Team:Groningen/Description

Description

Abstract

Bye-Monia, tackling Dutch nitrogen crisis: Reduction of Nitrogen emissions by alpha-amylase production in Saccharomyces spp.

The Netherlands is producing excess nitrogen, which is harmful to nature and biodiversity. One of the main culprits for this so-called nitrogen crisis is animal agriculture, a vital income source for the Dutch economy. Our project targets ammonia emissions contributing to the crisis and aims to convert them into a beneficial feed additive. Thus, we have engineered Saccharomyces spp. to synthesize alpha-amylase, an enzyme that optimizes cattles’ digestion. This way, their milk production and growth will be enhanced while ammonia emissions will be reduced simultaneously. Residual ammonia will be captured by a state-of-the-art filter device, a Metal-Organic Framework (MOF), and fed back to our GMO. Furthermore, insights from artificial intelligence will be employed to optimize the engineering process. Overall, we have designed a closed sustainable circle in which waste - excess ammonia, is converted into worth - a feed additive for cattle. To watch our promotional video, click here.

The problem

What is the nitrogen crisis?

In short: the Netherlands emits too much nitrogen-containing compounds, which endangers nitrogen-sensitive nature reserves. This is called the nitrogen crisis.

78% of our atmosphere consists of nitrogen (N2). Nitrogen on its own is not harmful for humans or the environment. However, certain nitrogen-containing compounds such as ammonia (NH3) and nitrogen oxides (NOx) can be[1]. Therefore, international guidelines have been established for the maximum emissions of these nitrogen-containing compounds[2].

Even though international guidelines have been set, the Netherlands still has a nitrogen emission surplus of 200 kg N per hectare per year, which is four times as high as that of the European Union[3]. These nitrogen emissions include both nitrogen oxide and ammonia emissions. However, for our project we focused on ammonia emissions specifically.

When ammonia is emitted to the atmosphere, it can mix with the air and later, in an area farther away, be deposited back onto the surface. Ammonia can either be deposited on to the surface with precipitation, this is called wet deposition, or taken up directly by plants and the soil, this is called dry deposition[1].

Difference between emission of nitrogen-containing compounds and deposition of nitrogen-containing compounds.

Figure 1: Difference between emission of nitrogen-containing compounds and deposition of nitrogen-containing compounds. Created with BioRender.com and adapted from [1]

The deposition of ammonia enriches the soil with nitrogen, which normally enhances plant growth. However, certain nature areas are sensitive to excess nitrogen, since they contain plants that grow better in nitrogen-poor soil. Excess ammonia deposition in these areas causes plants that grow well on nitrogen, to outcompete those that don’t. Consequently causing the disappearance of the animals that live off these plants from the ecosystem [1]. Moreover, excess ammonia emissions can lead to acidification and eutrophication, which can significantly harm (the quality of) ecosystems[3].

In 2016, 70% of nature areas in the Netherlands exceeded limits for nitrogen[3]. In other words: 70% of all Dutch nature areas were at risk of significant habitat quality degradation and a decrease in biodiversity.

Special attention should be given to Natura 2000 areas, which cover 15% of the Netherlands surface area as of December 2020[4]. Natura 2000 areas are areas that belong to the Natura 2000 network, which is a network of “core breeding and resting sites for rare and threatened species, and some rare natural habitat types which are protected in their own right”[5]. 118 out of the 160 Dutch Natura 2000 areas are sensitive to an excess of nitrogen[6].

Natura 2000 areas in the Netherlands: both on land and in water

Figure 2: Natura 2000 areas in the Netherlands: both on land and in water[7]

Agriculture and the nitrogen crisis

Nitrogen-containing compounds can be emitted by different sources. However, agriculture contributes to a big part of the nitrogen emissions, and is the source of the vast majority of ammonia emissions in the Netherlands. In 2019 the Netherlands emitted 125.8 Kton of ammonia, from which 107.5 Kton originated from agriculture[8].

Dutch nitrogen emissions per sector over time.

Figure 3: Dutch nitrogen emissions per sector over time. Adapted from [1]

Dutch ammonia emissions per sector over time.

