S



SI AppendixS.1: LCA case study and uncertainty resultsTable S1 Comparison of the LCIA results. Values are presented per functional unit (1 ton SO2 reduction).AmmoniaLimestoneActive cokeAmountGSD2AmountGSD2AmountGSD2Climate change (kg CO2 eq)2.74×1031.242.68×1031.28881.561.27Ozone depletion (kg CFC-11 eq)2.62×10-51.601.87×10-51.622.02×10-61.58Terrestrial acidification (kg SO2 eq)11.04 1.277.711.535.141.25Freshwater eutrophication (kg P eq)2.84×10-23.582.46×10-24.514.05×10-35.67Marine eutrophication (kg N eq)0.601.310.641.450.281.26Human toxicity (kg 1,4-DB eq)158.992.0949.341.5222.031.26Photochemical oxidant formation (kg NMVOC)14.471.3412.671.587.441.26Particulate matter formation (kg PM10 eq)8.01.4312.981.731.931.27Terrestrial ecotoxicity (kg 1,4-DB eq)5.18×10-21.992.28×10-21.475.39×10-31.29Freshwater ecotoxicity (kg 1,4-DB eq)0.182.560.133.382.69×10-22.62Marine ecotoxicity kg 1,4-DB eq1.313.010.221.664.38×10-21.38Ionising radiation (kBq U235 eq)188.023.7730.553.334.663.47Agricultural land occupation (m2a)5.452.942.852.910.332.27Urban land occupation (m2a)10.681.835.441.730.631.53Natural land transformation (m2)0.122.194.67×10-22.483.82×10-33.84Water depletion (m3)2.69×1031.74387.091.5061.211.46Metal depletion (kg Fe eq)190.732.7918.512.051.541.86Fossil depletion (kg oil eq)776.771.22451.321.33219.891.21Table S2 Comparison of the probability of the scenariosCategoriesProbabilityAmmonia ≥ limestoneLimestone ≥ active carbonAmmonia ≥ active carbonClimate change56.60%100%100%Fossil depletion100%100%100%Freshwater ecotoxicity91.30%100%100%Freshwater eutrophication81.9%100%100%Human toxicity100%100%100%Ionising radiation100%100%100%Marine ecotoxicity100%100%100%Marine eutrophication33.50%100%100%Metal depletion100%100%100%Natural land transformation98.60%99.30%99.80%Ozone depletion90.4%100%100%Particulate matter formation6.6%100%100%Photochemical oxidant formation82.0%98.8%100%Terrestrial acidification98.5%96.2%100%Terrestrial ecotoxicity99.1%100%100%Urban land occupation96.3%100%100%Water depletion100%100%100%Agricultural land occupation88.8%100%100% Table S1 shows the life cycle impact assessment (LCIA) midpoint scores. The ammonia scenario presents the highest median value in most categories, except for marine eutrophication and particulate oxidant formation categories; the limestone scenario presented the highest value. By contrast, the active carbon scenario exhibited the lowest median impact in all the categories. To further determine the level of confidence, the GSD2 is defined based on the Monte Carlo simulations. The Monte Carlo simulations yielded a GSD2 on climate change score of 1.24 for the ammonia, 1.28 for the limestone, and 1.27 for the active coke scenario. The results indicated that a 95% confidence interval corresponds to the integration over the range for the ammonia, limestone, and active coke scenarios on the climate change potential scores of 2.2×103 –3.4×103, 2.1×103 – 3.4×103, 694.1 – 1.1×103 kg CO2 eq., respectively. For the rest of the categories and scenarios, similar conclusions can be made with the GSD2 value shown in Table S1. Table S2 presents a comparison of the probability of the scenarios using Monte Carlo simulations. The probability that ammonia has higher climate change and fossil depletion scores than limestone is 56.6% and 100%, respectively. Thus, the climate change score of the ammonia scenario is similar to that of the limestone