Bioinspired Polymer-Bound Organocatalysts for Direct Asymmetric Aldol Reaction: Experimental and Computational Studies
Round 1
Reviewer 1 Report
1. What was outcome with D-Proline version catalyst? Provides exactly opposite selectivity (syn/anti)?
2. As per the mechanism (transition state, figure 2) or literature reports outcome of the selectivity depends on aldehyde (R Vs H), why authors think that introduction of the second linker (amino acid ) provide good selectivity and also it was proved by results in table 1 (entries 1 verses 2 and 5 verses 6), substation proline amide ( 3° amide) might be results difference outcome (selectivity).
3. It is highly recommended to include the results with simple proline amide (same out come like catalysts in currents manuscript).
4. Correct the all four structure of proline in figure2 (one line should be dotted in two doubles bonds).
5. What was the rational for electron rich aldehydes (table 4) provides poor results best results than electron rich aldehydes and authors not explained in section 3.4. More elector rich aldehydes needs to be studied to support the conclusion.
6. What was the role of the TFA? Results in the table 5 clearly indicates without any TFA (entry 2-5) improvements in selectivity and yield than with TFA. What is the rational or explain table 5.
7. References
Ref 5, no period after journal abbreviation
Ref 32, volume number is missing
Author Response
Response to reviewer 1:
1. What was outcome with D-Proline version catalyst? Provides exactly opposite selectivity (syn/anti)?
Answer: It has been agreed that enamines are the key intermediates for proline- and amine-catalyzed aldol reactions. A number of computational studies based on enamine intermediates were carried out that complement the experimental observations and interpret the resulting steroselectivities (see: J. Am. Chem. Soc. 2001, 123, 11273; J. Am. Chem. Soc. 2003, 125, 16). Our theoretical calculations are also in good agreement with the stereoselectivity observed in the experiment. As shown in Figs. 2, 3 and Table 3 in the manuscript (see below), TSa1 and TSc1 are respectively the lowest-energy transition states in the aldol reactions with L-proline version catalyst in the absence and presence of TFA, by which leading to anti-product (anti-1a). On the basis of computational results, it can be expected that D-version catalyst would not change the diastereoselectivity of the aldol reaction and the enantiomeric excess of main products. That is, anti-products are still dominant in this case. The difference in the stereochemical outcomes with L- and D-proline version catalysts is that the former gave anti-1a as the primary product while the latter would predominantly produce anti-1b, both of which are anti-adducts. This is reasonable considering that in the case of D-catalyst the transition state with the lowest energy should be structurally mirror image of TSa1 or TSc1. Also, for the “D-version catalysis” the relative energy barriers (ΔΔG) of TS’s should be the same as those in the “L-version catalysis”.
(TSa1) (TSc1)
2. As per the mechanism (transition state, figure 2) or literature reports outcome of the selectivity depends on aldehyde (R vs H), why authors think that introduction of the second linker (amino acid) provide good selectivity and also it was proved by results in table 1 (entries 1 verses 2 and 5 verses 6), substation proline amide (3° amide) might be results difference outcome (selectivity).
Answer: As demonstrated by Portnoy et al. (Chem. Eur. J. 2012, 18, 2290), the configuration of the favored diastereomer of the aldol adduct is dictated by the stereogenic center of the L-proline moiety rather than by that of the alkyl spacer connecting the catalytic unit with the polymer backbone. However, our observations indicated that the second linker indeed play the role of affecting stereoselectivity and catalytic activity. The fact that the catalyst P1 offered a higher ee values compared to P4 suggests that the aryl-substituted linker in P1 provided the catalytic site with a more beneficial environment for asymmetric induction, likely attributed to steric hindrances. Similar phenomena were also noted in previously reported organocatalytic systems (see: Chem. Commun. 2012, 48, 4011).
3. It is highly recommended to include the results with simple proline amide (same outcome like catalysts in currents manuscript).
Answer: Thank you for the helpful suggestion. We have synthesized such a simple prolinamide MC2 (see below) and conducted comparable experiments. Related results have been added in the revised manuscript. Overall, the small molecule catalyst exhibited lower catalytic efficiency compared to the polymeric catalysts under the same conditions.