Figure 4: Dutch ammonia emissions per sector over time. NEC 2010 and Gothenburg 2020 indicate the international guidelines for how much ammonia the Netherlands could emit in 2010 and 2020 respectively. Expected 2020 indicate how much ammonia the Netherlands expects to emit in 2020 if current policy is followed. Adapted from [1][9]

To combat the excess of nitrogen emissions several measures have been taken, ranging from lowering the maximal speed on highways to blocking plans for new homes, roads and airport runways. However, seeing that agriculture is one of the main sources of nitrogen emissions and the main source of ammonia emissions, measures against the nitrogen crisis have mainly focused on curbing agriculture. The most important measure has been the stalling of expansions of cattle, pig and poultry farms, although there have also been talks of shrinking, or even halving, the agricultural sector as a whole[10].

These measures have a far reaching impact. Not only do they threaten the income of farmers and the (financial) viability of their businesses, which lead to nation-wide farmers protests, but they also influence global food security. The Dutch farm animal sector is the densest in the world, with Dutch farms containing four times more animal biomass per hectare than the EU average[10]. So despite its relatively small size, the Netherlands still is the second largest agricultural exporter in the world[11]. The European Commission even remarks that “The Dutch agricultural sector is characterised as a productive, innovative and exportoriented sector with intensive agricultural production that is largely based on cost-price reduction and increasing economies of scale”[3]. Therefore shrinking the Dutch agricultural sector, causing production to have to move to less production effective areas, can greatly threaten global food supply.

How does ammonia get produced by cattle?

Within agriculture, the majority of ammonia emissions originate from cattle. And even though several measures have been taken to reduce the ammonia emissions by cattle, the reduction in ammonia emissions by cattle has stagnated over time.

Ammonia emissions to air in the Netherlands per emission source over time.

Figure 5: Ammonia emissions to air in the Netherlands per emission source over time. “Other” indicates all non-agricultural sources of Dutch ammonia emissions. Numbers derived from [8]

Ammonia is mainly produced by cattle through urea hydrolysis. When the cow ingests nitrogen-containing compounds through its feed, but does not manage to retain the nitrogen in its body, the majority of the leftover nitrogen will be excreted through the urine in the form of urea (CO(NH2)2). On the other hand are microorganisms in the feces of the cow, which produce urease, an enzyme that can break down urea. Once urine and feces mix, therefore, bringing urea and urease together, ammonia gets produced alongside with carbon dioxide. This process can be described with the following equation[12]:

Reaction equation describing urea hydrolysis by urease

Figure 6: Reaction equation describing urea hydrolysis by urease[12]

Urea from urine mixing with urease in feces causing ammonia emissions.

Figure 7: Urea from urine mixing with urease in feces causing ammonia emissions. Created with BioRender.com

Urea in urine thus plays a key role in ammonia production by cows. In order to work towards a new solution to fight the nitrogen crisis, for ammonia specifically, it is, therefore, important to understand how cattle produce urea.

Cows have four different stomachs. In the first stomach, the rumen, microbes are present. These microbes break down carbohydrates and proteins of the feed that the cow ingests through fermentation. Through this process the microbes produce their own nutrients, but also the nutrients that the cow takes up. Therefore, feeding the cow actually means feeding the microbes of the cow, which in turn feed the cow[13].

As can be seen in Figure 8, microbes in the rumen ferment carbohydrates to:

  • Oligosaccharides or disaccharides, which get broken down even further to monosaccharides, and lastly to ATP, which the microbes can use in their own metabolism
  • Carbon dioxide (CO2) and methane (CH4), which get belched out by the cow
  • Volatile fatty acids, which will enter the bloodstream to go to the liver, where they will be converted to ATP. This ATP will then be used by the cow to produce milk and to grow. Volatile fatty acids are the major energy source (70%)for the cow[13].

As can be seen in Figure 8, microbes in the rumen ferment protein to:

  • Peptides, which get broken down even further to individual amino acids. The microbes can directly take up these amino acids to use in their own metabolism, but they can also break down the amino acids even further to ammonia (NH3), which they can also take up and use in their own metabolism. Any ammonia that is not taken up by the microbes, will enter the bloodstream to go to the liver, where it will be converted to urea, most of which will be secreted through the urine[13].

Carbohydrate and protein digestion by cattle.

Figure 8: Carbohydrate and protein digestion by cattle. Created with BioRender.com

However, there is a balance between carbohydrate and protein fermentation by ruminal microbes. Microbes need both energy (from ATP after carbohydrate fermentation) and protein (from amino acids and ammonia after protein fermentation) for growth and multiplication. If either energy or protein is in short supply, microbial growth declines and so does the fermentation of the feed[13].