scenario, whereas the fossil depletion score of the ammonia scenario higher than that that of the limestone scenario. For the rest of the categories and scenarios, a similar analogy can be made with the probability results shown in Table S2. Active coke has a conclusive and significant effect on reducing the environmental potential impacts in all the categories. The ammonia scenario presents the highest environmental burden in most of the categories, except for climate change, freshwater and marine eutrophication, particulate matter formation, photochemical oxidant formation, and categories (Table S2). S.2: Sensitivity analysis of LCIA methodTable S3 Life cycle impact assessment results of the limestone scenario. Values are presented per functional unit.CategoriesUnitReCiPeIMPACT2002+TRACICMLClimate changekg CO2 eq2.68×1032.44×1032.68×1032.66×103Ozone depletionkg CFC-11 eq1.87×10-51.87×10-52.25×10-52.03×10-5Terrestrial acidificationkg SO2 eq 7.7165.899.397.19Freshwater eutrophicationkg P eq2.46×10-22.68×10-2Marine eutrophicationkg N eq0.64Human toxicitykg 1,4-DB eq49.39Photochemical oxidant formationkg NMVOC12.67Particulate matter formationkg PM10 eq12.9811.58kg PM2.5eq10.22kg PM2.5eqTerrestrial ecotoxicitykg 1,4-DB eq2.28×10-2Freshwater ecotoxicitykg 1,4-DB eq0.13Marine ecotoxicitykg 1,4-DB eq0.22Ionising radiationkBq U235 eq30.55Agricultural land occupationm2a2.85Urban land occupationm2a5.44Natural land transformationm24.67×10-2Water depletionm3387.09Metal depletionkg Fe eq18.51Fossil depletionkg oil eq451.3220.08 GJ primaryTable S4 Life cycle impact assessment results of the ammonia scenario. Values are presented per functional unit.CategoriesUnitReCiPeIMPACT2002+TRACICMLClimate changekg CO2 eq2.74×1032.51×1032.74×1032.72×103Ozone depletionkg CFC-11 eq2.62×10-52.59×10-53.01×10-52.85×10-5Terrestrial acidificationkg SO2 eq11.0480.5612.6910.48Freshwater eutrophicationkg P eq2.84×10-22.01×10-2Marine eutrophicationkg N eq0.60Human toxicitykg 1,4-DB eq158.99Photochemical oxidant formationkg NMVOC14.47Particulate matter formationkg PM10 eq8.06.09kg PM2.5eq4.54kg PM2.5eqTerrestrial ecotoxicitykg 1,4-DB eq5.18×10-2Freshwater ecotoxicitykg 1,4-DB eq0.18Marine ecotoxicitykg 1,4-DB eq1.31Ionising radiationkBq U235 eq188.02Agricultural land occupationm2a5.45Urban land occupationm2a10.68Natural land transformationm20.12Water depletionm32.69×103Metal depletionkg Fe eq190.73Fossil depletionkg oil eq776.7735.98 GJ primaryTable S5 Life cycle impact assessment results of the active coke scenario. Values are presented per functional unit.CategoriesUnitReCiPeIMPACT2002+TRACICMLClimate changekg CO2 eq881.56800.85881.56873.73Ozone depletionkg CFC-11 eq2.02×10-62.02×10-62.33×10-62.43×10-6Terrestrial acidificationkg SO2 eq5.1439.696.124.96Freshwater eutrophicationkg P eq4.05×10-34.65×10-3Marine eutrophicationkg N eq0.28Human toxicitykg 1,4-DB eq22.03Photochemical oxidant formationkg NMVOC7.44Particulate matter formationkg PM10 eq1.931.07kg PM2.5eq0.24kg PM2.5eqTerrestrial ecotoxicitykg 1,4-DB eq5.39×10-3Freshwater ecotoxicitykg 1,4-DB eq2.69×10-2Marine ecotoxicitykg 1,4-DB eq4.38×10-2Ionising radiationkBq U235 eq4.66Agricultural land occupationm2a0.33Urban land occupationm2a0.63Natural land transformationm23.82×10-3Water depletionm361.21Metal depletionkg Fe eq1.54Fossil depletionkg oil eq219.899.72 GJ primaryS.3 Methods of investigating the partial relation between industrial dust emission