(MC2)
4. Correct the all four structure of proline in figure 2 (one line should be dotted in two doubles bonds).
Answer: We've made amendments.
5. What was the rational for electron rich aldehydes (table 4) provides poor results best results than electron rich aldehydes and authors not explained in section 3.4. More elector rich aldehydes needs to be studied to support the conclusion.
Answer: Generally, an electron-deficient aldehyde is a good aldol acceptor because it acts as an electrophile in the aldolisation. Thus, the aldehydes bearing electron-withdrawing substituents are more active and provide better results when compared with the electron-rich counterparts. We have given a reasonable explanation for this in the revised edition.
6. What was the role of the TFA? Results in the table 5 clearly indicates without any TFA (entry 2-5) improvements in selectivity and yield than with TFA. What is the rational or explain table 5.
Answer: I guess this should be a misunderstanding, most likely stemming from the inappropriate notations in Table 5. Table 5 showed the results on recycling experiments. Recycling of P3 was demonstrated for five cycles in which the catalyst was readily isolated by precipitation in diethyl ether and re-used without further addition of TFA.
7. References: Ref 5, no period after journal abbreviation; Ref 32, volume number is missing.
Answer: These mistakes have been corrected.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Review on the article by Jiang Liming et al. submitted to Catalyst entitled: "Bioinspired Polymer-Bound Organocatalysts for Direct Asymmetric Aldol Reaction: Experimental and Computational Studies".
Jiang et al. report on the use of the pseudo-peptidic polymers (derivatives of 2-oxazoline) with pendant L-proline-amides as organocatalyst in asymmetric aldol condensation reactions. This experimental work investigates the catalytic activity of the polymer-bound organocatalyst, and the results are explained and supported by theoretical DFT calculations. The newly reported results are compared to the previously published by this group on similar polymer as well as to the catalytic activity of monomeric L-proline derivative as a control.
The well-designed experiments answer most of the questions.
The paper itself is well written and is scientifically sound.
Therefore I would suggest publishing it after minor revisions of the questions and comments listed below.
If the authors allow me, I have a few questions and some suggestions on the main text as well as on the SI:
The authors do not talk about the reproducibility of the results. What is the experimental error in determining catalyst efficiency?
Line 243: You talk about “The reaction cavity”, but you did not study in detail the 3D structure of the polymers with pendant L-prolines. Don't you think it would be wiser to use the notion of the “catalytic centre” “catalytic microenvironment” or something of that sort with a broader definition? Besides that, is a cyclohexanone a good or a bad solvent for the studied polymers?
Line 250: the phrase “…by a protonation-like hydrogen-bonding network in transition states” is very confusing. Could you rephrase it in a more straightforward way?
The SI may be homogenised better: in some places the triethylamine, in some places NEt3; in some places DCM and in some - CH2Cl2.
The SI, M2: what is the boiling point of M2 at 4 mbar? Should be indicated.
Figure S1. What do the arrows indicate? The caption is not very clear with respect to the figure. One could use two graphs or use the inset or somehow modify the figure to understand it better.
Figure S5 - S16. It is mentioned that the spectra are referenced relative to the TMS standard. One can see a signal at 0 ppm in proton NMR but not in 13C NMR. Could you explain this strange observation?
Figure S14: It is written: The grafting degree of L-proline = 2e/d X 100%. How was d integrated if it overlaps with c?
Some comments on text and figures:
Line 20: …the influence…transformation were? investigated.
Line 22: …changing the energy gap… maybe better: … changing the difference in energy… ?
Line 41: It looks like there is a problem with the font?
Line 50-52: The introduction of the second chiral subunit in the lateral group is to test for possible effects of the various configuration combinations on the stereoselectivity and whether the substituents at the stereogenic centre affect catalyst outcome. – Consider rephrasing to The second chiral subunit in the lateral group was introduced to test...
Scheme 1: first transformation above the arrow, ClCH2CH2NH3Cl better to write like in a Scheme S1 as (-NH2 HCl, or Scheme S2 as –NH3+ Cl-).
