Bioethanol processing from wheat straw: investment appraisal of a full-scale UK biofuel refinery

Abstract An indicative economic appraisal of a biorefinery investment has been undertaken for the case of an archetypal, full-scale process plant based on bioethanol production from wheat straw: a cellulosic co-product or ‘waste’ stream. A ‘life-cycle’, or ‘through-life’, biofuel chain was examined from the supply of wheat straw, through biochemical conversion processing, and distribution of product to fuel terminals for use in the transport network. The process technology investigated was dilute acid pre-treatment and enzymatic hydrolysis. Discounted cash flow (DCF) investment appraisal formed the key evaluation methodology, and the results were found to depend on the discount rate and lifespan of the biorefinery. Analysis of suitable UK locations, refinery scale or size, and logistics established potential low cost areas with good access to wheat straw. DCF investment appraisals of this type enable industrialists and policy makers to determine the implications of bioethanol production from wheat straw within a low carbon future.


Introduction
Background Biofuels are produced from organic materials, either directly from plants or indirectly from, for example, agricultural waste products. They can be characterized in two broad categories: being produced from woody biomass [including forest products, short rotation coppice (SRC, e.g. willow), or untreated wood products] and non-woody biomass [including high-energy crops (e.g. rape, sugar cane, and maize), animal waste, and waste from food processing]. The carbon dioxide (CO 2 ) released when energy is generated from biomass is roughly balanced by that absorbed during the fuel's production. Thus, it is sometimes regarded as a 'carbon neutral' process, and its use offers the opportunity to help meet CO 2 reduction targets [1]. Energy crops and agricultural residues are consequently considered good candidates for next generation of renewable energy technologies, particularly when obtained from marginal or 'waste' land. However, fossil fuels are used during their production (embodied in fertilizers), and during cultivation, transportation and processing. So that biofuels are not quite as low carbon as would otherwise be the case [1]. In any event, when biomass is refined to produce a biofuel it can help to meet the United Kingdom of Great Britain and Northern Ireland (UK) and European Union (EU) targets to gradually replace a proportion of fossil fuels in transport with biofuels. That would contribute towards the UK's legally binding target adopted in June 2019 [2] of achieving 'net-zero' greenhouse gas (GHG) emissions (or 'carbon neutrality') by 2050, and the aspiration of the European Commission to enshrine a similar 2050 climate-neutrality target into its first European Climate Law as part of a broader European Green Deal [3].
The UK transport sector has the fastest rate of growth in terms of primary (and end-use) energy consumption, and is currently responsible for 27% of total UK GHG emissions in 2019 [4]. However, Hammond et al. [5] suggested that there is only a modest potential for biofuel use in the UK automotive sector. Biofuels are nevertheless seen as 'sustainable' or 'renewable' fuels that do not require investment in new road transport infrastructure or vehicles; dispensed in current refuelling stations and able to be used in current vehicle power trains via blended fuels [5,6]. Indeed, the International Energy Agency (IEA) 'technology roadmap' on transport biofuels [7] suggested that, although First Generation Biofuels (FGB) will dominate the global market in the short-term (in line with the OECD-FAO projections analysed by Hammond and Seth [8]) advanced feedstocks will be required over the longer-term (as indicated by the IEA roadmap [7,9]). These advanced or Second Generation Biofuels (SGB) might constitute some 75% of biofuels production by 2050. Such SGB are generally produced from agricultural or crop 'wastes' (such as straw) and from non-food energy crops, which significantly reduces their negative effects [10]. The IEA [7] argued that the amount of global biofuels for transport could rise nearly sevenfold over the period 2020-2050 [to just over 30 ExaJoules (EJ) equivalent primary energy demand per annum]. That would represent some 27% of global transport fuel supply by the middle of the twenty first Century in contrast to only about 2% today [7,11]. However, the biofuels industry was strongly impacted by the Covid-19 pandemic with global transport biofuel production unusually declining in 2020. This is anticipated to rebound with total biofuel demand surpassing 2019 levels by 2021, and then growing by some 28% in around 2026 [12]. Bioethanola bio-based substitute for 'petroleum' or 'gasoline'produced from wheat straw could yield 6.8% of total world fuel supplies if all current cars were fitted with E85 ready engines: a blend of petroleum (or 'gasoline') and denatured bioethanol, containing up to 85% of the latter [1]. Certainly, in order to meet the need for road transport biofuels in Britain, and to create greater fuel security, both feedstocks and biorefineries need to be established domestically. Biofuels provided only 6% of total UK road and nonroad mobile machinery fuel in 2020 [4].