This means that if carbohydrate fermentation is suboptimal (compared to protein intake by the feed), that protein fermentation will be affected, as shown in Figure 9. If carbohydrate fermentation is suboptimal (A), less ATP will be produced and taken up by the microbes (B). If energy (ATP) is limited, microbes become less efficient at taking up and using ammonia (C). Therefore, more ammonia will go to the liver and be converted to urea (D). As a result, more urea will end up in the urine (E) and therefore more ammonia will be emitted once the urine mixes with feces. All in all, suboptimal carbohydrate fermentation, relative to protein intake, leads to more ammonia emissions by cattle[13].

Carbohydrate and protein digestion by cattle, if carbohydrate fermentation is suboptimal compared to protein intake.

Figure 9: Carbohydrate and protein digestion by cattle, if carbohydrate fermentation is suboptimal compared to protein intake. Created with BioRender.com

Conclusion

The Netherlands emmits too many nitrogen-containing compounds, one of which is ammonia. Once ammonia deposits on nitrogen-sensitive nature areas such as Natura2000 areas, it can greatly harm biodiversity and the quality of habitats.

Since agriculture is the main source of these ammonia emissions, current measures have mainly focused on curbing agriculture. However, these measures endangers the income security of farmers and can negatively affect global food security.

Within agriculture, cattle are the main source of ammonia emissions. In this case, ammonia gets emitted once urea from urine mixes with urease from feces. Cows produce more urea if carbohydrate digestion is suboptimal compared to protein intake.

A new solution is needed to protect the Dutch environment while taking into account the (financial) viability of the Dutch agricultural sector and its role in the global food supply.

More information on the scope of the problem and the involvement of agriculture in the nitrogen crisis can be found on our Human Practices page.

Our solution

Lowering ammonia emissions with a feed additive

With our project, BYE-MONIA, we aimed to find an innovative, safe and circular solution to the Dutch nitrogen crisis while fighting both the environmental as well as the economical problems caused by the nitrogen crisis.

Therefore, we decided to tackle the problem at the source by preventing the release of ammonia emissions already in the digestive tract of cattle. In order to do this, we have engineered Saccharomyces spp. (Saccharomyces cerevisiae and Saccharomyces paradoxus) to produce alpha-amylase.

Alpha-amylase is a glycosidase: an enzyme that catalyzes the hydrolysis of glycosidic bonds in complex sugars. More specifically: alpha-amylase catalyzes the endohydrolysis of (1->4)-alpha-D-glucosidic linkages in polysaccharides containing three or more (1->4)-alpha-linked D-glucose units, as shown in Figure 10. Alpha-amylase acts on starch, glycogen and related polysaccharides and oligosaccharides, and does so in a random manner. The term “alpha” relates to the fact that reducing groups are liberated in the alpha-configuration[14].

Hydrolysis of complex sugars containing three or more (1->4)-alpha-linked D-glucose units.

Figure 10: Hydrolysis of complex sugars containing three or more (1->4)-alpha-linked D-glucose units. Created with Chemdraw

Once alpha-amylase is purified and added to the cow feed, it will help break down the carbohydrates in the feed, thus reducing the formation of urea and ammonia and enhancing the growth and milk production of the cow.

As shown in Figure 11, alpha-amylase will enhance the breakdown of the polymeric carbohydrates (A), consequently making it more accessible to the microbes for fermentation. Once the fermentation of carbohydrates increases, more ATP will be produced and taken up by the microbes (B). Since more energy (ATP) is available for the microbes, they will become more efficient at taking up and using ammonia (C). As a result, less ammonia will go to the liver and be converted to urea (D). Therefore, less urea will end up in the urine (E) and less ammonia will be emitted once the urine mixes with feces.

Moreover, since alpha-amylase enhances the fermentation of carbohydrates, it is expected that more volatile fatty acids will be produced. As a result, more ATP will be produced in the liver, which enhances the energy supply. Therefore, the cow will produce more milk and grow better with the same amount of feed.

Figure 11: Carbohydrate and protein digestion by cattle, if alpha-amylase is added to the feed.

