Volume 14, Issue 14 e202200405
Research Article
Open Access

Insights into Substituent Effects of Benzaldehyde Derivatives in a Heterogeneous Organocatalyzed Aldol Reaction

Dr. Graziano Di Carmine

Dr. Graziano Di Carmine

Department of Chemical Engineering, The University of Manchester, Oxford Road, Manchester, M13 9PL UK

Department of Chemical, Pharmaceutical and Agricultural Sciences, University of Ferrara, Via L. Borsari, 46, 44121 Ferrara, Italy

These authors contributed equally to this manuscript

Search for more papers by this author
Dr. Fabio Pesciaioli

Dr. Fabio Pesciaioli

Department of Physical and Chemical Sciences, Università degli Studi dell'Aquila, via Vetoio, 67100 L'Aquila, Italy

These authors contributed equally to this manuscript

Search for more papers by this author
Dr. Simeng Wang

Dr. Simeng Wang

Department of Chemical Engineering, The University of Manchester, Oxford Road, Manchester, M13 9PL UK

Search for more papers by this author
Dr. Arianna Sinibaldi

Dr. Arianna Sinibaldi

Department of Physical and Chemical Sciences, Università degli Studi dell'Aquila, via Vetoio, 67100 L'Aquila, Italy

Search for more papers by this author
Giuliana Giorgianni

Giuliana Giorgianni

Department of Physical and Chemical Sciences, Università degli Studi dell'Aquila, via Vetoio, 67100 L'Aquila, Italy

Search for more papers by this author
Dr. Christopher M. A. Parlett

Dr. Christopher M. A. Parlett

Department of Chemical Engineering, The University of Manchester, Oxford Road, Manchester, M13 9PL UK

Diamond Light Source Harwell Science and Innovation Campus, Didcot, Oxfordshire, OX11 0DE UK

UK Catalysis Hub, Research Complex at Harwell Rutherford Appleton Laboratory, Harwell, Oxfordshire, OX11 0FA UK

University of Manchester at Harwell, Diamond Light Source Harwell Science and Innovation Campus, Didcot, Oxfordshire, OX11 0DE UK

Search for more papers by this author
Prof. Dr. Armando Carlone

Corresponding Author

Prof. Dr. Armando Carlone

Department of Physical and Chemical Sciences, Università degli Studi dell'Aquila, via Vetoio, 67100 L'Aquila, Italy

Search for more papers by this author
Dr. Carmine D'Agostino

Corresponding Author

Dr. Carmine D'Agostino

Department of Chemical Engineering, The University of Manchester, Oxford Road, Manchester, M13 9PL UK

Search for more papers by this author
First published: 22 April 2022
Citations: 4

Correction added on 25 May 2022, after first online publication: spelling mistake corrected in the title.

Graphical Abstract

Adsorption: Understanding factors that affect and govern heterogenous organocatalysis is undoubtedly critical to develop new and higher performing catalysts. NMR relaxation techniques provide very useful insights into adsorption effects of the aldehyde partner in an organocatalyzed aldol reaction.

Abstract

Organocatalyst immobilization onto solid supports represents a promising method for enabling asymmetric organocatalysis while retaining the advantages of heterogeneous catalysts, including catalyst separation, recycling, and the use of fixed-bed reactors. Understanding how such heterogenized catalytic systems work is fundamental to develop and tailor more efficient ones. Herein, we have elucidated the role of reactant molecular structure on surface interactions and reactivity for asymmetric aldol reactions between benzaldehyde derivatives and hydroxyacetone catalyzed by SBA-15 immobilized L-proline. NMR relaxation time analysis reveals that a stronger interaction between the aldehyde and the catalyst surface reduces catalytic reactivity, which is attributed to reduced access of hydroxyacetone to the L-proline surface sites, hence inhibiting the formation of the enamine intermediate between hydroxyacetone and L-proline. The results show that surface phenomena in these systems are important considerations for reactant selection, opening up new directions to explore in this area of research.

