Skip to main content
NCBI home page
ACS AuthorChoice logoLink to ACS AuthorChoice
. 2024 Mar 29;14(8):5550–5559. doi: 10.1021/acscatal.4c00985

General, Modular Access toward Immobilized Chiral Phosphoric Acid Catalysts and Their Application in Flow Chemistry

Michael Laue †, Maximilian Schneider †, Markus Gebauer ‡, Winfried Böhlmann §, Roger Gläser ‡, Christoph Schneider †,*
PMCID: PMC11036403  PMID: 38660609

Abstract

graphic file with name cs4c00985_0007.jpg

Chiral phosphoric acids (CPAs) are among the most frequently used organocatalysts, with an ever-increasing number of applications. However, these catalysts are only obtained in a multistep synthesis and are poorly recyclable, which significantly deteriorates their environmental and economic performance. We herein report a conceptually different, general strategy for the direct immobilization of CPAs on a broad scope of solid supports including silica, polystyrene, and aluminum oxide. Solid-state catalysts were obtained in high yields and thoroughly characterized with elemental analysis by inductively coupled plasma-optical emission spectrometry (ICP-OES), nitrogen sorption measurements, thermogravimetric analysis, scanning transmission electron microscopy/energy-dispersive X-ray spectroscopy (STEM/EDX) images, and solid-state NMR spectroscopy. Further, the immobilized catalysts were applied to a variety of synthetically valuable, highly stereoselective transformations under batch and flow conditions including transfer hydrogenations, a Friedländer condensation/transfer hydrogenation sequence, and Mannich reactions under cryogenic flow conditions. Generally, high yields and stereoselectivities were observed along with robust catalyst stability and reusability. After being used for 10 runs under batch conditions, no loss of selectivity or catalytic activity was observed. Under continuous-flow conditions, the heterogeneous system was in operation for 19 h and the high enantioselectivity remained unchanged throughout the entire process. We expect our approach to extend the applicability of CPAs to a higher level, with a focus on flow chemistry and a more environmentally friendly and resource-efficient use of these powerful catalysts.

Keywords: organocatalysis, heterogeneous catalysis, immobilization, chiral phosphoric acids, stereoselective synthesis, continuous flow chemistry

Introduction

Pioneered by independent studies of Akiyama1 and Terada2 in 2004, BINOL-derived chiral phosphoric acids (CPAs) established themselves as privileged Brønsted acid organocatalysts. Due to their high efficiency and versatility, a plethora of highly selective asymmetric transformations, e.g., Friedel–Crafts,3−8 Mannich,9−11 Diels–Alder,12−14 and Strecker15−17 reactions, was developed over the past two decades. Inherently, such catalysts combine a Brønsted acidic P–OH functionality and a Lewis basic P=O site. By exploiting the resulting cooperative activation mode, a large variety of organic reactions can be promoted under typically mild reaction conditions.18−20 However, CPAs are only obtained in complex, multistep syntheses, and in general, relatively high catalyst loadings are required. Combined with their poor recovery, this results in high costs and poor environmental performance, severely limiting their practical applications. Especially with respect to climate change and global energy and resource management, it is highly desirable to overcome these drawbacks by immobilization of CPAs. Transitioning to heterogeneous catalysis paves the road to broaden the use of CPAs because immobilized catalysts feature advantages such as facile separation, high recyclability, simple product isolation, and the possibility of an implementation into continuous-flow reactors.21−23

In contrast to conventional batch applications, flow systems generally benefit from enhanced heat- and mass transfer, improved safety, higher sustainability, and broad scalability.24−30 Moreover, flow reactions employing solid-state catalysts are perfectly suited to meet the high standards of recent “green sustainable chemistry” concepts.31−33

Although heterogeneous catalysis has played a crucial role in the chemical industry for decades, the impact of solid-state organocatalysts on the stereoselective synthesis of chiral products is often overlooked.34,35 In recent years, however, the field of heterogeneous organocatalysis has expanded considerably.

In 2010, Rueping et al. reported the immobilization of a polystyrene-supported CPA by cross-linking radical copolymerization, which was successfully used in an asymmetric transfer hydrogenation.36 One year later, Blechert and co-workers built up on this work and introduced 3-(anthracen-9-yl)-thiophene moieties in the 3,3′-positions of the BINOL backbone in order to obtain a microporous CPA by oxidative coupling under FeCl3 catalysis.37 The Pericas group has accomplished significant contributions in the whole field of immobilized catalysts for stereoselective transformations in batch and flow.38 In 2014, they prepared a solid CPA-based catalyst by immobilization of a BINOL-derivative on polystyrene, which was subsequently converted to the corresponding phosphoric acid over two steps in the presence of the support material. The catalyst was applied in a Friedel–Crafts reaction of indoles under batch and flow conditions and showed remarkable results in terms of activity, selectivity, and stability.39

An efficient strategy for the immobilization of the TRIP-CPA catalyst onto polystyrene and its use for batch and flow applications was also developed by the same group.40 This polymer-supported TRIP catalyst was successfully employed in different highly stereoselective reactions including allylation reactions of aldehydes40 and, very recently, a Pictet-Spengler cyclization.41 Further, Pericàs et al. could also show the preparation and application of SPINOL-derived CPAs immobilized also onto polystyrene.42

In 2022, the You group reported another copolymerization approach using external cross-linkers and applied their catalysts in an asymmetric transfer hydrogenation and dearomatization of β-naphthols, however, under batch conditions only.43

Conceptually, the reported immobilization methods can be divided into two different classes: (i) a copolymerization-like approach and (ii) the immobilization of protected BINOLs onto a solid support and subsequent introduction of the phosphoric acid on these modularly constructed immobilized diols. All these methods inevitably suffer from limitations regarding 3,3′-substituents, low-yielding precursor syntheses, difficult reaction monitoring during catalyst synthesis, undesirable swelling properties, and most importantly, their limitation to polystyrene as support material. A comparison as shown in Figure 1 clearly highlights the significant drawbacks of the known strategies. On the other hand, a direct immobilization of molecular CPAs is currently unprecedented but would allow for the preparation of immobilized chiral phosphoric acids (iCPAs) without all these limitations. We therefore envisioned to develop a conceptually novel strategy for the direct attachment of CPAs onto variable support materials that is completely modular, high yielding, and robust. Further, the catalysts should demonstrate their potential under both batch and flow conditions.

