Extraction of spruce hemicelluloses. J. V. Rissanen, H. Grénman, D. Yu. Murzin, T. Salmi

Transcription

1 Extraction of spruce hemicelluloses J. V. Rissanen, H. Grénman, D. Yu. Murzin, T. Salmi

2 Hemicellulose molecule

3 Long chained hemicelluloses Further processing 1 kda < Emulsifiers, Films etc. Extraction H 2, T Filtration < 1 kda Medium and short chained hemicelluloses Platform chemicals xidation Hydrogenation Fermentation Esterification Aqueous reforming Sugar acids, Sugar alcohols, Lubricants, Fuels etc.

4 Background Protonation of glycosidic bonds Wood. H H H H H H H H H H H. H H H H H H H H H 2 H H H H H [H Hydrolysis + ] H. HCl H 2 H H H H H H H H H + H H. H H H H

5 Starting material Non-extracted chips, mm Non-extracted cubic blocks, 1 mm

6 Collector vessel Reactor system PI TIC PI FI TIC TIC TIC TIC TIC Cascade reactor Flexible 5 point sampling of liquid and solid phases Accurate control and measurement of temperature, pressure and flow rate

7 Batch reactor versus cascade reactor Traditional batch digester Cascade reactor The chips do not differentiate between reactor type

8 Liquid samples, 17 C, -6 min 5 min 1 min 15 min 2 min 6 min The viscosity of the extract increasing The color becomes brown

9 Solid samples, 17 C, -6 min min 5 min 1 min 15 min 2 min 6 min

10 c / ( g / dm 3 ) c (g / dm 3 ) Experimental results The influence of temperature and chip size on the overall extraction rate 1,2 1,,8,6,4,2, T 17 C 16 C 15 C 14 C 13 C 12 C ,2 16 C, chip size 1,25-2, mm 1,8 16 C, chip size 1 mm cubic,6 13 C, chip size 1,25-2, mm,4,2 13 C, chip size 1 mm cubic Concentration of hemicellulose as a function of time (1 mm cubic blocks). Liquid phase concentration as a function of time with different chip sizes.

11 The behaviour of ph during extraction 6 5,5 T 6,5 6 T 5 ph 4,5 4 3,5 16 C 17 C C 14 C 13 C 12 C 5,5 ph 5 4,5 4 3,5 16 C 17 C 15 C 14 C 12 C 13 C ,25 2, mm wood chips 1 mm cubic wood blocks

12 Influence of chip size on ph The behavior of the ph during the reaction at four different temperatures 6, C 13 C 5, C 17 C ph 5,5 5 1mm cubic ph 4,5 4,5 1,25-2mm 1mm cubic ,25-2mm 3, The ph with 1,25 2 mm chips (open symbols) and 1 mm cubic blocks (solid symbols). The ph with 1,25 2 mm chips (open symbols) and 1 mm cubic blocks (solid symbols).

13 H 3 + vs total hemicellulosic sugars The diffusion of the acetyl groups is clearly faster than for the hemicelluloses i.e. higher ch 3 + is obtained at the same conversion for the larger chips

14 Modelling rate k k I k I I c e 1 ( H ) n solid E RT ai n 2 n 1 1 n 2 rate k I c solid T 1 T 1 T mean 1,25-2, mm chips 1 mm cubic blocks T mean 15 C Estimated parameters Estimated Std Error (%) Estimated parameters Estimated Std Error (%) E a [kj / mol] 122,3 12,2 k I [(L (n1+n2)-1) / (g (n1+n2)-1 min)] 9,57 1-3,3 5,55 1-3,3 The model of the complicated system can be simplified significantly

15 Sensitivity analysis and parameter correlations (E a and k I ) 1,25 2, mm chips k I 1 mm cubic blocks k I k I k I

16 Modelling results Fit of the model to experimental data (liquid-phase hemicellulose concentrations c) for the smaller chip size ( mm). Fit of the model to experimental data (liquid-phase hemicellulose concentrations c) for the larger chip size (1 1 mm).

17 So what did we learn from modelling exercise? Extraction and hydrolysis rate very temperature sensitive The chip size influences the overall extraction rate and the molar mass of the obtained hemicelluloses The extraction rate is not significantly influenced by ph The mathematical model for the complex system could be simplified significantly A good fit of the model to the experimental data was achieved Rissanen, Grénman, Willfor, Murzin, Salmi, Industrial & Engineering Chemistry Research (214), 53(15), Rissanen, Grénman, Xu, Willför, Murzin, Salmi, ChemSusChem 214, accepted, DI 1.12/cssc R1 Rissanen, Grénman, Xu, Willför, Murzin, Salmi, Cellulose Chemistry and Technology, submitted after revision

18 Mw ( g / mol ) 1 3 Molar mass of extracted hemicelluloses Molecular weight as a function of time for different temperatures and chip sizes 12 C 13 C 14 C ,25-2 mm 4 1,25-2 mm 3 1,25-2 mm 1 mm cubic mm cubic 1 mm cubic Mw ( g / mol ) 1 3 Mw ( g / mol ) 1 3 Mw ( g / mol ) C 16 C 17 C 5 8 1,25-2 mm 4 1,25-2 mm 6 3 1,25-2 mm 2 1 mm cubic 4 1 mm cubic 1 mm cubic Mw ( g / mol ) 1 3 Mw ( g / mol ) 1 3

19 Molecular weight as a function of conversion for different temperatures and chip sizes 12 C 13 C 14 C Mw ( g / mol ) ,25-2 mm 8 1 mm cubic Conversion / % Mw ( g / mol ) ,25-2 mm 1 mm cubic Conversion / % Mw ( g / mol ) ,25-2 mm 1 mm cubic Conversion / % 15 C 16 C 17 C Mw ( g / mol ) ,25-2 mm 1 mm cubic Mw ( g / mol ) ,25-2 mm 1 mm cubic Mw ( g / mol ) ,25-2 mm 1 mm cubic Conversion / % Conversion / % Conversion / %

20 Percentage of sugars in the extract, C

21 Conclusions Industrial applications need long chained hemicelluloses and monomers The extraction kinetics is crucial in order to tailor the product as well as the production The overall extraction is influenced by several interlinked phenomena Rissanen, Grénman, Willfor, Murzin, Salmi, Industrial & Engineering Chemistry Research (214), 53(15), Rissanen, Grénman, Xu, Willför, Murzin, Salmi, ChemSusChem 214, accepted, DI 1.12/cssc R1 Rissanen, Grénman, Xu, Willför, Murzin, Salmi, Cellulose Chemistry and Technology, submitted after revision

22 Conclusion The influence of the extraction parameters is complex and interlinked Solid-liquid ratio Temperature Detachment of compounds Chip size Hydrolysis verall kinetics of extraction Concentration in liquid phase ph Morphology Internal diffusion