Phenolic-rich bio-oil can be selectively produced from catalytic fast pyrolysis of biomass impregnated with K3PO4, K2HPO4, or KH2PO4. In this study, the catalytic effects of the three catalysts on the pyrolytic product distribution were investigated and compared via analytical pyrolysis-gas chromatography/mass spectrometry (Py-GC/MS) experiments. The results indicated that the three catalysts were all able to inhibit the pyrolytic decomposition of holocellulose to form volatile organic products, while promoting the formation of phenolic compounds from lignin. Hence, phenolic-rich bio-oil could be selectively produced. Among the three catalysts, K3PO4 and K2HPO4 possessed similar capability to increase the yield of the phenolics, which was better than KH2PO4. The phenolic contents among the total pyrolytic products steadily increased as the K3PO4 or K2HPO4 dosage increased. The maximal peak area of the phenolics reached as high as 68.8% (at 50 wt.% K3PO4) or 50.6% (at 50 wt.% K2HPO4) of the total peak area. Therefore, based on these results, K3PO4 was the best catalyst for the selective production of phenolic-rich bio-oil.
Selective Production of Phenolic-rich Bio-oil from Catalytic Fast Pyrolysis of Biomass: Comparison of K3PO4, K2HPO4, and KH2PO4
Zhi-Bo Zhang,a Qiang Lu,a,* Xiao-Ning Ye,a Ling-Ping Xiao,bChang-Qing Dong,a and Yong-Qian Liu a
Phenolic-rich bio-oil can be selectively produced from catalytic fast pyrolysis of biomass impregnated with K3PO4, K2HPO4, or KH2PO4. In this study, the catalytic effects of the three catalysts on the pyrolytic product distribution were investigated and compared viaanalytical pyrolysis-gas chromatography/mass spectrometry (Py-GC/MS) experiments. The results indicated that the three catalysts were all able to inhibit the pyrolytic decomposition of holocellulose to form volatile organic products, while promoting the formation of phenolic compounds from lignin. Hence, phenolic-rich bio-oil could be selectively produced. Among the three catalysts, K3PO4 and K2HPO4 possessed similar capability to increase the yield of the phenolics, which was better than KH2PO4. The phenolic contents among the total pyrolytic products steadily increased as the K3PO4 or K2HPO4 dosage increased. The maximal peak area of the phenolics reached as high as 68.8% (at 50 wt.% K3PO4) or 50.6% (at 50 wt.% K2HPO4) of the total peak area. Therefore, based on these results, K3PO4 was the best catalyst for the selective production of phenolic-rich bio-oil.
Keywords: Biomass; Phenolic compounds; Catalytic fast pyrolysis; K3PO4; K2HPO4; KH2PO4
Contact information: a: National Engineering Laboratory for Biomass Power Generation Equipment, North China Electric Power University, Beijing 102206, China; b: Beijing Key Laboratory of Lignocellulosic Chemistry, Beijing Forestry University, Beijing 100083, China;
* Corresponding author: firstname.lastname@example.org
Fast pyrolysis of biomass is a thermal decomposition process, with rapid biomass heating and decomposition followed with quick vapor condensation, to convert solid biomass into liquid bio-oil (Bridgwater 2012; Mohan et al. 2006). Chemically, bio-oil is a complex mixture of water and hundreds of organic compounds. There are many valuable chemicals in conventional bio-oils (Demirbas 2006), but most of them are in very low proportions due to the poor pyrolytic selectivity during the conventional fast pyrolysis process. Therefore, to obtain specific bio-oils rich in desired products, it is necessary to selectively control the biomass pyrolysis pathways toward the formation of target products. Currently, various selective pyrolysis techniques have been proposed to produce different value-added chemicals from different biomass components, such as levoglucosenone (Kudo et al. 2011; Lu et al. 2012; Wang et al. 2011) and 1-hydroxy-3,6-dioxabicyclo[3.2.1]octan-2-one (Fabbri et al.2007) from cellulose, furfural from holocellulose (Branca et al. 2012; Oh et al. 2013), and aromatic hydrocarbons from lignin or whole biomass (Carlson et al. 2011; Zhang et al. 2013).