and desulfurization The fly dust emitted from coal burning is known to be the most significant contributor to PM2.5 formation in China. The national dust environmental emission from coal burning can be calculated using Eq. (S1): QUOTE DE= (1-DDR) × Dc (S1)where DE, DDR, and Dc are the dust environmental emission, dedust rate, and dust generation amount from coal burning, respectively.Here, Dc =Dw + Dn = A×[(1-η) ×CCtotal×D + CCtotal ×S×DSR×Ds] ×0.5+(1-η) ×(1-A) × CCtotal×D = 0.5×A× (1-η) ×CCtotal ×D + 0.5 CCtotal ×S×DSR×Ds×A+(1-η) ×CCtotal×D-A×(1-η) ×CCtotal ×D = (1-0.5A) × (1-η) × CCtotal ×D+0.5×CCtotal×S×DSR×Ds×A =CCtotal×[(1-0.5A)×D×(1-η)+0.5×S×DSR×Ds×A] (S2)where Dw, and Dn are the national and provincial dust generation with and without FGD system form coal consumption, respectively; A, η, CCtotal, D, S, Ds, and DSR are the desulfurization system installation rate, dedusting rate of electric dedusting system, national and provincial total coal consumption, average dust environmental emission per ton coal consumption, SO2 generation per ton coal consumption, average dust environmental emission generated from the FGD system per ton SO2 removal, and desulfurization rate, respectively. When A, η, D, S, and Ds are considered as constants, Eq. (S2) can be rewritten as follows: QUOTE Dc= CCtotal×(b+c×DSR) (S3)Thus, Eq. (1) can be rewritten as follows:DE= CCtotal×(1-DDR)×(b+c×DSR) (S4)Controlling SO2 pollution in China is mainly implemented in industrial site; hence, a linear relationship holds between the national total and industrial coal consumption from 2003 to 2012 (Ref.3, Fig.S1). Assuming that the total coal consumption is proportional to the industrial coal consumption, Eq. (S4) can be rewritten as IDE=a×(1-DDR)×(b+c×DSR) ×CC (S5)where a, b, c are constants; IDE and CC refer to the coal consumption of industrial sites.Fig.S1 Relationship between the national total and industrial coal consumption from 2003 to 2012S.4 Effect of partial desulfurization on industrial dust emissionTo verify Eq. (1) using a regression model, we use the panel data for the above-mentioned variables from 2003–2012 for all provinces in China, except for Hongkong, Tibet, Macau, and Taiwan.1,2 The result is shown in Table S1.Table S1 Regression resultsRandom effectFixed effectOLSDSRConstantNo. of obs0.0635***(0.0000)0.0979***(0.0000)0.07792920.0633***(0.0000). 0.05522920.0563***(0.0000)0.1001***(0.0000)0.0669292Notes: DSR denotes the desulfurization rate, whereas OLS refers to ordinary least squares. The p-value of the two-tailed t-test is in parenthesis. One, two, or three asterisks indicate that a coefficient estimate is significantly different from zero at 10%, 5%, or 1% level (alpha level), respectively.We observe that industrial dust emission is positively related to the rate of desulfurization, ceteris paribus. Meanwhile, the effect of partial desulfurization on industrial dust emission is substantial at the significant level of 1%. For the robustness of the result, we also use the fixed effect and OLS (ordinary least squares) models. The sign and significance of the marginal effect are consistent with those of the random effect model. A Hausman test is used to differentiate the fixed-effects model and random-effects model. The statistical result is 0.0004 with a p-value of 0.9848. Hence, the null hypotheses cannot be rejected in all three cases. Therefore, the random-effects model is