Do Miscanthus lutarioriparius-Based Oriented Strand Boards Provide Environmentally Benign Alternatives? An LCA Case Study of Lake Dongting District in China

Abstract

Miscanthus lutarioriparius (M. lutarioriparius) in Lake Dongting District are in the situation of being discarded due to the government’s environmental policy, the decomposition of which will bring another pollution risk. The purpose of this study is to environmentally analyze the production of M. lutarioriparius-based oriented strand particleboards (M.OSB) as alternatives to the conventional artificial boards. The production systems were evaluated from a cradle-to-gate perspective using the Life Cycle Assessment (LCA) methodology. Our results showed that the M.OSB had an overall better profile than wood panels, identifying the production of starch adhesives and bio-fuels as the main environmental hotspots. It was also found that annual harvesting and utilization of M. lutarioriparius could ease the burden to the environment during the decomposition of this plant, and further improve the environmental performance of M.OSB. Sensitivity analyses were conducted on the key parameters, suggesting that there are opportunities for improvement. This study provides useful information for enterprises and policymakers on where to focus their activities, with the aim of making the future of M. lutarioriparius utilization more technically and environmentally favourable.

1. Introduction

Wooden panels are composite materials with modifiable properties, that are made from processed wood (or other biomass) and synthetic adhesives [1]. The most common artificial boards are particle board, oriented strand board (OSB), fiberboard, and veneer products including plywood and laminated veneer lumber [2]. China is the largest manufacturer (159,746,000 m³/year) and supplier (9,956,625 m³/year) of wooden panels in the world [3]. Particleboard output accounted for 20.8% of Chinese artificial board production in 2020 [3]. As China’s urbanization process continues, there is a favorable market perspective for OSB, but it also faces the pressure of reducing environmental burden and harm to human health. There is an urgent need to conduct studies on improving the environmental performance of OSB production. One effective approach is to use agricultural residues or other non-wooden biomass as raw materials [4].

Miscanthuslutarioriparius (M. lutarioriparius) is an endemic therophyte in China. It is distributed predominantly in the Yangtze River basin, and was widely used in the Chinese papermaking industry [5]. A large proportion of M. lutarioriparius originatedfrom 1.2 × 10⁵ ha of fields in the wetland of Lake Dongting, which is the second-largest freshwater lake in China and which plays an important role in the ecosystem of the Yangtze River basin [6]. In the past decades, these plants were reaped and transported to the paper mills around the lake. To prevent this industry from ruining the lake’s ecosystem, these factories have been shut down according to the government’s direction in the Plan for Ecological Environment Improvement of Lake Dongting(2018–2020) [7]. It was implemented in 2019 and left a great mass of M. lutarioriparius detained in the lake, which might cause another environmental problem. Currently, there is a debate in China over whether to abandon this plentiful plant, which means to let it grow without outside interference, or to use it in a more sustainable way than paper manufacturing.

There are a few scientists focused on evaluating the environmental impact of Miscanthus-related industries in LCA approaches. Meyer F. et al. performed comparative LCA research between three different conversion techniques for miscanthus biomass in Europe [8]. Kiesel A. et al. assessed and compared the environmental performance of the biogas production and utilization of Miscanthus, switchgrass, and maize in Germany by the ReCiPe method [9]. Hastings A. et al. considered the economic and environmental effects of introducing Miscanthus hybrids as an alternative to Miscanthus × giganteus in the UK [10]. For the potential applications, biomass energy has been a hot topic ever since the European government began to encourage developments in this field [11]. To the best of our knowledge, there is no published research about the environmental performance of OSB produced from M. lutarioriparius. Most of the related LCA studies are focused on the sustainability of wood-based particleboard [2], and there is also research on the environmental impacts of particleboard production using sugarcane bagasse [12] and coconut residues [13].

M. lutarioriparius is considered as an environmentally and economically beneficial way to produce particleboards in China [14]. The content of cellulose, hemicellulose, and lignin reached 39.07%, 8.90%, and 12.48% at the late growth stage of M. lutarioriparius, which was higher than the other three kinds of agricultural residues (rice-straw, cornstalk, and wheat-straw) [15]. It, therefore, has better physical properties as a base material for particleboards [16].

