWp] [kg/kWp] [kg/kWp] [kg/kWp]

4.24 1.95 0.47 0.50 0.72

[kg/kWp] [kg/kWp] [kg/kWp]

0.48 18.00 0.80

[kg/kWp] [kg/kWp] [kg/kWp] [kWh/kWp]

2.38 0.72 0.53 58.20

4300

M. Bravi et al. / Energy 36 (2011) 4297e4306

system, the impact method Eco-indicator ’99 (EI99) was used [18]. This method was developed by an international group of LCA and environmental experts and it is a damage oriented impact assessment method, developed on the base of the Eco-Indicator ’95 methodology [19]. The aim was to express the total environmental burden of a product in a single score. The damage function represents the relation between the impact and the damage to human health or to the ecosystem as impact assessment end-points. 2.1. Description of the evaluated systems and major assumptions The main topic of the present study is the LCA analysis of production and use of a novel PRAMAC MCPH grid-connected system compared to other TF and the more traditional crystalline silicon PV technologies. The reference system studied is the MCPH 125 W photovoltaic module (PRAMAC MCPH 125 W) system produced by the Pramac

Swiss Ecopower plant in Riazzino, Switzerland. This PV module is manufactured using the TF technology in dual silicon layers. This process combines a top layer of a-Si over a micro-Si layer (a-Si/mcSi) (see technical details in ref. [20]). The essential structure of a typical MCPH module is illustrated in Fig. 1. The surface cell absorbs and converts the visible solar spectrum, whereas the lower area absorbs and converts the near infra-red region. These panels can exploit solar radiation corresponding to a broader range of the electromagnetic spectrum than other PV technologies. Laboratory tests show that power losses are low for a temperature increase in the modules. Moreover MCPH modules have good energy efficiency under low irradiance and scattered light conditions and, due to the frameless glass, they are more suitable for architectural integrations [21]. The MCPH module manufacturing phase was analyzed by building an LCI for all the inputs (energy and material requirements) in the production process of 1 kWp of PV panels (8 units).

Fig. 3. Tree representation of GER contribution for 1 kWp of the grid-connected MCPH PV system (100% ¼ 1.1  104 MJ LHV; cut-off value: 10,5%).

M. Bravi et al. / Energy 36 (2011) 4297e4306 Table 3 Aggregated dataa for analyzed LCA studies available in literature (from 2000 to 2009). PV Technology

g CO2-eq/kWh

Source

poly-Si (average) confidence interval Mono-Si (average) confidence interval TF (average) confidence interval Not declared (average) confidence interval

180.3 (Min 42.1; 98.9 (Min 62.5; 39.2 (Min 27.7; 28.8 (Min 21.2;

[27e30], [32], [33], [35], [37], [38], [43e45], [47], [50] [27e31], [34], [35], [38], [3], [43], [45], [47], [50] [33e35], [38], [43], [45e47], [50] [35], [38], [50]

poly-Si (average) confidence interval Mono-Si(average) confidence interval TF (average) confidence interval Not declared (average) confidence interval

EPBT 3.5 (Min 3.8 (Min 2.5 (Min 4.9 (Min

Max 318.5) Max 135.3) Max 50.7) Max 36.4)

2.7; Max 4.2) 2.9; Max 4.7) 2.0; Max 3.0)

[27e30], [32], [33], [35e38], [41e45], [47e50] [27e31], [34], [35], [37e39], [43], [45], [47] [33e38], [40], [43], [45], [47e50] [35], [38], [40], [50]

2.9; Max 6.9)

a

Data reported in Table 3 are strongly affected by: irradiation conditions, modules performance, outdoor temperature conditions, modules tilt, inverter performance or performance ratio in general.

