Techno-economic optimization of renewable urea production for sustainable agriculture and CO2 utilization

Urea is the most used nitrogen fertilizer due to its ease of storage, transportation, and application. It is made by combining ammonia and carbon dioxide (CO2), both of which are produced predominantly from fossil fuels at present. The recent momentum behind ammonia production using renewable-powered electrolysis offers an opportunity to both make urea in a more sustainable way and utilize CO2 from external sources. In this work, we present a techno-economic optimization model to minimize the cost of making urea in this way. The model allows for time-varying chemical production in response to renewable variability by simultaneously optimizing production facility design and hourly operation. We performed a case study for Minnesota considering the use of byproduct CO2 from bioethanol production. We found that the present-day levelized cost of renewable urea is between $268 mt−1 and $413 mt−1 at likely implementable production scales up to 250 000 mt yr−1. This is within the range of historical conventional urea prices while offering at least 78% carbon intensity reduction. Projecting to 2030, there is a clear economic case for renewable urea production with levelized cost as low as $135 mt−1 due to technology improvement and electrolysis manufacturing expansion, facilitating a urea production scale increase to 525 000 mt yr−1. Optimal facilities use wind energy, with hydrogen and ammonia production operating in a flexible, time-varying way to minimize battery and hydrogen storage capacities. Urea production operates near steady state due to the relatively low cost of intermediate ammonia buffer storage. A mix of imported methane and locally produced hydrogen are used to provide heat for steam consumed in the urea synthesis.