Figure 1.

Figure 1

Open in a new tab

Comparison of the conceptually different methods for the synthesis of supported CPAs. (a) Copolymerization approach by Rueping36, (b) immobilization of protected BINOLs by Pericàs,39 and (c) our new strategy. Bonds highlighted in red are formed for the fixation of the catalyst.

Results and Discussion

Synthesis of Immobilized CPAs (iCPAs)

In order to avoid any perturbation of the catalytically active site, the CPAs were immobilized at a remote position, namely, the 6-position of the BINOL backbone.39

Starting from commercially available, enantiomerically pure (R)-BINOL, the (R)-6-bromo-[1,1′-binaphthalene]-2,2′-diol, was synthesized on large scale and in quantitative yield over three steps by a reported procedure.44 We reasoned that the modification of the BINOL-scaffold should be as minimal as possible, and therefore, (R)-6-bromo-BINOL 4 was converted to the methoxy-derivative in an Ullmann-type reaction (Figure 2). After MOM protection of the BINOL hydroxy groups, subsequent ortho-iodination, and cross-coupling, a suitable precursor 5 was obtained up to 69% yield over four steps, with the iodination step as the bottleneck.

Figure 2.

Figure 2

Open in a new tab

Schematic overview of the synthesis of immobilized chiral phosphoric acids according to our new route.

Different cross-coupling reactions could be employed to introduce all common 3,3′-substituents. In our hands, a Suzuki reaction and a Kumada cross-coupling worked best. Next, the MOM-protecting groups were selectively cleaved off in quantitative yield, and then, the phosphoric acid was installed. For the synthesis of silica-supported catalysts, the phosphoric acid moiety must be introduced prior to immobilization to avoid an undesired achiral phosphorylation of the silica surface, resulting in a racemic background reaction when used as a catalyst.

A key step of our new strategy is the deprotection of the 6-methoxy group. Ishihara et al. reported a boron tribromide-assisted CPA catalyst.45 Thus, it was assumed that chiral phosphoric acids are compatible with BBr3 deprotection. Indeed, 6-hydroxy CPA was obtained in quantitative yield and further converted to propargyl ether 6. The overall yield for the complete sequence was exceptionally high, giving rise to the immobilizable CPA 6 in 68% over 11 steps.

In contrast to known procedures,36,39 the complete synthesis of the CPA proceeded under homogeneous conditions and was easily monitored by standard analytical methods. Functionalized CPA 6 represents the first catalytically active phosphoric acid that can be directly immobilized on a solid support. This, in turn, highlights one of the main advantages of this strategy: the catalytic performance before and after immobilization is directly comparable. Therefore, the influence of the backbone modification and solid support can be evaluated separately for the first time.

CPA 6 was then directly immobilized via an alkyne–azide click reaction under copper(I) catalysis (CuAAC). All solid support materials were pretreated (see Supporting Information) and functionalized with an azide-linker-unit according to reported procedures.46

Optimal conditions for the click reaction employed CuSO4 as a precatalyst under aqueous conditions, NaOH as a base, and the THPTA ligand. Sodium ascorbate was added to generate the Cu(I)-species in situ.47,48 Even though click chemistry has been applied as a late-stage immobilization method for various organocatalysts,49 including the BINOL-skeleton,50 it has not yet been applied to CPAs directly because the corresponding precursors (e.g., CPA 6) were not accessible. Conveniently, unreacted CPA 6 could be completely recovered from the solution after separation of the solid catalyst by filtration, preventing any loss of the precious chiral catalyst. Three differently substituted iCPAs were synthesized with excellent loadings of up to 0.22 mmol·g–1. All common 3,3′- moieties can be installed, and further, the solid support can be freely chosen. The latter was highlighted by employing our procedure to polystyrene (PS) and aluminum oxide supports, as well. Both support materials were readily functionalized, yielding the corresponding iCPAs with loadings of 0.54 mmol·g–1 (PS) and 0.10 mmol·g–1 (Al2O3), respectively. Differences in loadings are based on the azide loading of the support. The immobilization procedure that was optimized for silica could be applied directly to aluminum oxide without further modifications. Polystyrene, however, had to be treated differently due to its swelling properties. In order to efficiently functionalize the polymer, it must swell properly in the solvent. Therefore, we employed a strategy that was originally used for the immobilization of a MacMillan organocatalyst.51 In a mixture of DMF/THF with DIPEA as a base and catalytic amounts of CuI, the polystyrene support was easily functionalized, whereas the click reaction under aqueous conditions was unsuccessful. This observation during catalyst synthesis already implies a drawback of polystyrene-based iCPAs: the solvent-depending swelling properties can affect the accessibility of the catalytically active sites and thus the performance of the catalyst. Swelling is also accompanied by significant changes in the material volume, which led to overpressure and reactor blocking during preliminary flow experiments (especially solvent screening in flow). For these reasons, we focused our studies on silica supports that do not suffer from such behavior.

In summary, a novel strategy for the direct immobilization of CPAs was developed that distinguishes itself conceptually from known methods. Extraordinarily high yields for the precursor synthesis were achieved as well as broad flexibility with respect to the support material and CPA substitution pattern.

Characterization

The properties of the iCPA catalysts were then investigated thoroughly. The catalyst loading was determined with high precision by three different techniques, with elemental analysis by inductively coupled plasma-optical emission spectrometry (ICP-OES) being the main analytical method.

Since the formation of the phosphoric acid occurred in the absence of the solid support, it can be safely assumed that only one phosphorus species is present on the surface of the material. Therefore, the phosphorus content derived from ICP-OES directly correlates with the CPA loading. All results were further validated by thermogravimetric and elemental analysis. Figure 2 shows the average loading for all three catalysts.

Nitrogen sorption was performed to characterize the textural properties of the material throughout the immobilization. The average pore width of ca. 100 Å was preserved until CPA was introduced. After immobilization, the pore width was reduced to ca. 35 Å, which is in accordance with the expected spatial requirement of the CPA (deduced from density functional theory (DFT) optimized molecular structures, see SI). Moreover, the mesoporosity of the material remained intact throughout the process.

According to the scanning transmission electron microscopy/energy-dispersive X-ray spectroscopy (STEM/EDX) images (Figure 3b) of iCPA 3a, the morphology was demonstrated to be very homogeneous. Material contrast imaging for O, Si, C, N, and P also revealed a homogeneous elemental distribution.