To expand the scope of this topic, we carried out a comparative LCA of industrial- scale M. lutarioriparius OSB (M.OSB) production, which also took the plant abandonment scenario into account. The findings of this paper are a supplement to previous studies on non-wood particleboard. In conventional processes, two types of chemicals are usually utilized: thermosetting resin, such as urea-formaldehyde resin, acting as a binder; and paraffin emulsion, which improves the hygroscopic properties of the panel. A particular modified starch-based adhesive was added in M.OSB, which was free of paraffin. The purpose of this study is to explore the environmental burden of M.OSB production and decomposition of M. lutarioriparius in water without annual harvest. The conclusion of the study will provide evidence for decision-makers to formulate the development strategy of relevant industries in the future.

2. Methods

The LCA method follows the framework of the ISO14040 [17] and ISO14044 [18] to evaluate the environmental impact of a product, process, activity, or service.

2.1. Goal and Scope

2.1.1. Objectives

The purpose of this study is to evaluate and compare the LCA performance of four categories: M.OSB, middle-density particle board (MDP), conventional OSB, and middle-density fiberboard (MDF). The differences between them are reflected mainly in the type of raw materials and sources of energy. To that end, the most important technical differences will be described in the next paragraphs, bearing in mind that this study is focused on common technology used in China.

2.1.2. Functional Unit

The functional unit is 1 m³ of uncoated M.OSB with a nominal thickness of 15 mm and an average density of 650 kg/m³. M.OSB is mainly utilized for packaging, furniture manufacturing, building formwork, and house construction. The physical performance comparison between M.OSB (tested by Hunan Product and Commodity Quality Supervision and Inspection Institute) and OSB is shown in

Table 1. Physical performance comparison between M.OSB and OSB.

Performance	Unit	M.OSB t = 15 mm	OSB 10 mm ≤ t ≤ 18 mm
Bending strength—major axis	MPa	22	20
Bending strength—minor axis	MPa	12	10
Modulus of elasticity in bending—major axis	MPa	4000	3500
Modulus of elasticity in bending—minor axis	MPa	1500	1400
Internal bond	MPa	0.35	0.32
Swelling in thickness 24 h immersion	%	15	15

, and it refers to the requirements of OSB in LYT 1580-2010 (Chinese Industry Standard of Oriented Strand Board).

M.OSB is composed of a modified starch-based adhesive matrix and a reinforcing phase of particles, which is solidified by hot pressing. This product is free of urea-formaldehyde resin and polydiphenylmethane diisocyanate adhesives. Instead, the latest biosynthetic adhesive technology has been adopted. Based on the design of the molecular structure, the adhesive forms a large crosslinked molecular network during hot pressing, which is integrated with the particles. This new type of OSB not only has a high bonding strength and is formaldehyde free, but also saves 50–58% of economic cost. The main components of the modified starch-based adhesive are corn starch, polyvinyl alcohol, and water. As the formulation is a non-patented technology, this is not an exhaustive list, and components of less than 1% mass are not listed. As the conventional crushing machines show no directivity and destroy the effective strength of primary materials, M.OSB manufacturers have developed equipment which crushes along the fiber in the same direction, maximizing the retention of shear strength. The technology of vertical intersecting directional multi-layer laying is used to preserve the physical strength of the substrate itself.

2.2. System Boundary

Figure 1 shows the system boundary of the M.OSB life cycle from cradle to gate. It covers the life cycle of M.OSB production, comparing it with MDP, OSB, and MDF in China, utilizing attributive LCA methods. The boundary of this system, which begins at the fields (cradle) and ends with fiberboard production (product gate), consists of two subsystems: agricultural and industrial. The agricultural boundary usually includes ditching, sowing, fertilization, insect control, and seeding [19]. However, for this study, we only consider the diesel fuel consumption of harvesting and feedstock transportation. The reasons for this are: (1) ditching is required every five years, and sowing at intervals of 15 years; (2) only a small quantity of fertilizer is used, because the fields, located in the wetland and tidal flats around Lake Dongting, are fertilized annually by seasonal flooding; and (3) the doses of herbicides and pesticides needed are small or even none, and as of 2017 should be in accordance with the environmental policy for Lake Dongting. For comparability between the plates, all the system boundaries were unified and carbon sequestration during the plants’ growth stage was not considered. For the biomass fuels used in the combustion phase, the generated CO₂ is approximately equal to the fixed. Consequently, the net CO₂ emission in this stage can be ignored. If fossil fuels were used instead of bio-fuels, the produced CO₂ is a net emission with no offset [20]. For these reasons, the actual CO₂ emission of production utilizing biomass fuel should be less than our result.