Figure 11: Carbohydrate and protein digestion by cattle, if alpha-amylase is added to the feed. Created with BioRender.com

Alpha-amylase is normally produced by a variety of organisms. However, several of these organisms, such as the Bacillus spp., are capable of nitrogen fixation[14], which is the process in which atmospheric nitrogen (N 2) is converted to more reactive nitrogen compounds such as ammonia (NH3) and nitrogen oxides (NOx). Seeing that our project aims to reduce emissions of these reactive-nitrogen compounds, we decided to clone the genes for alpha-amylase in an organism incapable of nitrogen fixation: Saccharomyces spp.

Moreover, Saccharomyces spp. are a known microbial cell factory for large scale industrial production[16], easily amenable to genetic manipulation[17] and modular toolkits have been produced to facilitate easy genetic engineering of this organism[18]. And most importantly: Saccharomyces spp. are able to take up ammonia and use it in their own metabolism for the production of glutamate and glutamine, which can then be converted to other amino acids[19]. Hence, working with Saccharomyces spp. gives us an extra way of reducing ammonia emissions: by giving excess ammonia back to our GMO as its sole nitrogen source.

Nitrogen regulation in S. cerevisiae. Protonated ammonia (NH4+) can be used to produce glutamate and glutamine

Figure 12: Nitrogen regulation in S. cerevisiae. Protonated ammonia (NH4+) can be used to produce glutamate and glutamine[18]

We have, therefore, expressed alpha-amylase in several different Saccharomyces strains to find both the optimal producer of alpha-amylase, as well as the strain that was able to grow the best in the presence of ammonia. Moreover, we have tested several different promotorts, signal sequences and alpha-amylase genes and used artificial intelligence to create over 600 different combinations and find the best genetic construct for optimal alpha-amylase production. More information on the genetic constructs and strains we tested can be found on our Engineering page, Modeling page and Results page.

Capturing ammonia emissions with a filter device

In order to feed any residual ammonia back to our GMO, while still keeping our GMO confined within the safe borders of the lab, we decided to use a state-of-the-art material: a Metal Organic Framework (MOF).

MOFs are a class of materials that consist of metal ions and organic molecules arranged in an ordered fashion. These materials are characterized by high porosity that arises from the way the metal ions and organic molecules are arranged in space on a molecular level. MOFs thus have a large number of tiny cavities that are several nanometers in diameter. This allows certain MOFs to accommodate ammonia in its gaseous state inside of the cavities[20]. Once in the cavity, ammonia is trapped and can be liberated by increasing the temperature[21] or decreasing the pressure[22]. In practice, these materials are powders.

Therefore, BYE-MONIA aims to place MOFs in the ventilation system of barns (poultry and pig barns specifically, for more information see our implementation page). After the MOF is filled up with ammonia, the MOF will be transported to the lab, where it is heated up to release the ammonia. The ammonia can then be fed to our GMO while the MOF can be reused in the barn.

With the implementation of the MOF, BYE-MONIA becomes a circular project in which a feed additive is used to decrease ammonia emissions, while any leftover ammonia emissions will be fed back to our GMO to produce even more feed additive, all while keeping the GMO confined to the lab.

BYE-MONIA as a circular project

Figure 13: BYE-MONIA as a circular project

Safe-by-design

To ensure the safety of BYE-MONIA for humans, animals and the environment, BYE-MONIA was designed with a safe-by-design approach. Not only did we design a kill switch, but various other measures have been put in place to prevent any harm from occurring due to our GMO, our alpha-amylase or the design and implementation of our project. More information on this topic can be found on our Human Practices page.

Conclusion

Alpha-amylase is a glycosidase, commonly expressed by nitrogen fixing bacteria. BYE-MONIA aims to reduce the formation of ammonia, and for this reason expresses alpha-amylase in a non-nitrogen fixing organism: Saccharomyces spp.. Different Saccharomyces strains and genetic constructs were used to determine the optimal combination for alpha-amylase production and growth in the presence of ammonia.

Once purified and added to the feed, alpha-amylase optimizes the fermentation of carbohydrates by ruminal microbes of cattle. This reduces the formation of urea, therefore, reducing the emission of ammonia by cattle. Moreover, the production of volatile fatty acids is expected to increase, consequently enhancing the milk production and growth rate of the cow.

A specialized filter device, a MOF, is used to capture any leftover ammonia and feed this back to our GMO. The GMO is therefore able to stay confined to the lab and use the leftover ammonia as its sole nitrogen source, while the MOF can be reused.