Introduction
The transition to lower carbon intensity alternatives in sectors such as power generation or ground transportation is already underway in an effort to mitigate anthropogenic impact on the planet.Certain industries will take longer to decarbonize economically (e.g.aviation fuel production) and could thus deploy carbon dioxide (CO 2 ) capture as an interim means of reducing carbon intensity [1].Other industries such as cement production will inherently include carbon capture in their sustainable operation [2,3].Widely used chemicals such as plastics are manufactured from carbon-containing feedstocks.This could be a future avenue for utilization of captured CO 2 .Specifically, using CO 2 and hydrogen to produce chemicals containing one (e.g.methane or methanol) or two carbon molecules (e.g.ethane, ethanol, ethylene) has been the topic of extensive research within the last decade [4][5][6][7].These chemicals are consumed across a range of end-uses while also acting as platform chemicals for longer-chain molecules.Continued research in this area is needed to ensure the viability of this CO 2 utilization approach which is not currently practiced at scale [8][9][10].
Urea production presents a unique opportunity for CO 2 utilization.Urea is synthesized from CO 2 and ammonia and its production is responsible for 57% of industrial CO 2 consumption [11].Urea is the most commonly used nitrogen fertilizer, making up approximately 50%-65% of global fertilizer application in a given year [12,13].It is produced in the form of solid granules which makes it both easier to use and less likely to volatilize than liquid fertilizers.Furthermore, careful urea application can reduce nitrogen runoff and subsequent eutrophication compared to other nitrogen fertilizers [14].Decarbonizing synthetic nitrogen fertilizer production is critical.It is currently responsible for approximately 2% of global CO 2 emissions [15].These nitrogen fertilizers are nonetheless essential for modern agriculture, providing nutrient nitrogen for approximately 50% of global crops [16] and total fertilizer demand projected to increase 50% -75% by 2050 [17].
Urea production for CO 2 utilization can leverage advances made over the last decade in making ammonia from hydrogen obtained via renewable-power water electrolysis.Recent studies have shown that the economics of making this renewable ammonia are promising in regions with rich renewable resources and should continue to improve as the involved technologies get less expensive and more efficient [18][19][20][21].The promise of renewable ammonia is not limited to theory; a number of projects with scales ranging from 10 metric tons per year (mt yr −1 ) to over 100 000 mt yr −1 have been announced in recent years [22].Further processing of this ammonia to urea could accelerate the transition to more sustainable fertilizers by ensuring as few consumers (e.g.farmers) as possible have to materially change their practices.
At present, the majority of urea manufacturing occurs directly at ammonia production facilities.Reactant CO 2 is obtained from the reforming process used to derive hydrogen from the fossil fuel feedstock.These facilities usually have annual nameplate capacities above 1 000 000 mt yr −1 to achieve economies of scale.Urea is traded on the global market.Its price can fluctuate depending on the price of the natural gas input and the food that urea is used to produce.The average Black Sea spot market price since 2010 is $335 mt −1 , with a range of $145 mt −1 -$925 mt −1 [23].For renewable urea production to be economically competitive, minimizing the cost of producing feedstock ammonia from renewable energy as well as that of obtaining feedstock CO 2 will be essential.Using renewable electricity to produce urea from this CO 2 also necessitates an alternative heat source because the urea synthesis process is conventionally heat integrated with upstream fossil fuel reforming.These factors must be carefully considered in order to achieve economic competitiveness.
There are relatively few studies in the literature which analyze urea production from renewable ammonia and CO 2 .Kim et al have conducted two recent studies in this area [24,25].They considered continuous hydrogen production via alkaline electrolysis powered by constantly available renewable electricity at prices of $50 MWh −1 -$70 MWh −1 and captured CO 2 with a price of $74 mt −1 CO 2 .They determined urea production costs between $1175 mt −1 and $1330 mt −1 in the 300 000 to 350 000 mt yr −1 scale range.For reference, they determined that methane-based urea production would cost $550 mt −1 under the same set of assumptions [24].Electricity costs were the predominant reason that renewable urea production was uneconomical.Fernando and Purwanto analyzed a similar urea production concept with CO 2 captured from a coal combustion power plant [26].Their studied facility sourced all energy from variable PV generation with 25% annual capacity factor (CF) and included batteries to enable continuous electrolysis operation.They found a present-day urea production cost of $2300 mt −1 at the 13 000 mt yr −1 scale, with a reduction to $1650 mt −1 using 2050 cost projections.Facilitating continuous hydrogen production led in part to the high urea production costs as batteries contributed over 30% to capital investment.
The research in this paper aims to achieve lower urea production costs by exploring flexible, time-varying chemical production to facilitate the use of low-cost renewable energy without incurring high energy and/or chemical intermediate buffer storage costs.We do this by developing a combined design and scheduling optimization model which minimizes the levelized cost of urea production (LCOU) while accounting for both hourly and seasonal variability in renewable energy availability.Specifically, this model determines the size as well as hourly production rates and storage inventories of each involved technology in the urea production facility.Integrating design and scheduling decision-making allows an annual production target to be met at lowest cost while ensuring that chemical production power and heat demands are met in each hour and the unique operating requirements of each technology are accounted for.To our knowledge, no previous analysis has employed such a whole-facility design and operation optimization approach in the context of urea production.
We use this optimization model to perform a case study for renewable urea production in Minnesota.There are numerous reasons to produce renewable urea in the U.S. Midwest.This region is the main American consumer of nitrogen fertilizer and has pockets of high quality wind resources with greater than 40% CF throughout.Furthermore, the ample bioethanol production in the area could serve as a CO 2 feedstock source.Highly pure CO 2 is already a byproduct of making bioethanol [27,28].A recent study determined there are no additional costs associated with utilizing this CO 2 [29].Finally, the U.S. Inflation Reduction Act includes production tax credits (PTC) for both renewable energy and hydrogen [30].These credits will improve the economic outlook of renewable urea.We specifically analyze urea production in three representative regions of Minnesota with different wind potential at scales between 75 000 mt yr −1 and 525 000 mt yr −1 .We use both present-day and 2030 projections for technology cost and performance.We also consider both imported methane and locally produced hydrogen or ammonia as heating options and allow the model to decide the cost-optimal fuel mix.Overall, in addition to optimizing flexible chemical production as described in the previous paragraph, this case study allows us to quantify the value proposition of near-term renewable urea production using easy-to-obtain CO 2 from an established industry.
The rest of the paper is structured as follows.Section 2 provides a description of the renewable urea production concept and the model for its optimization.It also includes the parameters for the Minnesota case study.Section 3 contains the results of the study and discussion with specific focus on the optimal facility design and economics.Section 4 outlines major conclusions and identifies directions for future work.