Primary data were given to the authors directly from PRAMAC Swiss and are presented in an aggregated form for confidentiality reasons where necessary. The flow-chart of the manufacturing process of MCPH Pramac module is illustrated in Fig. 2. Secondary data were selected from the available life cycle inventory in the Ecoinvent v.2 database [22,23]. The electricity mix employed for the production process, and throughout the entire study, was assumed to be that of the European Union for the co-ordination of production and transmission of Electricity (UCTE), Medium Voltage, and was taken from the Ecoinvent database. The following assumptions were taken into account: i. since all indicators expressed per kWh of electricity produced are directly dependent on the assumed lifetime of the system (L), the latter was assumed to be equal to 20 years. This is a period in accordance to the warranty given by PRAMAC Swiss and makes possible comparisons with other PV LCA studies on similar systems; ii. the average Southern European yearly irradiation (I) 1700 kWh/m2/yr, was chosen as a common basis;

4301

iii. for all systems, a 25% efficiency loss was assumed with respect to the nominal values (i.e. performance ratio ¼ 75%), in order to correct for the losses caused by the electrical components (inverters, cables and control electronic devices), by atmospheric dust deposition, not-optimal orientation, temperature fluctuations and other indirect factors that are not taken into account by the PV module nominal power rating; iv. the inverter was considered to be replaced only once during the PV system lifetime v. for all types of modules, the energy pay-back time was computed assuming the average UCTE electricity generation efficiency of 32%; The analyses of the operational phase were performed including the so-called balance of system (BOS) for a typical grid-connected rooftop installation. In order to assess the environmental impacts of the reference system use phase, it was necessary to quantify the electric energy output that can be produced by the MCPH system (modules and BOS) during its life cycle. The total electricity generation (G) by the PRAMAC MCPH system in one year was computed taking into account the actual conversion efficiency of the MCPH 125 W modules equal to 8.74% (E) and an assumed performance ratio (PR) of 75%, as specified above. Assuming a PV panels area(S) of 11.44 m2 necessary to generate 1 kWp, the electrical energy delivered in the first year of the PV system lifetime is 1275 kWh/year. The total amount at the end of its life cycle should be 23,214 kWh (assuming a 1% decrement per year in the conversion efficiency MCPH module during its lifetime). The total lifetime electricity generation (G) per kWp was calculated as follows (see Table 1):

G ¼ E  I  PR  S The reference system used to quantify the BOS main input flows to the whole system was the first grid-connected PV system which employed PRAMAC MCPH modules completely integrated on a tilted roof (tilt angle: 22 ) of a building in Santa Fiora municipality, Grosseto Italy. An inventory of the MCPH module and BOS components per 1 kWp for this installation is presented in Table 2. In general, the decommissioning of PV systems at the end of their life cycle was not included in the study because of lack of currently reliable scientific data. However, for the MCPH module the decommissioning scenario qualitatively considered that the PRAMAC MCPH modules are made

600 513.0

g CO2 eq/kWh

500 400 300 200 100

98.9 63.0

39.2

0 poly-Si

mono-Si

TF

28.8 not declared

20.9 MCPH System (module + BOS)

UCTE

Fig. 4. Comparison of GHG emissions for 1 kWh of delivered energy with different PV technologies. Data in this figure are referred to those reported in Table 3: bars show the confidence interval of data and the average value for each PV technology analyzed.

4302

M. Bravi et al. / Energy 36 (2011) 4297e4306

8 7 6

EPBT

5

4.90

4

3.81

3.47 3

2.63

2.47 2 1 0 poly-Si

mono-Si

TF

not declared

MCPH System (module + BOS)

Fig. 5. EPBT values for PV modules production with different PV technologies. Data in this figure are referred to those reported in Table 3: bars show the confidence interval of data and the average value for each PV technology analyzed.