The main steps of the industrial system include preparation of the primary material, drying, adhesive production, sizing, forming, and finishing the board. The material preparation phase involves stockpiling the raw materials and screening them to remove impurities such as leaves, joints, and dust. In the drying phase, M. lutarioriparius are chipped and then dried to meet the required water content, after which particles are produced and screening is repeated. The production process of the adhesive involves a non-patented technology, and as such will not be elaborated on here. During the sizing phase, bio-adhesives are prepared in advance, with particles and glue being processed automatically until the binder is kept evenly on the surface by mutual friction in the glue mixer. The forming phase involves forming the sized particles into slabs, which are preloaded to reduce the thickness and increase the tensile strength. The qualified slabs go on to be shaped and repaired, with unqualified slabs being recycled by repeating this process. The final step is board finishing, in which the formed boards are stacked for two days after unloading, drying, and cutting. These steps are further divided into four stages: material preparation (I), adhesive production (II), particleboard production (III), and heat generation (IV). In addition, there are a few more auxiliary processes (V), including field cleaning and internal transportation.

The process flow has been designed with the aim of cleaner production. The fuel used in the heat generation is partly recycled from the waste generated during the screening and cutting steps. The solid waste, including combustion remains, dust on the ground, and dust collected by cyclone dust collectors, is collected by the local fertilizer company. In addition, all wastewater is recycled during the manufacturing process of the adhesive.

2.3. Data Source and Assumptions

The inventory data of M.OSB (agricultural and industrial production) were obtained from the Hanchuang Particleboard Manufacturing Co. Ltd. in Anxiang County, Changde, China. The average value of M.OSB production was retrieved from the field measurement data and internal documents of a third-party institution (Jinke Testing Institution, Changsha, China). Data regarding the production of polyvinyl alcohol (PVA) was collected from a chemical plant (Ningdong Energy Chemical Investment Co. Ltd.) in Ningxia Hui Autonomous Region of China. The inventory of biomass fuel follows the production profile of a biomass solid fuel plant, which is only 2 km away from the M.OSB production field. The CO₂ emission during biomass fuel combustion was calculated according to IPCC 2006. The inventory data of transportation were derived from the literature [21]. The remaining background data were obtained from various databases in SimaPro 9.0, such as ELCD and Ecoinvent.

In previous studies on wood boards, it has been proven that the environmental impact of producing the capital assets is much less important compared to their operational stage [22,23]. Capital assets such as industrial machinery and infrastructure, as well as human activities such as daily consumption of workers and domestic sewage discharge, are not included within the system boundaries.

3. Results

SimaPro was employed for the LCA modeling in this study, as it is one of the most popular tools in research on environmental impacts [24]. The CML method is one of the most common approaches in studies on the non-comparative LCA of wood boards [25], however, the ReCiPe 2016 [26] and USETox [19] were also frequently used in comparative studies. Compared to the CML approach, ReCiPe 2016 provides the possibility that characterization factors are representative on a global scale, rather than Europe. It also preserves the possibilitythat characterization factors can be implemented on a national and continental scale by including a number of impact categories [27]. ReCiPe 2016 provides coherent characterization factors at both the midpoint and endpoint levels. Consistency in the development of the midpoint and endpoint models is enhanced by the use of the same time-range for each cultural perspective across different influence categories. Usingthe ReCiPe endpoint method can make the results of different schemes more comparable. Many studies used a combination of CML and USETox [28], but USEtox doesn’t provide characterization factors for terrestrial or marine toxicity [26]. There are also toxicity impact categories in the ReCiPe midpoint method. Based on the above reasons, the results were calculated by ReCiPe 2016 without USEtox as the supplement.