Renewable urea production concept
Figure 1 shows the connectivity of each constituent technology required to make urea from renewable energy and CO 2 .Wind and PV generation are used to specifically power PEM electrolysis to produce hydrogen, air separation to produce nitrogen, ammonia synthesis, and urea synthesis.This energy can also be stored in a battery to modulate its variability.Hydrogen and nitrogen are required to make ammonia via the following reaction in the Haber-Bosch (HB) process: ( Ammonia and CO 2 are then required to make urea (NH 2 CONH 2 ) via the following reactions in a two-step process where ammonium carbamate (NH 2 COONH 4 ) is first synthesized and then dehydrated: Chemical intermediates can be stored to decouple the production rates of successive processes.For example, such storage would enable electrolysis, ammonia synthesis, and urea synthesis to operate at different stoichiometry-adjusted rates.CO 2 is assumed to be available at a constant rate to reflect steady-state bioethanol production, so CO 2 storage can be used to facilitate time-varying urea production.Urea storage can then allow for a constant offtake.Urea production requires steam and thus heat to generate this steam.This is provided by boilers fueled by imported methane and/or hydrogen and ammonia produced at the facility.

Combined design and scheduling optimization model
The optimal combined design and scheduling model minimizes the levelized cost of urea (LCOU) by determining the size and operating schedules of all required process units i ∈ I. Specifically, it determines the size of wind generation x w and PV generation x pv , chemical production rates x c , storage x s , and boilers x q which gives the lowest LCOU for a given production scenario.Simultaneously, the optimal hourly chemical production m c (t), battery charge p + b (t) and discharge p − b (t), storage inventories L s (t), and heating rates using methane q m (t), hydrogen q h2 (t), and ammonia q nh3 (t) are determined in response to hourly variable wind and PV generation which have respective CFs π w (t) and π pv (t).This combined design and scheduling approach ensures that the annual urea production target σ u is met using the annually available CO 2 amount σ co2 , facility power and heat demands are met in each hour, and all units are operated within their specifications.The model equations are: Equation ( 4) is the LCOU; this is the objective function which is minimized by the model.Equations ( 5) and ( 6) respectively define capital and operating costs.In general, capital costs obey power law correlations with reference cost CAP ref i , reference capacity χ ref i , and scaling exponent γ i .Operating costs for renewable urea production arise from fixed operation and maintenance (O&M) costs which are assumed to be an annual fraction β u of the capital cost of each unit and purchasing methane at price α m .There are revenues from PTCs for wind energy, PV energy, and hydrogen production with credit values κ w , κ pv , and κ h2 respectively.These credits will be received for the first 10 years of production; their values are levelized for projects with a longer lifetime.
Equation (7) ensures on an hourly basis that energy generated from wind and PV resources as well as that discharged from the battery meets or exceeds the energy required for chemical production with specific energy requirements ρ c and battery charging.Equation (8) ensures on an hourly basis that heat produced via the boilers using methane, hydrogen, and/or ammonia as fuel with efficiency η q meets or exceeds the urea synthesis heat demand.
Equations ( 9) through ( 14) are inventory balances for battery, hydrogen, nitrogen, ammonia, CO 2 , and urea storage.Battery charge and discharge efficiencies are η + b and η − b respectively.The 3/17 and 14/17 coefficients which appear in the hydrogen and nitrogen inventory balances relate the ammonia production rate m nh3 to hydrogen and nitrogen consumption through the ammonia synthesis reaction stoichiometry.Similarly, the 4/7 and 36/49 coefficients which appear in the ammonia and CO 2 inventory balances relate the urea production rate m u (t) to ammonia and CO 2 consumption through the urea synthesis reaction stoichiometry.The quantity of hydrogen and ammonia consumed as heating fuel is related to the inventory of their respective storage by lower heating values λ h2 and λ nh3 .Equations ( 15)-( 17) are lower and upper bounds on chemical production rates, storage inventories, and the boiler heat production rate.The upper bound on these scheduling decisions is the optimal size of the relevant process.The lower bound is defined as a fraction of this size, specifically δ c for chemical production, δ s for storage, and δ q for the boilers.