Wheat straw as a bioethanol feedstock
Straw is an agricultural by-product; the dried stalks of cereal plant, after the grain and chaff have been removed [1]. These stems arise from crops (such as barley, oats, rice, rye or wheat) that can alternatively be fed to animals ('fodder'), used as a layer on the ground for cattle and other livestock to lie on ('bedding'), or for the making traditional objects (e.g. straw baskets or hats). Glithero et al. [13] estimated the potential supply of cereal straw as a lignocellulosic SGB feedstock via an on-farm survey of 249 farms (cereal, general cropping and mixed units) in Englandthe largest geographic part of the island of Great Britain (GB) which accounts for some 84% of the UK population. This study was linked with data from the English Farm Business Survey (FBS), which is conducted on behalf of the UK Department for Environment, Food and Rural Affairs (Defra). Glithero et al. [13] consequently found that there is a potential cereal straw supply of about 5.27 million tonnes (MT) from arable farms with a variety of co-benefits: 3.82 Mt is currently used for fodder and other off-field purposes, whilst 1.45 MT is chopped and incorporated into soil on the fields for their enrichment. If this chopped and incorporated cereal straw from arable farms were converted into bioethanol, Glithero et al. [13] estimated that it might represent 1.5% of the UK petrol consumption on an energy equivalent basis. However, the variation in regional straw yields across the countryit principally comes from East Midlands and East of Englandwould have a great effect on the indigenous English supply of straw. Notwithstanding these uncertainties, wheat straw offers the potential to yield a significant quantity of sustainable SGB feedstock in the form of bioethanol.
In a related farm business or market study of the above cohort of farmers, Glithero et al. [14] discovered that around two-thirds of farmers would supply wheat straw for biofuel use, with the most popular contract length and continuous length of straw supply was either one or three years. They found that arable farmers in England would be willing to sell 2.52 Mt of cereal straw for biofuel use nationally, including 1.65 Mt in the main cereal growing areas of Eastern England. Thus, cereal straw could be diverted from on-farm uses and from straw currently incorporated into the soil. In a follow-up structured postal survey undertaken jointly with colleagues [15] 516 usable responses were received. These indicated that a substantial proportion of farmers, particularly in the East of England, where significant quantities of cereals are produced, were unwilling to change from their current practice of straw incorporation in order to supply straw for biofuel purposes. Consequently, straw supply for biofuel feedstock is likely to be rather more limited than earlier propounded [1,14]. Glithero et al. [14] therefore suggested that policy interventions might be required to incentivize farmers to engage in this SGB market, and they argued that food and fuel policies must increasingly be integrated to meet societal goals.

Bioethanol processing
Bioethanol production from cellulosic feedstocks Bioethanol production from cellulosic feedstocks requires three distinct steps: pre-treatment, total enzymatic hydrolysis, and fermentation [16]. This is typically known as separate hydrolysis and saccharification (SHF), although a process of simultaneous saccharification and fermentation (SSF) is also available. The latter involves hydrolysis and fermentation steps being carried out simultaneously in the same vessel [17]. The pre-treatment step is needed to increase the substrate digestibility, because lignocellulosic biomass is somewhat intractable in terms of enzymatic hydrolysis, due to various structural factors [18]. Indeed, the structure of wheat straw is dependent on multiple harvesting factors [1], but is primarily comprised of cellulose, hemi-cellulose and ligninthe three basic components of crude biomass. According to Talebnia et al. [19] the composition percentages for these components are generally in the ranges 33-40, 20-25 and 15-20, respectively. The overall success of bioethanol production from wheat straw is largely dependent on the ability of the pre-treatment technique to improve the digestibility of polysaccharides, cellulose and hemi-cellulose contained within the structure of the straw. Cellulose chains in the epidermis of wheat straw are oriented along the growth direction of the straw. They are compacted into long polymer chains called 'microfibrils', which are linked by short chain, hemi-cellulose polymers [19]. Enzymatic hydrolysis of cellulose will rarely exceed 20% without pre-treatment, unless high enzyme concentrations are used [20].

Pre-treatment techniques
The aim of pre-treatment is to breakdown the lignin structure, and to disrupt the crystalline structure of microfibrils in order to release cellulose and hemi-cellulose polymer chains [20][21][22][23]. However, carbohydrate degradation can occur during pre-treatment, if conditions are too harsh, leading to the formation of inhibitors such as acetate, phenolic compounds and furfural [24]. These inhibitors then reduce the effectiveness of either of the following hydrolysis or fermentation steps. The pre-treatment temperature, time and pH must therefore be balanced to maximize polysaccharide output and minimize inhibitor formation [25]. Pre-treatment techniques can be classified into four categories; physical, physio-chemical, chemical, and biological. Taherzadeh and Karimi [21] lists available pre-treatments for all lignocellulosic feedstocks, whereas Talebnia et al. [19] provides a more updated description of pre-treatment technologies currently being adopted and/or studied in connection with wheat straw.

