Potential for solar industrial process heat in the United States: A look at California

Transcription

1 Potential for solar industrial process heat in the United States: A look at California Parthiv Kurup, and Craig Turchi Citation: AIP Conference Proceedings 1734, (2016); View online: View Table of Contents: Published by the American Institute of Physics Articles you may be interested in Comparing the net cost of CSP-TES to PV deployed with battery storage AIP Conference Proceedings 1734, (2016); / State of the art of performance evaluation methods for concentrating solar collectors AIP Conference Proceedings 1734, (2016); / Solar thermoelectricity via advanced latent heat storage: A cost-effective small-scale CSP application AIP Conference Proceedings 1850, (2017); / Review and future perspective of central receiver design and performance AIP Conference Proceedings 1850, (2017); / Solar thermoelectricity via advanced latent heat storage AIP Conference Proceedings 1734, (2016); / Long-term heating to improve receiver performance AIP Conference Proceedings 1850, (2017); /

2 Potential for Solar Industrial Process Heat in the United States: A Look at California Parthiv Kurup 1, a) 1, b) and Craig Turchi 1 National Renewable Energy Laboratory, Denver West Parkway, Golden, Co, USA. a) Corresponding author: parthiv.kurup@nrel.gov b) craig.turchi@nrel.gov Abstract. The use of Concentrating Solar Power (CSP) collectors (e.g., parabolic trough or linear Fresnel systems) for industrial thermal applications has been increasing in global interest in the last few years. In particular, the European Union has been tracking the deployment of Solar Industrial Process Heat (SIPH) plants. Although relatively few plants have been deployed in the United States (U.S.), we establish that 29% of primary energy consumption in the U.S. manufacturing sector is used for process heating. Perhaps the best opportunities for SIPH reside in the state of California due to its excellent solar resource, strong industrial base, and solar-friendly policies. This initial analysis identified 48 TWh th /year of process heat demand in certain California industries versus a technical solar-thermal energy potential of 23,000 TWh th /year. The top five users of industrial steam in the state are highlighted and special attention paid to the food sector that has been an early adopter of SIPH in other countries. A comparison of the cost of heat from solar-thermal collectors versus the cost of industrial natural gas in California indicates that SIPH may be cost effective even under the relatively low gas prices seen in A recommended next step is the identification of pilot project candidates to promote the deployment of SIPH facilities. INTRODUCTION Thermal energy and steam are ubiquitous needs in industrial processes. From the extraction of raw materials to food processing, heat is a vital part of the processing and manufacturing sectors. In the 1970s and 1980s there was a great deal of interest in collection of solar thermal energy for buildings and process heat applications [1]. Despite significant effort, very few projects came to fruition, mainly due to high solar collector costs and an associated inability to effectively compete with natural gas [2]. In recent years, the improvement and proliferation of solar collectors for electricity generation and the development of sophisticated solar collector modeling tools has regenerated interest in solar process heat applications. The number of Concentrating Solar Power (CSP) plants for electricity production has increased dramatically, with large plants constructed in the European Union (EU), Africa, Australia, and the United States. These concentrating technologies can achieve relatively high temperatures. For example, linear-focus collectors can reach temperatures up to about 500 C and point-focus technologies can reach even higher temperatures (e.g., up to 800 C). For solar hardware developers, expansion into industrial process heat (IPH) offers access to new markets for CSP collector technologies. In addition, greater market size can help drive down the cost of CSP collectors for both electrical and IPH applications through economies of scale in manufacturing and learning-curve advances in deployment. This paper looks at the potential for CSP collector technologies to supply IPH in California as a prime example of the United States. Global Activity in Solar Industrial Process Heat In 1977 the International Energy Agency (IEA) established the Solar Heating and Cooling (SHC) program to create an environment for the development and progression of SHC [3]. An EU-led collaborative project between the SHC program and the SolarPACES program known as Task 49/Task IV was created specifically to establish and SolarPACES 2015 AIP Conf. Proc. 1734, ; doi: / Published by AIP Publishing /$