Figure 3.

Figure 3

Open in a new tab

Analysis of the solid-state catalyst 3a. (a) Nitrogen sorption isotherms and pore width distribution for the iCPA 3a and the support material. (b) STEM/EDX images of 3a with the elemental distribution of O, Si, C, N, and P. (c) Solid-state-NMR of 3a (rotational side bands are marked with an asterisk).

Finally, solid-state-MAS NMR spectra on various nuclei (31P, 1H, 13C, 29Si, and 27Al) were recorded, proving a truly covalent immobilization of the CPAs.52 A representative 31P NMR spectrum of iCPA 3a depicted in Figure 3c shows a single peak at 4.29 ppm. To our delight, this observation gives clear evidence that only one phosphorus species is tightly anchored to the solid support, and further, the chemical shift is in excellent agreement with the data of the molecular precursor. Moreover, 1H and 13C NMR spectra confirmed the successful immobilization process (see SI for further detail). NMR spectra of the iCPAs immobilized onto PS and aluminum oxide (see SI) show similar results with single 31P signals in the expected range of chemical shift (3.73 ppm (PS) and 3.67 ppm (Al2O3), respectively).

Transfer Hydrogenation of 2-Substituted Quinolines

We first applied our iCPAs to an asymmetric transfer hydrogenation (TH) of 2-substituted quinoline derivatives 7 as published by Rueping et al.53

Batch Conditions and Recyclability

Prior to any flow experiments, the reaction was optimized under batch conditions (Table S1). A molecular version of catalyst 3a in chloroform was found to be optimal for this reaction. To our delight, the reaction proceeded smoothly at 60 °C giving rise to an excellent e.r. of >99:1 (S1, entry 1) at full conversion after only 0.5 h. With the use of co-catalytic amounts of HOAc as described by Antilla,54 the reaction temperature could be decreased to room temperature and the tetrahydroquinoline 8a was isolated with 92% yield and 99:1 e.r. (S1, entry 2).

To evaluate the influence of the linker, molecular CPA SI-3b was also tested under the same conditions and gave full conversion with a slightly diminished e.r. of 98:2. The same results were obtained by using iCPA 3a (S1, entries 3 and 4). It can therefore be safely concluded that the immobilization of the catalyst itself does not interfere with an effective stereoinduction.

Our initial concern was that the basic triazole unit could affect the catalytic performance of the iCPA. However, we did not observe any significant deactivation issues in our test reactions.

Once the optimal catalytic system was identified, one of the main objectives of this work was addressed, namely, the reusability of the iCPAs. In this respect, the silica-supported catalyst 3a was recycled ten times, giving consistently excellent results in terms of yield and enantioselectivity (Figure 4a). It is important to mention that no reactivation of the catalyst through an acidic wash between the recycling cycles was necessary. Thus, the immobilized catalyst was proven to be robust and highly recyclable. Using immobilized catalysts for this transformation saves considerable resources in the form of the precious CPA catalyst and, therefore, drastically improves both the environmental and economic performance of the catalytic reaction.

Figure 4.

Figure 4

Open in a new tab

Stereoselective transfer hydrogenation of 2-substituted quinolines with an iCPA under batch and flow conditions; TON in parentheses refers to the complete process (67 h). Conversion over time of the flow process was determined by HPLC. Optimized flow conditions: c(7,12) = 0.05 mol·L–1, V̇ = 250 μL·min–1 (τ = 4 min), rt, 10 mol % HOAc and tBu based dihydropyridine 11b. The absolute configuration of all products was assigned by comparison of optical rotation to the literature.53

Application of Additional Solid Supports

Further, we envisioned the newly developed method to be applicable to a broad range of support materials. To confirm this hypothesis, the influence of the solid support on the catalytic activity was investigated (S1, entries 5 and 6). Both the aluminum oxide- and polystyrene-supported iCPAs 9 and 10 catalyzed the transfer hydrogenation of 7a with the same yield and enantioselectivity as the silica-supported iCPA 3a (>98%, 98:2 e.r.). Consequently, this strategy is fully applicable to all three support materials, which in turn allows for an accurate adjustment of the solid support to the specific needs of any given transformation.

In order to conduct the reaction under continuous-flow conditions, tBu-substituted Hantzsch ester 11b was used instead of the ethyl ester-based derivative. This modification not only resulted in an improved solubility and shorter reaction times but also gave rise to even superior results of 97% isolated yield and >99:1 e.r. (S1, entry 7).

Flow Conditions

With the optimized batch reaction conditions established, the process was transferred to flow conditions (Table S2). In the continuous-flow setup, a commercially available HPLC column (ø 4 mm, length 100 mm) was employed as a fixed-bed reactor and packed with 770 mg of iCPA 3a (f = 0.21 mmol·g–1). Initially, a solution of quinoline and dihydropyridine was pumped through the reactor at a total flow rate of 100 μL·min–1 and 70 °C. Pleasingly, full conversion and an enantiomeric ratio of 96.5:3.5 were obtained. Nevertheless, e.r. decreased over time. This is likely the result of catalyst inhibition by the pyridine byproduct, deteriorating the catalyst performance. However, the addition of 10 mol % HOAc solved this issue.

After extensive optimization (SI), the product was obtained in 91% yield with an e.r. of 98:2 within only 4 min residence time (at 250 μL·min–1), highlighting the dramatic acceleration of this reaction with the new iCPA under continuous-flow conditions (Figure 4b). Interestingly, lowering the flow rate led to decreased enantioselectivities in this case, which is most likely attributed to less effective mixing.

With the optimized batch and flow conditions in hand, we next investigated the substrate scope. Variation of the aryl component in the 2-position of the quinolines was easily possible as demonstrated by introducing electron-neutral (Ph), as well as electron-donating (PMP) and -withdrawing (p-Br-Ph) substituents. All examples gave rise to the corresponding tetrahydroquinolines in over 90% yield and excellent enantioselectivity of 98:2 → 99:1 e.r. both in batch and flow experiments. In addition, 1,4-benzoxazine 12 performed likewise in this process with yields >90%, an e.r. of >99:1 (batch) and 96.5:3.5 (flow).

