Limits...
Waste Conversion into n -Caprylate and n -Caproate: Resource Recovery from Wine Lees Using Anaerobic Reactor Microbiomes and In-line Extraction

View Article: PubMed Central - PubMed

ABSTRACT

To convert wastes into sustainable liquid fuels and chemicals, new resource recovery technologies are required. Chain elongation is a carboxylate-platform bioprocess that converts short-chain carboxylates (SCCs) (e.g., acetate [C2] and n-butyrate [C4]) into medium-chain carboxylates (MCCs) (e.g., n-caprylate [C8] and n-caproate [C6]) with hydrogen gas as a side product. Ethanol or another electron donor (e.g., lactate, carbohydrate) is required. Competitive MCC productivities, yields (product vs. substrate fed), and specificities (product vs. all products) were only achieved previously from an organic waste material when exogenous ethanol had been added. Here, we converted a real organic waste, which inherently contains ethanol, into MCCs with n-caprylate as the target product. We used wine lees, which consisted primarily of settled yeast cells and ethanol from wine fermentation, and produced MCCs with a reactor microbiome. We operated the bioreactor at a pH of 5.2 and with continuous in-line extraction and achieved a MCC productivity of 3.9 g COD/L-d at an organic loading rate of 5.8 g COD/L-d, resulting in a promising MCC yield of 67% and specificities of 36% for each n-caprylate and n-caproate (72% for both). Compared to all other studies that used complex organic substrates, we achieved the highest n-caprylate-to-ncaproate product ratio of 1.0 (COD basis), because we used increased broth-recycle rates through the forward membrane contactor, which improved in-line extraction rates. Increased recycle rates also allowed us to achieve the highest reported MCC production flux per membrane surface area thus far (20.1 g COD/m2-d). Through microbial community analyses, we determined that an operational taxonomic unit (OTU) for Bacteroides spp. was dominant and was positively correlated with increased MCC productivities. Our data also suggested that the microbiome may have been shaped for improved MCC production by the high broth-recycle rates. Comparable abiotic studies suggest that further increases in the broth-recycle rates could improve the overall mass transfer coefficient and its corresponding MCC production flux by almost 30 times beyond the maximum that we achieved. With improved in-line extraction, the chain-elongation biotechnology production platform offers new opportunities for resource recovery and sustainable production of liquid fuels and chemicals.

No MeSH data available.


Heatmap of relative OTU abundances for the inoculum, the substrate, and bioreactor samples from Periods 1–5. OTUs were clustered hierarchically (average linkage) based on the Bray–Curtis dissimilarity index with sequence data for 12 microbiome samples, including: one inoculum sample; one wine lees substrate sample; three bioreactor samples from the batch phase (Period 1); and seven bioreactor samples from the phase of semi-continuous substrate addition with continuous in-line extraction (Periods 2–5). OTUs were grouped together based on both the average relative abundance and abundance profile. This resulted in grouping of OTUs with similar progressive shifts in relative abundances during the operating time. In addition, it also separated OTUs that were primarily abundant in the inoculum and the substrate. Each of the 49 OTUs listed comprised at least one percent of the relative abundance for one or more of the samples. Relative abundances of three OTUs (asterisks) were correlated (p < 0.05) with MCC productivities.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC5121217&req=5

Figure 7: Heatmap of relative OTU abundances for the inoculum, the substrate, and bioreactor samples from Periods 1–5. OTUs were clustered hierarchically (average linkage) based on the Bray–Curtis dissimilarity index with sequence data for 12 microbiome samples, including: one inoculum sample; one wine lees substrate sample; three bioreactor samples from the batch phase (Period 1); and seven bioreactor samples from the phase of semi-continuous substrate addition with continuous in-line extraction (Periods 2–5). OTUs were grouped together based on both the average relative abundance and abundance profile. This resulted in grouping of OTUs with similar progressive shifts in relative abundances during the operating time. In addition, it also separated OTUs that were primarily abundant in the inoculum and the substrate. Each of the 49 OTUs listed comprised at least one percent of the relative abundance for one or more of the samples. Relative abundances of three OTUs (asterisks) were correlated (p < 0.05) with MCC productivities.

Mentions: The microbial community was characterized based on 10 samples from the bioreactor broth during the first 84 days of this experiment (Phases I and II, Periods 1–5). In addition, we included one sample each from the inoculum and the substrate (Figure 7). For all 12 samples, we observed 2526 OTUs from high-quality sequence reads. For the bioreactor samples, 36 OTUs exceeded 1% relative abundance in at least one sample. For the inoculum and substrate samples, an additional six and seven unique OTUs, respectively, exceeded 1% relative abundance. The total of these 49 OTUs accounted for 86.9 to 96.6% of the total high-quality sequence reads for each sample. The number of OTUs within the community and their relative proportions (alpha diversity) did not vary considerably between microbiome samples from the operating period (Supplementary Figure S3). Moreover, the average Shannon index for bioreactor samples from the present study (3.7) was approximately equal to the average values for previous reactor microbiome studies with synthetic lactate (3.6) (Kucek et al., 2016a) or ethanol (3.4) (Kucek et al., 2016b) as electron donors.