Scheme S1: the sign “~” does not represent the hyphen “–“, maybe it is better to use the latter one.
Line 149: … which has smaller steric hindrance relative to cyclohexanone. You might consider a simpler rephrasing as …which is smaller than cyclohexane.
Line 228: …by precipitation in ethyl ether – better diethyl ether or simply ether.
Line 239: 1H NMR (superscript).
SI: Mp (melting point) – mp?
Author Response
Response to Reviewer 2:
1. The authors do not talk about the reproducibility of the results. What is the experimental error in determining catalyst efficiency?
Answer: We only examined the reproducibility of the representative aldol addition of cyclohexanone to p-nitrobenzaldehyde mediated by P3, not all the experiments. It was found that the catalytic reactions gave reproducible yield as well as anti/syn and ee values, the experimental error being less than 1%.
2. Line 243: You talk about “The reaction cavity”, but you did not study in detail the 3D structure of the polymers with pendant L-prolines. Don't you think it would be wiser to use the notion of the “catalytic centre” “catalytic microenvironment” or something of that sort with a broader definition? Besides that, is a cyclohexanone a good or a bad solvent for the studied polymers?
Answer: thank you for the helpful suggestion. Indeed, it is very difficult to elucidate the reaction cavity for the polymer-bound catalyst. We have taken your advice.
3. Line 250: the phrase “…by a protonation-like hydrogen-bonding network in transition states” is very confusing. Could you rephrase it in a more straightforward way?
Answer: We have taken your suggestion. The aldol acceptor is activated by the hydrogen-bonding network assisted by the proton provided by TFA in transition states.
4. The SI may be homogenised better: in some places the triethylamine, in some places NEt3; in some places DCM and in some - CH2Cl2.
Answer: We have taken your suggestion.
5. The SI, M2: what is the boiling point of M2 at 4 mbar? Should be indicated.
Answer: we have provided the boiling point of M2 in the revised edition.
6. Figure S1. What do the arrows indicate? The caption is not very clear with respect to the figure. One could use two graphs or use the inset or somehow modify the figure to understand it better.
Answer: we have modified Fig S1 according to your suggestion.
7. Figure S5 - S16. It is mentioned that the spectra are referenced relative to the TMS standard. One can see a signal at 0 ppm in proton NMR but not in 13C NMR. Could you explain this strange observation?
Answer: A higher sample concentration (~40 mg polymer/0.6 ml CDCl3) together with less scan time (usually 30 min) are the most likely explanation to the absence of TMS signal in 13C NMR spectra.
8. Figure S14: It is written: The grafting degree of L-proline = 2e/d X 100%. How was d integrated if it overlaps with c?
Answer: The integration value of c is twice that of e. So, if the peak d overlaps with c, the grafting degree could be evaluated by the following equation:
GD = 2e/(d + c - 2e) ´ 100%.
Also, the integration area of d is comparable with a, and the latter doesn’t overlap with other peaks.
9. Line 20: …the influence…transformation were? investigated.
Answer: the error has been corrected.
10. Line 22: …changing the energy gap… maybe better: … changing the difference in energy… ?
Answer: we have taken your suggestion.
11. It looks like there is a problem with the font?
Answer: it has been corrected.
12. Line 50-52: The introduction of the second chiral subunit in the lateral group is to test for possible effects of the various configuration combinations on the stereoselectivity and whether the substituents at the stereogenic centre affect catalyst outcome. – Consider rephrasing to The second chiral subunit in the lateral group was introduced to test...
Answer: the sentence has been modified.
13. Scheme 1: first transformation above the arrow, ClCH2CH2NH3Cl better to write like in a Scheme S1 as (-NH2 HCl, or Scheme S2 as –NH3+ Cl-).
Answer: Scheme 1 has been modified.
14. Scheme S1: the sign “~” does not represent the hyphen “–“, maybe it is better to use the latter one.
Answer: The corresponding modifications have been made.
15. Line 149: … which has smaller steric hindrance relative to cyclohexanone. You might consider a simpler rephrasing as …which is smaller than cyclohexane.
Answer: we have modified the sentence according to your suggestion.