with small amounts of material with respect to quantities used for modules composed of silicon wafers. At the end of their life cycle, they could be directly disposed of in landfills or sent to the foundry, where the glass, which accounts for 92% by weight of the module (192 kg on a total mass of 208 kg per 1 kWp MCPH modules), might be recycled. The analysis focused on this aspect by calculating the impacts of recovery of the glass contained within 1 kWp of MCPH 125W modules. Other parts of the system (structure, electrical cables, inverters and control devices) are not included in the decommissioning scenario, assuming that they are recycled or safely disposed of in landfills. Currently the decommissioning of PV modules is not governed by a special law in Europe nor in the U.S [24]; however, laws in force in those countries provide that PV modules are submitted for appropriate laboratory testing [25] for a period of 24 h to simulate conditions existing in landfill and to assess toxic substances dispersion. Modules that have passed these tests are classified as non-hazardous waste and may follow the same path as urban waste [26]. To be properly forwarded to a landfill, modules that have passed the appropriate tests should not produce any significant dispersion of toxic substances in the environment. However, authors wish to point out the need for tests that more accurately reflect the actual conditions of the landfill in order to effectively reduce the risk of harmful substances from reaching high concentrations. This is especially important in the

case of a high number of modules to be disposed of. With the market growth of PV modules, this seems likely to happen, therefore regulation of module disposal is both necessary and urgent. 3. Results and discussion In Fig. 3, the results from a graphic analysis illustrates in absolute values the contribution to the GER indicator of a 1 kWp of the grid-connected MCPH PV system working for 20 years. In an effort to evaluate the impacts on the environment which are to be avoided when using this PV technology, those associated with the production of 1 kWh of electricity were calculated. The evaluation was made by comparing the Eco-indicator values of the 1kWp MCPH PV system (considered to be 23,214 kWh for the total lifetime electricity generation) to those of UCTE electricity production mix. The analysis showed that the Eco-Indicator of 1 kWh of electricity produced according to UCTE energy mix (2.33  102 Pt) is 13.5 times higher than the energy produced by photovoltaic modules using the MCPH technology (1.72  103 Pt). Assuming an average lifetime of 20 years and a total production of 23,214 kWh delivered in this period, 1 kWp of the MCPH PV reference system avoids an atmospheric emission of 11.5 tons of CO2 when compared to the same amount of energy produced by the

4.00E+04

MJ/kWp

3.00E+04

2.05E+04

2.00E+04 1.45E+04

1.59E+04 1.10E+04

1.03E+04

1.00E+04

0.00E+00 poly-Si

mono-Si

TF

not declared

MCPH System (module + BOS)

Fig. 6. GER values for 1 kWp power for different PV technologies. Data in this figure have been calculated from those reported in Table 3: bars show the confidence interval of data and the average value for each PV technology analyzed.

M. Bravi et al. / Energy 36 (2011) 4297e4306

4303

1.4E+01 1.2E+01

MJ/kWh

1.0E+01 8.0E+00 6.0E+00 4.0E+00 2.0E+00 0.0E+00 UCTE

MCPH System

CdTe

CIS

a-Si

poly-Si

mono-Si

Fig. 7. GER values for different PV technologies.

current UCTE energy mix (513 g CO2/kWh). In order to compare the results for the MCPH PV reference system with other PV systems LCA studies, an analysis of data in the literature and data selected from the available databases was performed. In literature there are numerous examples of LCA studies on PV systems that express the environmental impact in terms of quantity of GHG emitted into the atmosphere and in term of EPBT. Based on the results collected from a sample of LCA studies applied on various PV technologies from 2000 to 2009, a model was constructed in an effort to standardize the data, which are affected by a considerable variability (see Table 3), as mentioned above. Figs. 4e6 show the results obtained in terms of GHG, EPBT and GER indicators for a variety of PV technologies. The comparison was performed using results derived from the analysis of scientific literature and results of the PRAMAC MCPH system operational phase in the reference system conditions. In the same instances, results relative to the frameless module and the BOS are reported in order to evaluate the impact of the two components in the reference system. From the analysis of these charts, it was clear that some inconsistencies occur for different analyses. The high variability in results can be mainly attributed to a different set of assumptions in the analyzed studies that make their comparisons difficult. In particular, through the analysis of literature sources, it is possible to define a set of parameters (related both to the method adopted in the study and to the definition of the analyzed system) that substantially affect the results of an LCA performed on PV systems. The list of parameters includes only the main ones and can be considered in no way exhaustive. The charts show the TF technology trend to a limitation of GHG emissions and to lower EPBT values.