Conversion Processes
Bioenergy conversion processes are the methods by which the energy stored within biomass can be released [5,26,27]. The complex and varied nature of bioenergy means that unlike other renewable energy sources, which have one set method of energy generation and mode of output, there are a wide range of bioenergy conversion processes that can deliver energy in many ways. These can result in solid, liquid and gaseous fuels, and provide energy across the end-uses of heat, electricity and transport [27]. Conversion processes for releasing the energy from biomass range from simple combustion (i.e. burning wood) to the complex formation of liquid biofuels for transport from lignocellulosic biomass feedstocks [28][29][30]. The conversion method used will primarily rely on the required end-use of the biomass; whether it is required to provide heat or power in-situ or to generate gaseous or liquid fuels for use elsewhere [26]. The development of increasingly efficient conversion processes into the future will be a key factor in deciding how limited biomass resources are best utilised. Such bioenergy conversion processes can be characterised as thermo-chemical, biochemical and physical-chemical methods.
Fermentation is the process of producing alcohol from sugars. In the case of bioenergy, fermentation is primarily used to produce bioethanol from sugar and starch feedstocks, such as maize, wheat, sugarcane and sugar beet. Due to the low sugar content in cellulosic crops, such as perennial grasses or straw [9,[28][29][30], bioethanol is more difficult to produce via fermentation. The sugar content of wheat straw is only about 1.4% by dry weight. Nevertheless, progress in that direction has recently been made [1,[31][32][33]. In order to produce bioethanol from cellulosic crops, the cellulose must first be broken down into sugar through hydrolysis [34]. It is then possible to ferment these sugars in order to produce bioethanol. It has been suggested that hydrolysis of lignocellulosic biomass could lead to low cost and efficient production of bioethanol that may consequently become competitive with fossil fuels within the next decade or so [35]. Extensive research on lignocellulosic bioethanol production has been conducted over recent years. This is reflected in a series of substantial published reviews, e.g. Chantell et al. [34]; Mabee et al. [36]; Mood et al. [28]; Gupta and Verma [37]; Khoo [29]; Kumar et al. [30]; Albashabsheh and Stamm [38].

The issues considered
The aim of the present indicative study was to provide an economic investment appraisal of an archetypal, full-scale biorefinery [39,40] for processing wheat strawa cellulosic co-product or 'waste' streaminto bioethanol across the life-cycle fuel chain. A biorefinery is a facility where low-value renewable biomass materials (such as crop residues like wheat straw) are the feedstock to a succession of processes where they are converted into several highervalue bio-based products (in this case bioethanol). Slade et al. [10] argued that the eventual cost of transformation of lignocellulosic biomass into road transport-grade bioethanol would be determined not just by the performance of the conversion process, but also by that of the entire supply-chain from feedstock production to consumption. A 'life-cycle' or 'through-life' biofuel chain has therefore been evaluated in this work from the supply of wheat straw, through full processing via a biochemical technology, and distribution of product to fuel terminals for use in the transport network. This life-cycle thinking is similar to that employed in the formalised procedure of environmental Life Cycle Assessment (LCA) to evaluate a range of environmental burdens from biofuel processing [41,42] and energy systems generally [43]. A series of international standards have been prepared that covers sustainability criteria for the production of biofuels and bioliquids for energy applicationstheir principles, criteria, indicators, and verifiers [44][45][46][47]. The present study generally follows the boundaries established in these standards (cited in terms of their UK specifications). Here the particular biochemical technology investigated was based on dilute acid pre-treatment and enzymatic hydrolysis. The system boundary is equivalent to 'cradle-to-gate' in LCA, or 'strawto-pump' in the present context. Discounted cash flow (DCF) investment appraisal formed the main element of the methodology employed. It represents a companion piece to the thermodynamic evaluation of bioethanol processing from wheat straw by Hammond and Mansell [1].
Analysis of suitable UK locations, plant scale, and logistics were undertaken in the present study in order to establish a potential low cost area with high access to wheat straw. Slade et al. [10] advocated the need for such location-specific assessments of feedstock availability and price. The base case examined was for a biorefinery processing 237,000 metric tonnes of wheat straw (dry weight) annually producing $75 million litres (ML) of denatured bioethanol per year. The wet straw was reduced by some 7% during pretreatment; effectively drying that takes place during the process. Sugar content was only about 1.4% by dry weight with a moisture content of 6.5% (in contrast to wheat grains that have a moisture content of about 15% in the UK context). Sensitivity analysis has been used in the present study to pinpoint the main factors that influence costs, including changes in yields (bioethanol yield, followed by wheat straw yields). The linear sensitivities analysed included farm gate, chemical and employee costs, discount rates, and electricity revenues. The market analysis was used to identify the market competiveness of bioethanol produced at the base case for the minimum ethanol selling price (MESP) at high and low bioethanol blends. These estimates will enable industrialists and policy makers to take account of some of the ramifications of bioethanol production from wheat straw in a low carbon future.
The remainder of the paper is structured as follows: the next section provides an overview of the Materials and Methods applied in this study, including the assumptions made in carrying out the DCF investment appraisal of the archetypal, full-scale biorefinery. This is followed by a section that presents the Results and Discussion flowing from this study, whilst the subsequent section outlines the role of crop residues (such as wheat straw) in a 'circular bio-economy' [40,48], and finally the paper closes with Concluding Remarks.

