• Ei tuloksia

Effect of pretreatments on methane and enzymatic hydrolysis

M = months, FA = formic acid, and n.d. = not determined

4.3 EFFECT OF PRETREATMENTS ON METHANE AND ENZYMATIC HYDROLYSIS YIELDS (I,III)

All the studied pretreatments led to clear changes in the chemical composition of hemp. (Table 11, III: Table 1). Steam explosion solubilized xylan but had less effect on lignin, although the lignin structure was likely modified. Presoaking in acid prior to steam explosion increased the solubilization of hemicelluloses. The alkaline treatment significantly increased the share of glucans by releasing lignin, hemicelluloses, and pectin. Xylan was solubilized to some extent. The amount of solubilized fractions during the pretreatments varied from 8% to 40%

(Table 11). The amount of cellulose decreased in acid-presoaked material, while the amount remained intact in other treatments. The enzymatic conversion of hemp showed strong correlation with the amount of pectin (III: Figure 2).

Table 11 Chemical composition of the studied hemp materials (fresh and pretreated) expressed as % of DM. Standard deviation is in parenthesis (n= 3).

Fresh SE SE with

acid Alkali treated

% of dry matter

Glucan 46.1 (1.3) 55.4 (0.4) 64.6 (1.6) 83.6 (1.2) Xylan 9.5 (0.1) 4.4 (0.1) 1.8 (0.1) 8.4 (0.2) Arabinan 1.2 (0.1) b.l.d. b.l.d. b.l.d.

Mannan 2.2 (0.1) 1.7 (0.1) b.l.d. b.l.d.

Galactan 1.6 (0.1) b.l.d. b.l.d. b.l.d.

Galacturonic acid 5.9 (0.6) 1.3 (0.1) b.l.d. b.l.d.

Lignin 18.0 (0.3) 19.8 (0.4) 29.6 (1.4) 7.2 (0.4) Total identified components 84.6 82.7 95.4 99.3

Loss of organic material 7.9 18.6 39.7 35.0 SE = steam explosion, b.l.d. = below detection limit (< 0.5% of DM).

The changes in the hemp structure were most significant after alkali treatment (Figure 14B) (III), visualized by SEM, and were comparable to modifications observed after pectin removal in enzymatic hydrolysis (Figure 10C).

Figure 14 Electron microscopy images of fresh hemp after extraction with water or with 1% NaOH. A: Bast fibers extracted with water; B: Bast fibers extracted with 1% NaOH; C: Woody layers extracted with water; and D: Woody layers extracted with 1% NaOH. The magnification of A = 2000, B = 400, and C and D = 100.

The visualized release of fiber bundles was efficient, and the effect on the wood layer part was remarkable. The exposed wood layer structures, spiral tracheids, became clearly visible after treating with NaOH (Figure 14D). Such clear visible changes were not observed in steam-exploded materials (III).

Steam explosion and alkaline treatment had a desired effect on the conversion of carbohydrates in hemp. The conversion of both glucan and xylan in standard enzymatic hydrolysis increased after all pretreatments (Figure 15). The steam explosion, however, was improved after soaking in 2.5% H2SO4 prior to the treatment and resulted almost to a theoretical yield in the enzymatic conversion.

However, the loss of carbohydrates was the highest as well.

Figure 15 Conversion of glucan and xylan to sugars in enzymatic hydrolysis of fresh, steam-exploded with and without H2SO4, and alkali-treated hemp, expressed as % of DM. Bar indicates ± one standard deviation of mean, n = 3.

0 % 20 % 40 % 60 % 80 % 100 %

Total Conversion Total Conversion Total Conversion Total Conversion

% of dry matter

Glucan Xylan

Fresh SE SE with acid Alkali

treated

0

Ensiled SE Fresh SE Ensiled Fresh Maize

Ensiled SE Fresh SE Ensiled Fresh Reduced particle size Hemp

Steam explosion was conducted also on fresh hemp and ensiled hemp as well as on fresh and ensiled maize. These samples were not washed prior to methane production or enzymatic hydrolysis. The results showed that the steam explosion improved the methane conversion of fresh crops (Figure 16), but impaired the conversion of ensiled crops (Figure 16), at least during the early stages of the AD. It could also be clearly observed that ensiling increased the methane yields of both crops more than the steam pretreatment (Figure 16). The effect of steam explosion and ensiling was clearer for hemp than maize, indicating the more recalcitrant structure of this material, requiring pretreatment.

Steam explosion resulted in increased hydrolysis yields, especially on hemp, and only the steam explosion of fresh maize decreased the amount of WSC, leading to the reduction of total fermentable sugars. Steam explosion decreased the amount of WSC in formic acid-ensiled maize as well, but the yield was compensated with the increased conversion of the structural carbohydrates. The effects of steam explosion on maize, however, were less remarkable than on hemp due to the softer, more accessible structure of the raw material, whereas the steam exploson increased the yields of all hemp samples, fresh and ensiled (Figure 17).

Figure 16 Methane yields of fresh and ensiled maize and hemp, without or after steam explosion (SE) after 30 days of AD at 35° C. Results are expressed as Ndm3 kg-1 TS. CH4 yield of inoculum is subtracted from the sample yields. Bar indicates ± one standard deviation of mean, n = 8.

Figure 17 Conversion of carbohydrates as reducing sugars in fresh and ensiled maize and hemp, without or after steam explosion, to sugars in standard enzymatic hydrolysis, expressed as % of DM.

The WSC indicates the amount of water-soluble carbohydrates in the samples; SE = steam exploded. Bar indicates ± one standard deviation of mean, n = 3.

In addition to steam explosion and alkaline pretreatment, the effect of milling as a physical pretreatment method was examined to enhance the accessibility of enzymes in the hydrolysis and AD and thus to increase the conversion of carbohydrates to sugars. Milling of fresh hemp and maize to fine particles of the size of about 1 mm (instead of chopped 10-20 mm) was carried out. The reduction of particle size, however, had no effect on the hydrolysis yield of maize, whereas milling significantly increased the conversion of polymers into sugars in hemp from 25% to 35% of DM (I: Figure 2) and the methane yield from 218 Ndm3 kg-1 TS-1 to 263 Ndm3 kg-1 TS-1 (Figure 16, I).

0 % 10 % 20 % 30 % 40 % 50 % 60 %

Fresh Fresh SE Ensiled Ensiled SE Ensiled with FA Ensiled with FA SE Fresh Fresh SE Ensiled Ensiled SE Ensiled with FA Ensiled with FA SE

% of dry matter

WSC Conversion

Maize Hemp

5 DISCUSSION

5.1 FRESH MAIZE, HEMP, FABA BEAN, WHITE LUPIN,