However, it must be pointed out that the lower efficiency of TF technology, compared to mono- and poly-Si technologies, implies a higher use of mechanical components to accommodate PV modules which partially reduce this gap [38,51]. In particular, Fig. 4 shows the CO2 emissions per kWh of delivered electricity for different PV technologies compared with the average CO2 emissions due to the energy production systems in Europe UCTE. Figs. 7 and 8 show the comparison among environmental impact values calculated for several PV systems: system PRAMAC MCPH, CdTe, CIS, multi-Si, mono-Si and a-Si systems. For the comparison data have been selected from the Ecoinvent database [23]. These data refer to the annual production generated by different PV technology installations in Switzerland for the year 2005. In this context, it was assumed that 1 kWp of installed power for the MCPH system in Switzerland is equivalent to an average production of 748 kWh per year. The analysis has been performed assuming a lifetime period of 20 years for all analyzed PV technologies. Moreover, the environmental impact values for the electricity production mix in Europe are included in the comparison. In order to increase the reliability of the results obtained from the comparison, a sensitivity analysis was performed to assess the impact of variation in solar irradiation level [52] on the GER and CO2-eq/kWh values. Tables 4 and 5 show the sensitivity analyses for all analyzed PV technologies. The results show that the MCPH system performs better even for variations in solar irradiation that cover the main geographical zone of Europe. The lowest emissions of CO2 equivalents are clearly achieved for the highest solar irradiation in Palermo (1.01Eþ01 g CO2-eq/kWh), as well as the lowest GER value (2.28E-01 MJ/kWh). In all cases, the sensitivity analysis

6.0E+02

g CO2 eq/kWh

5.0E+02 4.0E+02 3.0E+02 2.0E+02 1.0E+02 0.0E+00 UCTE

MCPH System

CdTe

CIS

a-Si

Fig. 8. GHG emissions values for different PV technologies.

poly-Si

mono-Si

4304

M. Bravi et al. / Energy 36 (2011) 4297e4306

Table 4 g Co2-eq/kWh sensitivity analysis for all analyzed PV systems.

Solar irradiation (kWh/m2/day) UCTE MCPH System CdTe CIS a-Si poly-Si mono-Si

Oslo

Dublin

Berlin

Zurich

Barcelona

Palermo

2.74 5.16E 1.83E 6.04E 6.28E 5.83E 5.88E 6.70E

2.52 5.16E þ 1.99E þ 6.58E þ 6.85E þ 6.35E þ 6.41E þ 7.30E þ

2.74 5.16E 1.83E 6.04E 6.29E 5.83E 5.89E 6.71E

3.01 5.16E þ 1.67E þ 5.49E þ 5.72E þ 5.30E þ 5.36E þ 6.10E þ

4.36 5.16E 1.15E 3.80E 3.95E 3.67E 3.70E 4.22E

4.96 5.16E þ 1.01E þ 3.34E þ 3.48E þ 3.22E þ 3.26E þ 3.71E þ

þ 02 þ 01 þ 01 þ 01 þ 01 þ 01 þ 01

02 01 01 01 01 01 01

þ 02 þ 01 þ 01 þ 01 þ 01 þ 01 þ 01

02 01 01 01 01 01 01

þ 02 þ 01 þ 01 þ 01 þ 01 þ 01 þ 01

02 01 01 01 01 01 01

Table 5 MJ/kWh sensitivity analysis for all analyzed PV systems.