To highlight the potential of this process for a large-scale synthesis, the reaction was conducted for a 26-fold extended time-on-stream. The system was operated in continuous-flow for 19 h, and the enantioselectivity remained excellent throughout the process with complete conversion. Starting with 14.2 mmol of 12, the highly enantioenriched dihydro-1,4-benzoxazine 13 was obtained with an overall yield of 93% (corresponding to 2.78 g) and 96:4 e.r. This translates into a remarkable turnover number (TON) of 82 and a productivity of 4.5 mmol(product)·mmol(Cat.)−1·h–1. In other words, the catalyst loading of the global process was as low as 1 mol %, representing a 5-fold decrease with respect to the batch conditions. We then attempted to push our system to its limits by further extending the time-on-stream to 67 h. After 19 h, the enantiomeric ratio started to decline slowly, as indicated by HPLC measurements (Figure 4b), which is potentially attributed to a slow accumulation of the pyridine byproduct and subsequent salt formation with the iCPA in the reactor. Therefore, a second fraction of the product was collected separately. The same overall yield of 93% was obtained for the second fraction, but the enantioselectivity dropped to 91.5:8.5 e.r. Nevertheless, we obtained another 6.98 g of the enantiomerically enriched product 13, leading to a total yield of 9.76 g and an accumulated TON of 289. Compared with the literature, the observed TONs are on par with other examples in the field of heterogeneous Brønsted acid catalysis. For example, Pericàs et al. reported an accumulated TON of 102 for their aza-Friedel–Crafts addition of indoles to N-tosylimines,39 243 for a desymmetrization of 3,3-disubstituted oxetanes with supported SPINOL-CPAs,42 282 for their asymmetric allylation of aldehydes,40 and 43 for a Pictet-Spengler cyclization41 (both latter reactions catalyzed by supported TRIP).

To further investigate if the slight deactivation during the long-term experiment is reversible, the complete setup was flushed for 30 min (at 250 μL·min–1) with CHCl3 (+HOAc additive). To our delight, a subsequent test run with the reactivated catalyst showed the same results as those observed at the beginning of the large-scale reaction. A reactivation phase in between flow reactions is a common solution to counteract reversible deactivation processes during upscaling of flow reactions.41 In addition, transferring the reaction to continuous-flow heavily impacts the space-time yield (STY). For the homogeneous batch process, a STY of 30 mmol·L–1·h–1 was observed (reaction time 1.5 h, tBu-substituted Hantzsch ester 11b was used). Based upon the short residence time (τ = 4 min) and small reactor volume, the STY is significantly increased by a factor of 23 to 700 mmol·L–1·h–1. It is important to note that the exact same catalyst material was used for all flow experiments, i.e., optimization, substrate scope, and large-scale reaction, leading to a cumulated total running time of over 100 h and a TON of 372. After this extended time of use, iCPA was again submitted to elemental analysis to quantify possible leaching. It was found that under our reaction conditions, the loading only slightly decreased by 5% (from 0.21 to 0.20 mmol·g–1). Besides, a 31P NMR of the used catalyst was measured, indicating no chemical modification of the CPA. In combination with the excellent performance of the catalyst even after it was used for this extended time, this proves that leachability is negligible.

Friedländer/TH-Reaction

We then extended our studies to a sequential two-step flow process, the Friedländer quinoline synthesis–transfer hydrogenation cascade affording a variety of 2,3-disubstituted tetrahydroquinolines. This reaction was developed by Gong et al., who used a mixture of Mg(OTf)2 and a CPA as catalysts for the two individual steps.55 For this scenario, flow chemistry is ideally suited as one can employ two different fixed-bed reactors with two immobilized molecular catalysts sequentially (Figure 5). After extensive screening, we found commercially available immobilized phosphonic acid 16 to be an ideal catalyst for the Friedländer condensation giving rise to complete consumption of the starting material in short time under metal-free conditions (Table S3). Since both transformations occurred separately in two different reactors, it was possible to employ a strong achiral acid for the first step without any impact on the stereoselectivity of the asymmetric transfer hydrogenation. Under batch conditions, this would require additional filtration and workup steps, rendering it rather inconvenient. For the hydrogenation step, 2,4,6-Cy3-Ph-substituted CPA was identified as the optimal catalyst (Table S4). Upon further optimization of reaction parameters, including Hantzsch esters, solvents, additives, and stoichiometry, we established the ideal conditions for the batch process (Tables S5 and S6) giving rise to the product in 76% yield, with 96:4 e.r. (major diastereomer), 91:9 e.r. (minor diastereomer), and 5.8:1 d.r.

Figure 5.

Figure 5

Open in a new tab

Friedländer-transfer hydrogenation sequence for the synthesis of 2,3-substituted tetrahydroquinolines in flow. The best results were obtained using 5 mol % HOAc as a cocatalyst and 5.0 equiv of dihydropyridine 11a in chloroform (c = 0.05 mol·L–1). a) The 2-Ph-substituted product 17d was identified to be cis-configured in accordance with spectroscopic data from literature.56

In the flow setup, the first column (ø 4 mm, length 250 mm) was packed with 2000 mg of immobilized phosphonic acid 16 to catalyze the Friedländer reaction of ortho-amino benzaldehydes 14 and β-keto esters 15 with a flow rate of 100 μL·min–1 and a residence time of 21 min at 60 °C. Subsequently, Hantzsch ester 11b was added to the stream at 100 μL·min–1 and the mixture was passed through the second column (ø 4 mm, length 550 mm) containing 4380 mg of iCPA 3b. A combined flow rate of 200 μL·min–1 led to a total residence time of 23 min for the transfer hydrogenation step. In this continuous-flow operation, product 17a was isolated in 68% yield, 95:5 e.r. (major) and 5.2:1 d.r. (S7, entries 3–4).

We next investigated the substrate scope (Figure 5) in comparison between reactions run with the molecular catalyst in batch and the immobilized catalyst in flow. The first examples reveal a broad variability of the ester substituents giving rise to the 2,3-disubstituted tetrahydroquinolines in typically good yields (batch: 27–76%, flow: 58–71%), very good enantioselectivities for the major diastereomer (batch: 96:4 e.r., flow: 93:7 to 95:5 e.r.) and moderate d.r. (batch: 3.5:1 to 5.8:1, flow: 2.7:1 to 5.2:1). Introducing a methyl ester in the 3-position (17c) led to diminished yields under batch conditions due to the formation of side products. In the flow process, the reaction proceeded more cleanly and gave rise to an improved yield of 71%. The 2-substituent could also be changed to a phenyl group. Notably, in the batch reaction of this quinoline, product 17d was obtained with only 49% yield, 2.7:1 d.r., and 94:6 e.r. (major diastereomer). In the corresponding flow process, however, this tetrahydroquinoline was isolated as a single diastereomer in 56% yield with 94:6 e.r. Thus, the migration of the process into continuous-flow resulted not only in an increased yield, but more importantly, the diastereoselectivity was significantly improved.