16. Line 228: …by precipitation in ethyl ether – better diethyl ether or simply ether.
Answer: it has been corrected.
17. Line 239: 1H NMR (superscript).
Answer: it has been corrected.
18. SI: Mp (melting point) – mp?
Answer: Yes, it has been corrected.
Author Response File: Author Response.pdf
Reviewer 3 Report
The manuscript by Du et al. describes the synthesis of a series of materials based on poly(2-oxazoline) connected to L-proline and their activity in the aldol addition reaction of cyclic ketones with different aromatic aldehydes. Besides the catalytic activity, the authors tried to explain the mechanism using DFT computational studies. Although the scientific idea presented in this manuscript is interesting, the manuscript requires major improvements and some additional experiments and products characterization.
My comments are the following:
1. Page 1, paragraph 1, lines 34-35: “Among them, synthetic polypeptides are interesting candidates for asymmetric syntheses because they are considered to be closely related to enzymatic systems.” This is a general statement and authors should state more clearly why the resembling of polypeptides to enzymatic systems is important in asymmetric catalysis .
2. Page 3, section 3.2 lines 111-113: “As was the case for those reported earlier [19], catalytic reactions were undertaken in neat ketones as the reaction medium in the presence of trace amounts of water and TFA.” This sentence has no meaning. Please reformulate.
3. The authors mentioned that the catalytic reaction is run in “neat conditions”, but they add water, even if it is in trace amounts. By definition, neat means without any presence of water. The authors should use the expression “without solvent”, as it is more appropriate for the present case.
4. What is the role of water in this reaction? This should be clearly mentioned in the manuscript.
5. Page 3, section 3.2 lines 113: “At a loading of 10 mol%,” with respect to which is the 10 mol% loading? This should be clearly mentioned in the manuscript.
6. Page 3, section 3.2 lines 116: “dr up to 1: 5.25 “. This is a single-time notation in the manuscript. The authors denote the diastereoselectivity excess in percent. Please be consistent in the notations used in the manuscript.
7. Table 1 at page 4 is very misleading. First of all, the reaction scheme found at the top of the table relates to the results presented in the entries 1-10, but the results in the entries 11 -12 are related to a totally different reaction: the acetone reaction with 4-nitrobelzaldehyde. The authors should create a new table for the last two results. Secondly, the authors should draw all 4 products: each diastereomer has 2 enantiomers (syn-R,R; syn-S,S and anti-2R,1’S; anti-2S,1’R).
8. When discussing the effect of different materials, the authors compared the catalytic results of reactions carried out at different reaction times. This is not acceptable. If a catalyst requires a longer reaction time to achieve a certain conversion, this should be clearly stated. For example, the authors compared the activity of material (S)-P1 with P3 (Table 1, entries 2 and 4), where the first reaction was carried for 48 h while the second one for 12 h.
9. The authors should test the activity of pure L-proline in the same reaction conditions as material P3.
10. Page 4, lines 125 - 127: “That is, the hydrogen-bond-accepting character of tertiary amides may exert synergistic effects on stabilizing the formed enamine intermediate in the aldol reaction, which in turn improved the catalytic efficiency of prolinamide.” The authors should explain what they mean through the “synergistic effects”.
11. The authors should state the exact quantities they used for each reaction they carried out. The phrase found at page 4, lines 133-134: “Reactions were performed on 0.27 mmol scale using cyclohexanone (0.4 mL, 14 equiv) in the presence of water (20 equiv relative to the catalyst loading) and TFA (0.8 equiv relative to the catalyst loading) at 20 °C for 12 h.” It is ambiguous and cannot be reproduced. The same for the other tables and for the supporting information.
12. Page 4, lines 140-142: “The overall performance of P3 is comparable to L-prolyl dipeptide catalysts reported in the literature [20,29,30], even superior to some silica- [31] and PS- supported [32] analogs in terms of yield, diastereoselectivity, and enantioselectivity.” The authors should also add the exact values, which would make the comparison much easier.
