BACKGROUND Lignin is an important component of marine organic carbon. It is also an important biomarker for extracting information on the evolution of the land and marine environment and tracking the source of organic marine matter. However, the existing analytical techniques are difficult to determine lignin directly. So, the content of phenolic compounds in the decomposition products of lignin in marine sediments were generally determined to indicate the content of lignin and the source of organic matter. The content of phenolic compounds in the decomposition products of lignin in marine sediments is often used to reflect the content of lignin. In addition, by calculating the diagnostic ratio of individual phenolic compounds, it also provides important information about the classification, source, and diagenesis of terrestrial organic matter in marine sediments. However, phenolic compounds in the decomposition products of lignin have the characteristics of strong polarity and low volatility, so they cannot be directly detected by gas chromatography and need to be derivatized first, which makes the sample processing complicated and often results in incomplete derivatization. Therefore, it is of great significance to develop a simple and reliable method for determination of phenolic compounds of the lignin decomposition products in marine sediments to explore the source of organic matter and understand the environmental evolution process.

OBJECTIVES To establish a simple and reliable method for the determination of phenolic compounds of lignin decomposition products in marine sediments using solid phase extraction (SPE) combined with ultra-high performance liquid chromatography-high resolution mass spectrometry, and to trace the content level and source of lignin in the sediments of Laizhou Bay in China.

METHODS Marine sediment samples were first decomposed with oxidative-alkaline CuO and extracted by solid phase extraction. Briefly, the oxidation was carried out in a polytetrafluoroethylene digestion tank. 1.00g of sediment sample, 500mg of copper oxide, and 100mg of ammonium ferrous sulfate were accurately weighed and placed in the tank. The components were thoroughly mixed with the sample and then the digestion tank was transferred to a glove box filled with nitrogen. 8.0mL of aqueous sodium hydroxide solution with a concentration of 8.0% (bubbled with N₂ to remove dissolved oxygen) was added to the tank. The digestion tank was covered tightly and transferred to an oven heating to 150℃ for reaction, which was terminated after 3h. After the digestion tank cooled to room temperature, it was carefully unscrewed, and an internal standard (ethyl vanillin) solution was added. Subsequently, the hydrolysate was transferred to a centrifuge tube, spun at 8000r/min for 10min, and the supernatant and reaction residue was separated. 2.0mL of 1.0% sodium hydroxide solution was added to rinse the residue, and centrifuged at 8000r/min for 10min. Combining the centrifuged supernatant obtained twice, the solution was acidified to pH=1 with hydrochloric acid. After the solution was left to stand for 30 minutes, solid phase extraction was performed.　　The SPE procedure was as follows: A hydrophilic-lipophilic balance (HLB) SPE cartridge (200mg, 6mL) was conditioned with 5mL of methanol and 5mL of ultrapure water. Sample solution was passed through the cartridge in a flow rate 1.0mL/min, and then the cartridges were rinsed with 10mL water, and dried under vacuum for about 3min. Phenolic compounds were eluted with 10mL ethyl acetate, and were evaporated by a rotary evaporator, reconstituted with sample solvent. Then, ultra-high performance liquid chromatography using ZORBAX Eclipse XDB-C18 column with packing particle size of 1.8μm was used to directly separate all target compounds at 28℃, with gradient elution. The mobile phase was composed of ultrapure water with 0.1% formic acid (V/V) and acetonitrile/methanol (9:1, V/V) , and the flow rate was set to 0.25mL/min. Electrospray ionization (in positive) time of flight mass spectrometry was applied to detect target compounds in full scan mode, and quantification was performed using an internal standard determination.