An X-ray crystal structure analysis of product 17e revealed the trans-configuration of the two stereocenters, which was adopted for all other 2-methyl-substituted products as well.

Mannich Reaction

Finally, the versatility of the iCPAs was assessed in a low-temperature C–C-bond-forming reaction. To the best of our knowledge, a continuous-flow reaction with an immobilized chiral Brønsted acid catalyst under cryogenic conditions has not been reported yet.

Based on our interest in Brønsted acid-catalyzed Mannich reactions,9,10 we studied the reaction of in situ formed aliphatic imines 19 and cyclic silyl dienolates 20. An extensive optimization under batch conditions showed that this reaction worked best in ethereal solvents, especially THF, with co-catalytic amounts of DMPU at temperatures below −50 °C (Table S8). It was found to be crucial to keep the temperature below −25 °C to avoid side reactions. The best results were obtained in the presence of a 2,6-Me2-4-Ph-Ph-based CPA (10 mol %) and 20 mol % of DMPU at −50 °C in THF, giving rise to γ-amino ester 21 in quantitative yield as a single diastereomer with an excellent enantiomeric ratio of >99:1. Under the optimized conditions, the reaction proceeded to full conversion in less than 10 min.

Drawing on preliminary experiments with conventional column-based flow reactors, it became clear that mixing of the imine and nucleophile 20 must occur in the presence of the chiral catalyst to prevent a racemic background reaction. Thus, a self-made flow reactor with an integrated mixing unit was designed to solve this issue. In the flow process, first, amine 22 (c = 0.20 mol·L–1) and aldehyde 23 (c = 0.27 mol·L–1) were pumped through a mixer at a flow rate of 100 μL·min–1 each to rapidly generate the imine at −50 °C (residence time 60 s). Afterward, the imine solution (combined flow rate 200 μL·min–1) was passed through the self-made T-shaped stainless-steel reactor (loaded with 5515 mg of iCPA 3c) at the same temperature. Using a third pump, the silyl dienolate was introduced inside the reactor (flow rate 200 μL·min–1, c = 0.4 mmol·L–1) to ensure that imine and the nucleophile react with each other only in the presence of the catalyst.

With this setup, the flow reaction was optimized regarding flow rates, stoichiometry, and temperature (see SI).

For the substrate scope (Figure 6), several aliphatic aldehydes were submitted to the optimized homogeneous batch and heterogeneous flow conditions. In general, all products were obtained as single diastereomers and with excellent yields and enantioselectivities from the flow process, which were very similar in comparison to the batch process (batch: 97:3 to >99:1 e.r., flow: 95:5 to 97.5:2.5 e.r.) To our delight, very similar results were obtained when 2-Me-THF, a green solvent alternative, was used, allowing for environmentally yet more benign reaction conditions. Our studies represent the first successful implementation of an immobilized chiral catalyst into an automated continuous-flow processes under cryogenic temperatures.57

Figure 6.

Figure 6

Open in a new tab

Vinylogous Mukaiyama–Mannich reaction with β,γ-bridged silyl dienolates and in situ generated aliphatic imines under cryogenic flow conditions.

Conclusions

In summary, a very robust, high yielding, and fully modular route toward immobilized chiral BINOL-phosphoric acids was developed. This conceptually unprecedented approach features prominently a broad flexibility with respect to the solid support material and catalyst substitution pattern. The iCPAs were fully characterized with ICP-OES and nitrogen sorption measurements, thermogravimetric and elemental analyses, STEM/EDX images, and solid-state NMR spectroscopy. Subsequently, the iCPAs were employed in three different continuous-flow processes including C–H and C–C-bond-forming reactions and a sequential C–C-/C–H-bond-forming reaction. In all cases, they gave rise to outstanding levels of productivity, space-time yields, and enantioselectivity. This strategy represents a breakthrough in the transition of stereoselective CPA catalysis to environmentally more benign and economically more valuable continuous-flow conditions and has the potential to expand the scope of applications significantly.

Acknowledgments

This work was generously supported by the European Social Fund (Heterogeneously catalyzed synthetic processes in flow systems) in the form of a doctoral fellowship awarded to M.S. M.L. is grateful for a predoctoral fellowship provided by the Deutsche Bundesstiftung Umwelt. The authors thank Dr. Peter Lönnecke (University of Leipzig) for obtaining the X-ray crystal structure analysis. The authors thank Prof. Dr. Burkhard König (University of Regensburg) for providing access to EA measurements and Dr. David Poppitz (University of Leipzig) for HR-TEM images. The authors thank Sebastian Witte for the experimental support. The authors also thank the glassblowers and fine mechanics workshop at University of Leipzig for support in manufacturing flow reactors.

Glossary

Abbreviations

CPA

chiral phosphoric acid

iCPA

immobilized chiral phosphoric acid

TH

transfer hydrogenation

DMPU

N,N′-dimethylpropyleneurea

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acscatal.4c00985.

  • Additional experimental details; NMR spectra of all compounds and materials; HPLC chromatograms; crystallographic details; details on DFT calculations (PDF)

Author Contributions

M.L. developed the catalyst synthesis and immobilization procedure, planned the cryo-flow reactor design, developed the Mannich reaction, and performed initial screening under flow conditions. M.S. performed the catalyst synthesis, investigated and optimized the reactions under batch/flow conditions, and performed the scope. M.G. provided the azido-functionalized, end-capped support material, and interpretation of nitrogen-adsorption and thermogravimetric analysis. W.B. performed solid-state-NMR measurements and interpretation. M.L., M.S., and W.B. wrote the manuscript with input from all authors. R.G. supervised the pretreatment and characterization of silica materials. C.S. conceived and supervised the entire project.

The authors declare no competing financial interest.

