Industrial biotechnology encompasses the application of industrial processes in the use of renewable feedstocks [11] , which is a starting point in the transformation from fossil to bio-based materials. Starting from a conven- tional industrial approach, recent scientific and technological advances in areas such as biocatalyst development, bioprocess design, interfacing with organic reactions, process analysis and purification technology [12] have led to an emerging paradigm shift, offering economic possibilities. Presently, attempts to transfer biomass to indus- trial useful and economically viable polymers through traditional biotechnological methodologies have led to limited successes, suggesting that a lack of effective interdisciplinary processes [13] , requiring both a technical and supply chain improvements.

The production of biomass into industry grade materials must follow the process of harvesting and collecting the feedstock, material flow and storage of the feedstock through the supply chain, transportation of the material, pre-treatment techniques and design of the production system into the end product [14] . For this to occur, feeds- tock cost must be viable, supply must be consistent and routine and supply chain actors and a system must be developed/be competitive. Today, there is no value/supply chain system in place for biomass in traditional com- parison, rather a number of segmented systems [15] . Lignocellulose as a feedstock is not an efficient material given its disparate geography [14] over forests, fields and factories, which present a logistical challenge.

Different from a mass market, conventional feedstock supply chain system, a bio-based “industrial ecosystem” must act as a complex system that interacts through a large number of agents, acting in a less structured manner [16] . In its onset, achieving supply chain synergies of reuse is often achieved through structured relationships via industrial symbiosis that aims at engaging separate industries to cooperate in a collective approach [17] in order to be competitive to the economics of conventional feedstocks.

Partially as a result of the burgeoning chemical and plastics industry that grew in ten years to over $95 billion annually¹, limited interest relative to the growth of lignocellulose as an alternative. The growing use of ligno- cellulose in industry has been considered for decades, yet research had only increased once commercial viability seemed possible in higher grade applications [18] . It has been recognized that valuable sugars (such as pentose sugars) exist within plant/lignocellulose materials, but the biological and chemical conversion of this biomass has been hindered by the recalcitrance of plant cell walls [19] , which is relatively refractory to direct degrada- tion followed by microbial bioconversion [1] . Obtaining sugars from these feedstocks requires both physical and chemical disruption, which leads to high conversion costs. In this process, the greatest obstacle is the lack of low cost technology for overcoming the recalcitrance of these materials [20] . Cellulosic materials are not readily fermentable due to the porosity of the waste materials, crystallinity of the cellulose fiber and low fiber porosity [21] .

Chemical pretreatments are expensive in the conversion process, but a more expensive alternative is no pre- treatment [22] . The challenge in pretreatment is to release the sugars from the lignin polymer without a signifi- cant impact on the valuable material; pretreatment generally refers to the disruption of the naturally resistant carbohydrate-lignin shield that limits the accessibility of enzymes to cellulose and hemicelluloses [23] . The fea- tures of successful pre-treatment include the breaking down of the lignocellulosic complex, decreasing the cel- lulose crystallinity, preserving the hemicellulose sugars, limiting the formation of degradation products that prevent hydrolysis and fermentation, minimizing energy and chemicals in the conversion process, achieving a high value lignin co-product, and minimizing toxins and hazardous wastes [21] . Beyond the lignin shield, there are challenges in the chemical and compositional ratio of carbon, hydrogen and oxygen that produce additional challenges, requiring novel catalytic conversion processes that are currently being studied in research [3] . From pretreatments, high yields of close to 100% should conceptually be achieved, with a general target of 85%² that is infrequently achieved. Novel biological pre-treatment techniques [24] are progressing to increase yields through training bacteria [13] , and/or genetically modifying lignocelluloses for better conversions [5] , but such breakthroughs are at their infancy in development.

While the conversion costs of cellulose has been reduced significantly over the past two decades with ad- vances in pretreatment, sugar fermentation, enzyme production, and enzymatic hydrolysis, further improvements must be achieved for market competitiveness [23] . Today, the conversion cost accounts for 67% of the total product cost, with pretreatment (18%) the most expensive element of the processing cost [22] . The present day challenge of higher conversion cost in relation to petrochemical feedstocks may be offset in the future in lower acquisition costs, as is shown in Table 1.

Brewer’s spent grain (BSG) is a high value material, containing hemicellulose, lignin, and high protein content [9] , monosaccharides of xylose, glucose and arabinose, and minerals and amino acids [25] . Although its compo- sition of barley grain husk, pericarp and fragments of endosperm is contingent upon the barley type, harvest time and mashing conditions [26] , carbohydrates comprise of around half of the dry mass of BSG, with the balance mostly being proteins and lignin. Challenges associated with using BSG as a feedstock includes the existence of a complex outer layer, making it difficult to separate and convert, and a high moisture content (80% - 85%), making it susceptible to microbial growth and spoilage within a 7 - 10 day period [6] . The moisture issue can be stabilized post production in the brewing process, deterioration of the feedstock through microbial activity can further reduce yields in the conversion process [27] .