BYE-MONIA is designed with the safe-by-design principle to minimize any risks to humans, animals and the environment.

Inspiration

To come up with a project idea, we started brainstorming in March about topics ranging from allergies to producing bio-odors. However, after a month of deliberation, our top contender for a project idea was to find a solution to the Dutch nitrogen crisis. We were mainly inspired to go for this topic after the 2019 nationwide farmers protests[10] in which farmers blocked off highways and entire city squares. These protests also inspired us to work on the part of the nitrogen crisis in which farmers are the most involved and by which they are the most affected: the ammonia emissions originating from farms.

Once we had chosen this problem as our topic, we first looked into the possibility of creating a biofilter that could convert ammonia back into atmospheric nitrogen. However, we soon found out that this project idea had too many downsides, including it being infeasible in the timespan we had and releasing harmful greenhouse gasses to the environment. More on this topic, including the background research we did, can be found on our contribution page. Due to these downsides of the first project approach, we chose to, instead of breaking down ammonia, go the opposite route and focus on assimilating ammonia to build something useful.

We soon found that certain organisms are capable of using ammonia as a nitrogen source in their metabolism, including Saccharomyces cerevisiae. Moreover, we found several feed additives that were marketed as being capable of reducing ammonia emissions in livestock. However, these feed additives were mainly produced in organisms that are capable of fixating nitrogen to ammonia, therefore creating more of what we wanted to prevent (ammonia emissions). Therefore, we decided to combine the best of both worlds and engineer a non-nitrogen-fixating organism that was not only capable of taking up excess ammonia but could actually use this ammonia to make a feed additive that would reduce ammonia emissions even further. Lastly, we found out about these new filter devices, MOFs, that would allow us to bring the ammonia from the farms to the lab, therefore allowing us to keep our GMO within the confined environment of the lab.

All in all, after extensive brainstorming and literature research, and going through several design-build-test-learn-cycles, we were confident that we found a promising project idea to fight the nitrogen crisis in the Netherlands.

Key values

Safe to use for humans, animals and the environment

Created with a circular design by feeding ammonia back to our yeast

Prevents ammonia emissions already in the intestinal tract of the cow

Utilizes the best genetic construct for alpha-amylase production and use only leftover ammonia as a nitrogen source

Provides farmers with a feed additive that enhances milk production and growth of cattle