Two environmental impact assessment methods of ReCiPe 2016 were considered in this study. Firstly, we employed the ReCiPe 2016 Midpoint(H) method to select characteristic factors, and reported environmental profiles in terms of the following impact categories: global warming (GW), stratospheric ozone depletion (SOD), terrestrial acidification (TA), terrestrial ecotoxicity (TE), freshwater ecotoxicity (FE), marine ecotoxicity (ME), human carcinogenic toxicity (HCT), human non-carcinogenic toxicity (HNT), and fossil resource scarcity (FRS). Secondly, we used the ReCiPe 2016 Endpoint(H) Method to compare the environmental impacts of M.OSB and other contrasting plates. This endpoint approach may help decision-makers carry out selections by turning multiple types of impacts into a single indicator [29].

Figure 2 shows the M.OSB’s relative contribution to each phase, evaluated by ReCiPe Midpoint(H). Detailed discussion on these results is provided in Section 3.1, Section 3.2, Section 3.3, Section 3.4, Section 3.5, Section 3.6, Section 3.7, Section 3.8 and Section 3.9.

3.1. Global Warming

With regards to global warming, heat generation, adhesive, and M.OSB production are the priorities (48.1%, 32.5%, and 12.1%), while the plant harvest andtransport, and the preparation stages, had a low level of influence (3.9% and 2.4%). The most relevant GW impact was associated with the heat generation, mainly due to the combustion of biomass fuel. Having only a short distance to transport materials contributed greatly to reducing CO₂ emission and conserving fossil energy, a detailed analysis of which is in Section 4.4.

3.2. Stratospheric Ozone Depletion

Production of the adhesive was responsible for 61.6% of the influence on SOD, followed by 33.3% from the heat generation. With the exception of the MDP production process (4.2%), the impacts of the remaining processes are negligible (≤0.5%). As for the adhesive production stage, the majority of the impact was related to maize starch production (94% of this stage).

3.3. Terrestrial Acidification

Most terrestrial acidification (TA) effects are generated by heat generation (48.9%). Compared to the harvesting and transportation stages of the raw material, approximately 95% of total TA was due to industrial production. In the heat generation phase, the majority of this influence was caused by the combustion of biomass solid (97% of this phase) and the resulting emission of air-borne sulfur derivatives. The influence of manufacturing the adhesive and the MDP production stages accounted for 25.2% and 17.7%, with the remaining stages accounting for less than 2%.

3.4. Terrestrial Ecotoxicity

The adhesive production stage was responsible for 55.1% of the influence on terrestrial ecotoxicity (TE), with the secondary source being the heat generation. Less than 1% of the total impact came from plant harvest and transportation, with the remainder coming from the MDP industrial subsystem. For adhesive production, the main sources of TE were maize starch and PVA production, which produced 58% and 41% of the impact at this phase. For the heat generation, biomass fuel production accounted for almost all of the impact (97% of this stage).

3.5. Freshwater Ecotoxicity

Of the entire freshwater ecotoxicity (FE) impact, 93.7% was related to manufacturing the adhesive, MDP production, and the thermal energy center (27.1%, 18.4%, and 48.2%, respectively). The starch-based adhesive utilized in this system was superior to other biomass matrix adhesives found in the FE index [27], and as a main source, biomass fuel production contributed 96% to the heating phase.

3.6. Marine Ecotoxicity

As for the marine ecotoxicity (ME) category, the heat generation, adhesive production, and MDP production subsystems were very relevant (47%, 29.1%, and 17.7%, respectively), mainly because of the release of copper and zinc into the water. According to the normalized results, ME was the most relevant category for potential impacts on the whole system. For the heat generation subsystem, nearly 96% of the impacts were caused by fuel production, due to both starch application and electric power consumption (16.6% and 83%, respectively, in biomass fuel production). For the adhesive subsystem, approximately 98% of the impacts were attributed to the maize starch and PVA production in the industrial plant.

3.7. Human Carcinogenic Toxicity

The influence of human toxicity is due to the release of pollutants, potentially toxic to human health, by human activities [29]. Figure 2 shows the M.OSB production file with the largest human carcinogenic toxicity (HCT) impact (99.7%) occurring in industrial processes, most of this being due to the manufacturing of adhesives (43.1%), heat generation (37.8%), and MDP manufacturing (12.1%). For this category, the manufacturing of the adhesives was the main contributor, especially due to the heavy metals (Cr and Ni) in the production of corn starch and PVA. The major contributor to heat generation was biomass fuel production. In other processes, the impact was mainly from electricity consumption.