Wheat straw availability in the UK
The initial value chain was established for research on currently available lignocellulosic demonstration plants in Europe was identified from the survey by Slade et al. [10]. The system boundaries were defined to cover the major processes and logistics, although these are not presently in place within the UK. The technology choice within the biochemical route encompasses a number of different process routes. In the present indicative study, a pre-existing route was selected based on a techno-economic assessment carried out at the US National Renewable Energy Laboratory [NREL] (see Humbird et al. [49]) in collaboration with partners at the Harris Group Inc. It examined dilute acid pretreatment and enzymatic hydrolysis of corn stover (another crop residue), and therefore required adaptation to evaluate wheat straw. Similarly, the findings of Humbird et al. [49] were converted in terms of scale, index prices, and currency in order to conform to economic conditions prevalent in the UK.
Given wheat straw was the chosen feedstock for the process value chain, an initial analysis of the available resources and distribution within the UK was vital. This was completed using official mapping data of UK wheat growing land gained from relevant Government departmentthe Department for Environment, Food and Rural Affairs ([Defra])most recently updated at the local (i.e. county) level in the 'Crop Map of England (CROME) 2019' report [50]. Wheat straw availability in England was obtained from the companion findings of Glithero et al. [13], who determined the quantity and regional distribution/yields. Information on availability was extracted from government sources [50] and one of the Defra executive non-departmental public bodies (the Agriculture and Horticulture Development Board (AHDB) [51]). Areas where large-scale current and future industrial CHP plants co-fire wheat straw were removed from the location choices. This was because long-term contracts (8-12 years) have typically been made with local farmers, based on supplying similar levels of agricultural material year-on-year. Little wheat straw would therefore be available for biorefineries in areas of planned straw-based energy facilities. Other regional uses, such as animal feed and bedding, were also identified, and then removed from the study in order to minimise the influence of regional competition for these bioresources. The basis of sourcing wheat straw was so as to not remove the resource from other established market uses, this led to only currently incorporated wheat straw being considered for the future biorefinery. Once availabilities of wheat straw across the UK was established using these criteria, the choice of location for the archetypal, full-scale biorefinery was deduced based on the region with the largest access to available wheat straw for bioethanol production.
The scale for the biorefinery was estimated based upon maximising the land access from the prescribed location. Similar collection radii to combined heat and power (CHP) plants in the UK were used with the aim of reaching between 25 and 50 metric tonnes per hour (MTPH). IEA Bioenergy Task 39 experts [52] indicated that the former Inbicon Commercial Demonstration refinery (that operated in Kalundborg, Denmark over 2009-2014) had adopted this target. Three different UK collection radii were selected to compare and contrast the increasing scale within a specific region with wheat fields distributed across it. All radii were created so as not to cross known current or future CHP plant collection radii. Geographic mapping information, as well as the regional wheat straw availability for the selected location, were utilised in order to determine the wheat growing areas, and then the associated collection radii to calculate the available tonnes of wheat straw. This established the overall scale of each potential biorefinery, and overall process is highlighted in Table 1 [13,50,51,53].