3 help meet the potential of solar for IPH [3,4]. (Note that the IEA uses the acronym SHIP for solar heat for industrial processes. The term industrial process heat is recognized in the United States, and this paper will refer to these applications as solar IPH or SIPH ). Much of the initial work in the IEA program dealt with the potential of nonconcentrating, flat-plate collectors. Flat-plate solar collectors are common in many countries, including the United States, where the overwhelming majority is applied to domestic home heating or water heating for swimming pools [5]. While these are excellent applications for low-temperature collectors, this paper deals with the growing interest in the deployment of concentrating collector technologies that can achieve temperatures needed within the industrial sector. Despite great potential, the worldwide adoption of concentrating collectors for SIPH generation is modest. For example, of the 155 SIPH plants listed in the IEA database, only 18 involved concentrating collectors [5]. The 18 SIPH plants are already in operation for industries such as Food (e.g. solar heating of steam) and Dairy Production (e.g. heat for sterilization of milk) [5]. Due to the excellent solar resource conditions in the United States (especially in the Southwest) and the ubiquitous need for IPH, the United States provides a sizeable opportunity for greater deployment of concentrating solar-thermal collectors with the associated benefits of increased solar jobs, lower carbon emissions, and potential cost reductions in collector technologies. For example, within the industrial sector in the United States the estimated consumption of energy for heat for applications such as washing, sterilization, and preheating was approximately 7.0x10 3 TWh th (24x10 15 Btu) in 2014 [6]. Depending on the specific industry in question, between 35% and 50% of the total energy consumption can be for IPH applications [6]. METHODOLOGY In the present study we adopt a top-down analysis approach that first examines the regional solar resource potential and the thermal energy demand characteristics of potential user industries. We next focus on specific industries that have the largest thermal energy requirements in the appropriate temperature range for concentrating solar collectors. This method provides an initial, high-level assessment of the SIPH potential (focused on California). The methodology has been developed based on other studies [7]. Understanding the conventional energy source (e.g., natural gas, waste heat, or electricity) is also examined to assess the economic potential. It is worth highlighting that there are different levels of potential. The raw resource potential considers only solar direct normal irradiance (DNI) and land area. The second level is the technical potential, where technologyspecific performance, topographic, and land-use constraints are applied, as defined in a prior study from the National Renewable Energy Laboratory (NREL) [8]. Realizable potential is related to technical potential, though further assessments and constraints are added to scale-back the technical potential to a probable target [7]. This paper considers only raw resource and technical thermal energy potentials and focuses California. Promising regions identified in this study would be prime candidates for follow-on, site-specific potential analysis (e.g., realizable potential and case studies). FIGURE 1. Methodology developed for this study to determine U.S. SIPH potential with an example of the food industry in California

4 For this study, researchers chose to undertake a top-down approach of the potential for SIPH (with a look at California), primarily because it gives a good overview of opportunity derived from the solar resource potential and relevant industries in the region. The purpose was to test the hypothesis that U.S areas with high DNI also contain an industrial need for steam and process heat due to the high consumption of natural gas. Figure 1 highlights the steps used to understand the thermal energy demand and the technical thermal energy potential for SIPH in the state of California. California was selected due to its strong DNI resource, natural gas consumption for industrial steam, and policies favorable to deployment of renewable energy technologies. The disadvantage of a top-down approach, especially for SIPH, is that the raw resource and technical potential can be quite large while industrial users of process heat may consume only a tiny fraction of the technical resource potential in a given geographic area. In such a scenario, the supply of solar energy is not a limiting factor and further research would be needed to determine whether a specific industrial site s IPH needs could be met by the potential supply. The proximity of the energy-supply and energy-demand locations is more important for SIPH than solar generation of electricity. Unlike electric power generation, the application of SIPH requires co-location of the solar field and the thermal-energy consumer. The majority of SIPH deployments are on land adjacent to the plant or on the facility's roof [5] because the distribution of hot fluids, such as steam, is difficult over long distances. For example, district heating systems rarely circulate fluids beyond an individual plant or campus site (hundreds of meters), whereas electric transmission lines can run for hundreds of kilometers. The top-down approach was selected as it provided a good overall assessment of whether further research should be undertaken for an area or facility without the commitment of time and resources necessary to undertake a detailed bottoms-up analysis. The bottoms-up approach as undertaken, for example in Australia [9], is a more accurate representation of site-specific potential, but it is most useful after performing an initial top-down analysis. RESULTS AND DISCUSSION Process Heat Consumption in the U.S. Manufacturing Sector The U.S. Manufacturing Energy Consumption Survey (MECS) is a national sample survey that collects information on the stock of U.S. manufacturing establishment, their energy-related building characteristics, and their energy consumption and expenditures [10]. Based on MECS data, the U.S. manufacturing sector has three primary energy sources: fuel, steam generation, and electricity generation. Considering the direct and indirect (e.g., onsite steam production) use of all energy sources, process energy consumed approximately 10,000 trillion Btu (TBtu, see Fig. 2). This represents approximately 42% of the total primary energy consumption in the U.S. manufacturing sector. (1 TBtu 293 GWh th.) From Fig. 2, the importance of process heating within process energy as a whole is clear. For 2010, the MECS industries utilized 2.1x10 6 GWh th /year (7,204 TBtu/year) for process heat generated from steam, electricity, and fuel or approximately 29% of the total primary energy in the U.S. manufacturing sector. FIGURE 2. Sankey diagram of process energy flow in U.S. manufacturing sector in 2010 [11]. Values are shown in trillion Btu