Solar irradiation (kWh/m2/day) UCTE MCPH System CdTe CIS a-Si poly-Si mono-Si

Oslo

Dublin

Berlin

Zurich

Barcelona

Palermo

2.74 1.14E 4.12E 1.14E 1.17E 1.05E 1.22E 1.39E

2.52 1.14Eþ01 4.49E  01 1.25E þ 00 1.27E þ 00 1.15E þ 00 1.33E þ 00 1.52E þ 00

2.74 1.14Eþ01 4.13E  01 1.15E þ 00 1.17E þ 00 1.06E þ 00 1.23E þ 00 1.39E þ 00

3.01 1.14Eþ01 3.75E  01 1.04E þ 00 1.06E þ 00 9.60E  01 1.11E þ 00 1.27E þ 00

4.36 1.14Eþ01 2.59E  01 7.20E  01 7.35E  01 6.63E  01 7.71E  01 8.77E  01

4.96 1.14Eþ01 2.28E  01 6.33E  01 6.46E  01 5.83E  01 6.77E  01 7.70E  01

þ 01  01 þ 00 þ 00 þ 00 þ 00 þ 00

1,00E+02 0,00E+00 -1,00E+02

Pt

-2,00E+02 -3,00E+02 -4,00E+02 -5,00E+02 -6,00E+02 -7,00E+02 -8,00E+02 MCPH System

Production

Operation

Decommission

Total LCA

4,00E+01

-7,31E+02

1,65E+00

-6,90E+02

Fig. 9. Eco-indicator (Pt) environmental impacts associated to the production, operational and recycling phases of 1 kWp of PRAMAC MCPH modules.

confirms that the MCPH system shows a good potential for green house gas mitigation and a minor employment of primary energy per kWh of installed electric power. To analyze the disposal phase for the reference system of this study, it was assumed that the decommissioned modules are transported from the installation site to a recycling centre placed 1000 km far from it. In this disposal scenario, the energy estimated for separating the glass from the MCPH modules, once it reaches the recycling centre, is similar to that used in virgin glass manufacturing. This estimate is conservative, because the databases assume that recycling has a lesser impact than virgin glass manufacturing. As shown in Fig. 9, the Eco-indicator value associated with the decommissioning phase of modules is 1.65 Pt and is due to the transportation phase, i.e. from the installation site to the recycling site. Fig. 9 clearly shows that the use of house PV systems has a positive effect on the environment. This benefit is realized mostly from the operational phase of the PV system since it avoids the effects caused by the UCTE electricity generation system. The negative sign of the Eco-indicator indicates the avoided impacts.

The analysis and the comparison with other PV technologies in this study, clearly show the advantages of TF modules. In particular, those produced with the PRAMAC MCPH technology provide a reduction in green house emissions, lower EPBT and a significant decrease in gross energy requirement. Moreover, the analysis performed on the Pramac MCPH modules allows for a very high level of details compared to the analyses of TF modules previously published in literature. For example, it is noteworthy that our assessment takes into account the polystyrene used for packaging during the production phase. It should be emphasized that this level of detail and the effort to have transparency establishes a benchmark for comparison with all other PV technologies analyzed with a minor level of detail. 4. Conclusions The in-depth LCA analysis developed in this study enables an appreciation of the performance, the real potentials and the market opportunities of the MCPH modules in the growing PV market scenario. The performance of the TF technologies discussed here,