Learn more about our project

Engineering

Engineering

Implementation

Implementation

Model

Model

Safety

Safety

Human practices

Human practices

Education

Education

References

  1. Rijksinstituut voor Volksgezondheid en Milieu, “Stikstof | RIVM.” [Online]. Available: https://www.rivm.nl/stikstof. [Accessed: 11-Oct-2021].
  2. European Environment Agency, “National Emission reduction Commitments Directive .” [Online]. Available: https://www.eea.europa.eu/themes/air/air-pollution-sources-1/national-emission-ceilings. [Accessed: 11-Oct-2021].
  3. European Commission, “Commission recommendations for The Netherlands’ CAP strategic plan Accompanying the document COMMUNICATION FROM THE COMMISSION TO THE EUROPEAN PARLIAMENT, THE COUNCIL, THE EUROPEAN ECONOMIC AND SOCIAL COMMITTEE AND THE COMMITTEE OF THE REGIONS Recommendations to the Member States as regards their strategic plan for the Common Agricultural Policy.”, Commission staff working document, 2020.
  4. European Commission, “‘Branding’ Natura 2000 goods and services,” Nat. 2000 Newsletter, Nat. Biodivers. Newsl., no. 50, Jul. 2021.
  5. European Commission, “Natura 2000 - Nature and biodiversity - Environment.” [Online]. Available: https://ec.europa.eu/environment/nature/natura2000/index_en.htm. [Accessed: 11-Oct-2021].
  6. A. M. Schmidt and R. A. Smidt, “Scientific analysis of the status of designated Natura 2000 areas and the protection of nitrogen-sensitive species and habitats Dutch contribution,” Wageningen Environ. Res., Apr. 2018.
  7. N. en V. Natura 2000 - Ministerie van Landbouw, “Natura 2000 gebieden .” [Online]. Available: https://www.natura2000.nl/gebieden. [Accessed: 11-Oct-2021].
  8. “Absolute emissiereeks ammonia (in [kg]) naar Lucht 2019,” Emmissieregistratie, Aug-2021. [Online]. Available: http://www.emissieregistratie.nl/erpubliek/erpub/weergave/grafiek.aspx. [Accessed: 21-Aug-2021].
  9. Rijkswaterstaat Ministerie van Infrastructuur en Waterstaat, “NEC-stoffen - Kenniscentrum InfoMil.” [Online]. Available: https://www.infomil.nl/onderwerpen/lucht-water/lucht/nec-stoffen/. [Accessed: 29-Aug-2021].
  10. E. Stokstad, “Nitrogen crisis threatens Dutch environment—and economy,” Science (80-. )., vol. 366, no. 6470, pp. 1180–1181, Dec. 2019, doi: 10.1126/SCIENCE.366.6470.1180.
  11. M. Dolman, G. Jukema, and P. Ramaekers, “De Nederlandse landbouwexport 2018 in breder perspectief,” Wageningen Econ. Res., Jan. 2019, doi: 10.18174/468099.
  12. G. J. Monteny and J. W. Erisman, “Ammonia emission from dairy cow buildings: a review of measurement techniques, influencing factors and possibilities for reduction,” Netherlands J. Agric. Sci., vol. 46, no. 3–4, pp. 225–247, Dec. 1998, doi: 10.18174/NJAS.V46I3.481.
  13. John Moran, Tropical dairy farming: feeding management for small holder dairy farmers in the humid tropics. Chapter 5: How the rumen works. Landlinks Press, 2005.
  14. “Information on EC 3.2.1.1 - alpha-amylase,” Brenda. [Online]. Available: https://www.brenda-enzymes.org/enzyme.php?ecno=3.2.1.1. [Accessed: 11-Oct-2021].
  15. Y. J, T. J, R. M, K. S, P. A. A, and A. A. MH, “Nitrogen fixing potential of various heterotrophic Bacillus strains from a tropical estuary and adjacent coastal regions,” J. Basic Microbiol., vol. 57, no. 11, pp. 922–932, Nov. 2017, doi: 10.1002/JOBM.201700072.
  16. O. JM, C. D, P. KR, P. SG, O. L, and N. J, “Industrial systems biology of Saccharomyces cerevisiae enables novel succinic acid cell factory,” PLoS One, vol. 8, no. 1, Jan. 2013, doi: 10.1371/JOURNAL.PONE.0054144.
  17. M. Parapouli, A. Vasileiadis, A.-S. Afendra, and E. Hatziloukas, “Saccharomyces cerevisiae and its industrial applications,” AIMS Microbiol., vol. 6, no. 1, p. 1, 2020, doi: 10.3934/MICROBIOL.2020001.
  18. M. E. Lee, W. C. DeLoache, B. Cervantes, and J. E. Dueber, “A Highly Characterized Yeast Toolkit for Modular, Multipart Assembly,” ACS Synth. Biol., vol. 4, no. 9, pp. 975–986, Sep. 2015, doi: 10.1021/SB500366V.
  19. B. Magasanik and C. A. Kaiser, “Nitrogen regulation in Saccharomyces cerevisiae,” Gene, vol. 290, no. 1–2, pp. 1–18, May 2002, doi: 10.1016/S0378-1119(02)00558-9.
  20. A. J. Rieth and M. Dincă, “Controlled Gas Uptake in Metal–Organic Frameworks with Record Ammonia Sorption,” J. Am. Chem. Soc., vol. 140, no. 9, pp. 3461–3466, Mar. 2018, doi: 10.1021/JACS.8B00313.
  21. A. J. Rieth, Y. Tulchinsky, and M. Dincă, “High and Reversible Ammonia Uptake in Mesoporous Azolate Metal–Organic Frameworks with Open Mn, Co, and Ni Sites,” J. Am. Chem. Soc., vol. 138, no. 30, pp. 9401–9404, Aug. 2016, doi: 10.1021/JACS.6B05723.
  22. D. W. Kim et al., “High Ammonia Uptake of a Metal–Organic Framework Adsorbent in a Wide Pressure Range,” Angew. Chemie Int. Ed., vol. 59, no. 50, pp. 22531–22536, Dec. 2020, doi: 10.1002/ANIE.202012552.
  23. Icons made by Freepik and Nhor Phai from www.flaticon.com