preferable to the fixed-effects model.S.5 Effect of SO2 removal rate on gypsum emitted from wet limestone-based FGD systemThe national and provincial SO2 removal rate and contribution of gypsum annual fly dust emission are calculated using Eqs. (S6) and (S7), respectively. r=R/(R+E) (S6)g=(G×R)/D QUOTE (S7)where the values of r, R, E, and D are the annual national and provincial rates of SO2 removal rate, SO2 reduction, SO2 environmental emission, and dust environmental emission, respectively; g and G are the contribution of gypsum particulate to dust emission and gypsum particulate emission amount for per ton SO2 reduction, respectively. The best case of the composition of gypsum particulates contained in desulfurized flue gas of 50% is used to calculate the value of G because more than 50% of the gypsum generally contained in desulfurized flue gas has been reported.3,4 The value of G is derived from the actual wet limestone-based FGD system running data (Table S7) and is calculated using Eq. (S8) or (S9), as shown in Fig.S2. The obtained median value of G is 9.78×10-3 t/t-SO2 reduction. Thus, the national removal rate of SO2 and the ratio of gypsum to annual fly ash emission in thermal power generation site can be deduced in Fig.5. G=0.5×d /[C×S×0.9×64/32)-O] (S8) QUOTE G=0.5×d/(I-O) (S9)where d, C, S, O, and I are effluent fly dust (kg/h), coal consumption (kg/h), sulfur content of coal (%), influent SO2 (kg/h), and effluent SO2 (kg/h), respectively. Fig.S2 Gypsum particulate emission amount per ton SO2 reductionTable S7 Wet limestone-based FGD system running data of coal-based power generation plants. PlantSulfur content (%)Limestone consumption amount (t/d)Coal consumption amount (t/d)Influent SO2 amount (kg/h)Effluent SO2 amount (kg/h)Effluent fly dust amount (kg/h)1#a0.911425268-4964.3-1778.7-59118.10.961525268-4053.1-10157.1-9260.50.911125843-Nondetected59.4-Nondetected74.3-Nondetected64.90.961255274-2150.1-Nondetected61.9-4156.62#b0.46564584142860.432--132662.536574440138255.1350.43574464147357.234554416146356.234--148659.435a Monitor data were taken from environmental report of Hubei Ezhou power plant in 2013.bMonitor data were taken from environmental report of Liaoning Shenxi power plant in 2013.S.6 The potential for environmental improvement using active coke-based FGD technology instead wet limestone-based FGD technologyThe national dust, industrial SO2 emission and its reduction rate, greenhouse gas, and SO2 emission, and standard coal consumption amount in 2013 were 1.28×107t (Ref.1), 1.84×107t (Ref.5), 72.2% (Ref.5), 1.03×1010 t （Ref.6），2.04×107 t (Ref.5), and 3.75×109 t (Ref.1), respectively. The dust and SO2 emissions and SO2 reduction rate in 2013 of independent thermal power industry sites were 1.84×106 t, 6.34 ×106 t, and 80.3% (Ref.5), respectively. Thus, the national SO2 reduction amount from the industrial and independent thermal power generation sites in 2013 can be calculated as follows: Industry: R=r×E/(1-r)=72.2%×1.84×107/(1-72.2%)=4.78×107 tIndependent thermal power industry: R=r×E/(1-r) = 80.3%×6.34 ×106 / (1-80.3%) = 2.58 ×107 tAs shown in Table S3, approximately 1.56 t CO2 eq, 20.4 kg PM10 eq, and 173.9 kg oil eq of climate change, particulate matter formation, and fossil depletion impact can be saved by the removal of every ton of SO2, respectively. Accordingly, the potential for national environmental improvement using active coke-based