*3.2. Benchmarking with Competing Technologies*

3.2.1. Comparison with ORC Technology (Both Fueled with Wood Chips)

Figure 6 shows the comparison between HBP technology and ORC technology both fueled with wood chips.

**Figure 6.** Comparison of HBP technology with ORC technology for 1 MJ of heat. The graph for 1 kWh electricity shows some minor differences due to a slightly different Carnot factor assumed for ORC technology. Values are normalized with respect to the impacts of the most impacting scenario. CC = Climate change, PM = Particulate matter, POF = Photochemical ozone formation, AC = Acidification, TE = Terrestrial eutrophication, WRD = Water resource depletion, MFRD = Mineral, fossil, and renewable resource depletion. WC = wood chips.

Compared to the same amount of heat produced by the ORC technology, the heat co-generated by the HBP technology shows lower environmental impact in terms of climate change (−42%) and much lower impact (−87%/−96%) in terms of particulate matter, photochemical ozone formation, acidification and terrestrial eutrophication. These differences can be explained by two main advantages of the HBP technology: (1) the HBP has higher energy and exergy efficiencies, and therefore less biomass is needed for producing the same amount of energy and exergy as outputs and (2) HBP avoids the external combustion of biomass occurring in the ORCs, and therefore releases less particulate emissions (the particulate matter impact of the ORC technology is for 97% caused by direct emissions of particulates). Although HBP technology has the same thermal efficiency as the ORC technology, its electric efficiency is three times higher. On the other hand, the HBP technology shows higher water depletion (+38%) and resource depletion (+79%). For water depletion, the high impact is caused by the replacements of the SOFC stack. For depletion of resources (minerals, fossil, and renewables), the main cause can be found in the production of the SOFC stacks. All these components are not present in the case of an ORC.

Similar results were obtained when comparing electricity production from HBP and ORC. The HBP shows lower impacts for climate change (−45%), particulate matter (−96%), photochemical ozone formation (−91%), acidification (−88%) and terrestrial eutrophication (−96%). On the other hand, the HBP technology shows an increased impact in terms of water (+31%) and depletion of resources (MFRD) +70%.

3.2.2. Comparison with Conventional Production of Heat and Electricity

Figures 7 and 8 show the comparison between the HBP technology operating with the three investigated biomass fuels and conventional separate productions of heat and electricity.

**Figure 7.** Comparison of HBP technology with competing technologies (for 1 MJ heat). Values are normalized taking as 100% the impacts of the natural gas boiler. CC = Climate change, PM = Particulate matter, POF = Photochemical ozone formation, AC = Acidification, TE = Terrestrial eutrophication, WRD = Water resource depletion, MFRD = Mineral, fossil, and renewable resource depletion. WC = wood chips, WP = wood pellets, MP = *Miscanthus* pellets.

The heat co-generated by the HBP technology shows a lower environmental impact compared to the heat produced by a condensing boiler burning natural gas. Even considering the least preferred fuel scenario in each impact category, the impact differences are at least −94% in terms of climate change, −70% in photochemical ozone formation, −37% in acidification, −43% in terrestrial eutrophication and −22% in depletion of resources. In particular, the significant difference in climate change is mainly generated by the biogenic carbon dioxide emissions (which are assumed to be carbon neutral) instead of fossil ones.

On the other hand, the HBP technology causes +28% impacts in particulate matter in the WP scenario (caused by the high particulate matter released when producing wood pellets) and significantly higher water depletion for the MP scenario (+13,000%), due to the water used for irrigation in the cultivation of *Miscanthus* (the only scenario with irrigation). When wood chips are fed instead of pellets, the HBP shows a much lower impact in terms of particulate matter (−59%) but still a relatively higher impact in water depletion (+19%), due to indirect water consumption in different life cycle activities.

**Figure 8.** Comparison of HBP technology with competing technologies (for 1 kWh electricity). Values are normalized taking as 100% the impacts of the German electricity mix. CC = Climate change, PM = Particulate matter, POF = Photochemical ozone formation, AC = Acidification, TE = Terrestrial eutrophication, WRD = Water resource depletion, MFRD = Mineral, fossil, and renewable resource depletion. WC = wood chips, WP = wood pellets, MP = *Miscanthus* pellets.

The electricity co-generated by HBP technology shows a lower environmental impact compared to electricity produced by the German electricity mix (EMIX). In particular, even considering the worst fuel scenario, the differences of impact are at least −86% for climate change, −43% for photochemical ozone formation, −56% for acidification and −63% for terrestrial eutrophication. Nevertheless, the HBP using wood pellets as fuel can lead to an increase in particulate matter (+7%). The HBP using *Miscanthus pellets* has much higher water resource depletion (+146%; caused by irrigation of *Miscanthus*) than the electricity from the grid mix. When operating with wood chips, the HBP shows a much lower impact than the EMIX, leading for example to −66% impacts in particulate matter, −98% in water depletion and −54% in depletion of resources (MFRD).

