Photovoltaic (PV) Panel Waste Volumes-3

PV waste management plans should be adapted to the specific conditions of each country or region. As noted in case study reports on Germany and the United Kingdom, different waste management frameworks have emerged from the national implementation of the EU WEEE Directive. These experiences can also provide various lessons and best practices that other PV markets can benefit from. Rapidly expanding PV markets such as Japan, India and China still lack specific regulations covering PV panel waste. However, they have begun preparing for future waste streams by conducting research and development and setting long-term policy targets. In the absence of adequate waste volumes or country-specific technical knowledge, regional markets for waste management and recycling facilities also help maximize value creation from PV waste.

Photovoltaic Panel Waste Model

This section describes future photovoltaic panel waste streams. Most waste is typically generated during four lifecycle stages of any PV panel. These are 1) Panel manufacturing, 2) Panel transportation, 3) Panel installation and use, and 4) End of panel service life. The waste estimation model described below covers all lifecycle stages except manufacturing. This is because manufacturing waste is readily managed, collected and processed by waste treatment contractors or manufacturers, and is therefore not a public waste management issue. Future photovoltaic panel waste streams can be measured according to the model described in Figure 3. Two main input factors indicate the probability of conversion and loss during the lifecycle of photovoltaic panels (steps 1a and 1b). The Weibull function is used to model two waste stream scenarios using a regular loss and early loss scenario (step 2). The next section provides a step-by-step guide showing the methodological details and key assumptions. To estimate waste volumes of photovoltaic panels, installed and projected future photovoltaic capacity (megawatt or gigawatt—MW or GW) has been converted to mass (metric tons—t) as shown in Table 2. The average PV mass ratio per unit capacity (t/MW) was calculated by averaging available data related to panel weight and nominal power. For past photovoltaic panel production, the nominal power and weight of representative standard photovoltaic panels are averaged from leading manufacturers at five-year intervals. Various current data sheets were taken into account. Data for future PV panel production are based on the latest publications. The model described in this section includes a correction factor to account for panels becoming more powerful and lighter over time. This results from optimization of cell and panel designs as well as weight reduction from thinner frames, glass layers and overall materials. The correction factor is based on exponential least squares fitting of the weight-to-power ratio for historical and projected future panels. Figure 4 shows how the weight-to-power ratio continuously declines over time due to further advances in photovoltaic technologies such as material savings and improved solar cell efficiency. The potential sources of failure of roof and ground-mounted photovoltaic panels were analyzed independently of photovoltaic technology and application area to estimate the likelihood of becoming waste before reaching estimated end-of-life targets for photovoltaic panels. Three main panel failure stages identified are shown in Table 3: [caption id="attachment_119698" align="aligncenter"] Figure 1. Two stages of the PV panel model[/caption]   Table 1. PV panel losses for stage 1a [caption id="attachment_119701" align="aligncenter"] Figure 1. PV panel weight-to-power ratio (t/MW) projection exponential curve fitting)[/caption]   Empirical data on the causes and frequency of failures in each of the stages defined above can be obtained from various literature sources. Regardless of these stages, Figure 2 presents a general overview of the main reasons for photovoltaic panel failure. Among the main causes of installation failures are light-induced degradation (observed in 0.5%–5% of cases), poor planning, inadequate installation work and poor support structures. Many installation failures have been reported in electrical systems such as junction boxes, string boxes, charge controllers, wiring and grounding. The causes of mid-life failures are mostly related to degradation of the anti-reflective coating of the glass, discoloration of ethylene vinyl acetate, delamination and cracked cell insulation. Frequently, after exposure to mechanical load cycles within the first 12 years of all stages, for example, failures such as (wind and snow loads and temperature changes) potentially induced degradation, contact failures in junction boxes, glass breakage, loose frames, cell interconnection failures and diode defects. In the wear-out phase, failures such as those reported in the mid-life phase increase exponentially, in addition to severe corrosion of cells and interconnections. Previous studies containing statistical data on photovoltaic panel failures show that 40% of the examined photovoltaic panels suffer from at least one cell with microcracks. This defect is reported more frequently in new panels produced after 2008 due to thinner cells used in manufacturing. These failures and loss probabilities, together with step 1a (conversion factors) data, are used to estimate photovoltaic panel waste streams (step 2). Based on steps 1a and 1b, two photovoltaic waste scenarios have been defined (see Table 4)—regular loss scenario and early loss scenario. Both scenarios are modeled using the Weibull function as specified in the formula below. Therefore, the probability of loss during the lifecycle of photovoltaic panels is determined by the shape factor α, which differs for the regular loss and early loss scenarios. In both scenarios, an average panel life of 30 years and 99.99% loss probability after 40 years are assumed. A 30-year panel life is a common assumption in PV lifetime Environmental Impact analysis (for example, in lifecycle assessments) and is recommended by IEA-PVPS. The model assumes that photovoltaic panels were dismantled for renewal and modernization over the past 40 years. Therefore, the durability of photovoltaic panels is assumed to conform to the average experience of construction and building products such as facade elements or roof tiles. These also typically have a service life of 30–40 years. Neither early losses nor early failures were included in the normal loss scenario. Results obtained from Kuitsche (2010) are used directly in this scenario assuming an alpha shape factor of 5.3759 (see Table 5). In the early loss scenario, the following loss assumptions are made based on literature analysis and expert judgment. [caption id="attachment_119707" align="aligncenter"] Figure 2. Failure rates according to customer complaints[/caption]   • PV panels with 0.5% damage are accepted to reach end-of-life in loading and transportation stages, • 0.5% of photovoltaic panels will become waste within two years due to poor installation, • 2% will become waste after ten years, • 4% will become waste after 15 years due to technical failures. The early loss scenario includes failures requiring panel replacement such as broken glass, broken cells or strips, and a cracked backsheet with insulation defects. However, only panels with serious functional or safety defects requiring complete replacement are included, while other defects such as those reducing power output or causing panel discoloration are disregarded. In the early loss scenario, the shape factor is calculated by regression analysis among data points in the literature and is also considered as early failures (see Table 3). The resulting alpha shape factor of 2.4928 for the early loss scenario is a lower Weibull function and due to differently assigned alpha parameters, the regular loss and early loss scenarios have opposite effects after 30 years. Therefore, the regular-loss scenario shows a higher loss probability than 30 years. F(t) = 1-e -(t/T) α Where t = time (years), T = average service life, α = shape factor controlling the typical S shape of the Weibull curve. In our next article, we will continue with the calculation of "regular-loss" and Weibull parameters. Cemil Koyunoğlu / Yalova University - Faculty of Engineering - Energy Systems Engineering Department