General comments

This article documents the early design stages of a creative, renewable Power-to-Liquid system. The article is written in generally good language. The technical aspects of the proposed energy ship are documented appropriately for a case study and assumptions/references are transparent. The economic evaluation of the concept is documented transparently too. The main critique point is specific comment no. (10). The critique refers to the methanol price projections and is decisive for the market potential of the proposed solution and eventually the conclusion of this article. I recommend this point being double-checked by another reviewer.

Specific comments 

1. Line 22: “the cost may be comparable to that of methanol produced by offshore wind farms in the long term” – see specific comment no. (10).
2. Line 35: It would be helpful for the reader if you shortly mentioned up to three main reasons for your choice of methanol, based on your referenced previous assessment.
3. Line 55 and following: As far as I understand, your proposed design has progressed and you provide comparisons/updates to previous estimates. This documentation in itself may be of value, as it showcases how weight or cost estimates develop throughout subsequent design stages. A short sentence highlighting this value could bring attention to this aspect.
4. Lines 67 & 88: you refer to eq. 2 from Barbarit et al. 2020 twice, hence it seems to be relevant for this study. Consider showing that equation explicitly here instead of only referring to the previous article.
5. Line 82 (Figure 4): You could indicate the vector of the propulsive force with an arrow in the left part of the figure. Potentially four arrows with lengths proportional to each FR’s force contribution.
6. Line 97: since the displacement has changed, I assume the hull shape has changed too. ‘The hull shape (Wigley hull) has been updated based on a more accurate displacement estimate’ could clarify this.
7. Lines 116-122: Consider mentioning the efficiency of the H2-to-methanol plant as well in order to increase transparency.
8. Lines 182 & 201: You could improve understanding by framing the annual methanol production capacity in terms of vehicles powered. E.g. units of 5000 dwt bulk carriers propelled:
   
   70,600t/year = 388,300MWh/year chemical energy
   
   assumptions annual energy consumption bulk carrier: 1,410kW x 24h/day x 180days/year = 6,091MWh/year
   
   6,091MWh / 50% thermal engine efficiency = 12,182 MWh/year chemical energy
   
   388,300MWh / 12,182MWh = 32 vessels that could be powered by the designed fleet
9. Section 4.2 and 4.3: Would it be more logical to switch the order of these two sections?
   
   A comparison of alternative carbon-neutral methanol production pathways first and market potential second (potentially only of the best candidate solution) seems more intuitive.
10. Figures 8, 9 and 10 and lines 360-364: If I understand the concept of learning rate correctly, you assume that the (levelized) cost of methanol decreases by 10% for each doubling in capacity. Many of the capital-intensive systems (shown in Figure 7) use existing technologies, and in particular technologies that are used in offshore windfarms and connected methanol production plants too. The cost for the same technology however will not develop significantly differently depending on whether the technology is installed onboard the energy ship or in offshore wind farms. Put differently, the cost decrease should be seen in relation to the worldwide installed capacity of the technology, not the energy ship (or fleet) alone. In that case, the costs of the energy ship would not fall as quickly as projected and the system thus not be competitive.
    
    On the other hand, it may be argued that the cost of offshore wind methanol increases with increasing installed capacity, as windfarms need to move to more distant offshore locations. The energy ship seems to be a rather robust solution to this issue, as it is relatively insensitive to shore distance and water depths.
    
    I recommend these cost projections being carefully double-checked. They do not affect the technical assessment, but have a significant effect on the market potential and hence the conclusion of this article.

Technical comments

1. Line 16: consider taking out the reference from the abstract.
2. Lines 16-17: you mention the “energy performance has been assessed”. Hence the statement “aim is to estimate the energy […] performance” seems confusing. ‘Revisit’ or ‘update based on design progression’ might clarify this.
3. Line 18: “wind-assisted propulsion experts” (without ‘s)
4. Line 30: consider replacing “low-carbon alternatives” by ‘climate-neutral’/’carbon-neutral’ or similar.
5. Line 32: ‘a sustainable fuel’ or ‘sustainable fuels’
6. Lines 38-39: consider replacing “sustainable” by ‘carbon/climate-neutral’ or similar to be more precise.
7. Line 49: Do you mean ‘levelized’ cost of energy? In that case, it can be advantageous to mention that explicitly.
8. Line 58: Figure 2 (not 3)?
9. Line 61: Consider replacing “Justifications” by ‘explanations’ or similar.
10. Table 1: Be consistent with using either H₂ or H2 and CO2 or CO₂
11. Line 71: ‘formulas’ or ‘a formula’
12. Line 230: Consider making an ordinary reference to this weblink.
13. Figure 7: an exploded pie chart (pieces grouped by CAPEX, OPEX and others) can improve the understanding of the figure.
14. Line 401: The title of this reference seems to be wrong.

# Summary

The present paper examines a novel wind-to-liquid power conversion system (energy ship) and an energy infrastructure (FARWIND) with respect to energetic and economic performance.

The FARWIND system comprises a fleet of energy ships that harvest wind energy far-shore and convert in on-board to methanol, a smaller fleet of tankers that provide feedstock and collect produce, and on-shore terminals.