5 Figure 3 shows the MECS 2010 end-use subcategories for natural gas, which is the fuel most often used for process heating, conventional boiler use, and combined heat and power (CHP) or cogeneration. In contrast, electricity is most commonly used for direct machine drive with some use in process heating. Data indicate that natural gas replacement is the biggest opportunity for SIPH. Situations where electricity can be offset will be rare although they may present favorable economics. The U.S. industrial and manufacturing sectors are the largest consumers of natural gas and electricity for process heat either directly or indirectly through steam production via a conventional boiler [12,13]. A representation of the potential U.S. IPH market size for steam is provided in Fig. 4 which depicts annual steam consumption for the manufacturing sectors of food, paper, petroleum, chemicals, and primary metals the five sectors that have the greatest usage of natural gas. The total consumption in the range between 100ºC and 260 C amounts to about 1.7x10 6 GWh th /year. To put this in perspective, the 64 MW e Nevada Solar One CSP plant produces about 350 GWh th /year so the thermal energy potential depicted in Fig. 4 represents the equivalent of about 4,800 such plants if all the sites were suitable for SIPH. All the sectors listed in Fig. 4 utilize steam in temperature ranges suitable for solar generation; however, the food industry presents a particularly appealing target. As noted previously, the food sector has been the application of choice for many international SIPH plants. FIGURE United States MECS overall natural gas breakdown by end use [12]. FIGURE 4. IPH annual energy use for steam generation for the industries utilizing the greatest amount of natural gas [13]

6 Solar Industrial Process Heat Potential in California While the MECS data do not provide information at the state level, a 2014 California Energy Commission (CEC) study listed natural gas consumption for industries in California [14]. Assuming the MECS 2010 national data to be representative of the state level, the natural gas consumption specifically for direct process heating and conventional boiler use can be estimated for California. For example, in the MECS 2010 data, the food industry in the United States consumed approximately 59% of its total natural gas consumption for applications of direct process heating and conventional boiler use. Assuming this percentage holds for the food industry in California, it can be estimated that about 10,200 GWh th /year were consumed for direct process heating and conventional boiler use in the state [14]. This methodology was continued across all MECS sectors listed in Table 1 below. Table 1 shows the GWh th used by the MECS industries identified as the biggest consumers of steam at less than 260 C in California. The CEC concluded industries in California such as food processing, chemicals, petroleum, and primary metals manufacturing represent prime areas of opportunity for reducing natural gas use [14]. TABLE 1. Estimated natural gas consumption for direct process heating and boiler use in California for select MECS industries MECS sector with North American Industry Classification System code Natural Gas Consumption for Process Heating (GWh th /year) Food Manufacturing (311) 10,200 Paper Manufacturing (322) 1,244 Petroleum and Coal Products Manufacturing (324) 31,211 Chemical Manufacturing (325) 3,526 Primary Metal Manufacturing (331) 2,134 Solar Resource in California 48,100 This study creates an initial outlook for the potential of SIPH in California. This state is part of a region that contains the highest annual DNI in the United States with a resource ranging from about 5 to 7.5 kwh/m 2 /day [15]. Focusing on California specifically, or the southwest more generally, does not imply they are the only areas suitable for SIPH, but that they could offer the best initial entry point for the technology. A prior analysis of CSP potential in the Southwest [8] was used as the basis for estimating the technical thermal energy potential for SIPH subject to several modifications. A commercial parabolic trough system was assumed, but without storage and with a solar multiple of 1.4. The following formula was used for determining the heat generation potential for a given region: MWh % 8760 (1) The thermal power density of 139 MW th /km 2 was derived from the electric power density used in the prior NREL study (49.43 MW e /km 2 ) divided by the thermo-electric power cycle efficiency of that study (35.5%) [8]. Energy density was debited for each grid square by using an estimate of capacity factor that was a function of resource quality. The analysis assessed the resource and available land area at 10-km by 10-km resolution and summed the results for California. The estimated technical thermal energy potential for California from solar was 23,000 TWh th /year [16], orders of magnitude beyond the demand of 48 TWh th /year shown in Table 1. California Food and Dairy Industries An objective for this paper is to tie specific industries known for high use of natural gas or electricity for IPH to their local solar thermal energy potential. For this study, the food industry and sample sub-industries in California were considered prime candidates. These include the MECS food sub-industries of animal-food processing, breweries, and dairy products. To highlight that the thermal demand of these industries can be theoretically met with the SIPH potential found for California, maps have been created overlaying the location of the industries in California and the estimated technical thermal energy potential. Figure 5 shows the co-locations of known animal