13. Page 5, lines 158-167: Why is the yield higher when 0.2 equiv. were used compared to the 0.4 equiv.?
14. Page 5, table 2: the authors should add the reaction scheme.
15. Page 6, lines 195-196: “Thus, it is theoretically expected that both catalysts will yield the dominant enantiomer with a (2S,3R)-configuration, which is in good agreement with experimental observations.” There is no experimental proof that the (2S,3R)-isomer was obtained.
16. Page 7, lines 220-221: “These results imply that the addition of an acidic additive in the aldol reactions catalyzed by either P3 or P5 should be favorable for the formation of (2S,3R)-adduct with higher enantioselectivity, as observed experimentally.” Again, there is no experimental proof that the (2S,3R)-isomer was obtained.
17. The authors did not explain why no more TFA was added in the recycling reactions.
18. There are several catalytic systems that are more active than the one presented in this manuscript (e.g. they require less reaction time). What is the advantage of this new system compared to others?
19. There are several English language mistakes, such as:
- Page 3, line 115: the expression “was less impressive in activity” is not a scientific expression. It can be changed to “far less active”
- Page 5, lines 176-177: “such computational studies have proven to be an important means to provide” should be “such computational studies have proven to be important means to provide”
- Page 5, line 177: “To achieve this end” would be “to achieve this aim”
Author Response
Reviewer 3:
1. Page 1, paragraph 1, lines 34-35: “Among them, synthetic polypeptides are interesting candidates for asymmetric syntheses because they are considered to be closely related to enzymatic systems.” This is a general statement and authors should state more clearly why the resembling of polypeptides to enzymatic systems is important in asymmetric catalysis.
Answer: we have made changes (Line 33-35).
2. Page 3, section 3.2 lines 111-113: “As was the case for those reported earlier [19], catalytic reactions were undertaken in neat ketones as the reaction medium in the presence of trace amounts of water and TFA.” This sentence has no meaning. Please reformulate.
Answer: We have taken your suggestion.
3. The authors mentioned that the catalytic reaction is run in “neat conditions”, but they add water, even if it is in trace amounts. By definition, neat means without any presence of water. The authors should use the expression “without solvent”, as it is more appropriate for the present case.
Answer: we have corrected related sentences according to the suggestion
4. What is the role of water in this reaction? This should be clearly mentioned in the manuscript.
Answer: In a preliminary survey on the water influence we found that the addition of some water has a beneficial to the enhancement of yield in the aldol reactions. It is generally to be understood that the beneficial effect of water is due to significant promotion of the rate-determining proton transfer step since the iminium–enamine catalytic cycle is intrinsically a multiproton transfer process.
5. Page 3, section 3.2 lines 113: “At a loading of 10 mol%,” with respect to which is the 10 mol% loading? This should be clearly mentioned in the manuscript.
Answer: “At a loading of 10 mol% with respect to 4-nitrobenzaldehyde”.
6. Page 3, section 3.2 lines 116: “dr up to 1: 5.25 “. This is a single-time notation in the manuscript. The authors denote the diastereoselectivity excess in percent. Please be consistent in the notations used in the manuscript.
Answer: We've corrected it.
7. Table 1 at page 4 is very misleading. First of all, the reaction scheme found at the top of the table relates to the results presented in the entries 1-10, but the results in the entries 11 -12 are related to a totally different reaction: the acetone reaction with 4-nitrobelzaldehyde. The authors should create a new table for the last two results. Secondly, the authors should draw all 4 products: each diastereomer has 2 enantiomers (syn-R,R; syn-S,S and anti-2R,1’S; anti-2S,1’R).
Answer: We've revised it.
8. When discussing the effect of different materials, the authors compared the catalytic results of reactions carried out at different reaction times. This is not acceptable. If a catalyst requires a longer reaction time to achieve a certain conversion, this should be clearly stated. For example, the authors compared the activity of material (S)-P1 with P3 (Table 1, entries 2 and 4), where the first reaction was carried for 48 h while the second one for 12 h.
Answer: In the P3-catalyzed reaction, the reactant (p-nitrobenzaldehyde) consumption was close to 90% in 12 h while for the P1 catalysis it consumed ~50% in 24 h. So, the former should be superior to the latter in terms of catalytic activity.