M. Bravi et al. / Energy 36 (2011) 4297e4306

compared to those of traditional crystalline silicon manufacture, are represented by the assessment of several advantages with respect to competitors: i. a shorter energy pay-back time thanks to the employment of smaller amounts of materials produced by energy intensive processes; ii. the decommissioning stage is more favorable because these modules are composed mostly by glass; iii. the lower level of CO2 equivalent emissions per kWh. In the short term, for the industrial initiatives oriented to establish the TF MCPH technology, it will be necessary to adopt effective strategies for penetrating the PV market and to make clear the efficiency and technological performances of the product. In the scenario of an increasing BAPV and BIPV market, all these elements can certainly represent advantages in terms of competitiveness with respect to crystalline and single junction amorphous silicon modules. Acknowledgments Authors wish to aknoweldge the PRAMAC Swiss Company for providing data for the MCPH photovoltaic panel LCA. In particular, Sergio Doneda (PRAMAC Swiss Technology Manager) and Luigi Gozzi (Director PRAMAC LAB) provided timely technical and organizational support, thus effectively improving the quality of this work. M.B. thanks the Province of Siena for the support of the Program “Ricercatori in Azienda” 2008e09 in the framework of which this study has been carried out in collaboration among the Department of Chemistry of the University of Siena and the PRAMAC Swiss Company. Careful reading and revising of the manuscript by Professor Emeritus Les Brooks, Sonoma State University, is gratefully acknowledged. References [1] Epia. Global market outlook for photovoltaics until 2013. See also, http:// www.epia.org; 2009. [2] Bagnall DM, Boreland M. Photovoltaic technologies. Energy Policy 2008;36: 4390e6. [3] Van Sark WGJHM, Brandsen GW, Fleuster M, Hekkert MP. Analysis of the silicon market: will thin films profit? Energy Policy 2007;35:3121e5. [4] Glen A. Thin film and organic PV: new applications for solar energy. NanoMarkets. Report No.: Nano-021. See also, http://www.nanomarkets.net; 2006 May. [5] Hegedus S. Thin film solar modules: the low cost, high throughput and versatile alternative to Si wafers. Progress in Photovoltaics: Research and Applications 2006;14:393e411. [6] Yang J, Banerjee A, Glatfelter T, Hoffman K, Xu X, Guha S. Progress in Triplejunction amorphous silicon-based alloy solar cells and modules using hydrogen dilution. In: Proceedings of first World conference on photovoltaic energy conversion. Hawaii; 1994 Dec 5e9. p. 380e5. [7] Yoshimi M, Sasaki T, Sawada T, Suezaki T, Meguro T, Matsuda T et al. High efficiency thin film silicon hybrid solar cell module on 1 M2-class large area substrate. In: Proceedings of third World conference on photovoltaic energy conversion. Osaka; 2003 May 11e18. p. 1566e9. [8] Green M, Emery K, King D, Igari S, Warta W. Solar cell efficiency tables (version 30). Progress in Photovoltaics: Research and Applications 2007; 15:425e30. [9] Meier J, Dubail S, Flückiger R, Fischer D, Keppner H, Shah A. Intrinsic microcrystalline silicon (mc-Si:H) e a promising new thin film solar cell material. In: Proceedings of first World conference on photovoltaic energy conversion. Hawaii; 1994 Dec 5e9. p. 409e412. [10] Shah AV, Meier J, Feitknecht L, Vallat-Sauvain E, Bailat J, Graf U et al. Micromorph (Microcrystalline/amorphous silicon) tandem solar cells: status report and future perspectives. In: Proceedings of the 17th EU PVSEC. Munich; 2001 Oct 22e26. p. 2823e30. [11] Shah A, Meier J, Vallat-Sauvain E, Droz C, Kroll U, Wyrsch N, et al. Microcrystalline silicon and micromorph tandem solar cells. Thin Solid Films 2002; 403e404:179e87. [12] Sinha P, Kriegner CJ, Schew WA, Kaczmar SW, Traister M, Wilson DJ. Regulatory policy governing cadmium-telluride photovoltaics: a case study

[13]

[14]

[15]

[16]

[17] [18]

[19]

[20] [21]

[22]

[23]

[24] [25]

[26]

[27]

[28]

[29]

[30]

[31]

[32]

[33]

[34] [35]

[36]

[37]

[38]