3.8. Human Non-Carcinogenic Toxicity

Compared with HCT, the human non-carcinogenic toxicity (HNT) contribution of the adhesive preparation (54%) was 9% higher. The contribution was lower in other production stages, specifically in the thermal energy center (35.7%) and manufacture of the MDP (7.5%). The corn starch contributes 83% of the effect of the binder phase, which was mostly due to heavy metal emissions in corn starch production (Zn, Cd, and Pb).

3.9. Fossil Resource Scarcity

Fossil resource scarcity (HRS), determined as fossil fuel potential (FFP in kg oil-eq), is defined as the ratio between the higher calorific value of the fossil resource and the energy content of crude oil [30]. According to our results, 94.1% of the impact on HRS was caused by the industrial subsystem. Approximately half of this category was due to the adhesive production phase, sourced from ethylene (35.9% of this stage) and the heat from steam in the chemical industry (53.9% of this stage). For the heat generation phase, the total impact (29.8%) was mainly related to the electricity consumption during biomass fuel manufacturing and heat generation. MDP production subsystems were responsible for 10% of the HRS impacts.

4. Discussion

4.1. Sensitivity Analysis and Improvement Opportunities

The actions to reduce the environmental burden focused on reducing inefficiencies, upgrading the composition, and changing the thermal plant technology. In order to find the improvement opportunities, a sensitivity analysis was conducted in the ReCiPe midpoint method. Figure 3 shows the result of sensitivity analysis in the form of the impacts deduction (%) in different scenarios. Six possible situations were considered: S0—original scenario; S1—reduce electricity consumption by 10%; S2—reduce starch consumption by 10%; S3—reduce PVA consumption by 10%; S4—reduce bio-fuel consumption by 10%; S5—reduce adhesive consumption by 10%.

In general, reducing electric power consumption did not make significant improvements for all the nine impact categories. The most prominent categories in S1 were ME and FE. There is a possibility of reduction of electricity consumption from the grid, that is, implementation of a cogeneration process, such as biomass integrated gasification combined cycle (BIGCC) system. In the BIGCC system, combustible biomass is vaporized to produce gas that is used to drive turbines, generate electricity, and process heat. According to Silva et al.’s analysis, about 90.0% of particulate matter and 50.0% of NO_x can be eliminated during gasification without carbon monoxide release [12].

The starch for the adhesives production was one of the main hotspots, especially in terms of SOD impacts according to S2. Another hotspot was PVA production. Results of S3 showed that PVA consumption reduction would bring environmental benefits, especially in order to prevent GW and FRS. Approaches to reduce the environmental impact of starch and PVA production may include: (1) minimizing their proportion in formulations; and (2) seeking out suppliers employing environmentally friendly alternatives, such as biomass fuels or other methods generating less toxic emissions than heat generated in chemical plants.

S4 contributed the greatest decrease in GW, TA, ME, and FE. Reducing biomass pellet fuel consumption (100%) to 90% can result in a reduction of impacts by 4.71% for GW, 3.30% for SOD, 4.83% for TA, 3.20% for TE, 4.63% for FE, 4.52% for ME, 3.55% for HCT, 3.48% for HNT, and 2.89% for FRS. Although, when comparing biomass fuels and other energy sources, ecological sustainability seems to be in favor of the use of natural gas instead of biomass [22]. If we incorporate the CO₂ absorbed during plant growth into the system boundary, the net CO₂ emissions from bio-fuel combustion would actually reach close to zero. The ash contents of Miscanthus (1.42%) are the lowest among the common stockfeeds of bio-fuel [16], so the bio-fuel prepared with them would have better environmental performance.

Bio-adhesive generation was classified as a hotspot for the following reasons: starch and PVA, as basic ingredients of adhesives, contributed greatly to the global environmental impacts, and the water consumption in this stage occupied the largest proportion of total water depletion. S5 provided the greatest impact decrease in five categories, which was responsible for 6.14% less SOD, 5.51% less TE, 4.31% less HCT, 5.40% less HCT, and 5.12% less FRS in comparison to the baseline. In addition to reducing adhesive consumption, another improvement opportunity is to capture evaporated water from the drying process instead of tap water. The collected water tends to be warmer than tap water, which also reduces the energy consumption of the boiler during the glue-making process.

