Identification and Quantitation of Potential Fermentation Inhibitors in Biomass
Pretreatment Hydrolysates Using High Performance Liquid Chromatography in
Combination with Ultraviolet Detection and Tandem Mass Spectrometry
Lekh Nath Sharma, Ph.D.
Mentor: C. Kevin Chambliss, Ph.D.
Conversion of lignocellulosic biomass to ethanol involves pretreatment of
biomass materials to increase the accessibility of carbohydrates in lignocellulose for
enzymatic hydrolysis prior to ethanol fermentation. The pretreatment hydrolysate
contains not only cellulose and fermentable sugars but also a wide variety of degradation
products, such as aliphatic and aromatic acids, aromatic aldehydes, and phenols. These
degradation products exert an inhibitory effect on downstream microbial processes,
reducing the overall efficiency for bioconversion of lignocellulosic materials to ethanol.
Therefore, development of a reliable quantitative analysis method for individual
degradation products is critical in order to advance a more fundamental understanding of
lignocellulose pretreatment as well as subsequent microbial processes.
Various analytical techniques have been applied to analyze degradation products
in hydrolysates, such as gas chromatography-mass spectrometry (GC-MS), GC-flame
ionization detection (GC-FID), and liquid chromatography (LC) with ultraviolet (UV) or
refractive index (RI) detection. Difficulties in derivatizing samples of unknown
composition for GC methods and incomplete analyte resolution in LC experiments have
caused researchers to typically employ LC methods targeting a limited group of analytes
(e.g., a single analyte class) in quantitative work. To address these limitations, we have
developed an improved high performance liquid chromatography (HPLC) method
utilizing photodiode array (PDA) and tandem mass spectrometry (MS/MS) detection.
The novel HPLC-PDA-MS/MS method enables simultaneous identification and
quantitation of 40 degradation products in hydrolysate samples in a single, 60-minute run.
Because a unique MS/MS transition is monitored for 37 of 40 target analytes, this
approach essentially alleviates resolution limitations of our previous HPLC-UV
approach. Upon successful development of the method, it was further applied to analyze
various biomass pretreatment samples.
Even though coeluting components do not interfere with the detection of analytes
using LC-ESI-MS/MS, they can significantly influence the ionization efficiency of
analytes at the ionization source. Thus, a series of experiments w