Introduction

Asymmetric organocatalysis1-6 is undoubtedly a pivotal technological platform that is routinely employed by industry.7, 8 Despite the great advancements and plethora of new reactions and activations discovered, the applications of organocatalysis have sometimes been hampered by its low turnover number (TON), the need for high catalyst loading, and tedious separation processes to isolate the product.9, 10 Scientists have devised ingenious means to address this low productivity: recycle and reuse, development of highly active catalysts,11 or combining different types of catalysts.12, 13 Another successful approach is represented by immobilization of organocatalysts over solid supports.14-19 The immobilization of proline, one of the privileged organocatalysts, has been frequently investigated.20, 21 However, unlike homogeneous organocatalysis, the reaction catalyzed by heterogeneous catalysts is heavily influenced by surface interactions between the support and the reaction species. Insights into surface interactions may help in improving and developing more efficient heterogenous organocatalysts. As such, a technique that can offer a direct means to observe surface interactions in a heterogeneous environment is undoubtedly useful and informative. In recent years, the application of NMR spectroscopy,22-24 diffusion25-28 and relaxation29-32 techniques to heterogeneous catalysts has uncovered novel insights, highlighting the important role that transport, adsorption and surface dynamics of reaction species confined within the pore space play in determining catalytic performances.

Surface interactions and surface dynamics of reaction species are also believed to play an important role in the case of organocatalysts immobilized on solid supports. In fact, in many cases, organocatalysts immobilized on porous oxides have much lower catalytic performances when compared to their homogeneous counterpart operating in solution.33, 34 Interestingly, if polymers are used as support materials, the immobilized organocatalysts tend to perform better than the ones immobilized on silica. In some cases, their activity is comparable to that of their homogeneous counterpart.35 Unlike inorganic porous oxides, which tend to have surfaces intrinsically functionalized with hydroxyl groups and potentially other reactive groups such as Lewis and Brønsted acidic sites, polymers are relatively inert. For example, in one of our recent works, we have observed that, by optimizing the choice of the solvent, N-heterocyclic carbene (NHC) organocatalysts immobilized over polystyrene particles show a catalytic activity that is comparable to the catalytic activity measured in homogeneous conditions.35

These observations suggest that the solid support plays a key role and its effect on modulating interactions between organocatalysts, reaction species and solvents may lead to significant changes in reactivity. Investigating these interactions is, therefore, essential to rationalize catalytic behavior.

Whilst most of the efforts in this area have been devoted to the study of solvents, the role of surface interactions of reactant species in organocatalytic reactions occurring over solid surfaces remained largely unexplored. This is particularly the case for the reaction studied in this work, whereby we investigate an aldol condensation between hydroxyacetone and a series of benzaldehyde derivatives, in the presence of L-proline supported on SBA-15. In particular, according to a generally accepted reaction mechanism, the formation of an initial complex between the aldehyde partner and the organocatalyst is suggested to affect the catalytic cycle by preventing the formation of the hydroxyacetone/L-proline adduct.36 Competitive adsorption between reaction partners becomes therefore an important parameter to consider, yet, to the best of our knowledge, no such data have been reported for this type of catalytic systems.

In this work, NMR spin relaxation is used for the first time as a means to investigate competitive adsorption of reaction partners in such catalytic systems. Results on the catalytic performance of different aldehydes are rationalized via a critical comparison with NMR relaxation time analysis, and within the context of the accepted reaction mechanism of the reaction, to provide insights between the nature of the aldehyde and catalytic behavior of the supported organocatalyst under investigation.

Results and Discussion

Building on our previous work on the effect of solvents in heterogenized proline mediated aldol reaction,33 hydroxyacetone 1 was reacted with electronically diversified benzaldehydes 2 ae to investigate the role of the acceptor. Proline supported on SBA-15 mesoporous silica 4 was selected as the immobilized organocatalyst for investigation; 4 was prepared according to a previously reported procedure.33 All reactions were carried out with L-proline (conditions A) and 4 (conditions B) and compared in terms of turn over frequency (TOF).

A summary of the results is presented in Table 1. Lower TOFs were always measured when the immobilized organocatalyst 4 (heterogeneous conditions) was used, as opposed to L-proline (homogeneous conditions). A similar diastereomeric ratio (dr) is observed in all entries with a relatively small variation. Conversely, enantioselectivity is slightly lower with 4; this could be attributed to additional non-covalent interactions between silica functional groups and species involved in the stereoselective determining step. We note that the solvent is likely to impact not only the TOF but also the enantioselectivity. However, such effects would be much more complex to unravel purely based on competitive adsorption effects as they would involve other aspects, such as adsorption geometry and changes in transition states. Though other approaches, including computational DFT, could be useful for investigating this behavior, these are out of the scope of this work.

