S. latissimi, an abundant macroalgae found in large numbers along the Norwegian west coast, was collected from Trondheimsfjord (N 63˚26'56", E 10˚10'48") near Trondheim, Norway in August of 2017. The macroalgae were subsequently washed using tap water to remove particulates from the surface. S. latissima was then milled using a tabletop blender with 10 mL of deionized water per 1 kg of macroalgae to produce a dense macroalgae pulp. The pulp was dried for 48 h at 30˚C and then stored in airtight plastic bags in a dry location for further use.

Commercial β-glucanase (G4423) from Trichoderma longibrachiatum (Sigma Aldrich, Germany) was used during enzymatic hydrolysis. This was an enzymatic mixture of β-1-3/1-4-glucanase, xylanase, cellulase, β-glucosidase, β-xylosidase, α-˪-arabinofuranosidase and amylase activities. The macroalgae pulp (10% w/v) was suspended in 0.15 M sodium carbonate (Na₂CO₃) solution to a volume of 25 L at a starting pH of 9.0 in a stirred tank (CE640, Gunt, Germany) for 2 h at 50˚C. After 2 h, the pH was adjusted to 6 using HCl acid solution. Then 5 mg of enzyme mix per g of macroalgae dry weight was added to the solution, and left in the stirred tank for 48 h. Samples were taken every 12 h and analyzed for glucose concentration using a hexokinase glucose assay kit (GAHK20-1KT, Sigma, Germany).

Total carbohydrates, reducing sugar and glucose content were both determined by acidic treatment of pre-hydrolysis dry biomass. Biomass (0.5 g) was treated with 5 mL of 72% (v/v) H₂SO₄ at room temperature for 30 minutes with constant stirring via a magnetic stirrer. The sample was then diluted to a volume of 50 mL with deionized water, then autoclaved at 121˚C for 30 minutes. Once cooled, NaOH was added to the sample to reach a pH of 7.5. The total carbohydrates in the sample were then determined by using a phenol-sulfuric acid method [29] . Reducing sugars in the sample were determined using a dinitrosalicylic acid method [30] . Glucose content of the hydrolysate was determined using a hexokinase glucose assay kit (GAHK20-1KT, Sigma, Germany).

Ethanol concentration was determined using spectrophotometric measurements at 267 nm in a potassium dichromate and perchloric acid solution [31] .

Saccharomyces cerevisiae was chosen for the fermentation experiments as it is a well understood fermentative organism. The organism was supplied with the fermentation equipment (Gunt, Germany). The 25 L of hydrolysate was used in the fermentation tank as the substrate (CE640, Gunt, Germany). 1 g of yeast per L of hydrolysate was added, along with 0.3% (w/v) of yeast extract and 1% (w/v) peptone and adjusted to a pH of 6.8 to support yeast growth. Fermentation was performed at 30˚C for 48 hours. The reducing sugar and ethanol content of the fermented hydrolysate was measured at 12 h intervals throughout the fermentation. Samples were centrifuged at 10,000 × g for 15 minutes at 4˚C, and the supernatant removed for analysis. Reducing sugars in the supernatant were determined using a dinitrosalicylic acid method [30] .

All experiments within this study were conducted in triplicate with the results displayed as mean values ± the standard deviation.

The total carbohydrate, reducing sugar and post-hydrolysis glucose yield was determined using samples collected in August of 2017 (Table 2). Before enzymatic hydrolysis, the total carbohydrates were 58% ± 2.6% of the dry weight of S.

latissima. The reducing sugar content was 37% ± 1.1% of the dry weight of S. latissima, whereas content was 11% ± 1.2% of the dry weight of S. latissima. This is within the expected range of carbohydrate composition based on similar research of S. latissima [32] , including recent research undertaken using S. latissimi from Trondheimsfjorden [33] [34] .

