More than 180 million metric tons of ammonia is produced globally with a possible 2.3% annual increase. Although, most used for fertilizer, ammonia has furture potential application of the gas is in the transport and power generation industries. However, the current Haber-Bosch process for ammonia production operate at high temperature and pressure and is responsible for Energy and CO2 reductions during ammonia production have necessitate the development of green process from bioresouce. We have successfully isolated 5 hyper-ammonia bacteria (HAB) that have ability to ferment soybean hydrolyzates to produce organic ammonia. Our current result porduced about 500 mM of organic ammonia but still have un-used substrate and requires two steps (hydrolysis and fermentation). The unused amino acids could potentially be fermented by HAB if order supplements (minerals, vitamins and co-factors) are present. Also, combining both hydrolysis and fermentation into one-pot system will reduce processing time and volumetric productivity thereby making the process more efficient. Therefore, these further studies are required to make organic ammonia production via HAB-fermentation of soybean industrially viable. Therefore, the objectives of this study are to develop soy-formulate suitable for HAB fermentation in a one-pot system. The outcomes of this research will be viable for developing ammonia biofuel production using ND crops and byproducts and will eventually expand the economy of ND agriculture.

The main objective of the research is to develop an efficient one-pot fermentation system for organic ammonia production.
The specific objectives are:
i. To develop soy-bean formulate suitable for HAB fermentation.
ii. To develop one-pot fermentation process for organic ammonia production via HAB fermentation of soybean.

- Information on supplements required for efficient soybean utilization.
- Knowledge of ammonia production efficiency using soybean in a one-pot fermentation system.

Updated November 30, 2021:
Development of Soy-formulate for organic ammonia production via hyper-ammonia-bacteria fermentation in a one-pot system

Ademola Hammed, Agricultural and Biosystems Engineering,
North Dakota State University

Objectives of the research
i. Development of a mild pretreatment for whole soybean
Completed work
i. Development of a mild alkali extraction of soybean protein
Preambles
Plant cell wall are rigid and prevent diffusion of protein needed during fermentation to produce organic ammonia. Therefore, we are developing chemical extraction processing to obtain protein from soybean. Since the targeted product is ammonia, we developed ammonia hydroxide extraction of soybean protein. Also, ammonia can be recovered and reuse making it an efficient extraction solvent.

Method
Whole soybean was grinded and sieved into different particle sizes >1, 0.425-1, 0.25 - 0.425 and <0.25 mm denoted, respectively as A, B, C, and D. Protein extraction was carried out on 5 g A, B, C, and D using 50 mL of different ammonium hydroxide concentrations (1, 2.5, 5, 10 and 15%). The mixtures were transferred into a shaker preset at 55 oC and 130 rpm. Extracts were collected at different time (0, 3, 6, 12, 24 and 48h). The extracts were centrifuged at 1000 rpm for 5 mins to separate the supernatant (extract) from the residue. The extract protein concentration was determined using the Bradford assay.

Results
Effect of particle size
The particle size affects extraction yield because increasing particle size reduces the solid surface area available for interactions with the solvent. Smaller particle size not only has high surface area that facilitates surface washing of extract but also has reduced diameter that aids extract leaching. However, reducing sample particle size is energy intensive, costly and prone to processing hazards including explosion. Therefore, we investigate the effect of particle size on the release of protein during protein extraction using ammonia hydroxide solution.
Figure 1 shows the result of protein yield at different particle sizes. Extraction yield increase across with reduction in particle size. At 15% solvent concentration and 48h, 50% protein extraction was achieved for sample A while for sample D, about 59% protein extraction was achieved. The results showed that particle size had a significant effect on extraction yield.

Figure 1: Effect of particle size at 48h and 15% solvent concentration
Effect of solvent concentration
Solvent polarity and pH affect extraction yield because high acidic/basic solvent disrupts plant cell wall and high OH- increase solvent polarity to dissolve polar solute. The degree of plant cell wall disruption and solute solubility could increase extraction yield. In order to determine the ammonium hydroxide concentration that favor soy bean protein extraction we used different ammonium hydroxide concentrations (0, 1, 2.5, 5, 10 and 15%) during processing. The result (Figure 2) shows that ammonium hydroxide solutions extract more protein than the control. Protein extraction increase rapidly (approximately 49%) with 1% solvent concentration. Further increase in concentration did not cause significant increase in protein.


Figure 2: Effect of solvent concentration at 48h
Analysis of variance
A two-way ANOVA was performed to analyze the effect of particle size, ammonium hydroxide concentration and time on protein extraction from soybeans. The ANOVA analysis in Table 1 shows that the processing conditions significantly affect the protein yield. The probability F is less than 0.0001 suggesting that the difference is highly significant.

Dependent variable: Protein Concentration
Source DF Sum of Squares Mean Square F Value Pr > F
Model 47 27650.59291 588.31049 50.43 <.0001
Error 72 839.89921 11.66527
Corrected Total 119 28490.49212

Extraction kinetics
In solvent extraction process, kinetic models are commonly used to simulate the extraction process. Kinetic modelling simplifies and facilitates optimization study design. Different kinetic models were used for the extraction and the best fit model (So and Macdonald’s model, Figure 3) was selected based on higher magnitude of linear correlation coefficient (R=0.94) and lower root mean square (RMS=0.24) value (Kitanovic, S et al., 2008).


Figure 3. Comparison of experimental (symbols) and fitted (line) yields for soybean pretreatment at 1% solvent concentration and 48h: The extraction kinetics was fitted by So and Mac Donald’s model.

Conclusion
The effect of extraction processing factors, ammonium hydroxide concentration, time and particle size on soybean protein yield have been investigated. Reducing particle, time and ammonium hydroxide concentration increases protein yield. The conditions for highest protein yield are 12 h, 1% ammonium hydroxide and < 0.425 mm particle size.

Work to be completed
i. Optimization of NH4OHaq aided extraction of soybean protein.
ii. Biochemical hydrolysis of protein extract.
iii. Inclusion substrate essential elements
References
Kitanovic, S., Milenovic, D., & Veljkovic, V. B. (2008). Empirical kinetic models for the resinoid extraction from aerial parts of St. John's wort (Hypericum perforatum L.). Biochemical Engineering Journal, 41(1), 1-11.

View uploaded report

This research will make soybean utilization for efficient organic ammonia production. This will eventually result into more reason to buy soybeans by the industries.