Table 1. Organocatalyzed aldol reaction of 1 with different aldehydes 2 ae.[a]

R

Conditions

TOF [h−1]×103[b]

dr[c]

ee [%][d]

Ph (2 a)

A

60.7

74 : 26

88 (42)

B

11.3

79 : 21

78 (29)

2-F-C6H4 (2 b)

A

93.3

77 : 23

86 (23)

B

18.9

72 : 28

80 (28)

3-Br-C6H4 (2 c)

A

104.0

76 : 24

68 (41)

B

15.9

73 : 27

73 (54)

4-MeO-C6H4 (2 d)

A

60.6

73 : 27

88 (48)

B

10.9

71 : 29

76 (31)

4-Me-C6H4 (2 e)

A

53.9

73 : 27

90 (43)

B

6.2

72 : 28

80 (32)

  • [a] Reaction conditions A: hydroxyacetone 1 (2.4 mmol, 24 equiv.), aldehydes 2 ae (0.1 mmol, 1 equiv.), L-proline (30 mol %), DMSO (0.8 mL), 25 °C, 24 h; Reaction conditions B: hydroxyacetone 1 (2.4 mmol, 24  equiv.), aldehydes 2 ae (0.1 mmol, 1 equiv.), 4 (41 mol%), DMSO (0.8 mL), 25 °C, 5 days. [b] TOF calculated from the averaged rate over the course of the reaction (24 h for the homogeneous catalyst and 120 h for the immobilized catalyst) as mmol-product/(mmol-cat×time in hours). [c] dr determined by 1H NMR analysis of the crude reaction mixture. [d] ee determined by HPLC on a chiral stationary phase; ee values of minor diastereoisomer are reported in brackets. See ESI for further details.

image

NMR relaxation measurements were performed to characterize surface interactions between benzaldehyde derivatives and the surface of immobilized organocatalyst 4. A typical data set of T1 and T2 data is shown in Figure 1, which shows the evolution of the NMR spectra for T1 inversion recovery (Figure 1a) and T2 Carr-Purcell-Meiboom-Gill (CPMG) (Figure 1b) experiments for 3-Br-benzaldehyde 2 c adsorbed within 4.

Details are in the caption following the image

(a) T1 NMR signal and (b) T2 NMR signal as a function of time delay steps of 3Br-benzaldehyde 2 c adsorbed in the immobilized organocatalyst 4.

Figure 2 shows the experimental data obtained from T1 inversion recovery (Figure 2a) and T2 CPMG (Figure 2b) measurements for the various benzaldehyde derivatives adsorbed within the catalyst pore space. The data set is obtained by integrating the area under each NMR peak (cfr Figure 1) as a function of the time delay associated with each acquisition step. The T1 data were fitted using Equation (S1) and the T2 data were fitted using Equation (S2) (see ESI). The agreement between the experimental data and fitting are excellent in all cases.

Details are in the caption following the image

(a) T1 inversion recovery and (b) T2 CPMG experimental data for the various benzaldehyde derivatives adsorbed within the SBA-15-immobilized catalyst 4. Solid lines are fitting to (a) Equation (S1) and (b) Equation (S2).

The obtained values of the T1 and T2 time constants are reported in Table 2, together with the values of the T1/T2 ratio. As expected for liquids adsorbed inside the porous catalysts, the T1/T2 ratio for all cases is greater than 1. This ratio is a robust indicator to characterize surface interactions of liquids adsorbed over surfaces, including organic liquids adsorbed over silica, metal oxides and supported metal nanoparticles.37 This has been confirmed by experimentally (temperature-programmed desorption) and computationally (Density-Functional Theory).38

Table 2. . Values of T1 and T2 NMR relaxation time constants and their ratio T1/T2 for different benzaldehyde derivatives adsorbed in the solid organocatalyst 4.

Reactant

T1 [ms]

T2 [ms]

T1/T2 [-]

2 a

51±1

13.4±0.3

3.81±0.09

2 b

36±1

11.0±0.2

3.27±0.08

2 c

87±2

24.5±0.5

3.55±0.09

2 d

85±2

21.3±0.4

3.99±0.10

2 e

111±2

25.5±0.5

4.35±0.10

Taking benzaldehyde 2 a as a reference, the results in Table 2 suggest that halogen-substituted benzaldehyde species, such as 2-F-benzaldehyde 2 b and 3-Br-benzaldehyde 2 c have a lower T1/T2, i. e., a lower affinity with the solid surface of the heterogenized catalysts, whereas 4-MeO-benzaldehyde 2 d and 4-Me-benzaldehyde 2 e have higher T1/T2 values, hence higher affinity. This may suggest that the electronic properties of the substituents are likely to affect the adsorption properties of the benzaldehyde derivatives, and that electron-donating groups (such as in 2 d and 2 e) tend to increase substrate/surface affinity.