9. The authors should test the activity of pure L-proline in the same reaction conditions as material P3.
Answer: We think the two don't seem to be comparable. L-Proline possesses the carboxylic acid group, which is a key contributing structural unit for activation of aldehyde acceptor, while the P3 is a prolinamide-type catalyst, relying on weak hydrogen bonding from the amide NH to activate the aldehyde partner in the transition state.
10. Page 4, lines 125 - 127: “That is, the hydrogen-bond-accepting character of tertiary amides may exert synergistic effects on stabilizing the formed enamine intermediate in the aldol reaction, which in turn improved the catalytic efficiency of prolinamide.” The authors should explain what they mean through the “synergistic effects”.
Answer: We gave a more detailed discussion to make the reader understand the implications of the “synergistic effects”.
11. The authors should state the exact quantities they used for each reaction they carried out. The phrase found at page 4, lines 133-134: “Reactions were performed on 0.27 mmol scale using cyclohexanone (0.4 mL, 14 equiv) in the presence of water (20 equiv relative to the catalyst loading) and TFA (0.8 equiv relative to the catalyst loading) at 20 °C for 12 h.” It is ambiguous and cannot be reproduced. The same for the other tables and for the supporting information.
Answer: We've revised it according to this suggestion.
12. Page 4, lines 140-142: “The overall performance of P3 is comparable to L-prolyl dipeptide catalysts reported in the literature [20,29,30], even superior to some silica- [31] and PS- supported [32] analogs in terms of yield, diastereoselectivity, and enantioselectivity.” The authors should also add the exact values, which would make the comparison much easier.
Answer: We've revised it according to this suggestion.
13. Page 5, lines 158-167: Why is the yield higher when 0.2 equiv. were used compared to the 0.4 equiv.?
Answer: It’s a pen error, and has been corrected.
14. Page 5, table 2: the authors should add the reaction scheme.
Answer: We've revised it.
15. Page 6, lines 195-196: “Thus, it is theoretically expected that both catalysts will yield the dominant enantiomer with a (2S,3R)-configuration, which is in good agreement with experimental observations.” There is no experimental proof that the (2S,3R)-isomer was obtained.
Answer: The experimental proof was provided in SI (Fig S18-S21). The relative stereochemistry of the aldol products was determined by chiral-phase HPLC analysis and comparison with the literature, which has been mentioned in the revised edition.
16. Page 7, lines 220-221: “These results imply that the addition of an acidic additive in the aldol reactions catalyzed by either P3 or P5 should be favorable for the formation of (2S,3R)-adduct with higher enantioselectivity, as observed experimentally.” Again, there is no experimental proof that the (2S,3R)-isomer was obtained.
Answer: same as the previous one.
17. The authors did not explain why no more TFA was added in the recycling reactions.
Answer: I guess this should be a misunderstanding, most likely stemming from the inappropriate notations in Table 5. Table 5 showed the results on recycling experiments. Recycling of P3 was demonstrated for five cycles in which the catalyst was readily isolated by precipitation in diethyl ether and re-used without further addition of TFA.
18. There are several catalytic systems that are more active than the one presented in this manuscript (e.g. they require less reaction time). What is the advantage of this new system compared to others?
Answer: Poly(2-oxazoline)s are regarded as pseudopeptides polymers due to their structural relation to polypeptides, which may hold promise as an ideal scaffold to construct effective artificial enzymes. As mentioned in the Introduction, our main aim was to employ the aldol reaction as a benchmark in order to explore the structure–property relationships, especially to understand the role of the tertiary amide backbone in the catalytic sites.
19. There are several English language mistakes, such as:
- Page 3, line 115: the expression “was less impressive in activity” is not a scientific expression. It can be changed to “far less active”
- Page 5, lines 176-177: “such computational studies have proven to be an important means to provide” should be “such computational studies have proven to be important means to provide”
- Page 5, line 177: “To achieve this end” would be “to achieve this aim”
Answer: The corresponding modifications have been made.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
Accepted
Reviewer 3 Report
The authors made the required modifications. The present form is much better than the initial one.