One of the valuable chemical fractions in bio-oil is the phenolic compounds, which can be used for the synthesis of phenolic resins. Phenolic compounds are mainly derived from lignin (Lou et al. 2011); thus, phenolic-rich bio-oil can be produced from fast pyrolysis of lignin or special lignin-rich biomass, such as bark (Ren et al. 2013). However, for common biomass materials, the lignin content is much lower than the holocellulose content. Therefore, to selectively produce the phenolic-rich bio-oil, it is essential to enhance the pyrolytic decomposition of lignin to form phenolic compounds while inhibiting the formation of volatile organic compounds from holocellulose. To date, only Bu et al. (2012; 2013) have reported a way to selectively produce phenolic-rich bio-oil through microwave-assisted pyrolysis of biomass using activated carbon as the catalyst. With fir wood as the feedstock, the phenolic concentration in the catalytic bio-oil reached as high as 66.9% (relative to the total peak area % detected in the bio-oil by GC/MS) at a catalyst-to-biomass ratio of 3:1. Phenol and 4-methyl phenol were the two most abundant phenolic compounds formed in the catalytic process. Recently, we found another way to selectively produce phenolic-rich bio-oil by catalytic fast pyrolysis of biomass impregnated with K3PO4 (Lu et al.2013). With poplar wood as the feedstock, the highest phenolic concentration (peak area % by GC/MS) was over 60% at high K3PO4dosages. Phenol and 2,6-dimethoxyphenol were formed as the two most abundant phenolic compounds.
In this study, K3PO4, K2HPO4, and KH2PO4 were selected as the catalysts for the selective production of phenolic-rich bio-oil. These catalysts were impregnated into poplar wood using the incipient wetness impregnation method, which differed from the K3PO4impregnation method in our previous study (Lu et al. 2013). Analytical pyrolysis-gas chromatography/mass spectrometry (Py-GC/MS) experiments were conducted to reveal and compare the catalytic effects of the three catalysts on the pyrolytic product distribution.
Poplar wood was used as the feedstock in this study. Prior to experiments, the poplar wood was ground and sieved; particles of 0.2 to 0.3 mm in size were selected and dried prior to pretreatment and further experimentation. The chemical composition of poplar was 49.8% cellulose, 24.4% hemicellulose, 23.3% lignin, 2.2% extractives, and 0.3% inorganic ash. Its elemental composition on a dry basis was 49.58% C, 6.33% H, 0.08% N, 0.09% S, and 43.62% O.
Pretreatment of poplar wood
The poplar wood was pretreated to impregnate the K3PO4, K2HPO4, or KH2PO4 catalyst using the incipient wetness impregnation method. Aqueous K3PO4, K2HPO4, or KH2PO4 solutions with different concentrations were prepared by dissolving prescribed amounts of K3PO4, K2HPO4, or KH2PO4 into 50 mL of deionized water. Afterwards, each solution was added to 10 g of poplar wood and agitated by ultrasonification for 12 h. The mixture was then dried at 105 °C in air for 4 h and stored in a desiccator prior to experimentation. A total of 18 pretreated poplar wood samples were prepared, six samples for each catalyst. The catalyst dosages (i.e., K3PO4, K2HPO4, or KH2PO4) examined were 5 wt.%, 10 wt.%, 20 wt.%, 30 wt.%, 40 wt.%, and 50 wt.% (on wood).
Analytical Py-GC/MS Experiments
Analytical Py-GC/MS experiments were conducted using the CDS Pyroprobe 5200HP pyrolyzer (Chemical Data Systems; Oxford, PA) connected with a Perkin Elmer GC/MS (Clarus 560; Waltham, MA). Experimental samples (0.30 mg of untreated or a prescribed quantity of pretreated poplar wood) were placed into a quartz pyrolysis tube. Quartz wool was placed at both ends of the tube to prevent the escape of solid wood particles during the fast pyrolysis process. The quantity of the pretreated poplar wood was precisely weighed to be 0.32 mg, 0.33 mg, 0.38 mg, 0.43 mg, 0.50 mg, and 0.60 mg for the samples impregnated with 5 wt.%, 10 wt.%, 20 wt.%, 30 wt.%, 40 wt.%, and 50 wt.% catalyst, respectively. Therefore, the polar wood quantity during pyrolysis, excluding the catalyst, remained constant at 0.30 mg for each experimental sample. An analytical balance with a readability of 0.01 mg was used for weighing. A detailed description of the sample preparation procedure has been reported in our previous study (Lu et al. 2013).