#### 3.2.3. Comparing with Other LCAs of SOFC CHPs

In the literature, 8 LCAs of SOFC CHPs have been conducted along with a review of LCAs on SOFC systems. In most of these LCAs, the fuels used in the SOFC CHPs assessed were natural gas and biogas and the capacity of the SOFC was only a few kilowatts (1–20 kW) of electricity. LCAs on SOFC CHPs of larger capacity (comparable to the one of the HBP) were conducted by [50–52]. Our results for the climate change impact of the HBP technology (0.03–0.09 kgCO2eq/kWhel depending on the fuel considered) indicate considerably lower impacts than for the SOFC CHPs assessed by these LCAs.

These lower impacts are especially found for the SOFC CHPs using natural gas as fuel because of the avoidance of direct emissions of fossil CO2 allowed by the HBP which is fueled with a biofuel instead of fossil fuel. In particular, among the LCAs of SOFC CHPs whose size is comparable to the HBP and operating with natural gas, Strazza et al. [50] assessed a 230 kWel SOFC CHP with electric efficiency of 53.4%. The resulting impact was 0.47 kgCO2eq per kWh of electricity, which is at least 5 times higher than for the HBP. An older study [52] assessing a 125 kWel SOFC CHP operating with natural gas, calculated an impact of 0.9–1.0 kgCO2eq per kWh of electricity, which is at least 10 times higher than for the HBP. Staffell et al. [44] assessed a 1 kWel micro-SOFC CHP fueled with natural gas and calculated an impact of 0.32–0.37 kgCO2eq/kWhel, which is a least 3–4 times higher than for the HBP.

The impact of the HBP is also at least 44% lower than for SOFC CHPs operating with biogas. In this case, the main reason can be found in the different fuel production processes and composition of the fuel used and consequent different composition of the direct emissions (e.g., methane emissions released with the exhaust gases). In particular, Strazza et al. [50] calculated an impact of 0.16 kgCO2eq for a 230 kWel SOFC CHP with 52.2% electric efficiency operating with biogas from sewage sludge. For the same type of system but with a capacity of 125 kWel, Sadhukhan [51] calculated an impact of about 0.19 kgCO2eq per kWhel. Concerning this last figure, Sandhukhan used a different functional unit (1 ton of sewage sludge processed through anaerobic digestion) and we derived it by applying exergy allocation on the energy outputs.

Similarly to the HBP, for multiple impact categories, Strazza et al. found that the impact of this system, independently on the fuel considered (natural gas or biogas) was dominated by the production of the fuel. The only exception was the climate change impact of the operation with natural gas, whose impact was mainly caused by the operation phase (mainly direct emissions of fossil CO2 of the system).

#### *3.3. Alternative Methods for Solving Multifunctionality*

Exergy allocation was used to partition the total environmental impact between heat and electricity, as explained in Section 2.2.1. In the literature, the two most applied alternative approaches to address the multifunctionality of SOFC CHPs are system expansion (enlargement) and economic allocation [20]. The first approach was only applied in studies where it was not necessary to differentiate between the impacts of heat and those of electricity.

Although the substitution method has a clear link with consequential analyses, it has been often applied in the literature for attributional LCAs with goals similar to the one of this study [53–55]. By the substitution method, the impact of the main product is obtained by subtracting the impact of the marginally avoided secondary products from the impacts of the overall system [21,56]. In particular, the main product is defined as the one providing the highest share of revenues within the analyzed product system (physical/economic significance) [21].

A sensitivity analysis was performed to understand the influence of the method on the results of the study. This analysis explored the variation of the results when applying economic allocation and the substitution method for the WC scenario, ORC scenario, and separate productions of heat and electricity.

When applying substitution, the first step is identifying the main product. Based on the economic heat/electricity ratio (see Table 2), electricity is the main product for the HBP. The production of heat by the HBP technology can marginally avoid the production of heat from natural gas on the market (Heat, central or small-scale, natural gas {CH}|heat production, natural gas, at boiler condensing modulating < 100 kW|Conseq from ecoinvent 3.4).

On the other hand, the HBP heat could also avoid the production of heat by an average biomass boiler (Heat, district, or industrial, other than natural gas {CH}|heat production, softwood chips from the forest, at furnace 300 kW|APOS from ecoinvent 3.4). The choice of a biomass boiler as substituted technology can be considered as an "alternative activity allocation" i.e., a form of "proxy-based disaggregation" [21]. This type of allocation is performed through the subtraction of impacts but differs from the substitution performed in consequential LCAs because it is not based on modeling of marginality [21]. Instead, this allocation takes as substituted processes the ones providing "primary productions of identical products and not of products that fall under different categories" [21]. This approach might, therefore, be an option also in attributional LCAs when reflecting the underlying physical relationship between the main and subsidiary products [21]. This sensitivity analysis considered both approaches, the substitution of a marginal activity (heat from a natural gas

condensing boiler) and the substitution of an alternative activity (heat from a biomass boiler, marked as (a) in Figures 9 and 10).