BSG is the most abundant brewing by-product, comprising of 85% of by-products generated, 31% of original malt weight and 20 kilograms per 100 liters of beer produced [28] . At present, the use of BSG as a feedstock is limited, primarily as animal feed, landfilled [28] , and more recently, bioethanol. With respect to animal feed, BSG has been found to be an excellent feed ingredient for ruminants, as well as benefits for humans [28] . Beyond its usefulness as an animal food by-product, some of its components are also being viewed as precursors for food grade chemicals or as energy sources in microbial fermentations [28] . BSG has also been used as an enzyme for removing organic materials from effluents and the immobilization of various substances [25] . Ener- gy recovery from BSG has been considered as viable relative to thermochemical and biochemical processes, yet most studies have focused solely upon bioethanol rather than other uses [29] . The sugars that are released can be microbially converted into various products, such as organic acids, ethanol, glycerol, food additives and butanol, among others [30] . In particular, lactic acid has found applications relative to food, fermentations, pharmaceuti- cals and chemicals [31] . Lactic acid is also a step before the development of poly-lactic acid (PLA), which leads to bio-plastic applications.

Based upon data received from the participant brewer, and validated through extant literature, the ratio between finished product beer and brewer’s spent grain is 5:1. Using this data within the context of the US beer market (200 million barrels a year), the US annual supply was approximated to be 2.4 million tons of BSG available to the secondary market, if one were to exist. Given the concentration of the US beer market between the two larg- est producers (Anheuser Busch Interbrew and MillerCoors) being nearly 70% of the total market, there exists an economy of scale market opportunity of 1.68 mm tons of BSG available from these two large beer manufactur- ers that would seem to create efficiencies through economies of scale. From our interviews, we found the current market price for BSG as a feedstock to be lower than that of other cellulosic crops (from extant literature) of $40.23/ton versus $50 - 150/ton, respectfully, which appears to be justified, given the high moisture content

of BSG, and assuming it is sold as-is.

Despite its low saccharide composition, and high moisture content that pose technical challenges to its eco- nomic value, there may be advantages related to this material from a supply chain standpoint. Given the US beer market (and that of other nations as well) is highly concentrated to a few actors, and has a centralized origin points (regional breweries in highly populated areas) it is conducive to supply chain optimization in contrast to decentralized agro-waste materials of other lignocellulosic bio-feedstocks. Through the participant brewery, supply chain costs were calculated (including logistics and freight) to be $43.20⁷ per ton; comparable supply chain costs were not able to be calculated for these other cellulosic crops, but it was assumed that these decen- tralized sources of “farms and fields” will lead to significantly higher logistics and freight costs, given a lack of accessibility to rail spurs in the forests and fields, as well as a lack of concentration (in comparison to the top two brewers in the US).

From the analysis of “BSG acquisition” as a biotechnological feedstock, the following conclusions are sup- ported:

1) Assuming 10% moisture content for other lignocelluloses, and 70% for BSG, the effective acquisition cost (in dry tons) for BSG is $134.10 in comparison to a range of $57.2 - $171⁸. This is assuming that the beer manufacturer will sell the co-product “as-is”.

2) If BSG co-product is assumed to be shipped “as-is”, the supply chain benefits of economies of scale (large US brewers) in well situated logistics origin points is negated through higher moisture rates (70% versus 10%), leading to a lower lignocellulose yield once dried.

3) On an aggregate supply chain standpoint, BSG is not found to be an optimal source of lignocellulose as a potential bio-feedstock.

As a raw material, the carbohydrate composition of barley is 65%, and is effectively reduced (via the malting process) to 49%. This potential feedstock has a lower carbohydrate yield than other lignocellulose crops, with a high moisture rate. Once this material is dried, the net result for our study is 504,000 dry tons (or 30% of the to- tal annual BSG). Starting with a BSG feedstock of 504,000 dry tons a year, a lignocellulosic “technical flow” is utilized from van Wyk (2001), shown in Figure 2, that identifies a “separation/conversion” process from raw feedstock to end products⁹ [37] [38] .