Case study data for renewable urea production in Minnesota
The average byproduct CO 2 output from Minnesota bioethanol production facilities is 185 000 mt yr −1 [31].This corresponds to a urea production potential of 250 000 mt yr −1 .This forms the baseline annual nameplate production capacity for this study.The range of urea production potential at existing bioethanol production facilities is 75 000 mt yr −1 -525 000 mt yr −1 .We therefore also investigate economics between these extremes to understand economies of scale.The location of Minnesota's 19 operating ethanol plants is given in figure 2. These plant locations are partitioned into three regions based on wind potential: Southwest (SW MN), Southeast (SE MN), and Central West (CW MN) Minnesota.The annual average wind CFs are 48.1% in SW MN, 48.8% in SE MN, and 41.3% in CW MN.The annual average PV CFs are respectively 15.8%, 16.1%, and 14.9% in these regions.The hourly resolution time series data for these CFs used as input to the optimization model were obtained from the NREL TMY3 database [32].
Capital investments are annualized using a scaled project lifetime of 10.23, which corresponds to a project lifetime of 25 years and a 8.5% weighted cost of capital.Revenue from renewable energy and hydrogen PTCs are scaled by a factor of 0.64 since they are only received over the first ten years of production.Thus, for the 25 year project, the renewable energy PTC has a levelized value of $16.96 MWh −1 while the hydrogen PTC has a levelized value of $1920 mt −1 H 2 provided the carbon intensity of hydrogen production is below 0.45 mtCO 2 mt −1 H 2 [30].
The capital and operating cost along with energy efficiency and operating flexibility for each constituent technology are described herein.Wind and PV generation have present-day capital costs of 1.27 MM$ MW −1 and 1.25 MM$ MW −1 with operating costs of 0.029 MM$ MW −1 and 0.02 MM$ MW −1 respectively.By 2030, capital costs are projected to decrease to 1.15 MM$ MW −1 and 0.97 MM$ MW −1 and operating costs are projected to decrease to 0.027 MM$ MW −1 and 0.018 MM$ MW −1 [33].Batteries have both energy storage (MWh) and charge/discharge (MW) components; each have associated capital and operating cost.Their storage cost is 0.28 MM$ MWh −1 and their power interface cost is 0.34 MM$ MWh −1 , with projected reduction to 0.24 MM$ MWh −1 and 0.29 MM$ MW −1 by 2030 [33].Batteries have an annual operating cost of 0.0072 MM$ MWh −1 and 0.0072 MM$ MW −1 for both energy and power components, with projected reduction to 0.0057 MM$ MWh −1 for the power component by 2030.The charge and discharge efficiencies are 90% and 95% respectively.Chemical production capital costs are determined with power law scaling relationships.Electrolysis has a reference capacity of 20 MW, a scaling exponent of 0.95, and a present-day reference cost of 1.2 MM$ MW −1 with projected reduction to 0.85 MM$ MW −1 by 2030 [34].Electrolysis has an annual O&M cost of 3% of capital cost.The efficiency of electrolysis also improves toward 2030, with total system energy intensity of 55 MWh mt −1 H 2 today and 50 MWh mt −1 H 2 by 2030 [34].Hydrogen storage capital costs are also projected to decrease slightly, from 0.8 MM$ mt −1 H 2 today to 0.7 MM$ mt −1 H 2 by 2030 [34].
Nitrogen production via cryogenic air separation has a reference cost of 3.16 MM$, a reference capacity of 1000 mtN 2 yr −1 , a scaling exponent of 0.52, energy intensity of 0.12 MWh mt −1 N 2 , and can ramp down its production to 60% of installed capacity [35].Nitrogen storage has a capital cost of 1× 10 −2 MM$ mt −1 N 2 .Ammonia production via Haber-Bosch synthesis has a reference cost of 1.55 MM$, a reference capacity of 1000 mtNH 3 yr −1 , a scaling exponent of 0.75, energy intensity of 0.6 MWh mt −1 NH 3 , and can ramp down its production to 10% of installed capacity [35][36][37].Ammonia storage has a capital cost of 1× 10 −3 MM$ mt −1 NH 3 .Urea synthesis has a reference cost of 4.05 MM$, a reference capacity of 1000 mt yr −1 , a scaling exponent of 0.58, electrical energy intensity of 0.18 MWh mt −1 , a heat requirement of 0.95 MWh heat mt −1 , and can ramp down its production to 50% of installed capacity [38][39][40][41][42]. CO 2 storage has a capital cost of 6.5× 10 −2 MM$ mt −1 CO 2 , while urea storage has a capital cost of 1× 10 −4 MM$ mt −1 .The annual O&M cost for chemical production and storage is assumed to be 3% of installed capital cost.Hydrogen, nitrogen, and CO 2 must be stored above atmospheric pressure; this corresponds to a inventory lower bound of 0.5% for 200 bar storage.Ammonia and urea are stored as liquid and solid granules respectively, so these inventories can be fully emptied.The boilers used for heat generation have a capital cost of $0.5 MM$/MW heat , a combustion efficiency of 90% based on lower heating value, and can ramp down its heat generation to 50% of installed capacity [43].For reference, the lower heating values of hydrogen and ammonia are 33.3MWh mt −1 H 2 and 5.2 MWh mt −1 NH 3 respectively.