7 food processing plants, breweries, and dairy products along with the technical thermal energy potential by county. These 3 sub-industries were selected as they were considered representative of the Food industry of California. Figure 6 zooms in on Fresno, California and shows the locations of the animal-food manufacturing, breweries, dairy product manufacturing, and fruit and vegetable producers overlaid with the solar thermal energy supply. Developed city areas and other exclusion zones are shown in white. While the majority of sites are within the Fresno city limits, there are clusters of specific industries that could potentially benefit from SIPH plants. As can be seen, clusters A, B, and C of fruit and vegetable manufacturing plants potentially have available land for solar developers to install SIPH facilities and provide heat to augment steam production processes. Clusters of multiple users near a potential SIPH plant site increase the likelihood of favorable economics. FIGURE 5. Locations of animal-food manufacturing, breweries, and dairy products plants across California along with annual solar thermal energy potential (MWhth/km2/year). FIGURE 6. Close-up of Fresno showing the solar thermal generation potential and potential user industries

8 Estimated Cost of Solar Thermal Heat The levelized cost of heat (LCOH) is a convenient metric for estimating lifetime cost of a solar collector system for process heat applications. LCOH is defined analogously to LCOE which conventionally refers to levelized cost of electric energy. In its simplest form, LCOH is defined as: & (2) where FCR is the fixed charge rate. The FCR depends on a range of financial parameters that can have a significant influence on LCOH. NREL s SAM model includes various ways of estimating LCOE. The latest release includes a procedure for estimating and using the FCR method that is used in this study [17]. A realistic installed cost of an SIPH solar field is about $200/m 2 [16]. As depicted in Fig. 7, at that cost, solar thermal energy is competitive with natural gas (i.e., has a lower LCOH) at its average California price of $7.6/MMBtu (in 2014) [18] for locations where the solar DNI is about 6.0 kwh/m 2 /day or greater. However, the $200/m 2 solar field cost is not competitive with natural gas at its 2014 U.S. national average price of $5.4/MMBtu even at the highest DNI level. The data suggest that economic SIPH applications can be found in California at existing solar hardware costs and market gas prices, with incentives in the marketplace further expanding the range of feasible projects. At the time of writing, the California Solar Initiative Thermal Incentive Program offers industrial sites up to $800,000 to set up an SIPH plant for the displacement of natural gas. Solar technology developers estimate that under this program, even with the relatively low gas prices of today, payback period on these projects could be less than three years. In summary, project viability will be strongly dependent on the specific solar project costs including any incentives and the specific gas pricing contract in place. The deployment of a few successful pilot projects would be expected to spur further utilization of SIPH with a concomitant decrease in project development costs. FIGURE 7. Estimated LCOH for different solar resource and solar field costs compared with two natural gas (NG) prices from Total installed project cost includes solar field, site preparation, heat transfer fluid piping, heat exchanger, and other project costs. Gas costs include $200/kW burner cost and 80% efficiency. Based on FCR = 0.101, WACC = 6.2% [16]. CONCLUSIONS This study considers the use of solar-thermal energy for application to IPH in the California. For most industrial applications, the steam or direct-heat temperature required typically is less than 260 C. This temperature is beyond the range that can be achieved with flat-plate collectors, but is ideal for concentrating solar collectors such as parabolic trough and linear Fresnel systems. A key part of this study has been to match known areas of high DNI to both states and industries that utilize high quantities of natural gas for direct process heating and conventional boiler use to generate steam. The investigation utilized previous research to identify the five industrial sectors that are the largest users of natural gas for these applications: food, paper, petroleum, chemicals, and primary metals manufacturing