RESULTS Firstly, chromatographic conditions and solid phase extraction conditions were systematically optimized. Ultra-high performance liquid chromatography was used for the chromatographic separation of phenolic compounds from lignin decomposition products in marine sediments. The separation effects of three mobile phase systems, namely, water-acetonitrile, water- methanol, and water-methanol-acetonitrile, were compared. When using a water-methanol -acetonitrile ternary mobile phase system, the resolution of various phenolic compounds was superior to the commonly used water-acetonitrile or water-methanol binary mobile phase systems in the literature. In addition, the effects of mobile phase acidity (trifluoroacetic acid, formic acid, and acetic acid were added into the mobile phase) on the separation of various phenolic compounds were investigated. The results showed that adding a certain concentration of all three acids to the mobile phase provided better separation results. Considering the compatibility with mass spectrometry, it was finally determined that adding 0.1% formic acid into the mobile phase achieved good peak patterns and resolution.　　In order to determine the ionization mode suitable for the analysis of phenolic compounds from lignin decomposition products in marine sediment, electrospray ionization (ESI) mass spectrometry was performed on each target phenolic compound in ESI⁺ and ESI⁻ mode, respectively. Under ESI⁺ mode, various target phenolic compounds were less affected by interfering substances in the sample matrix, and the MS response value for most of the phenolic compounds was higher than that found in ESI⁻ mode. Hence, ESI-TOF/MS in positive mode was selected to determine phenolic compounds of lignin decomposition products in marine sediment. Subsequently, the fragmentation voltage was optimized to obtain the highest sensitivity for all target phenolic compounds, which was the main mass spectrometric condition that affected the quantification accuracy and sensitivity. The effect of fragmentation voltage on the MS response signal of each target phenolic compound was investigated in the range of 80V to 200V. Overall, considering the detection sensitivity of the M+H⁺ ion peak of each target compound, 130V was selected as the optimal fragmentation voltage to determine phenolic compounds of lignin decomposition products in marine sediment.　　The effect of pH (1.0-2.5) of the loading solution for solid phase extraction on the extraction efficiency of various target phenolic compounds was systematically investigated, to ensure that the phenolic compounds of lignin decomposition products in marine sediments have a good recovery rate during the SPE process. When the pH of the loading solution was 1.0 and 1.5, the recovery rate of various phenolic compounds by using HLB solid phase extraction column was significantly higher than that of the loading solution adjusted pH to 2.0 and 2.5. When the pH of the sample solution was 1.0 and 1.5, although the recoveries of syringaldehyde and acetovanillone were relatively similar, the recoveries of other phenolic compounds were the highest at a pH of 1.0. Considering the recovery rate of all the target phenolic compounds and applicability of the method, the pH of the sample solution was confirmed to adjust to 1.0. In this study, HLB SPE column with 200mg of packing material was used to enrich phenolic compounds in sample extraction solution. Generally, 5-10mL of eluting solvent can ensure the full elution of all target phenolic compounds adsorbed on the SPE column. Therefore, based on the results of literature research, ethyl acetate was finally selected as the eluting solvent, with a dosage of 10mL.　　Under the optimum experimental conditions, the 11 main decomposition phenol compounds of lignin in marine sediments were well separated within 20 minutes. The proposed method had good precision (RSD was less than 9.0%), the correlation coefficient (R²) was not less than 0.9989 in the linear range, and the recovery rate of all spiked phenol compounds in blank marine sediment was in the range of 86.8%-93.2%, thereby indicating that the developed method would be suitable to determine the target decomposition phenol compounds of lignin in marine sediment. Subsequently, the method was used to determine the phenolic compounds of lignin decomposition products in the surface sediments of Laizhou Bay. The detection rate of 11 target phenolic compounds in 12 surface sediment samples was 100%, and the concentration of Σ8 in 12 surface sediment samples ranged from 0.001mg/10gds to 0.019mg/10gds. The value of C/V was between 0.18 and 0.81, with an average of 0.38; the value of S/V was between 0.18 and 0.45, with an average of 0.26; PON/P value was between 0.01 and 0.07, with an average of 0.03; P/(V+S) value was between 0.55 and 3.77, with an average of 1.44; (Ad/Al)_v value was between 0.12 and 1.07, with an average of 0.48; the value of (Ad/Al)_s was between 0.15 and 1.26, with an average of 1.02.

CONCLUSIONS The above diagnostic ratios indicate that the lignin in the surface sediments of Laizhou Bay originate mainly from the herbaceous tissue of angiosperms, while the proportion of organic matter in vascular plants is relatively low. The degradation degree of terrestrial organic matter in most sampling stations is medium or high, but there is still a small amount of fresh plant debris. The proposed method has the advantages of high efficiency, simple for sample pretreatment, and is a powerful technique for the determination of main decomposition product phenolic compounds of lignin in marine sediments.