DCF Investment appraisal
Economic appraisal evaluates the costs and benefits of any project, programme, or technology in terms of outlays and receipts accrued by a private entity (household, firm … etc.) as measured through market prices [54]. Financial appraisal is used by the private sector and omits so-called environmental 'externalities'. In contrast, economic costbenefit analysis (CBA) is applied to take a society-wide perspective, with a 'whole systems' view of the costs and benefits [55,56]. It accounts for private and social, direct and indirect, tangible and intangible elements; regardless as to whom they accrue and whether or not they are accounted for in purely financial terms [54,55]. A further distinction between financial appraisal and CBA is in the use of the discount rate to value benefits and costs occurring in the future [55,56]. Financial appraisal uses the market rate of interest (net of inflation) as a lower bound, and therefore indicates the real return that would be earned on a private sector investment.
This approach takes account of the 'time value of money' and discounting in order to obtain the appropriate investment appraisal criteria [55][56][57][58]. The net present value (NPV) of the sum of the capital cost, maintenance and operational costs, as well as (potentially) plant decommissioning, is calculated over the life of the project, along with the NPV of the total fuel produced. Consequently, by using this method, different technical options with a variety of lifespans, capital costs, and efficiencies can effectively be compared so that the most cost-effective option can be determined. In the case of public sector investments a so-called Test Discount Rate (TDR) is utilised in the UK. It is typically derived from a comparison with private sector discount rates fforming the Weighted Average Cost of Capital (WACC)g. In the UK, HM Treasury [59] recommends that the TDR for projects with durations of less than 30 years should be taken as 3.5% in real terms. An exception is applied for projects involving risk to life, when a lower rate of 1.5% is employed [59].
An investment appraisal method similar to that used by NREL [49] was adopted for the present study. It employed a DCF rate of return to calculate the minimum bioethanol selling price [when the net present value (NPV) is £0]. The NPV was considered zero when the bioethanol price was just positive to five significant figures. Once all the costs were established and modelled, the resulting MESP was normalised to break the overall MESP into smaller cost elements, such as the fraction of the MESP due to depreciation. Sensitivities were then used to assess the impact of each key parameter on the minimum price of bioethanol. This established which parameters were the most sensitivity to further cost reduction and therefore to decrease the minimum price of bioethanol. These assessments were carried out in the context of the prevailing UK investment environment. Finally, a market analysis of the MESP in a base case was used to test market sales opportunities. The main economic assumptions adopted for the present study were:-All costs were indexed to 2020 £ 1 [e.g. using the Chemical Engineering Plant Cost Index (CEPCI) [60] for CAPEX]. The plant is 100% equity financed. Working capital requirement is 5% of the fixed capital investment. A 21% corporation tax rate is assumed to apply over the life of the biorefinery. All capital investment occurs during the build duration (1 year). Land costs were estimated as £1,000,000. Depreciation of all fixed capital investment was based on a 20% declining balance. Start-up financial position during initial 3 months.
The boundaries for the analysis were defined as the feedstock cost ex-farm, finishing with the denatured bioethanol cost at the fuel terminal.
Ex farm wheat price: Large-scale uses of wheat straw, specifically CHP plants as a co-firing material, typically employed the farm gate cost for each tonne of wheat straw was set at a single price. This is commonly written into long-term contracts for supplying a given quantity of wheat straw. The price is index linked, so that if the market price increases so also does the contract price. The price includes all the farm-side activities, such as harvesting and storage, until collection. For the base case, an indexed 2020 value of £52 per tonne was assumed for the analysis (after Glithero et al. [13]) Ex terminal ethanol price: The fuel terminal was used as the end-point for the analysis, as this is a well-defined price within the current fuel supply chain. It is similar to the exrefinery price, with most liquid fuel being transported by pipelines which are low cost. The distribution network from the terminal to the forecourts is then typically undertaken by highly efficient and established logistics companies. The cost for the logistics and mark up to the forecourt is about 7-8 p (Sterling pence) per litre for petrol as of 2020 [61]. Bioethanol has a hygroscopic nature (it attracts water) and is corrosive; consequently, it must therefore be handled separately until late in the value chain. This may result in increased investment from fuel terminals leading to higher margins required compared to petrol and diesel. It was assumed that the fuel was blended at the fuel terminal [62].

Outline of the selected of biorefinery technology and sub-processes
The technology choice was taken directly from the NREL study [49], given that it was the most complete and detailed account of inputs/outputs and costing of all process requirements. The research used established a number of processes as the basis for analysis (using 'nth-plant' assumptions), removing additional cost risks and learning curve uncertainties associated with starting a refinery process [49]. Individual processes can be broken down into nine areas in the report:-Feedstock handling: Wheat straw is delivered to the feed handling area in bales. Storage is kept to a minimum with the bales been broken down and fed directly into the pre-treatment reactor. Pre-treatment and conditioning: The pre-treated wheat straw is treated with a dilute sulphuric acid catalyst as it is heated. This breaks down the straw and releases the hemicelluloses sugars. Ammonia is then added to reduce the acidic conditions as the mix was fed into a reactor for enzymatic hydrolysis. Enzymatic hydrolysis and fermentation: Enzymatic hydrolysis is executed in a fed-batch process using cellulose. The partially hydrolysed slurry is then moved into one of several batch reactors. Hydrolysis is completed in the batch reactor, and then cooled as the micro-organism Zymomonas mobilis is added for fermentation of the released sugars. This process is carried out sequentially over five days until most of the cellulose and xylose has been transformed to bioethanol. The fermented mixture is then sent to product recovery (see below). Cellulase enzyme production: Enzymes were produced on-site to reduce costs. It was produced from the fungus Trichoderma reesei in fed batch reactors. This entire reactor broth containing the produced enzymes was added to the enzymatic hydrolysis reactor (see above). Product recovery: The fermented mixture is separated into different streams of bioethanol, residual liquids, and residual solids by a mixture of distillation and solid-liquid separation. Bioethanol is distilled and Removal of areas with current and near future large-scale wheat straw uses (CHP co-firing).
Step 1, 2 4 Analysis of regional wheat straw yield and uses within general location to establish available resources.
Glithero et al. [13] 5 Creation of general collection radii from general location to maximise land coverage while minimising competition with other uses.
Geographic distribution map (Defra [50]) 6 Calculation of maximum scale of refinery from the collection radii.
Step 4, 5 7 Calculation of total costs and costs per unit of wheat straw logistics (£/T) and bioethanol delivery (£/L).
Road Haulage Association (RHA) cost tables [53] purified to 99.5% purity using vapour phase molecular sieve adsorption and sent to storage. Recovered solids are collected and sent to the combustor and other recovered liquids are sent for wastewater treatment. Wastewater treatment: Wastewater is treated by both aerobic and anaerobic digestion. The treated water is recycled, and any produced gas and residual sludge from treatment is transferred to the combustor for burning. Storage: Bulk storage is provided for the generated bioethanol, as well as all major chemicals for the production process (such as sulphuric acid, water and ammonia). Denaturing of the bioethanol occurs there. Combustor, boiler and turbo generator: All waste solids and gases from the production process are burnt in the combustor to produce high-pressure steam, along with generated heat and electricity. Excess steam produced is converted to electricity for plant use and the excess was sold to the grid.