For further insights, it is interesting to correlate catalyst activity and adsorption properties as probed by the NMR T1/T2 ratio. Figure 3 shows a plot of the TOF of the heterogeneously catalyzed reaction against the T1/T2 ratio. The plot shows that the higher the affinity of the benzaldehyde derivatives 2 ae with the catalyst surface (i. e., higher T1/T2), the lower the TOF recorded for the reaction. To rationalize this correlation, the reaction mechanism, including possible side reactions need to be taken into account. Kinetics and computational studies, that were performed to elucidate both the additive effect of water and the rate-determining step of the reaction, are of paramount importance in this regard.39 Experimental evidences suggest that interactions between proline and aldehyde could be detrimental for the reaction. In fact, according to Blackmond et al.,36 the kinetics of the reaction is driven by the addition of the enamine adduct to the aldehyde and small amount of water was reported to be beneficial for the reaction by hydrolyzing the iminium, generated by condensation of aldehyde with proline.

Details are in the caption following the image

TOF vs T1/T2 for benzaldehyde derivatives 2 ae adsorbed within 4. The vertical red line indicates the T1/T2 for hydroxyacetone 1.

The dependence of reaction rate on the aldehyde's electronic properties is well-known and has been investigated via Hammett plot experiments by Benaglia et al.40 From proline catalyzed aldol reaction between cyclohexanone and benzaldehyde derivatives, they observed a positive correlation between the rate of reaction and electrophilicity (Hammett electrophilic constants) of benzaldehyde, i. e. electron-poor benzaldehyde derivatives with withdrawing substituents exhibit an increased reactivity. The undesired formation of the iminium adduct, by condensation of benzaldehydes with proline, has been demonstrated by Blackmond36 and Pihko,41 with the addition of small amounts of water suppressing this undesirable process. It is reasonable to assume that a high affinity between the aldehyde and the surface of the solid matrix can facilitate the interaction with the proline moiety anchored over the surface, leading to a deactivation of the active site through the formation of a proline-aldehyde iminium adduct, which in turn inhibits the formation of the hydroxyacetone-proline adduct and leads to a decrease in reactivity.

Indeed, a lower TOF is measured for T1/T2 of 2 ae closer to T1/T2 of hydroxyacetone 1 (vertical red line), and this can be correlated to higher substrate inhibition resulting in a decreased catalyst activity (Figure 3). Whilst previous studies using NMR relaxation in heterogeneous catalysis42, 43 and heterogenized organocatalysis33, 35 have focused exclusively on solvent/substrate interactions, by performing a systematic screening of solvents using the same reaction system, here we show that competitive adsorption between substrates bearing substituents with different electronic properties is also a critical parameter for consideration.

It is important to clarify that adsorption effects on solid surfaces are not the only effects to consider when explaining reactivity in such systems. Stereo-electronic effects also play an important role in reactivity, as indicated by Linear Free Energy Relationship40 experiments, such as Hammett plots, that provided a quantitative relationship between reactivity and structure, reported for proline-catalyzed aldol reactions. Electrophiles bearing electron-withdrawing group tend to exhibit higher reactivity. However, from Table 1 it is possible to observe some deviation from the expected behavior based on Hammett parameters. For example, 3-Br substituted benzaldehyde 2 c shows higher reactivity than 2-F analogue 2 b in homogeneous conditions whereas the trend is inverted in the heterogeneous ones. Furthermore, the reactivity observed for 4-MeO-and 4-Me- benzaldehyde derivatives (2 d and 2 e, respectively) in homogeneous conditions is comparable, whereas the TOF, in heterogeneous conditions is almost halved for 2 d when compared to 2 e. These results show that the electrophilicity of the acceptor (i. e. the aldehyde) influenced by the substituent does not fully explain the reactivity trend in heterogeneous conditions; reversible deactivation, due to electrophile adsorption, plays an important role, in line with previous observations reported in the literature, whereby the aldehyde acceptor-proline interaction is the main cause of the decreased reactivity.39, 41 Whilst these effects might be challenging to explore and decouple, it is possible to make some qualitative considerations by normalizing the TOF of the heterogeneous reaction over the one of the homogeneous reaction. When such a normalized ratio is considered against T1/T2 (see ESI, Figure S1), it is possible to observe that a qualitative trend still exists, which suggests that, indeed, adsorption effects of benzaldehyde substrates are an important factor to consider in the reaction system investigated in this work.