Results and discussion
In this section we discuss the optimal design and economics of renewable urea production in Minnesota today and projected to 2030.We then contextualize these results with respect to the current urea market from the perspective of both economics and carbon intensity.Table 1 shows optimal renewable production facility design trends, which are similar across the three studied locations.All renewable electricity is obtained from wind generation.Lower CFs for PV generation make its inclusion uneconomical, even though such inclusion would result in more hours of available energy due to the temporal synergy with wind generation.Electrolysis produces hydrogen almost exclusively when wind generation is available.Its CFs are the same or only incrementally higher than wind CFs in each location.This is expected because storing energy for a meaningful quantity of hydrogen production in a battery is much too expensive from a capital investment perspective.The air separation CF is much higher, in the range of 83%-88%.Nitrogen production uses minimal energy and it is thus relatively inexpensive to install battery capacity to provide its power requirement when wind energy is not available.The ammonia synthesis CF is in the range of 55%-63%.This is a consequence of exploiting dynamic flexibility to minimize expensive hydrogen storage capacity.At most 90 h of hydrogen storage are installed in terms of the minimum hydrogen consumption rate for ammonia synthesis (i.e. when operating at 10% of installed capacity).On the other hand, at least an order of magnitude more nitrogen buffer storage is required in terms of its minimum consumption rate for ammonia synthesis due to closer-to-constant nitrogen production rate.We previously observed these design and operational paradigms in the context of renewable ammonia production [21].In contrast, optimal urea production is close to steady state.This enables high utilization of the required capital investment while also minimizing feedstock CO 2 storage capacity, with less than 20 h installed.Consequently, minimal urea storage is needed to enable a constant urea offtake from the facility.Matching this near constant urea production with variable ammonia production requires between 670 and 950 h of ammonia storage, but this only contributes minimally to the LCOU because liquid ammonia storage is relatively inexpensive.
The optimal facilities employ a hybrid heating approach using both imported methane and hydrogen made onsite.The annual heating fuel mix consists of between 10% and 25% hydrogen across the three studied locations and scales.This approach requires some additional capital investment for electrolysis, but this is outweighed by revenue from hydrogen PTCs.Using ammonia for heating is not economically optimal in this hybrid approach.The additional capital investment and energy requirement to make ammonia outweighs its lower storage cost in this situation where renewable hydrogen can be used when available and storage is not needed.The following discussion about the effect of production location and scale will focus on economics trends of urea production using this hybrid heating strategy.
Figure 3 shows present-day optimal LCOU for the three studied locations and production scales.Renewable urea production derives significant benefit from wind and hydrogen PTCs, which specifically provide an approximate value of between $307 mt −1 and $321 mt −1 levelized over the lifetime of the project.Without these credits, the LCOUs would be between $560 mt −1 and $720 mt −1 .These costs nontheless compare favorably to those determined for similar production in the literature, specifically $2300 mt −1 at 13 000 mt yr −1 [26] and $1150 mt −1 at 330 000 mt yr −1 [25].This demonstrates the benefit of flexible chemical production to effectively use variable wind energy which has an implied cost between $29 MWh −1 -$36 MWh −1 under the financial assumptions used in this work.Urea production in locations with higher wind CFs offers lower LCOU.Across each production scale, there is an approximately $85 mt −1 difference between the most and least expensive production in CW MN and SE MN respectively (see figure 3).The main driver of this higher cost is simply the need for additional investment in wind generation capacity to achieve similar annual energy production.Hydrogen and ammonia production also operate with lower annual CFs (see table 1), so further oversizing and thus additional investment is needed to ensure there is enough ammonia feedstock to meet the annual urea Net LCOU including levelized production tax credit displayed above top horizontal axis.production target.A major difference between the optimal facilities in each location is the amount of heat provided by hydrogen fuel: 25% in SE MN compared to only 10% in WC MN.More methane is favored in WC MN due to the higher cost of generating energy arising from the lower wind annual CF.The differences in optimal design for SE MN compared to the other two regions is also noteworthy.The optimal SW facility produces hydrogen, nitrogen, and ammonia production at higher CFs and requires hydrogen and nitrogen buffer storage to facilitate this.This is not necessarily intuitive given the similar 48.1% and 48.8% annual wind CFs for SW and SE MN.Evidently, other wind resource characteristics such as short-term variability and seasonality affect the optimal design and economics of these urea production facilities.This underscores the importance of using granular renewable generation time series data to optimize facility designs in each specific location.
Larger urea production scales offer lower LCOU as would be expected due to economies of scale (see figure 3).This phenomenon specifically manifests itself with lower capital intensity for nitrogen, ammonia, and urea production processes.On the other hand, there exists minimal economies of scale for wind generation and electrolysis, which are the two largest contributors to the LCOU.Overall, LCOU at the 525 000 mt yr −1 scale is approximately $90 mt −1 lower than at the 75 000 mt yr −1 scale in each location.There are however other considerations in determining the appropriate production scale.Only four of the 19 bioethanol facilities in Minnesota could support over 450 000 mt yr −1 of urea production: one each in SW and CW MN, and two in SE MN.Pursuing only the largest nameplate urea production may be to the detriment of logistical factors such as proximity to urea demand or the local interest in developing such a project.Furthermore, these >450 000 mt yr −1 facilities would require at least 800 MW and 700 MW of wind and electrolysis capacity respectively if developed using present-day technologies.At present, the largest wind farm in Minnesota has a capacity of 250 MW [44].Meanwhile, the global manufacturing capacity of PEM electrolysis today is 1.55 GW according to the IEA [45]; a single urea project at this scale would require 45% of that.Overall, economies of scale are important but should not preclude development of urea production at scales above 200 000 mt yr −1 , where LCOU is only roughly 10% higher than at 525 000 mt yr −1 .
Projecting the LCOU to 2030 is of particular interest because production must begin by 2032 to monetize the PTCs.Overall, technology improvements result in roughly $105 mt −1 LCOU reduction at each scale in SE MN and SW MN, and $115 mt −1 LCOU reduction in CW MN (see figure 4).Wind generation and electrolysis are less expensive for the same capacity and this offers more benefit in CW MN because these technologies have higher installed capacities to account for the lower wind potential.The installed capacities of all technologies are comparable to the present-day optimal facilities.More hydrogen is used as a heating fuel to increase the revenue from hydrogen PTCs; this is economical due to the increased electrolysis efficiency.Specifically, the optimal facilities obtain 46% (SW), 51% (SE), and 17% (CW) of their heat from locally produced hydrogen.Overall, projected technology improvements result in both lower costs and less fossil fuel consumption for optimal renewable urea production.Furthermore, the likelihood of higher electrolysis capacities increases as global manufacturing grows.This will allow for larger urea production facilities to better exploit economies of scale.The economic results in this work can be contextualized in terms of the current urea demand and market in Minnesota.From 2015 to 2020, the annual average urea consumption was 750 000 mt yr −1 [46].There is also a trend of urea increasingly becoming the dominant nitrogen fertilizer in the state.In the case where all anhydrous ammonia is replaced by urea, up to 1 060 000 mt yr −1 could be required.The total in-state urea production potential is 4 910 000 mt yr −1 .Thus, at most 20% of the byproduct CO 2 from bioethanol production would be required to meet Minnesota's demand.This gives an opportunity for additional urea production and export to neighboring states, or to use this CO 2 in other applications.The average final consumer price paid in Minnesota from 2010 through 2022 was $468 mt −1 [47].This price fluctuates across both weekly and annual timescales as it is by methane and food (see 5).The range of observed prices was $200 mt −1 to over $1000 mt −1 .Present-day production in the 75 000 mt yr −1 -250 000 mt yr −1 scale range would be competitive with the existing urea market, but a premium for sustainability might be required once transportation to demand and necessary profit margins for the renewable urea producer are accounted for.Renewable urea production projected to 2030 will be less expensive than all but the lowest historical conventional market prices, especially because larger production scales will likely be accessible as global electrolysis manufacturing grows.Renewable urea production also offers stable operating costs after the initial capital investment.This could translate to long-term, fixed price urea sale agreements, which would remove a major source of uncertainty for farmers.There is thus a multi-faceted economic case for the development of renewable urea projects in Minnesota.
The renewable urea production analyzed in this work is also favorable from a sustainability perspective.The carbon intensity associated with renewable production arises only from methane combustion for process heat.The present-day carbon intensity will be at most 0.17 mtCO 2 mt −1 .In comparison, the carbon intensity associated with state of the art methane-based urea production is 0.94 mtCO 2 mt −1 [48].From a broader life cycle perspective, there is also a benefit to using biogenic CO 2 byproduct from the ethanol production process rather than fossil fuels to make urea.According to the IPCC, using urea as fertilizer results in approximately 0.735 mtCO 2 mt −1 emissions after application due to the fixation of the contained nitrogen in the soil [49].In other words, using fossil-derived methane as the hydrogen source for urea production results in CO 2 emissions additional to those from the production facility itself.