4305

contrasting life cycle management with the precautionary principle. Energy Policy 2008;36:38e47. Fthenakis VM, Fuhrmann M, Heiser J, Lanzirotti A, Fitts J, Wang W. Emissions and encapsulation of cadmium in CdTe PV modules during fires. Progress in Photovoltaics: Research and Applications 2005;13:713e23. Bravi M, Parisi ML, Tiezzi E, Basosi R. Life cycle assessment of advanced technologies for photovoltaic panel production. International Journal of Heat & Technology 2010;28:133e40. ISO (International Organization for Standardization) 14040 standard. ‘Environmental management-life cycle assessment-principles and framework’; 2006. ISO (International Organization for Standardization) 14044 standard. ‘Environmental management-life cycle assessment-requirements and guidelines’; 2006. Prè Consultants. SimaPro 7.1. Available from, www.pre.nl/; 2008. Amersoort, The Netherlands. Goedkoop M, Spriesma R. The eco-indicator 99. A damage oriented method of life cycle impact assessment. Methodology report. Available from. 3rd ed. Pré Consultants, www.pre-nl/download/EI99_methodology_v3.pdf; 2001. Goedkoop M, Demmers M, Collignon MX. The Eco-indicator 95, manual for designers. Available from. Pré Consultants, www.pre-nl/download/ EI95ManualForDesigners.pdf; 1996. Pramac Solar MCPH Photovoltaic Module Datasheet. See also http://solar. pramac.com/images/stories/products/mcph-7_en.pdf NE Nomisma Energia. Due Diligence tecnologico-industriale thin film micromorph Pramac Luce. Final Report. See also, http://www.nomismaenergia.it/ cms/index.php?page¼ricerca&hl¼it_IT; 2010. Ecoinvent v2.0, Frischknecht R, Jungbluth N, Althaus HJ, Doka G, Heck T, et al. Overview and methodology. Ecoinvent report No. 1. See also. Dübendorf: Swiss Centre for Life Cycle Inventories, http://www.ecoinvent.org/; 2007. Jungbluth N, Dones R, Frischknecht R. Life cycle assessment of photovoltaics: update of ecoinvent data v2.0. See also. Uster: ESU-services Ltd, http://www. esu.services.ch/; 2008. Fthenakis VM. End-of-life management and recycling of PV modules. Energy Policy 2000;28:1051e8. Fthenakis VM. Overview of potential hazards. In: Markvart T, Castaner L, editors. Practical Handbook of photovoltaics: fundamentals and applications. Amsterdam: Elsevier; 2003. p. 857e68. Stigliani WM, Anderberg S. Industrial metabolism at the regional level: the Rhine Basin. Available from. In: Ayres RU, Simonis UE, editors. Industrial metabolism. Tokyo: The United Nations University Press. p. 119e62, http:// www.unu.edu/unupress/unupbooks/80841e/80841E00.htm; 1994. Alsema E, de Wild-Scholten MJ. Environmental life cycle assessment of advanced silicon solar cell technologies. See also, http://www.ecn.nl/ publications/default.aspx?nr¼ECN-RX-04-060; 7e11 June 2004. Presented at the 19th European photovoltaic solar energy conference, Paris. Alsema E, de Wild-Scholten MJ. The real environmental impacts of crystalline silicon PV modules: an analysis based on up-to-date manufacturers data. See also, http://www.ecn.nl/publications/default.aspx?nr¼ECN-RX–05-015; 6e10 June 2005. Presented at the 20th European Photovoltaic Energy Research Conference and Exhibition, Barcelona. Alsema E, de Wild-Scholten MJ, Fthenakis VM. Environmental impacts of PV electricity generationea critical comparison of energy supply options. Presented at the 21st European Photovoltaic Solar Energy Conference, Dresden, 4e8 Sept 2006. See also, http://www.ecn.nl/publications/default.aspx? nr¼ECN-RX–06-016; 2006. Alsema E, de Wild-Scholten MJ. Environmental impacts of crystalline silicon photovoltaic module production. See also, http://www.ecn.nl/publications/ default.aspx?nr¼ECN-RX–06-041; 31 Maye2 June 2006. Presented at 13th CIRP International conference on life cycle engineering, Leuven. De Wild-Scholten MJ, Alsema E. Environmental impacts of crystalline silicon photovoltaic module production. See also, http://www.ecn.nl/publications/ default.aspx?nr¼ECN-RX–06-005; 28 Nove2 Dec 2005. Presented at Materials Research Society Fall 2005 Meeting, Boston. Bernal-Agustın JL, DufoeLopez R. Economical and environmental analysis of grid connected photovoltaic systems in Spain. Renewable Energy 2006; 31:1107e28. Alsema E. Environmental life cycle assessment of solar home systems. A Study by Commission of the Netherlands Agency for Energy and the Environment (NOVEM). Report No.: E-2000-15; 2000. Alsema E, Nieuwlaar E. Energy viability of photovoltaic systems. Energy Policy 2000;28:999e1010. Azzopardi B, Mutale J. Life cycle analysis for future photovoltaic systems using hybrid solar cells. Renewable and Sustainable Energy Reviews 2010; 14:1130e4. Bossert RH, Tool CJJ, Van Roosmalen JAM, Wentink CHM, De Vaan MJM. Thin film solar cells: technology evaluation and perspectives. A study by commission of the Netherlands Agency for Energy and the Environment (NOVEM). Report No.: DV 1.1.170. Available from, http://www.ecn.nl/docs/ library/report/2000/rx00005.pdf; 2000 May. Fthenakis V, Alsema E. Photovoltaics energy payback times, greenhouse gas emissions and External costs: 2004eearly 2005 Status. Progress in Photovoltaics: Research and Applications 2006;14:275e80. Ito M, Kato K, Komoto K, Kichimi T, Kurokawa K. A comparative study on cost and life-cycle analysis for 100 MW very large-scale PV (VLS-PV) systems in