**Figure 9.** Sensitivity on allocation method for the generation of 1 MJ with HBP technology and competing technologies. Boiler running with natural gas taken as 100%. ORC = Organic Rankine Cycle, NG = Natural gas boiler, (a) = substitution of heat from a biomass boiler.

Based on the economic heat/electricity ratio of the ORC technology (see Table 2), heat is the main product for the ORC technology. In the case of the ORC, the electricity produced from the ORC avoids the production of marginal electricity from the electricity mix (this process is represented in the model by the ecoinvent dataset Electricity, high voltage {DE}|market for|Conseq).

The sensitivity analysis (see Figures 9 and 10) indicated that, compared to exergy allocation, the economic allocation method apportions more impacts on heat (+70% in every category) while it decreases by 23% the impacts of electricity. The same applies to the ORC technology (+40% and −36% respectively for heat and electricity). For CHPs, it is therefore important to show the impacts for both heat and electricity when an allocation method is applied, so that a full picture of its environmental impacts is provided.

**Figure 10.** Sensitivity on allocation method for the generation of 1 kWh with HBP technology and competing technologies (1 kWh electricity). Electricity mix taken as 100%. ORC = Organic Rankine Cycle, EMIX = Electricity mix, (a) = substitution of heat from a biomass boiler.

On the other hand, the conclusions of the comparative assessment did not change when applying exergy or economic allocation methods. This was true for all three comparisons: (1) between HBP with wood chips and ORC with wood chips, (2) between the three different biomass fuels scenarios and (3) between the HBP and the separate productions. For instance, the impact of the HBP per MJ of heat with both allocation methods was lower than for ORC in climate change, particulate matter, photochemical ozone formation, acidification, and terrestrial eutrophication, but it was higher in the two categories concerning the depletion of resources (see Figure 9). On the other hand, the percentages of potential environmental impact savings or intensifications compared to separate production can change significantly. For example, for climate change, the savings of impact of the HBP compared to ORC was 42% when using exergy allocation while it was decreased to 30% with economic allocation. However, for particulate matter, there was no difference.

Concerning substitution (see Figures 9 and 10), the variations compared to other allocation methods were small or large depending on the impact category considered and the type of substitution applied. Moreover, both types of substitution approaches and the alternative activity method led to negative results in some impact categories. This last aspect highlights that the modeling was not consistent with the attributional goal of the study, which is not aimed at assessing a change in demand, and therefore, it should provide negative emissions for a single product of a multifunctional process whose overall impact is positive [21]. When a physically/economically significant product (the substituted function was 42% of total revenues for HBP and 34% for ORC) is substituted in attributional LCAs (by assuming that its impact corresponds to the one that would be replaced in the market), the results are often not aligned with other allocation methods and contrasts with the attributional aim of the LCA. This aspect emerges clearly when multiple impact categories are investigated in the same LCA study resulting in conclusions in contrast with other allocation methods and of difficult interpretation.

## *3.4. Sensitivity Analysis on Potentially Sensitive Parameters*

#### 3.4.1. Internal Parameters

The results of the analysis indicated that the production of the biomass fuel (23–78% for the baseline scenario WC, depending on the category) and the SOFC stacks (10–43%) have a high contribution to the total impacts.

The ecoinvent dataset used for wood chips included both wood chips obtained as by-products of sawmill activities (15%) and from forest management (85%). To reduce the environmental impact, a scenario with only sawmill wood chips as fuel could be used. This type of wood chips presents a lower impact compared to wood chips from forest management because an important percentage of the impact of the upstream activities occurring in the forests is allocated to the main products of the sawmills. This scenario was assessed by sensitivity analysis to estimate the potential variation in the impact of the HBP (see the second column of Table 5). By using only sawmill wood chips, the environmental impact of HBP technology can be significantly reduced (indicatively by 10–40%).

**Table 5.** Sensitivity analysis on the reduction of the environmental impact of the HBP technology by either increasing the SOFC stack lifetime or using only wood chips produced as industrial by-products.


Alternatively, the impact of HBP technology could be improved by acting on the SOFC stack. Since the stack needs to be replaced every five years, the environmental impact could be improved by increasing the SOFC stack lifetime and therefore reducing the number of replacements over the plant lifetime. The second column of Table 5 shows the reduction of the environmental impact that could potentially be achieved by increasing the lifetime of the SOFC from five to seven years. This would lead to a decrease between 2% and 10% of the impacts of the wood chips scenario (baseline).