Waste Conversion into n -Caprylate and n -Caproate: Resource Recovery from Wine Lees Using Anaerobic Reactor Microbiomes and In-line Extraction
Heatmap of relative OTU abundances for the inoculum, the substrate, and bioreactor samples from Periods 1–5. OTUs were clustered hierarchically (average linkage) based on the Bray–Curtis dissimilarity index with sequence data for 12 microbiome samples, including: one inoculum sample; one wine lees substrate sample; three bioreactor samples from the batch phase (Period 1); and seven bioreactor samples from the phase of semi-continuous substrate addition with continuous in-line extraction (Periods 2–5). OTUs were grouped together based on both the average relative abundance and abundance profile. This resulted in grouping of OTUs with similar progressive shifts in relative abundances during the operating time. In addition, it also separated OTUs that were primarily abundant in the inoculum and the substrate. Each of the 49 OTUs listed comprised at least one percent of the relative abundance for one or more of the samples. Relative abundances of three OTUs (asterisks) were correlated (p < 0.05) with MCC productivities.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC5121217&req=5

Figure 7: Heatmap of relative OTU abundances for the inoculum, the substrate, and bioreactor samples from Periods 1–5. OTUs were clustered hierarchically (average linkage) based on the Bray–Curtis dissimilarity index with sequence data for 12 microbiome samples, including: one inoculum sample; one wine lees substrate sample; three bioreactor samples from the batch phase (Period 1); and seven bioreactor samples from the phase of semi-continuous substrate addition with continuous in-line extraction (Periods 2–5). OTUs were grouped together based on both the average relative abundance and abundance profile. This resulted in grouping of OTUs with similar progressive shifts in relative abundances during the operating time. In addition, it also separated OTUs that were primarily abundant in the inoculum and the substrate. Each of the 49 OTUs listed comprised at least one percent of the relative abundance for one or more of the samples. Relative abundances of three OTUs (asterisks) were correlated (p < 0.05) with MCC productivities.
Mentions: The microbial community was characterized based on 10 samples from the bioreactor broth during the first 84 days of this experiment (Phases I and II, Periods 1–5). In addition, we included one sample each from the inoculum and the substrate (Figure 7). For all 12 samples, we observed 2526 OTUs from high-quality sequence reads. For the bioreactor samples, 36 OTUs exceeded 1% relative abundance in at least one sample. For the inoculum and substrate samples, an additional six and seven unique OTUs, respectively, exceeded 1% relative abundance. The total of these 49 OTUs accounted for 86.9 to 96.6% of the total high-quality sequence reads for each sample. The number of OTUs within the community and their relative proportions (alpha diversity) did not vary considerably between microbiome samples from the operating period (Supplementary Figure S3). Moreover, the average Shannon index for bioreactor samples from the present study (3.7) was approximately equal to the average values for previous reactor microbiome studies with synthetic lactate (3.6) (Kucek et al., 2016a) or ethanol (3.4) (Kucek et al., 2016b) as electron donors.

View Article: PubMed Central - PubMed

ABSTRACT

To convert wastes into sustainable liquid fuels and chemicals, new resource recovery technologies are required. Chain elongation is a carboxylate-platform bioprocess that converts short-chain carboxylates (SCCs) (e.g., acetate [C2] and n-butyrate [C4]) into medium-chain carboxylates (MCCs) (e.g., n-caprylate [C8] and n-caproate [C6]) with hydrogen gas as a side product. Ethanol or another electron donor (e.g., lactate, carbohydrate) is required. Competitive MCC productivities, yields (product vs. substrate fed), and specificities (product vs. all products) were only achieved previously from an organic waste material when exogenous ethanol had been added. Here, we converted a real organic waste, which inherently contains ethanol, into MCCs with n-caprylate as the target product. We used wine lees, which consisted primarily of settled yeast cells and ethanol from wine fermentation, and produced MCCs with a reactor microbiome. We operated the bioreactor at a pH of 5.2 and with continuous in-line extraction and achieved a MCC productivity of 3.9 g COD/L-d at an organic loading rate of 5.8 g COD/L-d, resulting in a promising MCC yield of 67% and specificities of 36% for each n-caprylate and n-caproate (72% for both). Compared to all other studies that used complex organic substrates, we achieved the highest n-caprylate-to-ncaproate product ratio of 1.0 (COD basis), because we used increased broth-recycle rates through the forward membrane contactor, which improved in-line extraction rates. Increased recycle rates also allowed us to achieve the highest reported MCC production flux per membrane surface area thus far (20.1 g COD/m2-d). Through microbial community analyses, we determined that an operational taxonomic unit (OTU) for Bacteroides spp. was dominant and was positively correlated with increased MCC productivities. Our data also suggested that the microbiome may have been shaped for improved MCC production by the high broth-recycle rates. Comparable abiotic studies suggest that further increases in the broth-recycle rates could improve the overall mass transfer coefficient and its corresponding MCC production flux by almost 30 times beyond the maximum that we achieved. With improved in-line extraction, the chain-elongation biotechnology production platform offers new opportunities for resource recovery and sustainable production of liquid fuels and chemicals.

No MeSH data available.