Logistic Requirements
Logistics are organised around the need to maximise the use of vehicles to and from the facility. These were all located within the local area (30, 35 or 40 mile radius [1 mile ¼ 1.60934 km]) an average journey distance (x) is calculated for inbound and outbound logistics. The average journey is based on accessing half of the available area within each radius. A winding factor of 1.2 was used to account for indirect winding roads. Wheat straw can be harvested and gathered into bales using large balers, which creates many different sizes and density of bales. Common bale types used for transportation are large round bales (cylindrical 1.2 Â 1.2 m, 250 kg) or the large square bale (cuboid 1.2 Â 1.2 Â 2.4 m, 500 kg). The most efficient shape for transportation purposes was the large square bales, due to their high density and efficient shape for stacking allowing a greater straw mass to be transferred per journey. These bales were therefore assumed to be adopted for collection purposes.
Diesel Euro 4 type, flat-bed trucks were selected as being convenient to transport the straw, because of their reasonable size for travel along country roads to farms. These are 6 Â 2 axle vehicles with a gross vehicle limit of 26 tonnes (T), with typical load capacity of 16 T. 24 large square bales were assumed to be transported per journey, due to the constraints imposed by these bales. This leads to a maximum delivery of 12 tonnes per trip [63]. Overheads and other vehicle costs were sourced from the Road Haulage Association (RHA) cost tables [53].
Bioethanol transport is highly dependent on the relative distances between the biorefinery and fuel terminals, where the fuel can be blended and sold onto fuel forecourts. Because of the hygroscopic and corrosive nature of bioethanol, the fuel cannot be transferred through the many UK pipelines [62]. This left tanker trucks to transfer the bioethanol from terminals to the refuelling station forecourts across network. The fuel at that stage has a much higher energy density than the feedstock going into the biorefinery and has a lower impact over the distance traveled. The standard fuel distribution network from the fuel terminals are operated in Britain by tanker trucks, up to a maximum weight of 44 tonnes, or 18 tonnes in rural areas [64]. Typical road tanker volume was 36,000 litres (L); imposed by the UK maximum weight limits for 6 axle road vehicles. The weight of an unloaded truck cab and trailer was 15 T, and this left 29 T carrying capacity [64]. The density of bioethanol is approximately 789 kg/m, 3 although volume capacity is more likely to be the limit. Thus, a large tanker trailer of 36,000 L capacity (28.4 T) was used adopted for modelling purposes. HM Revenue and Customs requires that all bioethanol be denatured before leaving the point of creation when used as a road fuel. In order to denature the bioethanol, petrol of an octane rating greater than 91 is commonly used with a volume of at least 1% mix. Typically 5% volume of petrol (or 'gasoline') is used, and that was adopted for the current research.
Google Maps were used with the starting point as Ettington, Warwickshire in order to calculate the distance to the three closest fuel terminal locations. Birmingham and Oxford were both within a 40 mile distance using the M40 motorway (freeway in USA usage). In contrast, Bristol is approximately 80 miles away, and therefore a site North-South between Birmingham and Oxford fuel terminals were selected for bioethanol supply. The costs associated with the distribution were then determined from the number of journeys required per year, journey length, and associated costs. This information and data are provided in Table 2.

Process and costing estimates
Once the different scales of the three biorefineries were established from the availability of wheat straw in the collection radii, the new indexed process costs were calculated from NREL data [49]. The total installed equipment costs for the major equipment groups were scaled to the new full-size requirements. Each equipment grouping was scaled to a defined factor taken from Hamelinck et al. [65]. These scaling factors are displayed in Table 3 with final cost and scale factors for the three different collection radii (different feed rates).
Capital expenditure (CAPEX) costs were extracted from the NREL study [49], converted to GBP values, and then indexed via CEPCI tables [60]. They were also scaled to the required feed rate using the scaling factors as again shown in Table 3. The operational life of the biorefinery has a large influence on its economics as it spreads the capital cost over the total life. Life-times vary from 7 years to 30 years in techno-economic appraisals. The lower lifetime is due to the limited maturity of novel process technologies. A lifetime of 15 years was adopted for the current study in order to provide a medium life span within the two extremes [49,66].
The plant size change ranged from 23 to 33% the scale used in the NREL biorefinery, and so it was assumed that there was only a requirement for half the employees. Revenue was raised through the sale of bioethanol and the co-produced excess electricity. Working capital was calculated as a percentage of the fixed capital investment: 5% in the present case. This provides sufficient cash flow to cover 2 months of costs without sales revenue [67]. During the start up time, which was the first three months of operation, the revenues and fixed costs were reduced in line with typical refinery running up times. Finally, a financial 'base case' for the financial analysis of the designed process chain was established to ensure all results were directly comparable. The base case parameters are presented in Table 4.