#### 3.4.2. External Parameters

Since the technology will be deployed after 2025, it is important to explore how the comparative evaluation will change taking into account the current trends of decarbonization, which should lead to a decrease in the share of coal-produced electricity by shifting to renewables. In particular, the expected decarbonization of the European electricity grid will diminish the environmental benefits of HBP technology.

To assess this variation, the electricity mix based on two future scenarios for 2030 were considered: the EU reference scenario for Germany [57] and the IEA current policy scenario [58]. Due to the unavailability of the IEA current policy scenario for Germany, the IEA average mix of 2030 for the EU was taken as a proxy. This second scenario represents a more decarbonized electricity sector and includes other countries where the HBP could be commercialized. In particular, the IEA current policy scenario has only 13.7% coal and 44.7% renewables. The future savings of environmental impact allowed by the HBP is shown in the two columns on the right in Table 6. The environmental savings from the HBP technology will be only slightly affected (order of 5% overall) by the change expected in the electricity mix for 2030.


**Table 6.** Sensitivity analysis on the savings of environmental impacts of 1 kWh of electricity produced by the HBP compared to the grid electricity mix (EMIX).

## **4. Conclusions**

This article presented the first life cycle assessment (LCA) of a novel technology integrating biomass gasification and SOFC technologies. This technology is currently under development in the H2020 HiEff-BioPower (HBP) project and allows for the use of various biomass types as feedstock. This LCA assessed the environmental impacts when operating the technology with three different fuels: wood chips, wood pellets, and *Miscanthus* pellets. The impact of producing heat and electricity with the HBP technology was compared to the state of the art competing technologies. The results showed that most of the impacts of producing heat and electricity with the HBP technology are generated during the production (including transportation) of the biomass fuels (between 23% and 99% of the total impacts depending on the category and the fuel). The use of wood chips as fuel generates much lower impacts per functional unit than the operation with wood pellets (11–70% lower) and *Miscanthus* pellets (9–99% lower), in all impact categories. The next highest contributor to the life cycle environmental impacts is the SOFC stack, due to both the high energy intensity (especially in electricity consumption) and material intensities of its manufacturing processes, and its short lifetime (the stack should be replaced every 5 years). Beyond increasing the fuel efficiency of the technology and therefore reducing the consumption of biomass fuels, the main recommendation to technology developers would be to increase the lifetime of the SOFC stack. Increasing the SOFC stack lifetime could decrease the environmental impacts of 2–10%, depending on the category.

The comparison of the HBP technology with separate productions of heat and electricity (from natural gas condensing boilers and the German electricity grid) indicated significantly lower impacts for the HBP technology, especially in climate change (86%/94% lower), photochemical ozone formation (−43%/−70%), acidification (−37%/−56%) and terrestrial eutrophication (−43%/−63%). Overall, HBP showed also better performance than ORCs, as they have higher exergy efficiencies and almost zero particulate emissions resulting in 86–96% lower impact in the category particulate matter.

The sensitivity analysis on the allocation method for heat and electricity provided useful insights for the choice of allocation methods in CHP plants, and led to the following recommendations: (1) the attributional LCAs of CHPs should always provide the results for both heat and electricity to allow for better interpretation of results, independently of the allocation method, (2) LCA results from different CHP plants should not be compared if they assumed different allocation approaches and (3) substitution is not recommended in attributional LCA (especially if the substituted product is

not a minor by-product) because it provides results which are not in line with the attributional aim (e.g., negative emissions) and lead to conclusions in contracts with the ones from applying allocation methods which are proven to be a good proxy of physical causality for CHPs and therefore preferable.

**Author Contributions:** Conceptualization, C.M. and B.C.; data curation, C.M., V.R., T.G., T.B. and I.O.; formal analysis, C.M., B.C. and V.R.; funding acquisition, M.J. and L.S.; investigation, C.M., B.C., V.R. and L.S.; Methodology, C.M., B.C., M.J. and L.S.; project administration, B.C., M.J., T.B., I.O., and L.S.; resources, B.C.; software, C.M. and V.R.; supervision, B.C., M.J. and L.S.; validation, B.C., T.G., M.J., T.B., I.O. and L.S.; visualization, B.C., M.J., and L.S.; writing—original draft, C.M. and V.R.; writing—review and editing, B.C., T.G., M.J., T.B., I.O. and L.S. All authors have read and agreed to the published version of the manuscript.

**Funding:** This LCA study has been developed within the HiEff-BioPower project, which received funding by the European Union Horizon2020 Programme under Grant agreement number 727330.

**Acknowledgments:** The authors express their gratitude to the other project partners and all the members of the consortium.

**Conflicts of Interest:** The authors declare no conflict of interest.