Pyrolysis was carried out at temperatures from 300 to 700 °C, with the heating rate of 20 °C/s and heating time of 20 s. The pyrolysis vapors were directly analyzed by GC/MS. The GC injector temperature was kept at 300 °C. Chromatographic separation was achieved using an Elite-35MS capillary column (30 m × 0.25 mm i.d., 0.25 μm film thickness; PerkinElmer; Waltham, MA). Helium (99.999%) was used as the carrier gas with a constant flow rate of 1 mL/min at a 1:80 split ratio. The GC oven was heated from 40 °C (3 min) to 180 °C with a heating rate of 6 °C/min, and then to 280 °C (4 min) with a heating rate of 10 °C/min. The temperature of the GC/MS interface was held at 280 °C, and the mass spectrometer was operated in EI mode at 70 eV. The chromatographic peaks were identified according to the NIST library, Wiley library, and data from previous studies (Dong et al. 2012).
For each poplar wood sample, the experiments were replicated three times to confirm the reproducibility of the reported results. The peak area and peak area % values of each pyrolytic product were determined, and the average and standard deviation values were calculated. It is known that the analytical Py-GC/MS technique could not provide direct quantitative analysis of the compounds due to the complexity of the pyrolytic products and the lack of commercially available reference standards. However, the chromatographic peak area of a compound is considered to vary linearly with its quantity, and the peak area % is considered to vary linearly with its concentration. Therefore, for each product, its yield changes can be determined by comparing its average peak area values obtained under different reaction conditions, and the changes to its content among the detected products can be determined by comparing its peak area % values relative to the sum of all the peak area in a chromatogram (Luet al. 2011a). It is noted that biomass fast pyrolysis vapors are composed of permanent gases, water, volatile organic compounds, and non-volatile large molecular compounds. The latter three fractions will be condensed as liquid bio-oil, while the GC/MS is only able to determine the volatile organic compounds.
RESULTS AND DISCUSSION
The non-catalytic analytical fast pyrolysis of poplar wood has been reported in our previous study, with detailed product distribution and formation characteristics (Dong et al. 2012). Figure 1a shows the typical ion chromatogram at the pyrolysis temperature of 500 °C. The major pyrolytic products included hydroxyacetaldehyde (HAA), acetic acid (AA), 1-hydroxy-2-propanone (HA), furfural (FF), and levoglucosan (LG), all of which were primarily derived from holocellulose; additionally, various phenolic compounds (e.g., phenol, 2-methoxyphenol, and 2,6-dimethoxyphenol) were detected that were primarily derived from lignin.
The pyrolytic product distribution was significantly altered after the poplar wood was impregnated with the K3PO4, K2HPO4, or KH2PO4catalyst; typical ion chromatograms are given in Figs. 1b-1d. K3PO4was found to be a promising catalyst for the selective production of phenolic compounds. The major holocellulose-derived products, such as HAA and LG, were significantly decreased or completely eliminated, leaving the HA as the only major holocellulose-derived product when utilizing the K3PO4 catalyst. Meanwhile, some lignin-derived phenolic products were remarkably increased and became the dominant catalytic products; these products included phenol, 2-methoxyphenol, 2-methoxy-4-vinylphenol, and 2,6-dimethoxyphenol. The results agreed well with our previous study in spite of the different K3PO4 impregnation methods (Lu et al. 2013). The K2HPO4 exhibited similar catalytic effects as K3PO4 on the selective production of phenolic compounds; however, KH2PO4 was less effective than the other two catalysts. The detailed catalytic effects of the three catalysts and their differences will be presented and discussed in the following sections.