Supplementary Material

cs4c00985_si_001.pdf (7.5MB, pdf)

References

  1. Akiyama T.; Itoh J.; Yokota K.; Fuchibe K. Enantioselective Mannich-type reaction catalyzed by a chiral Brønsted acid. Angew. Chem., Int. Ed. 2004, 43 (12), 1566–1568. 10.1002/anie.200353240. [DOI] [PubMed] [Google Scholar]
  2. Uraguchi D.; Terada M. Chiral Brønsted acid-catalyzed direct Mannich reactions via electrophilic activation. J. Am. Chem. Soc. 2004, 126 (17), 5356–5357. 10.1021/ja0491533. [DOI] [PubMed] [Google Scholar]
  3. Uraguchi D.; Sorimachi K.; Terada M. Organocatalytic asymmetric aza-Friedel-Crafts alkylation of furan. J. Am. Chem. Soc. 2004, 126 (38), 11804–11805. 10.1021/ja046185h. [DOI] [PubMed] [Google Scholar]
  4. Rowland G. B.; Rowland E. B.; Liang Y.; Perman J. A.; Antilla J. C. The highly enantioselective addition of indoles to N-acyl imines with use of a chiral phosphoric acid catalyst. Org. Lett. 2007, 9 (14), 2609–2611. 10.1021/ol0703579. [DOI] [PubMed] [Google Scholar]
  5. Feng J.; Yan W.; Wang D.; Li P.; Sun Q.; Wang R. The highly enantioselective addition of indoles and pyrroles to isatins-derived N-Boc ketimines catalyzed by chiral phosphoric acids. Chem. Commun. 2012, 48 (64), 8003–8005. 10.1039/c2cc33200k. [DOI] [PubMed] [Google Scholar]
  6. Jiang F.; Zhao D.; Yang X.; Yuan F.-R.; Mei G.-J.; Shi F. Catalyst-Controlled Chemoselective and Enantioselective Reactions of Tryptophols with Isatin-Derived Imines. ACS Catal. 2017, 7 (10), 6984–6989. 10.1021/acscatal.7b02279. [DOI] [Google Scholar]
  7. Zhou J.; Zhu G.-D.; Wang Le.; Tan F.-X.; Jiang W.; Ma Z.-G.; Kang J.-C.; Hou S.-H.; Zhang S.-Y. Remote C6-Enantioselective C-H Functionalization of 2,3-Disubstituted Indoles through the Dual H-Bonds and π-π Interaction Strategy Enabled by CPAs. Org. Lett. 2019, 21 (21), 8662–8666. 10.1021/acs.orglett.9b03276. [DOI] [PubMed] [Google Scholar]
  8. Sheng F.-T.; Li Z.-M.; Zhang Y.-Z.; Sun L.-X.; Zhang Y.-C.; Tan W.; Shi F. Atroposelective Synthesis of 3,3′-Bisindoles Bearing Axial and Central Chirality: Using Isatin-Derived Imines as Electrophiles. Chin. J. Chem. 2020, 38 (6), 583–589. 10.1002/cjoc.202000022. [DOI] [Google Scholar]
  9. Sickert M.; Schneider C. Enantioselektive, Brønsted-Säure-katalysierte, vinyloge Mannich- Reaktion. Angew. Chem. 2008, 120 (19), 3687–3690. 10.1002/ange.200800103. [DOI] [Google Scholar]
  10. Giera D. S.; Sickert M.; Schneider C. Brønsted acid-catalyzed, enantioselective, vinylogous Mannich reaction of vinylketene silyl N,O-acetals. Org. Lett. 2008, 10 (19), 4259–4262. 10.1021/ol8017374. [DOI] [PubMed] [Google Scholar]
  11. Guo Q.-X.; Liu H.; Guo C.; Luo S.-W.; Gu Y.; Gong L.-Z. Chiral Brønsted acid-catalyzed direct asymmetric Mannich reaction. J. Am. Chem. Soc. 2007, 129 (13), 3790–3791. 10.1021/ja068236b. [DOI] [PubMed] [Google Scholar]
  12. Itoh J.; Fuchibe K.; Akiyama T. Chiral Brønsted acid catalyzed enantioselective aza-Diels-Alder reaction of Brassard’s diene with imines. Angew. Chem., Int. Ed. 2006, 45 (29), 4796–4798. 10.1002/anie.200601345. [DOI] [PubMed] [Google Scholar]
  13. Momiyama N.; Tabuse H.; Terada M. Chiral phosphoric acid-governed anti-diastereoselective and enantioselective hetero-Diels-Alder reaction of glyoxylate. J. Am. Chem. Soc. 2009, 131 (36), 12882–12883. 10.1021/ja904749x. [DOI] [PubMed] [Google Scholar]
  14. Held F. E.; Grau D.; Tsogoeva S. B. Enantioselective Cycloaddition Reactions Catalyzed by BINOL-Derived Phosphoric Acids and N-Triflyl Phosphoramides: Recent Advances. Molecules 2015, 20 (9), 16103–16126. 10.3390/molecules200916103. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Rueping M.; Sugiono E.; Azap C. A highly enantioselective Brønsted acid catalyst for the Strecker reaction. Angew. Chem., Int. Ed. 2006, 45 (16), 2617–2619. 10.1002/anie.200504344. [DOI] [PubMed] [Google Scholar]
  16. Rueping M.; Sugiono E.; Moreth S. A. Metal-Free, Enantioselective Strecker Reactions Catalyzed by Chiral BINOL and TADDOL Catalysts. Adv. Synth. Catal. 2007, 349 (4–5), 759–764. 10.1002/adsc.200600405. [DOI] [Google Scholar]
  17. Zamfir A.; Tsogoeva S. B. Asymmetric hydrocyanation of hydrazones catalyzed by in situ formed O-silylated BINOL-phosphate: a convenient access to versatile alpha-hydrazino acids. Org. Lett. 2010, 12 (1), 188–191. 10.1021/ol9025974. [DOI] [PubMed] [Google Scholar]
  18. Parmar D.; Sugiono E.; Raja S.; Rueping M. Complete field guide to asymmetric BINOL-phosphate derived Brønsted acid and metal catalysis: history and classification by mode of activation; Brønsted acidity, hydrogen bonding, ion pairing, and metal phosphates. Chem. Rev. 2014, 114 (18), 9047–9153. 10.1021/cr5001496. [DOI] [PubMed] [Google Scholar]