Conclusions

Asymmetric aldol reactions between benzaldehyde derivatives 2 ae and hydroxyacetone 1 catalyzed by L-proline immobilized over SBA-15 silica 4 have been studied by combining catalytic reaction screening in homogeneous (L-proline) and heterogeneous (immobilized L-proline) conditions with NMR relaxation studies. NMR studies were instrumental in characterizing the strength of surface interactions between the various benzaldehyde derivatives and the catalyst surface. The results show that adsorption effects are an important consideration that can affect reactivity within the heterogenized proline catalyst 4. In particular, spectroscopic measurements of NMR relaxation times, T1 and T2 allowed the characterization of the affinity between the catalyst surface and aldehyde derivatives; the results indicate that aldehydes with higher affinity for the catalyst surface tend to limit the catalyst activity, leading to lower TOFs. This may be attributed to the inhibition caused by the presence of off-cycle species due to the formation of the iminium ion intermediate between benzaldehyde derivatives and L-proline.

It is worth noting that the current study has been carried out on proline immobilized SBA.–15 and an extrapolation to other supports19 is hardly reliable given the differences in surface properties of different supports. However, this work proves that surface phenomena in immobilized organocatalysts are important considerations for the selection of substrates, hence rationalizing these effects can help in the optimization of reactions.

Experimental Section

Organocatalyzed reactions: To a mixture of L-proline (30 mol%, 3.5 mg) or 4 (41 mol%, 78 mg), hydroxyacetone 1 (24 eq., 2.4 mmol, 165 μL) and DMSO (800 μL) aldehyde 2 ae (1 eq., 0.1 mmol) was added. After the reaction mixture was stirred at room temperature for 24 hours (with L-proline) or 5 days (with 4). The reaction was quenched with 5 mL of NH4Cl 0.5 M. The aqueous mixture was extracted with DCM (4×5 mL) and the combined organic layers were washed with brine and dried over MgSO4 and concentrated under vacuum after filtration. The crude reaction was dissolved in 1 mL of a solution of internal standard (durene) in CDCl3 (stock solution containing 0.5 eq. of durene with respect to the aldehyde). The products have not been purified. NMR yield was determined by 1H-NMR of the crude reaction mixture using durene as internal standard (NMR D1 parameter was set to 10 seconds). NMR experiments: Samples for NMR experiments were prepared by soaking the catalyst particles in the liquid reagent under investigation for 24 hours. The catalyst particles were then gently dried on a pre-soaked filter paper in order to remove any excess liquid on the outer surface of the particles and then transferred to 5 mm NMR tubes. The sample was finally placed into the magnet and left for approximately 20 minutes to thermally equilibrate before the start of the measurements. All measurements were carried out at atmospheric pressure and 26 °C±0.5 °C. NMR relaxation measurements were performed in a Magritek SpinSolve benchtop NMR spectrometer operating at a 1H frequency of 43 MHz. T1 relaxation measurements were performed using an inversion recovery pulse sequence, acquiring 16 experimental points for each experiment with time delay values between 1 ms and 1000 ms, 16–64 scans and a repetition time of 1000 ms. T2 relaxation measurements were performed with the CPMG (Carr Purcell Meiboom Gill) pulse sequence, with an echo time of 120 μs, using 16 steps with a number of echoes per step varying in the range 40–70, 64–128 scans and a repetition time of 1000 ms.

Licence Statement

For the purpose of open access, the author has applied a Creative Commons Attribution (CC BY) licence (where permitted by UKRI, ‘Open Government Licence’ or ‘Creative Commons Attribution No-derivatives (CC BY-ND) licence may be stated instead) to any Author Accepted Manuscript version arising.

Acknowledgements

C.D.A., S.W. and G.D.C. would like to acknowledge the EPSRC, grant no. EP/S019138/1, for funding the research activities carried out for this work. G.G. is grateful to PON-DOT13OV2OC for an industrial PhD fellowship. F.P. thanks PON–AIM grant number 1842894 for funding this research. UK Catalysis Hub is kindly thanked for resources and support provided via our membership of the UK Catalysis Hub Consortium and funded by EPSRC grant: EP/R027129/1 (C.M.A.P.)

    Conflict of interest

    The authors declare no conflict of interest.

    Data Availability Statement

    All data supporting this study are provided in full in the ‘Results and Discussion’ section of this paper as well as in the associated supporting information.