4.2. Uncertainty Analysis

The uncertainties in the LCA study, as in any other scientific investigation, should be thoroughly investigated before conclusions and recommendations are drawn [18]. Monte Carlo simulation, which was usually applied to estimate the uncertainties introduced in the LCI data, was operated by SimaPro with a 95% confidence interval, 5000 runs, and a stop factor of 0.005 in our research. Figure 4 depicts the uncertainty analysis for the global profiles of M.OSB. Results reveal that uncertainties were acceptable for all impact categories where the uncertainty was below 120% [31].

4.3. Comparative LCA of M.OSB, OSB, MDP, and MDF

4.3.1. Comparative LCA in ReCiPe Midpoint(H)

The comparative LCA profiles of M.OSB, OSB, MDP, and MDF are shown in Figure 5, with the same impact categories evaluated by ReCiPe Midpoint(H). To ensure the comparability of this analysis, we adopted the inventory of 1 m³ functional units of the other three products, rather than the results in the literature, and unified their system boundaries with M.OSB. As a reference, MDF is scaled to 100%, with the other values presented as relative scores. The inventories of OSB, MDP, and MDF were obtained from Fangwen Z. [32], Shilong X. [33], and Xuyao Z. [34], respectively.

It can be seen that M.OSB had the lowest environmental impact across three of the nine categories (GW, TA, and FRS), scoring second only to OSB in the remaining six categories. The impacts of M.OSB were no more than 20% higher than that of OSB in TE, FE, ME, and HCT, and twice that of SOD and HNT. Considering the normalization factor of SOD, the influence of both products on SOD was negligible. MDF had the highest environmental influence in seven of the nine categories, with MDP having the highest in the remaining category (SOD and TE). The results showed that the relative impacts of M.OSB resemble OSB in most of the evaluation categories, while MDP and MDF had a higher impact than both M.OSB and OSB for all categories evaluated. The similarity of these two environmental profiles may be due to similarities in the manufacturing processes of M.OSB and OSB, such as the use of continuous hot pressing and biomass fuel boilers. However, it is obvious that M.OSB had 27.6% less emissions in GW than OSB. This is mainly because of the higher requirement of biomass fuel and electricity in the production of OSB. Based on the noteworthy difference in energy consumption between the two, we further compared the production processes and found the following possible reasons:

• M. lutarioriparius is generally harvested from December to January, when the humidity of the yielded M. lutarioriparius ranges from approximately 40 to 60%; nevertheless, much of the yield logs used in OSB manufacturing are greater than 70%. The average humidity required in the next stage is 2.5% for M.OSB and 4.5% for OSB. Consequently, there is a significant difference in the energy required for adequate dehydration.
• M. lutarioriparius is initially cut into sections by a harvesting machine and does not need to be peeled like timber, resulting in power savings.
• Self-produced adhesives are used in M.OSB production, therefore, the transport distance and weight of feedstock are much shorter or lower than that of OSB, which uses outsourced PMDI as adhesives.
• One functional unit of M.OSB consumes 1010 kg of M. lutarioriparius, and the transportation distance is no more than 100 km for the developed transportation around the lake, while OSB requires 1994 kg of logs originated from remote mountains, so the distance is usually longer than that of M.OSB. Consequently, the transport from forest to factory, or between different processes during production, means that OSB requires more energy than M.OSB.

M.OSB had a double HNT impact of OSB, whose major contribution was heavy metal emissions from the production of corn starch. Furthermore, maize starch production was also reported as the hotspot for SOD, TE, and HCT, with these indicators showing a slightly higher value for M.OSB than OSB. The final priority was the energy production, which primarily contributed to FE, ME, and HCT due to the electricity consumption involved with the production of biomass fuel. Our investigation found that if a biomass pellet fuel boiler is used in the drying process of biomass fuel manufacturing, it will effectively reduce the consumption of electricity. As shown in Figure 1, this approach was not included in the system boundary of this study.

