On-line near infrared monitoring of glycerol-boosted anaerobic digestion processes: Evaluation of process analytical technologies

被引:76
作者
Holm-Nielsen, Jens Bo [1 ]
Lomborg, Carina Juel [1 ]
Oleskowicz-Popiel, Piotr [1 ]
Esbensen, Kim H. [1 ]
机构
[1] Aalborg Univ, Esbjerg Inst Technol, Appl Chemometr Analyt Chem Acoust Chemometr Appl, DK-6700 Esbjerg, Denmark
关键词
anaerobic digestion (AD); near infrared spectroscopy (NIR); on-line measurement; volatile fatty acids (VFA); glycerol; process analytical technologies (PAT);
D O I
10.1002/bit.21571
中图分类号
Q81 [生物工程学(生物技术)]; Q93 [微生物学];
学科分类号
071005 ; 0836 ; 090102 ; 100705 ;
摘要
A study of NIR as a tool for process monitoring of thermophilic anaerobic digestion boosted by glycerol has been carried out, aiming at developing simple and robust Process Analytical Technology modalities for on-line surveillance in full scale biogas plants. Three 5 L laboratory fermenters equipped with on-line NIR sensor and special sampling stations were used as a basis for chemometric multivariate calibration. NIR characterisation using Transflexive Embedded Near Infra-Red Sensor (TENIRS) equipment integrated into an external recurrent loop on the fermentation reactors, allows for representative sampling, of the highly heterogeneous fermentation bio slurries. Glycerol is an important by-product from the increasing European bio-diesel production. Glycerol addition can boost biogas yields, if not exceeding a limiting 5-7 g L-1 concentration inside the fermenter-further increase can cause strong imbalance in the anaerobic digestion process. A secondary objective was to evaluate the effect of addition of glycerol, in a spiking experiment which introduced increasing organic overloading as monitored by volatile fatty acids (VFA) levels. High correlation between on-line NIR determinations of glycerol and VFA contents has been documented. Chemometric regression models (PLS) between glycerol and NIR spectra needed no outlier removals and only one PLS-component was required. Test set validation resulted in excellent measures of prediction performance, precision: r(2) = 0.96 and accuracy = 1.04, slope of predicted versus reference fitting. Similar prediction statistics for acetic acid, iso-butanoic acid and total VFA proves that process NIR spectroscopy is able to quantify all pertinent levels of both volatile fatty acids and glycerol.
引用
收藏
页码:302 / 313
页数:12
相关论文
共 22 条
[1]  
AHRING BK, 1995, APPL MICROBIOL BIOT, V43, P559, DOI 10.1007/BF00218466
[2]  
Amon Th., 2006, INT C SERIES, V1293, P217, DOI DOI 10.1016/J.ICS.2006.03.007
[3]  
ANDREE H, 2005, BALT BIOR S 26 28 MA, P141
[4]  
Angelidaki I, 2003, ADV BIOCHEM ENG BIOT, V82, P1
[5]  
Angelidaki I, 2002, ENV BIOTECHNOLOGY, P1
[6]  
Bakeev KA, 2005, PROCESS ANALYTICAL TECHNOLOGY: SPECTROSCOPIC TOOLS AND IMPLEMENTATION STRATEGIES FOR THE CHEMICAL AND PHARMACEUTICAL INDUSTRIES, P1, DOI 10.1002/9780470988459
[7]  
BOE K, 2006, THESIS DTU
[8]   New heterogeneous process for biodiesel production: A way to improve the quality and the value of the crude glycerin produced by biodiesel plants [J].
Bournay, L ;
Casanave, D ;
Delfort, B ;
Hillion, G ;
Chodorge, JA .
CATALYSIS TODAY, 2005, 106 (1-4) :190-192
[9]   Continuous determination of volatile products in anaerobic fermenters by on-line capillary gas chromatography [J].
Diamantis, V. ;
Melidis, P. ;
Aivasidis, A. .
ANALYTICA CHIMICA ACTA, 2006, 573 :189-194
[10]  
Esbensen KH., 2001, MULTIVARIATE DATA AN