4306

[39]

[40] [41]

[42]

[43]

[44]

M. Bravi et al. / Energy 36 (2011) 4297e4306 deserts using m-Si, a-Si, CdTe, and CIS modules. Progress in Photovoltaics: Research and Applications 2008;16:17e30. Kannan R, Leong KC, Osman R, Ho HK, Tso CP. Life cycle assessment study of solar PV systems: an example of a 2.7 kWp distributed solar PV system in Singapore. Solar Energy 2006;80:555e63. Knapp K, Jester T. Empirical investigation of the energy payback time for photovoltaic modules. Solar Energy 2001;71:165e72. Nishimura A, Hayashi Y, Tanaka K, Hirota M, Kato S, Ito M et al. Life cycle assessment and evaluation of energy payback time on high-concentration photovoltaic power generation system. Applied Energy, in press. Perpiñan O, Lorenzo E, Castro MA, Eyras R. Energy payback time of grid connected PV systems: comparison between tracking and fixed systems. Progress in Photovoltaics: Research and Applications 2008;17:137e47. Raugei M, Bargigli S, Ulgiati S. Life cycle assessment and energy pay-back time of advanced photovoltaic modules: CdTe and CIS compared to poly-Si. Energy 2007;32:1310e8. Stoppato A. Life cycle assessment of photovoltaic electricity generation. Energy 2008;33:224e32.

[45] Fthenakis VM, Chul KH, Alsema E. Emissions from photovoltaic life Cycles. Environmental Science and Technology 2008;42:2168e74. [46] Fthenakis VM. Sustainability of photovoltaics: the case for thin-film solar cells. Renewable and Sustainable Energy Reviews 2009;13:2746e50. [47] Ito M, Komoto K, Kurokawa K. Life-cycle analyses of very-large scale PV systems using six types of PV modules. Current Applied Physics 2010; 10:271e3. [48] Meijer A, Huijbregts MAJ, Reijnders L. Life-cycle assessment of photovoltaic modules: comparison of mc-Si, InGaP and InGaP/mc-Si solar modules. Progress in Photovoltaics: Research and Applications 2003;11:275e87. [49] Sakuta K. Life Cycle Assessment (LCA) of photovoltaic power generation. AIST Today (International Edition) 2008;28:18e9. [50] Pacca S, Sivaraman D, Keoleian GA. Parameters affecting the life cycle performance of PV technologies and systems. Energy Policy 2007;35:3316e26. [51] Raugei M, Frankl P. Life cycle impacts and costs of photovoltaic systems: current state of the art and future outlooks. Energy 2009;34:392e9. [52] RETScreen 4 International. Clean energy project analysis-RETScreen engineering & cases textbook. 3rd ed. Clean energy Decision Support Centre; 2005.