  19. del Corte X.; de Marigorta E. M.; Palacios F.; Vicario J.; Maestro A. An overview of the applications of chiral phosphoric acid organocatalysts in enantioselective additions to C=O and C=N bonds. Org. Chem. Front. 2022, 9 (22), 6331–6399. 10.1039/D2QO01209J. [DOI] [Google Scholar]
  20. Schenker S.; Zamfir A.; Freund M.; Tsogoeva S. B. Developments in Chiral Binaphthyl-Derived Brønsted/Lewis Acids and Hydrogen-Bond-Donor Organocatalysis. Eur. J. Org. Chem. 2011, 2011 (12), 2209–2222. 10.1002/ejoc.201001538. [DOI] [Google Scholar]
  21. Jas G.; Kirschning A. Continuous Flow Techniques in Organic Synthesis. Chem. - Eur. J. 2003, 9 (23), 5708–5723. 10.1002/chem.200305212. [DOI] [PubMed] [Google Scholar]
  22. Kirschning A.; Jas G.. Applications of Immobilized Catalysts in Continuous Flow Processes. In Immobilized Catalysts; Springer, 2004; Vol. 242, pp 209–239. [DOI] [PubMed] [Google Scholar]
  23. Tsubogo T.; Ishiwata T.; Kobayashi S. Asymmetric Carbon-Carbon Bond Formation under Continuous-Flow Conditions with Chiral Heterogeneous Catalysts. Angew. Chem., Int. Ed. 2013, 52 (26), 6590–6604. 10.1002/anie.201210066. [DOI] [PubMed] [Google Scholar]
  24. Wiles C.; Watts P. Continuous flow reactors: a perspective. Green Chem. 2012, 14 (1), 38–54. 10.1039/C1GC16022B. [DOI] [Google Scholar]
  25. Ley S. V.; Baxendale I. R. New tools and concepts for modern organic synthesis. Nat. Rev. Drug Discovery 2002, 1 (8), 573–586. 10.1038/nrd871. [DOI] [PubMed] [Google Scholar]
  26. Geyer K.; Codée J. D. C.; Seeberger P. H. Microreactors as Tools for Synthetic Chemists—The Chemists’ Round-Bottomed Flask of the 21st Century?. Chem. - Eur. J. 2006, 12 (33), 8434–8442. 10.1002/chem.200600596. [DOI] [PubMed] [Google Scholar]
  27. Hartman R. L.; McMullen J. P.; Jensen K. F. Deciding Whether To Go with the Flow: Evaluating the Merits of Flow Reactors for Synthesis. Angew. Chem., Int. Ed. 2011, 50 (33), 7502–7519. 10.1002/anie.201004637. [DOI] [PubMed] [Google Scholar]
  28. Wegner J.; Ceylan S.; Kirschning A. Ten key issues in modern flow chemistry. Chem. Commun. 2011, 47 (16), 4583. 10.1039/c0cc05060a. [DOI] [PubMed] [Google Scholar]
  29. McQuade D. T.; Seeberger P. H. Applying flow chemistry: methods, materials, and multistep synthesis. J. Org. Chem. 2013, 78 (13), 6384–6389. 10.1021/jo400583m. [DOI] [PubMed] [Google Scholar]
  30. Plutschack M. B.; Pieber B.; Gilmore K.; Seeberger P. H. The Hitchhiker’s Guide to Flow Chemistry ∥. Chem. Rev. 2017, 117 (18), 11796–11893. 10.1021/acs.chemrev.7b00183. [DOI] [PubMed] [Google Scholar]
  31. Anastas P. T.; Warner J. C.. Green Chemistry Theory and Practice; Oxford Univ. Press, 1998. [Google Scholar]
  32. Tsubogo T.; Oyamada H.; Kobayashi S. Multistep continuous-flow synthesis of (R)- and (S)-rolipram using heterogeneous catalysts. Nature 2015, 520 (7547), 329–332. 10.1038/nature14343. [DOI] [PubMed] [Google Scholar]
  33. Sheldon R. A.; Arends I. W. C. E.; Hanefeld U.. Green Chemistry. In Green Chemistry and Catalysis; Wiley-VCH, 2008. [Google Scholar]
  34. Mishra K.; Guyon D.; San Martin J.; Yan Y. Chiral Perovskite Nanocrystals for Asymmetric Reactions: A Highly Enantioselective Strategy for Photocatalytic Synthesis of N-C Axially Chiral Heterocycles. J. Am. Chem. Soc. 2023, 145, 17242–17252. 10.1021/jacs.3c04593. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Rothenberg G.Catalysis: Concepts and Green Applications; John Wiley & Sons, 2017. [Google Scholar]
  36. Rueping M.; Sugiono E.; Steck A.; Theissmann T. Synthesis and Application of Polymer-Supported Chiral Brønsted Acid Organocatalysts. Adv. Synth. Catal. 2010, 352 (2–3), 281–287. 10.1002/adsc.200900746. [DOI] [Google Scholar]
  37. Bleschke C.; Schmidt J.; Kundu D. S.; Blechert S.; Thomas A. A Chiral Microporous Polymer Network as Asymmetric Heterogeneous Organocatalyst. Adv. Synth. Catal. 2011, 353 (17), 3101–3106. 10.1002/adsc.201100674. [DOI] [Google Scholar]
  38. Rodríguez-Escrich C.; Pericàs M. A. Organocatalysis on Tap: Enantioselective Continuous Flow Processes Mediated by Solid-Supported Chiral Organocatalysts. Eur. J. Org. Chem. 2015, 2015 (6), 1173–1188. 10.1002/ejoc.201403042. [DOI] [Google Scholar]
  39. Osorio-Planes L.; Rodríguez-Escrich C.; Pericàs M. A. Enantioselective continuous-flow production of 3-indolylmethanamines mediated by an immobilized phosphoric acid catalyst. Chem. - Eur. J. 2014, 20 (8), 2367–2372. 10.1002/chem.201303860. [DOI] [PubMed] [Google Scholar]
  40. Clot-Almenara L.; Rodríguez-Escrich C.; Osorio-Planes L.; Pericàs M. A. Polystyrene-Supported TRIP: A Highly Recyclable Catalyst for Batch and Flow Enantioselective Allylation of Aldehydes. ACS Catal. 2016, 6 (11), 7647–7651. 10.1021/acscatal.6b02621. [DOI] [Google Scholar]