According to Figure 5, MDP and MDF showed a significantly greater impact than M.OSB and OSB for each indicator, which can be explained by the vast quantity of materials and energy required. MDP caused at least twice the environmental burden in comparison to M.OSB for each of the categories analyzed, with SOD and TE reaching over quadruple the amount compared to M.OSB, whose main contributors were wood pellets, electricity, and urea formaldehyde resin. The HCT of MDF was hundreds of times that of M.OSB, which was related to fuel wood and wood fuel. Wood burning and electricity consumption were the hotspots for MDF manufacturing [35].

The other three plates analyzed as part of this study were manufactured with petrochemical binders. Despite the cost of commercial biological adhesives being relatively high, it is sustainable to develop self-made renewable adhesives as alternatives to those sources from fossil fuels [27]. According to

Table 2. The inventory of 1 m³ M.OSB production.

Inputs		Outputs
Feedstock	-	M.MDP	1 m3
M. lutarioriparius	1010 kg	Emissions to air	-
Starch-based adhesive	52 kg	Nitrogen oxides	0.0426 kg
Lubricants	0.0003 kg	Sulfur dioxide	0.00712 kg
Water	17.85 kg	Carbon dioxide	68.47 kg
Biomass fuel	41.9 kg	Particulate matter	0.017 kg
Energy consumption	-	Emissions to water	-
Electricity	65.75126 kwh	CODcr	0.0038 kg
Diesel	1.34 kg	BOD5	0.00094 kg
Transportation	170.5 tkm	SS suspended solid	0.0028 kg
		NH3-N	0.00032 kg
		Other waste	-
		Waste oil	0.01364 kg
		Recyclable solid	303.36 kg

, the performance of M.OSB when adding biological adhesives is better than the standard of conventional products.

4.3.2. Comparative LCA in ReCiPe Endpoint(H)

To compare the environmental impact of the traditional and ecological plates, an endpoint-based analysis was proposed to give each product a single environmental score. We considered the normalization and weighting factors of the ReCiPe 2016 Endpoint(H) method. The analysis report at the endpoint level was based on three indicators which quantify the relative severity of damage: human health (HH), ecosystem quality (EQ), and resource scarcity (RS). These indexes were the results of aggregation of midpoint categories through specific endpoint characteristic factors. In this part, only the impact categories previously selected for the environmental assessment at the midpoint level were considered for the generation of the single environmental score.

The contribution profile of the comparison is illustrated in Figure 6, and the size of the circle corresponds to the magnitude of their global environmental impacts. The single scores for M.OSB (19.3 Pt), OSB (28.2 Pt), and MDF (36.8 Pt) were approximately in an arithmetic sequence with 9 Pt as the difference value, while the score for MDF (97.8 Pt) was much higher than the others. Such a significant difference was strongest regarding the impact on human health at the endpoint, which is mainly due to the air emissions from wood fuel. Even though M.OSB outperformed OSB only on three midpoint indicators (GW, TA, and FRS), the former was more environmentally friendly in endpoint analysis. This was due to normalized absolute values and the weight given by the ReCiPe endpoint method. Given the fact that M.OSB has price superiority (53.7% of OSB in Chinese market) and has equivalent physical properties compared to the other materials, it can be expected that M.OSB will be more ecologically and economically efficient than the other three.

For all these panels, the influence on human health took a dominant position, and the scores for ecosystems and resources were no more than 26% and 1%. Thus, the demand of fossil resources (such as coal, gas, and crude oil) is negligible. EQ ranked second in all board profiles, with the proportion of total scores ranging from 3.9% (MDF) to 26.4% (OSB). Regarding the former, the impact on EQ was mainly associated with greenhouse gas and acidification emissions from background processes such as CO₂, nitrogen oxides, and sulfur dioxide (43.7%, 16.5%, and 17.6% on total contributions to ecosystem damage, respectively). Although MDF had the smallest EQ ratio, its absolute value is second only to that of OSB, lying in the large total score as the denominator. The percentage of EQ in the total score of OSB was the highest among all products, which was linkedto ecosystem degradation caused by the utilization of cleft timber from forests.