Energy ships are sailing ships with water-turbines attached at the hull to provide energy to a power-to-X process. In the present paper, methanol was chosen for energy storage. Tankers firstly provide cryogenic CO2 that is used in the power-to-methanol process on-board the "energy ships" and secondly collect produced methanol which is then discharged at the on-shore terminals.

A previously developed model and preliminary design form the basis of the analysis. In the first part of this contribution, the technical model and preliminary design are revised. In the second part, an economic feasibility study for the FARWIND system is carried out.

## technical model revision

The authors present a revision to a preliminary design presented in an earlier contribution to "Wind Energy Science". The design features Flettner rotors for propulsion, a catamaran hull, and two turbines attached on either side of the hull. Revisions to the design include height of the rotors and rated power of the turbines. Model revisions include

* improved formulae to estimate aerodynamic coefficients of the rotors based on empirical data at higher (more realistic) Reynolds number,

* consideration of the effect of spin ratio on rotor driving power,

* consideration of rotor-rotor interaction,

* consideration of atmospheric boundary layer,

* revised mass-scaling of the hull, resulting in twice the mass of their preliminary design, and

* a revised turbine mass-estimate based on expert advice.

As a result, the authors report 10 to 20% less power generated than initially predicted.

## economic model

Assumptions on service-cycle length and annual production rates are made including the power predictions from the technical model. The following analysis is formed on the basis of one tanker servicing 28 energy ships per week for 4 weeks until returning to a terminal at the shore.

Tanker weight and corresponding propulsion power are estimated from service time and required tank volume.

The authors estimate an annual methanol production of approx. 70 600 t/a if continuous production is to be ensured, while factoring in production downtime due to failures and maintenance.

CAPEX for individual components including cost reduction for the entire FARWIND system due to scale effects are estimated based on literature research. Expected maintenance and operation as well as insurance costs are assumed to be proportional to capital costs. Expected ranges are taken from literature, except in the case of hull auxiliary and tanks which are arbitrarily assumed to be 2%!

To assess economic performance, levelized cost of methanol are computed under uncertainty, yielding a range of 1.2 to 3.6 Euro per kilogram, which is reported to be three times higher than usual market prices.

With respect to model assumption and uncertainty, it is found that:

1. Even at a learning rate of 10% (scale effect) the FARWIND system would not be profitable for reasonable installed capacity at current market prices for methanol.
2. If the produced methanol was used as an alternative fuel source, prices could be competitive with current gasoline prices in the European Union.
3. When benchmarked against a hypothetical power-to-methanol wind farm, the FARWIND system is may become competitive long term for large installed capacity.

# General remarks

As is revealed in figure 6, the previous assumption on required power to drive the rotors (4 x 40kW = 160kW) deviates significantly from the new model for a number of TWS and TWA combinations! Similarly, predictions for generated power reduced significantly as a result of model improvement. This hints at the fact that it might be advisable to investigate other parts of the technical model for further possibilities of improvement. Even though part one of the article is seen as an update to previous work, the discussion of the energetic performance model is kept too brief, as it leaves a few open questions. For example, power generation is surprisingly steady for different TWA and const. TWS while the peak-power stagnates with increasing TWS, which seems counter intuitive at first. I suggest that either, behaviour of the system at different TWS and TWA should be discussed in more detail, a reference to such discussion is given, or reports on model revisions should be shortened to shift the focus of the analysis.

In general, the analysis is based on many broad assumptions that undoubtably include considerable uncertainty. The notion of uncertainty is addressed by considering ranges for most parameters. There is however, no mention of distributions within the identified ranges. The expected rate of production, on the other hand, is assumed without any notion of uncertainty. It remains unclear how uncertainty is propagated through the model! It should be clarified which method of error propagation was used. For example, without the notion of distributions, it remains unanswered if, based on the assumptions, it is equally as likely to yield lower or upper LCOM as reported in figures 8 to 10. Besides, the mean with error bars would arguably more appropriate presentation in this context.

For a preliminary case study, the method of determining economic feasibility is probably sufficient. As the analysis was based on a predetermined design, the validity of the results is at the current state questionable. If the design of the "energy ship" was optimized for the specific purpose of increased profitability, the proposed system might become significantly more competitive compared to the current design, as noted in section 5.

I suggest to accept with minor revisions (see below)!

# Revisions

* Section 2: behaviour of the system at different TWS and TWA should be discussed in more detail, a or a reference to such discussion should be given, or reports on model revisions should be shortened to shift the focus of the analysis --> see general remarks

* line 71: no definition of the Reynolds number is given

* Sections 2.4 to 2.6 list assumptions for the power-to-methanol plant, tanks and auxiliary equipment: No references were given! They might be included in the first part, but this isn't stated either. References are given later in section 4.1, it's unclear however, if those are the ones considered in 2.4 to 2.6 as well.

* Figure 6: polar plots are missing units of measure for power and speed!

* line 301: Please double check the units! The market price of methanol is given as 0.4 Euro per kilogram or 72 Euro per Megawatt hour. With carbon tax it is given as 6 or 13 Euro per Megawatt hour depending on the taxation, which is about ten times lower than the price given w/o tax.

* line 401. The title of the reference seems to have changed. Consider adding DOIs to your references where possible!