The enzymatic hydrolysis of S. latissima was undertaken to hydrolyze the storage carbohydrates into reducing sugars during a two h macroalgae lysis step, followed by a 48 h enzymatic hydrolysis step for later use in fermentation. An enzymatic mixture of β-1-3/1-4-glucanase, xylanase, cellulase, β-glucosidase, β-xylosidase, α-˪-arabinofuranosidase, and amylases was used. The macroalgae pulp was suspended in 0.15 M sodium carbonate (Na₂CO₃) solution at a starting pH of 9.0 in a stirred tank. Measurements of the reducing sugar concentration were obtained every 12 h throughout the hydrolysis process.

The initial macroalgae lysis step liberated 9.18 ± 1.21 g/L of reducing sugar. The following enzymatic saccharification during the hydrolysis process yielded a further 22.13 ± 1.43 g/L after 48 h (Figure 2; Table 3). This resulted in a total reducing sugar concentration of 31.31 ± 1.73 g/L after the hydrolysis was complete. With the total reducing sugar content observed to be 37 ± 1.1 using the dinitrosalicylic acid method [30] before hydrolysis, the calculated efficiency of reducing sugar release from the macroalgae lysis and enzymatic hydrolysis process was calculated to be 85%.

The maximum rate of enzymatic saccharification of the macroalgae was observed after 2 h of incubation (Figure 3; Table 4), with a gradual decline after that, a typical saccharification efficiency relationship observed in other similar studies [19] [35] [36] . It has been speculated that this decline in hydrolysis rate could be the result of inhibition of the enzymes by the products glucose and cellobiose [19] [36] [37] .

The fermentation of S. latissima was performed at 30˚C for a period of 48 h. The hydrolysate was used as the bioethanol fermentation substrate. The yeast Saccharomyces cerevisiae was used for the fermentation process, with the maximum theoretical ethanol yield of 0.51 g per g of reducing sugar. Measurements of the reducing sugar and ethanol (Figure 4) concentration were obtained every 12 h throughout the fermentation process (Table 3). The maximal ethanol concentration

was reached after 36 h, at 13.02 ± 0.61 g/L. With the initial glucose concentration of 31.31 ± 1.73 g/L, the maximum ethanol yield from S. latissima was 0.42 g ethanol per 1 g of reducing sugar, 84% of the theoretical yield.

The yield observed in this study is amongst the higher known yields as compared to other observations made when using macroalgae as the bioethanol substrate (Table 5). Furthermore, since the experiments were undertaken at a large volume (not batch tests), these results display the potential for such bioethanol production up-scaling to industrial levels. After the 36 h point, the ethanol yield was observed to decline very slightly, and this may be due to the metabolism of the yeast strain, which can consume ethanol [35] [36] [37] [48] .

The ethanol yield observed in this study is comparable with the ethanol yields observed in lignocellulosic materials. Maize was observed to produce 0.48 g ethanol per 1 g of glucose [49] , Prosopis juliflora 0.49 g ethanol per 1 g of glucose [35] , Lantana camara 0.48 g ethanol per 1 g of glucose [36] , and newspaper waste 0.39 g ethanol per 1 g of glucose [37] .

Our results from S. latissima fermentation observed when extracting with sodium carbonate that by using 1 kg of wet S. latissima as the initial biomass, the amount of reducing sugars available was 31.3 g, which can then be fermented into ~ 13 g of ethanol. This carbohydrate-rich macroalgae could be used as the

raw material for bioethanol production, opening up further economic opportunities in aquaculture. By using macroalgae for the production of bioethanol, the requirement for fresh water, fertilizers and agricultural land for bioethanol production will significantly reduce. Furthermore, Nordic countries will be able to produce bioethanol locally, previously unfavorable due to the limited amount of agricultural land available, from naturally occurring S. latissima.

The cost of processing macroalgae for ethanol production can be kept low by employing cost-effective processing methods as used in this study; however, the cost of harvesting large quantities of macroalgae, as well as its delivery to the fermentation plant, are still significant barriers that require attention for the implementation of this technology at an industrial scale.