The context
The three different scales of planned biorefinery were analysed on DCF basis to find how the greater economies-ofscale and increasing logistical costs affect the final value for the minimum ethanol selling price (MESP): see Table 5. As can be seen the lowest MESP was found with the 40 mile radius of collection. No other distances or scales were considered as the wheat land became sparser the further the radius was extended past 40 miles, as well as more competition with other users negatively affecting the economics [50].

Capital Investment
The refinery was found to have a total capital cost of $£105.2 million (M) [see Table 6], with the sum of fixed and variable costs at the full base case operation being a total of $£8.4 M. These figures are thought to be accurate to ±30% of this cost, due to the scaling and factoring equations used to calculate them [67]. The total capital cost with uncertainty included is therefore around £105.2 M ± £31.5 M. Capital costing information within this industry at these particular full scales are sparse, and real-world validation of these figure would be difficult. Normalisation CAPEX costs has been performed in order to provide a better comparative understanding of the cost break down for the biorefinery investment.

Normalised findings
The MESP calculated through the DCF analysis came to 51.7 p/L, when normalising all the costs separately against the total produced bioethanol the MESP came to a total of 53.3 p/L; see Table 7. This is due to the different methods used. The effects of the discount factor along with the changes in tax paid per year over the project due to the 20% depreciation capital allowances for plant and machinery. This reduced the tax paid over the initial 4 years to zero, while later years were discounted heavily pay full tax. This causes the discounted cash flow MESP to be the more realistic of the two. The overall difference comes to just 3% of the total MESP and so this is considered acceptable within the accuracy of the analysis.
The largest cost factor influencing the MESP was the feedstock costs at 37%, based on the farm gate costs plus the cost of delivery. This has also been found in several earlier studies (e.g. Gnansounou and Dauriat [66] and Humbird et al. [49]) indicating that a key way to bring the costs associated with SGB refineries down is via the sourcing of a low cost lignocellulosic resource. The second largest cost factor at 25% was the 'return on investment' (ROI) cost, and therefore the rate at which interest is charged on capital will reduce the MESP. This could be obtained by securing long-term loans, which typically are associated with a lower interest than equity, as well as reducing financial risks [49]. The third largest influence was of the capital depreciation at 18%. This relates to the total capital investment, as well as the number of years the biorefinery is operational. These three areas come to a total of 80% of the MESP establishing that decreasing the cost of these  locations is the best way to create a more cost-competitive biorefinery.
The largest of the variable cost parameter was associated with enzyme production; at 7% of MESP, which is of similar magnitude to the savings through the sale of electricity to the grid. This price per litre toward the enzyme production was of a similar value to that of all other chemical inputs required during the process, therefore of all the process areas, improvements in cost efficiencies of this area would result in the largest MESP reduction. It must be said that this enzyme cost comes in below the method of buying in the enzymes [49].

Sensitivity analysis
Sensitivity checks are desirable in order illustrate those factors that are sensitive to change in values of key factors. The chemical product, electricity, employee, and farm gate costs (along with the electricity price and discount rate) have a similar price impact in the positive and negative directions. The bioethanol and wheat straw yield in contrast were biased toward a negative influence. The bioethanol yield was compounded by the fixed costs of the chemicals being input into the plant resulting in less output bioethanol. In the case of the wheat straw yield, a typical decrease took the biorefinery down to 60% capacity. Clearly, it is desirable to ensure the supply to, and performance of the biorefinery in all yield terms, is kept near the design case [67].
The bioethanol yield was reduced in the present study in comparison to the NREL assessment [49], because a 70% conversion rate was used. Processes for the biochemical technology seem to suggest that conversion rate stays mostly constant over the life-time so this risk, although it has the largest effect of the sensitivities, has a very low risk level [68]. The wheat straw resource security had the second largest impact in terms of sensitivities with a much larger risk, because of the very variable resource base [10,69]. A lot of factors impact the growth and yield of wheat straw, the most obvious (such as weather and disease) are prevalent. Variations in the wheat straw harvest represent a large risk factor in both occurrence and impact. The only available methods to reduce this risk are obviously to base the biorefinery in very dense areas of the resources used (in this case wheat straw), as well as basing it in areas of limited competition for the resource [70]. Both of these factors were taken into account at the base stages of the research to reduce the risks.
The impact of lifespan from 10 to 30 years was $9 p/L. If the life span of the refinery is established to be longer than this, then the effect is reduced. Longer life spans typically lead to higher maintenance or increased investment reducing this influence in real terms [49]. Finally, the large impact of the discount rate was analysed. Here a 3.5% rate is typical of the base rate for interest provided by a bank, whereas a 16% rate is that for a highly risky project (according to the business consultants KPMG International [71]). The overall range of MESPs from altering the discount rate was $16 p/L with the effects the biorefinery MESP varying in a linear fashion. This shows a clear need to limit this factor to allow the process to be more investable. To do so in-depth investment appraisals require completion to establish all the risks.