Given that every separation/conversion process is different, and that technical researchers often focus on spe- cific materials, this technical flow framework enabled the lignocellulose technical experts to assist in the general understanding of a conversion/separation process for BSG into useful by-products of sugars, lipids, lignins, and proteins. From research, conversions costs (shown in Table 3) were found of various lignocelluloses into etha- nol for agricultural/forest feedstocks to compare to this BSG industrial residual.

To determine the conversion and separation processes, costs, and yields, a combination of extant literature and responses from the industry expert participants was obtained. From the industry expert participants, existing engineering models were used based upon other lignocelluloses, and completed a ratio between the model’s analytical relationship between saccharide content and conversion costs, and the saccharide content of BSG (49.1%) to calculate its sugar conversion costs, the study was based upon a lignocellulose with a saccharide weighting of 67.9%, which is in the range of materials that was illustrated in Table 2. From lipids, lignins, and proteins, engineering studies were utilized, and a pro-rated cost of 15.8 cents a pound ($316/DT) was calculated for these materials.

Table 4 was developed from an engineering model from one of the industry experts who processes lignocel- luloses commercially in the biotechnology field. Most of these calculations were based upon ratios from other lignocelluloses processed, as this entity did not have direct experience in processing BSG as a feedstock. The end result of the conversion process is assumed to be sugars not to same specification as raw or refined sugar, but rather specialty sugars of xylose, arabinose, galactose and mannose, which we are estimating at an 80% pur- ity level. In Table 4, the net conversion cost per dry ton was calculated in using the conversion cost and yield from interviews with ligncellulosic experts, and factored in the yield of dry mass (30%) and percent of the ma- terial in BSG. Based upon the lower percentage of sugars in BSG, the conversion cost of this material was as- sumed to be higher than that of other ligncellulosic feedstocks, as shown in Table 3 for bioethanol. On a cost per liter basis, assuming 1000 liters to a metric ton, the saccharification process (not including fermentation and distillation) for BSG is 26.8 cents/liter.

Based upon this technical analysis, the conversion process of BSG to useful bio-materials was found to be in- efficient in contrast to that assumed to be for alternative bio-feedstocks.

In our discussion with the participant brewer, it was found that there is no established market function for BSG, as there exists for conventional feedstocks, such as in the petrochemical industry. By-product revenue is deter- mined on a site by site basis without any established market price, and is often primarily purposed to dispose of these materials from the production process. Therefore, it is assumed that no centralized commodity market en- vironment exists, and all buy-sell agreements are completed on an ad hoc basis with a limited or non-exist an- treference to a hypothetical market price.

In the structured interview process with biotechnology experts, a “hub and spoke” model to integrate potential sellers and buyers was conceptualized, as is shown in Figure 3. In this wide area network approach to industrial symbiosis, large suppliers of BSG could build an economical scale future state “hub and spoke” model with a bio-refinery that focuses on BSG. From this consolidated model of scale and production, the bio-refinery would sell these materials on an established “biomass market” at market prices to biomass buyers, who will re-manu- facture these materials into useful products.

Through involvement from lignocellulosic researchers and commercial biotechnology entities, Table 5 pro- vides a pro-forma “business model” for this hypothetical bio-refinery achieved through industrial symbiosis between this bio-refinery and the two largest beer manufacturers in the US, who possess 70% of the market (thus 70% of BSG). Starting with the volume yields from Table 5, a baseline acquisition cost of $40 a ton, cal- culations were made to establish a “dry ton market value” (since BSG is often sold “as-is”). Using this baseline price of $40 a ton, the acquisition costs were adjusted to tie to the gross tonnage of BSG that was brought into the bio-refinery versus the weight of the finished materials, which is 22.65% of the gross BSG tonnage. The net cost is associated with the processing cost of the BSG to convert and separate to the materials as shown in Table 5,

including the drying and storage process of BSG. Next, the capital costs were calculated assuming $80 million cost to build the bio-refinery (including land acquisition), and a 30-year depreciation schedule, with depreciation costs tied to annualized volume yields. Logistics costs assume a rail spur from each brewery to the bio-refinery, with an average 1000 mile distance, and FOB costs (per railcar and fuel surcharge) as gathered from the partici- pant brewery. The net profit is assumed to be 5%, with the conventional biochemical industry at a margin be- tween 2% - 3%, at much higher volumes of production. From this, an average cost per ton and pound was cal- culated in order to compare these fully loaded costs to the market costs of these today, as will be discussed be- low.

From this analysis shown in Table 5, a “processed BSG market price” of $179.29 must be achieved, in con- trast to the current fetch price of $40 a wet ton. Therefore, the use of BSG as a biomass feedstock would require a 450% improvement from its current buyer-seller relationship, which does not appear to be market justifiable.