Fig. 1. Typical MS ion chromatograms from fast pyrolysis of raw and pretreated poplar wood: (1) hydroxyacetaldehyde (HAA); (2) acetic acid (AA); (3) 1-hydroxy-2-propanone (HA); (4) furfural (FF); (5) phenol; (6) 2-methylphenol; (7) 2-methoxyphenol; (8) 4-methyl-1,2-benzendiol (9) 2-methoxy-4-vinylphenol; (10) 2,6-dimethoxyphenol; (11) 1,2,4-trimethoxybenzene; (12) 1,2,3-trimethoxy-5-methylbenzene (13) levoglucosan (LG); (14) 3′,5′-dimethoxyacetophenone; (15) 3,5-dimethoxy-4-hydroxybenzaldehyde; (16) 4-allyl-2,6-dimethoxyphenol; and (17) 1-(2,4,6-trihydroxyphenyl)-2-pentanone
Effects of Catalyst Impregnation Dosages on the Overall Product Distribution
The pyrolysis vapors emitted from the analytical Py-GC/MS experiments were unable to be collected and quantitatively analyzed. Hence, the distribution (yields and concentrations) of the pyrolytic products could not be quantitatively determined. However, the changes of product yields and concentrations can be estimated by comparing the peak area and peak area % values of the products obtained under different reaction conditions. Figure 2 shows the total peak area values of the total phenolic products and the other products (mainly holocellulose-derived products). Moreover, the pyrolytic products were classified into eight groups, and their peak area % values are presented in Table 1. The eight groups were phenolics, anhydrosugars (e.g., LG), linear aldehydes (e.g., HAA), linear ketones (e.g., HA), linear acids (e.g., AA), furans, cyclopentanones, and others (e.g., hydrocarbons and linear fatty acids). The peak area % sum of the eight groups was less than 100% due to some unidentified chromatographic peaks.
Fig. 2. Total chromatographic peak area values of the phenolic products and the other products for different catalyst dosages employed during pyrolysis performed at 500 °C
According to Figure 2, the peak area values of the total phenolic products, which were proportional to the total phenolic yields, initially increased and then decreased as the catalyst dosage increased. The maximum peak area values of the total phenolic products were obtained with 10 wt.% K3PO4, 20 wt.% K2HPO4, and 5 wt.% KH2PO4, respectively. The three maximum values were 47%, 46%, and 24% higher than the value from the non-catalytic process. The results clearly indicated that the three catalysts were all able to promote the formation of phenolics when properly impregnated into the poplar wood. The K3PO4 and K2HPO4 catalysts possessed similar capability to increase the yield of phenolics, which was better than the KH2PO4 catalyst.
The other pyrolysis products were primarily derived from holocellulose; their total peak area values decreased monotonically as the catalyst dosage increased. This observation clearly suggested that all three catalysts could inhibit the pyrolysis of holocellulose to form volatile organic products. The K3PO4 and K2HPO4 catalysts performed much better than the KH2PO4 catalyst at all dosage levels; K3PO4 performed better than K2HPO4 at dosages as high as 40 wt.% to 50 wt.%.
Table 1. Peak Area % Values of the Eight Product Groups at Different Catalyst Dosages Employed during Pyrolysis Performed at 500 °C
According to Table 1, as the K3PO4 dosage increased, the peak area values of the phenolics group, which was proportional to the total phenolics contents, steadily increased, reaching as high as 68.8% at 50 wt.% K3PO4. At the same time, the peak area % values of the anhydrosugars, linear aldehydes, linear acids, and furans steadily decreased as the K3PO4 dosage increased. Similar trends were observed for the K2HPO4 catalysis, except for the linear acids group. The peak area values of the linear acids group significantly decreased at high K3PO4 dosages (40 wt.% and 50 wt.%), but increased at high K2HPO4 dosages. The highest phenolics peak area % value (50.6%) was obtained when 50 wt.% K2HPO4 was used. As the KH2PO4dosage increased, the peak area values of the anhydrosugars, linear ketones, and furans all changed differently from those of K3PO4 and K2HPO4. The highest phenolics peak area % value (41%) was observed when 40 wt.% KH2PO4 was used.