Towards a bio-circular economy
Both the UK and EU have viewed the adoption of liquid biofuels in the transport sector as a policy intervention for helping to meet climate change mitigation targets, enhancing regional fuel security, and contributing to rural development [5,26,40,72]. The latter could be aided by the provision of an alternative source of income for, otherwise depressed, agricultural communities from the production of biomass. Such  biomass resources can be converted into premium-quality liquid biofuels and biochemicals [39,[72][73][74] via a biorefinery: an industrial facility for the sustainable processing of biomass into premium-quality bio-based fuels, chemicals, and materials. However, such bio-based products may have significant deleterious impacts in terms of direct and indirect land use change, loss of biodiversity and the impairment of eco-system services [73,75,76], and competition with food production. These side effects may be offset by the use of, for example, bioethanol (a bio-based substitute for 'petroleum' or 'gasoline') produced from agricultural or crop 'wastes' (such as straw) and from non-food energy crops, which significantly reduces their negative effects [1,10]. It holds out the prospect of retaining the existing transport infrastructure (e.g. refuelling or 'petrol' stations), in contrast to other potentially low carbon options, such as hydrogen-fuelled or electric vehicles [26]. That has significant benefits in terms of limiting capital expenditure and the potential speed of take-up. Biorefineries can contribute to what has been called a 'bio-circular economy' [30,48]: a programme to 'reduce, recover, reuse, recycle' biomass resources. It is a bio-based version of the so-called 'circular economy' that is sometimes referred to as being associated with 'resource efficiency' improvements or 'supply chain collaboration' [by the UK Department for Business, Energy and Industrial Strategy (BEIS)] engaging producers and consumers [77]. Wheat straw is a feedstock that can be regarded as a waste stream, but it has been shown here that it can be recovered and recycled to produce bioethanol [1,38]. Patel et al. [78] in the Netherlands undertook detailed analysis of bioethanol production processes as part of a portfolio of 'white' biotechnology opportunities. Bioethanol production via lignocellulosic biomass was then (in 2006) regarded as being at the development stage, and had yet to have been demonstrated at an industrial scale. Indian researchers have been at the forefront of modern biorefinery developments [40,48]. Thus, a more optimistic assessment was recently produced by Kumar and Verma [40] in the light of technological advances in the processing, such as genetic and metabolic engineering, along with synthetic and systems biology. Similarly, Awasthi et al. [48] noted that the integrated production of biofuels and high-value products greatly improved the economics of processing. This may however require the development of new business models in order to secure the benefits for both producers and consumers. Other teams engaged in the BSBEC research programme in the UK (see below) also recognised these factors.

Concluding remarks
The UK Research and Innovation (UKRI) is a non-departmental public body currently responsible for funding research and knowledge exchange at UK universities and via the national innovation agency. Its Biotechnology and Biological Sciences Research Council (BBSRC) established a BBSRC Sustainable Bioenergy Centre (BSBEC)actually a research networkthat in turn supported a project on the generation of bioethanol from the lignocellulosic biomass, including excess straw, spent grains and waste generated from food production. The present contribution has formed part of that project, and is a companion to other studies on straw use and availability in England, the development of fermentation techniques for lignocellulosic residues to produce bioethanol, the environmental impact of biofuels, and the thermodynamic performance of bioethanol production from wheat straw (a cellulosic co-product or 'waste' stream). Here an indicative economic evaluation of an archetypal, full-scale biorefinery for processing wheat straw into bioethanol has been undertaken across the lifecycle or through-life fuel chain. The process technology investigated was dilute acid pre-treatment and enzymatic hydrolysis. DCF investment appraisal found that the minimum ethanol selling price was to 51.7 p/L for a biorefinery processing 237,000 metric tonnes per year of wheat straw (dry weight) producing $75 million litres (ML) of denatured bioethanol a year. Such a biorefinery would require a capital investment of $£105.2 M ± £31.5 M. Feedstock costs were a primary component of the overall price; representing 37% of the total. There is scope to reduce the estimated MESP further, especially the cost of feedstock and for producing the enzymes employed within the enzymatic hydrolysis process. Key additional financial factors identified were the discount rate, and overall investment depreciation as dictated by the biorefinery lifespan. The largest risk was found to be associated with the availability of the wheat straw. In addition, the operation of the biorefinery needs to be as close to the maximum capacity as possible in order to mitigate against compounding reductions associated with low throughput or yield. This study will help industrialists and policy-makers to understand the ramifications of this biorefinery technology in a low carbon future.