FGD technology, instead of wet limestone-based FGD technology, can be calculated as follows:National dust emission saving rate= 4.78×107 t ×20.4 kg/t/(1.28×107 t) = 7.62% ≈ 7.6%National greenhouse gas emission saving rate= 4.78×107 t×1.56 t/t /(1.03×1010 t) = 0.72% ≈ 0.7%National energy saving rate = 4.78×107 t×(173.9 kg/t × 42.6MJ/kg)/(3.75×109 t×103×29.27MJ/kg) = 0.32% ≈ 0.3%Dust emission saving rate from power industry sites = 2.58×107 t ×20.4 kg/t /(1.84×106 t) = 28.6% ≈29%Energy saving rate from power industry sites = 2.58×107 t ×1.56 t/t /(4.19×1012 kWh×3.21×10-4 /kWh) = 2.99% ≈3.0%S.7 Life cycle costing of three scenariosFig.S3 Life cycle costing of three scenariosTable S8 Market price of energy, raw materials, labor, and initial investment. UnitMarket price ($/Unit)Electricity MWh 118.3Water t0.5Steamt37.7Liquid ammoniat259.7Polypropylenet 965.6Limestonet9.0Coalt59.9Compressed airNm32.1×10-2HCl (purity: 31%) L3.0×10-2NaClO (purity: 10%) L8.2×10-2TMT15 (purity: 15%) L1.42Ca(OH)2kg 6.8×10-2Na2CO3kg 0.2FeCl3 (purity: 41%)L0.6PAMkg 1.20Gasolinekg0.6Dieselkg0.7Disodium hydrogen phosphatekg0.4Cementt43.3Woodm3553.9Sandm39.0Dynamitekg1.3Steelkg0.5Desulphur catalysism34.5×104Methyldiethanolaminem31.3×105Nitrogenm30.1Ammonia sulfatet70.4Desulfurization gypsumt7.5Sulfuric acidt38.9Initial investment of ammonia scenario kW17.6Initial investment of limestone scenario kW15.7Initial investment of active coke scenario kW52.4Laboryr/person8982 The market price of energy, raw materials, labor, initial investment (Table S8) are used to evaluate the life cycle costing (LCC). In addition, external market price (based on the September 12, 2016 exchange rate of USD 1.00 = 6.68 Yuan), including carbon trade cost ($1.47/t), arsenic ($216.05/t), COD($216.05/t), lead($216.05/t), mercury ($216.05/t), nitrogen oxides ($185.18/t), particulates ($92.59/t), and sulfur dioxide($185.18/t) in China, is involved. Fig.S3 represents the LCC results. For the ammonia, limestone, and active coke scenarios, the total economic impact is $189.96/t, $311.29 /t, and $244.09/t, respectively. This overall economic impact of each scenario was mainly attributed to the prices of steam, electricity, byproduct produced by each scenario, initial investment, active coke, ammonia, or limestone (Fig.S3). Approximately 39% and 21% economic benefit can be observed in the ammonia and active coke scenarios, respectively.References1) National Bureau of Statistics of China. China Statistical Yearbook, Beijing:?China?Statistics?Press (2004-2014).2) China Energy Statistical Yearbook. Energy Statistics Division of National Bureau of Statistics (2004-2013). 3) Liu, Q.Z., Sun, Y.J. & Sun, Y. Cause analysis and countermeasure of gypsum rain in coal-fired power plants. J. Environ. Prot. 4, 1-4 (2013).4) Meij, R. & Te Winkel, B. The emissions and environmental impact of PM10 and trace elements from a modern coal-fired power plant equipped with ESP and wet FGD. Fuel Process. Technol. 85, 641-656 (2004). 5) Arnaud, D., Nando de, F., Neil, G. Sequential Monte Carlo methods in practice. New York: Springer. ISBN 0-387-95146-6 (2001). 6) Huijbregts, M., Gilijamse, W., Ragas, A. & Reijnders, L. Evaluating uncertainty in environmental life-cycle assessment a case study comparing two insulation options for a Dutch one-family dwelling. Environ. Sci. Technol. 37, 2600–2608 (2003). ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download