4.4. Comparisons with Relative Studies

A direct comparison of our research with previous literature was complicated by differences in system boundaries, inventory integration, databases, and environmental assessment methods. Shang X. et al. [35] conducted a relative comparative LCA based on data from China. Both studies agreed that less GHG emissions and less non-renewable energy aregenerated and consumed in the production process of straw or M. lutarioriparius particleboard compared to that of conventional wood particleboard. This conclusion can be attributed to the similarity in physical properties of the two kinds of feedstock, namely: (1) lower moisture content compared to raw wood; (2) shorter transportation; and (3) less processing of raw materials required, for example, the omission of peeling (normally required in conventional wood particleboard manufacturing). It should be noted that LCA was carried out by eBalance software with the CLCD database in Xiaoyu S. et al.’s research. It is also worth noting that the inventories in their study onlyfocused on energy sources. Consequently, only global warming potential and the consumption of nonrenewable energy were used as impact assessment criteria in their LCIA. The optimization strategies (e.g., using granular fuel instead of sand fuel, or considering Miscanthus or straws as suitable alternatives to wood) suggested in their research have been adopted in the performance of ours. Together with the addition of biomass adhesives, these provided M.OSB with a significant environmental advantage over other types of conventional furnishing boards.

Silva D.A.L. et al. [12] analyzed the manufacturing of particle boards with the addition of sugarcane bagasse (PSB). Our study, although conducted using different methods, confirms some of the results they reported. Both results showed that the hotspots were fuel, electricity, and the adhesive supply chain. They also concluded that wood particleboard and PSB had the same trend in terms of relative impacts for most of the categories evaluated, which is similar to ours. With a gradually increased dose of sugarcane bagasse, the environmental performance of PSB progressively improved. According to both studies, it can be concluded that non-wood biomass is environmentally effective alternatives to wooden materials.

4.5. The Burden to the Environment in the Abandonment Scenario

M. lutarioriparius in this study grows in the mudflat area of Lake Dongting. During the non-rainy season, the fluctuation of water flow in this area is relatively slow, and the water exchange is small [36]. Without annual harvest, the decomposition of this plant, which usually happened in the non-rainy season, was a heavy burden to the water quality. The colleagues (Li Y. et al.) in our project carried out the in situ decomposition experiment of M. lutarioriparius, and the summary is listed in

Table 3. The summary of the in situ decomposition experiment.

	Leaf	Stem
Initial dry mass (g)	5	5
Dry mass after decomposition (g)	1.46	2.52
Leaf–stem ratio	0.1175:1	0.1175:1
Decomposition rate	70.8%	49.6%
Nitrogen content (mg/g)	4.15	1.4
Phosphorus content (mg/g)	0.48	0.14

. The dry biomass was collected from three typical plots in November and decomposed in five months.

The average humidity of M. lutarioriparius is 40%. Calculated with the humidity and the data in Table 3, the N, P, CODcr, and CO₂ emissions under the discarding scenario are theoretically 0.105 kg/t, 0.011 kg/t, 0.369 t/t, and 0.530 t/t, respectively. The environmental impact of this situation was calculated as 9.11 Pt/t in the ReCiPe Endpoint method. The environmental impact of M.OSB would decline to 10.10 Pt if the reduction of the environmental burden caused by harvesting were included in the system boundary. This means that M.OSB has a remarkable environmental advantage over other boards in this situation.

5. Conclusions

This paper reports on a comparative LCA case study of the manufacture of particleboard which considers two subsystems (agricultural and industrial). The potential environmental impact was assessed by two methods with regard to nine impact categories. LCIA in midpoint method was performed to determine the priorities to assist M.OSB producers in improving their environmental performance. The identified priorities were due to the industrial subsystem. The greatest impacts were related to the production of the adhesive and heat generation. This is mainly due to electricity consumption and the production of feedstock. Sensitivity analysis showed that the reduction in biomass fuel and adhesives brought the greatest environmental benefits. BIGCC systems, M. lutarioriparius-based fuel and recycling of distilled water from the drying process are also suggested as improvement opportunities.

In the endpoint method, M.OSB showed the lowest impact according to all indicators, so it was proven to be an adequate substitution for wood-based artificial boards. If the reduction in the environmental burden caused by harvesting was included in the system boundary, M.OSB showed a remarkable environmental advantage over other types in this research.

Further LCA research is required to address other sources of agricultural or forestry residues. This LCA research should be applied to the production of other types of panels such as high-density fiberboards (HDF) and veneer-based products (e.g., plywood and laminated veneer lumber). Cross-country comparative LCA studies which include data from China and other countries would also be beneficial.