9 Due to its high use of industrial natural gas, coupled with excellent DNI conditions, California is well suited to benefit from SIPH. Furthermore, at the time of writing, the California Solar Initiative Thermal Incentive Program offers industrial sites up to $800,000 to set up an SIPH plant for the displacement of natural gas. A complementary follow-on analysis to this study would be site-specific case studies that identify promising early-adopter locations to promote deployment of SIPH within the state. Within California, the process heat demand derived from the use of natural gas was calculated for the food, paper, petroleum, chemicals, and primary metals industries. It was found that these five key industries in California had an estimated demand of about 48 TWh th /year (shown in Table 1). This was shown to be orders of magnitude less than the state s technical thermal energy potential from solar-thermal collectors of 23,000 TWh th /year. Based on developer information, a realistic cost for an installed SIPH solar field is about $200/m 2. At that cost, and including other costs for piping, pumps, and heat exchangers, solar thermal energy is competitive with natural gas combustion at the 2014 average California price of $7.6/MMBtu for locations with solar DNI greater than about 6.0 kwh/m 2 /day. However, the same solar field cost is not competitive with natural gas at its reported national average price of $5.4/MMBtu up to the highest DNI levels in the United States. The data suggest that economic SIPH applications can be found in California at existing solar hardware costs and market gas prices. However, project viability will be strongly dependent on the specific solar project costs including any incentives and the specific gas pricing contract in place. The deployment of a few successful pilot projects would be expected to spur further utilization of SIPH with a concomitant decrease in project development costs. As shown, California has great potential to provide thermal energy via SIPH. The thermal yield analysis for California, if extended to the remainder of the Southwest, is likely to conclude that SIPH could provide sufficient thermal energy to meet the demands and requirements of large numbers of industrial sites. ACKNOWLEDGMENTS This work was supported by the U.S. Department of Energy under Contract No. DE-AC36-08-GO28308 with the National Renewable Energy Laboratory. Funding was provided by the Office of Energy Efficiency and Renewable Energy Solar Energy Technologies Program. REFERENCES 1. Kutscher, C.F. et al., Design Approaches for Solar Industrial Process Heat Systems, Solar Energy Research Institute, SERI/TR , August C. Carwile and R. Hewitt, Barriers to Solar Process Heat Projects: Fifteen Highly Promising (But Cancelled) Projects, National Renewable Energy Laboratory, NREL/TP , October IEA, Solar Heating and Cooling Programme, accessed July SolarPACES Task IV: Solar Heat in Industrial Processes, accessed July Solar Heat for Industrial Processes, SHIP Plants - Locations - United States, accessed July Environmental Protection Agency (EPA), Renewable Industrial Process Heat, renewable-industrial-process-heat#about%20industrial%20process%20heat, accessed August C. Lauterbach, B. Schmitt, U. Jordan, and K. Vajen, The potential of solar heat for industrial processes in Germany, Renew. Sustain. Energy Rev., vol. 16, no. 7, pp , September A. Lopez, B. Roberts, D. Heimiller, N. Blair, and G. Porro, U.S. Renewable Energy Technical Potentials: A GIS-Based Analysis. NREL/TP-6A , National Renewable Energy Laboratory, July A. C. Beath, Industrial energy usage in Australia and the potential for implementation of solar thermal heat and power, Energy, vol. 43, no. 1, pp , July U.S. Energy Information Administration (EIA), accessed June U.S. DOE Office of Energy Efficiency & Renewable Energy (EERE), Sankey Diagram of Process Energy Flow in U.S. Manufacturing Sector, accessed August U.S. EIA, Table End Uses of Fuel Consumption, MECS Survey Data, March D. B. Fox, D. Sutter, and J. W. Tester, The thermal spectrum of low-temperature energy use in the United States, Energy Environ. Sci., vol. 4, no. 10, pp , Sep

10 14. L. Schrupp, The Natural Gas Research, Development and Demonstration Program. California Energy Commission, CEC , March, Solar Power Prospector, NREL, accessed July P. Kurup and C.S. Turchi, U.S. Solar Industrial Process Heat Potential, NREL/TP-6A , in preparation, National Renewable Energy Laboratory, System Advisor Model (SAM), release version , LCOE calculator financial model, U.S. Energy Information Agency, Natural Gas Prices, accessed May