  41. Chaudhari M. B.; Gupta P.; Llanes P.; Pericàs M. A. Polymer-Supported Phosphoric-Acid Catalysed Enantioselective Pictet-Spengler Cyclisation for the Synthesis of Quaternary Tryptolines in Batch/Continuous Flow. Adv. Synth. Catal. 2023, 365 (4), 527–534. 10.1002/adsc.202201275. [DOI] [Google Scholar]
  42. Lai J.; Fianchini M.; Pericàs M. A. Development of Immobilized SPINOL-Derived Chiral Phosphoric Acids for Catalytic Continuous Flow Processes. Use in the Catalytic Desymmetrization of 3,3-Disubstituted Oxetanes. ACS Catal. 2020, 10 (24), 14971–14983. 10.1021/acscatal.0c04497. [DOI] [Google Scholar]
  43. Huang X.-Y.; Zheng Q.; Zou L.-M.; Gu Q.; Tu T.; You S.-L. Hyper-Crosslinked Porous Chiral Phosphoric Acids: Robust Solid Organocatalysts for Asymmetric Dearomatization Reactions. ACS Catal. 2022, 12 (8), 4545–4553. 10.1021/acscatal.2c00397. [DOI] [Google Scholar]
  44. Cai D.; Larsen R. D.; Reider P. J. Efficient synthesis of 6-mono-bromo-1,1′-bi-2-naphthol. Tetrahedron Lett. 2002, 43 (22), 4055–4057. 10.1016/S0040-4039(02)00730-X. [DOI] [Google Scholar]
  45. Hatano M.; Goto Y.; Izumiseki A.; Akakura M.; Ishihara K. Boron Tribromide-Assisted Chiral Phosphoric Acid Catalyst for a Highly Enantioselective Diels-Alder Reaction of 1,2-Dihydropyridines. J. Am. Chem. Soc. 2015, 137 (42), 13472–13475. 10.1021/jacs.5b08693. [DOI] [PubMed] [Google Scholar]
  46. Moni L.; Ciogli A.; D’Acquarica I.; Dondoni A.; Gasparrini F.; Marra A. Synthesis of sugar-based silica gels by copper-catalysed azide-alkyne cycloaddition via a single-step azido-activated silica intermediate and the use of the gels in hydrophilic interaction chromatography. Chem. - Eur. J. 2010, 16 (19), 5712–5722. 10.1002/chem.201000106. [DOI] [PubMed] [Google Scholar]
  47. Chan T. R.; Hilgraf R.; Sharpless K. B.; Fokin V. V. Polytriazoles as copper(I)-stabilizing ligands in catalysis. Org. Lett. 2004, 6 (17), 2853–2855. 10.1021/ol0493094. [DOI] [PubMed] [Google Scholar]
  48. Hong V.; Presolski S. I.; Ma C.; Finn M. G. Analysis and optimization of copper-catalyzed azide-alkyne cycloaddition for bioconjugation. Angew. Chem., Int. Ed. 2009, 48 (52), 9879–9883. 10.1002/anie.200905087. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Rodríguez-Escrich C.; Pericàs M. A. Catalytic Enantioselective Flow Processes with Solid-Supported Chiral Catalysts. Chem. Rec. 2019, 19 (9), 1872–1890. 10.1002/tcr.201800097. [DOI] [PubMed] [Google Scholar]
  50. Yadav J.; Stanton G. R.; Fan X.; Robinson J. R.; Schelter E. J.; Walsh P. J.; Pericas M. A. Asymmetric allylation of ketones and subsequent tandem reactions catalyzed by a novel polymer-supported titanium-BINOLate complex. Chem. - Eur. J. 2014, 20 (23), 7122–7127. 10.1002/chem.201400204. [DOI] [PubMed] [Google Scholar]
  51. Riente P.; Yadav J.; Pericàs M. A. A click strategy for the immobilization of MacMillan organocatalysts onto polymers and magnetic nanoparticles. Org. Lett. 2012, 14 (14), 3668–3671. 10.1021/ol301515d. [DOI] [PubMed] [Google Scholar]
  52. Ernst M.; Ernst R. R. Heteronuclear Dipolar Cross-Correlated Cross Relaxation for the Investigation of Side-Chain Motions. J. Magn. Reson., Ser. A 1994, 110 (2), 202–213. 10.1006/jmra.1994.1206. [DOI] [Google Scholar]
  53. Rueping M.; Antonchick A. P.; Theissmann T. A highly enantioselective Brønsted acid catalyzed cascade reaction: organocatalytic transfer hydrogenation of quinolines and their application in the synthesis of alkaloids. Angew. Chem., Int. Ed. 2006, 45 (22), 3683–3686. 10.1002/anie.200600191. [DOI] [PubMed] [Google Scholar]
  54. Li G.; Antilla J. C. Highly enantioselective hydrogenation of enamides catalyzed by chiral phosphoric acids. Org. Lett. 2009, 11 (5), 1075–1078. 10.1021/ol802860u. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Ren L.; Lei T.; Ye J.-X.; Gong L.-Z. Step-economical synthesis of tetrahydroquinolines by asymmetric relay catalytic Friedländer condensation/transfer hydrogenation. Angew. Chem., Int. Ed. 2012, 51 (3), 771–774. 10.1002/anie.201106808. [DOI] [PubMed] [Google Scholar]
  56. Chen Z.-P.; Ye Z.-S.; Chen M.-W.; Zhou Y.-G. Enantioselective Synthesis of Endocyclic β-Amino Acids with Two Contiguous Stereocenters via Hydrogenation of 3-Alkoxycarbonyl-2-Substituted Quinolines. Synthesis 2013, 45 (23), 3239–3244. 10.1055/s-0033-1339849. [DOI] [Google Scholar]
  57. The only other example entails a purely gravity-induced flow process with an immobilized quinine catalyst employed for β-lactam synthesis, see Lectka et al.:; Hafez A. M.; Taggi A. E.; Dudding T.; Lectka T. Asymmetric catalysis on sequentially-linked columns. J. Am. Chem. Soc. 2001, 123 (44), 10853–10859. 10.1021/ja016556j. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

cs4c00985_si_001.pdf (7.5MB, pdf)

Articles from ACS Catalysis are provided here courtesy of American Chemical Society

RESOURCES