Conclusion
In this work we optimized economics and facility designs for making urea from renewable ammonia and byproduct CO 2 from bioethanol production.To this end, we developed and employed a techno-economic optimization model which simultaneously sizes and schedules the hourly operation of all involved technologies.This whole-facility optimization approach allows systematic accommodation of renewable energy variability via time-varying chemical production as well as battery and chemical buffer storage toward minimized urea production costs.This model could be easily be applied to urea production using different technologies for chemical production or energy storage, other CO 2 sources, or different locations.
Economically optimal urea production facilities in Minnesota use only wind energy and do not include PV generation.Hydrogen and ammonia production operate in a flexible manner to mitigate the capacity of battery and hydrogen storage required to use this wind energy, whereas urea production is close-to-steady for better capital utilization because the ammonia intermediate is inexpensive to store.Today, it would be optimal for these facilities to obtain up to 25% of their heat from combustion of locally produced hydrogen and the balance from imported methane.The likely achieveable optimal LCOU is between $268 mt −1 and $413 mt −1 depending on production location and scale.Projecting to 2030, this cost can be reduced to a range of $135 mt −1 and $236 mt −1 owing to wind generation and electrolysis cost reductions as well as increased global electrolysis manufacturing enabling larger urea production scales.By 2030, optimal facilities obtain up to 51% of their heat from hydrogen.Overall, the outlook of renewable urea is promising today and compelling by 2030 from both economic and sustainability perspectives.The carbon intensity of the renewable urea production analyzed in this paper is at least 78% lower than conventional methane-based production.
Given the optimistic outlook for renewable urea production revealed in this study, future work will focus on renewable urea fertilizer supply chain optimization.Such an analysis will allow for better understanding of the best and logistics for production given specific spatially-resolved demands.A more broad nitrogen fertilizer supply chain could also be explored, simultaneously optimizing the production and use of urea, ammonia, or other nitrogen fertilizers based on economic and sustainability criteria.
The methods and conclusions from this work may also be valuable in the broader context of systems engineering for CO 2 utilization.The modeling approach employed herein could be used to analyze and optimize any type of chemical production which uses CO 2 and variable renewable energy as feedstocks.It is likely that facilities for other CO 2 utilization avenues would have similar underlying design motifs such as flexible production of hydrogen-derived chemicals (e.g.methanol) as final products or intermediates to mitigate hydrogen storage costs.

Figure 1 .
Figure 1.Conceptual flowsheet for urea production from renewable energy and bioethanol CO2.

Figure 2 .
Figure 2. Map of Minnesota with wind potential overlaid with bioethanol production locations.

a
Capacity factors calculated by normalizing optimal annual production quantity by annual installed capacity of technology.b Hours of storage determined based on specific minimum commodity demands: facility power for battery, ammonia synthesis for H2 and N2 storage, urea synthesis for NH3 and CO2 storage.c Hours of urea storage optimized to allow constant urea offtake from time-varying urea production.d Scope of carbon intensity limited to urea production facility, arises from combustion of imported methane for heating.

Figure 3 .
Figure3.Present-day optimal levelized cost of urea across studied locations and production scales (listed in 1000 mt yr −1 ).Net LCOU including levelized production tax credit displayed above top horizontal axis.

Figure 4 .
Figure 4. 2030 projected optimal levelized cost of urea across studied locations and production scales (listed in 1000 mt yr −1 ).Net LCOU including levelized production tax credit displayed above top horizontal axis.

Figure 5 .
Figure 5.Comparison of conventional urea market in Minnesota from 2011 to 2022 with optimal renewable urea production costs for present-day 75 000 mt yr −1 -250 000 mt yr −1 facilities and 2030 projected 250 000 mt yr −1 -525 000 mt yr −1 facilities.Larger production scales are assumed to be accessible by 2030 because more electrolysis capacity will be available.

Table 1 .
Key optimal design results for urea production in studied Minnesota locations averaged across production scales.